Chapter 18: Metallic Structures II: Topologically Close-Packed Phases

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.

Today we are taking a pretty dramatic step away from the comforting simplicity of standard metallic packing.

Oh yeah.

You know, the FCCs, the HCCs of the world, we are plunging into the deep end of intermetallic crystallography.

This diet is focused entirely on one concept, the existence and architecture of what are called topologically close packed phases.

Right, or TCP structures.

TCP structures.

And if you're looking at metals and alloys that showed, you know, high strength or specific magnetic behaviors or, and this is critical, superconductivity, you find out really fast that simple spheres stack neatly just isn't the whole story.

Not at all.

So our mission today really is to trace the journey from a fundamental energetic drive in metallic bonding all the way to the construction of these incredibly complex crystalline structures.

I mean, some of them have nearly 200 atoms per unit cell.

That sounds monumental.

I mean, we're dealing with density, crystal geometry, some pretty complex math, and the relationship between how atoms are arranged locally and the properties you get on the global scale.

What's the core thesis here?

The central idea, and we'll keep coming back to this, is really this.

In metallic systems,

the energetic drive to minimize the free electron energy, which is, you know, the most basic requirement for stability,

it forces atoms into an arrangement that is locally icosahedral.

Okay, icosahedral.

And that arrangement is so efficient that the crystal lattice will actually sacrifice simple translational symmetry.

It gives that up and the result is these highly complex low symmetry structures known as Frank Casper phases.

So we're going to guide you step by step through the chapter's order, starting with the electronic cause, defining the geometric building blocks, the polyhedra, and then dissecting the specific prototypes.

The A15 lov's sigma phases.

Exactly.

And even the quasicrystal approximants.

Okay, let's start at the very beginning then.

The fundamental energetic reasons for a metal to a certain structure.

This lands us right in section 18 .1, the electronic foundation for metallic structures.

Right.

And to do that, we have to start with the ideal system, which is addressed by the free electron theory, or FET.

And for anyone who needs a quick refresher, what's the core assumption of FET?

FET assumes that the valence electrons are completely delocalized.

They form an isotropic uniform electron gas.

A sea of electrons.

A sea of is zero everywhere.

The electrons are just confined by the boundaries of the positive ion lattice.

Which makes the math very clean, but it completely ignores the reality of a crystalline solid.

Precisely.

In a real crystal, the positive ions are arranged periodically.

And that periodicity creates a potential that the electrons have to respond to.

This moves us out of simple FET and into band theory, where the allowed electron energies are structured by the lattice itself.

So the free electron model is good for some things, like specific heat maybe.

But it can't explain why a metal chooses its specific crystal structure.

It can't.

So if that lattice potential matters, we need a better way to describe how electron energies are distributed within that structure.

And that brings us to two key metrics.

Yes.

Two metrics that are crucial for material stability.

First, the Fermi energy, which is written as epsilon f, and that's the highest electron energy level that's occupied at absolute zero.

And second, the density of states, or g of epsilon.

Okay.

I always think of the density of states as basically a histogram.

It tells you, for any given energy level, how many available quantum states an electron could possibly occupy.

That is a perfect analogy.

And g of epsilon is a crucial structure -dependent property.

It depends on the coordination number, the species of atoms, the atomic volume.

It's the shape of this curve, especially right near the Fermi level, that dictates the chemical and physical properties of the material.

And the fundamental goal for any solid is to minimize its total energy.

So for metals, that means filling the available electronic states up to the Fermi level in the most efficient way possible.

Exactly.

And to connect that abstract idea back to the actual structure, we need to look just for a moment at the underlying quantum mechanics, which is what's covered in box 18 .1, quantifying the free electron system.

Right.

We don't need to solve the equation here on air, but let's walk through the math conceptual, because this really is the foundation.

Okay.

So we start with the quantum mechanical relationship for electron energy, epsilon.

In the free electron model, energy is just proportional to the square of the wave vector k squared.

Okay.

And that wave vector k is itself related to the total number of valence electrons and the volume of the system.

Since electrons have to occupy states in pairs, that's the exclusion principle.

The calculation involves integrating that density of states function we just mentioned.

So conceptually then, the Fermi energy, epsilon f, is what you get when you integrate the density of states until the total area under the curve equals the total number of electrons in your system.

That's the conceptual calculation.

But here's the critical link to structure.

In a real crystal lattice, that periodicity imposes boundaries on the available electron states.

These are the Brillouin zones.

The Brillouin zones, or Jones zones for these more complex structures, yes.

And if the Fermi surface, which represents the highest occupied states,

can just touch or intersect the boundary of that Brillouin zone, it creates a reduction, a dip in the density of states right at the Fermi level.

Ah, okay.

A dip in the density of states means the system can accommodate all its electrons by shifting the Fermi energy down a little bit.

And that lowers the total electronic energy and stabilizes that specific crystal structure.

That's it, exactly.

So the drive to select a structure is really a search for a lattice geometry that optimizes the band structure in just the right way.

And to get that energy minimization, the system needs maximum spatial efficiency.

And maximum spatial efficiency.

That drive to minimize free electron energy leads us straight to a preference for a very specific high coordination local environment, icosahedral coordination.

This is the absolute crux of the entire chapter.

We all know that standard close -packed structures like FCC and HCP give you 12 -fold coordination.

The shape they typically form is called a cuboctahedron.

But the research by Sir Frederick Frank showed that a 12 -fold icosahedral arrangement, where the neighbors form a perfect icosahedron around a central atom, is energetically better.

Why is that?

It's pure geometry.

The distance between the central atom and its 12 neighbors in an icosahedron is about 5 % smaller than the equivalent distance in the FCC cuboctahedron.

5%.

That sounds small, but I'm guessing it's huge energetically.

It's a massive difference.

That small reduction in interatomic spacing results in a significantly lower total bonding energy, especially when you think about the Leonard -Jones potential for metallic bonds.

So a 5 % local density increase is the driver for every single one of these complex structures we're about to talk about.

It is.

This slight geometric advantage makes the icosahedron the most favorable local cluster available.

But when nature tries to build a crystal by maximizing the number of these low -energy icosahedral clusters, it finds it can't do it using the simple periodicities of FCC or HCP.

They just don't fit together that way.

They don't tile space perfectly.

So it has to find a more complex way to arrange these units.

And this is why icosahedral coordination is not just the key to TCP phases, but also to amorphous or glassy alloys where only short -range order matters.

That sets the stage perfectly for Section 18 .2, defining topological close packing TCP.

We get the energetic demand.

Now let's define the geometric solution.

The core geometric feature of these structures is their absolute commitment to high density, and they achieve it in a very specific way.

They are defined by having exclusively tetrahedral interstices.

Okay, this is the key distinction, so let's contrast that.

In normal close packing, FCC or HDP, you have two kinds of voids, right?

You have the big octahedral ones surrounded by six atoms and the smaller tetrahedral ones surrounded by four atoms.

Correct.

And those octahedral voids are larger and inherently higher energy places for atoms to be.

So TCP structures, like the LAVES phases or the superconductor and B3SN, they solve this density problem by geometrically organizing the atoms to completely eliminate the octahedral interstices.

They only allow the smaller lower energy tetrahedral ones.

So they are literally packed in the most dense way possible, using only the smallest possible voids.

And that must force the coordination shells around the atoms to adopt some very specific complex shapes.

It does.

If you demand exclusive tetrahedral packing, the geometry dictates that atoms have to arrange themselves into the fundamental building blocks of all TCP structures,

the Frank -Casper polyhedra.

Which brings us to section 18 .2 .1.

This section introduces these foundational shapes.

Frank and Casper identified four primary coordination shells, which are called CP, for coordination polyhedra.

And they are the four triangulated coordination polyhedra, CN12, which is the icurthohedron, CN14, CN15, and CN16.

The CN just stands for coordination number, so it's the number of neighbors, which is also the number of vertices on the polyhedron.

And that word triangulated is crucial, isn't it?

It's everything.

It means that every single face of these polyhedra is a triangle.

Each triangle is formed by three atoms that define the face of an underlying tetrahedral interstice.

And this structural requirement for only triangular faces is what allows us to define the geometry mathematically, using this really profound constraint detailed in box 18 .2, geometry in Euler's theorem.

Right.

So Euler's theorem, v minus e plus f equals two, is a classic way to relate the number of vertices, edges, and faces of any convex polyhedron.

But for these Frank -Casper phases, the constraint that all faces must be triangles forces a very strict identity on the vertices.

Tell us how that constraint plays out.

Well, because the packing is so tight, every single vertex on a Frank -Casper polyhedron has to be connected to either five or six other vertices,

any fewer, and you'd create non -triangular faces, which violates the whole tetrahedral packing rule.

So we can classify all the vertices as either N5, meaning five -fold vertices, or N6, six -fold vertices.

So by substituting the relationships that you get from having all triangular faces into Euler's formula, they derive this incredibly powerful universal rule.

They did.

The final identity is, well, simplifies beautifully.

Since the coordination queue can only be five or six, the math works out such that for every single Frank -Casper polyhedron, regardless of its total size, the number of five -fold vertices, N5, must be exactly 12.

Wow.

That is a profound geometric constraint.

It means that icosahedral -like local symmetry is baked into all TCP structures, even the most complicated ones.

It's the geometric signature of maximum tetrahedral packing.

And if you look at the examples in table 18 .1, you see this perfectly.

Okay.

Let's start with the CN12 icosahedron.

It has 12 vertices total.

And since N5 must be 12, well, it follows that N6 must be zero.

Perfect.

It's the highest symmetry, purely five -fold coordinated polyhedron.

Now, let's jump to the largest one, the CN16 Priusov polyhedron, which is the structural core of the Lave's faces we'll get to.

It has 16 vertices total, since 12 of them have to be five -fold.

The other four must be six -fold.

So N6 equals four.

Exactly.

And that makes the CN16 polyhedron much more distorted and complex because it's designed to fit into a larger crystal structure alongside other polyhedra.

The CN14 and CN15 just fall in between with two and three six -fold vertices, respectively.

So how do these complex shapes actually fit together to build a crystal?

That's section 18 .2 .2, connectivity and stacking.

Well, this raises a good question.

If the CN12 icosahedron is the most stable, most perfect shape, why does the crystal structure even need the distorted CN14, CN15, and CN16 shapes at all?

That's a great point.

If the whole goal is to maximize icosahedra, using a CN16 seems kind of counterintuitive.

It's the classic difference between a local preference and a global space filling requirement.

You can't tile space efficiently using only CN12 icosahedra.

They don't pack perfectly into a periodic lattice.

The CN14, CN15, and CN16 polyhedra are the necessary geometric distortions required to bridge the gaps and connect all those low -energy icosahedral regions into a full three -dimensional periodic crystal.

So they're the architectural connectors linking the stable CN12 regions together.

And how do they physically link up?

They connect primarily through the atoms that define those six -fold coordinated vertices, the N6 ones.

These atoms are so structurally critical they have special name, major ligands.

They form the structural skeleton of the entire phase.

These atoms are shared between the larger polyhedra creating these robust networks.

So you can think of the whole structure as a framework built by these highly connected six -fold vertices and then the larger icosahedrally dominated polyhedra are just filling in the space between them.

Exactly.

And when you look at the major atomic layers in these structures, the source material notes these often form 2D arrays, you see these incredibly complex tilings.

We're talking way beyond the simple all -triangle tiling you see in basic close -packing.

These often have intricate combinations of hexagons and pentagons.

And this complex layered structure is what gives many TCP phases their pronounced anisotropy.

Meaning their properties like thermal expansion or magnetic response vary strongly depending on which direction you measure Okay, before we dive into the specific structures, a quick note on section 18 .2 .3, metallic radii.

The source material uses these simple numbers for comparison.

Yeah, and metallic radii are just arrived from half the observed bond distance in the pure elemental crystal.

Now we know bonding in an alloy isn't purely hard sphere packing, but these radii, like the ones in table 18 .2, are an excellent guide for understanding the size mismatch between atoms that drives the actual bond distances in the alloy are usually pretty close to what you'd predict from the pure elements.

Perfect.

So we've established the energetic, geometric, and topological rules.

Now let's get into the tangible structures in section 18 .3.

Structural prototypes of Frank Casper alloy phases.

We start with the A3B structure, the A15 phase, which has the CR3Si prototype.

A15 compounds are foundational in material science.

They famously include NB3SN, which is a type two superconductor critical for high field magnets like an MRI machines.

The structure is cubic with the stoichiometry of A3B.

So tell us about the coordination environment inside the unit cell.

The structure uses two main polyhedra.

The larger B atoms, like the silicon or tin, sit at the cube corners and the body center, and they are surrounded by the CN14 Casper polyhedron.

The smaller A atoms, like chromium or niobium, are surrounded by the 8 atoms total.

8 atoms total, 2 B atoms, and 6 A atoms.

Now if we look at a visualization of this structure, like in figure 18 .4, what's the most striking feature of those A atoms?

The most striking feature, absolutely, is that the A atoms form chains.

These chains run parallel to the cube axes, the 100, 010, and 001 directions.

The key insight here is that the very close spacing of atoms along these linear chains is believed to be the source of the remarkable electronic properties.

So this brings us right back to the electronic foundation.

How does the geometry of the A15 structure lead to its high temperature superconductivity?

It connects directly back to that density of states curve, the G of epsilon.

The existence of these very close packed linear chains of A atoms creates this highly anisotropic, almost one -dimensional electronic environment, and this results in incredibly sharp localized peaks in the density of states curve positioned precisely at or very near the Fermi level.

So the geometry is literally sculpting the electron energy landscape.

It is, and a high density of states at the Fermi level means the system is highly susceptible to coupling between electrons and phonons, which are lattice vibrations.

That high coupling strength is the physical mechanism behind high transition temperature superconductivity in BCS theory.

So the complex geometry of A15 is the direct physical cause of And to verify a structure this specific, you need sophisticated experimental tools.

Which brings us to a conceptual dive into Box 18 .3.

Calculation of the structure factor and extinctions for the CR3SI structure.

This is how you use X -ray diffraction to prove the structure is real.

Right, the structure factor, FHKL, is the mathematical expression that tells you the intensity of the diffraction spot for the indices.

You calculate it by summing up the atomic scattering factors of every atom in the unit cell, each scaled by a phase factor based on its fractional coordinates.

And for CR3SI, we have eight atoms in the cell.

So we need the coordinates for the two silicon atoms and the six chromium atoms.

The two silicon atoms, the B atoms, are easy.

They're at 0, 0, 0, and 12, 12, 12.

A simple body centered arrangement.

But the six chromium atoms, the A atoms, are the ones that define the complexity.

They occupy positions like 14, 0, 12, and its symmetry equivalence.

So the summation for those six chromium atoms is what makes the final structure factor expression so complicated.

It is.

And when you substitute those specific fractional coordinates into the FHKL sum, the algebra leads to certain reflections canceling out perfectly.

These are the extinction rules.

We've mentioned this before, but explain the extinction rule here conceptually for us.

The final math shows that if the indices H, K, and L are all odd, the contribution from the silicon atoms and the contribution from the chromium atoms exactly cancel each other out.

They're perfectly out of phase, and you get FHKL equals 0.

So there's no reflection.

No spot on the diffraction pattern.

No spot.

But this is only true unless a secondary condition is met.

Something like over 2 plus K over 2 being odd.

So the absence of reflections like 111 or 311 isn't random.

It's the unique fingerprint of this specific space group geometry.

It's a direct consequence of the precise location of those six chromium atoms forming their linear chains.

Exactly.

That's the power of crystallography.

These complex geometric features like the CN12 chains are imprinted mathematically onto the diffraction pattern through these extinction rules.

We can also visualize the A15 structure by looking at it as stacked 2D tilings.

Yeah.

If you view the A15 structure looking down the 01 plane, you see this complex stacking.

The formed piling.

A mix of triangles and squares repeating periodically with secondary layers of atoms in between.

This layered view is often the best way to wrap your head around structures that are hard to visualize in 3D.

Okay.

Let's move to structures that are related by mechanical transformation.

Section 18 .3 .2 on shear related structures like AL2Z, AR3, and BF2.

Right.

The concept here is polymorphism.

How minor changes, often from mechanical stress or shear, can shift material from one TCP phase to another.

A conservative shear is a localized change, like one layer sliding over another.

A non -conservative shear is more drastic.

It involves atoms moving perpendicular to the shear plane, maybe even changing the composition.

The AL2Z, AR3 structure is closely related to A15.

How do we visualize its architecture?

AL2Z, AR3 is complex, but you can simplify it by seeing it as a stack of two repeating 2D tilings, which you can see in figure 18 .9.

It uses the famous Kagome tiling, which has repeating hexagons and triangles, stacked with a more intricate tiling.

The larger zirconium atoms form the backbone of these tilings, and the smaller aluminum atoms sit in the interstitial layers.

And BF2 is then related to AL2Z, AR3 by a non -conservative shear.

Yes, and this just demonstrates that these complex Frank Casper phases aren't isolated islands.

They're structurally linked.

Small changes can transform one into another, which is critical for understanding phase stability in real -world alloys.

We're transitioning now to maybe the most common and robust family of TCP structures.

Section 18 .3 .3, the LAVES phases, which are AB2 compounds.

LAVES phases are everywhere.

They're observed in thousands of intermetallic systems, and their stability is driven by three really clear criteria, which makes them highly predictable.

What are those three criteria?

First, you need a strong tendency for metallic bonding.

Second, the smaller B atom usually needs an incomplete outer shell, so it's often a transition metal.

But the third, and the most dominant criterion, is geometric.

The size difference between the A and B atoms has to be significant.

We're talking at least 20 % or more.

That large size mismatch forces the structure into an arrangement where the small B atoms can perfectly fill the voids created by the large A atoms, just maximizing density.

And this is all quantified by the radius ratio.

Exactly.

The ideal radius ratio for a hard sphere AB2 structure is the square root of 3 over 2, which is about 1 .225.

That ratio mathematically guarantees the highest theoretical packing efficiency.

But what's fascinating is that when you look at actual stable LAVES phases, the observed optimum ratio is closer to 1 .36.

Why the mismatch?

Why is the real -world optimum different from the perfect geometric one?

That difference is the proof that the bonding isn't purely a hard sphere.

The metallic characteristic, the role of those free electrons and the specific electron concentration, allows for a slight compression of the atoms.

It lets them optimize the packing at a slightly larger size ratio than pure geometry would predict.

So let's define the coordination environment.

What are the key polyhedra in LAVES phases?

The structure is defined by the extremely high coordination of the larger A atoms.

They are surrounded by 16 neighbors, that's the CN16 -Priusov polyhedra.

The smaller B atoms are 12 -fold coordinated, and they typically form those highly symmetrical CN12 icosahedra.

So the CN16 polyhedra build the architecture, and stability comes from filling their voids with the most efficient possible CN12 clusters.

That's it.

And section 18 .3 .3 .1 explains that the three main LAVES phase types, C15, C14, and C36, are distinguished solely by the way these CN16 -Priusov polyhedra are stacked on top of each other.

Okay.

Let's break down the three main types, starting with the cubic one.

C2MG, C15 cubic.

Right.

The C15 phase is often the most stable at high temperatures.

In this arrangement, the large A atoms form a skeleton that corresponds exactly to a diamond cubic lattice.

Like the carbon atoms in diamond.

Exactly like that.

And the smaller B atoms then fill the large voids created by this diamond The stacking sequence of the major layers is written as A -star, B -star, C -star, which is a three -layer repetition of complex Kagome.

Okay.

Next up, the most common type, the hexagonal LAVES phase.

MgZn2, C14 hexagonal.

The C14 is hexagonal, and it's defined by a two -layer stacking sequence, A -star, B -star.

So instead of the A atoms forming a diamond cubic lattice, in C14, they form a skeleton equivalent to a Wurtzite lattice.

It's a subtle distinction, but critical.

It is.

It's a slightly different layer orientation, which gives you a two -layer repeat instead of the three -layer repeat of C15.

And finally, the really long repeat unit, MgNe2, C36 hexagonal.

C36 is the least common.

It has this complex, long period, eight -layer stacking sequence.

You can really think of it as an intergrowth, a mix of the stacking features you find in C15 and C14.

The complexity just shows how tiny energetic differences can favor these massive unit cells to get the packing just right.

Now, before we move on to the notoriously difficult phases, we should touch on box 18 .4.

Self -assembled nanostructures.

This is a really interesting modern application of LAVES phases.

It really is.

It highlights the intrinsic energetic stability of the LAVES structure.

Recent studies on self -assembled nanoparticles, specifically things like lead selenide and lead sulfide quantum dots, have shown that they spontaneously adopt LAVES phase lattices, like the MgZn2, C14 structure.

That is fascinating.

So it implies that the LAVES geometry is so energetically favorable that atoms will organize themselves into this specific complex crystal structure at the nanoscale, even without a bulk crystal to copy.

Exactly.

It confirms that the size ratio constraint and the resulting high coordination are incredibly powerful, intrinsic structural drivers.

They can actually direct atomic self -assembly.

Okay, now we have to address the boogeyman of high -temperature metallurgy.

Section 18 .3 .4.

The sigma phase.

The sigma phase is infamous.

It's infamous because its formation, usually in high -temperature stainless steels and other alloys, is directly linked to catastrophic embrittlement.

It turns a ductile, tough alloy into something brittle that can shatter.

And its structural complexity just reflects its technological difficulty.

It's not just a high atom count.

It uses every tool in the Frank Casper toolkit.

It is a crystalline behemoth.

The sigma phase, with CREFI as the prototype, has a tetragonal unit cell with a remarkable 30 atoms.

And unlike A15, or the Louvre's phases, it uses the full range of Frank Casper polyhedra, CN12, CN14, CN15, and CN16.

And these 30 atoms are distributed across five distinct, non -equivalent crystallographic sites, M1 through M5.

How does that site complexity relate to the polyhedra?

Well, having five distinct sites means different atoms are experiencing vastly different environments.

Each one optimized locally.

For instance, the CN12 sites are often occupied by the smallest atoms to get that low energy icosahedral symmetry.

But then you need the larger CN15 and CN16 sites to bridge and link these icosahedral regions into the overall periodic structure.

So we're seeing the geometric consequence of that initial electronic drive.

If a simple structure fails to minimize the energy, nature selects a 30 -atom, low -symmetry tetragonal solution instead.

It's the ultimate compromise between local energy minimization and global periodicity.

And the complexity, believe it or not, only increases when we look at section 18 .3 .5, the mu, p, and r phases.

Okay, let's just briefly characterize these highly complicated cousins of the sigma phase.

These are found mostly in systems with heavy transition metals.

The mu phase, W6FS7, is rhombohedral and has 13 atoms.

The p phase has 32 atoms per unit cell.

But the r phase is maybe the most structurally challenging of all.

It has a massive 53 atoms in its rhombohedral unit cell.

53 atoms.

In a single unit cell.

Each one potentially in a different bonding environment.

That must make determining their crystal structure an incredibly difficult task.

It is.

These structures exist purely because the combination of valence electrons and size ratios in these specific alloys demands a coordination and density that only a structure of this immense complexity, using a vast mix of distorted Frank -Casper polyhedra, can possibly provide.

They are the strongest evidence that the drive for efficient tetrahedral packing is paramount, even if it means abandoning all notions of simple structural design.

We've gone from 8 atom cells to 53 atom cells.

And this exponential increase in complexity brings us to our final major topic, which connects these complex crystals to the non -periodic world, section 18 .4, quasicrystal approximants.

Quasicrystals, discovered in the 80s, possess long -range order but no translational periodicity.

They maximize the use of icosahedral symmetry, but they do it in a non -repeating, non -periodic way.

So an approximate is a crystal.

It still has translational symmetry, but its local atomic arrangement is designed to mimic the high degree of icosahedral symmetry you'd find in a true quasicrystal.

That's precisely right.

They are periodic crystalline structures whose massive unit cells are just the most efficient periodic way to tile space, using that local icosahedral geometry that the alloy system energetically demands.

And you often find these approximants right next to true quasicrystal compositions on a phase diagram.

Let's look at the classic examples in section 18 .4 .1, MG32, ALZEN49, and ALFA -ALN and PSI.

The MG32, ALZEN49 structure is just a spectacular example.

It's cubic, and its unit cell contains a staggering 162 atoms.

162.

Yes.

Its fundamental lattice is FCC, but every lattice point isn't just one atom.

It's an enormous cluster of atoms, a massive superunit.

To make any sense of a 162 atom structure, you can't analyze individual atoms.

You have to use the shell approach, which takes us to section 18 .4 .2, Shell Models in Pelluhedra.

We analyze these huge structures by defining nested concentric shells of atoms surrounding a central point.

The whole structure is built by repeating these shell models.

And the first major cluster is the Bergmann polyhedron.

What's its composition?

The Bergmann cluster is a 45 atom unit.

You can visualize it like this.

A central atom surrounded by a first shell of 12 atoms forming that highly stable icosahedron, and then an outer shell of 32 more atoms.

So that's 1 plus 12 plus 32 for 45 atoms.

And the 162 atom MGALZEN structure contains two of these 45 atom Bergmann clusters embedded within its unit cell.

And the second major cluster, which is key to understanding the alpha -al -NA structure, is the Mackay icosahedron.

The Mackay icosahedron is a bit larger.

It's a 54 atom cluster.

It starts the same way.

A central atom and an inner icosahedral shell.

But the outer shell, which has 41 atoms, is formed by decorating the faces of that inner icosahedron with transition metal atoms, like manganese.

This specific decoration introduces the necessary elements of the quasicrystal structure into a periodic framework.

So if we look at the alpha -almond NC structure, its primitive cubic cell is essentially built by linking these massive 54 atom Mackay icosahedra together, sharing vertices or edges.

And what's so incredible here is that the entire exercise of building these 162 atom, or 54 atom units, a level of structural complexity that challenges even modern computing,

is just a highly complex geometric solution to satisfy the most basic energetic requirement.

Minimize the free electron energy by maximizing the number of local 12 -fold icosahedral coordination shells.

This has been an incredibly detailed journey through the high -density world of intermetallic alloys.

Let's bring it all back together with a summary.

We began by establishing the electronic foundation, that the periodicity of the lattice imposes constraints on electron energies, driving the structure to minimize its total energy by creating features in the density of states near the Fermi level.

Right, and this energy minimization dictates the geometry.

It forces a system to adopt topological close packing, or TCP, which is defined by the exclusive use of low -energy tetrahedral interstices.

And we identified the key geometric constraint that comes from that.

All Frank -Casper polyhedra CN12, 14, 15, and 16 must have exactly 12 five -fold coordinated vertices.

That's a direct consequence of the triangulated geometry.

From there, we explored the structural prototypes, the A15 phase, where those CN12 chains create sharp peaks in the density of states crucial for superconductivity, the highly predictable Laves phases, defined by the radius ratio and stacking of CN16 polyhedra, and then the frightening complexity of the sigma, mu, p, and r phases.

Finally, we connected all this geometric complexity to the non -periodic world through quasicrystal approximants.

These are giant crystal structures like MG32 -AL, ZEN49, whose entire existence is dedicated to periodically repeating these massive icosahedral clusters like the 45 -atom Bergmann and 54 -atom Mackay polyhedra.

So if we were to zoom out and look at everything we covered, what's the single most important what for a material scientist listening to this?

The most important realization, I think, is that when you're dealing with these metallic alloys, structure selection is often decoupled from simple lattice mathematics.

The local preference for the icosahedron driven by electronic stability is the dominant structural force.

The lattice then struggles to find the lowest symmetry, most complex way it can to repeat that low -energy local unit periodically.

If you can predict the preferred local geometry, you are halfway to predicting the stable crystal structure, no matter how complex the unit cell ends up being.

That structural sacrifice, abandoning simplicity for complexity just to achieve local stability, that's really compelling.

Which leaves us with one final provocative thought for you to carry forward.

If the entire architecture structures with 30, 50, or even 162 atoms is driven purely by the need to tile space using these low -energy icosahedral clusters, does every element pairing and every valence electron concentration generate its own unique, highly complex approximate structure, many of which are still out there, waiting to be discovered simply because of the sheer difficulty of modeling them?

A great question to end on.

Thank you for joining us on this deep dive into metallic structures and the fascinating world of topological close packing.

We hope this step -by -step guidance has clarified these dense and intricate phases.

It was a pleasure to explore the geometry of high -density stability.

Until next time, keep digging into the details.

And a warm thank you from the Last Minute Lecture Team.