Diffraction Basics

Coherent scattering around atomic scattering centers

occurs when x-rays interact with material
In materials with a crystalline structure, x-rays
scattered in certain directions will be in-phase or
amplified
Measurement of the geometry of diffracted x-rays can
be used to discern the crystal structure and unit cell
dimensions of the target material
The intensities of the amplified x-rays can be used to
work out the arrangement of atoms in the unit cell
The qualitative basics:
The chief result of
the interaction of
X-rays with atoms
in the specimen is
scattering
Scattering is the
emission of X-rays of
the same frequency
(energy) as the
incident X-rays in all
directions (but with
much lower intensity)
v h p = ) cos (cos
The Generalized 2D Laue Equation:
(h is the order of the diffraction, here 0 or 1)
In the specialized case where the angle of
incidence is 90 the equation becomes:
v h p = cos
For a two-dimensional lattice array of atoms, the
Laue equations are:
v h a = ) cos (cos
1 1
v k b = ) cos (cos
2 2
The Laue diffraction cones for the A and B
directions are shown below:
Diffraction will only occur when the diffraction
angles define the same direction. In the case
below this is when the cones intersect to form the
lines OX and OY
In a three-dimensional lattice array, there will be
multiple Laue diffraction cones. Below a simple
diagram shows three first order cones in ABC
space
There are now three Laue equations requiring a
simultaneous solution (i.e., there must be a
diffraction direction common to all three cones):
v h a = ) cos (cos
1 1
v k b = ) cos (cos
2 2
v l c = ) cos (cos
3 3
A unique solution is difficult to obtain
In Laue diffraction, the crystal is fixed and oriented
with a lattice axis parallel to the beam
is varied by using white radiation
With monochromatic radiation, movement of the
crystal is required for diffraction to occur

Radiation Method
White Laue: stationary single crystal
Monochromatic Powder: specimen is polycrystalline, and therefore
all orientations are simultaneously presented to the
beam
Rotation, Weissenberg: oscillation,
De Jong-Bouman: single crystal rotates or oscillates
about chosen axis in path of beam
Precession: chosen axis of single crystal precesses
about beam direction

The Bragg Law
X-ray beam encounters a
3-d lattice array at left.
Assume the following:
A third-order cone
about OA
A second-order cone
about OB
A first-order cone about
OC
We assume these
cones intersect at a
common line satisfying
the diffraction condition.

The rays scattered by
adjacent atoms on OA
atoms have a path
difference of three
wavelengths
Those about OB have
a path difference of two
wavelengths
About OC, one
wavelength difference
These points of
coherent scatter define
a plane with intercepts
2a, 3b, 6c (A, B, C)
and a Miller index of
(321)
A diffraction direction defined by the intersection of the
h
th
order cone about the a axis, the k
th
order cone about
the b axis and the l
th
order cone about the c axis is
geometrically equivalent to a reflection of the incident
beam from the (hkl) plane referred to these axes.
The Bragg Law bottom line:
in other words:
Diffraction from a lattice array of points may be
functionally treated as reflection from a stack of planes
defined by those lattice points
u sin
d
AB =
) 2 (cos
sin
2 cos u
u
u
d
AB AD = =
) 2 (cos
sin sin
u
u u

d d
n =
) sin 2 (
sin
) 2 cos 1 (
sin
2
u
u
u
u
d d
= =
u sin 2d n =
On the previous diagram, the reflected rays combine to
form a diffracted beam if they differ in phase by a whole
number of wavelengths, that is, if the path difference AB-
AD = n where n is an integer. Therefore
and
In the Bragg Law,
u sin 2d n =
, n is the order of diffraction
Above are 1
st
, 2
nd
, 3
rd
and 4
th
order reflections from the (111)
face of NaCl. By convention, orders of reflections are given as
111, 222, 333, 444, etc. (without the parentheses)
The Reciprocal Lattice
How do we predict when diffraction will occur in a
given crystalline material?
How do we orient the X-ray source and detector?
How do we orient the crystal to produce diffraction?
How do we represent diffraction geometrically in a
way that is simple and understandable?
Problems addressed by this unusual mental exercise:
Consider the diffraction from the (200) planes of a (cubic) LiF
crystal that has an identifiable (100) cleavage face.
To use the Bragg equation to determine the orientation required
for diffraction, one must determine the value of d
200
.
Using a reference source (like the ICDD database or other tables
of x-ray data) for LiF, a = 4.0270 , thus d
200
will be of a or 2.0135
.
From Braggs law, the diffraction angle for Cu Ko
1
( = 1.54060)
will be 44.986 2u. Thus the (100) face should be placed to make an
angle of 11.03 with the incident x-ray beam and detector.
If we had no more complicated orientation problems, then we
would have no need for the reciprocal space concept.
Try doing this for the (246) planes and the complications become
immediately evident.
The first part of the problem
Part of the problem is the three dimensional nature of the
diffracting planes. They may be represented as vectors where
d
hkl
is the perpendicular from the origin to the first hkl plane:
While this is an improvement, the graphical representation is
still a mess a bunch of vectors emanating from a single point
radiating into space as shown on the next slide ----
The second part of the problem
Ewald proposed that instead of plotting the d
hkl

vectors, that the reciprocal vector be plotted, defined
as:
hkl
hkl
d
1
*
d
The units are in reciprocal angstroms and defines a
reciprocal space.
The points in the space repeat at perfectly periodic
intervals, defining a space lattice called a reciprocal
lattice
Figure 3.3 can now be reconstructed plotting the
reciprocal vectors instead of the d
hkl
vectors
The comparison is shown in the following slides
Any lattice vector in the reciprocal lattice represents a set of
Bragg plans and can be resolved into its components:
* * * *
c b a l k h d
hkl
+ + =
In orthogonal crystal systems, the d and d* are simple
reciprocals. In non-orthogonal systems, the reciprocals (since
they are vectors) are complicated by angular calculations
Because the angle | is not 90, the calculation of d* and a*
involve the sin of the interaxial angle.
2 / 1 2 2 2
*) cos * cos * cos 2 * cos * cos * cos 1 ( * * *
1
* | o | o + = = c b a
V
V
The table below shows the relationships between axes in direct
and reciprocal space. At the bottom is a very complex
trigonometric function that defines the parameter V used in the
triclinic system.

1
= CO
2
*
) 230 (
d
OA =

u
/ 1
2 / *
sin
) 230 (
d
CO
OA
= =
) 230 (
*
sin 2
d
u
=
) 230 (
) 230 (
*
1
d
d
u sin 2
) 230 (
d =
Figure 3.7 shows the arrangement where the (230) point is brought into
contact with the Ewald sphere.
By definition and
hence
substitution yields:
from the definition
of the reciprocal vector
The Bragg
Relationship!
The Powder Diffraction Pattern
Powders (a.k.a. polycrystalline aggregates)
are billions of tiny crystallites in all possible
orientations
When placed in an x-ray beam, all possible
interatomic planes will be seen
By systematically changing the
experimental angle, we will produce all
possible diffraction peaks from the powder
There is a d*
hkl

vector
associated
with each
point in the
reciprocal
lattice with its
origin on the
Ewald sphere
at the point
where the
direct X-ray
beam exists.
Each crystallite located in the center of the Ewald sphere has its own
reciprocal lattice with its orientation determined by the orientation of the
crystallite with respect to the X-ray beam
The Powder Camera
The Debye-
Scherrer
powder
camera
Debye
diffraction
rings from
the d*
100

reflection.
Note the
1
st
and 2
nd

order
cones, and
back
reflections
Some Debye-Scherrer Powder Films
The Powder Diffractometer
Think of the diffractometer as a device for measuring
diffractions occurring along the Ewald sphere its
function is to move all of the crystallites in the powder
and their associated reciprocal lattices, measuring
diffractions as they intersect the sphere
Because of the operational geometry of diffractometers,
there must be a very large number of small crystallites
(a.k.a., statistically infinite amount of randomly oriented
crystallites) for the diffractometer to see all of the
possible diffractions
By convention (but not by accident note Fig 3.7)
diffraction angles are recorded as 2u. Data are
commonly recorded as 2u and intensity
Conclusions
The geometry of powder diffraction is best
understood through the use of the reciprocal lattice
and the Ewald sphere
The powder diffractometer is a device for directly
applying these constructions to measure d-spacings
in crystalline materials
X-ray diffraction allows direct measurement of the
lattice
Much information about the crystal structure can be
obtained from variations in intensity (and the
complete some reflections in the pattern)
Next week:
Diffraction Intensity:
The rest of the fingerprint
Origin, Variations, Extinctions and
Error Sources in diffraction
experiments