SOLIDIFICATION AND IMPERFECTIONS IN SOLIDS

ISSUES TO ADDRESS...

• What are the solidification mechanisms?

• What types of defects arise in solids?

• Can the number and type of defects be varied

and controlled?
• How do defects affect material properties?

• Are defects undesirable?

The University of Sydney Page 2

 Reading

– Chapter 3:
- 3.13 Single crystals
- 3.14 Polycrystalline materials
- 3.15 Anisotropy
– Chapter 4:
- 4.1 Introduction
- 4.2 Vacancies and self-interstitials
- 4.3 Impurities in solids
- 4.4 Specification of composition
- 4.5 Dislocations – linear defects
- 4.6 Interfacial defects
- 4.7 Bulk or volume defects
- 4.9 Basic concepts of microscopy
- 4.10 Microscopy techniques
- 4.11 Grain size determination

The University of Sydney Page 3

 Crystals as Building Blocks
• Some engineering applications require single crystals:
--diamond single --turbine blades
crystals for abrasives Fig. 8.33(c), Callister 7e.
(Fig. 8.33(c) courtesy
(Courtesy Martin Deakins, of Pratt and Whitney).
GE Superabrasives,
Worthington, OH. Used with
permission.)

• Properties of crystalline materials

often related to crystal structure.
--Ex: Quartz fractures more easily
along some crystal planes than
others.
(Courtesy P.M. Anderson)

The University of Sydney Page 4

Polycrystalline Metals

Aluminium
Ingot
Optical Micrograph
Annealed Pure Zirconium

Optical Micrograph of
silicon steel used for the
Harbour Bridge
The University of Sydney Page 5
 Single vs Polycrystals

• Single Crystals E (diagonal) = 273 GPa Data from Table 3.3,

Callister 7e.
-Properties vary with (Source of data is R.W.
Hertzberg, Deformation
direction: anisotropic. and Fracture Mechanics
of Engineering
-Example: the modulus Materials, 3rd ed., John
Wiley and Sons, 1989.)
of elasticity (E) in BCC iron:
E (edge) = 125 GPa
• Polycrystals
200 µm Adapted from Fig.
-Properties may/may not 4.14(b), Callister 7e.
(Fig. 4.14(b) is courtesy
vary with direction. of L.C. Smith and C.
Brady, the National
-If grains are randomly Bureau of Standards,
Washington, DC [now
oriented: isotropic. the National Institute of
Standards and
(Epoly iron = 210 GPa) Technology,
-If grains are textured, Gaithersburg, MD].)

anisotropic.
The University of Sydney Page 6
 Ni Based Super Alloy: Turbine Blades

The University of Sydney Page 7

 Solidification of Metals

– Metals are melted to produce finished and semi-finished parts.

– Two steps of solidification
 Nucleation : Formation of stable nuclei.
 Growth of nuclei : Formation of grain structure.

– Thermal gradients define the shape of each grain.

Grains
Nuclei

Crystals that will Grain Boundaries

Liquid
The University of Sydney Form grains Page 8
 Formation of Stable Nuclei

– Two main mechanisms: Homogenous and heterogeneous.

– Homogenous Nucleation :
 First and simplest case.
 Metal itself will provide atoms to form nuclei.
 Metal, when significantly undercooled, has several slow moving
atoms which bond each other to form nuclei.
 Cluster of atoms below critical size is called embryo.
 If the cluster of atoms reach critical size, they grow into crystals.
Else get dissolved.
 Cluster of atoms that are grater than critical size are called
nucleus.

The University of Sydney Page 9

 Heterogeneous Nucleation

– Nucleation occurs in a liquid on the surfaces of structural material.

Eg:- Insoluble impurities.
– These structures, called nucleating agents, lower the free energy required to
form stable nucleus.
Liquid

Solid
θ Nucleating
agent

– Nucleating agents also lower the critical size.

– Smaller amount of undercooling is required to solidify.
– Used excessively in industries.

The University of Sydney Page 10

Solidification

Grains can be - equiaxed (roughly same size in all directions)

- columnar (elongated grains)
~ 8 cm

heat
flow

Shell of
Columnar in equiaxed grains
area with less due to rapid
undercooling cooling (greater
∆T) near wall
Adapted from Fig. 4.12, Callister 7e.

Grain Refiner - added to make smaller, more uniform, equiaxed grains.

The University of Sydney Page 11
 Imperfections in Solids

There is no such thing as a perfect crystal.

– What are these imperfections?
– Why are they important?

Many of the important properties of materials are due to the presence of

imperfections.

The University of Sydney Page 12

 Types of Imperfections

• Vacancy atoms
• Interstitial atoms Point defects
• Substitutional atoms

• Dislocations Line defects

• Grain Boundaries Area/surface defects

The University of Sydney Page 13

 Point Defects

• Vacancies:
-vacant atomic sites in a structure.

Vacancy
distortion
of planes

• Self-Interstitials:
-"extra" atoms positioned between atomic sites.

self-
interstitial
distortion
of planes
The University of Sydney Page 14
 Equilibrium Concentration: Point Defects

• Equilibrium concentration varies with temperature!

No. of defects Activation energy

Nv  −Q 
No. of potential = exp  v 
N  kT 
defect sites.
Temperature
Boltzmann's constant
(1.38 x 10 -23 J/atom-K)
(8.62 x 10 -5 eV/atom-K)
Each lattice site
is a potential
vacancy site
The University of Sydney Page 15
 Measuring Activation Energy

• We can get Qv from Nv  −Q 

= exp 
 v 
an experiment. N  kT 
• Measure this... • Replot it...

Nv Nv slope
ln
N N
-Qv /k
exponential
dependence!

T
1/T
defect concentration
The University of Sydney Page 16
 Estimating Vacancy Concentration

• Find the equil. # of vacancies in 1 m3 of Cu at 1000°C.

• Given:
ρ = 8.4 g /cm 3 A Cu = 63.5 g/mol
Qv = 0.9 eV/atom NA = 6.02 x 1023 atoms/mol
0.9 eV/atom
Nv =  −Q 
exp  v = 2.7 x 10-4
N  kT 
1273K
8.62 x 10-5 eV/atom-K
NA
For 1 m3 , N= ρ x x 1 m3 = 8.0 x 1028 sites
A Cu
• Answer:
Nv = (2.7 x 10-4)(8.0 x 1028) sites = 2.2 x 1025 vacancies
The University of Sydney Page 17
 Observing Equilibrium Vacancy Conc.

• Low energy electron microscope

view of a (110) surface of NiAl.

• Increasing T causes surface

island of atoms to grow.

• Why? The equil. Vacancy conc.

increases via atom motion from the
crystal to the surface, where they join
the island.
Island grows/shrinks to maintain
equil. vancancy conc. in the bulk.
Reprinted with permission from Nature (K.F. McCarty,
J.A. Nobel, and N.C. Bartelt, "Vacancies in
Solids and the Stability of Surface Morphology",
Nature, Vol. 412, pp. 622-625 (2001). Image is
5.75 µm by 5.75 µm.) Copyright (2001) Macmillan
Publishers, Ltd.

The University of Sydney Page 18

 Point Defects in Alloys
Two outcomes if impurity (B) added to host (A):
• Solid solution of B in A (i.e., random dist. of point defects)

OR

Substitutional solid soln. Interstitial solid soln.

(e.g., Cu in Ni) (e.g., C in Fe)

• Solid solution of B in A plus particles of a new

phase (usually for a larger amount of B)
Second phase particle
--different composition
--often different structure.
Micrograph of Ti6Al4V
The University of Sydney Page 19
 Imperfections in Solids

Conditions for substitutional solid solution (S.S.)

– W. Hume – Rothery rule
– 1. ∆r (atomic radius) < 15%
– 2. Proximity in periodic table
i.e., similar electronegativities
– 3. Same crystal structure for pure metals
– 4. Valence
All else being equal, a metal will have a greater tendency to dissolve a
metal of higher valence than one of lower valence

The University of Sydney Page 20

 Question Time

Application of Hume–Rothery rules – Solid Solutions

1. Would you predict Element Atomic Crystal Electro- Valence

more Al or Cu to Radius Structure nega-
dissolve in Ni? (nm) tivity
Cu 0.1278 FCC 1.9 +2
2. More Al or Co C 0.071
in Zn? H 0.046
O 0.060
3. More Ag or Pd Ag 0.1445 FCC 1.9 +1
in Ni? Al 0.1431 FCC 1.5 +3
Co 0.1253 HCP 1.8 +2
Cr 0.1249 BCC 1.6 +3
Fe 0.1241 BCC 1.8 +2
Ni 0.1246 FCC 1.8 +2
Pd 0.1376 FCC 2.2 +2
Zn 0.1332 HCP 1.6 +2

The University of Sydney Page 21

 Solubility

– Different solubility (percentage of material B that can be dissolved in

material A) for elements in other metals.

– For example
- Cu in Ni : 100% solubility  1 phase material
- Cu in Al: 19.6 %  if higher percentage, 2 phases
- Cu in Zn: 38.3 %  if higher percentage, 2 phases

The University of Sydney Page 22

 Imperfections in Solids
– Specification of composition
m1 m2
– weight percent C1 = x 100 C2 = x 100
m1 + m2 m1 + m2

m1 = mass of component 1 m2 = mass of component 2

nm1 ' ' nm 2

– atom percent C = 1 x 100 C2 = x 100
nm1 + nm 2 nm1 + nm 2
nm1 = number of moles of component 1

nm2 = number of moles of component 2

There are some relations to go from weight percent to atom percent

The University of Sydney Page 23
 Interstitial Solid Solution
– Solute atoms fit in between the voids (interstices) of solvent atoms.
– Solvent atoms in this case should be much larger than solute atoms.
– Example:- between 912 and 13940C, interstitial solid solution of carbon in γ iron (FCC)
is formed.
– A maximum of 2.8% of carbon can dissolve interstitially in iron.

Iron atoms r=0.129nm

Carbon atoms r=0.075nm

The University of Sydney Page 24
 Question Time

Would it be easier to have a carbon atom as interstitial defect in FCC or BCC

iron?

What are the consequences?

Interstitial sites in BCC Interstitial sites in FCC

The University of Sydney Page 25
Line Defects: Dislocations

Slip lines can be observed Dislocations are visible in electron

micrographs
by optical microscopy

Zirconium sample
Cobalt-Nickel Alloy McCabe et al. (2006) Philosophical Magazine 86 (23)

The University of Sydney Page 26

 Line Defects

Dislocations:
• are line defects,
• slip between crystal planes result when dislocations move,
• produce permanent (plastic) deformation.

Schematic of Zinc (HCP):

• before deformation • after tensile elongation

slip steps

Adapted from Fig. 7.8, Callister 7e.

The University of Sydney Page 27

 Imperfections in Solids

Linear Defects (Dislocations)

– Are one-dimensional defects around which atoms are misaligned
– Edge dislocation:
– extra half-plane of atoms inserted in a crystal structure
– b ⊥ to dislocation line

– Screw dislocation:
– spiral planar ramp resulting from shear deformation
– b || to dislocation line

Burger’s vector, b: measure of lattice distortion

The University of Sydney Page 28

 Imperfections in Solids

Edge Dislocation

Fig. 4.3, Callister 7e.

The University of Sydney Page 29

 Motion of Edge Dislocation

• Dislocation motion requires the successive bumping

of a half plane of atoms (from left to right here).
• Bonds across the slipping planes are broken and
remade in succession.

Atomic view of edge

dislocation motion from
left to right as a crystal
is sheared.

(Courtesy P.M. Anderson)

The University of Sydney Page 30

 Imperfections in Solids

Screw Dislocation

b
Dislocation
line
Burgers vector b (b)
(a)
Adapted from Fig. 4.4, Callister 7e.

The University of Sydney Page 31

 Edge, Screw, and Mixed Dislocations

Mixed

Edge

Adapted from Fig. 4.5, Callister 7e.

Screw

The University of Sydney Page 32

Dislocations & Crystal Structures

• Structure: close-packed view onto two

planes & directions close-packed
planes.
are preferred.
close-packed directions
close-packed plane (bottom) close-packed plane (top)
• Comparison among crystal structures:
FCC: many close-packed planes/directions;
HCP: only one plane, 3 directions;
BCC: none

• Specimens that Mg (HCP)

were tensile
tested. tensile direction
Al (FCC)
The University of Sydney Page 33
 Planar Defects in Solids

Grain Boundaries
– regions between crystals
– transition from lattice of one region
to that of the other
– slightly disordered
– low density in grain boundaries
– high mobility
– high diffusivity
– high chemical reactivity

Adapted from Fig. 4.7, Callister 7e.

The University of Sydney Page 34

 Planar Defects in Solids

– One case is a twin boundary (plane)

– Essentially a reflection of atom positions across the twin plane.

– Stacking faults
– For FCC metals an error in ABCABC packing sequence
– Ex: ABCABABC

The University of Sydney Page 35

 Twin Boundaries

Electron Backscattered Diffraction

Optical Micrograph
orientation map
Annealed twins in brass
Deformed pure Titanium
http://core.materials.ac.uk

The University of Sydney Page 36

 Microscopy Techniques
What do we want to observe?

- Grain size and shape

- Defects in the materials: stacking faults, dislocations, slip
lines, twin and grain boundaries, etc…
- Crystallographic orientation of the grains
- Different phases and their compositions

The University of Sydney Page 37

 Optical Microscope

http://www.zeiss.com
Optical Micrograph of steel
specimen
http://www.olympusmicro.com/primer/virtual/magnifying/index.html
The University of Sydney Page 38
 Optical Microscopy

• Useful up to 2000X magnification.

• Polishing removes surface features: the surface should
have a mirrorlike surface aspect.
• Etching changes reflectance, depending on crystal
orientation.

crystallographic planes

Adapted from Fig. 4.13(b) and (c), Callister

7e. (Fig. 4.13(c) is courtesy
of J.E. Burke, General Electric Co.
The University of Sydney Page 39
Optical Microscopy

Grain boundaries...

• are imperfections, polished surface

• are more susceptible
to etching, surface groove
• may be revealed as grain boundary
dark lines, (a)

• change in crystal
Adapted from Fig. 4.14(a)
orientation across and (b), Callister 7e.
(Fig. 4.14(b) is courtesy
boundary. of L.C. Smith and C. Brady,
the National Bureau of
Standards, Washington, DC
[now the National Institute of
Standards and Technology,
Gaithersburg, MD].)
Fe-Cr alloy
(b)

The University of Sydney Page 40

 Grain Size Measurements
180μm

5 8 9 9 7 9 9 10 9 6 11 10 6 8 9 9
The University of Sydney Page 41
 Grain Size Measurement

– Average number of grains in one line: NAV

NAV = 8.4
– The length of the line is: l = 180 μm

– The average grain size is: dAV

dAV = l / NAV = 180 / 8.4 = 21 μm

The University of Sydney Page 42

 Scanning Electron Microscopy

http://www.zeiss.com

The SEM uses electrons instead of

light to produce an image

The University of Sydney Page 43

http://mse.iastate.edu/microscopy/path2.html
 Secondary/backscattered electrons

– Secondary electrons
low energy electrons that originate within a few nanometers from the
surface
 well defined 3 dimensional images
– Backscattered electrons
high energy electrons originating in the electron beam that are reflected or
back-scattered of the specimen interaction volume
 contrasted image showing regions of different chemical composition

http://www.olympusmicro.com/primer/java/electronmicroscopy/magnify1/index.html
http://www.ammrf.org.au/myscope/sem/introduction/

The University of Sydney Page 44

 Imaging with a SEM More information than with Optical Microscopy

Backscattered electron image of Backscattered electron image of

a sandstone specimen a cast iron specimen
The University of Sydney Page 45
 Transmission Electron Microscopy

http://www.zeiss.com
The University of Sydney www.nanoscience.gatech.edu Page 46
 TEM

– Advantage of this technique

very high resolution can be reached
and materials features that cannot be observed otherwise can be studied
– Drawbacks of this technique
Sample preparation is tedious
Only small area of the sample can be analysed

The University of Sydney Page 47

 Material Defects

www-hrem.msm.cam.ac.uk

Stacking fault observed

in GaN with high
resolution TEM
Courtesy of E. Cerreta

Twin and matrix regions

with dislocations
http://www.ammrf.org.au/myscope/tem/introduction/
The University of Sydney Page 48
 Resolution Ranges

TEM

SEM

Optical Microscope

Naked Eye

10-12 10-10 10-8 10-6 10-4 10-2 1

Resolution Range (m)
Example: the way into steel: http://www.youtube.com/watch_popup?v=-sl2Zy_XKhw
The University of Sydney Page 49