I p = I e F2 M 2
This equation predicts that the intensity diffracted continues to increase as the crystal gets bigger i.e. as M
increases. "This is not observed to be the case!"
As the crystal gets bigger, the chances of a second scattering event occurring become greater and so the
kinematical approach fails,
the above equation will no longer apply.
This second scattering event gives rise to the phenomenon of EXTINCTION, and leads to a REDUCTION
in diffracted intensity.
the above equation for Ip will be valid when EXTINCTION can be ignored.
This is the case for small crystal,
How small? < 10-4 cm thick
i.e. < 1 m thick
note this may be the "effective crystallite size", i.e. we may have a specimen which has a "mosaic"
structure, i.e. that is made up of many small crystallites, each of which may be ~ 1 m across.
Note we have also ignored absorption of the incident beam, i.e. we have ignored any diversion of energy
from the incident beam.
we know already that 2 types of absorption process can occur:
(1) True absorption (or photoelectric absorption)
incident energy KE of ejected photoelectron + potential energy of excited atom ( fluorescent
photon)
(2) Scattering
Energy transfer from incident to scattered beam, this is made up of Incoherent (Compton) +
Coherent scattering
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 296
ESM 512, Structure of Materials Lecture Notes
Clearly as the crystal gets thicker, absorption (i.e. NORMAL ABSORPTION), which can be
considered to be made up from the sum of photoelectric absorption + Compton Scattering) becomes
more important, according to
I = I0 e-t
Note that when the absorption coefficients are measured, coherent scattering is probably not
occurring i.e the specimen is not set in a diffracting position.
To summarize:
I p = I e F2 M 2
Is an equation which refers to the ideal case,
i.e. (1) small perfect crystal
(2) set at the exact Bragg angle for diffraction
(3) by parallel, monochromatic beam
The calculation of the diffracted beam intensity in general is dependent on the particular diffraction
method used.
In the Laue method, this can be difficult (many wavelengths recorded on a film with sensitivity that varies
as a functional of ). Powder method ---- see later.
Note that expression Ip = Ie F 2 M2 refers to the maximum value of the intensity, attained when the
3 Laue Equations are exactly satisfied.
However the peaks in the function y = [sin2Mx] / [sin2x] have a finite breadth.
i.e. there is an appreciable intensity in the diffracted beam when the Laue equations are not exactly
satisfied (even for perfect crystal + parallel beam!)
i.e when (S - S0) / = Hhkl is not exactly satisfied,
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 297
ESM 512, Structure of Materials Lecture Notes
in other words when
(S - S0) differs slightly from Hhkl .
Also this breadth depends on crystal size (through M).
Note that the incident beam is never exactly parallel,
there will always be parts of the incident beam do not have the correct direction S0 to satisfy the
Laue equations.
Also in general, a crystal will possess a property known as "MOSAICITY" where different parts of
the same crystal do not have the same orientation.
even if we could produce a perfectly parallel incident beam, and set part of the crystal at the
correct orientation for diffraction to occur other parts of the same crystal would not be in the correct
orientation.
If we have (as we always do!) a beam of finite divergence, we will get diffracted intensity over range
of angles either side of the exact Bragg Law orientation.
to measure Ip as in our equation we would need an extremely narrow slit, set at the exact 2 angle.
This is not feasible practically
(i.e. all 3 Laue equations will not be simultaneously satisfied for all portions of the crystal and all incident
beam directions) we need a measurable quantity.
Such a quantity is the Integrated Intensity ---- this can be both calculated and measured.
In practice, we can progressively change the orientation of the crystal relative to the incident beam,
and continuously record the intensity of the diffracted beam.
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 298
ESM 512, Structure of Materials Lecture Notes
Start with a roughly collimated beam impinging on a crystal so oriented that the beam makes
approximately the correct for hkl plane
rotate the crystal about an axis parallel to the hkl planes, and normal to the primary beam, at
constant angular velocity, .
Start a few degrees on one side of the exact Bragg setting and rotate to a few degrees on the other
side.
Detector film
Or ionization chamber with slit set wide enough to pick up all of the intensity diffracted
by the hkl planes
All parts of the crystal make the Bragg law Angle with each direction in the incident beam at some
point during the rotation.
Detector surface is
normal to S, and Y’
is parallel to Y.
S0, S in XZ plane,
S’ not in XZ plane
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 299
ESM 512, Structure of Materials Lecture Notes
For convenience, we assume that we rotate the incident beam direction (rather than the crystal) by
angle , in the XZ plane.
Since we are considering all directions in which there might be diffracted energy,
let S' be a vector whose direction describes the new diffracted beam direction, and does not
necessarily still lie in the XZ plane, but is displaced from it (non-parallelism in either the incident beam or
crystal planes may cause this).
The shift of S' with respect to S can be described by the 2 angles and .
is normal to the plane of page
is in the orthogonal direction (in plane of page)
Intensity Energy crossing unit area / unit time
total energy striking the detector surface Y'Z' by integrating the intensity at that surface over time,
and over the area of the detector surface
Jhkl = Ip dt dA
Time it takes for S0' to get from to + d, is dt = d/, = uniform angular speed
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 300
ESM 512, Structure of Materials Lecture Notes
The area element, dA, on the detector given by
dA = dy' dz' = R2 d d
Total diffracted energy
Jhkl = Ip dt dA = R2 / Ip d d d
When beam makes angle with hkl planes, the reflection condition is satisfied
i.e. (S - S0) / = Hhkl
in time dt, (S - S0 ) is displaced by an amount S
such that (S' - S0') = (S - S0) + S
= Hhkl + S
intensity expression corresponding to (S' - S0') .
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 301
ESM 512, Structure of Materials Lecture Notes
(S' - S0') = (S - S0) + S
= Hhkl + S
sin2/(S` - S`0) M1a = sin2/ ( Hhkl + S) M1a
= sin2/ [ (ha* + kb* + lc*) + S] M1a
a a* = 1 a b* etc = 0
sin ( h M1 + / S M1a)
2
we can expand sin2 (A + B) = (sinA cosB + sinB cosA)2
sin2( h M1 + / S M1a) = [sin( h M1) cos(/ S M1a) + sin(/ S M1a) cos ( h M1)]2
M and h are integers
sin( h M1) = 0 cos ( h M1) = 1
= sin2/ (S M1a)
It is convenient to represent S as a vector in reciprocal space
S = (p1 a* + p2 b* + p3 c* )
p1 , p2 , p3 will clearly be very small
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 302
ESM 512, Structure of Materials Lecture Notes
substitute for S in the sine terms
e.g. sin2/S M1a = sin2[/ (p1 a* + p2 b* + p3 c* ) M1a]
= sin2 p1 M1
Jhkl =
In order to evaluate this, we must relate the integration variables ,, and to the reciprocal lattice vector
components pi.
Changes in the angles ,, and can be considered as equivalent to increments in the vectors - S0' and S'
(changes occurring normal to S0' and S' , respectively)
From diagram, displacement of S0 during
time dt is S.
Similarly, the displacement of S can be
described by the small vectors S and S
along two mutually orthogonal directions
lying in the detector surface, where S + S
= (S' - S)
Magnitudes of these 3 vectors are, from
geometry
S = S0 d = d S0 = S = 1
S = S d = d
S = S d = d
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 303
ESM 512, Structure of Materials Lecture Notes
vector displacement of (S - S0)
These vector increments S, S and S,
S = (S' - S0') - (S - S0) correspond to changes in S, i.e., they are
= (S' - S) - (S0' - S0) like three component vectors which go to
= (S + S) - S make S
These 3 vectors are oriented so as to define a small volume element,
e.g. Small change in angle causes S to rotate slightly (and perhaps change length), according to the S
component;
Similarly, changes in and cause changes in S and S components of S
i.e. the terminal point of the vector S traces out a volume element in reciprocal space.
S, S, Z, Z’ all
coplanar. S is
orthogonal to all of these
We can compute this volume (c.f. unit cell volume) as
V = S ^ S S
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 304
ESM 512, Structure of Materials Lecture Notes
= S S S sin2
substituting for these vector magnitudes
V = sin2 d d d
the small angular changes d, d, d
cause the terminal point of the displacement vector S to trace out the volume element V in reciprocal
space.
We can express this volume element in reciprocal space, since
S = (p1 a* + p2 b* + p3 c* )
we can express a small motion of S, in terms of small changes in the coefficients, dp1 , dp2 , dp3 , such
that
V = dp1 a* ^ dp2 b* dp3 c*
= 3 V* dp1 dp2 dp3
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 305
ESM 512, Structure of Materials Lecture Notes
remember V* = 1/V
V = 3 /V dp1 dp2 dp3
equating this with V = sin2 d d d
d d d = 3 / (V sin2) dp1 dp2 dp3
Now we are in a position to carry out the necessary integration.
Jhkl =
Jhkl =
These terms can't yet be integrated from - to +, since there are an number of peaks in this range.
We know that the diffracted peak is very sharp (width of the peak decreases as M increases), in other
words we only get appreciable diffracted intensity close to the ideal position, where the ideal position is
the reciprocal lattice point specified by hkl,
This is when p1 = p2 = p3= 0, Bragg law exactly satisfied.
the 3 quotients have appreciable values only for very small values of p. (i.e. even for large M values,
we only have appreciable values for the quotients when denominator sine arguments are ~ integer or
zero)
Here we only consider 1 peak, i.e. the one centered on p = 0,
p0
sin p p
Also since we only have 1 peak, we can integrate from - to +
definite integral
[sin2 p1 M1 / ( p1 )2] dp1 = M1
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 306
ESM 512, Structure of Materials Lecture Notes
integrated intensity, when Laue equations are closely satisfied
Jhkl =
M1 M2 M3 = M, number of unit cells in the small single crystal.
Multiply Jhkl by V/V -----
Let MV=V be the volume of the small crystal (i.e. the volume irradiated)
And also substituting Ie = (e4/m2 c4R2) I0 (1 + cos22)/2
Jhkl =
This is the total diffracted energy produced when a small single crystal is rotated at angular velocity ,
through the Bragg condition.
If we define a quantity ---- the "reflecting power per irradiated volume element.
we can rewrite the equation for Jhkl as
Jhkl=
it is possible to rearrange the terms in this equation to give a slightly more useful quantity called the
INTEGRATED REFLECTING POWER of a crystal
Units:
which we write as
Rhkl = Jhkl / I0 = Q V
integrated reflecting power, depends only on the irradiated volume, and the reflecting power.
shape is unimportant ---- our assumption of a parallelepiped was justified.
Note:
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 307
ESM 512, Structure of Materials Lecture Notes
We can always assume that a crystal of arbitrary shape, is made up of many small parallelepipeds (note
that the peak intensity does depend on shape)
Note that the integrated reflecting power does not depend directly on either I0 or separately
these two factors tend to cancel themselves out
e.g. if we double the speed , we halve the number of photons that have a chance to be reflected.
1 MAJOR OMISSION (again) is ABSORPTION
remember we have (1) true absorption (2) scattering coherent + incoherent. (1) leads to the further effect
of extinction.
our calculations must only apply to small crystals, where the above effects can be neglected.
we can only really get a measure of RELATIVE reflecting powers of different plan with same crystal.
We can get around this, by considering many such small crystals to be combined into a
POLYCRYSTALLINE aggregate
or a POWDER
DIFFRACTION BY POWDER
Prof. Michael Dudley, Dept. of Materials Science and Engineering, SUNY at Stony Brook 308
ESM 512, Structure of Materials Lecture Notes
