Analytical Methods for Materials Analytical Methods for Materials

Laboratory Module #5
Determination of Crystallite Size
d L tti St i and Lattice Strain
Suggested Reading Suggested Reading
C. Suryanarayana and M.G. Norton, X-ray Diffraction A Practical Approach, (Plenum Press, New
York, 1998), pages 207-221.
B.D. Cullity and S.R. Stock, Elements of X-ray Diffraction, 3
rd
Edition, (Prentice-Hall, Upper Saddle
River NJ 2001) Ch 5 pages 167-176; and Ch 14 pages 385-402
697
River, NJ, 2001), Ch. 5, pages 167 176; and Ch. 14, pages 385 402.
Objectives
Upon completion of this module, the student will be
able to determine the average crystal size and able to determine the average crystal size and
lattice strain in a specimen using X-ray peak
broadening analyses. broadening analyses.
698
Background
Braggs law assumes that ideal conditions
are maintained during diffraction.
The crystal is perfect
The X-ray beam is collimated and
monochromatic
Etc
In practice, this condition never exists.
699
Ideal Particle Size
With powder XRD, the ideal particle size
depends on the relative perfection of the depends on the relative perfection of the
polycrystalline sample.
Usually 0.5 m to 10 m
What are the implications of smaller crystals?
The number of parallel planes (for XRD) is too
low for a sharp diffraction maximum to build up.
Thus, XRD peaks become broadened.
700
, p
Crystal Size and Lattice Strain
Measurements depend upon:
XRD peak locations
XRD peak widths
Must take peak broadening into account Must take peak broadening into account
What causes peak broadening?
701
Crystal Size and Lattice Strain contd Crystal Size and Lattice Strain cont d
d-spacings directly influence peak width.
What can change d-spacing?
strains
compositional variations
instrumental variables instrumental variables
slits (filtering and focusing)
sample area
diffracting layer thickness
particle size
702
More on the causes of broadening More on the causes of broadening
Instrument effects: Instrument effects:
Imperfect focusing of diffracted beam
Unresolved o and o peaks Unresolved o
1
and o
2
peaks
Finite widths of o
1
and o
2
peaks when
resolved
Crystal size Crystal size
Lattice strain
703
Lattice strain
Some Terminology
Domain size:
Domains are the parts of an XRD specimen that p p
diffract X-rays coherently. They form substructure.
Domains are usually regions that are misoriented
relative to one another (usually by <1) relative to one another (usually by <1 ).
Subgrain boundaries, tilt boundaries, etc
Crystal (or crystallite) size: Crystal (or crystallite) size:
Often used interchangeably with grain size. Refer to
regions separated by large misorientation angles regions separated by large misorientation angles.
Grain size:
704
Single crystal regions separated by large
misorientation angles
Corrections for instrumental broadening
We must first determine the individual
contributions to peak broadening
Peak Broadening
contributions to peak broadening
Peak Broadening
instrumental component
=
instrumental component
+
strain/particle size strain/particle size
705
Corrections for instrumental broadening
Let | = peak width
|
b d
= |
i t t l
+ |
t i d ti l i
|
observed
|
instrumental
|
strain and particle size
or
| = | + | |
o
= |
i
+ |
r What we desire
|
o
is usually determined using well-
annealed powders
706
p
Corrections -- Contd
R it ti f l tti t i d ti l Rewrite equation for lattice strain and particle
size effects:
| | | |
r
= |
o
- |
i
Thi i h ld f L t i
*
k This expression holds for a Lorentzian
*
peak
profile, however, many peaks have a Gaussian

fil profile:
|
r
2
= |
o
2
- |
i
2
Most texts use this
expression for simplicity
When in doubt, however, use:
p f p y
707
( )
( )
2 2
r o i o i
| | | | | =
Corrections -- Contd
Lorentzian
*
peak profile:
The peak falls away slowly as (2-2
o
)
-2
from the center of the
k (2 ) peak (2
o
)
G i

k fil
Gaussian
Lorentzian
Gaussian

peak profile:
The peak is sharper.
Actually there is more to
FWHM
Actually there is more to
it than this.
Modern XRD systems generally include software to
2
o
708
Modern XRD systems generally include software to
assist you in fitting peaks.
Broadening due to crystallite Size g y
Scherrer Formula:
I
max
u |

=
cos
k
D
I
n
t
e
n
s
i
t
y
I
max
|
u

= |
cos D
k
u |cos
D = grain diameter
| = peak width in radians at FWHM after
2u
2u
2
2u
B
2u
1
u cos D
| p
correcting for instrumental broadening
k = 0.9 to 1.0 depending upon grain shape
Equation applies for particle/grain sizes down
into the nm range Works best for particles in the
709
into the nm range. Works best for particles in the
range 2 to 300 nm.
Scherrer Equation
This formula was derived based on the assumption
of Gaussian peak profiles and small cubic crystals of Gaussian peak profiles and small cubic crystals
of uniform size (where k = 0.94).
The constant k actually varies between 0.89 and
1 39 depending upon crystallite size and shape It 1.39 depending upon crystallite size and shape. It
is generally assumed to be around 1.0 which
provides a precision of 10%. p p
Be careful because strain also influences peak
710
Be careful because strain also influences peak
width.
Schematic illustration of influence of grain
size on peak width p
Big grains
(e.g., 1 m)
I/I
o
Small grains
(e.g., 50 nm)
o
2
Keep in mind that peak width can be influenced by
strain as well as particle size.
2
711
strain as well as particle size.
Broadening due to strain
Deformation causes local changes in atomic
spacing spacing.
This causes local changes in the 2 positions
for diffracted x-rays.
Lattice strain causes peak broadening. Lattice strain causes peak broadening.
Broadening due to strain can be represented
by the relationship: by the relationship:
tan
i
q | u =
712
Strain in material
tan
strain
q | u
Variation with Bragg Angle gg g
Amount of peak broadening increases with Bragg
l (u) angle (u).
Most materials exhibit
a
x
Most materials exhibit
broadening due to
crystal size & strain
t

H
a
l
f

M
a
H
M
)
crystal size & strain.
Must use low u peaks
|
particle size
l

W
i
d
t
h

a
t
(
F
W
H
p
to separate effects.
50 0 100 150
|
strain
2u
F
u
l
l
713
2u
Now we can do something!
Width of diffraction peak after subtracting p g
instrumental effects equals the sum of the widths
due to lattice strain and crystal size y
r particle strain
| | | = +
Substituting appropriate equations yields:
tan
cos
r
K
D

| q u
u
= + cos sin
r
K
D

| u q u = +
Thi i i i h
714
This equation is in the
general form of a straight
line (y = mx+b).
Now, plot |
r
cosu vs. sinu . Get straight line w/ slope
the lattice strain q and an intercept equal to K/D the lattice strain, q, and an intercept equal to K/D.
q
c
o
s
u
q
|
r
c
K/D
sinu
Basically, the larger the intercept, the smaller the particle
size.
sinu
size.
The method works best for smaller crystals.
Remember, there will be no peak broadening for
extremely large crystals.
715
Using these relationships you should be able Using these relationships you should be able
to calculate crystallite size and qualitatively
assess strain based upon peak broadening assess strain based upon peak broadening.
716
Steps
1 C ll t d i d diff ti tt 1. Collect and index diffraction patterns
A. Well annealed specimen
B. Deformed specimen p
2. In well annealed specimen, calculate FWHM for
each reflection and determine instrumental
broadening to get
i
3. Repeat calculations for experimental (i.e.,
deformed) specimen to get
o
4. Determine
r
(true width after subtracting
i l b d i instrumental broadening;
2 2
r o
i
| | | =
717
5. Calculate
r
cosu and plot versus sin u
Collect and Index
Well Annealed
Collect and Index
Well Annealed
Deformed
211
y

(
c
p
s
)
200
220
n
t
e
n
s
i
t
y
220 I
n
60 65 70 75 80 85 90 95 100
2
718
Steps
1 Collect and index diffraction patterns 1. Collect and index diffraction patterns
A. Well annealed specimen
B. Deformed specimen B. Deformed specimen
2. In well annealed specimen, determine
FWHM for each reflection and the W
instrumental broadening to get
i
;
3. Repeat calculations for the deformed p
specimen to get
o
;
4. Determine
r
(true width after subtracting
r
( g
instrumental broadening);
5. Calculate
r
cosu and plot versus sin u.
719

r
p
2 2
r o
i
| | | =
Determine FWHMfor Each Determine FWHM for Each
Well Annealed
211
Well Annealed
Deformed
y

(
c
p
s
)
n
t
e
n
s
i
t
y
I
79 80 81 82 83 84 85
2
720
Compile Data
Data
CuK
Annealed particles
CuK
radiation: 1.54056
Peak# 2 () hkl FWHM () FWHM (radians) =
i
1 64.9 200 0.5 8.73E-03
2 82.0 211 0.6 1.05E-02
3 98.5 220 1.1 1.92E-02
Deformed
particles
CuK
radiation: 1.54056
FWHM
Peak# 2 () sin hkl FWHM () =
o
FWHM
(radians) =
o

r
cos
1 64.9 0.5366 200 0.75 1.31E-02
2 2
r o i
| = | |
2 82.0 0.6561 211 1 1.75E-02
721
Remember to convert all peak widths to radians!
3 98.5 0.7576 220 1.5 2.62E-02
Steps
1 Collect and index diffraction patterns 1. Collect and index diffraction patterns
A. Well annealed specimen
B. Deformed specimen p
2. In well annealed specimen, determine
FWHM for each reflection and the
instrumental broadening to get
i
3. Repeat calculations for the deformed
specimen to get
o
4. Determine
r
(true width after subtracting
i t t l b d i instrumental broadening;
5. Calculate
r
cosu and plot versus sin u
722
Data
CuK
Annealed particles
CuK
radiation: 1.54056
Peak# 2 () hkl FWHM () FWHM (radians) =
i
1 64.9 200 0.5 8.73E-03
2 82.0 211 0.6 1.05E-02
3 98.5 220 1.1 1.92E-02
Deformed
particles CuK 1.54056
FWHM
Peak# 2 () sin hkl FWHM () =
o
FWHM
(radians) =
o

r
cos
1 64.9 0.5366 200 0.75 1.31E-02 9.76E-03 8.23E-03
2 82 0 0 6561 211 1 0 1 75E 02 1 40E 02 1 05E 02
2 2
r o i
| = | |
2 82.0 0.6561 211 1.0 1.75E-02 1.40E-02 1.05E-02
3 98.5 0.7576 220 1.5 2.62E-02 1.78E-02 1.16E-02
723
Plot these two quantities
1.40E-02
y = 0.0154x + 1E-04
R
2
= 0.975
1.00E-02
1.20E-02
6.00E-03
8.00E-03

r
c
o
s

6.0x10
-4

2.00E-03
4.00E-03
3.0x10
-4

|
r

c
o
s

u
[K/D] 1 39 x 10
4
0.00E+00
0.0000 0.1000 0.2000 0.3000 0.4000 0.5000 0.6000 0.7000 0.8000
sin
0.0 0.0 0.0
0.0
sin u
[K/D] = 1.39 x 10
-4
Slope of line is proportional to strain.
cos sin
r
K
D

| u q u = +
724
Y-intercept is related to grain/particle size.
D
Particle Size
4
(~1)(1.54056)
1.39 10
k
D D

= =
o
4
1.54056
11, 083 A or 1.1 m
1 39 10
D

= =
4
1.39 10
Strain
Slope of line = 0.0154 or 1.5%
725
This method only provides ballpark numbers
Lets stop for now Let s stop for now.
The rest of this packet includes information The rest of this packet includes information
on residual stress determination.
Well go through later it if we have
time. time.
726