# Special Cases: Miller Indices For Hexagonal Crystals

Planes and Directions in Cubic Crystals
• The justification for using four indices in the case of hexagonal
crystals relates to the fact that equivalent planes are not directly
apparent if only three indices are used (as is illustrated in the
following example).
• (0 1 0) (0 1 1 0)
(1 0 0) (1 0 1 0)
(1 1 0) (1 1 2 0) These two planes are equivalent
(1 2 0) (1 2 1 0 ) although it is not initially apparent
• Only in the case of cubic crystals is a direction [hkl] always
perpendicular to the plane (hkl) and the angle between two
planes or directions is given by :
cos( ) ·
h
1
h
2
+ k
1
k
2
+ l
1
l
2
h
1
2
+ k
1
2
+ l
1
2
h
2
2
+ k
2
2
+ l
2
2
Mathematical Approach to Crystallography
• Lattice is defined by translational symmetry:
L · n
1
a + n
2
b + n
3
c
where a, b and a, c and b, c are non-colinear
and a,b,c are non-coplanar
• Rotations, Inversions, etc. (Not-translation Symmetry Operations)
a
r’
r
c
b
X'
Y'
Z'

]
]
]
]
·
R
11
R
12
R
13
R
21
R
22
R
23
R
31
R
32
R
33

]
]
]
]

X
Y
Z

]
]
]
]
r ' · Rr
where R is a symmetry operator. If we move the space and leave
the coordinate axes fixed, the symmetry operation is an active
operation. If we leave the space fixed and move the coordinate
axes, we have a passive symmetry operation.
• Identity Operation:
no change in position
Symbol: 1 (E Schoenflies Notation)
• Rotation (proper) about c-axis by 180°
Symbol: 2 (C
2
) or 2[hkl]
• Inversion through a point
Symbol: 1 (i)
Inversion through a point (center of
symmetry) changes the chirality or the
handedness. Objects before and after the
operation are enantiomorphically related.
1 0 0
0 1 0
0 0 1

]
]
]
]
−1 0 0
0 −1 0
0 0 1

]
]
]
]
−1 0 0
0 −1 0
0 0 −1

]
]
]
]
• Rotation by 2π/n around z- axis
Symbol: n or n [hkl] (C
n
)
rotations do not change
the handedness of an object
• Reflection
Symbol: m (σ) or m [hkl] (σ
h
, σ
v
, σ
d
)
reflection with repect to a mirror plane
which is perpendicular to the direction
given by [hkl]. For (x, y) mirror plane:
σ
h
is perpendicular to the principal axis
(c-axis). σ
v
contains the c- and a-axes.
σ
d
contains the c-axis but falls between
the a and b axis.
cos
2
n
|
.
`
,
−sin
2
n
|
.
`
,
0
sin
2
n
|
.
`
,
cos
2
n
|
.
`
,
0
0 0 +1

]
]
]
]
]
]
1 0 0
0 1 0
0 0 −1

]
]
]
]
c
c
a a
c
b
σ
d
σ
v
σ
h
• Improper Rotations
rotation followed by inversion (International)
rotation followed by reflection (Schoenflies)
Note that an improper rotation may be a symmetry operation
for a given crystal, even if the two individual operations
(rotation and inversion or reflection) are not.
Symbol: n (S
n
= σ
h
C
n
)
• The determinant of all symmetry operators are either +1 or -1:
+1: proper rotation and identity
-1: improper rotation, mirror and inversion
• Given a Lattice defined by :
the volume of the primitive cell is defined by:
L · n
1
a + n
2
b + n
3
c
V ≡ a • b × c
( )
θ
v
w
Given the basis of a lattice composed of vectors a, b and c, how
does one calculate bond angles and bond length for atoms or
groups of atoms in a crystal ? The bond length will be defined
from the length of a vector connecting two points and the bond
angle as the angle between two vectors.
If the vectors a, b and c define a basis
for a rectangular coordinate system, then,
the length of a vector r (x,y,z) is given by:
r · x
2
+ y
2
+ z
2
The angle θ between two vectors v (v
1
, v
2
, v
3
) and w (w
1
, w
2
, w
3
)
is given by:
v • w · v w cos( ) · v
1
w
1
+ v
2
w
2
+ v
3
w
3
Then, since many lattices have a non-cartesian crystal lattice
basis, how do we define bond length and angles in general ?
Generalized Equations for Bond Lengths and Bond Angles
v • w · v
1
a + v
2
b + v
3
c
( )
• w
1
a + w
2
b + w
3
c
( )
·
v
1
w
1
a • a + v
2
w
1
b • a + v
3
w
1
c • a +
v
1
w
2
a • b + v
2
w
2
b •b + v
3
w
2
c • b +
v
1
w
3
a • c + v
2
w
3
b • c + v
3
w
3
c • c
· v
1
v
2
v
3
[ ]
a • a a • b a •c
b • a b • b b • c
c • a c • a c • c

]
]
]
]
w
1
w
2
w
3

]
]
]
]
Metrical Matrix
v • w · v
T
Gw
row matrix of
components
column matrix
of components
The G elements are the terms a
2
, ab
cos(γ), ac cos(β), etc,....which are
easily obtained from the length of
and the angles between the cell axes.
Finally the bond length and the bond angle are obtained from:
v · v • v · v
T
Gv
cos( ) ·
v • w
v w
·
v
T
Gw
v
T
Gv ⋅ w
T
Gw
Reciprocal Lattice
c
b
a
c/l
b/k
a/h
(hkl) plane
In diffraction experiments we
between crystallographic planes
of type (hkl). These spacings are
measured along the normal
having the same Miller indices.
The interplanar spacings are easily obtained by consideration of what
is called the reciprocal lattice. Any lattice defined by a basis of three
vectors, a, b, c has an associated reciprocal lattice defined by a basis
of three vectors a*, b*, c* wich have the following properties:
1. The reciprocal lattice vector g = h a*+ k b* + l c* is perpendicular
to the direct lattice plane with Miller Indices (hkl).
2. The interplanar spacing d
hkl
between
successive (hkl) planes is:
d
hkl
·
1
g
Determination of the Reciprocal Lattice Vectors
• a*, b*, c* must form a basis such that a* is perpendicular to b
& c, b* is perpendicular to a & c and c* is perpendicular to a &
b. That this must be so can be understood from the fact that
the reciprocal space vector g = 1.a* + 0.b* + 0.c* (g = a*) is
normal to the (100) plane. Therefore a* must be parallel to the
vector bxc which is perpendicular to both b and c.
a* = const (bxc) and
c
b
a
(100)
a*
a*
θ
const =
1
a • b × c
( )
Unit Cell
Volume
d
100
· a cos( ) a * ·
1
a cos( )

a * a cos( ) ·1 a *•a · 1
a * ·
1
d
100

b * ·
c × a
a • b × c
( )
and c * ·
a × b
a • b × c
( )
a * ·
b × c
a • b × c
( )
Therefore the reciprocal lattice vector a* is given by :
Following the same arguments, we can calculate b* and c*
Note that:
a • a * · b • b * · c •c * · 1
a • b * · b • a * · a •c * · c •a * · b •c * · c • b * · 0
If
g · ha * +kb * +lc *
g
2
·
1
d
hkl
2
· h k l [ ]G
−1
h
k
l

]
]
]
]
A given vector can be expressed in both the direct lattice and the
reciprocal lattice. The coordinates of this vector in the two bases
are related to each other through the Matrix H. As a homework
problem determine this matrix H.
r
dir lat
· xa + yb + zc
r
rec lat
· x' a * +y' b * +z' c *
r
rec lat
· Hr
dir lat
General coordinate transforms:
basis 1: (a
1
, b
1
, c
1
)
basis 2: (a
2
, b
2
, c
2
)
a
1
expressed in basis 2
b
1
expressed in basis 2
c
1
expressed in basis 2
r
basis 2
· Rr
basis 1
where : R ·
a
1
X
b
1
X
c
1
X
a
1
Y
b
1
Y
c
1
Y
a
1
Z
b
1
Z
c
1
Z

]
]
]
]
Physical Properties and Symmetry
• Goal: Use the crystal symmetry to determine the minimum
symmetry of a physical property.
Neumann’s Principle:
The symmetry elements of any physical property of a crystal
must include at least the symmetry elements of the point group
of the crystal. This implies the physical properties can have
more symmetry elements than the point group.
• Examples of physical properties:
j · E
P · E
q · −k∇ T
· S
current density = conductivity x electric field
elect. polarization = elect. susceptibility x elect. field
heat flux = thermal conductivity x temperature gradient
strain = elastic compliance x stress
• The property is a tensorial quantity as it generally depends on
the choice of the coordinate system. Physical properties are
described by symmetric tensors.
• Let us consider the example of the relationship between
conductivity and electric field in the case of a cubic crystal and
examine the effect of the crystal symmetry on the symmetry of
the electric conductivity.
A cubic crystal has four 3-fold axes of symmetry along <111>
axes. The electric conductivity must therefore have at least this
degree of symmetry.
We rotate the space by 120° around the [111]
direction (R is the rotation matrix). Note that
R
-1
ρ R is a similarity transformation as we
require the physical property (conductivity) to
be unchanged by the rotation (symmetry
element)
j · E
Rj
( )
· RE
( )
j · R
−1
R E
· R
−1
R
• Determination of R: R is the matrix associated with the rotation
around the [111] axis, i.e. it is a coordinate axis transformation.
basis 1: (a
1
, b
1
, c
1
)
basis 2: (a
2
, b
2
, c
2
)
a
1
expressed in basis 2
b
1
expressed in basis 2
c
1
expressed in basis 2
For example: The first column of R corresponds to the coordinates
of [100] in the new system.
r
basis 2
· Rr
basis 1
where : R ·
a
1
X
b
1
X
c
1
X
a
1
Y
b
1
Y
c
1
Y
a
1
Z
b
1
Z
c
1
Z

]
]
]
]
x
z
y
old new
x y’
y z’
z x’
3 · C
3
· R ·
0 0 1
1 0 0
0 1 0

]
]
]
]
11 12 13
21 22 23
31 32 33

]
]
]
]
·
0 1 0
0 0 1
1 0 0

]
]
]
]
11 12 13
21 22 23
31 32 33

]
]
]
]
0 0 1
1 0 0
0 1 0

]
]
]
]
·
21 22 23
31 32 33
11 12 13

]
]
]
]
0 0 1
1 0 0
0 1 0

]
]
]
]
·
22 23 21
32 33 31
12 13 11

]
]
]
]
The rotation matrix R is orthogonal and the inverse
rotation is equal to the transpose of the direct rotation.
R
.
R
T
= R
T.
R = 1 or R
T
= R
-1
.
Therefore ρ = R
-1
ρ R is equivalent to ρ = R
T
ρ R
The equality between matrices implies:
ρ
11
= ρ
22
= ρ
33
and ρ
12
= ρ
23
= ρ
31
and ρ
13
= ρ
21
= ρ
32
If we repeat these calculations for a rotation around another axis of
the <111> family, we would find other constraints, such as
ρ
12
= - ρ
21
so that we can conclude ρ
ij
= - ρ
ij
= 0
Therefore:
·
11
0 0
0
11
0
0 0
11

]
]
]
]
· .1
Note that to determine the inverse M
-1
of any matrix,M, you should
use the following approach:
M
-1
= Transpose (M(cofactors))/Determinant of M
where the cofactors of M
ij
= (-1)
i+j
(Minor of M
ij
) and the Minor of
M
ij
is obtained by calculating the determinant of the submatrix
obtained by eliminating the i-th row and the j-th column.

the symmetry operation is an active operation. c and b. If we move the space and leave the coordinate axes fixed.c are non-coplanar • Rotations. Inversions. (Not-translation Symmetry Operations) r’ c  X'   R11 R12 R13   X   Y'  =  R21 R22 R23  •  Y  r b        Z'   R31 R32 R33   Z  a r ' = Rr where R is a symmetry operator. . If we leave the space fixed and move the coordinate axes. etc. c are non-colinear and a. b and a.b.Mathematical Approach to Crystallography • Lattice is defined by translational symmetry: L = n1a + n2b + n3c where a. we have a passive symmetry operation.

. Objects before and after the operation are enantiomorphically related.• Identity Operation: no change in position Symbol: 1 (E Schoenflies Notation) 1 0 0  0 1 0  0 0 1    −1 0 0   0 −1 0  0 0 1   −1 0 0   0 −1 0     0 0 −1 • Rotation (proper) about c-axis by 180° Symbol: 2 (C2) or 2[hkl] • Inversion through a point Symbol: 1 (i) Inversion through a point (center of symmetry) changes the chirality or the handedness.

σ v contains the c. 1 0 0  0 1 0  0 0 −1   c σh c a σv a c σd b . σ v.• Rotation by 2π/n around z.axis Symbol: n or n [hkl] (Cn) rotations do not change the handedness of an object cos 2    n  sin  2    n 0   2 − sin    n 2  cos  n 0 0  0  +1   • Reflection Symbol: m (σ) or m [hkl] (σ h .and a-axes. σ d) reflection with repect to a mirror plane which is perpendicular to the direction given by [hkl]. y) mirror plane: σ h is perpendicular to the principal axis (c-axis). σd contains the c-axis but falls between the a and b axis. For (x.

mirror and inversion • Given a Lattice defined by : L = n1a + n2b + n3c the volume of the primitive cell is defined by: V ≡ a • ( b × c ) .• Improper Rotations rotation followed by inversion (International) rotation followed by reflection (Schoenflies) Note that an improper rotation may be a symmetry operation for a given crystal. even if the two individual operations (rotation and inversion or reflection) are not. Symbol: n (Sn = σ h Cn ) • The determinant of all symmetry operators are either +1 or -1: +1: proper rotation and identity -1: improper rotation.

the length of a vector r (x. since many lattices have a non-cartesian crystal lattice basis.Given the basis of a lattice composed of vectors a. If the vectors a. v2. then. b and c define a basis for a rectangular coordinate system.z) is given by: r = x2 + y2 + z 2 v w θ The angle θ between two vectors v (v1.y. w3) is given by: v • w = v w cos( ) = v1w1 + v2 w2 + v3 w3 Then. how does one calculate bond angles and bond length for atoms or groups of atoms in a crystal ? The bond length will be defined from the length of a vector connecting two points and the bond angle as the angle between two vectors. b and c. how do we define bond length and angles in general ? . v3) and w (w1. w2.

etc....Generalized Equations for Bond Lengths and Bond Angles v • w = (v1a + v2 b + v3c ) • ( w1a + w2 b + w3c ) = v1w1a • a + v2 w1b • a + v3 w1c • a + v1w2 a • b + v2 w2 b • b + v3w2 c • b + v1w3 a • c + v2 w3b • c + v3 w3 c • c  a • a a • b a • c   w1  = [ v1v2 v3 ] b • a b • b b • c  w2   c • a c • a c • c  w    3  The G elements are the terms a2. ab cos(γ). ac cos(β). Metrical Matrix v • w = v T Gw row matrix of column matrix components of components v = v • v = v T Gv v•w cos( ) = = vw v T Gw v T Gv ⋅ w T Gw Finally the bond length and the bond angle are obtained from: .which are easily obtained from the length of and the angles between the cell axes..

1 dhkl = 2. b a/h a The interplanar spacings are easily obtained by consideration of what is called the reciprocal lattice. The interplanar spacing dhkl between g successive (hkl) planes is: .c/l c b/k Reciprocal Lattice (hkl) plane In diffraction experiments we will learn about distances between crystallographic planes of type (hkl). c* wich have the following properties: 1. Any lattice defined by a basis of three vectors. c has an associated reciprocal lattice defined by a basis of three vectors a*. b*. These spacings are measured along the normal between two adjacent planes having the same Miller indices. b. The reciprocal lattice vector g = h a*+ k b* + l c* is perpendicular to the direct lattice plane with Miller Indices (hkl). a.

1 a* = const (bxc) and a * = d100 d100 = a cos( ) 1 a* = a cos( ) c a b θ a * a cos( ) = 1 a * •a = 1 1 Unit Cell const = a • (b × c ) Volume a* (100) a* . That this must be so can be understood from the fact that the reciprocal space vector g = 1.b* + 0. Therefore a* must be parallel to the vector bxc which is perpendicular to both b and c. b*.Determination of the Reciprocal Lattice Vectors • a*. c* must form a basis such that a* is perpendicular to b & c.a* + 0.c* (g = a*) is normal to the (100) plane. b* is perpendicular to a & c and c* is perpendicular to a & b.

Therefore the reciprocal lattice vector a* is given by : b ×c a* = a • (b × c ) Following the same arguments. we can calculate b* and c* c ×a a ×b b* = and c * = a • (b × c ) a • (b × c ) Note that: a • a* = b • b * = c • c * = 1 a • b * = b • a* = a • c * = c • a * = b • c * = c • b * = 0 If g = ha * +kb * +lc *  h 1 2 = 2 = [h k l ]G−1  k  g d hkl l    .

rdir lat = xa + yb + zc rrec lat = x' a * +y' b * +z' c * rrec lat = Hrdir lat General coordinate transforms: rbasis 2 = Rrbasis 1 X a1 Y where : R =  a1 a Z  1 b1X b1Y b1Z c1 expressed in basis 2 b1 expressed in basis 2 a1 expressed in basis 2 c1X  Y c1  c1Z   basis 1: (a1. The coordinates of this vector in the two bases are related to each other through the Matrix H. b2.A given vector can be expressed in both the direct lattice and the reciprocal lattice. c2) . As a homework problem determine this matrix H. c1) basis 2: (a2. b1.

• Examples of physical properties: j= E P= E current density = conductivity x electric field elect.Physical Properties and Symmetry • Goal: Use the crystal symmetry to determine the minimum symmetry of a physical property. susceptibility x elect. Neumann’s Principle: The symmetry elements of any physical property of a crystal must include at least the symmetry elements of the point group of the crystal. field q = −k∇T heat flux = thermal conductivity x temperature gradient =S strain = elastic compliance x stress . polarization = elect. This implies the physical properties can have more symmetry elements than the point group.

The electric conductivity must therefore have at least this degree of symmetry. j= E ( Rj ) = ( RE) j = R −1 R E = R −1 R We rotate the space by 120° around the [111] direction (R is the rotation matrix). Note that R-1 ρ R is a similarity transformation as we require the physical property (conductivity) to be unchanged by the rotation (symmetry element) . • Let us consider the example of the relationship between conductivity and electric field in the case of a cubic crystal and examine the effect of the crystal symmetry on the symmetry of the electric conductivity. A cubic crystal has four 3-fold axes of symmetry along <111> axes. Physical properties are described by symmetric tensors.• The property is a tensorial quantity as it generally depends on the choice of the coordinate system.

i.e. rbasis 2 = Rrbasis 1 X a1 Y where : R =  a1 a Z  1 b1X b1Y b1Z c1X  Y c1  c1Z   basis 1: (a1. b2. it is a coordinate axis transformation. c2) c1 expressed in basis 2 b1 expressed in basis 2 a1 expressed in basis 2 For example: The first column of R corresponds to the coordinates of [100] in the new system. c1) basis 2: (a2.• Determination of R: R is the matrix associated with the rotation around the [111] axis. . b1.

Therefore ρ = R-1 ρ R is equivalent to ρ = RTρ R     11 21 31 21 31 11 12 22 32 22 32 12 13  0 1 0    0 1  23 = 0      1 0 0 33 23 11 21 31 22 32 12 12 22 32 23 33 13 13  0 0 1    1 0 0 23    0 1 0  33   31  11  21  =    0 0 1    1 0 0 =  33   0 1 0    13   .RT = RT. R = 1 or RT = R-1. R .z old x y z new y’ z’ x’ y x 0 0 1 3 = C3 = R =  1 0 0   0 1 0   The rotation matrix R is orthogonal and the inverse rotation is equal to the transpose of the direct rotation.

ρ21 so that we can conclude ρij = . we would find other constraints. .ρij = 0 Therefore:  11 = 0 0  0 11 0 0 0  = . such as ρ12 = .1  11  Note that to determine the inverse M-1 of any matrix.M. you should use the following approach: M-1 = Adjugate(M) /Determinant of M = Transpose (M(cofactors))/Determinant of M where the cofactors of Mij = (-1)i+j (Minor of Mij) and the Minor of Mij is obtained by calculating the determinant of the submatrix obtained by eliminating the i-th row and the j-th column.The equality between matrices implies: ρ11 = ρ22 = ρ33 and ρ12 = ρ23 = ρ31 and ρ13 = ρ21 = ρ32 If we repeat these calculations for a rotation around another axis of the <111> family.