Crystal Structure Analysis

X-ray Diffraction

Electron Diffraction

Neutron Diffraction

Essence of diffraction: Bragg Diffraction

Reading: West 5
A/M 5-6
G/S 3
218
REFERENCES

Elements of Modern X-ray Physics, 2nd Ed. by Jens Als-Nielsen and Des McMorrow,
John Wiley & Sons, Ltd., 2011 (Modern x-ray physics & new developments)

X-ray Diffraction, by B.E. Warren, General Publishing Company, 1969, 1990

(Classic X-ray physics book)

Elements of X-ray Diffraction, 3rd Ed., by B.D. Cullity, Addison-Wesley, 2001

(Covers most techniques used in traditional materials characterization)

High Resolution X-ray Diffractometry and Topography, by D. Keith Bowen and Brian
K. Tanner, Taylor & Francis, Ltd., 1998 (Semiconductors and thin film analysis)

Modern Aspects of Small-Angle Scattering, by H. Brumberger, Editor, Kluwer

Academic Publishers, 1993 (SAXS techniques)

Principles of Protein X-ray Crystallography, 3rd Ed. by Jan Drenth, Springer, 2007
(Crystallography)

219
SCATTERING
Scattering is the process in which waves or particles are forced to deviate from a
straight trajectory because of scattering centers in the propagation medium.

X-rays scatter by interaction with the electron density of a material.

Neutrons are scattered by nuclei and by any magnetic moments in a sample.
Electrons are scattered by electric/magnetic fields.

Momentum transfer: p'  p  q Energy change: E'  E  h

Elastic (E’ = E) For X-rays: E  pc
• Rayleigh (λ >> dobject) Elastic scattering geometry
• Mie (λ ≈ dobject)
• Geometric (λ << dobject)
• Thompson (X-rays)

Inelastic (E’ ≠ E)
• Compton (photons + electrons)
p 
• Brillouin (photons + quasiparticles) q  2 sin
• Raman (photons + molecular vib./rot.)  2
 Compton
COMPTON SCATTERING
Compton (1923) measured intensity of scattered X-rays
from solid target, as function of wavelength for different
angles. He won the 1927 Nobel prize.

X-ray source
Collimator
Crystal
(selects angle)
(selects
wavelength)

θ
Graphite
Target

Detector
Result: peak in scattered radiation
shifts to longer wavelength than
source. Amount depends on θ (but
not on the target material). A. H. Compton. Phys. Rev. 22, 409 (1923).
 COMPTON SCATTERING
Classical picture: oscillating electromagnetic field causes oscillations in positions
of charged particles, which re-radiate in all directions at same frequency and
wavelength as incident radiation (Thompson scattering).
Change in wavelength of scattered light is completely unexpected classically

Incident light wave Oscillating Emitted light wave

electron
Compton’s explanation: “billiard ball” collisions between particles of light
(X-ray photons) and electrons in the material

Before After p 
scattered photon
Incoming photon
θ
p
Electron
pe scattered electron
COMPTON SCATTERING
Before After p 
scattered photon
Incoming photon
θ
p
Electron
pe scattered electron

Conservation of energy Conservation of momentum

hˆ
h  me c  h    p c  m c
2 2 2

2 4 1/ 2
p  i  p   p e

e e

From this Compton derived the change in wavelength:

h
    1  cos 
me c
 c 1  cos    0
h
c  Compton wavelength   2.4 1012 m
me c 223
 COMPTON SCATTERING

Note that there is also an

unshifted peak at each angle.

Most of this is elastic scatter.

Some comes from a collision
between the X-ray photon and
the nucleus of the atom.
h
    1  cos   ≈ 0
mN c

since mN >> me

224
 COMPTON SCATTERING
Contributes to general background noise

Diffuse
background from
Compton
emission by
gamma rays in
a positron
emission
tomography
(PET) scan.

Fluorodeoxyglucose (18F)

225
X-RAY SCATTERING
X-rays: λ (in Å) = 12400/E (in eV)
• 100 eV (“soft”) – 100 keV (“hard”) photons
• 12,400 eV X-rays have wavelengths of 1 Å,
somewhat smaller than interatomic distances in solids
Diffraction from crystals!
Roentgen
1901 Nobel
elastic (Thompson, ΔE = 0)
• wide-angle diffraction (θ > 5°)
• small-angle diffraction (θ close to 0°)
• X-ray reflectivity (films)

inelastic (ΔE ≠ 0)
• Compton X-ray scattering
• resonant inelastic X-ray scattering (RIXS)
First X-ray: 1895
• X-ray Raman scattering
226
DIFFRACTION
Diffraction refers to the apparent bending of waves around small objects and the
spreading out of waves past small apertures.

In our context, diffraction is the scattering of a coherent wave by the atoms in a

crystal. A diffraction pattern results from interference of the scattered waves.
Refraction is the change in the direction of a wave due to a change in its speed.

Crystal diffraction
I. Real space description (Bragg)
II. Momentum (k) space description
(von Laue)

diffraction of plane waves

W. H. Bragg W. L. Bragg von Laue 227

OPTICAL INTERFERENCE

perfectly in phase:
δ = nλ, n = 0, 1, 2, …
δ: phase difference
n: order

perfectly out of phase:

δ = nλ, n = 1/2, 3/2, …
 BRAGG’S LAW OF DIFFRACTION
When a collimated beam of X-rays strikes pair of parallel lattice planes in a crystal,
each atom acts as a scattering center and emits a secondary wave.
 All of the secondary waves interfere with each other to produce the diffracted beam

Bragg provided a simple, intuitive approach to diffraction:

• Regard crystal as parallel planes of atoms separated by distance d
• Assume specular reflection of X-rays from any given plane
→ Peaks in the intensity of scattered radiation will occur when rays
from successive planes interfere constructively

2Θ 229
BRAGG’S LAW OF DIFFRACTION
No peak is observed unless the condition for constructive interference
(δ = nλ, with n an integer) is precisely met:

AC  d sin 

ACB  2d sin 

n  ACB

Bragg’s Law: n  2d sin 

When Bragg’s Law is satisfied, “reflected” beams are in phase
and interfere constructively. Specular “reflections” can
occur only at these angles. 230
 DIFFRACTION ORDERS

1st order:
  2d sin 1

2nd order:
2  2d sin  2

By convention, we set the diffraction order = 1 for XRD.

For instance, when n=2 (as above), we just halve the d-spacing to make n=1.

2  2d sin  2   2(d / 2) sin  2

e.g. the 2nd order reflection of d100 occurs at same θ as 1st order reflection of d200
XRD TECHNIQUES AND APPLICATIONS

• powder diffraction
• single-crystal diffraction
• thin film techniques
• small-angle diffraction

Uses:

• phase identification • crystal size

• crystal structure determination • residual stress/strain
• radial distribution functions • defect studies
• thin film quality • in situ analysis (phase transitions,
• crystallographic texture thermal expansion coefficients, etc)
• percent crystalline/amorphous • superlattice structure
POWDER X-RAY DIFFRACTION
• uses monochromatic radiation, scans angle
• sample is powder → all orientations simultaneously presented to beam
• some crystals will always be oriented at the various Bragg angles
• this results in cones of diffracted radiation
• cones will be spotty in coarse samples (those w/ few crystallites)

no restriction
on rotational orientation
relative to beam

crystallite

  2d hkl sin  hkl

233
Transmission
geometry
234
DEBYE-SCHERRER METHOD

  2d hkl sin  hkl

…or we can use a diffractometer to intercept sections of the cones

235
BASIC DIFFRACTOMETER SETUP

236
 DIFFRACTOMETERS

General Area Detector Diffraction System (GADDS)

THIN FILM SCANS

4-axis goniometer

238
THETA-2THETA GEOMETRY

• X-ray tube stationary

• sample moves by angle theta, detector by 2theta
239
THETA-THETA GEOMETRY

• sample horizontal (good for loose samples)

• tube and detector move simultaneously through theta

240
POWDER DIFFRACTOGRAMS
In powder XRD, a finely powdered sample is probed with monochromatic X-rays of a
known wavelength in order to evaluate the d-spacings according to Bragg’s Law.

Cu Kα radiation: λ = 1.54 Å

peak positions depend on:

• d-spacings of {hkl}
• “systematic absences”

Minimum d?
increasing θ, decreasing d d min   / 2 241
 ACTUAL EXAMPLE: PYRITE THIN FILM
FeS2 – cubic (a = 5.43 Å)
Random crystal orientations Cu Kα = 1.54 Å

“powder pattern”
Intensity

200
210
111 211 311
220

2 Theta
2θ = 28.3° → d = 1.54/[2sin(14.15)] reference pattern from ICDD
= 3.13 Å = d111 (384,000+ datasets)

On casual inspection, peaks give us d-spacings, unit cell size, crystal

symmetry, preferred orientation, crystal size, and impurity phases (none!)
d-SPACING FORMULAS

243
EXAMPLE 2: textured La2CuO4
Layered Cuprate Thin film, growth oriented along c axis

(00l) 2d 00l sin   

c = 12.2 Å 2 theta d (hkl)

Epitaxial film is textured. 7.2 12.2 (001)

(It has crystallographic
orientation). 14.4 6.1 (002)
Many reflections are “missing”
22 4.0 (003)
244
POWDER DIFFRACTION

Peak positions determined by size and shape of unit cell

Peak intensities determined by the atomic number and

position of the various atoms within the unit cell

Peak widths determined by instrument parameters,

temperature, and crystal size, strain, and imperfections

we will return to this later…

245
GENERATION OF X-RAYS
X-rays beams are usually generated by colliding high-energy electrons with metals.

X-ray emission
spectrum
2p3/2 → 1s

Siegbahn notation
Generating Bremsstrahlung

Generating Characteristic X-rays

247
 GENERATION OF X-RAYS
Side-window Coolidge X-ray tube

X-ray energy is determined by anode material, accelerating voltage,

and monochromators:
E  h  hc / 
Co Kα1 : 1.79 Å
Cu Kα1 : 1.54 Å (~8 keV)
Moseley’s Law:  1/ 2
 C (Z   ) Mo Kα1 : 0.71 Å
248
ROTATING ANODES

• 100X higher powers possible by spinning the anode

at > 6000 rpm to prevent melting it → brighter source

249
SYNCHROTRON LIGHT SOURCES
GeV electron accelerators

• brightest X-ray sources

• high collimation
• tunable energy
• pulsed operation

SOLEIL
Bremsstrahlung (“braking radiation”)

250
MONOCHROMATIC X-RAYS
Filters (old way)

A foil of the next lightest element

(Ni in the case of Cu anode) can
often be used to absorb the
unwanted higher-energy radiation to
give a clean Kα beam

Crystal Monochromators
Use diffraction from a curved
crystal (or multilayer) to select
X-rays of a specific wavelength

251
DETECTION OF X-RAYS

Detection principles
• gas ionization
• scintillation
• creation of e-h pairs

• Point detectors

• Strip detectors

• Area detectors

252
DETECTION OF X-RAYS
Point detectors

Scintillation counters

Gas proportional counters

253
X-RAY DETECTORS
Area detectors
• film
• imaging plate
• CCD
• multiwire
Charge-coupled devices

254
X-RAY DETECTORS
Imaging plates
photostimulated luminescence
from BaFBr0.85I0.15:Eu2+

255
X-RAY DETECTORS
Imaging plates
photostimulated luminescence
from BaFBr0.85I0.15:Eu2+

tetragonal Matlockite structure

9-coordinate Ba!
256
The Reciprocal Lattice and the
Laue Description of Diffraction

Reading: A/M 5-6

G/S 3

257
PLANE WAVES
A wave whose surfaces of constant phase are infinite parallel planes of
equal spacing normal to the direction of propagation.

|k|=2π/λ

ψ: wave amplitude at point r

A: max amplitude of wave k
k: wave vector
r: space vector from arbitrary origin

Amplitude is constant in any plane normal to k because k•r is a constant

for such planes:
k k
wavefront k•r1 = kr1
r1 45° r2 k•r2 = kr1√2(cos45) = kr1
origin k•r is indeed constant on wavefronts
THE RECIPROCAL LATTICE
The reciprocal lattice of a Bravais lattice is the set of all vectors K such that

e iK R
1
for all real lattice position vectors R.

Reciprocal lattice: The set of all wave vectors K that yield plane waves
with the periodicity of a given Bravais lattice.

Direct lattice position vectors: R = n1a1 + n2a2 + n3a3

Reciprocal lattice vectors: K = hb1 + kb2 + lb3
where the primitive vectors a2  a3
b1  2
of the reciprocal lattice are: a1   a 2  a 3 
a 3  a1
b 2  2
a1   a 2  a 3  and {ni} and {h,k,l}
are integers
a1  a 2
b 3  2
a1   a 2  a 3  259
 is satisfied when K•R = 2πn, with n an integer

To verify that the {bi} are primitive vectors of the reciprocal

lattice, let’s first show that bi•aj = 2πδij

a2  a3 a1   a 2  a 3 
b1  a1  2  a1  2  2
a1   a 2  a 3  a1   a 2  a 3 

a 3  a1
b 2  a1  2  a1  0 (since cross product of two
a1   a 2  a 3  vectors is perpendicular to both)

a1  a 2
b 3  a1  2  a1  0 Indeed, bi•aj = 2πδij
a1   a 2  a 3 

so, K•R = (hb1 + kb2 + lb3)•(n1a1 + n2a2 + n3a3)

= 2π(hn1 + kn2 + ln3) = 2π × integer
K is indeed a reciprocal lattice vector
 WHAT IS A RECIPROCAL LATTICE VECTOR?
The reciprocal lattice is defined at the lattice generated from the set of all
vectors K that satisfy

e iK R
1
for all direct lattice position vectors R.

What is K?
a wave vector of a plane wave that has the periodicity of the direct lattice

The direct lattice is periodic (invariant under translation by R)

Reciprocal lattice vectors = wave vectors of plane waves that are unity
at all direct lattice sites 261
THE RECIPROCAL LATTICE
• the reciprocal lattice is defined in terms of a Bravais lattice

• the reciprocal lattice is itself one of the 14 Bravais lattices

• the reciprocal of the reciprocal lattice is the original direct lattice

e.g., simple cubic direct lattice

a1  axˆ a 2  ayˆ a 3  azˆ

a2  a3 a2 2
b1  2  2 3 xˆ  xˆ
a1   a 2  a 3  a a
2 2
b2  yˆ b3  zˆ → simple cubic reciprocal lattice
a a with lattice constant 2π/a
→ b1 parallel to a1, etc. 262
Crystals with orthogonal axes (cubic, tetragonal, orthorhombic)
b1, b2, b3 are parallel to a1, a2, a3, respectively.

a1  axˆ a 2  byˆ a 3  czˆ

b3 reciprocal lattice 2
b1  xˆ
a
a3 2
b2 b2  yˆ
direct lattice b
a2
2
b1 a1 b3  zˆ
c
263
RECIPROCAL LATTICE OF FCC IS BCC
FCC primitive vectors:

Note: not orthogonal

a2
(yˆ  zˆ - xˆ )
a2  a3 4 1
b1  2  2 4 3  (yˆ  zˆ - xˆ )
a1   a 2  a 3  a a 2
(2)
8
4 1 4 1
b2  (zˆ + xˆ - yˆ ) b3  (xˆ + yˆ - zˆ )
a 2 a 2
→ BCC reciprocal lattice with lattice constant 4π/a 264
RECIPROCAL LATTICE OF BCC IS FCC
BCC primitive vectors (not orthogonal):

a2
(2yˆ  2zˆ )
a2  a3 4 1
b1  2  2 4 3  (yˆ  zˆ )
a1   a 2  a 3  a a 2
(4)
8
4 1 4 1
b2  ˆ ˆ
( z + x) b3  (xˆ + yˆ )
a 2 a 2

→ FCC reciprocal lattice with lattice constant 4π/a 265

RECIPROCAL LATTICES

• simple orthorhombic → simple orthorhombic

• FCC → BCC

• BCC → FCC

• simple hexagonal → simple hexagonal (rotated)

266
 r. l.

= b1
= b2
β ≠ 90°
= b3
d. l.
267
α,β,γ ≠ 90°

268
Note that these formulas
are missing a factor of 2π

269
FIRST BRILLOUIN ZONES
The Wigner-Seitz cell of the reciprocal lattice is called the first Brillouin zone
(FBZ).

Wigner-Seitz cell: primitive cell with lattice point at its center

d.l. FCC d.l. BCC

r.l. BCC r.l. FCC
1st Brillouin zone: 1st Brillouin zone:

enclosed region is W-S cell

for 2D hexagonal lattice

rhombic dodecahedron
truncated octahedron
270
 L

X (2a 0 0)

271
FIRST BRILLOUIN ZONES

Greek letters: points within the FBZ

Roman letters: points on the FBZ surface
272
3D BAND STRUCTURE
Brillouin Zone of Diamond and
Zincblende Structure (FCC Lattice)
Notation:
• Notation: <=>[100] 
– Zone Edge or  direction
surface : Roman  X<=>BZ edge 
alphabet along [100] 
– Interior of Zone:  direction
Greek alphabet
<=>[111] 
– Center of Zone or 
direction
origin: 
L<=>BZ edge 
along [111] 
direction273
274
Electronic band structure is calculated within the 1st Brilluoin zone

Electronic Band Structure of Si

Eg

<111> <110>
<100>
275
K and LATTICE PLANES
Theorem:
For any family of lattice planes separated by distance d, there are
reciprocal lattice vectors perpendicular to the planes, the shortest of
which has a length of 2π/d.

Conversely, any reciprocal lattice vector K has a family of real-space

planes normal to it, separated by d.

here, g = K

hk in 2D
hkl in 3D

276
MILLER INDICES OF LATTICE PLANES
Orientation of a plane is determined by its normal vector

It is natural to pick the shortest perpendicular reciprocal

lattice vector to represent the normal

Miller indices: coordinates of this reciprocal lattice vector

i.e., A plane with Miller indices hkl is normal to

the reciprocal lattice vector K = hb1 + kb2 + lb3

→ Definition #2: directions in k-space

(Definition #1 was inverse intercepts in the real lattice)

277
 Proof that K = hb1 + kb2 + lb3 is normal to (hkl)

a2 If K = hb1 + kb2 + lb3 is normal to the plane at left,

its dot product with any in-plane vector is zero.
k
Consider vector AB that lies in the plane.

a3
a1
(hkl)
By vector addition,   AB
h l
a1 The dot product,

a3
a1
a

b1

b2

b3
AB h  
3

AB K =    (h k l )
l  h l 
= 2  2  0 etc.
So the reciprocal vector formed by using the Miller indices of a plane as its
components forms a vector in space that is normal to the Miller plane.

Furthermore, the length of the shortest vector K is equal to 2π/dhkl.

a1
K
In the figure above, the spacing between the planes is the projection of on :
a

h K
1

K h 2 2 2
d hkl    → K   K0
h K h K K d hkl 278
REMINDER on ELASTIC SCATTERING
scattering vector
Momentum conservation: p'  p  q
elastic scattering: p'  p  p

Elastic scattering geometry

p 
q  2 sin
 2
von LAUE DESCRIPTION OF DIFFRACTION
• reciprocal space description, equivalent to Bragg description but
more powerful for crystallography & solid state physics

Equivalence to Bragg picture:

2
2d sin   n  n K

k
2
2k sin   n  nK 0  K
d
since scattering is elastic and p  k ,

p 
q  2 sin  2k sin   K
 2
von Laue: “Constructive interference occurs when
qK scattering vector is a reciprocal lattice vector.”
DERIVATION of von LAUE CONDITION

Consider two scatterers:

Path difference between the rays: d cos   d cos ' = d (nˆ - nˆ ' )
Condition for constructive interference: d (n ˆ - nˆ ' ) = n
Multiply through by 2π/λ: d (k - k' ) = 2 n
For the Bravais lattice of scatterers: R (k - k' ) = 2 n

i 2 n
Multiply by i and raise to e: e i ( k' -k ) R
=e 1
Diffraction occurs when the change in
So, k' - k = K wave vector, k’-k, is a vector of the
reciprocal lattice. 281
 K = k'  k Laue condition
k-space Bragg plane
(per. bisector of K) Alternatively,
ଶ

Reciprocal lattice vectors are perpendicular to direct lattice planes

Bragg: diffraction when path length difference = nλ

Laue: diffraction when scattering vector = recip. vector
equivalently, when tip of wave vector lies on a k-space Bragg plane
EWALD (“e-val”) SPHERE
A geometrical construction that provides the relationship between the orientation of
a crystal and the direction of the beams diffracted by it.

A sphere of radius k centered on the base of the incident wave vector

k drawn to the origin O (hkl = 000) of the reciprocal lattice.

Projected Ewald sphere (Ewald circle)

origin of reciprocal space
radius = k

Laue condition:
θ k
K = k'  k k’ θ O
K

(-2,-1)
direction of reciprocal lattice
real space
diffracted beam 283
origin of diffraction
 Diffraction occurs only when a reciprocal lattice point
lies on the surface of the Ewald sphere.

Laue condition:

K = k k'

θ k
k θ
θ O
k’
θ
K K
k’
(-2,-1)

K = 2 / d 210
--
In this case, hkl = -2,-1,0 so diffraction occurs from the (210) planes
and the diffracted beam moves off along k’.

284
285
 In general, a sphere in k-space with the origin on its
surface will have no other reciprocal lattice points on
its surface:

No Bragg peaks for a general incident X-ray!

In order to record diffraction patterns, we must either:

• use polychromatic radiation (vary the sphere size) → Laue method
• rotate the crystal (and thus the reciprocal lattice) → rot. cryst. method
• use a powder sample (equivalent to rotating reciprocal
space about all angles) → powder method
286
287
LAUE METHOD
• fixed crystal & angle, many wavelengths

288
The Laue method is mainly used to determine the orientation of
large single crystals.
When the zone axis lies along the symmetry axis of the crystal, the
pattern of Bragg spots will have the same symmetry. 289
ROTATING CRYSTAL METHOD
• single wavelength
• aligned crystal is rotated about one axis to rotate reciprocal lattice
• produces spots on layer lines

k’

http://escher.epfl.ch/x-ray/diff.mpeg

290
291
POWDER (DEBYE-SCHERRER) METHOD
• single wavelength
• fixed powder sample
• equivalent to rotating the reciprocal lattice through all possible
angles about the origin

every point in Each shell with radius K < 2k

reciprocal space intersects the Ewald sphere to
traces out a shell form a circle.
of radius K
All the diffracted beams from a
powder lie on the surface of cones
292
PEAK INTENSITIES

Peak intensities depend on (in large part):

1) intensity scattered by individual atoms (form factors)
2) the resultant wave from atoms in unit cell (structure factor)

In many cases, the intensity from certain planes (hkl) is zero.

Possible reasons: • symmetry of crystal causes complete cancellation of beam

“systematic absences”
• happenstance

Other factors that affect intensity: • scattering angle

• multiplicities
• temperature factor
• absorption factor
• preferred orientation
293
MONOATOMIC BASES
up to now we have considered diffraction only from Bravais lattices
with single atom bases (i.e., atoms only at the lattice points R).

We found the diffraction condition: ei (k' -k ) R = ei 2 n  1

iK R
which is the same as: e =1
The scattering amplitude FK is the sum over the lattice sites:

FK   f R (K )eiK R
R
where fR(K) is the “atomic form factor” for a given atom (disc. later).

The scattered intensity is proportional to the absolute square of the

amplitude: 2
I K  I 0 FK
…this is what is actually measured in an experiment.
n-ATOM BASES
Crystals with n atoms in each primitive cell must be further analyzed
into a set of scatterers at positions d1, d2 … dn within each primitive
cell.

The positions of the atoms are: A j (R )  R  d j

FK   f j (K )e
iK ( R+d j )
making the scattering amplitude:
R j

  e iK  R  f j ( K ) e
iK d j

L   e iK  R R j

R
“Lattice sum”
 K   f j (K )e
iK d j

j
“Structure factor” of the basis
*If the structure factor = 0, there is no diffraction peak.
STRUCTURE FACTOR
The structure factor gives the amplitude of a scattered wave arising
from the atoms with a single primitive cell.

 K   f j (K )e
iK d j

For crystals composed of only one type of atom, it’s common to split
the structure factor into two parts:

 K  f j (K ) S K

“atomic form factor”

SK   e
iK d j

j
“geometric structure factor”

S = 0 gives a systematic absence (i.e., absence of expected diff. peak).

296
STRUCTURE FACTORS
The amplitude of the rays scattered at positions d1, …, dn
are in the ratios:
iK d1 iK d n
e … e
The net ray scattered by the entire cell is the sum of
the individual rays:
n
Geometric
SK   e
iK d j
-Adds up scattered
structure
waves from unit cell
factor j 1

2
I ( hkl )  SK -In particular, no
peak when SK = 0
297
SIMPLE CUBIC

For simple cubic: one atom basis (0,0,0)

d1 = 0a1 + 0a2 + 0a3

iK 0
SK  e 1
Same result as simple monatomic basis

298
MONATOMIC BCC
For monoatomic BCC:
we can think of this as SC with two point basis (0,0,0), (½,½,½)
2 a   
iK  ( x  y  z )
SK   e
iK d j
 eiK 0  e 2

j 1
2
For SC, K (hxˆ  kyˆ  lzˆ )
a

 1  ei ( h  k l )
h  k l
 1  (1)
S = 2, when h + k + l even
S = 0, when h + k + l odd (systematic absences)
299
 e.g. consider the powder pattern of BCC molybdenum

Powder card shows only even hkl sums b/c Mo is BCC

Why?
- Diffraction from other (hkl) results in destructive interference:

(100) (200)

d100
d200

Beam cancels b/c body center atoms Strong reflection b/c all atoms lie on
scatter exactly 180° out of phase 200 planes and scatter in phase
MONATOMIC FCC

For monoatomic FCC:

SC with four point basis (0,0,0), (½,½,0), (0,½,½), (½,0,½)

4 a  a  a 
iK ( x  y ) iK ( y  z ) iK ( x  z )
SK  e
iKd j
 eiK0  e 2
e 2
e 2

j1

2
For SC, K (hxˆ  kyˆ  lzˆ )
a

SK  1  ei ( h  k )  ei ( k l )  ei ( h l )

S = 4 when h + k, k + l, h + l all even (h, k, l all even or all odd)

S = 0 otherwise. 301
POLYATOMIC STRUCTURES
Atoms of different Z in the unit cell have different scattering
powers, so we explicitly include the form factors:

Total
 K   f j (K )e
iK d j
structure
factor j

{fj }: atomic form factors

 # of electrons

302
CsCl STRUCTURE
Cesium Chloride is primitive cubic
Cs (0,0,0)
 K   f j (K )e
iK d j
Cl (1/2,1/2,1/2)
j

i ( h  k  l )
 K  fCs  fCl e
Φ = f Cs + fCl when h + k + l even

Φ = f Cs - fCl when h + k + l odd

Cs+ and Cl- are not isoelectronic

→ systematic absences unlikely

but what about CsI?

303
Cs+ and I- are isoelectronic, so CsI looks like BCC lattice:
(hkl) CsCl CsI
(100) 
(110)  
(111) 
(200)  
(210)  h + k + l even

(211)  
(220)  
(221) 
(300) 
(310)  
(311)  304
 Diatomic FCC Lattices
Sodium Chloride (NaCl)

Na: (0,0,0)(0,1/2,1/2)(1/2,0,1/2)(1/2,1/2,0)

Add (1/2,1/2,1/2)

Cl: (1/2,1/2,1/2) (1/2,1,1)(1,1/2,1)(1,1,1/2)

305
 i ( h  k  l )
 K  [ f Na  f Cl e ][ SK , FCC ]

 K  [ f Na  fCl ei ( h  k l ) ][1  ei ( h  k )  ei ( h l )  ei (l  k ) ]

Φ = 4(fNa + fCl) when h, k, l, all even

Φ = 4(fNa - fCl) when h, k, l all odd
Φ = 0 otherwise
306
Once again, there are more systematic absences for
isoelectronic ions (e.g., K and Cl)
(110) always absent in RS
(hkl) NaCl KCl
(100)
(110)
(111) 
(200)  
(210)
(211)
(220)   (111) sometimes absent

(221)
(300)
(310)
(311)  307
For RS, we expect the intensity of the all odd reflections
to increase with increasing ΔZ between cation and anion:

I111,311 : KCl < KF < KBr < KI

Less complete destructive interference

between cation and anion sublattices.
308
 DIAMOND STRUCTURE
Diamond: FCC lattice with two-atom basis (0,0,0,), (¼,¼,¼)

a  
iK ( x  y  z )
iK0
SK,diamond  [e e 4
][SK,FCC ]
i( /2)(hk l )
 [1 e ][SK,FCC ]

Only for all even or all odd hkl is S ≠ 0. For these unmixed values,
Additional condition: S = 8 h + k + l twice an even number
S = 4(1 ± i) h + k + l odd
S = 0 h + k + l twice an odd number

IFCC : all nonvanishing spots have equal intensity.

Idiamond : spots allowed by FCC have relative intensities

of 64, 32, or 0. 309
 FCC diamond
(hkl) Al Si
(100)
(110)
(111)  
(200) 
(210)
(211)
(220)  
(221)
(300)
(310) What about
zinc blende?
(311)  
310
SUMMARY OF SYSTEMATIC ABSENCES
crystal structure condition for peak to occur
SC any h,k,l
BCC h + k + l = even
FCC h,k,l all even or all odd
NaCl h,k,l all even,
or all odd if fA ≠ fB
diamond h,k,l all even and twice an even #,
or all odd
HCP any h,k,l except when h + 2k = 3n
and l is odd

 K   f j (K )e
iK d j

j 311
SIMPLE ANALYSIS OF SIMPLE PATTERNS
What will we see in XRD patterns of SC, BCC, FCC?

h k l
2 2 2

SC: 1,2,3,4,5,6,8,9,10,11,12,…

BCC: 2,4,6,8,10,12,...

FCC: 3,4,8,11,12,16,24,…

Observable diffraction
peaks for monoatomic
SC FCC BCC
crystals

We can take ratios of (h2 + k2 + l2) to determine structure.

SIMPLE ANALYSIS OF SIMPLE PATTERNS

1
2d sin   n sin   2
2

d hkl
a
For cubic crystals: d hkl 
h2  k 2  l 2

sin   (h  k  l )
2 2 2 2

sin  nth peak

2
(h  k  l ) nth peak
2 2 2


sin 1st peak
2
(h  k  l )1st peak
2 2 2

313
SIMPLE ANALYSIS OF SIMPLE PATTERNS
α-Fe is cubic. Is it FCC or BCC?
BCC!
110
sin 2  2 sin 2 33
 2
sin 1 sin 22
2 2  211

200

What about Al?

111 FCC!
sin 2  2 sin 2 22.5
  1.33
sin 1
2 2
sin 19 

200 311
220 331 420
222
400
314
Ex: An element, BCC or FCC, shows diffraction
peaks at 2: 40, 58, 73, 86.8,100.4 and 114.7.
Determine:(a) Crystal structure?(b) Lattice constant?
(c) What is the element?
2theta theta normalized (hkl)
sin 
2
h k l
2 2 2

40 20 0.117 1 (110)
58 29 0.235 2 (200)
73 36.5 0.3538 3 (211)
86.8 43.4 0.4721 4 (220)
100.4 50.2 0.5903 5 (310)
114.7 57.35 0.7090 6 (222)

BCC, a =3.18 Å  W
315
ELASTIC X-RAY SCATTERING BY ATOMS
Atoms scatter X-rays because the oscillating electric field of an X-ray sets
each electron in an atom into vibration. Each vibrating electron acts as a
secondary point source of coherent X-rays (in elastic scattering).
The X-ray scattered from an atom is the resultant wave from all its electrons

Particle picture:

• zero phase difference for forward/backward scattering

→ scattering factor (form factor, f ) proportional to atomic number, Z
• increasingly destructive interference with larger scattering angle (to 90°)
• for a given angle, intensity decreases with decreasing X-ray wavelength

1
Thomson relation: I  (1  cos 2 2 )
2
• max scattering intensity at 2θ = 0 & 180°
• gradual decrease to 50% as 2θ approaches 90°
SCATTERING OF X-RAYS BY ATOMS
scattering angle probabilities for a free electron:

Klein–Nishina formula

Low energy: Thomson

High energy: Compton

1
Thomson relation: I  (1  cos 2 2 )
2
ATOMIC FORM FACTORS
Form factor f = scattering amplitude of a wave by an isolated atom

For X-rays, f depends on: • Z (# electrons)

• scattering angle
scattering vector q
• X-ray wavelength
General elastic formula:

3 Cl-
f j (q)    j (r )e  iq r
dr Cl
0 K+
4 sin 
with, q

For θ = 0 (forward scattering),
 O
3
f j (0)    (r )dr = # electrons
0
θ = 37°

consequences: • powder patterns show weak lines at large 2θ.

• light atoms scatter weakly and are difficult to see.
ELECTRON DENSITY MAPS
The electron density as a function of position x,y,z is
the inverse Fourier transform of the structure factors:

1
 ( xyz ) 
V
 hkl
 hkl e  i 2 ( hx  ky  lz )

The electron density map

describes the contents of
the unit cells averaged
over the whole crystal (not
the contents of a single unit
cell)

319
PEAK WIDTHS
Peak shape is a Voigt function (mixture of Gaussian and Lorentzian)
• Gaussian component arises from natural linewidth and strain
• Lorentzian component arises from coherent domain size

Pure
Gaussian

Pure
Lorentzian

Peak width (broadening) is determined by several factors:

• natural linewidth of X-ray emission
• instrumental effects (polychromatic λ, focusing, detector)
• specimen effects
1) crystallite size
2) crystallite strain 320
FULL WIDTH AT HALF MAXIMUM (FWHM)

321
 Instrument and Sample Contributions to the Peak Profile
must be Deconvoluted

• In order to analyze crystallite size, we must deconvolute:

– Instrumental Broadening FW(I)
• also referred to as the Instrumental Profile, Instrumental
FWHM Curve, Instrumental Peak Profile
– Specimen Broadening FW(S)
• also referred to as the Sample Profile, Specimen Profile

• We must then separate the different contributions to specimen

broadening
– Crystallite size and microstrain broadening of diffraction peaks

322
SIZE BROADENING
Small crystallites (< 200 nm) show broadened diffraction lines

Nanocrystal X-ray
Diffraction

323
324
 Which of these diffraction patterns comes from a
nanocrystalline material?

Intensity (a.u.)

66 67 68 69 70 71 72 73 74

2 (deg.)
These diffraction patterns were produced from the same sample!
• Two different diffractometers, with different optical configurations, were used
• The apparent peak broadening is due solely to the instrumentation in
this case
325
The finite size of real crystals results
in incomplete destructive interference
at Bragg angle,  B
over some range of angles

 B1 phase lag between two planes =

perfectly in phase, constructive

2
At some angle 1   B
1
2
Phase lag between two planes:   
3
4 At (j+1)th plane:
Phase lag: 
   j    2
j-1
j
j+1

T = (2j-1)d • Rays from planes 1 and j+1 cancel

2j-1
2j
• Ditto for 2 & j+2, … j & 2j
• Net diffraction over 2j planes = 0
Crystal with 2j planes
Total thickness T The angular range θB to θ1 is the range where
diffracted intensity falls from a maximum to
zero (half of Bragg peak profile).
 Same arguments apply to 2  B
So we see diffracted X-rays over all scattering angles between 2θ1
and 2θ2.

– If we assume a triangular shape for the peak, the full width at

half maximum of the peak will be B = (2θ1 – 2θ2)/2 = θ1 – θ2

327
If we have more than 2j planes: If we have fewer than 2j planes:

 B1  B1
2 2

1 1
2 2
3 3
4 4

j-1 j-1
j j
j+1 j+1

2j+1 2j-3
2j+2 2j-2

Rays from new planes are canceled Rays from planes j-1 & j not canceled
still zero intensity at θ1 nonzero intensity at θ1
Peak sharpens! Peak broadens!
Thinner crystals result in broader peaks! 328
Let’s derive the relation between crystal thickness T and peak width B:

2d sin   
Considering the path length differences between X-rays scattered from the front
and back planes of a crystal with 2j+1 planes and total thickness T:
2T sin 1  (2 j  1)
Here, T = 2jd
2T sin  2  (2 j  1)
1   2 1   2
sin 1  sin  2  2 cos sin
If we subtract them: T (sin 1  sin  2 )   2 2

1   2 1   2
Using trig identity: 2T (cos( ) sin( ))  
2 2
1   2 1   2 1   2
Since  B and sin( ) ,
2 2 2
1   2 1   2 
2T (cos  B )( ))  . But, B  2( ) , so T
2 2 B cos  B
SCHERRER FORMULA
A more rigorous treatment includes a unitless shape factor:

K T = crystallite thickness
Scherrer Formula (1918) T λ (X-ray wavelength, Å)
B cos  B K (shape factor) ~ 0.9
B, θB in radians

Accurate size analysis requires correction for instrument broadening:

B  B B
2 2
M
2
R

BM: Measured FWHM (in radians)

BR: Corresponding FWHM of bulk reference (large grain size, > 200 nm)

Readily applied for crystal size of 2-100 nm.

Up to 500 nm if synchrotron is used.
330
SCHERRER CONSTANT
K 0.94
T T
B cos  B B cos  B
• The constant of proportionality, K (the Scherrer constant)
depends on the how the width is determined, the shape of the
crystal, and the size distribution
– the most common values for K are:
• 0.94 for FWHM of spherical crystals with cubic symmetry
• 0.89 for integral breadth of spherical crystals w/ cubic symmetry
• 1, because 0.94 and 0.89 both round up to 1
– K actually varies from 0.62 to 2.08
• For an excellent discussion of K, refer to JI Langford and AJC
Wilson, “Scherrer after sixty years: A survey and some new
results in the determination of crystallite size,” J. Appl. Cryst. 11
(1978) 102-113.

331
Suppose =1.5 Å, d=1.0 Å, and =49˚. Then for a crystal 1
mm in diameter, the width B, due to the small crystal
effect alone, would be about 2x10-7 radian (10-5 degree),
too small to be observable. Such a crystal would contain
some 107 parallel lattice planes of the spacing assumed
above.

However, if the crystal were only 50 Å thick, it would

contain only 51 planes, and the diffraction curve would be
very broad, namely about 43x10-2 radian (2.46˚), which is
easily measurable.

“Incomplete destructive interference

at angles slightly off the Bragg angles”
332
 What do we mean by crystallite size?
“A single-crystalline domain that scatters coherently”

• A particle may be made up of several different

crystallites (also called grains)
• The crystallites, not the particles, are the coherent
scattering units

333
CRYSTALLITE SHAPE
• Though the shape of crystallites is usually irregular, we can often
approximate them as:
– sphere, cube, tetrahedra, or octahedra
– parallelepipeds such as needles or plates
– prisms or cylinders

• Most applications of Scherrer analysis assume spherical crystallite

shapes

• If we know the average crystallite shape from another analysis, we

can select the proper value for the Scherrer constant K

• Anisotropic crystal shapes can be identified by unequal peak

broadening
– if the dimensions of a crystallite are 2x * 2y * 200z, then (h00) and (0k0)
peaks will be more broadened than (00l) peaks.

e.g., a nanowire 334

STRAIN EFFECTS
L
Strain: 
L

335
SOURCES OF STRAIN

336
 Non-Uniform Lattice Distortions

• Rather than a single d-spacing,

the crystallographic plane has a
distribution of d-spacings
• This produces a broader
observed diffraction peak
• Such distortions can be
introduced by:
– mechanical force

Intensity (a.u.)
– surface tension of
nanocrystals
– morphology of crystal shape,
such as nanotubes
– interstitial impurities 26.5 27.0 27.5 28.0 28.5 29.0 29.5 30.0
2(deg.)
337
THIN FILM SCANS

338
339
 EPITAXY - “above in an ordered fashion”
when one crystal grows on another with a well-defined 3D crystallographic
relationship
Homoepitaxy: epitaxy between identical crystals (e.g., Si on Si)
Heteroepitaxy: the two crystals are different (e.g., ZnO on Al2O3)
requirements = lattice symmetry & lattice constant matching

340
Molecular picture – Si growth on Si (100)

341
Dan Connelly
NTNU 342
Rock salt PbS “nanotrees”

Jin group – U. Wisc. branches grow epitaxially –

each tree is a single crystal 343
 A polycrystalline sample should contain thousands of
crystallites. Therefore, all possible diffraction peaks should
be observed.

2 2 2

• For every set of planes, there will be a small percentage of crystallites that are
properly oriented to diffract (the plane perpendicular bisects the incident and
diffracted beams).
• Basic assumptions of powder diffraction are that for every set of planes there is
an equal number of crystallites that will diffract and that there is a statistically
relevant number of crystallites, not just one or two. 344
 A single crystal specimen in a Bragg-Brentano
diffractometer would produce only one family of peaks in
the diffraction pattern.

2

At 20.6 °2, Bragg’s law The (110) planes would diffract at 29.3 The (200) planes are parallel to the (100)
fulfilled for the (100) planes, °2; however, they are not properly planes. Therefore, they also diffract for this
producing a diffraction peak. aligned to produce a diffraction peak crystal. Since d200 is ½ d100, they appear at
(the perpendicular to those planes does 42 °2.
not bisect the incident and diffracted
beams). Only background is observed.
345
 Wurtzite ZnO nanowire arrays on glass

Poor
uniaxial
texture

Good
uniaxial
c
texture

General route to vertical ZnO

nanowire arrays using textured Biaxial
ZnO seeds. texture
Greene, L. E., Law, M., Tan,
D. H., Montano, M., Goldberger, (growth on Al2O3)
J., Somorjai, G., Yang, P. Nano
Letters 5, 1231-1236 (2005). 346
ROCKING CURVES

347
ROCKING CURVES

348
ROCKING CURVE EXAMPLES

Thickness, composition, and strain state of epitaxial single crystal films

ROCKING CURVE EXAMPLE

(1° = 3600 arcsec)

Thickness, composition, and strain state of epitaxial single crystal films

350
PHI SCANS

sit on a reflection,
then rotate in-plane

351
k-SPACE GEOMETRY

for rotation around [001]

of cubic crystal:

two examples:

monitor {011}: expect 4 peaks separated by 90° rotation.

monitor {111}: expect 4 peaks separated by 90° rotation.
(ignoring possible systematic absences) 352
PHI SCAN EXAMPLE

1 um GaN (wurtzite) on Silicon(111) (1011)

(002)

2-theta scan proves

uni-axial texture phi scan proves
bi-axial texture (epitaxy)
In plane alignment: GaN[1120]//Si[110] 353
Epitaxial YBa2Cu3O7 on Biaxially Textured Nickel (001):
An Approach to Superconducting Tapes with High Critical Current Density
354
Science, Vol 274, Issue 5288, 755-757 , 1 November 1996
 phi

omega

Epitaxial YBa2Cu3O7 on Biaxially

Textured Nickel (001):
An Approach to Superconducting
Tapes with High Critical Current Density

Science, Vol 274, Issue 5288,

755-757 , 1 November 1996

355
TEXTURE MEASUREMENT (POLE FIGURES)

356
POLE FIGURES
• Preferred orientation of crystallites can create a
systematic variation in diffraction peak intensities
– can qualitatively analyze using a 1D diffraction pattern
– a pole figure maps the intensity of a single peak as a
function of tilt and rotation of the sample
• this can be used to quantify the texture

10.0 (111) 00-004-0784> Gold - Au

8.0
Intensity(Counts)

(311)
6.0 (200)
(220)
4.0

2.0
(222)
(400)
3
x10
40 50 60 70 80 90 100
Two-Theta (deg)

357
POLE FIGURE EXAMPLE – PHI ONLY

Example: c-axis aligned superconducting thin films.

(a) (b)

 

Biaxial Texture (105 planes) Random in-plane alignment

358
SMALL ANGLE X-RAY SCATTERING

359
SAXS: diffraction from planes with > 1 nm d-spacing

360
 Small Angle
X-ray Diffraction

Direct Visualization of Individual

Cylindrical and Spherical
Supramolecular Dendrimers
Science 17 October 1997; 278: 449-452
361
 HCP

Triblock Copolymer Syntheses of Mesoporous Silica with Periodic 50 to 300 Angstrom

Pores Science, Vol 279, Issue 5350, 548-552 , 23 January 1998 362
GRAZING INCIDENCE SAXS (GISAXS)

363
IN-SITU X-RAY DIFFRACTION

364
 UCI XRD

Rigaku SmartLab XRD

 0D, 1D, 2D detectors
 In-plane & Out-of-plane
 Thin-film XRD
 High resolution XRD
 SAXS
 μ-XRD
 Capillary transmission
 1500°C heating stage
 1100°C dome stage
366
ELECTRON DIFFRACTION

367
 Why ED patterns have so many spots
Typically, in X-ray or neutron diffraction only one reciprocal lattice point
is on the surface of the Ewald sphere at one time.
In electron diffraction the Ewald sphere is not highly curved b/c of the
very short wavelength electrons that are used. This nearly-flat Ewald
sphere intersects with many reciprocal lattice points at once.
- In real crystals reciprocal lattice points are not infinitely small and in a
real microscope the Ewald sphere is not infinitely thin

λX-ray = hc/E = 0.154 nm @ 8 keV

λe- = h/[2m0eV(1 + eV/2m0c2)]1/2 = 0.0037 nm @ 100 keV

368
DIFFRACTION FROM DISORDERED SOLIDS

amorphous solids

369
DIFFRACTION: WHAT YOU SHOULD KNOW

Index planes
Calculate crystal density
Calculate d-spacings
Derive/use Bragg’s Law
Index diffraction peaks
Determine lattice constants
Reciprocal lattice
Ewald sphere construction
Calculate structural factors, predicting X-ray diffraction pattern
(systematic absences)
Use of Scherrer relation

370
 Term Paper and Presentation
Choose a contemporary materials topic that interests you. For
example:
• Organic LEDs • Synthetic Biomaterials
• Metamaterials • Infrared Photodetectors
• Multiferroics • Conducting Polymers
• Graphene • Inorganic Solar Cells
• Photonic Crystals • Plasmonics
• Amorphous Metals • High κ Dielectrics
• Colossal Magnetoresistance • Quantum Dots

In ≥10 pages of double-spaced text (+ figures and references):

1) Quantitatively explain the basic principles at work
2) Summarize the state-of-the-art in synthesis, properties, and apps
3) Identify a key challenge facing the field and propose an original
solution to this challenge
371
 Term Paper Due Dates

February 27: Topic selection. E-mail me with a one-paragraph

abstract describing your topic, challenge, and proposed solution

8-10 am March 20: Paper submission and 5-minute presentation to

educate the class on your topic.

372
 Midterm Exam

• February 20
• full class period
• short answer questions
• topics up through the end of diffraction
• calculator, pens, pencils, nothing else
• focused on conceptual understanding, not memorization
(b) 4 pts. The graph below shows a single Bragg reflection from a powder sample of large crystallites.
On the same graph, sketch the peak expected for a sample of this solid consisting of small crystals
under substantial uniform compressive stress.
intensity

Peak will be broader and

shifted
to higher 2θ (smaller d-
spacing)

2 theta 373