Chapter 12: X-Ray Diffraction: Intensities

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome back to the Deep Dive.

For anyone following along our journey into material science,

you've already conquered the geometry of X -ray diffraction.

You know exactly where the beams land, that predictable order.

It's imposed by the crystal lattice and it's all governed by Bragg's law.

And that's a huge part of the story.

It is, but knowing the coordinates of a spot on a detector that really only gives you half the picture of a crystal structure, doesn't it?

That's the key distinction we're making today, absolutely.

The geometry, it tells you the size and the shape of the unit cell, sort of the box the atoms live in.

The framework.

Exactly, the framework.

But the internal chemistry, you know, what atoms are in there, their precise placement and all the symmetry that relates them, all of that manifests in one single crucial quantity.

And that is the intensity.

The intensity, the brightness of those spots.

So the crucial question we are diving into is,

how bright are they?

Why is the 111 reflection of a material maybe, I don't know, 10 times stronger than the 200 reflection?

And more importantly, why do some reflections that Bragg's law says should be there, like the 100, in a body centered cubic metal just completely vanish?

They're missing.

Right, where do they go?

Our mission today is to really take the fundamental physics of an x -ray hitting one single electron and scale it all the way up through the atom, through the unit cell.

And all the way to the final measurement.

All the way to deriving the total integrated measured intensity.

This deep dive is all about translating that static blueprint of the crystal into the big equation that every crystallographer uses to solve structures.

So we're building it block by block.

Block by block.

Defining every mathematical concept and every physical correction factor along the way.

Okay, let's unpack this journey.

And we have started the most irreducible level, right?

The interaction between an x -ray and just one single solitary electron.

Let's set the stage then.

An x -ray is electromagnetic radiation.

When it encounters a single electron, what is the basic physical mechanism that generates a signal?

What's actually happening?

So the mechanism is what we call x -ray scattering.

The electric field that's associated with the incoming x -ray beam, it acts on the electron.

It forces it into an oscillatory motion.

Makes it wiggle.

It makes it wiggle, exactly.

And a core principle of physics is that any accelerated charged particle, like our oscillating electron, it has to re -radiate electromagnetic energy.

So the electron absorbs and then re -emits.

It re -radiates x -rays.

That's the origin point.

That is the fundamental event of all diffraction.

And we can quantify this.

But I think the first thing to note is that this initial event is incredibly weak.

We use the Thomson equation for this.

What does that tell us, conceptually?

It quantifies the intensity scattered by that single electron.

And you're right, the value is just astronomically small.

We're talking in the range of 10 to the minus 26 or 10 to the minus 27 square meters.

Which is functionally zero for most purposes.

It's effectively nothing.

And the critical implication of that tiny number is that x -ray scattering is an inherently weak process.

This is why you can't use x -rays to image a single atom or a single molecule in isolation.

The signal is just too faint.

It's lost in the noise.

This weakness forces us to rely on a collaborative effect.

We need massive numbers of atoms, like you finding a bulk crystalline material, to all scatter in phase and achieve a measurable constructive interference.

So it's a testament to the power of that crystalline order that we get any signal at all.

Absolutely.

The crystal acts as a huge amplifier for this incredibly weak interaction.

Okay, but even at this most basic level, we have to immediately introduce a correction factor.

And it's related to how the x -ray is produced in the first place.

Polarization.

Yes, absolutely.

The x -ray beam, as it comes out of a typical source like an x -ray tube, is generally unpolarized.

What does that mean practically?

It means the electric field vector, which is the direction the electron is forced to oscillate in, is pointing randomly in any direction that's perpendicular to the beam's path.

So it could be up, down, left, right, diagonal, anything.

Anything in that plane.

And when the electron scatters the x -ray, the intensity of that radiation isn't uniform in all directions.

It actually depends on the orientation of that initial electric field vector.

So if we're trying to measure the scattered intensity, we have to think about the incident beam as having two independent components.

That's how we model it, yes.

We consider one component that's parallel to a specific reference plane and another that's perpendicular to it.

And they don't scatter in the same way?

They don't.

They scatter differently.

When we mathematically combine and average the effects of these two independent components across all possible scattering angles, we arrive at what's called the polarization factor, p of theta.

And the math for that is p of theta equals one plus the cosine squared of two theta, all divided by two.

So this factor has to be applied to our measurement.

What does that relationship tell us about the diffraction pattern itself?

It's a fundamental angular constraint.

It's built into the physics of the experiment.

Because of how the electron re -radiates the x -ray, that factor, p of theta, it reaches its minimum value when the scattering angle, which we call two theta, is exactly 90 degrees.

Okay.

So at 90 degrees, something specific happens.

Right.

At that angle, the cosine of 90 is zero.

So the cosine squared term just vanishes.

And p of theta drops to one half.

So the intensity is mathematically guaranteed to be cut in half at 90 degrees.

What does that actually mean for an experimentalist who's collecting data at that specific angle?

It means the experiment is fundamentally less efficient at 90 degrees.

It's a blind spot, or at least a dim spot.

If you collect data for the same amount of time at, say, 30 degrees and 90 degrees, the 90 degree peak will inherently appear weaker just because of this geometric polarization effect.

So it's not that the structure is scattering less.

It's that your measurement is less sensitive there.

Precisely.

To get equivalent signal to noise across the whole angular range, you often have to compensate.

You have to spend more time counting x -ray at or near 90 degrees to make up for that fundamental intensity loss.

It's a correction that's just baked into the physics of the interaction.

Okay.

Moving beyond the geometry of it, we also have to make a crucial distinction about the type of scattering event that actually produces the peaks we want to measure.

Yes.

We rely entirely on what's called coherent scattering.

Sometimes you'll hear it called This is the ideal event.

The clean one.

The clean one.

It's where the incident photon retains its wavelength and its energy.

It only experiences a 180 degree phase shift.

And crucially, because the wavelength is preserved, all these scattered waves can interfere constructively with waves from neighboring atoms.

That's the source of our ordered diffraction pattern.

That is the signal, but there's also noise.

Right.

The scattering event we don't want.

That's incoherent scattering or Compton scattering.

This happens when the x -ray photon actually loses some energy in the collision with the electron.

Like a billiard ball collision.

A bit like that, yes.

And that results in a change in wavelength.

Since the wavelength changes and the phase relationship is random, these scattered x -rays can't contribute to the ordered diffraction pattern.

They can't build up a peak.

So what do they do?

They just contribute to the uniform background noise on the detector.

It's a haze that we have to subtract out.

For structural analysis, we focus 100 % on that coherent component, hashtag, tag, tag, tag, ca -tag 1 .2, scattering by a single atom, 12 .1 .2.

All right.

So now we scale up.

An atom has an atomic number z, which means it has z electrons.

They're distributed throughout its volume, not just one point.

How do we quantify the combined scattering power of all those electrons?

We use a term called the atomic scattering factor, and it's symbolized by a lowercase f.

It's defined very simply.

It's just the ratio of the amplitude scattered by the entire atom, all z electrons, to the amplitude scattered by our reference point, which is a single electron.

So if f equals five, the atom scatters five times as strongly as one electron.

Exactly.

Now, if all z electrons scattered perfectly in phase with each other, then f would just simply equal z, the atomic number.

When does that happen?

When do you get that perfect maximum value?

Only during what we call forward scattering.

That's when the scattering angle theta is zero.

So straight ahead.

Straight ahead.

If the x -rays are scattered directly forward, the path difference between waves scattered from any electron in the cloud from the inner shell to the outer shell is zero.

They all add up perfectly constructively.

So at zero degrees, f equals z.

It is maximized and exactly equal to the atomic number.

But the moment you look away from zero degrees, even slightly, the path lengths start to differ.

And that's where interference begins to happen within the atom itself.

Exactly.

As the angle theta increases, the path length difference between x -rays scattered from the electrons near the nucleus and the electrons in the outer shell, it becomes large enough to cause cancellation.

This is internal destructive interference.

It is.

As the scattering parameter, which is sine beta over lambda,

increases, the f value drops and it drops rapidly.

And you can visualize this dependency.

If you were to look at a graph of f versus sine theta over lambda, the curve always starts up at z on the y -axis and then curves downward.

Yes.

And crucially, the steepness of that drop -off tells you something about the size of the atom.

How so?

Well, think about a heavy element like gold.

Z is 79.

Its electron cloud is large and spatially distributed.

It's puffy.

The electrons are spread out.

They're spread out.

And because they're further apart, the path differences are greater, even at small angles.

This leads to more immediate destructive interference.

So the curve for gold would drop off very steeply.

Much more steeply than the curve for a light element like carbon, or z is only 6.

Its electron cloud is tightly compact around the nucleus.

This reveals that the spatial distribution of electrons is fundamentally encoded in the angular dependence of f.

This f value is clearly essential for any kind of structural analysis.

But it seems like it would be a really complex calculation, involving quantum mechanics and electron density distributions.

How do crystallographers actually deal with that?

They don't calculate it from scratch every time.

That would be wildly impractical.

They rely on parameterization.

So look up tables.

Essentially, yes.

The f values are calculated once using complex methods.

Then they're fitted to a mathematical function and the parameters of that fit are what get tabulated.

Researchers like Doyle and Turner developed these tables that list curve -fitting parameters, these little a and b coefficients, for a sum of exponential terms.

So these parameters are basically a mathematical model for the shape of the electron cloud.

You just plug the coefficients into the fitted equation, along with your specific scattering angle, and out pops the precise f value you need.

Precisely.

So in practice, if you're solving a structure, you first use Bragg's law to determine the value of sine theta over lambda for the reflection you care about.

Then you look up the parameters for the specific atom in question.

Can you give us an example?

Sure.

Let's say you're looking at the 222 plane of tungsten using copper k alpha x -rays.

First, you calculate the required sine theta over lambda value from the geometry.

Then you look up the tungsten parameters in the table, you insert that geometric value, and you calculate f.

And what happens to f?

In that case, f drops all the way from the atomic number, z equals 74, down to around 38 .8.

Wow, that's almost half.

It's a dramatic reduction.

And it confirms that at that specific angle, the internal destructive interference within the tungsten atom is severe.

That detailed calculation really shows that the intensity of a reflection depends intimately on both the identity of the atom, through z and its parameters, and the exact direction you're measuring, through the angle theta, hashtag, hashtag, hashtag 1 .3.

Scattering by a unit cell, 12 .1 .3.

Okay, we've established how an electron scatters and how an entire atom scatters.

Now we move up the scale one more time to the unit cell.

This is where the magic really happens.

The unit cell contains multiple atoms at specific fractional coordinates, and this is where the geometry of the structure truly dictates the final amplitude.

Right.

The scattering from the unit cell is the summation of the scattering contributions from all the atoms inside it.

But here's the difference.

Unlike the electrons inside one atom, the atoms inside a unit cell can be spaced far enough apart that their scattered waves have really significant path length differences.

And that path difference is what determines everything, whether the waves interfere constructively, destructively, or somewhere in between.

That's it.

And this path difference is what causes this unique phenomenon of extinction, where the physics seems to be cheating the geometry.

It really does feel that way.

Yeah.

Let's walk through that classic scenario.

We start with a simple primitive lattice,

and let's say the Bragg condition is met perfectly for the 100 planes.

We should see a peak.

We should.

Now imagine we add an identical atom at the position one half zero zero.

So we have an atom at the origin corner,

and an identical atom exactly halfway across the A -axis of the cell.

That's the setup.

Now when x -rays scatter off these two atoms in the direction corresponding to that 100 reflection,

the path length difference between the two scattered waves is exactly half a wavelength, lambda over two.

A 180 degree phase shift.

A perfect 180 degree phase shift.

This results in perfect destructive interference.

The wave from the origin atom completely cancels out the wave from that extra atom.

And the resulting reflection intensity is zero.

The peak vanishes.

It's gone.

Wait, just to be clear, the x -rays are hitting the planes at the exact angle required by Bragg's law.

But because of one little atom shifted inside the cell, the resulting peak is extinguished completely.

That's remarkable.

It demonstrates that the internal arrangement of atoms within the unit cell, which is the true definition of the crystal structure,

it overrides the simple geometric condition defined by Bragg's law.

So Bragg's law tells you where reflections can occur.

But the arrangement of atoms, what we'll soon call the structure factor, determines whether they will occur.

But then if we look at the 200 reflection for that same structure, the peak reappears.

Why does just doubling the indices change the interference pattern so drastically?

It's all about the path difference again.

The plane spacing for the 200 planes D200 is half that of the 100 planes.

Right.

So if the plane spacing is halved, the corresponding path difference for the wave scattered by that atom at one half, zero, zero is doubled.

So it goes from lambda over two to a full lambda.

A full wavelength.

Yeah.

And when the path difference is a full wavelength, the waves are entirely in phase again.

And the 200 reflection will yield a strong diffracted beam.

This beautifully connects the real space positioning of atoms to that abstract concept of the reciprocal lattice we've talked about before.

We need an elegant mathematical shortcut to describe that phase difference, which we call phi.

Indeed.

And it turns out that the total phase difference phi between a wave scattered from the origin and a wave scattered from an atom at some position vector r, it simplifies dramatically when you express it in terms of the reciprocal lattice.

How does it simplify?

Phi is just two pi multiplied by the dot product of the real space position vector r and the specific reciprocal lattice vector g h k l that corresponds to the h k l reflection.

And that leads to the really elegant expression we use in practice.

If we define the fractional coordinates of the atom as u v w and the reflection indices as h k l l.

Then the phase difference simplifies all the way on a phi equals two pi times the quantity h u plus k v plus l w.

That expression is the mathematical bridge we need.

It is.

It means we don't have to calculate complex path differences in real space anymore.

We just multiply the indices of the reflection by the coordinates of the atom to get the exact phase relationship.

And this allows us to build up the total scattering amplitude from the entire unit cell.

Okay, so we now have the two key ingredients.

We have the atomic scattering factor f sub j, which tells us the strength of scattering for a given atom j.

Right, it's scattering power.

And we have the phase difference phi sub j, which tells us the direction of that scattering vector relative to the origin.

We combine these to define the master quantity for this entire deep dive.

The structure factor f h k l.

F h k l is the absolute descriptor of the total scattering amplitude from the entire unit cell for one specific reflection h a l.

It is the mathematical encapsulation of both atomic identity through f and atomic position through u v and w.

And crucially, f h k l is always expressed as a sum of complex numbers.

Why do we need the complex plane here?

Why can't we just add and subtract?

Because scattered waves are vectors.

They're defined by both their amplitude, which is their magnitude, and their phase, which is their direction.

The complex plane is simply the natural mathematical tool for summing vectors correctly.

So the magnitude of the vector is f j.

The magnitude is f j.

And the phase difference phi j is represented using Euler's formula in that complex exponential term, e to the i phi j, or more specifically, e to the 2 pi i times h u plus k v plus l w.

The structure factor f h k l is just the summation of these individual scattering vectors for every single atom j in the unit cell.

So we can actually visualize this.

There's something called an Argand diagram, where we're basically performing vector addition in two dimensions.

Exactly.

Imagine the complex plane.

Each atom contributes a vector.

The length of that vector is proportional to its f j.

So how strongly it scatters.

And the angle of the vector is dictated by that phase term phi j, which comes from its position.

f h k l is the final resultant vector that you get from summing all those individual atomic contributions head to tail.

So if that final resultant vector f h k l l is large, we get a strong reflection.

You get a bright spot.

And if all the individual vectors happen to cancel each other out perfectly, pointing in opposite directions, the resulting vector sum is zero.

And you get an extinction, a missing peak.

The measurable intensity, i h h k l, is proportional to the square of the magnitude of that structure factor.

So proportional to the absolute value of f h k l squared.

Since f h k l is a complex number, its magnitude squared is found by multiplying f h k l by its complex conjugate f star.

Correct.

And it's this squaring process, this measurement of only the magnitude.

That's the reason we lose all the phase information of the individual atoms.

That's a fact we're going to come back to later when we discuss Friedel's Law.

Hashtag, tag, tag, tag, 2 .2.

Extinctions due to lattice centering, 12 .2 .1.

Okay, so extinctions, these systematic absences, they're not just mathematical quirks.

They are critical diagnostic tools.

They're often the very first thing a crystallographer looks at to figure out the fundamental Bravais lattice of a new material.

Absolutely.

We can use that structure factor definition to derive the conditions, the rules, for the three major cubic lattices.

Let's start with the simplest, the primitive p lattice.

In a p lattice with one atom type, you only have the origin atom to contribute.

So the structure factor is just f.

There are no other atoms to introduce any cancellation.

So f h k l is generally non -zero for all reflections.

No systematic absences.

None from centering, correct.

Now let's look at a c -centered lattice, c.

This adds an identical atom at the coordinates 1 half, 1 half, 0.

The center of the c face.

When you plug the origin atom and that centering atom into the structure factor equation and you simplify the math, the terms cancel out to zero whenever the sum of the first two indices, h plus k, is an odd number.

So h plus k must be even for a reflection to be seen.

h plus k equals 2 n.

Therefore, reflections like 100 or 120 are systematically absent.

Gone.

Okay, next up is the body -centered lattice.

We add an atom at 1 half, 1 half, 1 half, the very center of the cell.

i centering imposes an even stronger constraint.

The math works out such that the terms cancel when the sum of all three indices, h plus k plus l, is an odd number.

So h plus k plus l must be even?

That's the rule.

And this systematic pattern immediately identifies the lattice as body -centered.

Reflections like 100, 111 in this case.

Oh wait, no, 111 is 3, odd.

No, 110 would be odd.

110, right.

Sum is 2, that's even.

My mistake.

Let's see, 100 is 1, odd, that's out.

110, sum is 2, even, that's in.

111, sum is 3, odd, that's out.

200, sum is 2, even, that's in.

Okay, so you just sum the indices.

If it's odd, the reflection is extinguished.

That's the simple rule for body -centered.

And finally, the most restrictive, the face -centered lattice f, where we add identical atoms at all three face centers.

This is the toughest filter.

The three additional centering atoms, when you add them to the origin atom, they cancel out all reflections where the indices h, k, and l have mixed parity.

Meaning where some are odd and some are even.

Correct.

So only reflections where all three indices are entirely even, like 200 or 220, or entirely odd, like 111 or 311, are present.

Any mixed index reflection, like 100 or 211, is systematically absent.

Systematically absent.

The power of this is just incredible.

Simply by identifying the pattern of missing peaks, you can rapidly and unambiguously determine if the material's lattice is p, i, or f.

It's a beautifully concise application of the mathematics of interference.

Hashtag, tag, tag, tag, tag, 2 .3.

Extinctions due to symmetry elements.

12 .2 .2.

So once we've identified the lattice type from those general conditions, we have to look deeper.

We have to look for additional systematic absences that are caused by the internal symmetry elements that define a specific space group.

Right, things like inversion centers, screw axes, and glide planes.

Let's start with the simplest, inversion symmetry.

If the unit cell has an atom at some position r and an equivalent atom at the inverse position minus r, this relationship greatly simplifies the math of the structure factor.

When you plug both of those positions into the complex structure factor equation, the imaginary parts of the exponential terms, the sine components,

they completely cancel each other out.

Which means if the structure has an inversion center, the structure factor FHKL must always be a real number.

It can be positive or negative, but it has no imaginary component.

Precisely.

This is a critical simplification for people who solve crystal structures.

Now for the more dynamic symmetries, screw axes.

These combine a rotation with a little translation parallel to the axis, and it's that translation that causes the extinction.

It is.

If we have, say, an n -sub -m screw axis that's parallel to the c -axis, we need to analyze the reflections that fall along that specific axis in reciprocal space.

So the zero -zero -pound reflection.

Exactly.

The translation component of the screw axis shifts the phase difference in a very particular way.

The rule that falls out is that the zero -pound reflections are systematically absent unless the index l is a multiple of n.

So for a four -sub -one screw axis, l must be a multiple of four.

Right.

So you would look for zero -zero -one, zero -zero -two, zero -zero -three, and they would all be missing.

But then zero -zero -four would appear.

This is a definitive fingerprint that a four -fold screw axis is present in the structure.

And a similar thing happens for glide plane symmetry, right?

These combine reflection across a plane with a translation parallel to that plane.

A glide plane is similarly diagnostic.

Let's consider a b -glide plane that's parallel to the zero -zero -one plane.

For this, we would analyze reflections that are perpendicular to that plane, like the 8 -TRL reflections.

And what's the rule?

We find that the intensity vanishes for these reflections when the index k is odd.

These systematic absences caused by screw axes and glide planes, they're known as special reflection conditions.

Why special?

Because they only constrain very specific types of reflections, like zero -zero or eight -zero -pound.

This is different from the general reflection conditions we saw from lattice centering, which applied to all HEL reflections.

This systematic analysis of symmetry and its consequences leads us directly to a really fundamental constraint in the entire field of crystallography, which is Friedel's law.

Yes.

Friedel's law is a bit of a tricky one.

It states that the entire set of measured diffraction intensities, i .e.

HKL, will always display a center of symmetry, even if the actual crystal structure does not possess one.

So the pattern of dots looks symmetric, even if the thing that made it isn't.

That's a good way to put it.

In other words, the intensity of the HKL reflection is always equal to the intensity of the inverse reflection, the H -bar, K -bar, L -bar reflections.

Which means we can't tell left from right or up from down using a standard X -ray diffraction experiment.

Why does measuring the intensity cause us to lose this crucial information?

It comes back to what we said earlier.

We're only measuring the magnitude squared, the absolute value of fXKL squared.

We lose the phase.

It's like taking a black and white photograph.

You capture the intensity, the brightness, and the contrast, but you lose all the color information.

And the phase is like the color.

In this analogy, yes.

You can't tell if the light that made the picture traveled from left to right or right to left.

That lost phase information is precisely why we cannot distinguish between a structure and its mirror image or an enantiomer without resorting to more complex techniques.

Standard diffraction can measure the atomic spacing with incredible precision, but it cannot determine the handedness or the polarity of structures that lack inversion symmetry.

Okay,

let's bring this to the practical application.

When a crystallographer gets a new sample, gets a diffraction pattern, how do they actually use these rules to determine the space group?

They work like a detective, systematically filtering the results.

These general and special systematic accents are all meticulously codified in what are called the international tables for crystallography.

It's the Bible of the field.

And what's the first step?

You first identify the Bravais lattice type P, I, or F by applying those general reflection conditions to the entire set of reflections.

That narrows the field of possibilities down from 230 space groups to a much smaller subset.

A much smaller set.

Then you look for the special reflection conditions.

The specific absences along certain axes or planes, which are the unique fingerprints of the screw axes and glide planes.

You check which combination of special conditions in the tables matches the extinctions you observe in your data.

Okay, for instance, the text gives the example of space groups CMMA.

It's C centered.

Right, so the first filter is that H plus K must be even.

Any reflection that fails that test is out.

But CMMA also contains another specific symmetry operation, and a glide plane.

And that a glide plane imposes an additional condition.

It means that for certain types of reflections, the index H must also be even.

So the only reflections that survive and appear in the pattern are those that satisfy both the C centering filter and the glide plane filter.

So the combination of multiple symmetry operations creates a very unique set of observable reflections.

And that unique set points directly to a very limited number of possible space groups.

By carefully analyzing the pattern of what's there and what's missing, you can zero in on the exact space group that defines the crystal symmetry.

This is where we really see the payoff.

Let's run through a few examples where the structure factor tells us something profound about the relationship between atoms, not just the lattice type.

Let's start with example one.

The cesium chloride CSCL structure.

This is a body centered cubic arrangement, but with two different atoms.

Cesium at the origin 0 0 0 and chlorine at the body center 12 12 12.

So because the atoms are different, we have to use FCs and FCL separately in our calculation.

Correct.

Now for a standard BCC lattice with identical atoms, we would expect extinctions whenever H plus K plus L is odd.

But here the atoms are different.

What happens to that rule?

The extinction is partially lifted.

It's not a hard zero anymore.

And this reveals two critically distinct types of reflections based on the parity of H plus K plus L.

What are they?

First, if H plus K plus L is even like for the 200 reflection, the structure factor FHKL is proportional to the sum of the atomic factors.

FCS plus FCL.

So these will be strong.

Very strong.

We call them fundamental reflections.

But second, if H plus K plus L is odd, like for the one reflection,

FHKL is proportional to the difference.

FCS minus FCL.

Ah, so because the structure factor now depends on the difference between the two scattering factors,

these odd index reflections will be very weak, especially since cesium and chlorine have relatively similar numbers of electrons.

Their witness is their strength.

These weak odd reflections are called superlattice reflections, and their very existence is foundational proof that the structure is chemically ordered.

What do you mean by that?

It proves that cesium and chlorine occupy fixed, distinct sites.

If this were a simple random alloy where the two atoms were just randomly on a BCC lattice, you'd use an average atomic scattering factor.

And in that case, the odd reflections would vanish completely, adhering strictly to the BCC extinction rule.

So the fact that they appear at all, even weakly, tells us the material have long -range chemical order.

That's the power of it.

A tiny weak peak confirms a massive structural fact.

Chemical ordering over random distribution.

That is profound.

Okay, what about example, yeah, the NACL structure?

It's a similar story.

It's based on a face -centered cubic arrangement where both the sodium and the chlorine atoms form their own interpenetrating FCC lattices.

So the F -centering rule still applies, absolutely.

Any reflection with mixed parity, like 100 or 210, is completely extinguished.

Gone.

No question.

But for the reflections that survive the F -centering rule, the ones with all even or all odd indices, the structure factor, again, depends on either the sum or the difference of FNA and FCl.

Let me guess.

If HCl are all even, it's the sum.

You got it.

The phase terms add, and FHCl is proportional to the sum, FNA, plus FCl.

Strong peaks.

And if HCl are all odd, it must be the difference.

The phase terms oppose, and FHCl is proportional to the difference, FN minus FCl.

Weaker peaks.

Again, the pattern of strong and weak reflections reveals the complexity of the structure, proving it's an ordered ionic solid and not just a simple single -atom FCC metal.

Okay, finally, let's tackle the really tricky one.

Example 3.

The diamond structure.

This is an FCC lattice, but it has two identical atoms, usually carbon, where one is related to the other by a translation vector of ¼, ¼, ¼.

This is a complex but beautiful case.

It perfectly illustrates how very specific atomic coordinates can impose extreme restrictions on what you can see.

We have the origin atom, and we have the atom shifted by ¼ along all three axes.

And when we calculate FHCl for this diamond structure, what do we find?

We find that the phase relationships, because of that very specific ¼ wavelength displacement,

introduce additional cancellations that go far beyond the standard F -centering rules.

So there's another, even more restrictive condition.

What is it?

The indices must first satisfy the F -centering rule, so they have to be all even or all odd.

And on top of that, the sum H plus K plus L must either be a multiple of 4.

So 4N?

Or it has to be 4N plus or minus 1, so an odd number.

Let's test that.

The 200 reflection is allowed by F -centering, because all the indices are even.

It is.

But the sum 2 plus 0 plus 0 is 2, which is not a multiple of 4, and it's not odd.

So it fails the second rule.

And that means the 200 reflection is systematically absent in the diamond structure.

That's incredible.

That specific, precise ¼ -cell shift ensures that the scattering from the origin atom and scattering from the translated atom are perfectly out of phase for reflections like 200 or 420 or 620, even though the general lattice rule allows them.

This level of prediction where mathematics dictated by a structural model perfectly matches the observed absence of peaks is why X -ray diffraction is such an exact and powerful science.

We've done it.

We've successfully calculated the intrinsic intensity, the absolute value of FHKL squared.

This is the maximum potential scattering power of the unit cell structure.

The theoretical ideal.

But when we move from that theory into the lab, the measured intensity, IHKL, is never solely dependent on the structure factor.

That's right.

The structure factor tells us about the atomic blueprint, but the physical reality of the experiment, things like temperature, sample size, the geometry of the setup, all of these things will attenuate or in some cases boost that signal.

So to match our theory with what we actually measure on the detector, we have to multiply our beautiful FHKL squared by a series of critical correction factors.

Hashtag, tag, tag, tag, 3 .1.

Description of the correction factors, 12 .3 .1.

OK, let's start with the one that's always present, the effect of EAT.

This is the temperature factor, also known as the W.

Waller factor, E to the minus 2m.

The physical principle here is just basic thermodynamics.

Atoms are not static points in space.

They vibrate around their fixed lattice sites because of thermal energy.

They're constantly jiggling.

Constantly jiggling.

And this thermal vibration causes the electron density cloud to be spatially blurred out over time.

And as we established way back in part one, any blurring of electron density reduces the effectiveness of the scattering because it introduces random phase shifts and more internal destructive interference.

Correct.

So our atomic scattering factor, F, has to be corrected.

We multiply it by this W.

Waller factor, E to the minus m.

This factor is always less than one, which means thermal motion always reduces the intensity of the observed peaks.

And the magnitude of that reduction m, what does it depend on?

It's proportional to the mean square displacement of the atom from its ideal lattice site, and it's obviously very temperature dependent.

Higher temperature, more vibration, bigger m, more intensity reduction.

So if the intensity is always reduced by temperature, where in the diffraction pattern is this reduction most severe?

It is significantly more noticeable at high scattering angles, high two theta.

At low angles, the x -ray wavelength is relatively large compared to the small displacement from thermal movement, so the blurring effect is kind of negligible.

But at high angles, where the measurements require extremely precise path length differences to get constructive interference,

even a tiny thermal displacement introduces enough randomness to cause substantial destructive interference.

So the result is that the high angle peaks get squashed down?

A dramatic decrease in the height of high angle peaks compared to what the structure factor alone would predict.

Okay, next correction.

We have the very practical issue of material transparency or lack thereof, the absorption factor A.

Right, x -rays are electromagnetic radiation, and they get partially absorbed as they pass through any material.

The absorption factor A just accounts for the total reduction in intensity due to the path the x -rays have to take through your sample.

The absorption coefficient itself, mu, is a characteristic of the material and the x -ray wavelength.

I imagine this calculation changes drastically depending on the sample.

A thin film versus a big chunk of metal.

Absolutely.

For a flat plate sample, like in a standard powder diffractometer, the path length is simple, and A can be calculated pretty easily.

But for single crystal analysis, where you might have a sphere or a cylinder, or even an irregularly shaped little piece.

The path length is different for every angle.

Different for every angle and every rotation of the sample.

This requires complex geometric modeling, and often leads to the most complicated and error -prone correction in a single crystal refinement.

Okay, our third factor actually boosts the intensity, and it's purely due to the geometry of the crystal system itself.

This is the multiplicity factor.

The multiplicity factor is simply the count of crystallographically equivalent planes, that all contribute to a single measured diffraction angle, two theta.

So in highly symmetric crystal systems like cubic, many different plane orientations just happen to have the exact same interplanar spacing D.

Exactly.

And because they all satisfy the Bragg equation at the exact same angle, they all pile their intensity into a single peak on the detector, which boosts its overall measured power.

Can you give us a quick example?

Sure.

In a cubic crystal, think about the hundred plane.

It has six crystallographically equivalent permutations, 100 minus 1, 0, 0, 0, and so on.

So for the colon of family of planes, the multiplicity P is 6.

But now consider the 11 plane.

It has eight equivalent permutations, 111, 1, minus 1, 1, and so on, plus their inverses.

So for the family, P is 8.

So purely because of this geometric multiplicity, the 11 peak will naturally be about 33 percent stronger than the 100 peak, even if their structure factors were identical.

That's it.

This is why the multiplicity factor is absolutely essential.

We have to normalize our calculations by P to make sure that a big peak intensity is because of a large structure factor, and not just because there are a lot of contributing plane orientations.

Okay.

The final major factor is a two -for -one deal.

The Lorentz polarization factor, LP.

This combines the polarization factor we already talked about with the Lorentz factor itself.

Yes.

LP combines that known polarization attenuation, P of theta, with the purely geometric Lorentz factor, L.

The Lorentz factor is specific to powder diffraction experiments.

And what does it account for?

It deals with the reality of using a random polycrystalline sample.

It basically accounts for the fraction of little crystals in the powder that just happen to be aligned at the exact right angle to satisfy the Bragg condition for that specific 2 theta.

So it's a measure of the rotational probability.

How likely is it that a grain is pointing in the right direction at any given angle?

That's a perfect way to describe it.

The geometry of the powder diffractometer setup means that some angles are just more probable for crystal alignment than others.

When we multiply L and P of theta, the resulting LP function has this complex angular dependence.

It drops off at very low angles.

It peaks, and then it tapers off slowly.

So it's another purely geometric and physical factor that has to be applied to every measured intensity to standardize the results?

It has to be.

Hashtag, hashtag, hashtag, 3 .2, peak shape analysis.

We've been talking a lot about integrated intensity, the total area under the peak.

But physically, the peaks we see on the detector are not infinitely sharp lines.

They have a distinct shape, a profile.

That's right.

The peak profile is typically described by mathematical functions, most commonly a Gaussian or a Lorentzian function.

The most essential measurable quantity we extract from this profile is the full width at half maximum, or FWHM.

And that's just what it sounds like.

It's just what it sounds like.

We denote it by beta.

It's simply the width of the peak measured at exactly half of its maximum height.

And the width of that peak, that beta value, is incredibly important because it tells us about the physical state of the material itself.

It's not about the atoms anymore.

Exactly.

It leads us directly to Scherrer's formula.

This famous formula relates the measured peak width, beta, to the average size of the crystallite grains, which we call D.

The relationship is D equals 0 .9 lambda over beta cos theta.

And what's the principle here?

The principle is that the measurement in reciprocal space, which is the peak width, is inversely related to the dimension in real space, which is the crystallite size.

So if you have a massive, perfect single crystal, the peaks are incredibly sharp.

Beta approaches zero.

And D, the crystal size, is huge.

Conversely, if you synthesize a material that's composed of nanocrystals, maybe the grains are only five nanometers in size, the peaks will be extremely wide and diffuse and blurry.

A large beta.

A large beta, which gives you a small calculated D.

Scherrer's formula, is a cornerstone of nanoparticle and ceramic characterization.

It lets material scientists estimate particle size directly and very rapidly from the diffraction pattern.

Hashtag, hashtag, hashtag, 3 .3.

Expression for total measured intensity, 12 .3 .2.

Let's summarize the entirety of this deep dive.

We started with the gentle oscillation of a single electron and we've built up this entire integrated system, accounting for every single geometric and physical hurdle along the way.

What is the grand final relationship that governs the brightness of a reflection?

The total integrated measured intensity for any given reflection, hkl, which we'll call ihkl, is proportional to the product of every single factor we've discussed today.

Let's list them.

Okay.

It's proportional to the absolute value of the structure factor squared.

That's the core.

Times the Lorentz polarization factor, times the multiplicity factor p, times the Debbie Waller factor e to the minus 2m, and finally times the absorption factor a.

That is the complete expression.

The structure factor, fhkl squared, provides the foundational intrinsic scattering power.

That's determined by where the atoms are and what they are.

And then that intrinsic ideal value is scaled by all the geometric and experimental factors.

The Lorentz polarization factor and the multiplicity factor boost the intensity based on the geometry of the experiment and the crystal.

While the Debbie Waller factor and the absorption factor both reduce the intensity based on thermal motion and how much the sample eats the beam.

That's the whole picture.

And we can see this formula validated in classic textbook examples like polycrystalline tungsten.

When you take the time to meticulously compute all these factors, the structure factor, the temperature factor, the LP factor, the multiplicity, refers several peaks.

What you find is that the resulting theoretical relative intensities match the experimentally measured relative intensities for the metal incredibly closely.

And that alignment between theory and experiment, that's really a profound scientific moment, isn't it?

It really is.

It proves that this entire model, which starts with a single electron and accounts for quantum mechanics, thermodynamics, and solid state geometry, it works.

It works perfectly to describe the atomic blueprint of bulk materials.

Hashtag tag tag outro.

Okay, let's review our essential takeaways from this really deep dive.

We saw how the intensity of a reflection is anything but random.

It is determined by the precise size and position of every single atom in that repeating unit cell.

And then modulated by temperature and the geometry of the experiment.

So what are the non -negotiable points of intensity analysis?

I'd say there are three.

First, the structure factor, FHKL, is the absolute core of the whole process.

It's that complex vector sum where the phase term holds all the positional information and its squared magnitude dictates the intrinsic intensity.

Second,

systematic absences are crucial signals, not mistakes.

The extinctions we observe are the direct signature of the crystal's lattice type P, I, or F, and its internal space group symmetries, like screw axes and glide planes.

They're the first clues to solving a structure.

And the third takeaway.

And third, to convert that calculated intrinsic intensity into something that matches reality, you rely on a handful of essential correction factors that account for thermal blurring, which is the Debye -Waller factor, and the specific experimental conditions, which are the Lorentz polarization, multiplicity, and absorption factors.

We saw throughout this how the intensity can reveal these really subtle but critical structural relationships, especially those examples of cesium chloride and sodium chloride.

The weak superlattice reflections, they were dependent on the difference between the atomic scattering factors, FA minus FB.

That tiny difference in electron count between the two atoms, which is then magnified by their perfectly ordered arrangement,

is what produces that weak superlattice peak.

If that weak peak vanished, it would fundamentally change our interpretation of the material.

It would prove it was a random alloy rather than an ordered compound.

That small, weak superlattice peak carrying massive structural information.

That brings us back to the Debye -Waller factor.

We know the Debye -Waller factor is strongly temperature dependent, and we know it always reduces intensity, especially at high angles.

Consider this for a moment.

What would happen to our ability to detect those weak, but structurally critical superlattice reflections if we performed the entire X -ray experiment at very high temperatures?

Something for you to mull over.

Thank you for engaging in this deep dive into the intensities of X -ray diffraction.