Symmery and Structure in The Solid State

INDEX
S. No Topic Page No
Week 1
1 Symmetry in 3D World 1
2 Two Fold Axis Representation with the Help of Esher Diagrams 13
3 Pure Rotation Axes 24
4 Properties of Crystal 40
5 Point Group Generation 52
Week 2
6 Combination ofSymmetry Elements 65
7 Arrangement of Symmetry Equivalent Objects 80
8 Introduction to Plane Lattices 93
9 Bravais Lattices 106
10 Details of Stereographic Projections 121
Week 3
11 Stereographic Projections (Continued) 132
12 Point Group and Crystal Systems 1 142
13 Point Group and Crystal Systems 2 155
14 Point Groups to Space Groups 169
15 Translation componentsin Monoclinic System 181
Week 4
16 Additional Symmetry Elements 195
17 Additional Symmetry Elements (Continued) 210
18 Space Groups 1 223
Week 5
22 Additional Information on Space Groups 273
23 Details of Space Groups 1 291
25 Details of Space Group 3 319
Week 6
27 Crystal Structure of Calcium Carbonate 347
28 Crystal Structure of Some Minerals 360
29 Atoms in the Crystal: Positions and Relevant Properties 376
30 Crystallographic Directions and Planes 390
Week 7
31 Interference of Waves 405
32 X Ray Scattering ; optical Analogy 419
33 X Ray Scattering ; Fourier transforms 433
34 X Ray Scattering; Deriving Laue Conditions from scattering theory 446
X Ray Scattering ; Laue conditions to Bragg â€˜s Law, Introduction to
35 Reciprocal lattice 459
Week 8
36 Bragg's Law in Reciprocal Space 1 474
37 Bragg's Law in Reciprocal Space 2 488
38 Calculation of Intensities 1 502
39 Calculation of Intensities 2 516
40 Conversion from Direct to reciprocal space, the inverse relations 531
Week 9
41 Diffraction and Reciprocal Space (Continued) 544
42 Limits of Resolution 557
43 Concept of Structure Factors 571
44 Systematic Absences 1 583
Week 10
47 Friedel's Law and Laue classes 630
48 Experimental Aspects of Data Collection 647
49 Structure Determination 1 661
50 Structure Determination 2 677
Week 11
51 Data Reduction 696
52 Fourier Syntheses 710
53 Patterson Method 1 726
54 Patterson Method2 740
55 Direct Method 749
Week 12
56 Powder Diffraction 1 769
59 Quantum Crystallography 1 812
60 Quantum Crystallography 2 827
61 Intermolecular Interactions 839
Symmetry and Structure in the Solid State
Prof. T.N. Guru Row
Department of Chemistry and Biochemistry
Indian Institute of Science, Bangalore
Lecture – 01
Symmetry in 3D World
Welcome to this course on Symmetry and Structure in the Solid State. This would be a
12 week course which is offered under NPTEL which means that there is a total of 30
hours which we spend to cover the essentials of symmetry and structure determination.
(Refer Slide Time: 00:35)
And of course, since we say it is solid state; it will be referring to solids and particularly
crystalline solids. The questions if you have any during this entire course and so on can
be addressed to me on the email, which the address of which is given here this is
gururow@iisc.ac.in
Let's make a beginning; in fact, we have to always begin at the beginning so that we can
reach an end. And this total of 30 hours now will be covered by understanding the 3
dimensional world in which we live in. There is one issue that is associated with the 3
dimensional world that is symmetry and both man and nature have a love for symmetry.
Symmetry is something which exists all around us and we grow as a part and parcel of
symmetry. And therefore, getting a basic idea of what is symmetry is all about is
1
essential. And of course, we have to formulate laws, rules and of course, the mathematics
that is behind understanding symmetry. So, that any type of symmetry which occurs both
in nature and in the man-made materials will be understandable in terms of these issue of
symmetry.
Man always loved symmetry because whatever constructions we did and whatever
constructions we continue to do; there is always a angle which is given to understand and
appreciate symmetry in any construction we make. Nature also kept this in mind as we
see from the way in which the trees grow, the plants grow and so on. But if we go to the
very closer details of symmetry, we find there is a considerable amount of asymmetry.
In fact, it is the asymmetry which makes symmetry more appealing and this is of course,
observed in many cases. Like for example, in particularly in south of India; they put a
black dot on the face of the child during the birthdays or first year birthday celebration
and so on. Now the fact, that there is a black dot introduces asymmetry, but this allows
you to appreciate the symmetry.
So, the baby is supposed to be appearing more and more beautiful; of course they say it
is the blessing of god and also the darting of the evil; which is why they put this black
dot. But essentially the black dot is enhancing the symmetry that is associated with the
baby’s face for example. And that's how symmetry is an issue which has to be
understood both in terms of symmetry and the asymmetry which as a consequence
appears.
In materials it so happens that the presence of asymmetry is the one which is required to
give a property for a material. The most symmetric material is always not very
interesting from its property point of view. So, if you want to imbibe a property to a
material; it is essential that you introduce the asymmetry associated with a given
material. So, material design, material construction also involves the loss of symmetry.
What is more interesting is symmetry encompasses more or less all aspects of science.
Be it biology, be it material science, be it physics; in fact I teach a course where people

come from all parts of the institute, all departments who attend this particular course
because understanding symmetry is a fundamental issue which is required in order to
pursue any kind of research in future. So, that way this is a very fundamental course and
2
I will keep in mind that the participants who are in this program also are coming from
different disciplines.
There may be people coming from biology who probably had mathematics even at the 12
standard level and therefore it is essential that these people who are at the attending this
course will also get a flavour of the entire understanding of symmetry. Now, symmetry
also as I mentioned dictates the structure of the material and therefore, symmetry and
structure go hand in hand. And the concept associated in this course will be to build the
symmetry and structure in order to understand the solid state associated material.
Now what do we mean by a solid state? In fact, matter exists in 3 forms gases, liquids,
and solids. These are the 3 forms of course, there is a fourth state of matter which is the
plasma; which we will not touch in this particular course, but other than that there are 3
states are matter. In gases, it so happens that the particles which constitute the gas or not
in contact with each other; they are very independent of each other.
And therefore, the particles are free to move for example, like air in this room. And the
air in this room is spread the particles are all over this room; suppose we have opened the
door, the air goes out of the room and then balances its pressure with the external world
outside air and therefore, there is an issue of particles being completely free. So, if one
wants to study the structure of this particle; then it is possible with gases we because we
can study the structure of one isolated particle.
The methodology and the experimentation; however, becomes extremely difficult in

order to study one single particle of a gas, particularly with the fact that a given
temperature and pressure conditions let us say the normal temperature and pressure
conditions; these gas particles are moving with enormous speeds. And therefore,
arresting them in one place and studying the structure becomes an issue. It is not; I am
not saying that it is not possible to study gas structures, in fact people have studied gas
structures by variety of methodologies. By for example, creating a very high vacuum and
in that particular work vacuum; this one single particle can be analyzed and there are
methodologies that are developed.
When gasses now are taken and compressed; let us say we apply pressure on gases or we
apply reduce the temperature. So, 2 parameters the thermodynamic parameters,
temperature and pressure can now convert gases to liquids; in most of the cases not in all,
3
but in most of the cases gases can be converted to liquids by application of pressure or
reduction in temperature.
As we go to lower temperatures the particles now which are totally free to move around
become somewhat connected to each other and that is how the particles in a liquid are
connected to each other and liquids flow. So, when liquids flow the particles move along
with it, and that is how they find their own level. For example, the water in this bottle
finds it on level; that is because the particles inside the liquid are connected to each
other.
However, even at in the normal temperature and pressure conditions; if we want to study
a liquid, it is still difficult because we have the so called Brownian motion which we
have studied in our high school days which keeps these particles not at rest, they will be
moving at a certain kinetic energy. And this therefore, then again becomes a challenge to
study these materials in the liquid form. Now we take the liquids and further compress it
with application of pressure or we can also reduce the temperature and again as I
mentioned most of the materials go from liquid to solid state.
Now, in the solid state the particles are in contact with each other and they are also so
close to each other that they get arrested in their positions; respective positions. So, their
movements are restricted; except for a certain amount of vibration which it can have
about a mean position. So, all the particles therefore, find their mean positions and if
these mean positions are found in such a way that one particle is connected to another
particle at a distance a; let us say in 1 dimensions.
Then if this next particle also comes at a distance a from the second particle and so on in
both directions, then we say there is a 1 dimensional lattice. A 1 dimensional lattice
which is occupied by these particles and these; particles therefore, which occupy this 1
dimensional lattice or now arranged in such a way that there is a periodic repetition of
these particles at a distance a; so, if we consider that in 1 dimensions then we see that the
repetition of a units.
For example, if I say here there is a particle here, there is a particle here, there is a
particle here; the distance between these two is 'a' and the distance between those two is
'a' then this periodicity we call it, this particular definition defines the periodicity. So, the
4
periodicity is a; for this particular lattice of repetitions which are occurring terms of
particles.
So, it is now essential therefore, to see whether we want to stay the idea is to study the
structure of this particle. So, if you want to study the structure of the particle we know in
a solid that this part in this particular type of solid; where there is 1 dimensional
periodicity; we know that there is an order. Now because of this order the particles are
arrested in their positions; they are not now moving around freely. And so it is possible
now to focus on that point and determine the structure of that particular particle.
Now if this repeats now itself in 2 dimensions; lets say we go in this direction and this is
b and the particle repeats again periodically in that direction. Then we have a 2
dimensional lattice where the particles are occupying the lattice points. And in third
dimensions it so happens that since we live in a 3 dimensional world as I already
mentioned; it so happens that in that particular direction also if there is a repeat pattern,
then this forms now a unit which we always not write in 3 dimensions, but write in 2
dimensions, because we are used to 2 dimensions as I will mention in a minute. We are
used to 2 dimensions and so we write the 3 dimensional object in this form.
So, we say that this now is a 3 dimensional crystal; this is a 3 dimensional crystal
because this now repeats a along the one direction, b along the second direction and c
along the third direction. So, the repetition a, b and c now will keep going through the 3
dimensional space and develop it into a object. Now this object if it happens like that
with a, b, c as the periodicity repeat periodically repeating distances; we call this a
crystalline solid. And so we now restrict our discussions for some time until we
mentioned later that the discussions will now be centered on crystalline solids.
So, we consider in this particular course the symmetry and the structure associated with
crystalline solids. See one of the restrictions that automatically come on to the nature of
the particle; the particle is not necessarily let us say a dot, it could be an object, it could
have some shape like that let us say. So, the dependence of the particle is now the same
particle will repeat here again like that, but then the distance between these two is
maintained.
So, the crystalline state therefore, will arrest these particles with the periodicity that is
required to repeat in all 3 directions. And this now becomes an essential factor because
5
this is the one which puts certain restrictions on the way in which the particles are
arranged. So, the restrictions come from the nature of the; what we call as this box which
eventually we will define later as the so called unit cell.
So this now is referred to as a unit cell and if this repeats in all 3 dimensions to form the
full crystal. So, it is therefore, essential to study what is the content of this unit cell rather
than study the entire content and that is what we will do when we go to the structural
aspects of these materials. So, as I mentioned that when we are looking at 3 dimensional
objects; we try to write the 3 dimensional objects in a 2 dimensional projection.
This is the mistake we made by joining school because when we went to school on the
very first today itself; grim looking teacher took a stick in one hand and the chalk in the
other hand and they started telling us what is a straight line, what is a bent line. So, they
said right a straight line, horizontal line, a vertical line and then they took another
vertical line and another vertical line; put them at an angle and put third vertical line and
said this is alphabet a.
So, we learnt in this particular way whatever we have learnt in the school. And therefore,
what happened is the 3 dimensional; we were born in a 3 dimensional world, but the 3
dimensional understanding which was there probably till we went to the school is lost
completely. So, we now are comfortable with 2 dimensions. So, there are 2 ways in
which we can visualize in 2 dimensions and see still try and understand the symmetry
aspects.
One is of course, what is known as stereographic projection; we will see in a minute how
it looks like. And the other one is the so called pattern recognition in 2 dimensions. So,
we will spend some time on both these issues; stereographic projection we will do it in
detail later on when we go and try to study the symmetry elements in more detail. What
are the topic typical symmetry elements which occur in crystalline solids; when we look
into that we will look at stereographic projection. The pattern recognition in 2
dimensions we will look at in the following slides.
6
So, this particular slide is an interesting painting which was drawn by Rubens Aguilon
and this is a picture which I took from the library museum the Louvre Museum in Paris
and this is hanging on the wall. And you see that it is not just us who are having
problems in understanding the 3 dimensional structures even the Gods had problems.
So, here is the picture of Atlas who carries in Greek mythology the earth on his shoulder.
And then they want to find out let us say the distance between Bangalore and Mumbai on
the surface of the earth. So, what is happening is that this angel is shining some light and
that light is creating a projection and the 2 angels are measuring now carefully with a
divider the distance between Bangalore and Mumbai.
So, essentially what this is done is showing is that it is possible to steady 3 dimensions in
2 dimensions to start with. And then we have to therefore, find out the rules and
regulations which govern the stereographic projection which we will study later on not
right away; we will study later on probably in the coming classes. So, in this particular
class we will also look at the visualization of symmetry in 2 dimensions.
7
And that will come from the next slide which is showing us the actually it is a diagram
which is drawn by a person called EC Escher; who of course, unfortunately passed away
recently.
He made these abstract paintings of course, he is not a crystallographer, but he made

these abstract painting with which we can understand all aspects of symmetry. In fact,
more or less all the aspects that we need to describe the symmetry and structure in the
solid state. And so I am going to show you several of these diagrams and then if people
are interested in looking at the detail of this; this particular set of pictures there is an
IUCr book which is issued which is available which gives you a list of all the EC Escher
diagrams.
And all these diagrams are somewhat abstract and somewhat imaginative. For example,
here you see the flying horses, you have a set of white flying horses and a set of red
flying horses; the white ones here and the red ones there and these flying horses now,
repeat themselves in this 2 dimensional motifs. So, if you now start this as a 0 of the of
our mapping; then you call this as the a direction and this as the b direction for example.
Then in this a direction and b direction we have this motif and this particular motif has a
repeat pattern. For example, we get this white horse; repeat itself at this position and
repeat again here, repeat again here; this is at a certain fraction of a and this is at a certain
fraction of b. Now this fraction of a and that fraction of b with respect to the a and b
8
which we have marked here is essentially representing the unit which will now repeat
itself in 3 dimensions; in 2 dimensions to describe the entire motif.
So, we take lets say the stomach of this horse; the stomach of the horse and probably I
have drawn it a little closer it should be the stomach of the horse here and the stomach of
the horse there. If we take that, that unit if we take; it will now we take this unit displace
it to this position, displace it to that position and so on; we will generate the entire motif.
One of the beauties that is associated with this is also the fact that there is no empty
space in this.
So, it is completely closed packed; so when we are looking at crystalline materials also
when materials crystallized like you know the flying horse can be replaced with a
molecule, the molecules try to now arranged in such a way that they are close packed.
This is the property of the fact that we need to have a solid when the particles have to be
in contact with each other and if the particles have to be in such a way that they repeat in
both a and b directions uniformly; then we will see that it has to have a close packing.
So, if we call this as a prime and this as b prime then we have this unit a prime b prime
representing the repeat pattern. What is very interesting also is this is the fact that instead
of taking the centre of this horse; as I have written we can take any part of the white
horse. We can take the nose of this horse, the nose of that horse, the nose of that horse
and the nose of that horse and define the so called unit which repeats itself and that unit
will be identical to this unit.
So, we can take any part of the white horse; likewise we can take any part of the red
flying horse. So, if we take any part of the red flying horse and then show from red
flying horse to the next line red flying horse that repeat will also be the same and this is
known as translational periodicity. So, the periodicity with which it repeats in a direction
and b direction is now controlled by an operation a symmetry operation which is now
referred to as translation operation.
So, this entire motif is now dependent upon the fact that we built these units and translate
these units by a certain amount a prime along this direction and b prime along that
direction. We can go either in the positive way or in the negative way; it remains one and
the same. So, therefore we now therefore describe this 2 dimensional motif in this
particular manner.
9
The next slide will tell you how to do; how to see that 2 objects are identical. We have
been telling that this white horse is identical to this white horse and the red horse is
identical to that red horse. When can we say that 2 objects are identical? This is an
operation which this is a explanation which comes from serious mathematical
considerations and that is the theory of isometric transformations.
So, we are not going into the detailed study of theory of isometric transformations
because that is the one which tells you when you can, given 2 identical objects placed in
random positions and orientation; which operation should be performed to superpose one
object onto another. Now this is a very basic question, when we say 2 identical objects;
how do we define this identity with respect to the 2 objects?
So, there are 2 types of identities we can define one and there referred to also as
congruent objects. Now we say this is one object, this is another object; these 2 objects
are what we call as congruent objects because in all principle definitions these 2 will be 2
identical objects. Now we have drawn 2 arrows one in this direction one in opposite
direction; so the directions are different.
So, the way in which we have to prove that this is identical to that one is to use this
concept of movement. So, we use this concept of movement; so by moving we have what
we call as the; in case of direct congruence we move it using the translation which we
described; we translate the object. We can also rotate the object about a certain defined
10
axis; we can also do the rotation combined with translation. So, these operations which
we do; the moments which we do the whatever movements which we do these are
referred to as symmetry operations.
So, we have object 1, object 2; so what we do is we translate this object in the in the
given direction and then rotate it about a position about an axis and then the rotation plus
translation combined will orient this object along this direction. And these 2 then will be
called identical objects; provided they are congruent with respect to each other. So, if we
cannot have the shape of this object like the shape of the object is immaterial in this
particular case because the way we define it is the following.
Suppose we take a point here on this object and another point there on the object and
draw the distance the distance between these 2 is d 1; that is a vector which we draw.
And we take another point here and draw and join these 2 points and call it as d 2 and the
angle between these 2 we call it at alpha 1. And then a similar definition of identical
identity will come if we take in this object on this particular object; similar positions
occur in such a way that we have a d 1, we have a d 2 and an angle alpha 1. We call this
as let us say alpha 1 prime this is d 1 prime and d 2 prime.
So, by definition the d 1 and d 2; so d 1 prime and d 2 prime. So, if d 1 is equal to d 1

prime and d 2 is equal to d 2 prime; alpha 1 is equal to alpha 1 prime, then we say that
these objects are congruent and there of direct congruence. So, in all case cases of direct
congruence the moments will be either translation or rotation about an axis or a
combination and of rotation and translation; which is referred to as the screw.
So, these movements will now bring in the objects onto direct congruence. The opposite
congruence the second type of congruence that can exist is the so called opposite
congruence where again d 1 will be equal to d 1 prime d 2 will be equal to d 2 prime, but
alpha 1 will be equal to minus of alpha 1 prime. Such objects are referred to as objects
which are identical and they are of opposite congruence.
As we see here the moment for opposite congruence will be involving an inversion;
inverted the object about itself there is a reflection across a plane. And then there is a
rotation followed by reflection; a rotation followed by inversion and the so called glide
plane where we not only have a translation, but we reflect it about a plane. Now the
details of this we will work out as we go along in this particular series of lectures. So, if
11
we are talking about independent objects here; 2 identical objects; so if the 2 objects are
identical we call these moments as symmetry operations; if they operate on the entire
space and then the symmetry elements therefore, will be either points or axis of rotation
or planes of rotation.
So, the entire space therefore, now can be considered as a symmetric space with the
following symmetry operations. The symmetry operations could be with respect to a
point we have an inversion and with respect to axis we can have a rotation; we can have
a reflection about a plane and the combinations of these. So, points axis and planes
therefore, form the so called symmetry elements.
I think this brings us to more or less the end of the first half hour. And so I have in this
half hour tried to cover the basics of what are 2 identical objects. And in what way these
identical objects will be of congruence and how this congruence can be direct and
opposite. And what are the symmetry operations which take an object from of direct
congruence onto itself, opposite congruence onto itself.
So, in principle the entire space therefore, is consisting of these symmetry operations. So,
a 3 dimensional space now is not just a 3 dimensional space; this 3 dimensional space let
us say if there is inversion, it has a centre of inversion, a point at which the inversion
takes place and so on.
12
Prof. T. N. Guru Row
Solid State and Structural Chemistry Unit
Indian Institute of science, Bangalore
Lecture – 02
Two Fold Axis Representation with the Help of Esher Diagrams
So, therefore we go back to this particular diagram, if this is the simplest of the diagrams
where we show only the translation periodicity. So that means, to say that the types of 2-
dimensional lattices we can have depends upon the nature of what kind of symmetry
elements we can associate with these objects. So, if there is a translation periodicity,
there is 2-dimensional lattice. There is a 2-dimensional lattice which has a 2-fold
symmetry, there is a 2-dimensional lattice which has a 3-fold symmetry and so on.
So, how can we go further and in what way we relate one object to the other in terms of
the symmetry. See we are talking about only translation symmetry; that means; that the
object can be 360 degree rotated that is a 1-fold symmetry. We have talked about the 2-
fold symmetry presence of 2-fold symmetry, we were talked about 3-fold symmetry, we
also talked about the 6-fold symmetry. And, similarly we can talk about the so called 4-
fold symmetry. One of the conditions which we have always been telling very quietly is
that these diagrams which are drawn by Escher have if, have a close packing and if we
13
have a 5-fold symmetry object and want to do a close packing; we will show later on that
it is not going to be possible.
Because, we will have some holes left in this 5-fold symmetry packaging and because of
the fact that 5-fold symmetry packaging will have holes; it will not satisfy the conditions
that we need for defining a 2-dimensional lattice. And therefore, the number of rotation
axis get restricted to 1, 2, 3, 4 and 6. We will mathematically prove in a later class that it
is only possible to have 1, 2, 3, 4 and 6 and this restriction comes because, of the
presence of the translational periodicity which is in the solid state. So, the presence of the
symmetry to 1, 2, 3, 4 and 6 is only there in case of solid materials and particularly in
case of crystalline solids.
So, that way the amount of effort we have to spend in understanding the symmetry of
objects get reduced because, of the presence of the translational periodicity. Since, there
is translation periodicity we do not get 5-fold symmetry operations for example,
similarly 7 8 9 and so on are also not possible. We are going to prove it mathematically
in the next coming class may be in the next class or so. But, for our current
understanding we need the fact that we need to note the fact that once we require close
packing, once we have translation periodicity and we have the presence of the symmetry
elements; we can now appreciate the Escher’s diagram in terms of all these presence of
symmetry elements.
And, replace these imaginary of you know abstract object by molecules then we can have
these molecules also following the same symmetry operations. So, we can have
molecules packing them also in a 2-fold symmetry situation, we can also have the
molecules packing together in a 3-fold symmetry situation and so on. One of things we
also have to notice is in this particular case is the definition of what we called as a prime
and b prime; see here we called this as a prime and b prime. And, we define this so called
block and we called it as a unit cell; if you recollect that part which we did in the
previous class half an hour class.
So, the angle between these two is 90 degrees. Now this therefore, is a type of a lattice
where the lattice points are located in these positions and we have a 90 degree angle
between these two. It is not necessary that a prime should be equal to b prime. Say if a
prime is equal to b prime you will get a square, if a prime is equal to is not equal to b
14
prime, you will get a rectangle. So, the presence of a square and a rectangle can sort of
generate this particular motive depending upon whether a prime is equal to b prime or a
prime is not equal to b prime. Now, these two units which we have defined in a this red
blocking are referred to as lattices and these are refer to as a plane lattices. So, we have a
square lattice and we also have a rectangular lattice.
If we got to the next diagram where we had this 3-fold symmetry, you see that this if you
take this 3-fold point and join in to that 3-fold point and join this two to this 3-fold point.
And, join it to the next 3-fold point something like this, you see now that you get a unit
which is now having a different than 90 degree angle.
So, this will now define a different kind of a plain lattice ok. So, we will see as we go
along how many plane lattices are possible. What is very interesting is to note the facts
that in solids because, of the translation periodicity restriction come to this in terms of
the shape of this so, called the unit cell which we are going to define. So, we can define
two types unit cells now; three types unit cells. When a is equal to b and alpha is equal to
90 degrees a not equal to b alpha is 90 degrees. And, here you can find out what should
be this angle; I want you to think about what could be this angle.
If this angle is 60 degrees or 120 degrees both are possible for this particular unit and
therefore, you will get a 3-fold as well as 6-fold rotation that can be describing this
15
particular lattice. So, we will come back to this translation periodicity again and in this
particular case as we have already notice, there is close packing and also there is
translation periodicity. So, this comes because of a fact that we this object will repeat,
this object will repeat itself at that part of object and that object will repeat itself in this
object and so on.
So, we also defining the some kind of a unit cell, we will expand on that little more
before we go and understand the symbols that we are going to use. Because, the symbols
that we are going to use which that will become a very critical issue.
The types of symbols which we are going to use are listed here. The reason why I have
given this particular slide for which to about which I will refer again later on is to tell
you the representation of symmetry. For example, if you have symmetry like 1 bar you
represent it by closed circle. So, if there is a 2-dimensional lattice and there is a 1 bar
symmetry in that one that is, the inversion symmetry, then you represent the point at
which the inversion is occurring by showing a small little open circle to represent where
exactly is this particular inversion center located. So, if you now consider the 2-
dimensional object which has an inversion center then that particular 2-dimensional
object will have largest inversion center indicated, it will also have translation
periodicity.
16
So, if it has translation periodicity it will have a plain lattice, if it as a plain lattice then it
will have either no fold symmetry or a 360 degree rotation 1-fold or 2-fold or 3-fold or 4-
fold or 6-fold. It would not be having any other symmetry elements which will coming.
We are going to mathematically do that in the next class, but at this particular point we
will see what are the symbols which are universally used in literature. Because, this is
what will be used in crystallography and we are going to use it repeatedly as we go
through the course. These are the symbols for example, for the 2-fold we have this
representation and for the 3-fold we represent it that way. This is as long as the axis or
normal to the plane of projection.
So, we have this is the 2-dimensional projection and we are looking at the normal to the
projection; that means, we are looking down to this screen. And, when we do that these
are the representations of the inversion symmetry. If there is no symmetry, there is no
representation which is 1 of course; the 360 rotation there is no need to represent a 360
degree rotation. So, we do not have any representation. It essentially tells you that the
object is on to itself. So, the object is on to itself in 1 bar, it is on to itself into 2-fold, it is
on to itself in 3-fold and so on. There is also additional symbols which we will describe
in more detail as we go long and these symbols represent the so, called screw motion.
If you go back to the previous class you will see that we talk about the objects of direct
coregents. The objects of direct coregents can have these kinds of symmetry elements.
These sort of symmetry elements which are refer to as 2 1 31 32 41 42 etc. These are called
this screw axes. What 21 represent is interesting; it represents the presence of a 2-fold
symmetry. And, it is not just presence of a 2-fold symmetry, but there is a translation
associated with a 2-fold symmetry of one half the unit cell.
Let me explain again that when you say 21 axes you have the presence of a 2-folds
symmetry and then there is a translation the 2-fold symmetry alone is not finishing the
operation. So, the object now will see the 2-fold and it will also see the half translation
along the unit cell. It is something like you know you driving a screw through the wall,
then you your screw send through the wall, what you do is a rotation which you give on
the screw driver. So, there is a screw fit fitting into the wall you give a rotation in this
direction. When you do this rotation the forward point of the screw will go inwards.
17
If you do the opposite rotation the forward point of the screw will come back and it
design in such a way these 21 screw means, that when we rotate it by 180 degrees; the
advance that is made is equivalent of half the unit cell in that particular direction and that
is therefore, it is effectively 1 divided by 2. So, this is referred to as this screw. We will
be seeing examples in more detailed for this kind of operations. Similarly, we have the 3
bar, 4 bar and 6 bar. These are 3-fold rotation with inversion centers, 4-fold rotation with
inversion center and so on.
The representation of axes which are 21 21 parallel to the plane of projection ; that means,
a 2-fold or a 21 screw axes which are parallel to the plane of projection. So, this is
parallel to this plane than we represent them like this. This is the 2-fold representation,
this is the 21 representation, there is a 21 representation. So therefore, all the possible
symmetry elements which occur in 2-dimensions under eventually the 3-dimensional
issue will come up in these representations. So, these representations for example, of a
mirror is essentially telling us that the axes parallel to or inclined to the plane of
projection. So that means, that there is a mirror plane which is like this passing through
this plane; there is a mirror plane which goes through like this.
If it is perpendicular that it is shown in the right side. So, if you study this table carefully
this are the various symbols which will appear in the description of this crystals structure
and also this so, called paint groups and space groups which will describe in a coming
class. You will see the symmetry elements displayed on the side or on the object itself.
These are known as symmetry diagrams and when we talk about symmetry diagrams, we
will read this into more detail. At this movement I showed essentially to tell you that we
will be using some of this symbols for the next part of some of our understanding, of
how to locate the symmetry elements on a 2-dimensional object and that is why I have
given this table.
This table will repeat again during the course where, we will go individually and explain
the performance each of these rotation operations with respect to a point in our molecule
or material or the object in general. As you should always remember that even though we
talked about the objects and similarity of objects, the once we refer to the 3-dimensional
space and associate the symmetry with that 3-dimensional space; it is always the fact that
3-dimensional space will have that symmetry. Suppose there is a 2-fold rotation
associated with the object which we have described like the flying cat in our previous
18
example, then the entire space has 2-fold symmetry whether there is a cat or not the
entire space has 2-fold symmetry.
And therefore, this space is now itself is gaunt by the presence of the 2-fold symmetry;
anything you place in any point in this 2-dimensional space therefore, will follow the 2-
fold rotation. So, if you are have an object coming at a particular point that point will
now undergo a 180 degree rotation to represent 2-fold. The only important thing is the
direction of the axes and that is why this table is shown here. The direction of the axes if
it is perpendicular to the plane is represented this way; if it is along the plane, it is
represented that way. So, which is perpendicular to the plane it is represented to this way
and if it is along the plane it is represented that way. So, these are representations which
we will see associated with the so called space group diagrams. And, eventually this will
be the one which will be describe the nature of the object inside the unit cell and how
these object repeat themselves using the symmetry concept. So, that is what we will
study towards the next few classes, we will get an understanding of how these all
happens.
Now, let us go further. Here is an example of a presence of a 2-fold rotation axis which
is now perpendicular to the plane. So, in the plane this is the Escher’s diagram again you
see some results here, white and black results. This is very beautiful representation even
though Escher himself was not a crystallographer; all of crystallographer is there in a
19
Escher’s diagram. So, this is very beautiful if you look at these diagrams and try to
understand the concepts. And, that is why I bring in this issue of Escher’s pictures. For
example, here if you take this white lizard and take this white lizard, these two white
lizards are related to each other with respect to this particular point where the noses are
interesting for example, and they have I have marked a 2-fold symmetry at this particular
position. That means, any point in this object will have an equivalent point in that object,
please note the word equivalent.
So, these if you take a point x y z here, if you take a point x y z which now represent the
2-fold axis. See this entire 2-dimensional space now is filled with 2-fold axis. And, I
have given a representation of a unit cell. This unit cell now can is like a frame, I can
move this unit cell this is attempt this is showing this is let us say a. The distance at
which this translation periodicity repeats itself; that means, these two motives repeat
again in these two motives and black ones; for example, and the white one again repeats
here. So, this is the so called unit cell which now repeats itself in both directions to
generate the 2-dimensional lattice. Now, we have identified the position of the 2-folds
screw axes here, the 2-fold axes here not screw sorry 2-fold axis here; this is a 2-
dimensional object.
And, then we also see that in addition one once we have identified 2-fold symmetry here
automatically there is a 2-fold symmetry comes up there which is due to the unit
translation. And, the 2-folds symmetry comes here due to unit translation, 2-fold
symmetry comes here due to unit translation. So, if you take any point in space x y z that
in the 2-dimensional space it could be here, here, here, here, here, here, anywhere. It will
take the object that particular point x y z will use the rotation axis; now the rotation axis
is perpendicular. So, every where you can think of 2-fold rotations coming in this space
so, any point here will now undergo the 2-fold rotation. So, mathematically x y z now
will go to minus of x, minus of y z in this particular case we are looking at the x y
projection; that means, this is now 0 and this is a and that is b let us say.
Or in fact, we should write here x and this is y and the unit cell vector in this direction I
am using the word vector, I am assuming that there is a basic knowledge of vector
analysis which all of you have a undergone in your mathematics classes around standard
12. So, I am assuming that you to have an idea of what a vector is just to recollect or
recount your idea of a vector. A vector is one which we represent not only the
20
magnitude, but it will also have a direction. So, if we take this point as 0 and then, we
move in this particular direction like that and reach this point which is now again, the
end of the unit cell. And, I will call this as a and this is along the x direction and along
the y direction I will call this as b. So, this representation we will not call this a here, we
will not call that a here we will not call that b here because, a and b now represent the
dimensions of the unit cell and this x and y represent directions.
So, you see that the any points in space something like this point which is let us say x y
z. Now, we will see the 2-fold symmetry, it will see the 2-fold symmetry let us say at this
point. This is at 0 0 0 of our unit cell, than if you see this 2-fold symmetry x y z will now
generate a minus x minus y z where, do you think it will come. It will now undergo this
2-fold rotation. So, it will go like that and come some where there ok. So, there is we a 2-
fold rotation about that particular point. What is very interesting about this lattice is that,
one once we have the lattice which is generated with respect to a and b as unit cell
dimensions 2-fold symmetry is exits at every one of this point in every point in the 2-
dimension space and, certainly at these points where you have the 2-fold rotation in this
particular example.
And therefore, you it will also see the presence of 2-fold at this point. These are
additional 2-fold symmetries; they come due to the translational periodicity. What you
see here is that this centre of symmetry, this 2-fold symmetry is with respect to the
presence of the intersection of knows of each one of these lizards. Whereas, here you see
that it is at the intersection of the 2 legs of the lizards. So, the object is so oriented that
you will have at presence of additional symmetry is which appear. So, whatever
symmetry appears here at half the distance along a and half the distance along b an
additional symmetry exists. Like for example, you have this point and this is that 0 0 0
this is the 0 0 0 because, we taken this as the origin.
Then at a we get a 2-fold so, this is that 1 unit translation, than again one more unit
translation in this direction will give you then next a and so on. In addition to the
presence of this 2-fold rotation because, of the fact that we have a 2-fold here and a 2-
fold here an additional to 2-fold will represent will develop at this point. So, the objects
therefore, which arrange themselves in which the 2-fold symmetry in this particular
motive has to sort of remember. For example, this lizard now cannot move its leg away
from here. It has to be keep itself the leg in that position in order to get the 2-fold
21
symmetry, otherwise it would not fit into this motive. So, if Escher had made a mistake
and shown this leg as different leg then it will not be close pact feeling motive, therefore,
the restriction comes now on the objects.
So, the object now therefore, must have the 2 legs intersecting at this particular point, if
it has to keep that shape. And, it is very interesting that lizards can do that kind of an
operation and therefore, we can set the lizards into the 2-dimensional motive. So,
molecules when they fit in they also have remember the fact that apart from the presence
of the 2-fold symmetry, they also have the invoking of the 2-fold symmetry at this point.
And, again at half this distances along a, half this distance along b and also half of a half
of b which is a different 2-fold symmetry, that is why I shown it in different colors. The
ones at the ends I have shown by blue, the ones in the middle that half distance is shown
by red and the one in the very middle are shown by green.
We will see the significance of these when we actually look into the way in which
molecules pack when we discuss this space groups. And, they are presence of additional
symmetry which appear because, of the fact that we have lattice and they lattice has to
obey certain rules which we are going to formulate as we go along. Right now, we are
looking only at the symmetries that are possible with respect to this motif. The presence
of additional symmetries is a crucial thing because; the presence of additional symmetry
will bring in the issue of defining this 2-fold axis in the same way. That means, if you
have an x y z and an minus x minus y minus z this point will now, also see this 2-fold
symmetry.
And therefore, this point will now go over there with a 180 degree rotation something
like this it will rotate by 180 degrees and go there, and therefore, we will have these
equivalents of x bar y bar z bar. So, in this particular 2-dimensions space therefore, we
define x bar y bar z bar x bar y bar z rather as the operation which takes the object on to
itself using the 2-fold axis; the presence of the 2-fold axis in this motive given there x y z
directions. Suppose, that is the 2-fold axis which is in the x direction than we see that the
fold axis is associated with the x direction than your x y z will than go if this is the
direction x direction. And, we look at now they y z plane than x y z will go over to what,
to x y z will now become x y bar z bar.
22
So, it will go to x minus y minus z and it is the way I pronounce it we always use this x y
bar z bar and in crystallographic notation; we will represent that has x ̅y ̅z . So, will I
write a bar over y and z and pronounce it as bar. So, if the axis about we so, suppose x is
coming towards out of we plane of this diagram than the x y z will go to x -y and -z.
And, that represents the 2-fold symmetry along the x direction; similarly along the y
direction we can define the 2-fold symmetry. We will see very interesting observations
which come out of the fact that we can have 2-dimensional, 2-fold axis intersecting with
each other and so on. Right now, we are discussing only the presence of one 2-fold axis
in this diagram and we will restrict ourselves to one 2-fold axis.
So, I think at this stage we come to the end of the second part of the talk; this is the
second half hour. So, what have what I will do is I will conclude what we have done so
far. What we have done so far is that we have looked at motives and we have looked at
the possibility of arrangement in 2-dimensions. And, we have looked at the arrangement
in such a way that we identify the presence of symmetry elements. And, we have taken
one example of the presence of a 2-fold axis which is now generating the so, called
equivalent points.
So, these are refer to as equivalent points. So, in a in a 2-fold symmetry containing
objective if there is a 2-fold axis, the equivalent points are x y z, x bar y bar z; if z is the
direction of the 2-fold axis. In other words if, z coincides with the 2-fold axis as we have
shown in this particular picture; I think that come to the second end of the second half of
a the talk. We will now go to further and examine from this stage how we now can look
at the other axis which can come up, other symmetry positions which can come up.
So, I think we will stop here for today.
Thank you.
23
Lecture – 03
Pure Rotation Axes
So, what we will do next is to see the presence of a 3-fold axes. We saw it already with a
different diagram of course there we have different objects, but in this particular case we
have a 3-fold axes. Now, the 3-fold axes is illustrated with respect to the presence of a 2-
fold two dimensional motif having a 3-fold axes. Now, here I want you to carefully
observed this, and identify the nature of the 3-fold axes. There are again in just like in the
2-fold axes situation, we have to put the objects in such a way in this particular motif that
they are related by 3-fold symmetry.
So, if they are related by 3-fold symmetry, and there we still demand the close packing,
then we have the in this particular case three types of objects. The one in red, the one in
white, and the one in black, so three objects. And now I am marking the 0 0 of the 3-fold
axes at this position, and I will call it as the origin. So, you see that that the origin there is
a 3-fold intersection of this, go back yourself and go to that previous diagram, which I
think I will show you again.
24
We had this picture, we saw that there is a 3-fold axes which now relates these three
objects the three fish. We also have the relation of the three birds, we also found an
additional factor that is the presence of the 6-fold arrangement of the tortoise, but then
the tortoise are arranged again in with respect to this 3-fold.
So, I have taken this example, which is a much simpler example to illustrate the same.
You see that they are identical objects here, they are not different kinds of objects except
that the colors are coded differently showing that the orientations are different let us say.
So, as a consequence you see that the presence of these three fellows with their tails
intersecting at the 3-fold, and the presence of these three white ones inked with the nose
intersecting, they generate a 3-fold. And if you now look at the black positions 1, 2, 3 4,
5, 6, they are 6-fold related. And the 3-fold relation we observed with respect to the three
black ones here. So, there are two types of possible 3-fold rotations, which can generate
this motif.
25
And therefore, it is not just the 3-fold rotation at the positions where we have this is
different from the previous case, where we had this results. Here again you see that the
job of the 2-fold rotation is the same to rotate the object by 180 degrees, but the object
orientation decides how the objects themselves orient.
For example, in this case the noses were related by 2-fold, in this case the legs were
rotated by 2-fold, and here the elbow joints were related by 2-fold. And similarly, in this
particular case you see that the relationship with respect to the black representation of the
object is a 3-fold representation here, and a 6-fold representation here. So, remember the
3-fold and the 6-fold go together. And infact this is something which will observe even
in three dimensions.
And therefore, we will get very special cases of the nature of the unit cell when we go to
systems, which we are going to define later called a trigonal system, and the hexagonal
system. And these this is the kind of origin to the existence of them together in a given
unit cell.
26
So, you see with two dimensions what we have tried to do so far is to understand the
concept of 2-fold, the concept of 3-fold, we can also introduced the concept of 4-fold.
Just like the 3-fold and the 6-fold existed together, we also have the 4-fold and the 2-fold
existing together in this diagram. This is again a different kind of results different sets of
results, we have shown here. And you see that the tales of these three fellows intersect at
that point, the tale of this tale of that tale of this, there are four of them now. The angle
between them is 90 degrees.
And so this particular unit cell, which I have marked is a unit cell which is a square.
Earlier the 2-fold unit cell was at rectangle, that 3-fold unit cell was whatever we had
before that is representing the three it was a trigonal unit cell of course we are talking
about two dimensions.
So, in two dimensions it has a 3-fold symmetry, it has a 6-fold symmetry. So, 3-fold
symmetry, 6-fold symmetry, 4-fold symmetry, and 2-fold symmetry also decides what
should be the nature of the lattice, what should be the nature of the unit cell. So that
means if a is equal to b and angle is 90 degrees, then you have what is called as a square
lattice. So, the square lattice now can be unique in the sense that we have the 4-fold
symmetry one once we have this symmetry represent here, we will get the presence of
the 4-fold symmetry here as well. But, apart from that we will now generate the 2-fold
symmetries at these points, just like we had 3 and 6 there, we have 4 and 2 here.
27
So, the presence of the 2-fold is inherent in a square lattice, but the 4-fold now dominates
by occupying the unit cell edges. Now, these are take home points at this particular point,
we are not making any statements the final statements, because we are studying number
one the objects in two-dimensions. We are looking only at projections.
And the second one is we are now looking at some abstract objects, we are going to
replace these with molecules later on. And look at the molecular symmetry, how the
molecular symmetry and the crystal symmetry at simulate together. Just to make life
easier at this particular moment, we see that this fellow this lizard has no symmetry. And
even though the lizard that does not have any symmetry, the way in which it gets
arranged in this lattice produces this 90 degree angles. So, we have this 90 degree angle
because of the presence of the lattice.
And therefore, these objects which have no symmetry by themselves can use this 4-fold
symmetry, and arrange it themselves in this. There are some time ago there was lot of
work done, and imaginations floored you know people imagined various things of how
to arrange objects into different kinds of unit cells and so on.
And there was the person by name Kitaigorodsky who found that if the object itself as a
certain symmetry, the object itself has a certain symmetry. Suppose, this object has a 4-
fold symmetry suppose, it instead of the lizard we have a cubic block ok. And then we
now look at the projection down the given axes, let us say it is projected down the cube
axes. Then he says he is he found out that if you examine a large number of such objects
if the molecule by itself has a symmetry, the molecular symmetry will be utilized by the
crystal symmetry.
So, when it goes into the crystalline lattice, you need to goes into this unit cell, the unit
cell will now ensure the presence of the molecular symmetry that means, we now not
have four of these, but we our cube as mount to sit here at this point. We have also made
a remark that each and every point in this particular case will have the cube. So, this will
also have the cube, this will also have the cube, this will also have the cube, so that
means the molecule now becomes special, we call these are special positions.
And we will be going into the more details one once we understand this symmetry fully,
we understand the basics of symmetry, and then we understand how translational
periodicity and symmetric can be combined, and how the various crystal systems can be
28
generated. And therefore, we will put all those restrictions together and talk about the
general positions and special positions, when we derive the space groups in a much later
class.
So, at this particular moment we see that if the molecule as a symmetry Kitaigorodskys
finding is that if the molecule has a symmetry, the same symmetry should be maintained
in the lattice as well. So, this particular two-dimensional lattice should have that
symmetry if the molecule has that symmetry in which case the packing becomes
compact, then only we will have the close packing of these objects in such situations.
On the other hand if the object is not having the 4-fold symmetry also or any kind of
symmetry as in this example, where there is a lizard sitting here. If they are arranged in
such a way that there is a 4-fold symmetry, then the 4-fold lattice can be fitted on to this
particular system. So, the objects now are a sitting what we call as general positions.
So, just to tell you about this a little more detail, we now have to derive the so called
equivalent points. So, in the 2-fold rotation, we got x, y, z, and x bar, y, z bar. And then
we have to now look at what happens, when we have a 4-fold symmetry, what happens
when we have a 2-fold symmetry and so on. So, this is the next step to which we go, and
this step which we take up to do that will come a little later, because I want to introduce
you to the opposite congruence objects.
29
So, we so far we have looked at only the direct congruence objects. So, the objects of
opposite congruence, and I am giving you an example from literature and that of a
molecule. In this particular case, I have shown the molecule ferrocene and not ferrocene,
this is a diphenyl iron is not ferrocene. Ferrocene will have a five membered ring system,
this as a six membered ring. So, we have the diphenyl, and iron is at the centre.
Now, the way the molecule appears the molecular symmetry appears is the presence of
given any x, y, z, it can go to -x -y -z. So, this is an inversion symmetry, centre of
symmetry, we call this has the centre of symmetry being present at this particular
position. So, the molecule is centre of symmetry. So, by the logic which we use now so
far in case of direct congruence objects; if the molecular is central symmetric, then the
molecule has to go into a crystal system which is also a center of symmetric.
So, the lattice we chooses should also have a center of symmetry, not just that the lattice
is chooses should have its center of symmetry coincident with the center of symmetry in
the molecule. So, the molecule has a certain symmetry, and there is no further freedom
given to the molecule, the molecule still sits in a position with which it will generate the
center of symmetry.
So, the molecule symmetry, and the lattice symmetry will have to have a center of
symmetry, because this space in which this molecule is now going to be put either it is
two-dimensional or three-dimensional will have x, y, z to -x -y -z every point in this
space will be going from x, y, z to -x -y -z. So, the molecular symmetry, therefore
dominates this is how Kitaigorodsky observed it. So, I thought I will give an example
with a object of indirect congruence.
30
Object of opposite or indirect congruence can also have mirror planes, this is an example
where Escher's diagram again I have used, in order to illustrate the presence of a mirror
plane. You see we talked about points centre of symmetry, we talked about axes of
rotation about which the rotation takes place 2, 3, 4 and 6; 1, 2, 3, 4 and 6 or the axes of
rotation. We have the point of symmetry which is the centre of symmetry.
In this particular case, we had this molecule just now discuss the iron compound, where
iron is sitting at the centre of symmetry. So, we have the rotation axis and points as we
said in the very first class that we have symmetry elements, which can be points, which
can be axes, which can also be planes. So, this particular plane is a plane which is now
defining the x, y, z if we define in this particular mirror in which we have shown here,
the coordinate system you see that the there is a change in the y axis, in the y direction,
so x, y, z has gone to x, minus y, z.
So, what happens in this particular case is that there is a plane, which passes through the
centre of this beetles, and you see that the right hand side of the beetles is identical to the
left hand side of the beetles, and this goes through the entire motif. So, we define what
we call as a mirror plane. So, this mirror plane is now, in the plane of this is the y
direction. So, the mirror plane is coming in such a way that we also invoke the presence
of mirror planes with the other object, which is now the black object the black beetles
31
that also is related by a mirror plane this also is related by mirror planes. So, the mirror
plane propagates through the entire lattice generating a close packed motive.
So, in this particular case what is required is in case of the centre of symmetry, we
needed a point. Here we in the case of the rotation axes, we need an axes of rotation
about which the rotation occurs, it could be a 2-fold, 3-fold, 4-fold, 6-fold, whereas in
this particular case mirror stands out by itself. So, in this case the mirror is perpendicular,
and we call it a mirror plane which is perpendicular to the in this particular case the x, z
plane. So, mirror is cutting through this. So, we have this side is mirror image of that
side.
So, it is like I put my left hand here, and my right hand here, and this now are related
through this mirror plane. So, you see that the overlap of the objects will occur that way
only if we now take this image, and take the image which is the left hand. So, with the
right and left hand define what are known as and enantiomorphous objects. So, the object
in fact, these two are not going to be identical otherwise. See nobody is left hand is
identical to anybody is right hand, then it will be a terrible issue is it matter. If both of
them are identical and thereby real space, they are related to each other, then our hands
will be in a different I do not know what orientation I have to show you, but it would not
be like that.
So, it has to therefore have the two objects, therefore my left hand and the right hand or
enantiomorphous objects. So, the symmetry that is associated with a mirror is an
enantiomorphic. What do you think is the symmetry that is associated with the centre of
symmetry? This should also now be an object, where the right hand side this side of the
object is now going to -x, -y, -z. So, when we there is a change of value of x, y, and z, all
three changing from x, y, z then also I have an inversion.
So, the inversion is also an operation, where the object can be also not necessarily, and in
fact it is not a enantiomorphous objects. So, enantiomorphous objects are very special,
and therefore the enantiomorphous behaviour in the arrangement of molecules will also
be different. And this happens in case of many of these amino acids particularly the
naturally occurring amino acids. The naturally occurring amino acids are have a rotation,
the optical rotation, there have the levo or dextro rotatory, and most of the naturally
occurring amino acids are levorotatory.
32
And therefore, when you arrange these molecules into a protein the structure of the
protein always can never go into a centre of symmetric system, it has to go into a non-
centre of symmetric system, because it has to have the so called enantiomers behavior.
So, the objects that compound a protein will be enantiomorphous object, and is
something which is very interesting in biology and in fact the most of biology is a matter
of fact consequence of the presence of a enantiomeric objects in protein structures. That
important point is something which will cleans the whole issue that is suppose you have
let us go back to the centre of symmetry or for that matter we can go to this mirror.
So, in the case of the mirror operation, we take x, y, z to x -y z. So, the point from here
went to the other point. Now, suppose I operate a mirror plane again on this fellow x, y
bar, z, what will happen x, y bar z will go back to expression. So, x, y, z and x, y bar, z
are the two equivalent points or the two equivalent points, which are related by this
symmetry operation, we call this as the symmetry operation the related by the symmetry
operation. These are the two equivalent points. And no other equivalent point exist in
this diagram for this time that means, x, y, z goes to x, y bar, z and x, y bar, z goes back
to x, y, z.
So, the whole this thing the closing, it is referred to as a closure property. And in this
brings in the issue of group theory, so in group theory we call such operations has
operations of a group. Now, there are various ways in which we can approach this
problem, we can go into a conventional group theoretical way of doing things, but then I
thought that it is not the way we will go about. Instead, we will keep this in mind that
when we have these equivalent points, operation of the equivalent point once again will
take it back to the original, depending on what kind of situation we are in, if there is only
one mirror plane operation, x, y, z goes to x, y bar z, and x, ybar z goes back to x, y, z.
Now, this kind of a thing will therefore close the whole issue that means, we cannot have
any more objects. If I have an object here, object here, these two objects only are there in
this particular unit cell, beyond that we cannot have any more objects. The next unit cell
is only a translational periodicity repeated thing, and therefore this will now repeat itself
in the next unit cell, it will repeat in the both directions a and b.
And therefore, the existence of points x, y, z, and x, y bar, z they define such a situation
in mathematics is referred to as a group. So, they define a closure property, and therefore
33
this defines a group property. So, the points now belong to a group. There is a clear cut
mathematical procedure through which we can understand, what is group, there is a
group theoretical principle. And we can design now the so called methodology for using
this group theoretical information to identify what are known as point groups, and space
groups.
In fact, the presence of point groups and space groups, we will bring it in a totally
different contacts. And then relate what we have seen in these diagrams once again to
whatever we have achieved so far, so that you know you get a one to one
correspondence, without going into the rigor of mathematics, which is involved in group
theoretical methodology. Even though that is probably not the method to follow the
group theoretical behaviour.
In this particular case, particularly for chemist and biologist which will be very useful to
understand what we are doing in terms of these pictures. And also now eventually when
we do the point groups and space group, how we derive this point groups and space
groups or rather than how mathematics arrives at that, because the mathematically
deriving thing is straight forward, and very easy from the mathematics point of view.
But, from the understanding point of view, the way we are going to do in terms of group
theory, and the particularly the point group derivation and space group derivation will be
a better way of understanding. The three dimensional architecture in molecular crystals
crystalline materials, which are molecules or ions or whatever it is easier to get a picture
of that .
34
So, we saw the glide plane. So, what is important here is something which we defined
already we have been talking about it, but we have not put it in a real nutshell. So, what
you know it is like as I always tell in my regular class, what we have done today in the
last few classes, I think it is now the class number-3 as far as time is concerned.
So, in the last three classes is to get the alphabets is it something like developing a
grammar, we are now trying to developed the grammar of crystallography, and the
grammar of symmetry and structure particularly. So, to develop the grammar, we are
now looking at various alphabets. So, what we have got so far is in terms of the
definitions of close packing, the presence of symmetry. The presence of symmetry in two
dimension, and eventually we will extend to three dimensions. And we have use the two
dimensional drawings of (Refer Time: 23:05) to get a feel for what are the types of
symmetry elements, they that can be there to describe the two dimensional motives
which are already displayed.
So, in this particular context, we have been talking about the definition of what we have
been referring to as the unit cell. The unit cell is something which repeats itself in both
the directions, in this case it is two dimensional. So, you this marking which I have done
in red will define the so called unit cell that means, whatever is there inside, which is
now atoms and molecules and so on, instead of these objects. Whatever is there inside
this box will repeat exactly the same way in the next box, it does not mean that I have to
35
fix the origin here, I can fix the origin wherever you want. So, it is like you know the
window frame. If you want to fix window frame, and in a building all the windows are
identical, the size the dimensions of the windows are identical. So, we have to now
generate frames.
Now, the frames which we generate or the unit cells. So, we can take the unit cell fix it to
this window on the right, window on the left, window in the back, window in the front,
because as far as the dimensions are concerned they remain the same. So, similarly unit
cells therefore can be defined in such a way that they are like frames.
So, we can move this along anywhere in the two dimensional space, it will fit exactly the
symmetry that is required for this two dimensional motif, so that way we define the unit
cell here. In terms of let now I can call this as 0, the 0 are the origin can be anywhere in
this diagram. So, we have decided to fix it there, then this distance will be a zero to a
along the x axis x direction, and in the y direction we see that this is b.
Now, if we go back to the 3-fold case here, you see suppose I call this as 0 and this angle
is very crucial now. So, if these are the vector, so this vector is now a, and this vector is
now b ok. So, effectively what happens is this now defines the unit cell 0 b, and 0 a. You
see that this is not if we take this as on the x axis, if the 0 a along the x axis, b need not
be along the x axis. So, the presence of the b will be in such a way that we have to now
define the direction of b, it is not with respect to the 90 degree angle. And so this
becomes an issue in case of 3-fold and 6-fold, it is not an issue in case of the 2-fold and
the 4-fold ok, so remember this, we will see why it is so important later.
So, in this particular example coming back to the definition of a unit cells. So, we define
this unit cell as a repeat unit, it repeats itself in two dimensions to describe our entire
motif. So, the unit cell is drawn in all other repeating wall pattern patterns, so just to
show that the unit cells can be drawn for all the diagrams, we have shown in the (Refer
Time: 26:37).
36
Now, coming to the last part of today’s this particular class, we will see that the two
dimensions can be extended to three dimensions, because now we are in a situation to go
towards three dimensions. So, just like we have the concepts built in two dimensions, the
extend this to two three dimensions.
And we therefore have the direction a, direction b, and direction c. Remember a, b, and c
need not always be rectilinear coordinates, they need not be rectangles of 90 degrees
depends upon the symmetry. If this symmetry allows for a, b, c to be at 90 degrees with
respect to each other, then only then and then only we can have the unit cell definition
ok, this is very important to remember at this moment.
So, this produces now a box. So, the box can be it like this with all 90 degree angles.
And if a is equal to b is equal to c alpha, beta, gamma the angle between these a, b, and c
or 90 degrees, then we call this a cube. So, this will be one of the possible unit cells,
which can we can have in a three dimensional description of the crystal structure. So,
with crystals can grow into cubic cells.
We will see how many possible cells are going to come in the next class, and in what
way they appear. And the consequences of how the angles play a very important role,
and the concept of vectors that means, we are now it depends upon where the origin is
define the concept of vector a, vector b, and vector c the way in which we define a, b,
37
and c. And the angle interracial angle between them becomes crucial to define the crystal
geometry.
And the three dimensional unit cells, which will develop cannot be all kinds of unit cells,
because it has to have the information about the symmetry we have understood. So,
combination of symmetry, and the requirement of the three dimensionality in the unit
cell, which preserves the symmetry it will allow only certain set of possible unit cells.
And that is how we end up with seven crystal systems ok. So, today’s class will end at
this particular moment, and we will go further.
Now, before I go further, I just thought I should give you three reference books. One is
the book of fundamentals in crystallography, it is an IOCR text, and this is by Carmelo
Giacavazzo. And some of the contents of today’s talk is taken from there. The (Refer
Time: 29:25) diagram, and the way in which it develops and so on is my own
imagination. And so it is dependent upon how you view it, and how one use views it. So,
I thought that that will be the best approach to give you the concept clear.
And then the next class, we will go and try to understand the concepts of how to
represent these symmetries, how the symmetries accumulate. And we now then go into
the aspect of what we called as point group symmetries. So, the next book is by Jack
Dunitz, it is X-ray analysis and structure of organic molecules. This is a book where we
you will find some very basic understanding of the X-ray diffraction theory, and then
38
how it refers to the molecular species. The fundamentals of crystallography will cover
overall all the aspects.
There is another book by Stout and Jensen. This is a book on X-ray structure analysis
practical guide, I did not write it down here, but it is X-ray structure analysis a practical
guide. And that book is bit ancient, but at the same time it gives as the required basics.
So, what you will see in the coming classes will be that I will be using these concepts in
and out, on and off.
So, books of this these books will be referred to once in a while. And whenever I refer to
that I will mention I have taken it from this particular books, so that when you have to
practice what you want to do in future with this course, then and you are guy got hold of
these books, it will help you to find out where we are now referring to at that particular
moment ok. So, I think we will stop here for today.
Thank you.
39
Lecture – 04
Properties of Crystal
Towards the end of the last class, we had realized that there are three basic operations
which occur in symmetry elements. One is an operation about a point, the other is an
operation about an axis in a certain direction that will introduce different kinds of axis
that are possible. In general there are n-fold axis, which can exist that is n equals 1 to
infinity. The infinity axis can be considered as the one which generates a sphere, which
is the ultimate most symmetric object.
So, the number of rotations axis that are possible are quite a large in quite large in
number. On the other hand the operation about the point is restricted, and also the
operation about it perpendicular to a plane is also restricted, it depends upon where we
put the plane. So, when we consider the crystalline state, this becomes extremely
important in the sense that the crystals have this particular property that is the property of
periodicity.
So, since we have this periodicity, it has a main issue in a crystalline material. The
periodicity introduce a certain restrictions. The restrictions come basically on the rotation
40
axis; because the point is unique the reflection across the plane is also unique, given the
direction. Whereas, the given the direction the axis of symmetry can be very different
from each other, it could be a 2-fold axis where 180 rotation is there. A 3-fold axis where
120 degree rotation is there which, we saw actually with examples taken from Esher’s
diagrams.
So, the Esher’s diagrams were kind of indicative that we can have these rotation axis
possible in a two-dimensional lattice, we were examining only a two-dimensional lattice.
What we will now do is to see, why this now becomes only five possible rotation axis in
case of crystalline objects. It becomes only five possible rotation axis. The rotation axis
being one of course in full rotation 360 degrees 2, 3, 4 and 6 that will be corresponding
to 180 degrees, then 120 degrees, 90 degrees, and 60 degrees respectively. And no other
axis is possible. We shall examine why it is so, and this is a very big advantage, when we
look into crystalline objects. In fact, this restricts the possible number of symmetry
positions we can have in a crystalline object.
And the whole issues rotates around the concept of periodicity. So, the periodic
repetition of objects or for that matter the periodicity that is associated with the three-
dimensional space decides what kind of rotation axis, we can have for a given system.
Let us say examine this with a one-dimensional lattice. The one-dimensional lattice is
shown here in terms of the red dots. And then the possibility of creating a two-
dimensional lattice is through the rotation either about A or a rotation about B. The
rotation angle can be defined as delta, and delta can be any value in principle.
So, we can now generate the lattice, but the lattice should have the periodicity imbibed in
it that means, the distance between A and B should be at a distance t, which should be
the same distance to the next point to the next point and so on. And after rotation the
point, which we generate should also obey the laws of periodicity that means, the two-
dimensional positions of A prime and B prime, which we are going to generate. A' being
generated by rotation axis operating on B, and taking A with an n-fold rotation. The n-
fold rotation can be defined as 2 π divided by 𝛿 where 𝛿 is the angle of rotation, 2 pi
being 360 of course.
So, if we now sit at B and look at A, and start generating the lattice points, which
eventually occur because of the two-dimensional periodicity in the lattice. Then we can
41
generate n such lattice points, the (n-1) th lattice point will be A'. One lattice point,
which is just before generating again the point A. A similar situation can be imagined
when we sit at A, and start at point B, and take the point B around. And the point B,
when it goes to the first position will be B' and so on in the n possible rotation axis. So,
effectively we therefore generate the two-dimensional lattice. The condition being that
the object or the red dot which represents the object should be repeated at periodicity
associated with it being t. So, by definition therefore, BA' will be also t, the translational
periodicity. So, also AB', which is translational periodicity.
So, if we now calculate the distance between A' and B', the A' B' will be a multiple of t,
because by definition of a periodic lattice. It has to be a multiple of t, it could be one
times t, two times t, three times t, but that value of m is an integer. It is only an integral
multiple of t. In that situation, we define a two dimensional lattice. Therefore, the
definition of a two-dimensional lattice comes into the picture here. And the presence of
the two-dimensional lattice puts a restriction on how we can get the value of delta are
organized. Let us examine it now from a simple arithmetic point of view or a
trigonometric point of view.
So, let us take this distance A' B, A' B has by definition is A B, because now we said that
this is the n-1 th lattice point. So, one point before the nth point, which happens to be B.
And therefore, we have A B', and therefore it is equal to AB and that is t. And the angle
which is BAB' So, this should be this angle which is referred to the angle ABA', which I
should be referring to here. The angle ABA' will be equal to 2 π divided by n, because n
is the fold of rotation. So, if n is equal to 1, it is 360 degrees; if n is equal to 2, it is 180
degrees and so on.
So, BAB' is 2 π divided by n, so also ABA'. And both of them define our delta value.
Now, the question is that these two values remain t, and B prime A prime defines mt, but
what we can do by geometry is we can write B' A' as equal to AB, which is the
translational periodicity minus AB' cosine 𝛿, AB' cosine 𝛿, and B A' cosine 𝛿.
So, this expression for B' A' will ensure that we have now B' A', which now can be
written in terms of t. So, A B is t, the A' B' is t, A' B is t, and therefore this is with two
times 1 minus 2 cosine 𝛿. So, 1 minus 2 cosine delta times t defines B' A'. And by the
42
laws of periodicity, this should be a multiple of t, which happens to be the translational
periodicity.
Therefore, we can write an expression mt is equal to t times 1 minus 2 cosine delta. So,
this would give us the value of cosine delta as 1 minus m divided by 2, we all know the
property of a cosine function. And the cosine function will now be an oscillatory
function. And the values corresponding to 1 minus m divided by 2, therefore is a
restricted value. The restrictions which come on m will be between -2 and +2, so m
cannot take any other value other than these -2, -1, 0, 1, and 2 representing different
types of rotation axis. For example, in this table we have given the value of 1 minus m is
equal to minus of 2, which corresponds to a cosine delta of -1. And cosine delta of minus
1 is the value part, which is 180 degrees.
And therefore, it represents a 2-fold axis, and then the value of B prime A prime in such
a situation turns out to be m value becomes 3, so it is 3 times A B. Now, this defines a
unit cell, because now we now define our unit cell in terms of these periodically
repeating patterns. And this periodically reporting pattern will now have to repeat in all
two dimensions to generate the two-dimensional lattice. So, the basic unit cell therefore
can be defined as a quantity which has a direction A, and a direction B in two
dimensions. And the angle between them is not necessarily 90 degrees, it could be any
angle.
So, this therefore defines one possible type of conventional unit cell, which we can have
with a compound which is or a or a material which is displaying a 2-fold rotation axis.
Likewise, when the value of 1 minus m is minus of 1, and we have a cosine delta value
of minus of half, the value of delta can be calculated and therefore n turns out to be three.
This is a 3-fold rotation axis in which case the value of B' A' is 2 times AB. And the
possible type of lattice, which can be generated is A equals B gamma is equal to 120
degrees.
Now, I would like you to dwell upon the fact that when such a situation occurs, when we
have a 3-fold rotation axis. And then the B'A' turns out to be 2 A B in what way we are
saying that the conventional units mesh should be A equals B and gamma one 120
degrees. This is something, which I would not leave it to your imagination.
43
The clue for this finding out what it is left in the Esher’s diagram, which we saw earlier
in the previous classes. Go to the Esher’s diagram which showed the 3-fold axis
symmetry. If you see the 3-fold axis symmetry, you will see that the unit cell which you
can define in that situation will be something like A equals B gamma is equal to 120
degrees.
So, I want you to figure out likewise. The case when the 1-m value becomes 0, the
corresponding value of m times t, which is B' A' will be AB. And this represents a 4-fold
rotation, and the lattice will be a square lattice. We have also given an example of this
kind in the earlier class in the Esher’s diagram So, the idea is now that you have to go
back and forth to the Esher’s diagram and bring in the mathematical relation, which we
have derived here to see one to one corresponding.
And the value of m is equal to 1, we get a 6-fold rotation axis. And you see now A B is
given as 0, I want you to find out why it is turns out to be 0. And in which case A is
equal to B and gamma is 120 degrees, it again defines a cell which is referred to later we
were going to talk about plane lattices. These now in fact define the plane lattices, these
1, 2, 3, 4 types of plane lattices, we will get.
In fact, we get a total of five plane lattices, we will discuss it when we bring in the
discussion on different types of lattices that are possible in crystalline materials. At this
juncture we are now trying to tell you in this particular example of this slide that in this
particular slide, what we are in stating is to show that the number of rotation axis that are
possible are restricted to 1, 2, 3, 4 and 6 in crystalline objects which are the periodicity.
And the same logic can be extended to three dimensions.
And therefore, the rotation axis are 1, 2, 3, 4 and 6 only, and no other rotation axis is
possible. This can be easily seen in a physical situation. Suppose, you are given a 5-fold
type tiles, and you are asked to fill this entire floor of this room, you will see that it is not
possible to do a closed packing. As we have been discussing earlier all our diagrams in
the Esher’s pictures are close packed. There is hardly any space that is left between them.
And the closed packing is are a consequence, which has to be built around periodicity. If
you want periodicity, then we will have closed compact packing. If we do not want
periodicity, we can have a 5-fold symmetry that is how we have the so called 5-fold
symmetry objects, which are referred to as quasi crystals which is of course an area of
44
interest over the last few years particularly, since it got the Nobel price of 2013 or
something like.
So, since the Nobel price was obtained, this particular aspect of 5-fold symmetry the
presence of 5-fold symmetry is actually also very important. But, for conventional unit
cells which we are going to deal with in this particular course, we will restrict ourselves
to close packed periodic objects, periodicity now driving the fact that there are only five
types of rotation axis.
So, even though our symmetry axis, which we identified in the previous classes to be
with respect to a point, with respect to an axis about which they rotate and with respect
to the planes about which they can reflect. The possibilities of types of such they rotate
the symmetry elements get restricted to a point, which will give us the so called centre of
inversion. The rotation axis which is 1, 2, 3, 4 and 6 only, and then of course the
reflection across the mirror; so, this is as far as the periodicity issue is concerned.
We will be introducing later on two other symmetry elements, which go with this that is
the screw axis under glide planes. I have already shown a Esher’s picture in which this
has been illustrated with respect to the glide plane, but we have just seen the picture. And
we have tried to see how the mirror followed by translation exists in order to preserve the
periodicity. But, I will introduce this two one screw axis, and then the screw axis can be
associated with the 3-fold axis, screw axis can be associated with 4-fold axis, and also 6-
fold axis. We shall see those independently, later on as we go with our discussions.
So, at this particular moment what we say is that in a two-dimensional lattice, the
number of symmetry elements are restricted. And they are restricted to a point about
which inversion can take place a point in which we have a 2-fold rotation, 3-fold
rotation, 4-fold rotation, and 6-fold rotation. And then of course, the last issue which is a
mirror plane. What is very important is when we talk and consider the issue of inversion
symmetry, the inversion symmetry is not a 2-fold symmetry, it is not restricted to the
two-dimensional objects. Two-dimensional objects whatever inversion we get will be the
2-fold axes rather than the inversion center. Because, the third dimension the so called x
bar, -x, -y, -z brings in the issue of third dimension. So, what we are discussed here is
only in two-dimensions.
45
And therefore, the presence of the inversion centre in two dimensions does not exist, this
must be borne in mind. And therefore, we will not see an inversion centre in the
diagrams which we have shown in the earlier Escher’s diagrams. If we want to show the
inversion centre which, we showed with respect to your molecule in the earlier issue in
earlier diagrams. We took the case of a di-phenyl iron compound in which case we saw
that the centre of symmetry indeed represents a three-dimensional object rather than a
two-dimensional object.
This is a take home lesson from this slide that the centre of symmetry will come into
picture, the inversion centre will come into picture only in three-dimensional objects. If
the object is purely two-dimensional, it will not have the inversion centre. And this is an
issue which becomes extremely important, when people are dealing with liquid
crystalline material. When people deal with liquid crystalline materials, there is order in
two-dimensions, but there is no order in the third-dimension. Even though there exists a
third dimension.
So, in such situations there are different kinds of phases, which come up as a
consequence of liquid crystalline moieties, and this becomes an area of study which is of
extreme importance in liquid crystals. Particularly with the importance of liquid crystal
displays, which have now become a common practice it is a common household
phenomena the liquid crystal screens which you see in the latest TV, TVs and so on, are
essentially based on the symmetric objects which are two- dimensional in nature rather
than three-dimensional as far as the crystallinity is concerned right.
So, having seen this particular slide, we will go and examine how we represent these
symmetries, this is something which we saw quickly in the last class. But, in this class
we will go into the detail of how these representations are made, and this is the
representation which is universally accepted, and it is the one which is followed by the
international union of crystallography.
46
In fact, the course which I am now giving is essentially considering the fact that there are
certain rules and regulations, which have been formed by the international union of
crystallography, and that we must follow through the entire course. So, the way in which
we represent for example a 1 bar symmetry which is the symmetry where we invoke the
inversion centre that is represented by a open circle. And you see that that open circle
does not exist in any one of these except at this position, which is the 3' position.
And this 3 bar position has a centre of symmetry, which is inscribed inside the 3-fold
symmetry. We will describe this in more detail, when we take talk in terms of the
objects, and how the objects get generated with respect to the various symmetries. But, in
general you see the presence of centre of symmetry is not invoked in any other B, C, D,
E which are explained in this about the various types of projection, and the axis which
we consider in this diagram.
So, essentially 1 bar is shown by an open circle a little open circle, because the bigger
open circle would represent an object, we will see that in a few minutes. So, this is a
small open circle which represents the centre of symmetry or the inversion centre. 2-fold
which is now coming out of this projection, out of this board is essentially represented in
this form. And then it is black and inside to show that it is a full-fledged 2-fold rotation
axis that means, there are certain relationships which have to be followed by the objects,
which are located in this space. Do not forget that one once we define these symmetry
47
elements, the symmetry elements are applying to the entire three two or three
dimensional lattice, which we are going to take into account.
So, therefore every point in that space will be obeying the two-dimensional two 2-fold
rotational axis. So, any point you put in that box will have the 2-fold operation validated.
So, we will always have the 2-fold rotation axis being validated for every point in that
space whether it is two-dimensional or three-dimensional. Similarly, the 3-fold is rotated
the notation is a triangle, and for 4-fold it is a square, for the 6-fold, it is hexagon and so
on.
I have also illustrated here, they so called a screw axis which we did not define me as
yet, but then we mentioned it as in the passing. And also there was an example which
was given in the earlier slide, and later we will probably give it in your questionnaire to
check out what is how you interpret this screw axis. Essentially, a 2-fold screw axis is
one in which the object undergoes a 2-fold rotation followed by one half the unit
translation.
So, if our unit cell is defined as the A and B with respect to A and B. And if this 2-fold
axis is with respect to A let us say, then the 2-fold axis is now doing a rotation of 180
degrees from the position which is now representing the object. So, the object is here, it
rotates by 180 degrees it comes over there. And then it gets translated by half a unit, this
translation of half the unit in fact is very important, because what it does now is to take
that position which is now at a half distance away from the plane of this board of the
illustrated board, it is half the distance away in terms of the unit cell dimension.
So, suppose the unit cell dimension is let us say in that direction is some c. Then it is half
that of c that means, the translation component associated with this screw axis is one half
1 divided by 2. And this now then goes through again the procedure that means, this now
rotates again by 180 degrees, translation is again half the unit, so that we get the
translational periodicity restored.
See, the question here is the understanding that is required in all these screw operations,
which are shown here with all the symbols which we are going to employ in later classes.
It is essentially to tell you that the periodicity again plays a role in defining these screw
axes. If there is no periodicity, there is no screw axes, if there is no periodicity, there are
no what we call later as the glide planes which are shown in this part of the illustration.
48
So, effectively what it means is that the periodicity now is restored, but not within the
one unit cell, but in the next units, because they there is no differentiation between one
unit cell and the next unit cell because of the translational periodicity. So, whatever is
there in the first unit cell repeats exactly the same in the second unit cell, the third unit
cell, the fourth unit cell and so on.
And as a consequence if the object is located in the plane here, then that particular object
will repeat itself one unit away in a direction perpendicular to this board. And therefore,
we will see that two one screw axis is a valid symmetry element. Now, what do we mean
by a valid symmetry element, this brings us to the issue of defining what are known as
point groups.
So, each one of them which are illustrated here defines what we would later realize and
explained as point group symmetries, except that the translation components involved
descriptions, which we have here are not for point group symmetries, because they are
not keeping the point information intact that means if I take for example a 2-fold axis.
The 2-fold axis will have an x, y, z. And then if the 2-fold axis is to the y axis, we will
have a - x, y, -z.
What you notice is if you operate the 2-fold axis once again on x minus, x y minus z, you
will get back x, y, z. And whenever we have these kind of operations, these operations of
symmetry are referred to as point group of symmetries. So, a 2-fold symmetry, therefore
represents an operation of a on any point x, y, z in space which can go over to -x, y, -z,
and that -x -y -z will now go to x, y, z. So, this is the relationship in to represent a 2-fold
symmetry axis.
So, the 2-fold symmetry axis which is representing this is represented by this symbol will
take any point x, y, z in space and take it to minus of x, y minus of z, if it is a three-
dimensional space. And then we see that this particular two-dimensional 2-fold
symmetry operation can be repeated on -x, y, -z to get back x, y, z. Now, such issues
where we had going from x, y, z and operating the symmetry coming by again operating
the symmetry, and coming back to x, y, z to the original position. This then defines what
is called a point group.
And we will see how many possible such positions can come in, and how many such
point group symmetries can come in. So, the property of the point group is that when we
49
operate this symmetry more than once, depending upon how many such operations are
allowed by the symmetry which is present here.
For example, in the case of a 2-fold axis x, y, z goes to - x, y -z. And this again when
operated will come back to x, y, z. So, there are two positions which will be generated by
a 2-fold symmetry operation, and these two positions are also referred to as equivalent
points. Equivalent points, they are equal to each other. So, there is no distinction between
what is there at x, y, z, and what is there at -x, y, -z. So, therefore they become this so
called equivalent points. Regarding the issue with the screw axis, which are shown here
the two one screw, three one screw and so on. There is always this translation
component, this one half which is associated with it.
So, one once we bring in the translation component, it is no longer in the within the point
group operation. The point group operation will involve another translation of one half.
So, since the translation of half is involved in case of a two one screw axis, it would not
be called a point group, you know what we have to do is the corresponding point group
for two one will be a two four. And therefore, we will have no additional point group
coming due to these translational symmetries.
The other issue which is shown down here is the issue associated with more than one
symmetry operation. We will discuss this in more detail in the coming slide, but more
than one symmetry operation would mean that we have let us say a 2-fold symmetry a 2-
fold operation. And on the twofold operation, we will also have an inversion centre that
means, we now combine the 3-fold rotation with inversion centre. This please note that
this is not going to involve any translational component. And therefore, this by itself can
be a point group, let me repeat.
Since, the operation of a 3-fold and a 4-fold for example. Let us take the 4-fold or let us
take simply the 2-fold rotation axis. The 2-fold rotation axis now, if it is operated along
with the presence of an inversion centre. What happens is this x, y, z will go to -x, y -z
that is the 2-fold operation. Suppose, we also have an inversion centre associated with it,
then what happens is that the operation will now generate an inversion operation as well.
So, effectively what one would live finally is an operation, which can become what is
known as a mirror operation.
50
So, in the case of a mirror operation x, y, z will go to x, -y, and z. And the x -y, z will
come back to x, y, z and that would be a mirror operation, so that is why there is no 2 bar
shown here in this part of the slide we are showing the 3 bar, 4 bar, and the 6 bar. The
combination of a 4-fold symmetry along with a centre inversion centre, therefore
becomes an independent point group, because there is no translational component that is
associated with it. Whenever there is a translation component associated with it, it is not
a point group.
So, therefore all the translational point associated information like in the screw axis as
well as in the so called glide planes, which we will discuss more and more as we go
along, these glide planes also will not be independent point groups. So, the point group
therefore is a is defined in such a way that the symmetry operations are confined without
the presence of translational components. Even though the requirement of a point group
is that we should have it in translationally periodic system. The point group validity is
there only if there is a translational periodicity in the system, but the translational
periodicity that is present in the system is reflected not in terms of the generation of the
so called equivalent points.
So, when the equivalent points are generated. Those equivalent points should be within
the first unit cell. And whenever they are within the first unit cell, they are just operating
on that particular point about which we have the axis. And therefore, there is no doubt in
the mind that to call that as point group symmetry. But, remember point group
symmetries which we are going to restrict in terms of 2, 4, 3, 6 rotation axes and so on.
The point groups will be restricted to a total number of 32 as a consequence. So, before
the next few classes, we will know how these 32 point groups come in, and I think we
will now end this particular class, and then go to the next one.
51
Prof. T.N. Guru Row
Lecture – 05
Point Group Generation
So, just to recapitulate what we did in the last class, we say now that the symmetry
elements are associated with a point or axis and or planes.
So, the ones which are associated with the axis we have shown they are limited to 1, 2, 3,
4 and 6 in case there is periodicity; that means, in case of crystalline materials we do
have only these axes that are present. We of course, have a mirror plane which could be
along different directions; mirror plane perpendicular to an axis, mirror plane along a
given direction and so on.
We will bring in the issue of various orientations of the planes, but the operation is just
that of a mirror plane. So, there is only one symmetry element that will come which is
the mirror plane. As far as axis are concerned we get 2, 3, 4 and 6 and as for as the
inversion centre is concerned we get this so called 1 bar symmetry.
52
The next slide I am going to illustrate how these now become point groups and how
many such point groups we can generate. So, the first set of axis we will consider or they
are limited to the 1, 2, 3, 4 and 6 because it is a periodic lattice and these 1, 2, 3, 4, 6 are
referred to as proper axis because they represent if you recollect again the previous
classes they represent direct congruence objects.
So, whenever these operations are there 1, 2, 3, 4 and 6 they represent the direct
congruence objects. So, objects of direct congruence only will be corresponding to these
proper axis systems. So, we can have a 1, 2, 3, 4 and 6. So, a total number of 5 such
proper axis alone point groups that we will generate in a given crystal system. So, in
crystals with or without translational symmetry as we pointed out so far operations may
coexist. See, if there is no translational symmetry like in the case of a 2 1 screw axis we
have a translation symmetry.
So, if there is no translation symmetry that is present like in the case of a operation like
the inversion centre, then we can combine the one with the inversion centre which will
give us the so called 1 bar. What now happens is that the object which is at x y z goes to
-x -y -z; that means, there is an inversion axis that is present. The moment we have an
inversion axis we if you recollect the previous classes we have objects which are present
with this improper axis being present these objects will be of opposite congruence. So,
53
they will be of opposite congruence; that means, to say that we can have therefore, the
value of this angle between the two directions which we define being negative.
So, the improper axis therefore, now represent the combination of a proper axis with an
improper axis which happens to be the inversion centre. So, if we now combine a
twofold proper axis with an inversion centre we will get a mirror. So, we will work out
how this happens, below here. Now, we will do the operation of a twofold which will
give us x y z which goes over to -x y -z assuming that our twofold axis is along the y-
direction. So, the value of x and z now became -x, -z this is the twofold operation and
this is a proper axis operation. So, the symmetry that is defining now the point group
symmetry is to we take x y z to -x y -z.
In case we have a 2 bar symmetry; that means, we operate a twofold and that twofold
operation will now undergo a centre of symmetry inversion. So, there is an inversion
centre which now, therefore, x y z is no longer staying at -x y -z, but it goes to its
corresponding inversion centre operation; that means, x goes to x , -x goes to a value of
x, y becomes -y and z becomes as such z. So, -z becomes z, -x becomes x.
Now, this operation therefore, is the operation of a ̅1 symmetry because we have taken
this value of x y z which is now x̅ y z̅ and we have operated the centre of symmetry to
go to x y̅ z. So, the so called equivalent positions with this improper axis operation or x y
z and x y̅ z. Examine these two carefully; if you examine these two carefully it means to
say that you have a mirror plane now and there is a reflection across the y-direction.
So, if the axis was coming perpendicular to the board, if the axis was coming
perpendicular to the board the plane of symmetry is with the board and so, we have a
point x y z which is above the board now we will go below the board x y̅ z and therefore,
this now is referred to as a mirror symmetry operation. So, what we therefore, see is that
̅2 now gets represented as equal to a mirror. So, we have 1 which is a proper axis, ̅1 is an
improper axis, 2 is the proper axis, 2̅ which is a mirror is an improper axis.
Now, we have to consider the issue that we can at, since the operation is about a point
you see this is referred to as the point groups. So, the operation is about a given point and
this particular point is identified with respect to x y z. So, the point we have identified is
x y z. The space in which this object, which is identified as x y z, is located is governed
by this particular symmetry.
54
So, if the symmetry is 2 we will get xyz equals x̅ y̅ z̅. If the symmetry is that of a mirror
symmetry we get x y z is x -y -z and nothing else. It is a point group because of the fact
that if we operate the mirror symmetry again, we will get back x y z. If we operate the
twofold axis again we will get x y z and therefore, we now define these as point groups.
So, 1, 2, 3, 4 and 6 which were they which are the proper axis defined 5 point groups.
There are total of 32 so, 5 of these are now accounted for. Similarly the improper axis
combination with 1, 2, 3 and 4 will give rise to these combinations and these
combinations will add another 5. So, totally it has now become 10. We should also
realize that since the operations do not involve translational periodicity, both these
operations the proper axis operation and the improper axis operation; proper axis is 1, 2,
3, 4, 6 and improper one is the 1 bar symmetry, the inversion centre.
These two operations do not involve any translational component, like we have in the
case of a screw or a glide plane and therefore, these now can be combined with each
other. So, when we combine for example, 1 and 1̅ we can show that it turns out to be
equal to 1̅ only. So, there is nothing new which will get created. In other words the
presence of the improper axis remains the same, whether you combine it with 1 and
make it 1 and 1̅ simultaneously operating you will get one bar only. Similarly, 2 and ̅2
now will introduce a combination of the proper axis twofold followed by a mirror
symmetry.
Now, what happens to the equivalent points? The equivalent points now will change I
will show you in a minute how it will change, the equivalent points will change.
55
So, what happens is now x y z, we are now considering the symmetry operation 2 by m
ok. So, we are considering, this operation, 2 by m. So, there is a twofold symmetry and
there is a mirror perpendicular to the two fold symmetry. So, this is a different operation
than just a mirror or just a twofold axis. So, this is an additional symmetry additional
point group symmetry which is illustrated.
So, this now becomes a new point group where what happens is that x y z will now go
over to, first there is a twofold operation which will take it to -x y -z, then on the x y z
we will also operate a mirror symmetry. So, it becomes x y̅ z. So, we call minus y as y̅
and then we operate on this the inversion centre. So, we operated the twofold axis we
operated the mirror. Now, we can operate the inversion centre. Inversion centre is also
operated on the original x y z.
So, you will get -x -y -z. So, the number of equivalent points has double, in case of the
twofold axis there were only x y z and x̅ y z̅. In case of the mirror plane it was x y z and
x y̅ z those were the two equivalent points. Now, the operation 2 by m that by m means
that 2 is the twofold axis, so, let us say about the y axis about the y direction the mirror
plane is perpendicular to the y direction. So, 2 by m therefore, defines a new point group
because the number of equivalent points now become 4 instead of 2 and this represents a
different kind of space altogether, the three-dimensional space.
So, in three-dimensional space therefore, or even in two dimensions for that matter, this
becomes a new point group symmetry. So, remember whether we are now talking about
56
two dimensions or three dimensions it does not matter, because the operation is about a
point. This point is located in space. The space could be two dimensional, three
dimensional or a multi dimensional. It does not matter because we now considering only
the point on which the operation is being performed and we have three coordinates x y
and z which identify that particular point and that particular point now goes through
these operations of twofold followed by mirror and the combination of twofold and
mirror giving rise to the centre of symmetry.
This x y z and x̅ y̅ z̅ are equivalent points of a centre of symmetry or we can in fact,

better define it as an inversion centre. This means that this particular point group has an
inversion centre in addition to the presence of two fold and mirror. Please note this
statement that in addition to the presence of two fold and mirror. It will also have a
centre of inversion, in this particular case there was a centre of inversion only and that
centre of inversion insisted that the twofold went into a mirror symmetry.
In the case of ̅3 where, however, it is different because the ̅3 is an operation which goes
first the threefold operates and then followed by an inversion centre. I am not going to
write down the equivalent points, but I suggest that you work out the equivalent points
and find out how the 3 bar now is different because we first operate 3 and then ̅1
symmetry. Likewise the ̅4 is similar to ̅2 , but we will have to find out in this particular
case there is no separate operation like 3 and ̅1; in ̅6 there is a separate operation like 3
divided by m; that means, threefold symmetry for which there is a perpendicular mirror
and that results in a ̅6 symmetry.
So, some of these things can be worked out by looking at various kinds of pictures or
diagrams or looking at structures which go into these kind of point group symmetries.
We shall do that in a later time when we actually look at molecules and relating the
molecules to the space group and so on in which it crystallizes. It is a little premature to
discuss those issues right, now. So, we will restrict ourselves to the combination of
proper and improper axis. So, 3 and ̅3 is not going to give us you point group; however,
4 and ̅4 , 6 and ̅6 will give rise to two different point groups 4 by m and 6 by m.
So, we cut off these two fellows; these two fellows are 1 and the same 3 ̅3 equals ̅3
and 1 ̅1 is ̅1 . So, we generate three new additional point group symmetries and
therefore, we have 5 plus 5 plus 3 total 13 point group symmetries which are single
57
rotation axis operations. We have considered only 1, 2, 3, 4 and 6 and in three-
dimensional space it is not necessary that we should have only one axis. We can have
more than one axis and they can intersect at a point.
So, the point group symmetry is associated with the point. So, at that particular point
which is not necessary that we should have a single axis of symmetry, we can have more
than one axis of symmetry. One axis could be twofold, another axis could be threefold
and so on. One axis could be twofold; another axis could be a mirror. Now, the mirror is
not perpendicular to that axis, but the mirror is with respect to some oriented direction
which is different from the original twofold rotation.
So, effectively what we can say is look at this particular diagram. This particular
construction is referred to as the Euler’s construction in mathematics. So, suppose you
take a point P and then there is a point O about which we now consider the presence of
combination of symmetry axis. So, right now we looked at the combination, no
combinations we looked at independent individual rotation axis now we look at
combination of symmetry axis.
So, if we look at combination of symmetry axis suppose let us say there is a point P, now
it sees this rotation axis. This rotation axis now is centred at O and that rotation axis let
us say is a two-fold axis; that means, the if P represents x y z, Q will represent assuming
that this is the y direction. It could be any direction for that matter, but assuming it is y
direction we will have x̅ y z̅ as the position Q.
Now, the position Q may look at another possible rotation axis and that particular
rotation axis is represented by l2; l1 and l2 are the two rotation axis. So, then Q will go to
R by that particular symmetry operation. It could be any of these, it could be a two-fold,
it could be a three-fold, it could be a four-fold, six-fold and such combinations we will
see how many such combinations are there as a matter of fact in a minute from now.
So, when we do this operation then Euler’s theorem states that if you have one operation
which is taking the rotation from P to Q and another operation Q to R operating both at
the same point O, then with respect to this point O there will be a third axis which will
take R to back onto P into in order to define a point group. Remember the point group
definition is that we have to get back to the original. If we have a x y z we have an x y̅ z
58
then operation of the mirror symmetry will bring the x y̅ z onto xyz. So, that definition is
very crucial.
So, keeping the same idea in mind if we have a point in space about which there is
intersection of more than one symmetry axis, invariably the third symmetry axis is
invoked automatically, and therefore, we should have three symmetry axes. Now,
because of the fact that we have 2, 3, 4 and 6 we have to consider all combinations of
these and when we consider the combinations of these it will end up giving us the angle
between these two axis and that is what is given in this particular table.
So, this particular table tells us that if there is a intersection of three twofold axis 2 2 2
for example, one twofold axis going along the direction l1, another two fold axis of going
along l2, then invariably I should be drawing it I will draw and show you in a second. It
should be an axis which will go here such that now R gets related to P by another two-
fold axis, ok. So, R and P now get related to each other by another twofold axis and that
is the point which we have to think of and we should remember.
So, whenever there are two rotation axis intersecting at a point the third rotation axis is
automatically generated because of the fact that we can have proper axis, improper axis
and proper-improper combined for a given direction of axis. So, this particular axis may
be a proper axis, it may be an improper axis, it may be a combination of proper and
improper. So, all these we have to consider.
So, considering these we now look at only the proper axis combinations in this particular
table below. When we have only proper axis for example, a proper axis rotation and a
second proper of axis rotation will never give the third one as an improper axis rotation.
This is the extension of the Euler theorem. See, Euler theorem originally stated that if
you have a rotation axis P goes to Q and Q goes to R in another rotation axis the third
rotation axis gets created, that is the first part of the Euler’s theorem. The second
theorem is that when we have this as a proper axis the l 1 is proper, l2 is proper, then l3 has
to be proper. It cannot be an improper axis, if there is an improper axis we will see what
are the combinations that are going to come as we go further in our discussion.
So, at this particular moment what we have therefore, is the 2 2 2 where we have a two-
fold axis, a two-fold axis and a two-fold axis. In case of 3-dimensional space if these
three two-fold axis are present simultaneously intersecting at a point the requirement is
59
that these three now define three directions and these directions now are with respect to
each other at 90 degrees with respect to each other. They have to be with respect to each
other.
So, that means, to say that there is some amount of restriction on the nature of the lattice
which you are going to generate based on this. There is an angle of 90 degrees between 2
and 2, 90 degrees between 2 and 2, 90 degrees between 2 and 2. The combinations of
which the other rotation axis 3, 4 and 6 with 2 2 2 with 2 2 2 is possible. So, there are 4
possible rotation combinations which we can think of and they will give rise to the 90
degree angle between 2 and 3 in case of 3 2 2, 90 degree angle between 2 and 3 and 60
degree angle between 2 and 2.
Now, I want you to digress your mind and go back to the in your notes, go back to the
Escher’s diagrams and in those Escher’s diagrams you will see the relevance of 60
degrees if you are looking at threefold axis. You will see the relevance of 45 degrees, if
you are looking at 4 at four-fold axis combining two other twofold axes. In fact, in the
diagram which we showed which related the lizards together we had the indication of
both the four-fold rotation as well as the two-fold rotation. So, you have to just go back
digress your go back a few classes and see how those diagrams are so relevant for the
understanding of the combination of symmetry axis which we have just now discussed.
Apart from these 4 combinations only other two combinations permitted under the
Euler’s scheme; that means, the Euler’s rule has to be followed; that means, the P has to
come back again itself. So, if we go and operate some rotation axis and go to Q, Q to R
then R to P must be repeated. So, if you put that condition no other combinations are
allowed except two other combinations which gives rise to 2 3 3 and 4 3 2. There are
other combinations like n if there is an n-fold axis, but remember we are restricted to 1,
2, 3, 4 and 6 only because of the translational periodicity.
So, because in the crystalline state only 1, 2, 3, 4 and 6 are allowed, crystals cannot have
any other combinations other than these 6 which are represented here and therefore, the 2
3 3 combination and 4 3 2 combination give rise to these kind of angles. You notice that
these kind of angles look very scary to you in the beginning, but then once we get a hang
of it you will see that these two represent the octahedral and tetrahedral arrangement in
cubic systems and we will we will get a better look at this as we go along.
60
But, at this particular moment, please note that Euler’s theorem has to be validated and
we in this particular example which we have shown as combination of symmetry axes,
we have considered only the proper axis and there are 6 possible combinations of these
proper axis and those are illustrated with respect to the angles between them. So, the
angles between them also get restricted; that means, the nature of the three-dimensional
lattice you are going to get or the three-dimensional unit cell you are going to get will be
the same. It cannot be more than these types remember that as we will go further to the
next slide.
In fact, this slide is almost immediately relevant, will we have illustrated here the 2 2 2
how the three 90 degree angles come with respect to each other. So, if you look at 2 2 2
here the angles are 90 degrees with respect to each other.
The next slide is shown here they 2 2 2 orientation. You see that the three are oriented at
90 degrees to each other. You see the relevance of the presence of the threefold axis in
terms of the 60 degree angles because you now get 1 2 3 4 5 6 a hexagonal arrangement
of the points on which the point group operation occurs; similarly, the 4 2 2 and the 6 2
2. As I mentioned the relevance of the axis with respect to 54 degrees and some minutes
and some seconds and all that issue will tell us about the arrangement of the symmetry
axis which is shown here. At this moment we are not going into the detail of this
discussion, except to tell that it is not necessary in a crystal lattice if you have a b c
61
directions, it is not necessary that the symmetry should be restricted to those directions.
The symmetry axis can be anywhere.
So, these for example, in this case the symmetry axis is not along any given in this case
there is it is along the given axis, this axis the twofold axis. But you see that the threefold
axis is along the diagonal and in fact, threefold axis along the diagonal is a condition that
is required to have cubic symmetry. And, therefore, the symmetry decides now how
these proper axis should be oriented, how the improper axis will be oriented in case there
is improper axis.
So, we have to therefore, consider not just the PPP, but we have to consider the
possibility of improper axis also as combinations in the Eulerian case. So, for example, if
there is a proper axis and an improper axis; the first one is a proper axis, the second one
is an improper axis, the third one must be improper, it cannot be another proper axis.
So, there is no PIP possible that is because one once we have a proper and an improper
axis combination, let us go back to the previous slide. One once we have a proper and
improper axis combination and we generated an improper axis in there as a consequence.
So, suppose we have a 3, 4 and a 4 bar the third axis should be improper. So, the
combinations therefore, get restrain to PII, IPI and IIP. So, these are the things which
essentially restrict to four.
So, four possibilities are there P PPP, PII, IPI and IIP and these combinations therefore,
are illustrated below and this now define the remaining part of our point groups. See
remember we had 13 point groups proper improper and proper improper combine that
gave to 13. And we have to now reach 32 which is the 32 point groups that are allowed
in crystalline materials and therefore, we have to can have these combinations.
So, if you have all proper combinations as we saw just now with respect to the angles
which are these, the 6 of these. So, we will give rise to six of those. Now, the
requirement that it has to be a PII combination, will generate these 6. It so happens that
all these six are not independent of each other and in fact, for that matter the 2 ̅3 will
generate system which is 2 by m ̅3 ̅3. So, here you see that all these six are again
degenerate. There are different kinds of point groups. These 6 are a set of point groups,
these 6 are a new set of point groups; that means, 6 plus 6 now 12 point groups
additionally to the 13 we have already seen.
62
Now, these are independently generated point groups and therefore, these combinations
will have to find a place in the crystalline systems which we are going to have. So, this is
the reason, this is the basis of why there are different crystal systems. If there was no
differences in the crystal systems we would not have had these combinations. Now, these
combinations therefore, decide how our lattice should develop into different crystalline
motifs. The so called seven crystal systems which you have studied in starting almost
from your standard 12 is a consequence of this.
So, you would have never asked this question in school that when the teacher said there
are 7 crystal systems you believed it like well, god is there, right and then we never
asked any questions. Now, the 7 crystal systems come as a consequence of the presence
of the point groups and the combination of these point groups in three-dimensional
space, which obeys periodicity. So, very interesting, crystalline state therefore, is the
most interesting way in which we can study the objects because now the objects are kind
of restricted. It is like you know in a class the students are all sitting in their respective
places and we get them arrested in those places so that attention comes to you.
So, the attention if we want to get of the students one once the class is over is almost
impossible. So, when they are restricted to those positions, these positions which we are
restricted will generate the so called point groups and the presence of the satisfaction of
the periodicity is a fundamental requirement and that induces the possibility of seven
crystal systems. So, the 7 crystal systems and thirty two point groups now are very easily
understandable in terms of the requirements of symmetry which govern the periodicity in
the given system.
So, this is the state at which we see that we do now start getting the information on all
possible point groups. Again to define the point group you should remember that the
point group has this condition that if you have a point on which you operate this
symmetry, the repeated operation of the symmetry should bring back the same point and
unless that happens you cannot call it a point group. Putting that condition and also the
condition that there is periodicity in the lattice that is a repeats along the a direction x
direction b repeats along the y direction and c repeats along the z direction we have these
restrictions of having 1, 2, 3, 4 and 6.
63
So, 1, 2, 3, 4 and 6 will give rise to the possible crystal systems, and then the
combinations of these with respect to PPP, PII, IPI and IIP will give rise to the other
crystal systems. So, these direct crystal systems which we can get or the orthorhombic,
the trigonal, the tetragonal and the hexagonal systems and then these two now give rise
to the cubic system. If none of these symmetries are there and there is simply a 1
symmetry or a ̅1 symmetry that defines the monoclinic system. And, in this therefore,
the case where there is twofold symmetry or mirror sorry, it defines the triclinic system
correction. And, in case we have the mirror plane or the twofold symmetry it defines the
monoclinic system.
So, therefore, we have seven crystal systems, the triclinic system representing 1 and 1
bar the monitoring system representing twofold, mirror and 2 by m and therefore, these
things which we say now 1 and 1 bar belong to triclinic, 2 mirror and 2 by m belong to
monoclinic and then 2 2 2 to a 2mm belong to the orthorhombic and so on.
So, therefore, we have the orthorhombicity, the triclinic, the trigonal nature, the
tetragonal nature, the hexagonal nature defined in these cases and then the cubic
symmetry which is what also referred to as the isometric system because this system
does not have any restrictions along given a, b, c directions. Now, since there are no
restrictions and on given a, b, c directions we will see what kind of restraints we can put
to define these seven crystal system as we go to the next class.
So, at this moment we will stop at this particular point and go over to the next class.
64
Lecture - 06
Details of Stereographic Projections
So, in the last class, we stopped at the discussion of having three mutually perpendicular
2 fold axis generating the 222 combination as a possible point group.
2mm is another point possible point group and so on and with respect to 3 fold, 4 fold, 6
fold and then also the combination of 233 and 432 giving rise to access in different
directions. We find that there are 6 plus 6, so, 12 additional point group symmetry is
which come up. We have to therefore, consider the other combination. We see here it is a
proper, improper, improper axis, here we at considered an improper axis with a proper
axis; obviously, by the satisfaction of the Euler’s theorem it has to be IPI, we can also
have IIP.
So, that these combinations now we will give rise to a newer possible point groups and
those are listed here. We are not going to find out how they come up and how to derive it
and so on. The best way to do that would be to take a course on group theory and if
someone is interested, there are group theory courses also available in this program. So,
if they take the group theory course, there will be a possibility to understand how to get
65
to the other. This is not in the scope of our course. But what we note is here is that this
will introduce four new symmetry. So, this 2mm or m2m is one on the same; see again I
will repeat, 222 gives rise with PII, we get 2mm.
Now; that means, that there is a twofold axis along this direction let us say and m and m
in these two directions. Now, if I go to IPI; that means, I should have m2m. So, the
symmetry will be m2m because there are 90 degrees with respect to each other and we
have a very nicely define rectangular coordinate system in this particular case with 90
degree angles between them whether we have a mirror here or a mirror there or a mirror
there, it does not matter and. So, also the presence of the 2 fold axis.
And therefore, m2m and here in this case mm2, they become one and the same as 2mm.
So, there is no need to have additional point groups coming in this combination. On the
other hand, in this combination and additional one comes because ̅3 is something
different from ̅2 because ̅2 becomes m where as in case of ̅3 it is different and it is a
standalone symmetry and therefore, you get ̅32m.
Similarly, ̅42m, ̅62m and you also have in this case ̅43m. So, there four additional things
will come in. On the other hand, if you have IIP combination, all these combinations PII
will be there will be repeated here and in this case also it is repeated and therefore, we do
not get any additional point group symmetries. So, we have therefore, 13 plus 16 that
comes to 32. So, that comes to 29, not 32.
So, what is happening what is it we are missing because we, we want 32 point groups
and we have now getting 29, what happened to the other three? The other three will
come as a simultaneous presence of proper and improper.
66
You see in the previous slide, we had this proper and improper axis simultaneously
present. So, we have to therefore considered the combination of proper and improper
axis.
So, being simultaneously present and if we consider that, we get these additional three
possible point groups. For example, if you take 2/2, 2/ ̅2 , 2/ ̅2, 2/ ̅2 this is proper-
improper, proper- improper, proper- improper, you get a new point group symmetry
67
which is 2/ m 2/ m 2/ m which is not present in this combination, because 2/m, 2 / m, 2/
m is not generatable by these operations.
And therefore, 2/m 2/m 2/m is a new point group. So, on the other hand, 6/m 2/m 2/m is
also a new point group and with 622 and with 422 also we will get an additional point
group, the rest of them do not appear. And as a consequence, we have 3 more
crystallographic point groups with more than one axis each axis being proper and
improper simultaneously, that is what this table says and this particular table therefore,
now concludes that we have 29 plus 3 which is now 32 point group.
So, these are the various ways in which we represent the 32 point groups. So, let me
recount. We can go back here with set of proper axis 1, 2, 3, 4 and 6, we get the
combination of ̅1 ̅2 ̅3 ̅4 ̅6 which is now the improper axis associated with the proper
axis. Whenever we associate an improper axis with proper axis, we get a new point
group and therefore, this is 5 plus 5.
We can also have proper and improper axis simultaneously operating at a point and that
gives rise to three more which is 13 and then we see the combinations that are possible,
we consider not just the PPP, we consider the additional PII, IPI, IIP and these
combinations will give rise to another 16 of them. So, 13 plus 16 give us 29 and then
they aspect that we can combine both proper and improper axis is simultaneously gives
us the three more and that for we get the total 32 point groups.
Now, comes the beauty. Now, comes something which we have not anticipated or not
bother to understand over the years. When people said that there are 7 crystal systems,
we believe that there are 7 crystal systems and that was a belief rather than actually a
proof and we also knew of course, there are 32 point groups. Most of us in the college
days, we learn to presence of 32 point groups, but now what happens sees that if you
now take this point groups and clarify classify them, classify them if you classify them
into various classes where the importance of the angles between them becomes crucial,
sorry the importance of the angles between the axis becomes crucial, then you see that
we can clarify classify them into different crystal systems.
So, this gives rise to these so called seven crystal systems. The first one is the triclinic
system where we have a non-centrosymmetric possibility; that means, there is no centre
of inversions that is present with this unit cell, but this unit cell can exist by itself. So, the
68
symmetry that is associated with this is one. And then, the when we combine this with
the centro symmetry combination the center of inversion is put in we get a ̅1 .
We will not write now worry about this so called Laue class or the lattice point group nor
worry about the geometry of these seven crystal systems because the geometry of the
seven crystal systems we will automatically get as we see in the next class maybe or
even today later in the today’s class. But what is important is that the 7 crystal classes
have the point group identification with them associated with them is identified here. For
example, if you take a monoclinic system, you can have a twofold symmetry or a mirror
symmetry which is ̅2 .
Now, both of them do not represent a centrosymmetric system. Now, I am going to

clarify this issue because you may say that, you know the ̅2 is a combination of 2 fold
and an inversion centre. So, how come it is non-centro symmetric? It is non
centrosymmetric because you now examine your equivalent points. If you examine the
equivalent points, sorry if you examine the equivalent points in case of the mirror
symmetry, it is x y z and x y̅ z.
We operate x y̅ z mirror again, we go back to x y z we do not see the presence of the

inversion centre. Even though mirror in principle is a combination of the 2 fold and the
inversion centre, it does not generated inversion centre. That is very important and
therefore, we see that the so called non-centro symmetric systems where there is no
centre of symmetry automatically coming into the crystal system, this is either 2 or
mirror and in case centrosymmetry it is 2/ m.
Orthorhombic symmetry, we already saw that it is 222 and the combination mm2 is
allowed and mmm is allowed, mmm is actually 2/m 2/m 2/m, we will prove all these
issues with respect to the fact that the geometry of these crystal systems why these
geometry now defines incorporating the point groups. So, the presence of the point group
forces the crystal system to go into these crystal systems, forces the molecule to
crystallize in these crystal systems.
So, the molecule when it crystallizes is first is the point groups, hopefully, we do not
know is a chicken and egg problem whether the molecule decides to do what kind of
crystal it should go in to or the crystal system decides what molecule should come into it
is crystal system, that is an answer problem even though there are indications that the
69
way in which the molecules associate themselves will decide ; that means, the functional
groups or molecular molecules will decide what kind of a crystal system they go in to.
But that is something which is not an equivocally established. What happens in a when
you crystallize a given material is that there is what is known as the nucleus which is
formed initially and this nucleation how it is formed and why it is formed at what
rapidity it forms and so on has never been understood.
So, crystallization phenomena even though you control temperature pressure and so on
and crystallized the materials. It is a kinetic way of crystallization is governed by the
kinetics of the situation kinetics of the way in which we grow the crystals either from
made it or from solvent evaporation or whatever techniques we used to grow crystals, it
is a kinetic process. This is something which is not fully understood because we do not
know what is the nucleation process that involves the crystallization protocol.
Anyhow, let us go further. So, we therefore, now distribute the point groups, this is a
very important list very important table in which we now classify the point groups into
various crystal systems. As we saw in this diagram, the moment we have the 222, we
identify orthorhombic system. So, we there is no other system which can have a point
group symmetry is associated with 222, tetragonal the trigonal systems will have 322,
tetragonal systems will have a four-fold symmetry and hexagonal system will have a 6
fold symmetry.
So, when we actually decide how these symmetry elements distribute themselves in the 3
dimensional lattice, we will see how the nature of these orientation of the axis decides
the nature of it is crystal system ; that means, in the case of just give you an example,
you already know this result from your earlier college days, but we have never put 2 and
2 together. For example, in the case of a triclinic system, we say a is not equal to b is not
equal to c and interactional angles are not equal.
Now, the only symmetry element that can be present in a unit cell of that type is either 1
or ̅1 . That is because it is a point operation. The moment we bring in an axis operation,
we will have to bring in additional symmetry that is associated with the lattice. So, in a
triclinic system in other words, you cannot have any axis of rotation. In a triclinic
system, you cannot have a mirror plane. In a triclinic system, you cannot have three-fold
rotation, fourfold rotation and so on.
70
Because triclinic systems are restricted to either by themselves or the maximum they can
have maximum symmetry, they can have a is a centre of symmetry as is shown here and.
In fact, that is what is referred to as the Laue class. This is the maximum symmetry to
which it refers to and this is what you will find when you do the x ray diffraction
experiment which we will learn later on. And then, what also is indicated here is the
maximum symmetry that will get associated with this system which is this so called
lattice point groups. We will work out these two issues as I mentioned later on one once
we do the diffraction theory.
So, at this particular stage therefore, we have now done the following we have found that
in a periodic system, we get this periodicity only because of the fact that we restrict our
ourselves to 5 axis of rotation and then the restrict ourselves to a point about which we
can have a centre of inversion and we also restrict ourselves to the fact that whatever
combination of symmetries, we can have is restricted to these many.
71
And as a consequence, we found out that these many operations in crystals with or
without translational symmetry, living out the screw operations and the glide planes.
Their operations will coexist and this gives rise to the so called point groups. So, if the
operations are independent like this, the proper axis we have five point groups, five point
groups coming with improper axis proper axis and improper axis combinations will give
rise to 3 of them and then if you have more than one, they follow the Euler’s symmetry,
more than one symmetry intersects at a given point O, then the third one is automatically
invoked and based on that, we will get additional possible point groups and therefore, the
there are totally 32 point groups. You see that as the as the lattice nature changes, you
will see more and more increasing number of point groups which can be associated with
it.
So, the moment you have a tetragonal system, you always have a fourfold symmetry.
You have a trigonal system, you always have a threefold symmetry, hexagonal system
will always be a 6 fold symmetry. Cubic system, as we saw there is no 90 degree angle
information here. So, you will have a threefold symmetry along the diagonal. So, the
presence of the 3 fold symmetry along the diagonal distinguishes a cubic system from a
trigonal system where the trigonal system has a threefold axis about one of the axis.
So, these are some of the issues we will discuss more and more as we go into the 3
dimensional lattices and then from this point groups, how we can go to 3 dimensional
72
space. What is that which generates combination of point groups with the crystal systems
and the behavior of the translational periodicity that is associated with the crystal
systems, which generates eventually what we call as the space groups.
So, we will then see how the distribution of the symmetry elements goes to the symmetry
in order to this space groups information. Keeping this part in mind, we should always be
reminded and that is why it is next slide.
We should always be reminded that the crystal symmetry and the molecular symmetry
are different from each other. It is not necessary that a molecule with a certain symmetry
should crystallize with the same symmetry in the crystal system. For example, if there is
a system which is now like this, this has a very high symmetry, I this is the point group
symmetry which are which is indicated for each of these molecules.
So, we can identify the symmetry that is associated with each one of these molecules.
Take for example, this case this particular compound which has a Fluorine in up there
and a fluorine down here is having a centre of inversion at that point. The presence of the
centre of inversion at that point will make this molecule process the centre of symmetry.
What crystal system it goes into is unknown, it depends upon what kind of interactions
that this particular molecule which has a centro symmetry develops with it is neighbors
in the three dimensional lattice.
73
So remember, that in a crystal when you grow a crystal it is a three dimensional unit, so
the basic repeat motif is this molecule. So, this molecule now let us say crystallizes in
some space group. It is not necessary that it should crystallize only in a triclinic system.
It can have a higher crystal system depending upon the way in which this is packed
inside the unit cell. This takes you back to the (Refer Time: 17:32) diagram, you know
you remember we talked about the (Refer Time: 17:33) diagram and in the (Refer Time:
17:35) diagram, one issue we put in very strongly that is the close packing.
So, when these kind of molecules back into the crystal lattice there, remember that they
should pack closely and the nature of the close packing is crucial and way in which the
close packing is decided depends upon the interactions this particular molecule develops
with it is neighbours and these are known as intermolecular interactions. We will be
spending some time towards the end of the course on understanding these intermolecular
interactions based upon the nature by in which these molecules are assembled
themselves.
At this stage, I was already referring to the work of Kitaigorodsky. Kitaigorodsky said
that if the molecule has a symmetry and this particular symmetry will be found and must
be found in the crystal structure as well. Invariably the molecular symmetry will find
itself very comfortable is settling down in a crystal system which has one bar symmetry
it does not mean that it should crystallize this particular molecule should crystallize in a
monoclinic system. I would suggest that you refer to literature and see this particular
molecule and see what crystal system it goes into. Just to keep the curiosity alive.
Similarly, what happens to this planar 6/m m m which is the symmetry that is associated
with benzene? Benzene is not going to crystallize in a hexagonal space group. Find out
where benzene crystallizes under what conditions benzene crystallizes. So, this opens up
the possibility of molecules crystallizing in more than one possible crystal system and
this is where the interest comes in pharmaceutical industry.
Because pharmaceutical industry wants molecules to crystallize in different crystal

systems because the physical properties associated with the crystal depend upon how the
molecules pack inside their crystal lattice. The physical properties by which I mean the
melting point, the solubility, this so called bioavailability which is very crucial
74
pharmaceutical materials, they need to have the molecular symmetry and the crystal
symmetry go hand in hand.
Because of the fact that the molecules, they are very clever the keep that symmetry,
benzene cannot cannot be other than having a planar 6/mmm symmetry, but for example,
a cyclohexane can have boat it can have a sofa configuration, it can have a half chair
configuration, it can have a twist chair configuration, it can also have a chair
configuration. So, depending upon the type of substituents you put on this particular
crystals they of this molecule, it will decide which kind of a minimum energy
configuration it would like to adapt. So, it has this flexibility whereas, here there is no
flexibility.
Now, because of the fact that there is no flexibility here, it does not mean that it should
crystallize in a system which has this 6/mmm space group 6/m2/m2/m space group. So,
this space group in which it crystallizes once again, I suggest you refer to the literature
find out in which crystal system benzene goes into and under what circumstances
because we although that benzene is a liquid at room temperature. So, when you
crystallize, you have to do some tricks either by reducing them reducing the temperature
or by applying pressure and then find out what happens to the crystal structure.
So, this therefore, gives rise to the possibility and innumerable number of possibilities
particular line pharma industries to design what they called as polymorphic
modifications. So, each and every molecule shown here can have different kinds of
polymorphs and the presence of polymorphism is crucial for the pharmaceutical industry.
So, there is an application potential right away coming up for the industry people this is
also true with materials because depending upon the type of molecule and the molecular
property, the material property will be decided once again by the nature in which it packs
itself inside the crystal lattice.
So, the crystals therefore, not necessarily should have the property of the molecule, but
the molecular property can be tuned to the property of that could be displayed in the
crystal system. So, this therefore, opens up the possibility of studying more than one
crystal structure of a given molecule. And so, polymorphism issues both in material
science and pharmaceutical industry become example important. So, that is why I
showed some examples of molecular symmetry here, just to tell you that for example, the
75
presence of mm2 symmetry in this does not mean that this molecule will crystallize only
in the orthorhombic symmetric. It can crystallize in many other symmetry. So, the
symmetry of the molecule is not necessarily the symmetry of the crystal structure.
So, the this is what I was telling that if there is a tetrahedral carbon with A B A B
substitution, there is a tetrahedral combination with all a is kind of substitution, you see
that the point group symmetry are different, the crystal symmetry may be more very
much different. So, this may crystallize in a tetragonal system, this may also crystallize
in a tetragonal system, but the molecular symmetry is still kept intact. When the
molecular symmetry is found to have a certain information which it can use in the crystal
system, when the crystallization takes place the crystallization will be in such a way that
the molecular symmetry is kept intact.
And therefore, it is very likely that these symmetry in the molecule is not destroyed.
What is destroyed is the way in which the molecules are arranged inside the unit cell. So,
the question of minimization of energy comes in. This is how prediction of crystal
structures has become a very big issue because you know unless you make a crystal and
solve the crystal structure. It is not possible to a priori you predict what crystal system it
is molecule can go in to and this has been an exercise which has been practiced over the
years and it I must mention that this stage that we have not achieved anything
significantly appealing in order to say that wow, I can predict the crystal structure given
the molecule. People have done this exercise, this is area is called crystal structure
prediction and this crystal structure prediction area is now a very flourishing area where
people are taking the molecules and then predicting the possible crystal structures into
which they can go based on various thermodynamic and kinetic parameters. They are
trying to control both thermodynamic and kinetic parameters as because I already
mentioned that crystallization is a kinetically driven phenomenon.
So, based on that energy minimization of energy is performed in a possible unit cell
which then is modified such that the minimized energy configuration eventually results
for the closed packed crystal structure where we have the molecules closely packed in
the unit cell. So, the unit cells are define and the possible structure are predicted. For
example, structures of very many pharmaceutical compounds are earlier predicted, but
the problem is that it either goes into one of the possible predictions or it will never go in
to one of the possible prediction.
76
So, the conclusions drawn from crystal structure prediction are still to be testing. So, this
approach of crystal structure prediction using molecular dynamics simulations and so on
is a upcoming field, but it is not a field where we can have hundred percent success at
this moment. So, there is a lot of improvement required, there is a lot of advances that
they one has to look at in terms of calculations particularly to get to the final prediction
of how a molecule can go into a crystal structure.
Now, we come to the operation itself, the symmetry operation itself. We are still not out
of the understanding of the symmetry operations. So, the way in which this particular
slide is prepared and demonstrated is to show what happens to the objects when they find
themselves in a situation where they see these point groups. For example, if a if an object
the object is referred to by this big circle with a positive and the object is referred to a
circle with a comma, whenever this subject in wards itself we put a comma inside the
object. So, whenever the object is inverted we put a comma inside the object and when it
is coming towards you, this is shown as plus when it is going inwards it is shown as
minus.
So, in this example for we have illustrated the suppose we take the coordinate of this
point as x y z, then the coordinate of this particular point is -x -y -z; that means, these
two now are related by a centro symmetry and the centre of symmetry symbol is used
here the little circle, open circle and the symmetry is illustrated below. So, all possible
77
symmetries which we have discussed with respect to the 2, 3, 4 and 6 are given in this
list. So, this particular list therefore, has all the possible ways in which you can generate
now objects in this 3 dimensional lattice. Here is only issue which I will discuss and
probably stop today that is to discuss, what happens if this object is here?
So, suppose the object now decides to come to the position of the centro symmetry, then
what happens to this particular object? So, this then will be moved here and then it has to
generate x y z to x̅ y̅ z̅. Suppose, this is now also the origin of the crystal system 0 0 0,
then it will make an image of itself on 0 0 0 ; that means, we will have to show this
something like what is shown here ok.
In the case of a ̅6 symmetry, you take this picture imagine this fellow to be moved at to
that particular position and that is exactly what will happen; that means, now if this is
representing a molecule, the molecule now must have a centrosymmetry because it has to
sit on the centre of symmetry and then this particular molecule which sits on the centre
of symmetry is related to itself.
So, the diagram now will look like taking out this fellow and pushing it to the one bar
position. Such a position such a situation is referred to as a special position in this
condition. So, whenever there is a object which sits on the symmetry element itself, it
becomes a special position in any given system. So, any point group which we have
discussed so far, if this sits on a special position, then it becomes a sits on the symmetry
element it becomes a special position.
So, in this particular case, the number of equivalent points generated will become only
one because 0 0 0 will give rise to -0 -0 -0. So, it is half and half. So that means, the
occupancy of this particular position when we are sitting at the centre of symmetry
becomes half.
So, in this case, the occupancy is full this becomes half. So, there are two independent
possible positions for this particular compound or the object in this particular case
molecule; one is related to the other by inversion centre; however, the number being 2
will reduce to 1 in case this sits on this is the inversion centre and that is how the
possibility of having a centre of symmetric molecule occupy central a molecule with a
with an inversion centre, how it can go and sit at the inversion center.
78
So, the molecules now will be repeated on through itself and therefore, we will get now
only one equivalent point here. There were two equivalent points, it becomes only one
equivalent point and the occupancy of that position now becomes half. So that means,
only half the molecule if there is a molecule of the type we discussed before, suppose
this particular molecule is sitting at the at that ̅1 position, then this half of the molecule
will be related to the other half. So, when you have to worry about is to determine only
one half of the molecule.
So, that way this is called the occupancy. So, the occupancy of that position becomes
half. So, please note the whole idea of the first few classes which you have gone so far as
I mentioned also earlier is to learn the grammar; we are actually learning the grammar
one once we have identified the alphabets the so called point groups and all that will
form the alphabetical notation in crystallographic language.
And what we are trying to do is to develop the grammar, the various ways in which we
can have an adverb, the way in which we can have an adjective in a language. So, the
crystallographic language therefore, has several new words introduced. Today for
example, may introduce equivalent points we introduce the various symmetry elements
called the inversion centre, we set special positions, we set the occupancy, all these are
individual grammatical terms which are necessary to develop a crystallographic
knowledge. At this stage, we will stop today and then we will go for forward in the
coming classes.
Thank you.
79
Solid State Structured Chemistry Unit
Lecture - 07
Arrangement of Symmetry Equivalent Objects
So, in the last class, we looked at the types of symmetry operations and the symbols that
we give for the type of symmetry operations. For example, in this very first illustration
here, you see the presence of an inversion centre. The presence of the inversion centre is
indicated as ̅1 which actually tells us that if a point x y z is taken in 3-dimensional space,
it goes to the x̅ y̅ z̅; that means, -x -y -z.
And now we operate the same one bar operation on this x̅ y̅ z̅. It will come back to x, y,
z. Now, as we defined in the last class such an operation is referred to as the point group.
So, this now represents a point group symmetry and that is represented by a small open
circle which represents ̅1 . So, in literature, you may find just this return or in literature
you may find only ̅1 written. And when you want to display the objects and their nature
on which they are present in this using the ̅1 symmetry, it is normally represented this
way.
So, the object is written as a bigger open circle and the sign positive tells that it is
coming towards us, towards me in this example and then when we operate the ̅1
80
symmetry this x and suppose this is represented in the x y plane, then this goes to minus
x minus y and then of course, z goes to -z that is illustrated by the presence of a 'comma'
in the middle of the open circle. So, the operation x̅ z y̅ is indicated by a negative sign
here which is taking the indicating that this object which is in front has gone to the
backside of the plane of this representation.
So, essentially we therefore, generate two equivalent points with a symmetry operation
which is the centre of inversion. And if the centre of inversion in the crystal context is
also referred to as the centre of symmetry because the object here and the object there are
related to each other by a centre of symmetry that is how we interpret. It also tells us that
if by chance the object sits on the centre of symmetry; if it is located at the ̅1 position,
then you see that it will be represented by let us say in this particular example x is 0, y is
0, z is 0 that defines the origin. And the centre of symmetry sits on the origin, then you
see that 000 and -0 -0 -0 will represent the same point. And therefore, the number of
objects that will get repeated in this point group operation will remain the same. In other
words there will be only 1 symmetry operation and that is referred to therefore, as this
special position.
So, in this special position, you will have the object both positive and negative indicated
at that particular position. In fact, the if there is a nomenclature which has to be indicated
here, you can remove the open circle and put the closed circle and put a representation
which is something like this which is shown here; I just show you the example. This unit
which is half plus and half minus with a negative can be taken up and put over there
saying that half the objects of object will occupy this position.
So, the number of equivalent points in general is 2 and the number of special positions in
this particular example is 1. So, there is special positions are a special situations with
respect to the position x y z and if the x y z sits at a centre of symmetry which could
which is shown in this diagram at this position and it could be anywhere in 3-
dimensional space. Whenever there is a centre of symmetry that is present in the three
dimensional space, it will occupy the so, called special position and therefore, we will
have half the molecule or which if you consider this as a molecule; half the molecule will
be present at that position.
81
So, this therefore now tells us the importance of symmetry representation. It tells us the
representation of equivalent points. It also tells us how the objects are oriented with
respect to each other this representation. Notice that this representation also is effectively
bringing a 3 dimensionality because we are talking about this object being in the front
and that object being in the back thereby indicating the third axis which is perpendicular
to the plane of the representation here.
We now go to the next example which is the so called the two-fold symmetry, and in this
particular two-fold symmetry, this is the symbolic representation it is it is marked with a
close figure instead of a open figure like that; it is a closed unit here.
And again you see that these symmetry operations now, are our for any given operation x
y z. It is -x y -z or x̅ y z̅ which indicates that the object which is shown here undergoes a
180 degree rotation. So, x changes to -x, z changes to -z. This is about the operation is
about the y axis. This also tells us that the y axis is now coming perpendicular to the
plane of the representation which we have here and therefore, the y axis is now
perpendicular and its coming towards me and going inside the inside this particular
representation. So, if that is the case there is no change in y coordinate and so the object
which is in front remains in front, object which can go to the back remains in the back.
So, if you have an object which is now a open circle with a negative sign it will now be
an open circle with a negative sign. So, both possibilities occur in case of a two-fold
symmetry; that means, to say that the two-fold symmetry, suppose let us say for the sake
of argument, that there is the object which is present on two-fold itself. I want you to
think about it and see how you write the equivalent points. In other words if x y z is 0 0
0, what would be the representation of x̅ y z̅ ? Will that be a special position or not? I
think it is better you think about it as an audience and the participants of this particular
workshop, rather than I explain.
It is also another issue which comes up here that since there is no origin definition here
because if we have a object in front, this object is also in front; if there is an object in the
behind, the object is also going to the behind part and therefore, the location of the two-
fold axis could be anywhere along the y axis. This is a very important point. It is not
necessary that the two-fold axis should coincide with 000; it can be anywhere along the y
axis. Hence this kind of a representation is still valid that is both plus and both minus
82
depending upon where the two-fold is located which means to say that whenever we
have a two-fold axis about a direction that particular two-fold axis can be anywhere
along that direction and there is no fixed origin for that.
This we will see more in detail when we look at the monoclinic symmetry and the way in
which the space groups developed, we have not introduced the world space group as yet,
but when once we introduce that we will see how it how it operates. I also want you to
look at this particular representation and see that it is represented as ̅2 . I also want you to
look at this and make sure that ̅2 is equal to m where, m is the mirror symmetry.
So, the equivalent points there will be x y z and x y̅ z and that is something I leave again
for you to work out as a home exercise. So, if you look at this particular representation,
in this representation the mirror plane is indicated in this fashion showing that the mirror
is now coincident with the plane of the representation here. So, that is coincident with the
plane of the board and the operation will take this part of the object to the other part of
the object on the other side of the zero and it will be therefore, plus and minus and you
have a 2 bar axis which is nothing, but a mirror.
So, we therefore, have covered it now one. two and three symmetry operations and these
3 symmetry operations are all point group operations because by definition they define
the requirement of a group and therefore, these 3 point group operations that are now,
considered among the 32 point groups which we will display later on.
The next operation is the screw axis and in this screw axis operation, what we will do is
to see the operation is done in such a way that now we have to consider an axis which is
coming perpendicular to the plane of the board and that particular axis has a certain
dimensionality. Now if that axis has no dimensionality remember when we described the
two-fold symmetry, we said two-fold can be located anywhere along the y axis. Now it is
important to define the 0 of that two-fold axis because there is a translation component
associated in that direction; that means, this operation will exist only in crystalline
materials. So, the so called translation involved the symmetry operations like the 2 1
screw symmetry, we will exist only in crystalline operations, crystalline objects and
crystalline materials.
Here again the object which is shown here will now undergo a 180 degree rotation and
then there is a half unit translation along the direction of the axis of rotation. Let me
83
repeat, the axis of rotation is associated with the y direction; the x and z values will now
go from +x, +z to -x, -z. Now you might see the difference between this representation
and the representation over here where the value of y is also changed and therefore, we
have put a 'comma' inside the circle. Here we do not put a comma inside the structure
because this has only undergone a rotation of 180 degrees.
So, this plus remains the same therefore, 0+ and to that we add a unit of half which is a
consequence of the presence of the axis in a crystal. So, this particular symmetry
operation, they will occur only in crystal systems; that means, to say that this 2 1 is not a
part and parcel of the point group symmetry. This is a very special symmetry which
exists only in crystalline materials. So, please note that point.
So, we will not now cover all the 2 1, 32, 41, 42 etc because these are all operations which
will come as a consequence of the rotation axis having a translation component elongate
direction in a crystal and we do not want to consider crystalline materials, at this moment
we want to consider only operations about a point. And when we talk about operations
about a point translation symmetry is not the component which we have to invoke and
therefore, we will consider only this part of the diagram.
So, we now go to the three-fold symmetry. In a three-fold symmetry suppose this is our
main object, this undergoes rotation of 120 degrees goes to that particular point and then
another 120 degrees will bring it to this point. So, this is a three-fold symmetry operation
indicated pictorially. I have not purposely written the equivalent points because that will
come as a part of your assignments later on.
So, this equivalent points there will be now 3 of them and these three will be generated.
So, these assignment will be for you people as well who are in the audience here. So, this
now operates the three-fold symmetry and this three-fold symmetry now gives 120
degree. So, there are 3 objects now generated. Now we take any object in this among the
3, then you operate the three-fold symmetry; it will come back to the same object. So, the
point group for definition is intact even in this particular case.
As we go further down we have the four-fold symmetry and the six-fold symmetry
which will follow likewise. In the case of the four-fold symmetry we will get 4 objects
and in the case of the six-fold symmetry, we will get 6 objects and this therefore, ̅1 2 3, 4
and 6.
84
Now, define the point group symmetry, the ̅2 operation which is also the mirror
symmetry operation is also indicative of the a point group symmetry. It also tells us the
fact that if we have a point group operation, the operation which is now taking that on to
itself across the centre of symmetry or the inversion centre will still be a part of the point
group because this has no translational component. Please note, that if we now consider a
two-fold axis and then operators inversion centre about the two-fold axis, these 2 will
now contribute to a point group symmetry that is because there is no translation
component involved in any direction.
This is just the change of the axis coordinate that is x to x̅ in this particular case y to y̅.
So, there is no change in, no additional component coming like it came in this example
where we had x y z and x̅ 1/2+ y and z̅. So, whenever this additional half comes, it is not
a point group symmetry. So, you should note that, but; however, this operation is valid in
crystalline materials and that will become an additional points group symmetry which
can operate, even though it is not classified under the point group symmetry.
So, if you consider now ̅3, this now takes the positive to the negative, positive to the
negative, positive to the negative. We generate 6 equivalent points. These 6 equivalent
points will become 6 equivalent points in case of ̅6 which is demonstrated in this way.
Once again this could form a part of your assignment, I want you to see why ̅3 which
generates now 6 of these and ̅6 which also generates 6 of these equivalent points are
different from each other.
The symbols inside are already indicating why they are different from each other. Here is
a three-fold with a centre of inversion at the middle and here is a six fold rotation with a
three-fold symmetry in the middle and this makes the difference. And the similarly we
have of course, the ̅4 symmetry which can appear which along with the 4 symmetry.
There is no additional concept associated it to our additional feature associated with it.
It is of interest to note the difference between ̅3 and ̅6 and to see how ̅3 and ̅6 are
different from each other. So, these therefore, are some of the point groups and the
representations in terms of equivalent points, in terms of position of the objects inside
this operational unit and at the same time were indicating how many objects are
generated inside this one unit operation. So, these are therefore, representatives of the
85
point group symmetry whereas, these on this side are representatives of the symmetry
whenever we have a crystalline system.
And now we will go to the next slide where I have indicated the equivalent point
operations. In case the operations are about in a crystal and there about various
directions, but this particular slide we will refer to later on. At this moment we will just
notice that the two-fold axis we will generate x y z if it is along the a, direction x y̅ z and
if it is along the b direction, it will generate x̅ y z̅. This will be x y̅ z̅ sorry this is z̅ and if
it is x y z; the c and the two-fold axis is parallel to c, we get x y z and x̅ y̅ z. I am not
introducing the concept of a b c at this moment because we are yet to define the concept
of a unit cell and that is something which we will take up next and then we will be
referring back to this particular slide again and then we will see how these operations
develop.
86
So, this brings us to the issue of defining lattices and crystal system. What are lattices
and what are crystal systems? As we just now saw from the symmetry operations and the
point group symmetry operations that the operations involving 1 2 3 4 and 6 which
represent the proper axis operations are point groups.
So, there are 5 point group symmetries which come up from this context and another 5
point group symmetries which come from ̅1, ̅ 2, ̅3, ̅4 and ̅6. So, totally we have
therefore, 10 point group operations which can come from here and 3 additional ones
proper and improper axis which comes from this. This we have already seen in the
previous class.
So, you see that in the case where we have a no symmetry or a ̅1 symmetry which is also
the proper and improper axis generating the same ̅1 symmetry. It is now required that we
have to define a crystal system associated with this symmetry operation. It is also
required to define a lattice symmetry associated with this operation. Now what is a lattice
and what is a crystal system? A crystal system is one which represents essentially the
dimensions of the unit cell. Now we saw in the very first couple of classes that we saw
the Escher's diagrams and there we defined a unit cell.
Unit cell is the one which now repeats itself in all 3 directions a, b and c to generate the
entire crystal, but whatever is present inside one unit cell is exactly the same as in the
second unit cell as far as crystalline materials are concerned. There are cases which differ
87
from each other, but essentially we are restricting ourselves in this course to crystalline
material which display unit cell symmetry and this therefore, is referred to as the unit
cell. The dimensions of the unit cell define the so called crystal system. The dimensions
of the unit cell are defined with respect to the 3 axis a b and c. The 3 axis a, b and c and
the inter axial angles alpha, beta gamma.
So, these 6 parameters define the nature of the unit cell and therefore, this now refers to
the way in which a, b, c, α, β, γ come up refers to the unit cell and therefore, it refers to
the crystal system. The lattice repeats these unit cells in all 3 directions. So, the
essentially when we talk about the lattice, it is like you know a framework.
The framework is built up of a, b, c, alpha, beta, gamma which is the unit cell and now
we can move the unit cell along the lattice points. So, the edges of the unit cell if we are
considering them as lattice points, the edges of the unit cells now we will define the
lattice points which now can be put across with translations along a, translations along b
and translation along c directions. And therefore, we will generate these unit cells in such
a way that we have the a, b, c directions repeating the translational periodicity.
The fact that we have translational periodicity in crystalline lattices and crystal systems
produces the lattices. So, the lattice sometimes is described in textbooks as an imaginary
concept, because the lattice can be present anywhere whereas, the unit cell one once we
define the origin, will be associated with the origin of that unit cell. And therefore, the
value of a, b and c alpha, beta, gamma define the unit cell. So, the crystal system
therefore, which now defines in term these define in terms of the unit cell and with
respect to an origin is unique with respect to the arrangement of molecules or atoms
inside that particular moiety.
Once whatever is there in the unit cell gets repeated in the following unit cells that is
controlled by last lattice translations. These lattice translations are again a, b and c in
three mutually perpendicular directions, but; however, this lattice can be infinite. So, the
points which we consider now need not be the ones which are associated with the origin
which we have defined, it can be anywhere in the 3 dimensional space, but the repeat
unit will still be the same and that is why the term lattice. It is something like you know,
we have a window and we have made a frame for the window, the same frame can fit all
88
the windows which have been put in that room because they have symmetry and so, we
can take the framework from one point to put the framework on the other point.
So, the movement of the framework is independent of how and why these lattices are
made, this can be considered a lattice. So, that is why in some textbooks lattices are
defined as imaginary concepts. There is nothing imaginary about it; it is just a framework
which can be slides along they already well defined unit cell concept which has the
translational periodicity.
So, the presence of the translational periodicity makes it a realistic definition. So, we
very often refer to in crystallography to unit cells and determine the unit cell. And one
once we have the unit cell determined on the crystal system identified, we will say that
we have already got the information above the lattice. The nature of the lattice is evident
in terms of determining the position of the location and directions of a, b, c, alpha, beta,
gamma and therefore, the definition of the units. I think that we will become clear as we
go along when we define it in terms of pictures, but what I see here is, what I show here
in colour is the crystal systems which develop by single proper axis or improper axis or
proper improper axis being present in the system.
Suppose there is a crystal system into which an object goes into and that particular
crystal system has only 1 axis of rotation of course, 1 is 360 degrees. The ̅1 operation is
the presence of the inversion centre and therefore, that is if we now want to define a unit
cell unit cell being of course, in two-dimensions, I will represent the unit cell like this.
We will see the plane lattices in a minute.
89
Suppose I say that I have a unit cell, this is the origin. Suppose I say this is a, that is b
then you see the value of a, the value of b and value of c depend upon what where is the
axis located and in what way the objects operating objects move around that particular
axis. So, since there is no axis with respect to 1 is a 360 degree rotation, and the only
possibility is the presence of a centre of symmetry. Let us say, it is located at this
particular point O, we do not have any restrictions on the value of a and b and we also do
not have any restrictions on the value of γ, the angle between a and b in this particular
plane.
So, by convention the angle between a and b is γ, the angle between a and c is β and the
angle between b and c is α. So, that is the convention. So, the a, b, c, α, β, γ have no
significance when we have only a single operation which is now\ 360 rotation. So, if the
axis of rotation 1 fold, then there is no restriction on the values of a, b, c, α, β, γ and that
defines what is known as a triclinic system. When we go further and go to the presence
of a two-fold axis, the two-fold axis now has the restriction of x, y, z should be going to
x, y, z; should be going to -x y - z assuming y is the direction of the two-fold axis about
which the two-fold is located.
So, x y z and x̅ y z̅ or the 2 equivalent points; if this restriction is put because of the
presence of the two-fold and because of the presence of the two-fold width, an inversion
centre which makes it a mirror symmetry or because of the combination of two-fold and
90
a mirror which is perpendicular to the twofold, then the values of a, b, c, α, β, γ will
change and later on we will see this then becomes what is known as a monoclinic
system.
Now, in a monoclinic system therefore, we have no restriction on a b c values, but α and

γ have to become 90 degrees if we consider y axis as what we call as the unique axis. We
will see them in terms of pictures as we go later on to make it very clear. So, what I want
to show here in this particular part of the talk is to say that there are independent single
axis operations and along with these single point operations if we make them also
improper axis and consider the combination of proper and improper axis which
effectively makes it the consideration of 13 point groups.
Among these 13 point groups, we can have 1, 2, 3, 4, 5 crystal systems. And in fact, the
triclinic system the monitoring system are more or less unique with respect to these
allotments whereas, the trigonal tetragonal hexagonal systems as I have indicated here,
the trigonal tetragonal and hexagonal can also have more than one axis in addition.
We have already seen this one axis operation, more than one axis operation and if it is
more than 1 axis, the Euler’s theorem takes over and based on the Euler theorem, we
must have a third axis coming in. All these details we have already seen in the earlier
classes and therefore, trigonal, tetragonal and hexagonal can have more than 1 axis. So,
the presence of the proper axis and improper axis single pole single axis operation
uniquely generates triclinic and monoclinic systems.
Trigonal, tetragonal, hexagonal have these along with that we can also have more than 1
axes that may be present and orthorhombic and cubic systems on the other hand have
more than 1 axis intersecting at a point and that is why the 2 systems have some special
properties. What special properties get introduced onto these crystal systems starting
from trigonal, triclinic to the cubic system, we will examine when we examine the seven
crystal systems in the next coming two classes.
However, we take home lesson from this last half an hour discussion is essentially the
following. We now looked at we have looked at the operations of the point group
symmetries, we have seen how the point group symmetries take the object to different
object and again bring it back to the same position, obeying the definitions of a group.
We have also seen the operations like the ones which involve a translational part of the
91
periodicity; in case of b symmetry it is half and so on. And we also have seen how they
generate these lattices and crystal systems in these generating seven crystal systems. So,
we will stop at this point.
92
Lecture - 08
Introduction to Plane Lattices
Yes. So, in the last class we saw that there are 7 crystal systems and we associated both
the single axis operation and also the operations which are more than one. Following the
Euler’s theorem, it has to be all 3 axis having rotation symmetry. So, we discussed those
points and then we said there are 7 crystal systems. Now how do the shapes of the 7
crystal systems develop?
So, we here is something which I have written crystals use different shapes of regular
box to build up into a crystal. This is very interesting statement; that means, that the
different shapes of a regular box. You see the world regular means that it cannot have
dimension which is not available for the atom to occupy or the molecule to occupy. In
other words every atom has a certain volume and every molecule will have a molecular
volume and we know that by the definition given so far that we have to have close
packing in a crystal.
So, when these conditions are put in the definition of a box, the shape of a box will have
to be regular, we cannot have an irregular shaped box. So, therefore, we have to have 3
93
directions and 3 interaction angles between these 3 directions defining a regular box.
And this regular box there are 7 different ways in which we can develop them into
crystals because remember these boxes will have to follow the translational periodicity.
So, if they have to follow the translational periodicity and still be regular boxes, in
otherwise these boxes have to be packed in such a way that the repeat along a direction
like that, b direction like that and c directions like this. Imagine the situation where these
boxes which are shown here. For example, the box which is shown here for the triclinic
system. So, the presence of the a direction here will repeat itself in the next unit cell, the
next unit cell and so on; b also will repeat itself like that and c also will repeat in that
particular direction. So, the repeat units a b and c.
So, if we now consider this as a unit a, this unit a now repeats in that direction in that
direction and so on, it forms eventually the crystal. If we consider this as the unit b, this b
goes like that and like that and like that. So, it will extend and define our lattice and so
also this c direction right.
So, if that is the way in which we want to have a unit cell, there are 7 different shapes
and these remember the shapes come because of the nature of the rotation axis which we
are using. So, in the case of the triclinic system, there is only one rotation axis which is
the one-fold rotation axis; that means, it is a 360 degree rotation. And then the only other
operation which can exist in a triclinic system is the centre of symmetry operation or the
presence of an inversion centre.
The inversion centre can be anywhere in this box. In fact, it can be anywhere in the 3
dimensional symmetry that is repeated by this box and then we can always taken origin
which is generally represented at this point in this diagram. So, that we have a different
the value of a, defined value of b and a defined value of c. The angle between a and b is
it the stated as γ here. So, as I mentioned earlier in the previous class the angle between a
and c will be β and the angle between b and c will be referred to as α.
So, the conditions which define a triclinic regular box will be a not equal to b not equal
to c and α not equal to beta not equal to the gamma and also not equal to 90 degrees; if
you which you can write that because the values of alpha beta gamma are not equal to
each other. Whenever these values are not equal to each other, the box length is a, b and
c which are again not equal to each other we have a triclinic symmetry.
94
So, if you identify somehow by some method which we are going to develop during the
next series of lectures, somehow identify the method in which you can now see that the
box the unit cell which can be describing this particular crystalline product could only
the triclinic. That means that there is no other symmetry that is present other then the
possibility of the presence of an inversion center.
So, other than the inversion centre there is no other symmetry that will be present accept
of course, the 360 degree rotation axis which is the 1 symmetry. The moment we bring in
the aspect of a two-fold symmetry things changed rapidly. The shape of the box in order
to have the 2 two-fold symmetry equivalent satisfied; that means, the equivalent points
now will be x, y, z the equivalent points; now will be x, y, z and assuming that b is the
unique axis in this particular case b is shown as the unique axis. So, then it will be -x, y -
z.
So, in this particular unit cell the equivalent points are x y z , -x y -z and you see that the
diagram which is now shown here is like a squished book. You take a book and then
proceed it up like that it will become a monoclinic system. The values now will be a not
equal to b not equal to c, alpha and gamma are 90 degrees. So, the fact that we have a
two-fold symmetry makes 2 angles 90 degrees with respect to each other.
Whenever you have a two-fold symmetry in whatever direction, you have in higher
symmetry systems as we go further we can have the two-fold symmetry in any other
direction other than the a b and c. In such situations we can therefore, find out the fact
that whenever such a thing happens the two-fold symmetry introduces to 90 degree
angles across the direction of rotation.
So, if the rotation axis is along b we therefore, get alpha and gamma is 90 degrees beta
will be non 90. So, this is the condition for a monoclinic system, then this is the second
regular box, which we will have which represents the symmetry therefore, of 2. Along
with 2 the other point groups which go as we saw in the previous class 2 ̅2 which is
mirror and also 2 by m which is a combination of proper and improper axis. So, there 3
point groups go with a monoclinic symmetry. The triclinic we have already seen it is 1
and ̅1, in this particular case it is 2 2 by m and m. So, these 3 symmetry elements, these
3 point group symmetries are associated with monoclinic systems.
95
So, a crystal forming a monoclinic going into a crystal system monoclinic will always
have the presence of either a two-fold axis or a mirror plane or a combination of a two-
fold and a mirror plane. Anytime such a combination occurs and nothing else no other
symmetries are developing, the crystal system will uniquely be monoclinic. So, the
definition of a unique triclinic and a unique monoclinic, we have already seen.
Now we will go further to the third system which is the orthorhombic system. In the case
of the orthorhombic system, the values of a b and c need not be equal. This is like a Bata
shoe box polis, I should not be saying Bata a shoe there are so many companies, but Bata
is a company, but you are familiar with a Bata shoe box because most of you by Bata
shoes, I do not know why. Whenever you by a shoe ok; let us make a general.
Whenever you buy a shoe it is put in a box, I do not know why they do it its very flimsily
made and we come home and threw it away, but it is put in a box and that box is given to
you along with your shoe or chapels are whatever you buy. And that dimension is a not
equal to b not equal to c. It so happens that there are 3 mutually perpendicular two-fold
axis.
Now, this is bringing back the Euler’s theorem because when we have a two-fold axis
along, let us say a direction here and a two-fold axis along the b direction here. Let me
they point out with a laser pointer. Suppose we have this is the origin, this is the a
direction and this is the b direction and that is the c direction. So, if you have this
particular direction as the a direction, there is a two-fold axis here.
The moment we have the two-fold axis, we told that if the two-fold axis is along the b
direction, then 2 of the angles should be 90 degrees. So, if it is along the b direction,
what are the 2 angles that have to be 90 degrees? In this particular case, the b axis was
the unique axis, the 2 angles for alpha and gamma. So, if this is the a axis, now the 2
values will be beta and gamma.
So, with along the b direction, we will have the alpha and gamma 90 degrees, along the c
direction we have the other 2 angles 90 degrees and therefore, we get a condition a not
equal to b not equal to c and alpha beta gamma are 90 degrees. So, this is another regular
box which is representing a unit cell. So, crystals which go into this will have to obey all
the three, two-fold axis symmetries which are operating on that. So, every point in this
box now will have a 2 2 symmetry, a 2 m m symmetry or 2 by m 2 by m 2 by m which is
96
referred to as m m m symmetry. And these 3 therefore, will refer to the point groups 3
different point groups. So, we have 1 and ̅1 here 2 m and 2/m here and 222, 2mm and 2/
m2/m2/m which is represented as mmm. So, we therefore, have covered 3 plus 3 plus 2 7
point groups symmetries, by going from triclinic to orthorhombic.
Now, as we go to the next slide, I will show you the remaining part of the unit cell of the
7, we have 3 of them covered; we will have 4 more to cover. And these 4 more are the
tetragonal, the cubic, the hexagonal and there is a very special system which is referred
to as the rhombohedral. So, we therefore, have a total of 7 crystal systems. In the
tetragonal symmetry, we will have the presence of the four-fold rotation.
So, the tetragonal symmetry can have a simple single fourfold rotation. The four-fold
rotation will ensure that if certain conditions on the type of the regular box, we can have
it is like a column or a pillar, the base will be a square. So, a is equal to b and then this
particular direction; the third direction about which we have the four-fold symmetry need
not be equal to a and b.
So, the presence of the four-fold symmetry therefore, introduces the concept that a
should be equal to b. So, whenever you have a square lattice, we have already seen that
whenever we have a square lattice the square lattice will effectively introduce a to be
equal to b and it is not equal to c in this direction and alpha, beta, gamma are 90 degrees
with respect to each other.
97
So, this can also be treated as a stretched cube and this essentially you take a cube and
stretched it in a direction perpendicular to the break base plane, then you will get a
stretched cube. This now define set a triangle system and in the case of a cubic system,
now what we do is we compress the system and we get to a, b, c, alpha, beta, gamma a in
equal to b equal to c, a equals b equals c alpha beta gamma are 90 degrees are very
symmetric isometric system. In fact, these are also referred to as isometric crystal
systems.
So, in the case of a tetragonal symmetry, what are the point groups we get? We get the
point group definitely 4. Now, other than the point group symmetry 4, what other point
groups we can get? So, we will now study the point groups in full detail as we go further
in the next few classes; so, that you are now getting ready to do all the assignments
which will come in your way, you can anticipate the assignments. The assignments will
be that given a tetragonal system what are all the point groups which you can associate
the tetragonal system with. For example, four-fold symmetry, a mirror perpendicular to
the four-fold symmetry which gives you 4/ m and then the operation 4/ m 2/m 2/m and
also the operation 4 2 2 which is aloud and things like that.
So, there is a large number of find group symmetries that will get associated with the
tetragonal system. Remember that we have only a total of 32. So, we have to take the 32
point groups symmetries distributed among the individual crystal systems. So, what you
have done is we have given it 2 to triclinic, 3 to monoclinic, 3 to orthorhombic and
several of them to tetragonal and so on. When we come to the hexagonal system to
extend this discussion instead of the c axis having a four-fold symmetry, the c axis can
be associated with a six-fold symmetry.
Since this is a six-fold rotation axis, we should have a 120 degree angle and therefore,
gamma now is 120 alpha and beta have to be 90 degrees in order to have the six-fold
rotation axis coming perpendicular to that. And therefore, we have a condition a equals b
not equal to c alpha beta 90 degrees and gamma 120 degrees. So, this defines now a
hexagonal system.
A special case of a hexagonal system gives rise to the. So, called rhombohedral system
where we essentially have a hexagonal system, but then this hexagonal system is very
special it becomes in such a way that a is equal to b equal c alpha beta gamma are equal
98
to each other. The difference between the rhombohedral system and the cubic system is
the fact that this angle is not equal to 90 degrees.
The moment it becomes all 3 equal to 90 degrees, we will have a cubic system and
therefore, effectively we have 7 crystal systems and these 7 crystal systems are
associated with the type of rotation axis which we can give to them. The type of rotation
axis we can give here is a threefold rotation axis, but that is a very special direction in the
case of a rhombohedral symmetry.
Same is true with the cubic symmetry we give the three-fold rotation axis along the
diagonal and then only we can define a cube. So, some of these features are right now up
store and the little bit a confusing as we keep discussing. But therefore, what we will do
now is to go away from this three dimensional thinking process because we already said
that 3 dimensional thinking has not been with us since we went to school. Because we
made the basic mistake of joining school and losing the 3 dimensional thinking and went
down to 2 dimensions. We will now examine the possibility of 2 dimensional symmetry
and that brings us to a very interesting discussion on what kind of plane lattices we can
think off in 2 dimensions.
So, these are therefore, the yeah 7 crystal systems the triclinic: the monoclinic, the
orthorhombic and so on tetragonal, cubic, hexagonal. I am repeating it just to drive home
the point of the types of point groups symmetry is which you can distribute to these 7
boxes.
So, we have 7 crystal systems and 32 point groups and nothing else. The nothing else
issue comes up mainly because we have translational periodicity. If translational
periodicity was not there, then things would have been very different because they
translational periodicity as put restrictions on the types of rotation axis we can we have
1, 2, 3 4 and 6. And because of that and also because of the fact that we have periodicity
in a, b and c directions in these boxes, we have only 7 types of boxes. And remember
that the 3 dimensional space which is generated by these any of these boxes will have to
have that particular symmetry.
In other words if there is a triclinic symmetry, triclinic crystal and it has only one-fold
rotation and all the points in that particular unit cell will have one-fold operation; any
point x y z will undergo one-fold rotation. On the other hand if there is a associated
99
centre of symmetry or inversion centre with respect to the triclinic system, then they
associated symmetry system will also have to be present in every point in that 3
dimensional box.
Suppose we have a tetragonal system for example, every point in this unit cell and also
the entire lattice which we can generate by translational periodicity operation will have
the four-fold symmetry; it must have the four-fold symmetry. So, the presence of the
four-fold symmetry ensures it is a tetragonal system, the presence of the six-fold
symmetry ensures, it is a hexagonal system and so on. So, these definitions therefore,
now make a solid material study become more and more systematic. So, if materials
crystallizing these systems it does not matter if a molecule we saw the molecular
symmetries in the previous class.
So, if the molecular symmetry is five-fold symmetry it does not matter as long as it can
be fitted into any one of these 7 boxes with the symmetry obeying the obeying the
symmetry rules which are put on the box and on the system. It is not necessary that the
molecules should have the same symmetry that is associated with the crystal system. So,
we the molecular symmetry and the crystal symmetry can be very different. However,
Karagodsky had demonstrated that if the molecular symmetry has a presence of a
inversion centre for example, that inversion centre will be utilized in the unit cell
associated with that particular compound.
So, suppose there is a compound with an organic compound or a inorganic compound

which crystallizes with the presence of a inversion centre at its center, that center will
now coincide with the center of inversion associated with the a crystal system.
So, the crystal system therefore, invokes the possibility of occupation of these molecules
at the special positions. So, the special positions will occur in other words on situations
where we have the presence of the molecular symmetry. The the occupancy of the
special positions are invariably done only when the molecule has the similar symmetry.
And this is a point which we will have to keep in mind when we study crystal structure.
So, if you have a knowledge of what molecule it is and if you have a knowledge that that
particular molecule has us 1 bar symmetry then in if it goes into let us say a triclinic
system, then we know it has to be P ̅1. Just to extend these argument; if in case we have
only an l amino acid because l amino acids form a protein.
100
So, if you have a protein which crystallizes in a triclinic system , it cannot crystallize in
P ̅1 it has to crystallize in a P 1 system. So, in another words, the protein crystals with
naturally occurring amino acids being present in a protein or for that matter peptide units
which have all l units or d units which our way if we have synthesized. These kind of
molecules we will not go into systems which have a centre of inversion. They have to go
to non center symmetric space groups.
So, when we discuss about how these systems go in to different kinds of groups, space
groups and eventually crystal systems, we will bring in the restrictions which come as a
consequence of this, that we will do later.
So, what we thought was we will now look at as we mentioned again that we our
understanding in two-fold objects is much easier the 2 dimensional objects is much
easier. And therefore, we will look at 2 dimensional plane lattices. It is, we can
demonstrate that there are five 2 dimensional lattices or 5 plane lattices.
The first lattice by definition of it you know; we have already seen the definition of a
triclinic system is made from a the value of a here is not equal to the value of b. So, if
you now call this as a and this as b and this is the angle gamma, then we see that a is not
equal to b and gamma is not 90 degrees. So, this defines the so called oblique p-lattice.
What is very interesting is this oblique p-lattice has the lattice points at the unit cell
edges.
101
The lattice points at the unit cell edges will now have because it is a plane lattice, there is
no third dimension. So, because there is no third dimension, the symmetry that is
associated with this point will be a twofold. So, the minimum symmetry associated with
an oblique p-lattice is a two-fold symmetry and this two-fold symmetry has the will
repeat itself at all these four corners wherever there are lattice points. In addition
because, there it has a two-fold symmetry, the presence of other two-fold symmetries
halfway between a direction, b direction and also along the diagonal will automatically
appear.
And that is because of the fact that we have x y z in this particular case x y 2 dimensions
x and y we representing a and b. So, if there is a point x y here; if there is a point x y
here, this particular point we will see the two-fold axis up there. So, it will undergo a
two-fold rotation and go to that point and this now when translated by 1 unit, we will
come down here because of the lattice periodicity we can translate it by 1 unit.
And this now, again translates into 1 unit and comes over here; that means, the object
which was here, we will also generate an equivalent point there. And because these 2
equivalent points are related by these two-fold axis the presence of the central two-fold
axis becomes invoked and therefore, the presence of two-fold axis at these points
symmetry at these points represents of the two-fold symmetry introduces two-fold
symmetry is at half along a, half along b, half along a and b. And therefore, we have their
the presence of additional two-fold symmetries.
This happens not only in 2 dimensions, it will also happen in the 3-dimensional unit
cells. Whenever you associate a symmetry with a lattice point that symmetry the same
symmetry more often there, not unless otherwise specified in. When we go to higher
symmetries, we will have more issues coming up. But in most of the cases whatever
symmetry is put at these origin 1 unit translation; in this direction 1 unit translation in
that direction, 1 unit in this direction 1 unit in that direction.
Automatically generates the same symmetry same symmetry at half way points and
therefore, we get the two-fold symmetries at these half points. This is a criterion which
we have to considerably worry about because when we have objects which have
symmetry associated with an oblique p lattice, it will invariably process this feature. So,
102
an atom which is up here will have to appear at this point and these 2 now define the 2
equivalent points associated with the oblique p-lattice.
Very often than not we think that oblique p-lattice has a third direction, but we are
considering only the plane lattice on the third direction is undefined. Such symmetries
we will occur in what we call as liquid crystalline objects. In liquid crystalline objects,
they took the plane symmetry only remains the lattices are forming two-fold symmetries,
but the third symmetry is undefined or it will have a certain property which is not a
periodic in nature. So, the periodicity in the third direction may not be present. In such
situations it is good to know to represent the 2 dimensional periodicity in this fashion and
that is how the 5 plane lattices are very crucial to understand fully.
So, having seen this that the fact that the oblique p-lattice now will generate 2 equivalent
points and then all these symmetry positions, these are the symmetry axis coming up at
we various positions. This is therefore, the property of the 2 dimensional oblique lattice.
Now, what happens in case we have a line of symmetry? Because we do not now have to
have a situation where we have an axis of symmetry where we have to have a plane of
symmetry because if we have to have a plane of symmetry that plane of symmetry will
not define a plane lattice.
So, whatever we consider as a plane now can be considered as a line lattice. So, we have
therefore, the line axis; so, the line of symmetry. So, this is a periodic repetition in one
dimensions. This is a one dimensions lattice which has a line symmetry. So, this line
symmetry now is represented by these points which are now repeated at let us say a. So,
this is a one-dimensional lattice now the one-dimensional lattice may have a point which
is associated with the line of symmetry or it can also have a point which is bit way
between the lines of symmetry.
So, there are 2 possibilities, only in this particular case, you have the line of symmetry
the point which we are associating could be on the line of symmetry or half way to the
lines of symmetry. And because of the fact that these we have either on this or that it
opens up 2 possibilities; one possibility a is a 2 dimensional lattice which we can
generate as indicated in this figure c, which now tells you that this is the 2 dimensional
lattice and that is a 2 dimensional primitive lattice which is now a rectangle as you can
see that this represents a rectangle because this distance is different from that distance.
103
So, we have a situation where we get this primitive lattice and that is indicated this way.
So, if you have this value as a and this value as b this is the gamma value. So, what
happens therefore, to gamma? Gamma becomes 90 degrees, gamma becomes 90 degrees
and therefore, we have a not equal to b and gamma is 90 degrees. This now represents a
primitive rectangular lattice, rectangular p-lattice. Notice that when we want to say
primitive; that means, the total number of that is very important we have not defined yet.
I think it is time to define what is meant by a primitive lattice.
A primitive lattice is one in which when we take all the lattice points in the unit cell like
for example, one translation, one translation, one translation; it again repeats itself in all
3 directions. So, you take the oblique p-lattice. In the oblique p-lattice, we have points
which will be coming at these 4 positions. If you take the next 4, it will come there, the
next 4, we will come there, the next 4, we will come there and so on.
And therefore, every point therefore, a shared by 4 unit cells in the plane, every point is
shared by 4 unit cells, one unit cell here, one unit cell there, one unit cell there, one unit
cell there with respect to this fellow and therefore, the occupancy of this will be one-
fourth. And therefore, if you add all the lattice points in the unit cell you get one-fourth
plus one-fourth plus one-fourth plus one-fourth; you will get a total of 1.
So, in any primitive lattice, the number of lattice points will be 1 ok. This is a point
which we must notice and the so in this particular case in the case of the rectangular
primitive lattice, again the number of lattice points turns out to be 1. You can see it here,
this is the unit cell; you take this as the unit cell. This point is shared by this unit cell,
that unit cell, this unit cell and that unit cell and therefore, you get the lattice point value
here is one-fourth and therefore, this particular unit cell will have a total of 1 and this
repeats itself in all 2 directions.
Generating the so called 2D lattice in which the property of the line of symmetry
remains; that means, we have this line of symmetry that line of symmetry and at this is
along the a direction. Along the b direction will have a line of symmetry there and a line
of symmetry there. And just the way in which we define the earlier case the two-fold axis
of course, has to be there in this particular case.
So, the two-fold axis is already located at these positions and the line of symmetry will
also repeat has we mentioned that if we have a symmetry element at the point 0 0 0 and 1
104
translation, there is of the same symmetry element. The corresponding symmetry
element will repeat halfway distance. So, therefore, you get this kind of a grid which
now represents the rectangular p-lattice.
So, any point in this particular position we will see not only the two-fold axis, it will also
see the line of symmetry about which it can move around. And this a equals b and the
angle gamma therefore, will represent this so called rectangular p-lattice. So, this is one
way of representing the rectangular p-lattice. This is another way of representing the
rectangular p-lattice where we have the points in the middle, but we still have there are
rectangular p-lattice and therefore, we have the symmetry that is identified with respect
to the primitive lattice. I think we will stop here for today’s class.
105
Lecture - 09
Bravais Lattices
So, we have been discussing the 5 plane lattices in the last class.
And we discussed the oblique p lattice, we also discuss the; we discuss the oblique p
lattice, we also discuss the rectangular p lattice. And we found that there are two ways in
which we can represent this rectangular p lattice. One is to show the points along the line
of symmetry and show the points half way between the lines of symmetry and both these
will now give rise to the rectangular p lattice. So, if you consider this point, these four
points now we will define our unit cell, these four points will define our unit cell and this
is a primitive lattice. And the two fold rotation axis will appear here, here, here, here
because any plane lattice will have a twofold symmetry.
And therefore, the plane lattices will always have a twofold symmetry at the origin and
that one unit translation. So, that is already represented along with that we have the lines
of symmetry. So, the lines of symmetry will repeat at these positions and as we
discussed, whatever is the symmetry at the origin and one unit translation will
automatically appear at halfway point because of the heat translation periodicity. So, the
106
similar thing happens here. Even though these points are in the middle of these lines of
symmetry, the unit cell is still having one point which is which is now again a primitive
rectangular lattice.
So, we can also have another type of a lattice where we have the overlap of the lines of
symmetry the way in which these are repeated. So, we take this unit, the one dimensional
unit, overlap this one dimensional unit with this one dimensional unit and that is
represented here. So, we have this line of symmetry, this is the repeat distance a and this
is the repeat distance b; here the points are with respect to this dimension. So, we have
this point, this point, this point and this point and between them we inter leave this one
now. Because of the fact that we inter leave this fellow, this one has come in between
and that also has a line of symmetry by definition which we have given here.
And therefore now, we see that there are lattice points are these four points and also a
lattice point in middle and this now defines a different kind of a rectangular lattice which
we call as the rectangular c lattice. So, the rectangular lattices there are two types;
primitive as well as the central lattice and this is represented by this picture here. So,
what happens is because of the fact that we have a centred position here, the property of
this repeats at the centre and therefore, the property of whatever is happening here will
also repeat at the centre.
And so, we get additional two fold symmetries now placed at one fourth between one
and half the additional symmetries are placed at one fourth, but there displaced by
another one fourth. So, at one fourth, one fourth, we get a twofold axis and therefore, this
now defines a new type of a plane lattice which is a rectangular c lattice. On the other
hand, if we just have these alternate ones, but no occupancy of the middle, this is similar
to the one above here and therefore, we will get a rectangular p lattice.
So, f represent a rectangular p lattice corresponding to b; on the other hand, the e

represents a rectangular c lattice where we have a centre. I think I should describe this
again for the convenience of people who have a difficulty in understanding this. We have
in a rectangular system, a equals b a not equal to b, but gamma is 90 degrees. The fact
that gamma is 90 degrees will ensure that these two fold symmetries which came here
without the presence of gamma being 90 will always represent at these positions. So, the
moment we have a plane lattice, the two fold symmetry is will appear at the origin one
107
translation along a, one translation along b and they will also appear along the half
distances.
Apart from that, the fact that a is equal to b and gamma is 90 degrees introduces a line of
symmetry and this line of symmetry can be now represented in this one dimensional
lattice by these points. The same one dimensional lattice now can move the lines of
symmetries halfway and the points of representation could be halfway between the lines
of symmetry. Both of them represent the one dimensional lattice and the corresponding 2
d lattices are shown in c and d.
Both c and d now represent a rectangular lattice, but this now is a primitive lattice
because the total number of points add up to 1. So, the number of lattice points is 1 and
therefore, we have a p lattice. We can always interview inter leave this a and b together
and therefore, we can have alternate layers of this line of symmetry and that line of
symmetry existing parallel to each other as I will just shown in the diagram e.
Here we show that the a type is existing in these alternate lines and the other alternate
lines are represented by this b type. And then this happens you see that this defines the
unit cell with a equals b and gamma is 90 degrees which is a rectangular representation;
however, we get a point at the centre. In fact, if you look at the most primitive cell that
can be generated in this unit cell, this could be here. I will show by marking with the pen
what I mean by this.
So, if you see this, this, this and this: this refines a rhombus and in this case a dash and b
dash are the two distances and gamma dash is the angle and this represents a primitive
rhombus. The problem with the definition of a primitive rhombus a is that we do not
have a value for gamma prime. Gamma prime is not any value. It has to take this value
corresponding to the rhombus and what is the value associated with the rhombus? The
value associated with the rhombus gamma can be calculated as a prime plus b prime
divided by 2, etc, etc. And therefore, this is not a uniquely defined gamma prime.
So, even though there is a possibility of a rhombus being defined here, the most
convenient way of representing because all the symmetry elements associated with this
will represent the presence of the rectangular lattice. So, the rectangular c lattice if
centred will define now this unit cell. This will now involve the possibility of including
this particular centre as well. So, we have therefore, here a and this is a this is b and this
108
will now represent our unit cell with an atom with a lattice point of this centre. So, the
total number of lattice point 1, 2, 3, 4 will define one and this will add additional lattice
point. And therefore, the number of lattice points is two and therefore this becomes
rectangular c lattice.
So, the occurrence of the rectangular c lattice is because of the presence of an additional
lattice points. Since we have two lattice points here this can be represented as a centred
rectangular lattice and so, we have two types of plane lattices in a rectangular system.
So, what is the implication of this? The implication of this is that effectively there are
only four plane lattices, but then among the four plane lattices which you can call which
we can could have called as two dimensional crystal systems.
There is an additional lattice that is possible which is the so called c centred rectangular
lattice. So, it now makes it possible to have therefore, 5 lattices. The fourth one is a
equals b, here it is a and this is b and here a equals b because it is a square lattice in
square the gamma is the angle here which is 90 degrees. And therefore, we get a
definition of a square lattice which is something like this.
Now, what is very interesting is to see how these fellows all developing. How the
symmetry elements develop in a situation like this? If you look at this particular point
this represents now because it is a square lattice it represents a fourfold symmetry. So, as
a presents of the fourfold symmetry of here is repeated at unit cell lines. Notice, that in
the case of the oblique lattice and also in the case of the rectangular lattice the same
symmetry element repeated.
Here, the fourfold symmetry now is not repeated because of the fact that these two now
which are diagonals they intersect at the centre and there we get the definition of the
presence of a fourfold symmetry which implies that we get additional two fold
symmetries at these half points. Now, why is that it is because of the fact that if we take
now assume that this is also fourfold symmetry, then this fourfold this fourfold together
will not generate the central fourfold.
If we want to generate the central fourfold, these have to be only two fold. Think about
it, if we have to generate the central fourfold and also they are the fourfold symmetries at
this are ledgers, this has to be two fold it cannot be any other value. If this is also
fourfold, then we will not be able to repeat the four fold here. The redundancy comes
109
from the fact that these diagonals will now represent the central point and these
diagonals now automatically carry the two fold information and since these diagonals
now carry the two fold information.
This central now becomes the four fold symmetry, and as a consequence these directions
will have now at the midpoint two fold symmetries. The same argument goes for the
hexagonal p lattice where a is equal to b and gamma is 120 degrees and you see that the
sic fold symmetry goes up in the edges and three fold symmetry is develop in these
positions where we have shown and the central between the six and the six is always a
twofold.
So, this now of course, will eventually as you can anticipate will come as an assignment
and you will have to work it out why this now has additional symmetry elements are
different points like 1/3 and 2/3. Why the 1/3 and the 2/3 develops could be in the form
of an assignment and this therefore, tells us that there are 5 plane lattices what we have
learnt in this particular set of discussion is that there are 5 plane lattices, the oblique p
lattice the rectangular p lattice, c lattice. So, we have an additional lattice with a centring
which is associated with the rectangular lattice. We also have this square lattice and the
hexagonal lattice.
Remember, the rotation axis that can be associated with this are the two fold axis of here,
two fold axis of their, four fold axis here and the six fold axis there. You also see that the
presence of the three fold axis is automatically invoked in the hexagonal p lattice. So,
there is no trigonal p lattice, the trigonal p lattice plane because it is already there.
Suppose, you consider this as the origin then you can define three fold symmetry
primitive lattice associated with the plane lattice. So, it is redundant. So, the six fold is
the one which now dominates in this particular occupation.
Therefore, there are only 5 plane lattices. So, even though there are we talked about
seven crystal systems, here there are only five, four crystal systems and 5 lattices. So, in
among the plane lattices, there are four crystal systems and in two dimensions and 5
lattices. Now, we will see now what happens to the seven plane lattices, seven crystal
systems and how many lattices we can generate with the same logic.
110
Before we go further we will examine in detail the definitions of the 5 plane lattices. This
is something which is very important and crucial you see that indicates of an oblique
lattice, the lattice type is p, the point group associated with the lattices 2. This is the
maximum point group symmetry that can be associated with the lattice. But, when you
come to the case of a crystal it depends upon what kind of an object is sitting inside this
oblique axis. If there is a material which is now containing an object let say molecules
that molecule has no two fold symmetry, it can still go into an oblique p lattice with a
value of 1.
So, the possible crystal point groups are 1 and 2, both 1 and 2 are possible in an oblique
p situation where a is not equal to b and gamma is a general value other then 90 degrees.
Having understood that we have the remaining p and c, we point group symmetry the
maximum symmetry is 2 mm and then depending upon the type of object we are going to
put in the crystal it could be 1m or 2mm. So, in logic appears to the square symmetry as
well as the hexagonal symmetry, you see that in the case of the square lattice we get a 4
and a 4mm, 4 mm is a possibility because we remember you remember that we have the
Elures theorem and one once you have the proper, improper, improper that combination
is allowed and therefore, we will have 4 mm.
So, in general the point group symmetry of this lattice a is a proper, improper, improper
axis combination in the value is a equal b gamma 90 degrees and when the value of
111
gamma becomes 120 degrees you have the hexagonal symmetry. The hexagonal
symmetry as you see can also have the threefold symmetry as we saw from the previous
diagram of here it is also got the presents of the threefold symmetry; one third removed
from the six fold symmetry, two third removed from the six fold symmetry, we have the
presents of the three fold symmetries and this surroundings of this suppose, we have we
call this as 0 0 0 at this particular origin there is a six fold symmetry, then the three fold
symmetry is removed by one third. We can always move the origin to the three fold
symmetry position.
Then, we will invoke a six fold symmetry because of the presence of the two fold at the
centre of these unit cell and therefore, we have these both the possibilities, it is coming
up here the presence of six fold as well as the three fold. So, we can therefore, have
objects which belong to the hexagonal symmetry, we can have objects belong into the
trigonal symmetry, both going into a hexagonal small p. The small p and the small p the
tell us that these belong to now two dimensional lattices; p represents a total lattice point
of 1; c represents a total lattice point of 2. So, the number of lattice points in rectangular
c is 2. The point symmetry is different from the number of lattice points in the in the
lattice. So, the lattice now decides how many combinations of symmetries it can have
because of the presence of either p or a c.
In addition, we have his presence of symmetry elements and therefore, the objects now
which find themselves inside these unit cells will have to look at both the point group
symmetry as well as the lattice symmetry. So, the presents of the lattices symmetry in
conjunction with a point groups symmetry and of course, the nature of the material
which goes into this lattice; all these combinations will decide what kind of a symmetry
one has to take in this particular case.
112
Extending the same logic of finding these symmetry elements and we saw that there are
5 plane lattices. Since there are seven crystal systems, the seven crystal systems now will
generate 14 lattices just like the 4 crystal systems in plane lattices generated 5 lattices,
here the 7 crystal systems will generate 14 of them and these are referred to as the
Bravais lattice this after the name of the person who discovered this. And this there for
now represents the three third dimensions, the three dimensions and then we see here the
cubic system in the case of the cubic system we therefore, have these points these are the
lattice points please note that.
These are lattice points are not atoms is always a confusion. So, we should remember
that we have not brought in any realistic atom or molecules in the in our discussion so
far. So, the points which are appearing here are the lattice points. So, the lattice points
therefore, now represent the presence of a primitive lattice because here now which is
three dimensions. So, each point will now represent one eight. So, the total number of
this will be 1, 2, 3, 4, 5, 6, 7, 8 multiplied by one eight. So, that will make it 1. So, the
primitive lattice always has one lattice point we can have the centre of the cube centre of
the cube located in such a way that we will have what is known as a I-centred the body
centred lattice.
So, the cube is appear which represents one lattice point, this 1 represents the second
lattice point. So, the number of lattice points saying a body centred cubic system is 2. If
113
we can have also the each and every face centred, we will have four possible positions in
which we can have the occupancies defined and therefore, we have the lattice points at
the corners during one lattice point, in addition the faces give half lattice point. So, since
there are three faces; half, half, half will add up to 3. So, 3 plus 1 therefore, there are four
lattice points associated with the f lattice. Now, other possibilities exist in case of a cubic
system which is the highest symmetry system.
In fact, I should have shown you the other way around, this 14 Bravais lattices; I
purposely so the cubic one to show the complexity of the situation. Now, we will look at
it in a simple way. The simple ways is that we have a primitive lattice which is
represented by P , we can have the C side centres lattice which is the case with a
monoclinic system. Because, the monoclinic system here now can have since the angle is
90 degrees, the alpha and gamma are 90 degrees because of that fact we will have this as
a possible C centred primitive lattice.
So, since it is a central primitive lattice, it could be alone say C or A or B. But in this
particular example, we have a C centred lattice; that means, there are two lattice points
associated with the monoclinic symmetry. So, triclinic symmetry always has only one
lattice point, monoclinic symmetry depending upon the case whether it is primitive or
centred will have one and two; hexagonal system always has only one lattice point, a
trigonal system always has only one lattice point.
114
The most common one is the orthorhombic system where we have four possibilities; one
is the lattice centring of course, in this particular case a, b and c are not equal to each
other. So, when we have a C centred lattice, we will not have the A and B centred and
therefore, we can have a single face centred lattice which is represented as C. We can
have also face centred as well which is referred to as F. We have the body centred lattice
which is I in the orthorhombic primitive is also possible. So, if we count the total number
of all these which now we will account to 14 lattices. So, we have seven crystal systems
classified into 14 Bravais lattice.
I how got a another illustration which is much more simple and straightforward and this
is drawn in projection. So, I have drawn the 14 Bravais lattices in projection and this
draw drawing will clearly tell us how we can arrive at the 14 Bravais lattices in to you
know we always want two dimensional representations. So, here you see that in case of
the triclinic system, we have only a primitive lattice. In case of the monoclinic system,
we have the primitive lattice. Now, why did I show like this because I am now showing
the angle beta this is a c diagram. In fact, this is to show that the c centring can be done
in this particular projection.
So, this is P, the primitive lattice. This is the in this case it is a here and c there and this
angle is beta in this case, this is c here and the a there and that angle is beta. The c
direction is centred with respect to the lattice centring. Therefore, we have two plane
115
lattices associated with monoclinic P and C and therefore, we have P and C lattices
coming in three dimensions as well. So, triclinic as one one Bravais lattice, monoclinic
as two Bravais lattices, orthorhombic will have four Bravais lattices represented in two
dimensions here, tetragonal has two, trigonal has three, tetragonal has three sorry,
orthorhombic the tetragonal has two cubic has three, hexagonal has one.
And this is a very special case as I mentioned when a equals b equals c, alpha beta
gamma are same value then we have what is known a trigonal system. This is in, this
diagram is in conjunction with the diagram we saw just a few minutes ago in the respect
to this one, you see that there is a one third and the two third, three fold rotation and that
is what is depicted here in three dimensions as well. So, there is a one third and a two
third, three fold rotation that is present in a rhombohedral system.
So, there is a threefold axis which goes along the diagonal. So, if you take a
rhombohedral system and put it along the diagonal and rotate that one, in such a way that
we can rotate it in such a way that the angle becomes 90 degrees we will have a cubic
system a equals b equals c, alpha beta gamma are 90 degrees. So, this therefore, now
defines the 14 Bravais lattices. So, we have 32 point groups, seven crystal systems and
14 Bravais lattices. So obviously, the next step is very clear.
We have to take these symmetries into account we have to take the translational
periodicity into account, we have to take the crystal system into account, we have to take
the 14 Bravais lattices into account and then distribute the 32 point groups among all
these crystal systems. When we do that, the prime groups now will alien with respect to
these 14 Bravais lattice and generate what is known as the 3 dimensional space groups.
That is because now in space, these symmetries get redistributed how in space the
symmetry is get redistributed we will see as we go along.
116
And now, before we go further, you remember that we decided to see everything in two
dimensions, but it is not always possible to see everything in two dimensions. Because
we saw the meiosis diagram, we understood the concept of periodicity from meiosis
diagram and we have those objects repeating themselves generating the periodicity,
generating the symmetry elements and so on. And, then we have to look into the aspect
that how one can really look into three dimensions and this is in real situations. In real
situations, we will have a crystal the crystal have different kinds of faces. So, these
different kinds of faces in the crystal have to now be inscribed into a highly symmetric
spherical system. So, sphere is the most symmetric object.
So, we take this sphere which is the most symmetric object. Take the crystal and insert in
in it inside a sphere in other wards you describe a sphere passing through these points
which I will stated here and you will get this sphere. Now, we take the centre of the
crystal and draw directions have to various planes about which the crystallize has this is
a very beautiful ground crystal, not always the crystals are so good but in this example,
you see that we are drawn from the centre here perpendicular radial direction.
So, the radial symmetry of the crystal you gets have represented here. Why do we
represent the radial symmetry because the radial symmetry is the one which is most
common with respect to spherical objects. For example, if you put a centre here all the
117
symmetry elements now associated with this particular sphere can be along the radial
directions.
So, if the crystal has a certain symmetry, the external morphology has a certain
symmetry that symmetry can be fitted inside the stereographic projection. I will just
diagnostic little bit at this particular point because we have to see why stereographic
projection is very crucial in looking at crystals. And this is in fact, the way in which
crystallographer practiced and identified all the symmetry elements associated with the
crystals. Nobody did x-ray diffraction in those days because extra diffraction was
discovered much much later.
All these concepts of point group symmetries and the space groups and so on were kept
and derived way back in the 18 century the in 18 century 1735 to be exact. In that
particular year, it was already shown that there are 32 point groups and because of the
fact that we have now crystalline objects which will not displace these point group
symmetries in the lattices, the combination of the lattices and the point group symmetries
will give rise to 230 space groups.
So, all these were worked out much much before. So, we have to find out how they
would our did out, not that we are going to do this way, we are going to actually use x-
ray diffraction techniques which is a experimentally reliable and develops technique in
recent years. In fact, x ray is well discovered only in 1900 by Rontgen. So, only after that
time, but before that all the space groups and all the presence of point groups space
groups etcetera you are already established.
So, how was it done? It was done by looking at crystals and people got all the very large
crystals or size crystals so that they can hold it in their hand and then examine the
perpendicular faces the perpendiculars to the faces, make an impression of that on to
sphere and on that sphere, we can mark the intersection point. So, if this is the point find
which is they appear we mark the intersection point by a open circle.
Now, I will define what is this open circle and close circle in a minute. Before that, as I
mentioned we will digress a little bit. We have a similar situation in case of earth we
have a three dimensional object; earth is a three dimensional sphere and when you want
to study the geography associated with the earth, we do a different kind of a projection
that is not a stereographic projection. We do what is known as the Mercator projection.
118
In that projection, what we do if we take they were earth the surface of the earth hold on
to it is diameter and then press the earth. It is like taking an orange and press it of course,
though did because it will push and spread and your dress, but essentially you take this
orange diameter and press it like this and when you press it like this, it flatulence out and
flat and out is your map.
So, then you take out the reasons of what you would like to inspect in this flat and out
map. If this is on this side, we will exam in those ones we both northern hemisphere and
southern hemisphere units will be represented and also the northern hemisphere or
southern hemisphere units will be represented on the other side. Essentially, we have to
then open it and spread it, this is the diagram we will see in all atlases. So, you press it,
open it and spread it around and therefore, you see the Americas on the left side and the
rest of the world on this side, Americas are always separate. So, that is what we will see
and now we want to determine in the distance between Mumbai and Bangalore as we
discussed earlier in with respect to this diagram.
So, this diagram is telling as is it clear this diagram. So, this diagram is telling as that the
story of Atlas. This is a Greek mythology the Atlas is carrying his the earth and you
shoulder and these two angels want to measure just like as the distance between
Bangalore and Mumbai. So, there holding this divider and then this angel, the third one
is helping them out why shining light on this so that we get a projected image of the
surface of the earth and this is on the ground and therefore, it is two dimensional and
they now measure this distance and have a scale factor which will work out to the
corresponding distance between Bangalore and Mumbai.
Now, there is a flow in this particular type of projection because it depends upon on what
point on the surface of the earth we have done this projection. So, that way this
projection is not a very convenient way of looking at actual systems. In other words,
suppose instead of taking this equator and pressing it suppose I take let us say in fact,
time when I visited New Zealand, this is what I we saw you take New Zealand as this
part and compress the rest of the earth and top of New Zealand.
New Zealand as you all know is the size of our state Karnataka but when you do this
kind of an operations New Zealand will appear very very huge. With respect to New
Zealand, India will appear very very small it may be a little bit of a short length. And of
119
course, U K will appear as a point that is not true, but this is what happens in this kind of
projections. So, we need to find out a proper way of generating the projection diagram
and that is where the stereographic projection comes in handy. So, what we therefore do
is, we draw all these radial lines with respect to the centre passing through the plane of
various plane so that the crystal is made up of various plane. This is let us say diamond,
many of you where diamond many of the students wear diamonds and the diamonds has
got very nice lustrous planes.
So, from the centre you draw this normals, the normal to the plane is the one which is
now collected on the surface of the sphere. So, we are not now compressing or pressing
this sphere which is remaining as such and then we do this projection. So, when we do
this particular projection, what happens is that we will have points with respect to the
northern part of this crystal, we also have points corresponding to the southern part of the
crystal.
So, if now we see the northern part of crystal as we shown here, if there is a point P on
northern hemisphere, we take the south pole and connect it to the north to the point p and
represent this little p in the equatorial plane as a point as a close the circle a little close
circle as you see here. This is the point p which is now the representation of the
projection of this particular point on to the south pole. A similar construction is done
with respect to the north pole.
And I think we will stop here.
120
Prof. T.N. Guru Row
Lecture – 10
Details of Stereographic Projections
So, in the last class we stopped that what we did with respect to the points in the northern
hemisphere.
So, you consider these as the points in the northern hemisphere. These are the points
which are drawn here in the northern hemisphere on the sphere and we take the South
Pole and connect it to the point which intersects the sphere on the northern hemisphere,
this point P let us say, we correct we take it here and then we take the equatorial plane.
So, it is again an orange. Now, what we do is we take the central part of the orange cut it
open halfway through and take this slice out, that will define the equatorial plane. So,
this slice is the representation of the equatorial plane.
So, when we do this South Pole to the North Pole points, all the points in the North Pole
are joined from the South Pole and their intersection with the equatorial plane is
indicated by the point P. So, the little point P here which is a closed circle now represents
all these intersections. The intersections of the South Pole region, the southern
121
hemisphere region with respect to the North Pole, now are joined, likewise and that point
is indicated by an open circle.
So, when we have a crystal with this kind of symmetry and do this operation then we
will now slice open this equatorial plane and take it out, take it out and spread it on the
plane. When we spread it out on the plane we will get diagrams of this kind. So, this is
now the equatorial plane and depending upon the nature of the crystal and the symmetry
that is associated with the crystal, we will get points associated with the intersection with
the North Pole which is indicated by x and the intersection from the South Pole indicated
by open 0 or vice versa, it does not matter. As long as we identify the points with respect
to the North Pole and with respect to the South Pole we keep that identity the same.
So, this equatorial plane now represents two possible symmetries you see we can read
the symmetry directly because this now represents the 360 projection. So, we have this
sphere which is a 360 sphere and we therefore, now represent the 360 degree
representation and we see there is only one point. If there is only one radial point which
is emanating into the crystal from that point we say that the point group symmetry is 1.
So, this is the projection diagram and these are called stereographic projection diagrams
of various point groups. For every point group therefore, we can draw these kind of
diagrams by taking examples of crystals which belong to different crystal systems and
therefore, different point groups.
So, when such studies are made we will see that in case of the ̅1 system we will have an
intersection with respect to the North Pole and the South Pole, the open circle and the x
point will now indicate this. It not only indicates this it indicates the relationship between
this point and this point. If this point is xyz, this point is xyz, let me tell you that clearly.
If this point is taken as xyz, now this point now is on the opposite side, this is northern
hemisphere, this is on the southern hemisphere of the sphere. So, therefore, you get -x -y
-z. So, when we get a distribution like this we identify the position of the presence of a
center of symmetry or an inversion center. So, this therefore, belongs to ̅1.
So, we have therefore, the stereographic representation of 1 and ̅1 as we see here. So, the
obvious extension is now to take the crystal systems associate them with respect to the
stereographic projections of symmetry elements and associate therefore, the point group
symmetries to each and every crystal system.
122
As we saw in the discussion before, this can be done in this particular table. This is have
the final result which I will show, but we will discuss each and every point separately
with respect to the points which we are going to do, with respect to the diagrams which
we are going to do.
So, what we will do is we will do this stereographic projection on the crystalline systems
and then from the crystalline systems, we can take individually the presence of the
symmetry elements with respect to the projections we got from the North Pole, the
projections we got from the South Pole. Conglomerate these results and see what kind of
a symmetry this distribution of open circle and the x points indicate and based on that we
can decide on the symmetry that is associated with the crystal system.
So, we have now brought in the crystal. The crystal is now 3-dimensional. We have
drawn radial lines with respect to the crystals, so that they can be represented as a
stereographic projection inscribed inside a sphere. So, effectively the crystal is taken and
put inside this sphere and we have taken the intersection points in the northern
hemisphere of points are joined to the south a South Pole and those are indicated by open
circle and the ones which joined the North Pole from the southern hemisphere are
indicated by x points and based on that we can generate and identify the equatorial plane
projection which is the stereographic projection and based on the stereographic
123
projection we can identify therefore, the point group symmetry that is associated with
crystal.
This is the way in which ancient crystallographers used to do people who study crystals
will look at the crystals and do these experiments and now eventually we will develop a
methodology which will tell us how to get these symmetry information on a direct basis.
So, this now tells us the systems to which we can go into. So, this is a general listing
triclinic system. The triclinic system can have 1 and ̅1 , a monoclinic system 2, ̅2 and 2/
m and so on.
So, we will see how each one of them develops with respect to the stereographic
projection as we go further.
Here is a diagram which shows you the stereographic projections of all types of crystals
belonging to these systems. So, here are the crystal systems triclinic monoclinic
tetragonal, trigonal, hexagonal cubic and this is the axis of rotation. So, the axis of
rotation can take values 1, and then it can take the value of 2, it can take the value of 3,
it can take the value of 3, it can take the value of 4, they are grouped together and the
way in which we are organized is triclinic, monoclinic, tetragonal, trigonal, hexagonal,
cubic. These are the these five in fact, trigonal, monoclinic, tetragonal, trigonal,
hexagonal these five systems have single axis symmetries as well as in case of
124
tetragonal, trigonal and hexagonal you can have more than one axis that is present which
is indicated here by individual stereographic projections.
So, our job now is to study each one of them. We have to study each one of them and
understand how the equivalent points are generated. So, we will make an attempt today
to generate to study a few of them and in the next class we will take up a few more of
them and then in general we take a few examples from the more complicated looking
ones on the right hand side.
And, then we will go into a representation of each one of them separately as we saw here
and as we discussed in the case of stereographic projection and then distribute them to
the various crystal systems.
So, this looks very complicated, but one once we do it and one once we revise what we
have done we will see that it can all be fitted into one single understanding of these point
group symmetry, the crystal system and then the distribution of these systems with
respect to generation of equivalent points. We have already discussed the generation of
equivalent points in case of 1 bar symmetry for a given xyz there is an x̅ y̅ z̅ bar and that
is in fact, illustrated here you have an xyz and an x̅ y̅ z̅.
Since this line now represents the line of demarcation and the location of the centre of
symmetry is here and because the location of the centre of symmetry is here and this is
125
the projection which has been done we see both of them as open circles you might be
asking why if this is one is open and one should be closed, right because if there is a
center symmetry and that projection is 90 degree oriented with respect to this projection.
In this projection we are considering the axis of rotation to be 1 and then the axis of
rotation ̅1 . So, the axis of rotation is the one which is now shown along this direction.
And, therefore, you see this is xyz, this is x̅ y̅ z̅ or vice versa because of the fact that the
central symmetry sits here it does not matter whether you have the xyz or x̅ y̅ z̅. So, that
therefore, tells you the two point group symmetries that can be associated with the
triclinic system. So, a triclinic system therefore, has two equivalent points in case we
have presence of a center of symmetry or more appropriately an inversion center and that
inversion center is present at this point where the pointer is shown.
Then we go to the monoclinic system and in the monoclinic system we see that we
generate again these two. Remember, now this is a projection which is showing in a
perpendicular direction two of the angles are 90 degrees with respect to each other we
can also show this as a two-fold rotation here. Now, you might be wondering what is
going on here because this is shown nothing here and this is shown as a picture where we
have one below the other and in this case one away from the other.
So, what is shown here are the two diagrams; it is not necessary to show two diagrams in
triclinic because there is no their axis of rotation. The moment you have an axis of
rotation we show these are the equivalent point positions and this is the symmetry
position. So, for every point group therefore, we show two diagrams and therefore, these
are the two diagrams which will be shown for each crystal system.
126
So, then we will go further into the presence of the twofold symmetry as I mentioned that
two diagrams now I have to be given for each one of these and this is now showing the
equivalent points, this is showing the symmetry position. So, this is the mirror operation
and you see that in the mirror operation I do not know whether you see it very
thoroughly here, this has to be a bold circle. This circle is lighter than this circle so; that
means, the mirror is now flat with respect to the diagram we are showing.
So, if you see the original textbook you will see it better this is from the textbook of
(Refer Time: 11:45). In fact, there is a proposal from my side that we will give you the
handouts and we will discuss with the NPTEL authorities and see we give a set of
handouts for each and every lecture to follow and that is something which will be very
useful for this symmetry class because symmetry and structure is very fundamental for
all areas of science and engineering and therefore, I want you to get a clear hang of these
symmetries. So, a handout will be accompanying the lecture series.
So, this now will be the darker one. So, it shows that the presence of this. In the case of
the orthorhombic system now we show the presence of 2/m this is the presence of a two-
fold followed by one bar rotation. The twofold followed by one bar rotation generates
two possibilities; one is the presence of these equivalent points now you see the mirror is
in the middle and therefore, this is a presence of a two-fold and now this is a darker
circle. So, the twofold axis is coming towards us the darker circle is the perpendicular to
the twofold axis. So, it is a 2/m symmetry and ̅2 which is mirror and 2/ m symmetry now
represent all the possible point groups associated with a monoclinic system. So, triclinic
has two point groups monoclinic has three point groups.
We go further now into orthorhombic symmetry in case of the orthorhombic symmetry

we can have three twofold axis mutually perpendicular to each other that is because we
have an a axis, an b axis and the c axis. The angle between a and b, b and c and c and a
are all 90 degrees with respect to each other. Because we have these 90 degree angles we
will have their two-fold axis along a, two-fold axis along b, two-fold axis along c. This
generates the equivalent diagram of this kind where you have a open circle and a closed
circle open circle and a closed circle. So, the number of equivalent points you generate in
a 222 point groups will be xyz then you first operate the twofold symmetry operation in
one direction. Let us say along the x direction, then you will have x y̅ z̅, then you operate
the twofold along the y direction which will be x̅ y z̅ and then you operate the twofold
127
along the z direction which will be x̅ y̅ z. So, which means that, it generates four
equivalent positions.
In case of the triclinic system, for example, you will generate two equivalent positions;
xyz, x̅ y̅ z̅. Please note that this picture I do not know how clear is there up there this is a
close circle and this is a open circle you do you do see that? No. So, this diagram is not
that very good I will be showing it more detailed in the next set of diagrams.
So, the presence of the three intersecting twofold axis now generate the symmetry
disposition. These are the symmetry disposition diagram where you see there is a two-
fold along this circle, the two-fold along the direction along x and direction along y. So,
all three directions have twofold symmetries. So, this represents 222.
Now, the combinations of 222 can be 2mm because we can have a proper axis and
improper axis and followed by another improper axis. No other combination is possible
which we have discussed much earlier with several classes ago. So, this can be 2mm or
2mm is same as mm2 and m2m because a b and c in an orthorhombic system are
interchangeable. We can always call a as b, b as c and c as a because the only condition
we put in orthorhombic systems is the condition on alpha beta gamma being 90 degrees.
So, we can interchange a, b and c.
So, this represents a common representation where we have a and b directions associated
with the mirror planes and the c direction associated with a two-fold. So, the symmetry
diagram is up here it tells now these are darker as you can clearly see that represents the
mirrors. So, one mirror here, another mirror there and the c direction which is coming
towards you has a two-fold symmetry.
So, the equivalent point diagram again is shown here. This should have therefore, how
many do you expect as the number of equivalent points? In this particular case of 222 we
are expected four equivalent points, in case of mm2 we expect again four equivalent
points. On the other hand when we have a symmetry which is mmm; mmm actually
represents 2/ m 2/m 2/ m and therefore, you will have number of equivalent points as 8;
four up and four down in this diagram, four up and four down that is totally 8 and this
now becomes all dark in the sense that there are three mirrors perpendicular to each
other.
128
The presence of the mirror and the angle 90 degrees will invoke automatically the
twofold axis around all three directions. So, there are perpendicular two-fold axis coming
towards us then the mirror planes which are intersecting in all three directions and that
represents the highest symmetry that is possible in case of an orthorhombic system.
So, today we have covered the triclinic, the monoclinic and the orthorhombic systems we
have shown how the equivalent points developed. The next thing is we will have to see
how the tetragonal system develops. In case of the tetragonal system we now look at the
possible symmetry positions which are coming with respect to the operation. So, as we
go back here x represents the rotation axis.
So, there is a fourfold rotation. There is a four-fold rotation with a mirror combination
that is with a inversions combination which is ̅4 symmetry and we also have a four-fold
rotation which accompanies a mirror symmetry. These are single point operations; that
means, these are single symmetry operations the moment you have a four-fold
symmetry, you will have also the presence of an inversion center associated with it just
like the monoclinic system we had a twofold symmetry, you had a ̅2 symmetry. So,
with 4 , we will have a ̅4 and we also have a 2/ m therefore, with for symmetry we have
a 4/ m symmetry.
Now, these represents again the equivalent points on the left side and the symmetry
diagram on the right and this therefore, represents the tetragonal system. The tetragonal
system can also have more than one axis. The fact that we have these orthorhombic
systems which are more than one axis PPP, PII and so on applies to these systems as
well, and therefore, in a tetragonal system we have 422 and 4mm; just like we have 222,
mm2 we can have it 4mm because we always represent the z-axis to be associated with
the four-fold rotation. The moment the z-axis is associated with fourfold rotation we will
have this issue of 4mm and that is unique that has here mm2 could be 2mm m2m and so
on.
So, because we have 422 and 4mm and we have just ̅4 symmetry. 4 bar symmetry will
give rise to ̅42m and also 4/m 4/m 4/m where we have the combination 4/m sorry 4/m 2/
m 2/m which is now representing the overall highest symmetry that is associated with a
tetragonal system. So, the tetragonal point groups are 4, ̅4 , 4/m, 422, 4mm, ̅42m and
4/mmm. So, 1 2 3 4 5 6 7, so, seven point groups can be associated with a tetragonal
129
system, monoclinic has three, orthorhombic has three and triclinic has two. So, this left
hand side of this slide now covers 1 2 3 4 5, 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15. So, we
have covered fifteen of the possible 32 space groups by going from triclinic to
monoclinic to orthorhombic and lastly to tetragonal system.
Now, we on the right hand side is the remaining crystal systems. The remaining crystal
systems you not we already see that hexagonal and trigonal are a little complicated in
terms of having both the six-fold as well as the three-fold. So, it depends upon where the
origin is and where the threefold and the six-fold are located in these compounds. It is
not necessary to have a threefold symmetry and the six-fold symmetry looking one and
the same; obviously, it is not as you see from the equivalent point diagram and also from
the diagram which we have drawn on the right side. So, this gives rise to the threefold
symmetry, a single rotation axis symmetry which will give rise to ̅3 .
Now, as far as the 3 and ̅3 are concerned these operations now introduces these
translation symmetry which is not 1/2 or 1/4. Because the translation symmetry is not
half or one fourth, the issue comes up with the point group symmetry because we have to
have the definition of a point group; that means, when we start generating the points with
respect to these operations they should come back to the original point, that is the
definition of a point group. So, if you start from in this example if you start from xyz you
got a x̅ y̅ z̅ you have to come back to xyz by operating it again. So, that repetition of
operations will occur in case of 3 and ̅3 , but they will not occur if we consider the case
of 3 or ̅3/ m or 3/m.
So, that is because of the fact that the three-fold axis appears at a rotation distance at a
uniform distance of 2/3 and 1/3 from the axis. And, the location of the threefold axis in
the case of the six-fold axis again there will be three-fold axis which will be rotated at
1/3 and 2/3 and therefore, we will have in this particular case no possibility of extension
of 3 and 6 being same and therefore, ̅3 and ̅6 are different from each other.
Since ̅3 and ̅6 are different from each other we can have a combination of 6 and a
mirror perpendicular to. A combination of the three fold and a mirror perpendicular is
not allowed. This is something which we probably will give you as an assignment I do
not know, but I think it could be a heavy assignment in such situations we will worry
130
about that fact whether we can you will be able to digest that as an assignment as we go
along.
But, the fact remains that with the trigonal as well and along with the hexagonal system
we should have the more than one axis intersecting at a point. When we bring in that
issue we will have these additional point group symmetries 32, 3m and ̅3m and the
equivalent point diagram is on the left and the symmetry diagram is on the right. and,
that is illustrated for all these hexagonal systems as well as from the trigonal systems
because of the fact that 3 and ̅3 create problems we do not have a ̅3m2 here. And,
therefore, we have 6 in fact, we have five trigonal point groups and 3 plus 3 6 plus 1, 7
hexagonal point groups.
Cubic system is quite complicated as we saw in the earlier classes also because of the
fact that the axis do not coincide with a, b or c. The only axis which coincides with a or b
or c is the two-fold axis which is a redundant operation in a cubic system and the issue
that the cubic system axis should not appear either along a or along b or along c the
minima the symmetry that is required is also very crucial and therefore, we generate very
complicated diagrams, stereographic diagrams as is shown here.
We will probably spend a little time on this or not we are not sure as we go along that
these can be studied with respect to the tetrahedral and the octahedral distribution we
discussed in several classes ago. We did talk about the way in which the distribution
occurs when we discuss individual rotation axis and then brought in the issue of the
location of these retention axis and with respect to that therefore, we now have a total of
32 point groups. And, this I think is more or less the stage at which I would like to end
here.
131
Prof. T.N. Guru Row
Department of Solid State and Structural Chemistry Unit
Lecture – 11
Stereographic Projections (Continued)
So far we have seen the classification of point groups.
We have seen there are 32 point groups and they can be put into the 7 crystal systems
which we have defined earlier. On the 7 crystal systems, they qualify themselves in
terms of their dimensions in terms of the 6 parameters a, b, c, α, β, γ and in case of a
triclinic system, it has 1 and ̅1 as the point groups. In case of a monoclinic system, we
have 2, ̅2 which is mirror 2/m and so on.
If we read any text book, the textbook will tell you only about how we can represent
these stereographic projections of each one of these point groups and then, straightaway
classifies them into these crystal systems. What I would like to do and I would like you
to understand is how this classification takes them into the individual crystal systems and
in what way we can generate the objects inside the stereographic projection which will
give us the relationship between the equivalent points, so that we can understand how
many equivalent points get generated in this stereographic projection of the point group
for a given crystal system.
132
So, each and every crystal system therefore has to be examined independently and
individually with respect to their distributed point groups and that is what we will try to
attempt.
What I was trying to say is that you now if you look at this picture which will essentially
we found in many of the textbooks, they will simply give the point group stereo graphic
projection and show where the objects are and then, they say triclinic and x being the
rotation axis, this is by far the best description you can find. This is taken from the
textbook of, but then you see the understanding of these has to come only if we go
individually and discuss each one of them and how the crystal systems go in to it and I
thought that would be the job of a course which describe symmetry and structure in the
solid state.
So, from that particular point of view I will now take you through the crystal systems and
the corresponding point groups as we go along.
133
So, let us start with the simplest of the systems Triclinic and well in some sense it is not
the simplest. It is probably the most complicated from the structure point of view
because a is not equal to b not equal to c, alpha beta gamma are not equal. So, a not
equal to b not equal to c alpha not equal to beta not equal to gamma and in fact, you can
even say they are not equal. It could be 90; one of them could be 90. So, the way in
which this particular unit cell is now described, there is no question of any directional
preferences of the symmetry operations in this.
That means, we cannot invoke any rotational symmetry in a triclinic system or for that
matter as the consequence of which is that we cannot any plane of symmetry in a triclinic
system. So, when a crystal system, when the object goes into the triclinic symmetry, it
can now only have a centre of symmetry or an inversion centre and therefore, when we
do the two projection diagrams, we will have this as x y z. Let us say the cross mark (x)
will identify a coordinate system and this coordinate system could be with respect to
these a b c and α β γ. So, the coordinate system will be x y and z.
Now, for all practical purposes in a all of crystallographic literature these x y and z
represent what we call as fractional coordinates. So, x y z is a fraction of a fraction b and
fraction of c. Remember in this particular case, it is not a system with 90 degree angles
as well. So, these are referred to as fractional values of a b and c. So, normally the x
value will appear as let us say 0.1 which means it is 0.1 of the distance a or the length a.
134
So, the fractional coordinates are representations of the 3-dimensional coordinate system
in the given particular crystal system. So, if the system is triclinic, we have this particular
way of representation that is to see that x y and z now represent fractions a, fractions of
b and fractions of c. This is a very important point because when we talk about the
coordinate system in a triclinic system, we should not assume that the x y and z are
coming from a rectangular coordinate system.
That means, the 90 degree angle is not there in this system. So, if you want to really look
at the and particularly people who do energy calculations and theory calculations based
on these kind of information, they should convert the x y z to the proper set up of
coordinates before they use their theoretical approaches because most of the theoretical
approaches assume a the system to have 90 degree angles and a b and c will then
correspond to x y and z axis.
So, that is why the nomenclature is to be carefully noticed here when we say little x, little
y, little z. These represent the 3 directions x, y and z, but these x y and z now represent
actually the cell dimensions a b and c. So, we should remember this issue and one once
we have in a triclinic system, the presence of a inversions centre is a possibility because
inversion centre is with respect to a single point. There is no question of involvement of
any rotation axis or mirror plane or whatever with that and therefore, only these two now
represent the point groups in the triclinic system.
Now, the way in which we generate the equivalent points in that situation are that if you
call this x y z, then if you call this as x y z this point, then the corresponding point which
is a open circle will be -x -y and -z which we pronounce as x̅ y̅ z̅ that actually is the
inversion of x y z about itself across the centre of inversion.
So, this gives us the symmetry operations which come inside the triclinic cell in the case
of the point group symmetry. We can also put in another information on this slide and
that is the information about the bravais lattice we have already seen the 14 bravais
lattices in earlier class. So, this particular situation in case of triclinic systems because of
the fact that there are no rotation axis and no plane of symmetry and so on, no other
possibilities there except a primitive system.
So, we have a capital P. In fact, conventionally it is written in a way which is like this a
italic P. This is the international union of crystallography representation an italics P,
135
capital P represents a primitive lattice. So, when we write a capital P and say it is a
triclinic system, it can have only two of the symmetry elements which can be associated
with this lattice. So, the first one will be there for P1 and the second one will be P ̅1 . So,
there are two possibilities therefore to combine the lattice symmetry with the symmetry
that is associated with the crystal system.
So, the triclinic system therefore will have the triclinic system. Therefore, we will have
two possible space groups. These are refer to as the space groups P1 and P ̅1 . So, here
we introduce the notation, the next point in the grammar that is the definition of a space
group.
So, we have now in a triclinic system a full description that is available on this slide. Any
triclinic system is characterized by a not equal to b not equal to c, α not equal to β not
equal to γ. So, this gives us the crystalline nature or the three dimensional nature of the
triclinic system and in that the coordinate fractional coordinate is presented as x y z.
If it is a symmetry associated with 1, it gives you only one position. That means, the
number of points or number of what we call as equivalent points in this particular case is
1 and that is normally refer to with a symbol z. So, z is equal to 1 that means, there is one
molecule. If there is a molecule sitting inside this triclinic cell that is 1 molecule in the
unit cell because this is the unit cell.
So, very interestingly whatever we have learnt so far is more or less caught up in one
slide. We have a triclinic system, we have the a, b, c, alpha, beta, gamma defining the
dimensions of the unit cell and then, this in this particular unit cell if there is only a
symmetry 1, then there will be one equivalent point which is identified as x. This could
be anywhere in this space. This is a stereographic projection.
The corresponding stereographic projection of the ̅ 1 is shown here where x is x y z and

this point which is the symmetry related across by the presence of the inversion centre is
-x -y -z and therefore, there are two in this particular case. So, the value of z in this
particular case will be 2 and the value of z will be 2. So, there are two units. So, suppose
there is a molecule or an atom or an ion ionic species which is in the position x y z and
equivalent atom or an ion or a molecule will be generated at this point.
136
So, therefore this particular unit cell with the dimension of a, b, c, alpha, beta, gamma in
the triclinic system will carry two molecules in the unit cell. Therefore, the unit which
appears independently of the unit cell, the so-called asymmetric unit is in this particular
case is 1 and also in the other case of P1 is also 1. This is referred to as the asymmetric
unit.
So, the asymmetric unit therefore in is 1 in both these cases and that is referred to as a
quantity called that z' and that is equal to 1. So, in a literature if you see is z equals 1 and
triclinic system, then it is automatically seen that it can have P1 that is not the correct
statement because if z is equal to 1 and it is a triclinic system, it can also be P1 bar in
which case the molecule atom or the ionic species will be associated with the inversion
centre what we then refer to as the special position special position.
So, the special position in this particular case is the inversion centre is the invention
centre. So, this inversion centre will hold now 1 unit whether it is an atom or a molecule
or for that matter an ionic species. So, let me go through this again just for the
convenience because and also for the fact that as we go along to the next set of crystal
systems and symmetries, all these issues will not be explained again, but I will just tell
you what is the z value, what is the number of equivalent points under parabolity way in
which the symmetry develops and so on.
So, if we now are given a triclinic system, these are the parameters which we go
automatically along with it. The first one is the unit cell which is characterized by a not
equal to b not equal to c alpha beta gamma are not equal to each other which defines the
unit cell. In the unit cell we have a representation of an object which could be identified
with respect to a coordinate x y z which is fractional coordinate.
Now, in a triclinic system there are two possible point groups 1 and ̅1 . The point group 1
represents just one position and therefore, the number of objects in this unit cell is equal
to 1. In the case of the P ̅1 , the number of objects in unit cell will be 2 defined by x y z
and x̅ y̅ z̅ if it so happens that the object sits at the centre of inversion, then the number of
the objects is z' which is equal to 1. So, the one in which we do not represent are the unit
which does not contain information about the point group symmetry is they called so-
called asymmetric unit. So, in both these cases the asymmetric unit that is the value of z
is equal to 1.
137
So, this in a nutshell describes the entire features which we have developed so far
including the feature that was associated with the bravais lattice that is the primitive
lattice. So, triclinic system is only primitive and nothing else. So, all the information you
want to know about a triclinic system appears here in this one single slide. We shall now
go to the next slide which is now going to take us to the Monoclinic system.
In a monoclinic system we will have definition of the unit cell, the definition of the
number of possible point group symmetries the nature in which of the equivalent points
which will develop and so on which is shown on the left. So, a monoclinic system
therefore has the following features just like the triclinic system. We will write down
now the features associated with the monoclinic system.
The first is that a not equal to b not equal to c and alpha and gamma, the two angles are
90 degrees with respect to each other which will essentially hold a twofold symmetry
and then, not equal to beta as you can see since a b and c are not equal to each other. It is
the angles which decide the direction of the two fold axis. So, in this particular case since
alpha and gamma are 90 degrees and beta is non 90, the value of b the cell dimension b
is the one which will now have the two fold symmetry. That means, the two fold axis
which is shown up here, I will show you in a second the two fold axis which is shown
here is now along the b direction and this is known as the unique axis in a monoclinic
system. So, this is referred to as the unique access.
138
So, in fact as you can see that this is a definition which can be a variable definition. That
means, if the two fold axis is along the x direction, the two fold axis along the a axis or it
could be along the c axis. So, those are referred to as non-standard settings. A standard
setting is the one in which the unique axis is b as far as monoclinic systems are
concerned. So, generally the monoclinic system is defined with respect to b axis being
unique the equivalent points are x y z and I want you to figure out what is that equivalent
point given the two fold operation.
Obviously, two fold operation will now keep the identity of the object same. That means,
there is no inversion here. So, what will happen is the value of x will go to -x, the value
of z will go to -z because these are the b axis being unique there is no change of the y
value. So, you will therefore get equivalent points corresponding to the two fold
operation as follows and that will be x y z and x̅ y z̅.
So obviously the number of the objects which can come in this particular crystal system
monoclinic with two-fold point group symmetry will be z equals 2 in the units cell. So,
there are two molecules in the unit cell and these two molecules are related to each other
by x y z and x̅ y z̅. Needless to say that there is not necessary it is not necessary that we
have to have one cluster like what we show here. We can have more than 1 plus. I will
just explain that in a second.
If you look at this situation here, you can have x, you can have a y, you can have a z on
this side and the corresponding y z will appear after the twofold symmetry operation,
with this operation x y z, x̅ y z̅ and they will appear on either side of x. So, you may have
a y, up there y will come down here, z up here down here that will come as I said up
there.
So, this could be clusters of molecules or different molecules or their molecules which
can be related through hydrogen bonding in a co crystal whatever in such situations also
the two fold axis is the prominent axis that is present in the monoclinic system and if the
crystal system turns out to be a monoclinic one with a twofold axis, then x y z, x̅ y z̅
would be the relationship between various aspects, various sizes and different objects
which may be present in this.
The objects may be connected to each other and may not be connected to each other.
Also, they could be just in van der Waals interactions between these units. So, the overall
139
volume of the unit cell now we have not defined that yet. So, the volume of the unit cell
will be defined now. The volume of the unit cell the unit cell of course is now there are
two possibilities for the unit cell. So, let us first define the unit cell possibilities. The unit
cells can be a primitive lattice as well as C centre lattice. So, this space group can be P2
as well as C2.
So, there are two possible space groups which can be associated with the two fold axis
shown here, the two fold point group symmetry shown here. So, the primitive nature of
the bravais lattice and the C centred of the bravais lattice both of them represent the
monoclinic system as we saw in the previous class will essentially tell us the possible
number of equivalent points.
Now, here is a situation where we will introduce how the equivalent points get generated
when you have C centred lattice and I think it is better that is described in terms of
understanding the bravais lattices rather than just write down the values here. So, at this
particular point I will say that the presence of the C fold axis, this is all we have to note
at this moment, but we will clearly derive it with respect to the understanding of the
lattices and the and the types of lattices we can have in a given crystal system. So, at this
particular time what we will say is that we will add a half along x and a half along y. So,
we represent the so called C centred lattice.
So, what it means that in the case of the space group this is now the space group. So, in
case of the space group P2 we have two equivalent points. In case of the space group C2,
we will have 4 equivalent points. Each one of these half plus half plus will add on to
each one of these coordinates. So, we will in effect have 4 equivalent points.
So, the presence of the 4 equivalent points in C2 tells us that these equivalent points can
be generated either by the operation of the point group or by the presence of the
translational periodicity which comes as a consequence of this centering associated with
the lattice. We are going to examine this in more detail when we discuss the lattices and
types of bravais lattices which occur in each one of the crystal systems.
So, at this moment C2 is also a space group which will have now 4 equivalent points.
The definition of special positions might come up or they may not come up in a situation
like this. For example, in P2 system if the object has a twofold symmetry, then the object
can sit on the twofold symmetry and as a consequence in that situation the number of
140
equivalent points will turn out to be equal to 1. So, that will be a definition of a special
position in C centred lattice as well there will be special positions. This special position
can occur both due to the point group operation and also due to this centring operation
and therefore, that needs to be examined more thoroughly which we will do in the
coming classes.
The next system which we will take up is the point group symmetry which is
corresponding to the mirror plane. So, in this particular case as you will see here there
are different kinds of notations. So, if you write the mirror symmetry like that, that
means that the mirror symmetry is now with the stereographic projection equatorial
plane so which means that the mirror is in the plane of our presentation and you see here
that the x y z and the generated mirror component which is x y̅ z, they are on top of each
other.
So, the mirror symmetry now can be either represented like that or we can make this in
some text books they make this a darker circle to indicate that the mirror symmetry is
resident in the plane. So, you have to be careful when you see the text books and see
whether they have darken the image of the equatorial plane.
So, in this particular case the equivalent points will be x y z and x y̅ z. That is why this
operation which is shown here is also shown as two bar. So, we operated twofold
rotation followed by an inversion. So, the operation of the two fold rotation with x y z to
x̅ y z̅ and then, we invert that we go to x y̅ z. So, x y z and x y̅ z therefore or the
equivalent points. So, again z is equal to 2 here and this also z equals to. So, we have
therefore included now in the monoclinic system two possible space groups P2 and C2.
141
Prof. T.N. Guru Row
Lecture –12
Point Group and Crystal Systems 1
So, when we come to the case of use of a mirror symmetry, here we therefore have the
operation 2 followed by inversion centre or immediately on acting on the two fold and
therefore, we get a mirror. So, in a mirror symmetry we have equivalent points at x y z
and x y̅ z. So, the axis again is the unique axis again is the b axis. So, x y that goes to x y̅
z and the b axis being unique, if the c axis is unique, the equivalent points will be x y z
and x y z̅ and if a axis is unique, it will be x y z and x̅ y z.
So, again you see that there are two in the unit cell and therefore, one in their symmetric
units. So, the value of z is equal to 2 and the value of z prime is equal to 1. The system to
which it goes into again can be a primitive lattice which is now going to give us the
space group Pm and it can also give the space group P sorry Cm Cm. So, we can
therefore have two possible space groups; the primitive with a mirror as the, so this is in
fact the way in which one represents the space group. We are going to do more seriously
the space groups later on, but this is the way. So, one once we know the crystal system,
we tell first the lattice symmetry. The first we tell what the type of the lattice we have
142
and when once we know the lattice type whether it is P or C centred or whatever as we
go further in case of the triclinic there is only primitive, but in this case it is primitive and
C centred lattice. The C centring can also change to a centring or B centring depending
upon the nature of the unique axis. In case the b axis is unique, the standard convention
is to have the correct centred lattice.
So, as I mentioned that one once the C centring comes up, you will have addition of half
plus along x and half plus along y for these two equivalent points thereby generating 4
equivalent points in which case z is equal to 4. So, you will generate 4 equivalent points.
So, whenever we represent there for a space group, the first we represent the lattice and
then followed that with the indication of the point group symmetry. So, the point group
symmetry with the lattice now will tell us this space group. So, a space group
representation is a lattice plus the point group.
So, this will tell us in fact something very interesting that we are now dealing with three
dimensional object a crystal is a three dimensional object. We have identified the crystal
system into which it goes into with a, b, c, α, β, γ based on that we have decided the
nature of the crystal lattice that crystal lattice can have the crystal system. The crystal
system will have a different kinds of lattices depending upon the bravais lattice
conditions you remember that in a primitive two dimensional rectangular system, we
have a primitive lattice and c lattice, the little p and the little c.
So, similarly we will have a capital P and a capital C in case of the monoclinic
symmetry. We have the lattice plus the point group now representing this so-called space
group. So, the representation of any space group therefore will tell us the complete three
dimensional information associated with the system. So, when once we have the space
group information, we automatically know the nature of the lattice, the nature of the
point group the equivalent points, the number of equivalent points in the unit cell and so
on.
So, that way the entire representation of any crystallographic symmetry along with the
presence of systematic, sorry along with the presence of symmetry operations are given
in a monoclinic system and in this particular case. So, similarly monoclinic system with
a two way have already discussed, monoclinic system with a mirror we have now
discussed we can go to these system which is. Now a combination of a proper and an
143
improper axis. So, this corresponds to a proper axis, this corresponds to an improper axis
operation and this is a proper plus improper.
So, simultaneously operating at a point. So, at this particular point which is up here the
symmetry operates in such a way that we have this point two fold is located. So, what
happens now is x y z will go to x̅ y z̅ and then, each one of these points now we will see
the mirror plane which is perpendicular to it, so that the proper and improper operations
are operating simultaneously. So, the proper operation will take x y z to x̅ y z̅ which is
the twofold rotation and each of these points now we will see the mirror plane and
therefore, x y z will go to x y̅ z and this one which is x̅ y z̅ will go to x̅ y̅ z̅. Let me write
it down, so that you are not confused.
So, if you have an x y z will go to x y z first will go to x̅ y z̅ . This is the two fold
operation. Now, we have generated therefore two equivalent points. Now, these two
equivalent points will now see the mirror which is perpendicular to the two fold axis, the
improper axis. So, x y that will carry over to x y̅ z and this one which is x̅ y z bar will
carry over to its mirror reflection which is x̅ y̅ z̅.
So, therefore the value of z now is equal to 4 and what are the types of space groups we
can have here? We can have P2/m and C2/m and these will have now 4 equivalent points
z equals 4 and C2/m. We will have z is equal to 8. So, there will be 8 equivalent points.
Now, I want you to go home and work out what are all the possible special positions that
can come in this situation. For example, we can locate the object on the twofold
symmetry, we can locate the object with respect to the mirror. So, in what way these
operations will come up is something which we will see as we go along, but at this
particular point we have covered the monoclinic symmetry.
We have covered the monoclinic symmetry in such a way that we look at all the three
possible point groups which belong to the system. We have looked at the crystal lattices
which are now P and C, P and m, we have looked at the lattices which are possible P and
m. Let me go through it with a lazer pointer. We have the three projection, the
stereographic projections shown here. The first one corresponds to 2, the second one
corresponds to mirror, the third one to 2/m and we see the symmetry operations
generating the equivalent points just to keep the interest of the chemist, the pharmacists
144
and maybe even some of the faculty members and also those who are not so very
inclined to mathematics biologists for example.
So, just to keep their interest going at this stage what I can say is that if there is a system,
if there is a biological system or a chemical system which goes into a monoclinic
symmetry. And it crystallizes in such a way that we have P monoclinic P lattice which
will appear, we will not know at this moment how we can identify whether it is P lattice
or C lattice that will come much later when we do the diffraction experiments. at this
particular stage. We now identify whether it is a primitive or C centred lattice.
Then, let us say the representation of x here is now replaced by our biological molecule.
It could be a small amino acid, it could be a peptide chain or it could be a protein in the
long run, it could even be a virus or a collection of viruses particles proteins more than
one protein and things like that protein crystallizing with a steroid molecule. So, you
have that stored in this particular x and if the space group turns out to be P2, then what
will happen is there will be such two units in the unit cell here. There is a small
suggestion to the biologist that is if there is a biological molecule and that biological
molecule has what they define a subunits in the biological molecule. Like for example
myoglobin and haemoglobin, myoglobin as a single subunit whereas, haemoglobin has 4
subunits.
Now, the presence of these subunits need not utilize the symmetry. So, the asymmetric
unit of myoglobin can still have 4 molecules of individual myoglobin units and then, it
may go into a space group to which it belongs to. Let us say it goes to P2, then we will
have 4 plus 4, 4 units of myoglobin accumulating as a haemoglobin molecule, but there
will be 2 haemoglobin molecules in the unit cell and as a consequence we will
effectively have the 2 big units of haemoglobin crystallizing in the unit cell.
So, it is not necessary that when we point out an x, it is one single point. It is a collection
of all possible atom positions. So, what I write as x y z could be x1 y1 z1, x2 y2 z2, x3 y3 z3
and xn yn and zn, where n is the total number of atoms representing that particular
molecule. So, the asymmetric unit if you now consider a symmetric unit will have this all
the four units of myoglobin defining the haemoglobin molecule. The haemoglobin
molecule has an internal symmetry is a molecular symmetry. So, that particular internal
symmetry when it crystallizes may or may not be used in that situation.
145
In the case of a realistic situation, haemoglobin does not use that centre of symmetry or
the symmetry that is present in the 4 subunits. So, therefore that is referred to as a
noncrystallographic symmetry. So, for biologists when they read their papers you should
remember that is a non crystallographic symmetry. In fact, the presence of non
crystallographic symmetry is utilized in structure determination of very large molecules.
There are various ways in which people can use this information to the advantage of
solving the phase problem and therefore, determine the crystal structure.
So, this I thought is a point which I should mention at this situation, so that there is no
confusion when we discuss with respect to x y z and say the x y z goes to x y̅ z in a
mirror operation. It means that the entire set of atoms associated with the molecule will
go to the other molecule. So, that way we have a situation where it is possible to now
identify the molecular symmetry and that molecular symmetry need not be associated
even though Kitaigorodsky said that molecular symmetry should be associated that is a
symmetry which is associated with a single molecule. Here there is a subunits which are
related by a symmetry. So, that is a non crystallographic symmetry.
So, there is a difference between a regular symmetry in the molecule, single molecule
and the non crystallographic symmetry associated with individual components. So, this is
something which I thought you should bear in mind and that is a crucial point which we
should discuss when we do the structural determination protocols. So, we have now
covered the point group descriptions of triclinic monoclinic.
146
And we will now proceed to the orthorhombic symmetry and then go for the description
of individual point groups. Here what I did here was to also overlay the diagram which
you have already seen which tells you the orientation of the axis when we have a
situation where there is more than one axis of rotation.
So, when we go to the orthorhombic system, we have three axis of rotation which will
intersect at a point and Euler’s theorem makes sure that if we have two of them, the
third-one will automatically come. So, the way in which the orthodontic symmetry
therefore comes is an intersection of 3 two-fold axis. So, when we represent the
corresponding point group, we should be able to represent all 3 two-fold axis in this
diagram and that is how 222 is a diagram where we have shown the two fold axis which
comes here at the centre which is along the direction which is perpendicular to us.
There are two-fold axis of rotation also associated with the x axis direction and also the y
axis direction. So, that way the two fold axis operation is quite interesting here. Suppose
you take this point x, this point x now goes over to that point x. This is the first two
operation which is let us say about the x or y or that it does not matter which one in a
case of an orthorhombic system.
So, we will say x y z now goes to x̅ y z̅. That means, it is a two fold perpendicular to the
y direction, then the two fold perpendicular to the x direction and the two fold
perpendicular to the z direction will also simultaneously operate on that and as a result
147
you get the 4 equivalent points. So, the 4 equivalent points I want you to take it as a
home assignment where you write what are all the 4 equivalent points which I have
written here. It is very obvious by looking at the nomenclature how they should be
defined, but then this is interesting to see how you will define it. For example, how
would you write this to that if I give that example, then you will write all the four. So, I
would not give the example except to say that you do see the relationship between these
two turns out to be a relation which is the inversion, kind of because you have a open
circle and x mark here, but what are the equivalent points of these, how do I get that
point to this point notice.
The circle which is going around here is also specially marked. It is there is a darkening
of that circle. So, you have there for a two fold axis perpendicular to that a two fold axis
along the x direction, a two fold axis along the y, so mimicking this representation. The
same representation is carried over down here and you see a difference in the equivalent
point position.
For example, in this particular case the two fold axis which is along the z direction is
shown here. So, obviously this is telling us that x y z when operated by this two fold
rotation will go to -x -y and z. So, here the rotation for x y z going to -x -y z can be
written, then you end up with two mirror symmetries; one of the mirror symmetries will
generate a mirror symmetry which is perpendicular to this direction which is the x
direction and the mirror symmetry perpendicular, let us say this is the x direction
conventionally and this is the y direction.
So, both x and y directions represent the mirror. Now, the question comes in the case of
the mirror symmetry. You had earlier an x y z and the mirror goes, it goes to x y̅ z. Now,
here again if you see the x coordinate, here the x y z if you take this as the x y z how will
you generate that point. you will generate that point using the two fold rotation using this
particular two fold rotation and then, inverting it.
So, because of the fact that you do the inversion operation twice because of the presence
of two mirrors here, you will get the same object repeated here. So, you write down the
equivalent points of mm2 that will be the next home assignment for you. When you write
the equivalent points of mm2, compare the equivalent points of mm2 with that of 222.
So, these two space groups are very rare in occurrence, but when they do occur they will
148
show some certain special properties and those special properties give rise to systems
which may show some non-linear effects and so on.
So, these space groups are very crucial for such situations. The next point group is the
mmm symmetry where we have the three two fold axis now becoming simultaneously
mirrors. So, this is in fact equivalent of representing the corresponding full symmetry as
I can call it is 2/m 2/m 2/m.
So, there is the simultaneous operation of two fold as well as the mirror which we saw in
the monoclinic case earlier on. So, a similar operation occurs here and you now see that
the number of equivalent points in case of 222 is 4. Number of equivalent points in mm2
is 4, number of equivalent points in mmm is 8. That is because we have the two fold
operation perpendicular to the mirror will take it onto itself and therefore, you will have
8 equivalent points. All these 3 symmetry operations correspond to this 222.
So, this now orthorhombic symmetry is identified with a minimum symmetry of 222, the
point group symmetries therefore become 222 mm2 and mmm. The 2 by m 2 by m 2 by
m is the highest possible symmetry that can be associated with an orthorhombic system.
The number of equivalent points here is 4, the number of equivalent points again in mm2
is 4 whereas, the number of equivalent points in mmm happens to be 8. The restriction of
course which we have not written, but which you know by now very clearly is the fact
thatwe have an a not equals to b not equal to c. None of these are equal whereas, the
angle between them alpha beta and gamma are equal to 90 degrees.
So, the three angles are 90 degrees to each other. They are mutually perpendicular to
each other and we have this system which now represents the orthorhombic system. So,
just like the monoclinic system was able to generate the 2, the mirror and the 2/m
orthorhombic generates the 2, mirror and 2/m in all three directions because we have the
three mutually perpendicular directions. So, therefore orthorhombic symmetry represents
an additional set of symmetry elements which can go along with it mainly because of the
fact that the angles alpha beta gamma are 90 degrees with respect to each other.
So, even though the geometry associated with the orthorhombic symmetry is simpler, the
corresponding geometry which results in the accumulation of all the molecular symmetry
inside the unit cell is more complicated. Let me make this sentence a little more clear say
if you take a triclinic system. In the case of a triclinic system, you get only two
149
equivalent points. If it is P ̅1 bar if it is P1, you get only 1 equivalent point. So, the
complexity associated with the molecule is less in the unit cell. So, in the unit cell either
there is one molecule or two molecules, the relation between these two being only a
centre of symmetry. If you go to the monoclinic system again it is not so very
complicated because again you will get two equivalent points in two fold equivalent
points in mirror and four equivalent points in 2/m.
So, the number of molecules you can put into the unit cell therefore will be either 2 or 4
or it sometimes depending on the special positions, it could be even 1 or 2. So, the
distribute the way in which the molecule now looks at it is to see that it wants to arrange
itself in a comfortable way. And as the molecule wants to arrange itself in a comfortable
way in the unit cell, the triclinic and the monoclinic ones are preferred. Then, you have
very flexible molecules and that is the reason why many of the organic systems with
their flexibility and functionality is attached to the organic moieties and so on.
They want to go into the crystal systems which are triclinic, monoclinic, orthorhombic.
Very rarely they go to higher crystal systems and in when they do go to higher crystal
systems, they will have some exotic properties to display otherwise generally most of the
organic systems crystallize either in a monoclinic or in an orthorhombic system because
of the fact that the functionalities have lot of freedom associated with them.
On the other hand, if you look at the inorganic materials these inorganic materials are
basically built up upon tetrahedral and octahedral units. It is only an association of
tetrahedral and octahedral units with linkers which are again not so very flexible and as a
consequence most of the inorganic materials they go to higher crystal systems, higher
that is crystal systems with higher symmetry. So, tetragonal and beyond tetragonal,
hexagonal cubic these are the preferred space groups in inorganic systems.
So, it is very interesting to note this point that the geometry associated with the
surroundings of an atom effectively decides what crystal system it should take and
therefore, the question of whether the molecule decides the crystal symmetry or crystal
symmetry decides the molecule is still unanswered in the sence that the nature is
eventually trying to minimize the energy.
So, the minimum energy configuration is what will occur in a crystal. So, in fact if you
look at a crystal therefore the way in which the molecules assemble what we eventually
150
referred to as supramolecular assembly in a crystal molecule is the best way in which
molecules can arrange in a supramolecular assembly and this is referred to very often as
supramolecule molecular assembly par excellence, the crystal represents supramolecular
assembly par excellence.
I mentioned it in between the discussions of our point group just not to keep the
continuity going, but also to give you advance information of what we are going to do
when once we understand the methods of structure determination. So, this is very crucial.
So, now we have finished one of these multi-axis situations we can go to the further
multi-axis situations.
So, we will go to the next slide which is now the trigonal system three fold system .
Interestingly when we come to the three fold system, it is it is always decided you know
the three fold system can exist by itself which is what is shown here the three fold the 3
with a mirror and ̅3 system. Notice the difference between this point group 3m and the
previous point group which we discussed with a monoclinic system, it is 2/m. The 2 slash
m represents a two fold symmetry and a mirror perpendicular to that whereas, in this
particular case the 3 and the mirror symmetry are operational together. That means, there
is a second axis that can be associated with the three fold symmetry.
So, three fold symmetry therefore can exist by itself or a three fold symmetry can exist
along with another. In other words, we can have a improper rotation or a proper rotation
151
associated with it and that will generate different space groups, sorry point groups.
Obviously, it is different space groups depending upon the type of lattice we associate
with the trigonal system. What we will do here after is bringing the space group
information associated with this much later when we discuss the Bravais lattice along
with the combination of Bravia lattice and the point group symmetry.
Mainly to avoid confusion because the number of space groups will now increase more
or less exponentially because up to orthorhombic, we have a large number of systems. In
fact, orthorhombic itself can have a primitive lattice, a body centred lattice A B C centred
lattice and also a face centred lattice.
So, since orthorhombic symmetry can generate so many lattics, the corresponding space
groups which can generate are quite large. So, instead of going into the discussion of the
space groups with the orthorhombic system right now, we will finish up the assignment
of the point groups with the rest of the crystal systems and then, come back to the
discussion of associating lattice symmetry and the point group symmetry with individual
crystal systems. That will also revise whatever we have done now for you once more and
as a result, it will never be forgotten.
The whole idea is that this concept I do not want anyone of you to forget whether you are
a mathematician, whether you are a physicist, biologists, chemists. The way in which the
symmetry operations happen in these crystal systems because of the geometrical
restrictions which get put due to the presence of the symmetry, we should be able to
appreciate it the moment we see it.
So, we do not have to study this again once we embedded this in our mind, then we can
describe any crystal structure and the description of crystal structure by someone else
than in a let us say a paper or a review we can easily comprehend when the reviewer for
example in his paper says you know the molecule is generated due to the presence of the
twofold symmetry. We now know how these are generated and to that extent the
understanding of the point group symmetry is crucial. So, we will go and understand all
the point group symmetries before we go further.
So, in a trigonal system the threefold axis can exist by itself that is just the threefold axis
as is shown in this place and this diagram here. So, this will now take this x y z and then ,
take it to this position which is 120 degree rotated and another 120 degree rotation brings
152
it to the 270 position and the 270 position comes back here and since it is a simple
threefold rotation by nomenclature, we write this triangle and a darkened triangle that
represents the three fold symmetry. And also, the fact that in the three fold symmetry we
also take into account whenever we now associate the three fold symmetry with respect
to the unit cell, trigonal unit cell we always associate that with respect to the z axis.
So, the three fold is referred to the z axis. This is a very crucial point because when you
now see the location of the threefold symmetry in a trigonal system and the location of
the three fold symmetry which you must have that minimum symmetry, the information
that should be present in a cube. For example when you go to a cubic system, you will
not find the three fold symmetry in the first position of the description. In other words,
the first position of the description will now represent the threefold symmetry along the z
axis in a trigonal.
So, there is a space group like for example I will there is a point group like for example
here 32 and there is also a point group 23.
153
So, the point group 32 represents a trigonal system whereas, you will see here, a point 23
represents a cubic system.
154
Lecture – 13
Point Group and Crystal Systems 2
Because now the threefold symmetry is not along a given axis, but the threefold
symmetry you see here is along the diagonal and this you must bear in mind because
when you talk about the point group symmetries, you will have to worry about this. Now
let us come back to the Trigonal symmetry which we were discussing.
155
So, the Trigonal symmetry have therefore must have is represented with the z direction
to have the threefold symmetry. So, the other two symmetries which can occur in this
particular case are either along this given x direction or along the 120 degree angles
because there is a threefold symmetry. The threefold symmetry can take this point to that
point to that point and also, it can take this point to that point on that point. So, 120
degree rotations on either side with a twofold symmetry will be representing the 322
symmetry for example.
So, what we consider here is only the 3 symmetry. So, that means there is no symmetry
associated with x and y directions. Now, how will you represent a trigonal system, in
terms of the a b c ? The trigonal system is represented in terms of a b c, It is actually a
similar representation to the hexagonal system and the values of a b and c and alpha beta
gamma will be represented slightly differently and how differently it is represented, we
will discuss later when we go to the discussion on the crystal system, the bravais latices
and the assignment of the space group.
So, at this moment we just consider the operation of a threefold symmetry on an object.
The object now is representing threefold symmetry. So, that means the object is located
here represented here. It rotates by 120 degrees. No I am on the wrong one. This object
rotates by 120 degrees and again rotates by 120 and again rotates by 120 and comes back
to the same position and therefore, this is a point group symmetry 3. There are no
156
additional symmetry elements. So, there are 3 equivalent points here and in this
particular situation, there is no special position either because we do not invoke any
possibility of a special position. On the other hand if we have a 3 system as it is shown
here where which will see that this point which is xyz now rotates by 120 degrees goes to
that point and rotates again by 120 degrees and comes to this point.
The second symmetry that is operating is an improper symmetry. So, when the first is a
proper symmetry, the second one is an improper symmetry. What should be the third
one? The third one should be also either a proper or an improper rotation, so which is the
one which is possible. So, if there is a proper rotation and then an improper rotation, the
third one has to be improper. So, that means that is also a mirror symmetry in the 3rd
direction. So, these 2 twofold axis which are shown here in 322 case now will become
3mm. In fact, it is 3mm we write it as 3m for simple purposes because we know that the
other mirror will automatically appear. There is no possibility of any other symmetry to
appear other than the mirror symmetry and that is why you see that each one of them is
related by a mirror symmetry. So, we generate 6 equivalent points.
Now, what happens to if the object is sitting on a mirror in any of these positions? Then
the object will mirror reflect onto itself and therefore, there is a possibility of special
positions which can come in this particular point group symmetry. The next symmetry is
the operation where we have a threefold symmetry followed by an inversion. So, there is
an inversion which is associated with the threefold symmetry and both of them are along
the same axis. That means, this is the axis which is 3. Now, it becomes ̅3 it carries the
inversion operation along with it and there for it again generates 6 equivalent positions in
the projection diagram, in the stereographic projection. The threefold symmetry will
generate these 3 axis and then, the corresponding inversions occur and so, we generate
these 3 open circles. So, if this is x y z will you get an x̅ y̅ z bar in a ̅3 symmetry, this
could be taken as an assignment. In fact, probably it will appear in the assignment sheet.
So, what is 3 bar? 3 bar is different from 2 bar. See in case of a ̅2 , we have a twofold
rotation followed by an inversion and the two fold it is not even followed by an inversion
the two fold itself inverts. The option for the two fold itself to inwards is given because it
is 180 degree rotation. In a threefold symmetry, we have 120 degree rotation. So, the
inversion about 120 degrees itself is not activated if you have a 3 followed by ̅1. ̅1 can
take only x y z to x̅ y̅ z̅. So, if this is at one-third position, then going to minus one-third
157
and minus two-third is not going to put them on top of each other. So, ̅2 degenerates to
mirror where as ̅3 will be actually an operation 3 followed by 1 bar. So, 3 followed by ̅1
is not going to generate any mirror symmetry and therefore, this becomes an independent
point group and this particular point group has 6 equivalent points.
So, the trigonal system therefore is now described with respect to the possibility of
having one single axis in these through 3 diagrams. In these two diagrams, 3 and ̅3, 3m
is a situation where it can have a the presence of 2 twofold axis as is shown here or it
could be two mirrors and what is shown here is the presence of 2 mirrors. So, it is a
proper improper improper rotation which essentially generates again 6 equivalent points.
You will see that the mirror planes are now 3 of them associated with each threefold
rotation.
So, that means they are mutually perpendicular to each other, the mirror operations as we
seen here. So, there is a mirror operation up here, the mirror operation there, mirror
operation there instead of the twofold and therefore, you indicate all those 3 in these
three dark lines, the dark lines representing the mirror operation.
So, we will go further on the trigonal system. The next trigonal problem will come up
because of the following. Let me erase this unnecessary stuff here. So, you see here two
space groups which are now containing operations. What is bothering you at this
moment I am sure is that where is the 3rd axis. Say you are saying there is a ̅3, you
158
earlier said ̅3 is a threefold and a ̅1 operation and then, there is 2/m. There must be one
more operation.
Now, the 2/m operation is also a center of symmetric operation. 2/m by itself is a center
of symmetric operation. Now, that centre of symmetry comes up there. So, you have a
threefold along with that a center of symmetry. So, the presence of the three fold and the
centre of symmetry takes this, this particular fellow well it rotates. Let us take this one it
rotates by 120, again 270 and then, comes back that is the threefold and the
corresponding inversion axis will take them to these open circles, they are also related to
each other by an inversion. So, therefore, we have ̅3 and 2m generated automatically.
So, what would be you think would be the next direction? So, you have 3 directions in a
hexagonal system. Let me go back to the previous slide. In this slide you see that we
have a threefold rotation and then, you have two fold rotation axis which are at 120
degrees to each other and they are intersecting with each other at this center here. The
threefold rotation is the one which has generated these axis and then, we can associate
with these axis either we can associate a rotation axis or just leave it ourselves. If we
leave it to our self such we get the threefold rotation or the ̅3 rotation.
If we do not leave it our self such we can have option of a twofold and a twofold which
means we have two proper axis. Now, if you have two proper axis under three fold
rotation, is it possible to have a space group? Sorry is it possible to have a point group ?
That is possible here in the one which is shown below. So, this shows three fold along
the z direction, two fold in a direction which is x and that is enough because the 3rd
direction will automatically get defined and that 3rd direction will have to be a proper
axis and in this particular case this has to be an improper axis. So, you will have
effectively a mirror here and a twofold here. And in order to represent that in a simpler
manner, the accepted representation of point group symmetries are 3 bar 2m and 32.
So, here in fact the full symmetry associated with this will be ̅3 , 2/m and m and in this
case it is 322. So, the moment you have a combination of a proper and improper axis, the
3rd axis will be automatically improper whereas, if you have two proper axis and one
remaining improper is not possible and therefore, you will have proper axis. So, it will be
322 and ̅3, 2mm. Those are the actual representation of full symmetry and that is not
necessary to describe. So, we just say it is ̅32/m and in this case it is 32. So, these two
159
trigonal systems therefore now we will define a set of equivalent points. In this particular
case, you see 1 2 3 4 5 6 7 8 9 10.
So, there are 10 equivalent points and this is interesting because what will happen when
it is put in three dimensional lattice? In other words what will happen to this point group
symmetry when we describe it with respect to your lattice? This question you keep
thinking about we will discuss it when we describe the space group associated with the
trigonal symmetries. Trigonal symmetry therefore is always a very interesting system
and not many molecules go into this trigonal symmetry and if they do go into the trigonal
symmetry, there will be some very interesting properties which will develop. So, we will
just keep that in mind and we go further to the Tetragonal system.
Tetragonal system is a little simpler system than the trigonal system from the point of
view of understanding. So, we will make an attempt to understand the tetragonal system.
What do we how do we define a tetragonal system? We define a tetragonal system the
following way. We say that in a tetragonal system a is equal to b not equal to c alpha
beta and gamma are at 90 degrees. So, a know a equals b not equal to c. So, it is the c
direction again which is unique just like in the trigonal system the threefold symmetry
was the unique axis, the fourfold symmetry is the unique axis for this system and
therefore, you see that if you have just the presence of a fourfold symmetry and no other
160
symmetry intersecting with each other, then you have a fourfold symmetry which will
generate four equivalent points.
These equivalent points get generated at 90 degrees with respect to each other. What is
important to notice here is that since a is equal to b, they are degenerate. So, you can call
a as b or b as a, it does not matter because a equals b defines a square planar lattice and
that square planar lattice will be associated with 90 degree angle at so happens that the c
direction which is now coming perpendicular to this square lattice, the value of the c will
now decide that it is going away from a cube because in the case of a cube a equals b
equals c.
So, now this c lattice the presence of c which is different from a and b tells us that we
take a cube and pull it out, it is a stretched cube in the direction of the z axis that is what
is shown here. You see that the fourfold is stretching out away from their axis. So, this is
the point of intersection.
So, at the point of intersection we can still invoke the presence of other axis, but the
point group symmetry can exist as itself as 4 and therefore, we can invoke ̅4 which is
associated with a center of symmetry. Now, what do you think are the equivalent points
in ̅4? I have left this equivalent point positions unattended or unmarked. So, these
equivalent points therefore will now bring in a separate concept altogether.
So, I want you to take it as a home as assignment and mark the equivalent points for ̅4. 4/
m symmetry which is simultaneous existence of 4 along with a mirror will create a center
of symmetry. These two are non-center symmetric. There is no centre of symmetry that
is involved in this. These are non-center symmetry and therefore, you have now 1 2 3 4
and the corresponding fourfold rotation as well as the mirror symmetry.
In fact, this so you will have 8 equivalent points here. You have 4 equivalent points.
Therefore, 4 equivalent points here and 8 equivalent points there, so what it means is that
if we take the value of z the value of z now varies considerably. So, if you have a
molecule which can be put into a tetragonal system, obviously the presence of the four-
fold system will invoke the presence of more than one molecule and in fact, it will be
more than one.
161
In fact, it is 4 or 8 or even higher. So, depending upon the nature of the crystal system,
therefore the molecules which will now go into these higher symmetry systems this can
be considered a higher symmetry crystal system. The number of molecules can increase,
but what is interesting to remember and note it is just a note at this particular point is that
when we are now dealing with the trigonal systems as I already mentioned most of the
tetragonal systems are inorganic rather than organic and the reason is that the tetragonal
system now can accommodate and invoke the presence of the presence of the possibility
of 90 rotation.
So, the molecule therefore can go 90 degrees, another 90 degrees and further 90 degrees
before it can come and overlap. So, the symmetry like the trigonal, the tetrahedral and
octahedral geometries are most suited for molecules to go into the tetragonal system. So,
if the system does not go into a cubic system, the inorganic systems prefer the tetragonal
symmetry. So, if you look into literature you will see a very large number of tetragonal
systems which take the octahedral tetrahedral molecules into them and that is something
which we should note because tetragonal symmetry occurs very often in inorganic
systems.
So, the presence of the fourfold, the ̅4 symmetry and 4/m symmetry the point group
symmetries are shown here, the equivalent points are generated and as I said the
assignment will be for you to generate the equivalent points of ̅4. Apart from that since
a is equal to b and not equal to c, the number of types of lattices as a consequence will
reduce. It will be now only a primitive lattice and body centered lattice.
Only these two are possible in a tetragonal symmetry. We are going to prove this later on
in when we discuss the possible bravia lattices that can be associated with crystal
systems. Even though we have listed it out in the previous classes, the detailed analysis
of why only P and I are possible will be done in the coming classes.
So, please notice that as far as the tetragonal system is concerned, there is no question of
having F centered lattice or A B C centered lattice because of the fact that the geometry
sort of forbids any other lattice that can exist other than primitive and I centered lattice.
So, the number of space groups associated with the tetragonal as a consequence will
reduce to that extent. However, the number of point groups that are associated with the
fourfold system increase and so, we get a very large number of systems that are possible.
162
So, for example in this particular illustration you see that we have now the possibility of
the fourfold rotation along with the possible perpendicular rotations associated with the
fourfold rotation. That means, the fourfold rotation can go 90 degree, 90 degree and
another 90 degree. So, the total number of possible symmetry elements are indicated
here.
So, 4 can go with a twofold as well as a twofold. That means, 4 can go with a mirror and
a mirror. So, we have to consider all combinations of proper improper and proper proper
axis combinations and so, we get additional point group symmetries which is ̅42m and
422. Again in this particular case we have illustrated the equivalent point positions of the
422.
You see that this is the fourfold position and this fourfold position will generate 4 of
these equivalent points and with the operation of the twofold which is now associated in
a direction of 90 degrees with respect to each other. As you see that there is an axis
which will go like that which is a twofold axis the perpendicular axis like this and then,
two diagonal twofold axis which can go with the fourfold symmetry. Let me show it
again clearly. You have an axis going like that and axis coming like that and both these
are twofold axis. Perpendicular to that are the two axis which go along the diagonals, this
is the twofold axis and that is the twofold axis.
163
So, because of that we will end up with the 8 equivalent points that are associated with
422. ̅42m now, we will have additional mirror symmetries which now go along the
diagonal. We have the fourfold symmetry and then, we in addition we also have the
twofold symmetry which will get embedded with respect to this.
So, the last m is actually 2/m. So, we will also have a twofold symmetry at the centre of
it. So, tetragonal systems therefore are defined with respect to the presence of the
fourfold axis on the various possible twofold symmetries that can be associated with the
fourfold axis. The fourfold axis again in general is representing along considered along
the z axis just like the trigonal axis where we consider the three fold along the z axis.
The same story goes with the hexagonal lattice as well which is 622, we will go to the
hexagonal system now after showing the remaining part of the fourfold rotation.
So, we can have the 4, for this 4mm is there and we also have the 4/m 2/m 2/m. Both
these systems point group systems will generate in one case it is 8 equivalent points, in
the other case 16 equivalent points. Along with the possibility of these now going into
more than one type of lattice that means, they can go both into the primitive lattice as
well as let me write it down it can go into the primitive lattice as well as I centered
lattice. So, the moment we have I centered lattice for every x y z every coordinates
which we generate, of course every equivalent point which gets generated by the P lattice
operation, you have to add half plus along x half plus along y and half plus along z.
164
So, for every x y z we have to add half plus, half plus, half plus in order to take it to the I
center. So, if there are 8 equivalent points in this 4 mm and if we now take it to the I
system where the I centering is introduced, you will have another 8 additional equivalent
points and therefore, there will be 16 equivalent points. What it essentially means is that
you as I mentioned already you can pack 16 molecules, but what is also interesting at the
same time is that when we really crystallized the inorganic systems particularly because
of the presence of the tetragonal symmetry and the octahedral symmetry, these symmetry
elements can adjust themselves across the fourfold symmetry.
And because they adjust across the fourfold symmetry, the number of molecules will
reduce considerably. In other words, the molecules are the molecules which go into these
kind of systems generally utilize the fourfold special positions. The special positions
which are offered in these crystal systems will be utilized and as a consequence the
number of molecules in the unit cell will anyway reduce. So, as we go to higher crystal
systems as I mentioned already, already the number of equivalent points will increase. In
fact, when we go to the cubic system, there will be 192 equivalent points, but when you
crystallize systems into these only highly symmetric systems crystallize.
And that is the reason why organics prefer only the triclinic, monoclinic and
orthorhombic systems and here the inorganic systems tend to crystallize. And therefore,
if you things of the recent organic, inorganic, hybrids for example. The hybrid structures
can go into higher symmetry space groups because they maintain the inorganic
symmetry, inorganic crystal symmetry intact and then incorporate the organic molecules,
so that the organic molecules now are forced to obey these symmetries and therefore, the
property of the organic crystal system therefore can be controlled and this is how the you
know metal organic frameworks structures and the covalent organic framework structure
is and so on the and all that are becoming more and more exciting.
Because they have a highly, they are put in a highly symmetric atmosphere and
therefore, they are constrained in their conformational flexibility. As a consequence
some special reactions can be carried out inside these moieties and that is how these
compounds become extremely important.
This is just a side point I thought I will mention.
165
Now, we will go to the hexagonal system. Again in the hexagonal system you see that
the sixfold symmetry is now perpendicular to that direction. In fact, this is a six fold
symmetry, it looks like a twofold in this diagram, but it is a six fold symmetry and six
fold by itself can exist just like the threefold and the fourfold. So, we get the ̅6 and 6/m
and in this particular case the equivalent points appear at 60 degree rotations because it is
a hexagonal symmetry and in fact, the value of a b and c will be a equals b not equal to c
alpha equals bet,. and gamma is 120 degree. So, this kind of a relationship exists for a
hexagonal system and we have there for 6 points which will be generated by a six fold
symmetry. ̅6 bar generates you see also 6 points one above and one below each other
because do you see here there is a six fold and then, a threefold incorporated into that.
This is what I discussed when I actually initially talked about the 32 point groups
symmetries I told you to find out the difference between the 6 bar symmetry and the ̅3
symmetry. The ̅3 symmetry has 6 possibilities. Here also there are 6 possibilities, but in
the 6 bar symmetry because of the higher symmetry nature associated with a 60 degree
rotation in the case of the 3 bar threefold rotation, it is a 120 degree rotation. You will
generate 6 points there whereas; here you will generate the six 3 points up and 3 points
down. Effectively again here the 6 points are generated, but their 1 on top and 1 on the
bottom. The 6/m symmetry generates again a mirror symmetry which is now attached
along with this and you now see apart because of the fact that there is a mirror symmetry
166
which is perpendicular to the six fold symmetry, you will get now 6 equivalent points
and 6 below.
So, the hexagonal symmetry therefore is again representable in terms of the other
intersecting axis. So, we have additional point group symmetries ̅6 m2, 6mm and the
equivalent points are marked here.
Then we are three additionally have 622, 6/m 2/m 2/m. This is the highest symmetry that
can be taken up by a hexagonal system. What happens to the hexagonal symmetry with
167
respect to the lattice? It has only one type of lattice in fact, that is a primitive lattice. So,
we only have a primitive lattice and the presence of the primitive lattice therefore is only
invoked in a hexagonal system.
There is no other possibility and so, even though the number of point groups as we
discussed in case of the tetragonal system are more, the number of lattices we can
generate with hexagonal symmetry are less. And this now is coming is he going to help
out to find out why only 230 space groups exists. So, we have only primitive lattice in
the hexagonal system and as I was telling you that there are 230 space groups as a
consequence. So, the number of space groups are limited to 230 and I do not know
whether it develops some static or something.
So, the six fold symmetry is therefore get define the presence of the six fold symmetry. It
now generates these many point group symmetry 6, ̅6, 6/m, ̅6m2 6mm, 622 and 6/m 2/m
2/m.
168
Lecture - 14
Point groups and Space Groups
So, in the last few classes we have been looking at the Point Group Symmetries and also
looking at how the point group symmetries can be combined with the 7 crystal systems.
Because since we are going to deal with only crystals and crystalline lattices, we have
introduced several grammar, grammar words. As we went along we introduced point
group, we introduced space group and we discussed 7 various crystal systems. We also
introduced equivalent points and then, we introduced the number of molecules in the unit
cell and so on.
So, what I thought was in the last two classes even though you have I have got a
reasonable gist of the whole thing I thought I will make a set of slides which will
completely revise what all we have studied, so that this is something which you will
never forget. And in that sense this class will be somewhat repetitive, but then it is
constructive in the sense that we will now put all the knowledge which we have gained
in order to understand how we take a crystal system. And in the crystal system we have a
unit cell and in a unit cell repeats itself to form a three dimensional crystal structure
overall and the points which repeat themselves in three dimensional space or refer to as
the lattice and things like that.
169
So, let us begin at the start with the least symmetric system which is the triclinic system.
In a triclinic system, there are no axis of rotation and therefore, there are no planes across
which reflections can take place. The only possibility that can happen in a case of a
triclinic system is a point about which we can have the centre of inversion. So, the
possibility of having a centre of inversion is the only issue which you have to consider in
the case of a triclinic system.
So, the two point groups which fall into this category are shown on the left. The one is
the space group the point group 1 with a equivalent point x y and z and the other is the
point group ̅1 which now takes any point x y z in space to any other point -x -y -z.
So, even though we are shown only one point here, this can represent a collection of
points and those points could be atom positions, it could be positions of ions, they could
be positions of fragments of molecules. Whatever that is they will always be following
this particular point group principle, that is in the case of the point group 1 there is only
one set of such things and in the case of ̅1 , we will have a the axis sorry the centre
symmetry related equivalent point for every point. That means, the centre of inversion
will be present on the set of objects which are represented by x and the set of objects
represented by the open circle here will represent that we do have a centre of inversion.
So, therefore the there will be two objects here and that is why we say this system has
two molecules or two units of repeat which are in the so-called asymmetric unit of the
170
unit cell. So, the unit cell of a triclinic system as we have been refined defining again and
again is given by this rule a not equal to b not equal to c α, β, γ are not equal to each
other. What we have to remember also which we revised several times yesterday is to
remember that x y z or the fractions of a b and c and this is very important because the
value of x y and z is always in fraction numbers.
So, suppose there is an atom here in a triclinic system sitting here, we will express it as
fractions of a, fractions of b and fractions of c. So, if we say there is an atom at 0.1, 0.2,
0.3, it means 0.1 along a direction and 0.2 along b direction and 0.3 along z direction
considering the full length as 1 1 1 each. That means, this 1 unit is taken as 1.0. So, the
value of a will now appear at a distance of 1.0. Suppose there is an atom at the just about
the end of a, then it is at 0.95.
Let us say what it means is it is just away from the full length of a. So, these fractional
coordinates therefore represent the positions of atoms or molecules or units or whatever.
Now, what we also define is what we call as a lattice. In this particular case, the lattice is
made up of points which are at these edges, the cell edges and they will also repeat
uniformly in all three directions keeping the definition of a unit cell. So, in this case of
triclinic system, we can have only one type of lattice where only the edges are occupied
and therefore, we call it a primitive lattice. So, if you count the number of lattice points
in a primitive lattice suppose you sit at this point and then, count the number there are 1
2 3 4 5 6 7 and 8 points.
Now, each of these 8 points are occupied by one-eighth because the other unit cells
around that surrounding that in three dimensions, they will also have this point as a
common point and therefore, each and every point here corresponds to one-eighth. So, if
we sum them up 8 colon points, one-eighth occupancy of each the total number comes
out to be 1. Whenever the total number of lattice points is 1, we call such a lattice as a
primitive lattice and this is the primitive Bravia lattice. We have already seen there are
14 Bravia lattices with 7 crystal systems
So, this is the first of the case where we have a triclinic system with a Bravia lattice P.
So, if we now combine the lattice with the point group symmetry that defines what we
call as a space group, so the space group now represents the three dimensionality
associated with our crystal. It will also tell us about the three dimensionality of the object
171
that is inside this unit cell. So, whenever the objects are put inside the unit cell, there
three dimensionality is the expressed in terms of various values of x y and z and at the
same time the lattice combined with the point group symmetry will define the so-called
space group.
So, this particular space group comes up because of the definition of a group. For
example, if you take ̅1 symmetry and combine it with the primitive lattice, it becomes P1
bar. Now, by the rules of the group definition when we operate x y z, use the centre of
symmetry and take it over to x̅ y̅ z̅ we operate the centre of symmetry. The inversion
centre takes x y z to x̅ y̅ z̅. The rules of the group say that one once we have this
coordinate system, if we operate on this coordinate system again the same operation the
same symmetry operation, so x̅ y̅ z̅ in this case the same symmetry operation we will
come back to x y z. So, that way there is a closure property we do not go out of this.
When we have x y z, we get x̅ y̅ z̅ and when we have x̅ y̅ z̅ we get back to x y z and we
not do not go out of the system.
And therefore this is known as a group and so such property is seen both with respect to
a point here and with respect to this space which is now made up of a lattice plus the
point group symmetry combined and which is referred to as the space group. So, the two
space groups we can realize in a triclinic system are P1 and P ̅1 . So, P1 gives rise to one
such unit in the unit cell and so, we call it the asymmetric unit as well and in this case Z
is equal to 1. We identify the number of molecules or number of units which occur in the
inside the unit cell let us say number of molecules.
When we consider a molecular structure, then the number of molecules is 1. If the

number of molecules is 2, it generally goes to P ̅1 position because it prefers to have a
inversion centre utility, but suppose we say that the number of molecules in the unit cell
are 2 and it is P1, then it means that there are two different molecules, but not related by
any symmetry. So, that means the presence of the so-called x y z in this particular case
refers to what is known as a asymmetric unit.
So, in an asymmetric unit there is no invocation of symmetry whereas, one once we

invocate invoke asymmetry it becomes now a equivalent point x y z becomes -x -y -z.
These two are equivalent of each other from the symmetry operation. Whereas, x y z is
by itself independent of it. So, therefore that represents the asymmetric unit that unit in
172
which there is no symmetry operation and then, the symmetry operation when we
operate it becomes now the equivalent point in the space group P ̅1. So, this generally
describes the overall view of any crystal system.
So, when we talk about a crystal system, we talk about the crystal system, we talk about
the point group symmetry associated with the crystal system and the corresponding space
group that gets generated by combining the nature of this lattice which is either primitive
centred or body centred or face centred. In this particular case of a triclinic system, it is
only primitive and therefore, we have the point group symmetry, the space group
information the triclinic being the crystal system. So, whenever there is a material which
crystallizes in this particular crystal system, all these issues which I have put on this slide
should be identified and discussed. One once we discuss all these issues, then we say we
have described the crystal structure.
Now, the x y z can be more than one as I mentioned it could be x1 y1 z1 , x2 y2 z2 each

representing different atoms inside this unit cell. Now, if we find out the positions of x y
z that is the next step, in this particular step which we have discussed. So far we are not
able to find out where x y z is. If x y z is at x y z, then all these rules apply, but where the
where is x y z? We do not know and that is what we want to eventually determine when
we talk about the structure determination. We want to determine the structure of the
molecule where the atoms are sitting inside the molecule.
And then we have to see how these molecules orients itself inside this unit cell making
use of the available symmetry information in the point group symmetry and therefore,
generate the molecular dimensions in the space in the three dimensional space which is
allowed by the spacecraft. So, again to conclude this discussion on this slide I will repeat
a crystal system triclinic if we had to describe this in full, we have to first describe the
unit cell, the nature of the unit cell a b c not being equal α β γ not being 90. And then, we
had to describe the corresponding point group into which the system goes into and then,
talk about the equivalent points which we can generate the presence of the symmetry
element inside the unit cell.
And as you know once ̅1 is the point group symmetry, the entire three dimensional
aspects associated with the unit cell every point in this unit cell will see the centre of
symmetry. And the centres of symmetry location is at this possibility as we have drawn,
173
it could be at the possible edges of this three dimensional box. So, the centre of
symmetry is here the moment there is a central symmetry here 1 unit translation along a
will create another centre of symmetry, 1 unit translation along b will create another
central symmetry 1 unit translation along c will create another centre of symmetry.
So, there will be 8 centres of symmetry that get created at all the edges. We have seen in
the two-dimensional space group discussion the what we call as the plane groups. We
have seen that whenever there is a symmetry element at the origin and one unit translated
a same or a similar in many cases, one in this particular case the same symmetry
develops halfway between the two.
So, we will therefore have centres of symmetries at these positions as well and that will
tell us all the aspects that are associated with this space group, that point we will take up
separately when we discuss the 230 space groups which form in the 7 crystal systems
under 32 point groups. At this moment we consider only the space group and the
corresponding point groups. So, we have 32 point groups, we consider only the crystal
system and when we want to look at the space groups and look at the details of the space
groups, we will discuss them separately in a later class. So, this is the information I
wanted to give you in a nutshell about the triclinic system.
Let us see what happens when we go to the monoclinic system. When we go to the
monoclinic system, if we look at the unit cell to start with the unit cell will have this
174
condition a not equal to b not equal to c alpha gamma are 90 degrees and beta is not 90.
So, one of the angles is not 90 and it is if beta is not 90, then it means that the twofold
axis which is a compulsory symmetry that has to be present to describe a monoclinic
lattice monoclinic cell will have to be along the y direction.
So, the y axis is the one which is then referred to as the unique axis, it is not necessary
that we should always have the y axis as the twofold rotation axis, but convention says
that it is better to keep it as the unique axis. There are cases where there are certain
special properties. See on crystalline systems along other directions and it so happens
that it necessitates the definition of a twofold axis not being along y, but either x or z. In
such a circumstance, the definition of the monoclinic cell will change instead of the beta
non 90 we can have either the gamma not 90 and this one alpha being not 90 degrees.
So, one of the angles therefore is not 90 degrees, the other a b c are all independent of
each other, there is no relationship between a b and c , a is not equal to b and not equal to
c. This brings up 3 point group symmetries as we can see here the first one is the so-
called twofold point group operation, the next one is the mirror operation that is so-called
m operation and the third one is these 2/m operation. As we see here that there is only
one axis of symmetry. So, we are now considering the case of one axis of symmetry
being present and therefore, we have a twofold, ̅2 which is m and the operation of both
two and mirror simultaneously there is a twofold and there is a perpendicular mirror to
the twofold axis.
So, these are the 3 point group symmetries we will consider and the equivalent points
you can write. In fact, we can see here the equivalent points are x y z and x̅ y z̅. So, you
can write it down x y z and x̅ y z̅ as the two equivalent positions. Similarly, in this
particular case you see that the mirror symmetry is notified like this we have seen this
convention. That means, that the mirror is in the associated with the plane of our
projection and therefore, you see both the points x as well as the other point which is
generated by the mirror symmetry which is an open circle. They will fall on top of each
other.
So, in this case considering b as the unique axis, it will be the equivalent points are x y z
x y̅ z, x y z and x y̅ z. It also amounts to an operation of a twofold symmetry followed by
an inversion centre that is exactly what we have done here. We have taken the twofold
175
symmetry and then, we have operated the centre of symmetry on that one. So, suppose
you take an object here which is x, you operate the twofold, you will go over there and
then, you inverted it, then you come back here and that is how we have a overlap of x
and an open circle.
Let me repeat. If you have this point you operate the twofold, you go over there, but then
you do not stop there you invert it. When you invert it, you will come back here with the
open circle and that is what is shown here and that is how you have a mirror symmetry
invoked at this situation. In the last diagram which we have here the 3rd point group
symmetry which is the only other symmetry point group symmetry which can exist with
monoclinic systems.
You see that we have 4 equivalent points two corresponding to the twofold operation and
two corresponding to the corresponding mirrors because there is a mirror which is
perpendicular now to the twofold axis. So, we get 4 equivalent points and you can write
down as a as an exercise all the 4 equivalent points.
Now, we have the point group information on the left, we have the crystal system in the
middle and now we have to see what kind of a lattice this can generate. As you can see
that we have a not equal to b not equal to c so, we have therefore two such situations;
one is a situation where we can have this beta angle. This is the projection diagram of
this where we have projected the unit cell and in this particular case this is the x
direction. So, this is the a cell dimension and this is the c direction which is the z
direction. So, this is the ac projection and in the a c projection you can get a primitive
cell which corresponds to occupancies of the lattice. These are the lattice points these are
not atoms; these are lattice points. So, the lattice points now add up in this projection if
you take it is a projection. So, it is 1/4 times 4 which is equal to 1. So, we will therefore
have one total number of lattice points and therefore, it is a primitive lattice.
So, this is one possibility, the other possibility is because we have the in the
perpendicular direction a rectangular cell, we remember you did the plain lattices and we
found that there are two possible plane lattics, the primitive as well as the centred lattice
which we called as the little c. Now, here also it is possible to have the lattice centre at
this point. That means, you have now the lattice points at 000, this is 001 and this is the
position at 100, 1 unit along x, 0 0 along y and z and this is the position where you have
176
a value of 1 for x and value of 1 for z. This is the z direction and this is the x direction.
So, this will be 101.
So, if you consider these four now defining these four points which are exactly the same
we will have the primitive lattice, but it is also possible to now have another set of two
points as shown here; one in this line and one in associated with that line essentially
there is only one. In fact, and that is removed by half along this direction. So, it is
coming away from the like that it is coming away from the origin and at a distance of
half we will have the presence of a lattice point. So, the lattice points therefore, add up
one these four will give us one and this will be half and half each which will give us one
more.
So, the number of lattice points therefore, is two in case of a C centred lattice. So, these
are the two possible types of lattices we can have in a monitoring system a primitive
lattice and a C centred lattice and this gives rise to several possible space groups and the
possible space groups are also listed here for your convenience. So, these slides will be
of immediate reference value as well if you look at the previous slide you have every
information about triclinic systems you have every information we want about the
monoclinic systems on this slide.
So, if you take for example, the space groups to be primitive the lattice to be primitive
combine it with the point group symmetry to you get P2 and you will get a space group
Pm and P2/m likewise. In all these cases the number of molecules in the unit cell in
general is Z equals 2, Z equals 2 in the case of Pm as well and Z equals 4 as is illustrated
here in the point group diagram. When we come to the case of a C centring we can again
have a C2, Cm and C2/m it. So, happens that when we have to generate the C centring
lattice what we have to do is to every x y z we add a half plus x and a half plus y
considering that it is C centred that is a C direction is centred.
If the a direction is centred then you add half to y and also half to z, if C is the centred
the if B cell direction is centred then you add half to x and half to z. Let me repeat you
add half to x and half to z in case you have a C centred lattice and in case of A centred
lattice you leave out the a that is corresponding to the value of x and add half to y and
half to z and in case you have a B centred lattice add half to x and half to z leave out b.
177
So, this way you generate double the equivalent points. So, when we have a primitive
cell it doubles in the case of a C centred lattice. So, you therefore, have 4, 4 and 8
respectively given the example of C2 the equivalent points in C2 are x y z, -x y -z
because its a twofold operation. Now these are the two operations which are generated
by the point group symmetry the lattice symmetry now generates two other points. So,
this gives us 1/2+x 1/2+y z and 1/2-x 1/2+y -z.
So, you will have 4 equivalent points in case of the C centred lattice. What you have
noticed also as we have discussed this issue of equivalent points is the fact that you are
keeping the positions of x y z and the other equivalent points within the unit cell the way
we have written here for example, if you take x y z you take x y z let us say at this point.
Then if you operate a twofold symmetry on this one it will take you to -x which is the
other side of x right outside of the unit cell, now you will be wondering how you have
four equivalent points in the units.
Let me demonstrate one example here that is you have now the twofold let us say this is
the position of x y z you operate the twofold symmetry the value takes it to minus of x
and minus of z them. So, the twofold operation which is shown here this will now
correspond to x y z and this will now correspond to -x y -z. So, you take this point kept
keep it inside the unit cell let us say we take the primitive cell to start with we have this
point and then it point the other point goes out of the unit cell.
Now, here comes the information which we already have on the translational periodicity
this is the advantage in crystalline materials and that is why we study crystalline
materials. The advantage is the fact that you from here if you add one translation along
the z direction you will come there and add one more translation along the x direction
you will come there. So, you see that the equivalent points now are up here and up there
and they are related by a twofold symmetry. Now, therefore, there is a central centre of
symmetry twofold symmetry created at the centre.
So, that is how you get additional symmetry elements inside the lattice. So, even though
the lattice is primitive; lattice is primitive the symmetry elements will insist on
generating additional twofold positions the lattice is still z equals 1, but you generate
these positions if you have a twofold symmetry operation. In the case of the presence of
the C centring lattice you will have additional points the same logic can be used by using
178
translational periodicity associated with x and z and y we can bring all the four points
inside the unit cell.
So, if you have the this point which is x y z you will have the corresponding x̅ y z̅ up to
here. So, there will be a position which is now here which is also a twofold symmetry.
Now because of the presence of the C centred lattice we have to have if there is an x y z
we will have 1/2+x 1/2+y.
So, the presence of 1/2+ x 1/2+ y and z will create a position up there corresponding to
the value of z notice that minus x y -z will take it there whereas, the 1/2+x then 1/2+z
will take it to the position which is different. So, we will have therefore, 4 equivalent
points after we do unit translations
So, in order to see all the equivalent points inside the unit cell you may have to perform
translation periodicity included operations. So, remember that do not get the way fetched
with the fact that one once you operate the symmetry it goes outside the unit cell, you
can always bring the point inside the unit cell by translational operations which are
allowed only in crystalline materials and that is why we study structures in crystals.
Because we now have the arrested these molecules inside the unit cell which must obey
the conditions we put on based on the space group and the crystal system right.
The other aspect associated with this of course, which we have not discussed fully is the
fact that we have these three space groups, which of the space groups are centre
symmetric and which are non central symmetric needs to be analysed. And that is
something which we will look again carefully because P2 and Pm will generate
equivalent points which are not involving -x -y -z and therefore, they are non central
symmetric systems. And the x y z and x̅ y̅ z̅ a coordinates whenever we get, that
represents the central symmetric system.
So, the threes point groups we have considered or 2, m and 2/m and we have now
discussed the possible space group. So, this has generated 6 space groups triclinic at 2.
So, we therefore, have covered 6 plus 2 total 8 space groups out of the 230 possible
space groups considering all the seven crystal systems all the types of lattices and so on.
It does not mean that as we go along we derive all the 230 space groups that will take
another 300 hours that is not the idea.
179
The idea is to give you an information on how these can be generated and probably we
go through each of the crystal systems and one once we reach a stage where we have
understood the aspects associated with one or two crystal systems the same logic will
apply.
So, whatever logics which have logistics which we have to change from one crystal
system to the other is the one we have to keep in mind. In monoclinic systems we have
another possibility that possibility sorry that possibility is to use additional symmetry
information. The additional symmetry information we have can come from screw axis
and glide planes we already mentioned about screw axis and glide planes these are
symmetry elements with translational periodicity. The screw axis is represented as 2 1 and
the glide plane is represented as a c or n, we will discuss this independently and
individually in the next coming slides.
180
Prof. T.N. Guru Row
Lecture - 15
Translation components in Monoclinic System
Continuing our discussion on the Monoclinic Symmetry.
We now have to examine two other symmetry elements which come due to the presence
of translational periodicity. We have mentioned earlier somewhere in the passing that,
there is possibility of what is called a screw axis motion which is denoted as 2 1 and then
there also, what are known as glide planes. I think we gave some examples from Escher's
diagrams and discuss the possibility of the nature of glide plane and how it acts and so
on. But now we will examine them thoroughly in terms of the monoclinic symmetry.
And one once we understand this symptoms of the monoclinic symmetry it can be
extended to higher symmetry crystal systems and therefore, higher symmetry space
groups.
181
So, let us go to the next slide where I am describing the screw axis as simple as possible.
Let us take for example, the 2 fold symmetry which we are all very familiar with now.
So, if this is the point x y z it undergoes a 180 degree rotation. So, suppose the 2 fold
axis is along the unique axis being y are b then we will have this as -x y -z we pronounce
it has x̅ y z̅. So, we have an x y z we have an x̅ y z̅ related by these two operations at then
it comes back again. So, it is a point group 2. We can invoke the presence of a point
group 21 likewise, where we take this point x y z; now instead of the 2 fold symmetry it
is a twofold screw operation.
So, what we mean by a 2 fold screw operation is that it now we rotate this by 180
degrees and then not stop there, but translate it by half a unit along the y axis which is
the b direction. So, if you have a b c as the unit cell dimensions and b is our unique axis
then the translation is along the b direction and this will give us an equivalent point at
half plus x̅ y z̅. So, this half adds on to y. So, the equivalent points therefore, will be x y z
and maybe I have to use pen now just to indicate it. So, I will use it x y z, x̅ 1/2+y z̅ so,
this is x y z. So, x y z now goes over to x̅ 1/2+y z̅ and now if we operate the 2 fold
rotation again and this x̅ 1/2+y z̅ object that object now will go to another 2 fold
182
rotation. So, it will goes that way, but it will now come to an equivalent point which is x
1+y z.
So, that means, we have now introduced a 1 unit translation along the y direction. So, in
crystal structures in general the monoclinic symmetry systems it does not matter if we
add a one translation as we saw in the case where we want to generate equivalent points
such that they come inside the unit cell we do a 1 unit translation. So, this 1 unit
translation therefore, gets tolerated and so, you have therefore, equivalent points x y z, x̅
1/2+y z̅ which will define a new point groups symmetry 2 1. And this point group
symmetry 21 does not exist in general it exists only if you have a crystalline system with
this possessing a twofold axis.
So, translational periodicity will generate half translation component if you have this 2 1
axis. So, when we say the 21 axis what we do is we rotate by 180 degrees and the
translation is 1/2. So, it is a half translation. So, the translation here is 1/2 half the unit
cell about which the axis of rotation is located. It may be in the y direction, it may be in
the z direction it may be in the x direction. So, suppose what happens here if it is in the z
direction. Suppose it is in the z direction this diagram is given so, this is x y. So, if z axis
is the unique axis what would be the angle between x and y the angle between x and y
will be what?
So, the angle between x and y, a and b so, the angle will be by convention γ. So, this
angle therefore, is the γ angle and so, we have 2 fold rotation; the 2 fold rotation can
come back. So, this is the representation of a non unique axis operation of the 2 fold in a
monoclinic symmetry. Now, what happens if the non unique axis has a screw system. So,
it is now possesses a screw periodicity. Then we have a screw point group symmetric,
then we have this plus z and the 2 fold rotation takes it up there and we add a half to the
z value. So, this is now this 1/2+ z and then we operate this again and add another half to
z it will come 1+z, but 1+z is same as plus z.
So, therefore, we get two equivalent points corresponding to 2 1 screw operation and
therefore, the number of objects that get generated either by the 2 fold rotation or by the
183
21 screw axis will be 2. So, the same object now generates the screw axis operation. So, it
is like for example, driving a screw through a wall. So, you have the screw and you want
to put it through the wall what you do is you rotate with a screwdriver. So, it can be
designed in such a way that one rotation of the screw will generate half unit translation in
the given direction. So, if you rotate the screw in the opposite direction it will come out
by half the unit.
So, you can do this operation back and forth and therefore, that is why this screw axis
operation is a valid symmetry operation and so, we get the equivalent points x y z, x̅
1/2+y z in the case of 21 axis and this is how we describe the screw axis and this is the
only possible additional symmetry element which can come in monoclinic system. So,
the monoclinic system has we saw before has the 2 fold the 2̅ and the 2/ m symmetry and
therefore, now two can have a 21 symmetry. So, we will have therefore, a screw axis
which is replacing the 2 fold here that will give us the new point group and a new point
group can be present here as well it could be 2 1/m; that means, a 21 operation with a
mirror perpendicular to that that is also a possible operation of the point group symmetry.
So, additional symmetry elements means additional space groups that two Bravais
lattices P and C will both see these additional space groups and therefore, we will have
space groups like P21and C21, P21/m and C21/m but we will later on show that C21/m is
nothing, but C2/m. So, that way we will have a control on the number of space groups
which we can generate, infact I have listed it somewhere later on in one of this slides. So,
this screw axis operation I think is now understood fully.
184
We will go to the next slide which now brings in the issue associated with glide planes.
This slide tells us a few more extra information than what we require, but this is the all
the information which we can give on a glide operations.
So, here is for example, an indication of a mirror symmetry. The mirror symmetry m we
have already seen that is the ̅2 operation and it has no translation. On the for example, if
you have a have an object here, this object will undergo a rotation sorry reflection across
the mirror, the mirror is perpendicular. So, you will get this operation taking the object
on to the inverted image of the object this is like you look through a mirror when you
look through a mirror or your face will appear of course, right becomes left, left becomes
right. So, you know that is the enantiomorphous behavior of objects.
So, it is like taking the left hand and trying to fit the right hand. These two are
enantiomorphous objects with respect to each other because the left hand and right hand
combination will now we give as an identity operation between this and that which
happens to be the enantiomorphous relationship. So, similarly the objects related by So,
if this is x and if this is the reflection along the y direction then you have x y̅ z which will
185
generate that. Now what I have shown here is one unit translation. So, this is your unit
translation along the which axis? along this is the reflection along the y.
So, it has to be either a or c. So, either way we have therefore, the next object which will
also appear to have the mirror symmetry. So, the mirror symmetry operation is one unit
translation also will appear. So, this is how we will be generating the objects when
mirror symmetry operates. So, having made sure that this is clear enough for us. So, we
have x y z, x y̅ z, x y z, x y̅ z ok. Now, we can have two types of possible operations and
I have shown you purposely the operation of the b glide. The reason why I gave the b
glide here is for a reason, the b glide can you know the mirror has to be perpendicular to
the axis of symmetry.
The axis of symmetry let us say is y that is the b axis, if the b axis is the axis of rotation
then the perpendicular direction to b will have either a or c. On the other hand, if I show
the b glide what it essentially mean is that it has to be the unique axis has to be either
along a or along c. So, this therefore, is the non unique access representation. I purposely
show this because you might come across situations where you had to deal with non
unique axis information. So, when such an information is required therefore, it is better
to know about the b glide, it does not matter whether it is a glide or a b glide or a c glide
it is along the axis. The only thing which we have to remember is that there are mirror
reflection occur perpendicular to the axis of rotation.
So, the axis of rotation either is with a or with b or with c. When we talk about
monoclinic systems the axis of rotation is always b unless otherwise specified because
we take unique axis as the b axis. So, here what you see is therefore, x y z now it
undergoes a mirror reflection mirror and after the mirror it translates by half the unit in
this case along the b direction. So, this is b/2 representing the b glide ok. So, the b glide
is therefore, can be perpendicular either to a or to c. So, the translation now is associated
with half the distance of b. Now we operate this glide again a mirror symmetry and add
half to that.
186
So, the half is in the direction along the line parallel to the mirror. So, the half
translation we add is in the line parallel to the mirror. So, your axis of rotation is
perpendicular to the plane and that is why we say it is gliding down that particular
direction and that is why this operation is referred to as glide operation.
If you go back to the very first class or the second class you will see that we did discuss
about this in terms of Escher’s diagram and you can go back and check the Escher’s
diagram again and see how this operation is coming up, the glide operation there it is
very clearly indicated with respect to an object. So, systems which will have the glide
plane operation will also now fall into the definition of our required space group
information because of the fact that we can now have one translation unit and that
translation is in the periodicity repetition of the object in the crystal and therefore, since
it is a crystal we can have this kind of operation. We can also have the operation because
we now referred to a plane and the axis is now perpendicular to a plane like in this
particular case the axis is coming towards you and a mirror operation is perpendicular to
that.
So, when we take such a situation, then the possibility is that it can also go along the
diagonal direction. So, there are three types of diagonal glides that are possible (a+b)/2,
(b+c)/2, (a+c)/2. Now my suggestion is that, you draw similar diagrams as I have drawn
on the right side to see an n glide even though I have given an n glide, I would like you
to mark the axis about which the n glide operation takes place. Suppose I show an n glide
like this where we have the operation taking it half plus , now it is half plus means that
we not only have the half translation in this direction we also have a translation in the
direction perpendicular to that. But in the plane associated with the plane you see this is
the plane.
So we have the translation we go half the translation along the plane direction
perpendicular to the plane we go. So, effectively we are going along the diagonal. So,
you have this translation half unit and then we go half above which is now parallel to this
plane. So, the plane is like this so, we have two translations one down here and another
going up there. So, effectively you have this half plus also coming in here. This half plus
187
tells us that it is an n glide and the notation for the n glide is line and a dot that we know
already from our early classes. So, therefore, we have again a translational periodicity
restored by this operation.
So, for the glide planes therefore, we can have either in an a glide or a b glide or a c
glide depending upon the unique axis definition in monoclinic, but if we go to
orthorhombic and higher systems we do have this all possibilities of a, b and c. The only
thing we should remember is the b glide cannot be perpendicular to the b axis, a glide
cannot be perpendicular a axis, c glide cannot be perpendicular to the c axis. There is a
special case of (a+b+c)/2 which is indicated with a star here that will occur only in cubic
systems because in a cubic system and probably it can occur in rhombohedral systems as
well.
In rhombohedral and cubic systems, the direction of the axis is not along either x or y or
z and as a result we can have this kind of glide planes which are very unique they do
exist and they create some different kinds of space groups, but we are not going to deal
with them because they become reasonably complicated for this course, but we should
know that such a thing also exists. The glides can be along all three directions because
your definition of the glide plan is only with respect to half translation and you are a, b
and c are not necessarily the axis in case of rhombohedral and the cubic systems.
There is another thing called the diamond glide which will occur in tetragonal systems
and this could having again 1/4 of translation and this will combine both a and b
directions. Again we were not going to discuss this because this probably generates only
one or two additional space group among the 230. So, we are not as I mentioned already
not going to study all the 230 space groups that is not our intension, but we should see
how this space groups develops use of this combinations which we are now trying to
attempt. For example, we have already seen the types of space groups which can occur
from non translational point group symmetries associated with monoclinic watch my
words non translational point groups; that means, there are no translational symmetry
associated with those point groups. So, got this P2, Pm, P2/m, C2, Cm, C2/m six of
them.
188
Now, we are associating the 2 1 screw axis and also the a glide, c glide and n glide and
therefore, we will get additional space groups. We will see how many additional space
groups we get and what kind of properties they will have. We will then formally study
space groups in later classes that is not the, it is not going to be the end of the show here.
We will just see how many space groups are coming, but how the space groups really
relate and get to the equivalent points and then make sure that these equivalent points
which we have generated form a realistic definition of a space group that we will see in
the later classes. But at this particular we just see how many space groups are possible
with the addition of the translation involved symmetry point groups which are involving
21 screw a, c and n and that will tell us something else.
Now, this is a slide taken from a book. So, it looks ugly, but I think you can live with it
because we just took a scanned copy of it. So, here you will see all issues associated with
monoclinic systems. So, we have taken only the monoclinic systems and their
corresponding equivalent points of course, diagonal glides are also given here for that
applies only for the tetragonal system. So, we will not worry so much about the d glide,
but all other glides will apply to monoclinic as well as orthorhombic systems.
Right now we will discuss it with respect to monoclinic because we understood only up
to monoclinic we have not discussed the orthorhombic as yet, but these are the
equivalent points which will come to your rescue and tell you how the group symmetry
189
has developed. For example, if you can axis 2 which is parallel to a by parallel to a I
mean along the a direction or along the x axis. So, if there is a 2 fold along the x axis
what happens, the x axis value will not change. So, you will have x y z, x y̅ z̅. So, the y
and z will change. Similarly if it is along the b direction then we know then it will not
change along the y axis the y value will not change.
Similarly, the z value will not change for the c direction. So, having notice that see for
these equivalent points which are operating on the 2 fold, that 21 axis can be
perpendicular sorry it can be along a , can be along b and also can be along c. So, if the
21 axis is along a we have x y z and wherever the 2 1 screw operates that would be the
direction where we have to add a half. So, for example, if you have x y z we add a half to
x if it is along the a direction half we add along the y if it is along the b directions. So,
for a unique axis monoclinic system the equivalent points will be x y z, x̅ y+1/2 z̅.
So, as a consequence we now see the possibility of existence of 21 axis in all three
direction. At this moment since we are discussing only monoclinic symmetry it will be
either along a or b or c if it is a long b that is the convention which we use for monoclinic
systems. Likewise, now we go over to the plane, in case there is a mirror plane
perpendicular to a. So, when we describe the mirror plane it is either perpendicular or
normal. In fact, this table is a little funny because if you say parallel here we can say
perpendicular , but if we say along there then we can say the plane is normal to the a b or
c.
So we can say twofold is along a b or c and the plane mirror plane is normal to a b or c or
we can say 2 fold parallel to a as is written here and perpendicular to a as is written here.
This is just the way in which you can talk about the existence of 2 fold and mirror
different text books give different notations. So, I am telling you both. So, many of the
textbook say m the mirror plane is normal to a then you will start wondering what is this
normal to a it is nothing, but perpendicular to a.
So, what happens is only the value of the coordinate associated with this axis a b or c
will change. So, x y z will go to x̅ y z, x y z will go to x y̅ z and x y z will go to x y z̅. If it
is a glide the a glide as I mentioned cannot be perpendicular to a it can only be
perpendicular to either b or c. So, depending upon to which direction it is perpendicular
190
that perpendicular direction will change sign and the addition will be always with respect
to the a glide.
What I mean to say here is if you take x y z, you see that you have an a glide
perpendicular to the b axis then, you are a value will be x+1/2 the equivalent point and
then y will change to y and z will remain the same. If it is along the c directions this z
will change the sign, but the addition is always with respect to the axis about which the
glide plane occurs. Similarly the b glide and the c glide are illustrated here. You notice
for the n glide perpendicular to a there will be additions both along y as well as the z
directions because now the n glide can be since it is perpendicular to a the translations
will be in directions perpendicular to the axis of that axis.
So, since n glide is perpendicular to a, the translation components can be along both b as
well as c. So, b and c will get the additions of y+1/2 and z+1/2. So, that is how you can
evaluate the nature of the glides as well as the nature of the 2 1 screw axis. The diamond
glide is little more complicated I will not go on and describe this now and I do not think
we need to worry about it at this moment at the level of the course we are offering now.
These will of course, generation additional space groups in very few crystal systems
cubic probably and tetragonal to some extent other crystal systems will not have the
diamond glides. So, as a result very few space group if any day you come across any of
those space groups you can specially analyse them and see whether you can get
information on this.
191
So, what happens with the presence of the 21 screw and with the presence of the glide
plane you get additional space groups. Now, the additional space groups you will get
from the P21 or with 21 screw axis or P21you can also get C21 we will later on show it is
nothing, but C2 there is nothing called C21, we will show that in terms of derivation of
the equivalent points which is probably in the next class. And then we have the space
group P21/m and since 21 cannot exist with the c glide we will have C2/m. So, these are
the additional space groups over and above the six space groups we saw which have no
translational component. Here the space groups have translational components.
The additional space groups coming from glides plus combining with screw axis are
given here, I want you to think for a moment and see whether something is missing here.
I say that something is missing here we have missed a few space groups and those space
groups involve what? They involve the diagonals glides. So, I am going to write them so,
I am trying to see whether I can write them I will get the pen across here and write them
down here.
So, it will be now Pn and P2/n and P21/n these are additional space groups. So, we can
have the combination of the screw axis as well as the glide plains as you can see here and
along with that we have the P2/n and Pn; Pn will be similar to Pc accept that now the
translation are along two different directions x as well as z, assuming y axis is the unique
one, then we have the in addition to that can you have Cn can you have C2/n. Can you
192
have C2/c we have to add those we have already add C2/c here, can you have C2/n.
These are the things which you have to explore yourself at home and you will see that
those possibilities are already incorporated into the space group here accept that it
depends upon how you do the additions.
See as I as we saw here that C21 can be shown equal to C2 the operations with the n glide
with the c centering is an issue which we can reduce it to a translation along the c
direction by doing some changes in the a b c directions that is a complicated issue either
we will do it later as an assignment part or we will see how we can show. In fact, we can
also do this as an assignment part later during the course. So, C2/m equal to C2, C2/m
equals these two issues we can take care as we go along.
So, we see that we have the 21 screw axis shown here. Now the glides and the screw axis
can occur in other crystal systems also we have so, far considered the monoclinic system
only of course, triclinic cannot have any of these systems, but when once we have the
monoclinic system which is associated with a 2 fold axis or a mirror perpendicular to the
2 fold axis we can have these additional translational component associated either with
the twofold operation or with the mirror operation and therefore, we will have these.
So; obviously, the other space groups like or other crystal systems orthorhombic and
higher symmetry, the moment we talk about higher symmetry we will have higher
symmetry information in them. So, we can all the threefold rotation, we can have
fourfold rotation, we can have six fold rotation then; obviously, we will have
translational components associated with them. And those translational components
associated with them will give us the information about the possibility of having 3 1 and
32. Now you notice that the three fold axis can be 1/3 rotation as well as 2/3 rotation.
So, 1/3 is when we the screw, now the screw is let us say a threefold geometry then we
can have one turn which gives you one third , second turn will give you two third, the
third turn will give you the full turn. Similarly, the 4 1, 42, 43 and 61,62, 63, 64, 65. So, all
these possible rotations can be located and therefore, we have shown that on the right
side I have shown it on the right side where you see the type of operations we have
discussed it already once.
So, these will now they was the 31, 32, 41, 42, 43 the equivalent points which get generated
by these operations or also indicated here. So, effectively we have covered the possibility
193
of glides and screw axis and we will have them in trigonal, tetragonal, hexagonal
systems and can we have it in a cubic system I have asked this question to you. So, that
you can think about how to do this cubic operation and how the cubic system can have
these kinds of operations.
You can think of a situation where you have a 3 fold axis in a cube. So obviously, the
possibility of 31, 32 in a cube is indicated. You also have the 4 fold rotation in a cube. So,
those are also indicated, but the fact that it is a cube has certain properties. So, which
property of the crystal system dominates and which does not dominate is very important.
So, far that is how for example, when we talked about this C21 and C2, it is the C
operation which dominates.
So, when once we have a C centered lattice we will take that she centered lattice as the
main concern for generating equivalent points. So, you will have x y z, ½+x ½+y z. Any
other component which involve translations in either of these two x or y is secondary.
So, therefore, secondary interaction can be ignored because it is already built in to the
primary interaction.
So, when we say a space group is let us say Cc the operation of this capital C is more
crucial than the operation of the small c. That is how for example, we will not have a
glide associated with C centering and that is where we see that the importance of the
dominance of the lattices type comes first. This is something I thought you should notice
before we go further.
194
Department of Chemistry
Lecture – 16
Additional Symmetry Elements
So, the presence of glides and screw axes in other crystal systems can occur. We have
the trigonal axis; we have the tetragonal axis and then the hexagonal axis. So, these three
axes correspondingly can have the screw translations symmetry.
So, the screw translations can be one-third, two-third with respect to 3 fold axes, 1/4, 2/4
and 3/4 with respect to the 4 fold axes and so on. So, that means the trigonal tetragonal
and hexagonal systems we will also have the presence of glide planes and screw axes
that is shown here the screw axes are shown here. And the corresponding symmetry
informations which we have discussed before as well are appearing here.
Now, as I said in the last part a few minutes ago we have to see what happens with the
cubic system I already gave a hint that it could be along the 3 fold direction. So, we
could have the cubic symmetry, but then I was discussing about the hierarchy of
operations. The hierarchy of operations comes in such a way that we talk about the
nature of the latice first and then we talk about the presence of the screw axes on the
glide planes.
195
This will have a very important repercussion when we see the x-ray diffraction patterns,
because the x-ray diffraction patterns we analyze with respect to the various values we
see from the presence of these reflections. And the presence or absence of these
reflections is governed by the way in which this space group develops. And therefore,
some of these elements the presence of the C centering, the presence of the I centering
we will introduce what are known as systematic absences.
And so, also the presence of the 2 1 screw axes on the mirror on the glide planes, but the
screw axes on the glide planes we will affect only the so called projection reflections
whereas, the general reflections are affected by the centering operation. And that is why
the priority is given to the lattice type and then the 2 fold and the translation involved
symmetries like 21 and n and so on, right.
We go to the next slide where we in shift the system now to orthorhombic. Orthorhombic
is more or less a expansion of a monoclinic discussion we had so far to three dimensions.
This is very straight forward, because in case of the monoclinic system we have a single
2 fold axes whereas, in case of the orthorhombic symmetry. We have three 2 fold axes
intersecting at a point, and these three 2 fold axes come at 90 degrees with respect to
each other.
So, because of the fact that whenever there is a 2 fold axes two of the angles should be
90 degrees. If there is another 2 fold axes perpendicular to the given 2 fold axes, the
196
other two angles will be 90 degrees the consequence of which is all 3 angles should be
90 degrees and that is what happens here α, β, γ are equal to 90 degrees. So, a ≠ b ≠ c, α,
β, γ are 90 degrees defines the so called orthorhombic system.
The way in which these three 2 fold axes are disposed with respect to each other is
illustrated here, as you can see this is the intersection point, so you have along the x
direction, you have along the y direction, you also have it along the z direction. The
minimum symmetry therefore, is 222 and that is the point group symmetry 222 which is
described here. So, we have a 2 fold axes associated with direction coming towards us
and as a consequence we have a 2 fold axes in plane and there are two 2 fold axes
intersecting with each other in plane and therefore, this would be the illustration of that.
So, you also have this circle a little darker than necessary to show the 2 fold, 2 fold, 2
fold there are three 2 fold intersections. So, suppose there is an object here it sees the 2
fold up here and goes over to that. And after having gone to that position it sees the
presence of the 2 fold, but then this 2 fold is in the back side of it which is at an angle of
90 degrees with respect to that and as a result it goes over to this position. And this now
since the 2 fold here and comes back to this position and vice versa. So, it keeps
therefore, 4 equivalent points and those 4 equivalent points now you are experts you can
write them down yourselves the 4 equivalent points were 222.
You can also write the equivalent points for the next set of symmetry elements because
we have been discussing this issue many times already, so maybe I should just write ones
to remind you what we are talking about. So, here this now represents a PPP axis; here it
represents a II sorry not primitive proper; proper; proper rotation axis improper;
improper correction; it is not primitive proper, proper, proper rotation axis improper,
improper, proper rotation axis and then the last one which is improper, improper, proper.
A combination of three mmm will represent the 3.
Now, how does mmm come about? Mmm comes not about 3 improper rotations because
we already showed by Euler’s theorem that 3 improper rotations will not go in
conjugation. So, mmm therefore, is not mmm, but it is actually 2/m 2/m 2/m So, it is a
combination of a proper and an improper rotation that is the point group which is which
we have discussed before. So, instead of writing 2/m 2/m 2/m consistently we can just
make it simplified to mmm. So, please note that mmm did not come from 3 improper
197
rotations combining with each other than you would be violating the Euler’s theorem.
So, therefore, it is 2/m 2/m 2/m.
So, whenever you see more than 3 mirror combinations coming, like for example, later
you will see in the tetragonal system 4/mmm, it actually means it is 4/m 2/m 2/m. So, if
you remember that carefully in fact, it is a part and parcel of your understanding this
whole course then you will never make a mistake ok.
So, 3 proper rotations 222, now will take us to the point group symmetry 222 and now
we have to see how many lattices which we have to consider combining 222 with. In the
case of the orthorhombic system we have all possibilities, we have the primitive one, we
have the body centered one where we have at the center of the body half position. So,
please count the number of lattice points for every one of these systems. In case of the
primitive system it is 1, in case of the body center it is 2. What happens in case of the F
center? The in case of the F center we have 1 2 3 4 and then two half ones 5 and two half
ones 6. So, totally how many are there? 1 corresponding to the edges and 2 more
corresponding to the centering with respect center of these axis and 1 at the center. So,
totally it will be 4.
So, the number of lattice points for a F centered orthorhombic lattice will be 4 and of
course, we can have A or B or C separately centered like in the a monatomic system and
then we will have therefore, the C centering. So, C centering means also it could be a
centered it could be B centered because it just depends upon how you define the values
of a, b and c. You see in the case of an orthorhombic system you can always interchange
a, b and c. So, when you interchange a, b and c the correspondingly the A centering, B
centering, C centering will change. So, they are one and the same. So, therefore, we have
4 types of lattices with which we have to combine the 3 point groups. So, 4 lattices 3
point groups with no translational symmetry associated with them.
And therefore, you have generating system like P222, Pmm2, Pmmm. So, these are the 3
crystal systems which you can generate with P. With I again you can generate these
three, with F again you can generate these three and with C centering also you can
generate these three and you can keep on doing that. So, you see the number of space
groups will increase considerably now. We had something like 17 space groups or so on
till now it is adding up considerable number.
198
So, orthorhombic in fact, is one of those systems which has the highest number of
possible space groups because it has the combinations of P, I centered lattices, F
centered lattices as well as A or B or C centered lattices. So, the possibility of a molecule
to crystallize in orthorhombic symmetry must be the maximum and so the possibility of
the occurrence of orthorhombic symmetry in overall crystal systems if one analyzes the
database organic orthorhombic systems are many more compared to the rest of the
crystal systems. It does not mean that orthorhombic is the preferred crystal system. It so
happens that it allows for several possible space groups. This is the point which you
should remember.
So, point group symmetry as well as the lattice symmetry combinations therefore, give
us the space group that you are very clear now. Now, we also have possibilities of 2 fold
screw axes. The 2 fold screw axes now is the with respect to a with respect to b with
respect to c and therefore, we have glide planes with respect to all of a b c. We can also
have combinations of screw axes and glides. So, you see now that is the 3 proper axes
can take for example, screw axes.
So that means, we can have a space group like P2122, P21212. We can also have a space
group like P212121, in fact, all 3 are possible and it depends upon how the molecule
decides to pack itself inside the unit self. So, we can therefore, invoke the in insertion of
the 2 fold screw axes by getting space groups P2122, P21212 and P212121. P212121 is the
most favorite space group for organic molecules particularly if they carry an asymmetric
carbon and that is because it allows for the maximum flexibility in the system.
When we study the space groups in detail we will derive those equivalent points. What
are the equivalent points that can come for the P212121 and see what I mean by that. You
remember this class and then we will discuss P212121 when we discuss this space group I
will make clear why there is so much adjustment space available for rather flexible
functional organic functional molecules organic molecules.
Glide planes; glide planes again you know we can have with the primitive lattice. We
have three 2 fold axes perpendicular to each other. So, we can have all 3 glides a, b and
c. So, we can have combinations of the 2 fold axes with the glides like P2ab and things
like that. We can also have all glide combinations like Pnab or Pbca and so on. So, you
199
see the number of possible space groups we can generate in a case of an orthorhombic
system increases enormously.
And so, this is I am talking only with respect to the primitive lattice, we can go to the I
centered lattice, we can go to the F centered lattice, we can go to the C centered lattice.
For example, we can have a space group like C2ab. So, we may all these space groups
now become possible and now you see the flexibility that is available in the solid state.
Until now we were talking about the restrictions available in the solids. We were talking
about the fact that translational periodicity locks the molecules into their respective
positions, but now the way in which the space groups can manifest themselves as we go
further into the higher symmetry space groups starting from triclinic monoclinic and now
in orthorhombic we see there is a now considerable amount of flexibility. So, the
molecules are very happy now, because the molecules now can crystallize into their
choice space groups.
In fact, the molecules can have more than one choice space group and that is where the
issue of polymorphism comes. And this is an issue which is a big issue for pharma
industries; we will discuss that towards the last few classes. Now, the combination of the
screw and glide also is very critical, which screw can combine with which glide plane.
Of course, we have to keep our Euler theorem in in mind because we cannot have
different kinds of combinations which we will give rise to the IIP for example is a
possibility, but anything beyond that is not a possibility and therefore, we have to
therefore, have something like Pmm2, Pmca for example and things like that which will
now correspond to the mmm point group symmetry.
Pmca is mmm point group symmetry which means it corresponds to 2/m 2/m 2/m and
when we say Pmca a it will be 2/m 2/c 2/a ok. So, the glide planes will always invoke the
presence of 222, so that is why the minimum symmetry you should have in an
orthorhombic system is 222. Unless you have three 2 fold axes intersecting with each
other in general you will not be able to generate orthorhombic systems. And so those are
the restrictions all, right, but the flexibility is immense, so using the immense number of
flexibilities the solid systems or in such systems in solid state can arrange themselves.
200
But there is always systematics which is associated with these arrangements and that is
where we have an advantage to identify and locate the positions of the atoms in any
given crystal system and in any given space group.
Now, comes the next set of crystal systems, we go to the tetragonal system. In the
tetragonal system of course, there are several possibilities, one of them is illustrated here
this you will so called 4 fold symmetry. That is also illustrated here. Remember that in a
tetragonal system we have a equals b not equal to c alpha beta gamma are 90 degrees
with respect to each other. The moment we have these three angles at 90 degrees to it
each other we invoke automatically two intersecting 2 fold axes, in addition to the
presence of the 4 fold axes.
The presence of the 4 fold axes generated will make the axes unique with respect and we
fix that as the c direction. So, we fix the c direction to be the unique dire axis direction
and that is why it is written here Z. So, we take this to be along the 4 fold axes and then
the innumerable number of 2 fold axes can be present here. But the simplest point group
we can consider is the 4 fold just the 4 fold which takes the objects into 4 different
positions as a as illustrated here along with the 4 fold notation which we have in here.
So, this generates 4 equivalent points.
Having generated 4 equivalent points we can also generate 4 equivalent points, if we take
the combination of the 4 fold symmetry which the presence of 4 1, 42, 43. So, if you take
201
for example, 41 the from this point to that point we will have a one-fourth translation
added, from this to this we will have a two-fourth translation added, from this to this
point we will have a three-fourth translation added.
On the other hand, if you have P43, from this point to that point we will have a three-
fourth addition added this to this it will be two-fourth addition added and then in this to
this will be one-fourth. So, the handedness associated with this we will change, in one
case it is going this way this screw axes in the other case it goes rather way. So, this
screw axes operations P41 and P43 are opposite of each other. If you consider P42 it is
equivalent of if equivalent of a 21 screw axes effectively. So, it is just a single operation
it is dichotomy in that particular case whereas, in case of 4 1 and 43 they go on opposite
directions the screw operation.
If you look at the possible Bravais lattices associated with the tetragonal system, one can
show P and I only exist which probably we will do in the form of an assignment later.
But if you consider the two lattices that are allowed only P and I our load. Any other
lattice like F for example or A B C center, because of the fact that we have a equals b as
an additional condition which makes the basal plane square, the moment there is a square
we cannot have it centered effectively and therefore, the centering of the phases is out.
So, whenever you have a square base you go back to your previous classes where we
discussed the 5 plane symmetries, in that we had a squares symmetry the that is the little
p, the primitive lattice in the in the 4 fold symmetry. There we cannot have a center
because if we have a center you can generate another 4 fold there and so it is degenerate.
So, whenever you have a is equal to b you cannot have that phase centered. So, because
this phase cannot be centered, that phase cannot be centered, likewise these are at 90
degrees with respect to each other. So, the 90 degree angles can also be space centered.
Unless you have these 4 centered we cannot have the other two phases centered and
therefore, F is ruled out. This is the logic which I am trying to give you. We can one can
work out work it out mathematically, but I thought that logic is good enough to for our
current discussion. So, we have the primitive lattice and the body centered lattice where
you will have a additional lattice point at half. So, the z=1, sorry the lattice number;
number of lattice points is 1 here, number of lattice points is 2 there.
202
Because of the fact that a is equal to b we will also not allow one single phase to be
centered. So, A B and C separately being centered is also ruled out and therefore, we
have only two possible lattices in the tetragonal system. So, the job now is to combine
these lattices with the possible point groups the first point group is with 4, so all the
possible combinations are shown here. The number of equivalent points initially is in this
cases of primitive lattice is 4 whereas, in case of the body centered lattice it will be
double that, so we will have z equals 8. So, as you all know that we had a half along x,
half along y, half along z.
So, having seen the possible space groups here and the I centering is adding this we see
if by an example of I4 is given here. Now, the type of rotation axis which we just now
discussed can be described here 41, 42, 43, you see that exactly what I talked about 41 if
you start, 43 if you start from this point that point will be three-fourth added and this
point will be one-fourth added. So, rotate if you now look at the 4 3, rotation it is this way
the left handed rotation and in this particular case it is the other way around this is a
anticlockwise and this is the clockwise rotation.
So, 41 is anticlockwise, 43 is clockwise rotation and that is also illustrated by this

combination here. 42 automatically generates only the half point, so there is no
distinction between one direction and the other. This will be more interesting to discuss
when we go to the trigonal system, but the trigonal system is an issue which we will see
later if time permits we will go into it, because we have to cover a lot of other things in
this course. So, we may not specially discuss that. Except that I would like when we go
to the trigonal system, I will mention it of course. So, let us see what other possible point
grooves that can occur with the tetragonal system, that is the next slide.
203
Next slide we will take us to 4/m. So, here it is written tetrad that essentially means a 4
fold rotation. This is from a text book where they use tetrad and dyad for a 2 fold and so
this has been taken from the textbook. But we will say this is the 4 fold. The 4 fold can
be proper and are of inversion that means, proper or improper axis along the z direction
and that we specify it first. So, the 4 fold is rotation is along the z direction all the time,
and the mirror is now going to be perpendicular to that operation. So, if you see here the
mirror is indicated perpendicular to the 4 fold.
So, the presence of the 2 fold axes is immaterial its 422. So, in this particular case the
system can be just 4 by m the other 2 fold axes will be automatically implied when we
derive the so called equivalent points. For example, here I4 is shown to have x y z, x̅ y z,
y̅ x z and y x̅ z that will be the 4 fold operation and along with the 4 fold operation if we
add half plus half plus half plus to each of x y and z we will generate 4 more points. So,
there will be 8 equivalent points for I4.
The possible space groups with 4/m combination however, are P4/m, P42/m, P4/n, P42/n,
I4/m, I41/a and there are additional ones which you can add depending upon the nature of
the rotation axis remember we can have both 41 42 and 43. So, what you see here is only
the 42 ones given 41 and 43 are also possibilities. Now, when 41 and 42 are considered
possibilities there is probably not a possibility of an n glide. So, we have to work out
why only these are given and that is something which is which could probably be taken
204
as a home work by you guys so that you can have a look at it and then see how it all
depends upon.
So, there is a 2 fold axes proper or improper along the x that also see. When we describe
the tetragonal system since a is equal to b, the two directions are degenerated so that
means, a can be b, b can be a. And because of this degeneracy issue we is represent only
one of the axis to be associated let us say with the a direction.
So, on once we associate that with the a direction the second 2 fold axes will be in a
direction which is perpendicular to 110 which is along 110. See, the directions and
crystal systems are normally given by 110, in fact we are going to discuss that in the
coming classes or in an assignment later on. So, 110 if we show it in square brackets it
represents a crystallographic direction. So, 110 is a direction which is along at one unit
along a and one unit along b, so 110 therefore, is the direction about which the 2 fold
axes gets a represented in this case. So, we can have 4/m and the possible space groups
are written here and this is an illustration which shows how the mirror plane can be
perpendicular to the 4 fold axes.
Having seen this, I think we can go to the further discussion on the P4 have space groups
possible space groups are listed here you see many number will develop because of the
presence of the 4 fold axes proper and or inversion because we can have 4 we can have ͞4
we can have 41 43 and so on 42 43.
205
Now, the presence of the ͞4 system is interesting because we already saw in the previous
days 4/m. We also saw 4, but we have not seen the ͞4 operation. So, the ͞4 operation will
also introduced will have the possibility of proper and this is an inversion axis on the 4
fold symmetry. So that means, in the z direction we can have not just the 4 fold rotation
alone we can also have the inversion associated with the 4 fold rotation. So, that
generates the equivalent points accordingly.
I do not think there is enough time for us or space for the us to discuss that in this
particular course how to get the equivalent points and so on. So, what we will therefore,
do is we will just illustrate as a particular space group like 422 let us say. Now, this
space group 422 has additional 2 fold symmetries. So, you just compare this diagram
with the diagram we have here this has only one 4 fold symmetry and that is an allowed
point group symmetry in a tetragonal system.
That means, there is no symmetry which is imposed in the direction perpendicular to the
z-axis. So that means, the x and y even though there is the 2 fold axes is implied in them
there it is not specifically mentioned because in this particular case when the mention is
made of 422 you see the number of equivalent points we will increase from 4 it will go
to 8 because there are additional 2 fold symmetries which occur due to the 4 fold
operation.
So, this 4 fold operation we will take 4 of these 2 fold screws and the between these two
screws we develop additional 2 fold screw operations. So, the number of available
symmetry operations we will start increasing in fact, they will multiply. So, as a
consequence you will end up with several possible space groups written like this and
therefore, tetragonal has also a very large number of a space the space groups that can
gen be generated.
206
Trigonal still, tetragonal still continuous to have different kinds of symmetry

combinations because we have not discussed the proper improper, improper axis
associated with it. So, this is the one which I mentioned earlier 4/m 2/m 2/m which is it
can also be written as 4/mmm and that is the maximum symmetry we can have in a
tetragonal system.
So, the tetragonal system all therefore, has a combination of point groups which are 1 2
here, 3, 4, 5 and 6. So, there are 6 possible tetragonal combinations 4, 4/m, ͞42m, 422,
4mm and 4/m 2/m 2/m. So, all these points groups combinations can be now combined
with the possible Bravais lattices which are two P and I centered. So, therefore, we get a
plethora of possible space groups which we can generate in a tetragonal system.
207
Now, we go to the issue of a trigonal system. We will discuss a little bit on the trigonal
system at this point. The trigonal system is just like we have a Z-axis representation of
the 4 fold we have a now a Z-axis representation of the 3 fold. So, the 3 fold now is
associated with the Z-axis.
I purposely wrote this “ulta” arrow (opposite) to show that we can either have it in this
direction or in that direction it is immaterial. So, it so happens that people generally write
that and people think it is only along that direction we should have the 3 fold. So, I
thought I will write it opposite you show that the 3 fold rotation axis can also be in the
opposite direction.
So, we can therefore, have only these point two combinations we can have 3 fold which
is just generating these three and then 3m which generates 6 of these because it is 3
followed by a mirror. Notice here it is not a ͞3m, it is 3 followed by a mirror operation.
So, first we operate the 3 fold and then mirror invert it. And ͞3 operation which is again
an operation which takes the 3 fold object on to an inverted 3 fold object. So, you see
that there is a symmetry between the 3 open circles and a symmetry between the 3 closed
circles.
So, consequently the 3 fold operations generate different kinds of equivalent points
compared to the so far discussed in terms of tetragonal and orthorhombic. But what is the
most important point is to consider the direction of rotation axis, so the direction of
208
rotation axis when we refer to the point group symmetry and eventually in the space
group is taken to be the z-axis. So, the z-axis representation of the presence of the 3 fold
is a consequence of the presence of the lattice. So, when we say the 3 fold is along the z-
axis we are already implying that there is a lattice corresponding to that. And how many
lattices are possible in the trigonal symmetry? We will take it up in our next discussion.
So, at this point we can sort of conclude that the 3 fold, the 4 fold or along the z direction
in case of the tetragonal and the trigonal symmetry and the tetragonal symmetry has P
and I and with the I we will get an additional set of points added with half plus along x,
half plus along y, half plug along z to every relationship which we generate in terms of
equivalent points associated with the primitive lattice.
So, the other thing is we have seen in the case of the orthorhombic system is that there
are three 2 fold axes perpendicular to each other. So, whatever property we associated
with one axis and the monoclinic system can be associated with the other two, and there
and also the possible existences of 4 types of lattices we will generate a large number of
space groups. So, now if we can add up all these space groups we are going very quickly
towards the 230 possible space group. We are not going to count them now, but
eventually we will reach a stage where we will finish up all the 230 space groups.
209
Solid State Structured Chemistry Unit
Lecture – 17
Additional Symmetry Elements (Continued)
So, let us look at the tetragonal and how the tetragonal representation takes place. This is
a bit of a change from the triclinic, monoclinic and the orthorhombic. In case of a
triclinic of course, we do not have to worry about a b or c, because a is not equal to b, b
not equal to c, so we can take any of those three crystallographic directions. And In fact,
there is no question of any rotation axis associated with that, and therefore, as a
consequence we need not have to worry about the direction in which we fix the axis,
because there are no options in case of trigonal
In the case of the monoclinic system we have that choice and the choice is either a b or c
depending upon what two angles, we would like to take as 90 degrees and technically,
and from convention it is always the b axis which is taken as the unique axis.
So, the Y direction is the direction in which the 2 fold axis exist in case of the
monoclinic system, and therefore, we normally refer to this as the unique axis the b axis.
It may be possible to have a different setting in a monoclinic system where we can also
either a or c, in which case we have to mention the uniqueness associated with the 2 fold
210
axis. When we come to the orthorhombic system we have no choice, because a b and c
all three are 2 fold axis and. So, the nomenclature is that first when we say a point group
like 2 2 2, the direction is now along a, along b and along c, the three directions.
So, as far as up to the point of orthorhombicity is concerned, the order is a b and c. When
we come to the tetragonal system the order changes and this is something which is not
mentioned in any textbook and people forget about it, or with fact people are ignorant of
it and that is where I thought I should mention that the 4 fold axis associated with the
tetragonal system always comes in the middle beginning.
So, when you say a point group symmetry, when you see a point group symmetry
starting with a 4 or any of the proper and improper inversion axis associated with the 4
fold axis, like 4/m or ͞4 or 41 42 43, then it means that we have a tetragonal system. So,
tetragonal system is characterized in terms of the first number associated with the Z
direction, we always have this along the Z direction.
The 2 fold is then in since we have 422 as a possible combination of 3 axis. See in
tetragonal we can have only one 4 fold axis alone and that itself satisfies the condition
required for a tetragonal symmetry; that means, a not equal to a equals b not equal to c, α
β γ 90 degrees. So, that is a single axis operation.
We can also have the 3 axis operation as per the rules of the Euler’s Theorem, then we
have something like 422 which have already seen in the previous class, and in this case
we have a proper improper, improper axis which is 4mm. So, this combination is also
allowed and then we the second 2 fold; therefore, or the mirror symmetry a represents a
proper and or of inversion along the X direction. So, the 4 fold is along the Z direction,
the next 2 fold or any associated 2 fold with mirror inversion and so on is along the X
direction.
And then we the, that leaves us with the third 2 fold axis which is again proper or
inversion, and that proper or improper and or inversion and that goes into the direction
which is 110 in the crystallography nomenclature. What we mean by this 110 is,
whenever we have a nomenclature like this in square brackets, it means that it is a
direction in the crystallographic point of view. Suppose we have a crystal structure with
a b and c as the directions, X Y Z are the directions to which a b and c will now coincide
211
in which case we will say [110] means 1 unit along the X direction, 1 unit along the Y
direction and 0 along this. So, this represents therefore, a crystallographic direction.
So, this is a vector which is pointing out in a direction which is [110], characterized by
[110]. Just to complete discussion on the nomenclatures if you have open square
brackets, it is a crystallographic direction and if you have them in normal brackets, it
represents crystallographic plane or planes the family of planes in that particular
orientation. We will be discussing this in more detail when we go over to this space
group nomenclature and beyond.
And particularly when we discussed the scattering coming from X ray diffraction, I think
that context we will postpone the discussion on the nomenclatures here, just to, you
should remember, and when we discuss it we will recollect that [pqr] in brackets, square
brackets represents crystallographic direction; (pqr) in brackets, normal brackets
represents a crystal plane and if you have this flower bracket, it actually talks about
family of planes. And In fact, it is it is coming very often in higher symmetry point
groups and we will discuss this again with respect to the crystal and the X rays
diffraction conditions which follow as a consequence of it.
So, just remember that these are the nomenclatures you will find in text books and one
has to be very cautious in looking at these brackets carefully, and then the meaning
associated with those brackets when someone reads a, writes a technical paper in the
area.
So, the tetragonal system therefore, can be 4mm which is essentially a proper improper
improper axis, and in this particular case down here, it is 4/m 2/m 2/m. In fact4/m 2/m 2/
m can also be written as 4/mmm and that is the shorter form of, this is essentially it
means that we have a combination of a proper and an improper axis associated at each
direction.
So, for this 4/m 2/m 2/m means that there is a 4/m symmetry associated with the Z
direction and then a 2/m symmetry associated with the x direction and a 2/m symmetry
associated with the direction [110]. So, this is how a tetragonal point group
representation comes out. Having seen this let us go further into the trigonal system.
212
So, we see now that a similar nomenclature is utilized in the case of trigonal symmetry as
well. So, the 3 fold axis proper inversion along the Z direction. So, now, the 3 fold takes
the Z direction in that a representation, as we discussed in the case of a tetragonal
symmetry the 3 fold will also be along the Z direction and it can stand alone, or it can be
with a combination of two other symmetries like 322 for example.
So, here is an example where we see that there is actually standalone 3 fold, standalone ͞3
and then we have here 3m. Essentially it means that it is a proper and an improper
combination, so; obviously, the third one will be also an improper axis which is not
represents generally, and ͞32m now represents the presence of ͞32/m which is proper and
improper simultaneously and then the last one which is 32. I am just writing a question
mark here, it was then saying that it is a proper improper improper or proper proper by
improper or improper by proper, whatever that is, I want you to find out what could be
the symmetry in the third direction given this nomenclature.
Similarly, in the case of 32, it is probably easier, because I have already written proper
proper proper. So, that will tell us the issue associated with 322. So, basically what we
are now representing, or the 5 possible point group symmetries that can be associated
with a trigonal system and the Zdirection, is the 3 fold axis. The 2 fold axis is again
along the X direction and the 2 fold which is proper and of inversion or whatever it just
noticed that there is a change between the tetragonal and the trigonal system.
213
In the trigonal system it is 1͞10, in the tetragonal system, it is 110. Let me point it out
with a laser, it is 1͞10, in the case of the tetragonal system, it was 110.
So, having seen the trigonal and a tetragonal which are represented in the corresponding
3 fold direction and the 4 fold, it is obvious that the hexagonal system will again have a
Z axis, representation of the 6 folds symmetry. Remember when we did the point group
symmetries which can exist just along 1 axis, we discussed the possibility of 2 3 4 and 6
and the 2 corresponds to a monoclinic system, 3 corresponds to a trigonal system, 4
corresponds to a tetragonal system and 6 corresponds to a hexagonal system. And
therefore, we have 6 and ͞6 which are indicated here and along with that goes the 6 by m
symmetry.
So, these are stand alone single rotation axis symmetries, but we can also have the
combinations like 622 for example, which means that the 2 fold axis. Now again along X
and the 2 fold proper are inversion again along 1͞10. So, there is no much difference in
their direction aspect, except that the symmetry is now controlled by the presence of the
6 fold axis compared to the trigonal system, where it was controlled by the 3 fold axis. In
fact, there is not much of a difference between the trigonal and the hexagonal, we do not
have enough time to go into the discussion of how the systems can be, for example,
hexagonal can be also viewed as a rhombohedral system, may be now in one of the ta
classes, this can be discussed the hexagonal to rhombohedral system.
214
And then how the rhombohedral system develops and both of them can be converted one
to the other. This is something which maybe can, may can be done in one of these
classes, but I generally feel that it is not immediately required for the continuation of this
course. So, we will assume that hexagonal and rhombohedral are convertible to one to
other.
I just want to take give you a take home lesson at this point. Find out the conditions of
the rhombohedral system; that means, what are the relationships between a b c and alpha
beta gamma and then you see that there could be a relation between the rhombohedral
and the hexagonal system.
So, these seven point groups now belong to the 6 fold hexagonal symmetry, and these
point groups; therefore, represent the stand alone symmetry axis and then the
combination of 6 and 2 and 2 and the arrangement of the axis are shown here. So,
depending upon the 6 fold symmetry we will develop 6 2 fold orientations on either side
of the 6 fold symmetry direction which is the Z axis.
So, having seen this system, we now go to the system which is the most simple system,
but most complicated to understand and that happens to be cubic system. The reason why
it becomes a little more complicated to understand is, because that of the fact that the
association of the rotation axis is not in principle with an origin and then the X direction,
Y direction, Z direction. As you can see here in this particular case, you have a 233
215
represented in this particular diagram, which essentially outline say tetrahedral type of an
arrangement and the cubic which is 432, it actually represents an octahedral type of an
arrangement. So, here what we do is, we refer t; the center of the cube rather than one of
the cube edges.
So, the symmetry information, the symmetry axis information will go with respect to the
center here and as a result we have a 3 fold rotation, along the diagonals. And these 3
diagonals are shown here with a 3 fold rotation marked the on them, and then we will
also have the 2 fold axis or the 4 fold axis in this particular case, which is along the
direction, along the 2 fold or 4 fold; that is the symmetry along the X direction and then
there is a 3 fold along the [111] direction.
So, what you see here is that the representation of the space group; therefore, which is
now shown in the stereographic projection. So, the stereographic projection therefore, is
with respect to the center of the cube and effectively you take this cube and put it inside
this sphere, and mark these up and down points as per the definition of the stereographic
projection as open and closed circles.
Then you will get the entire symmetry that is possible to be marked out on the
stereographic projection and that is why the stereographic projection looks a little not so
very easy to understand than the previous case. But if you look and compare this diagram
with this diagram you will start saying the intersection points which are appearing here
as different kinds of rotation axis.
So, in a in a typical cubic system we can have a 2 fold along the X direction or a 4 fold
along the X direction. We will also have a compulsory 3 fold along the diagonal, the
diagonal now is [111]. The presence of the compulsory 3 fold makes it unique associated
with as a unique symmetry that is associated with a cubic system.
So, when we describe these possible 1 2 3 4 5 point groups with the cubic system, we
have to consider the fact that we have the center now is associated with the center of the
cube. So, the representation here is 23; of course, obviously, the third axis is also 3 fold
axis which is not necessarily written here.
So, suppose you come across a space group which is 32 and then there is a come across
the point group which is 32 and a corresponding point group 23, it is now easily
216
designable. So, if the axis number starts with 3 in the beginning; that means, it is a 3 fold
axis which is along the Z direction. So, it belongs to the trigonal system in this particular
case, the 2 fold is with respect to the X axis.
So, the first one is again in the cubic system, the first representation is along the X axis
or the or the a direction, because a b and c are same in a cubic system we can call it as a
rectangular system, because alpha beta gamma are also 90 degrees, so we can use X and
a interchangeable here.
So, the 2 fold symmetry or the 4 fold symmetry in this particular case, the different point
group symmetry in both these cases we use that as the symmetry, identify along the X
direction. Now the fact that we get a 2 fold and a 4 fold along the X direction does not
just stop that, because we have them. The curve following 3 fold symmetry here and a 3
fold symmetry here, once again you see there is a 3 fold symmetry and a 2 fold
symmetry which comes along the [111] direction.
And this is the way in which we now identify the cubic point, point group, because if you
have a compulsory 3 fold along in the second position throughout all the point group
symmetry, the presence of the 3 fold are of course, the related inversion center or the
related mirror symmetry. As in this particular case this is a ͞3, 3 and ͞3 bar are both
associated with a single direction axis.
So, we have therefore, the presence of the 3 fold as a very common and unique
representation and that will come always in the second position in the point of symmetry.
And we will see that when we look at the associated space groups which will be a very
large number in the case of a cubic system, because cubic system can support different
kinds of lattices.
A cubic system will support a primitive lattice, it will re support a phase centered lattice,
it will also support an eye centered lattice and as a consequence he get a large number of
possible, lattice symmetries that can be associated with these point groups.
So, to cut the long story short, cubic is the most symmetric system one can think of. The
distribution of the symmetry elements in the cubic system looks too very complicated,
but it is fairly straight forward in the sense that it is either a 2 fold or a 4 fold axis along
the X direction. The 3 fold compulsorily is along the [111] direction and then the third
217
direction could have either a 2 fold or a 4 fold. As we can see here it could also be; of
course, a combination of a proper and an improper axis depending upon the rules which
have been laid out by the Euler’s Theorem.
So, it has to be PII or PP/I IP/I or whatever, so those combinations therefore, or allowed
combinations to represent the cubic point with symmetry. So, this essentially now covers
what we would like to have covered for the point group symmetries. The 32 point groups
now are distributed to the seven crystal systems, and the seven crystal system
characterization we have already analyzed with respect to their shell dimensions a b c
alpha beta gamma. They are identified with respect to the crystal system.
There corresponding lattice symmetries have also been identified. For example, there is
always a primitive lattice that will be associated with triclinic and nothing else. With a
monoclinic symmetry we can have a primitive as well as the possibility of a C centered
lattice, c could be either A or B depending upon the definition of the unique axis. If our
unique axis is b, then it is always refer to the C centering of the lattice as the possible
Bravais Lattice.
So, we therefore, have the next slide shows us the possibilities of what kind of lattices
which we can associate with each one of the seven crystal systems. So, we have for
example, here triclinic. Triclinic is associated with the lattice symbol P which is an
218
accepted symbol and then we have a b c not equal alpha beta gamma, not equal that is
the definition of the unit cell. And then the number of lattice points.
We have one lattice point in the case of a primitive and the point group symmetry. See
this is very interesting, we can have two associated trigonal symmetries with the point
group; that means, two point group symmetries 1 and ͞1 are possible, but notice that the
symmetry that is associated with the lattice is always ͞1 . So, if you just look at a triclinic
lattice and you are given the information only about the tricliniclattice by some
experimental procedure which is always a ͞1 symmetry; that means, the lattice associated
with the triclinic system is ͞1 .
On the other hand the point group symmetries are 1 and ͞1 , so this happens throughout,
in these cases the highest symmetry associated with. You see here it is mmm is 2/m2/m2/
m and as we go down, these highest point group symmetry is the symmetry of that
lattice.
So, the lattice symmetry always keeps the highest possible point group symmetry
associated with the seven crystal systems. So, individually we can have more than one
point group symmetry associated with each crystal system.
On the other hand, we can also have the types of lattices differ. So, here for example, in
the case of a monoclinic system, it could be a primitive lattice or a C centered lattice. So,
this is the definition for the unique axis being b. So, β is always different than 90
degrees. The convention says that β angle is preferably greater than 90 degrees, modern
day crystallographers, particularly people who just solved structures without thinking
about what is going on.
They even report β angles less than 90, but crystallographic convention demands β to be
greater than 90. There are lots and lots of publications in literature, particularly by
chemist practice in crystallographic or somebody who takes crystallographic for granted
or the mistake of the manufacturers and also the software people who have developed
these packages, beta is all some many times reported less than 90.
But as a quality crystallography expert one should always have β greater than 90. In a
monoclinic representation, this is a must and people should take care of it. So, those pay
structures even which β is less than 90. Well there are certain things which they will be
219
missing particularly in terms of data completion and so on. This is something which we
will discuss when we come to the experimental methods of structure determination.
So, the number of lattice points associated with a primitive moniclinic is 1 and a number
that is associated with a C centered lattice is 2 and a point group symmetry, as we saw is
the highest symmetry that can be associated with the monoclinic system which is 2/m.
Then we go to the orthorhombic system. In case of the orthorhombic system we have

four possible lattices; the primitive lattice, the C centered lattice which could also be
represented as A and B, A centered, as well as B centered the because a is equal to b is
equal to c. Then we have the I centered lattice which is the body centered lattice and then
the phase centered lattice.
Notice that the lattices are independent of the point group symmetries. So, the point
group symmetry of the lattice is very high. Whereas, the lattice symbols will generate
different types of lattices; that is number of lattice points. The number of lattice points
associated will decide on the nature of the Bravais Lattice, but the point group symmetry
of the lattice is the highest point group symmetry associated with the crystal system.
So, just by looking at the point group symmetry of the lattice, one should not conclude
the point group symmetry associated with that particular crystal system, the crystal
system still has 1 and ͞1 . For example, in case of the triclinic lattice these are issues
which one has to thoroughly remember and understand. In fact, understand it when once
you understand it, you will; obviously, remember it, it is not given in any textbooks, so
this is something which has to be learnt by practice and at this time you are learning from
this course.
So, the four types of lattices that we can associate with an orthorhombic system the P C I
and F, we will carry the number of lattice points like this. 1 for primitive, 2 for C A or B
and 2 for the body centered which is X corresponding to X Y Z and half along X, half
along Y, half along Z to be added to. So, that means, ½ ½ ½ is an additional lattice point
along with the 0 0 0 which will generate one-eighth of the points around in the lattice.
The F centered lattice will have 4; 1 corresponding to all the corners and 1 each for each
of the phases A B and C which is centered.
220
So, we then go to the tetragonal system, in the case of the tetragonal system because a is
equal to b it is a square lattice. We cannot have any independent centering associated
with it, because of the fact that a equals b and not equal to c. We cannot have a F
centered lattice, because the face centering will automatically move the A B C, because
we you know the translational periodicity is operational here and therefore, we get only a
P and an I centered lattice in case of the tetragonal system, we do not get any other type
of lattices.
And this therefore, gives us the number of lattice points 1 and 2, and the maximum
symmetry that is associated with the tetragonal symmetry is 4/m 2/m 2/m which is
simplified and written as 4/mmm.
Obviously, you know we cannot have a combination of this kind, it has to be 4/m 2/m 2/
m which will have both primitive and inversion associated with it. Then we have the
cubic system, we have the case of the cubic system as P I and F and 1 2 and 4 and you
see the symmetry is m3m; that is the highest symmetry one can have, 2/m 3 or 3/m
associated with it and then (Refer Time: 27:16) symmetry. So, m 3 m is the highest cubic
symmetry we can have. So, 3 and ͞3 are possibilities, but m3m is the symmetry; that is
associated with the lattice.
Then we have the trigonal and the hexagonal system, as I mentioned that the relationship
between these two is very subtle and it can be converted. For example, if you see the
rhombohedral, there are two settings; one setting is actually the setting associated with
the hexagonal system a equals b not equal to c ,alpha beta of 90 degrees and gamma is
120 degrees.
On the other hand the hexagonal system is a primitive lattice which is a equals b not
equal to c alpha beta 90 degrees, gamma is 120 degrees. You see in the rhombohedral
setting we have a equals b equals c alpha beta gamma are all equal and this should be
generally less than 120 degrees and the highest point group symmetry; that is associated
with this is ͞3m, and in this case it is 6/m 2/m 2/m. So, this generally tells us how we now
distribute the crystal systems, the Bravais Lattices, the type of the lattice we define that is
the unit cell convention or unit cells, the number of lattice points and the highest point
group symmetry which is associated with a lattice.
221
Of course we can have one more column up here which will tell us the all the point
group symmetries which we can associate, but I think we have already seen the separate
table for that, which is always good to have within a separate table, which we have
already seen several times.
And therefore, this now generally tells us that in a crystal system when compounds
crystallize they have a 3 dimensional crystal; the property of the 3 dimensional crystal
we have seen so far is the periodicity; that is associated with the crystal. The translational
periodicity which therefore, invokes the presence of a unit cell and there are several
possible unit cells associated with each one of these crystal systems which we have seen
and with them are associated the Bravais Lattices.
So, in order to actually represent an object inside this unit cell we need to have all these
satisfied. That means, let us say a molecule goes into a one of these crystal systems. The
molecule will now face the possibility of having any of these Bravais Lattices, depending
on any of these Bravais Lattices, the number of lattice points will change.
So; that means, that the lattice; that means, associated with, let us say the C centered
lattice will have to have the same object, also appearing with respect to the C centering
here. Again in the case of the F centered orthorhombic lattice, there must be four of
those and, this is the way in which we have to associate the objects and these objects
when once associated with the lattice points. Now we will go the corresponding point
group identification, they will be already symmetry associated with the lattice.
So, now the possibilities of the other independent space groups will come up and
depending on that the compound reorganizes itself. So, when a compound crystallizes in
a 3 dimensional lattice of this particular kind, it faces all these restrictions. And because
it faces all these restrictions it is now possible to find out the position of the atom or the
molecule or the association the molecular assembly. Whether it is a small atom or a big
atom or a big molecule or a small molecule, it does not matter, when once we are in a
crystal system these are the rules which have to be followed without fail.
222
Prof. T N Guru Row
Lecture – 18
Space Groups 1
So, one other little point which in fact, was mentioned to me by one of my students while
I was winding up the discussion on this table; is the fact that in trigonal, we represent it
as R which is the rhombohedral lattice. And, then we write a equals b equals c alpha beta
gamma are equal to each other. Of course, obviously, we cannot make it three-
dimensional closure objective. Any of these angles are greater than 120 in such a
situation, so, this has to be less than 120. The point that is also important is if alpha beta
gamma are equal 90 degrees, then it will go to the cubic system. So, there is a very subtle
difference just like between trigonal and hexagonal as we see from the notations here
there is a subtle difference between this and the trigonal and the cubic system.
The trigonal system also has a three-dimensional rotation axis, the threefold rotation axis
sorry. And the presence of the trigonal axis which is non 90 degree makes it a
rhombohedral system. So, even if the angle instead of 90 degrees is 89.9 or something
like that it will be pseudo cubic. So, most often than not trigonal symmetry is (Refer
Time: 01:49) into the cubic symmetry and many of the materials which go into the
223
trigonal symmetry. Therefore, have a near cubic symmetry. So, for all practical purposes
they are cubic symmetry, in the sense that the properties associated with those materials
will be cubic. So, it is an issue which I thought I will mention before we go further.
So, now let us get into the as to say, because these are the main issues of how we look
into the so, called space groups. So, far we saw all the 32 point groups, how the axis
developed with respect to individual point groups and so on. As the definition of the
point group refers to it is with respect to a point, but what happens in a, when once we
associate that with a Bravais lattice this is that they get into a crystal system. And now
with point of discussion takes us to how the objects inside this particular unit cell get
generated by the symmetry operations that are associated both with the Bravais lattice as
well as with the point group symmetry which is associated with that lattice.
So, this is therefore, one of the most important part of our course where we will see how
the objects relate themselves to each other across the symmetry in the unit cell and the
contents of the unit cell and how the equivalent points get generated. So, we will start
from of course, P1 is obvious. So, I did not take P1, because in P1 you will have a
system which is represented. Suppose, we look at this ab projection; that means, this
particular angle between a b is γ and this gamma value is non 90 degrees.
We will have no none of the symmetry elements associated with P1, because it is only a
one fold symmetry; that means, 360 degree rotation. So, there will be an object
224
somewhere and that remains there in the unit itself no other relationships were possible.
And so, there is no point in indicating the diagram for P1 even though the international
tables and crystallography will also list the P1 as the first space group. When we go to
the discussion on the international tables for crystallography, we will look at the
possibility of P1 as well.
So, the next point group symmetry which we can have which we try to make system as
this is indicated here is P͞ ͞1 . So, ͞1 is the symmetry operation and P represents the lattice,
the moment we represent a lattice by P, we know that the number of lattice points in this
particular unit cell is 1. So, there are effectively in projection, we are looking down the c
axis. So, this unless, otherwise specified the convention is always like this, down this is a
the down the origin, origin is taken as the left hand top corner. And with respect to the
left hand top corner, you mark a downwards and b across, c is perpendicular to us.
So, unless otherwise specified in all the diagrams which we are going to see in the next
few discussion half hours, we will see that the projection is always down c. If it is
different, it will be mentioned; we will have some discussion on what happens if it is a
different axis. So, in a triclinic system, we have a b c not equal to each other; alpha beta
gamma also not equal to each other. And therefore, we have the object which is sitting
here which I have marked as 1. Now, here after we will represent the object to show that
we have an atom or a molecule or a collection of atoms or an ion or whatever in the form
of a comma.
So, comma now represents our object. It could be a protein structure, it could be a virus
structure, it could be a single atom and so on. So, comma(,) is the object on which we do
the symmetry operations, the advantage of using this nomenclature comma also allows
us to turn it around and show the different orientations that the object can check on the
operation of the symmetry. And at the same time, we can also indicate whether it is in
the direction towards +c or -c by indicating them with the plus and the minus as is shown
here. For example, if you take this point 1 which is now represented as x y z the
coordinates of which are x y z. So, there is an object, here the object is put inside the P ͞1
symmetry.
When you put the object in the P ͞1 symmetry, which sees the center of symmetry at the
origin; remember that every point in this three dimensional box which is now shown as a
225
two dimensional projection for our understanding, every point will have the center of
symmetry. So, it means that if the object is here or object is there, object is there they
will all see the center of symmetry. The location of the center of symmetry is with
respect to this projection is shown here, when once it is at 0 0 0 that is our definition of 0
0 0. So, when once we are defined the origin, the location of the symmetry is fixed. So,
the location of the symmetry is fixed at 0 0 0, it will of course, repeat after one
translation along a, because of the periodicity that is required in a crystal. It will also
repeat one unit along b, because of the periodicity and one unit along c.
So, if you now consider this particular object and say it is coming towards as from the
board then we have the x y z as the coordinate. So; that means, if you now take this as
the x value and that has the y value, z will be the value which is coming out of the board
then we have the position of this object located. And that is located at x y and z, we can
have any number of objects which can be at different values of x y and z and what
happens is that now it will see the center of symmetry at the origin. When it sees the
center of symmetry of the origin, it undergoes the symmetry operation which we all now
very thoroughly know. If x y z goes to -x -y -z which we pronounce as x̅ y̅ z̅.
So, we have the second objected generated which is marked as 2. So, we have the object
1, we have the object 2, you see that the object now inverts and also goes down in the c
direction. So, x changes to -x y changes to -y z changes to -z, the problem here is that we
have generated a point outside of the unit cell, we want all the points generated within
the unit cell. So, in order to do that what we do is we now apply a translational
periodicity to this point 2 which will take it to that point; translation periodicity
associated with the a direction. And then, if we now apply the translation periodicity
associated with b direction, we will get to this point which is inside the unit cell.
Notice that this point and this point, this object and this object are related by a center of
symmetry which is located at ½ ½ ½ half along x, half along y, half along z, it is also
located at ½ ½ 0. So, there is a issue which one we will take. In this diagram which I
have shown as a projection, we can take it as 1/2 1/2 0 at this position. So, we have
therefore, a this as x y z, this will be x̅ y̅ z̅. What all you have done is to now see that this
particular origin has moved over to this position, we are not going to call this as 0 0 0.
226
However, 000 is still associated with that one; if that is 000 and this is 100. We have
already seen that the fact that there is a translation periodicity, the symmetry element will
repeat itself half way through this. So, we get a symmetry element at that point,
symmetry element at that point, symmetry element at this point and so, on. It will also
happen at this 0 as well as ½ and also to full unit. So, the symmetry that is represented
here is at ½ ½ 0. So, each one of these symmetry positions as you can see are
independent of each other; that means, if you have this position you have over here, you
see these two are operations now.
This is one unit translated along the x direction, this object translates one unit sorry,
along the b direction you get this point. Now this point will see the center of symmetry, it
will also generate that point which is outside of the unit. So, you see the consistency of
the definition of the space group that is this point now translated by one unit will go
there. Now, these two objects are related by a center symmetry, these two objects by
translation, these two by center of symmetry, these two also by center of symmetry.
So, the translational periodicity which has come here, we now look at the 100 center of
symmetry and go outside this point keeping the center of symmetry relationships intact.
So, this is one center of symmetry that is the second center of symmetry and so, on. So,
effectively we get two equivalent points one at x y z the other at -x -y -z. So, the system
is triclinic the number of objects that are generated now will be 2 and this we have
already seen is a nomenclature which we use as z is equal to 2 in the unit cell.
Now, let us look at a situation where this objects associates itself with the center of
symmetry; obviously, if it associates itself with the center of symmetry, it cannot take the
form of a comma. Even if it takes the comma, shape of a comma the comma has to be
overlapping with each other; that means, this of this image and that image will overlap at
that point. Half of the image of 1 and half of the image of 2 will overlap to generate the
object. In other words, if there is a molecule we cannot change the molecule, we cannot
say atoms are different now, because of the fact that it goes to the center of symmetry.
So; obviously, the center of symmetry should have a center of symmetric object
occupying that particular position.
So, the object by itself then is required to be center of symmetry. A non-center of

symmetric object; that means, in terms of the molecule the molecular symmetry now
227
takes into take important role; if the molecular symmetry has a center of symmetry , it
will coincide with the origin of the units. This is the point which you should know,
because very often they are not these problems will come up when a molecular
symmetry and the crystal symmetry are one and the same. If the molecular symmetry and
the crystal symmetry are different then it will sit in the so, called general position.
So, these two positions are referred to as the general position. So, if x y z is 000 then the
second position is also 000; that means, one half of this object and the other half of that
object will jointly be present on this position; in such a way that the object shape is never
changed, the objects shape remains the same. So, if the object has a center of symmetry,
it cannot change its center of symmetry, I mean this is the point which will come as a
very important issue when we discussed the, the various possibilities of what we call as
special positions.
So, 000 is a special position. Now, if the object is associated with 000 and that is the
special position then the other centers of symmetry which are in this diagram, they are all
different centers of symmetry. So, each one of the center of symmetry has to get an
identity. Now, this becomes an issue particularly when we go to higher symmetry space
groups, because in higher symmetry space groups, this kind of degeneracy may be many
more many more positions can come. So, effectively the center of symmetry position is a
degenerate position. So, each one of these centers of symmetry therefore, are
independent of each other. Now how do we identify these?
In order to identify their they are different from each other a convention defined by a
person by name Wyckoff has been formulated, we will discuss the Wyckoff notation and
Wyckoff formulation, when we looked at look at the entries in the international table,
because most of you are non-practicing crystallographers from the crystallographic point
of view, but you are practicing chemists who used crystallography, practicing physicist
who used crystallography and so on. By biologist who use crystallography, so, for those
people who use crystallography they a ready reference point for identifying. And then
looking at the details of the space grouping is the international tables for crystallography.
So, we will look at the internal table of crystallography entries and at that time we will
define what is known as a Wyckoff position. So, the Wyckoff positions; obviously, are
referred to the special positions and also to the general positions. So, the way in which
228
Wyckoff designed this methodology was to look at the most symmetric position in the
space group which according to him in this case the P ͞1 bar is 000, he calls it as a small
symbol a. So, a small letter a is associated with this point, then ½ 0 0 is b and so, on.
And the general position therefore, will be after all these eight positions identified.
So, you go by the alphabetical order a b c i e f g h and so, on. So, this kind of a
nomenclature is referred to as the Wyckoff notation and many many publications who
particularly in the field of inorganic chemistry complex materials and so on, people use
the Wyckoff symbol to identify the position of the atom. So, they will just say that atom
the Br atom or the or for that matter the Zr atom is sitting in a Wyckoff position a in this
particular space group. So, if they say it is sitting in Wyckoff position a in P ͞1 then we
know it is at 000. So, the position of the heavy atom is identified like that any atom for
that matter which sits in a special position is identified with respect to its Wyckoff
position
So, in a large molecule for example, it so, happens that some of the atoms might sit at
special positions. Their occupancies is the new word, I am introducing their occupancies
will not will be a, a fraction of the general occupancy. Let me define now the occupancy
if there are two general positions, in this particular case of P ͞1 there is x y z and x̅ y̅ z̅.
Then there are two molecules or two objects in this unit cell the object which is left alone
and as one here is referred to as the object in the asymmetric unit. So, we say that the
asymmetric unit has one object, but due to the symmetry operation, we generate two
objects which are symmetry related to each other.
So, these two positions are referred to as the equivalent positions. Now, if there is this
object occupying each of these equivalent positions then we say that the number of
equivalent position, we need two the occupancy is full associated with x y z and x̅ y̅ z̅.
So, we say occupancy is 1. The moment we put this into a special position, the let us say
x y z is 0 0 0 then x y z and x̅ y̅ z̅. will be one on the same and the occupancy now
becomes half of the total occupancy. So, the total occupancy here is 1 with respect to a
general position object, with respect to a special position object, it becomes half. So, the
occupancy of half therefore, is a issue which we will have to discuss further when we get
into this special position discussion.
229
So, just to put all the words we have introduced with respect to description of the space
group. These are the words one is of course, the crystal system then the point group
symmetry, then the space group information like P ͞1 , then the positions of the
symmetry elements like the center of symmetry in this case and then the position of the
object. So finally, it is the object we are interested in, we want to determine the structure
of this object whatever it is that is the final goal. But, this object now is put into this unit
cell with this information about the space group which will tell us how the object is now
symmetry related and arranged itself inside the unit cell and also in the translational
periodicity situation.
As a consequence the overall structure therefore, that can be determined for this is
represented in terms of two objects related by this center of symmetry. So, this particular
value of x y z now generates an x̅ y̅ z̅. So, we have introduced the other word which is
called the occupancy that is association with the position. So, if there are all these
general positions in the unit cell are occupied then we say the occupancy is 1. The
special positions have an occupancy in this particular space group of 0.5. The special
positions can have lower and lower occupancies, depending on the symmetry that is
associated with that point. Suppose, we have let us say a three fold axis, a six fold axis
and things like that the occupancies will be correspondingly reduced from 1.
Because we, if we have an object sitting at a six fold symmetry, there will be six objects
that is generated around it. Now if all these 6 become only 1, then the occupancy of that
may be 1/6 of that position. So, the occupancy of this here is they are the number of
equivalent points is 2, occupancy becomes 1/2. If number of equivalent points is 4
occupancy becomes one fourth and so, on. And therefore, the definition of the occupancy
is given as a additional word which we have introduced in our grammar. The one more
thing which we have introduced in our grammar is the Wyckoff notation. So, this
Wyckoff notation is the highest symmetry position is given a little a the symbol alphabet
a. And as we go to lower and lower symmetry positions associated with this position, we
call them as b c d and so, on. Finally, the general position will gets its alphabet.
So, in case of P1 for example, what would be the Wyckoff notation think about it and
that will tell you what would be the Wyckoff notation for P1, will there be any more
alphabet than a ? think about it, right. So, we have now seen the if this triclinic system 2,
space groups P 1 and P 1 bar the equivalent points, how we generate equivalent points
230
where are the symmetry positions. And, then we have always now looked at with respect
to how the object gets placed inside the unit cell.
So we have now a complete picture of what to expect in a crystal structure with respect
to an object which was crystallized in that particular crystal structure; obviously, triclinic
is one of the simplest of space groups which we can understand. So, if you understand
this all the 230 space groups could be understood in terms of the same logic. So, the
logic is that we find the number of equivalent points, the number of symmetric positions
that are present, the translational periodicity, the nature of the lattice, the lattice centering
information and the point group symmetry that is associated in that.
If we know all these, then we know the crystal structure followed by the molecular
structure. So, this our overall ambition in this is to find out where the atoms are sitting in
the crystal so; that means, we want to determine the molecular structure, because we also
see that the atoms join together with bonding features and so, on. And they are the
definition of the object we have here. So, if the object is now a molecule there will be
several atoms, but all these atoms now will be identified with respect to positions x 1y1z1
x2y2z2 and so on. All of them will see the second equivalent point. In this case of space
group P ͞1 bar and so, we will be generating the objects 1 and the object 2 and we will
identify the symmetry.
So, is there a way in which this can be investigated? Suppose you are given a crystal and
you are asked to find out the position of the atom, is there a way ? it is something which
we will be discussing through the course; obviously, there is a way we of course, are we
do not have the power to look inside the crystal and see where the atoms are, because the
size of the atom is not visible to the naked eye. Obviously we can use probes; the probes
which we use can be of the wave length which will probably correspond to the size of the
atom. The wave length which we use in normal light when we use this light and look at
the objects we will, we will be allowed to see for example, this particular screen which is
in front of me or the key board which is in front of me using this particular wave length,
but and in fact, the wave length is also now at that resolution, where I can also read the
alphabets which are written on the key board. Otherwise I will not be able to type these
things. So, to that extent there is the resolution.
231
But suppose I want to see the way in which the springs are is connected inside this
keyboard. And how this connection now goes to this that is not possible by a naked eye.
So, what is, is needed is a probe. Most of the probes are essentially driven by the fact that
we need the resolution that is associated with observing those. So, if the size of an atom
is about 10-8 cm or 10-10 mm, In fact, it is one order of magnitude lower than the
nanometer. If you are able to see these objects through a microscope then in principle we
should be able to see the positions of the atoms, but that is not what is going to be. What
is going to be is to find a probe by means of which we can probe the distances
corresponding to the distance between atoms and the size of the atom.
And it so, happens that in electromagnetic radiation, we do have such a range and that
particular range where we can see this are the X rays. So, therefore, we up we use X
rays, one might ask a question can we use other matter ways? Certainly like matter
waves which will make from let us say neutrons or matter waves which come from
electrons. So, we can also use neutron diffraction and electron diffraction, but X ray
diffraction is the one which is the most preferred. And the most conveniently usable
wave, because electron ma diffraction and the neutron diffraction need very specialized
equipment, very specialized conditions under which we can use those equipment.
Whereas, it so, happens that X rays diffraction which was discovered soon after the X
rays were discovered in 1900, the it was discovered in the year 1912.
And, the fact that if there is a regular arrangement of atoms inside the unit cell, then there
should be some regularity in the diffraction pattern which comes out from this
experiment. And if we reconstruct the image of the diffraction pattern, you should be
able to see the positions of the atoms and that was the idea which triggered Max Von
Laue. And later Bragg and Bragg to look at the possibility of structure determination.
Then came a plethora of activity in X ray diffraction measurements. Particularly with
respect to small molecules initially and eventually the stage was reached such that we
can now conveniently look at structures or viruses and so on due to the technology
advancements.
Technology advancements high speed computing for example, and also high reliable
probed through which we can look into the crystal. And, also the reliability with which
we can measure the diffraction pattern which comes out; all these add up to the
advantage of looking at the atom positions. So, when we say that we are looking at a
232
crystal, we actually now do not look at what are the dimensions of the structure. So, the
way in which we take the stereographic projection was actually by observation. We took
the crystal, took the center of the crystal, drew normals to the crystal planes and then
collected all these points into a stereographic projection that was one way of looking at
it. But, that way of looking at it will give information only on the external morphology
associated with the crystal and its corresponding point group symmetry.
Beyond that if you want to look at where the atoms are and so on, we need probes and
such a probe is available in X rays. And therefore, with the future course classes, we will
look at how we do this experimental protocol. And, this is the experimental protocol
which is in fact, a requirement for many of the industries, many of the teaching faculty
and many of the students who are taking this particular course. But, remember we the
take home lesson is if we do not understand this part fully, then we are nobody to
understand the rest of the program which we have in the course. And that is why I am
spending an enormous amount of time in trying to describe the way in which objects
arrange themselves in a given space group associated with a point group in a given
system.
So, the equivalent points are important, the location of the symmetry positions is
important, the way in which we describe the unit cell is important. And so the so, what
we do therefore, in a structure determination is to determine the geometry of the object.
The geometry with respect to the unit cell, what are the dimensions of the unit cell and
that is the starting point. Once we have that particular starting point, the rest of the
protocol can be developed such that we now get the information about all these points
which we discussed here. And that is the one which will give us the information of where
the atoms are, how they are located and how to identify them.
Now, how to identify them is a protocol which is developed through X ray diffraction
techniques which we will go into at a much later time. So, at this moment, we have seen
one space group, we have talked so, much but we now know one space group. So,
today’s job further job in the next few classes is to look at the few more space groups.
Get a hands on training on identifying the symmetry positions, identifying the equivalent
points, special positions and so on.
233
And then we cannot do all the 230 that is not the idea of this course. We will get sample
points from triclinic, monoclinic systems then we go to a later situation where we just
straight way look at the international tables for crystallography.
234
Lecture – 19
Space Groups 2
So, now we will go further to the next crystal system, which is the monoclinic system.
Monoclinic system as we know is associated with three point groups. 2, m and 2/m these
are the 3 translation free point groups.
Because of the fact that we have 3 dimensional periodicity, we get into the introduction
of two other additional symmetry elements which are the screw axes as well as the glide
planes. So, we have to now consider the combination of the Bravais lattice with the
presence of the twofold and 21 screw the mirror and also the associated glide planes. So,
we will thereby generate several space groups.
So, the first space group among these is the twofold symmetry associated space group
P2, and here what happens is that if you have an object which is x y z which is sitting at
point 1, which it looks at and it undergoes a twofold rotation, it goes from x y z to x̅ y z̅.
The location of the twofold is now could be anywhere along the b direction. So, the
value of the y is arbitrary, whatever is the value of x and y it will change to -x -z, but
once we fix the value of y, then the value of y will remain unchanged.
235
So, the location of the twofold axes could be anywhere along this direction. As a result
we do not get any other positions associated in the b direction. Notice that this is a
projection of ab looking down the c axis and therefore, the angle between a and b is 90
degrees. So, this is a nice rectangle. You also notice that when once we have this twofold
axes associated with b direction, in the ac plane particularly with respect to the a value
which we have shown here in the projection, there will be an additional development of a
twofold axes at half position by definition.
Any periodic repetition as we have seen will repeat its symmetry at half the distance and
therefore, we have this at half the distance. And this half the distance repetition will have
allow for the possibility of this point being related by a twofold symmetry.
So, it not only generates that point why in the normal twofold operation the translation
per 1 unit is generated here and these 2 objects are related by the twofold symmetry and
these are the 2 objects we will see in this space group. And therefore, this is Z is equal to
2 and in this Z equals 2 we can always have an object associated and sitting on the
twofold symmetry.
Suppose you have a twofold symmetry objects just like the pointer here the laser pointer
here, it has a twofold symmetry it is a circular point. So, it has all kinds of symmetry. So,
twofold is also incorporated in the laser pointer. So, the laser pointer therefore, has a
twofold symmetry and suppose it sits anywhere along the length of this, you see that half
the object on the top and half the object on the bottom are related by a twofold symmetry
and as a consequence this will be the special position in P2.
And then we also have the possibility of the same special position coming at the repeat
distance along the a direction. So, other than that we have only the general positions x y
z and x̅ y z̅ and this generates 2 equivalent points and the equivalent points on one
translation take the objects outside of the unit cell and so, therefore, Z is always 2 in an
operation P2. Suppose we now want to consider this projection such that it is we show
the ac projection. If we show the ac projection you remember and you recollect that the
two fold axes now will be represented by a set of symbols, which are now perpendicular
to the 0 position and that you should keep in mind.
So, what is shown generally in all the projection diagrams including the international
tables per crystallography, unless otherwise specified the projection is down x is a and
236
down y is b and c is coming either in a perpendicular order going inside the board in a
perpendicular direction. So, it is not necessary in a case of a monoclinic system, it should
come in a perpendicular direction. So, this angle is β. So, the vector which comes out
from the origin is at an angle of β. So, which is not necessarily the value, that it should
be 90 degrees. So, that is a very important point, do not think that this is now coming in
the direction at 90 degrees this comes at an angle of β.
And that therefore, qualifies this complete description of this space group P2. The
addition of a half translation is invoked in the case of a 2 1 screw axes. So, instead of a
twofold axis if we have a 21 screw axis, then you see the operations very clearly that x y z
now will undergo a twofold operation and after the twofold operation we will see there is
a half translation along the direction of b. So, in this particular case you have x y z, the
rotation axes is y and therefore, the translation is along the y direction. So, x changes to -
x z changes to -z; so, you generate this by twofold symmetry operation plus a translation
along the direction about which the axes operated and so, we get to this position.
Now, this position on one translation will bring it down here, telling us that we do have a
twofold symmetry operation at half position and these 2 now become the equivalent
points in the unit cell corresponding to the coordinates x y z, x̅ ½ + y and z̅.
Notice that we can also have the twofold through axis symmetry at 0 as well as one the
presence of the translation periodicity has brought as the additional 2 1 screw axis just like
the logic we use for P2 and also to the triclinic system.
So, these are the 3 simplest of the space groups we can discuss with respect to the
equivalent points, with respect to special positions and with respect to the a generation of
the equivalent points and Z is equal to 2 and if there is a object which has a twofold
symmetry then Z will be equal to 1.
If there is an object which is in P21, then there would not be any special position because
the translation associated will destroy the possibility of having the special position; that
means, the we cannot have these object because this particular object cannot have a
twofold rotation followed by half translation within the rotation axis. That means, if it is
here and it goes there, you still have the 2 1 operation possible, but then the chances of
such objects being in the P21 symmetry are very little. So, twofold axes has a special
position 21 axis can also have a special position I agree, but then we have to see it
237
depends purely on the nature of the object. The object will not have its symmetry which
is now including the translational part of the unit cell and therefore, the special position
in P21 is not possible.
We go further now the possibility of now combining the presence of the second lattice
which is available in a monoclinic symmetry and that is the C centered lattice. We saw
the primitive lattice the C centered lattice now we are looking at the possibility of the
twofold rotation being associated. So, we are going systematically, we are looking at the
lattice symmetry first then the lattice symmetry changes from P to C then we look at
individual point groups. So, far we have seen 2 and 21 in case of the primitive lattice now
we see 2 and 21 in case of the centered lattice.
So, if you see the C2 symmetry that can be present in a monoclinic system; then what
happens is that you now have the twofold operation associated with the b direction which
will take the object 1 to object 2. Then object 1 will also have the C centered lattice. The
moment we have the C centered lattice we generate this object at ½ + x and ½ + y,
because it is C centered. The moment we say C centered the x and y will get the will be
adding half and half, if you say a centered we will be adding y and z half and half.
So, in this particular case of C centering we add half and half to x and y and therefore,
you have this equivalent point which is x y z. The x+1/2 y+ ½ z is an equivalent point
that is generated by the lattice symmetry, this is not an equivalent point which is
238
generated by the symmetry operation. So, the additional symmetry which we have got
from the centering should also see the symmetry operation, in order to be consistent with
the definition of the space group we have to have the closure property. And therefore,
this point number 3 will also see the twofold axis, which is now generated automatically
at half position. Because we have a twofold at 0 we have a twofold at a, but half position
we will generate another twofold.
So, this twofold will take the object 3 on to object 4. Now what is the relationship
between object 1 and object 4? Object 1 and object 4 will have to now go in a very
interesting way, I want you to find out what is the relationship between 1 and 4. And see
and satisfy for yourself that a combined centering operation along with the twofold
operation will generate these 4 equivalent points.
Now, having generated these 4 equivalent points of course, this can be now translated by
1 unit. So, effectively these are the 4 equivalent points. So, 2 translated by 1 unit sees
this particular twofold axes goes out of the unit cell, again this goes out of the unit cell
and so on so. Inside the unit cell if you count the number of objects you will have 4 of
them qualifying itself to be the C2 axis.
Now, comes the possibility of C21, what happens when we get to a space group curve
C21 ? In the number of 230 space group placed C21 will not find its position that is
because we can always convert C21 to C2, and that is the operation which is shown here.
So, first of all let us look at how the C21 operation comes up, the C21 operation comes up
from this point you take this as x y z, you rotate it by about the this is a 2 1 screw
operation at the origin. If this 21 screw operation is at the origin it is something similar to
P21. So, you do the symmetry operation P21 and you get this half translated at element
that when translated down by 1 unit you will get this element.
So, these two now are related by 2 1 screw symmetry. Other than that we also have an
additional point which will get generated due to C centering and that comes up here with
½ + x ½ + y z. Now these 2 objects are not directly related, but they are related through
the operation of the twofold symmetry at 21 symmetry again. See what is important here
is to look at where the 2 1 symmetry positions are located. Suppose we take this as the
origin, the 21symmetry is located with respect to the 0 position here and therefore, this
239
now defines the unit cell Oa, Oa is the unit cell and Ob is the unit cell that is associated
with the C21 space group.
Now, if we now consider the presence of the 2 1 screw axis, the 21 screw axis will appear
again at half position and so, this particular point which you see here now looks at the 2 1
screw axis here and because of that presence of this 2 1 screw axis this should undergo a
21 screw operation, which will take it to this point. So, as far as C21 is concerned, 4
equivalent points get generated inside this outer unit cell, the outer unit cell I mean the
ObOa cell the cell which I have marked by shown by the pointer. That is the unit cell
that is corresponding to the presence of this C followed by 21 axis.
Now, the fact that we have these 4 equivalent points inside this box, does not qualify C21
to be a standalone space group because what is happening is the presence of the 2 1 screw
at this point and the 21 screw half way and of course, the full way here they are going to
generate a perfect twofold symmetry, but one fourth. The reason why a perfect symmetry
at one fourth gets created is because of the centering.
Because we centered at half and half and then the corresponding 2 1 operation has taken
place such that we also get the -1/2 -1/2 kind of a situation, where these 2 points are
generated. The consequence of that is to generate there is a for example, if you look at
these 2 points you automatically see a twofold axes between this and that.
This is the x y z you will see the corresponding -x y z and therefore, this defines a
twofold axes. Now, this twofold axes we are now taking this as the origin O’ therefore,
we have to add one-fourth unit to this one, the 1 unit translation will now be defining the
a’. The distance O’a’ is same as distance Oa. So, there is no change in the a axis
similarly the O’b’ is the same distance.
The only thing that is changing here is the origin is shifting by one-fourth, once it shifts
by 1/4 you see that the 21 axis is no longer in practice, but with respect to this o prime we
have the twofold axes. This therefore, will generate another twofold axes at one unit
translation, it will also generate a twofold axes at half point which now relates these two.
So, if you now consider this unit cell the O’a’ O’b’ unit cell let me mark that with the
laser pointer this unit cell then you see there are again 1 2 3 4; 4 equivalent points in that
unit cell.
240
So, I want you to write down the possible equivalent points for C21 and also the
equivalent points of C2 and verify yourself that this can be now considered as C2
because its just a question of shifting the position of the origin by one fourth the unit.
And it so, happens that when once it shift by one fourth the unit cell, the 2 1 screw is
replaced by a twofold axes. So, C2 and C21 are one and the same. So, the equivalent
points will be exactly the same with respect to the C2. So, we do not have a space group
C21 anymore.
So, any combination of a C centered lattice with a 21 is ruled out; that means, when we
go to the other space groups like C21/m and possibly C21 upon C which is the 2/m
operation those 2 space groups will not be allowed. Will not be allowed in the sense they
are one and the same. So, C2 and C21 are one and the same space groups, now we go
further the twofold operations are completed. So, we have the twofold symmetry we
have the P2 symmetry described, we have P21 described we have C2 described and we
have shown C2 is equal to C21 and therefore, the number of space groups are in
monoclinic system so, far are P2, P21and C2 as far as the twofold operations are
concerned now we go to the mirror operation.
The mirror operation as we can see here from this diagram here it is defining the space
group Pm. We can work out of course, the corresponding equivalent points, I want you
to write it down I thought of writing it down, but I think now you are experts enough to
write down the equivalent points. So, you have the origin here and then you have a here
and the b here, where is the mirror located the mirror is perpendicular to the b direction
and therefore, you see that these lines are darker are higher width description of the black
line.
So, effectively the mirrors are perpendicular to the axis direction. The twofold axes is
still along the b direction to define the minimum symmetry associated with the
monoclinic system and therefore, we have a mirror plane, where this now undergoes a
mirror operation. So, we have x y z and x y̅ z. Now when once we have this generated
this can be translated inside the unit cell by one unit and you see that automatically a
mirror gets generated at half position, just like the twofold axes and the 2 1 axis which we
saw. And therefore, we have mirror symmetries at these 3 positions and this now
generates 2 equivalent points, x y z and x y̅ z and I want you to see that even though I
have not written it, it can be considered as the equivalent points.
241
The question is suppose there is an object with a mirror, the object has a mirror in itself
would it generate a special position. This is something I want you to find out yourself.
Having heard the argument so, far you have to find out whether the object can have a
special position if it has a mirror symmetry, can the object with the mirror symmetry
represent on the mirror, in which case what would be the number of equivalent points?
So, this you can figure out yourself its not a very difficult proposition and if the object
lies on the mirror what would be its coordinates?
Suppose it lies at x equals 0 and then; obviously, it is perpendicular to the b direction and
so, it can lie anywhere in the plane of a c and any point in the plane of a c, if it lies
would it be a special position. This is something which you figure out as we go to the
next crystal system.
The next is crystal system is a c glide that can be associated with the primitive lattice as
we discussed earlier, this is a consequence of the translational periodicity. So, again the
diagram is fairly obvious here even though I will go through the description for the
completion is sake. You have the position x y z and now what you do is you now operate
the mirror symmetry. The mirror symmetry will take it to the other side and then it will
not keep quite there it will either move along the which direction it will move? The
direction in which it will move is given by half here; that means, it is moving half the
distance along the c direction. So, in the space group is Pc. Suppose if space group is Pa
242
then it will move half the distance along the a direction. And I want you to generate in
your mind that diagram that equivalent point diagram where the space group is p a which
is also a possibility. The Pc space group has the equivalent point where you generate the
mirror and a half translation along the direction perpendicular to the mirror. In this
particular case it is perpendicular to the b direction.
We can also move the plane of a and c and if you move it in the plane of a and c along
the diagonal then it will become Pn an n glide operation. So, I want you to put some
thought into this, because there is no point in me telling all these space groups and all the
way in which the equivalent points developed, but I want you to think about how Pn can
be generated. And if it this particular diagram I want you to generate for a Pm may be it
will be given as an assignment later by our teaching assistants.
So, find out what are the equivalent points if the space group is Pm. In this particular
case the space group now generates an equivalent point which is x y̅ ½ +z. So, it is the
translation is along the z direction and that is why it is a c glide. If the translation is along
the a direction you will have ½ + x. So, this now generates again 2 equivalent points and
I want you to figure out whether this has any special positions Pc.
Next we go to the mirror symmetry, but now we will change the lattice type and we will
change the lattice type to C. So, now, it is a C centered lattice in a monoclinic system;
that means, for every point x y z you have a point at ½ + x ½ + y z. So, if your object is
up here you have a corresponding point which will correspond to ½ + x ½ + y z. which is
shown as number 3 here.
So, 1 and 3 are related due to the presence of the C centering and now of course, the
mirror operates. The mirror operation will take this point onto 2 a simple mirror
operation and this now point also will add half and half. So, you will get this point and
you see that there is a mirror which is automatically generated in the middle. So, if there
is a mirror the mirror operation associated with 0 we will have mirror at half we will
have mirror at full length.
Now, where is the c glide located? C centering located? The C centering is located at this
point ½ + x ½ + y z.. As a consequence if you now examine the equivalent points 1 and 4
you find a relationship between them. The relationship tells you that there is a glide
plane which is present at one-fourth the position. It is not in the space group the glide
243
plane is a consequence of the centering, because the we have the centering the x y z is
becoming ½ + x ½ + y z. that presence of the ½ + x ½ + y z. and the corresponding
mirror operation relates 1 and 4 by a c glide operation.
So, if you examine 1 and 4 you see that it is a c glide and this particular glide operation is
it a c glide or an a glide? You look at this equivalent points and see whether it is a c glide
or an a glide? It is an a glide because the glide operation is adding half to the value of x
and this is coming as a consequence of the presence of the mirror symmetry, which now
takes the C centered object operates the mirror and therefore, you will get a A centered
lattice associated with this. Of course, the presence of the mirror is half distance and you
see that half distance mirror is shown by a dark line here and therefore, you have the
mirror, when once you have the mirror at 0 you have the mirror at half, you have the
mirror full length and you double up a glide which is an a glide here.
In the case of the C centered lattice with a mirror symmetry, you invoke the presence of
an automatically available a glide. Even though it is a part and parcel it is not a part and
parcel of the description of the space group, because the description of the space group
the dominance is given to the C centering lattice. So, the lattice gets the highest priority
and then it generates the equivalent points corresponding to the lattice, and then the
symmetry operation takes over.
So, the presence of the C centering therefore, has given rise to a possibility of a glide
plane which is at one-fourth this distance. So, at one-fourth disposition you will get an
equivalent point. So, this brings us to the fact that, given the positions of the symmetry
elements given the positions this positions of the objects, we can derive the symmetry
points symmetry equivalence; that means, if this is the point x y z, we can derive the
position of the symmetry position here by looking at the symmetry operations that are
possible.
One is now we took this point 1 took it onto the point 3 operated the mirror plane. So, we
got this. So, if that is how we can generate x y z and ½ + x ½ + y̅ z. now we see that
these 2 are related by a c glide sorry a glide.
So, what it essentially tells us is that, given the diagram like this it should be possible to
find out the equivalent points. The other issue is given the equivalent points like this it
should be possible for us to find the space group. And if you find the space group we
244
should be able to generate the symmetry operation diagram. This is known as the
symmetry operation diagram along with the objects which are shown here, but you can
replace the objects and in international tables what is given is the objects are separately
given in one side and the symmetry operations alone are given on the other side. We will
get a look at it after we finish this discussion so, that we understand the operations of the
symmetry elements and also the presence of the centering of the lattice and later when
we go to the space groups, you will see that the international tables for crystallography
which is an enormous effort put by early day crystallographers.
And eventually now it is undergoing recent changes, it is available online and one can
actually look at 2 possible diagrams. One diagram corresponding only to the symmetry
positions the other diagram related the objects. So, that this complexity that is associated
if you think there is a complexity associated with this, I do not think there is any
complexity anymore associated with this. Because you have just now understood the
entire diagram, we have not only adjusted understood where the objects are coming and
how they are coming, you also have understood how the symmetry elements are located
with respect to displacement.
Having seen that I have given you the Cc diagram. So, the Cc diagram now its a C
centered lattice with a c glide. Now it can also be Ca because a and c are interchangeable
in a monoclinic system because a is not equal to b not equal to c and the two angles
which are at 90 degrees which is essentially considering b as the unique axis will allow
us to have either a or c. So, we can have a Ca or a Cc both space groups are possible and
here the diagram of Cc is given.
Now, it means that we have a C centered lattice with a c glide associated with a c glide. I
want you to think whether Cb is possible cb; that means, the b glide which is along the b
axis. There is no possibility of a b glide along the b axis because in a monoclinic system
b is given always to the twofold unique axis. Since the b is given to the twofold unique
axes it is not possible to have Cb. So, Cc and Ca are both possibilities and in the case of
Ca what we see here is that the object 1 now since the c glide perpendicular to the b axis.
So, the c glide perpendicular to the b axis is indicated here and so, we generate these 2
equivalent points.
245
So, 1 and 2 will be generated by the operation of the c glide. Now we also have the C
centered lattice the C centered lattice will generate point 1 and point 3 same logic as we
used in the case of Cm using the same logic as we used here we generate 1 and 3 by C
centered lattice 2 and 4 are the operations due to the c glide, now c glide being at 0 we
will automatically generate one at half position because of translational periodicity and
we have this as well.
So, we can now bring all 4 points inside the unit cell, I want you to write down. The
equivalent points this is a home exercise, I want you to write down what is the value of 1
what is the value of 2 in terms of x y z and this kind of equivalences, which we have
shown in case of Cm, modify the Cm equivalences to generate Cc. Its not a fairly
difficult job its a very straight forward job so, you should be able to do that. When once
you do that you will see something, very interesting just like what happened here. In case
of mirror symmetry we developed a c glide at one fourth remember a glide sorry a glide
at one fourth this position.
Now, what will happen is because of the operation of the C centered as well as the c
glide you will generate an n glide at one fourth. I want you to find out how this n glide
operation has come and write down those equivalent points as well which will be the
same as the equivalent points that is allowed by Cc.
So, at this time we will just consolidate the diagram here we did in fact, we started from
the twofold symmetry and then we went to the presence of the mirror symmetry, the
second symmetry operation in a monoclinic system and we then looked at the possibility
of having the c glide associated with the primitive lattice, then we brought in the C
centering which is essentially is operating on the mirror symmetry. We find an
appearance of an additional symmetry element which is not a part and parcel of this Cm,
but indeed it in some sense it is a part and parcel of this because of the fact that we C
centering takes x y z onto ½ + x ½ + y z equivalence and they as a consequence of that
we develop the a glide which is at one fourth preposition.
And if we look at this situation where we have a C centered lattice with a c glide then we
generate equivalent points which will result in the presence of an n glide which is at one
fourth and three fourths. So, the additional symmetry elements start cropping up
depending upon how we combine now the lattice symmetry with the point group
246
symmetry. So, the flexibility that is allowed now for the objects which go into these
sports groups kind of restricted, because we it has to satisfy the other symmetry elements
which are also developing. For example, in the case of the Cm symmetry you had the
mirror you wanted a you get a a glide. So, the equivalent points which we generate will
have to also satisfy the a glide condition and therefore, the object is now arrested.
It is like you know putting all of you into a class room and making you sit in your
respective chairs. So, it is essentially restricted by the additional symmetries. So, Cc will
also generate additional symmetry the n glide and therefore, there are more restrictions
on Cc. Now you can think of the situation when molecules crystallize in these kind of
space groups. The molecules when they crystallize in these kinds of space groups will
have to satisfy all the symmetries which are also coming in addition to the existing
symmetry in the point group and the lattice type combination and thereby they are
restricted to those places.
And these restrictions will therefore, tell us which compound can go into which space
group. This leads us to the possibility whether they can a priory predict a crystal
structure. A priory prediction of the crystal structure is also a possibility. It is also a
possibility, but whether we can unconventionally or unequivocally determine the space
group is an issue this is something we will discuss when we talk about crystal structure
prediction.
It is not a straight forward way in which we can predict a crystal structure otherwise you
know we would not be taking this course when doing x ray diffraction experiments to
find out where the atoms are. Because the moment we see the object if we can predict
which space group it goes into the job is all done. So, the solid will be described fully.
247
Lecture – 20
Space Groups 3
So, we have been looking at the Generation of Space Groups and the way in which the
equivalent points arrange themselves in the unit cell. The simplest of the space groups is
P-1; of course P1 being redundant as it has no symmetry operation.
In a triclinic system P-1 is only allowed space group other than P1 and this space group
has a center of symmetry. The presence of the inversion center is shown with respect to
the equivalent points here. So, we have taken the object to be in the shape of a comma, it
could be a molecule an atom ion or whatever. And the plus sign here indicates that it is in
the forward direction and the minus sign indicates it is in the backward direction.
So, we did the generation of the equivalent points found out the presence of centers of
symmetries at different positions other than the origin, one of course, due to translation
along "a", another due to translation along "b" and halfway points getting the centers of
symmetry. And as a consequence of that this symmetry operation which goes outside of
the unit cell can be brought into the unit cell by 1 unit translation along "a", 1 unit
translation along "b". So, it is actually representing two such units to such objects in the
248
unit cell we call them the number of molecules or whatever number of units number of
objects inside the unit cell in this case it is 2.
So, we extended this to the monoclinic system the point group symmetry is being 2 and
2/m. The point group symmetry 2 gives raise to in the primitive lattice 2 cells P2 and
P21. So, the presence of the two fold axes and the presence of the 2 1 axes we discussed,
and we see that these equivalent points get generated as are marked here.
Then, we also looked at the possibility of the centering of the lattice in a monoclinic
system and C2 and C21 get generated. We also indicated that C21 can be changed to C2.
This is just a recollection of the previous class.
Then, we introduce the next symmetry element after the point group symmetry 2; we
have the point group symmetry m. We looked at the monoclinic symmetry primitive and
then we find further and looked at the monoclinic symmetry Pc, with P is the primitive
and "c" is the glide plane. So, then we looked into the mirror symmetry operation with
respect to the C centering and then we had arrived at this point where we had the Cc.
249
Now, so just keeping the logic of arguments going if one looks at this particular diagram,
suppose you are given only this diagram and then the positions of the equivalent points.
In this case of course, both the positions of the equivalent points and symmetry
operations are given. But suppose you are given only these equivalent points in principle.
One should be able to get to the equivalent points, one should also be able to get to the
equivalent points when once we are given the symmetry diagram and that is what you
will find when we go and discuss the international tables later on.
But at this particular moment in this case suppose you are asked to find out the
equivalent points how would you go about it. Of course, I have already marked 1 2 3 4
for your convenience, so there are four equivalent points in Cc. So, the way they are the
objective way in which we go is to take this point which is marked as 1 as x y z,
corresponding to the coordinates x y z then because of first you operate the C centered
lattice symmetry; that means, the lattice symmetry gets the predominance prominence.
So, when once you operate the C centering that means, there is a 1/2+x and 1/2+y added
to the x y z. So, suppose we call 1 as x y z, then we have a point number 3 which is
marked here which is due to the C symmetry operation which is half this is the a
direction, so 1/2+x and then there is a "b" direction 1/2+y, so we get this point. So, these
equivalent points one and three are generated by the C centered lattice.
250
The glide operation which is now coincident with the origin, so that is the glide is
perpendicular to the origin and therefore, we get a glide operation which takes this mirror
symmetry and in the direction coming towards you along the C direction it adds a half.
So, the glide plane "c" now operates the mirror on one and then adds a half.
So, this is the point 2, so we therefore, get this as x and y this is the mirror symmetry, so
we get x -y 1/2+z. So, therefore, we can actually arrive at the equivalent point the first
equivalent point being x y z, this is 1/2+x 1/2+y z and this is x -y 1/2+z.
Now, what about what happens to this? This particular point this particular point can be
translated by 1 unit and if you translate by 1 more unit it goes out of the unit cell so, but
in this case if you translate it by 1 unit, it comes to this point, half plus this object. Now,
what a what you see here is that, there is a by definition we are never there is a symmetry
operation at 0 and the symmetry operation 1 unit translation the same symmetry
operation will develop at half position.
So, using this half position the c-glide we can generate its corresponding c operation c-
glide operation, so that means, this now goes into a mirror symmetry and then it adds on
half in the direction of C and therefore, we get to this point 4. So, one can therefore, in
principle given this picture write down the equivalent points.
And in the next slide I have shown what are those equivalent points.
251
Those equivalent points are x y z, x -y 1/2+z, x+1/2 y+1/2 z. So, as you see that this is the
c-glide operation and this third is the C centering operation and then the c-glide
operation on this object which is due to the C centering we now operate the C sorry this
is due to the c-glide we operate on the c-glide the C centering operation. So, now, this
goes to x+1/2 -y+1/2 and z+1/2. So, these are the 4 equivalent points. So, the take home
lesson here is that given this diagram in principle we should be able to write the
equivalent points.
So, this the next example we can take up is given the equivalent points like here, given
the equivalent points as is given here you will be asked to determine the space group. So,
if one ask that question in this particular case it is a very simple example, in principle
you should be able to identify the space group.
So, you see here point 1 and point 2 they are related by what? They are related by a
twofold axes perpendicular to the y direction. So, you have a twofold symmetry. And
you also do not have any 1/2+x 1/2+y or 1/2+y 1/2+z or for that matter 1/2+x 1/2+z, so
that means, that the centering is not seen.
Generally, since the monoclinic symmetry is C centered you should have equivalent
points 1/2+x 1/2+y and that since that is not appearing this is a primitive lattice. So, the
first conclusion is examine the equivalent points, find out if there are any half
translations associated with x or y or in this particular example of a monitoring symmetry
and then you see and conclude that it is a primitive lattice.
So, we have found that it is the primitive lattice. It has a twofold symmetry. Now, if you
look at the point number 2, it is a mirror operation perpendicular to the "b" direction. So,
we have a twofold symmetry along the "b" direction a mirror plane perpendicular to the
"b" direction; that means, it has and that also has a center of symmetry equivalent point
which is x y z going to -x -y -z.
So, we are now looking at a central symmetric space group. So, the way logical way to
go about when equivalent points are given is to first see whether it is a the like type of
lattice. So, suppose it is a primitive lattice like here, identify the primitive lattice. Then
look for the existence of x y z, -x -y -z that will tell us that it is a center of symmetric
system, and then you look for the operation of the twofold in the operation of the mirror
symmetry as we see here. So, if this is very easy to conclude the space group is P2/m.
252
So, when once we conclude the space group is P2/m in principle we should be able to
generate at equivalent point diagram, and also a diagram for the symmetry positions.
This is something which you will have to practice taking some examples of other space
groups, this I have taken an easy example for you to illustrate the whole thing and that is
given here. So, you see that there is a twofold axes about along the b direction.
So, the twofold notation is here the twofold notation is here because this is along the "b"
direction we are giving the ab projection. We are looking down "c", so this angle is 90
degrees it is a monoclinic system primitive lattice and then we have these four equivalent
points around the centre of symmetry. So, 1, 2 which is generated by the twofold axes
symmetry, then the mirror symmetry from here goes there and then the central symmetry
operation takes it over there.
So, therefore, these are two and these three of them can be brought inside by translational
symmetries as we discussed before that is point number 4, we will go undergo a
translation along the a and translation along "b", we get this and so also this one
translation will bring it here, this is one translation will bring it there. So, there are four
equivalent points in the unit cell.
We also have marked the mirror symmetries as the dark lines here. They appear at the
edge cell edges as well at the central point. And the axes is about the y direction, so no
other thing happens except that one once the centre of symmetry develops here there will
be a centre of symmetry up there.
Now, because of the presence of the centre of symmetry up there this will also see now a
2/m operation and therefore, we have these kind of equivalent points. I will leave it to
you as a home exercise to find out the presence of special positions in this space group.
Since, we have discussed the special positions in detail in the earlier classes it not be
very difficult, but then it will need something here why or do or special positions
allowed in this space group, if so how many of them are allowed. And when we actually
go into the examination of the international tables entries you will see that, all these
things are given in the entry and it is just a question of interpretation of the entries in the
right context.
So, this therefore, now brings us to the understanding of two of the symmetry elements.
In fact, we today we have seen the Cc the presence of the mirror point group symmetry,
253
we have also seen the presence of the presence of two as well as a perpendicular mirror.
So, this now this space group analysis now tells us about the fact that we are now
looking into the center of inversion related space group. All the earlier space groups Pc,
Cm, Cc belong to the non-center of symmetric space group.
So, what really happens is that when objects get into these kinds of space groups there is
always a possibility that they will have a homo central non-central symmetric presence in
this structure. Some of them for example, in the case of an organic molecule they may
even have a asymmetric carbon.
But it is not necessary that a asymmetry carbon should be present in these three
symmetry systems because we have a mirror. So, the presence of the mirror can always
invert a central symmetric or an asymmetric carbon as well. So, what we have to look for
is the fact that there is no -x -y -z coordinate system, in these three space groups and
therefore, these three space groups belong to the non-centre of symmetric space groups.
Crystals have a tendency to prefer and crystallize in a given space group depending upon
their symmetry as we already saw. The molecular symmetry decides what kind of a
crystal symmetry it should occupy. So, having kept that in mind we will now go further
and discuss the presence of the 2/m symmetry. We have already seen the one with the
primitive system, now we have to see the combination of the twofold axes with the
mirror.
254
Now, when we are talking about the presence of the twofold axes with the mirror
something very interesting develops here. So, now, you take for example, this is the "b"
direction and your twofold axes is along the "b" direction as is indicated by the half
arrow. And at the origin we have an intersection of the 2 1 axes and the mirror. So, this is
the origin O at which position we have the intersection of the twofold axes on the mirror.
Now, the presence of the twofold axes on the mirror generates equivalent points and we
have taken a one dimensional, 1 unit example along the "b" direction. You see that the
operation of this the mirror operation on this particular point will take you over there.
And the 21 screw operation will go up here twofold rotation and a half translation.
And then the twofold rotation and half translation will bring this object which is now 1
unit translated of that object, and these two are now you see that there is a centre of
symmetry which develops at this point. The point at which the presence of the center of
symmetry is found in the 21/m system is one-fourth removed because the mirror repeats
itself at half position along the axes. So, the centre of symmetry repeats at one-fourth.
The convention in crystallographic nomenclature is not to have a center of symmetry if

there is a center of symmetry in the system, it should coincide with the origin that is
nomenclature and therefore, the it is a rule that the centre of symmetry becomes the
origin. That means, we have to now shift this O to this position. When we shift this O to
this position you see that we now develop a one-fourth position mirror.
So, the mirror now suppose, this is our origin the mirror now comes at one-fourth
removed from the centre of symmetry and then we will also have another mirror which
comes at three folds as a consequence of the centre symmetry being present at the half
position.
So, let me repeat. If you have the 2 1 fold and the mirror intersecting with each other the
presence of the mirror symmetry and then 21 screw axes perpendicular to that you are
shown here in the direction of the "b" axes. The first operation is the 2 1 screw axes, so
you take this rotate by 180 degrees and translate by half the unit along the "b" direction
and you will get this object. You see that the object now is going down and this object is
on the top.
255
And this object now again sees the centre of symmetry, sees the mirror symmetry here
and there is a mirror reflected operation. There is a mirror reflected operation here, there
is a mirror reflected operation there at the half position and also there is a full position
mirror reflected objects. So, you have 1 unit translation along this direction which takes
this object this object sorry, this object onto that one and the mirror symmetry operation
brings it back in. So, these are the four operations which we get, four symmetry positions
we get inside the unit cell along the "b" axis.
The fact that this happens brings in the presence of a center of symmetry invokes the
presence of a center of symmetry at one-fourth removed from the origin. The origin
being the position where the 21 axes and the mirror plane intersect.
The crystallographic nomenclature and the crystallographic follow up of rules tells us

that whenever there is a centre of symmetry the centre of symmetry should be at the
origin 0 0 0. So, what we therefore, do is, we take this position O and move it to this
origin.
So, now O is in this particular position. If O is in this particular position there is no other
change 21 screw axes is going through this centre of symmetry at 0 0 0, so the presence
of the 21 screw is not altered because it is going through the centre of symmetry at 0 0 0.
What is altered is the position of the mirror plane the mirror plane now moves by one-
fourth.
So, if you have an operation which has a translation involved component, along with that
a component which is no translation involved component the no translation involved
component moves away by one-fourth from the origin. The origin now is identified with
respect to the presence of a center of inversion.
So, the center of inversion or the center of symmetry in the crystal system is now at 0 0
0, mirror comes at one-fourth and then you have the 2 1 screw are coinciding with the
center of symmetry. So, this therefore, gives us the 21/m symmetry and this is where we
have to be cautious in deciding the equivalent points, because the equivalent points now
will change their faces in the sense that if we have the 2 1\m operation here the mirror
operation will take using this as the origin. It will take x y z to x -y z.
256
On the other hand, the mirror operation which is now here if this is the 0 position, then
this particular mirror operation the coordinates will change. The coordinates will change
in such a way that the presence of the one, the mirror at one-fourth away from the center
of symmetry needs to be recognized. So, this is as far as the 21/m symmetry is concerned.
Now, we will go and discuss the possibility of 2/c. Again, you remember that c has a half
component, half translation component. So, if c has a half translation component then
what is the way in which you infer where are the axes that will be located. For example,
the twofold axes where do you think they will be located. The twofold axes now will be
coincident with the origin.
Let us examine this carefully. Notice, that now instead of showing the "b' direction and
showing the 'c" direction, this is very important. So, this shows now the 'c" axes not the
"b" axes. See for all descriptions of the monoclinic systems we have taken the "b" axes.
Here we are showing the "c" axes keeping in mind that the unique axes is still "b". The
unique axes is still "b" in this example as well.
So, if you now look at the way in which the symmetry operations get disposed you will
see that is a twofold axes, which will coincide with the origin and the presence of these
twofold axes will generate these two points. Now, this point 1 unit translation will go
there and this point 1 unit translation will come here. This will tell us that there is a
twofold axes at the half position as well
Now, the fact that there is a there is a twofold axes here at 0, and the twofold axes at half,
under twofold axes again at this particular position will generate the c-glide operation
which is now remember this is the c axes. So, the half translation is along the c axes. So,
you mirror reflect it and translate by half unit, and this point gets generated.
You notice that there is a relationship between this point and this point, and that is the
presence of the inversion center. So, the presence of the inversion center is automatically
invoked. And as we discussed in the previous example the presence of the centre of
symmetry now should become the origin. So, we shift the origin to 0 0 0, to the presence
of to the centre of symmetry. So, the O is now moved over here.
So, what therefore, happens is the, non-translation involved symmetry element. What is
the non-translation symmetry element that is involved here? It is the twofold axes. So,
257
the twofold axes now moves by one-fourth. So, this moves by one-fourth away from the
centre of symmetry that will also generate the symmetry at threefolds.
So, as is written here 21/m with screw in "b" direction and mirrors at b is equal to 0 and
half. The mirrors are now at 0 and half as is shown in the diagram, but we now know we
have to shift it to the centers of symmetry. So, the centre of symmetry up here now has a
mirror at one-forth and three folds. And so also the presence of the c-glide invokes the
moving of the twofold axes by one-fourth from the centre of symmetry. So, the non-
translational symmetry two is moved by one-fourth.
Now, what is the consequence of this when we generate the space group diagrams? And
in what way this influences the presence and description of the equivalent points. There
are two ways in which we can argue with result. See the one of the things is that we have
this condition of the definition of a space group. When we say the points belong to a
space group then when we keep operating the symmetries we should get back to the
original point.
Now, if you take this for as the origin the intersection of the mirror and the 2 1 you are x y
z will see both 21 and mirror and apart from that operation, we do not have any further
operation at that particular point with respect to 0. The fact that the center of symmetry
develops at one-fourth position is not going to give us back the x y z, in other words if
you take this point as x y z you operate the 21 symmetry on this one, you will go to that
point which is which is what -x -y -z. So, you get the -x -y -z you get the x y z and the fact
that this has happened is because of the operation of the 21 screw.
Then we have x y z and this point which is the mirror symmetry where which will take us
to x -y z. So, the presence of the symmetry at x -y z and the at the origin and the presence
of the symmetry 21, if we keep operating again one of them or operate them together, we
will not get back to -x -y -z ; -x -y -z will not get invoked and therefore, the space group
points will not complete the definition of a group.
So, mathematically in order to complete the definition of a group, it is required that we

move the centre of symmetry to the origin. When once we move the center of symmetry
to the origin, we already know the x y z and -x -y -z are two equivalent positions with
respect to the centre of symmetry, and the rest of the equivalent positions can be
generated as we can see when we go to the next discussion.
258
So, since this part is a little tricky and this is not explained in any detail in any textbooks
except for this one which is the textbook of Stout and Jensen. I thought that. In fact, in
Stout and Jensen there is a mistake on one of these equivalent points as well which we I
think we have corrected and therefore, it one has to be careful in looking at some of these
you think the correction is up here, in this particular position.
So, what we will do therefore is to see that whenever we have a translation involved
symmetry that is associated with the space crew along with another symmetry operation
which is a non-translational symmetry operation or for that matter if it is a symmetry
operation. Suppose, we go to 2 1/c we will see that in a little while, that 2 1/c will invoke
translations of both the symmetry elements.
So, the take home from this particular slide is whenever you have a non-translation
involved symmetry element that particular non-center the non-translation involved
symmetry element for the fact that it is a center symmetric space group because 21/m will
translate itself to a P2/m point group. Since the point group is 2/m, it will represent a
center of symmetry.
So, the presence of the centre of symmetry invokes the requirement that we should have
a center of symmetry in the system. The fact that we have a center of symmetry in the
system we have to associate that with 0 0 0. So, when once we do that operation what is,
what we have to remember is whenever there is a non-translation involved component
that is the one which moves by one-fourth and the translation component will stay along
with this center of symmetry.
The same is true in the case of 2/c you see that the twofold axes which is going
perpendicular to that stays along with the center of symmetry whereas, the sorry the glide
plane stays along with the center of symmetry, the non-translational symmetry element
the twofold is moved by one-fourth.
Now, this two observations have to be incorporated when we now discuss the space
groups which involve non-translational periodicity associated with them. Space group
like P2/m, space group like P2/c, space group C2/m which is also C21/m as we have seen
C21 and C2 are one and the same, then of course, C2/c. We will also have the
combinations of both these having translational symmetry because a point group 2/m can
259
generate a centre of symmetric space group 21/c. So, we can have two additional space
groups P21/c and also C2/c.
So, we will have to now examine these and that will in fact, complete all the
requirements of a monoclinic space groups. So, as you see from a single center of
symmetry when we go to the presence of a minimum symmetry of a twofold and a
maximum symmetry of 2/m in a monoclinic system, three point groups now will
generate several space groups.
So, the number of space groups are now increased enormously compared to the two that
is associated with the triclinic symmetry. Therefore, we have the condition here that the
21/m operation will now shift the mirror plane by one-fourth and the 2/c operation will
shift the twofold symmetry by one-fourth.
260
Lecture - 21
Space Groups 4
So, the next one is to see how these now translates themselves to the presence of the
newer space groups.
The next space group we discuss will be P21/m. So, what is the take home lesson in
21/m? In the case of 21/m, the mirror symmetry is the one we shift by one fourth. So, you
see in this diagram the mirror symmetry if you look at only the symmetry diagram, the
symmetry elements diagram we draw this unit cell and draw the centres of symmetry at
the origin then we know that the twofold 2 1 screw axis will coincide with this. So, we
write the 21 screw axis to coincide with the direction of the origin along the "b" direction
because the unique axis is b and we have shown the "ab" projection as before as always.
So, we see that because of the presence of the 2 1 screw and the center of symmetry at this
position, the same will repeat at half position and then of course, the full unit cell. In
addition, the centres of symmetry will develop at the half positions along the "b"
direction. Notice that since there is only one axis of rotation in a monoclinic system. We
should remember that the symmetry operations are restricted to around the y axis which
261
is the unique axis. So, we do not have any other symmetry elements which will come in a
monoclinic system. So, there is only one unique axis about which the symmetry
operation take place.
So, in other words if it is a twofold axis operation or a mirror alone or a combination of 2

and m, whether it is a 21/c operation all these symmetry operations are restricted to
operate only along the y direction and that is why it is called a unique axis along "b" in a
monoclinic system and that b axis therefore, is the one which shows the presence of the
twofold along with it.
Just as we saw in the previous slide, that the mirror comes one fourth removed from the
origin since there is a mirror that is generated here another mirror will generate at three
folds. So, the equivalent point diagram now. So, these are the this is the symmetry
diagram. In the symmetry diagram therefore, we have the centre of symmetry indicated
at all the corners of the unit cell as well as at the half positions along "a" and "b" the
presence of the 21 axis is illustrated along the "b" direction and also at the cell ledgers
and also halfway distance along the "a" direction and the mirror symmetry is now moved
away from the origin the intersection of the origin and the 21 screw axis by one fourth.
So, the mirror now appears at one fourth and three fourth and therefore, the symmetry
operations now become slightly different. So, if you take the point number 1, operate
now the presence of the 21 screw axis. The 21 screw axis will take this rotate by 180
degrees translate by this unit. The rotation of 180 degree is as usual -x and -z and the
addition of half unit is 1/2+y. So, you get to this point number 2, one unit translation of
the point 2 will bring it inside the units. So, we have generated 1 and 2 inside the units
cell.
Now, the next operation is the mirror operation. The mirror operation is not now
associated with the y equals to 0, but we have to move it by one fourth. So, the mirror
operation is moved by one fourth along the "b" axis and because of the fact that mirror is
one fourth removed you will add for example 1 now goes to the point 3 which is now x,
y, z going to x instead of x -y z it acts on a half unit with that 1 and therefore, we get x -
y+1/2 z as the third equivalent point. Why do we get this half? We get this half because
the mirror has moved by one fourth and there is a mirror at three folds as well which is
going to generate the further operations in the other unit cells.
262
So, you see now that we have point number 1, point number 3 and this point which was
translated all three in the unit cell. So, there must be one more, that one more is
automatically the centre of symmetry operation because if you now take point 1 operate
the centre of symmetry you go to point 4. 4 now can be translated by 1 unit in the a
direction and 1 unit in the b direction you will get this point. You notice now that the unit
cell is up here and you have 1, 2, 3 and 4.
So, there are four equivalent points again in P2/m just like P2/m which you saw just
now. P2/m has all the four around these centre here and because of the presence of 2 1
axis and the mirror now moving to one fourth position you will get the different space
group identification with respect to equivalent points, now the equivalent points are the
ones given below
Notice that you operate again -x -y -z, you operate the centre of symmetry again you will
go here, operate the mirror symmetry you will go there and operate the twofold you will
21 screw axis you will go there. So, the symmetry is still maintained and therefore, the
property of the group is still maintained that is the requirement. So, P21/m therefore, is a
space group we have where we have moved the mirror by one fourth and the moment of
the mirror by one fourth is the one which kind of indicates the presence of the translation
involved symmetry element. This is only happens; this only happens in crystal systems
because we are now required that we have a unit cell and therefore, a lattice in order and
263
the object should e confined within the crystal in order to have this symmetry. He said
start. So, I started.
So, the space group is P2/m and the equivalent points clearly indicate now the presence
of the mirror symmetry which is removed by one fourth. So, this operation which is
operation number 3, symmetry operation number 3 has a half translation associated with
-y that is the one which invokes the presence of the mirror symmetry. So, these are the
four equivalent points in case of P2/m.
We will now shift our attention to the space group P2/c. Now the moment we see these
space group as P2/c, we know the lattice is primitive the presence of 2/c in fact the full
symbol for this space group should be given as the symmetry along the a direction or the
symmetry along the x axis, symmetry along the y axis and symmetry along the z axis.
So, it should read P1 2/c 1, but it is normally the 1 symmetry is not indicated. So, when
we say P2/c it tells us two things, one is that whatever symmetry operations we are
having the 2/m point group symmetry is along 1 single axis and this presence of the
twofold indicates that it is monoclinic symmetry.
Monoclinic symmetry is the one which will have the twofold symmetry alone along the y
axis along a given axis and that we take as the y axis. So, the symmetry operation 2/c
indicates that we are dealing with a monoclinic system. So, if someone gives you the
space group in principle you should be able to automatically find the crystal system. You
not only find the crystal system of course, the moment you see the lattice identification
you see the lattice also the Bravais lattice that is associated with it. The examination of
the symmetry that is given here convert it to a point group symmetry.
Suppose you are giving 2/c, the corresponding point group will be no translation
involved that is 2/m. So, it is a centre of symmetric space group and we conclude that it
is a 2/m point group we have the primitive lattice and therefore, this is a monoclinic
system.
So, we say now uniquely the looking at P2/c we identify it is a monoclinic system, it is a
primitive lattice and then we now can also say the operation of 2/c just like 2/m operation
will give rise to 4 equivalent points. So, by just looking at this we can anticipate how
many equivalent points come to this in general and in the unit cell and also we tell us
about the way in which this operation takes place.
264
We have already seen that whenever there is a non translational component that is
associated with the symmetry operation that particular symmetry which is not involving
the translational symmetry that is the twofold will have to move by one fourth from the
origin. n order to accommodate the requirement of 2/m symmetry which is the presence
of the centre of symmetry.
Just to repeat this presence of the centre of symmetry we will necessitate the symmetry
which is not having a translational component to be associated with a distance movement
of one fourth along the direction perpendicular to the axis to the unique axis. And
therefore, you see that if you look at this twofold symmetry the one fourth symmetry is
indicated very clearly here; that means, the twofold is not with the origin 0, 0, 0, but the
twofold symmetry is one fourth removed, that is why it is coming up by one fourth.
So, the twofold axis will pass somewhere there it one fourth if this is the "c' direction and
of course, it can come at an angle beta. So, the symmetry operation therefore, now will
be these are the equivalent points which are given you see that the equivalent points are
very similar to P2/m if you go back and see the P2/m the symmetry points are all
clustered around this origin.
Once again the clustered around the origin except that the point 2 and point 3 you notice
here we will see the twofold axis one fourth from here and as a consequence they move
by half unit. So, these 2 positions therefore, have a half translation in a direction which is
perpendicular to this plane of operation and therefore, we it is not really perpendicular it
is at along the c axis. So, we have to remember that in a monoclinic system the third axis
is not at 90 degrees right
So, this is now, therefore, going along the c axis it is half translation along the c axis.
Please remember the half translation along the c axis will not involve a 90 degree angle
which is very important in monoclinic systems. So, it is moved by half the translation.
So, what is important is the unit cell dimension. So, whenever there is a glide operation
or a 21 screw operation what is important is the dimension of that particular axis not the
orientation of that axis about which the operation takes place.
So, we therefore, have the mirror operation which is one fourth removed, and then we
have the symmetry which is generating the c glide system on the equivalent points are
given below. So, you see here that the 1 to 2 operation involves the twofold operation,
265
but the twofold operation is now at one fourth distance along the c direction. So, it
become -z+1/2. So, it is -x y -z+1/2. -x -z will tell us that it is a twofold rotation the
presence of half will tell us that the twofold axis is located one fourth along the c
direction.
Now, if you look at this point x y z and see how point 3 gets generated, the point 3 is now
a mirror symmetry operation and it is not just a mirror symmetry we see that there is a
translation of half involved in this and so, you see x, -y+1/2 z you see glide translation
symmetry is added on here and so, you get this presence of the c glide. If you just look at
these 2 then it represent straight away the c glide symmetry elements. And then of
course, because of the fact that it is a 2/m point group symmetry P2/c will also have x y z
and -x -y -z the centres centre of symmetry.
So, a full description of the P2/c is given here, but what is normally required is that we
have to give this symmetry information separately, the symmetry operation information
separately, equivalent point positions separately and that is how it will appear when we
examine the international table a very soon. The international tables for crystallography
is the handbook that is required to identify all the symmetry elements that we can
generate in a space group taking into account the requirement of translational periodicity
requirements of moving the axis or the aeroplanes by one fourth distance in this
particular set of examples.
As we go to higher systems where we have three fold symmetry, we have four fold
symmetry, we have six fold symmetry the amount of translations can be lower than that.
In other words it start exactly half it could be lower than that. In fact, we will see one
such case in a monoclinic system very soon and that will tell us that the movement of the
axis need not always be half it can be even one fourth and that is the symmetry which
you will see in the case of C2/c we will go there in a few minutes from now.
So, what we see here is therefore, the case of the involvement of the translation involved
components and the way in which the symmetry operations occur by shifting the non
translation involved component in these cases by one fourth as we have seen here. Now
we come to the case of P21/c is by for the most important space group from the point of
view of particularly small organic molecules. Most of the small organic molecules which
266
do not have a symmetry they tend to go in to P21/c, because P21/c appears to be the most
comfortable space group to lie down in that sense.
Suppose you are given a choice of this chair on which I am sitting and a stool on the
across it is across the lecture hall and then a sofa across at the back of the hall you
obviously, prefer the sofa. So, the organic molecules they want to relax keeping, their
flexibility intact they want to relax and when they want to relax the flexibility is still
allowed with a relaxation.
So, when you are sitting in a sofa or when you are sitting on the chair it does not matter
you will still be moving around right, but the movement around this sofa is much more
comfortable than the movement in this chair because you are restricted here in a sofa you
can spread your wings. So, effectively organic molecules spread their wings with the
flexibility that is allowed to them and that is why P21/c is the most favourite space group
for most of the organic molecules particularly when there is no asymmetric common in
them.
It is not necessary that organic molecules must crystallise in P21/c, but that is a preferred
space group. Any other crystallisation effort can take it away from P21/c will give rise to
systems which will have a higher energy located with them. So, it could be disorder for
example, if you take it to a higher system like a tetragonal order, cubic system the
organic molecule tends to be disorder and that is an issue we will discuss when we
actually look at the details of structures and how intra and intermolecular interactions are
important to study at that time we will look at the energy tips that are involved. So, P21/c
therefore, is a space group we will study in enormous detail
So, what are the operations here? The operations here are 2 1 screw and the c glide
simultaneously at a intersection point. So, you quickly realise that 2 1 and c will have to
move from each other. They cannot be associated and at intersect with each other
because the point group symmetry happens to be 2/m. Because the point group symmetry
is 2/m you need to have a centre of symmetry and the presence of the centre symmetry
and the requirements of the definition of a group will invoke for every given x y z across
the centre of symmetry located 0, 0, 0 you will have to have -x -y -z.
So, one of the advantages is that, in this case we already have the 2 equivalent points
already identified in P21/c. One of the questions which I would like to pose while to you
267
of course, you have already shown it here there only 4 equivalent points even in the case
of P21/c and these four equivalent points can be derived based upon the diagram which is
given here.
One could have spent some time on checking out what happens if 2 1 and c intersection
point can be taken as an origin. This could be probably an assignment in your course
work which will be given by TAs. What will be the equivalent points if 2 1 and a; 21 and
c intersect at the origin, generate the equivalent points and see for yourself and convince
for yourself that the presence of centre of symmetry gets invoked just like what we have
seen in these 2 examples and see that the presence of the centre of symmetry is not then
at the origin, but somewhere else.
In fact, that somewhere else is very interesting I want you to work it out and find out
where is that somewhere else and that will be very interesting to find, that somewhere
else will be not at half, half it will be at one fourth, one fourth and this is something
which I thought I will derive, but I was thinking that we can as well derive it when we
take a little more complex space group like P21 21 21 where there are 2 twofold axis and
21 screw axis do they really intersect with each other that would be a more exciting
discussion than the discussion with P21/c.
But with P21/c I thought a good assignment would be to see what happens when the 2 1
and c are coincident with the origin. So, that would that could be given as a home
assignment or a assignment in the course to carry marks as well this is this something
very important. So, it can carry remarks as well.
So, now let us examine the situation where we have taken care of this with the fact that
the definition of 21/m and 2/c convinced us that whenever we have the centre of
symmetry the centre symmetry should go to the origin. So, we decided to put this at the
origin. So, if you put this at the origin then we have the point number 1 now goes to
point number 2 which is the 21 screw operation, now the 21 screw operation is coincident
here with respect to the presence of the twofold 21 screw, but you see that it is now
removed by one fourth.
What is that in this particular case also it was removed by one fourth and we argued that
it is due to the presence of translation associated with c; c glide here also we have a c
glide and therefore, we move the 21 screw axis one fourth away from the origin. So,
268
when we move this one fourth away from the origin then you see the 2 1 screw axis will
develop at one half this position again one fourth removed and so on.
So, the symmetry operation 1 therefore, we will take this position 1 to a mirror position
and then half translation. So, this half translation will generate this point 2 which is half
minus ok; that means, it is below. So, what would be the equivalent point? If there was
only a 21 screw axis, the equivalent points would have been -x 1/2+y -z.
Now, along the c direction we have to move by the axis moves by one fourth and
therefore, the position of the equivalent point moves by half because the next operation
on this particular symmetry element should take it to the next unit cell and it should
overlap with 1 that is the requirement of a definition of a group and therefore, we see that
the equivalent point now is -x -z ; -z also gets a half added to it along with the 1/2+y.
So, it is fairly straight forward to understand P21/c, the presence of 21 as well as c will
move both the axis by one fourth, one fourth along the direction which is perpendicular
to the ab plane in case of the 21 screw axis and in plane with respect to the c glide.
So, the c glide now as you have marked here we will move by one fourth. So, c glide
moved by one fourth the 21 screw moved by one fourth. So, what happens if you move
the c glide by one fourth. So, the point 1 now has to generate point 3; the point 1
generating the point 3 now will have a mirror half translation. So, it should have been if
that was to be origin it would have been what? It would have been 1/2+x -y and z.
Now, you notice that we have moved it by one fourth along this direction and therefore,
you have to add another one fourth to the z value. So, you have therefore, you have to
add half to the -y value and 1/2+z is the c glide operation. So, x and 1/2+z is the c glide
operation, -y operation now gets added on a half. So, therefore, we have x -y+1/2 and
1/2+z . The fourth equivalent point; obviously, gets generated because of the presence of
the centre of symmetry which is shown up here.
Once again we see that the translational periodicity will bring this into this position the
translational periodicity here the. So, point 1 and point 3 are already located here the
twofold operate the 21 screw operation has gone outside of the unit cell we add one
translation. So, it comes inside and there is a development of the one fourth c glide along
269
with that there is a development of the c glide. So, the c glide at one fourth and three
fourth will invoke the formation of four equivalent points inside P21/c.
So, anything further about P21/c we will see in a few minutes. What we will notice here
in this case of P21/c is that if you have a object which is located at this particular point,
you have the unit cell that is available to that particular object let us say it is a molecule,
now you see that there is a half translation to generate point 3 and there is also a half
translation to generate this point, because you are adding one fourth along the direction
perpendicular to this and therefore, the molecules are as I mentioned well spread out
inside this unit cell of course, you the unit cell dimensions now will adjust themselves so,
that the flexibility associated with the molecule can also be created in.
If you look at the literature and if you analyse the Cambridge structural database and
look at how many P21/c structures are there in a recent evaluation it is found that
something like 38 percent of the total more than that in fact, the 38-42 percent I think of
the all organic compounds all so, far go in to this space group the reason being that many
of the organic compounds have an asymmetric carbon.
The asymmetric carbon involved a components will always go into a space group P21 21
21. So, that is why I mentioned the space group now itself when I say P21 21 21 what is
the take home lesson? The take home lesson that now we have 3 intersecting twofold
axis; one along x, one along y, one along z; that means, that it has to be 3 directions
which are at 90 degrees with respect to each other alpha, beta, gamma are 90 degrees
with respect to each other; obviously, there is no restriction and you have a twofold axis
on any axis and therefore, a is not equal to b not equal to c. So, a not equal to b not equal
to c alpha beta gamma 90 degrees will define an orthorhombic system and therefore, the
space group P21 21 21 is in the orthorhombic system and because of the fact that it is in
the orthorhombic system we will have three 21 intersecting axis.
Now, the question comes up if you use the same logic as we have used in case of P21/c
where we have shifted the axis by one fourth and one fourth axis as well as the plane, if
there are 3 intersecting as axis what we do. Because the Eulers theorem tells us that if
there are 2 points which are intersecting at a point, the third symmetry operation will
automatically get generated, but that is third symmetry operation with respect to the
point group will have to become a twofold.
270
So, if you have a let us say two 21 axis intersecting we will discuss that when we actually
derive the equivalent points of 21 21 21 at a later date, but what really happens is if you
have a 21 screw axis and another 21 screw axis intersects with that based on Eulers
theorem the third invariably has to be a twofold axis. So, in which case we will have a
space group P21 21 2 or not 21 21 21 then how can we generate the space group 2 1 21 21
which forms the another 38 percent of the total number of organic molecules in literature
which have in a asymmetry carbon they go in to 21 21 21 preferably.
So, how does the chart 21 come in violation of the Euler's theorem? It is not violating the
Euler's theorem, the Euler's theorem has to be modified in fact, there is a corollary on
Euler's theorem which tells that if there are such situations where the two intersections
are generating a translation free component the if there is a translation component
involved then we will have to separate them from each other.
So, the origin now has to be arbitrarily fixed, but it has to be fixed in such a way that
effectively the three axis intersect if to generate a 2 2 2 symmetry from the point group
point of view. So, translational periodicity that is involved in crystals is something which
we will invoke in order to have 21 21 21; that means, only in crystals you can have the
space group P21 21 21 but what is very beautiful is that the organic molecules prefer to go
in the space group 21 21 21. Organic component go into the space group P21/c otherwise
and P21/c again as we have seen here is a fairly straightforward set of the equivalent
points here in this case of course, it is only one single axis and so, there is no question of
invoking the Euler’s theorem because the full symbol of this space group is P21/c. So,
that way we have a situation where we have the three systems P2/m, P2/c and P21/c
having translational involved components. How many translational involved components
come up is discussed here.
So, in a nutshell what we will see now is that we have discussed the primitive lattices
associated with the 2/m symmetry. We now have to worry about the presence of the C
centred lattices. So, if we now think of a c centred lattice we should have a space group
C21/m, we should have a space group C21/c and we should have a space group C21/c and
also C2/c.
We have already seen in the earlier class that C2 and C21are the same it is just a question
of shifting the origin by one fourth and therefore, we will not have all those systems
271
which I mentioned as we will see in the next few slides that the systems which we can
have are limited. So, the wherever there is a c operation coming up one once it is the C
centred lattice the 21 disappears and we will have only C2 and therefore, we have C2/c
and then C2/m we will not have C21/m and that is something which we will have to
discuss as we go further.
So, so, far what we have done is to see the way in which the space groups formulate
starting from the triclinic system to the monoclinic system and in the monoclinic system
we have examined the occurrences of non translation involved components the screw
axis as well as the glide planes.
Screw axis and the glide planes come only in the monoclinic system and definitely not in
the triclinic system because triclinic system has no axis of rotation that is something we
should remember and so, the triclinic systems have only 2 space groups whereas, there
are several space groups which developed in the case of the monoclinic system.
Further, we have also seen that the presence of the translation involved components will
involved translations of the individual components depending upon the nature of the non
translation environment participation in the space group.
272
Lecture - 22
Additional Information on Space Groups
So, from P21/c now, we will go to the other space groups which are possible in the
monoclinic symmetry.
I have indicated 2 space groups here, I given the equivalent points diagram overlapping
the symmetry points. So, I want you to examine this equivalent points and the location of
the symmetry points and make a guess of what could be this space group. The equivalent
points of course, are already given here. So, making a guess of the space group is not
going to be very difficult if we follow all the rules which we lead out in the case of P21/c
for example.
So, here is a case where you have a twofold axis associated with the centre of symmetry
and 1, 2, 3, 4 are marked here the four equivalent points are marked here and four other
equivalent points are marked at this point. Now this refers to half along y and half along
x. So, this therefore, now indicates the addition of 1/2+x 1/2+y tells us this is has to be a
C centred system. So, it is monoclinic and C centred.
273
So, the four equivalent points are surrounding this particular point and there is a twofold
axis coincident with this and a mirror plane which is coincident with the origin. What is
very interesting is to see that apart from the presence of twofold and this we also develop
a glide plane at one fourth that is because the if we work out the relationships between 1,
2, 3, 4 and 5, 6, 7, 8 we will see the presence of a one fourth glide plane operation. So,
what is the space group?
We have an indication that the lattices C centred. There are no marks of shift of the axis
with respect to this position, the twofold axis are coincident with the origin and there is a
mirror plane which is perpendicular to that. What is very interesting is that you also
develop a 21 screw axis between this point and this point which is halfway removed that
was the centres of symmetries you have the 21 screw axis also located. So, the at the
outset the diagram looks very complicated.
So, many symmetry elements are shown here, for example the 2 1 screw axis is shown,
the glide plane is shown, the mirror plane is shown, the twofold axis is shown. So, how
do we discern that the space group is whatever it is? One conclusion we immediately
came by looking at these four and these four is that it is a c centred lattice so; that means,
we should have any equivalent point corresponding to x y z which generates number
point 5 which is 1/2+x 1/2+y z so; that means, the c centring is clear. So, we have
therefore, a c centred lattice.
The fact that we have a point which is -x -y -z tells us it has to be a centre of symmetric
system. So, if it is has to be a centre of symmetric system and a c centred lattice the point
group symmetry should be 2/m it cannot be 2 it, cannot me m alone and therefore, it is a
C2/m kind of a symmetry operation.
In fact, these if you look at the first four equivalent points you will see a similarity
between this and the P2/m space group which we saw a few slides before and therefore,
this is 2/m and this is C centred lattice so, this space group must be C2/m. So, the
diagram here and the set of equivalent points now we can verify that with respect to the
occurrence of 2/m. So, since it is 2/m symmetry, there are no translations that are
required and therefore, the mirror the twofold and the centre of symmetry the intersect at
0 0 0 and the C centring generates the 1 at half along x, half along y.
What is very interesting is in this particular space group because of the presence of
274
centre symmetry as well as the C centring we develop 2 additional symmetry elements;
one is the symmetry element 21 screw axis which is one fourth at this point and the other
is the glide plane which is removed by one fourth. So, I want you to take it as a home
exercise where you find out in what way this glide plane generates the equivalences of
these four onto these four in other words operate the c glide operation on 1 and see to
which of the 5, 6, 7, 8 it corresponds to glide plane operation. Likewise you operate the
21 screw axis about this point and see which of the 5, 6, 7, 8 associates itself with the
operation of the 21 screw axis.
So, the interesting observation which we make here is, even though the space group is
C2/m. There are no movement of the translation involved components we generate
translation involved component as a consequence of the presence of the C centring. The
C centring where which gives rise to 1/2+x 1/2+y will automatically generate additional
symmetry elements.
Now this is the beauty of crystallography. So, when molecules crystallise in such a space
group, they get confined to not only obey the requirements of C2/m as we have seen here
it also should obey the requirements of the presence of a glide plane one fourth removed
from the origin and a screw axis which is again one fourth removed along the a direction.
So, one once we have any axis at one fourth we will have automatically have the same
axis at three fourths and that is what is marked here you have the one fourth and you also
have the three fourths. Similarly, the 2 1 screw axis appears that one fourth and three
fourths. So, therefore, the space group is C2/m. And the next space group which I have
indicated here I want you to again examine it carefully and see what is your conclusion
on this one.
The conclusion on this one is a little complicated, but its straightforward again if you
look at 1, 2, 3, 4 and look at 5, 6, 7, 8 it tells you it is a C centred lattice. The only
difference or a major difference which you see is that the twofold axis are now removed
by one fourth. Since this twofold axis get removed by one fourth what is your guess? The
guess is that there is a translation involved component in a direction perpendicular to this
and the component which can be perpendicular to this is related to a mirror. And
normally in a monoclinic system the component that is related by a mirror symmetry is
the one which is now it is so happens that it is the one which associates itself with the C
275
centred lattice.
And the way with respect to the C centred lattice it is the one which is associated with
the c glide plane. So, there must be a c glide plane now associated with this. So, the c
glide coincides with the origin twofold symmetry moves by one fourth, this generates
therefore, the space group C2/c.
So, as you see by just looking at the presence of the equivalent points and the locations
of the presence of the symmetry elements and the location of the symmetry element you
can determine this space group. And one once you have determined the space group you
can write the equivalent points and these equivalent point diagram which is shown here
along with the presence of the symmetry information will uniquely determine the space
group and so we get the space group C2/c. And that way at this juncture we will now go
and try to examine the space groups more thoroughly; to examine the space group more
thoroughly there is no other method better than observing them through what we call as
the international tables for crystallography.
So, we now crossover to the international tables entry. A little story about the
international tables because when more and more structures started getting accumulated
and it was required that the practising crystallographers will have a direct axis to the
existence of the 230 space groups, their corresponding space group diagrams, the
diagrams associating them with the objects, the diagrams that are generally associated
276
with the equivalent points and so on.
A listing of the equivalent point is also necessary and as a consequence the international
tables for crystallography was developed early 1950’s. In fact, first issue was I think
issued in 1951 and that was called Red Book because the colour of the international
tables was read. It has undergone many changes in metamorphoses and at this particular
time it is available online. So, one can download the international tables and then look at
the details of the space groups.
So, whatever we have discussed so far in terms of the space groups, in terms of the
equivalent points, in terms of the special positions, in terms of the Wyckoff notation
which we are going to introduce now all these issues will be listed out under one entry in
the international tables. So, this is more or less the bible for crystallographers in order to
get all the details of the given space group.
So, suppose I crystallises in a given space group, most often the not nowadays the
possible space groups are indicated by the automatic machines on which data collection
is carried out and it is not required that you have need to be a crystallographer. If the
space group is identified uniquely, but if this space group is not identified uniquely by
the automatic machines then I think a little knowledge on crystallography becomes
essential and this particular course is aimed at giving you that kind of knowledge as a
background.
So, that you will be able to use the automatic machines within coats automatically; that
means, you will now make use of these machines to the fullest extent possible. And in
case the machine is not able to decide the space group by examining certain details of the
machine data in principle you should be able to find out what is the space group.
So, in order to find out what is the space group we should know how this symmetry
elements are disposed in a given space group and how each and every object repeats
itself based on the symmetry that is present in a given space group. We have studied the
triclinic and the monoclinic systems thoroughly and the study of the triclinic and the
monoclinic systems. I have already enabled us to understand the logistics of how the
objects find themselves with respect to a given space group. The thing which is missing
from us at this moment is a complete knowledge of a given space group.
277
What are the things which we need of a given space group and can we make a list out of
that and that is provided by the international tables for crystallography. At this time we
also will start introducing a new set of symbols which are up here and these symbols are
the ones which are referred to as the Schoenflies symbols are you able to see it. So, these
are the Schoenflies symbols, spectroscopies they do not use the nomenclature which
crystallographers use.
For example, in crystallography we use the point group information and also the lattice
information and combine the lattice information with the point group information to
derive the space group and this will be the space group nomenclature. This is refer to as
the Hermann Mauguin notation and that is the notation we will follow throughout the
course.
I will provide you with a chart probably in the next discussion where we can convert the
Hermann Mauguin notation to what is called the Schoenflies notation and this
Schoenflies notation is the one which the spectroscopies follow. So, since P1 is a space
group where there is no symmetry at all. In case of spectroscopic symbols the no
symmetry is indicated by C1, C1 represents a rotation 360 degrees and the one on top is
the one which is now representing the space group.
So, if one wants to use the Schoenflies symbol and still use this nomenclature to
represent the space group, then it is a C 1 1 we will see what representation P-1 will get
when we go to the description of P-1. So, this has to be born in mind because
spectroscopies and more often the not the practising mathematicians who use group
theory they use this nomenclature more often than this nomenclature. This nomenclature
is utilised essentially by crystallographers who look into the determination of structures,
look into the details of structural motives. And then look at the structure of the molecule
inside the crystal lattice and then the interaction of these molecules within the crystal
lattice and so on.
So, the requirement is that we should stick to the Hermann Mauguin notation. So in the
rest of the course we will stick to Hermann Mauguin notation, but for those
spectroscopies and group theoreticians who would like to go into the detail of what is the
corresponding symbol using the Schoenflies notation they can refer to the international
table again or I will provide a diagram and probably a hand out also will be given by the
278
TAs which will show the conversion from the Hermann Mauguin to the Schoenflies
notation. So, this is the first entry of the space group P1 in the International Tables.
So obviously, the first number is 1 out of the 230 space groups this is the first space
group. P1 is the Hermann Mauguin notation as I mentioned and the point group
symmetry which is mentioned here is 1. So, the point group symmetry is 1, the space
group is P1 and this belongs to the triclinic system.
Now, something is written down here which we will discuss much later, but what it tells
us these that there is something called the Patterson symmetry which will have the same
space group converted to P-1. So, if you look at the Patterson symmetry associated with
P1 we will get P-1.
Now, what is Patterson symmetry? We will be discussing this Patterson symmetry when
we actually determine the structures of molecules inside crystal lattices in the later class.
So, at this moment let us say we will take this off from this projection and look at only
this, so, P-1 one fold symmetry triclinic.
The International Tables provides you 3 different projection diagrams depending upon
the axis which we are choosing; which we have chosen and generally this is the preferred
projection the first one, which is 0 a and 0 b, x axis down here an y axis that way. As I
mentioned that the positions of any object inside this is mentioned as a fraction of the b
value and a fraction of a value and those are referred to as the x and y coordinates in this
particular projection. It does not mean that this particular angle has to be 90 degrees.
So, what you see as coordinates given in crystallographic analysis you see the
coordinates of the atom positions and so on when you read literature, in the research
papers and the review articles and all that which you read the coordinates that are given
are those of fractional coordinates. Fractions of a and fractions of b now represent x and
y similarly the fraction of c represents z.
279
So, when we say x y z as a position which is indicated here in the diagram below here
this tells us that there is a position which is for one particular object. In this particular
stage just like the previous discussions we had instead of a gamma here circle is used and
there is a plus which indicates that the object is coming away from the board or away
from this projection.
So, if the object goes inside this we indicate it by a negative sign. We will see what
happens if we invert the object in a next projection where we discussed P-1 bar. So, in
P1 therefore, if you see the unit cell this is the unit cell it is divided into half and half
here and you see that the unit cell has only 1 object. So, that is the object which is a 360
degree operation brings a turn to itself as we say as defined earlier the requirement of a
space group is that the object has to come on to itself after the symmetry operation.
So, in this particular case it comes onto itself and so the origin can be arbitrary we can
choose the origin wherever we want in case of a triclinic system and then we define what
is known as an asymmetric unit. Now the asymmetric unit in this particular case is the
unit cell because there is no other object as we discussed then z is equal to 1 in this
particular case and there are no special positions, no symmetry positions in P1 and
therefore, the asymmetric unit is the unit cell. So, it is described as 0 less than or equal to
x less than or equal to 1. Remember these are the fraction coordinates we talked about.
So, this is now a, b and c the full value of a, b and c is taken. So, the value of x rangers
280
between 0 and 1 the value of y ranges between 0 and 1 and that of z also ranges between
0 and 1.
The symmetry operation that is present in P1 is only 1. So, this is the redundant set and
you see therefore, this entry as the first entry for number 1 in the periodic table.
Now, apart from that the types of translational periodicity is indicated here these are the
generator selected the translational along a, translational along b and translational along
the c direction or these are the ways in which we can represent the translational
periodicity in the unit cell.
Now, the additional details which come here are the ones which we have to learn
carefully. The additional information which comes here which is written here can you
see it in the that these details clearly multiplicity, Wyckoff letter and site symmetry. So,
these are the 3 issues which we will be discussing in more detail as we go to the higher
space groups.
But in this particular case, the space group tells us about the multiplicity value, the
Wyckoff letter and the site symmetry. We will define these 3 when we go to P-1 at this
moment there is no entry associated with except this one which tells us that the
multiplicity associated with this object is 1 which refers to x y z the Wyckoff letter is a
that is the highest symmetry which is present in the system and then the site symmetry,
281
the symmetry associated with any position in the unit cell is 1. So, there is no symmetry
its 360 degree rotation.
So, the equivalent points are written here that is x y and z. So, there is only one
equivalent point and therefore, the required other issues associated with coordinates are
not given. So, the coordinates are only 1 type which is x y and z. So, P1 is the number
one space group which has no symmetry other than the 360 rotation symmetry and all the
specifications which are given here fully qualifies this space group P1 or the diagram
here the symmetry diagram here.
And then of course, the equivalent point positions indicating the number of equivalent
points in the unit cell which happens to be 1, the symmetry operation is also 1
asymmetric unit is defined by the unit cell here and then we have the definition 1 a 1 x,
y, z telling us that there is no further symmetry operations that are possible in this case.
The special projections which are indicated here tell you very clearly that along 0 0 1
what is the axis, along 1 0 0 what is the symmetry, along 0 1 0 what is the symmetry.
This symmetry you notice is indicated with a little p and 1, the little p tells us that we are
now talking about two-dimensional projection or in other words two-dimensional lattice.
This particular two-dimensional lattice which is in three dimensions triclinic has a
symmetry 1.
Remember when we discuss the two-dimensional lattices, there was no third axis and
there the symmetry was P2 there was always a twofold axis associated with lattice point.
So, here the moment we introduce the c axis direction and say this is a three-dimensional
object, the corresponding projections projection diagrams will have only 1 object and
therefore, it is P1 all in all the 3 cases and the origin is taken at 0 0 z, z is arbitrary. So,
we can say z is equal to 0 then this represents the diagram that is required. So, this
diagram therefore, has the origin at this point by definition we have given z is equal to 0
it is arbitrary then this is the a value the cell dimension and that is the b value the cell
dimension along this direction.
So, we are referring to this particular projection and in this projection we have these
object which is represented as x, y and z the value of z is now coming up from the board
or from the projection.
282
So, this now describes more or less the whole thing that is associated at this moment for
us to note there are other entries in the International Table with at this moment we will
not worry about because, we have to now fully understand all the 230 space groups in
order to appreciate what are these entries. So, we will not worry about these entries at
this moment; however, we are also not going to see all the 230 space groups in this way.
We will take a few representative examples. So, that we will now learn how to read the
International Tables, you see for example, when you are try taught the grammar of a
language like in English, the grammar is start with respect to the nature of the words, the
way you can combine the words, the types of words, you can have is the noun is talked
about, the adjective is talked about adverb is talked about. So, similarly we will tell all
the language that is required in order to understand the grammar of the crystallography in
230 space groups.
The actual textbook which we write about each and every molecule will be different, but
each and every molecule will have to follow these grammar book which we have
prescribed. So, we cannot go away from the grammar book which we prescribe and
therefore, all the elements of the grammar the a way in which we make sentences, the
way we make the combinations of sentences, the phrases becoming sentences, they all
become very important and once we study the space groups. Afterwards every novel can
be different, every crystal structure can its own beauty, every crystal structure eventually
283
with the molecular species inside can have its own story it can have its own property and
so on.
So, you can write a numerable number of stories and that is how the data accumulates in
the understanding of a given molecular species. So, then we can classify these molecules
into various kinds of organic inorganic organometallic and so on the categories we will
go into those details later. But at this moment we are looking actually into the grammar
book, the grammar book which tells us how we make this combinations and how these
combinations will give rise to sentences which are meaningful.
So, the next step would be to go to the next space group which is number 2 which is P-1.
P-1 you see the symbol now is C of I, the moment we use a small i in spectroscopic
symbol or in group theory it tells you the inversion centre; i refers to the inversion centre.
So, therefore, we have the -1 which is indicated by the Hermann Mauguin notation, in
this spectroscopic notation is C of i. So, group theory and the Hermann Mauguin
notation followers in spectroscopic nomenclature they use C of i and both of them now
belong to the first crystal system so it is 1. So, this is C i of 1 and the point group
symmetry is -1as we already know and this belongs to the triclinic system.
Now, how does the projection diagram look like and where are the symmetry elements.
So, that is given in these projections again we will concentrate only on this projection
diagram. Now this particular projection diagram tells us the presence of a which is now
284
down the this direction and the presence of b which is now in the horizontal. So, the
vertical down is the a direction remember this is not 90 degrees this is whatever is the
angle between a and b which happens to be gamma for a triclinic system a not equal to b
and b not equal to c, α not equal to β and β not equal to γ and therefore, we have the
situation where we have now representing the projection diagram like this.
The centre of symmetry as we discussed all along until the last class, the centre of
symmetry if it is present in a crystal in a space group it should be at 0 0 0. So, we
therefore, move the centre of symmetry to the origin. One once we have the centre of
symmetry at the origin there is a centre of symmetry at cell edges and therefore, all these
cell edges develop the centre of symmetry and we have seen by the equivalent point
distribution which is shown here.
The equivalent point distribution which tells us this is x y z and this is -x -y -z. So, this is
the centre of symmetry which is up here. So, you see that in earlier diagrams where we
studied we put them together, putting them together was easy for our understanding and
that is why we did that. But now we want to understand the symmetry positions
differently from the positions of the objects and that is why 2 different diagrams are
given. So, these 2 diagrams therefore, are of crucial importance this as well as this which
will tell us the symmetry position symmetry that position diagram here and the
equivalent point diagram here.
285
So, you see that this is now centre symmetrically going over there. So, the negative value
associated with this open circle is indicated by the presence of a gamma inside. So, this
is now a gamma inside the round representation and a minus indicates that it is going
down the plane, plus is up the plane, minus is the on the plane and apart from that the
other values of x and y in this case because this is the value of x that is the value of y. So,
this represents x y z the value of x and y changes to -x -y and that is why it is shown with
a gamma.
So, whenever there is an inversion the inversion is shown with a closed circle with a
gamma inside. So, this now will also tell us how the translation symmetry takes them up
into these objects and invokes the presence of the central symmetry at this 1/2+x 1/2+y
which is up here and you see that these two this comes by translation along this direction
and translation along that direction and so these two are related by a centre of symmetry.
So, we will see the further details the asymmetric unit now is what? The asymmetric unit
now is half along the x; that means, they asymmetric unit is only that much. The
asymmetric unit is only that much because you have a centre of symmetry here; you have
a centre of symmetry here; you have a centre of symmetry here. Effectively you can
consider this as a unit cell which is a sub cell of this unit cell and in that sub cell there is
only 1 object and that is why we call it as the asymmetric unit.
So, here the z value with number of equivalent points inside the unit cell is 2, x y z goes
over to -x -y -z bar and each of these blocks therefore, now that is why these 2 lines are
drawn each of these 2 blocks therefore, now represent the asymmetric unit. So, the
asymmetric unit is 0 less than or equal to x less than or equal to half y is full and z is full
you can also interchange them you can have this moving over to y and this moving over
to z and you can have other types of asymmetric units both in the y direction as well as in
the z direction.
But this is the definition of the asymmetric unit with respect to this diagram and with
respect to the projection 0 a b that is ab projection, ok. Now you see that the symmetry
operations become 2 there is a symmetry operation 1 and there is a symmetry operation
-1 and this -1 symmetry operation is about the origin 0 0 0 and so the again the generator
selected as before along the x direction along the y direction along the z direction.
286
Now, the space group is number 2 as we have seen now what happens; this is a very
crucial stage where we have to fully understand the concept of multiplicity, the concept
of Wyckoff letter and the site symmetry. So, let us go through it a little carefully now
there is something on the right hand side for which we will return much later, this is with
respect to the experimental conditions which get generated due to the presence of the -1
symmetry right now we will not worry about it. So, forget about it at this moment.
What you have to remember is the fact that if there are there is a general position which
is now associated with 1 symmetry 1 then there are 2 objects x y z goes over to -x -y -z,
and that those are the 2 equivalent position. So, z is equal to 2 z prime which is now the
asymmetric unit is equal to 1 and the asymmetric unit we have already defined as the
half the value along x or y or z we can choose any one of them in this particular space
group. So, having given that choice we will now look at the so called multiplicity.
Now, what is multiplicity, what is the number of the multiplicity that comes on the x y z
suppose you take the coordinate as x y z how many of equivalent points we generate in
other words the number of equivalent points the value of z, the value of z is represented
here which is equal to 2. So, multiplicity is how many such objects are there in the unit
cell? We have an object located at x y z corresponding to x y z how many such objects are
located with respect to -x -y -z. So, there are 2 of them, so, therefore, it is 2.
The site symmetry still 1 wherever it is sitting it is 1 only because it is a triclinic system.
287
So, the site symmetry is 1, but the multiplicity associated with the site is 2 so that means,
if you have an atom at x y z it has to have an atom at -x -y -z it depends upon where you
define the origin. So, you go back to the figure here you can define the origin wherever
you want in this projection diagram correspondingly all these centres of symmetry is will
move the 0 will coincide with wherever we represent that and the x y z value will be read
based on the value of the origin which is 0 0 0.
So, suppose I move this to this origin with the representation is different and therefore,
the value of x y z will be different ah. With respect to the x y z here therefore, we have
invoke now several centres of symmetry. So, this particular x y z position which is right
here the position with respect to that can be expressed 8 different ways, the 8 different
ways are 1, 2, 3, 4, half removed 5 and then 6, 7 and then 8 so totally 8 possible ways in
which we can represent them and those are listed here a, b, c, d, e, f, g, h and the
positions of these centres of symmetry are indicated.
What it tell is also the fact that if there is an object which sits at any one of these
positions the object is independent of the other positions. So, if you have 8 equivalent 8
positions here these 8 positions which are generated due to the presence of the -1
symmetry, they have already built in them the -1 symmetry; that means, every x y z will
go to -x -y -z is built on to them. That means, if you take for example, 0 0 0 operate this
in symmetry operation -x -y -z you will get 0 0 0 only. So, the -1 symmetry therefore, is
already inherent at that point. It is true with everyone of these 8 points -1 symmetry is
inherent at that point it recites at that point.
Since the -1 symmetry recites at those points these are referred to as are special
positions. So, we have 1, 2, 3, 4, 5, 6, 7, 8 special positions and one general position. So,
the special positions importance goes as we have arranged here we start from a then we
take the z axis half as the b, the y axis half as the c and the x axis half as the d, this is the
nomenclature which is now a, b, c, d now is referred to as the Wyckoff letter. Wyckoff
used this symbol to identify the differences of atoms which can sit in these positions.
For example we can have an atom with 1 and some other atom sitting in let us say 0.5 0 0
these two atoms are not 1 and the same there different atoms. So, structurally they are
different. So, indicate the differences in structure a and d are getting invoked. So, we say
that suppose there is an atom 1 at "a" and atom 2 at "d" we say that this structure is
288
generated with the atom 1 at "a" and atom 1 at "d"; that means, the 2 atoms are different
from each other.
In fact, if a same atoms sits in these two positions also their properties are different that
is because if you sit at 0 0 0 operate the centre of symmetry you get 0 0 0 again, if you sit
at 0.5 0 0 operate the centre of symmetry you get 0.5 0 0 you do not get 0 0 0 or any of
these 8 and that way they are very special. So, these positions therefore, are refer to as
special positions.
Wyckoff decided to follow a certain protocol and that protocol is generally followed for
example, in this case we start from z, y and x and when we have two of these value is
equal to half we start xy, xz and yz in that order. And this order depends also upon the
other symmetry elements which are present in the space group if we go to higher space
groups this order need not be followed. We will see that as an when we examine their
space groups, but in a triclinic system we have 8 general special positions, 2 general
equivalent positions, 8 special positions and they are indicated with the Wyckoff
symbols a, b, c, d, e, f, g, h for the 8 special positions and i which is a general position.
So, the number value of z is 2 in case the atoms are or the molecules or whatever is
sitting in general positions. So, in general positions the symmetry is 1 and therefore, it
will generate its -x -y -z, so, we will have 2 objects generated. So any objects sitting in
any of these special positions will generate only one object; that means, the object much
itself possess the centre of symmetry. So, half of the object the other half of the object
they should be related by a centre of symmetry then and only then the atoms can sit at
those positions in this example.
So, P-1 therefore, already is looking very complex in the sense of understanding, but it
also tells us that it is now given special occupancy positions for special issues. So, for
example, let us say you have an organometallic component and you have a metal which
is sitting at one of these Wyckoff positions the rest of the atoms need not sit at that
position, but they should sit in such a way that the overall symmetry associated with the
molecule is centre of symmetry otherwise this heavy element the metal cannot sit at any
of these special positions.
So, imagine a situation where the metal atom is sitting else elsewhere the moment the
metal atoms sits elsewhere there even though there is a centre of symmetry which might
289
be associated with the molecule which is a very rare chance the metal now will generate
another unit which is going to give us z equal to and therefore, there will be two objects.
So, let me repeat if you have a metal organometallic compound so there is a metal and
then organic surroundings, the organic the metal will sit let us say at a special position
because it atom by itself has a centre of symmetry, it is not just the atom we should
consider we should also consider the bonding associated with it. The nature of the
bonding and the way in which it is bonded to other atoms the environment should also be
centre of symmetry.
So, the molecule may find itself in a centre of symmetric environment, but it is not
necessary that all these atoms have to sit in this special position. So, they will have they
will be sitting in their general positions. When we look at the realistic structures later on
we will see how this really happens, even though it is this very interesting to notice that
this can happen with respect to the way in which these things appear. Suppose, there is
an atom which is sitting here and let us say this occupies anyone of these centres of
symmetries, if that is a spherical atom let me make it perfect sphere if possible now its
nearly spherical ok.
So, let us say this is our organometallic element then, whatever is the surrounding
organic let us say there is some branch coming out like that ok. Now this branch should
be identical to the coming out branch. So, these two should be centre of symmetrically
related to each other, but you notice that if the coordinate of this is let us say 0 0 0 and
the coordinate of this is x1 y1 z1 then the coordinate of this will be -x1 -y1 -z1 bar. So, these
two are now related by a centre of symmetry; that means, there are 2 objects they are
sitting in general position and the x y z is equal to is and -x -y -z is getting generated this
is a possibility then the metal can sit in the special position.
Suppose there is a structure in which there is a perfect centre of symmetry they can sit in
any one of these positions, you notice that if this is not at 0 0 0 or some 0 1/2 1/2 then the
coordinates will accordingly change with respect to 0 1/2 1/2. So, this molecule now will
be only present with respect to 0 0 0 and that molecule will be present only with respect
to 0 1/2 1/2 which where the metal is going to sit. I think this is a place where is a good
place where we can stop today and then we will continue tomorrow.
290
Prof. T.N. Guru Row
Lecture – 23
Additional Information on Space Groups
In the towards the end of the last discussion I realise that the diagrams which were being
shown for the space groups they were not of reasonably high resolution. And in fact, they
were too very tiny particularly if you are going to view it on your YouTube or whatever.
So, having the after discussing with our technical people and also the students I have my
students have in fact, recast the whole thing. So, it is not the type of entry we will see in
the international table they have split the entry into different possible PPT slides. So, that
you see the entire thing in very comfortable viewing because that is very important
particularly concerning the fact that some of these for example, the directions of a and b
and all were not properly seen in the earlier part of the ending part of the previous
presentation.
So, this will make it a little more visible and clearer for your understanding because this
is more or less the what should we say the heart of understanding crystallography. What
happens to molecules when they get into the crystal structure?
291
What happens to atoms when you they are getting into the crystal structure and even
larger smaller whatever be the size of the species which gets into a crystal structure? It
has to obey all these 230 space groups. The idea is not to present to you all the 230 space
groups that is immaterial and irrelevant at this moment. But to give you the flavour of
how it is helpful to study the international table content so that you will understand how
to appreciate the formation of equivalent points and the identification of special positions
and so on.
I have started I am starting from P ī which is number 2 in the list of 230 space groups
because P 1 is redundant it has no symmetry at all. So, there will be only one object in
the unit cell. So, this symmetry the corresponding Schoenflies notation which I discussed
before is Ci and the point group symmetry is ī. So, the moment a space group of this type
P ī is given to you, these two pictures should come into mind immediately.
And in fact, these two pictures are sufficient enough to find out what space group it
belongs to. This is the particular representation. The first representation is a
representation of the symmetry positions. And the second representation is that of the
equivalent points. The first representation as you can see is the showing the presence of
the centre of symmetry at the origin. And of course, it will be generated in various places
as we have discussed in very many times before.
So, if we now look at the number of molecule in unit cell there are two in the unit cell
this represents now the unit cell there are two in the unit cell. And this part represents the
so called asymmetric unit whose dimensions are given below. So, this essentially tells
you that unit of the molecule or atom or whatever the specimen or the species which is
now put into this unit cell in such a way that its presence is felt in the unit cell, but not
the symmetry operations.
So, if we show the symmetry operations along with then it will become the number of
repetitions in the unit cell following x y z, x̅ y̅ z̅ in case of P ī. So, this represents
therefore, the asymmetric unit and this represents the full unit cell.
292
Now, the corresponding equivalent positions are shown here. What is important to notice
here is there are three issues. One is called the multiplicity, the other is the Wyckoff
letter and the third one is the site symmetry. So, these are the three issues we should we
already discuss, but we will go into it in more detail with this clear diagrams which we
have now got.
So, the first one is the multiplicity it tells you given a position x y z how many such
positions gets generated by this space group. And that happens to be 2 because it is x y z,
x̅ y̅ z̅ in case of P ī. The way in which we arrange this Wyckoff letter is to look at the
most symmetric possible position in the unit cell. The most symmetric possible position
in this unit cell of P ī is at 0 0 0 where there is a centre of symmetry located.
Now if we operate these two equivalent point operations it gives back the same 0 0 0.
And therefore, the multiplicity associated with this site is 1 and the symmetry of course
that is associated is ī. So, this gets an alphabet “a” which is the first letter of the English
alphabet and this is what was designed by Wyckoff.
And this is now to give us a clear understanding of where the atoms are located in what
way they generate equivalent points and so on. I will take an example in a minute. So,
the as we go to other symmetry positions we see that we go from the z y and x we will
293
say b c and d and then we will go to the half positions along x and y, x and z, and y and z
in that order.
So, these numbers therefore, represent different centres of symmetries that are present in
this space group P ī, 2 i represents the general positions. So, there are totally one two
three four five six seven eight equal the special positions and one is the i is there, one
single general position. Now, what is the meaning of this special position? Let us
interpret it with respect to the position which we show now by marking it.
So, let us say there is an atom which is located at this point. This atom now is associated
with this centre of symmetry. According to the special position nomenclature it will have
only one symmetry that is generated so that atom will not generate any other atom in the
unit cell. So, the occupancy now we are I am defining a new word which is referred to as
the occupancy.
The occupancy of that site therefore, is now half. So, we say this occupancy of this site is
half because half the atom is associated with one part of the symmetry the centre of
symmetry is the other half. Now if there is a molecule which is something like this which
we actually discussed in the other class yesterday or day before.
Let us say it goes it has a substitution like that on one side and the substitution on like
this on the other side. I am trying to make it as symmetric as possible which is not a bad
diagram in my hand writing. So, you see that this is the possibility of a position of a
metal. So, the metal sits here let us say and this is a ligand which is attached to the metal.
And you see that this part of the ligand is centrosymmetrically related to that part of the
ligand. Now suppose you take this position and call it as x y z, this has generated the x̅ y̅
z̅ over here. Let me show it on a using the laser pointer.
So, if you see here this atom, this atom now sees this centre of symmetry under goes the
symmetry operations. So, if this I call it as x y z then the corresponding one here is the x̅
y̅ z̅. So, there are two positions so this atom is now sitting in the general position.
The atom which is sitting at this special position is our metal atom let us say it could be
zirconium or something. So, some heavy metal atom can sit there and the ligands that are
attached to this are these positions. So, this molecule therefore, which is eventually
294
generated with this kind of a contact which is shown here has a centre of symmetry. So,
this satisfies the Kitaigorodsky principle that if there is a centre of symmetry associated
with the molecule then the molecule prefers to sit at the centre of symmetry.
If they if there is no centre of symmetry associated with the molecules suppose we put
another substituent here let us say in which case this will move away and there will be
two such molecules in the unit cell following the rules of the space group. So, that is how
the positions of the atoms come up. We will discuss it with realistic example as we go
along further.
Now these are the right hand side entries in the international table, we will discuss these
entries in more detail when we do this the scattering theory followed by the experiment
the diffraction experiment. So, in a diffraction experiment what we record are the so
called reflections. You are already probably familiar with this but have no idea how these
come up these are the hkl reflections.
So, this so called hkl reflections which represent a plane. So, in diffraction experiment
we send in x rays. They come off the hkl plane and we get the intensity of the reflection
at a certain point. So, we identify where the hkl is and also identify the intensity. That is
controlled the symmetry now controls that the way in which this diffraction positions
come, the reflections come. And that is why this reflection conditions is put here and in
this particular case of P ī there are no such situation.
295
So, there is no extra conditions and no general conditions either. So, all the reflections
that are possible in a diffraction experiment will appear with no changes coming up. We
will see as we go along into the scattering theory what fills up in these position
separately. We will not worry about it at this moment.
Now, we go to the next space group P2 which is number 3. The moment we say 2-fold,
the moment we say there is a 2-fold symmetry in the as a point group symmetry
associated with the object which is there in our crystal system. Then we realise that it has
to be a monoclinic a single 2-fold representation. That means, the single axis operation
2-fold represents a monoclinic as we have studied many times earlier.
So, the presence of the 2-fold in fact, is indicated in a full symbol in this form it says P 1
2 1. The 1 represents the symmetry along the x direction assuming that this unique axis is
b here, 2 represents the 2-fold axis associated with the b axis along the y direction. And
then the z value the z axis is again has no symmetry or 1-fold symmetry. So, instead of
writing the full symbolise P 1 2 1 we write this space group as P 2.
The equivalent points on the space group we will see in a little while from now. But what
you see here are the two diagrams which are of immediate importance. So, the discussion
now should be centred around. Suppose you are given only this particular symmetry
positions for your material. So, you have a material it crystallizes and somebody tells
you that this is where it crystallizes and gives you only the symmetry information.
296
Now from the symmetry information you can infer several things. One is of course, there
is a 2-fold rotation axis coming away from the plane of a and c which is shown here the
projection is down the b direction. So, this is the ac plane. Now if this is the ac plane and
b is the unique axis this angle will be the non 90 angle in monoclinic symmetry the so
called beta.
So, that angle a o c will be the so-called beta angle which is a non 90 angle. So, this now
represents that there are 2-fold symmetries which are coming out from the surface of the
projection here to achieve. And also going inside you inside the projection and these now
represent the 2-fold symmetries.
And so the moment you see a picture like this you can conclude that this is a primitive
system because there are no additional indications here of the appearance of additional
symmetry elements. The symmetry elements correspond only to the 000 and therefore,
there is no centring here associated with it. So, effectively you conclude that this is this
space group P2.
It may so happened that in some example you may be given only this diagram. This
diagram is an extremely useful diagram because we now already see the asymmetric unit
which is now in fact, the dimensions of which is shown here. The asymmetric unit it has
only one object. This object now undergoes the 2-fold rotation goes up there.
And then this translated once translated once these two now will be related by the 2-fold.
So, if you are given this diagram alone still it is possible to identify this space group so
the space group will P2. So, having seen that we will also see whether there are anything
like special positions in this particular space group.
297
For example, the multiplicity Wyckoff letter site symmetry which is listed here shows
that there are four special positions. Remember in P ī there were 8 special positions, here
there are 4 special positions and what is interesting is that these special positions have a
value of z. This is in a different orientation this is probably seen down the z axis. So, this
is something which should be seen down the y axis. So, I will correct it. This should be
½ y ½ and 0 y ½ then ½ y 0 and then 0 y 0.
So, this 0 y 0 essentially represents this 2-fold. So, why that confusion came I will
explain. It came because of the fact that in this representation from which we copied this
equivalent position which is not along the unique axis b, it is along the unique axis c. It is
so happens that while taking this various diagrams it has come along the unique axis c. It
does not matter, in a fact it is a education value to see this next one. In the next one you
see that the 2-fold is not now associated with the b axis, the 2-fold is associated with the
c axis.
So, this is a non unique axis setting. I am happy that this was this mistake we made
because I can now tell about both the b axis as well as the c axis. So, the unique axis here
as you see is c. So, suppose you are given only this part of the entry in the international
table then you will be able to see that you see that in this particular case it is the c axis
setting which is a non standard setting in a monoclinic system.
298
But again even if it is a non standard setting one once we have the 2-fold symmetry
before special positions develop. And these special positions developed along now the
unique axis only this particular case the unique axis is c. So, what would be the
corresponding unique axis in the diagram we have shown before. In the special positions
here will therefore be 0 y 0 which is corresponding to the 0 0 z here.
And ½ y 0, 0 ½ y and ½ y ½; that means, it corresponds to these directions in which the

2-fold is located. So, there three of them here which come directly and therefore, we see
that this therefore, represents the overall representation of the space group P2. So, we
have four special positions and one general position.
So, you see Wyckoff now let us now say a b c d and e. So, what is significant about this
Wyckoff notation is; the fact that particularly as we go to higher and higher space
groups. The possibility of materials crystallizing in higher space groups is restricted to
only inorganic materials and materials with larger atoms. Organic compounds do not
generally go in into higher systems, but even if organic compounds go into this kind of a
system.
Suppose there is a situation where we have a molecule which is now going to sit here let
me do that operation to make you feel comfortable. Suppose there is a same compound
which we took earlier but now the heavy atom sits here and now the connectivity will go
like that. Now tell me whether it will go up or it will go down? In this case it will go
down and in that case it will go up. So, these are the 2-fold operations.
So, this atom and this atom suppose we consider these atoms, suppose we consider this
atom and that atom; these two are not associated with the 2-fold symmetry, but they will
go now such that the overall symmetry of 2 is maintained. So, these two atoms
individually will again be 2, but at the heavy atom that is present here will be 1.
And this is something which also tells us the way in which the molecules arrange
themselves inside the unit cell. Particularly when they have a symmetry that is associated
with the molecule which also is the symmetry that is present in the crystal system. So,
when we go to higher crystal systems the molecular symmetry may still be either lower
or higher.
299
But the space group may have both higher and it may have all higher symmetries. So, the
higher symmetry position can be occupied by a lower symmetry molecular system also.
And that is how you can generate many possibilities in higher symmetry space group.
So, if there is a very highly symmetric cubic system for example, where a=b=c,
α=β=γ= 90 degrees the special positions are the ones which will which are preferably
occupied by molecules.
And that is the reason the reason being that it will give us give the flexibility associated
with the molecule. Otherwise you know you will end up with a large number of possible
positions in the space group; which is not really the case the molecules are looking for.
In other words the overall conceptual plan here is that a crystal, this is the definition; the
crystal is a super supramolecular par excellence.
You cannot have the molecular arrangement any better than in a crystal. The crystal
gives you the best possible arrangement. So, when you talk about the assembly of
molecules; you have several molecules assembling together, the most preferred and the
most energetically favoured one is the one in the crystal. So, the molecule now finds a
very comfortable environment to go into a crystalline environment. And that is how we
are studying the symmetry and structure in the solid state.
So, when it goes into this system it cannot be any better. At the same time it may go into
a system which is what is generally referred to as metastable systems; this happens very
often in pharmaceutical sciences. This metastable system is not the one which is
energetically favoured to go into that crystal system. And probably it will now revert
itself to another crystal system and this is the issue of polymorphism. So, it will revert in
a given time; this may take about 1 day, it may take about 100 years, it may take a
millions of years.
Like for example, I give you a simple example of carbon. You all know that carbon
crystallizes in the as diamond, it also can crystallize as graphite. And of course, we the
C60, C70 and all that recently discovered; these are all different allotropic forms of
carbon. If there is an atom and it goes into different forms we refer to them as allotropes.
If there is a molecule and it goes it goes into different crystal forms then we call them as
polymorphs.
300
So, that is the basic difference; so this is particularly useful for people from
pharmaceutical industry. Because they will always be wondering what is this
polymorphism about in types of crystalline material; so, this is polymorphism.
Polymorphism is not restricted only to crystalline materials; polymorphism can exist in
very many circumstances. It can be in liquids, it can be in gases, it can be in a single
biological molecule, you can have different polymorphic forms emanating from the
system.
It can be polymorphs in gene sequencing and things like that. Polymorph is a very
common phenomena but in crystal when it happens it happens in this particular way.
And therefore, it is of importance to know that crystal gives us the best possible
environment for assembling a molecular species. So, when large number of molecules
come together and crystallized, they prefer to go into the best possible form energetically
favourable form.
The energy that is associated with that will be very minimum and the energy that is
associated this with that if it is not the minimum then it can take the other forms. So, that
is how the prediction of crystal structures becomes difficult. Otherwise we can just you
know look at a crystal oh the molecule is here it is not that way. And so crystallography
and the study of whatever we are doing in this course is still alive and it is important.
Because of late cryo-crystallography has cryo-EM crystallography; electron microscopy

based crystallography has been established as a technique where we can see very very
large molecules. In fact the latest one is a protein isolated from sea urchin whose size is
millions and millions of Daltons. And have been able to look at the microscope and they
get a microscopic image of the molecules, so that is the latest. Anyhow we will not going
to the detail of it. We will go now and see the presence of the special positions. So, there
are five, four special positions and one general position in the space group P 2. And this
is not required now we will discuss it later on.
301
Now, we go to the space group P21. P21 is where we have now introduced a screw axis.
So, the diagram if it is given like this; now you are experts you can easily say that just
give me this diagram I will tell it is space group is P21. So, none of these are given
except this diagram we will now say it is P21.
Because the 21 axis is at the origin, is 1 unit 1 unit away that is translation and they all
appearing at half positions; so it is a 2 1 screw axis. If instead of this diagram this diagram
is given; again you are experts enough now to say that this is the P21 system. Because
you have this to 180 degree rotation and half translation the direction perpendicular to
that, you have two equivalent points in the unit cell. One in the asymmetric unit which is
the same asymmetric as before and full symbol is P 1 21 1.
This is the Schoenflies notation which is used by group theoretical specialists, and also
by a most of the mostly mathematicians use that nomenclature along with
spectroscopists. The reason why we prefer to use the Hermann Mauguin notation as we
have here; is because of the fact that we can definitely straightaway delineate the lattice
from the symmetry exhibited by the point group, and that is a very big advantage.
Because we can identify the lattice here and the point group symmetry there, and the
point group symmetry if it has a translation; obviously, the translation component has to
go.
302
And therefore, the point group symmetry here is 2. So, the understanding of this space
group therefore, in terms of the Hermann Mauguin notation is more straight forward
when we consider crystallography. Crystallography has a lattice, crystallography has a
unit cell, and these are not present in spectroscopy. And mathematicians do consider
these as special cases of the normal cases, where they look at only the point group
symmetries and that is all these notations are changing. So, the crystallographers prefer
to have this notation.
Now, are there special positions in this space groups. Now the problem here is that if you
want to have a special position in this space group one can think of a special position
associated with this axis. But then there are there are issues associated with the
translational symmetry periodicity.
So, we are going to translate it by half the unit. So, when we translate it by half the unit
you do not have please note; you do not have the 2 1 screw axis and above half distance in
that particular direction. You have only in this direction, because it is a single axis
operation as you see it is operating only along the y direction; you do not have it in this
direction. So, because the symmetry elements are not present you won’t get special.
So, in the 2-fold case they are present again here they are not present. The half
translation destroys the presence of the symmetry. And therefore, we end up getting into
the situation ok. So, there is no there are no special positions in this. So, this 2a and x y z
303
x̅ y+½ z̅. There is something which is entered here again we will discuss that in the after
we study the x ray diffraction techniques.
The next space group again we are still in the point group 2. So, as you all know that
monoclinic symmetry gives a 2, m and 2/m. So, the 2, m and 2/m will represent one
single axis operation that is the; twofold axis a mirror perpendicular to it. Or the presence
of a 2-fold axis and a perpendicular all three are represented.
And therefore, the possibility of now considering the centred lattice which is associated
with monoclinic systems becomes handy. So, the space group now is C2 and this goes
takes us to the next space group which is space group number 5. So, in this particular
space group we again have the diagram given here. Now this diagram will tell us that for
every position here you have these so the twofold positions will define the presence of
the 2-fold symmetry.
Apart from that you get indications of the presence of screw axis associated with this half
one fourth positions. Now how does that come apart? It comes about by examining this
diagram. So, we are going to now spend some time in examining this diagram and
relating this diagram to that diagram. We did that already in the previous classes, but not
with respect to the space group diagram and equivalent point diagram the symmetry
diagram and the equivalent diagram given in the international table.
304
So, if we now understand this and correlate with what all we have already understood in
the previous classes. Then our understanding in this whole concept becomes clear. So,
you know when once a compound crystallizes in a C2 system, where are all the
symmetry elements? In this particular diagram which we see that all the symmetry
elements are shown here.
The moment additional symmetry elements come at one fourth position. Either along this
direction or any other direction in this particular case since the unique axis is along b it
will come only in this particular direction. So, the if there is a one fourth appearance of
any symmetry element in this case is a 21 screw axis. The presence of the 21 screw axis
tells us that is a C centred lattice.
So, we automatically know what how this C centred lattice is coming. And this is
something we will discuss a little more in detail when we take up this diagram. If we
take up this diagram we have the position of the x y z here then we have a 2-fold
operation which is across here so this twofold now we will bring back this here so, we
got two of the two equivalent points here.
Now, because of the C centring that is present in this particular space group you see that
additional for this the additional one comes here at this point which is ½ + x in this case
the projection is “ac” this diagram “ac” projection as you see here this is a and this is c.
So, this will be ½ + x, ½ + y that is indicated by the ½ + up here because y is direction in
which it is going perpendicular and then the z value remains the same.
And therefore, we get the first position here, second position there, third equivalent point
where this second equal and point comes within the unit cell. And then there is a twofold
axis which is present at this position which takes this over to here. So, we get this half
plus object at this position. So, we have therefore, 1 2 3 4 equivalent positions; so
therefore 4 units in the unit cell. And this is something which tells us the way in which
we have to now choose the so called asymmetric unit.
So, you see the asymmetric unit dimensions are given below here. And this asymmetric
unit therefore, gives 0 < x ≤ ½. So, x which is this one we take only half this distance
and along the y direction we take only half the distance so this fellow will not come, and
then along the z direction we take the full distance. So, what does the asymmetric unit
consists the real asymmetric unit consists of two objects.
305
Now the two objects are C centred lattice related there is no symmetry so, the definition
of asymmetric unit is still remaining. When we say asymmetric unit; there is no
symmetry, the presence of the C centred lattice is a consequence of the nature of lattice.
So, the lattices develop these two equivalent points so this asymmetric unit therefore, has
two equivalent points in the asymmetric units; so there will be two objects in the
asymmetric units. The moment there are two objects we know that it has to be a C
centred monoclinic system.
306
Prof. T.N. Guru Row
Lecture – 24
Details of Space Groups 2
See we have been looking at the Space Group C2 unique axis b.
But what is one issue which we have to keep in mind is that the presence of the y axis is
not unique the 0 of the y axis that means, the origin can be anywhere along this direction.
So, suppose we move the origin to this 2 1 screw axis what happens? We have already
discussed this issue suppose I take this origin and move to this particular point, now it
becomes C21 the space group becomes C21. So, we already showed that C2 and C21 are
one and the same.
So, it is necessary to specify the origin and you say here origin is on 2; that is why the
international tables is very careful very cautious. So, they say the origin is on 2 if the
origin is shifted to the 21 position, it will become C21, but these two are one and the
same. So, the question of fixing the y value is crucial here and that y value is put at 0. So,
this will be the 0 0 0 that is the presence of these and the 2-fold axis.
307
If we look at the possibility of special positions here interestingly, special positions do

develop that is why I define where the origin is and when once the origin is fixed then
the special positions will come at any value of y. So, the value of y is dependent upon
where you fix the origin. The origin is fixed at 2 fold axis then two special positions
come up and those are a and b Wyckoff notations and these have 2 equivalent points
because the total number of equivalent points here due to 0 0 0 and 2 fold is this, due to
the presence of the C centring with this is that which we have already indicated in the
diagram. So, we have come to a stage where we have finish the all the possible
symmetries that can be associated with a 2-fold axis.
308
Now, the reflection conditions as I mentioned we will not worry about it now, we have
already prepared the slides. So, let it be there, but we will look at the slide in more detail
when we discuss the diffraction conditions.
Of course, one can discuss the other space groups like Pc, Pm and so, on, but that is not
the idea as I told you that it is not required here to learn all 230 space groups. So, instead
we now go to the space group to P21/c. P21/c have been mentioning again and again that
this is a space group where we find most of the organic compounds go in to. And
309
therefore, it is better we understand this space group a little more clearly. Again, this is
the space group number in this case is 14. So, we have now jumped from the space group
5 to space group 14.
The arguments which go with the definition of the asymmetric unit, the special positions
and so on, will be very similar in the other space groups from between 5 and 7. So, we
do not have to go into the detail of it. So, we go to this space group number 14 very
important space group the unique axis is taken as b and a full symbol is P 1 21/c 1. You
see that 21/c will convert itself to a point group symmetry of 2/m. So, the moment we see
2/m the system has to be centrosymmetric we have learnt it already. So, since this system
has to be centrosymmetric the origin must be at the centre of symmetry that also we
decided and in fact, we derived the equivalent points of P 21/c in the previous class based
on this.
So, here you see that the origin is centre of symmetry the presence of the 2 1 screw is at ¼
the distance and then you see the presence of the glide plane which is shown up here
because, it is coming in a direction perpendicular to that and that is also at ¼ that is also
at ¼. And therefore, we see that the 2-fold axis screw axis and the glide planes are
shifted by ¼ and ¼. This is shifted ¼ perpendicular to that and this is the nomenclature
which we learnt long long ago in the earlier classes and this is the 2 1 screw axis which is
now appearing at ¼ position.
So, if this diagram is given again and nothing else is given, is it possible for us to guess
the space group. It is fairly simple now very easy because we see the centre of symmetry.
So, we know it is a centrosymmetric space group 2/m symmetry is centrosymmetric. So,
we say that this system is centrosymmetric, and we see a 2 1 axis ¼ removed and a glide
plane ¼ removed. We have understood that whenever there is a translational periodicity
involved space group; the systems with the no translation symmetry they move by ¼. In
this particular case both the systems 2 1 as well as c has the ¼ movement and therefore,
both of them move by ¼; one moves with respect to the other by ¼ let us put it that way.
So, we therefore, have now this equivalent point diagram. Now looking at the equivalent
point diagram is it possible for us to identify the symmetry element. So, if you look at
this point to this point you get a centre of symmetry, you look at this point to this point
you will get a 21 screw axis straight away ok. So, you therefore, have this and a 2 1 screw
310
axis. Now the fact that there is a centre of symmetry here brings a centre of symmetry
here as well and these two therefore, are related by the centre of symmetry. If they are
these two are related by the centre of symmetry and these two are by the 2 1 screw axis,
then there must be a possibility of a glide plane which is placed at ¼ the distance because
these two points will not be generated otherwise.
And this is one of the points which is generated by the c glide system and therefore, you
have 4 equivalent points these two equivalent points again are related by the presence of
the centre of symmetry here, these two are related by the presence of the centre of
symmetry there, these two are related by the 21 screw axis compared to this diagram you
can we see that these two are related by a 2 1 screw axis, these two are also related by a 2 1
screw axis. The fact that we are adding ½ to the symmetry related one here and the half
to symmetry related one here tells us that there is a possibility of glide plane at ¼.
So, this is therefore a uniquely determined space group. So, we have the ¼ translation
along the direction which is down here which is the c direction the 2 1 screw axis
developing and ¼ which is along the b axis which is the unique axis in case of P 21/c. So,
the equivalent positions in fact, they are symmetric unit now is very interesting. The
asymmetric unit is the y value which is the value perpendicular to this stops at ¼ and the
other 2 are in the full unit; that means, that we again have 2 molecules in the unit cell
now that is interesting because we should get only 1 molecule in the unit cell.
Now do we get 2 or 1 we get. In fact, only 1 because x value is full. So, the x value
which is up here is full and then the y value is only one fourth, and the z value is also
full. So, we get therefore, this plus that as the unit cell identifiers, but then you see that
these two are centrosymmetrically equivalent objects and then there is a half translation
in a direction perpendicular to that. So, it goes again beyond the ¼ plane. So, this is only
one let me explain again. The number of equivalent number of points in the asymmetric
unit here in this case is only 1 it is because of the fact it is a primitive lattice. So, in a
primitive lattice you should have only 1 in a C centred lattice you should have 2, in a F
centred lattice you will have 4 and, in a body centred lattice you will have 2, the
symmetry positions then give us the equivalent positions in the entire symmetry.
So, here you see that this if you consider the asymmetric unit as this one, since we are
stopping at ¼ along the y direction this point will not come this point none of these
311
points will come and therefore, will get only 1. So, when we get only 1 object it means
that it is primitive. So, these are all fairly consistent with the mathematical theory of
group groups and group theory is straightaway proves all these points.
But then when we are visualising the crystal structures, we cannot think of group
theoretic theoretical principles instead we can think of where the molecules are, where
the atoms are, and in what way the symmetry elements are related. So, this would be the
best way to learn about the space groups in my opinion of course, you will you by now
you also agree because you are going to recourse anyway. So, you will agree this is the
best method anyhow.
So, now these are the equivalent points we get 4 equivalent points written as 1 2 3 and 4
these 1 2 3 4 go to the references here. In fact, it is not marked here, but we can mark 1 2
3 4 given 1 2 3 4, I would suggest that you mark the 1 2 3 4 positions in the projection
given here which is 1, which is 2, which is 3, which 4 for example, this is 1 and this is 4.
So, the clue is already given you mark which is 2 which is 3. So, these are the 4
equivalent positions and there are special positions again in this space group these
special positions come like this. We have a position which is and the when the object is
associated with the origin. So, if you take the origin and put x y z to be 0 0 0, then you
will get 0 ½ ½ as the second possibility this will again be 0 0 0 this will be 0 ½ ½ the
312
translation along y the translation along z. So, the 0 ½ ½ will be the second equivalent
point.
So, there will be 2 objects rather than 4 associated when they are sitting in this special
position. So, you go to the other equivalent points the other Wyckoff letters we have 4
possibilities and these possibilities come at ½ 0 0, 0 0 ½, 0 ½ 0 what is the very
important here to remember and notice is that, if there is a molecule associated with 0 0 0
it is a different situation because we have only molecules at 0 0 0, 0 ½ ½ . So, if there is
something which is associated with ½ 0 0; that means, this particular centre of symmetry
is different from this centre of symmetry that centre of symmetry is different from that
centre of symmetry and so, on.
And that is why the Wyckoff notation change if all of them were similar than Wyckoff
notation would have been a and all 8 would have been grouped under this, but that is not
the case because the way in which these operate this 4 equivalent points operate decides
what are all the special positions. And therefore, we get 4 special positions and these 4
special positions have a multiplicity of 2 and therefore, these are independent special
positions; that means, if there is a molecule let us say we take a cluster of molecules
there are 2 molecules 2 different molecules and they crystallize inside a P21/c and there
are not very well defined interactions between these two molecules. So, it is a purely
vanderwall interactions.
Then the molecule one can sit here and a molecule can 2 can sit here and they can be in
the 2 different special positions that is alone. And that way you know in when you make
the co crystallization experiments when you make co crystals when you make salts when
you make eutectics and things like that, it may so, happen that these are the positions in
which these molecules can sit and they may have an interaction they may not have an
interaction between them. And that interaction is a question of intermolecular feature
which is independent of the location of the molecules.
So, that way we can have a molecule in this position and a molecule in that position, then
you will have 4 of those molecules and 2 of these molecules which will give a different
stoichiometric ratio. I am mentioning this because particularly from the pharmaceutical
point of view and also from materials industry point of view, this become very very
crucial. The way the molecules and them and sits themselves in the allowed positions of
313
the Wyckoff nature will tell us what is the nature of the molecule eventually, which all
what is the nature of the material which eventually a comes out of this whole
arrangement of super molecular assembly. So, that way it is very important.
The next thing we do is to we will skip the reflection conditions because this will come
only after we do scattering.
So, we go to the next space group which is the C centred space group and C2/c. So, we
know that C21/c is same as C2/c we already know that. Now, by looking at this diagram
314
also you can verify if you want to make a C21/c you move the origin to this fellow then
you will make it C21/c that anyway it is not important. What is important here is that this
space group again the unique axis is b, the full symbol is C 1 2/c 1. So, if you now
convert this symmetry in to point group symmetry it will be 2/m and its a monoclinic
system.
So, the equivalent point diagram is shown here and the you see now the asymmetric unit
is halved in all 3 directions. So, it is half here half there and half there. So, the moment
that it is half this fellow will not come up and therefore, there are 2 of them present one
here and one there or for that matter we are taking half the unit. So, this is half the unit
and that is half the unit. So, there are 2 refer additional point due to the presence of the C
centred lattice. So, since it is C centred lattice the asymmetric unit will contest consists
of 2 independent molecules, they are not dependent on each other anymore and therefore,
there will be 2 of those.
Also, we notice here that the origin is located at ī, and the glide plane c glide plane c is
coincident with ī. What happens in that situation is that this particular notation tells us
that there is a new symmetry that develops and that symmetry which develops is the so,
called n glide. We talked about the n glide and this n glide will develop in the case we
have a C2/c symmetry and n glide as you remember will add not just half to one of the
directions where we are now considering.
For example, if you consider the glide plane associated with this origin is c and that is a.
So, the translation symmetry that will be associated with n glide will be half along c and
half along a. So, the equivalent points will be x ½ plus y ½ plus z kind of a thing and
therefore, we now have a invoking a new symmetry element into the system, which is
now the n glide and this n glide is now coming away from the origin and in a direction
which is along the diagonal of a and c. So, there is a diagonal relationship like that. So,
you see here that these two if you consider they are related by an n glide because it is
half distance moved here and half distance up there. So, that is how you get the diagonal
glide invoked
315
Now, the equivalent points the equivalent points. Now, therefore, obviously, will double.
So, will have a eight of those. And then we have a several sets of special positions in
this. You notice that all these special positions to generate 4 equivalent points even
though there are 2 types of special positions which come up. You see in the so, far in the
previous examples up to this point where we discuss P21/c, this special position had the
same symmetry in this case it was ī, but if you come to C2/c, you see that this special
position are of 2 kinds.
So, the molecules or the atoms or whatever they makeup this structure, they can prefer
the centrosymmetric positions for their special positions and get 4 of those or they can
prefer the 2-fold location which is with respect to 0 y ¼ and they can get again 4
positions. So, the presence of the 2-fold or the ī symmetry in other words the molecule
now can either process a centre of symmetry or the molecule can process also a 2-fold
symmetry. If the molecule possesses the 2-fold symmetry it will go and sit with respect
to the 2-fold axis which is now a special position.
So, you notice that the Wyckoff notation the highest symmetry is always with respect to
0 0 0. So, it is going up there and we see that 4e is a position where you can have a 2-
fold symmetry. So, objects which are which have 2-fold symmetry preferably sit there
rather than in any one of these positions because these positions insist on having the
centre of symmetry. So, if the molecule has a centre of symmetry it prefers to go into
316
these 4 positions and if the molecules has a 2-fold symmetry it prefers to go to that
particular system.
So, this analysis of the equivalent points you can do it very easily now because you have
the 8 equivalent points listed up there take any one of these. For example, you take 0 y ¼
substitute that into these eight positions, you will end up with 4 positions. And those are
the 4 positions corresponding to the Wyckoff notation 4 e and that is how we get the C2
by C space group
So, the discussion so, far has been now with respect to the monoclinic symmetry. We
have discussed the triclinic symmetry, we have discussed the monoclinic symmetry, we
have discussed the equivalent points, we have discussed the equivalent point diagrams
we have discuss the symmetry diagrams. So, much of knowledge now pumped in the last
maybe 3 or 4 classes, but what is very clear about this whole thing is that the symmetry
is the driving force. So, the moment we have understood the symmetry over the previous
classes, this now becomes a straight forward understanding.
So, looking at an international table it does not look anymore like you know Greek and
Latin it will look like something which is very grammatically correct. Since you have
understood all the grammar associated with the equivalent points with the space groups
with the point groups and so, on it becomes a very facile thing to read the international
tables. In fact, international tables I should have brought it here it is a big thick book it is
like looks like a telephone directory and initially if you look at it looks very ugly because
all these pictures are there, and you will think it is like as insipid as a telephone directory.
But then one once you have got all the grammar understood, now you can read from one
end to the other of the international tables which I do not suggest you do, because it will
take half a lifetime.
But what is interesting to observe is that all these were worked out before any
experimental method of determining structures were found out. So, people looked at
crystal sitting in 1 place like I am just sitting for your class always here. So, the on the in
the more or less same position there is no change in the position I am stuck here. So, like
this people use to be stuck and the fellow who was stuck in was in fact, sitting at minus
thirty degrees in Siberia. So, that poor fellow sat there and looked at the crystals and then
317
arrived at all the possible space groups identify the crystal systems looking at crystals the
shapes of the crystals doing stereographic projections.
He created the 32-point groups and this was way back in the 17th 18th century seventeen
thirty and this is remarkable, but then if you happen to visit Siberia you will know why it
is he did he could only do that. Because the temperatures there are on average very very
low and it is always iced up you cannot go out and have enjoyment unless of course, you
go skiing around and that is also very dangerous. And therefore, you sit in one place and
do it and what is remarkable is all these were derived and shown x ray diffraction
experiments when they came they showed some of them are wrong not all are wrong, but
some of them are wrong and therefore, it is a very very important that you now learnt the
grammar. So, you can now start reading the text.
What is the best way to read a text you know the you do not feel like reading your text
books regularly instead if you are given a harry potter book or a lord of the rings book or
some novel, some interesting spy novel or something you will prefer to read it. Of
course, nowadays reading itself is avoided you read again on this stupid thing I am also
now teaching it on this one, even though I personally prefer teaching with a board and a
chalk which is what I do even now in my other class when I teach here IISc.
But then while we are all victims of modernization and computers have taken over and
now eventually with artificial intelligence coming up in full swing with computers will
probably rule us computers will tell you do not know how to teach, I will come and teach
such a day will come very soon. So, it is completely automatized and automated
understanding of everything will be clear. So, then I do not think we need the grammar
or anything that is associated with we can always go ahead and then determine this.
318
Prof. T.N. Guru Row
Lecture – 25
Details of Space Group 3
So, at up to this point we have seen the one single axis operation that is the twofold axis
operation. We looked at the twofold the mirror which is also two bar operation and then
2/m which is a center of symmetry involved operation. The same logic can be applied to
the single rotation involved space groups. For example, if you go to a tetragonal space
group and if there is only one axis which is the four-fold axis.
The same logic of generating equivalent points and the asymmetric unit and the diagram
which comes up with equivalent point positions being marked and the symmetry
positions being marked can be generated. If you go to a hexagonal system or a trigonal
system again with single axis operation those particular space groups can be generated in
this way. In all the cases the unique axis in those cases will be along the z axis z
direction.
We have already seen that because a=b in most of these cases in all these cases.
Therefore, we take the axis along the z direction and make the γ angle 120 degrees in
case of hexagonal and trigonal systems and in the case of the tetragonal system of
course, it is still 90 degrees. So, what happens if there is more than one axis which now
intersects at a point. So, there are several possibilities the first possibility is the question
of two-fold axis intersecting at one point. And when two two-fold axis intersect at one
point, we have been discussing this very many times.
319
But I can just remind you that if there is a point here and then there is an axis which goes
like that a two-fold axis. And then there is an axis which goes like that that is another
two-fold axis and invariably a third twofold axis will be generated and that is Euler’s
theorem. So, a point at this position let us say p a point p goes to a point q and then goes
to a point r and then it defines a point group symmetry 2 2 2 because we can come back
from one on to the other.
So, this is the intersection of two three twofold axis. So, it generates point groups 2 2 2
and therefore, this space groups could be P 2 2 2 any of the say a b or c face centered
because now it represents an orthorhombic system. Because the intersecting three
twofold axis will ensure α β γ is 90 degrees with respect to each other a b and c need not
have any fixed values.
So, we define now the orthorhombic system with respect to the 2 2 2 symmetry. So, the 2
2 2 symmetry will automatically now involve 2 m m, 2/m 2/m 2/m we have been
discussing this over several classes, so we are all familiar with that. Now suppose we
take a situation where instead of two twofold axes intersecting at a point like here, we
have a intersection of two 21 screw axis.
So, now, this becomes a 21 screw axis, this part we will remove and let me remove that
and show that. So, a twofold axis then there is another two fold axis this also becomes a
twofold axis, then can three twofold axes intersect at a point this is the question. And
320
Normans theorem tells us that it is yes two 21 screw axes can definitely intersect at a
point. And that is what is illustrated in this diagram.
So, you have therefore, a system there are two intersecting sets of screw axis that is what
is given here. So, we take two intersecting set of screw axis parallel to a and b they what
they produce is for example, this is the “a” direction. So, we have a screw axis 2 1 screw
axis here and a 21 screw there a 21 screw here now this intersects at the origin this is our
origin.
So, it intersects the other 21 screw up there. So, when such a thing happens, we generate
points this point now will generate by the 21 screw operation that one and then this one
will generate this point and by translational periodicity we bring it up here. And then we
also generate from this point the 21 screw at 0 and ½. Now will turn around and bring
this point.
So, therefore, we have now systems in which we can have equivalent points as given
here where the two 21 screw axes are intersecting with respect to each other. What will
happen to the third direction? Obviously, when we have these two points intersecting
each other at 0, 0, 0 suppose we take this as 0, 0, 0 then we generate a twofold axis
because of the fact you remember that whenever we have a translational component
involved the corresponding axis should move by one-fourth.
So, as a consequence here the axis moves by one-fourth in one direction one-fourth in
the other direction. So, it goes to this particular position and that becomes a twofold axis.
So, what therefore, results is a space group which is now P 21212 we will examine this
space group in more detail in a few minutes. But at this moment the take home lesson is
that if we have two intersecting 21 screw axis because of the fact that they intersect with
each other and they have to obey the Normans theorem.
The third axis which will now get generated will be because of the translation involved
in the two directions 21 21 the half translations if the third axis will develop at only one-
fourth position. The position one-fourth along x one-fourth along y thereby generating
this twofold symmetry and you can see that the equivalent points are nicely related by
twofold symmetry.
321
And that is how we get to this space group P 2 1212. So, we therefore, have point groups
the space groups P 222 and P 21212. Now is there a possibility to have three intersecting
21 screw axis. And if so what will be the validity of Normans theorem because the
Normans theorem says that with the intersection should be at a point.
Now; obviously, we see that if they intersect at one point, the third axis that is generated
is only a twofold. We can never generate a 2 1 axis in the third direction, but we do want a
situation where we can have three intersecting 21 screw axis. Why not? Because if we are
associating a twofold symmetry with respect to a given direction, we have a ≠ b ≠ c that
is not a problem, but α = β = γ are 90 degrees.
So, if you consider the parallelism with monoclinic symmetry in a monoclinic symmetry
if you have y axis as the unique axis or the z axis as the unique axis. If you have twofold
symmetry, we will also have 21 symmetry. So, that therefore, allows us to think that there
must be possibility of having 212121, 212121 in each direction which is allowed because
each of the directions are quantified or qualified by the twofold axis information.
So, if you convert to the point group, point group symmetry is still 2 2 2. So, since the
point group symmetry is 2 2 2 it is an allowed symmetry. So, therefore, we should have
212121 also as a possible space group in which case how these get disposed with respect
to each other. This diagram nicely illustrates that one comparing to this one. What you
see here is that you will have two intersecting 21 screw axis.
And now the intersections are not at 0 and ½ as you see here the intersection is at ¼ and
¾. That means, your first screw axis is going this way the second screw axis is removed
along the unit direction by ¼. So, that means, these two are not intersecting with each
other they are non-intersecting. So, if the two axes are non-intersecting then this is not 0,
0, 0. Anyway we do not need a 0, 0, 0 because this is a non-centrosymmetric space group
and so we have one axis which is corresponding to 0 and ½ and the other axis is
corresponding to ¼ and ¾.
So, if we have this kind of a system one twofold this way one twofold this way removed
by ¼ earlier they were coincident the moment you coincide the third axis becomes a
twofold that is the space group 2221 which we have shown here. So, in order to get 2 12121
we have to move this by one-fourth, three fourths then what happens is if you generate
322
the equivalent points based on these two operations. You will get this point removed by
half because the twofold screw axis is at one-fourth the distance from that whatever we
call it as 0 and as a consequence we get ½ plus 0.
Now the relationship between these two clearly shows the presence of a 21 screw axis
where is the 21 screw axis these 21 screw axes is also removed by one-fourth and one-
fourth, one-fourth along this direction and one-fourth along this direction. So, this is a
very interesting space group very beautiful space group which can be generated, and the
three 21 screw axes are non-intersecting and the they are removed from each other by one
fourth.
So, if there is a direction “a” on which we have fixed the 2 1 screw axis direction “b” will
have a one-fourth removal and direction c will be again removed by one-fourth. So, three
non intersecting axis which occur at one-fourth, one-fourth, one-fourth from each other.
So, this one-fourth, one-fourth, one-fourth is very significant because that will allow for
flexible molecules to crystallize more comfortably.
And that is how most of the organic compounds with asymmetric centers will crystallize
in 212121 is a very commonly occurring space group. It is very interesting that symmetry
now makes it very complicated in its arrangement such that flexible molecules can also
be organized into crystalline lattices. Otherwise you know if these flexible molecules
could not go into crystalline lattices in principle, we will not be able to crystallize.
Now what do you mean by crystallization in this context? By crystallization in this

context is that we want to arrest the molecules obeying a certain rule of symmetry. And 2
2 2 is an allowed point group symmetry so therefore, the space group P 2 12121 is also
allowed one once we allow the space group 2 12121 the flexibility is associated with each
of the axis. So, the here is the very unique example where the three axes do not intersect
at one point they are removed by one-fourth with respect to each other.
So, “a” direction is removed by one-fourth with respect to “b” direction; “b” is removed
by one-fourth with respect to both the “a” and “b” direction. So, then we have therefore,
three non intersecting screw axis displaced one-fourth with respect to each other. That is
how we describe this particular space group. We will see the more details of it in terms
of the space group diagrams now.
323
And we see the first one which is 2 1212. You see now the space group number is 18. So,
we have graduated ourselves from 1 to 18 now remember there are totally 230. We are
not going to; obviously, go through all of them. But here number 18 is now P 212121,
21212 the point group symmetry is 2 2 2 and this is an orthorhombic system.
And in the orthorhombic system we have represented the twofold symmetry. So, if you
look at this diagram now you know that this is an intersection of two twofold axis
systems one along this direction and the other along that direction. So, this is a 21 axis
this is another 21 axis. And along the two 2 1 axis you see that the intersection point is
shown at 0.
In other words, the twofold rotation is simpler operation than 2 1 operation because the 21
operation has a half translation. So, when we have a space group like this when we have
realized that the intersection of two 2 1 axis will generate a twofold one-fourth away.
Then it is as well possible to push these two 2 1 symmetries away from each other by one-
fourth and put the two fold at the origin. So, that we get a definition of our origin.
So, in this particular space group we get an origin and at that origin we put the 2 1 screw
in this direction one-fourth away and 2 1 screw in this direction one-fourth away. And this
then generates this equivalent point diagram you see that the asymmetric unit now will
be x half and y half and z full. And you will see the relationship between these this will
324
be a twofold rotation, and this will be a twofold rotation. You see that this is a twofold
rotation and you see that is a twofold rotation and so on.
You notice that the 21 screw operation which is one-fourth removed from here will not
generate any half positions here because we are looking at the projection diagram down
the twofold axis. But what you see is the presence of the 2 1 screw axis here. And the
presence of the 21 screw axis up there the presence of the 21 screw axis there generates
what find out. I want you to study this diagram and see what 2 1 screw axis is generating
which of these equivalent points. Of course, it will become; obviously, clear if you look
at this list of equivalent points which are shown here.
Ah the Wyckoff tells us that there are two possible special positions in this space group
both of them corresponding to the twofold axis along the z direction. So, you since there
are three directions, we put a dot dot 2 and a dot dot 2 to specify that the two fold is
along the z direction. It is not along the either the x or the y direction. And therefore, you
have 0 ½ z, ½ 0 z̅, 0 0 z, ½ ½ z̅.
Remember these similarities between this and what we saw when we took this z axis as
the unique axis when we studied this space group P 2. So, there is some similarity
between the 2. So, we get these special positions and two equivalent points in this and
four equivalent points there. So, the four equivalent points generate x y z; x̅ y̅ z that is the
twofold operation.
325
You see the twofold operation is restricted to the z direction. And then the 2 1 screw
operation takes it to x̅ + ½, y + ½, and z̅ that is the 21 screw operation about the y. And
the half adds up because the screw axis operation is verified which is associated with the
twofold being associated with 0, 0, 0 otherwise we will not have at this equivalent.
So, then we get this other equivalent point x + ½, y̅ + ½, and z̅. So, the molecules
therefore, or the atoms or whatever the objects them they arranged in this particular
fashion generating those four equivalent points. So, there are four equivalent points
points for P 21212. So, P 21212 is a very rare space group, but it occurs in many biological
molecules. In fact, from the literature if you go to the database and examine the database.
You see many of the nucleic acid bases containing one unit of nucleic acid that is
something like what we call as a nucleoside. Many nucleosides they go into this space
group 21212 it is very interesting you can study those structures. And see why they are
going into 21212 and that that brings us the concept of the folding and the DNA folding
and the axis about which the DNA goes through and so on. See DNA has a helical axis
and the DNA components show 21212 symmetry if you take different blocks of DNA and
that is quite interesting. One has to study the database for this anyway that is not the part
of the course here.
Of course, reflection conditions we will not consider now and we go to 2 12121. Now
notice that in 212121 there is nothing at the origin no symmetry in what I mean is there is
326
an origin 0, 0, 0, but no symmetry element is associated with the origin. The symmetry
elements are removed by one-fourth along this direction and one-fourth along that
direction and one-fourth along that direction.
So, these are the three 21 screw axes again the asymmetric unit is half along x half along
y and full along z. And what is very important is what is written here origin at midpoint
of three non intersecting pairs of parallel to 1 axis. Now this has to be examined carefully
because we have three directions now three directions are 90 degrees with respect to
each other. So, every one of the 21 screw operations are happening parallel to each other.
And parallel to the 21 axis, but these 21 axis now are midpoint between 3 non intersecting
pairs. So, that means, that if you take this origin it is midpoint of one-fourth midpoint of
one-fourth midpoint of one fourth. So, the all three are removed by one-fourth from each
other. And therefore, you get the origin and if you define the origin that way then only
you can generate this equivalent point diagram with the equivalent points coming at xyz
as is given here.
There is a certain amount of symmetry here in the generation of the equivalent points
also you know. I will tell you what I mean by that; if you look at xyz this x it goes to x̅,
y+½, z̅ + ½. And then you see that this is now first the number 3 is x̅ number 2 is y̅ and
number 3 is z̅. So, you have the three 21 screw axes generating x̅ y̅ z̅. And since both the
327
other two are removed by one-fourth and one-fourth you will have one of them positive
one of them negative.
So, let me explain again you have the x y z coordinate general position. Now you operate
one of the 21 symmetries the 21 symmetry which you will take you to x̅ +½ y̅ z+ ½. Now
I want you to go back in this picture and identify where is that point. So, if you identify
where is that equivalent point then you will know what am, what I am talking about. You
see that there is an inherent symmetry in the generation of these four equivalent points,
and I want you to identify that inherent symmetry.
So, you go to each one of these equivalent points find out how these equivalent points
are now coming on the diagram which is shown before. Because the if you go to the
international tables all these will come together except that this goes any way into the
next page the equivalent point information. So, take the equivalent point information and
verify for yourself that we have these positions.
Obviously, there will be no special positions in 21212 that makes it very comfortable for
organic or fly very flexible molecules to crystallize 21212. Now one of the take home
lessons could be or for that matter and as assignment could be that the prove or show
why there are no two ones there are no special positions in 2 1212. So, 212121 therefore, is
very unique and that has these equivalent positions and there is only one Wyckoff
position in 212121.
So, that way we have seen now a situation where we have looked at the more than one
axis operation and we have taken 2 2 2. So, anyone other than the more than one axis
operation which will be give rise to tetragonal systems it will give rise to hexagonal
systems. And of course; obviously, the isometric cubic systems all these systems,
therefore, have to be analyzed in this way.
So, the presence of more than one axis of symmetry makes the equivalences related to
each other in a much better fashion than just having one particular direction about which
there is an axis like in monoclinic. So, the possibility of arranging more symmetric
objects in higher symmetry space groups is very high compared to the probability of
arranging most more symmetric ones in the higher space group is more appropriate.
328
So, it is very hard to find a coordination chemistry involved component in a ī system or a
two-fold system you will see them mostly tetragonal and higher. What I mean by
coordinates and system involved this is a very loose word what I mean to say is that if
there is a tetrahedral or an octahedral unit in a crystal structure, they prefer to go to those
higher symmetry systems. Tetragonal and higher are the preferred systems for tetrahedral
and octahedral geometry.
Notice the difference when I am talking about the geometry I am talking about
tetrahedral and octahedral. And when I am talking about the arrangement of the atoms I
am talking about octahedron and tetrahedron. And both these are possible as we have
seen from the point group analysis in case of the cubic systems, we have the 4 3 2 and
the 2 3 the two space groups which will operate both on the tetrahedral as well as on the
octahedral right.
So, having come to this particular stage of P 212121 we will now go to the mm2 system.
And again, we will not take a simple mm2 system we will take a case where we have
also a translational periodicity associated with the point group with the space group. The
point group is mm2 the space group is P n a 21 what does it mean?
How do we describe this space group P n a 21? P n a 21 is a space group which is

primitive of course, P represents the primitive axis; “n” is a glide which is perpendicular
to the “a” direction, “n” is a glide which is normal to the “a” direction there are different
329
textbooks which give different ways of interpreting it. But let us take these two
definitions which are correct definitions “n” is normal to the “a” direction.
Since we talked n as a glide plane which is also a mirror effectively in the point group
symmetry the mirror has to be perpendicular to something or it has to be normal to a
given direction. So, the result direction that direction is characterized in this particular
crystal system which is the orthorhombic system in terms of a twofold symmetry and m
represents the 2 bar symmetry.
And therefore, we have an n glide which is either referred to as normal to the “a”
direction or perpendicular to the “a” direction. Similarly, we have an “a” glide which is
perpendicular to the “b” direction or normal to the “b” direction. And we have the 2 1 axis
when we say 21 axis we either say it is in the direction of the z axis or more appropriately
we say it is along the z direction or adjacent to the z direction. So, 2 1 axis when we refer
to the twofold rotation axis we refer to the direction when we refer to the glide planes we
refer to the mirror plane or the plane.
And therefore, it is two dimensional. The mirror planes which we refer to are two
dimensional planes on the other hand the axis are single axis rotations. And therefore, we
say that P n a 21 is a space group which has a point group symmetry of mm2 and n glide
perpendicular to the axis and “a” glide perpendicular to the base axis and a 2 1 screw
along this “z” direction or the “c” direction.
In the case of the orthorhombic system a b c and x y z are one on the same direction
because we have α,β,γ as 90 degrees. So, we can use them interchangeably even though
technically the correct usage when we refer to the mirror planes as perpendicular to some
direction it has to be x y or z. Now the x y or z in other crystal systems could need not
have the 90 degree angle they can have different angles, but you refer to the x y z
directions.
The value of a b and c will only define your unit cell dimensions the value of a b c will
only define unit cell dimensions. So, we can also say in this context that “n” is
perpendicular to “a”, “a” is perpendicular to “b”. But the correct definition will be “n” is
perpendicular to the x axis, “a” is perpendicular to the y axis, and so on so that is the
nomenclature.
330
Now you see the these diagram the symmetry diagram the symmetry diagram is showing
the presence of the 21 axis which is coming towards you and the presence of the “n” glide
which is perpendicular to the “a” axis. Now this is “b” by definition and vertically “a”
axis and you see that you have the n glide perpendicular to the “a” axis. This is the “a”
direction down the down the picture and this is the “b” direction.
So, the “a” glide is perpendicular to the “a” glide is “n” glide is perpendicular to the “a”
axis. Let me repeat so, having come to this particular stage of we have the 2 1 screw axis
associated with the “z” direction agreed. And then we have the “a” direction which is in
this way and you see that the “n” glide is perpendicular to the “a” direction “n” glide is
indicated by a line and the dot and the line on the dot and so on, so this is the “n glide”.
Just likewise we have a “a” glide; that means, a in the direction of a we shifted by one-
fourth and that one-fourth shift is in this direction.
And you see that we have the “a glide” which is perpendicular to the “b” direction. And
therefore, all the three symmetry elements are described here. So, if you are just given
this diagram in principle you should be able to say it is P n a 21 by just examining the
positions of the symmetry elements by now. Because you have learnt it over the last one
hour or so in the previous classes that the way in which we interpret these diagrams.
So, you already are an expert to see that the moment this diagram is given you know the
space group is P n a 21. And if you get to that understanding of this then your
crystallographic knowledge is reasonably good. And then we also have a situation where
this can be interpreted in terms of the equivalent point diagrams.
Now you examine the equivalent point diagrams the way in which it is generally
recommended is that we keep both these diagrams together as we go to higher and higher
space groups in order to understand the space group definition clearly. And that is what I
will try to do in the next couple of minutes in this particular space group. And that is a
generalization of any interpretation of any space group.
Here is a diagram which shows you the symmetry dispositions the positions of where the
symmetry elements are and how they are coming and what is the nature of the symmetry
element; This is a good example to take it because this is a very straightforward example
the it is an orthorhombic system. So, a ≠ b ≠ c, α ,β, γ are at 90 degrees with respect to
each other.
331
The point group symmetry turns out to be mm2. So, we have a two-mirror system and a
one 2-fold system. And this, therefore, is the representation of the “ab” plane and this is
therefore, the direction which is coming out from the board is the z axis or the z
direction. So, the twofold therefore, is associated with the z they 2 1 screw. Since it is a 2 1
screw we now mark the screws at this position at 0, 0, 0 then half the distance will
automatically come because of the edges being also carrying the 2 1 screw axis by
definition.
So, we fill up this 21 screw axis first and then we see that we now look at this diagram.
And see in this diagram whether we generate these 21 screw axis. So, if you like a take
this particular point operate the 21 screw axis you go to that point. And you translate it by
one unit and then one more unit you get here, and you see that this 2 1 screw axis is the
operation between this point and that point. And therefore, we verify the space group
definition with the z axis getting the 21 screw axis in this particular projection.
Having got that information then we look at the other two possibilities which are the “n
glide” and “a glide” and the compare that diagram with this diagram with those. Now if
this is the n glide operation then the n glide is going here; that means, you should have a
translation of half and half. So, which is the point which you will get by symmetry
relationship?
The point is from here to that one and that is an n glide operation. And again, you see
that these two become a 21 screw operation. So, you got the n glide operation now you
have the presence of the “a glide” perpendicular to this direction. And you see that the
“a glide” perpendicular to direction generates this and this point. So, you see that these
this is the twofold generated position this is the n glide generated position. And this is
also the twofold relation position because of this definition of the group, and this is also
the position of the generation due to the “a glide”. So, you therefore, have four
equivalent points in P n a 21.
Now, what are these four equivalent points you can see next slide, but you can write it
down. So, some of the assignments may ask you to look at this diagram and write down
the equivalent points as we go along and that is a very straightforward thing. Because
you take this as x y z you automatically write the rest of the equivalent positions by
looking at the disposition of the symmetry elements in the symmetry diagram. So,
332
suppose you are given these two diagrams, you can write the equivalent points and what
are the equivalent points.
They appear here it is x y z and you have the twofold axis the twofold axis is associated
with the z direction. So, you get x̅ y̅ z+ ½. So, the second point which you generate from
x y z is x̅ y̅ z+ ½. Then you have the “n glide”, n glide will take you to x + ½. Now comes
the question which is this point number three in the diagram here.
So, if you look at what is that point number three in this diagram you have several
choices you know these two are not possible and what is the choice, we have for this one.
So, the equivalent point is x + ½ y̅ +½ z . So, if this is the “a” direction then x + ½. So,
if you take this point x + ½ will be what it will go there ok. And then you see that it is y̅ +
½. So, you will come here and so this is the point number 3 ok.
So, you can therefore, correlate this with the equivalent points and verify this the fourth
point where it comes. And if you just go through back and forth between these two slides
in principle you will be able to get it. Now because of the fact that that we have now
three non translational components just like 212121 here we have three non translational
components; n a 21, n a 21 is you know n involves a half translation, a involves a half
translation, 21 involves a half translation. So, therefore, there are no special positions in
this space group.
333
Lecture – 26
Details of Space Groups 4
So, as we now continue with the P n a 21 , one other aspect which we should notice here
is that the origin now, see everywhere here after the definition of the location of the
origin is important. The one once we draw these two diagrams these two diagrams refer
to some origin 0 0 0 and in case there is a centre of symmetry no problem. Because, the 0
0 0 will coincide with the centre of symmetry.
But in case we do not have a centre of symmetry, we have to specify where the origin is.
So, in this particular case, it is at 1 1 21. So, the 21 screw axis in the z direction is now
coinciding with the origin. So that means, the value of z is put equal to 0 at that point.
So, this is very crucial because we have to define the origin term very clearly.
334
Now, we go to the next system when the orthorhombic symmetry which is all three
mirrors. So, you see that the diagrams look start to look more and more complicated. But
one once you have got the gist of it and you have understood what is it all about, it
becomes fairly straightforward. Let us see how quickly and how comfortably we
understand the space group P b c a. This is another space group which is fairly common
among various organic and inorganic compounds, particularly when you have a flexible
inorganic system or organic inorganic framework structure and so on; P b c a is a
preferred space group.
Now, therefore we should know something about this space group you see the space
group number is 61; you are graduating very fast towards 230. In fact, after this I do not
think we will do any more. So, we can as well say we are now coming towards the end of
understanding the space groups, by looking at two diagrams. And then comparing those
two diagrams with the equivalent points which are appearing in the next part, next page
in fact, in the international table and next slide in our presentation.
So, let us see what happens in case we have a P b c a; so, one of the things which is
immediately obvious says that it is a primitive lattice. So, we will not have any any
centering information coming in, any here face centered information coming in and
things like that. So this therefore, is a situation where we have no problems primitive and
there is therefore, the number of lattice points will be one. So, we have to worry about
335
only one lattice point. The moment we bring in a C centered lattice let us say it is C m m
a or something or C m n a or something. In which case we ha_ve to now bring in two
lattice points, one at x y z the other at ½ + x ½ + y z.
If it is the face centered lattice we have to bring in four lattice points. If it is a body
centered lattice we bring in two lattice points and I am bringing that discussion here
because orthorhombic symmetry allows for all kinds of lattices. It allows for the
primitive lattice, it allows for A centering, B centering, C centering of course, A B C are
interchangeable. So, we can always say it is A centered or C centered, one once we say
that they are interchangeable with respect to each other because, the three angles are 90
degrees with respect to each other a ≠ b ≠ c.
So, but we can define abc as bca and cab. So, the symmetry is still maintained. The 360
degree rotation of the symmetry is still maintained and so, we can call them
interchangeably. The interesting fact is that since all these possible lattice symmetries
can come up in an orthorhombic system, it is very easy to interpret it at this stage. So,
whenever you have a primitive lattice you have only one lattice point to worry about xyz
and after that symmetry elements we generate is purely by symmetry operations. If you
have a F centered lattice there are four center related operations the face centered related
operations.
So, for an every x y z you will get a ½ + x ½ + y z, ½ + y ½ + z x and so on. So, four
of them you will get and if it is a body centered system you will get for every x y z, ½ +
x ½ + y ½ + z. So, whatever symmetry operations you are now going to perform on x y
z the same in symmetry operations you have to perform on the next lattice also. So, what
happens is, that the number of equivalent points will increase, but the number of
symmetry related points will still remain the same.
So, in other words if this is instead of P b c a it becomes I b c a, what will happen is,
that the number of number of equivalent points which occur here which is in fact, 8 will
become 16. So, the same 8 will exist with the same symmetry, but now all of them are
moved by ½ ½ ½ along the unit cell. So, additionals operations of symmetry appear, and
the signature of the point group symmetry remains exactly the same.
So, whether it is P b c a or I b c a the point group symmetry is still m m m. And this is a

point which should be noted and this is a point which is which is going to be important
336
when we actually try to solve the structures and fix the positions of the atoms and so on.
So, the lattice points we will see all the symmetry operations associated with them
independently. So, if the C centered lattice we will have two, I centered lattice two and F
centered lattice four. I am, I am a bit repetitive basically to drive home the point that you
have to worry about these issues when you derive the equivalent points and particularly
when you try to discern the difference between these two diagrams which are shown
here. So, this diagram shows you the symmetry points and this diagram now shows you
the equivalent points.
So, the space group number is 61 and you see the full symbol the full symbol is 21/b 21/c
and 21/a. Obviously, because we cannot have a m m m combination, we have already
proved long long ago in earlier classes that, a improper improper improper combination
is not allowed. So, it has to be when we write m m m that is why we are happy to write m
m m, because m m m actually represents 2/m 2/m 2/m and therefore, if you now see the
independent derivation for each of these in the full symbol you will see it is P 2/m 2/m 2/
m. In fact, that is what it is and we reduce it to P b c a.
So, when we say P b c a the two and the presence of 21 with respect to each of the axis is
already implied which is already there. So, the in the way in which we described this
space group is not we have a primitive space group, there is a b glide perpendicular to a,
a c glide perpendicular to b and a glide perpendicular to c. Suppose I have a space group
which is P c a b is that possible is the question.
So, this is a question which we can answer right away, suppose instead of P b c a I have
a space group, let me write it down for you so, that we discussed that, instead of P b c a
we have a space group P c a b is it a possible space group? If P c a b is a possible space
group is it different from P b c a or it is a new space group. We said there are only 230
space groups right and is it going to define a new space group? These questions come up.
So, a primitive lattice is a primitive lattice there is no change. What it represents here is
there is a b glide perpendicular to the a direction. Here there is a c glide perpendicular to
the a direction. And in this case there is a c glide perpendicular to the b direction, here if
there is an a glide perpendicular to the b direction. And in the last there is an a glide
perpendicular to the c direction there is a b glide perpendicular to the c direction. So,
what is the answer? The answer is inherent in this definition the way in which we now
337
invoke the presence of the glide planes. So, P b c a is a commonly represented space
group which refers to a b and c axis.
Now, what all you have done is you have taken the c axis to be b and the a axis to be c
and of course, the c axis to be a ok. So, since you have change the axis directions, the
glide plane directions have also changed. Remember in an orthorhombic system the three
angles are 90 degrees with respect to each other. a ≠ b ≠ c. So, we can call any one of
them as a any one of them as b and any one of them as c. So, we can have therefore,
equivalences of space group. So, P b c a is same as P c a b except that your a b c
positions are interchanged.
So, in this particular case the positions are a b c ok. And in this particular case the
positions are this position with respect to this is a and this is b and this is c, am I right c
perpendicular to b, a perpendicular to c. So, this is your crystal system now. It does not
matter whether b is c, c is a or a is b because we can interchange them with respect to a
b c. So, these two space groups P b c a and P c a b are one and the same. So, this is
something which we should remember when we are dealing with the orthorhombic
systems. In orthorhombic systems we can therefore, you it may so, happen that your
molecules they do not know the definition which you want to give. They do not know
which should be a which should be b which should be c your molecules will go on
crystallize and when you go to the machine and determine the space group believe it or
not this the computer says it is P c a b.
Then you will quickly go into your international tables P c a b, P c a b where is it where
is it. You want to see where the equivalent positions are, and you will never find it. You
will never find it because the standard for this is P b c a. The standard space group as we
call it or the normally accepted space group is P b c a. So; that means, you have to
change your axis directions, your a b c now has to follow the P b c a the space group. So,
the a b c should be changed such that we get a b axis perpendicular to a, c axis
perpendicular to b and then a axis perpendicular to c. So, when we do this particular
change then it is one and the same.
So, this is something which you should remember and then whenever you have a
situation like this where the molecule has gone into P c a b you can either change the
symmetry elements and make them into P b c a or you solve the structure in P c a b, but
338
eventually report in P b c a. Yeah all are allowed. It depends upon what you would like
to look in that molecule. Suppose that molecule has a very special behavior, suppose
there is one direction which is very very large and that is the direction which is very
crucial for the property that compound is showing, and that happens to tell us that the
space group is P c a b and that large direction is b. Then we would rather keep it in this
form then report as P b c a.
So, you will see lot of literature particularly in inorganic compounds and also in
materials oriented compounds the space groups which are not these normal space groups
are acceptable. But what people do is when you read that literature in the brackets they
will give the standard setting is P b c a. So, they will indicate that even though the
structure is reported in this space group, the standard setting is P b c a. So, this is how
the space groups are unique and therefore, there are not more than 230 space groups.
Otherwise the numbers would have increased enormously. For any given abc we can
define these interchangeably to the combinations.
So, the combinations will keep on increasing which we do not want to consider. So, P b
c a is number 61, and if you crystallize the compound and it tells you the machine tells
you and you do the analysis it is P b c a, you do not have to panic it is still the space
group 61. Either you convert all these things to P b c a, which would be the most safe
thing to do, but if you do not have to do it because you have something very special in
your compound which is along the b direction and that wants to be along the z axis. Keep
it along the z axis, suppose you have a crystal grown in that direction. And you want to
measure some property associated with it. Of course, if you have a P b c a crystal it may
not be very interesting. Because it is nicely beautifully, center symmetric.
But, if you have a system where the non centric behavior comes up; P n a the previous
space group which we discussed for example, P n a 21. So, what suppose you have an n
glide perpendicular to a axis a glide perpendicular to b axis 21 is along the z axis. You
may have a equivalent space group P 21 n b. So, this P 21 n b is equivalent of P n a 21; if
for all practical purposes because we can change a, b and c. So, this particular table
where we can have the standard representation given in the international table and all
other possible variations which come up due to a, b and c being interchanged. A table of
that is provided in the international tables towards the end.
339
So, we can go to the end of the international tables look for these tables. In case you get a
space group which is not listed in the 230 space groups. The computer will give you the
space group what whatever it finds with respect to the molecular system which is there in
the in your crystal. Because it is going to base it on the diffraction conditions which it
finds and one once it has happened then one has to go and look at the standard setting
and so, in the when you report in literature, you have to report that the space group in
which you determine the structure within bracket you should say standard setting is P n
a 21.
So, it may be very important that these 21 axis which is found in P n a 21 has a very
special property like ferroelectricity or ferromagnetism shown in that particular direction,
and you want that to be z axis because, your crystal is grown in that particular direction.
It is very clearly seen in the morphology of the crystal. Then we would rather maintain
that particular space group, rather than report the standard space group. So, this is
something which you should bear in mind and never get into troubles when you discuss
this.
So, when we go to P b c a therefore, these are the equivalent point positions. I want you
to do an a little exercise again here. The exercise is to compare these and see which are
these corresponding this of course, is very straightforward. Here you have a center of
symmetry and this to this is now you are all experts. So, this to this is now the glide
plane and also the 21 screw. You see that the 21 screw and the glide plane coincide
because you have a system where this is a direction and in the a direction you have a b
glide which is perpendicular to the a direction. You also have the 21 axis in that
direction.
So, the 21 and the glide they move by one-fourth so, this is the glide direction. And
similarly the c axis and the z axis there is b axis on the c axis are also having the glide
planes with a 21 screw which is now removed by one-fourth here. And in this case again
it is in plane. So, you notice that the whenever you have a glide and they associated 21
they are in the plane of the diagram. Whenever you have the center of symmetry and the
21 the 21 symmetry goes by one-fourth that defines the translational periodicity which is
½ unit along a, ½ unit along b, ½ unit along c. And therefore, you can now correlate this
diagram with this diagram and at the same time also look at the equivalent points.
340
And here there are 6 equivalent points. Here is something which you have to see 1 and 5
are shown as x y z , x̅ y̅ z̅. I want you to do a little exercise here. So 1, 5, 2, 6 you take
these four to start with and then generate these four and one 1 is x y z. So, we go back to
the previous slide you see this is one and then 5 is this one because 5 is the x̅ y̅ z̅. So, like
that you I want you to mark 1, 2, 3, 4, 5, 6, 7, 8 the 8 equivalent points in this diagram.
And in this diagram we already know there are eight equivalent points because z is equal
to 8.
So, I want you to isolate and identify the equivalent points which we have in the next
slide with respect to 1, 2, 3, 4, 5, 6, 7, and 8. Whichever is there inside the unit cell I
want to also tell you that it is still only one in the asymmetric unit and I want you to
verify doubly why it is still only one object inside the asymmetric unit, you notice that
the asymmetric unit have shrunk now. It is x= ½ y= ½ z= ½ and you see all other
generations which you are making is either “½ + or ½-”.
So, that does not go with the existing one. So, you will have therefore, only one. So, the
primitive lattice is dominating in this particular structure and that is how it is a primitive
lattice. If it is centered you will have 2 and F centered 4 and so on. Right so, the other
thing which you have to notice here is that this also has special positions and these
special positions both of them are center symmetry. Now this is something which I want
to discuss a little bit here, before we end this particular session that discussion is the
following.
341
Say when we look at you see this is the C 2/c system, right? C centered lattice a 2/c is the
point group which gives rise to 2/m. In fact, and you have these diagrams and you have
related these two diagrams beforehand. Now, what you do is you look at the special
positions. Here there is a special position associated with ī as well as the 2 fold. The
space group has the 2 fold and the ī. The space group has also got the c glide, but there is
no possible association with c glide because it has a translation component agreed. Now,
whenever you have a translation component symmetry element with a space group that
particular component will not have special positions with respect to that.
So, let us go now straight away to the space group we have been discussing because, that
is where a very important point comes up. If you see here the P b c a space group what is
your take home? You have 21 you have b you have 21 you have c you have 21 you have a
so, what would be the special positions? The special positions cannot be associated with
any of these symmetry elements individually. So, there cannot be anything associated
with 21 there cannot be anything associated with b, it cannot be anything associated with
c and a. So, what is it where is the special position coming in a space group like this. If
one examines that carefully you have a solution to the problem, in this particular case
you know that this is a centre of symmetric space group.
342
Centre of symmetry is not associated with any axis or any translational periodicity. The
periodicity that comes up from a centre of symmetry is only one unit along x y or z. And
then of course, this it appears also at the half position as a consequence of symmetry. In
other words this 21 position can be this centre of symmetry position can be also this
centre of symmetry position or this centre of symmetry position. That means, this
framework is the lattice and this lattice we can slide down in all three directions and the
again it have the position of the centre of symmetry fixed and since the centre of
symmetry is a fixed position we will have special positions then the atom or molecule or
the system which we are discussing sits at the centre of symmetric position ok.
It is very clear very simple and therefore, you will see now the special positions will be
associated only with ī. You have all the other symmetries, but only ī is allowed. So, if
you look at all these special positions which you come across in the 230 space groups.
All those special positions will occur only with symmetry elements which are not having
translation components. So, it will be ī it will be 2 it will be mirror and things like that it
will not be associated with any of those so, ī, 2 and m these are all these special
positions you will have. Of course, you can have combinations of those for example; in 4
fold symmetry you may have a 4/m symmetry.
343
The 4/m symmetry in principle is centre of symmetric. Even though you have a 4 fold
rotation 90 degree rotation coming up it is in principle so, it could be offering a special
position. So, one studies the international tables fully which we are not going to hereafter
we are going now a little forward and go eventually to the structure determination
protocols and so on. At this particular position what we have to therefore, do is to to
consists consolidate what we have learnt in space groups. And to consolidate what we
have learnt in space groups. The best example would be to take something which is a
realistic structure. Look at a realistic structure and see how the realistic structure
distributes itself in a given space group.
I have purposely taken a space group which is not a standard space group. The
corresponding standard space group will be what in this case P m c n find out. A mirror
perpendicular to a, a c glide perpendicular to b and n glide perpendicular to c; so, we will
have to find out what is the corresponding standard space group. So, we have to go to the
international tables and go to the (Refer Time: 24:00) end of the international tables look
at what is appear what is appearing as P m c n what is the translational change we have
to do because it is an orthorhombic system.
So, a b and c have to be interchanged. What interchange will give us the standard crystal
system. That is first thing we will do in the next class. So, to wind up this particular class
we will see that we have the space groups now identified up to orthorhombic symmetry.
We have seen the triclinic system, we have seen the monoclinic system, we have seen
orthorhombic systems. In orthorhombic systems we have seen all possible point group
symmetries we have seen 2 2 2, we have seen m m 2 and m m m.
Now, in that we have considered all possible variations we have considered the
possibility of introducing 21 axis and we have introduced the 3 non intersecting 21 axis
we have also introduce the positions 2 mirror positions and the 21 axis the 2 mirrors now
having glide planes. Like P n a 21, which we have seen. And we have also seen P b c a
which will involve the 21/m, 21/m, 21/m centre of symmetric operation. We have
examined all in all cases wherever a special position is possible and in what way the
symmetry elements rearranged themselves in those special positions starting from the
triclinic system. We started from P ī and ended up with this particular thing.
344
So, in essence we have covered now all aspects that are required to be known in space
groups. We have not gone to the higher space groups because of the fact that which is
not required for this course because now here onwards we have to find out the
methodologies by means of which we find given a crystal. How do we find the space
groups and how we assign the space group to a given material? And then how the
material arranges itself in three dimensional space. So, we have come to that level and
therefore, what I will do is I will take this system P m c m. And in this take an example
of a known compound and we are going to look at the known compound being calcium
carbonate.
So, in the next discussion we will look at calcium carbonate and see how calcium
carbonate the calcium the carbon and the oxygen atoms arrange themselves in this space
group P m c n. Keeping all the symmetry information intact, how calcium carbon and
oxygen decide to go into this unit cell. Now how will they go into the unit cell is a very
open question. So, we will have to therefore, in principle find out how many such
calcium carbonate units can go into this unit cell. So, how will we do that? We do it by
determining the unit cell dimension.
So, we know the a b c values. If we know the a b c values we can calculate the volume
of the unit cell. Having calculated the volume of the unit cell since calcium carbonate we
know the molecular weight we can now calculate given the volume and the molecular
weight the density of the crystal. Assuming that this particular space group has; so, many
equivalent points. So, P m c n we examine with respect to equivalent points, and these
equivalent points how can we distribute the density which we can have in this particular
case for calcium carbonate. It is also suggested and in fact, in our school days I mean
when we were studying for our PhD not school days.
Whenever we looked at a crystal structure and whenever we grew a crystal we were

asked to determine the density of the crystal. In fact, papers were not getting accepted
without the acceptance of the density of the crystal. And therefore, determination of the
density of the crystal is very important how are we going to determine the density of the
crystal. So, we will start the next class from understanding how we determine the density
of a crystal which is very crucial. And then we see one once we have the density we have
the molecular weight we have the volume from a b c values we can then calculate how to
distribute individual atoms inside this keeping the symmetry in mind.
345
Now, I am saying as though we have all solutions for all crystal structures, obviously not
because we are taking a very simple compound calcium carbonate. Suppose we have a
very large molecule suppose we have a protein I cannot use this approach which I am
going to describe to find out where these atoms are. This is a very simple structure and it
is going into a space group where it has to obey certain symmetries. And therefore,
whatever symmetry it is going to obey we will try to reconstruct the structure in this
particular space group.
346
Lecture – 27
Crystal Structure of Calcium Carbonate
So, we start discussing the space group P mcn as we concluded in the previous section.
That we have now an enough knowledge of the way in which the diagram which is
consisting of symmetry elements appear. And also the diagram which consists of
equivalent points appear.
We also discussed the fact that we an orthorhombic system a b c can be interchanged.

So, depending upon that the corresponding symbols associated with the point group
symmetry will also change. And therefore, you know the even though this point group
symmetry is here 2/m 2/m 2/m the order in which they come can be not exactly the way
in which it is written in the international tables. So, this particular space group, if one
structures in the international tables you will not find. However, this is the space group in
which the structure of calcium carbonate is reported earlier on. And that is why I thought
that I will take this as an example.
The other issue is that we have, we are now in a situation to appreciate the structural
aspects associated with the space groups. See we know that we have been discussing that
347
with respect to objects like here, but instead of objects now we have realistic atoms,
which are connected to each other to form a molecule or a compound which is
represented in the unit cell and so on. So, since we have come to that kind of a situation,
we will now take an example of the structure of calcium carbonate. And it is reported in
literature in the space group Pmcn. So, we will take into account this space group Pmcn.
You are all familiar now with the way in which P mcn develops and the way in which
the positions of the individual symmetry operations appear in this space group. We know
that one once there is a primitive lattice the number of lattice points is 1. The mirror
symmetry is now perpendicular to the a axis you see that this is the mirror and it is
perpendicular to the a axis. Remember this is 2/m 2/c 2/n and with our knowledge on the
way in which the symmetry element moves away from the origin when we have a
translation involved symmetry operations. This actually now represents the mirror at
one-fourth along the a direction.
So, it moves one-fourth perpendicular, this is the a direction it moves one this is the a
direction sorry, it goes to one one-fourth and then it is normal to the a direction. So, the
mirror plane is normal to the a direction, and it is removed by one-fourth in these
direction. Now if we take the c glide into account that again moves by one-fourth and
this moves now in a direction which is perpendicular to this. we also invoke the presence
of additional 21 screw axis, as we discussed in the case of Pbca additional 21 screw axis
will develop at various places along the three directions.
The fact that this is a centrosymmetic system, we now put the centre of symmetry at the
origin. The moment the system is centrosymmetic we have to put this symmetry centre
of symmetry at the origin. So, 000 therefore, is up here, and this is the a direction, this is
the b direction and we have now a mirror plane which is removed by one-fourth and is
normal to the a direction. A c glide removed by one-fourth and is normal to the b
direction.
And the appearance of the n glide is shown up here, because it occurs in the third
dimension and therefore, it is shown with a typical symbol which represents the glide
plane and this arrow which is coming in the diagonal says that it is the n glide. So, we
therefore, have the full representation of the mirror, the c glide and the n glide in this
symmetry diagram. So, this is a complete symmetry diagram representing the P mcn.
348
The way in which the equivalent points develop as a consequence are shown here and we
will have to refer to the next slide to understand what these positions are.
For example, there are 8 possible symmetry positions in the unit cell in the general
position. The point symmetry being 1 and those 8 positions are given here.
Since the crystal system is centrosymmetic we have plus x y z and also –x –y –z . So, x y
z and x̅ y̅ z̅ which are now indicated here. Now the presence of the mirror plane which is
now at one-fourth will generate the mirror plane is going up here now. And since the
mirror plane is going down here in this direction, we see that there is a mirror
relationship between these two objects and so, also the mirror relationship between these
two objects which are generated due to the presence of the centre of symmetry which
exists here.
So, we see that also the presence of these n glide and the c glide, will introduce these two
additional symmetry elements. So, overall we get a mirror symmetry at this position
under c glide at that position, and so the total number of equal end points after
transforming from outside of the unit cell to the inside will give raise to 8 of them. I want
you to learn and work out how these equivalent points come based on the table which is
given up here, I do not think we have time to discuss that again and again. We have
already seen the orthorhombic symmetry we have seen P mc21, we have seen Pbca and
349
we have of course, very clearly seen 2 1 21 21. So, we have a knowledge of how these
equivalent points get generated.
And since the coordinates are given here find out where these coordinates are with
respect to the diagram which is shown here. Then you will be convinced that there are 8
equivalent positions corresponding to these 8 general positions. Now the it so happens
that in this space group because of the additional possibility of the object having a mirror
symmetry or a ͞1 symmetry that is the centre of symmetry or a ͞1 symmetry which is again
a centre of symmetry. We get special positions. We are quite aware of how these special
positions develop.
So, I will not spend time again on this one, except to tell you that there are three possible
positions which all generate 4 multiplicity. So, z = 4 which means that in fact, we will
have only half the molecular allowed, with respect to the general 8 positions. So, these
therefore, define now the mirror position the ͞1 position and ͞1 position and the total
number of equivalent points are 8. Now we come to a situation where we look at the
realistic situation.
So, we are now going to take this structure of calcium carbonate, the particular there are
various ways in which calcium carbonate crystallizes and one of the ways in which it
crystallizes is referred to as the mineral is referred to the mineral phase which is called
Aragonite. So, this is a naturally occurring mineral, this mineral aragonite. And the
structure of this goes into the space group P mcn. The crystal structure has been
determined, crystal single crystals of aragonite have been grown in the laboratory the
structure has been determined.
Now, the one once the structure has been determined, we get to know the total number of
calcium carbonates that are present inside the unit cell depending upon the volume of the
unit cell and other factors which I will discuss in a minute. This gives us(Refer time:
08:15) to 4 calcium carbonate that is. So, z is equal to 4. So, there are 4 calcium
carbonates units inside the unit cell. That means, the unit cell can accommodate 8
equivalent positions, but it can have only four calciums.
So, since it can have only four calciums, the calcium atoms can be associated with a
mirror point symmetry, or a ͞1 point symmetry which is with 0 0 0 or the other ͞1 point
symmetry which is with 0 ½ 0. Because all these three possibilities can be taken into
350
account, and calcium can have a choice where it wants to sit. See this is how the
structures develop you see calcium carbonate we already know has certain restrictions
from the way in which it forms the structure. The calcium coordination site is the one
which decides how the coordination around calcium should develop in the crystal
structure.
So, the calcium therefore, since there are only 4 calciums, these four calciums can go to
m or ͞1 or ͞1 , it can take any one of these. And same is the story with the carbon, there are
4 carbon atoms. On the other hand there are 4 times 3, there are 12 oxygen atoms. So,
these oxygen atoms have a very clear liberty of how they could arrange in generating the
calcium coordination. Because these 12 can be, see 8 of them can be sitting in general
position. And 4 can be in any of these three special positions, that is one option. The
other option is three types of oxygen sit in each of these ͞1 , ͞1 and m positions. That will
also total up to 12. So, now, is a situation where we have a unit cell which is P mcn into
which the crystal structure has gone into, calcium carbonate has gone into that structure
generating the mineral aragonite which is a naturally occurring mineral.
So, when the naturally occurring mineral formed, these were the options that were given
for the compound to form. Calcium, carbon and oxygen these were the options which
were given for this thing to form. So, calcium therefore, can have four that can be 4
calciums, 4 carbons and 12 oxygens and as we discussed the distribution among the
various types of symmetry elements in the unit cell is possible. Now how do we find out
how many molecules are there in the unit cell? that is done by using this formula n M
equals N rho V. This formula n M = N ρ V.
Let me explain this formula in a little more detail. n is the total number of molecules in
the unit cell which could be the z value. M is the molecular weight associated with the
compound. Since we know what compound is crystallizing we can have the
identification of the molecular weight, we can determine the molecular weight of
calcium carbonate by knowing the atomic weight of each one of these atoms and adding
them up into the compounds so, we have the molecular weight. We can also determine
the molecular weight of this particular crystal using mass spec.
So, we will have a measure of the molecular weight, so experimentally or theoretically

we can simply calculate by using the atomic masses. So, we know the molecular weight
351
of the system. N is the Avogadro number. The Avogadro number is 6.023 into 10 to the
power of 23 as most of you know per mole. And that Avogadro number is a constant and
it can fix the molecules inside the unit cell. So, N is the Avogadro number. ρ is the
density of the crystal. We will talk about the density of the crystal in a little while, ρ is
the density of the crystal and V is the volume of the unit cell.
So, in this particular case one once we know the value of a b and c, this is an
orthorhombic system because the crystal system as you see is P mcn crystal system is
P mcn. So, it is an orthorhombic system. In the orthorhombic system the volume will be
a times b times c. The product of the three a, b and c. And therefore, if V = a.(b x c) we
can calculate the volume. So, one once we have this volume available to us, which is
from the experiment which has been determined to using x-ray diffraction to show that
there are four molecules in the unit cell. How do we find out that there are four
molecules in the unit cell is what we are discussing now.
So, we first determine the cell dimensions. We determine the value of a, b, c; α, β, γ . We

identify the space group and this has been identified as Pmcn. Then we go ahead and
calculate the volume of the unit cell. So, we have volume known in this expression, we
know the volume we know the Avogadro number, we know the molecular weight. The
two unknowns are the number of molecules in the unit cell and the density associated
with the crystal. the density associated with the crystal can be measured. In earlier days
as I mentioned in the previous discussion previous class, that we were asked to determine
the density whenever we do any crystal structure determination it was a compulsory
requirement.
But now people do not bother about it and they simply get whatever density the software
gives, they assume that is the density of the crystal. There is always a possibility of going
wrong in these in this assumption. Because the software gives the density of the crystal
based on the value of M and suppose let us say in this example n was assumed to be 8
then we will get a wrong density. Because we know that n is equal to 4 by eventual
determination of the density.
So, how one determines the density of a crystal? The density of a crystal is determined
based on an experiment which is known as the flotation experiment. It is basically the
principle of Archimedes. You know the principle of Archimedes that when an object is
352
immersed inside a solution the weight of the object equivalent will be taken out from the
liquid, so it will be displaced from the liquid, so the volume of the liquid is equivalent to
the weight of the sample which is inside. So, if a person gets into the bathtub as
Archimedes did certain amount of water flew from the bathtub and that is how the
Archimedes principle came up as you all know.
So, similar thing we assume here. So, what we do is, we take, we roughly take two we
take two liquids, one with a lower density and another with a higher density. And these
two liquids these two solvents for example, these two solvents could have a property that
this compound calcium carbonate will not dissolve in them. So, for example, in this case
we cannot take water. It will dissolve in water. So, we have to take two solvents, let us
say we take alcohol on one hand and then some higher density liquid which could be
methylene bromide or methylene iodide or something like that, these solvents are chosen
in such a way that calcium carbonate will not dissolve in them. We already know the
densities of these two liquids from literature.
So, we take the value of the literature for these two liquids and then we drop the crystal
one single crystal of calcium carbonate is taken and is dropped inside the liquid, in one
of the solvents. Let us say we take the solvent which has lower density and then we drop
the crystal inside the solution which has lower density than the anticipated density of the
crystal. Then what will happen? If you have a lower density and you drop the higher
density crystal on top of a lower dense liquid what will happen, will it sink or will it
float? What will happen? The density of the liquid is lower. So, will it sink or will it float
it will sink.
So, if the density is higher it will float. So, we take the mixture of these two nicely mix
them up and then put the crystal and adjust the ratios of these solvents, such that there
will be one stage we can visibly see the crystal in between the two. That means, if going
at their middle. So, we have these two solutions mixed. So, this crystal will float in the
middle. This is known as the flotation technique for determining the density. Then we
take that mixed liquid and then measure the density of that using a densitometer. We can
use either use a densitometer or use a specific gravity bottle and then use the density
determination methodology which is very well known from high school.
353
So, we determine the density of the crystal. So that means, we have the knowledge of N,
we know what N is, what ρ is and what V is. ρ is expressed in grams per cc or kilograms
per meter cube. then M is the molecular weight of the material and so we know now V
rho N and M, we should be able to determine N. And this is the way in which it has been
identified that there are four of calcium carbonate units inside this Pmcn units . So, to
repeat the whole thing once very quickly, we have the material calcium carbonate the
aragonite form that has been crystallized. The crystals have been subjected to x-ray
diffraction and the unit cell dimensions have been determined by some technique which
we are going to learn in the future.
So, we determine the value of a, b, c; α, β, γ . The value of a, b, c; α, β, γ tells us that it

is an orthorhombic system. We go further and determine the symmetry that is associated
with this orthorhombic crystal and we find it is Pmcn. The moment we know it is Pmcn,
we know that this is orthorhombic number one and number two there are 8 general
positions and 4 special positions have indicated here. There are 8 general positions 4
special positions corresponding to a mirror symmetry, a ͞1 symmetry and a ͞1 symmetry.
Then based on the knowledge of the density which has been measured using this
methodology we just described. We know the volume, we know the density and we
know the Avogadro number. The right hand side is evaluated and is divided by the value
of M which is the molecular weight, and that value will give us the value of n and it turns
out to be the value of n is 4 that means, we have four calcium carbonate units in this unit
cell.
So, that knowledge is gained. Now we will see when we actually determine the structure.
What kind of positions calcium took, what kind of position carbon took and what are the
positions oxygen took. As we already discussed there is there are only 4 calcium atoms
so obviously, these 4 calcium atoms should sit in one of these special positions. Because
8 positions cannot be generated for 4 calciums, there are only 4 calciums. well it can be
generated with half occupancies and so on. But that is unlikely in a very nicely
determined structure so, in an ordered calcium carbonate structure like that found in the
aragonite mineral. We will have four calcium atoms. So, they have to sit in one of these
special positions. And the carbon atoms also should sit in one of these allowed special
positions.
354
Oxygen atom as we discussed again, that 4 times 3 is 12. So, it can sit 8 of them in
general position and 4 in one of these special positions. Or all three oxygens can be
distributed in these three special positions are summing up to 12. So, 4, 4, 4 that will
make it one so these are the possibilities. Now this is decided by the nature of the
structure. The coordination around calcium the way in which the carbonate units join
together and so on. The bonding comes into the picture here, the nature of the bonding
the nature of the coordination site around calcium will now tell us how these calcium
carbon and oxygen go into this particular space group. Occupying the special positions
are the general positions as we have discussed.
So, here is a situation where we have now landed ourselves into that whenever some
crystallization takes place of a material or a compound. The compound can go and
behave in such a way that the individual elements now constituting the compound can
follow the required multiplicity and the special point symmetry and so on and therefore,
their coordinates are kind of predetermined when once they go into the space group.
So, the symmetry therefore, is the key factor here, it drives the elements in this particular
compound to go into their respective positions. So, we have seen that this is a Pmcn
space group. So, there are 8 equivalent points, but when they are in special positions they
it becomes 4 and therefore we have the possibility that 4 calciums and 4 carbons will sit
only in special positions. Whereas, oxygen has the has the freedom to go either there or
sit only here. And let us now look at the solid crystal structure. Because we now have an
information that n is definitely 4 from this expression.
Now, so, this is a universal expression, this can be used for any crystalline material. Then
once we determine the density, once we determine the unit cell dimensions and volume
and if we know the molecular weight we can calculate the number of number possible in
the unit cell. Suppose we are looking at a structure whose molecular weight is not
known, and it goes into a space group where n is fixed. The number of molecules in the
unit cell is fixed, this which probably a very good method to determine molecular weight
very accurately.
Nobody uses this method, but one can determine the molecular weight of a given
compound by using this formula and this determination of the molecular weight will be
extremely accurate because we have the a, b, c, α, β, γ determined at Angstrom level. So,
355
the accuracy associated with this molecular weight determination will be very very
reliable. But this method is quite expensive to do because we have to number one grow
the crystals, number two determine the volume by x-ray diffraction experiments.
Determine the density by flotation and then calculate the M value so, it is going to be a
tedious one instead one can use the routine methodology that is available for molecular
rate determination.
But if the compound is unknown, this is the question which we will come again later as
we go to the structure determination part if the compound is unknown we do not know
whether it is crystal which I have in my hand is calcium carbonate or not is there a way I
can find out that it is calcium carbonate. And that is your approach eventually we want to
get to the structure determination protocols. To determine where the atoms are and what
those atoms are, and in what way they decide to sit inside the unit cell.
So, in principle we have come to the level of first level understanding of how this
compound could be. So suppose you are given an unknown compound and you are has to
crystallize you crystallize the compound, but you do not know what are the atoms inside
that let us say. You only go to the x-ray diffraction machine and then characterize the
given unit cell dimensions. And let us say the scattering is very good the crystal quality
is very good you get a volume.
So, you use this volume information, you know the Avogadro number and since there is
a good crystal that is formed you can also determine the density. So, the right hand side
is available to you, for any compound and the left hand side will the right hand side will
also tell you what are the various possible general positions with which we can associate
this compound to two and therefore, we know roughly what should be the n value or the
z value if all the atoms in this structure are sitting in general positions. of course, if they
are sitting in special positions also we know the distribution that is possible.
So, in principle we should be able to determine the molecular weight and in the long run
in principle we should be able to identify which item is which by looking at the
scattering characteristics the details of the intensity analysis and so on. So, the take home
from this discussion long discussion rather, is to say that even if you do not know
anything about the structure. If for example, the chemists told you that this is the possible
formula and he may be totally wrong. It does not matter to the x ray structure specialist.
356
So, much that you know once in a in a Jack Dunitz made a comment who is one of the
top crystallographers of the world. He made a comment that you know a chemist keeps
his compound in a dark room he gropes in the dark he makes the compound and gropes
in the dark.
So, it could be any of these structures. So, uses various spectroscopic methods some
other methods to sort of apriori predict what could be this structure. But this is like
entering a dark room and trying to let us say find this monitor inside the dark room,
suppose everything is dark here, and we enter the room and we want to find where this is
it is going to be difficult.
So, what crystallography does? What crystal structure determination protocol does? In
fact, in general what crystallographers do is to come and switch on the light. So, the
moment we switch on the light we know where the object is, where the crystal is, where
the atoms are, where the atoms are going into the various parts of the unit cell and so on.
So, a chemist groups in the dark and the crystallographer gets in and puts on the light in
that particular room to find the object.
So, thereby the determination of the crystal structure by x-ray diffraction is

unambiguous. One once you do a thorough job of analyzing the x-ray structure then,
whatever you have determined is going to be the final structure in the solid state. So,
when once you dissolve the crystal in or allow the crystal to become a liquid and liquid
becomes gas and so on. The it is not guaranteed that the structure which is in the solid
state will be maintained.
Because as you all know the structure in the solid state is such that the molecules are
connected to each other by various intermolecular interactions. And these intermolecular
interactions could be strong they can be weak depending upon the situation. The
molecules arrange themselves to go into a given space group, as we discussed the other
day the possibility of polymorphism. They can go into different space groups.
When they go into different space groups different types of intermolecular interactions
will hold the molecule together. And as a result one once we disturb these intermolecular
interactions we do not know what kind of reorientation occur in the molecule. To a very
large extent most of these structures remain unaltered which is, because of the fact that
357
they are all bound by very strong covalent ionic that kind of bonds, which are strong
bonds.
So, energetically it is very difficult to disturb the molecule just by removing the
intermolecular interactions. But then there are possibilities particularly in organic
systems and so on, where there is a potential for intra molecular interactions to develop
in you of the intermolecular interactions. Then you have a possibility that confirmations
may change.
So, the conformation you may get from NMR may not agree with the conformation you
got from the crystal structure. This is particularly true with small molecules. The story
with large molecules particularly proteins and viruses and so on, is very interesting
because these molecules they do not crystallize with the intention of having only the
molecular species. They always crystallize with a large amount of what we call as
solvent water. And there is a lot of water around it. So, more or less the crystal structure
if we determine the crystal structure of a protein it mimics the actual structure of the
protein in the in the biological media and that is a very big advantage as far as protein
structure determination is concerned.
The protein structure also has a issue that it depends upon the various states of salvation.
So, you can have polymorphism in proteins as well, depending upon the nature of the
salvation dynamics which occurs around the protein molecule and the way in which the
solvent structure develops around the protein molecule. Now why we why are we
discussing all these issues? We are discussing all these issues to bring home the
importance of understanding where the atoms are, and how the atoms now are distributed
So, if we consider therefore, the case of this calcium carbonate it crystallizes in a Pmcn
space group the options that are given to the atoms have already been discussed. So,
now, we have to see how the atoms are located. The moment we know how the atoms are
located we will know about the geometry of that particular molecule as well.
358
So, all these issues will come up if we go to the next slide. And that particular slide will
tell you how the atoms in calcium carbonate are actually located inside the unit cell. And
we will discuss that in a little more detail. But at this particular moment we see that there
are 4 calcium carbonate atoms in the unit cell based on the density. And the atomic
coordinates are given in this table, the atomic coordinates tell us that calcium there are 4
calcium atoms the calcium atoms decide to sit in the mirror symmetry position, which is
generating the special 4 positions. z equals 4. z equals 4 generation position.
So, we will have therefore 4 calcium atoms. Carbon also decides to use the mirror
symmetry it could have used mirror as well as ͞1 . But it uses the mirror symmetry 4 of
the oxygens use the mirror symmetry. So, this now determines the geometry of the
calcium surroundings. And then we have of course, the oxygen to the second one
generally occupies all general positions. And at this stage we will discuss independently
how we go about understanding this ok.
359
Lecture – 28
Crystal Structure of Some Minerals
Let us see now the structure of aragonite in the space group P m c n. There are 4 calcium
carbonate units in the unit cell. So, we now based on the density it has been determined
there are 4 calciums, 4 carbons and 12 oxygen atoms.
When the actual structure is determined these are the cell dimensions a is 4.96 Å b is
7.97 Å; so, this is now the a direction this is the b direction. Remember that there is a
mirror symmetry at ¼ along this direction and then there is a c glide symmetry which is
associated in this direction please note that.
So, both those are the mirror symmetry positions. So, we have ab projection we have the
c value as 5.73. So, a . (b * c) will give you the volume of the unit cell from which we
can determine the density as we discussed before. So, you see that the calcium atoms sit
down the 4 m position; corresponding to the x value of ± 0.25 and the value of this z is
0.75, but what is more important is that they sit at on the mirror which is at ± 0.25. This
360
is the mirror which is now perpendicular to the a axis and so this sits now along the
mirror plane.
So, you see that there is one at 0.25 and there is one at 0.75 and these are the calcium
positions. So, the additional calcium positions will appear automatically in the space
group P m c n at these 2 positions; so, we have therefore, 1, 2, 3, and 4.
Let us go back and check the mirror symmetry this comes at ±(¼ y z) and we see ± (¼
½ - y ½ + z) notice that this is ½ - y ½ + z at ¼ again.
So, the positions of calcium are at 0.25, 0.75, 0.25, 0.75 this is now coming at the value -
0.25. So, this one is + 0.25 this is - 0.25 will be also equal to 0.75 by translation. So, for
example, if you take this atom onto the - 0.25; it is up there you translate it by one unit
translated by this unit you will get back to the 0.25.
So, therefore, we have 1, 2, 3, 4 calcium positions; the carbon positions now sit at ± 0.25
the same mirror. So, we know that there is a mirror perpendicular to the a direction, there
is a c glide which is perpendicular to the b direction and n glide is coming up with
respect to the c direction. So, we therefore, see here that the carbon atom also occupies
the mirror symmetry 4 equivalent points; that is ± (0.25 0.75 0.08); that means, that the
carbon atoms which are shown here appear at 0.08 and 0.58 respectively because this ½ -
z or ½ + z will account for the value of 0.58.
361
So, I want you to check the what you should do it as a homework is to see take these
coordinates and see whether they satisfy the symmetry requirements associated with this
4 m position. Then you will convince yourself that calcium carbon are sitting in these
special positions. It so happens that the oxygen atom will also sit with respect to ± 0.25
as is shown here in this diagram. So, therefore, these occur at 0.08 and 0.08; you see that
the oxygen positions are indicated in a bigger open circle which is shown up here.
So, the carbon atoms came here. So, happens the 2 carbon atoms sit on top of each other
this is at -0.08 and 0.58 and the carbon atoms here therefore, are ½ added to the z value.
So, they sit at 0.42 and 0.92. And these two values therefore, this is 0.92 is 1 minus 0.08;
and this value is 1 minus 0.58.
So, you see carbon atoms come up these positions. So, the carbon is one above the other
along the z direction; similarly the carbon is one above the other in the z direction and the
values of calcium are given here they give rise to these 4 positions. So, 4 calcium
positions are taken care of 4 carbon positions are taken care of, 4 oxygen positions are
represented by 0.08 and 0.48, but the positions of x and y are different x and y are
different; that means, that we have now a coordination around the calcium atoms located
in such a way that we do have the connectivity between carbon and oxygen and then that
connectivity now defines the calcium carbonate structure.
So, you therefore, have 8 general positions which are occupied by the second oxygen set
and the coordinates are given here and they are represented in this projection. Now this is
a projection diagram; now over the last few classes you have realized the fact that even
though we represent everything in 2 dimensions eventually the structure is 3
dimensional.
So, here is a challenge for you; it will not be given as an assignment or an examination,
but I want you to generate the 3 dimensional structure of calcium carbonate by yourself
as a home assignment may be; to find out what is the coordination around calcium. How
the coordination develops around calcium and what are the connectivity’s in the structure
of calcium carbonate by based on the diagram which is given here. If you could work
that out then you have become experts in analyzing structures; even before you go and
determine the structures using X ray diffraction or any other techniques neutron, electron
and so on.
362
So, effectively what is represented here is a realistic structure; the realistic structure
going into the orthorhombic system P m c n with 4 of those calcium carbonate units in
the unit cell and the way in which the atoms are distributed in the unit cell. So, the
calcium carbon and one set of oxygen’s all occupy the 4 m position and the 8 i position
which is the general position is occupied by 8 oxygen atoms. So, totally we have 12
oxygen atoms 4 and 4 calcium and carbon and this determines the entire structure in the
unit cell.
So, what are the assumptions we have made here? We have made the following
assumptions; we have assumed that we know the structure; we have assumed that we
have done at X ray diffraction experiments which we will do later and we have assumed
that we know the unit cell dimensions and then the density; so that we know that there
are 4 molecules in the calcium carbonate structure. So, these are the things which one has
to do.
So, this is of course, as they say it is better to think before you do the experiment we are
now thinking before doing the experiment. And this is in terms of where the atoms are
coming and where they are sitting with respect to the known calcium carbonate structure.
So, if we know the structure yes we can definitely think about it and determine the 3
dimensional structure right. So, this now takes us to the possibility of looking at many
structures which are already known. This is not our ambition anyway we want to find out
methodologies to determine the structures in the long run.
363
But we will look at one more which is the very simple structure of copper oxide. Copper
oxide goes into the is another mineral called tenorite and this is the structure of copper
oxide.
This tenorite structure crystallizes in this space group, I want you to find out what is this
space group. By looking at the equivalent points and also looking at the ab projection
which is shown here; what do you think is the system. You have for example, lot of ½ ½
½ ½ indicated here and you also see that this angle is β. So, the structure has to be a≠b≠c
and β is 90⁰. What is the system?
It is a monoclinic system and in this monoclinic system we have lot of ½ ½ ½ coming up

and therefore, it has to be a C centered lattice. We see here in this diagram if we now
carefully look at this particular diagram we get a c glide which is associated with the
origin O here. So, and then we also get a twofold associated with that origin. So, there is
a c glide and a twofold intersecting with respect to each other with a 2/m symmetry. So,
it is a centro symmetric monoclinic system apart from that we also have the fact that
these 2 and c are intersecting at this origin.
So, it has to be this space group C 2/c. So, we have inferred that the copper oxide tenorite
goes into the space group C 2/c; C 2/c boasts itself of 8 equivalent positions right.
364
So, we will list those 8 equivalent positions; those equivalent positions are generated by
± x y z; ± x̅ y ½ - z I think 4 there are two more sets which was missing in this in this
table; we will have to add that up ± x y z, ± x̅ y ½ - z; what is missing is the addition of ½
½ 0 to these 4.
So, it is not purposely missed it is not shown here the; what is shown here is with respect
to 0 0 0. So, there are 4 equivalent points corresponding to 0 0 0; 4 other equivalent
points corresponding to the addition of ½ ½ ½. It has several special positions; now the
several special positions sorry the several special positions can be of symmetry ī or of
symmetry 2; 2 possibilities again exist. Just like we saw the P m c n case where we had ī
and the mirror symmetry in the previous case; we had the position of ī available, but all
the atoms decided to occupy the mirror symmetry.
So, here we have four ī positions that are available and these are the corresponding
coordinates; you will have to generate the additional coordinates. Here only two are
given; that means, you have to add ½ ½ 0 to each one of these to get to the 4 equivalent
points. So, we know where the equivalent points are, we have very clearly and quietly
determined that this space group is C 2/c without having any information earlier.
365
So, this is C 2/c and then there are these distribution of equivalent points. Now what we
will have to see is how many at how many units of CuO are in the unit cell. So, we have
to go ahead and determine the unit cell dimensions; we will determine the a b c and we
know a β is non 90ᵒ; so, we have to determine the β value also.
So, one once we determine a b c β we have the unit cell dimensions. So, we can calculate
the volume of the unit cell; we can measure the density of CuO just like we did the
density of calcium carbonate in the previous example; so we know the density. Avogadro
number is known and then of course, with copper oxide given to you we know the
molecular weight. And it turns out that if we use the formula Z * M = N ρ V; we will end
up with 4 copper oxide units in the unit cell.
So, therefore, we can if you want to call it a molecule it is not really a molecule; it is a
compound. So, 4 CuO units in the unit cell; so the this means to say that the choices are
again clear and very very clear because nothing is in the general position in this structure
because anything in the general position will generate 8 equivalent points. So, we already
know that this is not occupied at all. So, general positions we can forget about; even
though we have the option of having the general positions, this particular compound
decides to sit in only special positions.
Now; that means, to say that this will now have the choice of 5 different possibilities; it
can sit the copper atom can sit in this position this position this and that and so also
oxygen. There is only 1 copper and 1 oxygen is a much simpler example than calcium
carbonate. So, both the atoms have no choice, but to sit in special positions in the unit
cell. So, the atoms now, therefore, have to sit in special positions; so we now go and see
the structure.
366
If we go and see the structure we see that the copper atoms are found to lie at centers of
symmetry; which is ¼ ¼ and 0; that means, it corresponds to ¼ ¼ 0.
So, this are highlighted it here; the highlighted one is the place where the copper atoms
decide to occupy. The oxygen atoms likewise decide to occupy the twofold symmetry;
the twofold symmetry which also generates 4 equivalent points. So, the oxygen atoms
decide to go and sit there the copper atoms decide to sit here. And so you get this
structure which is written here 4 copper atom CuO in the unit cell; the oxygen atoms lie
at 0 y ¼; the value of y being 0.416; remember this is a C 2/c space group; so, the origin
is fixed.
So, now the choice of fixing the y is not independent; you see you have a twofold axis
alone then the y axis is arbitrary where we fix the y value is arbitrary. But when once we
have a centered system like 2/m point group symmetry here the origin is fixed at 0 0 0.
So, the value of y therefore, is coming at 0.416 when you determine the structure. So, the
crystal structure has been determined a is 4.65, b is 3.41; c is 5.11 and β is 99.5.
So, this gives rise to our unit cell and therefore, we can calculate the volume the volume
of the unit cell in turn will allow us to calculate the number of molecules; in the unit cell
there are 4 CuO units in the unit cell. The positions of the atoms are marked in this
projection again the projection down a and b. So, c axis is coming perpendicular; so
therefore, this angle is 90ᵒ.
367
The c angle is coming in a direction which forms the angle β 99.5. So, you see therefore,
the copper atoms sit at this and that position because they want to occupy the ¼ along a
and ¼ along b and 0 along z. So, one once they come there the C 2/c operation the will
ensure this presence of the center of the C centering will ensure that if we get an object at
½.
And again that ½ is repeated here through the center of symmetry which is present at this
½ position ½ along x ½ along y and therefore, that will generate this ½ and this one; so,
the positions of the copper atoms are fixed. The oxygen atoms on the other hand occupy
the value of y equals 0.416. So, if the copper atom is at ¼ ¼ 0 that is 0.25,0.25 and 0.
And then this oxygen atom comes at 0.25, 0 y and ¼; that means, the value of z is 0.25
and the value of y is 0.4. So, we can have the coordinates of this atom at ¼ ¼ 0;
coordinates of this atom are at 0, 0.416 and ¼ that is 0.25. Since we know the cell
dimensions these are the fraction coordinates which I am talking about. So, ¼ ¼ 0 is the
position of copper; then the 0, 0.416 and ¼ are the positions of the oxygen atom.
So, we can calculate the length bond length between copper and oxygen; I want you to
do this again. Remember we are in a monoclinic system; so you cannot just directly
measure the distance between these two and say it is the bond length; you have to
calculate the bond length the value of β comes into the picture and so there will be a sin
β component which will come in your calculation.
So, one can calculate therefore, the bond distance between them. So, the geometry of
copper oxide can easily be calculated; either this geometry with respect to that or that
geometry or that geometry. Again remember that they are all gone by the C 2/c
symmetry. So, the symmetry will now allow only these possible positions for the axis in
atom. So, in a nutshell we have seen two structures; two independent structures and we
have seen the space groups into which they can go by based upon the symmetry
diagrams and also the equivalent point diagrams.
And we have determined the crystal structure particulars particularly the crystal
structural details a b c, α β γ and from that we have inferred the possible space group. So,
we know the space group we know the positions of the atoms where the space groups
allow the atom positions. So, all the allowed atom positions are identified with respect to
the molecule which we want to put into that.
368
For example, in the case of the copper oxide we want we can put this into 8 equivalent
points if they are all in general positions and 4 equivalent points if they are in special
positions. What we then determine is the density of the crystal one once we determine
the density of the crystal; they we will have 4 copper oxide units indicated. And there
moment we have 4 copper oxides indicated we know that there are no general position
occupation in this particular structure.
So, this way we can determine the structure. So, what are all the things we know how to
determine the we need to know to determine the structure? We need to know what is the
molecule ,we need to know how the atoms are positioned with respect to the unit cell, we
need to know whether how many special positions get occupied and how many positions
general positions get occupied and so on. So, suppose you have a structure like this let
me draw it suppose you have a structure which is let us say at C26 H12 is a hypothetical
structure O4 N2 and let us say this structure goes into a space group P 21/c by a b c α β γ
we find out it goes into P 21/c; this is fine.
Now, we also know the positions of the special positions unit cell where the size of the
unit cell we know. So, the size of the unit cell we know. So, we know the volume; we
know the volume we can also measure the density. So, we know the density and since
somebody said this is the molecular formula then we can calculate the molecular weight;
so, M is known P 21/c.
So, I can calculate the number of molecules in other words the value of Z or the number
of molecules what is the value of n we can calculate.
369
So, having calculated the value of n; I would like to see whether I can determine the
structure see the complexity which is associated with the structure. So, far what we saw
was copper and oxygen right and we also found that by density calculations there are 4
molecules in the unit cell.
Suppose this also by density tells us that there are 4 molecules in the unit cell then there
are 4 general positions in P 21/c. So, we therefore, have to determine 26 positions of
carbon 16 positions of oxygen 8 positions of nitrogen and 12 times 4; 48 positions where
hydrogen can sit. Now is it practically possible to follow this methodology which we
have discussed so far to determine the structure of this? Obviously, not; so the structure
determination therefore, is not so very straight forward.
We have taken two examples where we already know the structure and so we are talking
about where they are sitting. In these cases we do not know where the atoms are going to
sit 4 general positions; 26, 12, 4 and 2 atoms have to be distributed we do not know the
structure of how these atoms are connected to each other and so on.
Therefore we have to do the structure determination and there we use the X rays X ray
diffraction as a technique to determine this and we will see how we do that in the coming
classes. Now we have seen the structure of calcium carbonate and copper oxide which
are two simple examples. Just to illustrate the reality of our the fact that the information
370
about the space group, the information about the possibility of general and special
positions is a must.
Even otherwise if we know the structure we can fix where the atoms are if we do not
know the structure also, we have to have this information before we go on determine
make an attempt to determine the structure. And that is where the whole thing boils
down to; the whole thing boils down to the fact that we need to therefore, have an idea of
what the space group into which the crystals is not they evolve it into which the
compound goes into.
This is a simple example suppose you take a protein there will be hundreds of atoms;
now these hundreds of atoms suppose the protein let us say is crystallizing in a space
group we will see what are all the possible space groups into which proteins can
crystallize. We will see that the number of space groups will not be 230, but it will
reduce we will see it in a few minutes, but the fact remains that we should know the
space group information.
So, when we do the X ray diffraction experiment we will find the unit cell dimensions
we will also have to find out a method by means of which we can find the space group
information. Remember the space group information is built in through the symmetry
and we have we are able to identify what are all the symmetries that are present. Now
what the presence of symmetry does to X ray diffraction is something which we have not
studied right. So, what the symmetry does to the X ray diffraction is the next step and
that we will study in the coming classes.
So, at this moment we now have a understanding of the nature of the space group. So,
now, we can list out the 230 space groups straight away instead of going into the
individual details we have. In fact, examined the triclinic system thoroughly it can go
into 1 and ī the point group symmetries and the space groups are P 1 and P ī. In fact, we
have systematically analyzed also the monoclinic system where the point of symmetries
are 2, m and 2/m.
So, using the monoclinic point group symmetry we can construct the possible space
groups you have taken examples of P 2, P 21 and C 2. And we have taken examples of P
m and C m we have shown also C c. So, we have therefore, in a monoclinic system 1 2 3
371
4 5 6 7 8 9 10 11 12 13; 13 possible space groups that can go into this system and we
also showed in this case that the other there are no other possibilities; other than these
possible space groups. So, 13+2; so 15 space groups have been generated; when once we
come to the case of orthorhombic symmetry we have 3 point group symmetries 2 2 2, m
m 2 and m m m.
Now, here you see that there are so many possibilities because we have different lattices
which can come in. You see in the case of the monoclinic system we had a primitive
lattice under C centered lattice no other possibilities. This is in fact, we have not shown
that, but maybe you can (Refer Time: 25:38) in a in a one of the one of the TA session
we can show that that the primitive the C centered lattice can also be I centered and
things like that. And if we there is a B centered lattice it will go itself into a primitive
lattice this can be shown in a TA example.
So, the number of space groups therefore, are limited in case of the monoclinic system.
Because the number of lattice symmetries are lattice types are P and C only whereas,
orthorhombic has all possible lattices. So, it has the primitive lattice generating so many
of these possible space groups we have discussed 21212 and 212121 in great detail. We can
also have a crystal system the with a space group C 2 2 21; C centered lattice, we can
have a face centered lattice and a body centered lattice and of course, I 212121 is also
possible.
If we take the point group symmetry m m 2 we took some examples and we studied in
fact, I think one of these P c a 21 is this is this space group we studied in detail, but you
see the number of point space groups to which it can belong to.
So, there is a plethora of possibilities with m m 2 symmetry and more possibilities with
m m m symmetry; m m m remember is 2/m 2/m 2/m. So, this is the possible number of
centro symmetric space groups in the orthorhombic symmetry. The number of centro
symmetric space group in the monoclinic symmetry is corresponding to this point group
2/m or these and the number that is possible in triclinic with centro symmetry is that.
So, the centro symmetric crystals therefore, can generate only these many space groups
though the other space groups which are shown here or do not have inversion centers.
So, this brings us the restriction on the way in which crystallization can occur. The
crystallization can occur either to generate centro symmetric systems or non-centro
372
symmetric systems and it is of importance to note that if everything is centro symmetric,
the properties as described by that material is not very exciting.
In fact, the center of symmetry is a consequence of a little more perfect arrangement than
in the case of a non centric symmetry the very fact that we say centro symmetric and
non-centro symmetric is in that context. So, if the space groups to which it goes into are
non-centro symmetric; the possibility of anisotropic properties which might develop in
different directions are many more. So, the non-centro symmetric crystals are more
exciting and more important than these centro symmetric structures; even though centro
symmetric structures are necessary to get an understanding of the isotropic nature of
given materials.
So, the if a material displaces particular property along some direction; it is generally the
case with non-centro symmetric space groups than centro symmetric space groups.
Now, what we will do is we will go and finish off this table; you see when we got a
tetragonal you get enormous number of space groups. I want you to find out which of
these are centro symmetric of course, now we are experts in these areas there is a
fourfold symmetry which is of course; obviously, non-centro symmetric 4 is non-centro
symmetric. I wanted to find out your 4/m is centro symmetric and is 4/m centro
symmetric; we have to see 4 is it centro symmetric or non-centro symmetric.
373
So, these are issues which come up; in fact, 4 could be centro symmetric because 4 has a
property where you have the 4 fold along with the repetition of the centro of symmetry
associated with it. It is just like 2 is equal to mirror what is 4? So, these are questions
which I leave unanswered so that you will make some mental calculations about which is
centro symmetric which is and non-centro symmetric and so on. And therefore, in this
case you have therefore, the point group symmetries are more and there is also these
possible space groups.
What is very important to notice here is that in case of the tetragonal system because a ≠
b the basal plane is square. And because it is square the centering of that particular
square alone is not possible and at the same time if the base is squared then the other 2
perpendicular directions are independent of each other. And therefore, the only
possibility of having the other than a primitive lattice for the tetragonal system is the
body center.
So, for a tetragonal system we will not get the face centered space groups or the A or B
or C centered space groups because of the fact that a = b, but that decides that we cannot
have this and this is the case which we should remember. On the other hand if we go to
the cubic system now the cubic system has several point group symmetries and cubic is
unique because we have a = b = c, α β γ are 90ᵒ.
The only clear cut symmetry that has to be and must be present in a cubic system is the
threefold symmetry and that should be along the diagonal. If you see the threefold
symmetry here 3 3 3 3 and 3 all these symmetries are along the 1 1 1 crystallographic
direction. So, by 1 1 we will soon understand and try to see how these crystallographic
directions crystallographic planes and things like that develop when we now try to
understand the internal symmetry associated with the unit cells which we have not
touched upon so far.
What we have done is the internal symmetry with respect to these point groups and the
space group distributions; the generation of equivalent points. We have taken 2 realistic
examples and see where the atoms are and so on. But how do we identify the position of
the atom with respect to the internal structure of the unit cell is something which we will
study in a near future.
374
So, what we therefore, see here is the tetragonal the cubic and then of course, trigonal
and hexagonal are given here. In case of trigonal and hexagonal again we should have
only the primitive lattice in both these cases because you have the angle 120ᵒ in that case
γ is 120ᵒ.
So, we will not have any other body centered and so on and at the same time we also
have this rhombohedral symmetry we have not discussed much of rhombohedral
symmetry very rarely we come across rhombohedral symmetry compounds; even though
they become they are becoming more and more interesting with the fact that some of the
perovskite related conductivity conducting materials and so, on have a rhombohedral
structure, but that is something which probably is a little advanced for us. So, this from
here on we will see how the presence or absence of center of symmetry brings in
different natures in the space groups.
375
Lecture – 29
Atoms in the Crystal: Positions and Relevant Properties
We have been looking at the total space groups in the form of a table.
And this particular table is something which is very useful for cross references back and
forth reference and so on. As we saw that it tells us about the lattice symmetry as well as
the symmetry that can be associated with the space group the point group symmetry. It
also tells us about the types of distribution of centric and non centric systems which can
come with the triclinic monoclinic orthorhombic and the other crystal systems.
So, we see that in the case of the orthorhombic system all possible lattices are there. So,
we generate the P lattice, we generate the C lattice, we generate the F, as well as the I.
376
Now, all these lattices except the C lattice gets generated in the case of a cube. So, we
will have the P F and I allowed in the case of a cube the only condition is that the
threefold symmetry which is a basic symmetry in a cubic system appears along the 1 1 1
direction.
In case of the tetragonal system there is only possibility of primitive as well as the body
centered. And we see that 4/m is the one in which we will have the center of symmetry
getting in. And also in the case of 4/m m m center of symmetries space groups get
coming; the rest of the symmetry elements I put a question in the previous discussion
that if 2 is mirror then it is non centrosymmetric.
So, if what is 4? So, 4 is also non centrosymmetric and that goes into generating space
groups P 4 and I 4. So, 4/m and the 4/m m m are the ones we generate the; so there are
large number of possible non centrosymmetric systems in a tetragonal system. And if
one looks at the literature and looks at all the magnetic materials; the magnetic materials
belong to higher systems particularly the inorganic based magnetic materials; inorganic
material based magnetic materials. And many of them are the distorted perovskites and
so on; they all go into the tetragonal system and they go into non centrosymmetric
tetragonal systems.
So, many of the ferroelectric systems go into tetragonal systems. So, if an organic
ferroelectric is crystallizing; you know tetragonal system it will be very excited because
377
such a compound will certainly have a very high hysteresis behavior and so on. So, what
I am trying to tell is that the property that comes eventually for the material is already a
part and parcel built in due to the symmetry.
So, the symmetry now decides what should be the property in a given particular direction
in the crystal. So, that way a crystal is not always isotropic; the property in a crystal
therefore, depends upon in which direction we measure or in which plane perpendicular
to which we measure and so on the property. It also depends upon how the atoms align
themselves or arrange themselves in these space groups as we discussed in case of
calcium carbonate and CuO.
So, therefore, we have reached a stage where we now consolidate the information which
we have on all their space groups before we go further.
So, for example, the point group to which the space group belongs is easily obtained
from the space group symbol by omitting the lattice symbol and by replacing the screw
axis and the glide planes, with their corresponding symmorphous symmetry elements.
What it means to say is for example, if you have a space group like P 65 2 2 you delete
the P, make 65 into 6 and you get the point group 6 2 2. So, that is how we can determine
given the space group we can determine the point group.
378
Given the point group we can predict what are all the possible space groups ok; the
number of space group that are possible for a threefold symmetry is 1 2 3 4, if there is
only 3 associated with it. In case of 6/m 2/m 2/m we will get so many 4 of them as
possible symmetry for space groups.
for instance the space group P 42/m m c, P 4/n n c; so, we have taken the case of the
tetragonal system. So, if you take P 42/m m c it will be actually 4/m 2/m 2/m which
corresponds to 4/m m m. So, it is very easy to find out therefore, given the space group
the corresponding point group symmetry.
379
The other points that have to be noted are the following; see suppose you have a have the
Bravais lattices with symmetry elements with no translational components; that means,
there is no 21 screw axis, no glide planes and so on.
So, if you look at count the number of such space groups we have the space groups now
corresponding to the total number of 73. So, out of the 230; 73 are called symmorphic
space groups; the examples are P 2 2 2, C m m 2 F 2 3 etc., Notice that the lattice
centering is immaterial, what is important is the non translational symmetry operation
with respect to the point group operation.
So, the point group operation should not have the point group should not be now
accompanied by translational components like the screw axis and the glide planes. But as
far as the lattice nature is concerned we can have all the possible lattices. So, if we count
all of them out of the 230; we get 73 of them and these are called symmorphic space
groups.
Apart from that in the 230 space groups there are some eleven of them which are quite
interesting space groups; these are called enantiomorphous its like the right hand and the
left hand pairs. For example, if you take P 31; the 31 axis in the case of 31 is the
translations are one third, two third and full to get the P 31 operation. For the P 32
operation it is two third and then it is one third in the next unit cell which can be brought
back to the previous unit cell.
380
So, the handedness if the first one is considered right handed the second one will be left
handed. So, that is why they are referred to as an enantiomorphous pairs. So, we have
these many space groups which are enantiomorphous pairs and what happens in these
case of enantiomorphous pairs is very interesting when we deal with optically active
molecules.
So, when optically active molecules crystallize the let us say there are 2 isomers the first
isomer which we will call as positive can crystallize in one of the enantiomorphous space
group; then the opposite number will crystallize in the other case. We still do not know
either to identify the plus with the clockwise rotation or minus with the anti clockwise
direction.
But that is something we will be able to determine by making use of a technique which is
very very specially associated with X ray diffraction and that technique is known as
anomalous dispersion. I do not know whether we will have time to cover that aspect in
this course; it is very advanced methodology, but maybe we will mention it in the
passing towards the later part of the class.
So, therefore, it is there possible by X ray diffraction therefore, to determine the absolute
configuration. So this is if the plus isomer is in one form and minus isomer in the other
form; in principle we are now looking at the absolute configuration of the given material.
Whether it is levorotatory or dextrorotatory or whether it is R or S molecule and we can
also point out this is R molecule and this is S molecule if we do anomalous dispersion.
And that is how it that technique becomes extremely useful; particularly in case of a
compounds which can crystallize both in R and S forms, the isolation of R and S by
crystallization methodology can be accounted for.
Biological molecules these are also very exciting because there are more and more
accumulation of very large molecules. People are now looking at protein protein
interactions people are looking at protein steroid interactions and so on. So, these
molecules are enantiomorphous; it does not mean that they will go only into these 11
enantiomorphous pairs; they will of course, go into only crystals crystal systems and
space groups which have no inversion centers or mirror planes; there are 65 of them.
381
So, out of it 230 when you do the structures of biological molecules particularly in a
naturally occurring native proteins and so on; there are only 65 space groups that are
allowed. So, these are points that one has to remember with respect to 230 space groups.
Now, we will go to an area which takes us to the definition of what is a crystal. You may
be wondering we have been discussing what is the crystal so far and now I am saying
again what is a crystal. The reason why we again go back and ask a question like this is
to see that we have 7 crystal systems; they are decided by the nature of a b and c and the
inter-axial angles.
We have also said that there are atoms inside this the point group symmetry and the
space group symmetry will tell us how many atoms are there inside this unit cell for
example,. So, there may be depending upon the restrictions we put on each of these
crystal systems; the number of molecules inside the unit cell will change. We can have
one type of molecule, we can have two types of molecules we can have n types of
molecules in a triclinic system still with z is equal to 1; that means, the entire assembly
of all these molecules need not have any symmetry; still the crystal may crystallize in a
triclinic system and therefore, it is a crystal.
So, if you want to determine the structure of it we have to determine the positions of
every one of those atoms. So, in a monoclinic system there is a minimum twofold
symmetry; we can have any number of molecules any number of atoms, any number of
382
different kinds of molecules different kinds of atoms. So, that is what is meant by a
crystal; the crystal now consist of the distribution of atoms and when we talk about
distribution of atoms we talk about actually the way in which electron density distributes
itself to various atom sides.
So, now comes the very idea of an atom; an atom we always identify with respect to a
sphere; that means, in the sphere we put certain amount of you know we have the
nucleus we have the electrons surrounding it and so on. But essentially we characterize it
by the number of electrons which it carries; the system carries. And that is done in terms
of what we call as atomic number in the from the chemistry point of view; we have a
periodic table and in the periodic table we have various elements carrying so many of
these atomic numbers.
So, H equals 1; hydrogen has 1 atomic number, oxygen has 8 and so on ; that means so
many such; so many electrons are associated with these atom. So, as the number of
atoms increase, the atoms become more and more dense. The volume of the atom is
something which is determined based upon the nature of the nucleus and the surrounding
electron density which goes around it.
If we look at an isolated atom, but in a crystal we do not have an isolated atom; we have
connected atoms and these atoms therefore, or bonded to each other. So; that means,
there is a certain amount of electron density which gives from the atom to the next atom;
the atomic electron density is shared.
There is a lot of theory which has been developed over the years which talks about these
bonding features. It defines various kinds of bonds and therefore, the density is shared
between these atoms, but for our practical purpose of looking at the structure
determination and also eventually as you will see when we do X ray diffraction; we are
going to assume that these atoms are like billiard balls; they cannot be deformed. So,
when we have a carbon we associate 6 electrons with a carbon, the 8 electrons with a
with a oxygen and so on and with hydrogen there is only 1 electron.
We also know that these electrons are bound to the nucleus; so we have a central nucleus
and the atoms are bound to that except for the case of hydrogen there is no core. We call
that as the core the electrons that are distributed are distributed in various levels and we
383
know by atomic theory; how these electrons get distributed into various shells. We have
the K shell, we have the L shell, we have the M shell, we have the N shell and so on.
So, these atoms therefore, go and sit inside this box in wherever they feel like depending
upon the nature of bonding and the symmetry that controls where they should sit. We
have so far seen the symmetry that controls where they should sit with respect to the
space groups we have to say the 230 ways in which the atoms can arrange themselves.
But it does not tell us what kind of atoms can arrange in what way. And since we have a
choice of very many atoms in the periodic table; more than a 100 and way in which they
can bond to each other is also different from each other.
For example, we have the ionic species which can form ionic bonds, we have covalent
bonds, we have metallic bonds, we have different kinds of bonds. So, all these can occur
inside the crystal and that that is the one which gives the property of that material and
therefore, we have to see how these electron density is distributed inside the inside the
box. So, effectively we have to take the box and dissect the box we have to cut the box
into different sizes and different shapes or whatever and then look at the individual
pieces.
So, effectively if you want to see what is the inside a box; suppose you get a present in
your Christmas is going on now. So, if suppose you get a Christmas present which is in a
box what you do you; open the box you open the box remove all the packing material
and eventually you find the object which you want to find. Now can we open a crystal
and see like that? Obviously, not and if can we see the atoms; we cannot see atoms
because the atoms are of the size which will not allow us to use this light with which we
see things; we can see only objects which are of certain dimensions. And its so atom
happens that atoms do not fall into these dimensions they are much much much much
much smaller.
Say for example, we can see an elephant very easily, but we may not see an ant moving
on the surface of the back of the elephant because that is very very small. So, if you can
use a lens and go to the near the elephants and then look at the back of the elephant and
focus the lens; so, that we will see the ant on the back of the elephant. So, in order to see
objects therefore, we need lens; now this lens is the one which makes small objects to
become large.
384
So, in laboratory for example, one uses microscopes; so, you can use a microscope to see
what is there inside the crystal. Now if you use a microscope to see what is there in the
crystal again the microscope uses light. Now the crystals are of the dimension of 1 Å and
so we were should be able to see what we want to see in the dimension of an Å; not in the
dimension of 5000 odd Å which the light is made up of.
As you all know that we have the wave particle duality we initially in the case of
Newton’s; Newtonian laws we study the light as though it is made up of particles. And
then the wave theory brings in the quantum mechanics where we study this in terms as
waves. So, we can talk about matter waves and things like that. So, it is not the intention
of bringing in quantum mechanical aspects here except to say that if we want to see the
atom we should use a probe. Because we cannot just cut the crystal and destroy it; we do
not want to have the crystal destroyed, but still determine the structure of the molecule
inside the crystal.
So, if you want to do that we need a probe; now whenever we want to probe something
we need a lens as I already mentioned and in order to use the lens we need the light. So,
if what is the light which corresponds to the size of an atom; if we look at the
electromagnetic spectrum and examine the electromagnetic spectrum carefully, the light
which can be of the 1 Å size corresponds to X rays.
So, X rays therefore, or the radiation which have the wavelength of nearly 1 Å; it could
be 10 Å something in the range of 1 to 100 Å or in the range of 0.5 to 100 Å. So, one can
use therefore, the X radiation in order to see what is there inside; so, we send the X
radiation into the crystal. Now when we send the X radiation into the crystal the X
radiation are coming from a rarer medium and entering a denser medium and then they
are emerging out again back into the rarer medium ok. So, there is a little bit of physics
which we have to learn here in order to understand the process and that process is the
fact that process is called scattering.
So, we send in the light it happens in light scattering; we will study the way in which
light scattering occurs and the way in which the scattering can be reconstructed into the
image of the object inside. So, these examples will tell us that we need a probe and
therefore, in this particular next set of lectures; we will look at X rays as the probes
which will look into the inside the crystal.
385
When they look inside the crystal; for example, let us say we had the power or the vision
of X rays suppose I have a X ray vision ok; then I can see inside the crystal. And if I see
inside the crystal I will see the individual atoms I will see how they are connected to the
various molecules. I will see the presence of symmetry and I will see how the 2 1 screw
axis is relating 1 molecule to the other if there is a 21 screw axis inside.
So, we have to therefore, also determine the symmetry information. So, we not only have
to determine what scattering is coming out and so on we also need to understand the
symmetry that is contained inside the crystal. So, to understand what is there contained
inside the crystal we need the diffraction experiment done. I am using the words like
scattering, diffraction very freely today, but all these will have to be defined in a proper
physics manner so that we know the physics behind the whole thing and that will form
the next few lectures as we come along.
So, what we do today is that we have the 7 crystal systems so, how do we look into the 7
crystal systems? One of the ways to look into the 7 crystal systems is to do the job of a
surgeon. So, if there is a surgical blade which can cut the crystal to atomic sizes; then we
can cut the crystal and see where the atoms are and then find out which atom is what for
example, oxygen, carbon, nitrogen, bromine. So, if you have a surgical knife which will
allow for us to cut at 1 Å level, then we can take the crystal and cut it.
386
So, what I will do is towards the end of this section; I will bring in the discussion on how
we can cut it; one of the easiest ways in which to understand that is to see the loaf of a
bread.
So, you have a bread loaf here and this bread loaf which you normally buy in a bakery;
you see that you have the unit cell here, you start from this as the origin, you have a a
direction; you have a b direction and you have the inside c direction.
So, it is a 3 dimensional box; it is in fact, if the bread you buy in the shops is
orthorhombic. Sometimes it is tetragonal also, but most often than not it is orthorhombic
because somehow orthorhombic is the one which has attracted the attention of many
bakers in earlier days I do not know why, but it is orthorhombic. It can have different
shapes as I have shown here; it can have a shape like this it does not matter which shape
the bread slice is.
So, we have therefore, a unit cell and now we say this is 0 0 0; then this will be one unit
along a direction. So, this will be the position of the coordinate 1 0 0; then we have 0 0 1
and 1 1 0 all 3; I mean the x, y and z; z directions. We can also now take the diagonal
along this direction and we will see now that z direction is 1 1 1.
So, in the next set of discussions we will discuss how the crystallographic planes come
up and how the crystallographic directions come up. So, what we see here in the case of
387
a loaf of bread is that the loaf of bread is 1 unit here; suppose I cut this at half the
position I know you can slice it whichever way you want. You can slice it along the 1 1 1
direction; then you will get a nice triangular piece. If you cut it along this direction you
will get pieces along the a direction, if you cut it like that you will get it along the b
direction and if you cut it the third direction you will get the c.
So, if you cut it along the a direction and take the central piece it will be half the distance
of a. So, the coordinate of this point is 0 0 0 the coordinate of this point is 1 0 0, the
coordinate of the half point is ½ 0 0. So, now, let us say we cut this piece and take it out
then that particular piece will represent a plane because it is a 3 dimensional object; we
take this plane out your bread slices which are like this here which is shown here in this
slide or individual planes. So, they are they are a 2 dimensional object.
So, this 2 dimensional plane at half the position we call it as 2 0 0 at one third position
we can call it as 3 0 0 and so on. So, these values of 1 0 0; 2 0 0; 3 0 0 now are given to
the planes. So, when we say that we are looking at a 2 0 0 plane; that means, we are
looking at a plane which is at half distance along a and that slice which comes out with b
and c values being 0 and we call it as 2 0 0; this is also called the Miller indices. We will
look at it in more detail when we discuss further on the way in which we can now cut the
unit cell.
So, when we say we cut the unit cell; in principle we are now trying to see take out the
individual planes and see what is there inside that individual plane. And if atoms and
molecules and so on are distributed inside this unit cell therefore, when we do this
cutting operation and look at that particular slice, we should be able to see what is there
in that slice. It could be a part of an atom, it could be an atom itself or it could be several
other parts of different atoms. So, how are we going to see it? That is something which
will now excite our curiosity and we will keep this curiosity in mind until the next class.
I want to show you one more thing this is just something which is very patriotic to me;
this is this is what do you think this is? You all think that this is the slice which has been
taken out and placed on the plane of the projection because you may see this is let us say
the so called 2 0 0 plane. In fact, if that it is possible that it is a 2 0 0 plane and wherever
the atoms and molecules are, you will see these openings and holes and things like that.
If you go to a Ayengar bakery for example, and by a special bread this special bread will
388
have all these additives; he will add cashew, he will add badam, he will add this raisins
and so, on; the way he makes the bread loaf is actually he makes the dough and when he
makes the dough he mixes all these ingredients.
Let us consider these badam then cashew and things like that as possible atoms different
kinds of atoms depending upon whether we are adding badam or cashew or some other
ingredient different kinds of things can be added. And now when we mix the dough he
does not know where they are and then he puts it into the bakery and bakes this into the
oven and bakes the bread. When the bread comes out we do not know where these pieces
have gone and when we cut this slice it may so happen we may cut a piece of cashew, we
may cut a piece of badam and things like that.
So, this if you consider as the 2 0 0 plane; the half point and spread it around like this
then you see that these holes are the places where let us say badam and cashew and other
things are there. So, parts of atoms; it is not the full atom parts of atoms can be
associated with this. Now can we just look at it like this and see whether we can
determine the structure a big question mark. It may not be possible because we do not
know how many electrons are still associated with these little pieces that only will tell us
the nature of the atom.
So, what is in fact, necessary therefore, is to study all the planes and all the contents of
the planes put them together integrate them and see what comes out. Now before we
conclude the, this particular session; I will also tell you that, this is not the picture of the
slice of bread; this is in fact, the first picture taken by our lunar probe sent by ISRO. So,
it so happens that I have some contacts with ISRO people and one of them sent me this
first picture which was taken when our probe was about to crash land on the moon.
So, this is at a certain reasonably large distance from the moon, but you see this is the
moon surface. Very interesting the moon surface and a piece of bread loaf bread slice;
they look alike. What is important is that these holes and the way in which it is jagged
shows the peaks and the valleys which are present on the surface of the moon. So, this is
in fact the picture surface of the moon taken from a closer distance from our lunar probe.
I think we will close this at this moment.
389
Lecture – 30
Crystallographic Directions and Planes
So, we have been now you have enough knowledge gained now on Symmetry we have
the basics of symmetry understood to the extent that we can now deal with crystals. So,
in the last discussion we brought in the issue of a loaf of bread.
So, a loaf of bread now can be considered as a crystal and if we now look at the size of
the unit cell. Let us say the loaf of bread is the size of a unit cell if we take this point or
that point as the origin; when once the origin is fixed then this is the a direction let us say
this is the b direction and the backside is the c direction.
So, it will say define the one these cell dimensions; from that we can calculate the
volume of the unit cell by using the formula a . (b * c) and that will give us the volume
of the unit cell. So, we also know that there are only 7 types of unit cells we can think of
and those unit cells will give the dimensions according to the definitions of a b c and the
interaxial angles between them.
390
Having noticed that we now also in fact, can identify the various possible planes through
this unit cell volume. So, the here is the volume of the unit cell and then we now slice the
unit cells into a number of pieces like we slice the piece of bread. And suppose I take the
midpoint of a as a and then cut it and take that slice out that very thin slice out that slice
will be equal to ½ the unit distance along a. And of course, if we consider this particular
point which I am showing now that will be 0 0 with respect to b and c.
So, this is referred to as the crystallographic direction we will in fact, describe these
crystallographic directions in detail in a couple of minutes. So, the direction through the
unit cell is referred to as the crystallographic directions. So, if you go along this direction
it is 1 along the a direction. So, this particular point will be 1 0 0; on this direction of a
will define 1 unit along a.
And if we now consider the 0 0 the intersection associated with b and c and if we take
that slice out that slice will be 1 0 0 and the slice which is at ½ point is 2 0 0 at one third
is 3 0 0 and so on. So, these numbers which we give 1 2 3 etc.., they refer to the so called
hkl values and this we will define in a few minutes as the Miller indices. And these hkl
values define they play any given plane in a crystal. So, if we have these hkl values given
to you; you have to identify them with respect to your plane.
So, how do we identify the crystallographic directions then; what are the identities we
can give for a direction in a crystal?
391
So, that can be seen from this illustration down here. So, we have because you see the
single crystals which we grow are anisotropic. So, it is necessary to specify the simple
way in which we can define directions or eventually planes in which we have to observe
the physical property.
Suppose let us say with the property of a given crystal it shows ferroelectricity or for that
matter it shows some specific physical property in a given direction. We have to specify
the direction in the crystal; the reason is that in the other directions the same property
may not be expected may not be present.
So, this is a property of the material; so, if we are looking for the property of the material
into which we have of which we have grown a single crystal then the single crystal will
have to be specified with respect to the direction or the plane about which these physical
properties are measured.
For example, it could be a certain value of the dielectric constant which is now measured
along a particular direction. So, when we have single crystals therefore, we identify the
directions put those directions in the device which will measure that property and we can
measure that property in that given direction. So, these therefore, the definition of a
crystallographic direction and plane become extremely important.
These points are or probably not very well explained in any textbook, but in these
discussions we will go specifically with respect to the understanding of the
crystallographic directions and planes; with the view that we are going to measure
physical properties later on.
For example, if you are measuring the elastic constants of a given crystal the elastic
constant direction has to be specified. So, we will have a just like anisotropy in the single
crystal we will also have anisotropy in the elastic properties; these as anisotropy can be
expressed in terms of a matrix and those matrix values can be calculated.
So, to make it very simple we will now take any 2 lattice points let us take this diagram
here any 2 lattice points and then continuation in that particular direction defines what is
known as a lattice row. So, for example, this is a lattice row that is a lattice row and that
is a lattice row. So, a row of lattice points is referred to as a lattice row and this could be
392
along a direction it could be along b direction it could be along c direction; it could be
along any direction inside the unit cell and that needs to be particularly specified.
So, as you see that in a lattice there are in infinite number of parallel planes the
illustration here shows a few of them; so, you can draw any number of them. So, if we
take two different lattice points just to make points clear if we take two different points
Q 3 1 2 for example, Q represents the vector in the direction of 3 1 2 and that represents a
given direction. So, if you take the Q as the value associated with the direction 3 1 2; this
is a measure the vector distance is a measure of where the 3 1 2 comes with respect to an
origin; obviously, we need a definition of an origin.
Suppose let us say I define this as the origin and take this as 1 2 3 that will be a 3
position and then I have a one in the direction perpendicular to that; so, I can take this
and that as 3 1. Now if I want to represent a direction which is perpendicular to that I can
take this direction. So, for example, if the point here can be represented in terms of the
origin here as 1 along this direction, 2 along that direction and 0 along the third direction.
So, these therefore, now defines a direction way from here to here one can define a Q
vector. A Q vector therefore, is a direction dependent vector and that can in case of 3 1 2
can be represented within this diagram.
Suppose we take something like Q 9 3 6 which is far away in the unit cell; now we would
like to find out whether 3 1 2 and 9 3 6 represent the same direction. In fact, they do
represent the same direction if it is a primitive lattice and that is because of the fact that 3
1 2 and 9 3 6 have a common factor between them. So, you can take the common factors
that is 3 times 3 is 9 1 times 3 is 3 and 2 times 3 is 6.
So, once again 3 1 2 and 9 3 6 they represent the same direction. So, when we say the in
bracket [3 1 2]; it comprises of all possible directions in that orientation of 3 1 2. 3 1 2 is
a specified direction and we have to therefore, represent all the lattice points in that
direction. So, if we now put this square bracket it represents all possible directions in 3 1
2; if we just say 3 1 2, that will refer to the with within the bracket as we have shown
here the then it represents the direction 3 1 2 and all the family members which are
factorizable under 3 1 2 belong to this direction.
The points are different, but the direction is the same; so, when we say 3 1 2 that
specifies the direction. So, the specification of the direction is by crystallographic
393
nomenclature done by the brackets; the square brackets [ ]. You might look into a
situation when it is a non primitive lattice. A non primitive lattice is one in which we do
we have centering for example, we have a C centered lattice, we have a F centered
lattice, we have I centered lattice and so on in which case the lattice points may occur at
0 0 0 and also at ½ ½ ½ which is a fraction.
So, if the lattice points are fractional like is shown here where direction Q which is 1/2
3/2 -1/3. And then there is a direction which is a Q which is 5/2, 15/2,-5/3; both of these
represent a single direction in crystallographic direction and that crystallographic
directions happens to be [3 9 -2]. How do we get [3 9 -2] with these two? What we have
to do is we have to take the LCM of the lower numbers 2, 3 and 3; we take the lowest
common multiple and then we represent them then we represent the direction.
So, we see here for example, one ½ ; one ½ times this is 6. So, 1 by it; so if we multiply
this by 3 ok, if we multiply this quantity by 3 then the least common multiple is 6; so,
you get what? You get ½ . So, 3 by 6 is ½ and 9 by 2 and in this particular case it is 9 by
6. So, it will be 3 by 2 and so on. So, the same thing happens to the logic here you have
to take the LCM of these three numbers and these three numbers and these three numbers
in the denominator or common and therefore, we the representation of the direction is [3
9 -2].
So, we get to represent both the direct lattice points in the primitive lattice as well as in
the non primitive lattices by this kind of direction identification. A better way to look at
it is to take several fractions and then see what direction it represents by taking the least
squares LCM of these numbers at the as the lower part of the fraction the denominator of
the fraction. And then calculating the corresponding planes which will represent the top
part; so, which is the numerator of that. So, effectively what we do is we once we take
the LCM we cancel out that number and whatever number remains in the numerator is
represented here.
So, if you do that operation on these two cases you will get [3 9 -2]. So, this brings us to
an idea of how we represent crystallographic directions in a unit cell. So, now, it is
possible therefore whether it is a primitive cell or a non primitive cell; it is possible to
identify directions in terms of 3 numbers and 3 integers rather. And these 3 integers
could be both positive as well as negative depending upon where you define your origin.
394
So, the origin definition becomes crucial and one once we have 0 0 0 defined then all the
directions in the crystallographic unit cell can be identified. So, there are numerable
number of such things infinite number of such possibilities and all these possibilities are
accounted for. And also many of the lines many of the lattice points will form which will
lie on the direction of this crystallographic direction which we have defined all those
lattice points will also belong to the same direction. So, there may be a large number of
lattice points in a given direction there will be less number of lattice points in a given
direction depending upon the value of the unit cell in those directions.
So, the unit cell decides what should be the total number of crystallographic directions
we can have in a unit cell. Of course, the unit cell will itself have the possibility of 7
different types of lattices. So, the lattices both primitive as well as the centered ones or
the non primitive lattices can be covered under the definition of crystallographic
directions.
So, having learnt what crystallographic directions are we will now go to the
crystallographic planes. In case of a crystallographic planes we need 3 lattice points the
diagram below is shown here it represents the 2 3 6 that is because I have shown two
along this direction 2 units and then you see the lattice points two are along these
direction and here it is 1 2 3 4 5 6 along the vector r along the vector p it is 2 and along
the vector q it is 1 2 and 3.
395
So, this now let say tells us that if we join these 3 points p 0 0, 0 0 r and 0 q 0 from the
origin which is at 0 0 0 this now represents the plane 2 3 6; 2 units along the x axis 3
units along the y axis and 6 units along the z axis and that represents the plane 2 3 6. If
we write 2 3 6 in brackets [2 3 6] it represents all planes which are parallel to this plane
and depending upon whether it is a non primitive lattice or a primitive lattice; we can
have a number of lattice planes.
And these are all parallel planes as we have seen drawn here this is one plane this is
another plane which is passing through this lattice then there is a plane passing through
that lattice point and so on. And eventually we have a lattice point which passes through
these 3 this will be actually the family 2 3 6; even though we now see it is 1; 1 2 3 along
these direction and 1 2 along that direction.
So, the lattice planes belonging to 2 3 6 can be represented in this fashion. All the lattices
lattice planes which come within the 2 3 6 direction will be less than the 2 3 6 value of
the unit cell. Now the we have to develop an equation to this plane; so, equation to any
plane is the value of x in that direction and the total length a.
So, suppose we have a unit cell a b c then the equation to the plane is along a we can
have a x’/a fraction which can represent our. For example, if we take this as p then the
fraction is x’/ pa and x is the direction of the unit cell and if the value of the unit cell is
a; we restrict the direction to stop at the point a because we it repeats again because of
the periodicity.
Because of the periodicity nature we have therefore, the unit cell a unit cell b unit cell c
representing the unit the entire unit cell unit directions. And then x’/p y’/q and z’/r
represents the fractional units along those 2 3 directions. And therefore, this will be the
equation to the plane and that we equate equal to 1. If we now introduce the fractional
coordinates and that is why time and again we have mentioned even in our earlier
discussions; that when we refer to a position x y z it represents the fractional coordinates
associated with the system. It is because now we define the fractional coordinate x as x’/
a y’/b and z’/c.
So, therefore, you can write the equation to the plane as x/p+ y/q+ z/r = 1. So, this is the
equation to the plane; so, any plane in the crystallographic unit cell can be defined by
this particular fashion x/p+ y/q+ z/r = 1. Now we multiply both sides by a quantity called
396
“m” which is an integer; this is because of the fact that we want to represent the entire set
of planes in all these intersections. And in order to do that we have to now take the
common least common multiple of p q r.
So, we take the least common multiple LCM of p q r and that represents all possible
values of m which will now be equating to this plane; that represents the family of
planes. This is a single plane which is 2 3 6 in these example we have given here and the
hx + ky + lz = m; where m is an integer represents all these planes that are possible inside
this representation of the 3 lattice intersection points which hold a plane.
So, p q r now, therefore, define the plane 2 3 6. So, we have seen therefore, the way in
which the crystallographic directions and the crystallographic planes can be
mathematically calculated. But we want to make it as simple as possible in our
discussions we have kept it at a very simple level.
And so we will look at it this way. So, suppose you take in you know we always want it
in 2 dimensions it is easy to understand in 2 dimensions.
So, the we now have to see that the third direction is coming towards you. So, it is a 3
dimensional unit cell and in this unit cell the lattice points; let us say this is the a
direction and this is the b direction these are the lattice points ok; these are the lattice
points this is a and that is b, c is coming out towards us.
397
Now if we want to represent the so called one unit along a and one unit along c; now
these lines which we have we have drawn now represent the entire family of 1 1. The
family that is associated with 1 along a, suppose this is the origin this is the one value of
a and that is the one value of b.
So, the entire family of 1 1 is illustrated by this set of parallel planes of course, we at this
particular way position consider the line here and the c is coming perpendicular to that
and that is taken as 0 in these example. So, if we take this 2 dimensional case we have
the 1 1 and I still call it a plane because we have the c direction perpendicular to that; so,
this is the 1 1 plane in 2 dimensions one along this one along that. Suppose we look at
this picture in this picture you see they even though there are points which are now
corresponding to lattice a and lattice b; you see that there is an additional point which
comes ½ way along a and ½ way along b and these now all are also parallel lines.
So, when you consider these parallel lines; you see that now we have to represent this as
the as far as b is concerned we have divided into ½ and ½ , a is still the same as before
the b direction which has been cut into ½ and therefore, this becomes 1 2. In that
discussion in the same way we can now define the; so, called 1 0 plane because if you
just take the x direction and this is the value of one and then y will be like this. So, the 1
0 plane will be running parallel to the y axis.
So, all the planes which are running parallel to the y axis will belong to 1 0 plane. So, y is
the direction y is the direction of the unit cell and the planes are now parallel to that in
the sense that this 1 0 plane; the family of 1 0 plane 1 0, 2 0, 3 0, 4 0, 5 0 whatever be the
number all of them will be parallel to y axis. In this case all of them will be 1 2 planes of
entire family of 1 2 plane; that means, 2 4 and 3 6 all these will be also belonging to this
family of planes. So, the family of planes therefore, are represented in brackets; let us see
how we can generalize it a little more.
398
We can generalize it a little more by looking at this diagram. Now this diagram tells us
that we have a lattice this is the 0 point and these are the lattice points along the a
direction and these are the lattice points along the b direction and these are the lattice
points which define our unit cell.
And so, this now if you consider this point this point this point this point these 4 now
define the values associated with in 1 units of lattices. And let us say this is the primitive
lattice; if we consider that as a primitive lattice then we have this family of planes which
is one along a and one along b; you see this is the origin ok. So, you take one along a and
one along b you get this plane.
Now, you can draw a parallel plane now this parallel plane cuts at 1 and 2 along this
direction 1 and 2 along that direction; the value of c is equal to 0 because we are now
taking the projection a b and so, this is 1 1 0. So, the family of planes is now represented
as 1 1 0. So, the when whenever we write a bracket like this it represents the family of
planes. If we do not write any bracket and just write 1 1 0 by crystallographic
nomenclature 1 1 0 refers only to that plane.
So, suppose there you are trying to say where the atom is sitting in which plane a given
atom is sitting; this atom may be sitting at 1 1 0 then you do not write the brackets.
Because if an atom is sitting at 1 1 0 it does not mean it will sit at all the parallel planes
understand.
399
So, if there is an atom sitting at let us say 2 2 0 and then let us say there is a whole 5 or 6
atom sitting in 2 2 0 ok; then it is not necessary that any atom need to sit at 1 1 0, but
these are all parallel planes. So, since they are all parallel planes the family of planes is
represented in bracket and when we look at the scattering in a few minutes from now;
you will see that the scattering is dominated by these planes.
So, let me now bring that discussion now itself, but we will make it more clear as we go
along. What we do is we have a unit cell and that particular unit cell is like this a loaf of
bread; now this particular unit cell can be cut into a large number of planes; we can have
defined crystallographic directions we can have crystallographic planes and so on. And
so when we now send a probe which could be a electromagnetic radiation, it will go
through this loaf of bread and it will pass through all these planes all these directions; all
these planes.
Now, suppose we now populate these planes with atoms; let us say as we discussed in
the last part of our previous discussion. Suppose this is a special bread and you have
additives in this you may have cashews added, you may have badam added and so on.
So, we do not know where these additives are sitting inside the loaf of bread it is not
visible; so, but when we slice this and take a take a plane out of this slice, let us say we
take 2 0 0 in that particular slice we may find a piece of badam and a piece of the
cashew.
So, a same logic can be applied to atoms inside the unit cell; the atoms and the molecules
are whatever which exists inside the unit cell are now distributed inside the unit cell
among all these planes. But what we do in a diffraction experiment which we will do
soon is to now shine some radiation electromagnetic radiation. The electromagnetic
radiation will go through the crystal it will go through the crystal it is like entering a rarer
to a denser medium and then it will emerge to a rarer medium. So, there will be
reflection, refraction all kinds of possibilities that can occur with waves; same thing will
happen, but then if these planes are now populated with let us say these additives like for
example, atoms of different sizes and different electron densities and so on.
Electromagnetic waves are made up of electric waves the electric field is associated with
an incoming electromagnetic wave. These electromagnetic wave even the ordinary light
is an electromagnetic wave as you all know; so it has an electric component. So, when it
400
looks at an atom let us say that particular atom will have its own characteristics;
particularly it has an electron density around the central nucleus.
And that electron density responds to the incoming scattered radiation incoming
radiation and then it scatters the radiation. It finds it is something like you know you are
moving in a crowded town centre and you are going in a straight line, but then a group of
10 people come in front of you will have to either go away from them or shoulder them
and go through.
So, that is exactly what happens to the incoming radiation it finds the atoms and
therefore, electron density in its path and when it finds that it has to scatter it has to move
in a different direction and that is exactly what we will happen. And therefore, we have
to now see what is the consequence of such experiments which we are going to perform
on the crystal.
So, to continue our discussion at the moment we have defined the crystallographic
directions, we have defined the crystallographic planes we have to completely now
finalize how to define the crystallographic planes. For example, here I had shown you
the 1 1 0 set of planes; here the this shows the 0 1 0 set of planes that is the all those
planes which are parallel to the x axis ok; all those planes which are parallel to the x axis
and of course, we are taking 2 dimensional plane here. So, it is c is automatically set to 0.
So, 0 1 0, 0 1 0 represents this and suppose I now want to represent 1 3 0 you see that the
1 3 0; that means,1 along the a direction and 3 units along the b direction. So, this has to
move up here and this is 1 2 3 and those are now the parallel set of 1 3 0. And similarly 2
-1 0 is shown in order to for you to understand the way in which the planes develop with
respect to the crystallographic unit cell a and b with an origin defined.
So, you have the family 1 0 0 sorry 1 1 0, 0 1 0, 1 3 0, 2 -1 0 and when we have put these
normal brackets ( ) around it represents all the planes from 1 equals 1 to n and 1 to n and
in this case one goes from 1 to n.
Here 2 goes from the value of 2 to 2 times the next value that is again 2. So, it will be 4 -
2 0, 6 -3 0 and so on; so those are the parallel. So, all parallel planes get represented as
families; if we just want to refer to one particular plane as we discussed suppose there is
one atom only at this particular plane; then we say that the atom is associated with a hkl
401
plane. That hkl now you see has no brackets around it, so we are now identifying that
particular plane hkl about which the atom position is located.
So, if atom is located here we say that the atom is located with respect to the plane 1 0 0;
it is always possible to locate atoms anywhere in the unit cell, it is not necessary that
they should be associated with the lattice planes. But when we do the scattering
experiment as I already defined the scattering experiment is happening with respect to
the planes. So, there may be an atom associated with the plane, there may not be an atom
associated with the plane, but the when light comes and falls on that light will get
scattered and that scattering is independent of whether it has an atom or not.
So, if there is no atom it will pass through the crystal and there is no scattering occurring.
If there is an atom as I said there will be some disturbance and therefore, scattering
occurs. But if there are atoms which are situated between these 2; suppose there is an
atom here this atom now contributes to all the parallel planes, it not only contributes to
the parallel planes it also contributes to the planes which are in other directions as well.
So, that is something which we will evaluate in a couple of classes from now; what is the
contribution of the atom to a given plane.
So, to cut the long story short here is a gist of what I would like to say in the last half an
hour’s discussion and this is very very crucial. What we now want to do is to examine
the inside of the unit cell we have now understood that there is symmetry, we have
understood all possible symmetries that can occur inside crystals, 7 crystal systems 32
point groups 230 space groups; we now thoroughly know where the equivalent points
come in a given space group what is the special position and so on.
Now, these are all the possibilities that can occur inside a crystal; we have understood
also the way in which the lattice develops itself in 3 dimensions to define the structure to
define the crystal. So, having got that basic information and having understood all rules
of symmetry; we now consider situations where atoms and molecules can get inside this
crystal, but the way the atoms and molecules are getting inside this crystal is like the
additives getting added inside the bread loaf.
But what we are going to measure in X ray diffraction techniques which we are going to
discuss in a few minutes from now is the fact that it is coming from the planes. So,
whatever scattering is occurring; the incoming radiation now will see these planes and
402
these planes now will scatter the incoming radiation. So, whether a particular plane
which we identify here is having an atom or not; the atoms which are present inside the
unit cell will contribute to that.
So, we have to see a what in what way they contribute to that and in what way we get to
the scattering information from that particular plane. Obviously, if we get all the
information from the scattering planes of this kind; in principle we should be able to
guess where these little pieces or in fact, the pieces of whatever additives we have added
in this particular example what are the atoms and where they are sitting inside the unit
cell.
So, we can identify the atomic positions we therefore, can identify the bonding
associated with the atom positions and hence the structure that is associated with the
atom. So, the logic therefore, is that when we do the crystallographic measurement; now
which we are going to do on a unit cell which about which all the rules and regulations
about which have been framed by us already. So, we now know the grammar of the
situation, we can now construct a novel.
The construction of the novel now goes the fact that we have now to determine what is
there inside the unit cell; what is there in the unit cell or the atoms and the molecules and
so on which consists of nuclei and the electron density around it so happens that when X
rays fall on a given atom we will see in a few minutes. When X rays fall on a given atom
it is the electrons that scattered; X rays normally do not approach the nuclei they do not
have the power enough to approach the nuclei. And at the same time the X rays which
are used are about 1 Å in diameter in sorry in wave length and the atom has a diameter in
the range of angstroms and therefore, the electron density around the atoms or in the
range of angstroms.
So, essentially the electron density it is the electrons which scatter X rays; we should
never forget it when we do our X ray analysis. Because whenever you probe through X
rays you will see the electron density; whenever you probe through just to information
sake whenever you probe through neutrons, you get the positions of the nuclei. And
whenever do you do electron diffraction again you will get the electronic information.
So, these are issues which we will discuss more and more in order to understand where
the atoms are sitting.
403
So, the take home from this slide is these are the definitions of the planes, different kinds
of planes, we have family of planes, individual planes are represented like this. And there
is one more representation which I must mention here is the representation of hkl in
flower brackets { }. When you see them in flower brackets it refers to all the h, k, l; k, l,
h; l, k, h and so on. So, this is entirely dependent upon how symmetry dominates the hkl
values.
And in other words how symmetry controls the orientation of the planes and this kind of
restrictions come in only higher symmetries crystal systems. Cubic systems have this that
for example, in a cubic system since a = b = c, α = β = γ are 90ᵒ, 100, 010 and 001; these
are all one and the same because we cannot distinguish between a b and c.
So, when we now represent the hkl associated with it; all these get represented not just
100, -100, 010, 0 -10, 001 and 00 -1. So, it is degenerate to that extent because a = b = c
and so, such planes are represented in flower brackets. So, when you write in flower
bracket and say 100; it represents all the 6, I mentioned in the case of a cubic system.
So, these are the nomenclatures. So, hkl alone and represents one plane; hkl in brackets
( ) represent now the family of planes and hkl in flower brackets { } represent all
combinations of this depending upon the symmetry conditions which we impose. Now
once again to generally remind you that if I have a square bracket and say [110]; it
represents a crystallographic direction. Suppose I have a square bracket and say [111]; it
is the diagonal of the particular unit cell whatever is the unit cell.
So, the crystallographic direction is therefore, different from the crystallographic plane
and that is that is what we learned in the last half an hour or so. And so this is something
which is very crucial for us to remember and when we lead read literature; in literature
people normally refer to these without any introduction to what they are writing about
and therefore, it is important to know these nomenclatures.
404
Lecture – 31
Interference of Waves
So, we come back to this slide again because this is our bread and butter in the entire
course. So, we come back to this slide again. So, this is the unit cell, and we are suppose
to find out the contents of the unit cell. So, how did all, how did it all began, and then
how this the very beginning of understanding how to determine the structure.
405
It all started in when the science, what is very interesting is science evolved at a very
rapid place pace during the latter half of the 19 th and early 20th century. If you look at the
way in which science accumulated knowledge the accumulation surprisingly was in the
last part of the 19th century and early part of the 20th century and after words it has been
slogging alone.
But there was a sudden part in several issues, and one of the major once was that the
waves and particles duality that was identified and explained. At the same time Rontgen
discovered X-rays and then X-rays were used to probe mater and particularly crystals to
establish its structure, and so crystallography emerged. Now, in 1912 Max von Laue, and
in 1913 Bragg and Bragg gave the methodologies by means of which we can use X-rays
to probe crystals. And therefore, establish the structure, what I mean by structure is now
the positions of atoms and inside the unit cell.
So, we will now start to develop the methodology and then understand how this logic
came into the existence because we are now having an object which is opaque like a
crystal and we want to see atoms inside them. Obviously, we cannot use ordinary light.
So, we use X-rays because we know X-rays are of the wavelength of 1 Å , and the size of
the atom is also about 1 Å. So, there must be a way in which we can find out where the
atoms are, I think that was the thought process these three gentlemen had and we will see
how it went.
406
How it has now eventually resulted in. This heralded a special event in the history of
science, by far the most accurate technique to determine molecular structures, the area of
X-ray crystallography. Von Laue won the Noble Prize in 1914, and the Bragg’s won a
Noble Prize in 1915. And since then over a million structures have been determined a
various sizes, shapes and content, and this has made enormous input to a many, many
areas in physics, chemistry, biology, and many overlapping areas particularly of late
materials science and nano sciences.
407
So, we therefore, have to now get an understanding of how all these can happen. Now,
we said that when electromagnetic waves come in the electromagnetic waves for
example X-rays are associated with a certain wave length. So, a wave is represented in
this fashion, it is an oscillating electric field. And this oscillating electric field can go and
fall on to the object.
Now, we also said that depending upon where the atoms are inside the unit cell; that
means, wherever there is electron density inside the unit cell, the very fact that they are
electrons that are associated with the atoms, and these atoms are spherical in shape,
where there is a nucleus which is positively charged and an electron which is negatively
charged. So, let us write something like an atom. I hope I will be able to do that. So, let
me say this is the electron density around a nucleus at the center. So, this is now
positively charged, and the electron density is negatively charged.
And so what happens is when a disturbance of this kind we will come in. When we sent
in a probe this is the electromagnetic radiation, and now we are thinking about X-rays
which are of 1 angstrom in wavelength. They come and fall on this atom wherever this
atom is sitting inside the unit cell. Imagine there is an atom in a crystal and that is sitting
somewhere in therefore refer to the position x y z. If there is an atom at x y z based on
our symmetry and space group determination the definition, the positions of the atoms
are already available.
So, if there is an x y z there is an x̅ y̅ z̅ in case of a triclinic P ͞1 system. So, there are two
atoms and so what happens is this atom now we will see the incoming X- radiation.
Now, what happens is that because of the fact that we have positive and negative here,
these incoming radiation which is like this we will now fall on this. And then generate a
gap between the positive and the negative one. So, in other words, it creates a dipole.
This therefore, develops a dipole.
So, effectively what happens is the distance between the nucleus and the electron, the
electrons are perturbed. So, electrons go away, but they cannot leave the nucleus and go
away. So, you develop a dipole moment. Now, electromagnetic theory says that an
oscillating dipole will generates its own wave. So, what happens therefore is that from
this particular atom, now another wave starts to come and that wave is shown here as a
wave which is moving in.
408
Now, this particular wave generation depends upon where the atom is inside the unit cell.
So, if the atom is somewhere from there the disturbance starts, so the disturbance with
associated with this oscillating dipole will start from that point. So, if there are several
hundred atoms inside the unit cell, it will start at 100 different positions. So, when such a
thing happens, the waves can be the starting point of the wave can be variable.
So, in this case depending upon where that atom is or the set of atoms are these waves
now will start. So, this is a let us say the first atom, the first atom wave will now start to
appear depending upon where it is sitting. And then this incoming wave is still there. So,
these two now undergo what is known as an interference. So, the new wave now
interferes with the existing wave and what you get is a resultant wave depending upon
where the positions of the atoms are with respect to the unit cell. These waves now can
be constructive as you see here or destructive as you see now.
So, we can have intermediate stages as well depending upon the distribution of the
electron density inside the unit cell. So, we have the unit cell; we have atoms in different
parts of the unit cell. When incoming radiation comes with a certain wavelength that
incoming radiation, now meets this electron density surrounding the atoms, and these
now therefore the creates local dipoles. So, a large number of local dipoles get created,
and each and every dipole now is capable of sending its own electromagnetic radiation
because that is what electromagnetic radiation theory says an oscillating dipole generates
an electromagnetic wave.
So, that electromagnetic wave therefore, is now coming from different parts of the unit
cell depending upon from where it comes we can have either the wavelets adding up or
the wavelets subtracting or the wavelets giving rise to some intermediate levels of waves.
So, at the end of the diffraction, at the end of the experiment, when it comes outside the
crystal, it could be in any of these forms it could be a highly crest to crest competition or
a crest to a trough competition, depending upon that the value of the phase, the value of
the intensity changes.
We are making an assumption here that whatever is the incoming radiation, the scattered
radiation will also have the same frequency. This is a very important point which we
must remember that in all of the X-ray diffraction and X-ray crystallographic theory, we
will assume that there is no change in the frequency. So, the incoming radiation and the
409
radiation with scatters from atoms inside the unit cell, they will have the same frequency.
And this is called coherent radiation or coherent scattering. Obviously, we should be
knowing that other kinds of scattering can also occur which changes the frequency, and
we are going to ignore that in our discursion.
In fact, X-ray diffraction experiments ignore this change in the frequency, the change in
the frequency leads to different kinds of effects. One of them for example is the so called
Compton effect. There is a change in momentum and so on in this particular situation
where we discuss X-ray diffraction we consider only coherent radiation.
Now, the coherent radiation will be ensuring whether we will have these waves coming
in and out depending upon different kinds of positions of the atoms. And therefore, in
principle, we will get now these waves coming out of the system. So, when these waves
come out of the system, we should have a measure. So, we will keep a detector at the
position across this scattering experiment region, and find out what is this intensity or
what is the or what is the intensity of these wave, what is the strength of these waves.
So, any wave in this in general is characterized by two quantities; one is the amplitude of
the wave they extent to which it goes up and down that is the amplitude of the wave with
respect to 0. So, in this case, as you see depending upon where the atoms are, the
amplitude is going from the highest value to the lowest value, so that is the amplitude.
Now, the amplitude variation depends upon how the wave up there is interacting with the
wave down here and therefore, these depends upon where the origin of the second wave
is. If the second wave is starting exactly at this point, you see the maximum let me trace
it, when it is starting exactly at this point you get the maximum value here; and when it is
starting exactly here, you get the minimum value so that is referred to as the phase of the
wave.
So, there are two things which qualify a wave, one is the amplitude of the wave, the
other is the phrase associated with the wave. It is so happens that this amplitude which is
coming out can be detected on a detector, and the detector measures the intensity.
Intensity is proportional to the square of the amplitude. So, therefore from measuring the
intensity by taking the square route we get the amplitude modulus of the wave.
410
However, the phase that is associated with the wave cannot be determined
experimentally. And this is the so called phase problem in crystallography. So, we will
come to the detail of it later, but what is happening here is that these waves therefore, can
constructively and destructively interfere.
And I have shown a picture here a various kinds of crystals that can be artificially grown
or can naturally occur and so on. The logic here is essentially that we use X-rays to
probe, because X-rays are of the wavelength of 1 Å , which is essentially the size and
distance between atoms in inside the crystal. And therefore, that is very useful to see
where the atoms are in principle.
So, we measure therefore the intensity of the scattered wave which is coming out in
some sense. How do we do these experiments, because you see there is a unit cell the
atoms are distributed all over the unit cell of course with obeying the symmetry rules
which we have formulated depending upon the space group, we will have so many
number of equivalent points. So, the atoms and molecules will follow the alignment and
the assignment given to them in terms of the equivalent points.
And therefore, they are not in random positions they are in well defined positions with
respect to the symmetry which we have defined that is the advantage of studying a single
crystal. In a single crystal the arrangement of the atoms and molecules follow the rules of
the symmetry which we have fixed.
For example, the in a case of a space group like P ͞1 for every x y z we have an x̅ y̅ z̅ and
nothing else. These nothing else is very important because we therefore, know where to
relate these atoms which we find out from these experiment to the space group
conditions, and therefore, generate the content of the unit cell. So, X-ray diffraction their
X-ray scattering is the one which is recommended. Use the word diffraction too early I
will define that little later stage.
411
Let us go to the next slide, how is the all this happening, it is all of this is happening
because of the following. It is a very simple experiment which can be done with light, in
fact this experiment as an analogy which I will draw with respect to crystals in a little
while from now. Suppose, we have a light source, and then it has a certain wavelength.
The wavelength of the light can be taken as about 5500 Å . Suppose, you take sodium
light you know the wavelength of the sodium light.
Now, let us say it passes through a single slit in a screen. So, we block it up everywhere,
except that you allow it to go out of a slit. Then from this slit now we will act as though
it is a source of radiation, the slit will act as a source of radiation. And the scattering the
waves will now emanate from that particular slit point in a spherical fashion. So, the
distribution is spherical. So, there is a spherical wave front which comes out of this slit.
So, this slit is now an object and that object generates spherical waves. So, if the in the
place of a slit, there is an there is an atom let us say. The atom now acts as a source
because we saw that the dipole develops. So, the atom, atom acts as a source of
electromagnetic radiation, and it sends out spherical wave front, but at sufficiently very
large distances this becomes more or less a straight line.
So, the it is like you know take a stone, and then go to a pond, and throws the stone into
the pond. The central part where stone falls into the way into this pond, you will have
circle of waves emanating. And these waves become bigger and bigger, and bigger. And
412
if you see at a distance of about 50 feet or so, it will be weaker also, at that the same time
you will see more or less it is spreading entire length of the pond effectively it becomes a
plane wave.
So, we make an approximation that whenever we do this kind of an analysis where we

send an X-rays beam on to the unit cell or a crystal which is consisting of unit cells, the
scattering a vectors, the scattering waves will be emanated from each and every atom.
And each and every atom will generate a spherical wave front, and these spherical wave
fronts move further, and as they go to very large distances they make plane waves.
And therefore, these can be treated as plane waves, mathematically it essentially means
that the mathematical equation that can be fitted is exponential i k dot r,[exp (ik.r)] where
k is called a wave vector, r is a position vector. We will give more details of that as we
go along, but at this moment what is happening is that, if you have a single slit, you get
spherical wave fronts coming out and at sufficiently large distances you get a plane
wave.
Suppose, at a certain distance instead of very going to very large distances, suppose this
little distance later I the so each of these now represent the wave which is coming out of
this particular hole which is that of a slit though I will put more than one slit. Then I put
more than one slit let us say I put two slits, each and every slit as you see shown here
will generate its own set of spherical wave fronts. So, there is a spherical wave front
which starts from here, and another spherical wave front which starts from here because
these two now are the resources.
So, there are two position sources on this particular screen, and therefore, two spherical
waves will start. Now, when two spherical waves will start at this point and they start
going up in their position. You see that a similar thing to what is happening here will
occur. So, we have this wave which is coming from one source, this is the wave from the
other source, and therefore, these two will constructively or destructively interfere
depending upon the overlaps of the wave fronts at these positions as I shown here.
So, whenever we have an overlap, we get a bright line; and whenever we do not have any
overlaps we get dark lines. And this is the process of interference. So, if you now
extended to three dimensions, and imagine that there is a three-dimensional screen with
different slits put into that, then we will get this lines converting themselves to spots.
413
And therefore, we will get bright and dark spots. If you now look at this diagram a little
more carefully and say that I do not put these two slits here, I will put one slit up there
and one slit up there, then the two slits now will generate wave fronts, but these wave
fronts are now different from what is shown here in the diagram because they start from
different positions.
Let us say it starts at this position, then this fellow which is shown as a spherical wave
front will now be positioned there. And when it is positioned there and this fellows now
moved away to this position, that means, the d value the distance between the two slits is
increased to higher values. Then what happens is that the number of overlaps that will
occur between the two waves will now be opposite, it will now have more parts as you as
you increase the distance between these two, we will have more spots or less spots what
will you have.
So, I will increase the distance. So, here you had this direction in which you got a line in
this case, but you will get a spot in three dimensions you get a spot in three dimension.
And whenever there is a intersections occurring you will get light and dark lines here.
Now, if we increase the distance between these two what will happen to the distance
between those two. You think about it, you think about it. And so what happens is that
the d value the distance between the atoms is now increased, in this case let us say this is
an atom position, this is an atom position, and light source is now falling over right away
on these two are waves which are generated and overlap occurs.
So, what happens to these lines, do they come closer or they move forward is something
which you think, I am not going to tell you and that will make you understand the
process of interference much more clearly. So, if I can go and give you a hint, the hint is
in this diagram. Suppose, in this diagram, instead of this wave coming up here the wave
is move to some other position.
So, then what will happen, the intensities will it reduce or increase is a different story,
but whether the interference which is occurring between these two will it be same or will
it be different, obviously, you know it was going to be different. Suppose, I put these two
holes closer to each other then what will happen? If you put the holes closer to each
other, these will now spread further.
414
So, the distance between these lines will become larger, if the holes are closer to each
other. If the holes are farther away from each other, these will become closer to each
other. So, imagine a ultimate situation, in the ultimate situation I will bring these two
these two holes closer and closer and closer and I come to a single position, then what
will happen it will go back to this position.
So, if there is a single atom, it will go back to the scattering coming from the single
atom. Whereas, when we have a collection of atoms, the collection of atoms when they
are very, very close to each other, this distance will go further away, and when they are
farther from each other this will come closer. So, there is a inversion associated with this
or what we call as reciprocity that is associated with that.
We can see these from a very old experiment we have all done in our school it is a
typical high school experiment where we were given a convex lens, and we were ask to
find the wavelength of sorry the focal length of this lens. The focal length of the lens we
all knew the formula 1 over f equals 1 over u plus 1 over v (1/f = 1/u +1/v ). So, where u
and v are the distances of the object to the screen, and the object to the sorry the distance
of the object to the let us go to that picture.
The distance of the object to the lens is u and distance of the lens to the object is v. So,
here is a situation where there is an object in this case I have shown a picture. Now, in
this picture, now goes to the convex lens. So, lights is scattered from different points on
415
this, and all the lights scattered from different points will generally emanate at
sufficiently large distances. It imitates a plane wave which is appeared. So, all these are
now plane waves which intersect at this point, and this lens now does the job of what is
known as a image reconstruction.
So, this image reconstruction takes place and you get the image of this object. And then
you say that the focal length of this lens is 1 over f equals 1 over u plus 1 over v. And we
get the value of f. This was a high school experiment which we performed. In this case
there is a lot of lesson to learn, one is that light fell on these object and scattered in all
possible directions. And these all possible directions were caught in different by the
presence of a convex lens, and this convex lens had a good job of doing the image
reconstruction.
So, it did the image reconstruction and got as an image. So, what is happening is that
these object went through this focus as I see here, and therefore, this is the focus of the
lens and you get the object image. And therefore, you see the image of the object gets
formed in an inverted fashion to the existing object, and this is brings us to the, what is
known as the concept of reciprocal lattice. We will talk about the reciprocal lattice in
more detail as we go to the discussion of what happens with X-rays.
We can right now ask a question what happens with X-rays. Suppose, this, this particular
person we have shown here has broken his hand. Imagine a situation where he has
broken his hand and so we now sent in X-rays on him on his hand ok. So, now, X-rays
come and fall here. The wavelength of X-rays is now 10-8 Å or in that region angstrom
region. And instead of light which was about 5000 Å this convex lens was able to do the
image reconstruction.
If X-rays are used and we want to do the image reconstruction, there are two things we
will have to do, one is it depends upon the value of the refractive index of the material.
So, if the material refractive index in this case of glass, glass refractive index we know
the value of the refractive index. So, the refractive index is the one which we will do this
job of reconstruction, and so we get an image of the object.
Suppose, we are sending in an X-rays beam here then again it will be scattering in all
directions, but in order to find it to focus and bring it to a focal length (Refer Time:
25:14), we will have to make the lens slightly bigger than the size of the earth. So, it will
416
be a very huge lens if at all you make it from glass, and it will be slightly bigger than the
size of the earth, so it is almost impossible to make it.
And the image will be form somewhere far away in space, you will get an image you
will get the broken and indicated in this inverted image, but then you have to go
somewhere in space, use ISRO’s newest launches, and then go there in space and see
whether you can get an image. So, effectively this experiment is impossible to perform.
Imagine a situation where you have kept a crystal here. So, each and every point in the
crystal now becomes a scattering point depending upon the positions of the atoms inside
the unit cell, sorry, it is the atoms which do the scattering; it is not the crystal which does
the scattering. It is the atoms which do the scattering, it is the electron density associated
with the atom which will do the scattering.
So, therefore, this scattered electromagnetic waves will come in all possible directions.
So, we have to therefore find some ways and means by which we can do the image
reconstruction. One of the advantage disadvantages we will further have if we really do
this instead of bringing in an a convex lens, we do mathematically, we do what is known
as a Fourier transform.
If we do a Fourier transform, then it is in principle possible to construct the image of the

object. However, if you want to do that operation, we should know the phase angle of
each and every wave which is emanating. Unfortunately, the way in which we are doing
the experiment, we are actually stopping the out coming radiation and measuring its
intensity. The intensity has given us the amplitude of the wave to determine the phase
then we should know where the atoms are and that is our job anyway.
So, since in a unit cell we our job is to find where the atoms are. And if you want to
know the phase information, we should know where the atoms are, these becomes an
unsolved problem. And so we therefore, have to worry about how to determine the
phases.
This is a very big issue which will be taken up later on when we actually discuss the
determination of the structures. At this moment there is no convex lens that is why we
are having this course; otherwise if we had a convex lens of a suitable kind we could
417
have got the image of the object, and therefore, seeing these atoms inside the crystal that
experiment unfortunately cannot be done now.
So, instead what we do therefore is to do it mathematically, but we do it mathematically

we use the Fourier transforms, what to do the Fourier transforms we need desperately
need the phase information. The phase information is not available, because we know do
not know where the atoms are. Here we know where the hand is, where the where these
and so on in these object, but in the case of a unit cell and the crystals inside containing
electron densities, we do not, so that is why we have a problem. And this particular
problem we will discuss in the future classes.
418
Lecture – 32
X Ray Scattering; optical Analogy
So, this brings us to the issue of how we can see atoms; say as we discussed in the
previous slide, if we know the position of the object inside the unit cell in principal let us
say this is an object inside a unit cell and we sent in an X-ray beam. If we know the
positions of these atoms this edge of the elbow or leg or whatever, then it is possible to
reconstruct the image by getting the amplitudes of the scattered radiation and doing a
Fourier transformation. Lens is not possible, but we do the Fourier transformation; get
the information of the position of the atom which is being present already.
But what happens is that we do not know where the atoms are and that is the job. So,
there are two unknowns here. The positions of the atoms are not known. The associated
phases are not known. So, in fact, we cannot get the image. So, we have to either guess
the position of the atoms which is what was done in the earlier days of structure
determination or guess the phase position and then we have to get to the; get to the this
one the image of the object.
419
So, how do we go about it? We go about it in such a way that the image reconstruction
has to be monitored by actually solving the phase problem. Now, how do we solve the
phase problem? We solve the phase problem by guessing what the phases are. There is
the only way we can get to the image of the object which means the positions of atoms.
So, how do we really see atoms this is a picture which I have taken from the book of
Glusker and True blood; a remarkably good quality book which I make I suggest that all
of you go through it. Here, this is a picture from the older version of Glusker and True
blood as you can see that we here have written photographic film. In modern days, we do
not use the photographic film we use different kinds of modern day detectors like the
CCD cameras or the EOS and so on, the photon detectors and so on.
Nowadays, a very high quality detectors are there to measure this. So, what is the thing
we need to construct, we see if we have the knowledge of the phase as well as the
amplitude information like in the case of a normal microscope, where we have the visible
light as the probe and visible light falls on an object. The objects scatters radiation in all
possible directions and we take now this scattered radiations and put the convex lens in
front of it the so called objective lens.
We get an image of the object. So, the main object generates the image just like the
previous slide we had and this image of course, can be made to enlarge by what is known
as an eyepiece lens and this now enlarges the image.
420
So, suppose let us say this was a piece of the bread which we had and I put this here and
put it under a microscope. I will get an enlarged image of the object of that piece of
bread and you see some of these little pieces here that could be the pieces which we
talked about in terms of cashews and badam and things like that.
So, effectively, we see they enlarge the image and therefore, we get an information about
the nature of the object much better and this is how the typical microscope works a like
microscope. We can use of course, different kinds of radiation and make the
corresponding microscope, we can use electrons and make a electron microscope and we
can use the electron microscope to fall on the object scattering will occur.
Now, instead of the lens we can have a measuring device like a CCD or something. So,
in this particular example, we are using X rays, if the object is the crystal. Now one of
the property is which the crystal has which I have not mentioned so far which you have
learnt so far, but I have not mentioned it. Is the fact that there is a systematic
arrangement of objects inside the crystal and this is governed by symmetry and because
it is governed by symmetry the scattering they occurs not in all possible directions. But
scattering occurs in only predetermined directions; predefined directions depending upon
where the nature of the crystal.
So, we have let us say the crystal lattice known. We know where the equivalent points
are; we know what are all the atoms which are there in the system. In principle what we
measure or the scattering which is coming from individual crystal planes and therefore, it
is not in all possible directions like our interference experiment told, if you have two
slits; the two slit experiment gives you the interference pattern dark and bright images. If
it is a three dimensional one you get spots week and dark spots and therefore, you do not
get the continuous scattering as is as it happens here. But you get discrete directions.
Now, the fact that this happens in discrete directions helps us to identify the spots.
So, we can identify the spots; we can also identify the coordinates of the spots and we
when we identify the coordinates of the spots, then we identify the nature of the way in
which the spots are distributed and this is referred to as the reciprocal lattice because
what we get is before the image reconstruction and we get the intensities of each and
every spot which is the intensity of the reciprocal lattice point. Then, we solve the phase
421
problem by some means and do a Fourier synthesis using the help of a crystallographer
in earlier days, but nowadays its only computer which can do the job.
So, we do not need a crystallographer; we just do these computers, use the computer and
do the image reconstruction and then, one once we do that, we will get the image of the
object. The image of the object now just like here, we saw all the additives; we will see
the positions of the atoms. What we see are not the positions of the atoms per say we see
the so called electron density distribution. The heavier the atom the more is the electron
density; the lighter the atom the less is the electron density. So, you get what is known as
an electron density map.
So, this in a nutshell tells you what we do in X-ray diffraction techniques. So, we have a
crystal, the crystal obeys all the rules which we have formulated for the symmetry
information and therefore, the scattering occurs in only specified directions, these
specified directions now can be caught on a screen it could be a photographic film; it
could be a detector. Then, we do the image reconstruction through Fourier synthesis.
After solving the phase problem and we get the image of the object.
So, to recapitulate again, we have a crystal, X-rays are falling on the crystal. The crystal
now we will scatter only in certain specific directions. We will find out what are those
specific directions in a in the next couple of classes and then, these now create dark and
white images we can say the so called reciprocal lattice will be created.
Because we have a lattice in the crystal, the crystal lattice will now create the reciprocal
lattice, we have just seen why it should be reciprocal from the previous diagrams. And
therefore, we this reciprocal lattice information can be now used. We measure the
intensity of every one of the spots. We also estimate the phases associated with these
spots.
We measure the coordinates of these spots in reciprocal space and then, do a Fourier
synthesis after guessing what the phases are and then, we get an image of the output.
And the crystallographer we will do the manipulations and the changes that are required
based on the Fourier synthesis in order to get a final electron density map and this can be
now done through computers. We have the synthesis Fourier refinement programs,
where we improve the phase information such that we get to the final electron density.
422
So, what we get finally is the electron density map; the distribution of electron density
associated with atom positions, wherever there is an atom we have a large amount of
electron density as you all know. And therefore, depending on the strength of the
electron density, we can also find out which atom is what. Therefore, we get the entire
structure. And the structure now will come in a situation of a single units under the single
unit cell, we know the equivalent points and therefore, we can determine the structure
associated with the asymmetric unit.
So, what we eventually look for is the structure that is associated with the asymmetric
unit in the unit cell right. So, how is the whole thing done?
So, this is the whole logic of how this whole thing has being carried out. So, let us go
through it very slowly and systematically. So, when electromagnetic radiation passes
through a crystal. So, we now have a crystal we are sending electromagnetic radiation
now we know it is X-rays. So, X-rays are falling on the crystal or matter in general as we
discussed so far the electrons are the ones which are perturbed that are disturbed by the
rapidly oscillating electric field.
Since these electrons are now connected to the nuclei, the nucleus and the electron
density gets disturbed in such a way that the positive and the negative signs there is a
displacement and this displacement causes a dipole and this dipole now oscillates with
the incoming radiation.
423
And therefore, it emits its own radiation and we have already made an assumption as I
mentioned that it is coherent scattering. So, there is no change in the frequency after the
scattering experiment is done. So, we shall take the electric field as E 0 cos 2πυt and if we
take that as the electric field or the electric disturbance which occurs at the origin and
whenever it falls on that object, we have to see when and where it falls on the object.
So, what is happening is that these now emanate as spherical waves as we saw from the
previous diagrams, where we had an object and a slit experiment. So, it these emanate as
spherical wave fronts. In general instead of using only the real part we also bring in the
imaginary part of the axis which is mainly because electromagnetic radiation and the
theory of electromagnetism essentially involves complex waves; waves expressed as
cosine + i sine.
So, E0 is therefore, E0 ei2πυt. So, the ei2πυt can be expanded further into E0 [cos 2πυt + i sin
2πυt]. So, it is a cosine and a sine function and therefore, we write this what is known as
the organd diagram, where we have the imaginary axis along the y direction and the real
axis along the r direction.
So, suppose we have two waves E1 and E2 and they are now coming out from this system
and E1 is at a phase which is δ 1 and E2 has a phase which is δ 2, then these two now we
will combine and the way in which the resultant is taken is very well known in
mathematics. So, we take the resultant and the resultant of these two waves E 1 and E2 is
E which is now has a separate phase which is δ. This can be in fact, illustrated with a
diagram here.
424
So, if we now say that this is the E1 and this is E2. So, depending upon the phase
associated with E2, this wave will develop either constructively or destructively as it
happens here. So, this particular wave now is overlapping a different positions along the
incident wave or we can take this wave as E1 which is emanating from the origin which
is the electron density position.
And E2 is another wave which originates from some other atom, but we associate that
with the origin, then we can combine E 1 and E2 and get the average the resultant
associated with E1 E2. And therefore, you see that if this is the phase associated with E 1
δ1; this is the phase associated with δ2, then we see that the phase associated with this
will be δ which is a combination of δ1 and δ2.
So, if we consider it that way, we go back to that picture and we see therefore, we have
an E1 and then E2, the resultant vector is along the E direction and so, this is the
corresponding δ value. What is happening in a scattering experiment is one more issue
which we have to consider, that is the issue of what we see and where it happens because
if the electrical disturbance is associated with let us say E 0 ei2πυt and that is occurring at a
particular origin 0 0 0 and then, we actually look at what happens at a distance which is
R from the origin.
This particular wave which we are now talking about, which is a now a complex wave, it
spreads at a velocity of light. So, it spreads at a velocity of light and it spreads as a
425
spherical wave front; both we have seen and had sufficiently large distances r we have
already decided, it will be a plane wave. So, it will go as E i.(k.r), where k is the wave
vector and r is the position vector. So, that is associated with the amplitude E 0 and this
will be the phase.
So, the phase now if it is real is cos 2πυt, but since we are considering electromagnetic
radiation, the phase will be ei2πυt. Now, the frequency we have assuming is not going to
change and therefore, even if we now look at a distance R, we are not looking at the
disturbance which we are now looking at a disturbance which originated at a time t - R/c,
because we are now measuring the intensity at a position which is at a distance R from
the source.
So, if this is the source and we measure the distance at a distance R, then it is already
travelled that distance with the speed of light and therefore, what we are measuring is
that which occurred t - R/c before. Let me explain it a diagram. Suppose, this is our
position of the object; this is the position of the object from which these scattering is
emanating, we are now measuring it at a distance let us say R.
So, this distance is R now. So, the wave front is spherical to start with and somewhere
here it is almost a plane wave. So, as we measure it at a distance R, we are now sitting at
this position and measuring. Now, what we are measuring is not a t, but t - R /c because
the wave started t – R/c times before at that particular position. Let me make it clear
again, what is happening is that if we are measuring at a time t with respect to the
position R, we are actually looking at the wave which originated t – R/c, where c is the
velocity of light and R is the distance at which it has travelled.
So, light has already travelled by that distance and therefore, this is therefore, what is
known as the path difference. This introduces a difference in path and the path difference
is now can be expressed in terms of wavelength as well because c is equal to; c is equal
to υ/λ. So, therefore, we see that it is dependent upon λ. So, we introduce this expression
it is either 2πυ(t – R/c) or 2π(υt – R/λ) and this is therefore, the extra path which it has
traveled at any given time t.
So, the path that is travelled by the wave is covered by t - R/c. So, if that is the case then
we have to consider that as the so called path difference or the phase sorry so called
phase difference. So, if R becomes R +ΔR and this ΔR is very small let us say. So, if R
426
becomes R +ΔR, then there is a increase in the path since this is t - R/c; you see t - R
+ΔR /c. This is now the distance which is which has added a ΔR to that, we will make
the value of the phase associated with that wave a little longer.
So, there is a phase difference which is negative here and therefore, the path difference is
positive. So, for every additional path the phase difference is more. So, as we keep
adding, the phase difference will become more and more. So, it is inverse of the two.
And therefore, we see that the expressions can be worked out now, the way we have seen
here. So, we now look at the complex amplitude of the wave the so called E o. The Eo can
be equal to if there is an electron density ρ(r); if there is an electron density ρ(r) dr that is
present in space. Let us say and then you are sending in the X-ray beam the X-ray beam
is coming in these particular direction and we call this quality as s 0 which is the unit
vector of the representation of the so called wave front. Remember the wave front now
which originated as a spherical wave front has become a plane wave.
So, therefore, we are writing a single line here showing the wave front. So, as the wave
front approaches in this direction, we sit here at an origin and observe what happens to
the presence of a little amount of electron density at this point. So, we put a little bit of
electron density there and then the rays are coming in and we are sitting at this origin and
observing what happens to that.
427
As we have said earlier that the electron density now will change the direction of the
wave, it will start its own wave and that wave we will now originate in the direction P’.
So, the original one is going in the in the direction of P, this scattered one is going in the
direction of P’.
So, now P P’ will define the so called scattered direction. So, incoming direction is
referred to by the unit vector s 0. The scattered direction is referred to by the scattering
vector s. So, what is happening is the fact that whenever you find this electron density,
this electron density is located at a distance of r from the origin. So, we have an electron
density which is located at a distance of r from the origin and we see therefore, that this
wave travels an extra distance of r . (s – s0).
So, since this travels and additional distance of r.(s – s0), this is the additional distance.
This has travelled (r.s), it goes up like that; whereas, the r.(s – s0) is the additional
distance. This will be the one which causes the phase change. So, the phase change
therefore, is 2πυt that is 2 π i r.(s – s0)/λ.
We are assuming that there is no change in the frequency. So, the frequency factor does
not appear here. It is only the change in the path length. So, the change in the path length
is introduced as ρ(r) exponential 2 π i r.(s – s0)/λ and this is occurring due to the presence
of an electron density which is ρ(r) dr.
So, Eo is equal to this quantity. The (s – s 0)/ λ we replace it by a R which we call as the
scattering vector. We will look at some of the properties of the scattering vector as we go
along. The scattering vector has dimensions of reciprocal length; λ is the wavelength. So,
R has dimensions of reciprocal length; that means, this scattering vector is something
which could be a representation of the reciprocal lattice vector in some way. We have to
find out it what is the representation of R because it is proportional to 1/λ?
So, the fact that our image gets inverted after the scattering experiment is done, using the
analogy of a convex lens experiment of the light waves. We see R is proportional to 1/λ
and therefore, R could be utilized to define the coordinates associated with the reciprocal
lattice. I am going to do this again one more round so, that there is no confusion
associated with the understanding of this phenomena.
428
This is the basic phenomena of scattering. So, this quantity R can also be if you do a
construction of this kind shown below here, s0/λ is the incoming radiation s/λ defines the
scattering direction radiation and then, this scattering is due to the presence of the
electron density which is ρ(r) dr. Then, the scattering vector is in a direction
perpendicular to these two.
So, it is actually a bisector of s0/λ and s/λ and I take this angle as 2θ, then it is usually
shown that modulus of |R| is equal to 2sinθ/λ and this is the scattering vector. You see
the scattering vector modulus is proportional to 1/λ and therefore, this represents a
quantity in reciprocal space.
So, if r is in the direct space, where the electron density is located in our object; the R is
the representation of the reciprocal lattice vector which corresponds to the real space
vector r. The R arises because of the real space vector r having an electron density
associated with it and a phase which is defined as exponential 2 π i r.(s – s0)/λ.
And so, as a consequence the path P’, the extra path that is travelled is r.(s0 – s) this; that
means, when the wave is coming up here, it finds the electron density it travels an extra
path which is r.s0 that is because this is the perpendicular direction which we have taken.
From here, we draw another perpendicular.
So, therefore, r.(s0 – s) is the path difference and we have already seen the path
difference being longer now, the phase difference will be decreasing accordingly and the
decrease in the phase is 2 π i r.(s – s0)/λ and that is what is substituted here it. It is 2 π i r.
(s – s0), the decrease in the phase. The decrease in the phase can be represented as (s –
s0)/λ and therefore, this is the path difference which occurs on the path difference is a
consequence of the definition of the scattering length. So, R therefore, is a r.(s – s0)/λ
We will study this in little more detail and probably in the next class, but at this
particular point we notice R is (s – s0)/λ and now we have to consider the entire crystal. If
you consider the entire crystal there is electron density distribution depending upon how
many atoms are there inside the crystal. So, what we do is we assume that the electron
density is continuous in the entire crystal. It is not going to become negative, 0 and all
that.
429
So, we assume that even if it becomes negative, 0 and all that, it is going to be a
continuous function. Since, ρ(r) is a continuous function, we therefore, see that we can
take the ρ(r) is the continuous electron density distribution; what all we have to do is to
express the overall amplitude which comes out of the crystal.
See what is it we are looking for, we are now getting a we get back to this picture. We
are getting the scattering coming from the crystal in a certain direction and this particular
direction is due to the scattering which occur due to the presence of the electron density
inside the crystal. And the presence of the electron density inside the crystal has now
resulted in the change in the path length and any additional path length results in
decrease in the phase associated with the wave.
And so, we can calculate the complex amplitude which is in fact, if we measure the
intensity the square root of that intensity will give us the amplitude. The complex altitude
can be calculated by these expression and this is an expression which is associated with a
element of a space which is representing ρ(r) dr. But if the electron density is continuous
in the entire crystal space, this can now be obtained in the as the integration over the
entire space the entire crystal.
So, the integration over the entire crystal of ρ(r) dr this is the phase factor and this is the
quantity which contributes to the overall complex amplitude and therefore, we get F(R)
which is referred to as the scattering factor. Notice that this quantity F is a function of the
scattering length R and this particular scattering vector has the dimensions of reciprocal
lattice. So, when we measure the so called intensity we are actually measuring the
intensity coming due to the presence of the reciprocal lattice point.
So, the intensity is coming from a reciprocal lattice point identified with respect to the
scattering vector R and therefore, we get F(R) and this is a complex quantity again and
this is an integration over ρ(r) dr which is the overall electron density now inside the
crystal and this is the modification which gets introduced in ρ(r) dr due to the fact that
the object that is located inside the ρ(r) dr the atoms are located at different positions
So, you will get different phase changes causing this so, called path differences which
are represented by the scattering vector R. So, R is (s – s 0)/λ and we get F(R) is equal to
this quantity.
430
Let us go to the next slide and define find out the properties of F(R). We will do this
derivation once again a little later in a slightly different way so that it becomes very clear
of how the scattering is occurring. But before we go to that we just look at the possible
properties of this F(R). F(R) is a complex quantity as we see already and the intensity of
the scattering as I mentioned is given by F(R) times F*(R).
Now, this F*(R)is referred to as the complex conjugate F(R) and F*(R)in fact, essentially
define I(R). So, if we measure the I(R) and we take the square root of this quantity, we
will in principle get the magnitude of F(R). In principle, the intensity and hence F(R)
modulus can be measured experimentally; therefore, as a function of R. However, the
phase angle associated with F(R) is not an observable quantity.
So, then of course, a special property of a F(R) is that if there is a if ρ(r) is real. F(- R)
can be written as F*(R) and the scattered intensity is unaffected by reversing the
direction of the incident and scattered beams. Likewise the same intensity pattern is
produced by a given distribution of ρ(r) and by the distribution of ρ(-r) obtained by
inversion through the origin that means, when you have a inversion centre and the
electron density is measured, this inversion centre we will ensure that you will get a
quantity which is actually a real quantity.
So, F(R) in that situation of a centro symmetric positioning of the atoms will become a
real quantity. So, if the that is what is written here if the distribution ρ(r) is Centro
431
symmetric, if ρ(r) = ρ(-r) let us say in P ī; then F(R) is a real quantity. So, this again we
will now refer to the organ diagram and in the organd diagram we will represent what
happens when there is a centro symmetry and what happens when there is no presence of
an inversion centre and that is something which we probably we will take up in the next
class.
So, at this moment just to revise what we have done so far in this class; what I will do is
we have taken the scattering issue and we have now considered the scattering process in
detail. So, what we have done in this particular case is to look at two possible waves
which are interfering with each other; two possible ways which are interfering with each
other; one is represented by E1, the other represented by E2 and then, the resultant one is
represented by E. So, we depends upon what is the corresponding phase angles
associated with E1 and E2.
432
Lecture – 33
X-Ray Scattering: Fourier transforms
So, we discuss this scattering theory yesterday. And it looks like that from the feedback I
got from my students. It looks like an another explanation, and probably a thorough
explanation of the scattering is required, particularly with the view that there are people
from pharma industry, and possibly even teachers who are trying to go through this
course.
And so what I thought was I will prepare some slides reorganize them. And then go
through it once more, so that we get a clear feel for the scattering theory, because this is
this is the basis of all scattering, and eventually the diffraction, and what happens, when
X-rays fall on a crystal. And how the diffraction appears, and how the equivalent points
respond, and how the atoms and molecules respond and so on.
So, what we therefore say is that when electromagnetic radiation passes through a
crystal, the electrons are perturb by rapidly oscillating electric field. So, here is a wave,
the wave as we know can be characterized in terms of its amplitude. And this angle at
which it starts is referred to as the phase angle. The wavelength associated with the wave
433
is now given by λ, and that is the distance from the crest to the next crest. It is also equal
to the distance between the trough and the trough.
And therefore, when these electromagnetic radiation comes in and falls on an atom, this
is a bloated picture of an atom, I have taken helium as an example. So, what happens is
that there is a central nucleus, which is positively charged. And then there are electrons,
there are two in the case of helium and these electrons carry a negative charge.
So, when this wave comes and disturbs these atom, as we see here that it generates a
rapidly oscillating electric field, and this because the fact that the wave which is coming
in can be described in terms of an electric field, which is E0 cos 2πνt taking only the real
part or E0 ei2πνt, which is taking into account the complex nature of a wave; so these
electromagnetic waves in general or complex quantities.
So, to describe this wave we make use of the complex numbers, we will go into the detail
a little later. But, at this moment there is an incoming wave which comes and hits the
atom, and the atom is set to vibrate. Now, because there is a heavy central nucleus, the
electrons are disturbed much more the outer electrons particularly.
And as a consequence the positive and the negative charges develop, and therefore it is
becomes a dipole a positive and a negative is separated by distance is a dipole. So, this
dipole now oscillates with the same frequency as the incoming radiation. So, there is no
change in the frequency and so we call it as coherent radiation.
And according to electromagnetic theory any oscillating field of this kind will generate
its own wave as you shown here. So, this wave now comes out in this form, and this
particular wave now is from this particular atom. So, it may so happen that there are
several atoms in the crystal and all these atoms are connected to each other and so on.
As a consequence, we will get a large number of waves coming out with this interaction.
The advantage again is the crystal and the symmetry and that is why you know we keep
on stressing on the symmetry throughout this course. The fact that these atoms or inside
a unit cell, and the unit cell obeys the laws of symmetry.
The issue of the obeying of the laws of symmetry brings in the fact that these atoms now
or inside a lattice. And since, they are inside a lattice, the presence of the atoms follow
434
the symmetry rules. And since, they follow the symmetry rules the coming out of these
waves are actually in a direction, which is which is decided not just by the presence of
the atom inside, but also by the presence of where the atom is present depending on
where atom is present, different kinds of waves get to be coming out of the system.
And therefore, we have a several atoms giving rise to its own their own waves. And so if
you see this as the equivalent of a crystal at this particular point, we see that there is a
source which sends electromagnetic radiation has spherical waves, sufficiently long
distances, the spherical waves can be approximated to a plane wave.
And this now in turn we will generate its own spherical waves, so this point now
becomes if this is the crystals, now crystal becomes the source. So, from this we get lots
of waves coming out, and therefore we have the process of interference which we
recorded yesterday, and which tells us that only in certain directions, we will have the
bright light coming out and the rest of the directions will be dark. So, we get in a three
dimensional geometry, we get spots dark and white spots.
And these spots now will go to the receiver, which is essentially a detector. And we find
out as a consequence the nature of the way in which this these; what we called as
eventually reflections come up. We will get to a better understanding with as we go
along.
435
So, the electric field that is associated with the disturbance, which occurs due to the
presence of the wave coming in the incoming radiation this is the wave. And that
particular wave now introduces a certain disturbance at every atom position. So,
wherever there is an atom, these dipoles get developed, and then we get this oscillations
and these oscillations will inturn generate waves.
So, if we consider the electric field as E 0 cos 2πνt, this quantity E0 is the amplitude of the
wave, and cos(νt) now represents the so called phase associated with the wave. So, the
entire wave is described in terms of the amplitude, and the phase that is associated with
the wave. And this particular this electric field now spreads as a wave front, and that
wave front goes with the velocity of light the c is the velocity of light, because X-rays
are photons, and they also have the velocity of light they are essentially light. So, these
photons therefore will move with a velocity of light.
Now, just as we were mentioning as the spherical wave fronts, which spread from a
given point at very large distances at a particular time t, what really happens is that this
phase associated with that wave is at a time t, at a distance R. So, we now move to a
distance R from the position of the scattering scatterer. And at time t, if we observe the
behavior of the wave which is coming at a distance of R, it is actually the phase that is
emitted from the source at a time t - R/c, because it is travelling with a velocity of c, R is
the distance at which we catch it.
And therefore, whatever we measure at t is now that which originated at t - R/c, and that
is the phase which we get at t - R/c. And therefore, the phase will be 2πν(t - R/c) or 2π
(νt - R/λ), where λ is c/ν, this is something which there was a mistake yesterday while
writing down, even though I explained it as λ equals c/ν, there was a mistake. So, λ is
defined as the velocity of light divided by the frequency. We have also noted that the
frequency will not change, after scattering that is the coherent scattering.
Now, what we do is we consider a situation, where the R distance R is now becomes R+

ΔR. So, if ΔR is an additional distance that means, there is an increase in the path length.
Whenever there is an increase in the path length, there is a corresponding decrease in the
phase. So, as we go longer and longer distances, we the phases keep on reducing
associated with this scattering vector.
436
And therefore, we see that it leads to a decrease in the phase, and that decreases 2πΔλ.
So, if R+ ΔR, becomes now 2πΔλ will be the decrease in the phase. And since we
already said that these are electromagnetic waves, the electromagnetic waves are indeed
expressed as complex numbers. So, instead of E0 cos 2πνt, we have E0 exp 2πiνt.
So, this now is the representation, which is in the organd diagram below, we have the
imaginary axis, and we have the real axis. And suppose we take two such waves coming
out from this experiment, we have discussed only one here. Suppose, there are two such
amplitudes which come out from a single point at taken as the origin, then we have a E 1
which has a phase of δ1, and then we have an E2 which has a phase of δ2. Then what
happens is that when these two combined due to interference, we will have the resultant.
And the resultant is E, and their corresponding value associated with the phase is δ.
So, we noticed that as we go to higher and higher distances. So, if R becomes R+ ΔR,
there is a decrease in the phase by 2πδν. So, what are the consequences of these? So, we
have to now see that there is a wave, the wave is represented by a complex equation
which is E0 exp 2πiνt, this particular quantity represents the wave. The first part which is
E0 is the amplitude of the wave, exp 2πiνt is the phase associated with the wave. To
describe the wave fully we need both these quantities, the amplitude as well as the phase.
Now, what really happens is the amplitude is the one which carries the amount of energy
in the in the system. So, if we consider E02, which actually represents the total energy, it
is also refer to as the intensity of the wave. So, when the intensity of the wave has to be
looked at, we consider the E02 as the intensity of the wave; so the amplitude square.
So, what happens in a typical experiment, which we did in the last class about the
interference experiment, for example is the fact that when waves coming, and two waves
interact with each other, there is a interference. And that interference depends upon,
where the waves originated from that means, it depends upon the δ 1 and the δ2, which are
shown here.
Depending on the value of δ1 and δ2, we will get a resultant which is corresponding to the
phase δ and the value E. So, what is happening in this situation is that there are two
waves coming out, and these two interfere. And the interference results in an addition of
these two waves, which are coming in. The addition is with respect to the vector
directions E1 and E2.
437
So, what is the consequence of this? The consequence of this can be seen in the next
slide, where we see that we have the complex wave coming in. So, the incident wave
which is coming in this is a wave, as we have imagined before. So, it has an amplitude
and a phase. Now, this incident wave which is representing the energy associated with
the incoming radiation which is E02 is the energy, so that is referred to as the intensity.
So, there is a certain intensity of the wave that is how the light intensity comes up. And
that intensity is now incident on, let us say a small volume dr in the in a in matter. It
could be a crystal, it could be any other state of matter, but we will consider a crystal. So,
inside the crystal that is a small elemental volume dr which we consider, where it has
ρ(r)dr as the electron density that means, there is a certain amount of electron density,
which stays inside this little piece.
And because this electron density is there, when the incident radiation falls on that it
undergoes a the scattering because of the presence of the electron as we discussed earlier.
There is an atom, atom is attached to the electrons, and the electrons are the ones which
interact with the X-rays. Please notice that whatever interactions that occur in an X-ray
diffraction experiments are more or less or we can say almost entirely due to the
interaction of X-rays with electrons, this is the point which we should never forget.
And therefore, depending upon the electron density, we will get the scattering variations.
And that will eventually interfere and generate the scattering. So, the incident radiation
438
which is coming in which is generated from an X-ray source can be taken as in terms of
a unit vector s0, s0 is referred to as the unit vector. And this unit vector s 0 now represents
the incoming radiation, and this if we if we sit at an origin O (0, 0, 0), here we decide
that this is the origin.
And we look at the what happens, when these electron density presents itself in front of
the incident beam. So, in the incident beam comes up here, and falls on this ρ(r)dr, and
then it is scatters in this direction P-P'. So, this direction which in which it scatters can
also be considered in terms of a unit vector S. So, s 0 is the incident unit vector, and s is
the scattered unit vector.
So, if you consider now the two paths, the path traversed by this line, and the path
traversed by that line, you see that there is a difference in the path length. The path
length is in terms of so what we do to take this origin drop a perpendicular to this point,
which is now representing the incident radiation.
And take this point and from the electron density which is present at dr we drop a
perpendicular, so that is r . s and this is r . s 0. So, the path difference therefore will be
r . (s0 - s) that is the additional distance with this wave this line has traveled representing
the wave compared to this particular line. Since there is no electron density here, and
there is a electron density there the extra distance that is (Refer Time: 15:38) is the path
difference.
We already saw that a path difference of δ, we will introduce a if it is positive, it will

introduce a negative change in the path in the phase difference. So, we can always
calculate the phase difference in this particular case. So, the path difference is r . (s0 - s),
so the phase difference will be 2π, 2π(s-s 0)/λ, where λ is the wavelength. So, the phase
difference for the path difference of r . (s0 - s), the phase difference is 2π, 2π(s-s0)/λ.
So, when we now decide to represent, what is the amplitude associated with this electron
density ρ(r)dr. We bring in the issue of E 0 that E0 is ρ(r)dr as given here, and this now
becomes the so called phase term. So, the phase term you see is now saying s, (s - s 0)/λ,
which is now the path difference 2π(s - s 0)/λ. Since, this is a complex quantity, and it is
with respect to r . (s - s 0)/λ. So, this is r . (s 0 - s) the path difference. So, the phase
difference should be the phase difference should be smaller.
439
So, larger the path difference larger the phase difference, and therefore there is a change
of sign. The change of sign is incorporated by changing s and s 0 positions. So, we have
therefore the complex amplitude expressed in this following wave. So, ρ(r)dr this part
will explain the will consist of giving the E0 part the amplitude part, and this is the phase
that is associated with the wave.
So, therefore we can now express the complex amplitude the associated with the wave in
terms of ρ(r)dr exp 2πir . (s - s0)/λ is the phase difference. Now, we make an
approximation here is not an approximation, we give a nomenclature for this quantity (s -
s0)/λ. Notice that (s - s0)/λ is representing a distance, which is in inverse space or
reciprocal space, because λ is the wavelength, and it is measured as a distance. The
difference (s - s0) is also a distance.
So, therefore the we now replace (s - s 0) by a value R. And this R is referred to as the
scattering vector. And this scattering vector will be equal to (s - s 0)/λ. So, therefore this
quantity R has dimensions of reciprocal length remember this, because we are now going
to worry about what are the types of possibilities this vector R can take in a situation,
where we have the scattering occur occurring inside a crystal. So, this vector R and its
properties are what we will worry about as we go along.
So, to summarize this part of the discussion, we have any incident radiation. And then a
scattered radiation, it falls on a certain electron density ρ(r)dr, which is a small elemental
volume inside the crystal let us say. So, as a consequence we get a path difference
associated with the path in which the wave traverses, and that is r . (s 0 - s), which leads to
a phase difference which is - (r . (s0 - s)), which is (s - s 0) r . (s - s0)/λ multiplied by 2πi
indicating that this is a complex quantity.
So, we therefore get the expression for the complex amplitude as ρ(r) exponential this.
And we now define (s - s0)/λ in terms of a vector R. And this particular vector R,
therefore has dimensions of reciprocal length. So, what you noticed in this expression is
very interesting, there are two vectors r and R. The r is refer to the direct space distance,
the distance from the origin to the electron density. So, this is indirect space and R is in
the reciprocal space.
And therefore, r . R now is a product which represents a product of a distance and a

quantity which has the dimensions of reciprocal distance or inverse distance. And so r . R
440
will turn out to be a simple number, we will show that later on. But, in this particular
case you see that as I have written here, the caption to the figure the path p' is r . (s0 - s)
longer than the path p. And therefore, the corresponding to a decrease in the phase,
which is 2πr . (s - s0)/λ, and we take it take this as a complex quantity; so i gets enough
there.
Now, if ρ(r) is a continuous electron density, suppose we have a little element of volume
considered, and we said it is ρ(r)dr, but it is there now in the entire crystal. So, ρ(r) now
represents the continuous electron density distribution. Then the amplitude will be and
integration over this quantity. So, this complex amplitude which we are talking about,
the complex amplitude associated with the wave now coming out from a crystal, which
has a continuous electron density distribution.
So, we assume that inside the crystal, we have all these atoms and molecules and so on.
And the electron density that is associated with all these atoms, molecules, and so on or
distributed in such a way that there is continuous electron density distribution, this is an
assumption we are going to make. So, the whole entire crystal is now filled with electron
density. Of course, the electron density will be more, when there is an atom position.
Much more, when there is a bigger atom. Very less, when there is a smaller atom. And
between molecules and there is some space in which the electron density will be reduced
considerably.
But, essentially the whole unit cell now is filled with electron density. If that is the case,
then the complex amplitude for the entire crystal will turn out to be F(R) = ∫ ρ(r) exp (2
πi r . R) dr. So, this is an expression which we call as the structure factor. I have not
written it down here, because we are going to bring it again and again throughout our
discussion. But, please note that F(R) is referred to as the structure factor that is because
it represents the strength that is associated with the amplitude of the structure, so because
of the fact that there is electron density ρ(r)dr inside the crystal, this quantity F(R) gains
strength.
Now, we define what we call as a Fourier transform, we will define at this moment. And
we will see that the Fourier transform is essentially a value of ρ(r) taken outside now, so
that means F(R) can be converted to ρ(r) by this expression. And ρ(r) can be converted to
F(R) or F(R) can be converted to ρ(r) by this expression, ρ(r) can be converted to F(R)
441
while this expression. And these two together form what are known as Fourier
transforms of each other.
We will be spending considerable amount of time in understanding this quantity, but

essentially what we say here what we see here is that a ρ(r) is in the real space that is the
electron density. So, if we want to calculate electron density, we need to have the F(R)
values. And these F(R) values are in what we call as the reciprocal space.
So, the R vector is in reciprocal space R vector, and r vector is in the direct space. So,
you also see that there is a change of sign here exp (-2πi R . r) dR that is the
representation of the electron density. So, given the value of F(R) in principle we should
be able to get ρ(r), provided we know about this quantity which is the phase. On the
other hand if we are given the F(R) and the phase value, it should be possible to get ρ(r).
It is also indicative that if we know ρ(r) the electron density distribution, where the
atoms are sitting and so on. We can easily calculate the quantity F(R) exponential, this
the whole thing can be calculated.
So, if we know the structure, we can get the structure factor. If we know the structure
factor in principle, we should able we able to get the electron density. The only catch
here is that we need to find the phase. The problem is referred to as the phase problem,
because the finding of the phase is not going to be straight forward, because the waves
which are scattered emanate from different points in the entire crystal, and we do not
know from where it is coming from. It also depends upon where the atoms are sitting.
Even though we now can imagine the crystal to be consisting of a of several planes, and
then we can actually look at the possibility of sending an X-ray beam want to a given
plane, and trying to get the intensity out of that plane. The actual scattering is done by
the electrons, and therefore the atom positions. So, whenever we refer to as x, y, z, and
say there is an atom in that particular position or an object in that position. It is the object
which scatters, it is not the plane inside the crystal which scatters.
But, we will see as we go along, because this is something which we will always be
confusing, when people talk about the scattering coming from planes. So, we have to sort
out, what is the relationship between the scattering that comes from the planes. And what
is the relationship between that and the electron density that is associated with the entire
442
unit cell; so the entire unit cell has atoms and molecules and so on. And it is these
electron density associated with these atoms and molecules, which cause these scattering.
You have to have a physical presence of the electron density to scatter these wave, but it
is not necessary that these electron density should be associated inside the crystal with a
plane, because we have been deriving equivalent points in the earlier classes. And
equivalent points clearly tell us that the objects are generated by symmetry operations.
And these objects which are generated by symmetry operations either are associated with
general positions or they are associated with special positions. The general positions and
the special positions, we have refer to refer to the possibility of having atoms associated
with planes, it is not telling as or planes or axis or whatever.
So, if there is an atom sitting on a two old axis, we will get a special position. Atom
sitting on a mirror plane, we get a special position. This analysis we have done in the last
classes, but we have not done an analysis where the atom has been restricted to sit on a
given particular plane. And that particular plane need not be having a mirror symmetry,
but what we are going to do as a measurement of the scattered radiation because of the
fact that the scattering now is in very loose terms quantized with respect to the crystal,
because these measurements will come now with respect to where the scattering material
is located, and this is located by symmetry rules.
So, the laws of interference will tell us that where these spots will come, so it is not a
continuous distribution of intensities, we get discrete distribution of intensities. So, this is
where we have to get a very clear understanding. And we will make an effort to get a
clear understanding of this aspect as in the future classes, and in this class onwards. But,
what is very important is the or these two equations, these two equations now are the key
equations in structure determination.
So, either we get to measure F(R), we know that we can measure F(R) from whatever
discussion we have had now, because we can measure the intensity of the spot, which is
coming out after the diffraction, after the scattering. And scattering is now giving a spot
because of the interference rules, which we have put in on a three dimensional object.
So, the scattering is now coming on to that particular spot. So, we can measure the
intensity of the spot. The square root of the intensity will give F(R), but it gives only the |
F(R)|. So, we need to get the phases. So, we need to get the phases in, and we will put
443
these phases in one once we evaluate the phases, we should be able to get to the electron
density. And that is the way in which we approach. Of course, the other approach is to
have the electron density already known, we know where the atoms are and so on; we
can calculate the corresponding F(R).
Now, you see the advantage here. The advantage here is the following, see since we have
a problem of determining the phases experimentally, we cannot assure the accuracy
associated with the phases because of the fact that we cannot assure the accuracy of the
phases associated. The first calculation of using F(R) estimated phases to give a ρ(r) will
be very approximate. So, we may not have the correct electron density. We may have a
nearly correct electron density, but not the perfect electron density.
So, what we have to therefore do is use these two equations alternately. So, when we use
these two equations alternately, so one once we have this information of approximate
F(R) corresponding to the actual F(R). This approximate F(R), now is fitted into this
equation. The F(R) the |F(R)| is accurately measured by experiment. The approximation
comes only in the phase.
So, one once we lock into these phases, we can go back and calculate the electron
density; so the way in which we process this protocol is known as refinement; so we go
back and forth, and as we see go back and forth we can get more and more accuracies
associated with the ρ(r) calculated. And therefore, the ρ(r), the F(R) which is calculated
from the ρ(r), which is now going to be the calculated ρ(r) will be very nearly the correct
structure. And we keep on repeating until the ρ(r), which is measured and the ρ(r) which
is calculated eventually go towards 0. They may not go exactly to 0, but they go towards
0. And then we say we have determine the structure accurately.
So, therefore please remember that these two equations which are Fourier transforms of
each other or actually lifelines, during the structure determination protocol. So, we must
have an idea of how to use this, and everything originates from the fact that we have the
scattering come in from the electron density that is associated with the material in this
particular case with the crystal.
So, if there is large number of atoms inside the crystals that is one way in which the
scattering can come, of course it automatically increases the volume, because each atom
will occupy approximately 18 Å3 in a general calculation. So, if you have that kind of a
444
volume, if you depending upon the size of the molecule, and the size in which it has to
grow into the unit cell, the unit cell size changes.
So, if you have a single atom, that single atom now will essentially occupy 18 Å 3 that is a
space it generally occupy. So, here with the assumption is that atoms are spherical
number 1. And they remain spherical even inside the crystal, even after they become
parts of the molecule. And these molecules now going, and set in the unit cell. Under
these circumstances, therefore we have to be we are making an very serious
approximation, because we are neglecting the chemistry of the whole situation. The
atoms are not going to be spherical in a molecule situation, because they get deformed.
But, we are now telling that the deformation is so small as a first approximation, we can
still use these two equations to get to the final structure with respect to that ok. At this
time we can stop.
445
Lecture – 34
So, we just saw that the 2 equations which are very crucial to our structure determination
or these the structure factor expression and the electron density expression.
Now we will see what is the relation that comes in the geometry point of view because
the relation that comes from the electron density we now know because the more the
electron density the more is the scattering. Suppose you have a very big atom and a small
atom the contribution from the big atom is more to the scattering compared to that of a
small atom.
So, X rays therefore, now if one used whenever we use them; we will have the intensities
depending upon the strength of the electron density that is associated with the unit cell.
Now let us see what happens in terms of the geometries let us look at this figure; this
figure is essentially showing this that we have the incoming incident radiation. The unit
vector is associated with this is s0 we now normalize it with respect to λ; so, it becomes
independent of λ. So, we can use any λ remember that we are talking about atom sizes of
the order of 1 Å.
446
So, the wavelength can be anywhere between 0.4 to 2.5 Å; if we use that wavelength we
will be in a position to see the details associated with atoms and molecules and so on. So,
therefore, we prefer to use X rays in that range about 0.5 Å to 2.5 Å is the range of X
rays we like to use. It is recommended that since with the λ is coming up as a
normalization factor; it is recommended that we use a single wavelength.
So, at this particular stage I can introduce the way in which we generate X rays; there is
no slide or a diagram for that because I thought it is probably not necessary. Instead I
will just give a narration and that narration should be sufficient for our understanding of
how we generate X rays. You have all seen an X ray tube I suppose and you have used;
in fact, you have gone to whenever you broke a hand or a leg which most of you do you
go to a doctor and the doctors puts you through the X rays.
So, X rays are generated in a vacuum tube and these X rays are generated in such a way
that they come up from what is known as thermionic emission. That is you take an
element a filament in fact, a tungsten filament and that tungsten filament is heated with a
certain small current. As the tungsten filament is heated that is just like the light source
we have in olden days we had those incandescent lamps which still some people use, but
now we have replaced everything by LEDs.
But this incandescent lamp are used to have a filament and when once you put on the
switch when the power goes in the power is actually heating the element. The heated
element now generates luminescence and that is how you get the light. So, the light
becomes now on and as a consequence the filament glows and filament glows because
we have heated the element. So, the same heating we do and now we have the there is an
vacuum tube which encloses the filament.
Now, what really happens is in this case of the filament heated there is a certain region
around that where we generate electrons. So, electrons get emitted there is of course, the
usual heat the joule heat will be there. So, that is why if you put the light for about half
an hour the incandescent light and go and touch it your hand will burn; the reason is we
have generated a lot of Joule heat. Apart from that we have also use the concept of
luminescence which occurred in the tungsten filament.
Now, the process is now essentially generating electrons. So, these generated electrons
now are put in a large electric field. So, we put what is known as a cathode and this
447
becomes now the anode. So, the cathode is now made up of an element ok, So, that
particular element maybe I will have to write some picture for that. So, instead what I
will do is I will explain and tell you now in very simple terms that the electrons now get
accelerated, when we apply a very large field and these field could be 30 to 40 kV that is
the X ray generator you will see.
So, in the case of a typical medical X ray it is about 60 to 80 kV of power which now
accelerates these electrons. There is also a certain current which is involved in it we want
the electrons to be generated faster rate than we increase the current. In a typical
laboratory experiment where we expose the crystals that current is in mA; on the other
hand in a medical X ray the current is in A. So, it is about 60 A; so 60 A and something
like 60 to 80 kV is what is applied in a medical X ray situation.
In a when we want to expose the crystals as you know it is very small; the size of the
crystal one uses it is about 0.3 mm3; it is a very small crystal and therefore, the power is
not very high it is not required. In fact, we can calculate the power. So, if you multiply
40 by 20; it is about 800 right; so this is 800 W. So, the power that is generated will be
800 W.
So, depending on 800 to 1200 W is the power which we use in the case of when we
expose the crystals to X rays. On the other hand in the medical X ray it will be now 60
times 80 or something; so that is 4800 W its a very you know almost half a kW.
So, the X rays now what happens is that these electrons now hit the cathode which is an
element. Now this element has a certain characteristic X ray maybe we will discuss that
in a later class, when we actually see the experimental setup; when we actually see how
we set up the experiment to measure the diffraction, how do we use the detector, how do
we generate X rays at that time we will consider that.
But at this moment these X rays are accelerated to the cathode; the cathode is a particular
element. Now this element has some what we call as characteristic behaviour depending
upon where the k shell electrons are there with respect to these element we get the
intensity going up in X rays. Otherwise generally we get X rays normal X ray which is
known as the white radiation, like any other radiation you get you get a white radiation;
along with that white radiation there is a characteristic radiation corresponding to the
elimination of the kα elimination of the k shell electrons.
448
When the k shell electron goes out l shell electrons drop in and therefore, we generate the
X rays and those X rays in case of medical X rays are very strong. So, they pass through
our body there is no harm. But the X rays which we use in the laboratory is so low in the
power that we should not get ourselves exposed to X rays. Anyway let us say that we
have generated X rays now for our discussion; we will go to the experimental detail later.
So, we have the generated X rays; so this is the X rays which are generated with a certain
wavelength λ. So, we all these discussion came up because since we using this λ as the
normalization factor, it is better that we in most of these experiments we try to get the λ
to be a single value. In earlier experiments which Laue and others did which we are
going to discuss now they used white radiation.
However, it is now recommended that we use what we call as the monochromatic X rays.
So, there is a single wave length X ray. So, the incoming incident beam is normalized
with respect to that. So, this is the way in which it comes and falls on these electron
density ρ(r)dr let us say it is up here. So, if the there was no electron density this would
pass through and therefore, this is now called the incident direction; s 0/ λ is known in this
particular direction it goes out into the system.
In fact, depending upon the amount of electron densities present since the energy that is
associated with the incoming radiation is so large most of the radiation will go through
except a little bit which now we will go in the direction s/λ because of the presence of the
electron density it depends upon how much of electron density is there.
So, the amount of electron density decides how much of scattering intensity should
come; so this goes in this particular direction. Now you see that if you calculate this
angle if you call these angle as 2θ or now the vector R which is (s - s 0)/ λ is the bisector
of these two. So, there is s0/λ, s/λ this is the bisector; the vector R therefore, is the
bisector of s0/λ, s/λ. You can calculate what should be the value of R; a little bit of
trigonometry you have to do yourself. So, I will leave the choice to you how to get to this
expression.
So, we get |R| = 2sinθ /λ. So, we therefore, get the actual value of R depends upon now
the so called scattering angle; the more the scattering angle you see that the value of R it
depends upon the value of λ. So, for a given λ the value of the scattering vector depends
upon the value of θ; that means, the direction in which it goes the scattering direction in
449
which it goes, is not really indicative in this expression which only giving us the
|2sinθ/λ|. So, the direction in which it goes is following the s 0/λ, s/λ because that is the
bisector of that.
So, depending upon different values of 2θ this vector s/λ can go anywhere between this
to that; this can go from 0 to nearly 1800. Now what how does s/λ vary with respect s0/λ;
we can consider that detail by considering the possibility of having atoms put here which
we will discuss later.
So, at this moment |R| = 2sinθ/λ; what you notice immediately is therefore, this quantity
R as we have been telling already is inversely proportional to length. And therefore this
scattering vectors represent the vectors in reciprocal space. So; obviously, we have to
now define a reciprocal space which we will do in an little while from now. So, the
essentially the idea is now to see that we can get the value of the magnitude of R; the
phase angle anyway has to be determined and therefore, we still have the phase
determination remaining, but we can get the magnitude of R.
Now, we will go into the realistic situation until now we generally discuss scattering we
generally discuss scattering with respect to some electron density which is lying in some
space ρ(r)dr ; now we will say that the electron density is associated with an atom. And
more so we will make this atom a point atom; that means, suppose you consider a carbon
450
atom, there are 6 electrons associated with the carbon and of course, there is nucleus core
and valence electrons and so on.
Now, the whole thing we will condensate as though it has 6 electrons and say it is
belonging to a point x y z. Then what happens is that any scattering that will occur if we
go back here then is scattering that will occur; now we will become independent of 2θ.
Because this is now a simple point if you now have a certain volume associated with this
particular point like here small elemental volume we said; then we will have the
scattering angle changing, otherwise if it is a point atom you will just see the point atom
and the scattering will go; so it is independent of θ.
So, this is an very serious approximation, but it will allow us to imagine a situation
where we can calculate what is the quantity of what is the total information we can get
from this. In this particular case you see that the we define now because of the fact that
there is a scattering from a point atom; the point has all the electrons in it. So, therefore,
if we now calculate this quantity called δ(r) which is referred to as a delta function which
is written here this delta function will be equal to 0 everywhere else because atom is only
at one point; a single point atom we have taken the atom is at one point and let us say
this atom is located at r = 0.
So, if the atom is located at r = 0; the expression of here which we have written here ρ(r)
exp (2πir . R), there is no dependence on the phase. So, it will be ρ(r)dr . So, that will be ∫
ρ(r)dr . So, what happens therefore, is it becomes ∫ ρ(r) will be now written as δ(r)dr; this
δ will be the total number of electrons. The total number of electrons present in this
quantity ∫δ(r)dr therefore, equals to c; c is a number which represents the total number of
electrons and in case of carbon it will be 6 nitrogen it will be 7 and so on.
So, this allows us to identify the atom as a single atom single point atom and then prove
the delta function. So, it has certain special properties that it is equal to the ∫δ (r)dr = c at
r = 0, but everywhere else surround the surroundings; it is not actually a measurable
quantity it equals to 0. Now what happens if there is an atom which is situated not at r =
0, but away from r = 0.
So, we now see that F(R) can be written as ∫δ(r - r') now the phase factor comes in; how
does it come in? It comes in because of the fact now the atom is not the point atom is not
at the origin; it is removed by a quantity r'. It is located distance r' from the origin and
451
therefore, we get this expression away from the origin by r' distance. Then we have the
phase factor coming in and the delta function is at r' from the origin and then we will get
these express c which is still the total number of electrons multiplied by the phase factor
which is exp (2πi r . R').
So, this therefore, now gives us the expression for the structure factor. So, if we consider
a delta function which is at the origin; we get this expression, delta function is a special
case of F(R) essentially and therefore, we now define F(R) in the following fashion. So,
this happens if you have a point atom; now suppose we have a large number of point
atoms which is essentially the crystal. In a crystal we have let us say 100 atoms let us
consider all and each and every atom as a point atom if we consider each and every atom
as a point atom; they are located at r1, r2, r3 from the origin.
So, we sit at an origin consider r1, r2, r3, r4 etcetera the positions of the atoms then now
corresponding to each one of these points we have to add it up; the total F(R) therefore,
will be a summation and this summation will be your all the r 1, r2, ri. So, therefore, it
represents c(i); so c(i) therefore, is the total count of the electron density. So, if there are
let us say C10N2O5H13; that kind of a molecule and that are there are 4 molecules in the
unit cell.
Then c(i) is the total number of electrons of the entire unit cell and therefore, we see that
the summation is over the entire unit cell, but then the summation is overall only the
atom positions because these are point atoms. And these summation will now give us the
total number of electrons inside the crystal. And every one of them now as a phase factor
depending upon where they are sitting, if they are sitting at r 1 there is a phase factor, r2
there is a phase factor, r3 there is a phase factor and therefore, this now represents the
structure factor for a collection of point atoms.
So, what we have done here is we have considered all the atoms to have only all the
electrons and everything associated with the atom to be with a point. And this
approximation is a very very valid approximation from the point of view of calculating
the structure factors. In fact, this is what we are going to use when we determine the
structure by what are known as direct methods. So, we will assume that of course, this is
an impossible situation atoms cannot be associated with one point; they have their own
452
diameter in which they distribute the nucleus and the electron density around it and so
on.
For practical purposes to get the evaluation of the structure factor this therefore, is a
good and convenient method. So, we say that there are collection of point atoms, but
what is the reality? The reality is the following.
The reality is that there is a scattering coming from a periodic array. So, when the
scattering comes from a periodic array we have a 3 dimensional lattice; we have the a
axis, we have the b axis, we have the c axis.
Suppose we now consider the atoms to be associated with lattice points; it is not
necessary, but let us say each atom is now associated with the lattice point; then we can
generate this diagram. Now if we generate this diagram then we have to be able to
calculate what is happening with respect to the scattering.
So, any vector in this unit cell; let us say we start from the origin and let us say I take this
lattice point where let us say we have an atom in that lattice point because of the
periodicity it repeats itself. So, from this point to that point let us call this vector as r; the
little vector r a little vector r is in real space this is the actually the position vector the
position vector corresponding to that atom let us say. So, this now is r which is now
453
represented by (n1, n2, n3); 3 integers in the representing the 3 directions the a direction
the b direction and the c direction.
So, let us say if it is one unit cell you have the a, b and c values coming up here. So,
these are therefore, the integral values or rational values more appropriate. So, any vector
r therefore, can be expressed in terms of these rational numbers are integral values n 1a1 +
n2a2 + n3a3.
So, we have therefore, a lattice a lattice of points instead of point atoms being randomly
distributed as we discuss before; the point atoms are now associated with a lattice point.
When we do that we get the vector r to be n 1a1 + n2a2 + n3a3. The volume of the unit cell
is given by a1 . a2 x a3; this provides the volume of the unit cell.
At this stage we now define a set of vectors b 1, b2, b3 with very special properties; the
properties are that if we take aj and take the dot product with b j that is equal to 1. And aj .
bk = 0 whenever j ≠ k so; that means, when we have a 1 . b1; the value = 1, a1 . b2; the
value goes to 0. Now such set of vectors when we define using this expression these
vectors are supposed to be reciprocal of a1, a2, a3.
Now, it becomes; obvious to you why we are doing this? We are doing this in order to
get the representation of R which is our scattering vector and we also know it has
dimensions of reciprocal length. So, we are now trying to get how to express R given the
r in the in a 3 dimensional lattice. So, if when you have a periodic array we consider the r
and we now want to define the R. So, when we do this operation we can show that b 1 =
(a2 x a3 )/( a1 . a2 x a3); a1 . a2 x a3 by definition is volume.
So, this now becomes a2 x a3 divided by the volume similarly b2 and b3. So, when we
have these 3 vectors b1 ,b2 ,b3 following these vector expressions (a2 x a3)/V, (a3 x a1)/ V
and (a1 x a2)/V; defining the values of b1 ,b2 ,b3; then these set of vectors b 1 ,b2 ,b3 will be
able to quantify or will be able to represent these are called scattering vector R because it
defines now a lattice of vectors b1 ,b2 ,b3 just like this lattice.
Suppose we now set at an origin and now these are b1 b2 b3 we can define the values of b1
,b2 ,b3. I am not telling a1 ,a2 ,a3 = b1 ,b2 ,b3. The b1 ,b2 ,b3 is inverse of b1 ,b2 ,b3 or inverse
of a1 ,a2 ,a3 or the reciprocals of a1 ,a2 ,a3. So, if there are 3 vectors a1 ,a2 ,a3; the reciprocal
vectors that is 1/a1,1/a2, 1/a3 will be b1 ,b2 ,b3.
454
And therefore, these b1 ,b2 ,b3 in principle should describe our vector R which is (s - s0)/λ.
So, we therefore, define what is known as a reciprocal lattice which is built up of the 3
vectors b1 ,b2 ,b3. So, a1 ,a2 ,a3 are the real space vectors; b1 ,b2 ,b3 are the reciprocal space
vectors.
So, what we have done so far and this part of the discussion is to consider the atoms to
be point atoms. And then we have consider the collection of point atoms and then we
have distributed this collection of point atoms to various lattice points in real space. And
then we have defined the quantities which define now the reciprocal space such that
b1,b2,b3 are reciprocal vectors of a1,a2,a3. This will allow us to describe with the r which is
in direct space and the R which is in the reciprocal space.
So, r and R have dimensions of reciprocal length with respect to each other r is the
position vector in real space R in principle is the position vector in the reciprocal space.
So, if we have this knowledge we will see what happens when we have a realistic crystal
ok?
Now, let us go further this is what was done by Laue that is why I put his picture here
which you already saw in the earlier class because this is his contribution. Now, what is
it that is going to happen if we have a crystal, we have a lattice, we have a 1,a2,a3 defining
the vector r and b1,b2,b3 defining the reciprocal lattice vector. So, just like we describe
the r; n1,,n2, n3 as n1a2 + n2a2 + n3a3 where n1,n2,n3 are integers. Now we can describe the
455
vector R in terms of h1,h2,h3; these are again 3 integers which will define h1b1 + h2b2 +
h3b3.
So, effectively from the real space you have gone on to a reciprocal space where you are
able to express vectors in the reciprocal space. Now what we will do? We will do a little
small little mathematical operation that is to take the dot product of r . R. Now you might
worry why we are doing this r . R dot product we need this r . R dot product because you
see here you have in this expression r . R the calculation of the structure factor will be
accomplished if we get this quantity r . R ok?
So, we need to have this dot product between the real space vector and the reciprocal
lattice vector. What happens when we do this is that r . R will become h 1n1+ h2n2 + h3 n3
n1,n2,n3 are integers h1,h2,h3 are integers. So, sum of 3 sets of integers will be an integer;
so this will be an integral number. So, the structure factor F(R) which can be now
expressed see now we will take the total electrons in the unit cell outside; total number of
electrons which equals the total count of the number of electrons associated with all the
atoms in the structure total count is see.
Then summit over n1,n2,n3 sum over all values of n1,n2,n3; we then write the exp (2π i R .
r). So, this exp(2π i R . r), now is the expression which will now allow us to determine
the phase. This is the amplitude and this is the phase you notice that the amplitude is now
directly proportional to the total number of electrons. So, that why we are in a very
comfortable position; so, we now substitute R . r at this expression.
So, when we do this R . r substitution we will get exp(2π i (h 1n1 + h2n2 + h3n3)) and this
now represents every time a point in 3 dimensional space and that 3 dimensional space is
the reciprocal space. And therefore, we get c N, c is the total number of electrons N is the
total number of reciprocal lattice points and therefore, we get F(R) = c N.
Now, what is happening here? If in terms of light what is happening? So, the incoming
radiation is X ray. So, the X rays now come and fall on the crystal; the crystal now acts
as a periodic array and because it acts as a periodic array, it generates the vector the
every position vector r generates the R and therefore, the product R . r will be an integer
value. So, whenever the product R . r is an integer value you get light coming out
whenever the R . r is not an integer value; we will not get a light coming out.
456
So, what was continues scattering you know we talked about scattering occurring from
an object; it scatters in all possible directions now these directions get in sometimes in a
very loose term quantized. So, these therefore, now will occur in such a way that these
atoms therefore, will now become these sorry, these points will now become organized.
And they get organized in such a way that they have to satisfy these expression c exp(2π
i (h1n1 + h2 n2 + h3 n3)) this now being an integer.
And therefore, only when these integral values are there you will get the light out and
whenever the integral values are not there you will not get the light and this process is
the so called process of diffraction. And this has been done by some experiments in your
college days which you might recollect; you would have done it with light. You would
have done the light diffraction experiment and you were given a diffraction grating.
So, this is essentially a glass piece where there were grating marks; they were marked
with equal distances, this is essentially likely the discussion we had on the 2 slit
experiment. So, as we put more and more slits it is equivalent of putting more and more
gratings and therefore, the moment you put gratings; the light which is coming in we will
now get into go through the grating. And the diffraction grating will give the specific
directions in which light can pass through and therefore, you get a diffraction data.
This is a essentially coming because of the lattice because of the fact that we have a 3
dimensional unit cell which now repeats itself in all 3 directions we get this information.
So, it can happen only in crystals or crystalline materials. So, therefore, we get the
diffraction and this is what was discovered by Von Laue for which he was given the
Nobel Prize.
So, what did he do? How did he explain this? He explained it in this fashion if you take
R; R is now (s - s 0)/λ; we know this is the scattering vector. This is the direction of the
scattering vector it its magnitude is 2sinθ/λ we have just calculated that. Now, this now
becomes vector h and these vector h will not be all over the place; it is not a continuous
function now; it will happen only if this value is satisfied when it is h 1b1 + h2b2 + h3b3;
h1,h2,h3 being integers.
That means, these 3 conditions which are written down here (s - s0)/λ is in fact, you can
write it as H. H a 1; this is your unit cell dimension a. So, H . a 1 = h1, H . a2 = h2, H . a3 =
h3. You can replace h1,h2,h3 by h, k, l which is normally used in many text books which
457
you read and these are the so called Miller indices; a 1,a2,a3 you can replace by a,b,c
therefore, this is the expression which Laue derived.
So, the H(h1, h2, h3) represents what we call as a reciprocal lattice vector; the dimensions
of the reciprocal lattice vector is proportional to 2sinθ/λ. And therefore, the diffraction
will occur only when these Laue conditions are satisfied. So; that means, when you send
in the X ray beam on to the crystal only when all these conditions are simultaneously
satisfied then only you will get the diffraction; that means, when you have a 3
dimensional crystal the diffraction will occur only in certain given directions and those
directions now will generate a point.
And therefore, we get a reciprocal lattice point; how does it generate a point is still a
question which we will answer as we go along.
458
Lecture – 35
X Ray Scattering; Laue conditions to Bragg’s Law, Introduction to Reciprocal
lattice
So, the Laue’s conditions form the basis of all diffraction.
So, whenever Laue’s conditions are satisfied we get diffraction. So, you would have seen
the X ray diffraction patterns of crystals and you will see they come with various spots.
We also discussed that in the diagram earlier on when we discussed the way in which the
we can compare microscopy ordinary microscopy with X ray diffraction.
We showed some points which come out from the scattering experiment and these
whenever scattering experiments are done on crystalline objects if it is a 3 dimensional
crystal; we will get spots. And these spots therefore, now represent the values of h1,h2,h3
that is where the spot will occur with respect to the reciprocal lattice. And therefore, we
will be able to not only find the coordinates of the reciprocal lattice point in terms of
h1,h2,h3 which can be replaced later by h, k, l these are called miller indices.
459
Now, this is where Laue completed this work and he got the Nobel Prize in 1949. Then
the famous work of Bragg and Bragg father and son Bragg came up and what is done is
essentially we start from the Laue’s condition. So, we start with these 3 conditions ok.
And then after starting from these 3 conditions we divide the Laue conditions by h1, h2,
h3 respectively and subtract from each other. So, take the Laue condition divide by h1, h2,
h3 and you subtract. So, we get R .( a1/h1 - a2/h2) = 0.
So, whenever we have an expression like this 2 vectors a . b = 0; that means, a is

perpendicular to b; this is vector analysis. So, therefore, we see that the reciprocal lattice
vector h is perpendicular to a1/h1 - a2/h2. The same logic we can apply to the other two
sets that is R .( a2/h2 - a3/h3). So, reciprocal lattice will be perpendicular to a2/h2 - a3/h3
and the third one as 1.
So, which essentially means that the vector R is perpendicular to the plane containing a1/
h1; a2/h2 and a3/h3; so h1,h2,h3 are refer to as the miller indices. We can replace h1,h2,h3 by
h, k, l because most of the textbooks follow h, k, l but this particular textbook which I
have followed is that of Jack Dunitz, where this nomenclature has been used.
So, the fact that R is perpendicular to a1/h1, a2/h2, a3/h3 essentially tells us that if you have
now the s0/λ coming in here; you remember the picture we drew we have just
460
reorganized that picture, I have rotated the picture such that I now represent a plane and
that plane is in terms of h1,h2,h3.
So, this particular plane is characterized by the miller indices (h1 h2 h3) or (h k l). So, this
is a plane (h k l); so, s0/λ comes here and then it is scattered you remember we wrote the
angle 2θ for the s0/λ to s/λ direction. So, this angle is 2θ, but since this plane is
perpendicular to this vector R you see that these 2 is divided into θ and θ. And therefore,
we get the vector R to be perpendicular vector R is perpendicular to the plane containing
a1/h1, a2/h2, a3/h3 and that is this particular plane.
So, vector R therefore, now is being perpendicular the direction of s/λ is indicated here.
And if you look at this very carefully essentially what we see is that this plane is like a
mirror and we are actually seeing the X ray pattern because of the fact that the incoming
radiation and the scattered radiation or the diffracted radiation or essentially a reflection
from the plane.
So, what Bragg and Bragg called this as a reflection. So, the this can be approximated to
a reflection of course, radiation may go through into this crystal and so on, but from this
particular plane in the crystal we therefore, have this s/λ coming out as a reflection.
So, the diffraction now becomes reflection. So, whenever we now see a spot in a
diffraction pattern; we identify the spot with the miller index (h1 h2 h3) and we say that is
the reflection because of this particular plane which is at (h1 h2 h3); the reflection of the
incident ray is measured here. So, this is now coming as a point source from a point
source and this falls on the mirror. So, this again becomes a point and that is how we get
a collection of points.
Depending upon the various orientations of the planes we can think of inside the crystal
which we have already discussed in the previous classes. We can get diffraction say
these scattering or diffraction or reflection all three are one and the same now they come
in specific directions they obey the of course, the laws of Laue the conditions of Laue
and we get this one.
So, if we now call the spacing between planes; this placing if we call that spacing as dhkl
that is the distance of the plane where between the two planes dhkl then we can write this
expression a1 . H. Because you can now see that if you go to the Laue’s condition we are
461
essentially writing the Laue condition a1 . H is the Laue condition; this is a1 . H; this will
be equal to h1. So, a1 . H which is actually h1; a1 . H is h1; now this is divided by h 1 and |
H|.
So, this represents the distance between the planes. So, we can calculate dhkl as a1/h1 . h/|
H|. We know the |H| is 2sin θ/λ; a1 .H is h and therefore, we get an expression which is
1/ dhkl is equal to 2sin θ/λ or 2d sinθ = λ.
So, this is the so called Bragg’s law and this got the Nobel Prize in 1950. So, very
interesting it is just a mathematical jugglery a little bit of reorganization of the equations
which Laue derived, but why did it get also a Nobel Prize? It got a Nobel Prize because
of its simplicity number 1 and secondly, because of the fact that here after we can get to
the distances associated with the planes directly by measuring the scattering angle and
the wavelength.
So, if we know the θ value and λ we know the value of dhkl and dhkl is what we are
interested in terms of the spacing between planes. Now what you see here is 1/ dhkl; so
the so called vector R has still got the dimensions of reciprocal length. So, the
dimensions of reciprocal length associated with R will be associated with dhkl also. So,
the distance between the planes now is a real space quantity dhkl is therefore, a real space
quantity; this is something which is important and you should remember because this is
essentially telling that R is perpendicular to (h1 h2 h3). Now R is H the reciprocal lattice
vector and both the Laue conditions and the Bragg equation are satisfied when the
diffraction occurs.
The other advantage is that now diffraction can be very easily understood as though it is
a reflection. And the use of the Bragg equation will allow us for a direct determination of
the reciprocal lattice quantities, which was not probably possible if we has use the 3
conditions of Laue. And that way practically this equation is the one which now tells us
the entire geometry of the diffraction experiment which you have done.
So, we not only get the information about the real space geometry; we also get
information about the reciprocal space geometry and the 2 geometries are related to each
other. That means, if you have a well defined crystal with the symmetry which we have
defined over the year over the weeks then that particular symmetry is reflected; also into
462
the reciprocal lattice. We will have to examine that more carefully as we go along, but
essentially this is therefore, the one of the key equations.
In fact, along with this the key other 2 equations which we wrote earlier for the structure
factor and the electron density these 3 are the 3 key equations. Now why do you ask that
question? We asked that question because this now just gives us only the geometry
associated with the crystal; it will tell us in fact, with based on this equation we can
calculate the cell dimensions and therefore, the crystal system a, b, c α, β, γ can be
determined by using these expression. So, if we do an experiment using X rays; the
scattered X rays will now follow the rules of Laue’s conditions and as a result they
satisfy these equation.
So, using this particular equation we get information about the d spacing; we also get the
values of 1/d and 1/d we can call it as d* a vector in reciprocal space. So, we can get
solutions to a* b* c*; the values of the reciprocal lattice directly from these expression.
Once we have the reciprocal lattice information we can convert it into the direct space
information.
So, the geometry of the unit cell and its a, b, c, α, β, γ values and its corresponding
spacing of the planes inside the crystal both will be we will be able to get using this
equation, so the entire geometry of scattering can be obtained. The other 2 equations
which we discussed earlier on discuss scattering factor and the electron density represent
essentially the presence of the electron density inside this crystal.
So, a combination of this to give us the geometry and a combination of the Fourier
transform equation to give the electron density in principle therefore, should help in
solving the structure. Finding out where the atoms are what are the atoms is left to the
other 2 equations; finding out the environment in which that electron density is stuck
because of obeying the Laue’s conditions and following the Bragg’s law; we will get the
geometry that is associated with the crystal.
So, therefore, the beginning of crystallography was done by these 3 gentlemen Bragg and
Bragg and Max Von Laue. So, both of them; in fact, both groups all 3 of them deserved
therefore, the Nobel Prize in those days because this now forms the basis of structural
determination fine.
463
So, let us go further and see the more detail of this; how the diffraction is occurring how
the Bragg’s law is coming into the picture. Bragg’s law is now coming into the picture as
we see the intensities that come out from a diffraction experiment do not depend upon
the lattice points because the lattice point need not contain any atom.
So, if what in a unit cell when you are putting these atoms they form their own array. So,
suppose you have one type of atom in the crystal let us say the green type; they will form
an array of their own satisfying this as the a direction; this is the a direction and that is
the b direction. So, because of the periodicity they will repeat itself in both these
directions. So, this defines a lattice of atoms now it is no longer a lattice of points which
is a lattice of atoms. Now when we say it is a lattice of atoms, then the intensities should
come out when the scattering occurs.
Similarly, we can have another atom which defines a lattice of atoms and these 2 can be
put inside the crystal. So, inside the crystal we have now a diatomic molecule let us say.
So, this is one atom that is the other atom. So, we have a diatomic molecule defining the
crystal so; that means, now these crystal dimensions can be built in. So, this will be the a
direction of the crystal that will be the b direction of the crystal and inside that we have
the atoms located at (x1, y1, z1); (x2, y2, z2).
So, the presence of (x1, y1, z1) and (x2, y2, z2) therefore, is crucial with respect to the
location inside the crystal.
464
The location inside the crystal is with respect to the lattice points. So, the lattice points
will be now associated with this black dots which we have shown here; the black dot
here the black dot here. So, this now defines our lattice; so if these are the lattice that is
defined this is we are in the direct space, then when X rays are shown each and every
atom will follow the Bragg’s law.
So, there let us say 20 atoms associated with this one lattice point each of them will
follow the Bragg’s law. So, the red atom will also scatter at θ the blue atom will also
scatter at θ, but what you measure is the reflection which is coming from the plane. Now
each and every plane is now a point represented in this 2 dimensional diagram and
therefore, this particular point is a plane which satisfies the Bragg’s condition or the
Laue’s conditions.
So, therefore you get now from this point you have let us say the X rays coming in this
direction, the scatter from this point go in that direction. So, this is what we measure
what we measure is therefore, the black arrows, but remember the black arrows are
coming originating from the positions of the atoms. And they contribute even parallel to
the intensity that is coming out from the plane. So, when we measure the intensity from
the plane; we have a collective measure of all these atoms scattering in that particular θ
value.
465
So; obviously, the intensity will be dependent upon where these atoms are located and
that is the main issue. We have to find out where these atoms are located, but the fact
that we are getting it from the various planes which we can locate by Bragg’s law or for
that matter the Laue’s condition will tell us that it is in principle possible to find where
these atoms are depending upon the intensity now.
So, until now we dependent upon the directions to get the geometry the intensity
associated with the diffraction spot we will tell us where the atoms are; if for example, if
this red atom is associated with this lattice point then the intensity will be totally
dominated by the presence of that red atom ok. So, that is the logic which we will follow
in getting into the details of how this scattering occurs.
Here is a little example of the whole process for example, if there is an incident beam;
the Bragg equation is satisfied the diffracted beam D can be recorded as a reflection of
the incident beam I by a set of lattice points represented by (hkl). So, these now represent
the lattice points; remember when we say a (hkl) all the parallel (hkl) planes are
considered and therefore, we show the (hkl) in brackets.
Now what how each and every one of them contributes is very important because we
want to get contribution from every plane even if they are parallel to each other we will
see how that behaves in a little while from now. So, whenever the angle of incidents
satisfies the equation 2d sinθ = λ we get this scattering ok.
466
So, suppose we now consider 2 the spacing between these 2 is dhkl now we found that 1/
dhkl is related to 2sinθ/λ; how does that come about is shown in this construction. I have
left it unexplained because I want you to get an explanation for that and you already
know these from your college days, how these Bragg’s law comes up I will just give you
the idea about it. So, suppose you take this ray 1; it gets reflected from this plane and
goes off and here is a parallel plane, which is placed at a distance of dhkl.
So, now we can we can see that this second ray is up to here if we drop a point here
which goes to P. So, this drop a point here you get perpendicular here you go to Q. So,
the extra length that is travelled by this second ray is PB + BQ. Now, therefore, this ray
has travelled PB + BQ, but then the scattering occurs at an angle of θ the spacing is dhkl.
Now making use of the fact that the addition distance is PB + BQ; in principle you
should be able to find out what is dhkl because 1/ dhkl corresponds to 2sinθ/λ.
Now, you have to therefore, calculate the values of PQ and BQ in terms of the θ value.
So, you see that this is also θ that is also θ. So, you can calculate the length of PB and
length of BQ that will be equal to the dhkl. So, the additional distance that the ray travels
is equal to dhkl because these 2 rays are parallel to each other. So, this is purely a
geometrical phenomena and therefore, we should be able to find out and this satisfies the
Bragg’s condition.
So, the Bragg condition is satisfied because of the fact that λ is 2 d sinθ and how do we
get this? PB + BQ is what? You have to calculate PB + BQ. PB + BQ will give us 1/dhkl
equals 2sinθ/λ right. And therefore, that will give us the value of sinθ and this value of
sinθ take the two sinθ values and then you will get the value of 2 times dhkl how does
2dhkl come up?
So, discuss this diagram in your mind I want you to take it as a home lesson to find out
that this diagram now illustrates the Bragg’s law has 2 d sinθ = nλ. The reason why I am
not explaining it is because I want you to think about it with simple geometry how PB +
BQ add on and what are the values of PB and BQ, which essentially, now gives us this
parallel beam and that will satisfy the Bragg condition.
467
The other issue which I will discuss now and probably leave you after this is to
essentially consider 3 different planes. This will also give you a hint of how to look at
the previous diagram and arrive at 2d sinθ = nλ.
Now the hint is the following you see that you consider plane (210) alone; the plane
(210) is parallel to several planes. So, if you consider (210) only you will see that the
incident ray is here and the deflected ray goes like that, I have indicated + and y as the
directions. So, you will get a value of sinθ210.
So, since 2d sinθ = nλ you get a value of sinθ 210; d210 this value. In fact, is equal to 2 d420
because you are now making 2X2 and 1X2. So, this distance by reciprocity will become
half. So, if you double the (h k l) values by reciprocity it becomes half you have studied
it already. So, you have (420) you also have (630) which is by 3 times. So, what you see
here is the effect is to change the angle θ. the sin θ 210 is in fact, this is I made a mistake
here; I should correct it. So, you see that the angle here which is the 2θ angle or the sinθ,
the θ angle will now ensure that sin θ210 is half of 420 and one third of 630.
So, this means that if we consider the order of the reflection if we consider let us say
(210) as n = 1; (420) as n = 2 and (630) as n = 3 then the value of n comes up here. So,
we therefore, have 2d sinθ = nλ. So, this is the general expression which you find in
many textbooks; so, where does these n come from? n comes from the fact that there are
orders of reflections.
468
So, n = 1, n = 2, n = 3 and so on when we do the X ray diffraction experiment ;the X ray
diffraction experiment will individually give the reflection idea; that means, 2d sinθ = nλ
for each one of them separately because the angle θ is different from each other. So, this
essentially now will tell us that the; if this the equation 2d sinθ = nλ is the one which is
which is essentially covering the thing.
So, let me expand this and show this is what he was telling. So, (210), (420) and (630); I
am showing the diagram again because I have a not showing it in slide mode. So, I will
show it again and to confirm that d210 is 2 d210; so one half of sinθ, one third of sinθ. So,
the angle you see here is becoming different from each other. So, when you do an X ray
diffraction experiment, you can get these individually the intensities because of the
values are being different.
So, 2d sinθ = λ for all practical purposes, but when we deal with other issues like X ray
spectroscopy for example, then all the orders will appear together and therefore, we have
to use this formula 2d sin θ = nλ. So, basically n can take different values and that
represents the so called order of the reflections. So, the order of the reflections come up
as a consequence of the h and k doubling here and tripling here, which is resulting in the
variations in the sinθ angles.
So, this having seen this now we will see what happens in the diffraction condition. So,
this is a very important slide for all practical purposes I will use this slide to describe
469
how the scattering experiment is happening and how we get to the diffraction conditions
and how we get the reciprocal lattice points.
So, what we do is we put a crystal layer at the origin. So, the crystal is now having a
dimension a in this direction a dimension b in that direction. So, I am taking a 2
dimensional plane into account and this 2 dimensional plane therefore, will start from the
origin 0 here, I will also say that the reciprocal lattice origin and the direct space origin
coincide.
Now, what happens is that the value of this is a 2 dimensional example. So, d (12) is
represented in these direction that is because you see that the plane that is considered
here the second plane which we have drawn here this is (000) plane; the plane which we
have drawn here which is (120). So, one along the a direction half along the b direction.
So, this makes it (120). So, this plane is now (120) if you consider this plane, then it will
be (210) it is a parallel plane.
So, this is (120) and this is (210). So, if this is (120) this is (210) what would be the next
plane. So, these are parallel planes anyway. So, if you have (12) then this plane will be 2
this plane will be 2 along these direction and this in this direction it is 1. So, you see that
the bread analogy which we discussed. So, you essentially have the planes going through
this is the loaf of bread which is now having a orientation like this. So, we are cutting the
bread piece along the (12) direction in this case and in this case it is along the a = 2 and
b= 1 and in this case it is a = 3, b = 21/2.
So, the corresponding value will be the reciprocal value of that. So, we can read out the
corresponding planes associated with this. So, what happens is this is d (12) when we do
the scattering experiment the corresponding reciprocal point will appear at here position
and that represents d*(12) notice that, now this vector is in the reciprocal space. And since
this vector is in the reciprocal space we will therefore, generate corresponding to each
one of these parallel planes different points in the reciprocal space.
So, suppose I take this as (12) ok. So, this is now cutting at 1 and this is cutting at 1/2 the
position. So, this then will be what? It will be (24). So, this will be (24) which is not
shown here, but we can show the other example which is a essentially telling us that if
you take this point (000) and go along these direction by 1 unit ok; then you will have
470
the (10) a* coming up here and the b direction will be representing b*; it depends upon
what is the angle of β here the values between a and b which is γ.
So, the value of γ now decides what should be the value between a* and b*. So, the value
between a* and b* will be 180 - γ; so, because this whole thing has to be 360 degrees. So,
you have therefore, the γ value here and 180 - γ will define the a*b* direction. And once
we have the a*b* direction you will have the planes will be identified by points that is
how you get the diffraction in terms of spots. The spots which you see from the
diffraction experiments correspond to these reciprocal lattice points.
So, one once we have the dimensions of a, b and c in principle we should be able to find
out and index these spots. This process of finding out what are the values corresponding
to h and k in these 2 dimensional plane is referred to as indexing. So, each and every
point therefore, is indexed with respect to the value of h and k. So, you see here that this
particular point which is indexed is now corresponding to (12).
Now, please notice that these (12) is the same as (12) up there. So, there is no change
there so; that means, the value of h and k will still remain the same whether in the direct
space or in the reciprocal space. It is essentially the way in which the reciprocal space
develops with a value of which is proportional to 1/distance. So, in a in other words it
generates the reciprocal lattice point. So, planes in the direct space will generate points in
the reciprocal space and therefore, we call them as the reciprocal lattice points and each
and every reciprocal point therefore, correspond to the plane in the crystal.
So, the planes in the crystals will now, therefore, can be identified in the direct space by
looking at spots and a indexing those spots. So, when I index this spot as (12); that
means, I am referring to the plane (12) in the real crystal. So, that way we can do these
values find all these values and these values now represent the reciprocal space and the
diffraction pattern. So, this is the direct space and the crystal structure. So, if there is an
atom now sitting here; if there is an atom now sitting here its contribution to this plane
needs to be calculated in order to find the intensity of this spot.
So, if there is an atom sitting here for example, very close to (12) how are we going to
calculate the intensity that can be associated with this part and that is something we will
have to evaluate in the coming classes. So, at this time we have a few questions left
unanswered which you have to answer. One is the answer to this diagram and as a home
471
exercise you can work it out how to arrive at the formula 2dhkl, 2d sinθ = λ from these
expression from these diagram.
So, you have to find out the perpendicular distance from this point to this perpendicular
distance from this point to this. So, the path difference is now PB + BQ. So, the it has
traveled by that extra path there is a corresponding phase difference. The phase
difference will now introduce what? It will tell us at what θ angle the scattering is
occurring and therefore, the sinθ values can be calculated one once we know the value of
θ and if we know the sinθ value; then we can possibly use 2sinθ/λ which will give us the
value of dhkl.
So, dhkl is 1/dhkl sorry 1/dhkl will be 2sinθ/λ and that is how we calculate these particular
value. So, what I want you to do it as a home exercise; the second one is a is an
understanding of how 2d sinθ, nλ comes up, n is the order of the reflection these 3
diagrams will help you in principle to identify how these diagrams are coming and what
are the angles inter axial angles between them between I and D and that will tell us the
relationships and therefore, we get the different values of n going from 1,2,3 and so on.
So, essentially what we have just a quick revision of what we did until now we started
with the scattering from a periodic array. And then we derived the expression using their
Laue conditions and having done the Laue conditions we derived the Bragg’s law. And
having done the Bragg’s law we try to understand in terms of the diffraction as a
reflection from a plane.
And then we identified that the scattering is coming from atoms and not from the planes,
but then the atoms are now associated with the planes. We will do in the coming class
how this association with the plane will generate the intensities how what is the
contribution of each one of these atoms to this particular lattice point.
And then we saw the 2d sinθ = nλ and then 2d sinθ = n λ and then we are now trying to
get to an understanding of the reciprocal lattice points which come from the direct space
lattice. This is something we will discuss again and then see how we can arrive at the
Bragg’s law in reciprocal space. Because we have the recording in reciprocal space
whatever we get from a diffraction experiment gives us parts in reciprocal space.
472
And so the geometry associated with this can be identified one once we index this spots
and once we have this indexation done, then in principle we should therefore, see how
Bragg’s law behaves in the reciprocal space and that is the material for our coming
classes.
Thank you.
473
Prof. T.N. Guru Row
Lecture – 36
Bragg’s Law in Reciprocal Space 1
We have been looking at the Bragg condition. A few issues with Bragg conditions still
left undiscussed. I left this diagram for you to analyse, but I thought that maybe I should
explain it a bit more.
Other than that this new diagram I have put in. This new diagram will tell you that if you
consider a this that is say is the plane from which the diffraction or the reflection of the
incoming radiation is giving us the diffracted beam. Now, essentially the diffraction from
all practical points of view under the Braggs law we consider it as a reflection; that
means, the angle of incidence and the angle of diffraction they are one and the same and
because of the fact that this distance AB in fact, AB will represent now the wave front
associated with the incoming, that is the incident radiation and CD will represent the
wave front which is going out after this scattering.
So, that would mean that this angle is 90o and that angle is 90o. So, which essentially tells
us that if you take any point on the plane take any plane and take any point on the plane
the diffraction condition is still the same; that means, if you see this angle ACB and
474
angle CAD these two angles are one and the same which happens to be equal to θ and
because these two angles are equal to θ by construction it means that all this points any
point in this plane will scatter with the same phase. So, if you assume that there are
different kinds of atoms in different parts of a given phase all those atoms will scatter in
phase.
Suppose, you consider this as the plane passing through the origin and that along
different directions different distances different atoms lie on the on that particular plane,
as far as the phase associated with this plane is concerned it is always 0 because it is
passing through the origin. And if atom sit on this particular plane the phase that is
scattered by these atoms will also be 0. So, this is a very important point that this scatter
with the same phase if you consider any point on the plane.
So, which means to say that in a given plane any plane for that matter it need not pass
through the origin any plane there may be several atoms sitting. This is again taking you
back to the bread analogy where we take a slice out, there may be several pieces of
different things which we have added in the piece of bread and therefore, different
electron different electron densities associated with different atoms may coincide with
that particular plane. All these now will scatter in a with a single phase and this is a very
important phenomena because then we need not have to worry about the what are the
contents of that plane as for as phase determination is concerned.
Suppose, in this particular plane there is a carbon there is a nitrogen, there is a bromine,
there is a iodine and things like that which parts of it of course and all these electron
density therefore, will all scatter with the same phase ɸ associated with the plane. So, this
factor will help us in actually determining the structural details in a later point. So, what
we have to take home the take home lesson from this particular diagram is that these
scattering is associated with the same phase and that particular phase angle is now with
respect to the plane.
So, suppose these atoms are not in this plane and somewhere else the contribution of this
plane to the scattering will still have the same phase. So, the phase angle is therefore, a
property associated with the plane and the contents of the plane now whatever be the
contents the phase angle will still be the same and this is the phenomena which in fact,
will come in handy when we toss determine the structure later date.
475
Now, regarding this diagram which we already discussed a bit in a last class we see that
the wavefront again here in this case is a is shown as AP. So, this difference this defines
the incoming unit vector S0 and this the vector S which is the scattering vector and the
what is represented here are the two different plane separated by a distance dhkl.
So, whatever are the atoms on this plane they will all scatter in phase whatever are the
atoms in this plane they will scatter with that particular phase and if the first we consider
the first plane which is out of the; out of the origin suppose A is the origin position then
the first plane out of the origin will scatter with a path difference of λ and therefore, a
phase difference of π. So, that defines in fact, the Braggs law that is because of the fact
that this little geometry construction will tell us.
So, if you look at AP and AQ this AP and AQ are plane wave fronts the path difference
is that if you consider this particular wave on the second wave which is going up here,
the path difference will be additional PB+BQ. So, the additional path the second wave is
travelling is PB+BQ. Now, the value of PB+BQ can be calculated as based on this
diagram because this is the angle of incidence this is the angle of diffraction and
therefore, this is the angle of incidence and this is the angle of diffraction.
These two triangles are similar triangles. So, APB and ABQ are similar triangles in as a
property of the similar triangle is that the external angle will be equal to the internal
angle. So, these two angles will also be θ and therefore, this will be one side one d sinθ
the other side another d sinθ. So, it accounts for 2d sinθ. So, PB is d sin θ, BQ is also d
sinθ. So, PB+BQ is 2d sinθ.
So, in order for the reef reinforcement occur in order this two waves now overlap with
each other they should be a whole number; in other words if the diffraction has to take
place these two have to be a whole number of n number of wavelength. So, the first one
will be 1λ, second one plane will be 2λ, third plane will be 3λ and so on. So, therefore,
they satisfy this equation 2d sinθ is equal to nλ. So, you get the Braggs equation, 2d sinθ
is equal to nλ that is the condition which is derived from this particular diagram.
So, therefore, these two diagrams tell us two major things one is that when there are
points any point on this given plane will always scatter in the same phase, irrespective of
where the point is on that particular plane and when we consider parallel planes like this
476
with are separated, but a distance of dhkl we get the condition that 2d sinθ is nλ and that is
the way in which the Braggs law appears.
To see what is the value what is the role of this n which is referred to as the order of the
reflection we have already discussed this, but we will again look at it from the point of
view of how the angle changes with respect to the variation in the values of h, k and l.
Suppose, this hkl is 210 the value of 210 if you consider the value of 2 times d 420 where
you have now taken 2h, 2k and of course, you can call it 2l, but it 0. 3 times d 630 that is
the 3 times. So, we therefore, see that the value of the d 210; this is the value of d 210 the
spacing between these two planes is twice 2 times that of the plane distance between
these two and 3 times that of the distance between these two and therefore, d 210 is equal
to 2 times d 420 equals 3 times d 630.
So, as a consequence what will happen to the sinθ value? The sinθ value which is again
following this formula 2d sinθ = n λ, we see that sin θ 210 will be half that of sin θ 420 as is
illustrated here; this is the θ angle, this is the θ angle the angle of the of the between the
incident and the diffracted. So, this is the incident angle θ; this is the diffracted angle θ.
So, the angle between these two therefore, will be 2 θ.
So, this I and D now come at an certain angle sin θ 210. This now becomes one half of that
angle and this now becomes one third of that angle and if you consider the fourth order
477
reflection it will be one fourth of that angle. So, the order of the reflection therefore, tells
us what should be the angle that is at which the diffraction occurs. So, at which the
incident ray comes and falls on the plane and these are of course, all parallel set of
planes.
So, from the X-ray diffraction point of view and when we do an X-ray measurement
even though this is a general equation 2d sinθ is n λ your n value can be different from
each other when the diffraction spots come. So, when you have the reciprocal lattice
points these reciprocal lattice points can have different values of n. So, therefore, we get
2d sinθ equals λ for the first case, 2d sinθ equals 2 times λ, 2d sinθ 3 times λ and so on.
So, as a result the position of individual reciprocal lattice points will already put a
formula which is 2d sinθ equals λ.
So, the order of the reflections will appear in a diffraction geometry in such a way that
each of them will represent a separate position for the reciprocal lattice point because
this now will occur with this point, this will that point that will occur at that point and
that depends upon the angle between the incident and the diffracted rays which is given
in terms of this values. So, therefore, in general 2d sinθ is n λ, but for all practical
purposes in a diffraction experiment we take it n equals 1. So, we therefore, get 2d sinθ
equals λ which is what we will be using in the entire analysis associated with the
reciprocal lattice.
478
Now, we discussed a little bit of this diagram yesterday, but we will re-do that in the
sense that we have to now consider the relationship between the direct space and the
crystal structure and the reciprocal space on the diffraction pattern. Now, this is a factor
which is of crucial importance because what we will observe from a diffraction
experiment are these.
What we will observe from a diffraction experiment are these reciprocal lattice points,
so, when you put crystal in the X ray beam and diffraction occurs you will get spots
depending upon the quality of the single crystal and this particular spots will tell us the
points of the reciprocal space lattice. So, this therefore, defines the reciprocal lattice.
Now, what is the relationship between the direct lattice which is up here which is our
unit cell defined suppose you take this as a 2-dimensional crystal you see that this is
represented by the a value and this is represented by the b value therefore, this red square
here will be the unit cell. Now, this unit cell repeats itself because it is a crystal in all
both the directions. So, this therefore, now defines the crystal space. So, this is the lattice
representing our crystal.
So, the angle between a and b which is γ will also decide the orientation of these a and b
with respect to each other. So, in other wise this angle which is here could be γ. So, that
is because of the fact that this two are identical I can show the γ here, I can show γ here, I
can show γ here, I can also show it here. So, this you know already.
So, we have therefore, a direct space and the corresponding crystal structures. So, the
atoms are all lying here in this part of the diagram. So, there here wherever depending
upon where the atom wants to be and where the molecular structure develops or the ionic
structure develops and so on. The atoms are connected to each other as we discussed
yesterday, these scattering is done basically by the atoms; that means, the electrons
associated with the atoms. So, X-rays when they fall on this crystal they will be
diffracting based on where the electron density is found. So, the larger is the electron
density the more intense will be the diffracted spot.
But, the observations which we are going to make is with respect to the individual planes
which are defined in the direct space. Now, what are these individuals space planes we
define in the direct space? We will define for example, a plane which is not going
through the origin and therefore, as we discussed all the atoms which may lie on this will
479
scatter in phase in the phase we the scatter with the same phase a 1 with 0 and that
particular case ϕ will be 0 this the phase angle will be 0.
The plane which is coming out from here we have we can consider several planes which
come out from here, but what we have done is a set of parallel planes. So, we have drawn
here the plane which is now represented by a line because we are shown it in 2-
dimensions. So, plane is perpendicular to us. So, if you now consider this particular
plane a in two-dimensions it can be represented as 1,2. Now, why 1,2 because we see
that this blue line intersects these red unit cell at 1 unit along a and half unit along b.
So, we said already that the indexing is done in such a way that if there is a cut of half it
will be actually 2 in terms of the identification. So, therefore, d 12 go back to the bread
analogy where we took the central slice and said it is at half position along a, but that
would mean that it is a 200 reflection if we are considering the reflections along the a
direction. So, the same analogy brings up here the value which is the distance the black
line here indicates the distance here indicates the where position of d 12 the this the length
of d12 that is the value of the d value of 12; 12 being the plane which we are representing
here.
So, if this is the 12 plane the next parallel plane will be what? The next parallel plane
will be 2 and 1. So, this is 2 and 1, this is sorry this is 1 and 2 and this will be; this will
be 2 and 1 and this will be 3 and find out. Now, you have gone into the next unit cell up
here if you consider the intersection, but it is a parallel plane to this one. So, all parallel
lines with respect to 12; so, 12, 1 along the a direction and 2 along the b direction that
becomes half of that. So, this will therefore, be 13 right 23, 33 sorry. See, 1,2 is up here
the parallel plane to that will be what? This will be 2,1, the next parallel plane will be 3
something.
Now, if we consider this as 1, this will be now 3/2 one and half and therefore, this
intersection point will be what? So, if this is 3 that will be what? Find out and then you
see that these are now a set of parallel planes ok. These set of parallel planes let us go
further these set of parallel planes are represented by this vector in this case the plane 1 2
is represented by d12, the distance of 1 2.
Now, the corresponding reciprocal lattice vectors will occur with respect to a, 1/a this is
the vector direction 1/a vector direction represents a* and 1/b vector direction represents
480
b* this is now the value γ and therefore, this will be the value γ* to define the 01 and the
10. So, the reciprocal lattice points are now represented as 10 and 01 with respect to the
reciprocal lattice and this angle will become γ*; γ* is 180-γ where γ is this angle.
Suppose this is angle is exactly 90o this will also will be 90o. So, a will be 1/a, b will be
1/b. Here also as far as vector directions are concerned it will be 1/a, 1/b.
So, these therefore, now represents the a* b* axis and this is 10 in this direction and 01
this direction. So, if we now take all the reciprocal lattice points along the b axis b* axis
we will have 01, 02, 03 and so on and the corresponding values in the a direction are 10,
20, 30, 40. We will therefore, get the intersection of 1 and 1 as 11 then 1 and 2 as 21 and
so on. So, if we are now looking at a plane d 12 the corresponding reciprocal lattice vector
will be now we have to look for 12 that is one intersection along a 1 along a and 2 along
b.
Now, this what this represent is the reciprocal point that is d* 12. If this is the d12 we
represent this as d*12. So, this therefore, now defines the diffraction pattern, it also
defines the reciprocal space. So, if you sit down and study this diagram a little carefully
you will definitely see that the one to one relationship between the direct space and the
reciprocal space. And that one to one relationship will therefore, tell us that if this is a
lattice if this is the red one is a lattice, then the collection of all this blue dots should also
be a lattice and therefore, this we call as the direct lattice this now becomes the so called
reciprocal lattice.
So, in the reciprocal lattice we have two vectors a* and b* with an angle of γ* defining
this particular nature of this lattice. So, the distribution of the diffraction spots therefore,
should come at these blue points. So, keeping this in mind we will see how we can use
this idea of reciprocal as because we always see the reciprocal lattice in an experiment.
So, when we do the experiment we put an X-ray we send an X-ray beam onto the crystal
a collimated X-ray beam onto the crystal. The crystal now diffracts, so, it will follow the
Bragg’s law and therefore, we have an incident ray and a diffracted ray according to the
Bragg’s law satisfaction 2d sinθ = n λ and therefore, we will get to positions which are
now the reciprocal lattice positions and therefore, we will get spots.
Now, these parts which we will get in a single crystal are therefore, each and every spot
now is a representation of a plane in the direct space. So, every plane in the direct space
481
gets represented by a point in the reciprocal space and this a very big advantage because
we do not have to look at the plane we look at only the reciprocal lattice point. We also
know another factor that every plane therefore, if any atoms are lie on that particular
plane they will all correspond to the same phase and therefore, when we get this
reciprocal lattice point let us say 21 here; 21 will contain information of all the atoms in
that particular plane in the direct lattice. And therefore, 21 is something which is one
once we get the coordinates of 21 it is important to see what is the intensity that is
associated with 21.
So, before we go to the analysis of the intensity we have to do one other way in which
we can look at the Braggs law. Whatever Braggs law we derived earlier is in the direct
space, we took the parallel planes, we did the construction then we found the phase angle
and so on and showed it is 2d sinθ equals to λ or nλ where n is the order of the reflection
on the other hand when we do a diffraction experiment what we see is the reciprocal
lattice image; that means, we see the spots. So, it is always good to see whether we can
actually think of deriving the Braggs law in that space itself.
So, in other words, can we have a Braggs law in reciprocal space can we derive Braggs
law in reciprocal space? This was thought process which was put to develop what is
known as the Ewald construction. So, it was developed by a person by name Ewald. So,
it is known as the Ewald construction. You can directly visualize the diffraction
conditions by means of which reciprocal lattice points come and that is how it is done
which is shown in the next view graph.
482
Here you see something which we have to do a construction. What you see are the points
here which are the reciprocal lattice points. So, I have drawn the reciprocal lattice. So,
the reciprocal lattice is similar to the reciprocal lattice shown here. So, all these points
which come we know now how they are coming and knowing that how they are coming
we have taken these points and we have transferred it on to the plane.
Now, therefore, this is now a planar representation of the reciprocal lattice points. Then
what we do is we do what is known as a Ewald construction. So, each plane is a
reciprocal lattice point as we saw just now. So, what we do is we just take a circle, we
draw a circle of radius 1/λ; λ is the wavelength. Remember that the reciprocal lattice is
represented in inverse of length the dimensions of inverse of length.
So, we therefore, take a circle and draw the circle of radius 1/λ. That particular circle we
can draw with some centre here and you see that this is the circle which we have drawn.
This is a circle drawn with the radius 1/λ. So, if you consider the centre as here this to
this distance or that the distance is 1/λ.
Now, we define on this particular circle this circle can be anywhere ok, it can be
anywhere in 3-dimensional in the 2-dimensional space we have drawn here it is 3-
dimensional because we are now going to eventually change the circle in a sphere, but at
this moment this is a 2-dimensional diagram and so we have a circle. Now, this circle
can be drawn anywhere, but it should have a diameter of 2/λ or a radius of 1/λ. So, this
483
radius of 1/λ is with respect to some origin. So, we take this the one of the reciprocal
lattice points to be an origin. So, if you take this as a origin of the reciprocal lattice point
(0, 0) now this is the circle with a 1/λ radius which is drawn in such a way that the circle
passes through our (0, 0).
So, remember the circle is passing through our (0, 0) on the circumference. So, this is the
circumference of our 1/λ sphere. So, there is a centre up here. Now, let us say we send an
X-ray beam from the top. The X-ray beam from the top now comes and falls on this
particular origin which is a reciprocal lattice point so, it is scatters. Now what it
therefore, tells us is that any point which intersects with the circumference of the circle
because we set the circle is arbitrary it can be put anywhere. So, any point which
intersects with the surface of this particular circle will give us therefore, the diffraction
condition that satisfied. So, can we use this method to find out what is the diffraction
condition that has to be satisfied.
So, therefore, we see here in this particular example we have taken the reciprocal lattice
points and we see that this is the vector b1, the reciprocal lattice vector and this is the
vector b2 which is the second vector. So, b1 and b2 now represent the reciprocal lattice
vectors, reciprocal lattice dimensions. So, b1 and b2 are the reciprocal lattice cell
dimensions or reciprocal cell dimensions.
So, having seen b1, b2 being representing this is let us say one direction this is the other
direction. You see we go up in this direction drawing these points on, these are all the
points which are lying with their origin here and these are all the points which lie with
the origin with respect to the b2. So, we have an we have a b1 here and a b2 that is
defined, so, this is the origin.
Now, you see in the diagram which I have drawn it so, happens that the point which is
now 2 along 1 along the; 1 along the one direction that is this point this is the if you
consider b1 the and this is the point 1 along the b direction. So, we see that this coordinate
therefore, which lies on this is 2 this one in this case of b1 and b2 is 1 and 2. So, there are
2, this is a second position; that means, this vector H is represented by (1, 2).
So, since this vector is this is the crystallographic directions in some in some sense, so, it
is (1, 2). So, the diffracted beam is now shown as S in that direction. So, therefore, what
happens is this is the incident beam, this is the diffracted, beam you remember the
484
diagram with we draw the diagram will be this will be the angle 2θ between the incident
beam and the diffracted beam. Now, if you take the bisector of that we already know that
should be the direction of what we call as the scattering vector. So, this is written as the
trace of plane normal to H. So, this will be the direction of this scattering vector.
So, we therefore, have a θ here and a θ there and this is the centre of the circle and we
have the incident beam and the diffracted beam. Now, what happens is that as you see
here the there are several reciprocal lattice points which on this grid, but none of them
are intersecting except the one which you are taken as an example the reason is obvious
because we are keeping the reciprocal lattice fixed; that means, we are keeping the
crystal in one position and then sending in the X-ray beam.
So, if we keep crystal in one position and then send an X-ray beam very often there not
there may be no diffraction coming no reflection are observed or we may observe a
reflection by chance like what we have seen here and we will see this particular
reflection we do not see any other reflection; that means, to say that whenever the
reciprocal lattice intersects with the Ewald sphere on the surface. Now, instead of the
circle I am going to call it a sphere because we are now going into 3-dimensions. So, we
call it as the Ewald sphere.
So, anywhere anytime an Ewald sphere intersects with a reciprocal lattice point the
surface of the Ewald sphere then can we get the diffraction condition satisfied and this
diffraction condition is satisfied if H lies on the surface and if H lies on the surface of
this sphere then the scattering angle we know it depends on the length of H the Laue
condition; the length of H, S-S0/λ
So, the length of H therefore, is equal to 2sinθ/λ in magnitude. We have shown this
magnitude of H is 2sinθ/λ, so, it is 2sinθ/λ. Now, what is 2sinθ/λ? It is 1/d . So,
hkl
therefore, you equate these two, you will get the Braggs law 2d sinθ =λ
So, therefore, the Ewald sphere has a radius of 1/λ. So, what is the catch here? The catch
here is the following that we have kept the crystal in a single position sent in the X-ray
beam, by chance it so happens that in this diagram which we have shown purposely there
is a point which intersects with the surface of the Ewald circle or the Ewald sphere and
therefore, it satisfies the Bragg condition. See in order to satisfy the Bragg conditions we
have to have this equation satisfied that is because of the fact that the vector H which is
485
the reciprocal lattice vector should have a magnitude which is equal to 2sinθ/λ then and
only then we get diffraction.
So, we see here that in this entire diagram we have shown only one possibility of a
diffraction point. Then how do we get all these reciprocal lattice points to diffract? The
solution is very simple, you rotate the crystal. Now, as you rotate the crystal this
reciprocal lattice itself will start rotating. So, as the reciprocal lattice rotates different
points in the reciprocal lattice can come and intersect with the Ewald sphere here, the
surface of the Ewald sphere.
So, as and when any of these for example, here there is a nearest intersection it may
slightly turn around and then next point will be intersecting here this will intersect and
like that. So, all these points will start intersecting. Now, what is the limiting condition
for that? The limiting condition is that now this becomes the radius, 2/λ becomes the
radius and therefore, this circle which we are drawn outside or the sphere which we can
draw outside in 3-dimensions which is referred to as the limiting sphere, so, this has a
radius of 2/λ. And this 2/λ is now the limitation. So, all those reciprocal lattice points
which lie within the 2/λ, have a chance to intersect with the Ewald sphere which has a
radius of 1/λ.
So, the radius of the limiting sphere is 2/λ. So, the reciprocal lattice points which lie
within that particular sphere have a chance whenever when the rotation takes place the
360 rotation of the crystal will bring all these points which are shown within the limiting
sphere to intersect with the Ewald sphere and whenever there is an intersection with the
Ewald sphere you get diffraction. So, it may so happen that in as you rotate the crystal.
There may be positions of more than 3 or 4 reflections satisfying the Braggs condition
then you at that particular angle of rotation of the crystal you will get more than one
reflection. So, it is not necessary that one at a time should come, you can get any number
of reflections at a given angle theta in different directions see the directions are different
this point and let us say this point were to intersect. We will get one diffraction in this
direction another diffraction in that direction and so on, so, it goes in that particular
fashion.
So, what we therefore, do is to see that this set of points which will now present or which
are present in the limiting sphere will have a chance to intersect with the Ewald sphere
486
and this is a point which therefore, allows us to do look at Braggs line reciprocal space.
So, we now know what under what co conditions and under what circumstances a crystal
can give diffraction. So, we have a crystal, we get the diffraction spots and these
diffraction spots will follow this particular equation which is the Braggs equation. So, I
think at this level we have now just understood the Braggs law, both in the direct space
as well as in the reciprocal space.
487
Prof. T.N. Guru Row
Lecture – 37
Bragg’s Law in Reciprocal Space 2
So, we just see here that we can do Braggs law in reciprocal space the way in which the
number of reflections appear. So, it is always a wonder in many cases that how the how
come when we when we start rotating the crystal we get more and more reflections. And,
this particular construction helps us to understand how this really happens and how we
can get to the diffraction condition.
The one of the major requirements is that the scattering angle depends on the length of H
which is 2sin θ/λ. So, this is a major factor which we should remember. So, as the θ
angle changes the length of H will change and since length of H will change whenever it
satisfies 1/dhkl in this particular unit cell; so, you have a unit cell you have a b c α β γ and
then you have the d values calculated from a b c α β γ. So, those d values 1/dhkl whenever
that is equated you will get a reflection.
So, this now, therefore, is can be actually shown in such a way that as we rotate the
crystal different parts of this limiting sphere points they will all come and intersect with
the Ewald sphere and whenever the intersection is with the Ewald sphere, whenever the
488
intersection is on the surface of the Ewald sphere you will get to the satisfaction of
Braggs law and that is why our diffraction pattern is now looking like this.
We get a series of spots and these spots therefore, have an identity now. The identity is
with respect to the scattering angle and the scattering angle now decides the value of H
and H is 2sin θ/λ. So, we can therefore, index these parts in terms of 1/d hkl which we of
course, can be calling it has d*hkl. So, we have therefore, d in the real space that is the
spacing between planes. Now, it get represented as an identity of a reciprocal lattice
point and that reciprocal point will now represent the d*or 1/d hkl in the reciprocal space,
right.
489
Now, we have got to an understanding of this we will have now have to see how these
spots appear? What is it that is making this spots to appear and why one should be strong
one should be weak and so on. Most of us have seen the single crystal diffraction spots
or the single crystal diffraction pictures and if you go to diffractometer you will see that
being recorded on the CCD camera where you will see them as spots.
Now, this spots can be of different intensities one is strong, another is weak, something
else is hardly visible and so on. Even though the geometry is defined, in other words the
1/dhkl is defined when once you have a*b*c* their reciprocal lattice vectors when once
you have a, b, c you have a* b* c*. So, you have 1/dhkl which is the collection of d*
values. So, we know where the positions of d* will come depending upon the Braggs
law. But, whether the d* value which has come at that particular point has an intensity or
not depends upon what is it that is actually causing this diffraction and this now we will I
will take you through an exercise where we now understand what is it that is giving us
this diffraction.
What is it that is giving us the diffraction it is the interaction between the X-ray beam
and the electron density that is associated with that particular position x, y, z in real
space. Let me repeat, the position of the atom therefore, the atom now is consisting of
electrons as we know. So, it is the electron density which is associated with a point x, y, z
which is responsible for the X-rays to do the diffraction. However, the d hkl the 1/dhkl in
490
particular that is the positions where the reciprocal lattice points should appear is a
geometrical calculation.
So, because it depends only upon the a* b* c* values. So, the geometrically we can
identify the positions where the diffraction should occur. However, the intensity of that
depends upon how much of electron density that can be associated with that reciprocal
lattice point. So, if the reciprocal lattice point is now associated with a large amount of
electron density, then you will get the diffraction pattern. Let me draw a little picture to
show you what I mean.
Suppose, there is a electron density something like this, this region where there is a lot of
electron density and this now lies on a plane which is passing through that like that. So,
this could be a one plane, it could be another plane there may be another plane going like
that another plane which is going through this and all these planes now contain these
electron density.
So, if you now calculate the intensity or if you see now where these spots should come it
is already defined by geometry, but the intensity that is associated with the diffraction
on; in other words the energy that comes out after scattering depends upon how much of
electron density is associated with them. So, these three planes all these three planes will
have a large value for the intensity; that means, these will be brighter spots compared to
the other weaker spots.
So, if that is the logic that is to be followed we have to find out how this is all happening
it is all happening because the amplitude of the wave which comes at very long distances
at R, let me go back to the other mode. The amplitude of the wave at distance R which is
very far away in terms of the λ value, if the amplitude A of that λ wavelength is incident
on the electron with a mass m and a charge e, then the amplitude will be this value which
depends upon an expression of this kind. We have to understand what it all means.
So, you have an electron with a charge e and mass m. Now, these this one particular
there is no one single electron, there are several electrons in the unit cell, right we
associated with a molecule atom position and so on. So, the charge is e and the mass is m
the amplitude A that A0 which we would like to have from the incident radiation is
already know us.
491
So, the incident X radiation has an amplitude A0. It gets modified after scattering and we
observe this amplitude of the wave at a sufficiently large distance R is very much greater
than λ. Remember I am using this vector R to identify that we are observing this
amplitude corresponding to the vector R which is the scattering vector, which can be
replacing in a single crystal by the H which is our scattering vector which is essentially
2sin θ/λ. So, it has to satisfy the Bragg condition.
So, since it satisfies the Bragg condition we get the spot. The amplitude of the spot
depends on this quantity and this particular quantity is now A which is the incoming
amplitude divided by R. So, if you just send an X-ray beam light intensity for example,
light coming from a source in front of us; so, the light will be becoming weaker and
weaker as we go to very large distances. In fact, it vary such the square of the distance.
So, therefore, when we take it at a very large distance anyway the amplitude gets reduced
by that a particular amount proportional to 1 over R. So, amplitude is already reduced.
The fact that it is the electron which is involved in this scattering, we have this e 2/mc2
factor coming in and then this factor (1+ cos 2 2θ) / 2 where 2θ is the scattering angle you
see this is very crucial this 2θ scattering angle decides where your H vector should come.
And therefore, this factor is the one which modifies the amplitude and therefore, you will
get different intensities at different positions. So, we therefore, have the amplitude of the
value equal to this.
Now, what is it that is causing this amplitude? what is the electron density, right and
electron density in our case when we discuss atoms and molecules and collection of
molecules in the unit cell depend upon with what atoms these electrons are associated
with.
Suppose, we have a structure which as a carbon, nitrogen, oxygen on one hand or a

structure with a germanium or even higher elements like zirconium or for that matter
even uranium containing compound. So, the electron density associated with these atoms
there is a certain size of these atoms which of course, is quantum mechanical
calculatable. So, this quantity is also coming from the quantum mechanics evaluation.
So, we are not going into the detail of it.
However, it depends upon a factor which is now going to be very very crucial for our
calculations and that we call it as the atomic scattering factor. So, if there is a carbon
492
atom, carbon atom will have six electrons. So, at the scattering angle 0, all these six
electrons will contribute. So, it is equal to the atomic number. So, if the scattering angle
is 0 and the carbon atom is situated there all the electrons will contribute to that one and
therefore, it is proportional to 6. I am using the word proportional to 6 because it depends
upon where you define your 0 in your crystal system, where is your reciprocal lattice
origin. You saw it in the previous diagram it is variable, we can decide where the origin
is depending upon where we would like to take the reciprocal lattice point origin and
therefore, the atomic scattering factor is proportional to Z the atomic number.
So, depending upon what is the size of that particular atom whether it is carbon or
bromine or whatever the number of electrons associated with that particular point will be
proportional to Z. But, it so happens that it also with varies with respect to sin θ /λ where
in other words it varies with respect to this scattering angle. We already have the
amplitude variability shown here for a free electron. Now the amplitude variability is
now to be shown for electrons which are associated with the atom. So, we have a
position of the atom, we have the core of the atom, we have the surrounding valence
electrons and so on, but they are associated now with the atom.
And, therefore, we have to define the atomic scattering factor in a proper way and this is
the definition. Amplitude scattered by an atom in a particular direction the two amplitude
scattered by a free classical electron in the same direction. So, you consider a free
electron and consider the collection of electrons associated with that atom and then the
atomic scattering factor is the amplitude that is scattered. So, what happens therefore is,
if this is the free electron component we add we multiply it by the factor f which is
known as the atomic scattering factor.
So, this is the amplitude of the X radiation scattered by an atom. It is as I said ideally Z
because all the electrons should contribute if you just put one atom and send an X-ray
beam and this particular atom is now in this in a position like this where this scattering
angle is not seen 0 then we will see that the amplitude of the X-ray scattered by the atom
is proportional and in fact, more or less equal to Z ideally it Z it will never happen
because atoms are all in different positions with respect to some reciprocal lattice origin
and therefore, this value you see depends upon this 2θ value.
493
So, since this depends upon the 2θ value we see that it depends on the scattering angle.
So, since it depends on this scattering angle the scattering directions has to be considered
very seriously and that is shown in the diagram below. We will try to understand this
diagram carefully because this is going to tell us that unfortunately for us this is the
limiting step for us. If it were to be individual atoms contributing their entire
proportional to Z contribution then there would have been no problem. In fact, that is
what happens in the case of a neutron diffraction we will worry about it later.
But, at this particular moment when since X-rays interact with electrons and there is a
size of the electron cloud around a given atom and different atoms have different sizes,
different contents of electrons and so on. This particular quantity f will also vary with
respect to the angle the scattering angle. An illustration of this kinds tells us that if it is
not changing the direction if it is coming in and we find electron density at this point A
and as electron density at this point B it is not changing any direction going through; that
means, there is no change in the path.
The paths simply goes through, but we do find the electron density in its path; that
means, if this is atom all the electrons is effectively you know I have considered two
small elements of electron density, but there is electron density in the entire sphere which
is shown here and all these electron density will now be a part and parcel of scattering
and therefore, there is no path difference. So, DA - BC is equal to 0. So, there is no path
difference. So, if there is no path difference, there is no phase difference; φ is also 0 and
as a consequence the scattering that is coming out from this atom is proportional to the
value of Z. In fact, it is equal to Z at ideally 0 value of no change.
But, as we increase or change the scattering direction, we see that as we rotate the crystal
different parts of the reciprocal lattice will come and therefore, the angle has to change
and that angle changes behaving the Braggs condition. So, as the angle changes if the
angles are very small as it shown here in the second diagram which is referred to as b
then it is if the angle is small like 2θ indicated here is a small change. So, X-rays are
coming here, falling on the these two electron densities and these electron density now
decides to diffract the beam. So, it follows the Braggs law and then therefore, this angle
2θ comes into picture and this will be now this scattering vector wave front. So, this yes,
the scattered wave will go in this direction.
494
So, if that is the case then we see that there is a path difference which comes in and this
path difference DA - BC will be equal to δ which is the path difference if 2θ is small. So,
for smaller 2θ values it goes as DA - BC, but remember that this scattering can be in all
possible directions all 360ᵒ it can scatter, we do not know, right.
So, when it goes to very high angles what is happening is when you have to take a
incoming radiation with a certain intensity and want to deviate that particular incoming
radiation by a very large value 2θ as is indicated here; obviously, most of the intensity
that is contained in the incoming beam will be absorbed by this atom.
So, as a result the path difference now will become DA + AC. So, the path difference is
very very large and for 2θ values large the path difference δ becomes very large and
therefore, you get a very interesting behaviour of the variation of f with respective the
scattering angle and, the next view graph shows you how it varies.
So, we have therefore, plotting here the scattering atomic scattering factor f and its
variability with the sin θ /λ because we would like to say see this independent of the
wavelength which we are going to use. So, we show the sin θ values and we show the sin
θ /λ so that it is a universally valid curve for all wavelengths which we can use. So, f
versus sin θ /λ now varies in this particular fashion.
495
There are certain things which are associated with this plot which I will discuss as well,
but at this moment you see that the atomic scattering factor f falls off with respect to sin
θ/λ. So, it is falling off with respect to sin θ /λ and at very large values. It is becoming
very very small. So, if you will look at actually a diffraction pattern you will like observe
it whether it is a powder diffraction pattern or a single crystal X-ray diffraction pattern.
As you go in the 2θ direction to higher and higher angle the intensity keeps falling very
rapidly. In fact, in an organic molecule if you record a powder diffraction pattern you
hardly see anything at room temperature, of course; you hardly see anything beyond 35
degree in 2θ that is because these values will go down very fast.
See here the what is plotted is that of Chlorine ion and Caesium ion I have plotted the
scattering factors of ions and as a result the Caesium is getting 54 and the Chlorine is
getting 18; that means, one extra electron added to Chlorine and one extra electron
removed from Caesium and therefore, we get this scattering curve. So, what it also tells
is that we can represent a represent also ionized atom positions with respect to the
scattering behaviour. So, whether it is in the neutral position or in an ionized position, it
depending upon the level of ionization we can actually have the value of f different.
So, for Chlorine the curve will start from 17 at 0 value and then fall off exactly like
whatever it is shown here similarly in the case of Caesium. I have taken Caesium and
Chlorine specifically because I am we are going to discuss more of Caesium and
Chlorine in the coming slides and that is why I have taken Caesium, Chlorine, but it is
there for universally for all the elements.
Now, the only problem with hydrogen is that it does not have a core. There is only one
electron associated with the hydrogen and therefore, the variation with sin θ /λ with
respect that is hard to calculate. In fact, these calculations are done as I have mentioned
here the values of f at various intervals along the sin θ /λ are calculated from quantum
mechanics. So, we make use of quantum mechanical methodology to calculate the values
of every value; that means, we go in interval let us say 0.02 and calculate the values of
the how the f value varies with respect to the scattering angle. So, it will fall off with
respect to that and comes down fairly rapidly and then more or less stabilizes at higher
angles.
496
This kind of a scattering factor behaviour is there if for all the elements and it so happens
that in case of anions in case of anions we have the loosely packed electrons in anions
are loosely packed than in the case of cations. So, for example, in the case of Chlorine,
you see that the fall off is very very rapid; that means, the starts losing the electrons at
lower sin θ /λ angle compare to the Caesium which falls slowly. So, whenever the you
have an anion and cation the behaviour of the scattering factor is different from each
other. The electrons in anions are loosely packed than cations. So, the fall off with
respect to sin θ /λ is dependent upon how tightly the electrons are packed in a given
atom.
So, all these values of f versus sin θ /λ at intervals of whatever range we would like to
have can be calculated from quantum mechanics and that is not the job of this particular
course. So, what we will say is we have taken that value to be already available, so, f
varies with respect to sin θ by λ. So, at sufficiently large angles as we see the intensities
fall the intensity fall not because the atoms are not there, but the intensity that is
associated with the scattering is falling. So, they become poorer and poorer scatter as we
go higher and higher in the scattering angle and this is a very serious issue as far as X-
ray diffraction is concerned.
So, as I mentioned already hydrogen has no core, so, it cannot be determined generally
by a routine X-ray measurement. We have to do some tricks to identify the positions of
hydrogen. When we really look at the structure determination protocols we will actually
go there and try to see how these analysis can be done.
So, to just to summarize the last two slides now see that the amplitude of the wave is a
function which is dependent upon the so called atomic scattering factor. It also depends
upon what is the scattering angle. And it so happens that the atomic scattering factor
depending upon the angle of the scattering will vary with respect to sin θ /λ and that
variation with respect to sin θ /λ has been indicated here. I have taken Caesium and
Chlorine I keep this in mind because we are going to use this for a discussion in a few
minutes after this discussion.
497
Now, we go to a situation; you see the thing is we have discussed two issues. This is very
important point; the scattering is done by electrons. We have a crystal it is filled with
atoms and molecules and so on, the scattering is done by electrons and then we do a
diffraction experiment. From what we see from the geometry of the diffraction which we
have analysed already in terms of the reciprocal lattice, each and every spot we get is
from a plane in the crystal. So, the diffraction is happening, the measurements are with
respect to a plane in the crystal.
Whereas, the atoms may or may not lie with respect to a plane, it is not necessary that the
atom should be associated with a plane. It is certainly not necessary that the atom should
be associated with lattice points. In fact, that is how we have been introduced to the
subject in school days that is because it so happens that if we take simple atoms by
simple atom I mean let us say copper or you know gold or silver or something like that.
These atoms are indeed associated with lattice points in their unit cells. So, when you
crystallize an atom, the atom crystals of the atom will have lattice points associated with
them to be the positions for the atoms as well. So, that means, that the phase associated
with that diffraction which comes with the atom position is zero.
So, you it so happens that the belief that lattice point should have atoms has continued in
our mind. So, we have to give a way that belief. What we measure in a diffraction
experiment is a measurement which comes as a consequence of Bragg’s diffraction
498
which is occurring from a plane in the crystal and we know how to identify planes, we
know how to identify those as a reciprocal lattice points, we know the geometry. The
geometry is done by the Laue conditions and also by the Braggs equation. So, it is not a
continuous distribution it is certain set of regions where these the angle, the scattering
angle corresponds to the length of the reciprocal lattice vector whenever that happens we
get the diffraction.
So, we have these diffraction points those are coming from planes. The diffraction is
coming from the planes in the crystal and therefore, the it is single spots bright and dark
and so on. We now just analyse why some are bright and some are dark depending upon
the amount of electron density that is present associated with the atom. Now, the atom
goes and sits in the unit cell in some position x 1, y 1, z 1; x 2, y 2, z 2 and so on.
Now, this x y z is not necessarily to be associated with a lattice point or for that matter
any point that can occur inside the unit cell is a possibility of an atom to sit there with.
The only thing which is also have to be obeyed is the fact that it has to obey this
symmetry rules we have made; that means, if we sitting in a monoclinic system it has to
have a minimum twofold symmetry with which it generates the other atoms. So, if there
is an atom at x y z with a twofold symmetry, P 2 space group you know that x y z should
generate x̅ y z̅; that part we know. So, that means, there are two of those atoms inside the
unit cell at x y z and x̅ y z̅.
So, these both these atoms now will contribute. Now, both these atoms will contribute to
every reflection which is satisfying the requirement that the scattering angle should be
equal to the length of the reciprocal lattice vector. So, if that is the logic then how do we
get the atom positions. So, this is an issue which we will address here after because what
we are measuring from the diffraction experiment is the reciprocal lattice geometry by
means of which we identify the individual hkl values, the so called reciprocal lattice
vectors hkl and we also measure the intensity that is associated with a given hkl.
So, when you a crystal in a X-ray beam, there X-ray beam passes through the crystal
diffraction occurs and you will get various parts and these parts now can be identified
with a certain hkl. We also can measure the intensity that is associated with it
experimentally, at that intensity is the one which is associated with that lattice plane.
Now, the requirement that atom should be there or not is not there in this entire
499
experiment. The atoms can be at some point in space which is defined as x y z or x 1 y 1 z
1, x 2 y 2 z 2 depending on the number of atoms and therefore, this issue has to be
completely understood and how a position x y z contributes to a hkl reflection has to be
valid and we are now planning to look at that particular phenomena in more detail.
In order to do that we need to see the way in which the planes are oriented in a crystal.
We are quite familiar with this now. We can draw a diagram of this kind or we can draw
a diagram with respective reciprocal lattice points, but I think in direct space we have
still a very clear picture because it is always better to have direct space geometry and
therefore, we are drawing this particular diagram.
Now, this diagram tells you that there is a plane which passes through the origin 0, 0, 0.
Now, in the previous class we learnt that any point along this particular direction will
scatter exactly the same way. So, suppose there is an atom sitting at 0, 0, 0 and then I
will put an atom at this position which is moved along this direction with plane within
the plane. So, the plane now you remember, plane is a 2-dimensional plane which is
coming out towards us. So, any of these points lying with respect to the origin 0, 0, 0
they will all scatter in phase.
So, suppose there are ten atoms in this particular plane, all the ten atoms with scatter in
phase φ as the scattering angle the phase associated with this will be 0 and because there
is no difference in the path. The path length is 0, path difference as we saw in the
previous slide and therefore, we have Δ= 0 and φ= 0. So 0, 0, 0 if there is an atom here it
will now contribute its entire scattering power to this particular plane. So, this plane
could be some hkl associated with 0, 0, 0 and this the first plane away from the origin.
Now, we should consider the first plane away from the origin it has to satisfy the Bragg
condition and what is the Bragg condition 2d sin θ = λ. So, the path difference should be
λ. So, the from the origin every diffraction plane which will come will be a path
difference or multiple of path difference associated with d hkl we have seen that it depends
upon the value of n λ, n could be 1, 2, 3, 4, 5, 6 to whatever.
So, if we consider n is equal to 1, then the path difference is Δ equals λ and a

corresponding phase change will be 2π. How is that coming? The Δ path difference is Δ
is the path difference, the phase difference is 2φ you go back to the discussion where we
had the how the path length and a phase they come with respect to each other R going to
500
R+ΔR and that gave a path difference and that Δ R now gets multiplied by a factor of 2π
in order to get the phase difference which is a reduction in the phase. So, addition of a
longer path length will reduce the corresponding phase and therefore, we get a phase
difference which is 2π.
So, if delta is λ/2 then we get value a φ value of π. Now, how does that happen? It does
cannot happen here, but suppose it can happen then we will have to remember that Δ can
be then λ/2. Keep that in mind we will solve that mystery in a few minutes. So, if Δ is
λ/2, then φ will be equal to π remember that and so on. I mean if Δ is λ/3, λ/4 etc.
Now, under what circumstances we will have that kind of a phase contribution to the
given crystal because at this moment we do not have any such contribution. By definition
the first plane out of the origin plane; plane passing through the origin must have a path
difference of λ by Braggs law and therefore, since Braggs law says it is the only way in
which we can have this particular plane which is now at a distance of d hkl will give us the
diffraction.
The next path which we will next plane we will get will be the 2dhkl there the Δ is 2 λ, φ
is 4π. So, these numbers have to be now integers. We can of course, go in the negative
direction as well we can go up from the 0, 0, 0 plane and we get a phase difference,
which is minus 2π again the value of Δ is λ it is moved by d hkl, but since it is in the
opposite direction the phase change will be minus 2π. Now, what is 2π? 2π is 2 times the
π value. What is π? What is 2π? 2π is 360 degrees is it or 180 degrees, which is correct?
Think about it.
We will leave at leave you at this particular point and then one once you tell me what is
2π we will go further or once you realise what is 2π we will go further.
501
Prof. T.N. Guru Row
Lecture - 38
Calculation of Intensities 1
So, you see here now we have been discussing how this value is φ is equal to 2π. So, the
value of φ now goes as 0, 2π, 4π if you look at any of the wave motion and consider the
wave as a sin wave we have the trough to trough or the crest to crest distance as λ . So, if
that is the wave length then we see that the change in the phase that is from here to that
particular point is 180ᵒ, that 180ᵒ in the down and 180ᵒ in the up so it will become 360ᵒ.
So, therefore, we have these kind of phases which will phases which are developing with
respect to the individual planes. So, let me make it clear we have a crystal, in that
particular crystal we have a plane passing through the origin 000 and this particular
origin now will allow us the first diffraction to come out of this plane at a satisfaction of
the path difference being λ and therefore, the phase difference is 2π.
If you go in the opposite direction the path difference is λ and phase difference is -2π and
we can always develop many such parallel planes. It is also possible to have planes like
that in a opposite direction, in another direction you remember the bread analogy we can
cut the bread not always like this we can always cut it along 1 1 1 plane or something
502
like that. So, you can have a sandwich kind of bread. So, we can have these cutting of the
slices in various possible directions.
Now, the thing which we have to notice is that the amplitude of X ray scattered by each
atom in the unit cell is proportional to f. So, each atom now the amplitude of X rays
scattered is proportional to f and therefore, it depends upon the scattering angle. So, as
we go to the higher and higher angles from 0 0 0 as we go to higher and higher angles of
scattering the value of f falls with respect to sin θ/λ.
The atomic position in the unit cell dictates the final amplitude if the atom is at 0 0 0 the
phase is 0, so, this is a point which one should remember. Now what happens if the point
is not with respect to the plane and this is something which is of extreme importance and
we will discuss that in a minute. Now, suppose the we have a 0,0,0 plane just like that
plane here we have a 0,0,0 plane, the first diffraction will occur at a distance of d hkl, now
the second one will occur at 2d hkl and so on. Suppose there is an atom x, y, z at this
position in the unit cell so there is an atom here and that is in the unit cell.
So, what you do is you draw a plane which is parallel to the plane so, in two dimensions
we are drawing a line. So, this line which is there 1 d hkl away which is actually giving the
scattering associated with it, what we do is? We draw a line parallel to that, so the x, y, z
is an atom which is now not at a distance of d hkl, but at a distance of D hkl. So, D hkl is the
distance of the atom from the plane and we have drawn a plane which is parallel to 0,0,0
and therefore, parallel passing through x, y, z, but parallel to the first plane which is d hkl.
So, if that is the case then what I have written will happen scattering is from the atom
and not the plane we know very very well now.
So, phase difference of X ray scattered by atom x, y, z and an atom at the origin. So,
suppose there is an atom at 0,0,0 and another atom at x, y, z the atom at 0,0,0 there is no
problem because the phase is 0. The atom at x, y, z now will have a path difference of d
hkl this is where the atom is located, so the path difference of d hkl and this is where the
plane is located. So, we are actually now calculating wherever be the atom it does not
matter the contribution of that atom to the plane because we are actually getting the
experimental measurement in terms of the planes.
So, the diffraction spot we get is from d hkl which is not coming from x, y, z position, but
the scattering is done by x, y, z and therefore, we have to them consider (D hkl/d hkl)/λ as
503
the path difference and therefore, the phase difference will be 2π (D hkl/d ) and so we
hkl
can calculate the corresponding phase difference associated with the atom being at some
position x y z with respect to the plane.
So, now you can imagine the contribution of x y z to some other plane which may pass
through the origin and it may go something like that, then the first plane out will be
there. So, then we have to calculate that D hkl and the corresponding d hkl. So, therefore,
this atom now contributes to all the reflections go back to the very original idea we
brought in that is when we have more than a wave coming out from a position these
waves can interfere.
So, they can interfere constructively, they can interfere destructively, they can interfere
with partial overlaps and so on you remember that animation I showed of a sine curve.
So, that therefore, will tell us the contribution from x y z could be maximum to certain d
hkl planes could be minimum to certain d hkl planes and so on. And therefore, the intensity
that is associated with the d hkl in the contribution of this atom; x, y, z to that plane is very
crucial and that comes out from this calculation that phase difference will be 2π (D hkl/d
).
hkl
So, how does it help? It helps in the following way if we take an example. So, what we
are therefore, going to take do is to take an example to illustrate this one and that is why
I showed you this scattering factor curves of f versus sin θ/λ , corresponding to Cs + and
504
Cl - the ions because these two ions form a compound which is Cs Cl the Cs Cl, Cs Cl is
a salt we know that. So, it is exists as Cs + and Cl -.
So, the scattering factor that the atomic scattering factor that contributes to this particular
structure will come from the atoms Cs + the ion the ionic species Cs + and the ionic
species Cl - and that is why we drew the curve earlier on of f versus sin θ/λ .
Let me take it back to you so this is the curve which we have one for the Cs + the other
for this Cl -. So, we can get the value of the atomic scattering factor at whatever be the
scattering angle at which we find the diffraction spots. So, individually we can identify
what is the contribution of Cs, what is the contribution of Cl -. So, if that is the case then
we should be able to construct a intensity profile associated with any hkl plane.
See in the case of Cs Cl if the structure is known, so we are going to assume the structure
this the say Cs atom is sitting at the origin and it’s a simple cubic structure. So, you will
have it repeat again one unit away one unit away and one unit away. So, this is our unit
cell with Cs at 0,0,0 and Cl sits at ½ ½ ½. So, this is shown as at x equals ½ y equals ½
and z is equal to ½ ½ is indicated next to the atom.
So, these are the positions of the two atoms there is nothing else in this unit cell. So, we
want to see what are the contributions, these two atoms do or give to the individual
reflections, because what we get from the diffraction experiment is the individual
505
reflection that is associated with your hkl planes. So, how do we get that? So we write
the I(hkl). So, this is a very general expression for any intensity of hkl we can write this
expression, this is equal to I (hkl) = ( n cosφn)2 + ( n sinφn)2 .
That is because you see have discussed earlier on the scattering factor and then
eventually the structure factor. This structure factor was a complex number and the
square of the structure factor gives us a value which is proportional to the intensity and it
is expressed as cosine + i sine you know in the complex diagram since that is expressed
as cosine + i sine and that is our structure factor when we want to square that quantity
what we will get will be this expression.
So, therefore, the I (hkl) = ( n cosφn)2 + ( n sinφn)2. So, this is a general

expression for intensities any compound intensity to any given hkl can be calculated
based on how many atoms are there in the unit cell and what is their contribution to a
given scattering angle. So, if you see in this particular case there are only two atoms in
the unit cell this is summed over 1 to N all the atoms inside the unit cell, sum over 1 to N
all the atoms in the unit cell.
So, if you have a 100 atoms structure; obviously, this calculation is going to be very very
very difficult its; obviously, not the way to do it. So, at this moment we know what is the
contribution of the intensity the contribution of the individual atoms that is present in the
structure, it depends upon where it is the phase that is going to be associated will depend
upon where it is sitting. So, in this particular case the Cl atom is at 0,0,0 sorry ½ ½ ½ and
the Cs atom is at 0,0,0.
So, let us calculate the contribution to intensity 1 0 0. So, intensity 1 0 0 will be at a

distance of dhkl is d1 0 0. So, the dhkl now is equal to a the cell dimension. So, therefore, we
are considering the plane which is passing through the value of d 1 0 0. So, if you this is the
three dimensional structure you have this line this line will now be the representation of
d1 0 0 therefore, 1 0 0 plane is represented by a horizontal line.
So, this is your 1 0 0 direction this is your 010 direction and therefore, if you consider
this point it will be one along this and one along that. So, we therefore, consider the
contribution of d1 0 0 the intensity of d1 0 0. The intensity of d1 0 0 in this particular example
506
where we have only 2 atoms can be very easily written that is because we can write the
use this expression expand it. So, we get (fCs cos 0 + fCl cos π)2. Now why did I write cos
π here? You all know now by now what is phase difference which comes, if the atom is
removed by the ½ the distance.
So, we say that the path difference the path length for the first plane which is out here
which is now at 1 0 0 in this direction the path difference is λ and therefore, the phase
difference is 2π. Whereas, the atom now is coming at ½ position so we have to take the
ratio of D hkl to d hkl. So, what is D hkl with respect to Cl you have to draw a line here
parallel to 1 0 0 and that comes at ½ this position.
So, if this is the value for the path difference of λ then this will be at the point which is
equal to λ/2 that is because ½ (d 1 0 0) is (½ a), since it is (½ a) direction we will have the
path difference of λ/2 because with a full “a” the path difference is λ by definition of the
Braggs law.
So, the atom now is sitting ½ way between this, since we now draw a parallel line we
calculate the path difference or in fact, the phase difference as D hkl/d hkl and that will be
equal to π, because it the path difference is λ the phase difference is π, path difference is
sorry path difference is λ/2 the phase difference is π path difference is λ then the phase
difference is 2π which we have already seen.
And therefore, the value of I(1 0 0) goes like this (fCs cos 0 + fCl cos π)2 +(fCs sin 0 + fCl
sin π)2. So, this can be calculated we know the how the cosine behave, here takes place
and how the sine function varies with respect to the angular variation and so this turns
out to be (fCs - fCl)2. So, the intensity of 1 0 0 is the difference between this atomic
scattering factor of Cs ion - the atomic scattering of Cl ion.
We have to find out where this I(1 0 0) will come see I(1 0 0) is not coming to going to
come at the scattering angle θ equals 0, it will come at a certain θ value one can calculate
where that θ value can come and that particular value of f should be taken from this
curve. So, the it is not going to be 54 and 18 contribution it is going to be something
different and that something different will come from the value of sin θ/λ we will see
how it comes up.
507
So, what is happening here is that the d 1 0 0 represents a plane which is passing through
this Cs atom, similarly d2 0 0 which is a/2 passes through the Cs atom as the Cl atom. Now
Cs atom is now removed there is no Cs atom associated with 2 0 0, but what is the
contribution to the intensity this very interesting there the Cs atom is at 0,0,0 as you
know that any atom at 0,0,0 will always contribute with a phase change of 0ᵒ that is
because the path difference is 0 and the phase difference is 0.
So, we have therefore, I(2 0 0) we have the scattering factor of Cs cos 0, for Cl now it
becomes cos 2π because now d2 0 0 is the one which is the first plane out of the after 0,0,0
and therefore, it will now have a path difference of λ . So, the phase difference is 2π so
we get (fCs cos 0 + fCl cos 2π)2. So, when you put these values in and then you will see
that this turns out to be (fCs + fCl)2, the only interesting point here we have to notice is that
we have to see that corresponding sin θ/λ value which will come for d1 0 0.
And the corresponding sin θ/λ which will come from 2 0 0 and read out the
corresponding fCs, fCl values; that means, if you go back here the 1 0 0 may come
corresponding to these two points 2 0 0 may come corresponding to those two points. So,
the value of f will not be now 54 and 18 it will be much lower. So, as you go to the
higher angles from 1 0 0 to 2 0 0 the value of f Cs and fCl will reduce; however, you see
that the contribution to the intensity 2 0 0 comes out as (fCs + fCl)2.
So, it may be that even though fCs, fCl individual values are lower the contribution to the
I(2 0 0) will be much higher because they are adding up whereas, in this particular case
the contribution is subtracted. So, the individual values of the scattering powers of Cs
and Cl may be higher for 1 0 0 case because of this position in the sin θ/λ where we
have to read out the corresponding f value. On the other hand the 2 0 0 will be further
away in sin θ/λ , but since there is an addition here this value may be much higher than
that value we will see what are those values which will come as in a few minutes.
So, how does this happen? This happens because d2 0 0 is ½ of d1 0 0 and that is therefore,
2π times (½ a) this is Dhkl/dhkl, the Dhkl is (½ a) the small dhkl is also (½ a) because we are
now considering the reflection 2 0 0 which is ½ way between 1 0 0 and the origin. And
therefore, we have a 2π value corresponding to the phase difference and that is what has
appeared here. So, if you now study this carefully once more this particular slide you will
see that it is very very easy to identify how these contributions are coming.
508
So, when we consider the 2 1 0 plane Cl atom will lie midway between the lattice planes
to D2 1 0. So, it will get 3/2 (d2 1 0) let me indicate where 2 1 0 is it’s here. So, this is the d 2 1
0 we are now calculating the contribution from Cs as well as from Cl, we do not have to
worry about contribution from Cs because it’s sitting at the origin. So, it contributes
equally to all the reflections because the atom is at the origin.
So, since the atom is at the origin it contributes to all the reflections equally there is no
difference in path, there is no difference in phase. So, whether the plane is here parallel
like that d1 0 0 d2 0 0 d3 0 0 or a plane which is going now like this is now 1 unit along y and
(½ a) unit along x. So, therefore, this is the d 2 1 0 plane corresponding to d2 1 0 this is the
plane 2 1 0 and you see that the plane 2 1 0 is how the Cl atom now sits ½way between
these two and so if you calculate that value it turns out that it is at 3/2 (d2 1 0).
So, the X ray scattered by Cl atom at ½ ½ ½ into this reflection will be 2π times (3/2)
that happens to be 3π. So, 2 1 0 therefore, you can calculate you have f Cs cos 0 that does
not change, but fCl cos 3π now square of that and the sine part also go become square. So,
the value can be calculated as (fCs - fCl)2.
So, the reflections 1 0 0, 2 0 0 and 2 1 0 they do not appear as we again discussed earlier
on with a figures the values of f Cs+ and f Cl- are to be taken corresponding to the 2θ
values of these reflections in the f versus sin θ/λ curve these values. So, we have to pick
509
up those values of the f. So, if we do that then we will be able to contribute calculate
what should be the contribution of Cs Cl to these individual reflections.
Now, imagine a structure which has 100 atoms you have a unit cell let us say 8 * 10 * 12
an orthorhombic unit cell. So, you can actually generate how many possible planes can
be created because we now know how to generate the planes, the planes need not be
along a or b or c direction this is our bread analogy we can cut this loaf of bread in any
orientation we want and each loaf each slice will now represent a plane.
So, the plane the bread can be cut like this bread can be cut like that bread can be cut
along the diagonal and so on. So, if you cut the bread along the diagonal the first plane
out at the middle at the a = 1, b = 1, c = 1 will be the 1 1 1 plane. So, all planes which are
parallel to 1 1 1 can come if we keep slicing the loaf in that particular direction.
So, therefore, we now know how to calculate the contributions for the individual
reflection. So, suppose there are 100 atoms each and every atom wherever it is sitting
will contribute to this. So, it is not going to be a very easy task here we have taken 2
atoms we have taken a Cs at origin and Cl at ½ ½ ½, very easy to see how these are all
coming, but when we have 100 atoms in a xn, yn, zn positions randomly distributed, but
covered remember it is in the crystal, so it is always covered by symmetry rules.
So, we have our symmetry rules validated, so keeping the symmetry rules in mind we
will have these atoms, but still it’s a large number of atoms and this is not the way which
we should be calculating. However, if you want to see how the Cs Cl behaves, this is a
chart which I am giving you this chart tells me tells you the following this is a simple
cubic as we said.
510
So, a is equal to 4.123 it so happens the unit cell dimensions are all 1234 is by chance, so
a is 4.123Å, Cs is 0 and Cl at ½ ½ ½. So, we can calculate the hkl 1 0 0 what should be
the intensity for that. So, we get the d hkl which is equal to a in this particular case
because 1 0 0 is the reflection we are considering. So, if we calculate the sin θ/λ using
the formula 1/2d hkl we will get 1/2a = 0.12, so this is the value of sin θ/λ . So,
corresponding to 0.12 you have to read out the Cs and the Cl scattering factors. So, let us
go back and see how to do that.
So, here you take the value of 0.12 somewhere here, so you will have to now read the Cl
contribution and read the Cs contribution at 0.12. Now the 0.12 values are in fact, given
in this table, the table gives you the values to be coming in such a way that the difference
in these two the value of fCs is 50 for fCl it is 15. So, there is a reduction in the
contribution of the individual atom.
So, the even though we have a Cs+ and a Cl- in this structure the contribution from Cs
and Cl is coming as though the atoms have only 50 electrons and 15 electrons in these
two cases the ions have only 50 electrons and 15 electron. So, the this is something
which you must remember when you do the calculation and so we therefore, all have
already seen what is 1 0 0 in the previous view graph that is (f Cs - fCl)2. So, the difference
50 - 15 is 35 square. So, we get an intensity of 1225; we get an intensity of 1225.
511
So, similarly we can do for 2 0 0, 3 0 0, 2 1 0, 4 2 0 depending upon the value of d we
substitute this value of d up here we get the sin θ/λ value you see that the sin θ/λ value
keeps on increasing and correspondingly the value of Cs and Cl keeps on decreasing and
depending upon where the fCs fCl values are for example, 3 0 0 is a reflection which
comes after 2 1 0.
And therefore, you see the contribution is 41 and 10 here and 36 and 8 there, but what is
more important is the way in which the phase addition takes place and therefore, here
you have fCs + fCl giving you 2809 as the intensity. This intensity is an absolute value
because we do not have anything else other than these electrons. So, these are therefore,
showing the absolute value and the absolute values are now 1225, 2809 you see that even
though this is fCs - fCl square even though the values are 28 and 7 the contribution of 4 2 0
is very large, so you get a value of 1225.
So, what does it all mean what it means is now you take Cs Cl crystal put it in the path of
the X ray beam and do a diffraction pattern what I will show you instead of showing
these parts which will appear for these individual reflections I will show you a powder
diffraction diagram which is simulated.
So, we know where Cs is sitting and where Cl is sitting so I took it out from the web and
this is the XRD pattern of Cs Cl. So, you see now that this is the intensity and this is the
512
2θ, you see here the intensity values are given some numbers here. So, what why is that
given like that, it depends upon the instrument you are using.
So this is therefore, what we can call as the relative intensity it is not the absolute
intensity, but you see that the first reflection and the second reflection is 1 0 0 and 2 0 0
and you see the values were 1225 and 2800. So, you see the relative variations of the
intensities in the powder pattern. So, the powder pattern or the single crystal pattern
suppose this was single crystal.
In fact, the simulated pattern is very useful because you draw a single line in case of the
single crystal you get a spot here corresponding to this intensity a spot there
corresponding to that intensity and so on. So, this is therefore, the plot of the Cs Cl X ray
diffraction pattern. So, now, you see a one to one correspondence for the first time, you
see the diffraction occurring the diffraction therefore, as I made you believe and now you
know very well very thoroughly that it is now associated only with the planes atoms are
sitting at 0 0 0 and ½ ½ ½, but the X ray diffraction pattern is produced from the planes
and you have calculated yourself the contribution to each of these planes in case of Cs
Cl.
And you see that behaviour is shown in the X ray diffraction pattern and what you have
calculated of course, is a little more sacred than this in the sense that you have calculated
the absolute intensities. So, what you have to therefore, do is to convert the relative
intensity you have got in the X ray experiment to the absolute intensity there are so many
other factors which are involved to do that operation, we will see how we do it at a later
date when we actually determine these structures.
But at this particular point we see that the intensities can be calculated if you have a very
simple structure like Cs Cl where we know already the structure Cs is at 0 0 0 and Cl is
at ½ ½ ½ then it is very straight forward as we have done now and we also now know
how the pattern arises because of the positions of these atoms, but what is more
important is to take a case where we do not know where the atoms are.
Suppose we had a Cs Cl structure where we had not solved the structure then there is no
way I would have done this calculation. So, there is no way I would have got these
relative intensities; however, if I had recorded a powder pattern I would have got this
513
powder pattern this is for reality. So, we have the Cs Cl sample and you record powder
the Cs Cl crystal and recorded the powder pattern.
Of course, this is a simulated pattern, so it does not give you any widths, in a typical
powder pattern there is a certain width associated with it this we will all discuss when we
go to powder diffraction methodology, which is very crucial particularly for the pharma
industry people to remember that there will be a certain width with each of these peak
positions and that is very crucial.
So, effectively this is the intensity we will get when we put the Cs and Cl in known
positions and calculate the corresponding contributions to individual reflections. So, X
ray diffraction pattern therefore, is coming from the planes and the atoms are sitting need
not sit in any plane they can sit wherever they want, but their contribution to the plane
can be calculated best as we have done in the case of Cs Cl.
I think this kind of brings us to a stage where we can stop at this juncture, because what
we are going to do next will be a situation where we can get to the general expression to
calculate the intensities, see we cannot all the time calculate the intensities this way
which we did. For example, in this case we have the intensity expression which is a
general expression, but then we cannot get the every time where the atoms are sitting and
with respect to them calculate this. This is a very simple calculation where we had only 2
atom in the unit cell, but if there are more atoms in the unit cell and they are distributed
all over the unit cell how do we get to that position, how do we calculate the intensity?
So, we have to have a general formula. So, the next job would be to generate a general
formula by means of which we can calculate the intensity given the atom positions we
still given the atom positions, if the atom positions have to be determined that is a next
step which is our structure determination we do not know where the atoms are, but we do
have the experimental data now which gives us the I versus 2θ plot whether it is a
powder diffraction or if it is a single crystal we get the individual reciprocal lattice points
which can be indexed and the intensities can be measured.
So, we have reached a stage of let us say it’s a mid level stage of this particular course
where we have more or less covered all the aspects of symmetry, all the aspects that are
associated with space groups and so on, how we generate the equivalent points? Now we
514
have also found a situation how the atoms sitting in different positions contribute to the
intensities of individual reflections.
515
Prof. T.N. Guru Row
Lecture – 39
Calculation Intensities 2
So, what we have been looking at is to see, what is the contribution of the intensity , what
is the contribution towards the intensity, when ones atoms are found in some positions.
For example in the case of a caesium chloride, we discussed the issue that caesium is at 0
0 0, chlorine is at ½ ½ ½ and then we calculated the intensities associated with various
reflections like I (1 0 0), I( 2 0 0) and so on.
We found that in all these expressions, the intensity gets simplified mainly, because of
the fact that the caesium is sitting at the origin, which makes caesium to be at 0 0 0 and
chlorine is sitting at ½ ½ ½. So, it makes it contribute to various planes depending on it's
location.
So, remember we have to use the expression D hkl/d hkl that is the one which gives us the
ratio of how the position of the atom is with respect to the given plane. For example, if
you take 2 1 0 here, the chlorine atom is positioned exactly midway between the parallel
planes, which pass through this point, defining 2 1 0 and this is again 2 1 0, so, these are
516
parallel planes. So, this was a methodology with, which we developed that the intensities
can be calculated and if we indeed calculated several intensities.
But in a general case, it is not always caesium chloride we are worried about. We are
worried about hundreds of atoms and these hundreds of atoms by symmetry or
generating. Let us say more than 4 or 5 molecules in the unit cell. So, it is a very large
assembly of atoms and each and every atom now, from our discussion contributes to a
given plane. The interference issues with waves coming out of these atomic positions,
because it is the electrons which scatter X rays.
So, depending on the electron density, we get contribution into various planes and these
are now constructively or destructively interfering and therefore, the intensity associated
with a plane gets calculated. The issue is now, how can we get a general expression,
because we cannot just every time go and find out where the atoms are sitting and then
calculate instead there should be a very general expression depending upon wherever the
atom is sitting in the unit cell.
We should be able to calculate the contribution to the plane and that is the idea with
which we will now, develop a general expression as I have written here; a general
expression for the calculation of intensities. This will be applicable to all three
dimensions and therefore, it is in a unit cell depending upon where the positioning of the
517
plane is, we should be able to calculate the contribution to that plane from all the atoms
that are present in the unit cell.
So, this is therefore, a very important calculation, which will sort of generalize, how we
approach the various issues of how to calculate the intensity. So, what we will do is, we
will first derive the general expression and then incorporate the general expression for
finding the contribution to various planes the intensity contribution from the presence of
100’s of atoms or 1000’s of atoms or whatever it is. So, this in principle is easily
programmable, so, we can develop packages with which we can calculate intensities.
Whenever I am saying calculating intensities, it means that we already know the

positions of the atoms. So, it is only after the structure is determined we can calculate
intensities that we must remember. So that means, we should know the positions of the
atoms if before we calculate intensity, but the general expression for calculating intensity
will always be automatically coming and that is obtained by this obtained by this
diagram in which I show, a two dimensional projection, x along this direction and y
along this direction, this is the angle between x and y and of course, z is coming in the
direction away from the plane of the projection.
So, if you consider a plane passing through the origin as we have always discussed, this
particular plane will have a phase difference of 0, because there is no path difference at 0
0 0. We also have shown that all atoms lying on this plane will scatter in the same phase.
So, the phase is 0 associated with atoms lying in that plane. What then we will do is to
take by definition of the Braggs law. We take the first plane, which comes out of this
particular 0 0 0 position and let us take a general position, where we will have the first
position coming, which is shown here. This is the first plane out of it; that means, the
phase change will be 2π.
So, whenever the phase change is 2π, we have the λ as the path difference and therefore,
we satisfy the condition 2 dhkl sin θ equals λ. So, we satisfy that expression, which is the
Braggs expression. So, suppose, we have an atom at x 0 0 along the direction of x. We
locate an atom x 0 0 the contribution of this atom to this plane. We know already how to
calculate, what we do is we draw a line parallel to this plane passing through the x 0 0
like that and then we calculate the distance of this particular plane from the origin and
518
then we take the a/h of course, a/h b/k and c/l will be the intersections for the first given
plane.
So, depending upon what is the value of h k and l we can define this particular plane. So,
the first plane, which obeys the Bragg equation will come at a distance of a/h, where a is
the unit cell, h is the Miller index, b/k wherever k is the miller index, b is the unit cell and
of course, in the third direction it will be c/l. So, if we now consider this as the x 0 0 then
what is the phase that is associated with this, because that is what we want to calculate
the intensity. So, the first hkl plane out from the origin makes an intercept a/h, which is
what we are shown here with a phase change of 2π.
So, the first phase change is 2π. Atom is at x 0 0, so, what would be the D hkl/dhkl the D is
2πax, because that is the distance, which we which up to the x 0 0. 2πax divided by a/h,
which is the value here. So, if we take this 2/ax divided by a/h, this will give rise to 2πhx.
So, 2πhx is now, the plane, the phase that is associated with this particular atom its
contribution to the plane, which is the first plane out of 0 0 0.
So, the contribution to this plane is 2πhx, if there is a atom sitting at x 0 0 likewise, if
there is an atom sitting at 0 y 0. It will become 2πky, because it will be 2πby divided by
b/k. So, that will become therefore, 2πky and similarly, if it is in the l direction, it will
become 2πlz. So, in the case, where the atom is sitting at x y 0. Let us say we have an
atom here. So, what we have to do is, to line draw a line parallel to that calculate the D hkl
and the small dhkl is already available and therefore, that will be the atom is at x y 0.
So, the corresponding phase will be 2π(hx + ky), hx due to this direction, ky due to that
direction and therefore, we have x contribution y contribution both taken into account.
So, the phase change will be φ equals to 2π(hx + ky). So, this diagram therefore, is a
general picture of any crystal structure, which will have this kind of disposition. We have
considered the positions of three different atoms corresponding to their contributions to
this particular plane and the phase changes have been calculated. When once we have the
phase changes, we can always calculate the intensity that is associated with it, because an
atom at in a general position x y z will give rise to a phase change of 2π(hx + ky + lz).
And of course, the amplitude will be proportional to that of f, which is the atomic
scattering factor corresponding to the scattering angle θ about, which the reciprocal
lattice vector is located.
519
So, what is important here is, the location of the reciprocal lattice point. The location of
the reciprocal point is this associated with the plane and the amplitude therefore, has to
be read proportional to that. So, whatever is the value of f, at that particular
corresponding scattering angle, which is corresponding sin θ/λ has to be read out and
then we substitute it to calculate the intensity, this is something which we will now
consider and see whether we can do this in very simple terms.
Let us take a unit cell with N atoms. So, now, we have a unit cell with N atoms, we have
various planes passing through this crystal. It is any general crystal. So, the cell
dimensions are a b c α β γ. So, in such situations what we have is, we have a variety of
atoms in our structure. So, let us say we have N type of atoms at N different positions, a
very general discussion. So, the atomic scattering factors are f 1 f2 f3 so on and the
coordinates are x1 y1 z1; x2 y2 z2; x3 y3 z3 and so on. So, if there are N atoms we have
therefore, the fn corresponding to the atomic scattering for nth atom and xn yn zn as the
coordinates, then we write the expression for I(h k l).
Now, remember the I(h k l) expression, we should contain contributions from every one
of these items and therefore, we have f1 cos 2π. Now, this is the phase change, we have
already calculated the phase change to be 2π (hx + ky + lz). Now, the atom is at x1 y1 z1.
So, we get this first term as f1 cos 2π(hx1 + ky1 + lz1) + etc.., for second third fourth atom,
until the Nth atom, fn cos 2π(hxn + kyn + lzn)}2. Notice that it has to be whole square. This
520
is the cosine term, we will also have the sine term, because we know in our wave is a
complex number. We will come to the discussion of structure factors in a later time, a
few may be in about an hour and hour and half of discussions on this. So, what we
therefore, see here is the intensity consists of cosine functions and the corresponding sine
functions, cosine function square + sine function square. So, this is the intensity
associated with any h k l. So, what is the job we have at hand? We have a crystal, we
have done the diffraction experiment and we want to now, calculate the contributions of
atoms lying at various positions x1 y1 z1 , x2 y2 z2 and so on to the overall intensity that is
developing at a given plane h k l. So, for any given general plane, it should be possible to
calculate the contribution coming from various atoms. So, there are N atoms here, so, all
the N atoms are summed up here, in this contribution.
So, this can of course, be written as equal to sum over 1 to N fn cosine 2π(hxn + k yn +
lzn)2 + sum over 1 to N fn sin of 2π the corresponding sine term square. Let us take the
example of our caesium chloride again, to verify whether we are doing it correctly,
whether we are doing this calculation or whether this particular expression, which is a
general expression for any crystal any crystal with N atoms and it's contribution to any h
k l plane. So, therefore, I call it as the general expression for calculating the intensities.
So, we take the case of caesium chloride, caesium is at 0 0 0, chlorine is at ½ ½ ½. So,

what all we have to do is to substitute the values here. So, f cs the atomic scattering factor
of caesium cos 2π (h*0 + k*0 + l*0), because 0 0 0 is the position of x1 y1 z1 then fcl cos
2π (h*½ + k*½ + l*½) 2 and of course, the corresponding sine terms. What happens in
this situation is that the sine terms will vanish, we will in fact, make a general discussion
of why sine terms vanish in situations where which is, which are similar to the structure
of caesium chloride. In fact, sine terms generally, vanish in case you have a
centrosymmetric crystal.
So, if there is a inversion centre in the crystal and that inversion centre coincides with the
origin, then all the centrosymmetric structures therefore, will not have any contribution
from the sine terms. On the other hand, the cosine terms does do contribute in the
following way for example, the fcs since, it is at 0 0 0, there is no phase change. So, fcs
remains as such. So, fcl now gets modified with this change in the phase and the change
in the phase happens to be {cos 2π (h + k + l)/2}2.
521
So, this is a general expression for deriving the intensity in case of the caesium chloride.
So, all possible h k l planes, which come out from the caesium chloride diffraction
experiments can be calculated and this particular value is now absolute value, because
we are reading out the corresponding atomic scattering contributions from individual
atoms, chlorine and caesium. So, one ones we take an example here, we want to calculate
the intensity of some plane 4 2 0. So, what all you have to do is to substitute that
expression in this expression the values.
So, you take the fcs as corresponding to that value in our f versus sin θ/λ picture. So, we
cannot take the z value we have to take that value of f cs, which corresponds to the
scattering angle. So, if scattering angle is θ, the corresponding sin θ/λ value has to be
read for the fcs similarly for fcl. So, one ones we have those values fitted in then the cos
value of 2π now, h + k + l here is 6 so, 6 divided by 2. So, 6 divided by 2 happens to be 3
times.
So, it is essentially this. So, happens that it turns out to be a factor of cos of 2π. So, the
sign becomes +. So, you will have (f cs+ fcl)2. So, for any given I(h k l) you can calculate
again for 1 0 0 or 2 0 0, whatever contributions will come from any general h k l plane.
So, those h k l planes, which will contribute to this will the intensity can all be
calculated. So, if we now do an experiment, let us say with caesium chloride either a
powder diffraction experiment or a single crystal experiment. Let us take the case of a
powder diffraction experiment, then we will get a powder pattern corresponding to
caesium chloride and that particular powder pattern will be, we will able to index it in
terms of where the h k l planes are coming by looking at the x axis of the X ray plot. Let
me show you that plot.
522
So, let us say this is the XRD pattern is in fact, the simulated XRD pattern using the
coordinates of caesium chloride. We have calculated, what are all the contributions to the
to individual reflections. Since, it is a powder pattern you will see the lines. In fact, in a
typical recorded powder pattern each line will have a certain width. We will discuss
those issues later when we do powder diffraction, but for our discussion now, each of
these endpoints will represent our lattice points, if it is a single crystal. So, in principle at
various values of these 2θ values you will find the reflections appearing and those
appearance of reflections will have this amount of intensity.
So, this is the arbitrary intensity, because this is done on some equipment, whether you
do it in, an equipment in Bangalore or some other place or your place, wherever you
want to go and do the experiment, the intensities will change depending upon the
intensity scale will change depending upon which instrument you are using what are the
experimental conditions that are available and so on. So, this is actually a relative
intensity and this has to be scaled and brought onto the actual intensity to the absolute
intensity, which we get from our calculation, because in the calculation we are using the
fcs and fcl as they should be at that scattering angle and therefore, we get the absolute
value (Refer Time: 18:30). We will not worry about that aspect.
What we worry about this aspect is the following and what we should know about any
powder diffraction, we will when we do it in detail later we will also see that this is that
523
the geometry that is associated with the diffraction, that is the dimensions of caesium
chloride, the nature of caesium chloride, the appearance and disappearance of these
peaks in case of sodium chloride all that will be indicative in the x direction.
So, x direction therefore, gives you the geometry of the diffraction, the x direction is the
place where you can actually see the presence of 2dsin θ = nλ reflections, which are
satisfied. So, you do not get continuous pattern here, you get discrete patterns, whenever
there is Braggs law obeyed, we will get the diffraction and that diffraction, whether it is
constructive or destructive, it is decided by the positions of the atoms with respect to the
planes from which we are doing this locating the measurements. So, we therefore, have
all the planes which are coming out in the x direction. So, the x axis represents the
geometry of our diffraction condition, the y axis represents the intensity that is associated
with the diffraction measurement.
So, because of this we see that the associated corresponding intensities will come with
respect to the individual calculations, which we did for I(1 0 0) and I(2 0 0) and so on,
we calculated those intensities. So, they will come, because of the of fact that caesium is
at 0 0 0, chlorine is at ½ ½ ½. So, this discussion therefore, assumes that we know the
structure. This discussion also tells us that the way in which the XRD pattern can appear,
because the XRD pattern will appear as a relative intensity versus 2θ in all experiments,
which you have performed in a normal diffractometer. And therefore, the 2θ degrees,
which we plot in the x axis, the values associated with x axis will therefore, represent
various 2θ values.
Now, these 2θ values are the ones, which will tell us what are the d values
correspondingly, because we use the formula to 2dsin θ = nλ, where λ is the wavelength
we have use so; obviously, this pattern will be different for different wavelengths.
Suppose, we use copper radiation, it will have this pattern, if you use molybdenum
radiation the pattern will be different, the θ values will come at different positions. So,
you should remember that.
But if you plot let us say the relative intensity against the d value, d is already
incorporating the corresponding sin θ expression and so on, then the two patterns will be
similar and this is something which you must remember. So, when you what is plotted
along the x axis is important.
524
But basically what is given in the x axis direction, in this plot is the value that is
associated with the geometry. So, the geometry of the diffraction condition we can
determine; that means, we can determine the unit cell, the dimensions we can determine.
Therefore, the crystal system we can determine where the diffraction spots are coming,
and therefore, the geometry of the diffraction. So, entire geometry of the diffraction is
conserved in along the a axis along the x axis.
So, that way we know when we do a xy plot of this kind that all information on the
geometry is here, all information about the structure is along the y direction. Because the
intensity is the one, which is the, which is telling us, which item is where and therefore,
the electron density distribution is covered along the y axis. So, the information that is
present along the y axis therefore, incidentally tells us about the way in which atoms are
distributed inside the unit cell.
So; obviously, the appearance of these relative intensities depend upon the symmetry
associated with the material and therefore, when we have found out the space group and
in the unit cell dimensions, the number of molecules in the unit cell etc.., which about
which we are already reasonable experts. So, this will now tell us how that contributes to
the intensity measurements.
So, this tells us that these are the possible reflections associated with caesium and
chlorine and a is equal to 4.123 and this is the calculated values, which we get for sin θ/
525
λ, based on the presence of where the d values are and therefore, we can calculate based
on our general expression which we derived just now. This general expression will tell us
how to go about determining the intensities for any given X ray diffraction pattern. So,
this is a key equation. Now, if we take the square root of this I(hkl), we will get what is
known as the structure factor, it is the modulus of the structure factor.
We will now take up a discussion on the structure factor, but I taught that before we go
into the detail of the structure factors associated with any diffraction condition, which
essentially is an important factor, because we remember we had 2 expressions, we called
them Fourier transforms of each other. We showed that F(h k l) or F(R) at that time F(h k
l) is related by a Fourier transform to ρ(r) where ρ(r) is the electron density.
And vice versa that is if we from ρ(r) we can get to the F(R). So, those two equations are
the major fundamental equations, which we need in order to solve the structure. So, for
example, if we know the structure factors, we can determine the structure in principle
and if we know the electron density, we can determine the structure factors. So, this
Fourier transform equations tell us clearly that when we do a structure determination
protocol, the structure determination in order to determine where the electron density is,
should have structure factors not the intensities.
So, here what we have done is to shows the general equation by means of which we can
calculate the intensities, because that is what we experimentally measure in a diffraction
experiment. So, since we have the intensities that is available with us, we can use these
intensities and then determine the structure factors. The major problem which comes up,
there is the phase problem. We get the magnitude of F(h k l), we do not get the α
associated with h k l.
So, here how did we solve the phase problem in calculating the intensities? We solved
the phase problem by looking at the fact that if depending upon where these atoms are
the variation is with respect to 2π (h x + k y + l z). So, that gave us the phase issue this is,
because we know where x y z is since, we do not know where x y z is the modulus value,
which we get for structure factor is devoid of phases. No experimental way is available
with the way we do the experiment, no experimental approach can be developed to find
the phase.
526
So, therefore, we have to determine the phases or guess the phases that are associated
with the structure factors. So, you might ask a question, why not use the intensities and
do a Fourier transform. What is what happens if we do the Fourier transform of
intensities? If we do the Fourier transform of intensity, we will get what is known as a
Patterson function. We will be discussing details of the Patterson function, because that
is also an extraordinarily useful function provided, we know certain things about the
structure. See the I(h k l) gives rise to the so called Patterson function and that Patterson
function is one in, which we get the inter atomic distances as vectors. So, it is a vector
map and in that vector map, we get inter atomic distances.
So, in other words what we will get if we do the Patterson of a caesium chloride example
we have taken here it is not the positions of the atoms, but the position of the end vectors
between caesium and chlorine. We not only get the position of the vector between
caesium chlorine, we will also get vectors associated with caesium and caesium itself,
chlorine and chlorine itself, chlorine to caesium and then all the other equivalent points,
which get generated in this space group and their vectors.
So, if there are N atoms we will get N2 vectors out of which N will be self vectors. So,
the total number of independent vectors will be N2 - N. It is very easy in case of the
caesium chloride to guess what should be the Patterson map, because there are only 2
atoms in the unit cell. On the other hand, if there are 100s of atoms, we will get N2 - N if
100 atoms are there.
So, we will get N2 - N peaks and all these peaks will originate from the same origin and
therefore, Patterson is also a very useful tool, because we do not have the phase problem
there to solve; however, the tool that can be used there is to sort of extract information
from the vector map. Since, there is a severe, large number of vectors coming from a
single origin, it becomes a severe overlap between vectors and structure solution is not
very straightforward, it can be done, but it is not a straightforward methodology. So, the
generally recommended methodology is to measure the intensities, identify the
magnitude of the structure factors by taking the square root of the intensity and then
solve the phase problem to get to the structure.
So, this now is the general methodology that is available for the structure determination.
So, what have we done so far? What we have done so far in terms of the structure
527
determines. In fact we did not determine the structure, we assumed the structure of
caesium chloride and then did the substitution of various types to get to the intensity
calculations. We in fact, got the absolute intensities, because we have used the realistic
atoms inside the unit cell. The idea of this particular calculation was to show that, we can
get to the intensities given the positions of the atoms. So, even if you have 1000s of
positions you can write a computer program, which will use this for general formula for
calculating the structure factor, the intensity and it will calculate all possible intensities.
So, there is an advantage since we have fast speed computers and so, one once we know
the positions of the atoms we can calculate all possible intensities that come out from the
experiment, but that is not our motive. Our motive is to find out the positions of the
atoms. So, it is we have now got ourselves into a situation, where if we do not have the
positions of the atoms, we cannot get to the intensities. So, that is something which we
do not want to face, but we do get the intensities from the experiment.
You see, we have an experiment we performed an experiment we sent a beam of X rays

onto the crystal, the crystal did the diffraction. We know the geometry of the diffraction,
because it obeys the Braggs conditions, we know where there a particular peak has to
come with respect to the 2θ axis of the powder pattern we have a relative intensity
measured. So, eventually we should find out a way in which we can put these relative
intensities to absolute scale.
All that is possible, but will it lead to the structure determination is a question of is
question which is going to be very difficult to answer. Very-very simple structures can
be determined, but as we go to higher complexities, if we go to large organic molecules,
inorganic coordinate systems (Refer Time: 30:23) MOFs and then eventually to proteins
and viruses. It looks an (Refer Time: 30:28) arduous task.
So, the task is cut out here for us to use the intensities and use the information about the
geometry which we have derived from the Braggs law conditions and use these two
aspects to find out the structure, find out where the atoms are, the challenge lies there.
So, the rest of the structure determination protocol which we are going to develop will
depend upon, how we develop this whole theoretical approach to find out the positions of
the atoms.
528
What is the plus point? The plus point is that we have done an experiment on with X ray
diffraction techniques. So, we have the machine, we generate the X rays, I spent a little
time on how we generate the X rays, but maybe we will assume that we have the X rays
coming in, because of the number of classes limited in this course. Suppose, something
sometime is available with the TAs they may discuss how X rays are generated. So, the
X rays will be generated, they will fall onto the crystal, the crystal now diffracts and the
intensity are measured the positions of where the peaks come, the points come.
Now, can be can be indexed by in terms of the h k l values, because we know the
scattering is now giving us the reciprocal lattice. So, we use the h k l values using the
collection of h k l values, we determine the geometry of the diffraction. So, we now
know what is the crystal system and what is the various ways in which we can now
attempt to determine the space group using the information that is available on the
reciprocal lattice. `
So, that will be the next aspect, how are we going to use this information to find? What
is the space group that is available for us to handle and then we will know one, once we
know the space group you know how many molecules are there in the unit cell, because
we eventually have to solve for only 1 molecule the other molecules in the unit cell can
be generated by equivalent points, which we know very well.
So, this in a nutshell is what we will have to be doing and therefore, 2 things are
important; one is the accuracy that is associated with indexing procedure. So, if you go
wrong in that everything will go wrong, that is the first thing that indexing procedure
therefore, will tell us where the atom, where the h k l planes are coming and in what way
those h k l planes help us in understanding the geometry of the diffraction.
One once we have that handled, then we can look at the intensities, handle the intensities
by taking their square root and going over to the structure factor, the magnitude of the
structure factor. Guess the phases and then go and compute the electron density using the
Fourier transform expression, that will give us the final structure which we are looking
for. So, what then we will do is there is one issue which I always wanted to discuss
before we go further and that is something which we have been looking at earlier, in the
earlier classes the concept of reciprocal lattice.
529
Since, we have been now saying that the there are 2 lattices; the direct lattice and the
reciprocal lattice we have also formulated the relationship between the direct lattice and
the reciprocal lattice, but I think we should have a relook at this reciprocal lattice concept
a little more closely.
So, that we now are able to see how diffraction is occurring and how we can get an
access to the reciprocal lattice geometry by means of which we can index the pattern and
then the distribution of the intensities to individual reflections. Has it got any
information, regarding the space group information? Can we derive a space group by
looking at the intensities, which have come in our three dimensional reciprocal lattice,
because the main direct lattice has this information.
So, can we get that information out of our reciprocal lattice. So, we now go in over to
that region. So, before we go over to the region, where we now look at what is the
information the reciprocal lattice can give, with respect to the space group determination
and in what way we measure the intensities and how we put it put them on the relative
scale. All these aspects now are automatically done on diffractometer. I always keep
selling that telling all the time that crystallographers are their own enemies that is,
because they have made it.
So, very simple now a chemist or any other person who wants to determine a crystal,
what all he has to do is put the crystal on the machine and wait for some time and maybe
a few hours nowadays, in order to get to the final structure, by just sitting at the keyboard
and pressing enter buttons whenever it is asked for. So, effectively the thing is that we do
get information regarding the structure.
Without knowing any of the background we have here and that is where the
interpretation of the determined structures becomes a challenge to chemists, to physicists
to engineers and so on. And that is very important, because one once we have the
structure information available, it is always related to the property, the material exhibits.
So, the next step therefore, will be to see, what information we can get from the
reciprocal lattice analysis. So, we will go into the details of the reciprocal lattice as we
go further.
530
Prof. T.N. Guru Row
Lecture – 40
Conversion from Direct to reciprocal space, the inverse relations
We will now look into a few more aspects of reciprocal space and reciprocal lattice. So,
effectively we look at more on Bragg’s law and reciprocal space.
So, we have 2dsin θ = nλ, that can be written in a slightly different way by saying sin θ =
(nλ/2)*(1/d). Now why did I write like that, we will know a little while later, because it
is essentially to do, deal with how the appearance of the reciprocal lattice is, and how we
can get to these various values of sin θ from the d values for a given λ.
So, that what we have done is, we have already discussed the reciprocal lattice. So, some
of the points which we should note from the Bragg’s law are the following. So, in order
to keep this discussion going and also to keep you afresh, as we go further and further, is
to remind you of the following points; one is sin θ is inversely proportional to d, so this
is seen here. Sin θ is a measure of the deviation of the diffracted beam from the direct
beam, it is evident that structure with large d will exhibit compressed diffraction patterns
and conversely for small d. Let me take a little time to explain this. Suppose we have a
structure with a very large d spacing, this is equivalent of having a multiple slit
531
experiment which we discussed earlier on, we discussed of course, only a two slit
experiment, and then we decided to this increase or decrease the distance between these
two slits.
We can have any large number of slits, many slits, some 8 or 9 slits. So, I can show you
a little diagram of my own which will be very lousy, but I will try to draw it. Suppose
you have a source here which is a light source and then we keep a slit in front of it, keep
a screen in front of it and then put holes through the screen; 1, 2 and 3 and so on. At
equidistance, my diagram does not show equidistance, but you assume that they are all
equidistant to each other. Then as we discussed earlier on each one of them now
becomes a source, so you will get spherical waves coming out from each one of those
like that. And whenever there is an intersection here we will get a bright spot on the
screen, which we are going to put here.
So, we will get therefore, the bright and dark images corresponding to this, like that and
we know that it is an inverse relationship. So, if the distance is d between these two. If
that distance between these two points is d and that is repeated in all in the two
dimensional plane, then we will get a large number of indications of bright and spoil dot
lines here; that is the interference theory. So, what is said here is an issue where we can
have different spacings of these slits or holes. So, each hole is a source, so instead of this
we can have a diffraction grating which we have already discussed before, or we can
have a crystal which is essentially a three dimensional grating; and therefore, depending
upon the type of d spacing we have.
If we have a larger d spacing what does it mean? What does it mean if you have a larger
d spacing, suppose 1 0 0 is a very large value then 1/d hkl will be a very small value. If the
value of a dimension is very large, suppose it is 20 Å then the corresponding 1 0 0 plane
will be at 1/20. So, as the unit cell expands, the expansion of the unit cell can go
commensurate with the number of atoms which are put inside the unit cell. So, we can
put let us say a protein molecule, the cell dimensions could be 100, 200, 300 or
whatever. And they also go to their respective crystal systems depending on the
symmetry with which they crystallize. On the other hand if you have a just an atom, the
unit cell dimensions will be fairly small. So, for a smaller value, the 1/d value will be
very very large and for a larger d value the 1/d value will be very very small.
532
So; that means, for larger unit cells the number of points which you get will be a large
number compared to smaller unit cells, you will get a smaller number. So, that is what is
meant by this statement. So, the statement looks very profound, but essentially what it
means is that, we get a large number of reflections from a larger unit cell, and reasonable
number of reflections from a reasonable unit cell and less number of reflections from a
smaller unit cell. Obviously, this is not going to hurt our approach, because our approach
would be to see as many information, as many scattering directions possible we will
identify; that means, as many d values we can accumulate, the better is the result.
So, the idea of looking into reciprocal lattice further now will tell us how we can get this
information and how useful that information can be. So, we therefore, talk about the
construction of a reciprocal lattice. So, it is based on 1/d. So, we are talking about the d
spacing here; that is the spacing on the on the screen here, 1/d is the spacing which
corresponds to the reciprocal lattice. So, when we get the reciprocal lattice image, the
reciprocal lattice image is we know that it is proportional to 1/d; that is why the equation
here is written in terms of 1/d right.
So, let us see a little detail of how these issues come up, let us take a unit cell. So, let us
take a unit cell here, this is let us say represents the unit cell, this is the origin of the unit
cell. So, from the origin of the unit cell if we now look at the planes which appear; so,
suppose you take this as 0. So, 1 unit along this direction you have this plane, and 1 unit.
533
So, if you consider now the plane which is 1 unit away at 0 then we call this plane as 0 1.
We are looking into two dimensions, so this plane will be 0 1, the index of this plane will
be 0 1. So, we are looking at the diagram in the with respect to this background.
So, we start from 0, this is the direct space we go up to that point which is 1 unit and then
1 unit along this will represent the 0 1 direction. So, if you now consider 1 unit along this
and 1 unit along the second direction then you will get a 1 1 plane, that 1 1 plane is
represented in this fashion. So, 1 1 plane is now going through this point. So, the
similarly 2 1 plane is now half way along this. Remember the bread analogy, this is half
way along this and 1 unit down this, so it will be 2 and 1. So, if you consider this plane
this is 3 and 1 ok. So, we now, therefore, have represented 0 1, 1 1, 2 1, 3 1, the 4 planes
with respect to the direct lattice.
So, we have a direct lattice with as a have indicated the lattice diagram in the
background, we have drawn the lattice net. And using the lattice net we have identified 4
positions 0 1, 1 1, 2 1, 3 1. These are the planes passing through the with respect to the
origin of the main cell. Now, the next step that is required, is that. So, what we have done
is, next step is to draw normal’s to all possible direct planes from the origin.
So, we have drawn now 4 direct planes 0 1, 1 1, 2 1 and 3 1, and we are now going to
draw the all possible normal’s to these draw normal’s. What do we mean by normal, it
should come at 90 ᵒ to the diagram which we have drawn. So, you get a 0 1* a 1 1* a 2
1* and a 3 1*, so these are now perpendicular. So, you draw the perpendicular to 0 1,
perpendicular to 1 1, perpendicular to 2 1, and perpendicular 3 1.
So, these are now the star the stars are now the points which are normal, which are drawn
with respect to the various d hkl . Now where do we stop it? We have to have a stopping
point and the stopping point will be 1/d . So, the length of each of these now we
h k l
stopped at 1/d . So, we know the value of the d

hkl hkl now. And so at 1/d hkl from this
origin this is now the origin of the reciprocal lattice. So, the origin of the reciprocal
lattice is here, so we now draw these lines. Normal to the various planes which we have
shown intersecting at that point; so, this therefore, will give us the collection of all the
reciprocal lattice points which are indicated by star mark.
So, reciprocal lengths and angles are designated by a* b* c* α* β* γ* respectively,

similar to the star marks on these values, so these are the 1/d h k l values. So, what we
534
have to see also here is that whenever we consider planes in the two dimensional direct
space, the plane 0 1, the plane 1 1, the plane 2 1 and is a two dimensional case. The
normal’s now are ending up with a point and therefore, they are represented by points in
the reciprocal space.
So, when you get a diffraction pattern, we have already discussed that in the previous our
that way with a diagram. Here again this diagram is making it more clear that from a
diffraction from a plane, is essentially like reflection, and so when X rays fall on a plane,
any point on that plane on which it falls will scatter at the same phase and as a result the
scattering direction is determined and that scattering direction now is fixed at 1/d h k l
which now forms a point and that is the reciprocal lattice point.
So, if you take therefore, a collection of all these reciprocal lattice point. In principle it
also represents a three dimensional lattice and therefore, we call it as, not just the points
in the reciprocal space, we call it as a reciprocal lattice. So, given the direct lattice and
the planes associated with the lattice, since the planes are associated with the direct space
lattice. The points which are now normal’s to those planes which are now the collection
and the representation of the reciprocal space, in reciprocal space, they define the
reciprocal lattice.
So, therefore, the reciprocal lattice, reciprocal points also form a three dimensional
lattice. So, this is very important, they do not form a two dimensional lattice, they form a
three dimensional lattice, because this direct space, these normal’s now are with respect
to the planes. So, the planes are all in various orientations with respect to a b c. And
therefore, we have the reciprocal lattice characterized by a* b* c* and α* β* γ* or the
interaxial angles.
So, how does that develop? In what context we can see they are developing into these
three dimensional lattices in various crystal systems. So, we will examine that a little
carefully and closely, because that is something which will clearly tell us how this
concept of reciprocal lattice has developed. It is not just that you know if you have a =
100 Å, 1/a how is it getting represented, will it become 0.01 something. In fact, this
question has been asked when we have taught several times by students. So, we thought
we will make it clear in the context of how this is really coming up, and these two
pictures will tell you clearly how these develops.
535
The first picture tells us that we have a unit cell and this particular unit cell is described
with respect to the origin here. This is an orthorhombic direct space lattice. So, we have
therefore, Oa is the distance along the x direction, Ob is along the y direction and Oc is
along the z direction. So, by definition the plane up here is 1 0 0. So; obviously, by the
plane if you consider a planar half point, it will be half that is 0.5 0 0 the position and
this plane will be identified as 2 0 0; that is the Miller index revision.
So, we can therefore, have this as the unit cell. So, this represents now a b c α β γ. α β γ
in this example is 90 90 90. So, how does the reciprocal cell come up? Reciprocals cell
in this particular case comes up by this definition. The table d below shows, the
relationship between the direct space in the reciprocal space associated with a
orthorhombic unit cell.
So, the unit cell dimensions are a b and c, the corresponding dimensions will be a* b*
and c*. So, the point to note here is that the longest side in the direct cell; that is a here in
this example, will be the shortest in the reciprocal lattice. So, we take the same origin,
the origin is the same for the direct space lattice as well as the reciprocal lattice, and we
now show the picture of the reciprocal lattice; that is we have already discussed and
found out it is also a three dimensional lattice. And so if there is a behaviour of atoms
and molecules in such a way, that they are restricted to the orthorhombic symmetry. The
536
symmetry that is reflected in the direct space must also be reflected in the reciprocal
space.
So, we therefore, indicate this unit cell if this is a*, the short one becomes longer it is b*
and this third one is c*. So, this will be now the unit cell of our reciprocal lattice. The
fact that α β γ are 90 ᵒ, will ensure that these values are equal to 90 ᵒ. What is also
important to note is that they are interconvertible; that means, if you have a* b* c* as 1/
a 1/b 1/c, the corresponding a b c will be 1/a* 1/b* and 1/c*. So, what happens to the
volume of the unit cell. The volume of the unit cell is a . (b * c), and therefore, we can
now calculate the volume of the reciprocal lattice which is 1/V and that is a* b* c*.
So, the volume also is invertly inversely related to each other and that is how this
relationship develops; a very straightforward case in case of the orthorhombic symmetry.
Any symmetry higher than orthorhombic symmetry, particularly if it is cubic can also
have this kind of a situation or tetragonal will be a similar situation, because the angles α
β γ are 90 ᵒ with respect to each other. So, crystals belonging to orthorhombic tetragonal
and cubic will have this feature; a* is equal to 1/a, b* equals 1/b, c* equals 1/c and so
on.
Remember in the case of tetragonal and cubic in particularly in case of tetragonal a = b.

So, your a* and b* will be 1 and the same from the distance point of view, but they are
oriented at 90 ᵒ with respect to each other. Same thing they extended to cubic system a =
b = c and therefore, you will have 1/a 1/a 1/a as the reciprocal lattice values for the
reciprocal lattice in the cubic cell.
So, otherwise these crystal systems will have this kind of a behaviour. So, this is the
direct space lattice and that is the reciprocal space lattice and the relationship between
them is shown here.
537
(Refer Slide Time: 17: 27)
Now, we go to the discussion on the next crystal system which is a monoclinic system.
So, in the monoclinic system we have a little bit of a deviation, the deviation comes
because of the fact that β ≠ 90 ᵒ, considering b axis as the unique axis. So, when once we
consider b axis of the unique axis, the angle between a and c will be β. So, β is the angle
between a and c which is indicated in this diagram. So, we have the x, y, z directions
indicated here, the x direction is taken up by a, the y direction is taken up by b and the z
direction is taken up by c.
So, this volume of the unit cell will therefore, be a . (b * c), there is no change in that.
Remember I am using the vector notation, so when I say volume of the unit cell is a . (b
* c), you should remember that there is a certain relationship between the involvement of
β as well in this calculation. So, let us look at what happens to the reciprocal lattice in
this situation. The reciprocal lattice will be now differently oriented than the case of an
orthorhombic system, it is not going to be coincidental with the directions of a b and c,
because now what happens is, a* now will become the reciprocal lattice will now be 1/a
sin β; that is because we have to now consider the normal to the plane. You remember
the previous picture where you saw four normal’s being drawn to four different planes.
So, we have to consider the normal to the plane and therefore, if we consider the normal
to the plane, this is the plane which is now represented, this is ac projection and
therefore, if you consider this is the plane, this will be now the plane corresponding to,
538
the plane corresponding to the c axis. And therefore, if we now consider the d value
associated with it, the d 0 0 1 is perpendicular to this direction. So, therefore, this angle is β
and we get the d 0 0 1 which is the normal to the plane, not associated with the c axis.
The normal to the plane is 90 ᵒ with respect to the plane, and therefore, we come get this
issue that we have now 0 0 1 in this direction, 1 0 0 is in this direction, because this is the
1 0 0 plane. If you consider here this is the 1 0 0 plane. So, what you have done is you
have taken the origin and drawn a line like that and that is1 0 0. And similarly along the
b axis you have taken and drawn a line like that; that is d 0 0 1.
So, these two now well be making an angle and because of the fact that we have a β
angle between a and c, the angle between d 1 0 0 and d 0 0 1will also follow β, but now that
will be β*; that is because it is going to be 180 - β. The reason is obvious, the reason, it
would have been ideally β equals β if it were 90 ᵒ. Since we now deviate from 90 ᵒ this
diagram is showing an obtuse angle greater than 90 ᵒ, the β* will now be an acute angle.
So, that is because it is greater than 90 ᵒ, 180 - 90 greater than 90 ᵒ will be shorter than
90ᵒ. So, we therefore, get the β*. So, the conversion from a* to a to a* will be a* equals
1/a. Now since this is involving β we bring in the issue of sin β. How does that come
about? this is drawn at 90 ᵒ, so this angle will come into picture, the β angle and you see
that these two are at 90 ᵒ and these two are at angle β, and therefore, you get a * sin β,
because the 90 ᵒ angle is here and this is the value which we have to calculate, this is the
angle β.
And therefore, we get 1/(a sin β), b* will be remaining the same, because this is the
direction of the twofold axis; that means, that both α and γ are 90 ᵒ with respect to each
other. So, it will remain as such like in the case of an orthorhombic system. So, b does
not change b* will be 1/b. Whereas, c* will follow the same diagram as we have here.
So, we get c* is equal to 1/(c sin β). So, this now is the representation of the reciprocal
lattice associated with monoclinic systems. So, if this is the case then what happens to
the relationship between these two. The reciprocal relationship will still remain and in
fact, this should be β* is little error here, maybe I will correct it right away.
So, this will be β*, same as this. So, a is equal to now 1/(a* sin β*), 1/b* and 1/(c* sin
β*). So, you see that the volume of the reciprocal lattice is now somewhat shrunk with
respect to the volume of the unit cell, because unit cell volume is abc sin β. it is it
539
depends upon therefore, this diagram. In this diagram you will see that the volume of the
monoclinic cell with a β value greater than 90 ᵒ will have a certain volume, and that
volume will be different from the volume of the reciprocal lattice. I just want you to
think about it and tell me which of these two volumes are bigger in terms of the value
values you have, because V* will be a* b* c* sin β*.
Maybe you can take it as a numerical example, let us say a = 10, b is equal to maybe b =
5 and c = 12. So, take a unit cell a = 5, b =10 or a = 10, b = 5 and c = 12, and the β angle
you take it as 95 ᵒ or 98 ᵒ. So, let me give you the values a = 10, the unit cell dimension a
is 10, unit cell dimension b is 5 and unit cell dimension c is 12 ok. And β angle you take
it as 98 ᵒ calculate the volume. Now I want you to check whether the reciprocal lattice
volume is smaller or larger ok.
So, you calculate for calculate the a* value b* value c* value and find out what is the
value of V* and see whether you can get to the value of V*. So, what is important is to
remember that β* is of course, 180 - β, so that will help you in this calculation. The
reason why I want you to do it yourself at home is just to get an hang of how these
relationships come up between the direct space and the reciprocal space. And how the
unit cells now changeover in case when we have the angles other than 90ᵒ; I am not
going to discuss the next crystal system in full detail, because; obviously, you see with
one angle being away from 90ᵒ it gets a little complicated, the case of triclinic is
extremely complicated.
So, I will show you the diagrams which luckily for me is taken from the book of Stout
and Jensen which is a nice picture. In fact, all these pictures are from that book, but the
a* b* c* values and α* β* γ* values associated with the triclinic system, we will just take
it for granted and that is in the next slide.
540
The next slide therefore, shows you the triclinic direct and reciprocal cells, you see the
complications. I did not want to show you and then say it is complicated, you already
anticipated the complication. So, we therefore, have in this case a* b* c* values getting
values of this kind, where you see that the volume of the unit cell still controls it. I think
this diagram is already sort of giving way to what problem I gave you by just looking at
it, but I want you to work it out.
So, take the case of a triclinic cell which is, this is a this is b that is c, α β γ are all
different from 90ᵒ. And you see now again what we have to do is to draw normals to
planes. So, when we draw normals to planes these planes are not associated with any 90ᵒ
angle and therefore, they are acute or obtuse depending on α β γ being acute or obtuse,
you will have the reciprocal lattice appearing with something like this.
So, it is not that when you say reciprocal lattice. The idea is to, you know this nicely
done by is the book Stout and Jensen. So, nicely done in the book Stout and Jensen and
so because, the idea is to give you that there is a dimension for the reciprocal lattice. You
may think if there is a = 100 Å , b = 200 Å, then the reciprocal lattice volume will be
very very small, because its 1/a* 1/a and 1/b; that is the logic. The logic is shown here
the volume of the unit cell will be very different the of the reciprocal lattice cell. And
you can calculate the values of a b c a* b* c* by using these expressions.
541
And these involves so much of trigonometry associated with it, the orientation angles
come in and those go as products of cos α, cos β, cos γ and therefore, you will get these
expressions. You will see the volume now which is simply again written as a . (b * c) in
vector notation, will be equal to a b c and square root of this quantity which involves the
cosine square functions and also the cosine functions associated with α β γ. So, basically
the take home lesson here in the last few minutes which we have our discussions are, is
that there we have different. So, I want you to also work out the case of hexagonal and
hexagonal system ok, hexagonal and trigonal (Refer Time: 27:44), in both cases we have
an angle 120 ᵒ.
So, it is effectively like monoclinic, so that is why I have not shown you, it is almost
effectively like a monoclinic system, except that there are more conditions than in a
monoclinic system. So, the angle being now γ =120 ᵒ are not β, the unique axis is along
the c axis. So, the diagrams and the way in which you work out the direct space and the
reciprocal space vectors and their corresponding volumes will be like very similar alike
to (Refer Time: 28:16) what we do with a monoclinic system.
So, let me recapitulate what we have done today in this case. We started with the use of
the Bragg’s law and the reciprocal definition. We indicated that as we increase the
distance between in the between the planes, we find that the distances between the
reciprocal lattice points shrink and vice versa. And we also said that the one once we
have the 1/dhkl values we can directly generate the reciprocal lattice, and we showed with
examples the generation of the reciprocal lattices with respective various planes, and
then we said that in case of the orthorhombic system, the monoclinic system we
calculated the direct space and the reciprocal space relationships, and also assumed the
values associated with the triclinic system.
This is like the, all of it has been taken from the book of Stout and Jensen in this
discussion and I am very grateful to the that book, because drawing these diagrams will
be a very difficult task for me. The way I draw things you have already seen, it becomes
a very difficult task for me. So, what do we do next?
542
Next we will see what happens to Bragg’s law in reciprocal space. This is something
which we will take up in the next round.
543
Lecture – 41
Diffraction and Reciprocal Space (Continued)
So, having seen the relationship between direct space and reciprocal space in several
crystal systems, different kinds of crystal systems and the way in which we can derive
the values of the reciprocal lattices. We could do a few numerical problems, but then
instead of spending more time on numerical problems, because we have to cover the rest
of the course content, maybe we could give it as possible exercises during the course
evaluation. So, what I will now do is to redo this Bragg’s line reciprocal space, we have
already done it once.
We have indicated possible occurrences of the Ewald’s sphere and the presence of the
limiting sphere, and then we showed that so many reflections will come up in a
diffraction experiment. Only because of the fact that we just do not keep the crystals
steady, but we rotate the crystal. And as the crystal rotates the reciprocal lattice will
rotate with it, and then we have this, the limitations coming up due to Bragg’s condition.
So, this particular slide, we will tell you the entire construction procedure which we do in
order to understand the direction and in the reciprocal lattice of the diffracted beam. So,
544
we therefore, consider this as diffraction of X rays. In fact, I should put a, should have
put a bracket here Bragg’s law in reciprocal space.
Now, as I mentioned again and again that Bragg’s law in reciprocal space is more crucial
for us, because what we record is I vs 2θ and from 2θ we can get information on the
reciprocal lattice and eventually the direct lattice. So, we take now for example, a
reciprocal, a direct space lattice, sorry a reciprocal space lattice. Now this reciprocal
space lattice is represented by two vectors. This is a 2 dimensional reciprocal lattice. So,
b* is coming towards us in this from the plane. And so in the plane we have drawn the
reciprocal lattice. We have drawn the graph of the reciprocal lattice; that means, we have
taken O as the origin. And with respect to O as the origin we have two vectors; a* and
c*, and the angle between a* and c* will be γ*.
So, this is the representation of the reciprocal lattice in the background. So; that means, if
you want to find out and index this parts in the a* c* plane, how will you do that? We
will do it like this, you start from 0. Suppose you go along the a axis in one direction in
the positive direction, that particular index will be 1 0 0 in the reciprocal lattice, 2 0 0 in
the reciprocal lattice and so on. And in the c* direction you can have 0 0 1, 0 0 2 and so
on.
So, this point therefore will be 1 0 1 right, you got the point. So, this is how we can go
and index the various spots. So, they occur in different positions in the background, and
now we will see what happens when we send in X ray beam. So, there is an X ray beam
which is coming in this direction of XO and XO is go through this direction and I am
going to put a crystal at C. Origin I have taken here, but I will discuss that a little later,
but at this moment let us say we put a crystal at C.
So, if this is the direction in which the unit vector of the incident beam is coming, this is
S0 and this is the direction at which it goes out, it is S 0 and this is S, this scattering vector
the angle between them is 2θ by definition. So, we have given this as 2θ.
So, this is in fact, the CP is the direction of our R vector, the R vector. It is also the
direction of the H vector, if it cuts the reciprocal lattice point. So, this is as far as the
diagram is concerned. So, we have now CP and there is a reciprocal lattice point cutting
this. Apart from that, one more thing is cutting that reciprocal lattice point what is that?
545
That is the one which is the so called Ewald circle in this particular case, but when we go
for 3 dimensions it becomes the Ewald’s sphere.
So, we are now discussing this issue again in 2 dimensions, because the reasons which
we have been discussing on and off again. So, the 2 dimensional network is shown in the
back; that is the reciprocal lattice network. We have the cell dimensions a* and c* and
the angle γ* between them, X rays are passing through and there is a crystal kept here so
this scattering takes place; otherwise of course, we have the direct beam going through
and through. Now we sit at the point C and then take the radius of this circle as 1/λ,
where λ is the wavelength which we are using.
Let us say we are using copper kα radiation, it will be 1.542 Ås or 1.5418 to be exact. So,
you take that, the value as 1 divided by that quantity, take that as the radius of this circle
and draw this circle. Now this is known as the Ewald circle, this is called Ewald
construction and this known as the Ewald circle. So, you see that the origin of the
reciprocal lattice point is now taken at on the circumference of the circle at this moment.
So, the origin of the reciprocal lattice is taken at the circumference of this circle.
Now, we have said that the crystal is up here and this diffraction direction is that, with
respect to that the angle is 2θ. Now what is happening is that if you look at this circle and
see where are all the points where the reciprocal lattice is intersecting with the circle. We
find at least P is definitely intersecting with the structure, with the circle. So, whenever
such intersections takes place by definition, we will have diffraction. So, whenever there
is a reciprocal lattice point in this reciprocal mat, which cuts the circumference of the
Ewald’s sphere which is drawn with a radius of 1/λ, that particular point now will be a
point from which diffraction can come out.
That means it will be a point corresponding to the vector H, the H which we have
discussed. And therefore, we see that there will be a diffraction in this direction. And
since the diffraction is then this direction we have to work out what is the geometry that
is associated with this diffraction condition, and that should be the same as Bragg’s law,
because Bragg’s law said what, Bragg’s law said that if there is a plane and then incident
beam, the direction of the incident radiation characterized by vector S, the scattering
vector say characterized by a vector an incident is characterized by S0 and this scattering
546
by S. We would have had this kind of a discussion, and therefore, it will now give us the
diffraction condition satisfied, 2 d sin θ= λ.
So, what we have done now is to go to reciprocal lattice. The once we went to the
reciprocal lattice we decided to take 1/λ as our unit and the circle is drawn with a radius
1/λ. Now in this circle there are certain properties, the properties are the following. The
angle OPB is an angle in a semi-circle and angle in a semi-circle is always 90⁰ from our
school days. So, therefore, OPB is 90⁰.
Now you see that the sin of OBP, if you consider this as OBP, the sin associated with
OBP that is with respect to the angle θ. So, this will be the sin θ. If you consider this
triangle the sin θ is equal to, if you consider the triangle OBP. So, we are looking at the
triangle OBP; one of the angles is 90⁰ this angle is θ. So, what we are now looking at is
the sin of OBP, it will be OP which is this distance from the reciprocal lattice origin, OP
divided by the diameter of this Ewald’s sphere, Ewald’s circle which is equal to 2/λ. So,
it will be therefore, OP/OB which is OP/(2/λ).
Now, the thing is OP as by definition is a reciprocal lattice point and that is cutting this
Ewald’s sphere. So, this OP, the distance from the reciprocal lattice origin to P is 1/d hkl.
Therefore, we can write sin θ which is this as equal to 1/dhkl which is OP, and then you
see slash 2/λ, it can be written as a multiplication factor λ/2. So, this gives rise to 2 d sin
θ = λ. So, you therefore, derive the Bragg’s law in reciprocal space in the following way.
I suppose with the explanation given before and the explanation on this diagram it is
clear now, how to work out the Bragg’s law in reciprocal space. Now there are two
issues which we have brought up here; one is we said that we could keep the crystal at
the centre then we said the origin is in the on the circumference of the circle and so on,
instead of that we can keep one definition.
We can either keep the at this moment with the discussion we have done and with the
derivation we have done, we have assumed the origin of the crystal and the origin of the
reciprocal lattice to be at O. But we will see this diagram which will tell us how to
reorganise the argument with respect to the diffraction from a plane. So, what is
happening is now the X rays are coming in and let us say we have a crystal here.
547
The crystal is consisting of a large number of planes and one of the planes is shown here
which satisfies the Bragg condition. So, when once it satisfies the Bragg condition it will
generate a H vector in the direction OP. So, we have therefore, a scattering direction OP
which is now is the reciprocal lattice vector. So, this is with respect to O as the origin.
So, what will happen now is that since this angle is θ, the other angle is also θ because
this inside angle is 2θ. So, we have a plane, it is as though there is a plane and then X
rays went in there and a reflected. So, the reflection direction is along the OD. What you
see is that OD and CP are parallel to each other, and therefore, this now represents your
scattering direction D, OD could be now equivalent of CP.
If you are putting a crystal at C, CP and OD are the directions of scattering and that
therefore, is clarifying the fact that it is at an angle of 2θ. So, whether we keep the crystal
at O or whether we keep the crystal at C, it does not matter, because what we do is we
measure it at very long distances.
We are talking about here the Å inverse lengths in the case of the reciprocal lattice. So,
the distance here is in very very small quantities, and what we measure is a diffraction
which is coming out from the crystal is at a distance which is far larger than the 1/λ
which is involved here, and therefore, whether it comes from this point or that point is
immaterial. So, for all the rest of the discussion one this can say that crystal is at the
centre of the Ewald’s sphere. So, this is the derivation of the Bragg’s law in reciprocal
space
548
So, having seen this we will go further and see what are the ways in which we can now
utilize the fact that these reciprocal lattice points will cut the Ewald’s sphere, and
whenever this will happen at the Ewald’s sphere the diffraction condition is satisfied, so
we get a reflection. So, now, we have the 1/λ into the picture; that means λ into the
picture, the wavelength of light. The wavelength of light we use is anywhere between, let
us say 0.4 to 2.5 Ås, they will give similar results. But what is important is, what is the
influence of this wavelength on the amount of data which we can get, because we want
to get as much data as possible in our diffraction experiment.
So, if we go back to this position, we discuss the issue that this is a stationary a
reciprocal lattice, and in the stationary reciprocal lattice we have occasional intersections
of the Ewald’s sphere with the reciprocal lattice, and whenever that happens we get
scattering. But when we rotate the crystal about the centre of the diffractometer and the
crystal rotates, as the crystal rotates the reciprocal lattice also rotates with it.
So, as the reciprocal lattice rotates different points on the reciprocal lattice will have to
intersection with the Ewald’s sphere. So, what we have to therefore consider is, those
reflections which come in a sphere of radius, double this one. That means those which
come in a sphere of radius 2/λ , because the radius of the Ewald’s sphere is 1/λ. So, this
will be 2/λ and this 2/λ will be the, what we call? Excuse me.
549
So, what we were looking at is to see now what is the wavelength which we can use,
under what circumstances we can make sure that we get maximum number of
measurements, because the whole logic of this X ray structure determination depends
upon how many independent reflections you can get in a diffraction experiment. This is
the advantage of also the solid state and particularly the crystalline state, because it not
only gives the possibilities of diffractions parts, it gives you an option of having a very
large number of diffraction spots compared to what is it we have to determine. What I
mean to say is, suppose there are 10 atoms in a molecule and this molecule goes into a,
let us say a space group P 21/c, Z is equal to 4; that means, we have to determine 40
atoms.
The determination of 40 atoms will mean three coordinates associated with the atom.
And since atoms are not rest in a given crystal material, we have to determine their
thermal effect parameters also. So, the thermal parameters are not isotropic, because
atoms are connected to each other in all three directions, so they will be anisotropic.
They are not isotropic they will be anisotropic. And as a consequence there are 6
parameters which come up to describe the atom in the form of an ellipsoid. The
uncertainty that is associated with determining the position of an atom therefore, is
within that ellipsoid, and that is the one which controls the deviation or the effect of
thermal vibrations which are associated with atomic positions.
So, when you say that you have determined a position of an atom at x y z, there is a
certain uncertainty that is associated with it, because the atom is not sitting there in the
next moment, but your diffraction experiment is a static experiment and therefore, we
have to find those error bars. Those error bars are expressed in terms of thermal
parameters, so that will be 6 parameters. So, x, y, z plus 6 thermal parameters will be
about 9 per atom. So, we have to determine totally there are 40 atoms in the unit cell, but
we have to determine 10 position atoms, 10 atoms in the asymmetric unit. So, 10 times 9
will be 90 plus 1 which is now going to put the measurements on an absolute scale, the
so called scale factor.
So, totally 91 parameters have to be determined. The way in which we are going to
determine and improve the final structure depends on a procedure called the least square
refinement. And the requirement of the least squares refinement is that the number of
parameters which we have to determine must be very very small compare to the number
550
of measurements we can offer in this experiment. The measurements we can offer in an
experiment in a diffraction experiment or the intensities, and the location of them in the
reciprocal lattice which is the h k l.
So, the number of planes therefore, which we can locate and measure the intensity from
should be reasonably large. In fact, the ratio is anticipated to be atleast 10 for a very good
structure determination. That means, the number of reflections or number of observations
we should have, the number of I(h k l) we should measure which is independent of the
diffraction condition that should be at least 10 times more than the number of parameters
we have to determine. And this is a logic which therefore, tells us that larger the number
of measurements, the better we are. Now here is a situation since our Ewald’s sphere is
1/λ, the chance that is given to us to expand the reciprocal lattice is in the times of tuning
the λ value.
So, if one as synchrotron radiation we can tune the λ value, but in a typical laboratory
experiment we normally have a copper radiation. So, copper radiation will be 1.542 Å
sorry 1.5418 Å. So, that will give raise to a reciprocal lattice of a certain volume and
dimensions. And the corresponding Ewald’s sphere which we construct will be 1/λ, so it
will be 1/1.542. Suppose, we now reduce the value of λ, then 1/λ will keep on increasing.
So, the shorter the wavelength, the more reflections in the reciprocal lattice will fall. So,
therefore, the more reflections will fall and more reflections will satisfy the Bragg
condition. That means, if we use shorter wavelengths, shorter λ we will have a larger
sphere of reflection and for the smaller λ or a long λ, longer values of λ we will have
smaller sphere of reflection.
So, the larger this sphere of reflections the better it is. So, between copper and
molybdenum, molybdenum is preferred radiation for determining the crystal structures in
general. Of course, we can use silver radiation which is much more superior. We can
also use radiations of different kinds or we can go to synchrotron radiation and tune the
wavelength. There are other practical aspects when we do this kind of a thing; that is the
design of the instrument itself.
For example, let us say we, because of this logic we go to very very short wavelengths
ok, less than 0.2, 0.3, 0.1 and so on. Then the accessibility volume will anyway increase,
but our measurement device and the way in which we design the experiment, the
551
diffractometer construction, a mechanical device which is moving the arcs to measure
the distances and also the detector capability it limits that value at which we can collect
the data. So, normally it is recommended that we use wavelengths of that particular order
which will resolve the distance between the two atoms.
So, the most preferred wavelength if you are using the laboratory X ray source is
molybdenum. The other issue is of course, the absorption and things like that which we
will be discussing in the next few, after probably a few classes, is the fact that we will
have absorption, we have in total internal reflection leading to what we call as extinction
and such effects. And these effects will also destroy the quality of the intensity we get
number 1, and also the quantity of the intensities we can measure.
And therefore, these are all issues which will come up, but in general it is better to use a
shorter wavelength to access larger regions of their reciprocal lattice, so that we will get
more and more reflections that can be measured within the limits of our Ewald’s sphere
limitation; that is which will satisfy the Bragg condition. And that is why this
construction which we discussed in the previous slide and there the picture which is
shown here, is self-clear, is very clear to indicate that the shorter wavelengths are the
ones which are preferred.
So, the way in which we can express this will be in a slightly different way that will be
the way in which we can calculate. In fact, we can calculate how many reflections can
552
come in a given experiment. Suppose we take 1/λ as the Ewald’s sphere and 2/λ
therefore, will be the limiting sphere. All those reciprocal lattice points, so we put, let us
say we embed the reciprocal lattice behind this particular sphere or through the particular
sphere.
In this example let us say its 2 dimensional, so we put it behind this in the background,
that reciprocal lattice will rotate with the as the crystal rotates and different points in the
reciprocal lattice. When they intersect with the sphere of reflection we will get the
diffraction condition, satisfied and therefore, we will get to the measure of the intensity.
So, what is the number we can access? It is quite interesting to calculate the total number
of reflections we can access by this approach. The value of the number of reflections
which we can approach depends upon two things; one is of course, the volume of the
limiting sphere we have, and the other is the volume of the reciprocal lattice, because
volume of the reciprocal lattice is the one which will tell us how many Bragg’s spots we
have. So, the number of Bragg’s spots we have depends upon the volume of the
reciprocal lattice; 3 dimensional. And this therefore, is the sphere the limiting sphere
now, the volume of the limiting sphere is also a consequence.
So, the volume of the limiting sphere is four-thirds π r cube by definition, any sphere the
volume of the sphere is (4/3 π r 3). In this particular case the value of r is 2/λ, and
therefore, (4/3 π r3) will be this quantity, so divided by the volume of the reciprocal
lattice. The value of (4/3 * π) and this quantity is fixed ok. So, except that the variability
variable here is only λ.
So, we can recast this expression by writing N is equal to 33.5 which is actually
calculated as (4/3 * π) times 2 is 33.5 divided by the volume of the reciprocal lattice
times λ3 , because λ is in the denominator here, it comes down here, so it is multiplied by
λ3.
So, then what we can also rewrite this expression in terms of the direct cell, because we
can measure the volume of the direct cell or volume of, it does not really matter, but this
is 33.5 *(volume of the direct cell/ λ3). So, N is the total number of possible reflections
that can be observed, if there is only one point per unit cell that is of course, per unit
reciprocal cell. So, we can calculate the total number of such reflections that can come.
553
Now, actually we can do this exercise now. So, if we take for example, the λ value, let us
say is 1.542 or 1.5418 Å that of copper. And let us take a unit cell which is 8 * 10 * 20 is
orthorhombic let us say. So, a = 8, b = 10 and c = 20. So, what would be the volume?
Multiplication of a b and c if it is an orthorhombic cell; so, 8 times 10 is 80, 80 times 20
is 1600. So, the volume of the unit cell will be 1600 Å3 .
So, if you now substitute the volume of the unit cell here; 1600 Å 3, this will be therefore,
1600 * 10-24 that is the Å3 in centimetres. And the value of λ is also now expressed in
centimetres will be 1.5418 * 10-24 in Å ok. So, we can forget the 10-24, so 33.5 volume of
the direct cell divided by λ3 .
So, if we calculate that value of N, N turns out to be from the calculation you will see
yourself it is 14,700. So, it means that we can measure 14,700 reflections in principle. If
we use the Cu kα radiation in a unit cell, which is now consisting of 8 * 10 * 20 as the
unit cell dimensions; obviously, the unit cell dimensions 8 * 10 * 20 in an orthorhombic
system. Let us say we have ¼ all general positions and let us say this space group is
212121. If there are four molecules in the unit cell and the molecule size will be not
greater than number of atoms, will be probably around 20 to 25. So, if that is the number
of atoms we now know that for 20 atoms in the unit cell in the asymmetric unit; 20 * 9 =
180, plus 1. So, it is 181 parameters to be determined, and we have the in this example
the number to be 14,700.
Now, this property that is associated with a unit cell and the property that is associated
with the reciprocal lattice, which gives a very large number is referred to as over
determination. So, we have to determine 180 plus 1 parameters and we have 14,700
reflections. So, the measurements are over determined and that is in fact, what we require
to improve the structure to the accuracies which we eventually report. Mathematically
this is a very comfortable situation. The best possible situation would be to now change
the wavelength. So, if I change the wavelength of Cu kα from 1.5418 to 0.7107, let us say
its molybdenum radiation. If you calculate the value of N, you will be astounded,
because it will now be 1,29,000 possible reflections.
Same crystals, same unit dimensions, but now you see that the number of reflections you
get will be an enormous number. Of course, there will always the issue of symmetry and
what symmetry can do to these reflections. Not all 14,000 or 1,29,000 reflections are
554
independent of each other. Many of them will be equivalent of each other. So, there will
be a reduction of one-fourth or one-eighth of this total value, by the time we actually get
to the unique axis, unique number of reflections. We call it the unique number of
reflections.
So, by the time we get to the unique number of reflections, it will be reduced by a factor
of 4. Even then we have a heavily over determined site and therefore, this methodology
of using structure determination to get very accurate positions of atoms, inside the crystal
goes up by a very large amount. There is no other method which will give you this kind
of an accuracy. You might have seen the results which are published from X ray
diffraction studies, where carbon to carbon distance is 1.542 with a standard deviation of
± 0.003 Å. So, the accuracies with which we go, is a real measure of the quality and
quantity of data which we have got. So, if we have a good single crystal.
So, the rate limiting step in the entire analysis is the use of a good single crystal. By a
good single crystal I do not mean a good looking crystal. A good looking crystal is not
sometimes diffracting very well. In fact, a extremely good looking crystal will not
diffract at all, because the extremely good looking crystal has many reflecting surfaces.
Because it has many reflecting surfaces, there is a good possibility that reflection coming
from a plane, maybe again getting reflected within the crystal by another plane. It is
oriented in such a way that it will satisfy the Bragg condition again, and so eventually it
will be within the crystal and never come out of it.
In fact, there is very clear example of my own personal experience which was the case of
the Glucuronic acid. Glucuronic acid I manage to get crystals of very beautiful size and
very beautiful shape. Very excited I put it on the diffractometer and waited as usual 45
minutes, 1 hour. Then I lost my patience waited, almost lost my patience, but then I said
2 hours. No diffraction came out, but I was very sure it was looking beautiful. So, I took
it to the my polarising microscope, looked under the polarising microscope, there was no
extinction visible, because the parallel planes are so well organised that you will not see
any extinction of light.
One of the characteristics you should check therefore, when you mount a crystal, is to
see the crystal under the microscope. These are practical issues, because many of you are
going to determine structures and when you determine the structures you will be using a
555
single crystal. And when you are using a single crystal, first check you have to do is to
take the crystal of suitable size, normally the size which is suited for a diffractometer
experiment depending on the of course, the contents of your unit cell, assuming it is a let
us say a large organic molecule or a small inorganic molecule or whatever and you are
using Mo kα radiation.
The size of the crystal should be 0.3 mm 3 in a typical case. It may not be always possible
to grow crystals in cubic shape, 0.3 mm3 would mean it is a cube. It may be a needle, it
may be a plate, it may be a flat very flat plate or a thick plate. Depending on that the
average overall dimension should come to about 0.3 Å3 .
The whole idea is that when you sent in an X ray beam collimated on to the crystal, the
crystal should be fully bathed inside the beam, and otherwise you will get other effects.
So, the requirement is that the size of the X ray beam which comes in and the size of the
crystal should be about the same or slightly smaller than the beam coming, but definitely
not too very big. If you have too very big, a crystal that would be severe absorption and
in this case of the Glucuronic acid what happened was, that there were several parallel
reflections, so everything was internally getting reflected. I then decided let me cool the
crystal and when I cooled the crystal I thought that you know the diffraction may be
better may be it is large thermal vibration inside the molecule.
Well, it is not the large thermal vibration inside the molecule, but the diffraction itself
was now going within the crystal. So, what really happened is when we cooled the
crystal, the crystal exploded after the diffraction was done, because the diffraction
experiment increase the temperature inside the crystal, because the X rays were going
within the crystals so many times that the crystal could not withstand and the crystal
exploded on the goniometer.
So, an extremely good crystal is not what we are looking for. We are looking at crystals
which have the property of extinction, which have the right kind of dimensions and we
take this particular crystal and mount it on the goniometer. First check whether the
diffraction conditions are nicely satisfied and we are getting a good quality set of
reflections from the reciprocal lattice issue and then continue for the data collection.
556
Lecture – 42
Limits of Resolution
A couple of more points about this limit of the reciprocal space, particularly in terms of
how much we can cover in a typical experiment, this is crucial particularly as the size of
the unit cell increases.
Suppose you see we have talked about the change in the wavelength up here which we
discussed in the previous discussion and we found that changing the wavelength will
give you larger numbe of reflections. Of course, not all 1,49,000 will be independent. It
depends upon the symmetry associated with the system, but it maybe substantially
reduced by one-fourth or one-eighth of the total number; however, it is a very large
number of reflections.
And one once we have the access to all these measurements we can, we indeed can
measure experimentally all these individual reflections. And therefore, if they are
symmetry equivalent we can take the average of the 2 so the accuracy of the experiment
increases.
557
The other issue is that the limit of the limiting sphere is defined by the value of sin θ, sin
θ going up to 1, but it is not necessary that the diffraction is limited. The diffraction can
go beyond sin θ/λ and that is something which we will discuss towards the end of this
course where we can use such data to determine the bonding densities. We can also
determine what happens to the density, electron density between two molecules, if it is
let us say having intermolecular interactions, so that particular data is of extreme
consequence for analyzing accurate datasets.
However, in the context of what we are discussing now, the limiting sphere. Not only
limits, the number of reflections depending upon the wavelength. It also depends upon
the volume as you can see from this expression N is 33.5*(volume of the direct cell/ λ 3).
So, there are two variables here; one is the λ and the other is the volume of the direct
cell. In the example which you have taken here, we have taken the volume of the direct
cell to be 1600 Å3, but if you are now crystallizing a protein or a virus or whatever, the
volume of the unit cell will be something like 100 Å3 on an average, so the volume will
be enormously large. So, the larger the volume, larger are the number of reflections.
So, effectively when we now compare this to the multi slit experiment which we
discussed as the number of slits, increases the number of scatterers also will increase.
And therefore, in the case of a protein crystal structure the number of scattering points
associated with the electron density which happens to be the atoms in the unit cell, they
will increase enormously. And since they increase enormously, we will have the
diffraction pattern luckily for us, it is a reciprocal relationship. So, we do find the
diffraction pattern over crowded, but then at the same time reasonably separated from
each other. So, we have to have individual reflections, separated from each other in order
to estimate the intensities.
So, what really therefore happens, is that when we have to talk about such situations
larger, smaller molecules or even the case where we have very accurate datasets. We also
talk about an issue which is extremely important for experimentalist and this is
something based on which your results can be evaluated. For example, you are let us say
studying a very large molecule like a protein or a virus then it so happens that we have to
worry about this value dhkl is λ/2sin θ.
558
What also happens is that in the case of the biological molecules, particularly there is a
large amount of space which is occupied by what we call as solvent water mostly it is
water in many of the cases in biological molecules. And therefore, this solvent water has
some kind of a structure, but not a realistic structure associated with the actual crystal
structure of the protein and therefore, this solvent protein will give you some kind of
diffuseness associated with the diffraction. So, much that the actual even though the
crystal is of a very high quality the extent to which it goes in sin θ is severely limited.
In other words the extent to which we can get reflections beyond let us say a certain θ
value is limited. If the θ value is limited you see that is in the denominator. So, the
corresponding d value is therefore, dependent upon the value of sin θ. So, as we go to
higher and higher sin θ values we can get to better and better d hkl, because what is
relevant is 1/dhkl from the reciprocal lattice point of view and this 1/d hkl will identify the
distance from the reciprocal lattice origin to individual reflections and that d value is
very crucial.
So, we call this d value as the resolution. So, when we refer to this d value of a given
experiment and if the d value of the given experiment let us say is limited to 2Å; that
means, the resolution is limited to 2Å; that means, anything beyond 2Å level is not a
reliable conclusion. Suppose we are looking at let us say the distance between two
carbon atoms. The distance is 1.54 or something. If we are now talking about a 2Å
559
resolution data this we call as the 2Å resolution data depending on the d value that
depends again now only sin θ when once the λ is fixed.
So, one once the λ value is fixed we therefore, decide on the resolution of the
experiment. In a typical small molecule experiment these resolution values are not so
very critical as you can see down in this table here the minimum value of d hkl we can
reach with the wavelength that we let us say we use copper or molybdenum with when
we use the copper radiation the minimum resolution we can get is 0.77 that means, in
principle we should be able to see the carbon hydrogen distances just to give an example
because carbon hydrogen distance is around 0.9 also.
So, this is the resolution one would expect in a typical experiment on a smaller molecule.
The problem with the larger molecules is even though they also have the (d hkl)min
associated with the λ value. See the minimum value of dhkl is associated with the
wavelength we are using.
So, shorter the wavelength the lower is the dmin. For example, in this case of molybdenum
it is 0.35; obviously, using the molybdenum data to determine small molecule structures
are much more accurate than the much more reliable in terms of the resolution compared
to the resolution we obtain from copper. But there are other limitations because copper
gives higher intensities, the intensities which we get from the copper diffraction is much
higher in intensities therefore, the measurements become more and more reliable than the
measurements we can get from molybdenum at this resolution.
When we go to this resolution we need a very accurate measurement technique and most
often they are not since we now use high resolution CCD cameras or CMOS detectors or
whatever this kind of resolution limits is possible to reach one once you have a good
diffraction condition and a very stable λ and particularly a tunable λ, which can be
obtained from synchrotron radiation.
So, essentially what we are now discussing is the fact that we have now a diffraction
measurement possible. We have found out how the diffraction occurs in reciprocal space.
We know the Bragg’s law in reciprocal space. So, we can identify where these
reflections come from and how many reflections we can calculate also has been a
evaluated in terms of the limiting sphere response; however, the quality and the quantity
of data depends upon the value of the d, the minimum value of d we can reach. So, in a
560
typical small molecule experiment these are the values, but for larger and larger
molecules this value keeps going up.
So, when you see literature where structures of proteins and complexes of proteins and
all are reported you will see that in the title of the paper itself people say at what
resolution this work has been done. So, you will say 2 Å resolution structure, 4 Å
resolutions structure, 1 Å resolution structure which means that the accuracy with the 1 Å
resolution structure is enormously high. So, your reliability on the positions of the atoms
which you have determined becomes higher and higher. So, that way the value of the
limits of resolution becomes an important quantity to evaluate.
There are other definitions associated with the limits of resolution see we have said here
the maximum possible value of sin θ is 1 that depends and therefore, the value of sin
θ/λ , sin θ is 1, but sin θ/λ where λ is let us say 0.7 is much higher than the value of sin θ
and therefore, we can get access to diffraction beyond sin θ/λ, 1 and the minimum value
of dhkl as we have seen is λ/2.
The other measures of accuracy or the resolution which we can reach are in terms of this
table where we have the λ we also have a quantity called sin θ/λ. So, we can calculate the
sin θ/λ by taking this value down here and that will be one over of d hkl and that value is
given in terms of what we see here.
So, the maximum value of sin θ/λ corresponds to 0.64 in the case of copper so;
obviously, if you are looking at the measurements which you want to do for covalent
interactions and intermolecular interaction understanding in molecular crystals copper is
a “No”. You have to use the molybdenum and other lower wavelength diffraction
wavelengths, so, that we can essentially get to the accuracies which we are looking for.
So, you see that the sin θ/λ max is also a very crucial factor. So, if you are getting a data
up to 0.648 in copper, it is equivalent to getting a data of 1.407 in sin θ/λ in
molybdenum; that means, with the molybdenum data the reciprocal lattice is effectively
in principle expands and these are issues which one has to remember when we choose
the wavelength that is required for the given experiment. For example, if you use
molybdenum radiation and try to collect the data on a protein, more often it is very
difficult, because the intensities we get out from the molybdenum k radiation are much
lower than the intensity we get for the copper in case of the diffraction conditions being
561
satisfied under the conditions in which we do the experiment. The other issue which has
not been mentioned here is the fact that the temperature of data collection. So, that is also
a limit of resolution, because at any given temperature, let us say at the ambient
temperature when we measure the data the reciprocal lattice is distributed in such a way
that the a b c α β γ values or for that matter a*, b*, c* values we will have a certain
particular value.
Now, as we cool the crystal if we go down to let us say 100 Kelvin there is a little
contraction of the cell. So, since there is a contraction of the cell, there is an expansion of
the reciprocal lattice. So, the expansion of the reciprocal lattice will allow us if we can
get into these kind of limits associated with the d max , (dhkl)min then we can access more
and more data.
So, the fact that the temperature factor gets reduced is one advantage, the second is that
if we collect low temperature data with the molybdenum radiation we can reach higher
accuracies and also the higher angle of diffraction associated. So, these are all issues
which are practical, which are very important if somebody is practicing crystallography
as a tool just out of curiosity or crystallography for a living.
So, in both ways these points should be borne in mind and at the same time when if
people are in industry or using crystallogra φ c results they should ask the questions like
what is the resolution limit and what is the resolution of the experiment one has done
here particularly if they interact with some collaborators they should ask those questions.
The resolution of the data set, the temperature at which it is collected, the extent to which
they can rely on the results which they get in terms of identifying the atom positions,
finding the bond lengths angles and intermolecular interaction features. And in case of
protein structures the nature of the fold, the alpha helix, and the beta sheets and to extent
to which the reliability can be placed on both the secondary and the tertiary structures.
Quaternary structure is also an issue which can be addressed at this stage.
So, this kind of brings us to a state where we can wind up on the discussion on the let me
see what the next slide is yeah we can wind up on the discussion on the reciprocal lattice.
So, I think this is by far the extent to which anybody who have any textbook would cover
the reciprocal lattice information. So, the overall take-home is that when we have a
crystal which is kept in the path of an X-ray beam we get the diffraction; the diffraction
562
gives us the reciprocal lattice information. From the reciprocal lattice information we get
the information of the crystal, the nature of the crystal; the crystal system and then we
also get something more and that something more is what we will now address.
The something more is to find out the point group and the space group information from
the diffraction measurements because the diffraction measurements see the atoms and
molecules inside the unit cell they also see the symmetry that is relating one molecule to
the other or one part of the ion to the other and so on in the given crystal. And therefore,
it becomes crucial that we not only get the highest resolution data, the examination of the
reciprocal lattice values that is in terms of the number of h k l we have got the way in
which h k l distribute themselves in other words whether h k l are all in one side and –h,-
k,-l on the other side and so on.
So, the question is now what is the way in which we can get information out of the
measured reflections regarding the issue of a space group can we uniquely determine a
space group by looking at the reflections and the distribution of the reflections that
statistical distribution of the reflections? How many are in the projection reflections?
How many are in the axial reflections? What is the way in which the axial reflections and
the projection reflections come up compared to the overall reflections and in what way
the overall h k l planes behave with respect to the incoming radiation? So, that will take
us to another part of the discussion where we will wherein we will now see how this all
develops.
563
One of the immediate things which we notice is that for a centrosymmetric structure we
know that for every xn, yn, zn there is a corresponding -xn, -yn, -zn. So; that means, that the
fn if you look at the expression for the sine part of the cosine + sine which we wrote in
the intensities. If you recollect that the intensities are written as the f n cosine term first
and square of that, sine term first and square of that next.
So, the sine term we said it vanishes in case of when we took the example of caesium
chloride. Now that is because the sine value disappears depending upon where the atoms
are sitting, the atoms are sitting there at 0 0 0, ½ ½ ½ but when we say the atoms are
sitting at 0 0 0 and ½ ½ ½ we did not say at that particular state moment that the 0 0 0 is
also the the position for the inversion centre in caesium chloride. So, caesium chloride is
a centrosymmetric structure and therefore, in centrosymmetric structures you will have
this sine function always becomes 0 as is written here.
The reason why that becomes 0 is because of the fact that the f n sin 2π(-hxn - kyn - lzn)
which comes under the summation is equal to - f n sin 2π(hxn + kyn + lzn). So, when we
take this expression combine it with the fn sin 2π(hxn + kyn + lzn) they each other cancel
and therefore, we get to a 0. So, the summation over 1 to N all the atoms in the sin terms
in any centrosymmetric structure is equal to 0 which means to say that the phase
associated with centrosymmetric systems is always a real value, it cannot be in the an
imaginary value.
564
So, in other word in principle what we are talking about is if you now consider the
coordinate system to be like this and this is the real axis and this is the imaginary axis.
The component of the phase along the real axis alone will be there in case of the
centrosymmetric system. So, it will be either 0 or 180⁰, so, it is either 0 or 180. So, we
also associate that as + and -. So, the phase factor associated with centrosymmetric
structures can be only real quantities and therefore, in other words we use only the cosine
function in order to calculate the intensity.
We will examine the reason why this becomes very crucial and also of extreme
importance to solve the crystal structures later on, but at this particular moment we
should remember that in centrosymmetric structures the phases are either 0 or 180⁰
associated with every hkl reflection. So, on the other hand in case of non
centrosymmetric structures it can be anywhere in this 0 to 360⁰. So, it can be now having
an imaginary component as well because the value of the sin x in the expression of
intensity calculation will also have a certain value.
Now, one of the things we must go back and forth and always keep remembering are
these two equations. These two equations are the bread and butter of our study. The first
one is the structure factor, which now gives us the electron density and the second one is
the electron density, which gives the structure factor. These two are the so called Fourier
transforms of each other.
So, the way in which this develops is the fact that if we know the electron density and we
know where the atoms are sitting; that means, the vector R then it is possible to calculate
F(R). And if we know if F(R) and where the R vectors come that is our reciprocal lattice
vectors come from the reciprocal space measurements then we can calculate where the
atoms are sitting and the corresponding electron density distribution.
So, these two equations; therefore, as I mentioned are very important for structure
determination. So, it is also very clear that one once we have got this information about
F(R) it is in fact, the scattering structure factor and that structure factor is square root of
the intensity. So, in the typical X-ray experiment we send in an X-ray beam get the
diffracted diffraction done and we identify the coordination of each and every spot in the
reciprocal lattice in terms of hkl and we measure the corresponding intensities of hkl with
a detector.
565
So, this intensity of hkl will give us the modulus of F(R), but the phase associated with
the F(R) has to be calculated. We actually estimate these phases by various ways means
because phase by itself cannot be experimentally determined nor it can be accurately
estimated to the accuracy that we require in a structure determination. So, therefore, this
is always an approximation which will result in a nearby phase being estimated. So, in
that context the previous discussion we had on the possibility of a centrosymmetric
molecule having a phase of 0 or 180⁰ gains a lot of importance because we have only
two possibilities for every reflection and as a consequence the structure solution
protocols become easier.
We will evaluate them little more closely later on. So, the Fourier transforms we need to
use the structure factors F(R) to get to ρ(R). So, that is our aim when we say structure
determination in the solid state using the symmetry information and on the way in which
the diffraction pattern appears then we have to use the F(R) to get to ρ(R). The Fourier
transform will give us the final result. At this time I just want you to make a note you see
we have in an experiment we have the intensities measure.
Now, why cannot you use intensities directly and go to the structure? In principle it is
possible what we get when we use the Fourier to the in fact, the I(hkl) and P(uvw) this
Patterson function which is essentially a vector map, it depends upon the vector distance
between atom one and atom two. So, if there are two atoms the distance between the
atoms is the characteristic of u, v, and w. So, u is equal to x 1 - x2, v is equal to y1 - y2, w
is z1 - z2. So, the inter atomic distance is the function which goes and becomes the so
called Patterson function.
Here there is no phase problem, so we can use the intensities in principle to get to the
Patterson function and if we interpret the Patterson function properly then based on the
Patterson function because we have all the atoms there, but not as atoms, but as inter
atomic vectors. So, there are lots of limitations in this approach, but this particular
approach which was developed by Patterson in 1925 can also be utilized under very
special circumstances, which we will discuss when we go for the structure determination.
So, essentially Fourier transform is the key and the Fourier transform on F(R) is the one
which is required to get to the electron density which we are seeking. So, if you use this
566
I(hkl) we get P(uvw) then we have to treat P(uvw) in such a way that this eventually
reach to ρ(R).
So, how to go from P(uvw) to ρ(R) will be a part of the discussion when we discuss
structure determination protocols. So, at this moment therefore, we see that the phases
associated with the centrosymmetric systems become all real. So, that the functions
become cosine functions only when we do the summation to calculate the structure
factor.
Now let us understand therefore, the structure factor in more detail, because the structure
factor is the one which we need and the structure factor is a quantity which depends upon
the wave nature of the material. So, the suppose we consider a disturbance the you know
it is the electromagnetic wave, which is now related by this expression the e i (ωt + φ). We
are going to take ei (ωt) out this is equivalent to the expression we wrote earlier on in terms
of E0 ei(φhkl) the φhkl is the phase that is associated with individual reflections. Modulus of
a is nothing but the modulus of F(hkl) which we get from the intensity measurements.
The ei (ωt)
we are going to drop completely because the diffraction is assumed to be
coherent. There is no change in the frequency. The fact that there is no change in the
frequency we remove this. In fact, if we had considered this the possibilities are very
immensely interesting. If you had consider this and then involve the expressions for the
structure factor calculations what we will get is the behaviour of what is known as
567
Compton scattering and Compton scattering analysis will give us the momentum
densities and so on, but that is not a part of our discussion here. So, we are going to
remove this ei (ωt).
So, what we require is therefore, this quantity “a”. So, for a single wave which is coming
out from the experiment we have a modulus of “a” which is nothing, but our modulus of
F(hkl). We will bring in the F(hkl) later, but we will understand the mathematics in terms
of the disturbance which is associated with φ which you can call as a wave function if
you wish.
So, the wave function depends upon the modulus of the amplitude, “a” is the amplitude
here, eiφ , iφ represents the phase. So, this is the amplitude, this is the phase, amplitude
and phase together give us the wave function. So, therefore, we have a wave which is
coming out, this wave of course, can be a constructed based upon the overlap of several
other waves as we know depending upon where the atoms are in the unit cell and so on.
So, at this moment for our simple discussion and understanding we will assume that
there is a wave single wave which is coming which probably could be the resultant wave
of all the waves overlapping in a given direction. So, if we consider that as ψ then we
have a * eiφ forgetting ei (ωt) that will be the simplest expression. So, this expression is if
you consider this quantity “a”, the quantity “a” as we already saw has two components,
the cosine component as well as the sine component. So, we expressed therefore, the
value of a is equal to x + iy.
The value of x is modulus of ‘a’ cos φ and the value of y is modulus of ‘a’ sin φ . So,
the imaginary component is up here. So, this therefore, now gives us the definition of the
amplitude. So, the amplitude has got a real part and it has also got an imaginary part. So,
the real part has a certain modulus of ‘a’, the same modulus of ‘a’ is associated with the
imaginary part as well and ‘φ’ is the angle which is shown here.
So, if you take this simple Argand diagram this is known as Argand diagram by the way
you all know this. So, it is that this is the real axis and that is the imaginary axis. You
have this value of ‘a’ which is what we would like to determine and that depends upon
the value of ‘φ’ . So, whatever is the value of ‘a’ depends on the value of ‘φ’ .
568
So, in terms of the projections we have written ‘a’ is equal to x + iy. So, if you take the
projection of this onto this, this gives you the value of x, the projection of this onto the
imaginary axis gives you y, but since it is on the imaginary axis we have a = x + iy and
now what we do is this modulus of ‘a’ since it is x + iy we can take it take the square
each of these quantities. You remember how we did the intensity calculation?
When we did the intensity calculation this which took it and squared it and that value we
took it and squared it and therefore, i became - i and so we had this value, which is not
having any imaginary component. So, that is how the intensity developed. So, we if we
now, therefore, draw the analogy with that one, we have modulus of ‘a’ is equal to
square root of (x2 + y2) so; that means, it in mathematics we can write this as a . a*, this
a* is what we call as the complex conjugate, if ‘a’ is x + iy then a* will be x - iy.
So, the product of a.a* will give us the modulus of a2 which is our intensity. So, the
intensity therefore, consists of not just the structure factor alone it also consists of the
product of the structure factor with its complex conjugate. So, the structure factor with
its complex conjugate will give us (x2 + y2) which is the modulus of a2.
So, you also see that if you want to calculate the phase associated with this ‘a’ value then
you see that it is very straight forward. Now tan φ = y/x. So, what we have shown try to
show here is a simple mathematical way in which we appreciate and understand the ways
and analyze them when they are in the Argand diagram. And then see whether we can
express the amplitudes in such a way that we have the x component which is the real
value component, the y component which is the imaginary value component and then get
to the final value of the way in which we can describe the intensity.
So, the intensity is what we measure, but what we want is the ‘a’ value and intensity
happens to be a.a*; a* is the complex conjugate and the angle φ which is the phase
angle can be determined by this tan φ = y/x. So, this essentially tells us that we can go
ahead now and it study the structure factors with respect to how they develop and how
they come up.
So, there is a little more of mathematics coming your way and I will try to make it as
simple as possible. So, that everyone in this in this group will appreciate because if this
is understood then I think the basics of all diffraction will be understood fully. One once
we have a full understanding of it we do not have to depend upon what the pressing of
569
the button on the machine gives because what all you have to do as, as you have many of
the pharma industry and probably people who use diffractometers or have realized is that
you have to just put the crystal on the diffractometer measure the diffraction that is
coming and then analyze the diffraction in terms of what the machine says and press a
few times the enter button before you get to the final structure.
So, the background mathematics which is required here, the background understanding
of both physics and mathematics. The scattering processes, the diffraction process and all
that you have now started to understand fully. And one once you appreciate this I would
welcome you to do that and enter operations to determine the structure, but otherwise do
not determine the structure yourself because you do not know how it is coming and
obviously, you can make many mistakes and those mistakes can be quite costly there are
a large number of structures which are published in literature which go with an error and
one has to be extremely cautious when you conclude.
For example, particularly pharma industries when they conclude whether it is a conclude
cocrystal or a salt they will make enormous mistakes and then this leads to all kinds of
legal issues and problems and so on, which the pharma people will face. So, if some
pharma people are taking this course then it is addressed to them that this what you have
learned today and out what you are going to learn over the last few days and also what
you are learning in the next few days is rather looking very complicated, because it has
some mathematics.
I am trying to make it mathematics as simple as possible, but you will be able to

appreciate why we do this because one once you do that you will not make such mistakes
where you cannot do it not only costs a lot of money for your company, it also costs lots
of other problems, because you know people take your drugs. For the pharma people it
has to be taken with seriousness, but what we call as a structure as cocrystal or a salt or a
polymorph all these issues one has to worry about when such a thing happens.
570
Lecture – 43
Concept of Structure Factors
The fact that the magnitude of the amplitude is given by the expression, which is equal to
square root of x2 + y2 and the intensity therefore, can be expressed as a multiplication of
the real and the complex conjugate parts of a; a . a* giving us this modulus of a2 . With a
defined as x + i y and a* defined as x – i y, this is now telling us that we can get to the
intensities from any scattering. So, this is representing one single wave, but we can have
multiple waves. The phase that is associated with its tan φ = y/x.
571
Now, if there are large number of such amplitudes. So, because that is what happens in a
crystal than the overall amplitude which comes out the so, called resultant amplitude is
sum over individual amplitudes. So, the resultant amplitude is sum over individual
amplitudes and that is given over sum over 1 to N of modulus of |an| exp(i φn).
So, you see that this resultant is the representation of F(hkl). So, therefore, you see that it
is sum over 1 to N we can write it as a real part + ‘i’ times the imaginary part which is
associated with ‘i’. So, the real part and the imaginary part together will define our
resultant ‘a’. So, now, you can imagine a situation or you can create a situation where
you write here the overall structure factor F and then this will be F(hkl) individual h k l
going from 1 to N the total number of reflections, this also could be an individual h k l
value F(hkl); F(hkl) this is (Refer Time: 02:24) F(r), F of overall resultant, capital F.
So, for an atom at x y z the amplitude is f and the phase changes 2π which we have
calculated in the previous classes to be 2π (hx+ky+lz). This we worked out with the fact
that it can be done for any plane h k l and any position of the atom xn yn zn. So, if there is
an atom at x y z, the amplitude of course, is always f depending upon the content of the
electron density at that position vector x y z. So, if there is a carbon atom there will be 6
electrons in nitrogen atoms have 7 and so on.
So, we can therefore, write the expression for the structure factor. Remember until now
we were discussing the intensities because that is what we measured and now we are
572
writing the expression for the structure factor. So, the structure factor is 1 to N, f n exp 2π
(hx+ky+lz). So, what you therefore, see is that how the phase factor can also include the
imaginary component.
In the case of a centrosymmetric structure this imaginary component will not be there it
will become a cosine function only, but in case of the general crystal structures where we
have non centrosymmetric materials also this will be expression. So, this is another key
expression for our calculations in the future.
So, this is the expression for the structure factor you see that there is no modulus written
here. So, it is actual structure factor modulus and the phase together, and this is the
expression we need to dump it into the Fourier transform expression of ρ(r) so, that we
can get to the ρ(r). So, this is the value which we need to compute electron density.
So, this structure factor therefore, can be also written conveniently in this form cos + i
sin for the centrosymmetric system as I mentioned again and again only this term will be
there. And this term now will be 2 times that because this will vanish and this will be
now there 2 times over N by 2, but it is equal to 1 to N, f n cos 2π (hx+ky+lz) how does
that came about we will see as we go along.
We will see this right away. So, for the centrosymmetric structure atoms are at xn yn zn
and the -xn -yn -zn so, that is the definition; that means, we are taking the origin as 0 0 0
573
as coincident with the inversion center for the any crystal structure which is
centrosymmetric. So, there are so, many centrosymmetric space group as we have
already learnt and all these space groups will have this property of having xn yn zn , -xn yn
-zn. Apart from other equivalent positions which will come due to the operation of
symmetry elements.
So, if it is P ī these are the only 2 positions, but if we go to higher systems like P 21/c
and so, on you will get other equivalent positions. But the common factor is that we have
the xn yn zn , -xn yn -zn which therefore, defines our center of symmetry to be at the
origin. If the center of symmetry is at the origin then F(hkl) is sum over 1 to N/2 because
now we do not have to consider x y z separately and, x̅ y̅ z̅ separately and if we do
separate them than its summation is over only the half the number.
Because for every x y z there is a -x -y -z present in the crystal lattice or in the unit cell.
And therefore, this summation is over the unit cell and this now you see is N/2 f n {exp
2πi (hxn+kyn+lzn) + exp 2πi (-hxn-kyn-lzn)}.
So, if one evaluates this the sine terms vanish as we showed just a few in the previous
discussion and therefore, you see that this is equal to 2 times sum over 1 to N/2 f n cos 2π
(hxn+kyn+lzn). So, what you do is you consider half the number of atoms somethem over
2 times and get the cosine function or the most easy way to program it in (Refer Time:
06:37) of programming is F(hkl) is equal to summation 1 to N fn cos2π (hxn+kyn+lzn).
So, the complete disappearance of the dependence on the imaginary axis is gone. So, this
is no complex quantity that is associated with the with the centrosymmetric structures.
So, this value of F(hkl) can be plugged in to our Fourier transform equation and we can
get the electron density distribution. So, that is where we have arrived. So, we have
arrived at a very fairly after safer position now, we do an experiment we do an X ray
experiment calculate the intensities and having measured rather measure the intensities
having measured the intensities calculate modulus of F(hkl).
So, what is left out is to find out the value of cos2π (hxn+kyn+lzn) in case of
centrosymmetric systems and the exponential value corresponding exponential value in
the case of non centric systems. We also know how to calculate this cos2π (hxn+kyn+lzn)
with the example we took for caesium chloride. So, it is a fairly straight forward one if
574
you have only 2 atoms, in the unit cell and that is the centrosymmetric unit cell it was a
very easy calculation. But you can imagine that if there are 100s of atoms the
calculations will be a little more tedious, its not going to be a fairly straight forward.
The second issue is the unknowns here are xn, yn and zn what you have in this information
is the information about the modulus of F(hkl), but as long as we do not know x n yn zn we
cannot get the phase and therefore, we have only the magnitude of the scattering that is
coming up and that is why we say it is the structure amplitude. So, the magnitude of
F(hkl) is called structure amplitude. So, the structure factor is the one which includes the
amplitude as well as the phase. So, the amplitude as well as the phase included will be
the structure factor right.
We go further now the there are lots of equations which will come, but I will go very
slowly systematically of course, hopefully so, that we will follow each other. See again
we come back to the symmetry that is why you know we entitle this structure as in the in
course as structure and symmetry in the solid state or symmetry and structure in the solid
state in that order. Because symmetry governs the structure and structure governs the
symmetry and then both of them now coexist and handshake in the case of a solid state
and that is why we have given the title other than saying crystal structure determination
or chemical crystallography or whatever even though the last part of the course we will
cover mostly chemical crystallography.
575
So, systematic absences is something which we will now introduce. Its a very nice
statement systematic absences something which is absent systematically. Well I have an
example in my group there is one person who is systematically absent in all these talks
there others are attending this talks, but there is one fellow who is systemically absence.
In fact, using him I can determine the symmetry in the group and that as a joke I am
telling you this, but essentially systematic absences help in determining the space group
of a given crystal system.
So, we need to study it in more detail. See one of the things I will make a statement now
it is like making a statement and giving a proof like in mathematics you know we give
the proof and say QED (Quod Erat Demonstrandum) right. And similar thing we will do
here the statement that I am going to make is that if in your crystal system, the space
group involves translation components watch my words; if in your crystal system the
space group involves translational components whether it is a centering associated with
one direction or Face centering or Body centered.
So, in other words the translation is associated with a lattice or whether it involves the
translational symmetries in the presence of screw axis and or glide planes and if all these
operates simultaneously in the space group generating over space group like for example,
if you take P 21/c in P 21/c we have the presence of a primitive lattice. So, there is no
translational component associated with the primitive lattice; that means, primitive
lattices will have all the reflections with respect to h k l.
So, there are no systematic absences that will come. So, the fact that there are systematic
absences coming is a consequence of the presence of translational components in the
space group. So, it is not obviously, there in the point group, so, it is in the space groups.
Luckily for us crystallization ensures that we have three dimensional periodicity. So, the
presence of periodicity, the presence of lattice centering will ensure that will have
systematic absence. So, using X ray diffraction and finding out how the h k l values is
appear, it is possible to see the translational component associated with a lattice.
So, it could be a C- centered lattice, it could be an F centered lattice, it could be an I

lattice we will see how it develops. The other aspect is of course, the presence of 2 1
screw axis as well as the glide planes. So, let us now take the structure factor expression
576
which we have and see how the structure factor expression reorganizes itself when we
have a centering. So, we will start from a C- centered lattice.
So, in the case of a C- centered lattice what are the extra components you will get for
every x y z you will get a ½ + x ½ + y z. So, if there are 8 equivalent points, you will get
8 other equivalent points associated with the addition of ½ + x ½ + y no addition along
the z axis. So, this is the concept of C-centering in the lattice. We will take the
international tables later on probably later on during the coming classes it probably will
not happen now, but in the coming classes.
So, that we now understand the underlying mathematics behind this and the underlying
principle behind it and then we will not have to worry about it when ones we have this
understanding then we will go and examine the international tables for a crystallography,
which we have already seen with respect to equivalent point diagrams and the symmetry
diagrams.
We have seen the diagram associated where we see only the symmetry that is associated
with the space group on the left hand side and the appearance of the equivalent points on
the right and we are now reasonable experts to go one from one to the other given the
crystal system point groups symmetry and so on. So, here what we will do is we will get
the expertise on understanding how this systematic absence appear. In fact, what
international table gives is not the systematic absences, but it gives you the what we
could call as systematic presences.
So, whatever is present is indicated in the international table please remember that. Do
not look at the international table and look for systematic absences because what is given
in the international table are the systematic presences. So, you have to look for once
which are not going to follow what is given in the international table before we go to
that.
So, let us not hurry up and see and go to the international tables right now what we will
see is what happens to the structure factor when there is a C- centered lattice that is
offering itself in the crystal system. We sent an X ray beam we got the diffraction pattern
the diffraction pattern gave us the reciprocal lattice positions as well as the intensities. I
measured the intensities, I also got a list of all the h k l values I have.
577
Now, I am going to examine that list of h k l values what is important for us to remember
is that if there is a C- centered lattice, the n atoms just like we discuss the case of centro
symmetry just now. The atoms there are xn yn zn, for every xn yn zn there is a
corresponding -xn yn -zn. There are no translation components with that and therefore, the
presence of the centrosymmetry or otherwise is not going to be indicated by the so,
called systematic absences. We will notice that when we look at the international tables
and see what are written, what are what is all written in that systematic absence column.
So, if similarly here the n atoms are divided into 2 parts; there is one part which is
associated with x y z and the other part which is associated with ½ + x ½ + y z
representing the C centering in the lattice. So, what is the effect of that on the structure
factor? One is of course, each of them has half of N. So, if N is the total number of
equivalent points M is half of N.
So, I hope this equations are fairly clearly seen in the projection, you see it alright ok. So,
F(hkl) therefore, can be written as sum over 1 to N this is our general expression which
we just now derived fn exponential 2πi for any crystal system. Be it centrosymmetric or
non centrosymmetric for any crystal system sum over 1 to N is f n exponential 2π
(hxn+kyn+lzn) we now know this is the factor which is the phase factor and this is the
corresponding amplitude coming due to fn and we know also that fn falls with sin θ/λ.
So, it depends upon the scattering angle θ as θ increases the value of f n keeps on
decreasing, we know that and we have understand it thoroughly. So, now, what we will
do is, because of the fact that there is C centering we factorize this into 2 components.
So, we have sum over one to M the first part corresponds to x n yn zn. So, this is
exponential 2πi (hxn+kyn+lzn), the second part corresponds to ½ + xn ½ + yn zn.
So, we separate these two. So, this is sum over M now, but there are 2 components one
corresponding to half of N the other corresponding to the other half of N. So, we have
therefore, this expression exponential 2πi {h(½ + xn) + k(½ + yn) + lzn}. I hope the speed
with which I am going is ok. So, you are all with me. So, this is important particularly
when you are looking at the video and trying to get the hang of it if I hope I am trying to
adjust to the speed, with which we one can interact because in a class room you can sit in
front of me and I can see your heads nodding, but when a in a case where you are
578
observing it on a video, I do not know whether your head is nodding or whether you are
sleeping right.
So, I will go adjust the speed in such a way I hope that you will be with me all the time
and please follow this indicator which is a nicely red in color. So, I will be following that
with the equations. So, let me repeat this part to keep your concentration intact. So, when
we go to the general expression of F(hkl) this is the general expression for h k l. So, it is
1 to N; N is the total number of atoms in the unit cell.
Now, this N is becoming N/2 in a C- centered lattice and therefore, I have now the
summation over 1 to M, M is equal to N/2. And then I write the expression which is
associated with xn yn zn and this is the expression which is associated with ½ + x n ½ + yn
zn. So, there are n atoms associated with the entire system half of it with respect to this
and half with respect to that and that defines a C- centered lattice.
See this is the advantage of symmetry, this is the advantage of crystallization in a C-

centered lattice. Effectively what it means is that if once you have this 1 to M measured;
that means, the number of atoms which you are looking for is now halved because we
have to determine only the M positions because the other half is generated by this that is
advantage number 1. The advantage number 2 is something will happen to these F(hkl),
what will happen to these F(hkl) is what we see in the next part.
What will happen is I can write rewrite this the expression as 1 to M, f n exp 2πi
(hxn+kyn+lzn) So, what I do is I take the common factor out of this. So, once I have this
common expression fn 2πi exp 2π (hxn+kyn+lzn), then this will be multiplied by this now
becomes value equals 1. So, because we have this taken out, so, it is 1 + exponential 2πi.
Now, what is left in this expression is (h+k)/2. So, this is the additional half values
associated with h and k. So, this is exponential 2πi (h+k)/2. Notice that h and k are
integers and therefore, this is a mathematical quantity which can be evaluated the
exponential 2πi (h+k)/2 we can evaluate mathematically. And that is given here we can
expand this as cos π(h + k) + i sin π(h + k).
So, this exponential factor with the integers can be calculated straight away and it so,
happens that this has a value (-1)h+k. So, if you evaluate this expression mathematically
we get this as (-1)h+k. So, this (-1)h+k can replace now this quantity. So, we will have this
579
as 1 + (-1)h+k. So, the value of h and k therefore, will get restricted by this condition, how
they get restricted by this condition is shown here.
Since h and k are integers, F(hkl) is equal to this is clear F(hkl) is equal to 1 to M, f n exp
2πi (hxn+kyn+lzn) times 1+(-1)h+k. So, now, the value that comes up for a h k l reflection
depends upon whether this value (-1)h+k is evaluated the way it is has to be.
So, we substitute different values of h and k. So, if h + k this quantity h + k is even; that
means, if suppose you have a reflection like in the let us take an example, suppose there
we have a reflection in the diffraction experiment which is 1 1 0. So, it is h = 1, k =1. So,
h + k is what h + k is 2; that means, it is even 2 times n, n is equal to 1 here. So, h + k is
2.
So, when it is that value then this (-1) h+k will just give an additional 1 to this quantity and
therefore, you will have F(hkl) as equal to 2 times, 2 times this value that will be 1+1.
So, it will be 2 times sum over 1 to M, f n exp 2π (hxn+kyn+lzn). So, you notice that this is
now 2 times and we are summing it over only the half the number of total number of
atoms in the unit cell.
So, the effort is reduced to 50 percent and only for the even values of h k l; l is
immaterial in this calculation because its a C- centered lattice. So, in the C- centered
lattice therefore, we have h + k is equal to 2 n and whenever it is odd let us say we take a
580
reflection 1 2 0 in case of a reflection 1 2 0, 1 + 2 is odd. So, the value of F(hkl) is 0; that
means, there is no structure factor value; that means, there is no intensity. So, that is why
we call this as systematic absence.
So, all those reflections in among the h k l all the general reflections we call them as
general reflections, h k l reflections which are measured in the entire experiment if you
measure them, you will see that all those values for which h + k is odd will not be there.
So, if you just now display your reciprocal lattice in such a way, in which you look at the
way in which the reflections appear all the h + k odd values will not be there at all; that
means, you do not see any reflection that is coming out from the experiment.
So, experiment generally now dominates of half the set of reflections and they come
twice 2 times sum over 1 to M, fn exponential will allow us to calculate the F(hkl) values.
So, what is very interesting here? What is very interesting here is the fact that if we have
the C- centered lattice, the presence of the C- centered lattice will give us what we call as
systematic absences and those systematic absences occur when h + k is equal to 2 n + 1.
So, let us take a few examples and see how this happens. So, suppose I am now trying to
measure in a C- centered lattice, I let us say I want to get a value of F(hkl); that means, I
measure the intensity of h k l what I measure are the intensities of h k l. So, I measure
have the diffraction experiment done some spot appeared. So, I indexed the spot, so,
indexing again is a very crucial factor.
So, I indexed the spot and when once I indexed the spot, I will identify the h k l values. If
I know already that C centering is present or if I do not know anything about the
presence of this let us look at it that way. I have a data and this particular data tells me
that suppose I look at this reflection statement which you have made here, F(hkl) is
systematically absent for reflections with values of h + k odd by taking some examples.
Suppose I give you 1 4 0. So, this 1 4 0 will be h + k is 5 so; that means, it belongs to
this category.
So, whatever be the structure which crystalizes in a C- centered lattice please note the
words, whatever be the system which crystalizes in a C- centered lattice whether it is
monoclinic, orthorhombic and what other is possible in a C- centered lattice monoclinic,
orthorhombic, tetragonal does not have a C- centered lattice. So, also hexagonal and
581
cubic; cubic can have a C- centered lattice, but then it becomes a F centered lattice by
symmetry.
So, it is only in a case of monoclinic and orthorhombic systems, we can have this kind of
systems. So, you see that you can cross check the a b c α β γ which you have got in some
sense, so, 1 4 0 will be absent ok. Now if we look at the reflection like 2 5 0 that will also
be absent. So, if you look at something like 2 4 0 it will be present.
Now, reflection like 1 3 0 will also be present. Now why I am taking 0 it’s not necessary
to take 0. So, I can take for example, 1 3 1 what will happen to this reflection 1 3 1? That
1 3 1 will have to worry only about h + k even. So, therefore, for all h k l values whether
it is l can have any value including 0, these h + k odd will be absent and that therefore,
characterizes a C- centered lattice.
So, now you see the importance of how diffraction can be used to identify the symmetry
that is associated with the diffracting crystal. So, the system into which the diffracting
material goes in to. Now clearly says that this has to belong to a C- centered lattice and
depending upon the other the distances and angles it could be either monoclinic or
orthorhombic. Of course, the other symmetry elements which come are not associated
with the centering, they are associated with the a and b and c directions depending upon
the nomenclatures we have for different crystal systems.
So, that part we will discuss later. What happens with respect to those three components
which describe only the symmetry associated with directions. Those 3 or sometimes 2
sometimes 3 sometimes the same thing repeating 3 times those symmetry elements if
they also have translational component those symmetry elements will also show
systematic absences.
But please note that those systematic absences will not affect all of h k l, it will effect
only either projection reflections or axial reflections. This is the statement I am making
do not worry too much about it we will discuss it in full detail as we go along. So, right
now what we have discussed is a C- centered lattice.
582
Lecture – 44
Systematic Absences 1
You look the equations the moment, now some people see the equations they get worried
here there is nothing to worry about these equations because you are now experts in
understanding what we are talking about. C- Centered lattice gave you h + k odd absent.
So, suppose now you simultaneously see you cannot have A center and B center and
leave it at that because by symmetry C center will automatically get centering also done
therefore, it become say face centered lattice.
So, either you have C, A, or B centered lattice which is representing the monoclinic and
the orthorhombic systems or you will have a face centered lattice which can come in of
course, monoclinic special case F can be coming in, but we convert it, which will be
probably indicated in some TA class. Then we have the F- centered coming in
orthorhombic, F center will come in cubic as well.
So, depending upon the other symmetry elements, we can decide what crystal system it
has to go into and apart from the fact that we have a b c α β γ which is also indicators of
what crystal system the given the given crystal goes into given system goes into. So,
583
these are the four equivalent points in case of F- centered lattice we have xn yn zn this will
be equivalent of xn ½ + yn ½ + zn.
This represents the A centered lattice ½ + x n ½ + zn yn this represents the B centered

lattice, ½ + xn ½ + yn zn represents the C- centered lattice. So, when all these faces are
centered we have a face centered lattice. So, we again do the calculation. So, what we do
is again now we divide this into the 4 equivalent points, the number of atoms in the unit
cell. So, M now becomes N/4.
So, the summation now is over one-fourth of the total number of atoms in the unit cell.
So, the effort becomes lesser when we have a F- centered lattice. So, the F(h k l) is now
sum over 1 to M we have the general expression fn exp 2πi (hxn+kyn+lzn) and then to this
we have to add the centering of each one of them this represents the A centered part this
represents the B centered part and this represents the C- centered part.
So, A face is centered, B face is centered, C face is centered and therefore, the expression
now becomes sum over 1 to M of x y z associated with x y z that is taken as a common
factor from all these four just as we did in the case of the C- centered. We have more
expressions here and therefore, that becomes 1 + exp (k + l) + exp (h + l) + exp (h + k).
So, this expression comes up and this expression now can be substituted with (– 1)
(k + l)
as we calculated in the previous case, exponential with these two integers can be
easily evaluated mathematically and the mathematical solution tells us that this can be 1,
- 1, - 1, - 1 and therefore, we get this expression for F(h k l).
So, what do you think will happen when we have face centered lattices? Face centered
lattices as we see there are four equivalent points which developed because of the face
centering. So that means, where the set of atoms we are considering now is reduced to
one-fourth. So, these one-fourth atom now will repeat based on the nature of the
centering.
584
So, what will happen to this expression what would be the actual F(h k l)? F(h k l) if h k
l are all even or all odd then k + l, h + k and l + k will all be even right. So, for example,
if you take the value of h k l to be let us take an example, let us take these example of
reflection 1 1 1 ok.
So, this tells you h is 1, k is 1, l is 1. So, I take this reflection 1 1 1 now this reflection 1 1
1 has to be fitting into this equation and how it will fits into this equation we will see in a
minute. We will go to the next slide and come back again this slide tells me that if F(h k
l) are all even or all odd. So, our 1 1 1 is all odd then if you add h + k it is 2. So, k + l
will become 2, h + l is also 2, h + k is also 2.
So, they will be all even. So, if it is even let us say if it is 2 2 2, then also these quantities
will be even. So, reflections like 1 1 1, 2 2 2 will if h k l are all even or all odd, this value
will all be even and then the calculation will be 4 times this quantity. So, the structure
factor will have a value whenever this possibility occurs and for all other things for all
other things it becomes 0 for all other h k l values it becomes 0.
So, h k l value is like 1 1 1, 2 2 2, 3 3 3 or combinations of other kinds of odd values lets

like 3 1 4 cannot be there, 3 1 3 yes, 2 2 6 yes, 2 2 1 no so, such absences therefore, we
will develop. So, this is the statement if h k l are all even or all odd, then the value of h +
k, k + l and h + k will all be even then you get this systematic absences.
585
So, this is correct this is the way you have to look at it. So, if h k l are all odd or all even,
the summation of each one of the two components will become even and therefore, this
exists this expression exists. That means, this systematic absence now will be enormous
and the many of them will be absent in this F- centered lattice. In fact, this is something
which is given later to you in powder diffraction as a what should we call a thumb rule;
the thumb rule is that if you have a crystal system which is a simple primitive lattice or if
you have a system which is centered like may be face centered the other one is body
centered.
Suppose there is 3 polymorphs this is with respect to the pharma people as well as
materials people. If it is 3 different polymorphs one crystallizing in primitive cell another
crystallizing in face centered cell the third crystallizing in a body centered its quite
possible. If such a thing happens then if you take a powder diffractogram of this material
you have the same material if you take the powdered diffractogram of the same material
you will see more reflections from the primitive and less reflections for the body
centered and further less reflections for the face centered.
So, that is how if you record the pattern of sodium chloride you will get only 5 peaks in
the powder diffraction pattern. So, the more simplified the powder diffraction pattern is,
it represents the symmetry that can be associated with the centering of the lattice.
Centering of the lattice is the one which will now reduce the number of possible
reflections and therefore, the number of possible observations you can have.
So, more the symmetric the more the symmetric the space group is, the less is the
number of observations you will have. And this is an issue which we may have to tackle
when we are dealing with diffraction both in single crystal as well as in powder and we
will discuss that when we actually do the experimental analysis of these, particularly in
powder diffraction this will be very crucial right.
So, now we will take a realistic example. So, you can also say the English language is
rather difficult that is why there is always a problem in getting this statement correctly.
So, if h k l are all or neither all even nor all odd. So, we can use that phrase as well. So,
what happens? If h k l are neither all even nor all odd, then it will be the systematic
absence. So, the systematic absences therefore, will now if you add neither or nor here
instead of all and or then you get to the definition of the systematic absences.
586
So, what you should remember here is that in a face centered lattice you add the two
indices add h and k, add k and l and also add h and l. When you do that, they should be
always even whatever be the value of h k and l individual values; when you add 2 values
it should always be even and when those values are even you will get a reflection, you
will get the measurement of F(h k l).
So; that means, you will get an intensity I(h k l) all other intensities are 0. So, by just
looking at the diffraction pattern now and looking at after indexing, you just easily it is
so, easily now to find out the centering of the lattice. We will take an example to
illustrate this systematic absences by taking the structure of Zinc Sulphide. In fact, it is
the mineral Zinc blende; Zinc Sulphide exists in various forms. So, we will take the Zinc
blende, the mineral Zinc blende it has been found by crystallography, that there are 4
Zinc Sulphide units in the unit cell.
So, compound Zn S occurs four times inside the unit cell and that is the information we
have from determining the density of the crystal, then we determine the volume of the
unit cell by determining a b c α β γ. So, we do the powder diffraction experiment or
single crystal experiment, look at the x axis of the powder diffraction and find out the 2θ
values and from 2θ values you calculate the d values and from there calculate a b c.
So, the value of a b c will tell us now that the Zinc Sulphide system is going into the
crystal system, where there are 4 molecules in the unit cell so; that means, 4 Zinc and 4
Sulphur. So, when one solves the structure we will take the solved structure so, that we
can illustrate the occurrence of the systematic absences very clearly. So, when we solve
the structure and we show the projection of the Zinc blende. So, Zinc is at 0 0 0, Sulphur
is located at ¼ ¼ ¼. So, Zinc is at this position.
So, this Zinc appears at the end of the a axis as well as end of b axis and then it also
appears at the ½ the distances along the axis and then the Sulphur because it we are
already checking out that Zinc Sulphide is face centered ok. Now let us see what happens
to the systematic absences. So, what happens to the reflections which comes from Zinc
Sulphide? Zinc Sulphide is Zinc blende is a face centered cubic lattice ok. So, we have
Zinc at 0 0 0 Sulphur at ¼ ¼ ¼. So, if we look at the projection diagram down 0 0 1. So,
this is very clear now this is distribution of atoms. So, we will see what happens to the
structure factors or in fact, the intensities and therefore, the structure factors.
587
See what is very important now is that we are directly getting the expression for the
structure factor, which now can be undergoing a Fourier transform to give the electron
density, what we are not worried about at this moment still which we should worry about
is how to determine the face. We have the value of fn, we have the value of h k and l we
do not have xn yn and zn in this example we are taking we have x n yn zn. So, we can get the
expression for F(h k l), because we can get the expression for I(h k l) and from that we
can get the expression for F(h k l).
So, remember that part because what we are now trying to discuss is given the structure
of Zinc Sulphide which Zinc is sitting at 0 0 0 Sulphur is at ¼ ¼ ¼ how do we get the
systematic absences satisfying the F- centered lattice that is all we are now examining
and that we can see from this set of equations.
You see F(h k l) is = to 4 times fn exp 2πi (hxn+kyn+lzn) that we know now because from
the previous slides here F(h k l) is that expression.
So, we take this expression and we have written it down here; now how many atoms are
there in the unit cell that is the summation 1 to M total number of atoms is 1 to M. Now
the number of atoms here are only 2, there are 2 atoms they associated with the structure
mark my words the unit cell has Z = 4. So, there are 4 Zinc atoms and 4 Sulphur atoms
associated with the unit cell, but with the structure there is 1 Zinc and 1 Sulphur. So, the
summation has to be done over two of them 1 for Zinc and 1 for Sulphur.
588
So, here M is 2 therefore, F(h k l) is 4 times f Zn, Zn is at 0 0 0. So, just like our caesium
chloride example the phase associated with it is 0. Now f S is sitting at the position ¼ ¼ ¼
and you know how to do this calculation now it will come at (h + k + l)/4. So, this is now
the expression for Zinc Sulphide the structure factor expression for Zinc Sulphide.
So, there are 4 such Zinc Sulphide units. So, it is 4 times f Zn and so, on there is a little
error in the textbook here I have taken it from the textbook, M is actually 4 and when
once we write 4 here it is then M = 2 is correct that is not an error M = 2, fZn and fS is the
scattering factor associated with Sulphur.
So, there are 2 atoms with Zinc Sulphide Zinc and Sulphur and the summation is done
over those 2 atoms. But 4 times because we have now or going to distribute this into four
equal parts; one coming from x y z another coming from the ½ + x ½ + y another one
coming from ½ + y ½ + z and the last one coming from ½ + x ½ + z. So, therefore, we
get into four such values and then we calculate the intensity. See eventually what we see
in the diffraction pattern is the intensity or for that matter the powder pattern will give us,
the intensity I versus 2θ . So, we have to calculate the intensity.
So, when we calculate the intensity, it is F(h k l) 2. So, 4 becomes 16, now square of that
{fZn cos 0 + fS cos 2π(h + k + l)/4}2. So, this now this adds up in case of each one of
these. So, f of whole square + {fZn sin 0 + fS sin 2π(h + k + l)/4}2. So, what we have done
here is to express it as a cosine + sine function you remember we are looking at the
intensity. So, there is no imaginary component that gets associated with it.
So, this gives us the solution to the problem, we get the intensity value as 16, {f Zn + fS
cos 2π(h + k + l)/4}2 + this. So, this quantity therefore, is something which we will have
to now see what are the possible reflections that come? Now why I am showing this, I
am showing this because Zinc and Sulphur even though they are F- centered lattice
alright because of the fact that there are only 4 units of Zinc and 4 units of Sulphur they
are sitting in special positions.
Because they sit in special positions effectively you will see more systematic absences
the major systematic absences are coming from the F- centered lattice, but the fact that
Zinc and Sulphur are sitting in special positions, they will also add to the possibility of
589
So, remember we have been discussing systematic absences in terms of lattice centering
so, far. We will also discuss the systematic absences which will come because of
translations associated with the 21 screw axis or any screw axes and glide planes. This is
with respect to the symmetry elements, but when once we have this symmetry elements
identified in the space group, we already know that there are some what we call as
special positions. So, these special positions therefore, will have additional systematic
absences.
So, if atoms have to be confined to special positions then the structure will show over
and above the systematic absence required for the symmetry, it will also show systematic
absences because the atoms are sitting in special positions. This is something we will
write down now as the value. So, it is either 16 (f Zn + fs)2 whenever h + k + l is 4n and if
(h + k + l) is 4n + 2 then you will have this value and (h + k + l) is 2n + 1 you will have
this value.
So, these now are very special to Zinc and Sulphur, the general systematic absences
come from this expression. So, one can calculate this expression and we know that h + k,
k + l, l + h should be all even then only we will get these reflections coming up. Among
these reflections because Zinc is at 0 and Sulphur at ¼ ¼ ¼ that happens to be special
positions in that space group, you will get additional absences. And they look as though
they are systematic, but they are systematic only with respect to the atom positions. They
are systematic of course, therefore, to the special positions in the unit cell.
So, let us go back to the discussions on the special positions for a few minutes. Because I
think we discussed it sometime ago and probably you would have you know over the
days you would have forgotten. So, the special position when we discuss this special
position issue, we found that the number of equivalent points suppose P 21/c we take, 4
equivalent points are there.
Now, if the special position is there in the system which is there, it is let us say sitting on
the mirror symmetry, then the number of equivalent points will get reduce to 2 so; that
means, what was 4 before has now reduced to 2. So, the atoms are now restricted by the
symmetry again and that symmetry restriction is coming due to the presence of the 21
screw and the c glide.
590
So, whenever we have again translational symmetry associated with the space groups
will get special positions. We do not get special positions for example, in case of P 1, but
we get special positions in P ī, because x y z and x̅ y̅ z̅ now can give rise to several
centers of symmetry positions which are independent of each other. And therefore, if the
atoms sit there they are very special, in other words they give Z = 1 rather than Z = 2 and
that gets reflected in terms of the systematic absences we see.
So, these are not now to be called systematic absences; what we should call systematic
absences are with respect to the crystal system. So, if it is a F- centered lattice whatever
are the systematic absences associated with F centered absence has to be checked. But
then if we find additional systematic absences like what we see here the special
conditions there are three possibilities in which the reflections can occur, and those three
are indicated here. Only those reflections will come because Zinc and Sulphur are sitting
in special positions in the F lattice.
So, lattice anyway exists and over and above you will have additional as though they are
absences, but they are not really absences they are additional conditions over and above
the conditions that are present in the F lattice. So, when we reexamine the international
table entry, we will see what is the effect of special positions and how additional
symmetry elements develop because of that effect.
So, this is something which we have to keep in mind and then we will come back and re
visit it, when we reexamine the international tables. So, what we have done so far is
therefore, to take the face centered lattice find out how the systematic absences come up.
And then we took the example of Zinc Sulphide which is very special, because Zinc and
Sulphur are sitting in special positions and we worked out the possible systematic
absences as a consequence in the F- centered lattice.
591
So, now we will have to see how to evaluate these or observe these systematic absences.
So, let us now generalize it, when once we generalize it we see that the systematic
absences can be put in this form of a table. So, whenever we have a primitive lattice we
have no systematic absences. Primitive lattice can exist in all 7 crystal systems. So, in all
the seven crystal systems we go ahead and collect the data, we look at all the h k l
reflections try to align them with respect to h even, h odd, k even and so on. And when
such an arrangement is done we find that there are no systematic absences in those
reflections then we say it is a primitive lattice.
So, the primitive lattice therefore, will have no systematic absences. So, space groups
like P 1, P ī, P 2, P m, P 21/c will have systematic absences not because of the centering,
but it will have systematic absences due to the 21 and c which we will examine later. At
this moment whenever there is a primitive lattice, systematic absences will not occur in
general reflections by general reflections I mean h k and l all having values no projection
reflections and no axial reflections are considered in this evaluation.
Special the projection reflections and also the axial reflections come may come as a
consequence of centering and that was indicated to you when we discussed the C-
centered lattice. I told you that there is a suppose there is a h k l which is 1 4 and 0. So,
that is actually a projection reflection, but what is important is to consider the addition of
592
1 and 4 and I said 1 4 0 is absent. So, that 0 there is actually a referring to a projection
reflection 1 4 0, but now we are considering the systematic absences in the overall h k l.
So, we call these as general planes. So, under the general planes we also have the
projection reflections as well as the axial reflections. So, when you have a centering of
the lattice you will have both of these coming up and that you must remember. And that
is something which we have to evaluate. So, when we analyze the data the way in which
we analyze the data is we put the crystal, X ray X rays come through do the scattering I
will keep on repeating this whenever required. So, we get the scattering and when the
scattering occurs we get the reciprocal lattice points, we index the reciprocal lattice
points we measure the intensities of the reciprocal lattice points.
Then we organize the intensities in a certain order, let us say even reflections, odd
reflections and so on. Then if there is a systematic absence coming in the general h k l
reflections, then we will be able to identify the presence of centering. It could be one of
the faces centered or all faces centered it could be body centered and so on. So,
whenever such a thing happens we should have systematic absences.
So, this particular table which I have given here tells you what are the types of
systematic absences that occur when this coordinates where there is a mistake in the
textbook it should be lattice points, e is not there its c, lattice and there should be any
after that. So, c is missing here. So, its I have copied it from the textbook.
So, many almost all the equations I have copied in the textbook because I thought that if
I write the equations you know my handwriting and that will be terrible. And if I prepare
the equations that will be an enormous effort for me number 1 and number 2, it may not
be it may be more error prone when we prepare it ourselves. So, best is to take the make
sure that there are no other serious errors in the text book and take it from there. And so,
this has been taken from the textbook of Mackie and Mackie and in fact, most of the
equations which we have shown before are also coming from the textbook of Mackie and
Mackie.
So, let us now examine this table once more so, that we have now a complete hold on
systematic absences coming due to centering. And I think that will take us to one
particular stage where we can now diversify and see how this structure factor behavior
593
comes. And then take up the systematic absences which come due to the axial systematic
absences the 21 screw as well as the glide planes.
So, let us see how this appears. So, the primitive lattice no systematic absence you are
very clear about it. So, if it is A centered lattice we will have coordinates of lattice point
0 0 0 and 0 ½ ½. So, since the ½ is associated with the k and l direction we will have k +
l odd reflections absent. So, those only those values for k and l we will have the when it
is even we will have the reflection coming otherwise it would be absent.
I do not think I need to take an example to illustrate it, you can take some h k l examples
to find out how this systematic absences coming. B centered lattice likewise and C-
centered lattice likewise and F we have learnt just here I have written the English
explanation h k l neither all odd nor all even; that means, they have to be either all odd or
all even to get the systematic presence.
So, they will be present only this systematic absence. So, we can say neither all odd nor
even. So, the I centered is fairly straight forward (h + k + l = 2n + 1). Now these
systematic absences occur in all general reflections be it projection be it axial it will
occur in all h k l reflections. So, we have to consider all the h k l reflections in this
particular case.
I have given here 2 additional entries we have not discussed that all in detail the
rhombohedral and hexagonal systems, because I just thought that this is slightly away
from the plan of this course. However, we should remember that those symmetry also
occurs and in fact, there are many materials which go with a rhombohedral symmetry.
But there many systems which go into hexagonal symmetry; particularly when we when
people are talking nowadays about organic ferroelectrics and so, on there is a large
number of compounds which go with hexagonal axis, rhombohedral cell, but hexagonal
axis.
So, the hexagonal systems therefore, will have the coordinates of the lattice points are 0
0 0, 2/3 1/3 1/3, 1/3 2/3 2/3 in the case we have a rhombohedral setting of a hexagonal
lattice. What I mean by that is that we have a rhombohedral system where a = b = c, α =
β = γ ≠ 90⁰. So, effectively it is a distorted or pushed around cube it is not a distorted
cube, its a turned around cube about a given axis.
594
So, we will have therefore, three values coming up for equivalent points of the
coordinates and therefore, we have 3 lattice points and therefore, the systematic absence
will come up as - h + k + l = 3 n ± 1. So, this is a very special case we are not going to
discuss it in detail here, but you should know that systematic absences can occur in a
rhombohedral system not in a hexagonal system.
And particularly when the rhombohedral axis are referred to straight away if we describe
the system in rhombohedral axis, then there is no. But if we describe the rhombohedral
axis in a hexagonal system; that means, we have a rhombohedral system, but we say a =
b ≠ c, α = β = 90⁰ γ at 120⁰ that is the unit cell we fix.
That means we now have actually a rhombohedral cell, but we have fixed it in a
hexagonal lattice. If we have a rhombohedral cell and fix it in a hexagonal lattice then we
have this problem and this problem is because the rhombohedral axes now will show
coordinates of these three positions and that is the property of the rhombohedral cell. It is
a special case of the hexagonal lattice all hexagonal lattices will not show. So, in
principle we if we have a hexagonal system which is indexed as a hexagonal system you
have a = b ≠ c, α = β = 90⁰ γ at 120⁰ and then it shows rhombohedral symmetry, then
you will have the systematic absences.
So, this is very interesting suppose you want to find out the crystal system whether it is
hexagonal or rhombohedral, you use this test. So, you use this test if you see these
systematic absences it is rhombohedral axes, then you go down and set it to
rhombohedral axes then you have no systematic absences any more. So, this is a way in
which you can actual distinguish the rhombohedral axes with respect to a hexagonal
system. So, this in a nut shell is what we see a for systematic absences which are
displayed by conventional lattice types.
595
Now, let us go further and see what we have further to offer from X ray diffraction. You
see we are talking about structure factors the systematic absences in structure factors and
so on. So, what we now do is to see what is the way in which structure factors behave;
you see we have seen the intensities how they behave.
In other words we have seen that when you do an X ray diffraction experiment I have
already mentioned several times that the I(h k l) will be equal to I(-h -k -l) and that we
call as the Friedel’s law; that means, that we have the 2 systems where we can now take
the h k l reflections we have observed, it could be centered or face centered or body
centered those systematic absences will be there. But for any h k l measurement if you
also measure -h -k -l these 2 will always be equal for as intensities are concerned and that
is the Friedel’s law.
Of course Friedel’s law can also be violated and if this equation is not satisfied in a
certain circumstance. We now invoke a situation where the electrons are now inside the
crystal systems are now very close to the incoming radiation frequency the incoming
radiation has a frequency, we are ignoring the frequency change because we assume only
coherent scattering, but what happens is if the frequency is very close to this the atom
now absorbs the X rays.
596
Because its it does not know the difference you know its there is a frequency coming,
that frequency matches the frequency associated with the atom vibration or atom the K
shell of the atom and the and the nucleus transition.
So, if both are same the waves cancel each other out and therefore, you will not get any
diffraction you get severe absorption at that situation what happens is the electrons do
not know what to do. So, the K shell electrons you see we the characteristic X rays we
get for copper which we discuss as 1.5418 Å or Molybdenum 0.7107 Å they come due to
the fact that we say it is Kα radiation; that means, the K shell electrons are the ones which
get knocked out by the incoming X ray this is known as Moseley’s law I do not think we
will have time enough to do it in detail.
So, I will measure it now in the next few minutes I will do the orally I will explain how
the Moseley’s laws comes, but I think it is something which is very fundamental and
which can be very easily understood.
So, let us say we have an atom inside the crystal and the copper radiation is sent inside
the crystal, I will make it as simple as possible copper radiation is sent inside the crystal
and then there is a nickel atom which is the next fellow in the periodic table. By the way
you must all remember and appreciate that this year is the year of international tables.
So, United Nations have decided to call this year as the year of international tables.
So, this is a real recognition for the work done by Mendeleev and the periodic table.
Now the fact that we have scattering factors, structure factors intensities are all due to the
fact that we have electrons we have nuclei, we have different kinds of elements and
therefore, our scattering depends upon the different kinds of elements and the amount of
electron density that is associated with them. All these are present in the periodic table.
So, periodic table is in fact, the fundamental requirement for any of these courses which
we discuss this way.
So, its so, happens that this year is the international year for periodic table. So, lots and
lots of international levels activities will be organized may be in the next present part of
the presentation may be I will show you a picture of Mendeleev with his periodic table,
just to remember that it is the year of periodic table just at the passing. It is not relevant
to our course, but why obviously, it is relevant to our course because we are dealing with
597
atoms and the electron density associated with atoms. So, I thought just as a memory I
will show it.
So, we will see now what happens to the structure factors. Intensities we know become
equal. So, if you have a reflection plane which is 1 1 1 and a reflection plane which is -1
-1 -1, the intensity from this will be equal. I was talking about the violation of this
intensity the violation of the intensity comes because the absorption edge associated with
an atom inside the crystal is very close to the incoming radiation.
Then happens is that the K shell electron should get knocked out from the incoming
radiation and then there should be a transition from the K shell from the L shell to the K
shell and that transition gives raise to the characteristic radiation which is 1.5418 Å
which we use.
There are 2 electrons in the K shell and therefore, we have effectively 2 radiations that
should come one for the one electron the other for the other electron, but the 1 and the 2
electrons are not distinguishable and therefore, when we get this diffraction coming out
we will get the α1 and α2 which are very closely spaced. So, for all practical purposes we
decide to use them as an average α.
So, when this is happening and if it is close to the absorption edge, the 2 electrons which
are now getting knocked out they do not know what to do whether they should stay with
the incoming radiation and get out or they should go now and generate the characteristic
radiation. So, the consequence is these electrons now reach relativistic speed if there is
relativistic speed they do not know how to go about and this whole scattering now
becomes anomalous and this process this methodology is called anomalous dispersion.
And when anomalous dispersion occurs it will occur in any crystal.
It will occur to a very small extent so, we can ignore it, but if we do very accurate
measurements of I(h k l) and I(-h -k -l) every atom is associated with an anomalous
component and these anomalous component can also be with respect to the Argand
diagram. So, we can have both real and imaginary components associated with this
anomalous scattering, and anomalous scattering is very useful to determine the absolute
configuration of a crystal.
598
So, we will discuss it when we come towards the end of the course, I do not know
whether we will have time enough to discuss such details, but essentially we will
mention how to determine the absolute configuration for sure. And when we do that we
will bring in and invoke this variation or in other words the violation of the so, called
Friedel’s law. So, we will stop here.
599
Lecture – 45
So, we have been discussing the appearance of systematic absences depending upon the
lattice types. And here is a list of all possible lattices in the seven crystal system, which
will ensure the number of coordinates and coordinates of the lattice points which are
shown here. And the type of systematic absence, we get as a consequence of the
centering of the lattice.
So, the primitive lattice obviously, therefore will not have any systematic absences. What
it means is that all the primitive lattices, which come in all the seven crystal systems, will
not have any systematic absences associated with general reflections. So, if you take
reflections h, k, l, all values of h, all values of k, all values of l, then the systematic
absences associated with a primitive lattice will be none. In the sense that, in principle all
the reflections allowed are present.
Of course, the primitive lattice may carry information about the symmetry, which could
be having a translational component like a screw axis or a glide plane, we shall discuss
that later on. But, at the moment whenever there is a primitive lattice, and none of no
600
translational component associated with the symmetry elements like for example, P 1,
and P ī, P 2, P m, and things like P 2/m and things like that, there will be no systematic
absences.
So, systematic absences come up with the centering, so if there is A centering, we will
have the systematic absence coming in terms of the values which the which adds on to
the axis, so whichever axis to which the ½ adds on. Those axis will show the systematic
absences. So, we get k + l = to 2n + 1 for the A centering, B centering h + l is 2 n + 1,
and C centering h + k is 2 n + 1. We have discussed the appearance and how it influences
the expression for the structure factor in earlier classes.
And therefore, right now we will see that for the face centered, which we discussed with
the example of Zinc Sulphide, h, k, l neither all odd nor all even that means, they have to
be in such a way that all the three indicated here above should be simultaneously
satisfied for a reflection to be systematically absent. What is listed in the international
table, which we will have a look later on or not the systematic absences, but systematic
presences, so that is something which we should remember, when we look at the
international tables for reference.
So, having seen this we now go and see, what are the systematic absences, which come
due to the presence of translational components in the symmetry associated with the
601
space group? So, the space group the lattice type, we have seen. Now, let us say there is
a space group, which is associated with an a glide.
Let us take an example of a primitive lattice with an a glide, it could be an I centered

lattice with an a glide in a monoclinic system or centering could be also a reason, which
is introducing the a glide. So, if there is an a glide which is now in any of the directions
associated with x, y, z. Suppose there is a glide which goes through the origin, and is
parallel to 0 0 1 the plane 0 0 1 the z-axis.
Then, you see that x, y, z every point x, y, z, we will give rise to a ½ + x y and z̅, so that
means the a glide operation takes the point x, y, z to ½ + x y z̅. And, if we operate again
on this one, the a glide will go back to that, so that is the group which we are
considering. And what happens to the structure factor expression, so we write the F(h k l)
expression as before, it is 1 to N, fn exp 2πi (h x + k y + l z).
And, we also have due to the second equivalent position, because again these two are
equivalent of each other, we separate them into two parts. So, the summation goes from
1 to N/2 and fn exp 2πi (h xn + k yn + l zn) + exponential 2πi. This is ½ + x, so ½ + h adds
on here, {½ h + h xn + k yn - l zn}. And, this can be simplified and written as sum over 1
to ½ of N, fn exp 2πi (h xn + k yn) exponential this quantity + exponential that quantity.
Now, remember in the previous case, when we were discussing the C centered lattice, we
had expressions of on the other side of the common factor, which involved only
exponential to be raised to a integer. In other words, we had expressions of this current
exp 2πi (h/2) kind of thing, but we did not have any of the independent axis that is
associated with it.
So, therefore, when we have an a glide parallel to 0 0 1, we look at not the general
reflection, so there will be no general systemic absence. For example, P a if it is a space
group, primitive will not give any systematic absence associated with general reflections.
But, the a glide will give systematic absences, when l is put equal to 0.
What happens, when l is put equal to 0 is this expression will now simplify, so we write
the expression for F(h k 0) that means we are now looking at only the projection
reflections. We look at different values of h, different values of k, but 0 value for the z
602
value. So, whenever this is 0, it is a projection down the z-axis. And, the expression
becomes 1 to ½ of N, fn exp 2πi (h xn + k yn) * (1 + exp πih).
You see, now what I was mentioning has happened? In this case, we have therefore only
the integers in this part of the flower bracket. And, when you have a integers in the part
of the as flower bracket, we can calculate mathematically the value, and that turns out to
be 1 + (-1)h. So, depending on the value of h, this value will be either 2 or 0, so it will be
1 + 1 or 1 + (-1), which is 0 or 1 + (-1)h that will make it -1 or + 1.
And, whenever this is + 1, it will be that h = 2n + 1, that means all odd values of h, this
particular value 1 - 1 will cancel out, and therefore this will become 0, and therefore the
value becomes 0. So, the systematic absence associated with a glide passing through the
going is along the z-axis or parallel to the z-axis, a is perpendicular to the 0 0 1 plane.
Then, it F(h k 0) is 0 for which h odd is absent, so h = 2 n + 1.
So, this therefore that tells us that we can have systematic absences due to glide
operations. We have already seen how many glide operations are possible a glide, b
glide, c glide, n glide, d glide in a special situation. So, all these glides, therefore will
give rise to systematic absences depending upon the projection reflection. So, it is only
the projection reflections, which will show you systematic absences, and we will
examine it a little more closely.
603
Let us take all possible a glides. If you have all possible a glides the type of glide is a
glide, then the translation is a divided by 2. Whenever a is involved, it is h = 2n + 1,
which is systematically absent. For the b glide, it is b by 2; k = 2n + 1. And, of course,
we cannot have a c glide parallel to 0 0 1, we know that. And therefore, there is no c
glide associated with the glide planes parallel to 0 0 1.
But, we can have c glides parallel to the other two directions that means, two other two
planes 0 1 0 and 1 0 0 can have c glides. 0 0 1 which is which is a plane, which is
associated with the direction of z, it is perpendicular to the direction of z. So, c glide is
not allowed, because the translation has to be ½ along that particular direction.
So, we have therefore the a glide and the b glide, we can also invoke the presence of the
n glide, which is a + b by 2. Then, we will have the systemic absence h + k is 2n + 1.
But, all these systematic absences occur in only projection reflections, projection
reflections associated with these. So, in this case for example, it is h k 0. In this particular
case, what would be the thing, I want you to find out it will be h k 0, but this time the k is
equal to 2n + 1. So, the systematic absences are all in h k 0 reflections.
So, in the case of a glide, it is a h odd. In the case of b glide, it is k odd. In case of the n
glide, it is h + k odd. And in case there is a diamond glide, which we have not discussed
which is probably out of the scope of this course. We will not consider this, but it will
also have systematic absences as shown below it is just for our record, we are not going
to discuss how it turns out.
604
So, if that is the case, what are the types of glide planes, we can think of, we can think of
glide planes parallel to 1 0 0 direction, which is along the a axis. Glide plane parallel to 0
1 0, which means its b, and glide planes parallel to 1 1 0, which is which is along the
diagonal.
So, we have listed therefore the planes, which are in these directions, and the glide
planes which are along 1 0 0 can be b, c, n and d. 0 1 0 is a, c, n, d of course, we can
always have 0 0 1 which we already discussed, so we have not put it here. And, the glide
plane 1 1 0, it can be c, b, n and d. And, the set of reflections that will get affected in the
case of the glide plane, parallel to the plane 1 0 0 is 0 k l glide plane associated with 0 1
0 is h 0 l and glide plane this will be h h l.
And, so the systematic absences are listed here. And, these will be the condition, here
you see on the right side, we did not list the systematic absences, but we have listed the
systematic presences. So, k even only present, l even only present in the example of a
glide plane, which is which has a b glide, and c glide respectively it is k = 2n, l = 2n.
So, the conditions for getting the reflections measurable are given here, and this is what
you will see in the international tables. But, for the discussion sake which we have done
so far, the lattice centering, and then also the a glide which is which we have discussed
in detail. We have discussed the systematic absences, because it is always good to see
what is absent than what is present, I mean that is normally the logic you know the there
605
is a statement that if in an audience in the classroom for example, if all students are
present, their presence will be ignored. Those who are absent, they will be noted. So, it is
something like that.
So, systematic absences therefore are very crucial to notice, and that is why systematic
absences are discussed before. But, what is present is more important from the point of
view of the diffraction condition, and therefore the presences are listed here. So,
international tables for crystallography will list only the systematic presences, what are
present systematically alright.
So, this is as far as the presence of the glide plane is concerned. So, we have seen the
lattice centering, and now we have seen the glide planes. So, glide plane give systematic
absences associated with projection reflections. The general reflection systematic
absence give the lattice centering information.
So, the next obvious choice is the presence of the screw axis. So, what is the plane that it
can affect, it can affect not the planes, but axial reflections. By axial reflections, I mean it
will affect the h 0 0, k 0 0, and the l 0 0 with respect to hkl plane. So, it is therefore, the
actual reflections that define the h direction, the k direction, and the l direction, h = 1, 2,
3, 4 like that.
606
So, what happens in such a situation is again, we have to write the general expression for
the structure factor. And, then evaluate the structure factor features associated with it let
us let us look at it in a careful manner. So, in this case the systematic absence, we now
consider produced by screw axis. Please notice the nomenclature; I am just reminding
you that when we put in square bracket, it represents the direction.
When we put in ordinary brackets, it represents the plane. So, the previous representation
was with respect to the plane, and this is now with respect to the direction, so this is
along the z-direction. So, when we talk about the screw axis, the screw axis or parallel to
the 0 0 1. And, when we talk about the glide planes and we have written parallel to that
in the parallel to the planes 1 0 0, 0 0 1, and 0 1 0.
So, systematic absences produced by screw axis now the direction is 0 0 1. So, x, y, z,
now will become x̅, y̅, and you see since the translation component is along the z
direction, we have ½ + z in that direction. So, this screw operation, again operated will
take you back to x, y, z defining the point group.
So, whenever you have a crystal system like P 21 for example, a primitive lattice with 21.
Then you will have this systematic absence coming up, how it comes up, we will discuss
in a minute. So, again we write the general expression for F(h k l), you see again we have
to consider ½ the number of total reflections, because one ½ will obey this, the other ½
will obey that. And, so the expression now f n {exp 2πi (h xn + k yn + l zn) + exp 2πi (-h xn
- k yn + ½ l + l zn)
So, we now take out the common factor, and rewrite this expression the common factor
here between these two is exp 2πi (l z n) fn. So, this therefore becomes 1 to ½ of N, f n exp
2πi (l zn), and exp 2πi (h xn + k yn ) as far as this first expression is concerned. And, as far
as the second expression is concerned, it is (-h xn - k yn + ½ l).
So, the same logic as we defined in the previous case that we would like to have a
expression which will essentially depend upon integral values to take the exponential,
then we can calculate it mathematically, and so we separate it out. The way to separate it
out is to put h = and k = both equal to 0.
So, if h and k are both equal to 0, we have only reflections of the type 0 0 l. So, these F(0
0 l) now will be having a formula reduction, which is going from 1 to ½ of N, f n exp 2πi
607
(l zn), multiplied by this since h is 0, k is 0, this will go. And, therefore, you get (1 + exp
πil). And, we know this value can be (–1) l. And therefore, it has to be again the same
logic that whenever l is odd, F(0 0 l) will be absent. So, the systematic absences
associated with the with the screw axis essentially affects only the axial reflections, so,
the reflections which go along the definition of the axis either h or k or l.
And, so we can now write down those expressions, and see how it comes about. So, we
can have screw axis in variety of crystal systems as we know, we can have 2 1 screw, we
can have a 41, 43, 42 in a tetragonal system. We can have 3 1 and 32 in a trigonal system.
We can have 61, 65, 62, 64, 63 in an hexagonal system. So, these are all the possible screw
axis, we can have in a crystal system.
And therefore, we now look at the corresponding systematic absences. If, the screw axis
is parallel to the z direction, so that will affect the 0 0 l reflections only, and the
condition is on l value. And, that condition on l value for 2 1 screw axis, the translation is
along the z direction. So, it it moves along the c axis, c divided by 2, that is the
translation associated with the screw axis.
So, we get therefore the systematic absences to be l = odd, so all odd reflections will be
absent. What is interesting is the systematic absences follow a certain trend among the
tetragonal, and the trigonal, and the hexagonal systems that is because in all these cases
the unique axis is z. And therefore, the screw axis is parallel now to the unique axis.
608
Since, the screw axis is parallel to the unique axis or along the unique axis, the
systematic absences will develop differently depending upon the nature of the screw axis
operation which we invoke. For example, if you have a 4 2 axis, for all practical purposes
a 42 axis is effectively a 21 axis, and therefore you will have only l odd absent.
So, suppose there is a space group P 42, then the systematic absence in that particular
space group will be just 0 0 l, l odd absent. On the other hand if you have a screw axis 4 1
and 43, these two axis are now in two opposite directions, but representing the same
effective positioning of the atoms except that the handedness of associated with 1 and 3
will change.
And therefore, l is going to be ≠ 4n that means, whenever l = 4n, you will get the
reflection. So, n can be equal to 1, so you get reflections like 4, 8, 12 and so on.
Multiples of n, n can take values 1, 2, 3, 4 etc.., So, it depends upon the length of the unit
cell, suppose the unit cell length is about 20, then you will have the reflections coming
only for those reflections 0 0 l which is l = 4, l = 8, l = 12, l = 16, and l = 20. So, you will
get five reflections along the axial line.
So, you can easily observe, when you look at the reciprocal lattice image, which you
have measured. And, particularly if you have done a photographic measurement, it is
easily seen in earlier days. But, now a days you can always have a program which looks
into the possible systematic absences, when once you have a collection of all the hkl
reflections. So, you have done the experiment, you have put the crystal on the beam, you
get the diffraction done.
And, then you have the reciprocal lattice image, the images of the reciprocal lattice, you
examine, after indexing them with respect to h, k, and l. We will see how that is done in
a after probably in a couple of classes, we will know it the authoritatively how to do that.
And, when once we have this hkl information available to us, we examine the values of
hkl, and the systematic absences will show, if there is anything.
For example, in this case all values of l = 4n will be present. So, we can easily identify as
far as 0 0 l is concerned, whether it is 41, and 43, for both it is the same systematic
absence. Same is true with 31 axis, and the 61, and 65, you see l ≠ 3n, l ≠ 6n. And, in case
of 62, 64 also, it is l ≠ 3 n.
609
So, what it means is that if we by looking at the systematic absences associated with
these crystal systems, you cannot decide the space group, please note the point. Suppose,
there is a space group, which is P 31, and then a crystal system which goes into P 32, by
just looking at the systematic absences, you cannot say whether it is 3 1 or 32. Same is true
with 61, 65; same is true with 62, 64; and same is true with 41, 43.
So, only when you have the ½ the translation of the length, for example 6 3, for example a
42, because there is no ½ translation possible in the threefold-axis. So, 3/6 th is one
effectively, if you divide this by 3, it will become 2 1, 42, we divide by 2, it will become
21 and 21. So, the systematic absences are all identical, their l = 2n + 1.
So, having seen this, we will see what happens in case the systemic and screw axis are in
different directions. So, screw axis is parallel to a, screw axis is parallel to b, and screw
axis is parallel to 1 1 0. Whenever the screw axis is parallel to that direction that
particular direction only is affected that means h 0 0, h = 2n, and h = 4n are present in
these four crystal systems. In these four crystal systems, if the screw axis is b, we will
see 0 k 0 k = even or k = 4n systematically present. So, these are the present conditions,
this is what we mean by conditions, in fact that is what is written in the international
tables as well.
Screw axis parallel to 1 1 0 that is quite possible, and then you have therefore the
systematic absences h h 0, this is possible in tetragonal systems, because you have a = b
610
in a tetragonal system. So, you can have a 21 screw axis, parallel to 1 1 0 direction. Since,
the crystal systems could be ideally identified based on these systematic absences.
So, essentially we have now covered all the symmetry elements, which will have
translational components. The centering information that is with associated with a lattice
will also give ½ ½ additions and so on. And therefore, we have these systematic
absences. We have systematic absences due to the glide planes; we also have systematic
absences due to screw axis.
So, these translation components which are present only in crystals, you see that is a very
important point to remember. These systematic absences are present only in crystals, and
they affect only diffraction. And therefore, they it allows for a unique determination in
most cases of the space group, just to give you an example. Suppose, there is a space
group P 21/c, you for the primitive you get no systematic absences. So, you will have all
h k l reflections present in principle. It is not necessary that all hkl value should be
measurable, some of maybe weak; some of them may be strong.
And, we know when it is weak, when it is strong by our previous discussions. It depends
upon where the atoms are sitting in the unit cell, the contribution to the diffraction, the
intensity contribution to the diffraction is coming from individual atoms, which are
inside the unit cell, but the measurement which we do is with respect to the plane.
So, what is it that is coming out from the plane? And therefore, we draw we consider this
formula Dhkl/dhkl. And calculate the values that come to the intensities based upon 2π (h x
+ k y + l z), and therefore we have this general formula which will allow us to calculate
the intensity.
So, the intensities need not be present for all possible hkl, some of them may be absent,
some of them may be present. Some of them may be absent, because the waves coming
out successively interfere such that they destructively interfere ok. So, when they
destructively interfere, we will not get any intensity, it is not a systematic absence. The
systematic absence should be throughout the data, what I mean to say is watch my words
carefully. If h k l, h + k odd is absent, it will definitely tell you that it is a centered lattice.
But, by chance in a general h k l data, some h + k odd may be absent, but it is not
systematic that is allowed, then it is still a primitive lattice.
611
So, only when all the reflections of your h k l collected obey these conditions, then only
you can say this has a symmetry. This case becomes very very crucial. In case you have
a monoclinic system for example, in case of a monoclinic system, it may so happen you
are looking let us say at the unique axis b, and it may so happen the unique axis is about
4 Å. You may have a and b very large, but the c axis is 4 Å.
So, if there is a 21 screw axis associated with the b direction, then what would be the
systematic absences, the systematic absences will ensure that only 2 0 0, and 4 0 0 sorry
0 2 0 and 0 4 0 will be present. So, 0 1 0 and 0 3 0 will be absent that is the extent to
which we can go, because we can measure only two reflections. In fact, we can measure
four reflections 0 1 0, 0 2 0, 0 3 0, 0 4 0. So, we have four reflections at our disposal
along the c direction, along the b direction sorry correction along the unique axis b. So,
there are four reflections 0 1 0, 0 2 0, 0 3 0, 0 4 0.
So, if they are to be systematically absent, and give rise to the presence of a 2 1 screw
axis, we have to measure we should be measuring only 0 2 0 and 0 4 0, they should come
with some intensity. It may so happen the the atoms and the molecules inside the crystal
may so arrange themselves that 0 2 0 and 0 4 0 may also become very very weak, so
weak that we cannot measure them.
Then you will see all the four being absent or in the case of a system which has no screw
axis, you may measure let us say 0 2 0, 0 4 0 as possible reflections. It may so happen the
atoms again arrange themselves, even though there is no 2 1 screw axis, such that the
contribution to 0 1 0 and 0 3 0 is very very small, then it looks as though there is a
systematic absence. So, these are all pitfalls in uniquely determining the space group.
So, determination of P 21 with a unique axis b very small is a question. And in such
situations, we might have to look at the structure determination. And, when once we
determine the structure, in fact determine the structure in both space groups P 2 as well
as P 21, and then decide which is the space group into which it goes into.
So, the determination of the structure then becomes crucial to identify the presence of the
21 screw axis, so that is where we have to be a little cautious. But, routinely if you have
got a P 21/c, it will have 2 systematic absences 0 k 0, k odd absent, and h o l, l odd
absent. If h o l shows a h + l odd absence, then it will be an n glide. So, we can uniquely
determine the space group P 21/c or P 21/n, by just looking at the systematic absences.
612
As I said primitive will not give any systematic absence, so you will have all general
reflections present. Then you look at the projection reflection. So, systematically you
take the data; arrange it in such a way that you have all the general h k l, the projection
reflections and then the axial reflections.
And, the projection reflections if you look at now, they will have systematic absences
corresponding to the c glide that means, h o l, l odd absent. So, the glide planes will
affect the projection of the along the plane that means, it is 0 along that particular
direction where the axis is located that means, if you have in the case of a monoclinic
system, we have a 21 and c, the unique axis is b. So, about the unique axis, we have the
21 screw, so 0 k 0, k odd. And about the unique axis, we have the mirror plane
perpendicular to that, because the point group symmetry is 2/m for P 21/c.
So, we therefore have the systematic absence coming with respect to h o l, l odd absent.
So, space groups like P 21/c can be uniquely determined. And, this is an advantage we
have, because this particular feature will enable us to identify the space groups uniquely.
So, for a simple discussion many space groups belonging to different crystal systems can
be uniquely identified by an examination of the h k l reflections. It is only those in which
there is no translational component involved, these space groups are not identified by
So, those space groups of course have to be identified, whether they belong to they
belong to the system. And, what is the for example, what are the symmetry elements that
are associated with it. Take for example, P 2 and P m; it is very hard to distinguish
between P 2 and P m, because we do not have systematic absences.
So, of course there are ways and means in which we can identify probably one of the
space group to be correct, we will discuss that later, when we look at the intensity scale
statistics. So, since intensities come from the scattering overall of the unit cell, so it does
not matter in principle to the intensity the orientation of the planes. The intensity is
coming from the electron density, and the electron density is continuous in the space.
So, the way the intensities distribute themselves among various let us say values of sinθ /
λ, because we know that is the one which controls it, because scattering factor falls off
with respect to sinθ / λ asymptotically. So, since scattering factor falls off with respect to
sinθ / λ asymptotically, we have a dependence of the intensity on that value of that
613
particular atom. And, so the intensities will fall off with respect to sinθ / λ. Apart from
that the intensity is therefore distribute themselves into various regions of the scattering
angle, so there is a statistical evaluation that is possible for the intensities.
It so happens, for example if you have a space group P 1 and space group P ī, the
distinction between P 1 and P ī is fairly straightforward in the context of intensity

statistics. Because, the intensity distribution will be different in case of 1 compared to the
distribution in case of ī. And this we will discuss when we do the scaling operation and
how to put the intensities on an absolute scale. At that time we will discuss this issue,
because that is it is at that time we can distinguish in such cases, where systematic
absences are not available, we have to find out the nature of the space group.
So, whether P 1 and P ī there is a triclinic system, whether the crystals are now going
into a centric system or a non-centrosymmetric system can be identified only by an
analysis of the intensity statistics, which goes along with it. So, to summarize all the
issues done so far, we have identified systematic absences with respect to the lattice
centering. So, we have identified systematic absences with respect to the glide planes to
be with those of projection reflections, and with the screw axis with those of axial
reflections.
So, once we accumulate a data, collect the data, after doing the diffraction experiment.
We need to analyze the h k l reflections, for this systematic presence or absence. And, if
this is nicely satisfied, we can uniquely determine the space group. If, it is not nicely
satisfied, we can still have an ambiguity in the space group. We can still identify the
space group, but we can say this is this or that.
For example, we discussed just now the possibility of 3 1 and 32, so we can say well the
space group is P 31 or P 32, but we are not sure whether it is 31 or 32 as of now. And, that
those are in fact 31 and 32 are known as an enantiomorphous space groups. And it is
required to do the structure in both, and finally identify which is the realistic space group
associated with 31 and 32.
614
Lecture – 46
So, having got an idea of the Systematic Absences, it is now necessary to read it out of
the international tables for crystallography. This as we have already discussed this
equivalent point diagrams we have discussed the symmetry diagrams. So, we do not have
to go into detail of that again.
But, what I will do is to show you some of these space groups and see how the
systematic absences appear and we will see how to associate systematic absences with
respect to general and special positions.
So, the P1 of course, is a redundant set. So, I am not going to take into account P1
instead we will go to number 2, P-1. I just recollect for you, this is a full symbol is -1 and
this is triclinic and the point group symmetry is -1, it has a center of symmetry. So, what
happens is that the center symmetry is developed as per the symmetry diagram here and
the equivalent points developed as per the equivalent points here and this particular slide
we have discussed before in full detail.
615
Now, comes the centri I said we will discuss this later on and this centri tells us the
possibility of the multiplicity, Wyckoff letter and site symmetry associated with each one
of the types of reflection that types of coordinates we can have.
So, the first one is 2, is a general position atom x, y, z it gives raise to -x, -y, -z. This is
P- 1 and you see there are several centers of symmetry which will develop because of the
nature of the center of symmetry itself and also the periodicity associated with the lattice.
So, we have therefore, 1,2,3,4,5,6,7,8 Wyckoff positions and addition, the general
position taking the value of i and the occupancies are the multiplicities that are
associated with these are in these case it is 2 coordinates in all these cases is 1
coordinate.
So, this part also we have studied in detail and thoroughly. So, I will not repeat it. What
we see is on the right hand side? Right hand side there is nothing much to discuss in P-1
because the reflection condition. So, what are the conditions in which the reflections
occur? This is general; that means, this reflections which are referred to as the general
reflections the general reflections have no conditions and as a consequence this special
reflections, special positions which are shown here also has no extra conditions. So, P-1
is that way not very exciting to discuss any further, it has no systematic absences.
So, as I have already mentioned between P1 and P-1 therefore, there are no systematic
absences. So, when you determine the by the based on the diffraction condition when
616
you determine the space group you say this space group at this moment is either P1 or
P-1 and later we will decide how to distinguish between P1 and P-1 and then finally,
determine the structure.
So, let us now go further to the space group number 4. Again, P 2 is not listed because it
is not going to have any systematic absences.
P21: P21 has a 21 screw axis, the full symbol is 1 2 1 1 and you know what the symmetry
diagram as well as the equivalent point diagram. So, we will not go into the unnecessary
discussion of that again.
617
The point group symmetry is 2 and therefore, this has two positions no special positions
for this. So, what is important is this now; the reflection conditions h, k, l since there are
no systematic absences associated with primitive lattice, we have taken P21 as the
example, no systematic absences will be there. But, if we look at the systematic absences
due to the symmetry element the symmetry element is 2 1 screw and therefore, you get,
but the reflections which are now along the axis 0k0 they will be systematically absent.
So, k is equal to so and so.
So these entries will be there in the international table. I have blown it up so that we
discuss it more detail. Normally, this appears parallel in this direction to this in their in
the international table sheet which you have already seen. So, I do not want to go into the
detail of it. So, the equivalent points are 2 and the systematic absences 0k0, k equals
even, unique even.
So, therefore, we can uniquely determine the space group P21 provided as we discussed
earlier on the b-axis has sufficiently large you know cell dimension. It so happens many
organic compounds which show particularly photo reactions and they have the 2 plus 2
additions among ring systems and so on. They always have a normally in the sense that
unite cell dimensions or less than 4 angstroms because it so happens in a photo reaction
condition when two double bonds have to interact with each other then the distance
618
between the two double bonds should be closer than 4 angstrom this is the Smith rule.
And, in fact, that is the beginning of crystal engineering, we will discuss that later.
So, the two double bonds when they come close to each other they should come within 4
angstrom and therefore, the unit cell dimension there is a consequence will be short and
this short units cell, the confidence we can put in the observed systematic absence
becomes less and therefore, we can we can generally see whether this crystal system is
either P2 or P21 and in fact, in the systems the organic systems which undergo this deals
order 2 plus 2 addition reactions you do see that they form a four membered ring and it is
so happens that the crystal structure does not change much that is that is the aspect of
crystal engineering which we will discuss in a later class.
But for all practical purposes if we have a P21 symmetry for our current discussion 0k0,
k odd will be absent. So, we identify the actually absence and determine the space group
uniquely. So, it is P21 system which can be uniquely determined.
Next we go to the case where we have a centering I am going through the lower
symmetry space groups essentially because I want you to get this point very clearly in
your mind and how to also read it out of the international tables because when you do the
structure determination and the machines are used automatically these days, systematic
absences will be listed then. In fact, the programs will pick up the space group also for
you.
619
But, very often in a situation like what I discussed before about the 2 plus 2 addition
reaction compounds the machine may decide something which may be wrong and that is
where you are required to know what is going on here; and that is the reason why we are
discussing this in detail. And, if you if this space group is really P21 and you want to
solve this structure in P2 you will never solve this structure and that is the beauty of
crystallography.
So, the presence of the symmetry dictates everything and that is why we are studying
symmetry and structure in detail in this particular course. So, this is the diagram of C2.
There is no need to discuss this anymore. We have done it many times. So, it develops
additional 21 screw axis and so on and then these are the equivalent point dispositions.
So, because it is a C centred lattice we have four equivalent points, two due to x y z; -x y
-z that is the two-fold operation and then we have the additional 1/2 1/2 0 as a additional
lattice point. So, we get 1/2+x 1/2+y z; 1/2-x 1/2+y -z and these will be the four
equivalent points in general positions. And, in general positions we have the set of
systematic absences which will allow us to see this space group to be C2. We have to see
this a little carefully and you will see you will enjoy this discussion in some sense
because you see some of the reflections which are shown here corresponds to the
systematic presences associated with glide planes.
620
Now, you see the diagram here there is no glide plane in this diagram, right. There is no
glide plane in this diagram. It is only possibly presence of the new 2 1screw axis, but
when you look at the systematic absences you will see absences due to glide plane. You
may be wondering what is going on and this is very crucial to analyse this. So, we will
go through this systematic absences in this space group C2 a little carefully. So, the four
equivalent points and the centering causes the general reflection hkl to show systematic
absences. So, h+k odd will be systematically absent. So, h+k = 2n is only present this is
clear. So, we will have the general reflections, analyse them, you will get h+k = 2n.
In addition, you will get these many additional systematic absences. Now, these
additional systematic absences are not going to complicate your space group because the
crystal system is monoclinic you have already determine a, b, c, alpha, beta, gamma.
You have determined also it is the C centered lattice. So, C centering gets the preference.
So, it is always the centering of the lattice which gets the preference; that means, the
Bravais lattice information gets the preference over and above the glide plane systematic
absence and over and above comes the screw axis operations.
So, in this particular case h0l for h = 2n is listed. In fact, we should put in principle all
these in brackets to show this is a consequence of this absence, see because hkl. h+ k
even is absent is a present, sorry, even is present, odd is absent. So, we will discuss
systematic absences even though this systematic presences, we will discuss systematic
absences. So, hkl: h + k = 2n, h0l: h = 2n that is because you put k = 0 here you get h0l:
h = even. Same thing with 0kl: k = even. So, these two are a consequence of the main
systematic absence. So, this has no meaning what is. These two will not carry any
meaning you have to ignore this even though you will see this.
So, when you examine the systematic absences you will see absences due to glide planes,
but these are not absences due to glide planes, ok. So, projection reflections are absent
mainly as a consequence h plus k being even that is because you put the value of k
equals 0, h equals even value of h equals 0 you get k equals even.
The then of course, hk0 you see it is shows h + k even; that is also a consequence of that
because l is 0 here and it is a special case of l. See there are no systematic absences
associated with l. So, l will be also showing h + k = even. All l values will show h + k
621
even and therefore, l equals 0 will also show that. Same logic 0k0 now shows as though
there is a 21 screw axis h00 shows h equals 2n as though it is a screw axis.
So, what is very important is if you look at this particular diagram there are screw axis
one fourth removed from the origin and it is not a consequence of these screw axis which
come as a special case because of the presence of the centering. Sorry, I went to the
wrong one because of the presence of the C centering. So, this is the 21 screw axis which
will develop at one fourth distance as a consequence of C centered. This has nothing to
do with this systematic absences which you see here.
So, let me repeat it very quickly. So, that if there is no confusion we have a space group
C2. C is centering will ensure that we have hkl, h + k even only present.
So, the fact that this happens it gives raise to certain special additional conditions is not
special additional conditions. The additional conditions are h0l: h even, 0kl: k even, hk0:
h + k even, 0k0: k = even, h00: h = even. So, these are consequences of the fact that hkl
is h + k even. The possibility of a special positions do exists in this particular space
group. You see two possibilities, one is a presence of a symmetry element two-fold
which is associated with 0, y, half another is a two-fold which is associated with 0, y, 0.
But, it is not going to give us any special or no extra conditions that is because this is a
two-fold axis. This does not carry any translational component, alright.
622
So, having noticed this we will go to the next system which is the most common system
we come across when we are dealing with organics structure. So, P21/c. P21/c the full
symbol is P 1 21/c 1 and it has the symmetry positions like this. We have the center of
symmetry. We have the two-fold axis and we of course, have the one fourth at one fourth
represents of the c glide above the plane and then these are the equivalent point diagram.
This shows where the equivalent points are z is equal to 4.
And, therefore, the general positions are 4. The four general equivalent positions are
listed here and then there are some special positions which are present in this particular
space group and those special positions are now associated with a -1 symmetry.
Remember that, this is again has no translation component, but it has a -1 symmetry;
that means, it introduces and additional symmetry element these has two equivalent
points rather than four equivalent points.
So, let us see what is the effect of this on the general reflection and special reflections.
So, here the general reflections show the presence of 2 1/c uniquely. So, you can
determine the space group without any false because this will have h0l: l odd absent and
0k0: k odd absent. So, the h0l: l = 2n will also additional introduce a condition 00l: l
even because you put h = 0 there, l = even will be the reflection that will be present. So,
it shows a systematic absence associated with axis in two directions, but there only one
623
direction that is the 0k0; k equals unique is unique with respect to the 21 axis. So, this is
due to the c glide and this is due to the axis 21 and this is a consequence of the c glide.
So, you have to therefore, see when you get the systematic absences or presences
indicated in the international table. Like this you have to make sure that which is the
one ,which is a consequence of the earlier listed systematic absence. So, the preference
goes from hkl. No general reflections are no conditions on them and when you go to h0l
the projection reflection there is a condition and this projection reflections can put
additional conditions as though there are coming on the actual reflections which is a
special case of this. And, then 0k0: k odd will be re systematic absence that is due to the
presence of the screw axis.
Now, there is something interesting that is happening here. Apart from the fact that we
have the 21/c absence we get additional systematic absences with respect to each one of
these. Now, what happened to the logic we defined here? The logic we defined was that
if there is no translational component, no systematic absence, then how did this fellow
come up? This fellow again came up because of the C centering. See, because what
happens here is if you want to have the -1 to be evoked you see that the value of l here or
z here has to be 1/2 or 0. So, the value of z is always either 1/2 or 0.
So, the value of z being half or 0 will decide where these systematic absences should
come from. So, here everywhere it is k + l odd which has four equivalent points. It has
special positions and these special positions are associated with the center sub symmetry.
But, what happens when these are associated with the center of symmetry is that the
value of z also becomes special in the sense there it is now either 1/2 or 0 and then 1/2 or
0, 0 or 1/2, 0 or 1/2.
Now, since these are now the positions of the covalent points. You see that the presence
of the symmetry which is -1 will generate only two equivalent points associated with the
particular space group and therefore, we end up with conditions which are special on the
all these conditions now or with respect to the general reflections. So, along with the
presence of the systematic absences which are written up here the general reflections will
also show k + l = even, k + l = odd, k + l = even, h + l = even. All these are even when it
has to be present. And, therefore, k + l even will make it make the hkl. This is now you
624
notice that with respect to the hkl. The systematic absences are with respect to the
general reflections.
So, all the general reflections will show the presence of the -1 symmetry that is
associated with P21/c. So, if there is a molecule with a special symmetry which is
occupying this particular space group you will have these additional special condition
symmetry absences. It looks as though you know that general reflections have an
absence; that means, it has to be centered it is not necessarily that it has to be centered
this indicates the a centering, but it is not really A centering it is a consequence of the
atoms sitting in special positions. So, the fact that when we are analyzing is crystal
systems like P21/c we first look at these absences even though if the atoms are molecules
are associated with this special symmetry then we will get additional special absences.
So, in general P21/c can be uniquely determined by just looking at these systematic
absences.
So, having done P21/c we will now go to an orthorhombic system where we have more
than one axis intersecting with each other or not intersecting with each other as is this
particular case the crystal system P212121. We have studied this in detail and we find that
this has three non-intersecting screw axis removed from each other from a defined origin
by one fourth in each direction. Therefore, this symmetry elements are indicated here and
625
equivalent points are indicated here there we have studied it in details. So, I would not go
into the further discussion.
The equivalent points therefore, are 4 and there are no special positions in this crystal
system obviously, because even these regular symmetries that are present are not
intersecting with each other. So, therefore, you have four equivalent points and the
conditions which come are independently for each 21 axis. So, we have h00: h even only
present, 0k0: k even only present, 00l: l even only present. So, you can determine the
space group 212121 again uniquely.
So, this is an advantage because these kind of systems the symmetry elements do not
intersect with each other and this is a non centro symmetric system and in the non centro
symmetric system there is no question of having a center of symmetry and also its
association with the origin. So, the origin as no special significance in this space group
except that it has to have the three to four axis perpendicular to each other removed by
one fourth from this origin each in each direction. So, that will generate this systematic
absences. So, 212121again is a crystal system we can uniquely determine.
So, you see that the exercise we were doing before into understand the presence of
symmetry elements and the way in which equivalent points develop and so on practically
becomes redundant because if you do an X-ray diffraction experiment and record the
pattern, the recorded pattern will tell us clearly what is the space group if we look at the
626
systematic absences. So, the appearance of the diffraction parts now which follow the
rules of symmetry and therefore, the fact that direct space lattice and reciprocal space
lattices are related to each other we are looking at the reciprocal lattice points hkl and the
systematic absences associated with this hkl reflections will tell us this space group.
So, we do not have to go to the question of finding where the equivalent points develop
with respect to the symmetry operation. So, the previous exercise which we have done is
to get an understanding of how symmetry gets generated in this crystal systems and
space groups and what effect it has on the diffraction is shown here in terms of the
systematic absences. So, this is a very crucial issue in deciding the crystal system. If
someone goes wrong in identifying the space group, if the machine goes wrong in
identifying this space group which is quite possible then your structure will never be
determined or whatever structure you determine will have all kinds of short contacts
within atoms and so on and you will wonder what is going on.
So, for a practicing crystallographer who does not want to study crystallography or who
does not want to look into this systematic absences and so on sorry, the symmetry
positions and equivalent points the observation of systematic absences understanding of
it is crucial because if you have to determine this space group uniquely before you
proceed to determine this structure because this is therefore, the basement on which you
are construction has to be developed. So, determining the space group is a very important
factor.
So, you have three things here to remember one is the geometry of diffraction which is of
course, decided by the Braggs law conditions. So, if you imagine a powder diffraction
diagram you have the intensity versus sine theta, you have a powered diffraction which
intensity this is the intensity I here it is the 2 theta, this is the powdered diffraction we
record. So, your powdered profile will goes something like this. So, this will now the x-
axis will tell you where the 2 theta value should be for the given peak this peak of
course, assuming that it is all unique then we can determine where this peak positions
come and this gives us the information on the d value and therefore, the cell dimensions.
So, we can determine the a, b, c, alpha, beta, gamma using that information. The
intensity information now tells us the value of the intensity associated with this
diffraction which is coming up for that particular plane hkl. So, we determine therefore, I
627
of Ihkl. So, then after determining the densities we now examine the systematic absences.
When we examine the systematic absences we find that the you can determine this space
group some of them get it determined uniquely.
So, we determine the space group and that is the issue. Space group determination
therefore, involves both the x and y coordinates in its own way. Even though we do not
still have the structure information, we have the intensity distribution available to us. The
way intensity is distributed is decided by the nature of centering of the lattice or for that
matter the presence of symmetry elements in a given direction and therefore, we see that
the utility of all the information in an x-ray diffraction experiment goes into determine
the space group. And, therefore, the determination of the space group by examining
equivalent points is a very important step.
And, if there is some mistake done by the by the data collection protocol introduced into
your machine; in fact, many often you will see that when you are doing the data
collection in a practical sense you have crystal grown put it on the diffractometer, you
measure the intensities coming out and then you press the button determine space group.
More often then not it will say not possible to determine, no success it will say. Then you
have a serious problem and that is why this is this part of the course is very important
you should know what are the systematic absences you should look for in what way you
have to do this diagnosis.
The diagnosis follows by looking at the general reflections and then followed by
projection reflections, then of course, later the actual reflection. Based on that you
identify the symmetry elements which might come for example, in this particular case a
212121you examine three axial reflection systematic absences. And, if they do appear like
that then we have a situation where we have uniquely determine the space group. And,
therefore, the information that we get from the hkl reflections enormously important;
both the intensity as well as their appearance with respect to the geometry.
So, the full data associated with x-ray diffraction is incorporated into the determination
of the space group and therefore, is a very important step. People generally ignore,
particularly those practicing chemists and pharmaceutical people and also many of the
people who are just physical chemists and the organic chemists and so on, they ignore
this and if the space group is long they structure can never be determined. So,
628
determination of the structure therefore, solely depends upon uniqueness with which you
are determine the space group and this is a point it should bear in mind.
629
Lecture – 47
Friedel’s Law and Laue classes
Let us go to another example which is Pbca. This is because this is one of the see I have
taken all the common crystal systems, which come across which you come across
particularly chemist come across and also in pharma industry. You will come across only
lower symmetry systems, because you are dealing mostly with organic compounds. And
the pharmaceutical compounds have all these functionalities which make the molecules
very flexible. And so the crystals systems into which they go into a lower symmetry
space groups and therefore, I have considered the monoclinic triclinic monoclinic and
orthorhombic systems.
So, among these systems I have considered only the most commonly occurring systems.
So, we discussed P21/c as a most commonly occurring center symmetric monoclinic
system, the non center symmetric 212121 in the orthorhombic case is very often occurring
and Pbca is the one which is its counterpart as far as center of symmetry involvement is
concerned. So, it is actually P 21/b 21/c and 21/a; that means, it is a b glide perpendicular
to the a axis, c glide perpendicular to the b axis, a glide perpendicular to the c axis.
630
So, this in this particular system of course, we have studied the way in which the
symmetry elements come up, additional symmetry elements show up which are not
really indicated here they will also show up. And as a consequence we get eight
equivalent points in the unit cell.
So, those equivalent points are the general points and therefore, there 8 equivalent points,
which represents the multiplicity of this site x y z. And then the symmetry associated
with the site is 1. This particular space group has two possible special positions. So,
again associated with -1, as you see it is a consequence of the value becoming 1/2 or 0,
depending upon the generation of the symmetry associated with these 8 equivalent
points.
So, what are the conditions? The general conditions are 0 k l: k even, h 0 l: l even and h k
0: h even. So, what does this represent? This represents a, b and c glides. In the
association of these three with respect to the a b c makes it Pbca, but the systematic
absences are with respect to a b and c all three together. Now the special reflections to
actual reflections h 0, 0 and 0 k 0, 0, 0, l will also develop because it is the system is 2 1/
b 21/c 21/a the full system. Because, the point group symmetry associated with this unit
cell is m m m, 2/m 2/m 2/m and so these are the general systematic absences.
Ones it occupies this special position, the central symmetry in Pbca it takes up two
possible Wyckoff positions a and b, c be in the general Wyckoff poison. And this
631
Wyckoff position will generate 4 equivalent points 0 0 1/2 So, you see that the
conditions look as though what? It looks as though they are F centered lattices you
remember the conditions for F centered. So, this looks as though they are F centered
lattices. So, what you have to do is never look at these special absences, look at only the
general absences to determine this space group.
So, these are occurring mainly because of the way in which the atoms are positioned
inside the lattices. So, suppose you take a Pbca lattice and put these equivalent points
you will see automatically the generation of the face centered symmetry. Because, you
have a 0 0 0 you have a 0 1/2 1/2, you have a 1/2 1/2 0 and you also have a 1/2 0 1/2.
You see here this 4 positions essentially tell you it is a face centered lattice. The way the
atoms are located this is what we did in the case of Zinc Sulfide, when we took the
example of zinc sulfide in the earlier classes. And, we found that it belongs to C2/c, if
you want you can go back to that example you will see that the presence of the face
centering in Pbca comes because the atoms are positioned in such a way they mimic a
face centered lattice.
Since, this is now face centered lattice you will get face centered absences in the general
reflection and so it is not an F centered lattice. It is still a Pbca with special position,
which is associated with Wyckoff a and that is how we have to interpret the symmetry
indications shown in the international tables for crystallography. Now you see the
importance of analyzing the equivalent points their presences and absences and then
identifying the in a space group to the extent possible. So, again and again I have been
stressing that if there is an error here you will never get this structure determine. And if
the machine also makes the error which is quite possible then also you will never get this
structure.
So, this point has to be borne in mind and what you have learnt so far in this particular
course is very crucial for this, see once you have this unusual systematic absences
coming up, since you have not done the structure you see the these are coming up
because now you know where the atoms are sitting right in this special position. It does
not tell you the structure information you do not have still x y z that is what you have to
determine, but before you even determine the x y z you know in this particular case the
atoms have to sit in special conditions and follow the systematic absence. So, it is an
advantage. So, it this is the advantage of putting the atom user molecules inside the unit
632
cell and keep the symmetry information which is very crucial for determining this space
group.
I think I have stressed enough on this space groups until now and we will wind up the
discussion on the space groups at this moment. So, we will now look into the other
properties of the structure factors. So, we have actually here when we are discussing the
systematic absences we are actually looking at the presence or absence of structure
factors right. Because it is Fhkl we are looking on to.
So, the Fhkl as a special property, that is known as the Friedel’s law. So, the Friedel's law
is a very special property which makes the diffraction always cento symmetry. So, your
crystal system may be non-central symmetric let us say P212121 is your space group.
Then it has to be a non-center symmetric system, but when you do the diffraction and
when you do the scattering experiment and get it diffraction done you will get a center of
symmetry associated with your h k l distribution; that means, for every F hkl you have a
corresponding minus of F-h-k-l. And this is known as the Friedel’s law.
The disadvantage with Friedel’s law is the fact that you cannot identify the non center
symmetric feature, associated with your space group or for that matter the symmetry that
is also associated with your crystal system because the data always is centro symmetric.
So, when you do the X-ray diffraction experiment your number of point groups will
reduce from 32. In fact, it will become only 11 and these 11 are known as the Laue
633
group. So, diffraction experiment therefore, will not directly give you the 32 point
groups.
This is the revelation at this moment, but we will we will have to live with it because X-
ray diffraction is centro symmetric, why is it. So, intensity of X-ray diffraction is the
product of relevant structure factor and which complex conjugate. Mathematically Ihkl is
expressed as Fhkl times F*hkl and I have told written down below what is F hkl and what is
F*hkl. So, if you look at F hkl it is sum over 1 to n of the cosine term plus i times sum over
1 to n of the sine term. This is the general expression for the structure factor because now
you know we are not expressing in terms of the exponential we are expanding it as
cosine theta plus i sine theta.
So, this therefore, is Fhkl the corresponding complex conjugate F*hkl can be written as I
equals to 1 to n of the cosine part minus i times the sine part then if you take F hkl and
F*hkl and take their product you will end up with I and therefore, when you take this
square root of Ihkl what you get is not F hkl directly you get the Fhkl modulus. Because, of
the fact that there is a face information that is contained in this one we will have only the
modulus coming in the phase does not come up when you take this square root of the
intensity. And therefore, this expression is can be considered like this since cosine of
minus theta is equal to cos theta and sine of minus theta is minus of sine theta, if you
substitute you get Fhkl star as equal to sum over 1 to n, fn cosine 2π ( - h xn - k yn -l zn).
That means we are taken -xn -yn -zn as the possible coordinates of the atom instead of xn
yn zn so; that means, to say we are now considering the centro symmetry equivalent. So,
for every x y z there is a -x -y -z in a centro symmetric system. So, here we are bringing
that concept into the positioning of the F* hkl as fn cosine 2π ( - h xn - k yn -l zn) plus i
times sum over 1 to n of fn sine 2π ( - h xn - k yn -l zn). So, what does this represent this
represent F-h-k-l. So, F*hkl is equal to F-h-k-l and you see that Fhkl can be shown equal to F*-h-
k-l . So, the complex conjugates are related to each other by this mathematical fashion.
So, F*hkl is F-h-k-l and Fhkl is F*-h-k-l. So, we can rewrite this expression for Fhkl and then you
will see this equivalence coming up therefore, because of this feature of the complex
conjugate nature of Fhkl you will get Ihkl is equal to Fhkl times F*hkl and that is equal to F-
h-k-l because Fhkl is equal to F*-h-k-l from this expression and from the other expression you
634
write F-h-k-l and what does this represent if I hkl is Fhkl F*hkl then F*-h-k-l times F-h-k-l this
should represent I-h-k-l; that means, Ihkl is equal to I-h-k-l.
So, therefore, what we see is that the intensity that you measure is always has a centro
symmetric component. So, it is this is where you know we calculated the possible
number of reflections by making use of the formula which told us the volume of the
limiting sphere and the volume of the reciprocal lattice. And using that we calculated the
number of reflections that are possible and that turned out to be for a given wave length
lambda because it depends on 1/λ3. And we found out that in case of molybdenum
radiation it is 142000 something.
But you see that half of it is redundant because whatever you have measured as I hkl the
other half of I-h-k-l is identical. So, actually measured reflection twice and as we will see
later the symmetry associated with the system will also make some other equivalence.
And because of the fact that there are some other equivalences with high intensities the
number of reflections you collect start to reduce. The I hkl equals I-h-k-l is a universal
relationship. And it is always valid unless under special circumstances.
So, for all practical purposes this is a valid law and therefore, we end up with 11 Laue
classes. So, the 11 Laue classes can be followed into the 7 crystal systems and
corresponding point groups. So, the triclinic has 1 and -1 the Laue group will have only
-1 . So, the 2 systems that are associated to point groups will be essentially 1 point group
635
as far as diffraction is concerned. Similarly in the case of monoclinic system 2/m is the
only possibility, mmm is the only possibility for orthorhombic and so on. So 1, 2, 3, 4,
5, 6, 7, 8, 9, 10, 11, these are the 11 Laue classes.
So, for a diffraction experiment therefore, there are only eleven point groups symmetry
and, so how do we now find out the realistic point groups. The determination of the
realistic point groups will come in later, but in the diffraction experiment you will get
only the Laue symmetry. Now just in the passing I will also tell you what happens in
case the Friedel’s law is not valid. Under what circumstances Friedel’s law validity goes
off.
One of the things which we will notice in a practical experiment is that when we actually
measure the intensity of h k l and measure the intensity of -h -k -l bar, using a detector
the counts which we get on the detector will not be exactly equal. They will be nearly
equal if Ihkl is very strong, I-h-k-l will also be very strong, but we do not expect the
numbers to be; let say 10000 and 10000.
So, there will be a difference between these two, but this difference will be minimal the
difference is due to the fact that we have a crystal. So, X-rays are coming from the from
here or rarer medium and they are going to the crystal which is a denser medium and
then they are scattering. So, even though we treat this as essentially a reflection they are
going through a denser to a rarer to a denser to a rarer medium, before they are measured
and therefore, there will be differences depending upon where exactly in the crystal h k l
is located and where exactly in the crystal your -h -k -l is located with respect to the
defined origin and therefore, the intensity is will be slightly different from each other,
but Friedel’s law will still be validated.
So, when you put this data on absolute scale you will see that they are nearly equal, but
when you actually do this exercise with a compound where you have a very heavy
element. And it is going into a non centro symmetric system. It is possible that the
system which now is generated may show I hkl as not equal to I-h-k-l. So, what was my
example 10000, it may become 4600 there is a huge difference between these two it
cannot be mainly due to the absorption or the size of the crystal. It is due to some other
affect that some other affect is known as anomalous dispersion and this anomalous
dispersion is. In fact, a very useful tool this was discovered by a person called Bijvoet.
636
So, it goes according to his name let me write it somewhere, very inserting spelling
pronounced as by-foot.
So, you can say Bijvoet if you want that is what most of us spell it, but it is pronounced
by foot. So, they this gentlemen found out that the differences is due to something very
interesting and that is the phenomena of anomalous dispersion. So, what happens in
anomalous dispersion, I do not think this is within the purview of our course. So, I will
just describe verbally for a couple of minutes what this process is all about. The process
is that, when X-rays fall on the given element by the intensity which is falling on the
element it is possible that the energy that is associated with the X-rays in coming X_rays
is sufficient enough to knock out the k shell electrons.
So, there 2 electrons in the k shell for any element, so the 2 electrons go away and the
immediate action will be that to keep the stability of the atom the immediate action is
that there is a transition from the next which is the l shell to the k shell. So, the transition
from the l shell to the k shell is accompanied by emission of radiation and that radiation
is the so called characteristic radiation which will be a very sharp x radiation which is
what we use in our diffraction condition, which is what we use for input for copper
radiation or molybdenum radiation and so on, we pick up that characteristic radiation we
have discussed this in an earlier classes well.
So, what happens therefore, is that this particular wavelength has a certain ofcourse,
when we say wavelength it has a frequency associated with it. Now if there is an atom
which as a frequency or that difference between the k and l shell is about the same as that
of the wavelength which is coming in. Then it is does not know whether to stay there or
go out because the frequency matches and because the frequency matches these electrons
now they do not know what to do they reach relativistic speeds. As a result whatever
scatters from the atom becomes anomalous and this anomalous behavior is the one which
get causes the change in the observation of I-h-k-l being different from Ihkl.
People have estimated the possible effects and we will be discussing it at later time, if
time permits because this is the out of the purview of this particular course, but what is
important is that by observing these differences we can actually identify the absolute
configuration of a given molecule in the crystal. So, when we discuss this structure
determination protocol and the absolute configuration are that particular time we will
637
spend a little extra time to see how these Bijvoet difference as they are called the
difference between Ihkl and I-h-k-l refer to as Bijvoet differences.
The presence of the Bijvoet differences can be utilized to determine the absolute
structure of an organic molecule. In fact, it can also be used to determine this structure of
large proteins. Again due to this anomalous dispersion effect because what the net effect
will be is you can guess is the fact that your scattering factor F versus (sine theta)/λ does
not know how to behave, because they only when the atoms go out of the k shell and
generate that wavelength you will have this F versus (sine theta)/λ behavior associated
with a atom, but now the atom is also vibrating with the same frequency.
So, the problem is now it will therefore, completely absorb the radiation which is coming
in and let out something else and that something else is what we have measured as the
intensity and therefore, that information can be used so the your scattering factor get
modified. This scattering factor is which is real. so for will also become imaginary.
So, we will have a component of f. So, what was f which was a real quantity? Now f
becomes fo + Δf ' (a real component on the on the organ diagram) + i Δf ''.
We will have a slide illustrating this detail and how to do the absolute configurations
from anomalous dispersion when we discuss the structure determination aspect. So, at
this moment we end up with eleven Laue classes. So, in a nut shell we have a data we
have a crystal we have sent in the X-ray beam and now we want to collect the data and
when we collect the data we look at the intensities and these intensities.
Now can be analyzed with respect to systematic absences and the systematic absences
will tell us this space group the diffraction apart from that is also now essentially obeying
the Friedel’s law in most of the cases in general. And because of the fact that it obeys the
Friedel’s law, the number of point group symmetry is will reduce because we will get
only the centro symmetric information in an X-ray diffraction pattern. So, we will get
therefore, the eleven Laue classes.
So, the eleven Laue classes are essentially shown here and the scattering therefore, is
essentially this. So, the diffraction pattern of your h k l reflections which you get will be
always with respect to the center of symmetry. In addition the fact that we have higher
symmetry elements in monoclinic orthorhombic trigonal tetragonal as we go higher and
638
higher in symmetry for example, because of the fact that b axis is unique and we have a 2
by m symmetry in the data the h k l reflections will be equal to -h -k -l and also if you
have a h -k l reflection it will also be equal to h k l.
Therefore, there will be other equivalences which will develop and therefore, the
symmetry that is contained inside the data set will also sort of tell us which crystal
system the data will correspond to and that is something which we will see in a next slide
in the coming class. So, essentially therefore, we have covered the various aspects of
space groups, the reason why I am repeating it again and again is because this is the
crucial step and we should not miss any of these points.
And therefore, let me go through it once suppose you consider an a glide we have these
systematic absences when l is equal to 0.
639
We will end up with this list where we have the conditions to be satisfied for a glide
plane parallel to 1 0 0. Glide plane parallel to 0 1 0 these are the conditions to be
followed glide plane parallel to 1 1 0 these are these systematic presences.
Similarly, in case we have the screw axis symmetry we see these conditions coming up.
640
And then we have analyzed this with respect to individual space groups.
The conditions which come up as in general reflections take the precedence over the
conditions which come from special reflections in this particular example of C2 there is
no such worry.
641
But when we go to something like P21/c you see as though this is a glide plane additional
to the existing glide plane. It is not that way to be interpreted. In fact, this is the h k l. So,
it looks as though it is a centered lattice. So, this gives you the information that it is the a
centered lattice and that is because if you look at now the positions of the atoms in the
unit cell the positions of the atoms in the unit cell mimic a centered lattice they are still
in P21/c.
642
So, to determine the space group uniquely one as to refer to the general systematic
absences, and suppose you have recorded a data and then only these absences come in,
this is a point which we should now discuss. Suppose I record a data and that data gives
me only this condition what is your conclusion? Your conclusion is very complicated
now.
Because your crystal system is P21/c, you determine the a, b, c, alpha, beta, gamma by
the distribution of the d h k l values, you found out the a b c alpha beta gamma we are
going to find out in the next set of classes exactly the methodology which is used
because I thought that instead of describing it at this particular position. It is better we
describe it with respect to the experiment you perform because, now the logistics will be
from here on to go to the actual experiment which we perform.
So, we will take the crystal we will take the sample we will grow the crystal examine the
crystal put it on the machine and then the machine collects the data. So, what I thought
would be that with respect to the machine we will now see how things move, because
that is what we are practically going to use in the rest of your crystallographic adventure.
You are not going to use any of these things which we have learnt, but whatever we have
learnt is going to be extraordinarily useful when we want to do the experiments.
Otherwise we will do the experiment blindly and it can always lead to the wrong
direction.
643
So, as I mentioned earlier in one of the classes that jack dentist says that chemist makes
the compound and puts it in a dark room he does not know where is what. And the
crystallographer puts comes into the room and switches the light on and then you will see
where is what and that is thats how it is and therefore, what you have to therefore do is to
see that we get an idea of the way in which the systematic absences come up.
So, coming back to the question which I have asked which I have post I get a indication
that this system is P21/c, but I get systematic absences corresponding to a centered
lattice. So, what do I conclude? Do I conclude that my P21/c is wrong and say this is
now a centered monoclinic system which is quite possible to have what is your
conclusion? The conclusion will be that we had to look at always the general reflections.
But then the general reflections are also not shown only these are shown. So, what is the
conclusion? The conclusion is that if we have these special conditions that are coming
up. The special conditions that are coming up is because of the atoms being in very very
special positions. It is not coming up because the unit cell is this.
This is where the confusion comes, see foe example if this is a compound which is
crystallizing with all these atoms in these positions. Let say any of these positions. Let us
say the molecules are associated with 000, 01/20 now what is very interesting is only if
and only if the atom is sitting at these positions you will have this if it is an atom. If it is
a molecule only the molecules center symmetry will sit there and the center symmetry of
the molecule need not have any atom. Okay?
The molecule there may be one atom in the in the molecule which may sit for example, it
could be a organometic compound the metal atom may sit at this position, but the fact
that the other atoms are sitting in other general positions will not give you only these will
come only if the atoms are in these positions, but we are dealing with molecules we are
leading with compounds we are dealing with complexes.
So, in such situations we will straight away get these. So, only if there is an atom which
crystallizes, in the space group P21/c with which is sitting at 0 0 0, you will get this
systematic absence. So, if there is if there is a organic metallic compound for which you
have collected the data and you try to solve the structure by P21/c you will have
difficulties. You will have difficulties because the phase which you are going to
determine is dominated by the heavy atom position and therefore, then you will have
644
some difficulties to overcome and that is something which is very special to very few
compounds. So, we will not discuss that’s beyond the purview of this particular course.
However you should remember that if the conditions are such that we do get atoms that
are present in these positions, you will also have these indicated in the general reflections
So, the same logic we extended to 2 12121 and in particular Pbca where we get an absence
equivalent of a F centered lattice and therefore, we see that we have the following
information we have this space group information, we have the Friedel’s law information
we have the intensity data that is collected and we know that the diffraction is essentially
belonging to eleven Laue classes. So, what we do next? What we do next is we have now
the entire grammar book written up. We have all the necessary information which is
required to do a diffraction experiment.
So, it is at this stage we can go ahead and do the diffraction experiment which we will
do. So, the next step would be to do the diffraction experiment, grow the crystal to start
with check the crystal for diffraction quality, put it on the machine and then get the
diffraction data and one once we have the diffraction data collected using a counter of
different kinds we can use CCD various kinds of machines to find out the intensity that
comes in the form of a reciprocal lattice point reflection and that particular intensity is
and the position of hkl will allow us to determine the geometry, and also the structure of
the given material.
645
So, that will be the next step. So, we now have reached the stage which is the milestone
of understanding the symmetry and structure in the solid state the next one is therefore,
where we apply? Obviously, we apply to determine the structures. So, to determine the
structures we now have a situation where we have to now get to the compounds. So, get
to the chemistry of it, get to the actual experiment the idea of this course is not to
describe the experiment in detail even though we need the experimental results to gain
go and understand the symmetry and structure in the solid state. So, we will get some
basics of the experiments, we will also find out that the major issue as we have always
been discussing is the so-called phase problem the experiment is giving us only the
magnitude of the structure factor and we need the structure factor to do the Fourier
transform to get the electron density so; that means, we need the phase information.
We also know very in a passing discussing that if we use intensity of h k l the Fourier
transform of that will give us what is known as a Patterson function which essentially
gives us a vector map of the distance between each atom to the other. So, we get a inter
atomic vector display on the three dimensional projection and that will have some special
properties obviously, which we have to examine. So, the logic therefore, is now we will
go find out how the collect the data and so on.
Accumulate the data analyze the data put the data on an absolute scale because whether
you collect the data in Bangalore or Timbuctoo or somewhere in synchrotron radiation
the crystal or the material is still the same and therefore, you should get the structure of
the material you do not care what radiation you are using what diffractometer you are
using what kind of scaling is there in that and so on.
What you want is to put the whole thing on a common scale which is the absolute scale.
The absolute scale come from the fact that we can you see the scattering is done by the
electron density right, inside the crystal. So, it does not matter the electron density is
what matters is the total electron density, the total electron density does all these
scattering experiment.
So, based on that idea the total number of electrons matter. So, if you now consider the
total number of electrons in the crystal system that is the one way with respect to which
we do this scale and that is something which we will consider in the next class.
646
Lecture – 48
Experimental Aspects of Data collection
So, at this time we are in a position to look at the Experimental Aspects.
So, this is this is now becoming a little more critical from the point of view of what we
can do from knowing all the rules of symmetry and details of how the molecules
arranged themselves in the given space group and so on. Obviously, the job would be to
see where the atoms are and determine the positions of the atoms and eventually find out
how the atoms are connected to each other, look at their properties and eventually relate
the property of the crystalline material to the external physical property it exhibits.
So, there are crystals which exhibit different kinds of properties some of them may be
fast ion conductors some of them may be pharmaceutically important compounds some
of them may be ferroelectric materials and so on. So, basically now is we are we get into
the stage where we are now going to look at the experimental aspects in more detail.
What are the major requirements which comes across is to grow a crystal and we have to
grow a good quality crystal.
647
Now, growing a good quality of the crystal is in fact, can be considered a rate limiting
step because very often some of the molecules have the property of not giving us good
quality crystals. By good quality crystals I mean the diffraction extending to as much
extent as possible on the limiting sphere reciprocal lattice space. So, if you get as many
reflections then the issue of not having the required resolution limits becomes an issue.
So, growing a good quality crystal therefore is what we will concentrate on. Of course, if
a material does not grow into a good quality crystal, but remains polycrystalline then the
poly crystalline samples are also acceptable because one can use the so called technique
of powder diffraction. We will be spending a little bit of time on powder diffraction as
well.
So, apart from that what is the experimental setup we look for? We can look for
laboratory sources where we can have a tube x ray or synchrotron radiation which is
available in several centers around the world including the one in Indore in India then of
course, we need a diffractometer which is with a detector of a high quality. So,
diffractometer is the one in which we mount the crystal on a goniometer and things like
that and then of course, we also need non ambient attachments like low and high
temperature attachments, high pressure attachments and things of that kind so that we
can study the variation in the property with respect to these non ambient conditions.
So, particularly some of the compounds become highly conducting at higher

temperatures and some of the compounds at low temperatures shows specific properties
which are different from the room temperature property basically phase transition studies
associated with these materials will be of consequence. In context with the
pharmaceutical industry, the requirement that the structure is known fully and it is
behavior is known in non ambient conditions becomes an issue because it is it depends
upon how the how and why the crystals grow and the so called depiction of
polymorphism associated with crystalline materials will also bring in the issue of
dichotomy in pharma industry.
So, considering all these this is the basic experimental setup that is required all these
together and of course, we can either have a good quality single crystal or a
polycrystalline sample. Now what are the methods we use for growing crystals? One is
the solution methods where in principle we take the material and then dissolve that
648
material in a solvent and then eventually control the evaporation of the solvent. This is
the usual approach which is shown here.
So, we have a saturated solution and then the crystals will start appearing depending
upon the thermal reservoir. So, eventually the temperature associated with this we can
control so that the evaporation of the solvent is slow enough so that we get good quality
crystals.
This can be controlled in various ways we can have more than 1 availed more than 1
thermal reservoir and things like that. So, this is basically the solution method. There is
no time to, in this particular course we are not going to spend too much time on crystal
growth mechanisms and so on, that is not the purpose of this particular course. However,
we should know how we grow crystals.
So, we have solution methods when the solvent goes away or straight forward solvent
evaporation we can take the saturated solution heat it up or make a solution which is near
saturation and then just heat it up, evaporate the solvent and hope that the crystals will
form at the edges of the tube. The other approach is of course, using solving diffusion.
Allow the solvents to slowly diffuse out of the material so that crystals will form.
Another approach is the so called sublimation method where we heat the sample, the
sample sublimes, and then you provide a thermal gradient along this tube and this
eventually we send in a coolant here and then this is vacuum chamber. So, after the
sample evaporates, it now condenses on to the tip here which is carrying the coolant and
you will get crystals. So, this is the approach of sublimation because in case the
compound sublimes this is a very good method to prepare good quality crystals because
now we can control the coolant temperature in such a way that the temperature
associated with this can be monitored and the size of the crystal thus can be controlled;
size and shape of the crystal.
The other approach which has essentially developed in very few laboratories and one of
the laboratories happens to be our own where is the use of in situ cryo-crystallography.
Here what is done is the you know there are lots of solvents which are used in
crystallization process and the solvent structure becomes therefore, very important
whether it is having a certain property related to it like the associated dipole moments
649
and so on, if it is a sub polar solvent or a non polar solvent, what is the influence of those
solvents.
So, in order to know about that we have to crystallize the solvents themselves. So, one
can crystallize the solvent by this procedure. What is done is the solution is taken in a
capillary tube and there is a thermal gradient which is created in such a way that at one
time the solvent inside the capillary tube freezes and becomes a solid. Now, by repeated
reorganization of this like what they call is the zone melting we keep melting the solid by
heat it providing a certain amount of heat and again freezing it up. So, as this process is
repeated over and over again a good quality crystal will develop as is shown here. This is
a good quality crystal which eventually develops and the quality of the crystal can be
checked on a diffractometer by checking the diffraction spots which come up come out
of this experiment.
So, these are various methods and several others, but these are the basic methods which
are essentially used. In order to crystallize proteins there is a total new methodology,
there is a total clearly different kinds of approaches. They saw they saw is these the crisp
the material is taken in the form of a inside a solvent and then as drop of the solvent is
made to hang on upside down against the gravity on a glass plate and because of the
gravity eventually the material sort of sinks in and you will get the crystal growth in the
solvent drop method.
There are several other methods. In fact, we can also have the zone melting which is very
popular among the metallurgist, where they can heat and cool, heat and cool the system.
There are several apparatus that are available to grow quality crystals so much that some
of these methods will allow us to grow crystals which are very large in size that brings us
to the size of the crystal we want to use in our experiments. Most of the experiments
which use x-ray diffraction would like the size of the crystals to be about 0.1 mm3.
So, we do not want any bigger crystals because of absorption of the x-rays within the
material and at the same time we do not want too very thin crystals because the quality of
diffraction suffers. So, there is the optimum size estimated for either organic or inorganic
materials is about 0.1 mm3. In case the crystal is a big crystal then we have to apply
absorption corrections; in case the crystal is a very small crystal then hardly diffraction
comes. You see the whole idea is to get the reflections in the entire reciprocal lattice
650
frame; that means, the the all possible reflections which come within the limiting sphere
must be collected the larger, the data that we get the better it is. And, that is where we
decide on the quality of the crystal. The quality of the crystal is not by the looks of the
crystal, but by the extent to which it diffracts.
So, as we discussed earlier, we discussed about the resolution limits we calculated the
value of d from 2d sinθ λ and the value of d now decides what is the resolution the data
of the data we are going to expect by checking out what is the extent to which the
diffraction occurs in sin theta. So, the larger the sine theta up to 1 degrees 1.0 sinθ value
the better the data quality and also the quantity of data, that is both are required in order
to do a good structural determination protocol.
So, as we saw earlier also this slide that it is taken from the book of Glasker and Tubuler
it is a old book and that is how we have now shown here a photographic film in that
place we put now the CCD camera or the CMOS detector. So, what essentially it does is
comparison between a microscope and a x-ray diffraction experiment. In case of a
microscope the object now sends scattered radiation. Here the radiation is following the
Braggs condition because it is a crystal.
And, all these Braggs conditions will satisfy, we get the reciprocal lattice image. This
reciprocal lattice image the intensity can be measured and the measured intensity now
converted to the modulus of f of h k l, the structure factor. The phase problem is solved
651
either by the crystallographer or by the computer and then we do a Fourier synthesis. The
Fourier synthesis we will reconstruct the image of the object and in terms of the electron
density distribution.
So, we do get an electron density map, as we call it and this electron density map not
only identifies the positions of the atoms. In terms of its strength it also identifies the
nature of the atom whether it is a carbon or a bromine or a nitrogen or whatever. So, that
identity of the molecule can be obtained from what we call as the electron density map.
So, this is the exercise we will have to do in the experiment and normally therefore, this
is the protocol, you grow single crystals, pick the best of the single crystals. We will see
in a minute how we pick them up and how we mount them on the on this machine which
is a 4-circle diffractometer, that is a in-house diffractometer which we have and this
particular diffractometer we will we have several parts. The moving part is the this part
is the goniometer. The goniometer is mounted along this what we call as the omega axis
and on that omega axis the crystal will can rotate and therefore, we now identify the
position of the reflection in terms of 4-circles.
The circles are the rotation about crystal itself which is the phi rotation, the rotation of
the of the 2 theta arm which is the this is the 2 theta arm the rotation of the 2 theta arm
that is theta. So, we have phi, theta and then this is the angle which the goniometer forces
itself to go, that is the kappa angle and then the base plate rotation is the omega. So, we
652
have therefore, phi, omega, kappa and 2 theta. These are the four angles and that is why
it is called a 4-circle automatic diffractometer.
This gives us the reciprocal lattice image a good crystal will give very nice spots as is
shown here separated from each other. So, we now next have to do what is known as the
indexing of this pattern and when once we do the indexing we will get identity of each
one these h k l and we measure the intensity of each of spot. So, that is our data. So,
when once we have this data we take it through the phase problem analysis and then the
Fourier synthesis, we will end up with the electron density distribution and this
represents therefore, the crystal structure. So, this view graph essentially tells you the
basic experimental methodology one has to adapt for determining the structures.
Now, the crystals can be of different shapes and sizes here are some of the crystals which
are good crystals in appearance are shown here. These crystals are not so good in
appearance, but they are also shown here just to tell you the variety of crystallinity we
can impart on these materials. What is more important is not the appearance of good
crystals or a bad crystal; what is important is this part which is now the polarizing
microscope view.
So, if we put the crystal under a polarizing microscope and then move the polarizer, the
Nicol prisms in such a way that we will have an extinction of the light which is coming
from this crystal. So, the crystal is in this orientation. Different polarized light is now
653
sent in and as the polarization polarizer is rotated we will have extinction. So, if the
complete darkness comes up and again when we rotate another 180 degrees of the
polarizer, we will get the crystal picture like this.
So, this is very important before we decide to take the crystal on to the machine. So, if
this is not happening and this remains same as let us say same transparency as you seen
here with respect to light then it means that the crystal is not a very good crystal. So, the
crystal quality is tested under the polarizing microscope. It is not a guaranteed test, but
basically it tells you the possibility of using that as a quality crystal onto the
diffractometer.
Now, on the diffractometer, on the goniometer we can have different ways in which we
can mount the crystal. One of the ways in which normally it is mounted is to take a glass
fiber and that the tip of the glass fiber use the epoxy of some kind and then mount your
crystal. This is a needle crystal which is mounted up here. If you are doing non-ambient
crystallography and particularly variable temperature crystallography and so on, then the
crystal has to be protected because we are going to change the temperature and pressure.
So, for such a purpose we can enclose this crystal inside a capillary called the
Lindemann capillary. This Lindemann capillary has the property that it will not absorb
our diffract x-rays and as a consequence they will just be the containers which hold the
crystal. So, whatever comes out in the form of diffraction is that from the crystal. So, we
get different signs of these Lindemann capillaries which can be mounted up here. And,
so, that is these are the two ways in which you collect the data.
Of late the data collection particularly if the crystal is small and also if the crystal is
protein crystal or whatever we have these what are called the loops. So, this is a loop
mounting with the help of a oil. In this particular case we use different kinds of materials
when you are doing let us say a data collection at 100 K you can use a oil which is now
holding the crystal at this point and then at 100 K it will freeze. And, this is essentially it
is oil and the oil that we use is very special and that particular oil, paratone oil. So, it is
the paratone oil which we use in order to mount the crystal.
So, we make use of the paratone oil and we mount the crystal on this loop. So, this is
now the most common method which is used because there is no interference of either
glass or any other material with the system and also at the same time the crystal is now
654
completely bathed in the incoming beam which is a requirement for a quality diffraction.
So, we will not go into the unnecessarily further details of experimentation because that
is not the purpose.
The purpose of this is to see what happens when we do the diffraction. So, we do the
diffraction in this one of the CCD images which shows the reciprocal lattice images. So,
these are the reciprocal lattice points which are seen here dark and white. We know why
some of them are dark and some of them are not so very dark as are bright as we have
seen from our positioning of the atom with respect to the planes.
So, having got this is one of the several hundreds of images which we can collect for a
given crystal depending upon the orientation of the 2 theta, phi, psi and kappa. So, these
4 values, the 4 circles now decide the geometry of the reciprocal lattice which is now 3-
dimensional in space. So, we collect the reciprocal lattice image which is 3-dimension in
space and make use of these determine their intensities from the experiment by using a
measurement technique which is now converting these photons into electric fields in the
CCD detector and we measure the intensity directly.
The there are other detectors which one can use the detectors could be image plates, the
detector could be CMOS detector which is essentially used in space technology and CCD
which is very commonly used nowadays. Image plate is very commonly used in case of
macro molecular structures because it is possible to wipe the image on the image plate
655
before we re-expose. So, these are some of the advantages in an image plate where the
reflections are very closely spaced. Those materials which are not so very closely placed
can be collected both on CCD as well as CMOS. The time required and the amount of
time we spend to collect each data point is pre decided by us and that depends upon the
quality of the crystal and the extent to which we are looking for reflections in the
reciprocal plane.
So, if we want to go to the end of the reciprocal lattice dimension which is the limiting
sphere we would like to have a good time of exposure and these exposure time also
depends as we know as we have already seen on the f versus sine theta by lambda curve.
So, at lower angles we will have higher intensities at smaller angle the larger angles we
will have lower intensities. So, the exposure time accordingly we can change we can go
from smaller 10 second data collection to a almost a minute data collection if you are
collecting at very high temperature and thus very high angular range.
The other thing is as we increase the temperature on the crystal, the crystal the atoms in
the crystal vibrate a little more than necessary and those vibrations can cause the
intensities to go down. So, on the other hand if we cool the crystal the spots become
sharper so, because the vibrations of the atoms are minimized. So, all these are
experimental aspects which we have to optimize before we collect a good data set.
656
Having got this kind of a thing we start now with what we call as indexing this is the
backbone of the entire structure determination. So, this is a process which I have given a
flow chart from the textbook of William Clegg; edited by William Clegg, several other
authors have written this. William Clegg is the editor of this particular book which I will
of course, in the end of this course I will give all the list of all the books with their
publishers etcetera with very special acknowledgments to each one of these authors who
have written the books because I have taken their pictures, their equations and their
flowcharts as you can see here. This is something which we must be very happy about
because these are already available and we can use it for our presentation.
So, the indexing process will start by collecting the data for several small parts of the
reciprocal space. You saw in the previous picture a position a portion of the reciprocal
space and in this reciprocal space you see there are a few reciprocal lattice points which
intersect the Ewald sphere. And, those which intersect the Ewald sphere come up with
intensities like this that is how you see the spread in the case where there is cell
dimensions are very large you see the spread becoming closer closer. So, larger the cell
dimensions the spacing between these spots will be much much closer. So, it is a
reciprocal relationship. So, if the cell dimension is 100 angstroms like in a protein, then
these parts will be very close to each other; whereas, in this case it looks like a data from
a very not so very strong reflecting material and at the same time a smaller unit cell
damage.
So, you collect the data from several small parts of the reciprocal lattice. So, do not keep
the detector in one particular place you change the angles; the other three angles the phi,
psi and omega. So, if you change the phi, psi, omega and then keep changing also the 2
theta in go in steps of 2 theta keep the let us say collect data 10 degrees, 12 degrees, 17
degrees, 18 you can program it in when you do the data collection protocol. So, collect
the data from several small parts of the reciprocal lattice. So, we have a representative
number of reflections just as we saw in the previous viewgraph. We have a
representative number of reflections coming from the same material in different parts of
2 theta space.
So, having got that you generate a list of vectors between the reflections collected,
vectors between the reflections collected and then in such a way that you find three
shortest non-coplanar vectors in the list. They should not be in the same plane. So, it
657
should not be coplanar and then find among the set of spots which you have got. Find the
smallest or the shortest non-coplanar vectors from the list. It is believed that these three
vectors will give us the trial unit cell it may not be the correct unit cell, but you just
select the three shortest non-coplanar vectors. And, these three short non-planar it is can
be now used there must be a primitive subset of the cell if not the correct unit cell this is
something which we have to examine.
So, to find three vectors which are non-coplanar, identify these three vectors as the trial
units. When once you find that out you say well this could be the possible unit cell some
a, b, c, alpha, beta, gamma between them. So, these three vectors are taken as the trial
unit cell. If this unit cell is a correct unit cell using the trial unit cell assign hkl indices to
all other vectors. So, if you take this three as the trial unit cell, then you know the
endpoints with these 3 vectors are defining now a, b and c. So, the endpoints with these
three vectors are a, b and c.
So, we now have the unit cell dimension and the interactional angles between them
because we are identifying the vectors. So, using that you now assign hkl values to the
rest of them. So, suppose this is 1 unit, 1 unit, 1 unit along a, b and c and something else
which is coming farther away could be a multiple of this something which is coming
below this could be a fraction of that, but the chance of something coming shorter than
this is less because we have taken the three shortest non-coplanar vectors in from the list.
So, these are the three star shortest non coplanar vectors. So, it could be a correct cell, it
could be a not a correct cell that is that is checked by this questionnaire here or all hkl
indices integral value. So, then once we have this thus it come the rest of the spots which
we have. We may have a 100 spots. Let us say the 100 spots which we have are these hkl
indices also integral values. So, we check on these hkl indices integral values and if these
integral values all appear to be good in the sense you may not get exactly 1, 4, 5; you
may get 1.2 at the most and on the other side 0.8, then you can round it off to 1
effectively or you may get 4, 4, 7 and 8, 4, 3.9 and 4.1 and things like that.
So, you give a tolerance limit within which all the hkl integers are integral. If they are
really falling in that zone then you say that you are determine the unit cell dimension. So,
use this unit cell dimensions and do what is known as a least squares fit for this to
658
identify the unit cell a, b, c. So, unit cell obtained and you stop it here. So, this is the
indexing process.
So, in the indexing process the major requirement is that you have to now take every
point in the reciprocal lattice for which you have measured the intensities. Take those
spots and then make sure that these spots are generated by three non-coplanar vectors
which are the shortest set of vectors in this representative handled or reflections. This can
be easily done programmed and the machine will do this work and so, when once that is
done if they are all integers, it is fine. If they are not integers then what you do is if they
if they come out as simple fractions let us say one fourth, one half then we can multiply
by the 2 and 4 and then eventually decide this would be the cell.
So, if many indices are simple fractions then divide appropriate a, b, c by that fraction.
So, you will get the new trial cell. When once you have the new trial cell go back here
and then again using the new trial cell assign hkl and now they will become all integral.
If this process happens there is no problem. But, if this also fails if you do not get a hold
of it, then you can do two things – one is to collect more data and use the subset of
present data or try another indexing method then repeat the cycle. There are different
ways in which one can index the methodology there are programs which are available
different kinds of programs approaches it in different ways. And, these therefore, can be
now put in such a way that we go back in and then again generate the list of vectors.
So, this flowchart essentially tells us how to get to the cell dimensions. So, the way how
to get to the cell dimensions and how to get the hkl values assigned to individual
reciprocal lattice points. The appearance of the individual reciprocal lattice points thus is
very important. So, when you have a good crystal the appearance of the individual
reflections should be as spherical as possible. It may not be always of spherical, but as
spherical as possible that is because that we will tell us the that the crystal is rightly bet
in the beam of x-rays which are coming in because the normally the x-ray beam which
comes out is a is a circular path. It has a circular diameter, it has got a circular what
should we say circular shape which now falls on the crystal and therefore, the reciprocal
lattice points should mimic the shape of the incoming radiation.
The incoming radiation is collimated normally. So, the x-rays which are coming from
this various sources they pass through a collimator, the collimator will adjust it to the
659
size accordingly. So, we can change the size of the collimator. We can have 0.5 mm, 0.3
mm, 0.1 mm depending upon the size of the crystal. If the as the crystal becomes bigger
and bigger you can use bigger collimators, but then very big crystals are no no because
they will absorb all the radiation which comes in. So, there is an optimized value again.
So, normally the ones which are used on the routine diffractometer laboratory setup the
collimators are 0.05 and in some special circumstances 0.03 is also a used diameter for
the collimator.
So, the one once we have this hkl indices identified the initial reflection list is generated
we have the unit cell. So, when once we have the unit cell now you collect a few more
frames of data that is reorient in different 2 theta directions and collect some more
reflections. Use these cell dimensions, this is an additional check which we can do which
is not indicated in this flow chart use these a, b, c alpha beta values and then work out
their corresponding hkl’s. So, go backwards and determine the hkl’s and those should be
integral values. This will now doubly verify the nature of the unit cell dimension which
you have determined.
So, the unit cell three vectors non-collinear vectors will also have a, b, c, alpha, beta,
gamma information. So, the moment you have the first three shortest non-coplanar
vectors as identified as a unit cell you know the crystal system because you know a, b, c
you know alpha, beta, gamma, so, you know the crystal system. So, the crystal system is
already identified by just doing a diffraction experiment and selecting the reciprocal
lattice points in such a way that there are three directions in which the shortest vectors
which will generate the rest of the reciprocal lattice points are identified.
And, this process can be repeated over and over again until the final a, b, c, alpha, beta,
gamma values are depicted on the screen. One once that is done any data points you
collect you can go ahead and collect the full data full sphere of data, every lattice point in
principle should be indexable that means, we can give a hkl value based on this a, b, c,
alpha, beta, gamma. So, that way this is a very useful approach by means of which you
identified the cell dimensions.
660
Lecture – 49
So, after looking at this indexing process and identifying the unit cell dimensions, the
next job will be to calculate and improve all the a, b, c, and alpha, beta, gamma and so
on.
661
So, we can do that very comfortably by looking at this particular expression. So for
example, you see that the 7 crystal systems are listed here. The value of 1/d 2hkl as we saw
in the earlier classes also depends upon the nature of the crystal. Suppose it is a cubic
crystal, the 1/d2hkl will be (h2+k2+l2)/a2 because in a cubic system what we have is a=b= c
alpha, beta, gamma is 90 degrees. This is in a cubic system.
So, in a cubic system since we have these values the way in which the d 2hkl comes up, in
this system is 1/d2hkl is (h2+k2+l2)/a2. So, we have done this identification, we have
calculated the hkl values and the h2 k2 l2 square can be calculated and then, we know
where the d values are coming with respect to the reciprocal lattice. So therefore, we can
calculate the cell dimension here. The same way in a tetragonal system, we can calculate
the cell dimension a and c, these are straightforward equations and orthorhombic we can
calculate a b and c.
So, because the 1/d2hkl defines this equal to that; in the case of an orthorhombic system.
So, as you see here trigonal and then rhombohedral setting of the trigonal system, the
monoclinic system where the beta angle becomes important. We have discussed this
earlier on when we discussed the reciprocal lattice and it’s relationship with direct space.
So, we can use these expressions therefore; to kind identify the d values.
So, in principle we have the d values of every spot in the data collected identified with
respect to the cell dimensions a, b, c the three non co planar vectors which we were
662
talking about. And so, what we do in principle, is what the program does in principle is
to see first take those reflections which you have indexed as hkl. Take the three non-co
planar vectors which you have identified and then what it does is, it starts from fixing the
cubic lattice to the system and then see whether hkl remains integrals. If the hkl values
are not integral as we discussed in the previous viewgraph or the previous flowchart,
then it will see you all it is not cubic let me try tetragonal.
So, like that it will try the systems to bring down the symmetry and you see triclinic one
is fairly complicated because alpha, beta, gamma are not equal to 90 degrees. So all these
expressions it will fit in and see which is the most probable fail and then feed it back to
the flow chart we saw in the previous case. So, we can keep on doing the refinement
associated with the value of d. So, since we have a very large number of reflections
identified the 1/d2hkl is very very large in number with respect to the measurements which
we have done and so, these values of a, b, c will become extremely accurate.
The accuracy associated with the cell dimensions could be at the third decimal place or
sometimes even the fourth decimal place in a single crystal. In fact, its accuracy becomes
much more when we do powder diffraction, mainly because powder diffraction has a
large number of very small what we call as crystallites each one is a small crystal. Since
we have a random collection of all these crystals, there is an averaging out effect on
determining the cell dimensions. So, we can even go to the fifth decimal place.
So normally, for example in case of a cubic system if the a value is 6.123 let us say, we
can also find out from this over fitting of all the d values we can find out how far is the
deviation of this determination. So, it can be like 6.123 which are standard deviation of
2; that means, the standard deviation is 0.002 so; that means, within the error bar of
0.002 the determined cell dimension is valid.
So we can therefore, get very accurate cell dimensions and powder diffraction gives
more accurate cell dimensions than single crystal data. This is something which is to be
noted.
663
So, having got this up to this particular point we now go to the description of how we
determine the structure? How do we use the symmetry information which we already
have learnt so far and how the structure determination protocol per say goes. I will take
you through this whole series of determination of the structure in a few slides rather
rapidly. However, we will be visiting each one of them separately with respect to the
later date the next few lectures.
So, the first step is one once you have the data, we have determined the a, b, c, alpha,
beta, gamma you know the crystal system, you now go and look at the systematic
absences. The systematic absences will tell us the space group information. So we have
the Cell Dimensions and therefore, we also have the information on the Laue Symmetry
and then of course, we now have the intensities collected so we can do the Intensity
Statistics. In fact, the intensity statistics will tell us whether the crystal system belongs to
a centro symmetric system or a son-centro symmetric system. Please make a note of this
at this moment, we will discuss it later on probably in the next class or something.
So, the intensity statistics will be deciding the Centro Symmetric or Non-Centro
symmetric nature of the given material. Systematic absences of course, will determine
the space group, but in case the space group does not have any systematic absences like
P1 and P-1. How do we distinguish between p 1 and p 1 bar? We distinguish between P1
664
and P-1by looking at what we call as the intensity statistics. How the intensities are
distributed with respect to (sine theta)/λ.
So, we have an average intensity over the experiment. So, we calculate I/I average and see
how the (sine theta)/λ varies with respect to I/I average and that will give us a distribution
which can be plotted in the form of a curve and if that particular curves are two types.
One is the exponential type the other is the error function type. The error function type
tells us it is the centro symmetric system if it is an exponential type it tells us it is a non-
centric system. We will see that in more detail when we go for the discussion of scaling
and the intensity statistics which we will evaluate in the next few minutes.
So, the point group of the unit cell determines the symmetry of diffraction pattern. So,
that is, the Laue symmetry that will determine the symmetry of the diffraction pattern.
Then we translational symmetry elements result in systematically absent Bragg spots.
For example, this is centering, screws and glides which we discussed in the previous
class. So, we have all these information that is available. So, once you have collected the
data, once you have taken that particular data onto your system determine the a, b, c,
alpha, beta, gamma by means of which every spot which was identified with respect to
four angles: 2theta, phi, chi and omega is converted to hkl.
So we have therefore, the information about the indices and when once we have the
indices we have the information on the crystal system. We have the information on the
nature of space based on the systematic absences we also have information on the space
group. So, this is the next step which we covered up to the previous class. So, we have
therefore, the uniquely indexed pattern at our disposal from the experiment. We have
determined the space group to the extent reasonably reliable and in most cases systematic
absences as we saw will give us a unique space group information having gotten all that
we now go and go for the so called structure determination.
Here is a some issues which we have to discuss, but if before we go into the detail of it I
will just tell you the protocol that is followed since we are now doing, we are in an
experimental mode, so to say. Because we have talked about how what kind of crystals
we choose, how we mount and how we collect the data on the diffractometer, how do we
look at all those data then determine a, b, c and d of hkl and then of course, systematic
absence is giving us the space group.
665
So, I thought we will finish off with the protocol of a structure determination. Now any
structure determination is done in two steps. The reason is that we have to determine the
phases and this is the so called phase problem in crystallography. So, experiment. we got
the intensities, the square root of the intensities gave us the modulus of F hkl which is
coming from the square root of I hkl. So, if we want to calculate now the 3 dimensional
structure by using the Fourier Synthesis, which we will discuss in the coming class. The
modulus of Fhkl has to be combined with the phase information. To get the F hkl and then
the Fourier transformation takes it from the f of hkl information which is up here to the
rho of x, y, z which is the electron density distribution.
So, we plot now the 3 dimensional x, y, z coordinate system. The ρ value, the electron
density value which is calculated from these expression. You are quite familiar with this
expression Fhkl exponential {2πi (hx + ky + lz)}. It now depends upon n atoms. So, all
every atom will have this formulation divided by V the volume of the unit cell. So, here
xyz of course, are the fractional coordinates it can range between 0 and 1. The idea of
this part of the slide which is shown here is to bring in the importance of the
determination of the phase problem, a solution to the phase problem. How do we get to
the phases because we cannot get them from experiment, we cannot calculate them
directly and obviously, we have to make estimates of the phases.
So, the estimates of the phases is done by several methods. One is the direct space
method which is also referred to as the directs methods and the other is to go through the
vector space which we discussed earlier on. That is, take the intensity and directly do a
Fourier transform and that Fourier transform in turn will give us the information on the
inter atomic vectors that is to say let me draw it a little bit.
Suppose you have an atom x, y, z here, you have an atom at x 2, y2, z2, you have an atom
at x1, y1, z1. Then what you will get in a Patterson map is the vector between these two.
The problem with the Patterson map is the origin will be the starting point for these
vector. It will also give the vector between x2, y2, z2and x1, y1, z1.
So, both these vectors appear with respect to the origin. Both these atoms will also
appear at the origin because the self vectors will also be depicted in the Patterson. So
however, in principle we should be able to find where the atoms are and that is the so
called Patterson method. We will be discussing detailed analysis of the Patterson
666
approach particularly with respect to this symmetry and structure. You see the whole
idea of this course is symmetry in structure. So, we will bring in the importance of
symmetry in the vector space which will allow us to determine the structure.
So, we will discuss Patterson synthesis in that context. Of course, the direct phase
structural determinate or structure determination using the direct phase solution problem
is something which is based on probabilities and we will develop it very little but not in
full detail, but give us the basics of the physical properties that are involved in direct
methods approach. So, the physics of it we will try and understand. Again, in terms of
the planes and in terms of the electron density contained in the planes and so on.
So, that way the phase problem solution we will not spend too much time; however, we
will find the methods to solve the phase problem. Once we have the phase we will try to
get an understanding of what is this Fourier synthesis is all about and how do we get the
Fourier transform done from Fhkl to ρxyz. We already have seen that if we have the
information on the ρxyz which means the we know the positions of the atoms, we can go
and calculate Fhkl.
But what we get from the experiment now is the modulus of F hkl phase solved we have
the Fhkl values, this is the expression for the Fourier transformation. So, we go from F hkl
to ρxyz. From Fhkl to ρxyz using the Fourier transform.
667
Let us go further and see how this all happens. This is a picture which is a typical
electron density map which comes out after all these process is done; that means, when
you have done the Fourier transformation you get a picture like this. This is a sort of very
nicely featured picture you may not always get this kind of a picture, but for discussions
sake, I have taken a picture like this.
Now, this is the electron density map. So, what we have done is, we have plotted the
electron density map some of the details are given here. This is an output from the
program. So, we will see that we will get electron densities drawn in this particular
fashion. Now what it is what is it that is represented in this is the position of position x y
z of each one of these atoms up here, and then there is a contour which is drawn at a
certain specific intervals and those contours which are drawn at specific intervals have
some meaning; that means, they carry that amount of electron density.
So; that means, when you calculate the ρxyz you have a map you calculate the value of 0,
0 0.1, 0.1, 0.2, 0.2, 0.2, 0.3 and so on, covering the entire space. We normally cover the
symmetric unit of the unit cell and in this particular diagram of course, the isometric unit
is combined with the rest of the other symmetric unit in the unit cell so that we are
actually getting the full picture. Now the values of the electron density as we calculate
are mapped here.
So, we actually put the values of the electron density up there and draw a contour which
goes, let us say these values come out to be 10, 20, 30, 40, 50, 60, we draw as 60 value
here, 50 value there, a 40 value there and so on. So, effectively these electron density
which is drawn here will represent an atom. So, by counting the number of contours, we
can decide what atom is this one. In this case, we have lots of carbon atoms as you see
they are all looking fairly identical. This is a very well refined electron density map.
But, this is how the positions of the atoms will appear and one once they appear we can
draw lines showing the bonds between these atoms and so, we get a for example, in this
case a six membered ring which is representing a planar six membered ring.
So, it’s a phenyl ring and these are the substitutions associated with different kinds of
atoms. So, by counting the number of contours because we have already scaled it up, this
data is already scaled up data and therefore, the electron density will give you the atom
668
positions as they are and in terms of their strength, we can identify where the atoms are
sitting.
So, this is a typical electron density map which can be computed and one once we have
this map, we can draw this diagram.
This diagram is drawn from the structure of this particular compound whatever that is,
we have a ring system here, we have the substitutions, this corresponds to the diagram
here and these are the values which will appear and we see that based on the strength of
the electron density at these positions, we get the positions of the oxygen atoms. The rest
of the atoms are carbon atoms.
669
This is the beginning of the structure determination.
That is why I said the structure is determined in two steps. The first step covers basically
the solving the phases. So, the phases which we have determined or approximate and
therefore, the structure is also going to be approximate. In other words if we go back
here, the distance between these two may not correspond to the exact distance between
carbon and carbon. Some of them may be elongated, some of them may be short, some
of them may be long because there are errors that are associated with the phase.
So, the structure therefore, is first the initial estimate of the phases. You will get what
this is known as a starting model. So, we get a starting model and that starting model has
to be refined. In order to refine the starting model, we need to have a large amount of
data. The fact that we have a large number of hkl values and less number of parameters
to determine will allow us to do the so called least squares refinement.
I do not know whether it is within the purview of this course, but I will give some basic
details of the least squares refinement. So that we know how the structure which we
determine initially which is called which is giving us the trial structure. We will develop
into a final structure of this kind where the atoms positions are known, the distance
between atoms are accurate, the angle between atoms are accurate and this is a refined
structural model of that same compound.
670
So the way the way in which we have to do is refer to as the structure completion. So,
this is the second part of the structure determination. First part is to get this trial structure
by solving the phase problem and then using the Fourier transform on the F hkl to get the
initial ρxyz. You can call it the initial electron density map. Now after that we have to do
this protocol of structure completion. So, this lists out the protocol of structure
completion. So, extract the 3D coordinates of the atoms as we saw just now and assign
atom types, scattering type C, O etc. Assign additional parameters to model the thermal
motion of the atoms.
Now, here again a little digression because the as we always know that there is these
atoms in a structure or not at rest, they are vibrating about their mean positions. The
reason for vibration is purely thermodynamical and the reason is also accounted by the
fact that these atom may not be associated with that x, y, z which you have determined.
So, there is always a, there is always a area around the x, y, z which is possible for this
particular atom to occupy and this error in this Δx, Δy, Δz could be in all three directions.
And therefore, these thermal motion is expressed in terms of what we call as a thermal
ellipsoid. We will be analyzing this in a short while from now by one once we do the
Wilson plot which is essentially the plot which puts the relative measurements we got
from the experimental measurement onto a final scale value which is equal to the total
number of electrons inside the scattering material we are looking for. And this one once
we have this the thermal motion of the atoms will get additional parameters along with
the x, y, z.
So, if there is a carbon atom it is associated with the coordinates x, y, z. It is also

associated with the thermal portion parameters. Since the value of a, b and c are different
from each other in many crystal systems and also the alpha, beta, gamma, depending
upon the symmetry information the thermal ellipsoids, the ellipsoids that is associated
with the position x, y, z can have different shapes and the values of these major and
minor axis may be different. The orientation of the ellipsoid may be different.
So, all these now come as additional parameters. So, we have we will have six additional
parameters which are referred to as an isotropic thermal parameters. Then we have other
parameters which will come along with the data. It is due to the nature of the crystal and
also due to the nature of the material. The extinction and the twinning is essentially
671
associated with the crystal. When we have twinning associated with the crystal, the
crystal diffraction will be affected by twinning and there is a way to treat it which we are
not going into the detail of it.
The flack parameter which is called the x parameter will allow us to determine the
absolute configuration particularly in case when we have the anomalous scattering
effects. So, every one of the atoms in the structure will be anomalously affected, but if
you have a very heavy atom like bromine or something in an organic molecule, the affect
associated on the variation of Ihkl and I-h-k-l which we discussed in the last class. The
difference between them, the Friedel’s law will be violated. The Bijvoet pairs will be
created and those Bijvoet pairs will allow us to determine this parameter. The parameter
which will now, tell us about the absolute configuration of the system. We will go into
the detail with an example in the later classes.
So essentially, we build up a model. The model is based on F hkl it depends upon the scale
factor the fj value, the thermal parameters and so on, this should be under the summation
j equals 1 to n. S is to replace the summation. This is the Bill Gates effect on my
computer because on my computer this comes as sum here it comes as S. So, I should
have checked it before, but anyway this is summation over j equals 1 to n, f j is the
thermal parameter exponential 2π (hx + ky + lz). Surprisingly, the 2π has been retained
somehow.
The non-linear least squares parameter refinement until convergence, this is the next step
where what we do is you see we have a trial structure. The trial structure will give us a
ρxyz so let me write it down. So, we have now arrived at a trial structure and this trial
structure which we have is based upon solution to the phase problem. So, we solved the
phase problem we got a trial structure, this is the starting model.
Now, the starting model values we can use to calculate rho of x, y z. So, we have any rho
of x, y, z which is the electron density. Now using the electron density you go back and
calculate. This is now the calculated Fhkl. So, the calculated Fhkl will have errors because
we do not have the accurate determination of the phases. So, the F calchkl is now put back
here in this expression, you have to see hkl which is coming from the trial structure is put
here which is, Fhkl calculator and we already have the F hkl observed, which is the
experimental Fhkl.
672
So, Fhkl observed minus the Fhkl calculated whole square. This is a quantity which is
which can be called as ΔF, that is the difference between the actual experimental
observation and the calculated value which has come from a trial structure calculation.
So, this difference therefore, has to be minimized. So, we have to minimize the
summation sigma hkl where w is a weighting scheme we will not discuss it further.
There is some kind of a weight it has to be attached for each one of these reflections
because we do not measure every reflection under the same condition.
So, that will attach a factor called w, we will not go into the detail of it; but sum over hkl
you have k h observed minus F hkl calculated. This value has to be minimized from the
least square sense. So, this is the non-linear least squares refinement protocol we will
discuss it a little more. This is just to finish the protocol we have to follow for one once
we have put a crystal collected the data and so on.
So now, you have an agreement factor. Now this agreement factor is referred to as the R
factor. Now this R factor is equal to the sum over this should be sigma, sum over mod F
obs. I will correct it in the possibly not now, but I will correct it later because I don’t
have the necessary software here. So, sum over Fobs minus Fcalc that is the delta F divided
by sum over Fobs. So this gives us agreement factor or what is called a reliability index.
Now, this reliability index is expressed with respect to 100. So, if the reliability value is
somewhere around 10 or something, you say that your structure is fully determined.
There are other issues which will come up with the least squares refinement which we
will be discussing when we go in detail with the non-linear least squares refinement
protocol. So at this moment, what we have to do is one once we have the solved the what
we call as the phase problem we get a trial structure. The trial structure is now associated
with a thermal motion for each and every atom and other parameters which are
extinction, twinning, flack etcetera are covered along with it, Then we go and develop a
model and that particular model will give us the trial structure with an electron density
ρxyz.
Now, based on this ρxyz we calculate the structure factor again and subtract the structure
factor from the observed structure factor because that will tell us what is the difference
between the two which for which we have to account for and that particular difference
has to be minimized; and that minimization process is using what we call as a non-linear
673
least squares refinement protocol and we will discuss that a little more detail in the
coming classes right.
So, at this stage how do we go about this refinement there are automatic packages. Since
we are going through the protocol of structure determination, I thought I will take you
through this particular slide as well. So, what you do is, you use programs like SHELXL,
Crystals, XTAL. So many programs are available. So, use all these programs and what
you do is you improve on the positional parameters by doing the F obs minus Fcalc and
trying to minimize the value of Fobs minus Fcalc along with the isotropic thermal and later
it could be converted to anisotropic thermal parameters as well.
Now having done this, when once we reach a reliability index of a decent amount and
then that does not change much with changing the F. Fobs minus Fcalc values, you now
introduce the hydrogen atoms. Now the issue of hydrogen atom is very interesting.
Hydrogen atoms do not have a core. Since hydrogen atoms do not have a core, we have
an issue. The issue is whether the sin F will varying a (sinθ)/λ can be existing at all for
the hydrogen atom. Even if it exists, it will be very very small.
So, when we actually determine the structure in the first instance using the phase refined
a protocol and then eventually trying to get to the starting model, the starting model will
cannot give up hydrogen atoms. So, we have to now introduce the hydrogen atoms there
are various ways in which we can introduce the hydrogen atom. The detail of which I
674
will not probably go into in full swing, except to tell that hydrogen atoms can be
introduced because we can fix the hydrogen atom stereo chemically on the associated
atom or we can refine the hydrogen atoms with the x,y,z, U iso or riding on their carrier
atoms and so on.
So, this particular detail I will not tell you right now, but when we actually discuss the
refinement and the interaction of hydrogen atoms, we will bring in these details later.
After all these, we do what is known as an ORTEP presentation. This is the OakRidge
Thermal Ellipsoid Plot. This will tell us how the atoms are oriented with respect to each
other in the crystal structure along with their thermal features which are now the features
that have to be discussed.
Let me show the picture which will be clear. So here is a structure when in which we
have shown the thermal ellipsoid. So, each and every atom now is associated with a
thermal ellipsoid. Hydrogens are shown as circles because they are associated with a
isotropic thermal ellipsoid in this particular structure which you have taken as an
example. So, these therefore, tells us that the oxygen atom location O5 is with respect to
some x, y, z here and then the uncertainty in locating that particular atom because the
atom is vibrating at any given time. When we do the x-ray diffraction experiment, it may
not be at that point, but it may be distributed in any place, any space 3 Dimensional
space within that particular ellipsoid.
675
So, this tells us the limit to which we can rely on the determination of the position of the
atom and therefore, this is the thermal ellipsoid. We can calculate the thermal ellipsoid
for each and every atom and this therefore, now represents the structure of the molecule
as such, as it at that particular temperature. So if you reduce the temperature, the size of
the thermal ellipsoids will reduce. If you increase the temperature the size of the thermal
ellipsoids will increase. So the atomic vibration therefore, goes hand in hand with
temperature. So it is therefore, suggested that all the structures have to be determined at
the temperatures which are lower than the ambient temperature.
676
Lecture – 50
Structure Determination 2
So, we see that this is a representation of the structure drawn by a program, this is a
program called Pluton, there are several programs which can do this all type plot. And
this is the structure of a sucrose. So, the sucrose structure is determined then, then its
positions of the atoms the carbon atoms and oxygen atoms are shown and so on. I was
talking about the hydrogen positions, so in a situation like this there are two ways in
which hydrogen positions can be determined. One is to put all the atoms in their final
positions as we have done the refinement. And, once we have those in the final positions
do F calculation, that is you go back to this calculation and then do a Fobs minus Fcalc.
677
So, all the atoms are in an accurate final accurate positions and so you got a difference of
Fobs minus Fcalc. So, that will be now they representation of whatever is left out which is
reserved for hydrogen atom positions. Or, for that matter if you have an item which is
wrongly put or wrongly introduced in this process, then it will not start showing up when
you do a delta F based Fourier map. So, in the Fourier map now you do not use F obs you
used delta F. So, when you calculate the Fourier map you use delta F and that means,
you will use the whatever is the electronic density that is left out and that is known as a
difference Fourier plot because, we are taking the difference of the Fourier values of the
observed and the calculated.
So therefore, it is a difference Fourier plot. So, when we do this difference Fourier plot
the hydrogen atoms, many of them will show up, as possible small density values were
around 1 electron per angstrom cube. And this positions, therefore can be fixed now we
can identify their hydrogen positions. So, then we see also the geometry that is
associated with those hydrogen positions. In case it is not coming out clearly the
hydrogen positions then one can use the stereo chemical information that is available in
the molecule. In case of sucrose we know whether it is up or down whichever way the
hydrogen positions are going we call them alpha and beta in sucrose nomenclature. So,
those hydrogen positions therefore, can be actually fixed at that particular distance.
678
And angle which is which it is making with respect to the atom for example, this
hydrogen atom which is coming out of this carbon atom. The oxygen atom is going
down. So, the stereo chemistry around this particular carbon is tetrahedral. So, we have
now the 3 directions fixed the 4th direction can be now given to the hydrogen atom. And
at the corresponding distance which is normally a normal O-H distance is about one
angstrom.
So, we take 1 angstrom from there and fix the hydrogen position. So, then we do not
touch the hydrogen position, but we allow the carbon, oxygen and other surrounding
atoms to refine. So, whenever the new positions of these carbon comes up, this hydrogen
automatically is fixed again at 1 angstrom; that means, this is referred to as a riding
hydrogen. So, we call this is as a riding hydrogen because it rides along with these atom.
So, whatever happens to this particular atom hydrogen in any case will not leave this is
like a rider sitting on a horse.
So, as the horse keeps moving he is not going to be displaced, he is stucked to the back
of the horse. So, the hydrogen therefore, is riding on the corresponding atom with which
it is attached. So, this is known as the riding atom hydrogen atom refinement in which
case the distance is not altered. The hydrogen atom position is therefore, now depends
upon how the positions of the other atoms nearby change. The other thing which we can
do of course, is one once we have the difference Fourier hydrogen identified and if you
still have a good data at our disposal then we can try refining the hydrogen positions as
well.
But refine them with respect to the isotropic value. So, that is why you see only circles
here rather than ellipsoids, because this is now associated with what we call as an
isotropic thermal parameter; only one thermal parameter is associated with respect to x y
z and since there is only one thermal parameter you will have a circle instead of a
ellipsoid. So, these hydrogen atoms therefore, are represented by circles and this is now a
full representation of the particular molecular. So, this is how we get the molecule in the
in the asymmetric unit represented in a diagram like this.
679
So, just to recapitulate we got into a situation where we had the initial electron density
map of some kind. And this particular initial electron density map they could give us
some atoms in the structure. We took those atoms to be the correct atoms, calculated the
structure factors from the electron density associated with these atoms, that give us the
calculated structure factors.
The calculated structure factors and the observed structure factors are subtracted from
each other. And the corresponding agreement factor or the reliability index R is
calculated and this particular R value should be going down as we improve the Fobs
minus Fcalc. If we find that this F obs minus Fcalc after the so called revaluation goes and
makes the R factor higher; that means, either the atoms are in wrong positions or some of
the atoms are in wrong position some of them in the right positions. And all these will be
revealed in the difference Fourier map.
So, if you do a difference Fourier map it will come with ups and downs, we will study a
little bit of how these Fourier maps come up in terms of a numerical example. So, that it
makes life easier we will understand, how they associated electron density features come
up by taking an example of a Fourier series and checking out how we calculate the
Fourier series when we do the difference Fourier plot. So, in this particular case we do a
difference Fourier plot and try to improve the condition, right.
680
So, as and when these things are all done this is referred to as the refinement of the
structural model protocol, we now refined the hydrogen atoms or riding on their carrier
atoms as we just discussed. And therefore, we have now completed the refinement of the
structural model we go ahead and calculate the so called R factor. These R factors will
not go towards 0 they go towards 0, but they will not be 0 because that would mean F obs
and Fcalc are exactly equal to each other which will never happen, due to uncertainties
involved. We have so many uncertainties involved like for example, determining the
thermal parameters there are certain things determining the position of x y z also there
are uncertainties.
The thermal ellipsoid feature will not allow the R factor to go to at 0 because, we cannot
pinpoint the exact position of x y z. The only thing that will happen is R factors will go
lower if you collect the data at lower temperatures. If you collect the data at higher
temperatures then the R factors will not go down, it will have a certain value at which it
will sort of remain the same. So, then such situations we call it as convergence. So, we
go and find out the optimal value of R and when we reach the optimal value of R, we say
that we have reached convergence in the refinement protocol.
681
Now, let us see after having done the structure and having got a good R factor and
successfully determine the structure, in the space group which we have determined and
based on symmetry, we have now the structure information. So, one once we have the
structure information how do we analyze the geometry and intermolecular interactions?
This is the next that which we will take up and in this particular step we see that we have
to determine the geometry of the way in which the atoms are arranged inside the
structure.
So that means, we need to find out bond distances, bond angles, torsion angles ring
puckering geometry in the case of sucrose you see that we have the puckering of the ring
etcetera. So, all the molecular properties can be calculated here, in case there is an
inorganic system we can calculate the coordination distances and so on, with respect to
the inorganic atom which is present there. In case of organic metallic samples we can
also calculate the way in which the coordination sphere develops around a metal atom.
We can also calculate because of the fact that we have not just one molecule in the unit
cell more than one molecule in the unit cell, we can now construct the rest of the
molecules in the unit cell.
So, what we have determined by the structure determination is only the asymmetric unit
atoms. So, we can now take the asymmetric unit atoms use the symmetry information
which we have in terms of both structure and symmetry in terms of both space groups
682
and symmetry and then we can then calculate the contacts between molecules. This now
refers to the inter atomic contacts. So, we can determine the intermolecular contacts and
inter atomic contacts are available here at the bond distances angles and so on.
So, the complete geometry associated with this can be calculated using programs like
PLATON, PARST etcetera these are all readymade programs downloadable free of cost.
After that we have to see there are no funny contacts. So, suppose now you solve the
structure and then the structure develop some contacts which are very much shorter
between intermolecular contacts then you know something is wrong. So, anything wrong
at this stage of the geometrical calculation one can find out and redetermine the structure.
So, this is a place where the reliability of the structure determination is evaluated.
Having got the geometry and the intermolecular interactions evaluated, we have to
actually now validate the structure. The validation of the structure is done according to
the international union of crystallography rules; say everything is under control of the
international union of crystallography as far as structure determination and structure
reporting is concerned. So, the refinement results all the results which we have got they
are referred to as a crystallographic information file. So, this is referred to as the file CIF.
So, one has to have a CIF file available for any structure that is determined.
And that is CIF file will have so many factors, CIF file will consist of the a, b, c alpha,
beta, gamma the standard deviation that is in the a, b, c alpha, beta, gamma the
coordinates x y z and their six thermal parameters, if it is anisotropic refinement. The
hydrogen positions and their isotropic or anisotropic parameters all these features will be
reported under the CIF file. Along with the fact the CIF file will also consist information
contains information about the nature of intermolecular interactions in terms of how they
are bonded to each other. What is the symmetry operation which takes this molecule 1 to
molecule 2 through a hydrogen bonded interaction for example.
All such things are reported in the CIF file. So, one once CIF file is available one can
calculate all the geometry that is associated with the structure. Then we also the structure
validation is also seen in terms of the Fobs, Fcalc calculated data, that is the final values of
the observed and calculated data. And this is known as a FCF file. So, together CIF file
and FCF file now represent the geometry and the structure information together, because
683
one once we have an FCF file it is possible to calculate the electron density map directly.
And then verify for yourself that the structure is what is reported in the literature.
So, the CIF value and the FCF values are all available one once the structure is made and
the structure validation is checked under the IUCr protocol which is which uses what is
known as a check CIF tool. There is also what is known as a PLATON validation tool
which checks on other aspects including the aspects associated with intermolecular
interactions it will check. The various ways in which one can identify the hydrogen
bonds and other kinds of bonds; we will not it is not necessary for us to go into those
details because it does not come under the purview of the syllabus of this course.
But that particular course could be a course on advanced crystallography and the
information we get from the crystal structures. We can also check in the Cambridge
crystal structure database which holds houses all possible structures determined, so far
all the CIF files all the FCF files which are now available for all the structures
determined from day one. So, the entry is crossing hundred and 50000 entry so far and
its adds every day some 15 to 20 structures additionally. So, then we can check the
similarity between structures by using the Cambridge structural database; this we will
bring us to the issue of how we do crystal engineering analysis I will have some time
towards the end of the course to discuss about that.
So, we can also look at hydrogen bonds and various types of hydrogen bonds strength of
hydrogen bonds, we can calculate the energies that are associated with it separately. So,
essentially the structure one once determined has to be validated before it is put into
circulation. Before it is published one has to look at the CIF value and the FCF file
format values in order to make sure that they go through the officially approved check
CIF protocol. If check CIF says there are no errors then your structure seems to be in
order, ok.
684
There are certain problems which you will face any experiment we will have problems
one is of course, the way in which your diffractometer is set up the laboratory conditions
and all that considering all that deficiencies being overcome and you have a perfect
system, on which you can do a good data.
Then first major hurdle is the crystal quality. So, this is particularly so in the case of
macromolecules where the crystals of proteins and other complexes cannot be so easily
grown. It is mainly because of the nature in which they fold themselves and the nature in
which the interactions are coming in. So, the poor crystal quality can also occur in very
very small crystals as well. So for example if you get fine needle bundles all together,
that will not give a good diffraction.
So, that is the poor crystal quality. So, this is one of the technical problems one would
come across. And when we collect data on a very poor quality crystal it may be miss
leading and it may not gives the structure which is the correct structure; the second one
as we always been stressing determination of the correct space group symmetry.
Now if we have a wrong space group symmetry identified, then the structure
determination will run into serious technical problems. There is an issue of what is
known as a pseudo-symmetry and this issue of pseudo-symmetry can come up with
respect to the molecule, it can also come up with respect to the space group that is
determined.
685
So, this anomaly is going to put us into some serious issues I do not think I will have the
necessary time to cover the details of this, but this is a warning place where we can also
have this pseudo-symmetry as a technical problem. Since, I am just listing out the
technical problems. Absolute structure of light atom structures is a question mark we
used a flack parameter or whatever, but essentially we make use of the fact that there is
anomalous scattering effects.
But the anomalous scattering effects at a given wavelength like for that of copper
molybdenum they are so small for lighter atoms, that the determination of the absolute
structure is a questionable entity. If you do have a very heavy atom then the value of the
anomalous features will be higher and then the absolute structure determination is
reasonably accurate. So, this is another technical issue which will come up; the more
complicated technical issue which comes up is the so called Twinning. So, some of the
crystals will just like we have in humans twinning problems we also have twinning
problems in crystals, crystals can grow as twins.
And there are lots of different kinds of twins that can grow some crystals grow as
merohedral twins that some crystals grow with a one access unique, but they grow
together, all these problems will cause the nature of the diffraction then will be different,
because diffraction will be coming from both these crystals. And those two crystals are
joined to one another in some fashion unless we know in what fashion there are joined to
one another, we will not be able to understand and appreciate the analysis of the data
which comes from that.
So, twinning is not a is not a virtue it is a curse for it is a technical problem. Then there is
also another issue which will bother in many cases and this is a very serious issue
particularly for pharmaceutical compounds positional and substitutional disorder or part
or even the whole molecule. Sometimes the whole molecule is disordered; that means, it
is not in one position it can be in two possible positions.
The symmetry will allow it, the problem is the symmetry that is associated with the with
the determined space group will allow for the positional and substitutional disorder. And
this therefore, we will give rise to different kinds of diffraction conditions. It may look as
though because of this positional and substitution disorder, the molecule is going into
two different crystal systems which is the definition of polymorphism, but it is not the
686
case it is essentially the possibility of positional and substitutional disorder of the
molecule.
The other technical problem we normally face particularly in larger molecules is the
disordered solvent. So, the solvent maybe fully disordered or partially disordered
because most of the protein crystals crystallize along with solvent water. And this
solvent water has no structure of its own, it is disordered and because of that we will get
what is known as diffuse scattering. So, apart from your single crystal diffraction which
is coming out from the system we will also have diffuse scatter radiation coming out of
this disordered solvent. And as a result the actual evaluation of individual reflections
coming from the crystal system becomes in an issue and it is not going to be very easy.
There is another thing called the incommensurate structures that is possible because the
way in which sometimes the crystals grow. The cell dimensions which we get will be not
sufficient to describe the orientation of the molecules in different unit size. That means,
the molecules now can orient they have to be exactly the same in the next unit cell that is
the definition by translational periodicity, whatever molecule we are getting in a crystal
structure in the unit cell should repeat itself in the next unit cell should repeat exactly the
same way in the third unit cell. It may so happen that the second third and fourth unit
cells the molecules are slightly rotated from each other.
So, it's not an exact replica of the first unit cell, the exact replica of the first unit cell let
us say comes at the fourth unit cell, then we call it as the commensurate modulation. So,
the unit cell dimension now is not the one unit cell it is that of the fourth unit cell. So, we
have to go from first unit cell to the fourth unit cell fourth unit to the eight unit and so on.
And that is now the repeat distance, this is a technical issue which can be sorted out. On
the other hand if it is not repeating exactly at four unit cells, but somewhere around three
and half then it is called incommensurate structure. There are what are called
incommensurate way which go through the crystal and one has to determine that.
So, these are all technical issues which will create problems. For example, we will get
diffuse scattering, we can get streaks and diffuse layers of reflections and so on. And
these reflections can also cause serious problems. So, since we face these technical
problems, associated with crystals, one has to be aware of these pitfalls and these pitfalls
might also create bad data and one once we have a very bad data set, the structure
687
determination is going to be difficult. So, whenever we grow a crystal we should look at
all these possibilities. And if such possibilities do exist we have to get rid of them or
regrow the crystals such that they we get rid of all these. There are certain issues which
we cannot get rid of like twinning we cannot get rid of, like incommensurate modulation
we cannot get rid of, the presence of disorder and disordered solvents we can get rid of.
So, one has to be aware of these and see how we can account for the for these technical
problems, how we overcome or is there any way in which we can solve these problems.
Some of the problems which can be solved or for example, we can solve the twinning
problem we can solve the incommensurate structure problem. The substitutional disorder
positional disorder part we can evaluate, but we cannot solve it. It can still be a part and
parcel of the structure which has been determined the disordered solvent of course, will
remain disordered. And so also the diffuse scattering streaks diffuse layers and so on.
So, these are all part and parcel of any experimental protocol and experiments do have
these kind of difficulties. So, what I therefore, want to say is that I will go back in a few
slides and then tell you what we have done with respect to the data collection.
So, when we put a crystal and collect the data we get these reciprocal spots.
688
These reciprocal spots now can be indexed and we determine the all possible hkl values
from which we get the unit cell dimensions.
And one once we have the unit cell dimensions we can use this 1/d 2hkl formula to
determine the crystal system based on these and therefore, get the a, b, c, alpha, beta,
gamma value.
689
And then we determine the space group by looking at the systematic absences in their
structure, we can also get the cell dimensions and then of course, Laue symmetries there
already and the analysis of the intensity statistics, which will tell you the difference
between centro symmetric and non centric systems. We will also be possible to use this
methodology to determine the crystal system whether it is a centro symmetric or a non-
centric system.
Particularly in cases where systematic absences are not helping and then we of course,
we will go to the structure determination we follow this protocol, determine the electron
density get them the starting model the starting model is now put back into this
calculation. And therefore, we solve the phase this is the major problem in the phase
problem.
So, we get only the estimate of the phases. So, we get the starting model, the starting
model electron density is analyzed atom positions are identified and the Fcalc is made with
respect to that then we take the difference between the F obs and Fcalc and that delta F
analysis will give us the rest of the atoms. We go through this protocol until we do not
get any more of better value of the so called R factor. And at that stage we will now this
is the set of refinement procedures which we will follow and that is where we stop and
say refinement is complete and make a representation of that in an R type plot.
690
So, that is the overall protocol and then of course, we analyze the geometry the structure
validation of course, is very crucial. And this is the overall experimental procedure. Now
having seen the overall experimental procedure and the technical problems that are
involved, we will now go into the details of how we take the data, how we treat the data
and how do we reduce the data into an absolute scale before we subject it for structure
determination protocol.
And this is the one which follows here. We have already learnt some of these things, but
I will go through it in such a way that we get a hang of what we have to do. This is again
an experimental data.
So, experimental data will give us the intensity measurements, we have of course, hkl
identified. So, if the modulus of structure factor is proportional to root of I. So, it is used
in the calculation of the electron density map. This is I have written it again just for your
continued remembrance and also as a continuity. I will not be explaining it again and
again because you know what we have what we have written here. So, we convert the
intensities into observed structure amplitudes modulus of F0 by a data reduction program.
Now, this is where I told you that we can collect the data on any part of the surface of the
earth on any instrument, maybe we can even correct data in space it does not matter, but
what instrument you are using is deciding what intensities you will get because it
depends upon the input radiation, the scattered radiation, the efficiency of the detector
691
and so on. So, one once you have that data you have only a relative data, relative
intensities therefore, you get a modulus of F which we call as the observed structure
amplitude.
Now, you have to treat this data or what we call as a data reduction methodology, such
that we correct for all systematic errors that might occur on this particular value. And
one of the rigorous procedures which must be followed before we go any further is to
treat this Fhkl by this expression. The expression tells us that the F hkl modulus of course, it
is Ihkl to the power of half as it shown here and that proportionality constant is replaced
by K/Lp capital K by capital L and small p. Now, the quantity p is called a polarization
factor and the quantity L is called the Lorentz factor.
The polarization factor depends upon the incoming radiation and the scattered radiation
and is a simple function of 2 theta; so we can correct this p as equal to (1 + cos 2
2theta)/2. So, independent of the method used for data collection except when a crystal
monochromator is used we will not go into the detail of it except to tell that this
polarization comes up due to the fact that even the x ray radiation is coming they are all
circularly polarized. That means, the electromagnetic waves are uniformly oscillating in
all directions. And, when the impinge on the on the diffracting plane or the reflecting
plane on reflection it. So, happens that they go in such a way that if they are partially
polarized.
So, when we are measuring the intensities of partially polarized light and p is the factor
which corrects for that partial polarization. The factor L depends upon the way in which
we measure the data, remember we have this problem of the reciprocal lattice rotating
with the crystal. And the origin of the crystal is somewhere at let us say some central
point and as you go in the reciprocal space to higher and higher angles. The amount of
time that is spent by a reflecting geometry sphere; that means, the way in which the
diffraction occurs is that that particular point will intersect with the Ewald sphere.
Whenever the reciprocal lattice point intersects with the Ewalds sphere then we get
diffraction.
Now, the amount of time that is spent by that particular diffracting point will be different
depending upon the orientation and the distance of that point with respect to the origin.
And this is where we have a issue. So, the issue is in terms of L which is called the
692
Lorentz factor and this is dependent upon (1/sin theta). So, we use this L and p this is
independent of the method of data collection we have used it is independent of the
instrument we have used where the L and p is a constant factors. So, we calculate these L
into p and put it in this expression K is the overall scale factor, with which we have to
now put the Fhkl on take the Ihkl take the square root and K is the proportionality constant.
Lp correction is made on the every intensity which we observed.
So, the values of L and p are listed in international tables again in a different volume not
in the volume where we find the space group information, but in the second volume of
the international table. So, one can get the values of L and p as you see they depend only
on the scattering angle. And therefore since they depend on the scattering angle they
depend upon at what angle this scattering is coming, at what angle the h k l reflection is
coming rather than anything else.
So, therefore, it is possible to get the Lp corrections uniformly organized for every given
wavelength, it is going to depend on the wavelength; obviously. So, it is a wavelength
dependent correction Lp, but it is available on the international table.
Now, K is depends on the crystal size the beam intensity and a number of fundamental
constant it is normally a constant for a given set. That is the proportionality K we are
talking about which depends upon the way in which we collect the data.
693
So, effectively we have to treat the intensities and make the corrections associated with
the Lorentz and polarization factors. Before we process it further this whole processing
of the data depends upon various issues, it also depends upon the way in which the
amount of X-ray gets observed inside the crystal and way in which the total internal
reflections can take place depending on what we call is extinction factor and so on.
These are all parameters, these are all factors which affect the diffraction condition. And
these factors which affect the diffraction conditions obviously, are parameters.
So, we can refine these parameters overall in our structure determination protocol. So,
the absorption correction can be applied we can refine the absorption correction in the
way in which we would like to do. And also the case of the extinction we can apply the
extinction correction. So, therefore, the data has to be manufactured in such a way that
this particular data which we are going to use finally, is put on an absolute scale.
So, in order to put the date on an absolute scale we have a protocol on that protocol we
will have to depend upon the fact that we have already made the structure factors into a
proper position and a proper scaling. So, the amount of information which we have is
enormous, we have a very large number of data points and those data points now has
been treated for this data reduction protocol.
So, we now have a collection of Fhkl this Fhkl is now divide of all systematic errors as far
as possible. The only problem is it is still dependent upon the way in which you have
collected the data the instrument that is used the way in which the detector has been
place. So, it depends whether you have used a photographic film or you have used an
imaging plate or you have used a CCD detector or a CMOS detector. So, it all depends
upon the power with which you can convert the number of photons which are falling on
the detector to the electrical signals which eventually measures the intensity.
So, the intensity measurement therefore, is a totally relative phenomena and. So, we have
to therefore, put this intensity measured on an absolute scale. And the measured intensity
to be put on an absolute scale essentially comes from the fact that we need to have them
on the scale which counts the total number of electrons. So, what the logic of this whole
thing is that we have a crystal and in the crystal, we have electron density and the
electrons are all over the place it is distributed all over the place.
694
Of course, they are atom centered, but anyway for all practical purposes it is not the
position of the atom which matters now as far as these the scattering overall is concerned
it essentially matters that we have the total number of electrons. So, what we can do is
we can normalize all the structure factors with respect to the total number of electrons.
So, in other words the quantity F 000which is our standard quantity. So, the quantity F 000 is
equal to the total number of electrons. So, we can call it sum over all the z values of
every atom.
So, if you have i atoms in the structure i runs from 1 to n. So, sum over Z i. So, suppose
you have 20 carbons, 30 oxygens and so on. So, 20 time 6 plus 30 oxygens times 8 that
summation will be now the F000. So, that gives us the overall content of the of the unit
cell. So, the overall content of the unit cell is sum over z of i which we have to determine
this includes the symmetry related atoms as well. So, that provide us the required scaling
and that required scaling is the one which we will have to worry about. And how do we
find that scaling factor is something which we will discuss in the next class.
695
Lecture – 51
Data Reduction
So, as I was mentioning in the last class set of classes, maybe a couple of classes ago that
this is the Year of Periodic table.
So, the United Nations has declared this as the Year of the Periodic Table and what is
very crucial from the point of view of the periodic table is the way in which it is
arranged, you see that Mandelif who is shown up here. This is the plot which is in front
of the University of Bratislava in Slovenia.
And this is the one where we have the display of all the elements and you see that he also
loved symmetry. So, all the elements have been arranged in a symmetric manner. You
can see that all the rare earths are up here. A helium, neon, argon, krypton, xenon, radon
and you see that the this is the completion of the sequence. Fluorine is up here and you
can start from this side hydrogen lithium, sodium potassium, rubidium and so on. So, in
fact, this is the way he preferred to probably have the arrangement. We do not know, but
essentially now, periodic table appears in the current acceptable form, I mean universal
basis.
696
Now, most of the e m structures are made up of elements and it is the symmetry in these
elements which actually carries through to the symmetry of diffraction. As we have
already realised that we assume that the atoms are spherical and scattering that is coming
from the atom is essentially from the electron density which surrounds these atoms and
therefore all these atoms scatter and we know that the scattering factor falls off with
respect to sine theta by lambda. It also depends upon the value of z which is indicated in
this periodic table.
So, there is a very close resemblance and very close association of what we get from an
X-ray diffraction experiment; what we get as symmetry information with the
arrangement of atoms in the periodic table. Towards the end of the course, we will see
whether we can see some details of these individual elements and their properties from
the point of view of doing what is known as quantum crystallography; that will be the
last part of the discussion in this course.
So, what we will do now is to go back to our the scale factor issue and the one once we
have done the Lp corrections as is shown here the Lorentz polarization correction. We
can therefore get the relative values of the structure factors in terms of k prime and F
naught. The value of k prime is the one which we have to determine and it depends upon
crystal size, beam intensity and the number of fundamental constants. It is normally a
constant for a given set of measurements. So, we can actually calculate the value of k.
697
So, we can write once when we have the F relative we can write the I relative as in this
form. The F of relative square averaged over different ranges of sin theta by lambda and
for a unit cell when it contains n atoms, the theoretical average intensity is I abs is sum
over i to n of fi square.
What it essentially tells us is that the average intensity depends merely on what is in the
cell in terms of the electron density and not where it is located. So, the absolute value
depends essentially on the sum of i equals 1 to n of f i of square. For all the atoms that are
present inside the unit cell and so this gives us an idea of how we can go from I relative
and get to the actual I abs through the use of determining. This is so called Structural
scale factor. So, the ratio of I abs to I relative therefore should be the scaling factor as I
mentioned and this is allowed.
Because of the fact that we have now all the electrons in the scattering material
represented in terms of the absolute value of the intensity. So, as I mentioned in the
previous class, wherever we measure the intensities whichever instrument we use,
whichever radiation we use, we can actually put the observations on an absolute scale by
calculating this. I absolute from the I relative values, but working out the right kind of
scale factor. This process in which we do this analysis is due to Wilson and it is known
as the Wilson Plot.
B is Debye Waller factor, λ is the wavelength, amplitude of atomic vibration is u.
698
Of course, these scattering factors are also affected by the thermal motion of the atoms;
if the atoms are not at rest, they are vibrating in their rest positions and the magnitude of
the vibration therefore, depends on the temperature, the mass of the atom, the firmness
with which it is held in place by covalent bonds and other forces. So, higher the
temperature greater is the vibration, electron cloud spreads over a larger volume and
scattering power of the atom falls off rapidly with the ideal stationary model with the
formula e minus B sin squared theta by lambda square representing the fall off.
So, B is called the Debye Waller (Refer Time: 05:51) factor and it is actually
representing how far the atom is moving away from its mean position. The mean square
amplitude of such a situation is u average square. Then, B is equal to 8 pi square u
average square and this is known as the Debye Waller factor. So, the scattering factor
gets modified as a consequence; F becomes F not; e to the minus B sin square theta by
lambda square. So, the modifications that get introduced in F can be represented in the
form of a behaviour of sin theta pi by lambda and the variations with respect to that.
For example, if we put now e minus B sine squared theta by lambda square on the y axis,
plot it against sin theta by lambda. As the value of B increases, of course B equals 0 will
be very constant. There is no thermal parameter associated with it, but that is not very
ideal case which is impossible case. In fact, because the atoms are always moving they
are not at rest any time. So, as the B value increases you notice that this value which is e
699
minus B sine square theta by lambda, the so called thermal power factor. It influences the
curve in such a way that we get this fall off. So, essentially what therefore, happens is if
we now consider the scattering factor and multiply it by this thermal parameter, we see
depending upon the revival of factor the scattering factor fall off f versus sin theta by
lambda takes different asymptotic paths.
So, the extent to which the scattering occurs at somewhat an angle like 0.3 sine theta by
lambda, you see that it gets reduced considerably, if the thermal parameter is large. So,
the contribution from the atom now essentially has to be read out from this curve to
various scattering angles. So, thus the scattering angle increases, we have reduction in
the F value and the further reduction is caused by the presence of this so called thermal
parameter. So, having kept this in mind and looking at these two diagrams, we see that if
the temperature is a very crucial factor in our all our X-ray diffraction measurements.
So, unless otherwise there is a phase transition, it is suggested that the material under
investigation, the data is better collected at low temperatures. The lower the temperature,
the better is the approximation of F sine theta by lambda to ideality and as a
consequence we will get the entire contribution to the diffraction condition. So, the
diffraction; the reciprocal lattice point intensities will improve when we reduce the
temperature and at the same time of course, there is a little bit of contraction of the lattice
and this contraction of the lattice can be anticipated and evaluated. It is a straightforward
proportionality factor and therefore, it is not a serious issue. The serious issue will be
when there is a phase transition which we are not going to consider in this particular
course, we are not discussing phase transitions. So, as we go down to lower
temperatures.
So, it is generally suggested that the ideal temperature would be to collect the data at
helium temperatures; something like 10 degree Kelvin or something. However, you
know getting the 10 degree Kelvin is a very serious issue, we will have to spend a lot of
money and as a consequence most of the data is suggested that it is collected at liquid
nitrogen temperatures. Liquid nitrogen temperature, the boiling point is 77 Kelvin. So,
we collect the data around 100 K.
So, most of the data, it is suggested that one collects it at 100K to get better resolutions
number one and secondly, more reflections as a consequence because there is a
700
contraction of the lattice all right. But the reciprocal lattice even though it contracts more
and more observable diffraction spots will appear because the reciprocal lattice volume
depends upon the main unit cell. So, as a result we get the larger values associated with
the scattering coefficient and therefore, the overall scattering improves considerably.
Iabs is absolute intensity, Irel is relative intensity, fo is the atomic scattering factor at Sinθ/λ
equals to zero.
So, having noticed that aspect we will now see what happens when we want to put it on a
proper scaling. So, the I abs which we are talking about the absolute intensity therefore,
now not just depends upon the fi square, but it depends upon f which is at rest which we
call it as foi square that is the atomic scattering factor with respect to the sin theta by
lambda. The measure of the Z value, but that gets modified by the exponential minus 2 B
sine square theta by lambda square.
So, this modification of interaction of minus 2 B sine squared theta by lambda square
will take into account, the thermal parameter. So, therefore, we get the I absolutely as
equal to e to the minus B sine square theta by lambda square. We take it out; we will
assume that all the atoms of the same thermal parameter. So, we take it out and minus 2
B sine square theta by lambda square comes up and then, we have the summation over i
to n of foi square in which now varies with respect to the nature of the atom. So,
701
essentially what it tells us is that we now assume that all the atoms are vibrating at the
same temperature.
And therefore, the temperature effect is identical on all these scattering factors of all the
atoms which are in this structure maybe there are 30 atoms, oxygen, nitrogen, chlorine,
bromine, whatever. So, the fall off of f versus sine theta by lambda is fairly well defined
in terms of the behaviour of f and then, we modify it with a thermal parameter which is
now modifying every one of these fall off of f sine theta by lambda for every atom.
So, when we do this, we get a value of the absolute value of the intensity. Now, what we
have measured from the experiment is the relative intensity. So, we can write relative
intensity is equal to some constant times the absolute intensity. So, we can rewrite this
expression, substitute for the absolute intensity as e to the minus 2 B sine squared theta
by lambda square summation over i to n of foi square.
So, this now, represents the relation or the value that we should get for the I relative
which we have measured. So, this is the measured value, this is the absolute value we are
looking for. So, if you take the ratio of I relative to the sum over i to n of f oi square, this
ratio is essentially equal to a constant C times an exponential factor. So, this, now
exponential factor if you take logarithms on both sides. So, you take the logarithm of this
quantity, logarithm of this and logarithm of that you can write this in the form of a
straight line equation.
The reason is the following, we can write it like we can write the equation like y equals
ax plus b. So, this will be like a straight line that is because we take the logarithms on
both sides. So, this then will become log C minus two B sine squared theta by lambda
square. So, this now, looks like y equals ax plus b right and therefore, we get a straight
line and we plot that.
702
Here is the expression given logarithm of I relative by sigma, i going from 1 to n and so,
we get this expression and this is like y equals nx plus C equation. So, when we plot this
logarithm value against sin squared theta by lambda squared, we will be ending up
getting a straight line; the intercept is log C and the slope will be corresponding to 2 B.
So, we get therefore, the value of the scale constants which is k equals 1 over root of C.
So, from log C, we can calculate this quantity and the thermal parameter, the overall
thermal parameter associated with scattering can also be calculated by measuring the
slope of this curve. This is the Wilson plot. I will spend a few minutes on this Wilson
plot because this is something which is of crucial importance in structure analysis.
The nearness of the it is all not always ideally a straight line, it is nearly a straight line. It
depends upon to some extent where the atoms are located inside the unit cell, even
though we already told that the overall content of the scattering is what is there inside the
unit cell rather than where it is, but it also matters that where it is; also matters. In the
sense that suppose, we have several ring systems, then what happens is that we will not
get a straight line. We will get some kind of a deviation from this straight line;
somewhere around sin theta by lambda 0.18.
So, if we get a significant deviation at that point, we know that we have a ring system or
a collection of 6-membered rings or 5-membered rings in the structure. The other thing is
if the Wilson plot is really cranky, it is not looking like a straight line. It is never an ideal
703
straight line, but if it is not mimicking a straight line, then we know there is something
wrong in our data reduction. So, Wilson plot is not only used to find the scale and the
overall thermal parameter, it is also in a way checks the quality of our diffraction
measurements and also the data reduction procedure.
So, if we have done something wrong in the data reduction procedure, this will not be a
straight line. So, Wilson plot therefore, is a very essential part and parcel it is a part and
parcel of any structural determination. The advantage of doing this Wilson plot also is
that we get a overall thermal parameter associated with the structure. So, we could
associate this thermal parameter in the future improvement of the structure, when we do
the so called refinement.
As I mentioned in the last class, there are two steps in which we determine the structure;
the first step is to get the trial structure by trying to solve the phase problem, get enough
information on the phases which though approximate are sufficient enough to give us the
starting model and this starting model need not have the idealized geometries. So, what
we do is we use the over determinacy in the data, the number of reflections which are
available. Again, we put it back into the calculation.
So, we calculate the structure factor, we come take the observed structure factor and take
the difference between the observed and the calculated and that quantity, the difference
delta F is minimized and when we minimize that quantity, we get to a better and better
structure. Geometrically as well as from the point of view of accounting for the entire
electron density which is their inside the crystal structure.
So, the fit of the F not and Fc is very crucial and this is evaluated in terms of several
figures of merit; one of them we discussed in the last class which is the r factor which is
modulus of F not minus modulus of F t divided by modulus of F naught. So, these are
some of the issues which will come out in when we do the Wilson plot. So, in it in a
typical experiment, we collect the data, these data are measured the intensities are
measured there on a relative scale and these relative intensities can be put on an absolute
scale by this procedure which we just now discussed.
And based on the Wilson plot, we get the scale factor which converts the relative
intensity to absolute intensities. So, that we can get this expression satisfied F abs equals
704
k times F real and this is the rel is there real the relative intensity which we measured
taking the square root of the intensity which we measure the relative structure factor.
So, the structure factor; absolute structure factor is what we need for our future phase
solution and also for the Fourier transform which will then give us the electron density
distribution. In this context, if we now plot this is known as an NZ plot. N is the fraction
of the data which is represented in terms of the fraction Z. So, the number of data points
with respect to the fraction Z of the collected data number; the number of data points. So,
this is the fraction of data compared to the number of data points.
If we plot this; this gives us a curve which has the value of Z increases with respect to
NZ. This is known as the NZ curve. This essentially tells us the nature of the distribution
of intensities with respect to Non-centrosymmetric and with respect to Centrosymmetric
systems. So, here is the situation where we can use this to a large extent, when we have
no systematic absences which are depicted in the structure. And at the same time, we are
looking for the presence of centric or a Non-centrosymmetric space group. The one of
the easiest examples is from the triclinic system, when we have P1 and P-1 these there
are no systematic absences as we have discussed it thoroughly.
So, since there are no systematic absences, what we get is a situation where we analyze
the intensity if this is the intensity distribution with respect to various sine theta by
lambda ranges. So, the intensity distribution therefore, tells us that in the case of a non-
centrosymmetric structure the curve which is shown below is generally followed and in
the case of a central symmetric structure, the curve which is shown up is followed. So,
one is an error function; the other is an exponential function.
So, as a consequence we can distinguish between P1 and P-1. So, it is also true that we
can distinguish between P2, Pm and P2/ m and by looking at certain statistics which are
associated with projection reflections, it is possible to distinguish between P2 and Pm.
Because P2 and Pm will give same system and there are no systematic absences so, also
with P2/m. In other words, whenever we determine the structural features this way
whether it is centric or non-centric, it is based upon the overall intensity distribution.
And all these approximations which we have used to find out the scale factor and thermal
parameter evaluation is also incorporated into this. And therefore, it is not 100 percent
reliable. So, the way we do that is either solve the structure as a central symmetric
705
structure or preferably solve the structure in the non-centrosymmetric structure. Assume
that it is a P1 in fact any crystal structure take it from me, any crystal structure can be
solved as P1. Because if you solve the crystal structure as P1 and then, determine where
the molecules are, the molecules now show up the symmetry relationships between them.
So, if we can identify the symmetry relationships between the molecules, we can now
downgrade these or upgrade the P1 system to higher symmetry space group. So, it is
possible therefore, to solve a structure of P21\C as a P1, then we will find four molecules
which will show the relationships, which will fit into the P2 1\C. So, in other words, the
non-centrosymmetric structures in P1 and P-1, we can solve the structure in P1, but then
identify the 1 bar later on and put it across to P-1. The indicated deviation from non-
central symmetry to central symmetry, in this particular situation is only a guideline.
So, the intensity statistics is not a foolproof methodology to find out the whether it is
centric or non-centric, it is a guideline to see whether probably one is centric the other is
non-centric; we cannot conclude it. We can conclude only after the successful final
refinement of the structure in the two space groups P1 and P-1. So, we just have to
remember that when we come across such systems.
So, basically what we have done so far in the discussion now is to start from the fact that
we have now measured the intensities. We have identified the coordinates h k l in
reciprocal space so, we have the intensity measured. We collected it for the Lorenz and
polarization effects which we discussed in the last discussion and having done that we
now get the relative intensities. Now, we can also get the absolute intensity because the
average intensity depends on what is there in the crystal; what is the total electron
density that is there inside the cell, unit cell and not where it is and therefore, we have
this expression.
Now, only thing now we have to worry about is the fact that they are all affected by the
thermal motion. So, the thermal motion has to be introduced and that thermal motion
which when we introduce introducing the concept of the Debye Waller (Refer Time:
23:35) factor, we find that the behavior of the F what is the sin theta product changes as
this is shown here. So, therefore, we modify the existing is sum over fi square by the
corresponding expression which will give us the I absolute.
706
Now, the I absolute therefore, can be written as equal to this quantity. We will take out
the thermal parameter assuming that all our atoms in the structure are having the same
similar thermal vibration factor. It is not the true case, but we can make an
approximation because some of the light atoms may vibrate more, the heavy atoms may
vibrate less; hydrogens may behave funnily we do not know how they are regions
behave. So, all these things matter. But in principle we can have a overall thermal
parameter associated with the diffraction condition.
So, then we now can calculate therefore, a quantity which is now the logarithm of I
relative by sigma i foi square which now represents the actual; I absolute value and then,
we can get an expression which is something similar to a straight line which is y equals a
x the a x plus b kind of a situation. So, we plot this quantity logarithm of this versus sine
squared theta by lambda squared, we get a straight line. The intercept gives you the
evaluation possible for the scale factor by means of which we put the relative structure
factor onto an absolute scale. Only after putting it onto the absolute scale, we can attempt
the structural determination protocol.
So, this is a must and I already listed out the importance of the other details which we
can get from Wilson plot. The non centric distribution based on the intensity statistics is
only an indicator, but that is a useful indicator to tell us which is which. So, in a practical
situation when you are collecting data, the data is collected in such a way that we
accumulate all the data of the h k i and the h k l; make a list of all that and then we do a
statistical analysis. We calculate several parameters other than this essentially to see how
the intensity statistics varies with respect to sin theta by lambda.
Now, the factor sin theta by lambda is very crucial because our atomic scattering factor
falls off with respect to sin theta binomial. If data had not happened, then there was no
problem because then we would have had a constant of the atomic scattering factor
associated with one diffraction condition, of course modified by the thermal parameter.
But now it so happens that it falls off with respect to sine theta by lambda. So, what we
do is, we will consider regions of equal value of F with respect to a range of sin theta by
lambda. In other words, we assume we take the reciprocal lattice information, this
limiting sphere and then divide the limiting sphere into a large number of concentric
707
circles and these individual concentric circles now represent the reflections where the f
versus sin theta by lambda is more or less slightly in like behaviour.
So, in other words if we go to the previous projection, you see that here if we take this
region and go up to let us say 0.05, you see it is almost a straight line; then this part
which is almost a straight line, that part which is almost a straight line. So, we actually
take those regions where it is almost a straight line and that gives us the possibility of
using this expression in a straightforward way. So, we therefore, take our limiting sphere,
the reverse sphere related limiting sphere and divide this limiting sphere reflections
which are appearing as h k l which are appearing as spots and these spots now form in
different concentric circles, where the variation of F with respect to sin theta by lambda
is minimal and that is very very important.
Otherwise the validity of the Wilson plot calculation here which is followed by the plot
which we have done here becomes not very critical. So, this is now very important from
the point of view of determining the scale factor and also the thermal parameter
associated with it by looking at the slope which gives us minus 2B, where B is the
divisor effective.
So, at this stage, we can now go over to another aspect because now we what we are
done is we have the experiment, we have now taken the; from that we have reduced the
data after reducing the data, we have put the data on an absolute scale. So, what is our
data? Our data is modulus of Fabs absolute value of the modulus of the structure factor.
So, what we therefore need obviously is, are the phases.
If we multiply these by the phases then we can do a Fourier transform to go to electron

density. So, before we go to the Fourier transform, we should know something about
what this Fourier series is all about and how the Fourier transform now brings in the
issue of the presence of the variability associated with the relationship between F abs the
absolute value of the structure factor multiplied by the phase and then the Fourier
transformed taking us to the electron density. How does this come about? We should
know a little bit of background information on what we call as Fourier series and at the
same time, we should also know something about how we do this Fourier transforms. So,
the next set of slides which we will consider will be the one which we will discuss now
the concepts associated with the Fourier transforms.
708
f(x) is a function of x which can be expressend as combination of sinusoidal functions.
So, we see here that we first look at the Fourier synthesis, with the point of view that the
structure factor is a Fourier transform; it takes around two electron density.
709
Lecture – 52
Fourier Syntheses
So, what is Fourier series? So, first let us learn something about the Fourier series. This
is a very general expression which we have written here for Fourier series it can be f of x
equals a not plus a 1 cosine 2 pi x plus a 2 cosine 2 pi 2 x. So, we are now writing the
cosine part of the f of x and then we write the sin part of the f of x with coefficients b 1,
b 2, b 3 and b n. So, a 1, a 2, a 3 are the ones which are associated with the cosine the
real part of the function and the b 1, b 2, b 3 are the ones which are associated with the
sin of the function and you see that sin 2 pi x, sin 2 pi 2x, sin 2 pi 3 3x and so on sin 2 pi
nx. It essentially represents a periodic function.
So, this overall is a periodic function. One is a periodicity associated with cosine, the
other with the sin. Both of the cosine and the sin functions are anyway we know are
periodic functions and therefore, effectively what it represents is a function like this. You
see the resemblance of this with our structure factor expression in terms of the cosine and
the sin being present, the only thing which is not present in this is the i, the absolute the
710
imaginary component associated with that, but then we can also have the imaginary
component brought in here with by expressing this in terms of an exponential.
So, basically the Fourier series represents a periodic function. So, if a function f which is
a variable x associated with it and that value of x comes in these positions associated
with the cosine function as well as the sine function. And, this periodicity is expressed in
this particular form.
So, now, let us go over to a situation where we have h in this particular case is an integer,
a and b are constants and x is a fraction of a period. So, this is the periodicity x is the
fraction of the period.
Y is the periodic step function. F(hkl) is the structure factor, ρ(xyz) is the charge density
at given xyz. V is the unit cell volume.
So, when we have a situation like this it is better we understand a little better in terms of
actually considering a function which is periodic. So, what we do is we consider a
periodic step function as is shown here. This is from the textbook of Stout and Jensen of
course, what I will do is that towards the end of the course I will give acknowledgments
to all the textbooks from which I have taken the pictures. I will also give references to all
those textbooks and at the same time acknowledge the individual authors because, but
711
for them this course could have could not have been compiled. So, we will come to that
when we come to the end of the course.
So, this is a periodic function, the periodic function is a four term cosine approximation
of a periodic. So, we are dropping out the sin part, we are looking only at the cosine part
and this is the expression Y equals pi by 4. Pi by 4 now defines the extent to which the
step function goes. So, this is a step function. So, we step up here by pi one fourth, keep
it steady come down then this is 0 for some time and then again one fourth and again 0
for some time and it repeats periodically. So, the periodicity is in terms of this to this
distance. It could be any point here and any point here is a periodicity one as apply pi by
4 step associated with it.
So, this step function can get a mathematical form pi by 4 plus cosine 2 pi x minus 1
third cosine 2 pi 3x plus one fifth cosine 2 pi 5x; of course, we can add many many many
more terms. We are now making an approximation by using a four term cosine function
to fit this so called step function. So, we have a step function already given to us and we
want to fit the coefficients in such a way and organize the expression in such a way that
we fit it to this step function.
So, let us take the first part which is the pi by 4. So, if we take the pi by 4 this is
represented in the diagram b here which will tell us that this is at a height of y equals pi
by 4. So, this we call it as the first component y 1 which represents the period of the step
which is repeated here, the amplitude of the step which is repeated here. The step size in
other words is one fourth and this is told shown as this is pi 4 pi by 4 and that is shown in
this first plot. The second plot takes the plots the second function which is cosine 2 pi x.
Now, what we will do is we plot the cosine 2 pi x y 2 is cosine 2 pi x and that cosine 2 pi
x follows this part and similarly the minus one third cosine 3 2 pi times x is a plot which
is shown here. And, the fourth function which is one fifth cosine 2 pi 5x has a function
which is behaves like this in this particular plot. So, these are actual plots of individual
components of these four term case. So, this is term one, this is term two, this is term
three and this is term four.
So, now, we will represent a overall Y where we sum up all these four as is shown here.
So, when we take the sum of all these four expressions, we will get a function which is
shown here on top of the existing step function. So, our idea is to fit this step function by
712
an equation right and we have taken four terms here and these four terms almost nearly
fitted by this variation. So, this variation is showing the deviation of the fit with respect
to the actual existing one.
So, suppose this you consider as your f ups ok, this will be your f calculated and you see
that there is an approximation associated with it. Now, as we keep adding the functions
we can add now the function one seventh minus one seventh cosine 2 pi 7x and then we
can add the next function which is plus one ninth cosine 2 pi 9 x and so on. So, we keep
on adding it. Take two infinite series then eventually what it represents is this function
the step function.
So, the four term approximation is bringing essentially the near complete representation
of the periodic function. You see the periodicity in the repeat of the four functions up
there in pi by 4 region and down below in the pi by 4 region. So, even though for
example, if you add this at y one fourth throughout you see that the big dip in this
function the dips associated with these functions will take it at the 0 level. So, y one
fourth gets cancelled and therefore, you see that one can take additional terms we can
have ten terms. The larger the number of terms the better is the fit and that is how we can
now look at the Fourier series.
So, in the case of the resemblance of this Fourier series to what we are doing here the
Fourier transforms you see that we have a large number of observations, j it is a large
number this is over all the atoms, but the observations we have or in terms of the h, k and
l and we have a large number of hkl measurements from our reciprocal lattice
availability. So, from the reciprocal lattice we have measured a very large number of h k
l. What we have to fit of course, is are the values of x j, y j, z j which are associated with
the with the atomic scattering factor f j, but if you look at the expression below you see
that this is now the Fourier transform of the top expression and the Fourier transform of
the top expression now, gives us the electron density.
So, if you now consider this part this now should in principle represent the electron
density associated with the system. And this now is replaceable this part is replaceable
by this quantity which is practical. So, we write rho of xyz as equal to 1 over V because
we are now looking at a volume of the unit cell. See this again brings us back to the
symmetry and structural details we have because of the periodicity we do not have to
713
sum it over the entire crystal. The crystal can consists of a large number of unit cells
thousands of unit cells. The calculation can be restricted to the volume. So, whatever we
get in the volume will repeat in the next unit cell volume and so on. So, we can calculate
only for that.
So, therefore, this expression rho of xyz will get this 1 divided by V as we see in both
these cases. The presence of 1 divided by V essentially tells us that we are now in a
define single crystal and we have the unit cell that is defined. And, so, we then I will do
it over sum over all the hkl values you see you have thousands of measurements. So,
every measured F of hkl modulus you have calculated you have put it on an absolute
scale by the Wilson plot and then this exponential i phi hkl will give us the phases. How
do we get to these values is the next step which we will consider times exponential 2 pi i
hx plus ky plus lz which is a representation of the atom position xyz.
So, with respect to all the atoms we have this general expression for rho of xyz which
depends upon the summation over all h, k and l. In principle if we want a perfect fit to
this y one fourth curve it should be an integral value of all these terms. In other words we
should have replaced this by a triple integral integration going over h, k and l minus
infinity to plus infinity, in principle, but in a crystal we have the advantage of limiting
the number of h values to a certain values k to a certain value and l to a certain value.
So, it is now a effectively a triple summation replacing the triple integral; integration
over h integration over k integration over l is replaced by summation over h summation
over k summation over l. So, this is a 3-dimensional summation which now takes the
Fhkl value. It also takes the phase into account and therefore, the electron density can be
computed this represents now a complete representation or near complete representation
of the periodic function.
So, let me repeat this issue a little bit. What is happened here is that we have now a
function and that function is now chosen in such a way that it will mimic the axial
electron density distribution inside the crystal. If we want to remain mimic that one we
have to now have it in terms of what we call as the Fourier series and this Fourier series
expansion we have discussed in terms of an expression which is a four term expression
here. We see that it is almost there, if you add four more terms it will become better and
better. So, as we keep adding terms it will become better and better. So, the rho of xyz as
714
we keep adding over all possible hkl’s this will give the best possible identification of the
electron density. So, that is the logic with which we build up a Fourier series.
P(uvw) is the patterson function, V is unit cell volume, F(hkl) is the structure factor, Z is
the atomic number.
Now, we will see how to solve the structure. You see the we have in order to solve this
structure; let me go back to the previous slide in order to solve this structure, we need
this fellow and this is the phase factor. As we have been telling over and over again this
phase problem is a serious problem. In fact, it is so serious that there is no solution to this
problem we can only get approximate values of the phases. We cannot get the actual
values of the phases.
So, we can only make an approximation closest to the right phases and from there on go
and try to improve the structure. So, that is why we get a trial structure and the trial
structure is taken to the full better refined structure based on the fact that we have over
determined data set. So, the determination of the phase therefore, becomes very crucial
in order to that is the major step for toward structure determination. So, one once we
have this, we can then use this expression to calculate the electron density distribution
which will tell us where the positions of the atoms inside the unit cell.
715
So, the summation is over the unit cell and within the unit cell we will get the
information of where the atoms are sitting and since Fhkl depends upon fj the nature of
the atom will also be shown one once we compute this quantity. So, we will know
whether it is a carbon nitrogen oxygen depending upon the value that we can associate
with fj. So, if this is the procedure we have to now think of the methodology by means of
which we can solve the phase associated with this. So, the so called solution to the phase
problem.
One of the solutions to the phase problem was arrived at by a person called Patterson. Of
course, I have written it here that he worked in Eugene Warrens lab at MIT 1935 he
developed this expression. This expression you see has lot of similarities to the previous
expressions except that we are now using modulus of F of hkl square. When we use
modulus of F of hkl square after the Wilson plot, we are using absolute values of the
intensities. So, F of hkl square represents is proportional to the intensity with all the
constants and things like that.
So, we have therefore, F of hkl square has the component here times cosine 2 pi hu plus
kv plus lw where u, v and w represent the inter atomic vectors. What we mean by inter
atomic vectors? These are vectors which are between atoms. So, for example, u is
suppose there are two atoms in the structure u is equal to x 1 minus x 2; the x coordinate
difference. Similarly, we can get v equals y 1 minus y 2; w equals z 1 minus z 2. So, this
is the distance between the two atoms and that comes in as the inter comic vector and
that is the factor which comes here.
Otherwise you see the similarity between the expression we had for the structure factor
and the electron density. Here instead of the electron density we have the so called
Patterson function. P uvw therefore, represents the Patterson function and since these are
vectors this represents a vector space. So, these are this now defines a vector space we
have the direct space we have the reciprocal space and now we also have the vector
space.
So, Patterson represents the vector space. What you see here is the total absence of the
face. So, we do not have the phase information that is required here; it is become
everything is real. We will see the properties of Patterson in a few minutes, but here the
quantities are only real quantities because we are using cosine function and there is no i
716
sin component h is associated with it and as a consequence this can be directly evaluated
in real space. So, it is a vector space and vector space now represents the real space
vector space.
And, so, the way in which we go about the term using this Patterson function is
something which we will describe in the near in the next few slides. Even though
Patterson function looked as though it can solve any structure in fact, it in principle it
can. However, one once we have this Patterson expression it becomes extremely difficult
to evaluate individual atom positions from the collection of all these inter atomic vectors
because Patterson is by itself a centrosymmetric system and it represents a
centrosymmetric map and therefore, this vector map is centrosymmetric; that means, u
can be x 1 minus x 2 it can also be x 2 minus x 1.
Both are one and the same and they are expressed with respect to a common origin. So,
as a consequence there will be a large number of overlapping vectors.
So, the as the crystal size increases 1 or 2 atoms 2 or 3 atoms is no problem, but as the
size of the molecule increase suppose we have 20 atom, 30 atom, 40 atom structures the
number of vectors will enormously increase. We will see how it all happens in the
coming slides. But, one thumb rule which I will mention at the very beginning is that
when you want to determine the structure using the Patterson function. It is generally
recommended particularly for small molecule structures, for a successful structure
determination you should look at this ratio.
The ratio of the number of heavy atoms, the square of the Z value of the number of
heavy atoms; sum over all the possible heavy atoms. Suppose, there is only one heavy
atom then it is depends upon this Z square of the heavy atom, Z being the atomic
number. So, if we have carbon it is 6 square it is not a very heavy atom, on the other
hand if we have bromine we have 35.5 square which will be a very heavy atom. So,
compare it to the weight associated with it. The weight associated with bromine is much
much more than the weight that is associated with carbon.
So, it is generally recommended that when we want to use Patterson function and do not
go to any other phase determination protocol. The approach of Patterson is preferred
whenever you have this ratio satisfied, at least nearly equal to 1. It is not a not a
necessary requirement that is why I say it is a convenient thumb rule. But, whenever we
717
are close to this ratio with all the remaining atoms in the structure represented by this
denominator and then it is not a serious problem to use the Patterson function.
So, Patterson function will give us now the heavy atom positions uniquely and fairly
accurately because this information in the Patterson is after all a development of the
presence is due to the development of the presence of atoms inside the unit cell and that
is what we are looking for. So, it does give us the positions of the atoms particularly the
heavy atoms. And, once we have the data which is now giving us the heavy atoms we
can use the over determined data set to determine the remaining light atoms by the by a
procedure which we described just now the using the Fourier synthesis.
So, this is just a convenient thumb rule. Now, let us look at the other properties of the
Patterson function.
So, as I mentioned there is a vector between every pair of atoms in the structure and this
includes the self vectors that is the x 1 y 1 z 1 on x 1 y 1 z 1 itself which will generate the
origin peak. So, for n atoms if there are n atoms in the unit cell 20 atoms in the unit cell
we get 20 square minus 20 peaks. So, all these peaks now originate from the single
origin and therefore, there is a severe overlap of these vectors.
So, every pair of atoms to form fives two vectors A to B and B to A which are equal and
opposite. Therefore, Patterson has inversion symmetry. So, Patterson is always
718
centrosymmetric. What is also important is the fact that all the phases, we can set them to
0 automatically forces inversion centre. So, that means, Patterson does not worry about
the phase factor. What is important in a Patterson is therefore, we all these space groups
which we have discussed now, now we will bring in back bring back the symmetry. See
the discussion in this course is on symmetry rather than anything else symmetry and
structure in the solid state.
So, we are looking at crystals and in the crystals we can now do the Patterson synthesis
by calculating the Patterson map which is a vector map, but this particular Patterson map
is always centrosymmetric whether the crystal is centrosymmetric or non-centric, the
Patterson map is always centrosymmetric that is one issue. The second issue is that all
translation involved components of our symmetry like 2, 1 screw axis, the glide planes;
they will lose their translation component when we do the Patterson synthesis.
So, if it is a P21\c the Patterson symmetry is 2 by m. So, this is already indicated in the
international tables of crystallography and when we go back to the previous two classes
and look at those slides which we have on the space group representation we also
indicate the Patterson symmetry. So, the Patterson symmetry for P2 1\c will be P21\m. So,
the Patterson will have only P21\m. On the other hand, if you have a c centred lattice let
us say C2\c then the Patterson will be C2\m. It will lose the translation component
associated with the screw axis and the glide planes, but the lattice information is
preserved.
So, the number of possible symmetries that Patterson represents can easily be calculated
and I wish I want you to calculate that. I would not tell you how many possible Patterson
space groups are possible in principle. I want you to find out. So, what is happening is
that the primitive lattices remain primitive, the centred lattices remains centred; that
means, the C lattice remains C lattice, F remain F, I remains I and so on and all the
translation components like in a structure of P2 12121 the Patterson symmetry becomes P
2\m 2\m 2\m which is Pmmm.
So, the symmetry reduces the neglects the translation components associated with the
operations like screw axis and glide planes. However, it retains the centring information;
the lattice information is retained. So, this is something which one should remember that
is what is written in this paragraph. So, whatever I described just now is written up here.
719
So, the Laue class the Patterson syntheses is the same as the Laue class for the diffraction
pattern. However, the space group retains the lattice type.
Now, each peak resulting from a vector between two atoms as a size which is
proportional to the atomic number ah; that means, to say that; that means, to say that if
we have let us say a Z i, and Z j then the two atoms the product of the atomic numbers
will be now proportional to the height peak height of the peak value. So, if there are two
carbon atoms the Patterson peak will be proportional to 36, 6 and 6. If there are there is a
bromine atom it will be proportional to 35.5 and we are now considering the vector
between the carbon atom and the bromine atom this will be 35.5 times 6.
So, you see that the peaks that appear in the Patterson function are very sensitive to the Z
value and therefore, if you have heavy atoms that is the logic of using that Z heavy
square divided by Z square as a thumb rule because it is possible now when we compute
the Patterson map. The peaks associated with the heavy atoms will stand out particularly
if your if both Z and Z j.
Now we are looking at the bromine to bromine vector, it will be not only be longer as far
as the bond distance is concerned because between bromine and bromine it will be very
long distance compared to the distance between carbon and carbon which is a covalent
distance of 1.54. Bromine and bromine may not have a covalent distance C to Br will
720
have which is longer than C to C, but bromine to bromine distance will be more than 3
and a half angstroms.
So, this vector therefore, will not only appear at 3 and a half angstroms from the origin it
will also come with a Z i, Z j proportional to 35.5 square which is a very large value.
And, so, it is therefore, possible to use the Patterson synthesis to identify the vectors that
occur between bromine and bromine the vector that occur between carbon and bromine
and so on. So, once you identify those vectors in principle, we can find out where these
atoms are. So, this is known as the heavy atom approach and Patterson synthesis is
basically used when we have a heavy atom along with a large number of small atoms.
So, the position of the heavy atom can be determined and when once we determine the
position of the heavy atom we can put the heavy atom into the expression for F
calculated and therefore, the phase that is associated with the bromine atom the heavy
atom is now considered to be the phase associated with the rest of the structure, even it is
it is an approximation, but you see the phase modification will be maximum in case of a
heavy atom than compared to the lighter atoms and so, it will be a dominant factor.
So, one once we use that information and then go and do the F not minus F c the more
delta F synthesis or the difference Fourier synthesis then we will get the rest of the
structure. So, heavy atom position first determined and using the heavy atom determined
position we compute the F calc and then do a difference Fourier between F naught minus
F calc. The heavy atom position disappears. The rest of the atoms one now will show up
ok.
So, if you have for example, looking for a particular person who is thin among a large
number of a few heavy people you remove the heavy person, then you can identify the
thin person. So, just get rid of the heavy person and then you get the thin person. So, that
is the idea of locating the rest of the atoms and when we locate all these atoms in
principle we will get to the structure determination.
So, if the relay unit cell contains a relatively small number that is what I have written
here among a majority of lighter ones the peaks corresponding to the vectors between
pairs of these heavy atoms. Remember, we do not get the position of the heavy atom, we
get the vector distance between the atoms will be large and will stand out clearly from
general unresolved background level of the smaller peaks. So, it is possible therefore, to
721
use the Patterson synthesis in case you have a heavy atom and in case the heavy atom is
following our thumb rule. The presence of the heavy atom in the structure follows the
some thumb rule. It can be more than one heavy atom, it is not necessary just to have one
heavy atom, but this ratio should be reasonably satisfied.
However, very many complex structures have been determined by Patterson synthesis
because the development of the next method which we are going to discuss later in the
class is the so called direct methods. By definition itself direct methods mean that we
directly determine the phases. So, until the development of that took place and it matured
into a very useful and straightforward computable methodology Patterson was the only
preferred method. So, the preferred Patterson method therefore, is the one which in fact,
sort of took care of everything.
So, the preferred Patterson method was determining many many structures. So, in initial
days what people did particularly the organic chemists did was to have a heavy atom
derivative of that compound which they want to get this structure of. So, suppose they
make a structure and they want to determine the structure and there is no heavy atom
what they do is for suppose there is a phenyl ring and a nitro group attached to it. To the
nitro group they will attach a bromine atom, so that the structure of the molecule can be
determined.
Very little did they realize at that particular time, that the two compounds will be very
different in their properties. If one is looking for a property the heavy atom the
substitution is not a very good idea because it is changing the nature of the compound,
but you will get the structure, alright. So, initial days of structure determination if you
look at literature people put heavy atom derivatives and determine the heavy atom
position and thereby solve the structure. And, then they assume that without the heavy
atom also the structure may be very nearly the same.
So, many of the early days structures therefore, use the Patterson methodology to
determine the structure of heavy atom derivatives of organic compounds that is there
plenty in literature, but then one once they realize that the property which you are
looking for is completely disappearing there is no point in introducing a heavy atom
derivative. They had to live with light atoms. By that time the direct methods evolved
and direct methods evolved into such a situation that, it could be easily programmed in.
722
So, the development of high speed computers and modern technology for data collection
enable the takeover of structure determination procedure by the direct methods over and
over the Patterson function.
So, the Patterson methods are still used in various circumstances. In fact, they are very
much useful in determining the structure of proteins as we will discuss later on very
briefly because we are now coming towards the end of the course. So, we cannot now go
into the detail of that. However, we will discuss it briefly other than that direct methods
is the most preferred one because it is easy to program and what all you have to do now
in the modern day structure determination protocol is to present enter button after getting
these structures after getting the observed structure factors on an absolute scale and the
rest of it is done by the machine using the direct methods.
So, continuing with the Patterson we see that very complicated structures of course, were
determined. In fact, the structure of vitamin B 12 which got the Nobel Prize to Dorothy
Hodgkin was solved by the by the location of the cobalt atom. You see that my thumb
rule will not be valid here anymore because there are so many other atoms, but still the
structure could be done by identifying the cobalt position. In fact, this structure was done
by a group of people who are working under Dorothy Hodgkins and I take pride in
announcing that one of them happened to be an Indian and that Indian happens to be
professor Venkateshan with whom I did my PhD. So, there is a connection.
723
So, this is one of the very difficult structures which was done in those days and the
admittedly with lot of difficulty no it was not a straight forward structure solution and the
phases from the cobalt atom were used. The Z square ratio I have given here just for your
wonder the Z square ratio was 0.17. We said it should be nearly 1, but if you take the Z
square ratio here it is 0.17 and that shows the effort put by that group in order to
determine the structure of vitamin B 12. But, no matter what this was a very crucial
structure determined in those days, one of the biggest structures of those days which got
the Nobel Prize to Dorothy Hodgkins who did wonderful things later on, but then this is
where things worked out the way.
There is one little issue with Patterson and that issue is where is the symmetry? We have
been talking about symmetry and structure in 3-dimensions unfold and then we are
looking at projections we are looking at possibilities of translational periodicity is getting
involved and so on, but what we said was done in Patterson the symmetry associated
with to the 21 screw axis becomes twofold the glide planes become mirror. So, where did
where did the information of symmetry which we have in all these the crystal systems go
away? It did not go away. It stayed in the Patterson and that is known as a Harker line
and planes.
So, the information that is contained in Harker lines and planes will tell us the space
group information whether they have the translation components associated with the
space group which we have. So, even though the Patterson now has reduced the
symmetry the information regarding the translation involved components which are
present with that particular space group is incorporated into what we call as Harker lines
and planes.
So, Harker of course, also should have walked away with a Nobel Prize. He did so many
other things. In fact, he is the one who actually initiated work in the direct methods the.
So, called Harker Kasper inequalities actually heralded the protocol which was
eventually which eventually became direct methods, that is something which is
remarkable. Harker also contributed to. So, many other things and the one of the things
which Harker contributed was the structural determination of proteins.
So, I think we will discuss these issues with the Harker lines planes and contribution of
Harker because it so happens that I overlapped with Harker, when I was a postdoc in
724
SUNY Buffalo and he unless I talk about him no course on symmetry and structure is
complete because the fact that Patterson this took away the information on the symmetry.
However, the fact that it retain the symmetry was discovered by Harker and that helped
in solving many many structures.
Suppose, a heavy atom derivative gets into a P2 12121 structure and then the 212121
information disappeared in the Patterson. You determine the position of the heavy atom
all right, but how do you determine the rest of the atoms? Now, where is that information
that is associated with the heavy atom position? In fact, the heavy atom position the
location of the heavy atom position is contained in the so called Harker lines and planes.
In fact, we use these to determine the structure and that is why this contribution from
Harker is enormously significant. So, we can say that you know there are so many people
who are listed under those who did not get the Nobel Prize. So, Harker is one such.
725
Lecture – 53
Patterson Method 1
The symmetry that is required in a Patterson synthesis. As we have just seen that the
Patterson is always centrosymmetric, it loses all the information on the translational
components, which are associated with the space group. So, 2 1 screws is off, the glide
planes are off, so the space group now will be the one which trips of all these
translational components.
So, it is space group is 212121 as we already mentioned is the corresponding Patterson

symmetry will be P2/m2/m2/m abbreviated as Pmmm. But as well as the centering is
concerned, we see that the centering is still intact that means, if there is a C centered
lattice C2221, it is corresponding Patterson symmetry will be C2/m2/m2/m. So, the seen
centering information is not lost with respect to the lattice that is with respect to the
Bravais lattice. Bravais lattice information is maintained in the Patterson, the
translational components symmetry is lost.
However, Harker came up with a very nice suggestion that is to examine this Patterson
maps very very carefully. When the Patterson maps are examined very carefully and one
726
looks at projection as well as the axial regions of the vector map, since the vectors are
redefining, now u, v, w is defined as x 1 minus x 2, y 1 minus y 2, z 1 minus z 2 the inter
atomic vectors.
The property of the vectors will be in such a way that they will also follow the
symmetry, because the Patterson as retained the symmetry. The only issue is now is there
any way, we can find out whether the system was associated with a 2-fold screw or a 2-
fold axis. So, if one looks at the projections of these Patterson map, so we can do a three-
dimensional Patterson map and look at the projections of that along x, y and z.
So, if we put now a situation where we have the projection down the x axis, 0 v w, if one
looks at 0 v w, the 2-fold axis is parallel to a, b and c. So, as a result we end up with what
is known as a Harker plane. So, the Harker plane will have information about the 2-fold
axis, there will be a peak corresponding to v and w in the projection down 0 that means,
we now look at this plane 0 v w in the vector map, it will give us the solution for v and w
for every atom in the Patterson.
And if there is a heavy atom, the u w coordinates of the heavy atom or embedded in the 0
v w, if we are looking down looking at a case where we have a 2-fold axis which is
parallel to a. That means if the crystal system is P2 let us say and the 2-fold axis is now
parallel to a that means, we have now looking at the projection 0 v w. So, this is the
vector map.
So, in the vector map v is y 1 minus y 2, w is z 1 minus z 2, but in the vector map if there
is a 2-fold axis in the system. Let us take the example of P222 in orthorhombic system,
where we have 2-fold axis parallel to a, parallel to b and as well as parallel to c. In that
case, we have three projections 0 v w; u 0 w; u v 0. So, all these three projections will
give us the values of u and w corresponding to each inter atomic vector.
So, suppose there is a heavy atom in the structure the heavy atom in the structure will
give rise to a very heavy peak and that particular heavy peak will correspond to v and w.
And therefore, the when we do this projection of the Patterson map and then plot it up,
we will see that there is a particular value of v and a particular of value of w with now
satisfies the v w that should correspond to the heavy atom position.
727
So, we therefore get the solution to the two coordinates. So, we will get therefore v is y 1
minus y 2, w is z 1 minus z 2. So, we get information on not just the z 1 and z 2, we will
also get information on y 1 and y 2 and this will be uniquely found. So, the projection
plots will carry the symmetry information. So, if there is a 2-fold symmetry, the 2-fold
symmetry information even though last on the main Patterson, it will be present on the 2-
fold screw position corresponding to parallel to a, parallel to b, parallel to c we have
these positions.
So, if we now look at the section half, you will get solutions for u and w. So, we will see
that with examples as we go further, but essentially they Harker planes will now
correspond to positions associated with either 2-fold or 2-fold axis or any other axis,
which are now with respect to the Patterson, they have a projection possibility that
means, 3–fold axis is possible, 4–fold axis is possible and so on and along with the
screws. So, where did I have given only the 2-fold example here for our; for making it
simple, so what is done is we if we now have this system P222 here, we look at these
three projections and we get solutions for u w, u v w, u w and u v. So, if the system is
P222, then there are two 2-fold axis perpendicular to each other just as an example.
Then we have the coordinate the solution for v and w for the heavy atom position, it will
be very large value u and w for the heavy atom position and u and v for the heavy atom
position. So, you see that between the three projections, between these two u is common,
between these two w is common and between 1 and 3 v is common. So, we get a unique
solution for u, v, w. One once we have u, v, w; then it is in principal possible to get the
coordinates of the heavy atom.
So, the heavy atom presence will be easily identified by looking at these projections and
also the identity associated with the symmetry information. So, even though symmetry is
not there in the main Patterson, the overall Patterson; the projection reflections carry and
information associated with the translational component. So, if there is a 2-fold, we have
to examine the sections passing through x, y and z at half, half, half.
So, take for example 212121 you will uniquely solve for u, v and w, because you look at
these three now projections and you get solutions for u, v and w. So, if there is a heavy
atom, then the heavy atom position in terms of u, v, w vectors is possible to be found. So,
therefore, this gives the examination of Harker planes therefore, gives us an opportunity
728
to identify, where the heavy atoms are. And also to associate the symmetry that is
associated with that particular space group. So, if the space group is P2 12121 in we
compute the Patterson. The Patterson we know is P2/m2/m2/m, but the three projections
associated with the x, y and z directions which is parallel to a, b and c will tell us about
the possibility of locating the heavy atom positions to start with.
And secondly, it will also tell us what is the possible translational component that is
associated with that particular symmetry operation. So, if we take 2 12121 it is quite
uniquely all three 21 axis can be determined by looking at the three projections and
looking at a high values of the u w, u w and u v which appear in those projection plots.
The similar argument goes for the Harker lines. So, in case of the mirror and the glide
planes, you will instead of getting Harker planes, you get a Harker lines.
So, there will be lines of symmetry about which you will get information about for
example, if there is a mirror plane, perpendicular to a, perpendicular to b or
perpendicular to c; the corresponding Harker lines will be u 0 0, 0 v 0, 0 0 w. So, you
have to look along the axis u v and w. And if there is an a glide, it will now have to have
a half component here. So, we look at half v w, if the a glide is parallel to perpendicular
to b; and if the a glide is perpendicular to c, we get half 0 w and so on.
So, basically what they take home lesson from this table is that we compute a Patterson,
and using the Patterson, we can now determine the space group as well as the heavy
atom position if we make use of Harker lines and planes. In fact, the Harker idea we
came up for the location of the heavy atom in a totally different context. It came up with
the idea that if we have a protein solution and in that protein solution; we now introduce
a heavy atom containing die. So, like for example, we have a green sweater and blue
sweater. Essentially, you put dice into these sweaters, sweater is a fibrous protein, it is
made up of material, which is a protein.
There is no change in the structure of the protein, the wool remains wool, but what
happens is the colour changes, because your put a dye; that means, the dye has been
incorporated into the silk structure in the into the wool structure, which is the protein. In
such a way that the actual structure and associated properties of the material has not
changed, because of the fact that in a protein we have lots of empty space, which is
occupied by solvent water and so on, it is possible to introduce dies without changing the
729
overall configuration or confirmation of the protein molecule. So, this process where we
can put a dye into the system and depending upon the nature of the dye, the nature of the
structure of the dye and so on, it may occupy different positions around the protein and
still leave the protein as such.
Suppose the dye contents of heavy element. The heavy element will dye will now also be
going into the protein; and that particular heavy element therefore, can be located by
Patterson synthesis. The process, which was develop by Harker is known as isomorphous
replacement method. And using that one can find out the structure of the protein, we can
solve the structure of the protein is not within the scope of our current discussion. But
essentially one should know that it is possible to determine the structure using the heavy
atom approach of a big large protein. And their of course, the issue of z square ratio will
be very ominously low. So, still one can determine the structure using this isomorphous
replacement.
So, essentially depending upon the different kinds of heavy atoms and their derivatives,
which are now going into the protein structure. So, what you do is you take the dye
solution, soak the protein in the dye solution, then you will dye will get incorporated. So,
the colour will change, but the crystals which come out of this now will contain the
heavy element as well along with the structure of that small molecule into which we have
put the protein in.
Similarly, depending upon the size and nature of the heavy atom and the surroundings,
the position at which the heavy atom goes will also change. So, this is known as multiple
isomorphous replacement, because you will have different positions for the atoms to
occupy within the framework of the protein structure. In such a way that the crystal
structure of the protein still remains the same that means, the space group remains the
same, there is no alteration in any other expect except that we have now different
positions of the heavy atoms that can be used to reconstruct the protein structure. And
this is known as the multiple isomorphous replacement that is again due to Harker.
So, Harker as contributed immensely to the Patterson synthesis do the normalize this
isomorphous replacement method and MIR which is multiple isomorphous replacement
method, which is very commonly used by protein cryptographers around the world now.
Programming and things like that also become easier to determine the heavy atom
730
position following which one can determine the structure of the protein. The other aspect
which Harker has contributed to we will discuss after when we go to the direct methods.
There he developed the required inequalities to start with which eventually became of
course equalities of probabilities and those expressions where utilized in order to develop
the so called direct method by means which we can solve the structure.
So, essentially we have now a Patterson calculatable probably the possible Patterson can
be calculated. The Patterson will be divide of the face problem. If there is a heavy atom,
it is possible to use the Patterson to locate the heavy atom. And the symmetry that is
associated with the space group is sort of (Refer Time: 14:28) into these Harker lines and
planes through, which we can extract the information not only about the position of the
heavy atom very straightforward and easy way, but also identify the translational
components that are associated with this particular space group.
So, the symmetry again as you see plays a major role in controlling how the Patterson
vectors appear, and how they get aligned in Harker lines or in Harker planes with respect
to the heavy atom position determination. These now serve as guidelines to solve the
crystal structure to get the heavy atom and then one once we have the heavy atom
difference Fourier techniques will allow us to solve the structure right.
So, we will take some examples. So, the case of the first one is P-1, I will take the case
of P-1equivalent points x, y, z, minus x, minus y, minus z. So, this is the position of the
731
atom let us say in P-1. Then what happens we, we make a matrix of this kind on the right
side where we take x, y, z combine it with itself, which is the self vector and that is 0, 0,
0. Then take the second vector and combine it with this one which gives us minus 2 x,
minus 2 y, minus 2 z; and with x, y, x minus x, minus y minus z or the x bar, y bar, z bar,
we get 2 x, 2 y, 2 z and 0, 0, 0.
So, that means, if you compute the Patterson now in the Patterson diagram, we will get
suppose there is a heavy atom in the structure, the heavy atom is associated let us see
with x, y, z and x bar, y bar, z bar is automatically there in the space group P-1. The x, y,
z position can be determined, because we get two vectors one at minus 2 x minus, 2 y
minus, 2 z, and the other at 2 x, 2 y, 2 z. So, since we get two vectors one at 2 x, 2 y, 2 z,
other at minus 2 x, minus 2 y, minus 2 z we could look for a solution for x, y, z.
So, in principle we are able to find out the position of the heavy atom. So, we have these
two corresponding peaks appearing one at 2 x, 2 y, 2 z and the other at minus 2 x, minus
2 y, minus 2 z. So, use this two to find out x, y, z so that means, you divide this by 2 that
by 2 divide that by 2. And this one again divide by 2 divide by 2 divide by 2, but do a x
bar, y bar, z bar to go over to x, y, z then you will get a unique solution for the heavy
atom.
So, p 1 bar system is fairly straight forward. So, we can use this in this right hand side 2
by 2 matrix in this particular case the equivalent points and the 2 by 2 matrix is
developed this way. Using this matrix you will get solution for x, y and z. So, we find the
heavy atom position. The discussion now on the Patterson is limited only to the heavy
atom method, because we are not going into the possibility of using Patterson as a
general technique to determine the structure which can be done, but that is again for a
way into the advance crystallographic course which we are not now going through.
732
So, the next space group we can consider is P2 1, where we have equivalent points are x,
y, z; x bar half plus y z bar. Remember that there is a P2 1 here. So, when once you see
P21 as the space group, you expect a Harker line. So, you expect a Harker line or a
Harker section, which one you expect, you expect a Harker section. Harker lines comes
for glide planes and mirror and Harker section comes for the screw axis. So, you expect a
Harker plane. So, you see here you get a solution to the heavy atom, there by taking
vector between general positions, x, y, z, x bar half plus y z bar I am writing general
position because P21 has a special position 2.
So, general positions x, y, z, minus x, half plus y, minus z so, if you do the same
operation as we did before with P-1, we will get two sets of vectors 0, 0, 0. And this
gives me minus 2 x, half, minus 2 z, notice the half here ok and then you get 2 x, half, 2
z, because y is equal to half in both these examples, you get solution to x and z. So, in a
space group like P21, y can be anywhere, y is arbitrary, because it is along a 2-fold axis.
So, we can fix your y coordinate everywhere along the 21 direction. So, 2 x and 2 z will
solve for x and z. You easily identify now the Harker sections which are this as well as
that corresponding to y equals half.
So, you have now therefore in the projection, which you do down the y axis at half you
see that you will get a solution corresponding to 2 x, 2 z, minus 2 x, minus 2 z. So, you
plot now the value of v equals half ok. And then you will see a centrosymmetric related
733
two peaks which are very heavy corresponding to the heavy atom. So, those will be 2 x,
2 z and minus 2 x, minus 2 z, that is why say centro symmetrically in the plane. So, the
plane which you are considering now is half which tells you automatically that there is a
21 screw axis that is present in the system. So, this is the Harker section.
What you can do also is one once you have found the solutions for x, z and then arbitrary
y, y you can fix anywhere in P21. So, let us say we fix the y value at 0, then you can look
for 2 x, 2 y, 2 z, there will be a p corresponding to 2 x, 2 y, 2 z for the solutions which
you have arrived for x and z. So, here we get solution for x and z, we find the
coordinates of x and z. The coordinate of y can be fixed arbitrarily, because there is no a
region fixation along the 21 screw axis, where it can be anywhere all along the line of the
screw axis, it can still have the screw property.
So, you fix decide to fix where the y should be conveniently and therefore, we have this
option. So, you see the importance of the Harker section which is developing here with v
equals half. We straightaway see the solution 2 x and z, we get 2 x, 2 z and therefore, we
get x and z. So, we have the x and z coordinates of the heavy atom, y we are fixing. So,
we get the corresponding 2 y with depending on where we fix the y value, and therefore,
we get the position of the heavy atom.
(Refer Slide Time: 2 1:10)
So, P21is a fairly straight forward case, we now go to P21/c which is the most common
system we come across. So, here we have four equivalent points x, y, z, x bar, y bar, z
734
bar, the x, half, minus y, half plus z, and x, half plus y, half minus z. So, we have we
have to now do a Patterson between them. So, x minus x is 2 x, y minus y is 2 y, z minus
z is 2 z, which comes here. So, you write the four equivalent points, you write the four
equivalent points and make the matrix of this. So, this matrix will give you 0, 0, 0 here;
0, 0, 0 here; 0, 0 along the diagonal it is all 0’s. And then you get a symmetric this
position of 2 x, 2 y, it is 2 z, 0 half plus y half, 0 half minus y half, 2 x half.
So, corresponding to the 2 1 screw axis yeah. So, we now are looking at the vectors
which come from symmetry operations of P21/c. So, if one looks at P21/c, we have four
equivalent points. These two equivalent points are generated by the center of symmetry.
These two equivalent points are generated by the center of symmetry, These to these is a
glide operation, then these to these is also incorporating the center of symmetry
equivalent glide operation. So, there is a relationship between these four, which tells us
the presence of 2 1 as well as the c glide perpendicular to the 2 1 axis.
How does it reflect on the Patterson vectors, this is where the beauty of Harker sections
and Harker lines come up, because using this we get a unique solution for the heavy
atom position in P21/c. So, we compute the Patterson, Patterson of course, as we know
will going to be P2/m. So, it is not going to give us much information on the translational
component as we know. However, if you look at this matrix; examine the matrix where
we go now write the four equivalent points here and four equivalent points there and then
make the combinations of those.
So, if you take x, y, z and itself, it is 0, 0, 0; you take x, y, z to minus x etcetera we get
minus 2 x, minus 2 y, minus 2 z. You see that this position to this position, the it is
related by a center of symmetry, the corresponding vectors are also centrosymmetrically
related, so minus 2 x, minus 2 y, minus 2 z will generated 2 x, 2 y, 2 z. So, these two are
now centrosymmetrically related. So, we get solutions for x, y and z by examining the
come to two components 2 x, 2 y, 2 z and 2 x, 2 y, 2 z minus of the two vector. So, we
look at this vector end and this vector end and see whether the peaks that are associated
with this are corresponding to the product of the heavy atom.
So, if the heavy atom is bromine, these two vectors should be very close to 35.5 square
after the scaling factor is taken into account. It will not be exact, but it will have a very
high value compared to the rest of the vectors, you find in the Patterson space. So, you
735
pick up these two. So, the way you practically do Patterson is you compute the Patterson
of a heavy atom structure and then arrange the vectors in a descending order and see
whether they follow any kind of symmetry.
So, if the follow symmetry, then you identify the symmetry and also try to solve for the
x, y, z from the set of vectors. So, we have therefore, here 2 x, 2 y, 2 z, 2 x, 2 y, 2 z and
we get a x, y, z keep it aside. And now take the third one, which will now give us with
respect to this 0, half plus y, half; now this is a glide operation and therefore, we have to
look here under the glide operation you get 0, half minus 2 y, half. So, this is a Harker
line and this is also a Harker line.
So, if we look at the Harker line, you get a unique solution for y. This should be half plus
2 y here; there is a mistake, it should be half plus 2 y. So, you get half plus 2 y, half
minus 2 y, you get a solution for y; you have solution for x, y and z already here,
compare that y with this y, it should be the same. So, you therefore, now conclude that
the value of y is for the heavy atom system has been completely determined. You go to
the last one which is 2 x, half, half plus 2 z, you will find this minus 2 x, half, half minus
2 z, which is now the previous one was a Harker line this is a Harker section. This is a
Harker section, because we are looking at the c glide operation, this is a Harker section,
because we are looking at the two one screw operation Harker line this Harker line is due
2 1 screw operation, this Harker section is due to the c glide operation.
So, we see all the symmetries which are coming here. So, you do not have to worry that
Patterson is not giving you any cemetery information. The symmetry information that is
coming up is essentially coming from the equivalent points, which are present in the
crystal structure and they are depicted in terms of their vector relationships in the
Patterson map. So that way we can identify these matrix set up this matrix get all these
coordinates and we can uniquely solve for x, y and z. So, we get the position of x, y, z
which is corresponding to the heavy atom in a very straightforward way if we analyze
these sections, the Harker section as well as the Harker planes, along with that look at the
vectors corresponding to 2 x, 2 y, 2 z.
So, you would see that there is no other set of vectors which will be present here either
they are related by center of sub symmetry or by the operation of the 2-fold like here. For
example, if you look at these 2 x bar, y bar, z bar and the third one, you see that it is 2 x,
736
2 y, 2 z and here it is minus 2 x, 2 y, minus 2 z. So, that is the operation of a 2-fold about
the y direction. And here it is half plus y therefore, it adds on a half value to the y 2 half
plus 2 y it should be.
So, essentially we get a unique solution in case of 21/c. There is one other example in all
these cases one thing which we have talking about is that we are looking at only one
heavy atom. It is not necessary that we should have only one heavy atom, we can have
more than one heavy atom. I have given this table and I have given the Patterson vectors,
which between general positions in Pbca. You can verify yourself that you can put this
into a kind of a matrix, which we put in the earlier two cases. I am not going to put that.
You can this is an exercise, which one goes through we will get a hang of how to set up
those matrices, so which are essentially required to identify and solve the vector position.
737
The for the 212121 anyway I have given this matrix. So, Pbca you can find out what
should be the matrix. So, it means you have to reorganize this in the form which is
coming up here. So, you write a eight equivalent points and then generated in case of
Pbca. So, the size of the matrix keep going up, but these solutions become more and
more unique. So, you get a solution for example, from looking at the peak 2 x, 2 y, 2 z,
you get a solution from looking at 2 x, half minus 2 y, 0. We will also get information
from 0, 0, half plus z and so on.
So, essentially you get x, y, z from various possibilities and the common x, y, z, which
comes up from a looking at all these vectors will be the one which you identify as the
heavy atom position. Obviously, all these peaks will have to stand out, so these are
therefore, the highest peaks in the Patterson. So, it is becomes very easy computationally,
because one once you compute the Patterson you rearrange the peaks in a descending
order. And look at the possibility of assigning these peaks to the various where the peaks
appear and when you see where the peaks appear, you will get the various Harker
sections and Harker lines. So, essentially you can solve the heavy atom position fairly
straight forward way.
738
So, this is a gist of overall operation of the Patterson synthesis. And I know I have
restricted through, but here is a practical example to make you convinced and how you
get exactly the heavy atom position.
739
Lecture - 54
Patterson Method 2
So, we take it as a practical example. So, in a Patterson data, this is the structure of
locating the bromine atom in methyl micromerol bromoacetate. This is a compound
whose structure has been determined with the space group P212121. So, it was located
from three Harker sections, each calculated over only it is unique area; that means we
calculated the Patterson’s maps. We, we did not have to calculate the entire three-
dimensional Patterson map, because we have already determined the space group earlier
on by looking at the distribution of the hkl reflections when once we had the data. So, the
space group is P212121. And so we know that there are three Harker sections.
So, we look at the three Harker sections. so we look at the section at u equals 48 by 96.
What is 48 by 96 is half? So, we look at u half, v is now going from 0 to 50 up to 100, it
just goes 50 slash 100 and w goes also from 0 to 50 slash 100. So, we look at this
particular section of the Patterson, we compute only this particular section of the come of
the Patterson. We get a peak at 48 by 96, 11 by 100, and 21 by 100. I think I should
illustrate that with a small diagram.
740
How we do this? What you do is, you have a Patterson section. So, you take u equals
half. So, this is now u 48 by 96 ok. We are taken the top off at u equals 48 by 96. So, this
is the origin here. You now go from let us say this is the v direction, and this is the w
direction. So, you mark this as 0, you mark this 0 0. So, we now go along the v direction,
we go up to half, which represents 50 by 100, you go here again to half which represents
50 by 100.
Now, you compute the Patterson value at every point one let us say, just to give you an
example or every 0.05. So, you get several points along the w direction, you get several
points along the v direction. And then find out the peak positions, the value of the
Patterson. The value of the Patterson you already know is varying as with respect to this
as zi zj right that product of zi zj.
So, if it is a heavy atom, in this case bromine, it will vary as 35.5 square. So, you will
start getting the peaks. The way you will get the peaks will be like this. You draw the
contour diagram. This is the contour diagram of a Patterson map. So, you take let us say
for example, here at this point you got a very high value let us say you got some 90. And
then adjacent points you got 85 and then you got something like 70 here, then something
like 40 there, and something like 70 there ok.
And then all the other points also will come like that. The various values corresponding
to the value of the w as well as that of v, we are now looking at u equals half that is 48
by 96 section. And you are going from 0 to half here and 0 to half here. So, what you do
is now draw the contour. The way you draw the contour is I am going to draw the
contour of 80 ok. So, 80 is about this it is 85; so 85. So, I will draw this contour which is
80. And 80, 40 is midpoint, 80 will come beyond this 70 and this goes beyond the 70. So,
you get a contour like that.
Then you draw the contour of 85. So, it will go something like that; and contour
belonging to 90, it will go like that. So, this way we plot the contours. Taking the points
of the adjacent, so the next point will be 20 let us say and there is something like 80 here.
That will be another peak in the u v section corresponding to some other atom let say.
So, we take the highest peak, the highest peak now, is at 48 by 96, 11 by 100 and 21 by
100 in this map. So, we what we have done is 0 to half is expressed as 50, 0 to 50 slash
100, that means, half right, 50 by 100 is half, 50 by 100 is half. So, we will calculate the
741
corresponding value of y and z as 11 by 100, 21 by 100. This is only a computational
ease, so that we plot this section. So, we have plotted the section u equals half, and we
have taken the top of the u and v and w is spread out. So, you have a sheet, where you
have u and w values you mark 0.1, 0.2, 0.3 and so on. At every point you calculate what
is the Patterson value and that is proportional to z i zj, so that value you put it here. And
then you draw the contours depending upon the strength or the weight associated with
that position as far as peak is concerned.
So, you draw what are known as contours. And these contours will tell us the location of
the u v and w, and that happens to be 11 by 100 and 21 by 100. So, it is 11 by 100 along
the v direction, and 21 by 100 along the w direction, so that is the peak. So, you pick up
that peak. You assign this peak as half, half minus 2y, 2z that means this is now a Harker
section. And so you solve for y and z. Now, what will be the value of y and z? So, this is
v is what, v is 2y, and w is 2z. So, you get the value of y as 19 and a half by 100 and 10
half way. So, you get the value corresponding to y as well as z.
Then you go to the section v similarly, you go to the next one which is the section along
the y direction, which is v equals half and plot u and w in this following way. And pick
up the peak. The peak found is at 34/96, 50/100 and 29/100, this is corresponding to 2x
half, half minus 2z. See here the peak was found at what 48 by 96 is half ok, and 11 by
100 is half minus 2 y. Say what you got you may be wondering, what is this going on,
because you said 11 by 100 is the peak 11 by 100 corresponds to 2y. So, you have to take
half minus 2y, so half minus 2y will be 19.5 100 when you solve for y.
So when you solve for z, you again take the value 2z. So, 2z is 20y by 100, and
therefore, this will be 10.5 by 100, half of that. So, 2z is 21 by 100 that is half of that and
this is half minus 2y; so 11 by 100 now gives you 19.5. So, the peaks corresponding to y
and z are at 19.5 by 100, 10.5 by 100. You get a solution in the second section at two
axis. Now, you get a solution for x. Here we got a solution for y and z. So, here we get a
value for x. And we also get a value for half minus 2z from this position, and the value
for half minus 2z is now 29 by 100, which is a requirement coming from up they up here,
because z value is 10.5 by 100. So, half minus 2z if you calculate, it will be 29 by 100.
So, there is a peak corresponding to that as well.
742
So, similarly you work out the w section. And therefore, you get the value 14/96, 39/100,
50/100 as the 2x, 2y, 2z are the solution as a half plus minus 2x. half plus minus 2y and
half, and then you get the values corresponding to that. So, what I suggest is you go
through this table. And then you consider the first one as a Harker section corresponding
to x equals half. The second one is the value corresponding to y equals half. The third is
corresponding to z equals half. And you have the corresponding values of y and z in
every case in this case y and z sorry in this case, x and z and in the third case it is x and
y.
So, effectively two of them are common between these. In other words, you have the
value of 2z here and a half minus 2z here. So, z can be uniquely determined between
these two. Similarly, the value of x can be determined between these two, and the value
of y can be determined between 1 and 3. So, you get the value of x, y, z, which is unique.
So, the complete set of symmetry, then you get you see remember this is a P2 12121
system. And therefore, there will be not just one bromine atom; we will have 4 bromine
atoms. And therefore, you have to find the positions of 4 bromine atoms. And those four
bromine atoms should be related by the equivalent point relationships.
So, when you get this solution as 17 by 96, 19.5 and 10.5. 10.5 came from here; and 19.5
came from there; and 17/96 came from where? 17/96 came from here the value of x. So,
then we had the second position of the bromine. So, when once you have this solution
coming from the two three peaks which we have located, you can now generate the
equivalent points in the space group, and look for peaks corresponding to those values.
These three peaks will correspond to these values, that means you have let us leave out
the lower part, then we will keep the 96 or 100 let us say for arguments sake.
So, we have a bromine sitting at 0.17, 0.195, 0.105, then you use half minus x minus y
half plus z half, then generate these points which is coming now to be 0.31, minus 19.5
and minus 60.5, minus 0.605 sorry. So, these values now will correspond to each one of
these bromine, and these bromines are symmetry related bromines, these only one
bromine in the structure but you get four bromines in the entire unit cells.
So, this corresponds therefore to the 4 bromine atoms in the unit cell. So, you can
determine all their positions. So, Harker section utilization along with the symmetry that
is available in the Patterson map will allow us to find the positions of the heavy atoms
743
fairly uniquely. And particularly in systems which offer unique finding of the position of
the atom. I hope I have made it clear, but just for the sake of revision, let us go through it
once.
Let us take the case of P212121 and that is where we have taken the example right. So,
P212121 has four equivalent points. The equivalent points are x, y, z, half minus x, this
one and this one before recalling. So, you generate the matrix. So, when you generate the
matrix, you see that you get both Harker is all three Harker sections will appear
corresponding to the 21 axis. So, with respect to x, there is a Harker section coming at x
equals to half; with respect to y, there is a Harker section coming y equals half and z
equals half.
So, if you take the Harker section coming with respect to x equals half, what are the
equivalent points we find in that particular Harker section? One is this half, half plus 2y,
minus z, another is let us me point a put an x here, half where else is half here; half, half
minus 2y, 2z; then half half minus 2y, 2z; and then half, half plus 2 y, 2 z. So, these four
will correspond to the Harker section 1. So, in principle, these are the 4 bromine atoms
related by equivalent points So, you see that each one of them now in separate columns,
each separate column is under each separate equivalent point. So, you get the Harker
sections corresponding to x equals half in these 4 marked as x.
744
Now, what about y? y will come for example here. I am marking it as y and then this is
half y equals half then y equals half is here, then y equals half is there, and y equals half
is there. So, once again you see it belongs to 4 equivalent points.
Now, the question comes up which x corresponds to which y and which y corresponds to
which z that can be easily sorted out when once we now find out what are the z values.
So, these z values are here; this is a z value, this is a z value, because z equals half, and z
equals half, and z equals half. Apart from them as the requirement of the Patterson, you
get 0 0 0 along the diagonals. So, therefore, you have x, y, z, all three located with
respect to all 4 bromine ok.
So, now you can go to the next slide and see these four are the four bromine which we
have located based on these three Harker sections. So, I want you to spend some time off
you on your own rather than me going and explaining it, because these two are very
clearly given slides through which you can now go back and forth and check up the
coordinates, which we have got with respect to the Harker sections and Harker lines. So,
P212121is going to give us a unique solution. What we want is only 1 bromine position,
the rest of them we can generate by symmetry operation. This is just to see that Patterson
will contain that symmetry information, we have illustrated this particular example.
Now, that is about all I want to say about Patterson method 6. So, the there is lot of
things which one can say about Patterson methods, but there is no time which is left now
we are coming towards the end of the course all together. So, I have to go to the next
methodology, which is the direct methods. Once again at this particular point, let me
stress that I am not going through the entire direct methods protocol. In fact, I am going
to really rush through the direct methods, because of lack of time, but I will give you the
basic physical meaning associated with direct methods.
And this is extremely important, because what you see when you do a structural
determination is when you go to the stage where you have to determine the phases or
solid structure, you essentially put up a inter mark and they were there it is written either
Patterson of direct methods, you just say direct methods and put the enter button. So, the
enter button now will take you to the direct methods. After a few minutes, you will get
the solution to the problem. And you see the map the so called e-map we will discuss
what is e the e-map which will tell you the positions of the atoms which you are rotating
745
now as point atoms. So, the all the electron density associated with that atom will be
there, but it is located as a point atom.
So, what is the logic of direct methods? These two this particular entry here will tell you
what we are doing. Suppose, we take modulus of h and do what is known as a Fourier
transform on just the modulus of h, it is known as amplitude synthesis. If you take
exponential, I filed and use the Fourier transform, it will be the phase synthesis. So, this
will do the amplitude synthesis, this will be do the phase synthesis, but we want to do
these two together, and then do it on the F of h we get a Fourier transform which will
give us the electron density.
A very beautiful experiment was done by a Ramachandran and his associates in this case.
What was done was we took the structure of a compound, whose positions were known.
And then they use the modulus of F of h, and did then amplitude synthesis of compound
A. They also did the phase synthesis, because they know the structure of A, they did the
face synthesis using the A, and then kept it aside. So, we have the amplitude synthesis,
the phase synthesis and of course, the electron density of compound A. For compound B,
what they did was to take the amplitude of compound B and put it into the phase of A.
This was done in 1963 fantastic experiment. G N Ramachandran was a genius. I think he
should have got a Nobel prize in India, you know he was never given. He also did the phi
site map as most of the people know for the protein structures; he did the Fourier
746
synthesis of different kinds of Fourier analysis. One of them is this and he also did
several other things in crystallography, which included some work in anomalous
dispersion and so on.
So, he did this exercise. So, you have a compound A, and you have the amplitude as well
as phase, and then you calculated electron density. You now have the compound B,
compound B only the amplitude was taken, but the phase of A was put on that. And what
happened when you did the electron density calculation, you got back the structure of A.
You took the amplitude of B, and put it on the structure of A, it did not give us any
meaningful Fourier.
So, what it means is the importance of phase is very input very crucial. So, the structure
essentially depends upon how the phases are rather than on the amplitudes. And so you
got the structure of A in B, because you use the phase associated with A very beautiful
experiment, which tells you that both these quantities are equally important, but the
phase information is extremely crucial to determine the structure. And so we have to
solve the phase problem, unless we solve the phase problem it is not possible to get to
the structure.
So, what is done in direct methods is to take instead of F of h, we use the value of what
we call as normalized structure factors E of h. And of course, with the phases it will be
giving us the point atom structure. Now, this is the logic, which we use in case of the
direct methods that we use E of h, which is the normalized structure factors. How do we
do the normalization, that expression is given here. E of h square is equal to F oh square,
which is our intensity that we have measured divided by the sum over i equals 1 to N of
all the atoms whose atomic scattering factor is known at that under that particular
structure.
So, if we know the number of the variety of number of atoms or molecular formula, we
can calculate this quantity. This epsilon h is something which is referred to as the
symmetry balancer. So, we actually use this to balance the symmetry, so some reflects
some of the reflections are having a weight of 1, some have a weight of half depending
upon the symmetry associated with the crystal system. So, at this moment, let us say
epsilon h is equal to 1, but it can be 2, it can be 3 and so on depending on the
equivalences of the h k l reflections.
747
We have not discussed that in this course. So, I am not going into the detail of it except
to say that there is a value for epsilon x. For our discussion, we can take it as 1. So,
effectively the normalization of the structure factor is done with respect to all the
electrons in the system in principle, and therefore, we get E h of square. So, this now
when we do a Fourier transform on that, it will give you a point atom structure as though
the atoms are all points, which is wrong we know that it has a thermal ellipsoid, it has
distribution of the electron density and so on.
But we will have a situation where we will have E of h it go undergoing a Fourier

transform to get to the point atom structure. If we now multiply this by atomic scattering
factor F of h and do a Fourier transform, we will get the real atom. So, because this part
now we again multiply, so this Fourier transform therefore, will now have the F naught
of h with the phase solution, and so we get the real atom that means, if we take F of h do
the Fourier transform we get the electron density as h. So, this is what we are going to do
in case of direct methods, but what do we need we need the phase of course. So, how do
we find the phases? So, we will stop here.
748
Lecture – 55
Direct Method
So, the requirement is to find get the E values from the observed structure factors the
observed intensities we take the square root of the intensity, this epsilon h as I said is
known as the extinction factor. There is nothing to do with extinction effect which we
discussed earlier on its an extinction factor for example, if we have a P2 1/c structure and
we look at h o l reflections remember that l reflections are all absent by systematic
absences. And therefore, the weight edge that gives is given to h o l is double the weight
of a general h k l reflection.
Let me repeat, a general h k l reflection if we give a weight age of 1; that means, it has
whatever measurement we have is if the weight of one the h o l reflections because l odd
is absent and only l even present it gets double the weight. And so when we are dealing
with h o l reflections for this E calculation. By the way I should say that it is a h here is a
vector which refers to h k l.
So, this E of h square now will be you will have to be divided by two times this quantity
so; that means, the division the weight edge is now reduced by half. So, you give a
749
weight value of epsilon h equals 2. So, depending upon the space group and the
systematic absences you vary the value of epsilon h and therefore, you get a normalized
structure, this structure factor is now effectively coming from point atoms. That means,
atoms now are in a point and all the electron density is associated with that point and as a
result you know you remember we define the delta function earlier on the diffraction is
from a point atom we get a delta function. So, the contribution comes all of it will come
from the point where the x y z is located with respect to the E value.
So, then what is the advantage, the advantage is now that it is not going to vary with
respect to sine theta by lambda. So, the scattering angle it is independent of the scattering
angle in principal. And so if it is independent of the scattering angle; that means, there is
no change in the face that is associated with a point atom. So, we it becomes very easy
now to put up a logic, the logic which we have is explained in this particular diagram this
is a very crucial diagram see you look at three planes ok. This represents the one set of
planes h k l let us say this is represented as h, it actually represents a h k l plane parallel
planes you see.
So, the dark lines here of the parallel planes and the dotted lines between the parallel
planes represent the halfway point; that means, if this is the distance at which the
Bragg’s law is validated if you now consider a cosine curve for example or a sin curve
the crest is at these points and the trough at this point ok.
So, the crest and the trough are identified as plus and minus bias in this particular table, I
will come to that table in a minute. So, now, we have a set of planes which are h k l and
you see that in this h k l plane there are 4 regions of high electron density; that means,
atoms are contained in these 4 regions A B C and D. So, these 4 regions contain atoms;
that means, the value which you get for the structure factor and therefore, the E of h is be
very large.
So, these we are now dealing with the highest possible values. So, after we get the E
values we arrange them in the order of the value. So, the highest value is put on top the
next highest next and so on. So, one once we order that we have to take the highest
possible E values and take three planes, one plane is now this represented by h, the other
plane is represented by k which is now this the origin. So, we have a k here and the k
750
there and similarly the half points are denoted by dotted lines; that means, if they are
now the troughs of the wave.
So, if you have a wave let me make it clear because that is an important thing to happen
to observe. So, if you now take wave which is like this ok. So, this to this is now the
wavelength right, one single wavelength is the point here to that point or is it the point
from here to that point. In fact, that is correct the point at this point to that point is the
wavelength.
So, if this is the wavelength then you see that this is represented by solid lines and this is
represented by dotted lines. So, the solid lines will now correspond to one single lambda.
So, to d sin theta lambda is be satisfied and so, we will have these coming in the crests of
the wave and this comes in the trough of the wave. So, the alternates so if you have a h k
l plane which is shown here let me go back there, if you have a h k l plane which is now
going here with respect to this origin. One unit away this is the next one which is going
here, between them there is a set of dotted lines which essentially represent the trough of
that particular wave which is associated with the plane h, which could be h1 k1 l1, we
call it as h let it be h1 k1 l1.
Similarly, this is the value of k which is now represented perpendicular to that direction.
So, let us say this, this and that and then the alternate dotted lines are indicated, and we
take the third set of planes h minus k which is in the direction of the diagonals. So, we
essentially represent all three directions here three so if the logic is the following. You
consider three very strong E values h k l values which are intersecting with each other to
represent this plane and they are intersecting in such a way that they passed through
electron density positions A B C and D.
Now, what happens when it passes through A? When it passes through A the value of h
as we see here is a full line; that means, it is positive the value of k is positive down here
and the value of h minus k is also positive.
So, the three values now which intersect at a will give rise to plus plus plus so; that
means, if you are looking at a central symmetric structure these three now represent a
value equal to 0, the phase value equals 0. So, this is sin of h sin of k sin of h minus k.
Now the product of these three s h s k s h plus k will therefore, become positive this is
what is important, when we have three very strongly diffracting planes which pass
751
through either at the trough or at the crest through these points A B C and D where there
is electron density. That means, atoms are located at these positions, then you will have
the product of s h s k h k and s h minus k always positive and one can associate a
probability with this and that is how direct methods are carried out.
So, we now, therefore, take three strong highest E values, these E values are defined in
such a way that we have four points A B C D which are electron density rich and these
three planes intersect such that they pass through these electron dense regions. So,
whenever they pass through the crest we call them positive and whenever they passed
through the dotted lines or the troughs we call it negative. So, the B position you see is
except for the h which is now this way, this is the h value except for h which is positive
at b these two are going to be negative, but the product is still positive.
Similarly, in the case of C you see that the value of l is positive the other h and k are
negative. And so the product is still positive, the point D you have again one of them
positive which is the k value which is positive here and these two are going to be
negative.
So, essentially if you look at this diagram you have possible combinations where with
individual signs associated with these planes, it hardly matters because there is rich
electron density associated with these three planes and that electron density therefore, it
gives for all four positions where there are atoms, the product of the signs is positive. So,
that would mean that wherever there is an atom if you take intersecting planes those
intersecting planes should be having a positive value. And this is the logic with which
direct methods work because as you go to lower and lower E values the reliability with
which you can assuredly say that these three products are positive becomes less and less.
So, you can calculate a certain probability that is associated with it, we do not go into the
detail of this discussion because this is not in the purview of our current course. This
needs an advanced crystallography course, but I will tell you how we normally do the
direct methods.
752
So, the direct methods now calculate normalized structure factor amplitudes, set up these
phase relationships I showed it in terms of science, but they can have different values of
phi depending on whether it is a central symmetric or a non-centrosymmetric structure.
So, instead of sign you replace it by the phase. So, the phase of h is nearly equal to the
phase associated with h minus k plus the phase associated with k. So, this is the
relationship which we are going to use and find the reflections to be used for phase
determination. Based on this you generate phases for let us say something like a 100
reflections you do not have to generate phases for all the reflections. The top E values are
taken top 200 E values and within that 100 E values any combinations of this kind which
we can develop, we can now associate this equation with them and find the reflections to
be used for phase determination.
And then you assign starting phases to these, you can if you assign positive sign for
everything in a central symmetric case you will get nothing you will get a very strong
peak at the origin, because it is all phases cannot be all positive. Some of them must be
positive some of them should be negative like the way we I indicated in this diagram, in
the previous diagram.
So, based on this logic we calculate now what is the probability that this product is plus.
So, in this example where I have shown the probability is 1.0, but we can calculate the
probability as we go down in the E values the probability value keeps on going lower
753
and lower. The confidence level with which we can assign these values become lower
and lower; that means, the confidence with which we can assign phases for reflections
become lower and lower, but then we can put a cutoff on the probability limit, let us say
we cut it off at a value of 0.9.
Then we will get the solution for the phases for the first 10 top E values that may not be
sufficient, what we want is reasonable number of reflections based on which we can
calculate a Fourier map which is called the E map, and that Fourier map will give us the
approximate positions of the atoms it will give us the trial structure. So, there is no
involvement of whether it is the atom should be heavy or light it does not matter now
because it is independent of the z value because we have taken this in the denominator
we have normalized the structure factors. Since we have normalized the structure factors
there is no dependence upon the nature of the atom, it is only dependent upon where the
point atoms are situated and as a consequence this probability development of phases
will depend upon how many starting phases we can have.
So, at this stage there are different ways in which one can do it, there is what is known as
a symbolic addition method, there is what is known as a phase development method then
phase criterium development method and so on. So, an estimate of each phase
determined is given by this expression. So, this expression depends upon how strong
these three reflections are and what is the total number of atoms in the structure.
So, we assign starting phases to a few reflections. For those reflections which we do not
know the phases we can either give symbols which is known as the symbol symbolic
addition method and try to solve the sign of the symbol if it is a central symmetric
structure. Or you take about twenty such reflections, strong reflections give random
phases to them.
So, use a random number generator and assign phases to them compute the E value, that
will be wrong anyway you will not get any completely map it will be a junk. So, then
what you do is you allow this to refine and you find the newer phases making use of the
fact that we should always have this reliability on the strength of the E value, the
stronger they we have E value the better is the phase determined.
So, keeping that in mind we can do an E map calculation based on the phases we
generate we give random phases to about 20 of them, keep changing the phases using a
754
random number generator every time calculate the E maps. So, 1000s of E maps can be
calculated eventually, we will hit a combination which is closest to the final structure
then the E map will give you the required I can positions.
So, this is why this method is called direct methods because it is, it can be directly
calculated the phases and so we have therefore, the structure information. The ease with
which programming can be done using random number generators and so on is
mathematically very simple. And therefore, the computer programs which are available
or extraordinarily fast, the computation is also fast because we use what is known as a
fast Fourier algorithm FFT fast Fourier transform and that FFT will make sure that your
E maps are computed at a very quick time.
So, one once you have that you calculate and interpret the electron density map which is
referred to as the E map you will get the trial structure. I do not think beyond this we
should worry about the phase determination protocols because that would make it an
advanced crystallography course which we are not now really interested in. What we are
interested in is the symmetry.
Now, where does the symmetry come into the picture here the symmetry came
automatically into the picture, it came automatically into the picture because we are
talking about the electron density and the associated phases with the planes which are
strong a strongly reflecting plane. Therefore, in principle should have electron density in
it or the contributions of various atoms to that particular plane should be very very
positively significant.
If that has to be significant then the nature of the atoms that are arranged inside this
particular space group should be related by a proper symmetry relationship. So,
symmetry again now controls the way in which this propagation of the phases occur and
therefore, again we have to see that our requirement of symmetry and the star in the
structure is very very crucial. That is all I can tell about the methodologies in this
particular course because it is otherwise becomes out of scope.
755
We now change gears; we now change gears and go to a new area and this is an area
where many of you are interested because it is not always possible to grows good quality
single crystals. In many cases there are reasons why good quality single crystals will not
grow at all; however, it is possible to get poly crystalline phases. So, poly crystalline
growth is quite possible; that means, the size of the crystals or the crystallites as we call
them are so small that each one is a single crystal, but then they come in the form of a
powder. We cannot look under the microscope and separate if one single particle.
These particles are of course, crystals. So, a large number of crystals now are assimilated
in the form of a powder. So, they if we get a powder sample particularly an organic
chemist synthesizes and gets a solid form of a in the form of a powder. The first check he
will do is to put it with XRD, the reason why it is done is that if it gives a sharp set of
reflections like this then it is a poly crystalline sample. It is possible that we do not get
this kind of a sharp diffraction line when we do the X-ray measurement we may get a
large hump like this.
Then the conclusion is that it is an amorphous material. So, it is possible to have

amorphous materials also looking very shiny like for example, glass is not as crystalline
material it is an amorphous solid. So, essentially what we therefore, do is in a three-
dimensional single crystal which we discussed so far we have isolated h k l positions, but
all these h k l information is contained in this one dimensional profile.
756
So, you have a three-dimensional reciprocal lattice which we have discussed so far.The
three dimensional reciprocal lattice gets condensed into a one-dimensional profile like
this. Whatever is the information here is also contained here, but there is a severe overlap
of these reflections, because we are now plotting intensity versus 2 theta. This 2 theta is
the scattering angle and as we go from 0 to let us say 90 degrees or 100 degrees the
reflections which dominate are appearing here.
It may so happen that under each every peak there may be several reflections we do not
know, but a typical XRD pattern with a good polycrystalline material will look like this.
Now how do we use this information to get to the structure, this is something which is
very crucial for pharma industries particularly because very often the growth of good
quality single crystal is a issue, in many other cases also when materials are made and
synthesized normally poly crystalline forms are available very quickly, but growing a
single crystal of the material is a issue. This is a problem in organic inorganic or any
other area of solid state chemistry.
So, we did discuss how we grow the crystals in the last couple of classes, but then even
with we make all these attempts it may so happen that we may not get good quality
single crystals. So, we have to deal with polycrystalline samples. So, if we have
polycrystalline samples how do we get to the structure? That is the next issue which we
will take up.
757
So, there are a certain set of cautionary points one has to take, which nobody takes when
we do the powder diffraction. So, in case of a single crystal you do not have to take any
cautionary steps this is another big advantage. If you get a good crystal single crystal you
look at through the microscope and the microscope shows the extension effect, then you
assume that the crystal is good put it on the diffractometer it will give you the diffraction
spots. And if the spots are good and you have enough information on the reciprocal
space, you will now know how to go to the structure determination and also final
refinement and so on which we have briefly looked into.
So, we know the structure determination protocol for a single crystal. What happens in
the case of a powder? We can just not take a powder and throw it on the diffractometer
which most people do and you put it on the on the glass plate or whatever is the plate
given by the company manufacturer and put the sample and then collect the data and
then you will see a picture. And that picture if it is sharp enough peaks you say oh I have
a polycrystalline sample.
And very often in industry what they do is to look at the diffraction pattern and then if
the if they know roughly what the material is all about they get the diffraction pattern
already they available in the databases and then put one on top of each other and
compare the peaks and say oh we have made the same compound. These are all more or
less criminal acts as far as diffraction is concerned, because X-ray diffraction needs a
little more, it has a little more honor than what we people give.
So, it gives us more information than what is required, as I already said powder
diffraction has all the information of the three-dimensional reciprocal lattice imaging and
therefore, we should in principle be able to use powder diffraction even for structure
determination. It is not just that we compare two samples and say oh they are one and the
same one or if you get a slight differences in the peak positions we cannot say we have
made a polymer or we have made a new compound and whatever. So, those conclusions
which are done sometimes in industry have landed them in serious trouble and it should
not be done.
So, what do we do, we first have to worry about how we prepare the sample for the
diffraction experiment and of course, then we have to worry about what is the
wavelength which we are going to use sample preparation. It is generally recommended
758
that even though we get experiment to give us a powder, it is better that we make sure
that the powder is nicely prepared, by prepared I mean we take the sample in a pestle and
mortar and we grind the sample. We after grinding the sample in principle, in principle
nobody seriously does it, but in principle we have to pass it through a sieve which has a
certain dimension.
So, let us say we pass it through a sieve which has hundred microns and it is like you
know getting the atta prepared for your chapatis. So, you just play with the sieve and
then get the powder of 100 micron sizes only assimilated in one place. And take that as
the sample and then you prepare the sample by the rule, by the method which you are
going to use, the way there are three different ways in which you can collect powder data
this is something which probably is not very well described everywhere.
The routine X-ray diffraction equipment which we get for powder diffraction is the
reflection mode. You will have a sample on a horizontal glass plate you spread the
sample on top of it then X-ray will come from here and then fall on the sample and the
Bragg diffraction is satisfied. So, it will go at an angle satisfying the 2d sine theta and
lambda and you will get the reflection and you may put a counter or a measuring device
there the detector, and the detector will catch the intensity that is coming out and then
you move the 2 theta circle so that the detector moves with it and collects with respect to
two theta the data.
So, we get intensity versus 2 theta that is the usual geometry we have in many of the
standard setups one can also collect the data with what is known as that if the
transmission mode where the equipment is such that we now allow the X-rays to pass
through the sample. The sample now can be rotated as well and as the sample rotates the
X-rays pass through it and then they come out the diffraction is measured in 2 theta with
respect to the detector.
So, this is a flat horizontal goniometer system where you have the goniometer put in a
direction of the incoming radiation and that will now catch carry a container in which we
have the sample. So, the sample is now exposed to the x rays in a perpendicular manner,
it is no longer the reflection geometry it is called the transmission geometry.
The age old method of the powder diffraction which was discovered and used in the
earlier days was to take the powder and fill it up in a capillary tube, these capillaries are
759
very special and these capillaries have a property that they are not absorbing the X-ray
radiation they indeed scatter the radiation very little other than the scattering that is
coming from the sample. So, we have to now have the sample fully filled inside the
capillary and now the capillary is rotated and this is known as the device shutter mode
due to the two people who found this methodology.
So, we have a capillary, the capillary rotates in a direction perpendicular to the X-ray
beam and we measure the diffraction coming out and that will be the 2 theta circle. So, in
all these cases the data recorded is intensity versus 2 theta. So, suppose we have a
compound this is the example of a compound you may not be able to see it so I will read
it out, this is the sample which is, the sample is I will write it down what the sample is
the sample is mica.
So, you have all seen mica sheets they are all layered and they are layered on top of each
other. So, for the reflection geometry what you can do is take a sheet out, put it on your
standard diffractometer setup and then do the reflection mode, when you do that you will
get a pattern like the one shown on the top. This is the pattern you will get.
So, you are going to get effectively 1, 2, 3, 4, 5, 6, 7 peaks the externality is of course, 0
to 60 degrees which is shown here, but you will get this as the eighty degrees sorry 0 to
80 is shown here. So, you will get peaks corresponding to these h k l positions. So, you
can find out h k l positions because 2 theta you get the d values from d you can find out
what are those h k l values one once you know the a b c alpha beta gamma. You know
that methodology, if you do the transmission mode you will get a different pattern
altogether, it is not it is the same mica, but your XRD pattern is different.
Now, this is something which is to be explored very carefully the reason is that the atoms
in mica are all in a given plane and so the scattering is dominated by that particular
plane. So, you get very strong reflections of the planes here and in the case of a
transmission mode that plane may be associated with this where you see no diffraction
coming in the reflection mode you get a diffraction which is very strong here.
So, this tells us that these two methods are now indicating the mica to be two different
compounds altogether, but it is not the case. This kind of a thing happens in many
layered compounds, that a lot of people who work on layered in organic materials.
Organic inorganic layers they are parallel pack them, they are hybrid structures. In all
760
those situations these two methods are not the recommended methods for data collection,
the recommended method is the device under approach the capillary method where you
rotate the sample so it the crystallites inside the capillary they orient themselves in all
possible directions.
So, X-rays now fall on those crystallites in different directions. So, you get diffraction in
from all different orientations. So, you see a very complicated pattern for mica here, but
what you see for these two modes is very different from each other and therefore, when
we are dealing with two layered compounds one has to be extraordinarily careful. So, the
method that is generally suggested to be used is the device shutter mode. In fact, device
shutter mode is the best mode of data collection for any powder data, but this is not the
case with all molecules and all structures. So, it is when we get a XRD to be taken we
use the reflection geometry.
So, reflection geometry is the preferred mode of data collection because it is cheap, the
most of the commercially available instruments are using the reflection mode geometry.
The entry in the database for mica is given below and you see that for every point here
there is a corresponding h k l. So, this is the calculated pattern for mica. So, you can
solve the structure of mica by growing a single crystal of mica and then determine the
possible h k l positions and those h k l positions are shown as lines here and this is the
therefore, the simulated XRD pattern and this is available in the database.
So, you see there is a one to one correspondence between this and that showing that all
the peaks that are coming in the case of capillary are not impurity peaks they are all
coming due to the presence of peaks which are in the structure of mica. But those now do
not appear because of the fact that mica has the orientation effect this particular thing is
known as the orientation effect this phenomena which we described in case of the refine
the reflection mode as well as the transmission mode.
The highest peaks of these are coming due to the exposure of that particular plane from
which the diffraction is coming and only that dominates because we are not either in this
particular case of reflection mode we are not rotating the sample either. And so it is
coming from one particular position. So, all these therefore, tell us that the mode of data
collection is something which is very important it should be remembered, particularly
when we are dealing with layered compounds.
761
Now, what is the quality and quantity of data one would like to have. If you want to
determine a structure you should get as many data points as possible the same logic as in
case of single crystal analysis where we had to use as many data points as possible; that
means, the maximum extent of data which we can acquire in the two theta range. So,
suppose let us say we go in the range of 1 to 100 if we go in the range of 0.02 degrees in
2 theta we get nearly 5000 data points.
So, we should never measure every data point in a powder pattern. So, we call it oh this
is in the background this is in the peak and all that it is our decision, but basically we are
measuring the intensity which is along the x axis y axis here at every point along 0.02
degrees in 2 theta.
If we go to synchrotron radiation we can measure it at 0.005 degrees in 2 theta. So, you

get something like 20000 data points. So, one would say why we do this routine XRD let
us go to synchrotron radiation. So, if you are determining the structure of sodium
chloride why go to synchrotron radiation. So, you know you do not need a need a missile
to kill a sparrow right, in Kannada we say [FL].
So, we do not have to go to find the structure of smaller molecules, smaller compounds
to actually go to a synchrotron radiation; synchrotron radiation becomes a requirement
when there are some special issues concerned with small molecules. What about protein
structures? So, in case of protein structures we need large number of data points.
So, suppose we want to determine the structure of a protein somehow by the powder
diffraction method which is nowadays a possibility even those are not many structures
which have been determined people have done the powder diffraction pattern of
structures which are known and try to fit the parameters associated with the structure to
the observed diffraction pattern. There we can definitely use synchrotron radiation
because we need lot more data points.
So, the requirement of the data points is one issue the second issue is of course, lies with
the accuracy with which we measure. So, since we now have a sample properly prepared,
we have a proper lambda used. So, we can change the lambda we don t have to use
copper K alpha as is believed everywhere to be the mode of data collection we can use
molybdenum we can use silver radiation. So, the more lower is the lambda value we
know already about resolution we get better resolution.
762
So, use of silver radiation for example, will give us much better resolution than use of
copper K alpha radiation you remember that formula where we calculated the number of
possible reflections from may single crystal diffraction. We saw that in case of copper K
alpha we had 14700 in case of molybdenum we had 142000 reflections. Silver radiation
will probably give us more than 2 lakhs of reflections.
So, we can always use the right source for the right analysis. So, it is not always
necessary to go to synchrotron radiation for powder work, if we are doing some other
special observations with respect to let us say non-ambient work high temperature low
temperature powder diffraction or for that matter we are looking at a in-situ reaction
process which is occurring in a powder photochemical reaction and so on, then going to
synchrotron radiation would be of enormous use. So, synchrotron radiation of course,
can be used for powder data as well that is that is the in a nutshell the gist of the whole
story.
So, what do we do first further with the powder data? See most of us do the powder data
collection like this and if you go to pharmaceutical industry or for that matter any
materials scientist they want the following information, they want the structure
information if possible, they want to know the crystallinity the extent of crystal into this
particularly. So, in case of pharmaceuticals they want the phase purity which is very
important, they want the crystallite size and of course, the instrumentation characteristics
763
become very crucial for aligning all these things, the positions of the peaks and
intensities are very crucial.
So, we have the x axis giving us the geometry the y axis giving us the structured
information. So, the x axis is the one which is most crucial for most of the industries
because they want to get to the cell dimensions and the possible crystal structure
information that can be derived from this. Other than that in pharmaceutical industry
they want to distinguish with the they find out the API the structure of the as prepared
ingredient and pharmaceutical ingredient excipients if any other required materials for
tablets capsules and so on.
And they also want to qualify in case of aerosols and then of course, the major thing lies
in terms of polymorphs and co-crystallization. Co crystallization can give rise to solid
solutions it can give rise to co-crystals it can give rise to eutectic phases and so on. So,
all these issues come up and the answer comes from powder X-ray diffraction believe it
or not and.
So, powder XRD therefore, is a very very powerful tool as I already mentioned earlier in
some context that the cell dimension you determine from powder diffraction this is much
more accurate than the cell dimensions you get from the single crystal measurements that
is because the powder sample has a large number of crystallites, many many thousands
of crystals whereas, a single crystal has only one crystal. So, that therefore, allows for
averaging effect and therefore, you get very accurate cell dimensions and therefore, the
determination of space groups and other things become fairly straightforward in that
context.
764
Let us take the case of a standard sample you know most of the materials, most of the
manufacturers give us this particular compound LaB 6 and if you record with copper K
alpha you get a pattern like this a very beautiful pattern and you go from 20 degrees in
this case to 100 and 40 degrees nearly and you get a beautiful pattern by beautiful I mean
you see the background the background is steady throughout. So, there is no deviation in
the background and you get peaks which are fairly sharp, if you look closely they have a
certain width associated with them I do not have time to discuss with all; within this
course what is the reason for the width and so on. That is again a very advanced course
in powder diffractometer.
However, we see that these are the peaks. So, if the compound now belongs to a cubic
see we discussed this already in case of single crystals, we found out how the d square h
k l values can be determined in case we have the information on the h k l as well as the
cell dimension. The cell dimensions can be determined one once we have h k l and d of h
k l available to us and it varies from various systems cubics tetragonal hexagonal an
orthorhombic and so on.
Most of these indexing is done through automatic programs which are available in free
domain space, the program called TREOR, there is a program called DICVOL there is a
package called free chris fire, but I am told by some of my students that chris fire is no
longer active. Obviously because of the fact that chris fire probably was on a previous
generation computing system and.
765
Now, the computers have changed their architecture. So, I did not nobody did make an
attempt to write that program fresh possibly, but there is this JANA package which is
again a free downloadable package which does everything it does single crystal
refinements, it does standard powder diffraction analysis, it also done the does the so,
called rietveld method which we will briefly touch upon as we go along further.
So, how do we index the spots. So, now, you see here is a pattern ok, there you have 1, 2,
3, 4, 5, 6, 7, 8, 9, 10 etcetera so many peaks ok. So, I will take the first 12 peaks
associated with this and these first 12 peaks I will start giving the least possible h k l
values; that means, I will start with 100, then the second one is 110, third one is 111 this
is a cubic system. So, in a cubic system what you have is this expression h square plus k
square plus l square is equal to half to the power half right.
So, what are the h k l values we can think of? We can think of the h k l values to be h
equals 100 and in a cubic system 100 is same as 010 is same as 001. So, we do not need
to consider those and that is why the next one we consider is h equals 1, k equals 1 so
that is why it becomes 110 then of course, 111 we go to 200, 210 or 211 then 220, 300,
310, 311, 222 in a cubic system. So, all these now we will h square plus k square plus l
square is equal to 1 in this case h square plus k square plus l square is equal to 2 in this
case.
766
So, you give these numbers 1, 2, 3, 4, 5 when you come to number seven you cannot
have a combination of h k l which is integer and it will generate the number 7. So, we
will not have seven in other words h square plus k square plus l squared equals 7 is not
possible check it out. So, give different values of h different values of k different values
of l you will not get to 7. So, we have to leave it out and then the rest is 220, 300, th310
etcetera so, this is the process of indexing.
So, if you have a cubic system a simple cubic system and you have the formula h square
plus k square plus l square and you are dividing this by a square to get to the value of d
of h k l what you get from this is the I versus 2 theta from 2 theta you calculate sin theta
and then 2 d sin theta and lambda will give you the d values those d values. Now you
associate with these planes, the very highest index could be 100, the next one is 100, 110
and so on.
So, when you give these values these will now start to fit with each other. If this is a
cubic system, if it is not a cubic system we will not get integral fits of this kind, then you
have to think that this system does not belong to a cubic system. So, the process of
indexing therefore, is all automatized so we are not in a position to go to the discussion
of how we index it. So, we have to go to these program packages which will do these
indexing whatever be the crystal system.
767
So, what I will do is instead of going into the detail of powder diffraction I will discuss
the powder diffraction uses relevant in pharmaceutical industry. This is also we can
generalize it in chemistry for example.
768
Lecture - 56
Powder Diffraction 1
As we have been discussing the x axis contains all the information that is required of the
geometry of the diffraction and the y axis contains all the information about the
structural aspects. We see a typical pattern XRD pattern which is shown here, where we
we have marked the possible peak associated with these; and the dashed lines here are
the ones which are calculated structure factors. We will see how calculated intensities,
we will see how it comes up.
And then the one below here shows the difference between the observed and the
calculated, essentially helping us to determine the structure of this material, by the so
called read filled assignment. So, in the next maybe half an hour or 1 hour, we will try to
cover all aspects related to this powder diffraction techniques, because this is something
which is quite a relevant in industry and also it is quite a relevant in teaching modules.
And at the same time of course we do not forget about the symmetry that we have and
we can associate with powder diffraction patterns. So, the questions which we can
769
answer, in fact, by just a record of I versus 2 theta on a powder sample we discussed
yesterday the three modes of data collection.
So, let us say that this is a typical standard X-ray data, which is coming from the
reflection mode. So, we take this pattern and the what are the contents of this pattern?
The contents of this pattern or it contains information about the structure. It also tells us
about the crystallinity. If you look at the background associated with this recorded
pattern; you see the background is slightly falling, but essentially remaining more or less
a constant.
The more the constant it is as we go to higher angles it suggest that we have a very
crystalline material because, the peaks are quite sharp and the width of the peaks are not
so very different from each other. They are normally about the same size. In fact the
techniques of analyzing the peak widths and so on also gives us several other
information associated with it; apart from crystalinity the other aspect is one once we
have a compound like this we can also question and answer the phase purity questions,
pure sample will have an unique powder diffraction patterns.
So, powder diffraction pattern is almost the fingerprint of a given material. So, that way
the presence of the databases the in organic crystal structure database, they came re
structure database it has all the data sets associated with single crystalline structures. On
the other hand ICDD database, the international centre for diffraction data has
information and all the powder diffraction data which has been collected over the years.
So, very closely related compounds will resemble each other as far as powder diffraction
is concerned, if you have just done for example, some changes in the alloying conditions
or some changes in materials. Or if you have made some changes by substitutions by
changing one atom to the other and so on. The powder diffraction pattern will not be
very different, but it will be different from each other I said because it is the finger print
of the given material.
So, phase purity is one possibility which we can analyze, more than one phase also can
be analyzed doing a powder diffraction analysis we will see as we go later. And we can
also determine the size of the crystallite by measuring the full width at half maximum,
that is what is known as a Scherrer formula which will give us that information. We of
course, have to worry about the instrument characteristics we discussed already the
770
modes of data collection in the previous class, but apart from that we they depending on
the instrument the width and the so called asymmetry associated with the peak, because
the peaks may not be exactly symmetric as we see in this example I have given here.
It may become asymmetric and that probably is due to the way in which we collect the
data. So, the instrument characteristics also play a very crucial role. The peak positions
and the intensities of course, we already said that peak positions are the geometry the
intensity has the structure.
So, when we come to pharmaceuticals particularly the questions come about the as active
pharmaceutical ingredient, one once they make this compound ready they would like to
ask what is the characteristics of this API? Whether this particular API is identical to the
API which has been marketed by someone else or this particular API has the issue of
what we call as a polymorphism which means that it may have a different crystal
structure.
So, questions on polymorphs cocrystallization which would mean the formation of co

crystals the formation of salts the formation of eutectic those kind of phases, will also
come into the picture. We can also work out in the case of tablets, we can analyze the
tablets directly other required materials for tablets apart from excipients and the API and
then we can also analyze the capsules and aerosols.
So, these are some of the advantages with powder diffraction which probably we will not
have with single crystal diffraction; other than that of course, all the information that is
contained in a single crystal diffraction is available in this one dimensional profile of the
powder. So, in some way or the other there is one dimension profile has to be broken
down into three dimensional reciprocal lattice information. One once we do that we can
determine the structures even by using the single crystal approach.
771
So, this is the basis of it. So, we will now go systematically to how we index a given
pattern. LaB 6 is the standard compound which is supplied by many of the companies.
They give us routinely this compound and its uses copper K alpha radiation with this;
this is the diagram. This is the diffractogram we will get; always call this as the
diffractogram or X ray pattern do not ever call these as a X ray spectra.
Because X ray spectra means something else anything to do with spectrum or a spectra
means something very different; because what it means is that you have a ground state of
the material and it is taken to a higher excited state. And then there is a there is a
radiation which is emitted from the excited state to go back to the ground state, because
it cannot stay in the excited state forever.
There is a time lag in that particular period or in which it comes down to the original
ground state. And when that happens whatever comes out as a radiation is the spectra
which contributes to the spectroscopy. So, all spectroscopy is therefore, have a time
component associated with them. And therefore, we can straight away do Fourier
transformations in spectra, because with the time component is built in there in
diffraction we cannot do the straight forward Fourier transform unless we solve the phase
problem.
The phase problem is overcome in case of spectra so; the important point to remember
here is that this is not spectra, it is essentially the scattering that occurs we have
772
discussed enough of the scattering over the days. So, it is not spectroscopy which is
involved. So, what we should do therefore, always say the as a diffraction pattern or a
diffraction diffractogram and X ray spectra is result for some other phenomena which
can also be done then can do X ray spectroscopy, different kinds of X ray spectroscopy is
using electron beams and so on.
So, depending upon the crystal system, we have already seen this table sometime ago the
value of d of hkl can be calculated depending upon the nature of the crystalline material,
that is whether it is a orthorhombic crystal or a hexagonal, tetragonal cubic of course,
monoclinic and triclinic will have more complex expressions. So, in the case of a cubic
system it is fairly straight forward calculation.
So, suppose let us say the given diffractogram belongs to the cubic system, we will be go
through a very quick process of how we go and ahead and index that cubic picture.
So, d of hkl is a which is the scull dimension divided by h square plus k square plus l
square to the power of half. Now this quantity h square plus k square plus l square we
can write it as s and that has to be always an integer because hk all or integers. So, the h
square plus k square plus l square should also be an integer. So, it can take values 1 2 3 4
etcetra.
773
So, let us take a situation where we use this let us take this example which is given here
this is the intensity versus 2 theta. So, we take this example to show how the indexing
process is done. So, what we do is we can we have of course, 2d sin theta lambda from
which we can write the hkl in this particular form, where S is the value of h square plus k
square plus l square. We say S can take the various values, so the values of S can be 1 2
3 4 5 6 as it written here and then we assign the possible hkl planes in a cubic system.
Remember in a cubic system a equals b equals c alpha beta gamma are 90 degrees.
So, whether it 1 0 0 or 0 1 0 or 0 0 1 it is the same. So, this is the associated multiplicity

because of the symmetry. So, again you see symmetry is the dominant factor in any
diffraction experiment even in the case of powder diffraction. So, when S is equal to 1
the only value we can assign to hkl is 1 0 0. So, when it is 2; the only value we can
assign is 1 1 0 because h square plus k square plus l square has to be S and that has to be
an integer.
So, 3 4 5 6 therefore, has well defined hkl values. So, we can therefore, assign hkl values
to the peaks which we find in the diffractogram. So, the values of the hkls' can easily be
identified in a cubic system, by just giving the value of S and giving each and every peak
the corresponding value of S; S is 1 2 3 4 5 6 and you see that there is no way we can
assign 7, because h square plus k square plus l square if we take three integers it will not
add up to 7 so, 7 is forbidden.
So, as a consequence we now have uniquely indexable pattern and so, we get these hkl
values; one once we have these hkl values we can calculate d of hkl and one once we
have d of hkl we can get to the a of hkl; because d of hkl can also come from one say d is
equal to lambda by 2 sin theta. So, we can get the d hkl value which is lambda by 2 sin
theta from the 2 theta value here; we calculate the sin theta value and lambda by 2 sin
theta will be the d of hkl and this is equal to a divided by h square, plus k square plus l
square.
We have the hkl assigned here. And therefore, we can get to the value of a which is what
we want because we want to find out the cell dimensions of the given crystal. So, this is
the easiest example with which we can illustrate ah, but normally there are lots of
programs which are available in opens webs; websites and these are all automatic
programs for indexing. The logic in which it goes is that any given pattern is considered
774
to start with is a cubic system, the one of the requirements that it should have is that it
has to satisfy this conditions of the integers being there for we are associated with it. So,
if we cannot assign the peaks which come in the diffractogram to these values of S then
we know it is not a cubic system.
So, slowly we go up in the reduction of symmetry so we reduce the symmetry to

tetragonal, hexagonal, trigonal, orthorhombic, monoclinic and triclinic. In that process
we get evaluation of how well the pattern is fitting the assigned hkl values. And as a
consequence we get to the cell dimensions and then we look at the possibility of
identifying a unique cell dimension that can be indexed.
So, this process is known as a indexing and this is now very automatically done in many
programs and the standard programs that are generally used are they TREOR, which
essentially means TREOR trial and trial error is what is it trial and error method. Trial
and error method and we also have the compound program called DICVOL we have
several other programs there is a package called Crys fire which is available.
But apparently of late the Crys fire has some problems. So, you have to be careful when
you download Crys fire because Crys fire used to work on earlier operating systems, the
current a improvements in technology in computer science and also the incorporation of
now the modern day memory boards, will not allow Crys fire the old one to run. But then
there are other programs the best package that can be used for every purpose starting
from indexing the powder diffraction to structure solution is the JANA and this package
again downloadable free and this can be accessed through the web.
So, these are the programs which will automatically index a given pattern. So, whatever
be the crystal system in principle following this methodology roughly, which we
mentioned we can determine the unit cell dimensions. Having determine the unit cell
dimensions what are the questions we come across in X ray we already discussed what is
the content of a the x r department in a in the powder x r department. So, I thought I will
sort of highlight the uses that are relevant in pharmaceutical industry and also we general
it should be chemistry.
775
So, one is the phase composition of a sample this is very critical particularly if you want
to generate a new pharmaceutical material. So, we therefore, do a quantitative phase
analysis which is possible to determine the relative amounts of phases in the mixture. So,
suppose we have a mixture of phase one and phase two both of them will be inside this
particular diffractogram and as I mentioned that, each and every material is like having a
fingerprint associated and that is the powder diffraction pattern. So, if you have two
different compounds mixed into a system the two different compounds will give two
different types of XRD patterns.
By type I mean even the full width at half of maximum will be different the shape of the
peak will be different and so on. So, effectively the profile of the peak which comes in a
diffraction experiment is characteristic of the material. And therefore, if you do the
analysis of the profile in principle we should be able to find out the possibility of more
than one phase being present in a sample.
So, the way in which we go about is we determine the unit cell parameters and of course,
Bravais lattice symmetry index peak positions and lattice parameters can vary as a
function and therefore, give you information about alloying doping solid solutions strains
etcetera these are all very critical in pharma industry.
The other most important aspect is since we say the profile is characteristic of the
material and its profile fitting is a issue which we can sort out in order to get the all these
776
issues completely analyzed. One can also determine the crystal structure by what is
known as a Rietveld refinement, which about which will spend a few minutes as we go
along. And this is the refinement of the entire diffraction pattern see whether we when
we make an analysis and identify a phase, and then we want to doubly verify that it is a
single phase. The best approach is to be always use the whole pattern the complete
pattern rather than use patches of with this reflection that reflection and so on.
Remember when we did the indexing with read out the peak positions we did not bother
so, much about the way in which the peak is appearing the shape of the profile and so on.
But it is important that we use the shape of the profile because the shape of the profile is
definitely represents the integrated intensities. So, it is a integrated intensity we will need
in order to calculate in order to determine the structure, because we already know that
structure is along the y axis of the powder diffractogram. And therefore, we have to
analyze these profiles extremely carefully after indexing those patterns, it is not
necessary that we have to index only some 10 or 12 reflections we have to, in fact in
principle indeed a index the entire diffraction pattern.
Some of the peaks maybe weak some of them may be strong and we have to identify all
possible peaks which appear one once we have the crystal system which has been
analyzed. So, these are some of the issues which are essential for pharma industries and
this is where careful analysis has to be done particularly when we are looking at the
identification of possible new polymorphs or comparison of synthetic protocols, lesser
efforts may be you know you have a two step procedure to make a API.
Somebody else as I tensed a procedure to make an API, but have we really made that
API. So, such questions come up and the better of course, the other issue is better particle
size for better solubility; solubility is a major issue. In case of pharmaceutical materials
particularly most of the pharmaceutical materials are delivered or administered in solid
form. And therefore, the powder diffraction of these materials becomes very very crucial.
So, this is a very basic technique that has to be employed in pharma industry in order to
characterize the solid pharmaceutical materials. So, what are the accesses the access
points which we are which we have got to do this kind of an analysis? Of course, we can
always record a powder XRD of a given material is very easy procedure now, it is not
like you know we spend a lot of time to grow a single crystal and get the single crystal
777
diffractometer powder is already formed in our experimental protocol. But is this powder
a pure powder is this powder a crystalline powder, how crystalline is the powder how
much of non crystalinity is associated with this powder or is the powder amorphous?
These are all questions which come and when we want to answer those questions there
are certain plus points which are available already.
So, first thing is we have to check if you want to determine this structure check if the
structure is already solved. So, there are several websites which are going to help you out
one is the Inorganic Crystal Structure Database ICSD the web address of which I have
given. And in fact, four percent of this is available for free online as a demo. So, one can
use that and get an inc link of how to use the ICSD. And then it can be put on order
crystallography open database which is available in this website where you can get some
information or if not all about the structure.
Mincryst which is having this website information, we can also go there and get to the
datas. American mineralogist is a another source where we have all the datas on minerals
which have been and in fact, the classification is also available among the minerals if one
goes to that site. The web mineral again is a another sub set of American mineral it just a
kind of thing and there again we can get information on the minerals; other than that we
have the databases as I already mentioned the international centre for diffraction data is
the one and the most for most available ICDD package, which will be using a program
778
where the package called PDF 4, which can be bought at a cost. And then of course, we
have the Linus Pauling File for ASM International. It is actually a sub sector of sub
section of ICDD.
The best is to of course, access ICDD because that is the one which has all the possible
data collected so far on the surface of the earth ah. In fact, they have some data sets
collected in space as well on some powder data. So, they have everything in stored and it
is easy to access the ICDD database. And of course the if you want the single crystal data
and then ICDD is also linked to the CSD now.
So, you can link these two get all the date of both the powder as well as the single crystal
information. You get information on what is called the crystallographic information file
which we discussed when we discuss the structure determination in the powder in the
single crystal case. So, we also have the availability of a Cambridge structure database.
The literature of course, we can always go to literature in the end earlier days it was the
only available source, but now people go to literature only after they do not find anything
in these; the way in which now modern day science done is go to the website. In fact, a
way in which anything has to be done is go to the website you want to know how to
make sambar? You go to the website you do not do it the way we used to do it earlier
days. So, we first go to the website get all the ingredients and follow the protocol. So,
website is the become a part and parcel of our life. So, we go to the websites get
whatever information we want.
So, website therefore, have the information you want for all crystallographic data as
well. Then of course, the databases these are storage places where you store all the data
which is available, it is like your brain you have you store the data. So, what you learnt
in class three, in general you should be able to recollect when you are reminded of it, but
when you do not have to be reminded of what movie you saw when you are a 10 year
old? The moment you are activated you will remember it. So, it depends upon what kind
of databases we can access, but databases are in the necessity to get to this. And the last
result happens to be literature still remember when we went to when we were doing our
PhD studies way back in the 70s none of the top two were available and so, we have to
go only to literature.
779
I still remember going to the library and climbing up the stacks and removing the dust of
the old books and then, take it the old journals and see whether the relevant literature is
available there or not. It was quite a painstaking methodology which we used to use. So,
normally when we go to the library we will be coming out very tired and hungry,
because we will be climbing up the ladder to access the books which was stored way up
there; and then taking it out and opening it closing it with removing all the dust again go
up come down. And each access point will be a different parts of the library and depends
on the size of the library you are accessing it would be quite a painstaking exercise
anyhow.
So, the we can use the PDF to search ICSD listings and follow the references. So, ICSD
also helps you to ICSD and also ICDD both they help to follow the references because
they give a list of references. This will help you in looking for isostructural materials
similar structures isomorphs, isostructural those materials can be accessed. So, this is the
access point which we have and therefore, powder diffraction is not just a standalone
recording of the PXRD, it now has a lot of information that is hidden in it which we can
isolate and identify.
So, there is the main important thing which comes up in a pharmaceutical industry is the
phase identification the so called search and match. So, we there are packages that are
available we associated with the machines which we use. So, the diffractometers which
780
you use if you use STOE and CIE you have a automatic search on peak list or manual
matching with raw data or the peak list.
If you lose Bruker you have this package called DIFRAC EVA which is gives you access
to automatic search on the background sub subtracted raw data automatic searches, peak
list and so on. The most recently developed package which is the high score plus by
Panalytical allows for automatic search of raw data background subtracted automatically
and also it allows for the search of the peak list. So, all these increments now come with
the facility that you had you cannot you do not have to worry too much, you just ask the
machine or the computer which is connected to this diffractometers to do the phase
identification it gets automatically done.
It not only searches for the required phase, it also matches the phase with the databases.
And so we now get a very clear idea of a whether we have made a new compound or not.
Now, here is a situation where we now jump on to the determination of the structure, see
the number of peaks that are produced in a diffraction experiment, whether it is a powder
diffraction experiment or a single crystal experiment, are limited to the number of access
points we have on the limiting sphere the reciprocal that is information. So; that means,
that all the reflections that we can collect on a single crystal should also be possibly
collected in a powder, depends upon the what range of 2 theta we are covering.
781
And so, if you have a pattern like this particular pattern you see is if you expand the
region from 39 I hope you are able to see it 39 to 45. If you expand that region of the
powder diffraction pattern you will see not very comfortably shaped peaks, but a real
distribution of peaks in this region. What one can do is one once you have index the
pattern you know the crystal system, it is not generally possible in a straight forward way
to determine also the space group unless of course, you do a search match and find out
the corresponding crystal system in the database. Otherwise you only have the crystal
system.
Now, one once we have the crystal system identified a b c alpha beta gamma identified
for the given compound, then you can calculate how many hkl values are possible and
where they can appear in 2 theta. Use the Braggs equation. So, when you do that you will
mark the ones which are marked as horizontal the vertical lines here, are the ones which
actually tell you what are all the peaks that can form; if you look at region like this you
see that there is a severe overlap of one or two peaks could be 2 peaks, 4 peaks, 6 peaks
and so on.
This is the problem in analyzing the X-ray powder pattern. X-ray powder pattern is as we
see it is a one dimensional profile which is extending our the 2 theta range. And
therefore, the individual reflections now may appear under one peak may appear under
several individual peaks may appear one under one peak. So, this is the peak overlap
problem unless we solve this peak overlap problem we cannot get the intensity is of
individual hkl planes. So, to do that there is a procedure called the pattern
decomposition.
So, this is a illustration of how one can do this? I am not going into the detail of how it is
done because of lack of time I think we are almost coming to the last couple of hours of
the entire length of the course. So, we cannot just spend time on that, but essentially I
will tell you how the breakup is done? The breakup is done by actually trying to develop
the shape of the profile. So, we do what is known as a profile analysis.
So, the shape of the profile now will be able to be fitted under this particular curve. So,
this is essentially a curve which is going from 39 to 45 and under that each and every
position of hkl is associated with the peak. The shape of the peak how we assign I will
tell you in a minute where that is referred to as the profile refinement protocol. So, we
782
therefore, now what we do is we distribute the peak positions for example, under this
position which you have here, let us take an example this one particular this peak for
example, at 42 it is a unique peak there is only peak. So, we cannot see that the peak can
be fitted with a function which goes under it the width of the peak and the shape function
associated with the peak can be generated mathematically.
And how we do that? We will discuss in a minute; so we therefore, see when there is a
severe overlap, there are large number of peaks for example, in this region if you
consider there is a peak here 1, there is another peak 2, another peak 3, another peak 4.
So, all these four peaks together will give rise to this profile which you have got in your
recorded pattern. So, the recorded pattern why it was looking so funny is because it has it
is made up of four reflections. So, when once we identify those four reflections, we can
attach individual intensities to that and adjust the intensities in such way that we mimic
the observed pattern.
So, we have the calculated intensities the shape of the profile is evaluated, then you
mimic the shape of this; one once you mimic the shape of this you know now you have
fitted the profile. And this process is known as profile fitting and we can refine this
profile with respect to various parameters, which are used to describe this function
profile function and therefore, we can do a profile refinement and when a good
refinement occurs we know that whatever cell dimensions we determine for the structure
is very very accurate. So, it is not enough just to index the pattern and say this is our
compound, we have to also do your profile refinement to verify the indexed pattern.
And then get the information on individual peaks see. So, far we have not brought the
idea of structure, we are not telling where the atoms are and how the atoms are located
we are just trying to fit the profile and therefore, this process is known as profile fitting
which will give you an evaluation of the geometry of the situation. If you want to put in
the intensity information that is a different story which we will see and that is the one
which takes us to the Rietveld refinement protocol.
783
So, let us go further and we see here the types of profile functions we can fit. So, we
have to get a function which is looking like this right. So, we have to get a function
which will look like that then let us take this function this is the profile so associated
with this reflection.
So, in order to get that profile function we need to have several parameters and those
parameter are supplied mathematically by these following functions. One can use a
Gaussian function if one is collecting neutron data, the peaks coming out from a neutron
diffraction will always be Gaussian in shape. The reason is obvious because the
diffraction is not from the electron density around, but it is from the nucleus. And
therefore, the size of the nucleus is about 10 to the power of 12 centimeters minus 12
centimeters and the size of the atom size is about 10 to the minus 8 centimeter. So, its 4
out of some magnitude smaller and the incoming radiation is X ray and incoming
radiation is also one extra.
And therefore, we shape of the profile is generally a Gaussian in other words the atom
positions are now very accurately available through the location of the nucleus and
therefore, they all become individually delta function. So, the Fourier transform of the
delta function will be a Gaussian or vice versa. So, we therefore, have Gaussian
functions, very easily fitting neutron data. So, it is recommended that when we have a
accessibility to a neutron source powder diffraction data of a neutron is very very useful.
784
In fact, a powder diffraction data can also help us in finding one once we solve the
structure we can also find the hydrogen atom positions. The only problem is that the
neutrons are very very even though they are moderated from a nuclear reactor; they are
very very strong.
And therefore, the amount of depth powder that is required is quite a large amount, we
need gram quantities of the material to do neutron diffraction by powder; where as you
can you need only a few milligram sometimes even a 5 to 10 milligrams you can get a
very good X ray diffraction data. So, this is the limiting you see the two techniques go
hand in hand. The technique of neutron diffraction by itself is not a standalone even
though one could say that one can make it a standalone technique, depending on the
particular compound which is designed for neutron diffraction.
Otherwise it just goes as a hand in hand in glow technique associated with X ray. So, X
ray technique is the paramount it is the first one which is supreme, the only issue about X
ray is as we know is the falling of the atomic scattering factor which sin theta by lambda
which does not happen with neutrons. So, even though in the syllabus I mentioned we
will discuss neutron diffraction it looks like that we are running short of time. So,
whenever possible I will bring in the advantages and did disadvantage of using neutron.
And at the same time mention a how superior the data from a neutron would be for
determining the hydrogen position, because there is no variation of sin theta by lambda
fall of this scattering factors. There is what is known as a scattering length which is
characteristic of the given element and this value can be both plus and minus we can
normalize it to 1. So, it is be either plus 1 or minus 1 and the range of these values can be
occupied by various elements.
But these elements now will not show any change with respect to sin theta by lambda.
So, it remains the same value throughout the range of the data collection. The other
possible fits which are given there are we have given four of them it depends on the
instrument on which you are collecting the powder data you are. So, if you are using a
Lorentzian data; Lorentzian function you use these parameters, I am not going in to tell
you what are these parameters except to tell you that, if you take this as the initial
position of theta; and this is the final position of theta. The difference between these is
785
sort of minimized with respect to these functions the C 1 and C 2 constants will adjust
the functional parameters.
So, we have Lorentzian type in a module Mod 1 Lorentzian, Mod 2 Lorentzian and very
often we use this so, called pseudo-Voigt function. I am not explaining individual values,
we have used for example, here H associated with this and we have used G associated
with it these are all functions which are readily available ah. So, these functions are
already available in all the instruments. So, so in instrument determined; the instrument
determines the profile function which we normally adopt. So, for a routine powder
diffractogram where we get from the machines like panalytical or Bruker we normally
use a pseudo-Voigt function for the profile fit, just remember that I do not think is very
particularly important.
786
Lecture - 57
Powder Diffraction 2
So, let us look at what are these refinable parameters we are talking about in a profile.
So, let me do one small my specialist diagram and draw a profile. It is not looking very
bad, but it is a very bad diffractometer on which this profile has been obtained, but it is
we can discuss that. So, we have a profile and this profile has to be fitted now. So, it is it
has a certain x direction gives us the 2 theta value, the y direction gives you the intensity.
So, the x value we read out and it could be 1 peak, it could be 2 peaks, it could be 3
peaks.
So, what are all the refinable parameters which will fit this particular shape of our
profile. One of course, is the structural parameter which we do not have we do not know
the structure. So, we will not touch it. So, we will keep it separately, but those are
refinable remember if you know the structure if you know the positions of the atoms then
these are also refinable parameters in the way we discussed in the case of a single crystal
analysis the so called refinement procedure.
787
The profile parameters can be separated out from the structure parameters and those are
the 2 theta-Zero which is specific to the instrument. Sometimes the instrument alignment
is such that there is a little deviation from what we call as 2 theta-Zero. So, that has to be
normally it can be estimated or the instrumentation instrument people will give us that
particular value. It can also be ideally determined by collecting a standard data.
For example, if you put silicon and you have to get a reflection at 2 theta 28.65, if the 2
theta is at 0, so, you put silicon on the machine and find out where the peak is coming
and that will give you the 2 theta-Zero. If it is coming at 28.6 then you know there is a 2
theta is shift of 0.05 in the in the negative direction. So, that correction has to be applied.
So, and it also is a refinable parameter. We can vary it with respect to the length the
extent to which we can carry out the structure determination in using 2 theta the profile
refinement using 2 theta.
Then there is a typical instrument for profile one of them is caused by me just now. So, it
can have a shape throughout like this is a issue which we can be corrected, then profile a
symmetry which is very obvious in this profile which I have drawn. Then you know if
you now consider the powder pattern to be going from a certain 2 theta value to certain
other 2 theta value. So, let me extend it like that. Then there is a background which goes
like that.
So, the background which is goes going like that need not be a straight line. It can be
falling off with respect to sin theta it may be a first order function, may second order
function. So, we have a polynomial which takes care of fitting the background. So, we
can now use those as the refinable parameters. Refine the functions of the polynomial to
account for the background and this is a very important factor. How you account for the
background now will tell you how many how accurately you can get the peak positions
and their intensities. Suppose, you put the background way up here then you lost all the
reflections.
So, essentially background selection is a very crucial factor and the way it is done is to
choose the background along the 2 theta directions starting from one of the low angle
regions to the highest one and choose some 20 or so points which are definitely not
coming under any peak indication even if there is a little kick here we do not take it as a
background point. So, thus background points now are fitted with a polynomial so that
788
the peaks and the peak position stand out. And, so, this is this is a little tricky issue, but
again these are refinable. So, profile refinements will handle this kind of a phenomena.
Then of course, crystallite size and microstrain and once we have the profile parameters
analyzed we can look into those issues. Specimen profile breath parameters this is due to
the nature of the specimen itself as I mentioned in the beginning the choice of the sample
preparation. So, if there is no such preparation done somebody throws the powder on the
diffractometer. Each powder particle you can call it as a particle will have different sizes.
So, the size of the grain will be different and if different size grains are there it will also
introduce a certain breadth associated with the peak. So, the peak become wider. It has
nothing to do with the fact that it has the size effect. It is because of each crystal has a
different size.
So, if you have the sieve experiment which we discussed yesterday, then it is possible to
look at the sieve experiment and then get uniform sized particles and then it is it will not
have this difficulty of specimen profile breadth parameters, but if that does exist we can
refine them. And, of course, the most important thing is the lattice parameters see we
determine some a b c, alpha beta gamma by indexing hkl peaks. We identified and we
identified maybe 20 maybe 30 depending upon the quality of your diffraction data. That
is probably not an indicator of the correct lattice parameters. So, lattice parameters also
can be refined under the profile parameters.
So, profile parameters are so many, specimen transparencies, specimen displacement

these are all issues and of course, the major thing is the absorption. Because it depends
upon how many heavy elements are there in your material if there is a heavy element.
Particularly when you are doing a pharma analysis and your excipient is let us say
titania; it has titanium in it and as a consequence one has to be worried about the
absorption effects. So, this absorption is also a refinable parameter and so, when once we
refine all these parameters we can do the profile fit.
789
And, there are packages that are available to do both the pattern decomposition as well as
the structure determination. Now, what I will do is I will take the profile refinement
examples when we discuss the polymorphic issues. Here I am going to take you directly
to the structure determination protocol. This is ab initio structure determination protocol;
that means, you have no idea of the structure you recorded the powder, you did a pattern
decomposition using this package called EXTRA, the program you can do the pattern
decomposition. And, this pattern decomposition is done in the following way. We
consider the data from 2 theta 3 to 100.42 degrees and a given some particular compound
we are we are collected the data I think the picture of that is somewhere here.
790
So, this green line is the powder diffraction pattern. The green line which you see here is
the powder diffraction pattern of a compound. This is bismuth vanadate and vanadate
different kind of bismuth vanadate. It is not a direct bismuth vanadate, it is a substitute as
bismuth vanadate a compound for which the structure was not known and you see it is a
very complicated pattern. In fact, it indexes as a triclinic system.
So, we have this powder pattern. So, we now want to determine the structure of that
compound. How are we going to do that? What we do is we now go back and look at the
2 theta values. We take the 2 theta values from 3 to 100.42 and the data are collected at
0.02 degree intervals; that means we have a very large number of observations and what
we do here is to consider each one as an observation.
So, whether it is a peak or otherwise each measurement is now considered an

observation. So, we have so many measurements that are available with respect to 2
theta. And, so these measurements now which appear here with on the green line or the
powder diffraction pattern of that particular compound.
So, the example here which is given here is that we that we were able to isolate and
identify 498 reflections. So, it is not just the peaks if you count the number of peaks here
it is 1 2 3 4 5 6 like that if you count maybe about 35 to 40 you will get. But, if you do
this fact that we if you do this analysis of where each and every hkl can come in this
crystal system when once you have indexed the pattern, the pattern has been indexed.
791
The a b c, alpha beta gamma is known. So, we can calculate how many reflections will
come up. So, each and every reflection is assigned to the corresponding 2 theta value and
that particular 2 theta value is now considered a reflection and the intensity is considered
as a measurement. So, when we do that as a intensity measurement, then we will get the
498 reflections generated. Let me take you back here to show what I mean by that.
So, if you look at this one if you look at the counter peaks 1 2 3 4 5 6 7. So, in this case
there were seven peaks on the powder pattern, but if you look down here and count the
number of reflections that are possible you will see 1 2 3 4 5 6 7, a overlapped peak 8 9
10 11 12 13, overlapped peak 14 15 16 17. So, you have a very significantly larger
number than what you have there and of course, the overlap peaks will remain overlap.
So, we have to do the pattern decomposition to identify individual ones and we have
done that through profile analysis. So, if we have done that exercise we will end up
getting 498 reflection. So, it is as though you have a single crystal data now which has
given us 498 reflections.
Then you calculate the profile R. The R factor associated with the profile, how do you do
that? The R factor that is associated with the profile can be obtained in the following way
you have the powder pattern. Now, based on the idea of where the reflections should
come we have assigned the intensities. The assigned intensities will generate reflections
and those reflections are marked as plus here. So, all these are possible positions of
792
where the intensities can come in this powder diffraction pattern. So, the points plus is
therefore, the I calculated and this is the I observed. So, for every point we have a I
observed and I calculated. So, what did we do in the least squares refinement in the
single crystal case we took the modulus of f of hkl minus modulus of the observed f of
hkl minus the calculated f of hkl the modulus and then we made it a minimum quantity
by doing the refinement protocol and then we calculated an R factor.
So, similarly we calculate an R factor here and that R factor is now associated with the
profile. So, we call it as the profile R factor. Of course, the each and every measurement
is now benefit is has a different advantage because at lower angles we have more
intensity at higher angles we have less intensity. So, we need a modifying function and
that function is the weight factor. So, we attach what is known as a weight factor and
therefore, we have two R factors the profile R factor which is 11 percent here 11.55
percent which is pretty good for a powder diffraction pattern and the weighted R profile
is 15.80 percent.
So, when we have this kind of a situation we can believe that the 498 reflections are
reasonably well generated. So, we now have a list of hkl and the intensity 498. It is still
not sufficient you know in a single crystal as you all know by now you have thousands of
reflections for a unit cell of a defined value. So, here we get still not that kind of
numbers, but 498 reflections is not a bad idea. So, using these 498 reflections we can
either use the Patterson methods or direct methods to hope and get the positions of the
atoms.
So, in this particular case luckily for us with the example which we took, in fact, this is
the structure which was determined as the first ever structure in India for a ab initio
measurement and this was done in our laboratory by a person by name Digambar Porab
who is now teaches in Goa. So, this compound we know is a bismuth vanadate ok. So, it
has two bismuth atoms and one vanadium atom according to the molecular structural
formula and using that information we now went on to use the Patterson synthesis.
So, 498 reflections we use the Patterson method we also use the direct method and then
we in both cases we were able to identify two bismuth atoms and one vanadium atom
uniquely because these are standing above over and above the oxygen atoms which are
present. So, when once we have this information we cross checked with using both the
793
methods in Patterson gave directly the two bismuths and one vanadium. We verified this
by running the direct methods also and found out the two bismuths and vanadiums are
reasonably well located with a good R factor over all R factor associated with this 498
reflections.
Then what was done was to use this approach to do a difference Fourier. So, what we did
was we accounted for two bismuth and vanadium and did a difference Fourier map and
this difference Fourier map on this program that was used is program called GSAS that
was used for the refinement of the structure later. But, at this moment Patterson and
direct methods gave the positions of the heavy atoms. These every items were now
subjected to difference Fourier analysis; that means effectively we remove the bismuth
and vanadium and see what is left in the density.
And this picture which is shown here gives us the difference Fourier map and you see the
two bismuths here and the vanadiums here, the vanadium here, the bismuths there and
so, this is a 3-dimensional plot of the difference Fourier and whatever is the density
which is coming up here represents now the oxygen positions. We then identify these
atoms, got the oxygen positions and put all the heavy atoms as well as the light atoms
together and theta refinement. The refinement was a profile refinement, but with the
structure information available with the structural parameters put in.
794
So, what are the structural parameters we are talking about? The structural parameters
are the same as the structural parameters we always refer to that is the position
coordinates xyz and then the thermal parameter which could be isotropic B or
unisotropic B 11 to B 23; the 6 parameters and of course, the scale factor. So, these now
define our structural parameters.
Now, what we are doing is to put the bismuth and the oxygen positions into this one and
do a refinement and then do the difference Fourier map. And, that difference Fourier
map of course, every time as we do the structural parameters we should also refine the
profile parameters. The way it is done is in fact, very well organized and that is known as
the Rietveld refinement due to a person called Rietveld.
So, what we should remember particularly when people loosely talk about Rietveld
refinements in many cases because Rietveld refinement is in the phrase itself is a
refinement technique; that means, you should have a starting model the starting model in
this case came from our position of the bromine, the position of vanadium and then the
difference Fourier map given the positions of oxygen’s. So, we are not doing any other
aspect. See we took the powder pattern and decompose the powder pattern and identified
some 498 reflections use the Patterson approach and the direct method approach to get to
the heavy atom positions. After that the heavy atom positions were put in and refinement
was done.
So, before we could do all these things we have done the profile parameter refinement in
the beginning that is we have fitted the every powder diffraction peak with a profiles and
we have refined the parameters that are associated as is indicated here. So, essentially
what I am trying to tell is the way in which we have to do the Rietveld refinement, it is
not a structured determination protocol Rietveld refinement will not help you to solve
structures there is a misnomer and so, I want to remove that misnomer because Rietveld
refinement is only a refinement technique and it uses both the profile as well as the
structural parameters.
So, the number of parameters are many more because you have to have 2 theta-Zero, you
have to have instrument profile, you have to have background, you have to have lattice
parameters everything to be refined. Before that; you in before you introduce the
structural parameters into the protocol. So, when you do this refinement therefore, the so
795
called Rietveld refinement you need a starting model. The presence of the starting model
can be generated the way we discussed or the starting model can be obtained from the
database.
So, if we go to ICDD database and look at different kinds of bismuth vanadates, the data
that is available on those there is a good possibility that there may be a structure very
closely related to the structure with the cell dimensions agreeing nearly the same. So, in
such a situation we can take those coordinates and put those coordinates as the structural
parameters and go ahead and refine, then it will be a successful refinement.
So, in other words Rietveld refinement can do only small changes in the starting model.
It cannot determine a structure find a position of an atom in the in space and so on. And,
so, so, the refinement technique which is used under the Rietveld protocol is essentially
referred to as a toothless technique because it can only grind it to a certain extent. It
cannot grind it to a final structure refinement position final structure refined positions of
the atoms.
So, keeping that in mind it is a very extremely useful technique because any powder
diffraction pattern in principle its structure can be determined ab initio like the way we
discussed or we have to now find ways and means in which we can find the starting
models. If we can do that then this Rietveld refinement protocol will do a wonderful job
and we will get a good fit as you have seen here all these are the calculated positions
marked in marked as red in this particular case.
So, all these marked in red as plus or the calculated positions the values of the intensities
I calculated. I observed is from this and then you see the pink display here telling how
well it is the well is the feet. So, we can calculate the corresponding R factors and this is
a fairly successfully determined structure.
796
So, the structure is shown here. It is not a simple structure; it is a fairly complex structure
this is the biggest structure of in those days determined by ab initio structures. Of course,
there is there was a structure of zeolite which was determined around the same year by
Lynne McCusker and that is a much much bigger structure and the much more
challenging structure. But, this is one of the bigger structures which came from our
laboratory. So, this is the structure of bismuth vanadate. The positions of the atoms are
very clearly seen here. It is as though it is a single crystal structure. All the details
associated with the single crystal structure can be derived from this.
797
Lecture – 58
Power Diffraction 3
So, what is the method and strategy we use? We know now Rietveld refinement will help
if we have a static model, but how do we get this static model? One is to do in any case
we have to do the pattern decomposition and there are two mainly available methods one
is due to Pawley and the other due to Le Bail and these two methods are there already in
the program extra which we used.
The structure solution of course, we can develop a model either from direct methods or
from Patterson methods and that can be now taken to structure completion by the
difference Fourier technique and then we follow it put the structure in and do the
Rietveld refinement where we refine the profile first take the profile to the final possible
refined values of Rp and Rwp and then introduce the structural parameters put x y z
etcetera and then do the structure refinement. It is not the straightforward procedure like
in a single crystal where you just press a button and the least squares refinement gets
done using all the parameters.
798
Here we have a certain specific wave in which we have to release the parameters because
we are dealing with very subtle things. Remember the overlap in the powder pattern and
the one d profile, so we have to be very cautious when we do the Rietveld refinement.
So, initially we release the we first do the scale factor refinement and then we release the
x y z, so there is an excellent book in fact, I should give the reference to the book here,
this is the powder method where we describe the Rietveld refinement etcetera what is the
edited by R. A. Young.
So R. A. Young edited this book called powder methods; methods and this particular
book is IUCR text and available through oxford university press, so one can buy this
book or read the book and this will now cover all these full approach because it is out of
scope of this particular course. I just thought, I will mention the possibility of
determining the structure other than these methods which is the traditional approach and
that is referred to as the direct space approach. So, here we are directly trying to get a
trial structure.
So, the trial structures are generated independent of the powder pattern, so we do not
worry about the powder pattern. We can use techniques like Monte Carlo simulations,
simulated annealing, grid search and of more recent origin is the genetic algorithms. So,
these methods will allow you to give a starting model for the structure, these are again
based on the knowledge which we have in the databases.
So, making use of the knowledge in the databases we can build a possible starting model
for the structure run it through these theoretical approaches to get to a situation, where
we are nearly correct in our structure. And then we can calculate compare the calculated
pattern with experimental pattern and do the Rietveld refinement to solve the structure.
799
This direct space approach is undergoing a lot of new changes in fact, there are not many
structures which have used this method and structures have been done, but it would be
the most preferred approach if people are interested in pharmaceutical industry to
determine the final structure, the complete structure of the compound with PXRD
studies. So, the methodology the basic methodology is described here, first we get a trial
story crystal structure need to be generated in direct space.
You can also use for example, other experimental techniques like NMR for example, you
can get a starting model from NMR, a starting guess model from NMR put it into this
procedure and then try to get to the structure the. So, once you have a trial structure you
calculate the powder pattern compared with observed powder pattern. So, your R value
and the Rwp; Rp and Rwp will be the guidelines.
We already know how and why? Because Rp and Rwp are now the signatures of the
nature of your material, these are the fingerprints of the nature of the material where the
atoms are sitting. How the structure is developed decides the shape of your profile and
also where the reflections appear.
And therefore, Rp and Rwp give the guidelines and the aim is to identify the trial
structure with rho a star value this is known as R factor search this is already practiced
even in single crystal structure determination in earlier days before direct methods and
Patterson methods took cover.
800
Particularly direct methods took over people were experimenting with this again the
contribution of G. M. Ramachandran to this area has to be remembered. He did what is
known as an R factor search using the vector search methodology, using the Patterson
function he did what is known as an R factor search and identified the lowest R value
which could give us the trial structure and then did the routine structure refinement this
was with single crystal data.
Any technique for global optimization may be used Monte Carlo simulated grid search
and genetic algorithm and cell dimensions and experimental of course, accurate cell
dimensions and good experimental pattern are the prerequisites. So, the structural
formulas as to be known and the molecular dimensions then are optimized if they are
known partly or fully.
So, this is the approach which generally allows the structure determination directly
without making any assumptions. These are undergoing a lot of changes there is lot of
structures which are being solved now, but not to the extent of the number of structures
which come every day from single crystal diffraction which is obvious because the
methods are yet to be tested and checked out.
So much that my friend Lynne McCaskey who practices this art of structure,
determination over the years made this maze. Now, this is in fact, maze the structure
solution by powder diffraction is indeed a maze, so you start from the you start from.
801
The polycrystalline sample and you have to end up in crystal structure. So, this is like the
maze you have to find the way the shortest to the proper path through which we are to
traverse to reach to the crystal structure.
So, what you do is first you collect the data using the radiation which could be laboratory
X rays, synchrotron X rays, and neutrons all three can be used. And then we do the data
collection and after the data collection we need to do the indexing, so we get the cell
dimensions and then we find the space group we just the space group because in powder
diffraction a unique determination of the space group is not possible.
So, if it is a monitoring system we tried and it is there is an indication of the space group
identification, we can identify the all possible space group try the structure in all the
space groups, so it is a little tedious, the approach is a bit too tedious. Then the intensity
extraction so the intensity extraction Le Bail and Pawley is methods are available. One of
the major problems here is from the Pawley crystalline to go to indexing and space group
is a real ordeal and once you get over it you can intensity extract them with Le Bail and
Pawley methods.
And then you have to treat the overlaps there are various mathematical ways which are
available to treat the overlapping peaks and this can be done by collecting multiple data
sets for example, at different wavelengths maybe and then see whether we can
differentiate between the overlapping peaks. And then we use the method of structure
determination, so model based methods, reciprocal lattice methods which we discussed
the Patterson and the direct methods.
And then the direct space methods the use of Monte Carlo simulation and simulated
annealing and so on and genetic algorithm. So, using any of these we now solve this
structure, having solved the structure we have to now do a whole profile refinement, the
whole profile has to be refined this is very very important. So, it is not just refined
anything I hope this is visible the whole profile is it visible because the background is
very similar to the color I have used here, so the whole profile is a must. So, whatever we
do we have to consider the entire diffraction pattern.
So, as we said 3 degrees to 100 degrees into theta all that whole thing has to be
considered and then we of course, from there go to the chemical information which we
802
have this is a must, we should know what are the contents of the material in which it is
made of number of phases etcetera; etcetera, then we go and complete the structure.
So, this is the first part of it, so we have the trial side, so you know in the same logic any
structural determination is done in two steps, first we do all these things and get a trial
structure. The trial structure now uses the data the extent of enormous amount of data the
over determine data that is available and then refines the positions the thermal
parameters etcetera is the same thing we do here, but we do it with Rietveld refinement;
that means, we do both the profile as well as the structural data and then we go to the
crystal structure.
So, this is the overall powder diffraction technique to start from the data up down here
somewhere here and go to the structural determination there. So, what is required for this
entire operation is a poly crystalline sample. So, if you see that if it is a single crystal
data this maze is now FLT is very much reduced, so we do not have a maze, we can have
a direct path to determine the crystal structure whereas, in powder diffraction you have to
go through this maze of various steps before you come to the actual crystal structure.
Now, the next issue which we consider is issue which normally comes up in pharma
industry. I will take two examples where which were sorted out in our laboratory using
powder diffraction and profile refinements. So, a question came from a company, so I am
going to use technical numbers now companies do not allow us to use the name of the
803
compound because these are all come under the, what we call as the CDA agreement the
former people who are participating in this course they will realize what I am talking
about.
So, this is the technical report on issues related to a compound which was made by a
company and they called it BIPL 005 25. So, if we do not know what the compound is
even today I do not know what the compound is, but it is I know now of course, because
it is marketed now. So, the known structure which is reported in literature for which
there was a patent which was taken by a company is called GSP PYMT-17-04
technically, now a by technical it means that they have the pure compound which they
obtained from the company.
So, the idea is to see whether this is different from that to start with. So, what is normally
done? You just record the powder pattern and put one powder pattern on top of the other
this is what most of the pharma people we will do and when they did that we they found
that there is something like extra peak here and some peaks here are looking different to
them and then there is additional peak here which is probably is due to lack of
crystallinity this is not really showing that.
So, when you record the two powder patterns, so you cannot immediately conclude that
whether it is a polymorph of a drug molecule or not, you see that there is a lot of
resemblance also between the two patterns, so the question came up whether it is a real
polymorph. So, these are some of the problems which come which can only be solved by
the knowledge which you have in symmetry, the knowledge which you have in
diffraction and the knowledge which you have in structure determination otherwise these
problems can never be sorted out ok. So, this is something which is very important that is
why the course which we have studied in understanding symmetry and structure gains
enormous importance in your future carrier.
804
So, what was done? The indexing was done we used a package called fullprof I told you
GS as in the previous case and fullprof and Jana. So, many I said, but we are using
fullprof and Jana in this protocol we can use whatever program we would like to use, but
we use fullprof and use this program DIVCOL which I already mentioned to determine
the unit cell dimensions. It so happens that the cell parameters of the company compound
or 8.1, 23.8, c is 10.8, beta is 99.7 and the volume is 2077. And the figure of merit which
tells you how well these cell parameters have been fitted using the profile refinement.
So, what we did was to do a profile refinement.
Got the cell dimensions the usual way, indexing solving for the cell dimensions and then
we calculated this so called figure of merit which is essentially telling you how reliable
this value is. It should be reasonably high value it is 19.8, the larger the value the better
is the fit. So, let us not discuss this FOM because I have not talked to you about how this
is coming about, but what is very important is this fact which is highlighted here all the
peaks have to been in the indexed.
So, when you do a profile refinement it is also your job to make sure that all the peaks
which appear are all indexed; that means, we have the hkl associated with every peak,
there is no unindexed peak in the list with which you have determined the unit cell
dimensions and this is a must when that is a must we know now that this is very uniquely
determined unit cells for the system.
805
Then the other compound was also taken up using the same package, the same machine,
the same characteristics associated with the machine, same conditions we find that the
cell parameters are these and the volume is 2441. Now the two volumes are different and
of course, the cell dimensions are different the beta is different, so do you conclude that
it is a polymorph well we do not know. Because the indexing of course has been done
uniquely, so that says that these two are single phase compounds anyway the marketed
compound has to be single phase otherwise it will not be patented and patents will not be
permitted.
So, the first excitement of this is to say that it is a new polymorph, but we cannot just say
new polymorph and wind up the whole thing. So, what we have to do is to do the profile
fitting, we do not need to do the structural determination we do only the profile
refinement and the profile refinement gave the following values R p is 10.08 and R wp is
15.36 and there is a measurement of the accuracy with which it is fitted and that is 22.29.
These two values 10 and 15 are very reliable given the nature of the part the powder
pattern we have. Normally in a very good powder diffraction pattern and particularly the
one we discussed previously which was inorganic in nature we will not have that kind of
quality powder diffraction coming in this case.
And therefore, what we see here is the two R factors which are reasonably reliable, so
we say that this cell dimension has been reliably determined. So, also the next one we
have to do the experiment on that as well and do the profile fitting using the same
condition, same evaluation parameters.
We find there it is a slightly better fit it is more crystalline as you saw, here this material
is more crystalline than this material which these people have made in the company, that
depends upon the method of preparation the conditions in which they made and so on;
obviously, they were claiming that the methods were different. And therefore, it has to be
a new polymorph that is was their conclusion, but the conclusion cannot be unless it is
absolutely verifiable it cannot be concluded that these two are the different compounds
and therefore, polymorphs of the same material API.
806
So, the profile fit results are shown here the this is the company compound and this is the
profile fit for the available compound which is marketed. And these this is these two are
of course, the APIs active pharmaceutical ingredients they are not the materials in tablet
form. And, so the profile fit clearly shows that these two fits are very very accurate.
And in addition at that moment we came across a US patent which provided d spacing of
two other modifications of the drug and we found that one of them is this the already
patented material, the other one already marketed material. The other one the second
form which is the drug is also different from that of behavior, so we also did the whole
operation again on the second one. Even though the second one the pure API was not
available, so we could not do a thorough analysis like what we did in this case.
We find that the pattern is different and the profile fit which we could do with whatever
was available was not a not agreeing with the company sample. So, we concluded that
the company sample therefore, is a new polymorph. I am told that the company has got
this happened last year, so am told that the company has gone ahead and filed a patent I
do not know what the results are, but this is something which we did for the that
particular company.
807
Then there is another case, there was the pure API which was made by a company, there
were tablets made by the company and there were also tablets from two other companies,
who were actually manufacturing and selling that product so, the company which we
have, which approach to us wanted to know whether they have a new polymorph.
So, this was a little more challenging task because two companies have made these
compounds and those two companies the powder diffraction pattern is shown here, if you
look closely of course, we will overlap it you may say that they are one and the same. So,
what was shown here in fact, below is a more clear cut description of how these peaks
are developing, maybe the crystalline quality of one is slightly better than that of the
other but these two are one and the same.
And this in fact, is the polymorph among several other polymorphs already reported. So,
for this particular compound there are five polymorphs four polymorphs. So, the
company thought that they have made a polymorph which is different from, these two
fellows, these two fellows have made a polymorph which is characterized as polymorph
II.
So, the company which came with the data I wanted to know whether this particular
polymorph is different their particular compound is different from these two. Obviously,
from the comparison here and also if one does the profile fit we see that these two
components are one and the same and they belong to the so called polymorph II.
808
So, what was done was the data of all the polymorphs are listed here. So, you can see
that the in the patent it is a Indian patent, there are 1 2 3 4 polymorphs I listed them
slightly differently just for comparison sake. You see the polymorph I here, polymorph II
here, III here, polymorph IV there and polymorph II here this is the reported in the
patent. And these are the so called the 2 theta values for the wave length which is
common in all of them they are given the 2 theta values, you see for polymorph I it is
5.3, 8, 9 and so on.
Polymorph III is different it is 5.6, 12, 14 and so on; polymorph IV is 4.6, 6, 9, 11. So,
they are different from the point of view of comparing the individual peaks and
polymorph II is 5.5, 6.8, 10.8 etcetera. Now if you look at the pure form of the this name
of the compound is this Ticagrelor, this particular compound was given to us by the
company and this is the values of 2 theta. You see that the Axcer and Brilinta also have
the same or nearly the same values of 2 theta; we did a profile fit, but we knew
beforehand that they do not have a new polymorph.
They have a polymorph two all three companies have the same polymorph two this
disappointed the company, but this is this shows the power of analyzing the structures by
powder diffraction particularly in identification of the polymorph. So, the world of
caution is that we should not take two samples overlap with each other and say we have
that new compound have a new polymorph.
809
So, the conclusion should be done by doing a profile refinement, if structural data is
available to the Rietveld refinement. And of course if we can go the single crystals, no
doubt that you can get the structure to the accuracy with which you can argue that is a
new polymorph.
So, what I will try to do in powder diffraction therefore, in the last maybe an hour or so
is essentially to give you the basics of powder diffraction the indexing procedure and
based on the indexing procedure. We have gone ahead and talked about the nature of the
profile the way in which you can do the profile fitting and then the inert the this the
decomposition of the pattern and followed by the profile refinement and also eventually
the Rietveld refinement if structure is known.
We have also discussed where the sources for structures can come from the databases,
from literature and also from the fact that there are similarities between earlier observed
structures and the new newly made structure. If you are aware that, we have done only a
doping or replacement or a substitution on a given compound and if the cell dimensions
are nearly the same, we know now that there probably isomorphous in nature, so we can
use the same structure information.
So, basically we need a starting model either we do single crystal structure or we do this
we need a starting model. So, on once we have the starting model we can go and do the
refinement. So, the refinement procedure that we use therefore, either in single crystals
or in powder, in principle also tells whether we have this right starting model or the
wrong starting model. Suppose we have the correct starting model then the refinement
will proceed smoothly we can keep on improving the parameters, so you are accuracies
in positions and thermal parameters will improve.
The structural details which you can get in terms of bond lengths angles etcetera will also
improve, but that is not the situation which will exist if the starting model is wrong or for
that matter if your assignments of atoms is wrong what you called as carbon is not
carbon, but it is nitrogen and things like that this can happen in organic systems you can
probably replace the position of nitrogen with a carbon in your structure and R factor is
looking very good.
So, what you have to therefore, do is you always have to do after the final structure is
finalized a difference Fourier map. There are two advantages of doing the difference
810
Fourier map, the first advantage is it will now check whether all the positions of the
atoms we have determined with the thermal parameters are so accurate that you will not
be left with any density which has not accounted for in the neighborhood of that atom.
So, if there is a density which is left in the neighborhood of that atom, the assignment
you have given to that atom is probably wrong.
So, if there is a mistake made between a carbon atom and a nitrogen atom it is very
easily seen in the difference Fourier map. Difference Fourier map will also assist you in
identifying the positions of the hydrogen atoms because hydrogen atoms now we will
start so are showing up one once the structure is accurately determined. There are
situations where hydrogen atoms probably are not determined accurately even by this
approach; then there is stereo chemical way of fixing the hydrogen’s which we discussed
already along with the so called riding hydrogen refinement process procedure.
So, as they add the heavier atoms change positions the hydrogen’s will ride with them.
So, considering all this we atomic identification and the determination of positions and
of the atoms and all that are now available and we always have the Fourier analysis to
verify the accuracy that is associated with the structure.
811
Lecture – 59
Quantum Crystallography 1
So, having seen the issues related to powder diffraction and the way we can use powder
diffraction data. We will look at another aspect you see we have always been stressing
on the fact that the diffraction which we observed in a diffraction experiment is coming
solely due to the electron density.
So, in principle it should be possible to map the electron density, once we solve the
structure which is what we do when we plot the Fourier diagrams and then from that
derive the positions of the atoms and do the so called R type diagram and so on. So, this
heralds the area which has been named as Quantum Crystallography in 2017 in fact, this
is the one would think is the youngest member in crystallography regime because it has
been christened in 2017 IUCR meeting and this is an area which is now pursued
vigorously.
What is very interesting is the very first idea in crystallography, when people like Bragg
and of course, before that form Laue indicated that the diffraction is a consequence of the
electrons electron density around the atoms which are inside the crystals. When once that
812
idea came up it was almost taught in an immediate fashion that one should be able to see
electrons in orbits. In fact, that was the ambition, it was predicted also that one can see
electrons in orbitals.
And so there were some experiments which were done by physicists at that time to see
whether we can actually plot the electron density and see the distribution of electron
density associated with where it orbits; because at that time the theory of the orbital’s
and all that work there means Bohr was there who gave the K L M shell concepts. So,
people were thinking that maybe it is possible to see the electrons inside the K shell,
inside the L shell and so on.
So, they thought this is the ideal experiment which will allow it to happen. So, the idea
of mapping the charge densities from those days 1913 onwards has been there. The
problem was that as we already know that if we want to see very subtle electron density
distributions particularly, the distribution of electron density in bonds and molecules and
also in the inter molecular region and so on.
Because of the fact that the amount of electron density that is present is so small that it is
necessary to have very high quality experiments performed and that level of experiments
could not have been performed in those days, so it has to wait a long time. The other
aspect is of course, the fact that we should have very accurate data sets in order to see
those very subtle effects of remaining electron density other than the molecule.
And so these things took a long time, so long that the actual electron density ideas that
particularly charge density mapping and so on were developed in the early 70s and since
80s. So, early 70s saw the resurgence of this activity in the area to map the electron
densities and this originated from the fact that we can have both X ray experiments as
well as neutron experiments performed on the same crystal.
If we perform the electric X ray experiment as well as the neutron experiments we have
certain advantages. Because the neutron experiment will give us the variation of the there
is no variation of sin theta by lambda of the scattering length, on the other hand by X
rays give the variation is sin theta by lambda.
Now, we know in what way this is going to help? We will have a quick look at that. I
will just draw a small diagram to show how this would have happened. Say for example,
813
if you now see f and sin theta by lambda you have sin theta by lambda, the it will fall off
asymptotically and go toward 0 at higher angles, If we now separate the core and the
valence that is associated with a given atom, so this is let us say carbon and that means, it
will starting point is 6 electrons, we will put two core in the core and four in the valence
because we know carbon has a valence of valency of four.
So, these four electrons now if we now follow the fall of f with respect to sin theta by
lambda it will fall off and in fact, go towards 0 somewhere around 0.65 and the core
electrons which are here two of them, they will continue to fall, but then they will sort of
have some nonzero value beyond that particular point. In the case of a neutron diffraction
of course, we have a straight line which is associated with the scattering length this not
cannot be plotted on this diagram obviously. So, it is depends on the scattering length,
but that is not varying with respect to sin theta by lambda.
So, suppose we conceived of experiments let us discuss it from the experiments point of
view and then bring in the quantum concept a little later. So, if we can conceive of
experiments where we can measure this region and that region as you see will have very
little contributions from various atoms and therefore, the diffraction is weak. Now, we
have to make this region to have better diffraction. How are we going to make this region
to have better diffraction? We have to then have two issues one is of course we can
collect low temperature data.
So, when we collect the low temperature data it is better to have estimates of higher
angles and then we have to measure these reflections at higher angles. In a typical X ray
experiment what we do is we collect the data of let us say copper radiation we go to 2
theta up to 60 degrees and that represents, the limit of the limiting sphere or nearly the
limit of the limiting sphere. That would mean that we are going to values of sin theta by
lambda only around this value of 0.6 or 0.65. So, that would take the contributions
coming from the core completely away from the calculation.
So, effectively since we now consider that structure of scattering factor which is starting
at a value of 6 and falling off to 0 toward 0, there is inclusion of the core in some sense
in the plot which we have done and therefore, we can get the values. But realistically
these values are not very strong, so the reflections the reliability with which we can
measure the reflections is not very strong.
814
So, the third process was now to cool the crystal to below nitrogen or towards helium
temperatures, in which case some of these measurements probably can be made.
Effectively the data which we collect now should go far beyond the 60 degrees; that
means we have to look for sin theta by lambda which are going beyond 1 or even 1.2, 1.3
and so on.
So, now we are considering a situation where we are trying to get the maximum data
which we can get from a given material which is in the form of a single crystal and that
single crystal therefore, has to be an extraordinarily good quality single crystal and in
terms of diffraction. And then we collect the low temperature data and the low
temperature data effectively brings in the possibility of measuring these what we call as
X high order reflections.
815
Now what is the example of the X high order reflection? The advantage of this X high
order reflection is that we have now the contribution only coming from the core number
1, and number 2 the contribution is not varying with respect to sin theta by lambda. So,
effectively it is like the delta which is coming out from a neutron diffraction, only factor
is that we have to consider is that these reflections are beyond, the normal range of data
collection which we normally do for single crystal work. So, that is why we call it the
high order data.
And using this idea we can actually look at the possibility of measuring the electron
density beyond the region where the valence contribution stops and therefore, effectively
you are now looking at the inside of the atom. And So if you now do the difference
Fourier plot, that difference Fourier plot in principle should give us the electron density
which is coming from this high angle region, which is coming essentially from the core
region and therefore, in principle we should be able to see the bonding densities.
So, that was the logic which was used in those experiments and such experiments were
initially performed with lot of care and people also used the neutron data on the same
crystal that is because with the case of a neutron data we have no variation with respect
to sin theta by lambda. So, we could get as much diffraction as possible from the crystal.
Again as we mentioned in the morning regarding the powder diffraction the amount of
sample that is required.
816
In case of neutron diffraction also the crystal size should be almost centimeter size
because the strength which these moderated neutrons come out they will just pass
through the crystal if we want a normal crystal which we use for XRD, the signal crystal
X ray diffraction which is 0.3 millimeter cube. So, if we go towards 1 centimeter; that
means, we have to grow very; very large crystals and that becomes a rate limiting step.
Instead we use the same X ray diffraction on the same small little crystal, but get the
high order data augmented we can replace the neutron results from the X high order
results and this was known as the X minus X high order analysis and people calculated
these maps. Now, what do these maps give when we subtract the overall data from the
data which is coming only from the higher angles, the higher angles are essentially
representing the core electrons and therefore, we can determine the positions x y z more
accurately.
So, if we use the X high order data and calculate the structure factors, those structure
factors will be now coming more or less from the core and therefore, they will be very
accurate. So, now, we subtract from the observed data, the data which is coming from
higher order, then what is left out will be the electron density in the valence region. And
since we are now looking at the electron density in the valence region we are able to see
the what is the amount of electron density that is there between the bonds and the density
which is between molecules inter atomic interactions.
So, that is the logic which was used in the experimental aspect of charge density
analysis. And at the same time collecting the data at low temperatures ensures that the
thermal vibrations are minimized and as a consequence we do not have to worry about
the fact that, the thermal vibrations which also give wrong positions or rather positions of
the atoms which are a little removed than what they are where they are you know
because of the lattice vibrations that gets reduced.
So, the accuracy with which we can determine the atomic positions increases extensively
when we collect the high angle data and use the so called X minus X high order
refinement. So, the early days of charge density 70s in fact, 60s, 70s and early part of 80
also you will see large number of literature associated with X; X high order refinements
X minus n maps X minus X high order maps and so on where people have tried to look
at the bonding densities.
817
And the success of that is mainly because of the fact that when we used the high angle
data, the data is devoid of the fall of f with respect to sin theta by lambda it is more or
less a straight line.
At this stage quantum mechanics and quantum chemistry came into prominence over the
years and people are became aware of various kinds of ways in which the contributions
from atoms can be calculated and so we one could actually calculate wave functions and
the square of the wave function if we take we are aware of that aspect we will get the
approximation towards the intensity associated with the X rays and we now know that
takes the intensities come from the planes in a crystal, so I of h scale will be the source.
So, keeping that in mind when the models have to be constructed one of the thoughts
which I occurred was we will discuss it when once I once I highlight the basics of
quantum crystallography.
So, just like the you know we have the quantum chemistry fundamental laws and rules
we have fundamental laws and rules followed by form for quantum crystallography and
that you see is formed only very recently in the year 2017. So, it is a very infant area of a
very old thought process, it is a area is infant, but the thought process is very old because
it was very beginning people thought that it should be possible to see the electrons in
orbits at the time X ray scattering was realized.
So, there are two possible approaches in quantum crystallography as has been formulated
and in fact, it is at this moment spearheaded by a group of people several groups around
the world are working on it, one of these groups is the consist of three people there is
Simon Grabowski, Alessandro Genoni and Hans Beat Burgi. So, these three gentlemen
wrote this article which is interaction to the new area of quantum crystallography.
What they discussed was that we can look at this area into two part possibilities; one is
we look at models that we can develop for charge densities in crystals and of course, this
model will incorporate electron density distributions, atomic positions and atomic motion
using high resolution diffraction experiments which we just now discussed can be
improved based on quantum chemically calculated, tailor made electron density.
So, what was suggested was in this approach, that we now calculate the electron
densities and this calculated electron density is essentially the form factors which is also
the scattering factor which we have discussed just now it is also referred to as the form
818
factor. So, we can actually from quantum mechanics calculate these form factors and
these form factors which we are going to calculate will include the concept of questions
like what happens to an atom when it is put in a which is a part of the molecule and what
happens to the molecule when it is put inside the crystal?
So, two fundamental questions get asked what happens to an atom when it becomes a
part of a molecule. Obviously, the electron density of atom one and electron density of
atom two gets shared, it could be a covalent sharing, it could be an ionic sharing or
whatever. So, there is a certain amount of electron density which is going from one atom
to the other; that means, the carbon atom will no longer be having six electrons some
amount of electrons or either donated or transferred to the region between the two carbon
atoms or it is transferred to the next atom which now becomes an electronegative atom.
So, that way the propagation of electron density takes place and we form a molecule.
So, sharing of the electron density between atoms in general will allow for sharing to
form the molecule and therefore, these molecules now exist in crystal structures and they
obey the rules which we have formulated in the entire course the rules of symmetry. So,
depending on how the symmetry directs these molecules, these molecules arrange
themselves inside the unit cell and there will be a possible interactions between one
molecule and the other.
These interactions in fact, lead to the area of crystal engineering where we can actually
identify the nature of these interactions and see whether either they are highly directional
or whether they are highly attractive or repulsive depending upon the nature of the
interaction, which allows for us to understand the concepts of how crystals get
engineered through molecules how these molecules arrange themselves inside the crystal
lattice. And so the concept of hydrogen bonding, the concept of halogen bonding and
other kinds of bonding issues come up.
So, basically the idea is to use the quantum mechanical modeling, one of the early
models which we are going to discuss in detail in the next few minutes will be the one
which is called the multipolar refinement approach where we develop a multi polar
model associated with an atom. Obviously, now atom cannot be a spherical atom in this
modeling, it has to be distorted.
819
The atom should be distorted in such a way that it is now oriented it’s itself to the
connecting bond a bonds it should be directed towards connecting bonds and then certain
amount of electron density should be going out, certain of amount of electron density
should be coming in. So, that the overall molecule remains neutral. So, the electron
neutrality of the molecule is result and as a consequence there is a redistribution of the
electron density and that can be estimated from quantum chemical calculations.
So, the possible approach is that we have the charge density model, the charge density
models will be developed, incorporating the electron density distribution, atomic
positions and atomic motions using high resolution diffraction experiments. So, this is
what we do to get the charge density models and these can now be improved with
quantum chemical calculations. This is one way of redefining the behavior of f versus sin
theta by lambda. So, it will change accordingly when the atoms bond to each other.
The second approach is we can calculate wave functions from quantum mechanics and
those wave functions always are approximate, they are not final. We know that apart
from hydrogen atom the other atoms quantum chemically quantum mechanically we
cannot fully solve for them and therefore, there are approximate wave functions.
These can be improved by constraining them to reproduce observations derived from

diffraction and scattering experiments. So, we have these experiments done, we have this
higher resolution experiment done where we are collected data beyond sin theta by
lambda, we have done the Fourier analysis, we have located the atom positions and so
on.
So, these are the observations that are already a part of the diffraction experiment and
therefore, they represent the chemical structure. So, the chemical structure is available
and therefore, these wave functions now can be improved by including this information
from the experiment. So, one is the improvement of quantum mechanical observations
using the experimental data which are very accurate, the other one is the use of the
models the correction to the models that can be introduced when once we have quantum
mechanically calculated electron density.
So, quantum mechanics and crystallography these two get combined and therefore, this
area has the name quantum crystallography. In 1995 in fact, it was defined
crystallographic information to enhance quantum mechanical calculations and the
820
information derived from them. So, we are using the crystallographic information to
enhance the quantum mechanical calculations in a given molecule. So, how best we can
use this approach how best we can do this we will see in the coming slide.
So, two approaches as we just now discussed one is the first aspect where the
crystallographic data are integrated into quantum mechanical-chemical calculations and
the information content of the wave function. So, we have quantum chemistry, we have
crystallography, they build it together into quantum crystallography which will allow for
experimental wave function. See one of the things which we can we have the problem
we have when we get the experimental wave function, this will have all kinds of errors
that is because we have made the first of all we have made the spherical atom
approximation.
Secondly, we have these errors in phases we do not have the actual phases. We estimated
the phases the phase problem in crystallography persists and therefore, the experimental
wave function which we get has to be going towards the actual true wave function. So,
we can use this data and then try to see whether we can take this over to that that is one
approach.
And the other approach is now we have the crystallography and quantum chemistry
coming together to quantum crystallography. Here what happens is that we end up with
improved dynamic charge density, see what we are doing from the charge density
experiments is to map the electron density after removing the presence of the atoms
which participate in forming a molecule.
So, as I mentioned what happens to an atom when it becomes a part of a molecule, there
are certain amount of electron density is shared between the atoms. So, when we do this
charge density mapping, we remove the atom contributions and look at only the
accumulation of the charge density in the regions between the two atoms the topology of
the electron density is what we analyze. And that is subject to again considerable amount
of errors because it depends upon what model we are going to use.
So, that with quantum mechanics input will give us the improved charge density. We put
here dynamic because that is very crucial to have as let us say the change of if there are
some changes which are brought about by shining light onto the crystal, photo
821
crystallography is an area which has developed in recent years. So, we can see the energy
changes and then therefore, calculate the dynamic charge density in such systems.
So, effectively this will now go and lead us to the truth charge density in the model this
is what is our approach. Experimentalist look for this theoreticians look for that and you
see that the marriage between quantum chemistry and crystallography therefore, leads to
either of these two successes. So, that is the idea of quantum crystallography and so we
see that this is the way it gets defined.
In fact, quantum crystallography and the quantum chemistry and crystallography are not
very far from each other, this particular diagram shows how quantum chemistry and
crystallography have contributed to various aspects of theory and experiment. So, here
you have the X ray diffraction data, Compton scattering data, polarized neutron
diffraction data and then you can go up to a the single determinant wave function, energy
minimization many determinant wave functions.
So, there are lots of experimental and theoretical calculations which I will develop in
between and therefore, this is a very straightforward link which brings in quantum
chemistry and crystallography or quantum mechanics and crystallography together right.
ρatom is the charge density distribution of an atom which is divided into ρ core (core electron
density) and ρvalence(valence electron density).
822
Having seen that how does this all happen and how are we coming into the picture as
crystallographers. What we do is we most of the structure which we determine is based
on assumption that atoms are spherical this is what we have done in the structural
determination protocol using both the single crystal methodology and the powder
diffraction. So, we assume that the atoms are spherical and they keep their asperity even
when they are bonded which is a rather not so acceptable concept, but we believe in that.
And then we so therefore, model the electron density as though it is made up of a

electron density from the core and electron density from the valence. So, this is a
spherically average density and it is referred to as the independent atom model. So, the
atoms are all independent of each other this is the routine X ray diffraction.
ρat(r): Atomic density at a ponit r.
Pc and Pv: Population of core and valence electrons.
κ and κ': Expansion and Contraction coefficients of core and valence electrons
repectively.
Now; obviously, it is not true so we have to put in the information about the bonding
features. How do we do it? One of the methods which is very popular which was
823
developed by Hansen and Coppens way back in 1978 and this is in fact, the way in
which most of the charge density analysis until now are being done. So, a very few
which are using other methods, but this is by far the most powerful and well documented
report because the charge density analysis is easily approachable by and is easily
programmable by this approach.
So, here you see that the electron density at the atom now is no longer just core plus
valence, the core is associated with a certain population Pc that is the number of
electrons with the core and Pv is the number of electrons with the valence. So, we
therefore, now have an expression for the electron density which has the part associated
with this and then the spherical part. Now when the bonding is taking place the atom will
expand and contract.
So, it will contract in the region where there is contribution coming from some other
atom and it will expand when it has to push the contribution to some other atom
eventually to form the molecule. And this expansion and contraction is referred to as the
radial distribution or radial part and that is this part. And this part of the expression now
joins the definition of the electron density and this part of the expression tells us this
value of kappa tells us how the expansion and contraction is associated with this
behavior.
Now, apart from that we also have the reorientation, so it is not spherical the spherical is
a big approximation again, but definitely not so much of an approximation as
independent atom model where we assume solid atoms atomic spheres and they are like
billiard balls and you cannot bend them. In this case the expansion and contraction
parameter will allow for the possibility of forming bonds.
In the early days that was what was developed again from Philip Kaufman’s group and it
so happened that I happened to be a part of that group at the time, so I was also involved
in the development. But that was essentially the expansion and contraction of the atom
it’s known eventually as a radial refinement protocol. But then we also realized very
quickly that there is a as spherical component. So, to build in the spherical component
this additional term is added which is their spherical part.
So, effectively now the atom is described as a realistic atom, so this is now a realistic
atom in a molecule. So, when the atom becomes a part of a molecule this will be the
824
value of the electron density and not the value of electron density that we use always for
determining the crystal structures. So, this is just ρcore plus ρvalence whereas, as you see
here ρcore plus ρvalence plus this radial term and then this is known as the spherical
harmonic term which essentially takes into account the asphericity that can be associated
with the atom.
Unfortunately there is not a enough time to go to the fuller details of this except to give
you the highlights of what kind of results we can get from this kind of an analysis, one of
the things of course, is we get very highly accurate crystal structures. So, the atomic
positions now and the atomic displacement parameters, the thermal parameters are
shown in this r type diagram you see that we have a low temperature data, so the r type
plots along is also showing the presence of the hydrogen atoms also shown as
anisotropic.
So, this kind of a feature can come when we fit the equation of this kind which is known
as a multipolar expansion components. So, we have included the multipolar components
along with the polar components in some sense. So, what is happening here is that the
atoms now are actually represented of the way in which they are situated in a molecule
and so this now develops using the symmetry rules in this space group it puts the other
atoms together. So, this is the structure of a realistic molecule, I will not give you the
detail of what it is, but one once we now do this we also do what is known as a
difference Fourier of this.
Now, this difference Fourier of this is slightly different they define than what we already
defined. This difference Fourier is the electron density as is obtained from here minus the
electron density assuming that the atoms are independent atoms. So, you have atomic
positions here, so you have removed the atoms effectively and what is left out in the
electron density after we allowed for their spherical refinement is plotted here this is
known as the deformation density map.
So, it tells you how the atoms are deformed and how much of electron density gets
shared between atoms and so on. Let us look at this part you see this is the six member
ring if you look at this part you see this is an atom position here and an atom position
there, you see there is electron density which develops between these two. In fact, by
counting the number of contours these are measured plotted at 0.01 electrons per
825
angstrom cube so very small electron density, but by counting this you can also get any
idea of how strong is the C-C bond.
And you see that all these six bonds in this particular ring are about the same value as far
as the deformation density is concerned. So, this deformation density also tells you that
we are fairly accurate in our ideas about the types of atoms and the contribution of
chemistry to these type of atoms. If you look at this particular atom which is the oxygen
atom, this oxygen atom now appears with the two lone pairs. So, the lone pairs on the
electrons can be seen here, the lone pairs on the of the on the oxygen can be seen here.
The lone pair electrons have a orientation which we believe is the direction of bonding as
we know.
So, earlier days Pauling and others define these as the bonding directions, the directions
where a nucleophile can approach the direction where a new electrophile can approach,
all these ideas now can be visualized. See X ray diffraction is always as I kept on
mentioning is for visualization; visualization and realization of the reality of what is
happening to atoms. So, we are now putting our lens which we started by looking at the
image of an object you remember long ago we put a convex lens and then did the focal
length experiment.
Now we are actually looking at focal with a lens where we are looking at the density
between the bonds. So, the density between the sorry between the atoms we can also
look at the density between the bonds which are also plotted here and these red dot; dot
lines show negative contours. So, essentially we have therefore and now possibility of
viewing these electron densities at this level. What is required is a high resolution data as
I mentioned sin theta by lambda 1 and above 1.1 angstroms in this case is a minimum
requirement to do these experiments.
826
Prof. T. N. Guru Rao
Lecture – 60
Quantum Crystallography- 2
So, one of the good data sets is illustrated here so, when we do the data collection and
collect the data over the on a crystal. It is not just a few 3-4 hours dataset. The data that
is collected is now has to be complete and we have to see whether we have got the data
on every reciprocal at this point which comes in the limiting sphere and in fact, beyond
the limiting sphere in some sense.
We also measure all possible reflections we do not leave out the reflections now and as
you see from this; this is the structure of epsilon which was done in our laboratory.
Ebselen polymorph II it is supposed to be a drug. And the temperature at which the data
was collected is 100K this was done by my student who is now his name is Dr. Sajesh
Thomas and he is now somewhere in Denmark. So, he measured the data at 100 Kelvin
and the wavelength that is used is molybdenum. And as we saw from the from the table
we made for resolution limits and so on lambda by 2 is the resolution to which we can go
in principle right.
827
So, it should be half of 0.7. So, we should be able to see these bonding densities fairly
clearly in this data. Sin theta by lambda he managed to go up to 1.08 and the total
number of reflections collected as you see is enormously large for a very small crystal
unit cell dimensions. I think the unit cell dimensions are not given here, but I remember
the unit cell dimensions it is a monolithic system and it is 8 by 12 by 14, it is a very
small firm.
So, you see that we have 112503 out of which of course, all the all the redundant ones
are also measured; that means, we measure the Friedel equivalent also. So, the Friedel
equivalents do not contribute freshly to the data, but they will allow you to average the
data. So, if you now reduce this redundant data to unique reflections we have 11087
reflections. So in fact, if you look at the sphere you all sphere you will see that it is 100
percent complete data, it is all possible data points have been taken into account a very
high quality data and we are measured some reflections more than 10 times. That means,
you we have measured let us say in this space group we have measured 10 percent 10.2
times is the redundancy; that means, a given hkl is measured in all possible orientations
10 times so and then averaged out.
So, you are accuracy of the data measurement is extraordinarily high and that is indicated
by this what is known as a fractal plot. So, I have no time to go into the detail except that
this has to be symmetric with respect to the electron density the residual electron density
which is shown here. So, based on this now we go to the refinement so, we now refine
the unit cell in such a way that we consider this is at routine data; that means we do as
spherical atom refinement.
So, when we do this spherical atom refinement we reach a R factor which is 0.02 very
accurate data and rated R factor is 0.08 and the goodness of fit which should be ideally 1
it is 1.038; that means, it is an extraordinarily high quality data. And as you can see from
here if you do look at the what is left after the difference Fourier map you see that an
amount of electron density that is left is minus 0.79 electrons per Angstrom cube to plus
0.79 both are nearly equal; that means, it is a very flat map after you account for the
entire data which the entire crystal and the atoms in the crystal. Then we go and do these
multipole refinement, you see the number of reflections now used are 8038 we have
made some more filtration of that.
828
Because we now consider reflections above the threshold of 2 sigma I sigma represents
the threshold. So, the ratio see what is important in this data again is the over
determination, just like any structural refinement we had over determination. Here for
every variable which is now increased in number because now the variables are not just
xyz we have to now refine many other parameters to define this asperity.
In fact, every atom earlier had x y z and 6 thermal parameters. So, that made it 9
parameters per atom. Now, along with this 9 parameters we now consider the atom to be
a multi pole; that means, the atom is now associated with a monopole which contains the
number of electrons, then it is also associated with a dipole so which we will have 3
parameters 1 parameter for the monopole, 3 parameters for the dipole and then 8
parameters for the octopole and so on. So, we will keep increasing the number as we go
higher and higher to describe the atom. So, effectively what we do is the following. You
just consider the atom to be made up of a monopole and this particular monopole.
Now, covers the diameter of the atom and then in that way have a dipolar contribution
which will go like that with respect to the center and 3 directions, then the quadropole,
the octopole and the hexa pole upon. So, effectively we fill in all the possible space here
and the shape being aspherical; the space being aspherical we will have the
representation of the atom in terms of multipoles that is why we call the multiple. So,
829
there will be 26 additional, 9 plus additional parameters are accounting to total 26. So,
they we therefore, define refined 26 parameters per atom.
So, considering that the number of reflections to a number of variables ratio is 19. So, it
is generally recommended it should be at least 10 for a successful determination of
bonding density. So, it is quite over determined in that sense. So, the R factor here you
see is 0.016; that means, the error that is involved is less than 1 plus 1.6 percent, you can
express the R factor as 1.6 percent; that means, the 1.6 percent is your R factor the
weighted R factor is 3.7 percent goodness of it is 0.87.
In fact, if when once we compare this to F of x minus F of calc you see this line where in
which we have plotted every reflection that has been measured and you see if I have the
beautiful straight line saying that there are not much deviations in the individual F of
measurements and calculations. So, this essentially represents a very accurate data it is
P21/c as I mentioned it is a monotonic system.
What are the experimental requirements? How do we collect the data in such an
experiment? The experimental requirements are, we can use a conventional sources good
laboratory source and the preferred radiation is molybdenum. Of course, we can also use
silver because the idea is to get the sin theta by lambda maximum to as high value as
possible. Obviously, use of copper is disallowed because it is just up to 0.65, we need to
830
have data beyond 1.0 of sin theta by lambda and so molybdenum is the choice if you are
doing charge density analysis.
We can use synchrotron radiation and tune the radiation in any range between our silver
and molybdenum. So, that is also good news we can use rotating anodes as well the
detectors could be CCD image plate I must be adding here CMOS detector which is
nowadays the most suggested this detector on diffractometers. The low temperature data
we can use either nitrogen or helium we use the nitrogen which is a about 100K because,
in all these there is 1 issue there are 2 issues in fact. The 2 issues are they were 2 which
are ticked. Only when we can tick these two we can do the charge density analysis and
therefore, aid in the development of the ideas of quantum crystallography.
So, we have to have good quality crystals that is the first requirement, the second
requirement is that we should have a good electron density model. So, when once we
have these two, this is the multipole model which we just now discussed, if we have
these two properly done then we can go further.
And now we can bring in the aspect of combining theory and experiment. One of the
early ways of doing this was sorry about this overlap it is university of Torino this
reference is 2006. But we have now, what is known as crystal 17, I told my students to
get it, but they have not shown any further interest on that. Here we have crystals 17 and
831
the crystals 17 is the one; which we should be having and the theoretical calculations can
be performed and we can use the package XD which is the charge density package.
So, when we take the charge density package and use this program crystal from the
charge density package we feed in the coordinates the lattice parameters and the basis
set, then periodicity and symmetry are brought into the system, we can use Hartree Fock
and DFT etcetera to calculate the way functions and reflections. So, the advantage here is
we end up with theoretical structure factors.
So, if we get theoretical structure factors they are devoid of all systematic errors we do
not have to worry about absorption, we do not have to worry about extinction
corrections, we do not have any issues related to WLP and so on. All whatever errors we
have had in the routine structure determination are avoided. So, the theoretical structure
factors of the ones which we now use and of course, we use the XD protocol refinement
that is the charged density refinement and process the data.
There are two ways in which we can do it one is of course, we have the wave functions
this is the quantum crystallographic approach. So, we have the wave functions, we can
go back here recalculate these values and then feeding to crystal and then develop the
theoretical structure factors of it. So, or we if we have the reflections these theoretical
structure factors are essentially the reflections. We can therefore, cycle these out such
that we get finally, a high quality theoretical structural factors which can be taken to
refinement. So, this is the approach which is there in crystal package.
832
And now we have to see how to see these atoms. So, you see we have done the map the
multi pole modeling, we have accounted for the overall electron density and so on. But
how do we see these atoms? There is a theory due to Bader which is referred to as atoms
and molecules there is a book which is atoms and molecules A Quantum Theory
published by Oxford University Press in 1990, this was written by Richard Bader.
So, in this approach what we do is, we do not look at you see nobody told us that when
we say a bond between an atom let us say one carbon and another carbon. Nobody told
us that there should be a straight line between these two, which we call as the bond
length, but there is nothing called a bond length ok. There is an electron density between
the 2 atoms and it is shared and the electrons do not know that they should go in a
straight line nobody told me to go in a straight line.
But we measure the straight line distance and say it is bond length and then is the same
thing is true with the bond angles then we take the 2 lengths and then say this is the bond
angle. What really happens is that the electron density gets distributed in this region. So,
it is now a 3 dimensional issue of the electron density distribution and therefore, we call
it as the topology associated with the bond.
So, effectively you see here if you take this as a carbon atom and the end other end as a
carbon atom this is the C-C bond, the electron density is distributed inside this cylinder.
It is not necessary that it should be a straight line it can be any anywhere in this
833
particular cylindrical entity which is characterized by three quantities lambda 1, lambda
2, lambda 3; which are essentially elements of what we call as a Hessian matrix.
Theoretically what it means is that we have the true chemistry of this bonding is
represented in terms of an expression which is the second derivative of the electron
density at the so called a critical point; the critical point is one in which the first
derivative of the electron goes to electron goes to 0. So, between these 2 atoms the first
derivative of the electron density goes to 0 at a critical point and whenever there is a
critical point between 2 atoms in principle we call it a bond.
And if there is a bond that the critical point del square rho value should be lambda 1 plus
lambda 2 plus lambda 3 and there are different kinds of bonding features and those are
indicated here. You do not have time to explain every one of them except to tell you that
these are actually tenses of rank 3. So, if it is 3 minus 3 it represents an atom position, 3
minus 1 represents a bond position and then 3 plus 1 represents a ring and 3 plus 3
represents a cage.
Effectively we are now trying to see what is the chemistry in this bonding? Rather than
look at the straight line distance between atoms and so on which is important which we
which is important to say find out how accurately we can determine those by a straight
line distances. But what is important is now to see the electron density and the how the
electron density is distributed between these two bonds.
So, whenever this value of del square rho c which is called the Laplacian is less than 0, it
represents that the 2 atoms here involved are sharing the electrons, it is a covalent bond
what is defined as a covenant bond by Linus Pauling is exactly the fact that the del
square rho c is less than 0. If del square rho c is greater than 0 it represents closed shell
interactions; that means, if you have let us say a molecule here and a molecule there and
these two molecules are interacting through a hydrogen bonds or some other Van der
Waal interaction.
This essentially represents the limit of the Van der Waal interaction; that means, there is
no density developed at the critical point which is now del square rho c greater than 0. In
both the cases there is no density of the first derivative of the electron density that goes
to 0. We also now defined because we know these values lambda 1, lambda 2, lambda 3,
these are the eigenvalues of the Hessian matrix as I mentioned.
834
These are all terms which I am using freely, these things need a full another 30 hour
NPTEL course in which we can systematically go and define all these factors. But what
is important from the chemistry and what is important from this particular course
structure point of view, structure and symmetry in the solid state? We are now trying to
see the structure in the bonding regions and the structure in the regions between
molecules and see if symmetry is still there and the how the symmetry is directing the
electron density distributions. So, from that point of view we are discussing this issue of
quantum crystallography.
So, if we define a value epsilon which is referred to as the bond electricity, as I

mentioned again that as we mentioned again and again it is not the central straight line
distance between the 2 carbon atoms. So, this bond ellepticity tells us the nature of the
bond. So, if the epsilon is exactly equal to 0 it represents a sigma bond, if epsilon is
greater than 0 it represents the nature of a pi bond.
The fact that we can use multiple atom modeling and use the multiple atom modeling to
look into the topological distribution is a very nice idea because, we now know whether
the given chemistry can be represented for a molecule and so on. We will take examples,
but before we go for such examples I just wanted to introduce the other aspect which is
now becoming very popular this is known as the Hirshfeld atom refinement. In this
835
particular case what is done is to determine the structural parameters from single crystal
diffraction data with very accurate possibilities of course, independent atom module.
So, the structural parameters are from the independent atom module. And then we use
the aspherical atom partitioning of tailor made ab initio quantum mechanical molecular
electronics densities a big word big phrase. So, let me tell you what it does. Now what it
does is now that we have a molecule and let us say we are looking at inter molecular
space. So, there is a molecule a and the molecule b and we want to see what is the type
of interaction this molecule a can have with the molecule b.
So, there are different kinds of atoms, there are atoms which are electro negativity atoms,
which are electro positive atoms which are neutral and so on in this molecule and so also
in the other molecule. So, if you now consider even the case where we have only 2 atoms
and for a diatomic structure let us say and this diatomic structure the interaction between
2 atoms. So, effectively you replace the molecules by single atoms. The atoms between
these two the distance between these two of course, is our bond length, but the fact that
there is a certain amount of overlap of the electron density from atom 1 to atom 2 and a
certain amount of overlap from 2 to 1 in the topological region defined.
When we use the multi pole atom modeling the problem is that the multi pole atom
modeling does not restrain the extent to which electron density can be penetrated
between the 2 atoms position so; that means, there must be some position. So, the
position is defined by what is known as atomic basin in which we have let us say an
atom in this case we can use this diagram.
So, we can have an atom here and an atom there and del square rho the first derivative of
electron density goes to 0. So, if you now plot the electron density along this particular
direction the first derivative of the electron density, when it goes to 0 that represents a
maxima. So, the actual amount of electron density that is at the middle of the bond is
maximum and then it falls off towards the other atom it falls off towards the other atom.
On the other hand when we do the multi pole atom modeling, the multi pole atom model
will have the populations of the higher terms, higher multi poles which are extended
beyond this region where there is a there is an atomic basis. So, we have an atom we
have an atom and between the 2 atoms. So, let me write it quickly because of lack of
836
time. So, we have an atom here and an atom there. So, the electron density will go like
that. So, this is the so called critical point.
So, the first derivative goes to 0 and this essentially means that these two atoms are held
with respect to the electron density region in this topological space. So, the expressions
which we have for the multi pole refinement modeling will extend beyond this particular
point which is strictly not allowed because this is the place at which these two should be
separated from each other. So, whatever contributions that are coming from multiple
modeling has a tendency to penetrate beyond this.
And therefore, there is an error that is associated with multi pole modeling this is gets
avoided when we do a different kind of modeling and that again has been developed by
using mechanical molecule electron density. So, if we now consider the molecule as a
whole and calculate its electron density, there is a certain form factor which can be
straight away attached to the molecule as though it is a some kind of a thermal
parameter. And so, we have a thermal parameter surface 1, thermal parameter surface 2
which now interact with each other.
So, the atomic based in now can be better defined. Between the 2 methodologies there is
not much of a difference except that we have therefore, I have written here the difference
between multiple expansion and Hirshfeld surface it is in the partitioning of space
between atoms. And this in fact, allows for the competition of hydrogen atom position.
So, we can actually get compute the where the hydrogen atoms are lying and also they
can get their corresponding atomic displacement parameters.
So, this is a limitation in the X-ray experiment we said that we cannot determine the
hydrogen positions, but we use this approach and use the either the multipole modeling
or this one preferably the Hirshfeld surface modeling, then we will get the hydrogen
atom positions. So, in some sense this methodology of Hirshfeld atom refinement is
similar to that of neutron diffraction.
837
I think I should not go into further detail of this or for that matter multipole model,
instead what I will do is, I will start taking examples which will make your life
comfortable because you will understand it better and in of course, you all know the
story of Gulliver.
838
Prof. T. N. Guru Rao
Indian Institute of Science Bangalore
Lecture – 61
Intermolecular Interactions
Of course you all know the story of Gulliver right and this fellow was a big man
compared to the small Lilliput fellows who are all here and what happened is when his
boat crashed and he landed on the beach. When he really came to his senses he found
that he cannot get up, he could not get up because he has been tired with so many of
these his hair was tied and tied it up here so that he so many contacts were made and
these people tied it up in such a way that he cannot get up.
That means, what happens to a molecule in a crystal is this is exactly what it what
happens because of the crystal forces the interactions that are involved between
molecules and so on the weak bonding which are essentially the way in which they are
tied is hair here. For example, these are now the strong medium and weak interactions
and weak interactions essentially occupy the intermolecular space.
839
So, since we are talking about the modeling that we can do for the molecule as a whole
now in terms of multiple refinements as well as the Hirschfield atom surfaces, it should
be possible therefore to look into or take a peep into the intermolecular space. If we look
into the intermolecular space we will be left with a certain amount of electron density
and this electron density is crucial to keep the molecule in it is position which is defined
by our loss which we have formed in this course the symmetry and also keep it in that
particular space group.
So, the way in which this is tied up; this fellow is tied up can be modified and you may
have a different way of tying up molecules with using different interactions then the
crystal structure will change and so you can get polymorphic modifications. So,
essentially the weak and strong interactions, therefore based on energy criteria can
actually be analyzed if we do the charge density analysis.
Here is an example, a very quick example now we go through quick examples because of
lack of time, we will have to wind up the course and then I will stress essentially on
some weak interactions which look very critical and crucial for understanding the weak
interactions and their importance.
So, here is the structure of ferulic acid it was solved long ago and it has very strong
hydrogen bonds, we are not worried about the strong hydrogen bonds because they arrest
the molecules in their positions corresponding to the unit cell requirements and also the
840
symmetry requirements. But in addition to the fact that they are located interacting this
way we find some additional interactions which are essentially of the type C-H O, there
is a lot of initial discussion about whether C-H O can be a hydrogen bond or not, because
the polarization that is associated with the carbon atom is minimum.
So, if I said C-H O interactions exist at all was a question and if it does exist does it
follow the hydrogen bonding features like directionalities, because a very strong
hydrogen bond will have this angle 180 degrees and this particular angle let me show
that correctly. The angle C-H...O will be 180 degree for a very strong hydrogen bond and
normally such bonds do exist in crystal structures where we have for example,
carboxylic acid diamonds, carboxylic acid diamonds ensure very strong hydrogen bonds.
So, whether C-H...O will have such an interaction was discussed and finally with a lot of
arguments involving crystal engineering principles it was arrived at that C-H...O can be
called a hydrogen bond. Now our interest is that can the C-H...O would really hold the
molecules together and what is the energy that is associated with it, if we can analyze
that in terms of the charge density analysis. So, we have looked into this region there are
3 possible C-H...O hydrogen bonds, the lengths and the angles are given here and we
wanted to see the distribution of electron density in this.
When we analyze the distribution of electron density in this as I told that Ferulic acid is
held by a very strong carboxylic hydrogen bond, this decides the orientation of the
841
molecules in the crystal essentially this is at the center of symmetry of the crystal. So, we
have this very strong hydrogen bonds and the strength of it is minus 47.48 kilocalories
per mole, this calculation has been done based on charge density analysis I am not going
into the detail of it but essentially we will see it is a very strong bond, the larger is the
negative it is stronger it is.
Then in addition to that there is a between the 2 molecules there is a strong reasonably
strong O-H...O hydrogen bond and then of course there is a simple C-H...O bond with
the methyl group. So, we have calculated the energies for all these this work was done by
again Thomas in our group along with power these two gentlemen graduated of out of
my group recently.
See the you see that when you consider the energy that is associated with this trifurcated
C-H...O bond there are 3 bonds do you know this is the Gulliver idea, see if the lipids
can tie him down with many such small interactions this is exactly what it does. So, these
2 molecules are now tied down with this very weak interactions they are weak
individually, but collectively they generate an energy which is minus 15.87 much much
higher than the energy that is involved in a typical O-H...O bond.
So, one would think that a typical O-H..O bond would be stronger when in fact it is
stronger than C-H...O, but the case that these 3 together the handshake between these 3
essentially holding it together will is the so called Gulliver effect of weak interaction. So,
we call this as the Gulliver effect of weak interactions and this is how we could analyze
the lone pair directionality is in the relative strengths of C-H...O interactions.
842
So, this is one example where charge density analysis can be done. In fact, at this stage
we also did a little earlier on in fact, one of my students Munshi, he analyzed the all
possible hydrogen bonds and also all possible bonds which are in the Van der Waal
interaction region, that means hardly any interactions between them and he calculated the
electron density from the available data in literature and we had a very reasonably large
number of such available interactions.
So, we find that we have interactions like for example N-H...O, O-H...O and N-H...O
these are representing strong hydrogen bonds. So, we found that there is a universal
curve between the way in which electron density occurs and this Rij now is the region
between one atom center and another atom center it is not the straight line which is
joining the bond, but this is known as the bond path. This is the path the electron density
will take, so if the epsilon value is larger the path will be bent, if the epsilon value is very
small the path will be straight.
So, depending upon the value of the ellipticity associated with the bond we find that Rij
value can be determined. So, this is a universal curve where you see that the interactions
are involved in these for example, I will discuss a little bit on sulphur-sulphur in a
minute. But these kind of interactions also have an electron density this is the electron
density plot electron density value, so they do have some electron density and they all
cluster around this region.
843
So, much that we can now actually think of drawing a line at 2.75 angstroms for Rij
which we would like to call as a limit of a hydrogen bond, so anything beyond this is not
a hydrogen bond and anything before this is a hydrogen bond and it could be classified
into weak and strong hydrogen bonds and this. In fact, you see the very strong hydrogen
bonds go up to a distance of one point this is in oxalic acid dihydrate the hydrogen bond
is the strongest.
It also tells us one other feature that is if you now keep it is this is going asymptotically
towards the infinity that means it is going parallel to more or less parallel to the electron
density vector. But as we keep going higher and higher, if we go through the end of this
one and go further we will get the other types of normal bonds we will get the covalent
bond the ionic bonds and so on.
So, any bonding therefore, in general has this continuum, so that means when 2 atoms
come closer they come closer to the Van der Waals interaction region, they are happy to
stay there, if they are happy to stay there they will be in this region. If there is a weak
interaction it will stay in this region and if there is a strong interaction they will stay in
that region and the even stronger like covalent bonds, ionic bonds and so on they go
through on to the other side of this projection they will go into the room away from me.
But they will all follow this curve which is a universal curve, so we call it as the
continuum of strong hydrogen bonds to weak interactions. With this we could also define
the radius of the hydrogen bond I do not think I have time to go into those details.
844
Except to go to the next one where I talked about Sulphur-Sulphur bond, this is
something which by chance was done by me in 1981. So, what I will do in the next 15-
20 minutes having made you sit through this entire course over the last so many days I
think I will highlight the work which has been done in our group on charge densities.
This is something of a propaganda for our group as well as something which has been
done in India for the first time and also there are not many groups in India who are doing
this kind of work. And this is the some kind of a basis for the futuristic outlook of
quantum crystallography. So, I thought I will generally discuss the work which has come
out from our group and probably it will take another 10 minutes at the most.
So, what I will do is here is what is done here is the we have looked into all crystal
structures you have seen this is 1981 there were not too many structures, but whatever
structures were available in the Cambridge structural database. We looked at that and
found elected it and found out that shorter distances between sulphur and sulphur. So,
there is there are molecules which have sulphur atoms on them and these sulphur now
come due to the crystallography interactions the symmetry and so on, they come closer.
But when they come closer what kind of interaction develops between them? This is
purely Van der Waal in nature, but does it give any indication of what is the nature of
sulphur atom. Say I mentioned a couple of days ago about a couple of day classes ago
the importance of this year which is the year of periodic table. So, sulphur is a very
845
crucial element in the periodic table, now we assume that the radius of sulphur is so
much.
But it shows in this particular way of analyzing, we see that the radius of sulphur is no
longer spherical it has got a non spherical shape, the reason is that whenever there is an
atom which is approaching in this direction which is in fact a nucleophile approaching
sulphur. If the other sulphur is a part of a molecule which makes it look like a
nucleophile, so the approach of the nucleophile to this sulphur atom is concentrated in
the open circles. And the concentration of the closed circles will tell us that it is
approaching from a direction which is about 20 degrees to the normal, this is the
approach of the electrophiles and this is the approach of the nucleophiles.
So, effectively when you take at sulfur atom it is not a hard sphere sulfur atom is now
ready to accept electrons and give out electrons as well and so now like it is more likely
you know you put a you have a pomegranate and you put a hole through the
pomegranate.
So, pomegranate hole and you press the pomegranate it compresses in that direction ok,
when it compresses in the direction the overall shape of pomegranate is going to change
and that is what happens to every atom, not just the sulfur atom and therefore there are
interaction regions which develop. And these interaction regions have been theoretically
calculated and those interaction regions are the so called sigma holes and pi-pi holes.
We do not have time to go into the detail of it, but those can be characterized from the
quantum mechanical calculations as well as from charge density measurements of very
highly accurate datasets. At this particular time there were not many accurate data sets
available, but we could predict that the sulfur into sulfur interactions will generate non
spherical shape of sulfur in crystals, this is something which is just out of the hat kind of
explanation in those days, but then this defined they so called bonding in groups all
groups of atoms of 14 to 17.
846
So, they add the groups 14 to 17 are shown here we have therefore, the carbon atom, we
have the nitrogen atom, we have the oxygen atom and the fluorine. This now belongs to
the halogen group. This belongs to the sulfur group that is the chalcogen group this
belongs to the pnicogen group and this belongs to the carbon group. So, the all these
atoms now will have therefore a behavior which is like what we described in case of the
sulfur. So, the non spherical shape is associated it with every one of these elements in the
periodic table and therefore they can have these kind of directed interactions and these
directed interactions come as a consequence of what we have defined here.
For example, a halogen bond it is the definition which has been accepted by all these
crystal engineering community as well as charge density community, that is the
definition, a halogen bond occurs when there is evidence of a net attractive interaction
between an electrophilic region associated with a halogen atom in a molecular entity.
And a nucleophilic region in another or the same molecular entity I think this part is cut I
will just try to see.
So, you see this is the definition, halogen bond of I moved this one away specifically it is
it happens when you give let us make it smaller right. So, yeah this is the definition a
halogen bond occurs when there is evidence of a net attractive interaction between an
electrophilic region associated with the halogen bond.
847
So, we are now concentrating on halogen bonds and this diagram clearly shows you that
there is a donor and then there is an acceptor donor can be any of these atoms acceptors
or halogens and the value of the Y atom could be again carbon, nitrogen, halogen so it
could be anything. So, the halogen bonding occurs between these which is now shown as
a something which is ready to open mouth and a gulp this part and then this association
takes place.
And this was further characterized theoretically by remarkable work of Politzer who
actually found out these kind of regions where now bromine in a CF 3Br situation is now
exhibiting what is known as a sigma hole. And this sigma hole is something which now
tells us that this bromine is ready to have this sigma hole behavior and ready to accept
interactions. So, the interactions will develop in that region so they become a highly
directional, so halogen bonding is highly directional, chalcogen bonding is highly
directional. In fact, we have also demonstrated the pnictogen bonding involving nitrogen
is also a highly directional interaction. All these carry this issue of a sigma hole and
things like that.
848
This was extended in the case of halogen bonding.
This was the work of Venkatesha who did this work and calculated the energies that are
involved and showed that in different kinds of geometrical situations we can show that
sigma hole.
This by the way these are not orbital’s as one they look like, but they are not orbital’s
these are 3D deformation density maps. So in fact, one cannot see the electron insane one
cannot see the orbital’s because orbital is a quantum mechanical phenomena. So, at one
849
time people thought that they are looking at orbital’s, but they are not realistically
orbital’s there are orbital their deformation density has 3 dimensional deformation
density map. So, you can see that there is a sigma hole here which is now facing the
electron dense annulus region which is blue. So, the red blue portion interacts with the
blue portion and that is how you get an interaction between Cl1 and Cl2 in this particular
case.
So, there are other examples which have been worked out as far as then came the
realistic understanding of this sigma hole. And then the bonding features when the work
is purely from IASC this theoretical prediction was made by Professor Arunan of IPC in
organic physical chemistry department. See there is always this hydrophobic interactions
which occur and how do they occur because, we already have a tetrahedrally satisfied 4
bonds with carbon. So that means it is a very stable atom still the nucleophile can attack
on this one and we can form this so called SN 2 reactions.
So, the basis of incentive reactions tells us that we can have interactions where we can
have a whole of the carbon exposed for possible interaction profile. So, what was
predicted by Arunan was the fact that he proposed what is known as a carbon bond, it is
an analogy with hydrogen bond, an halogen bond and he said that for example, in these 3
cases which were theoretically calculated, they theoretically found find the presence of
finds, the presence of the critical point which is shown as a green dot here.
850
So, the green dots are critical points and if since you find a critical point by definition we
gave it defines a bond. And therefore there is a shortage interaction in this particular case
it is a F-H...O hydrogen bond in this case it is a F-Cl...O halogen bond and in this
particular case you have a carbon and you see that the oxygen atom now has a bond path
which is shown like that.
So, it is not going to be an interaction associated with these hydrogen atoms and then
coming in which would be a C-H...O bond it is essentially a direct interaction, that means
the carbon atom now has opened out its a sigma hole and this interaction is now accepted
and that is the reason why you get a fifth bond with carbon. Because there are already 4
bonds the fifth bond with carbon now becomes reactive and so you proceed with the
reaction.
So, this was the prediction which was made and then we wanted to verify it
experimentally. So, we made 2 compounds, 1 compound which was we had already
studied at this short interaction between chlorine and carbon which is 3.25 and there was
another case which we designed so that we have this carbon oxygen distance of 3.02 this
again is the work of Thomas and Pavan. So, there are 2 possibilities, one possibility is
the definition of carbon bonding which is given by Arunan, the other possibility is the
trifurcated hydrogen bond C-H...O right. So, A is being the acceptor atom in this
particular one case it is chlorine, the other case it is oxygen; agreed.
851
So, what we did was to do the charge density analysis and plot the deformation densities.
What is found is that in the case of the first compound which is Fenobam we find that it
is actually going to the hydrogen atom, the bond path when we calculate the critical point
is up here and the bond path goes to the hydrogen atom nothing exciting. So, this is
essentially a C-H...O on the other hand in this compound DMA 4 HBA we see that this
now goes to the carbon atom directly, avoiding the 3 hydrogen atoms which are readily
looking for it. You know it is like in cricket there are 3 slips on the galley and somebody
pushes the ball between the 2 slips.
So, which one is, which fellow is going to catch it and of course, nowadays Indian team
has improved their performance otherwise what will happen is if the ball goes between
the second slip and the third slip both of them dive on each other and the balls escapes.
But it may so happen that one of them make a slip on it.
So, in this particular case this fellow caught the ball which is nothing exciting which is a
allowed interaction, but this is an interaction which we were looking for. So, effectively
if you see this carbon now it has 1 2 3 4 and 5 bonds and therefore it is an intermediate to
a possible reaction site.
852
So, we found for the first time the bond path which represents the carbon atom and we
almost forgot it, but then it became very famous it came in CND News. So, the last part
of the course I thought we will give some publicity to my group. So, this came in CND
News it became very famous, people said we defined a new carbon bond and electronic
effects that produce halogen bonds. We did non covalent interactions in other elements
and so on and so forth.
And there is a big paragraph written about what we have done in this case, sigma hole
bonds are also highly directional and occur in line with the covalent bond that causes
them. So, again the universal plot which we did in 2003 is verified even in this particular
case, so this was a quite an excitement at that time.
853
The next excitement came with a title which we see here sigma meets pi for a hole lot of
bonding and this was the work done by Rumpa in our group with me and then we find
that in this particular case it is a compound which is biologically important and there is a
S-C-N a bond it is just linear bond and then there is an interaction between the sulfur and
the center of this NHC bond NCS bond and also there is a center interaction.
So, it is its space group is P41 and so it is quite an interesting organic molecule and when
we analyze this structure we had some interesting phase transitions. So, I do not want to
touch upon that but what we found is that in one direction it has the sigma hole
interaction dominates and in the other direction the pi hole interaction dominates.
So, one goes this way forward direction sigma hole and the other goes from the
perpendicular direction which is the pi hole and then of course Jason who is a editor of
chemistry world he wrote that we found the co operative sigma hole and pi hole
responsible for rolling molecules in a isothiocyanate based peptide. So, this compound of
course we got from Professor Balaramas group. So this work was praised again truly
smashing and acts considerable scientific value to bond theory this was written by one of
the pioneers in charge density analysis Dietmer Stalke.
854
And thereby we essentially see that a typical X-ray diffraction experiment with of
course, a little more analysis using quantum crystallography providing chemical
bonding, characteristics crystal dynamics etcetera and it gives various ideas about how to
go further is emerging.
So, again the rules of symmetry play a major role, the way in which the sigma hole
orientation takes place depends upon where the symmetry element is located in the
atomic arrangement. It also depends upon where how many molecules are there in the
unit cell and in what way the molecules direct themselves with involving these very
weak interactions, these are all very small in terms of electron densities unless you do a
very high resolution X ray diffraction data you cannot see them experimentally. And
then of course theoretical modeling is also equally difficult.
But once the modeling is done there is a nice handshake, this is the only area where there
is for once the theoreticians and the experimentalists shake hands and agree with each
other. Normally theoreticians are always thinking they are somewhere above space and
they are doing wonderful work, experimentalists are very humble and they say well we
have found this. But this is the first time an experimentalists are dictating the
theoreticians and vice versa. So, this is therefore an area of future and you see that
improved structure determination in protein crystallography we are looking for that.
855
Determination of hydrogen atom positions hydrogen ADPs is already possible and
modeling of accurate. So this part which is which we should have discussed a little more
when you had time is essentially it depends upon the way in which the atoms are and the
way in which the atoms are interacting with each other and the way in which molecules
are put inside the crystal. So, again the symmetry probably will add on to the
contributions of modeling electron density distributions, we do not know the issues
associated with symmetry if they are brought along with the concept of modeling that
will definitely help. In fact, the multi pole model also when we do the expansion about
Monopole, Dipole Quadrupole and so on.
Depending upon the symmetry of the system some of the values associated with these
polar components will either appear or disappear effectively like systematic absences.
So, there will not be any contribution coming in certain directions because of the fact
that there is symmetry that is associated with the crystal structure.
So, that way the study of symmetry at the basic level which we have done in the entire
course and then the way in which we have carried out the understanding of the
development of point groups, the space group, the systematic absences through
experiments and the type of experiments we perform using X-ray diffraction. And at the
later stage when we have looked at the accuracies with which we can analyze the results
and then the risk based refinement, the powder diffraction. All these in a nutshell are a
part and parcel of the basic understanding of symmetry and structure in the solid state.
And I think in this particular course we have been able to cover that aspect to a
reasonable extent.
We have expanded this into the recent areas we have seen many a potential applications
that are possible, but since it is a short course of this kind we cannot go into the detail of
that. So, maybe there is an advanced course which we will now use the ideas which we
have learnt from this short course. And then go to the advanced course where you will
learn more details of all these methodologies which again require the basic
understanding of symmetry and structure, so symmetry and structure is different
foundation for any of these discussions we have had until now.
856
So, I say this I always like to say something at the end of a lecture in this like this series
of talks, this series of classes never ends, because it has to go on and on and on this is a
statement from Tolkien in the Lord of the Rings, he says that the road goes ever on and
on is what we came from structure and symmetry we went to various directions.
Down from the door where it began, it began at a point it began with what it began with
the definition of point groups and it began at a point center of symmetry it then went to 2
fold symmetry it went to the plains and so on. Now, far ahead the road has gone so we
have gone far away in those directions we have understood all the symmetry principles
and so on and I must follow if I can this is the problem I must follow these whatever I
have understood I must follow pursuing it weary feet obviously because we have
learned. So, many things I have talked for so long in fact I have coughed many times
some of it may be recorded also.
So, it went on and on it is quite vary to go through this until it is joined some larger way
finally we have put them all together we are looking at very finer aspects of the small
molecule, structure determination bring in quantum chemistry into the picture. Of course,
we have not seen the other aspect of large molecules that is also very fascinating. So,
there are 2 major developments in the last decade or so one is understanding small
molecules to the best possible extent leave, for the first time we are seeing holes in atoms
in the periodic table in a loose sense. We have identified the sigma holes and the pi holes
857
associated with atoms and how they provide electrophilic and nucleophilic interactions.
So, pursuing it with weary feet until it joins is larger way there are many paths and
errands meet. So, all these things sigma all meets pi hole something we wrote that fellow
wrote I did not write and without them I cannot say so we are reached the main road
now.
Now, after reaching the main road what we will do whether we will take a taxi, we will
take a bus, we will take a bullock cart I do not know this petrol is getting over in this
world. So, we may even end up with a bullock cart, but whatever we cannot say right
now and that is the statement of lord of the rings which I will end this course. And I must
acknowledge because I have just left and right copied the beautiful equations and
diagrams which are there in these text books.
I do not know how to thank these guys because, unless I had these books at my disposal
and of course you know copyright issues being violated left and right these days. So I
could get the scanning done and get the diagrams out. So, many of the diagrams are
taken directly from these books and so I thought I should acknowledge all the people
who wrote those books there are wonderful books, for example this one the
Fundamentals of Crystallography which is edited by Giacovazzo. But you see there are.
So, many high quality authors who have contributed to this; this is a more recent book
858
published by IUCR the Oxford University press we have brought it out that is what
science publications.
So, you can see that all the modern methods of structure determination are covered here.
So, this is the most recent book, one of the older books is the essentials of
crystallography Duncan Mckie and Christine Mckie these 2 have written a wonderful
book this probably the best book to understand symmetry. If anyone wants to get a
complete understanding of how symmetry develops. What are the various symmetry
elements in terms of both physics as well as chemistry? I have not gone into the detail of
how mathematically we can come to these understanding of point groups and space
group, but that is an issue which is not for our course.
So, but we could have one could study the group theory associated with it and still arrive
at all the details, but this is a book I must acknowledge this is by far the most referred
book in our course. The other is the Gurus book this is Jack Dunitz I consider him my
Guruji because, he is the one who explained all the scattering phenomena in such a way
that a very simple person like me can understand the entire structure the entire scattering
theory. So, whatever scattering theory we did and how we arrived at the Bragg condition
and the Laue conditions and so on were all taken from this textbook of Jack Dunitz.
And then we have got the other 3 books the next book of course is the X-ray structure
determination practical guide Stout and Jensen as the title says it is a practical guide. So,
859
most of the tables which we have in the presentation are from the textbook, which tells
you how to get the equivalent points how together it the equivalent diagrams how to get
the symmetry points and then the equivalent points and so on. Then of course, the book
of the crystal structure analysis which is now one of the current books which are in
which are available in many bookshops and this is for the modern day crystallographers
who used the machines.
We do not have to learn too much of crystallography but depending on the machines, but
having depended on the machines what are the principles and practice one has to go
through for crystal structure analysis is beautifully compiled in this, it is again this is
edited by William Clegg. So, this has articles by many of the top crystallographers
currently in the world, so this is another textbook which we have referred to it is also an
IUCR textbook. And then of course the Bible or the Bhagavath Gita or the Quran
whatever it is that is our international tables for crystallography.
Now, without this the course could not have been done at all and the contents of this
book the tables have been displayed many times and we have discussed every point with
respect to the international tables of crystallography. And so I think I am very grateful to
the availability of these 6 textbooks and the tables without which the course could not
have been completed. I come to the end of the course now with a last slide.
860
The last slide is very crucial because if you want to do something like this you need the
help of people right and I have got some 4 mug shots or 4 guys 4 people who are in my
group Diptikanta Swain is a post doc, Titas, Ajana and Ravi. In fact, these 4 fellows will
continue to be with you during the TA programs and also the monthly discussion
programs and so on throughout the course. And of course, IISc without which we cannot
do this NPTEL is also situated they have an office in IISc and if any time you are there
you can see that this is a very beautiful symmetric building surrounded by of course a lot
of nature around it.
But if you look at it carefully you can see the symmetry that is associated with the main
building, I as a challenge I want you to find the symmetry associated with this at the end
of the course. I also thank all my other postdocs and project assistants who now
participated very actively in this course program and also helped me out to make some
correction here and there particularly English corrections and at the end I would like to
Thank you all for participating in the course.
861
THIS BOOK IS
NOT FOR SALE
NOR COMMERCIAL USE
(044) 2257 5905/08 nptel.ac.in swayam.gov.in

Symmery and Structure in The Solid State

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Symmery and Structure in The Solid State

Uploaded by

Copyright:

Available Formats

INDEX

(Refer Slide Time: 00:35)

Be it biology, be it material science, be it physics; in fact I teach a course where people

The methodology and the experimentation; however, becomes extremely difficult in

He made these abstract paintings of course, he is not a crystallographer, but he made

So, by definition the d 1 and d 2; so d 1 prime and d 2 prime. So, if d 1 is equal to d 1

(Refer Slide Time: 00:31)

(Refer Slide Time: 05:15)

(Refer Slide Time: 07:19)

(Refer Slide Time: 15:21)

So, I think we will stop here for today.

(Refer Slide Time: 00:39)

(Refer Slide Time: 10:35)

(Refer Slide Time: 29:11)

(Refer Slide Time: 01:37)

(Refer Slide Time: 00:42)

(Refer Slide Time: 08:50)

This x y z and x̅ y̅ z̅ are equivalent points of a centre of symmetry or we can in fact,

(Refer Slide Time: 22:58)

(Refer Slide Time: 00:33)

(Refer Slide Time: 04:09)

Now, both of them do not represent a centrosymmetric system. Now, I am going to

We operate x y̅ z mirror again, we go back to x y z we do not see the presence of the

(Refer Slide Time: 13:35)

(Refer Slide Time: 16:05)

Because pharmaceutical industry wants molecules to crystallize in different crystal

(Refer Slide Time: 25:33)

(Refer Slide Time: 00:54)

(Refer Slide Time: 16:20)

Now, in a monoclinic system therefore, we have no restriction on a b c values, but α and

(Refer Slide Time: 00:57)

(Refer Slide Time: 10:37)

So, suppose there is a compound with an organic compound or a inorganic compound

(Refer Slide Time: 20:49)

(Refer Slide Time: 00:36)

So, f represent a rectangular p lattice corresponding to b; on the other hand, the e

(Refer Slide Time: 19:17)

(Refer Slide Time: 21:03)

And I think we will stop here.

(Refer Slide Time: 00:33)

(Refer Slide Time: 06:47)

(Refer Slide Time: 08:10)

We go further now into orthorhombic symmetry in case of the orthorhombic symmetry

So far we have seen the classification of point groups.

(Refer Slide Time: 00:34)

(Refer Slide Time: 01:56)

The corresponding stereographic projection of the ̅ 1 is shown here where x is x y z and

(Refer Slide Time: 15:21)

(Refer Slide Time: 00:29)

(Refer Slide Time: 25:00)

(Refer Slide Time: 29:47)

(Refer Slide Time: 00:29)

(Refer Slide Time: 07:27)

(Refer Slide Time: 11:42)

(Refer Slide Time: 20:35)

This is just a side point I thought I will mention.

(Refer Slide Time: 26:30)

(Refer Slide Time: 26:42)

When we consider a molecular structure, then the number of molecules is 1. If the

So, in an asymmetric unit there is no invocation of symmetry whereas, one once we

Now, the x y z can be more than one as I mentioned it could be x1 y1 z1 , x2 y2 z2 each

(Refer Slide Time: 13:55)

Continuing our discussion on the Monoclinic Symmetry.