Wood's Notation

Describing the precise arrangement of atoms on a crystal surface presents a significant challenge. How can we concisely communicate the intricate patterns formed by overlayers or the self-reorganization of a surface? Wood's notation offers an elegant solution, providing a standardized language for the field of surface science. It bridges the gap between theoretical models and experimental observation, allowing scientists to catalog and comprehend the atomic world. This article addresses the need for a clear descriptive framework by exploring the principles and applications of this powerful notational system. In the following sections, you will first learn the fundamental rules and mechanisms of Wood's notation, including its variations and the more universal matrix formalism. Then, you will journey into its practical use, discovering how it helps decipher experimental data and unlocks our understanding of everything from semiconductor growth to the quantum phenomena of 2D materials.

Imagine you're trying to describe a new, intricate wallpaper pattern that has been placed over an existing one. How would you do it? You wouldn't describe the position of every single new flower or swirl. Instead, you'd look for the repeating unit of the new pattern and describe how it relates to the old one. Is it twice as big? Is it twisted by a certain angle? Is it shifted? This is precisely the game we play in surface science, but our "wallpaper" is a grid of atoms, and the new pattern is a layer of other atoms that have settled on top. To bring order to this atomic world, we need a language, a simple yet powerful notation. This is where Wood's notation comes in.

At its heart, Wood's notation is a beautifully simple system for describing how an overlayer—the new layer of atoms—arranges itself on a substrate—the underlying crystal surface. Let’s say our substrate has a primitive unit cell defined by two vectors, $\mathbf{a}_1$ and $\mathbf{a}_2$. The simplest way an overlayer can arrange itself is by creating a new unit cell that is just a scaled-up version of the substrate's cell. If its new vectors are $m$ times the length of $\mathbf{a}_1$ and $n$ times the length of $\mathbf{a}_2$, we write this as ​​$(m \times n)$​​. For example, a $(2 \times 2)$ overlayer has a unit cell that spans two substrate cells in one direction and two in the other.

Of course, the overlayer might not be perfectly aligned with the substrate; it could be rotated. We account for this by adding an angle, giving us ​​$(m \times n)\text{R}\theta$​​, where 'R' stands for "rotated" and $\theta$ is the angle of rotation relative to the substrate's axes. This single angle $\theta$ is a powerful descriptor. It's a real, physical rotation that we can see directly in real-space images from a Scanning Tunneling Microscope (STM), which maps out the atoms one by one. And because of the wonderful duality between real space and reciprocal space (the space of waves and periodicities), a rotation of the atomic lattice by $\theta$ causes the exact same rotation of the diffraction pattern seen in Low-Energy Electron Diffraction (LEED).

But there's one more piece to the puzzle. The rectangular box we've just described, with sides $m|\mathbf{a}_1|$ and $n|\mathbf{a}_2|$, might not be the smallest possible repeating unit of the overlayer.

• If this box is the smallest repeating unit, we call it a ​​primitive​​ cell and denote it with a p, as in $\text{p}(2 \times 2)$. It contains exactly one lattice point when all the corners are properly counted.
• Sometimes, however, the overlayer places an additional atom right in the center of our notational box. This forms a ​​centered​​ lattice, denoted with a c, as in $\text{c}(2 \times 2)$. In this case, our $(2 \times 2)$ box is a conventional cell, which is convenient to visualize but contains two lattice points (one at the corner, one in the center). The true primitive cell of a centered lattice is smaller and has exactly half the area of the conventional cell.

Here we come to a profound point about physics: nature has a single reality, but we may have several different, equally valid ways of describing it. The choice of description is a matter of convention and convenience, often chosen to make a particular feature, like symmetry, more obvious.

Consider a square substrate, like the (100) surface of a salt crystal. An overlayer forms that we could describe as $\text{c}(2 \times 2)$. Its conventional cell is a square, twice as large as the substrate's in each direction, with an extra lattice point in the middle. But what are the primitive vectors for this lattice? One valid choice of primitive vectors connects a corner atom to its nearest-neighbor center atoms. If you do the simple geometry, you'll find that these primitive vectors are equal in length and are at a $45^\circ$ angle to the substrate axes. In fact, their length is $\sqrt{2}$ times the substrate lattice constant. We have just discovered another name for our $\text{c}(2 \times 2)$ structure: $\text{p}(\sqrt{2} \times \sqrt{2})\text{R}45^\circ$!.

These are not two different structures; they are two different names for the exact same physical arrangement of atoms. So why have both? The $\text{c}(2 \times 2)$ notation is appealing because the square shape of its conventional cell instantly tells us the overlayer respects the square symmetry of the substrate. The $\text{p}(\sqrt{2} \times \sqrt{2})\text{R}45^\circ$ notation is more fundamental as it describes the smallest true repeating unit, but the rotation might make the overall symmetry less immediately obvious. This choice highlights a beautiful tension in science: the search for the most fundamental description versus the most intuitive one.

Wood's notation is elegant, but it's like a specialized dialect. It works wonderfully for overlayers that are simply scaled and rotated versions of the substrate cell. But what if the overlayer is sheared, like a rectangle that has been pushed into a parallelogram?

For this, we need a more universal language: ​​matrix notation​​. Any overlayer primitive vectors, $\mathbf{b}_1$ and $\mathbf{b}_2$, can be written as a linear combination of the substrate vectors $\mathbf{a}_1$ and $\mathbf{a}_2$:

This whole transformation can be neatly summarized in a single $2 \times 2$ matrix, $M$. For example, a simple $\text{p}(2 \times 2)$ structure corresponds to the matrix $M = \begin{pmatrix} 2 & 0 \\ 0 & 2 \end{pmatrix}$. A more complex structure, like the one generated by the matrix $M = \begin{pmatrix} 2 & 1 \\ -1 & 2 \end{pmatrix}$ on a square substrate, can be shown through a bit of geometry to be equivalent to the Wood's notation $\text{p}(\sqrt{5} \times \sqrt{5})\text{R}26.565^\circ$.

This matrix formalism gives us a wonderfully powerful and simple rule. If you want to know how the area of the overlayer unit cell, $A_{\mathrm{ov}}$, compares to the substrate's, $A_{\mathrm{sub}}$, you don't need to do any complicated geometry. You just calculate the determinant of the transformation matrix! The area ratio is simply:

This is a beautiful piece of pure mathematics—a property of linear transformations—that finds a direct, tangible meaning on the surface of a crystal. For a $\text{p}(m \times n)$ structure, $\det(M) = mn$. For a $\text{c}(m \times n)$ structure, the area ratio is $\frac{mn}{2}$ since the primitive cell has half the area of the centered conventional cell.

This descriptive language is not just for classification; it allows us to immediately deduce important physical properties.

One of the most crucial is the ​​surface coverage​​, $\Theta$. This is the simple ratio of the number of adsorbate atoms to the number of substrate surface atoms. If we assume the simplest case of one adsorbate atom per overlayer primitive cell (and one atom per substrate cell), the coverage is just the inverse of the area ratio:

So, a $\text{p}(2 \times 2)$ structure has a coverage of $\Theta = \frac{1}{4}$, while a $(\sqrt{3} \times \sqrt{3})\text{R}30^\circ$ structure on a hexagonal surface has a coverage of $\Theta = \frac{1}{3}$. The notation directly quantifies how densely the surface is packed.

Another fascinating consequence of the interplay between overlayer and substrate is the formation of ​​domains​​. A substrate can have certain symmetries—for instance, the hexagonal (111) surface of a gold crystal has a three-fold rotational symmetry and mirror planes. If an overlayer has a lower symmetry (e.g., its unit cell is a rectangle), it can exist in several distinct but energetically identical orientations on the surface. Each of these orientations is called a ​​domain​​. You can have ​​rotational domains​​, related by the rotation operations of the substrate, and ​​mirror domains​​ (or enantiomorphs), which are mirror images of each other. For a general overlayer on a substrate with $C_{3v}$ symmetry (like fcc(111)), one expects to find $6$ different domains, corresponding to the six symmetry operations of the substrate. This isn't a sign of a messy or imperfect surface; it is a beautiful and direct manifestation of the underlying crystal's symmetry.

So far, we have been living in a tidy world of ​​commensurate​​ structures—patterns that lock into a repeating superlattice with the substrate. This happens when the overlayer's basis vectors can be described by a transformation matrix $M$ with purely rational (and usually integer) entries.

But what happens if the natural lattice constant of the overlayer material is just slightly different from the substrate's? The atoms can't perfectly align everywhere. Instead, they fall in and out of registry with the substrate, creating a large-scale "beat" pattern. This is a ​​moiré pattern​​, the same effect you see when you overlap two window screens or fine-toothed combs. This is an ​​incommensurate​​ structure.

Here, the simple Wood's notation with small integers fails us. An irrational relationship cannot be captured by a finite ratio of integers. What do we do? We adapt! While there is no perfect, small-integer Wood's notation, we can find a ​​rational approximant​​—a very large supercell that almost perfectly matches the irrational ratio. For example, a moiré pattern formed by two slightly mismatched hexagonal lattices might be approximated by a giant $(53 \times 53)$ supercell. This tells us that over a distance of 53 substrate units, the overlayer has fit in 54 of its own units, bringing the two lattices almost back into alignment.

This journey from simple scaling, to rotations, to matrices, and finally to the vast, shimmering landscapes of moiré patterns shows the power of a good physical notation. It's a language that not only describes what we see but also reveals the deep principles of symmetry, geometry, and order that govern the atomic world.

Now that we have grappled with the principles of Wood's notation, you might be tempted to see it as a piece of formal bookkeeping, a dry cataloging system for the geometric arrangements of atoms. But to do so would be to miss the forest for the trees! This simple notation is, in fact, a powerful key that unlocks a vast and fascinating landscape of physics, chemistry, and materials science. It is the language used on the front lines of discovery, a shorthand that condenses entire symphonies of atomic behavior into a few characters. It is not just a label; it is a starting point for deep questions and a guidepost for technological innovation.

Let us embark on a journey through this world, to see how this notation comes to life in the laboratory and in the frontier of modern science.

Imagine you are a materials scientist, a modern-day alchemist trying to build a new semiconductor device, atom by atom. You are using a marvelous technique called Molecular Beam Epitaxy (MBE), laying down layers of Gallium and Arsenic to create a flawless crystal. How do you know if you are succeeding? You can’t just look; the atoms are too small. Instead, you use a technique like Reflection High-Energy Electron Diffraction (RHEED), which skims a beam of electrons off your growing surface. The way they bounce off and form a pattern on a screen tells you the atomic arrangement.

Suddenly, a pattern emerges that your knowledge of Wood's notation allows you to read instantly. You see a pattern you identify as a $(2\times4)$ reconstruction on your Gallium Arsenide (001) surface. What does this simple label tell you? It tells you that the surface atoms, left to their own devices, are not content to sit in the same boring grid as their brethren deep inside the crystal. To heal the "wounds" of being at the surface—the dangling, unsatisfied chemical bonds—they have ingeniously rearranged themselves. They have formed a new, larger, and more stable repeating unit cell, one that is twice as long as the ideal cell in one direction and four times as long in the other. This isn't a sign of a mistake; it's the signature of a high-quality, stable surface reaching its lowest energy state. The notation is your real-time feedback, a message directly from the atomic world.

What does this rearrangement actually look like? With another incredible tool, the Scanning Tunneling Microscope (STM), we can produce breathtaking images of these surfaces. We might find, for example, that a $(1\times2)$ reconstruction on the (110) surface of a metal isn't just a mathematical curiosity. The STM image could reveal a beautifully simple structural change: every other row of atoms is simply gone. The periodicity in one direction is doubled because a line of atoms is missing from each unit cell. The elegant notation finds its expression in a stark and simple physical reality. These are not just abstract scalings; they describe tangible, physical structures like terraces, trenches, and new atomic chains.

And nature, it turns out, can be far more creative. One of the most famous and complex surfaces in all of science is the silicon (111) surface. When cleaned in a vacuum, it performs a staggering feat of self-organization, a structure so complex it took decades to solve. Yet, this intricate atomic dance is known by the deceptively simple name: the Si(111)-$(7\times7)$ reconstruction. This simple label describes a huge unit cell containing dozens of atoms in a complex arrangement of "adatoms", "stacking faults", and deep "corner holes". It is a universe of complexity captured in a name, a testament to the power of a good notation.

So far, we have spoken of what the notation is, but how do we deduce it from an experiment in the first place? When we shine waves—be they electrons in Low-Energy Electron Diffraction (LEED) or X-rays—onto a periodic structure, they diffract, creating a pattern of spots. This pattern is a direct picture of the surface's ​​reciprocal lattice​​.

Now, "reciprocal space" sounds terribly abstract, but the idea is wonderfully intuitive. Think of a picket fence. If the pickets are spaced very closely, the frequency of pickets is high. If they are far apart, the frequency is low. Reciprocal space is simply the "space of frequencies". A diffraction pattern maps out the fundamental periodicities (the frequencies) of the atomic lattice. There is a simple and beautiful inverse relationship: a larger spacing in real space corresponds to a smaller spacing in reciprocal space.

If our surface reconstructs into a $(2\times1)$ superlattice, its real-space periodicity has doubled in one direction. Consequently, in the diffraction pattern, the distance between the spots in that direction must be halved. An experimentalist sees the new, closer-spaced spots appear in their LEED pattern and immediately knows the surface has doubled its periodicity in real space. They read the reciprocal space map and translate it back into the real-space language of Wood's notation.

The puzzle can get even more interesting. Suppose you see new diffraction spots appearing exactly halfway between all the original substrate spots on a square lattice. You might quickly guess this is a $(2\times2)$ reconstruction. But wait! There are two main possibilities: a simple, or "primitive," $\text{p}(2\times2)$ cell, and a $\text{c}(2\times2)$ cell, which has an extra atom in the center. Can we tell them apart?

Here, the wave nature of our probes provides a wonderful clue. In the $\text{c}(2\times2)$ structure, waves scattering from the corner atoms and the center atom can interfere destructively, systematically canceling out certain diffraction spots. A careful analysis shows that for a $\text{c}(2\times2)$ structure, the half-order spots that lie on the axes of the pattern vanish; only the diagonal ones survive. But if you observe all the half-order spots—both on the axes and on the diagonals—there are no systematic absences. You can therefore conclude with confidence that the structure must be the primitive $\text{p}(2\times2)$. This is a beautiful piece of physical deduction. The very absence of a signal becomes a powerful piece of information, allowing us to determine not just the size of the unit cell, but its internal symmetry.

Nature is not confined to simple rectangular grids. Many of the most important surfaces, like the basal plane of graphite (graphene's parent) or the close-packed faces of metals like copper, platinum, and gold, have a beautiful hexagonal symmetry. On these surfaces, adsorbates and reconstructions often form structures that are not only scaled but also rotated relative to the substrate.

One of the most famous and ubiquitous examples is the $(\sqrt{3}\times\sqrt{3})\text{R}30^\circ$ structure. The name itself looks more intimidating, but the physics it describes is elegant. It tells us that a new hexagonal superlattice has formed with a periodicity that is $\sqrt{3}$ times larger than the substrate's, and this entire new lattice is rotated by $30^\circ$. This specific arrangement is a favorite of nature because it represents a particularly stable and high-symmetry way for atoms or molecules to pack on a hexagonal grid.

Again, this notation is more than a label; it is a quantitative statement. Knowing the notation allows us to calculate the exact dimensions of the new unit cell. But more profoundly, it tells us about density. By a simple geometric argument rooted in the mathematics of lattices, we can deduce that the area of this new unit cell is exactly three times that of the substrate's primitive cell. If we assume that the substrate atoms themselves are the ones rearranging and that no atoms are lost or gained, this means that each reconstructed unit cell must contain exactly three of the original surface atoms. The notation encodes not just geometry, but a precise atom count.

We have arrived at a deeper, more philosophical question. When we see a new superstructure appear upon, say, exposing a clean metal surface to a gas like Carbon Monoxide (CO), what has actually happened? The new diffraction pattern, which we might label $(\sqrt{3}\times\sqrt{3})\text{R}30^\circ$, tells us the new symmetry. But it doesn't immediately tell us the cause.

Did the CO molecules simply land on the pristine, unchanged metal surface and arrange themselves into an ordered overlayer, like chess pieces on a board? Or did the interaction with the CO molecules cause the metal atoms themselves to tear up their old bonds and rebuild the top layer of their own crystal into a new $(\sqrt{3}\times\sqrt{3})\text{R}30^\circ$ structure, upon which the CO then sits?

This is a critical distinction, and surface scientists have developed ingenious methods to answer it. One powerful strategy is to perform a reversible experiment. You start with the clean surface and measure its diffraction pattern—specifically, the intensity of its spots as you vary the electron energy, creating a unique "fingerprint" called an $I(V)$ curve. Then, you introduce the CO and observe the formation of the new superstructure. Finally, you gently heat the crystal just enough to make the CO molecules fly off (desorb), and you measure the surface again.

If the surface returns identically to its original clean state—if the new superstructure spots vanish and the original fingerprint $I(V)$ curves are perfectly restored—then you have strong evidence that the substrate was merely a passive stage for the CO overlayer. But if the substrate does not return to its original state, you know that the CO has induced a true, irreversible reconstruction. Similar stories can be told using the atomic-scale precision of X-ray diffraction (SXRD) or the direct imaging power of STM to watch the surface heal—or fail to heal—after the adsorbates are removed. This is the scientific method in its purest form: observation, hypothesis, and a clever experiment to distinguish between possibilities.

The concepts we've explored have exploded into one of the most exciting frontiers of modern physics: the world of two-dimensional materials. What happens when you place one atomic sheet, like graphene, on top of another, like Ruthenium, whose natural atomic lattice spacing is similar but not quite identical?

The two grids, slightly mismatched, interfere with each other, creating a new, larger-scale "beat" pattern, much like the beat frequency you hear when two guitar strings are slightly out of tune. This beautiful, large-scale periodic pattern is called a ​​moiré superlattice​​. This is, in essence, an incommensurate superstructure.

While a simple Wood's notation with small integers may not strictly apply, the underlying physics is the same. And sometimes, the system finds it energetically favorable to have the top layer slightly stretch or compress to "lock in" to a true commensurate superstructure with the substrate. For graphene on Ruthenium, the lattice constant ratio is very close to the rational number 11/10. The system can choose to form a truly periodic structure where exactly 11 unit cells of graphene match perfectly with 10 unit cells of Ruthenium. This would be a genuine, commensurate superlattice, one that could be described, albeit with large numbers, in the spirit of Wood's notation. This subtle interplay of lattice mismatch, strain, and energy lies at the heart of the novel electronic and quantum phenomena seen in twisted bilayer graphene and other 2D heterostructures.

From the quality control of semiconductor growth to the fundamental detective work of surface chemistry and the engineering of new quantum materials, the simple language of Wood's notation is the common thread. It is a powerful reminder that in physics, the most profound ideas are often hidden in the most elegant and compact of descriptions. It is the poetry of the atomic plane.