Graphite, the material used in pencil lead, has a structure consisting of multiple stacked layers of carbon atom sheets. Graphene is a single, atom-thick layer extracted from this material. Within this extremely thin sheet obtained from such a familiar substance, low-energy electrons behave as if they are massless particles. Why does this property emerge from a single sheet of carbon?
The wave-like behavior of electrons is strongly influenced by how the atoms are bonded together. In graphene, each carbon atom bonds with three neighboring carbon atoms, forming a honeycomb structure. Expressing this atomic arrangement and its periodic regularity mathematically is the first step toward calculating electron waves and energies. Let us begin by unraveling the geometric structure of this honeycomb lattice.
The standard method for investigating the periodicity of a crystal is to translate the entire pattern and check whether it perfectly overlaps with the original. If the pattern overlaps completely, it means an identical atomic arrangement exists at the translated position. First, let us choose a carbon atom on the honeycomb lattice and translate the entire lattice by the carbon–carbon distance $a$ toward a neighboring atom. While half of the atoms overlap with other atomic positions, the remaining half move into the empty centers of the hexagons. Why do they not all overlap at once? By examining the three bonds extending from each atom, we find two types of atoms: those with one bond pointing upward and those with one bond pointing downward. Attempting to overlay an atom with an upward-pointing bond onto an atom with a downward-pointing bond fails to reproduce the overall pattern because the bond orientations are reversed. What happens if we translate an atom with an upward-pointing bond toward another atom that also has an upward-pointing bond? This time, the entire honeycomb pattern overlaps perfectly. Extracting only these upward-pointing atoms reveals a network formed by seamlessly tiling equilateral triangles. This is a triangular lattice. Similarly, extracting only the downward-pointing atoms yields an identical triangular lattice shifted by a small translation. Here, let us refer to the upward-pointing atoms as A-sites and the downward-pointing atoms as B-sites. Although both are carbon atoms, their surrounding environments, such as bond directions, differ. Thus, graphene is not a simple single lattice, but rather a structure where a pair of two atoms, A and B, is placed at each point of a triangular lattice.
The entire honeycomb lattice can be constructed by repeating a pair consisting of an A-site and the B-site directly above it. To visualize this, imagine a stamp engraved with this pair of A and B sites. Every time the stamp is pressed, one pair of A and B sites is added to the paper. Starting from an A-site at the origin, let $\mathbf{a}_1$ be the vector pointing toward the upper-right A-site, and $\mathbf{a}_2$ be the vector pointing toward the upper-left A-site. We construct a rhombic stamp using these two vectors as adjacent sides. Aligning the A-mark on the stamp with each A-site and pressing it repeatedly without rotation seamlessly tiles the rhombuses, completing the entire honeycomb lattice. The positions where the stamp is applied can be specified by the number of steps taken along each of the two vectors. When moving $m$ times along $\mathbf{a}_1$ and $n$ times along $\mathbf{a}_2$ from the origin, the position $\mathbf{R}$ of the A-mark is expressed as follows:
$$\mathbf{R}=m\,\mathbf{a}_1+n\,\mathbf{a}_2 \tag{1}$$
Here, $m$ and $n$ are integers, with negative values representing movement in the opposite direction of the vectors. For example, setting $m=1$ and $n=-1$ yields $\mathbf{a}_1 - \mathbf{a}_2$, reaching the neighboring A-site located directly horizontally. Combining these two vectors alone allows us to specify the positions of all A-sites. In physics, the set of all stamp positions $\mathbf{R}$ is called a Bravais lattice, and the reference vectors $\mathbf{a}_1$ and $\mathbf{a}_2$ are called primitive translation vectors. Translating the entire crystal by a vector $\mathbf{R}$ yields a structure identical to the original arrangement; this operation is called a lattice translation. Furthermore, a group of atoms placed at each lattice point, such as the pair of A and B sites carved on the stamp, is called a basis. The smallest repeating unit area corresponding to a single stamp press is called a unit cell. In graphene, each unit cell contains one A-site and one B-site, totaling two carbon atoms. Now, let us calculate the primitive vectors $\mathbf{a}_1$ and $\mathbf{a}_2$, which form two sides of the stamp, using the carbon–carbon distance $a$.
The primitive vectors $\mathbf{a}_1$ and $\mathbf{a}_2$ representing the two sides of the stamp point from the A-site at the origin to the upper-right and upper-left A-sites, respectively. Tracing the diagram reveals that one can reach the next A-site from an A-site by following two consecutive bonds via the B-site directly above. In other words, combining the two bond vectors for the path A $\to$ B $\to$ A yields the primitive vectors. Let the $x$-axis point to the right and the $y$-axis point upward, expressing vectors as pairs of $x$- and $y$-components. The measured carbon–carbon distance in graphene is $a = 0.142\text{ nm}$. Let $\boldsymbol{\delta}_1, \boldsymbol{\delta}_2,$ and $\boldsymbol{\delta}_3$ be the vectors pointing from an A-site to its three adjacent B-sites. Their lengths are all $a$. From the geometric properties of a regular hexagon, these three vectors are oriented at $120^\circ$ to one another. Choosing the first vector $\boldsymbol{\delta}_1$ in the upward $+y$ direction gives:
$$\boldsymbol{\delta}_1=a(0,1) \tag{2}$$
Rotating this vector clockwise by $120^\circ$ gives $\boldsymbol{\delta}_2$, with components $a(\sin120^\circ, \cos120^\circ)$. Rotating counterclockwise yields $\boldsymbol{\delta}_3$, obtained simply by flipping the sign of the $x$-component. Substituting $\sin120^\circ=\frac{\sqrt{3}}{2}$ and $\cos120^\circ=-\frac{1}{2}$ gives the remaining two vectors:
$$\begin{aligned} \boldsymbol{\delta}_2&=a\left(\frac{\sqrt3}{2},-\frac12\right)\\ \boldsymbol{\delta}_3&=a\left(-\frac{\sqrt3}{2},-\frac12\right) \end{aligned} \tag{3}$$
Since $\boldsymbol{\delta}_j$ points from A to B, the vector pointing back from B to A is $-\boldsymbol{\delta}_j$. Moving from the origin A-site to the B-site directly above by $\boldsymbol{\delta}_1$, and then to the upper-right A-site by $-\boldsymbol{\delta}_3$, yields the primitive vector $\mathbf{a}_1$. Similarly, changing the second step to $-\boldsymbol{\delta}_2$ yields the primitive vector $\mathbf{a}_2$:
$$\begin{aligned} \mathbf{a}_1&=\boldsymbol{\delta}_1-\boldsymbol{\delta}_3 =a\left(\frac{\sqrt3}{2},\frac32\right)\\ \mathbf{a}_2&=\boldsymbol{\delta}_1-\boldsymbol{\delta}_2 =a\left(-\frac{\sqrt3}{2},\frac32\right) \end{aligned} \tag{4}$$
Using Equation (4), let us calculate the magnitude of the primitive vector $\mathbf{a}_1$:
This is the lattice constant of the triangular lattice, that is, the distance between neighboring lattice points. While the carbon–carbon bond length is $0.142\text{ nm}$, the lattice constant representing the minimum repeating period is slightly longer at $\sqrt{3} \times 0.142 \approx 0.246\text{ nm}$. Note that while $a$ denotes the carbon–carbon bond length in this explanation, specialized literature frequently uses $a$ to denote the lattice constant $\sqrt{3}a$. Care should be taken regarding which physical quantity $a$ represents when comparing equations across textbooks and research papers. The atomic arrangement remains completely unchanged under a translation by a lattice vector $\mathbf{R}=m\mathbf{a}_1+n\mathbf{a}_2$. How, then, does an electron wave propagating through this crystal change under such a translation?
Imagine placing two floats on the surface of water and sending a straight wave toward them. Depending on their locations, the two floats might rise together, or one might rise as the other falls. The quantity that describes the timing of oscillation within a single period is the phase, and the difference in phase between two points is called the phase difference. A wave whose wavefronts propagate as parallel straight lines is called a plane wave. The form of a plane wave at a given instant can be written as $\cos(\mathbf{k}\cdot\mathbf{r})$, or using complex notation as $e^{i\mathbf{k}\cdot\mathbf{r}}$. Here, $\mathbf{r}=(x,y)$ represents the position coordinates, and $\mathbf{k}$ is the wave vector indicating the direction of propagation, with its magnitude defined in terms of the wavelength $\lambda$ as $|\mathbf{k}| = \frac{2\pi}{\lambda}$. At a position $\mathbf{r}+\mathbf{R}$, shifted from $\mathbf{r}$ by a lattice translation vector $\mathbf{R}$, the phase changes from $\mathbf{k}\cdot\mathbf{r}$ to $\mathbf{k}\cdot(\mathbf{r}+\mathbf{R})$. The difference is $\mathbf{k}\cdot\mathbf{R}$. If this phase difference is an integer multiple of $2\pi$, the peaks and troughs of the wave align perfectly before and after the translation. If it is not an integer multiple, the wave amplitude shifts even though the atomic positions overlap. Expressing this phase shift as a complex exponential $e^{i\mathbf{k}\cdot\mathbf{R}}$ yields the phase factor. Even if phase differences differ by integer multiples of $2\pi$, Euler's formula ($e^{i 2\pi n} = 1$) ensures that the phase factors remain identical. Consequently, two waves with different wavelengths or directions can produce identical phase factors for all lattice translation vectors $\mathbf{R}$. This means they are indistinguishable on the lattice points. Under what conditions for the wave vector difference $\mathbf{G}=\mathbf{k}'-\mathbf{k}$ does this phenomenon occur? Writing out the phase factors for two waves gives:
$$e^{i\mathbf{k}'\cdot\mathbf{R}} =e^{i(\mathbf{k}+\mathbf{G})\cdot\mathbf{R}} =e^{i\mathbf{k}\cdot\mathbf{R}}e^{i\mathbf{G}\cdot\mathbf{R}} \tag{5}$$
For both factors to coincide across all lattice translations $\mathbf{R}$, the rightmost term $e^{i\mathbf{G}\cdot\mathbf{R}}$ must evaluate to $1$:
$$e^{i\mathbf{G}\cdot\mathbf{R}}=1 \quad\text{for every }\mathbf{R} \tag{6}$$
To visualize Condition (6), consider a wave $\cos(\mathbf{G}\cdot\mathbf{r})$ described by the wave vector difference $\mathbf{G}$. Condition (6) means that the wave peaks align precisely on every A-site lattice point. Let us search for special vectors $\mathbf{G}$ that align wave peaks across all A-sites by varying the wave orientation and wavelength. For instance, setting the wavelength to $\lambda=1.50a$ and aligning wavefronts along $\mathbf{a}_2$ causes a wave peak to intersect every row of A-sites. Let us label this wave vector difference pointing toward the upper-right as $\mathbf{b}_1$. For the wave $\cos(\mathbf{b}_1\cdot\mathbf{r})$, shifting the position by $\mathbf{a}_1$ moves to the adjacent peak, advancing the phase by one full period: $\mathbf{b}_1\cdot\mathbf{a}_1=2\pi$. Conversely, moving along $\mathbf{a}_2$ keeps the position on the same peak, yielding a phase difference of $\mathbf{b}_1\cdot\mathbf{a}_2=0$. Keeping the wavelength constant, let us switch the wave orientation along $\mathbf{a}_1$, calling this upper-left wave vector $\mathbf{b}_2$. Translating by $\mathbf{a}_1$ now remains on the same peak with zero phase shift, whereas translating by $\mathbf{a}_2$ reaches the adjacent peak with a shift of $2\pi$. Summarizing these properties using inner products yields:
$$\begin{aligned} \mathbf{b}_1\cdot\mathbf{a}_1&=2\pi, &\mathbf{b}_1\cdot\mathbf{a}_2&=0,\\ \mathbf{b}_2\cdot\mathbf{a}_1&=0, &\mathbf{b}_2\cdot\mathbf{a}_2&=2\pi \end{aligned} \tag{7}$$
What happens if we halve the distance between peaks for the wave $\mathbf{b}_1$? An additional peak appears between original peaks, but the condition that peaks align on all A-sites remains satisfied. Because wave vector magnitude is inversely proportional to wavelength, this wave vector is expressed as $\mathbf{G}=2\mathbf{b}_1$. Thus, infinitely many vectors $\mathbf{G}$ satisfy Condition (6). To analyze the spatial relationship among these special vectors $\mathbf{G}$, let us introduce a coordinate plane with axes $G_x$ and $G_y$. This plane is called reciprocal space (or wave vector space), where the tip of each vector $\mathbf{G}$ is plotted as a point. Vectors $\mathbf{b}_1$ and $2\mathbf{b}_1$ lie along the same direction from the origin, with $2\mathbf{b}_1$ twice as far. Points farther from the origin correspond to waves with smaller peak-to-peak spacing. Plotting all vectors $\mathbf{G}$ satisfying Condition (6) on this plane forms a periodic array of points called a reciprocal lattice, and vectors extending from the origin to these points are called reciprocal lattice vectors. The vectors $\mathbf{b}_1$ and $\mathbf{b}_2$ serve as primitive vectors for this reciprocal lattice. Each reciprocal lattice point represents a wave whose peaks align perfectly across all lattice points. Conversely, selecting a wave vector off these lattice points results in phase mismatches across different lattice sites. Just as real-space positions are represented as integer combinations of $\mathbf{a}_1$ and $\mathbf{a}_2$, reciprocal lattice points can be expressed using integers $m'$ and $n'$:
$$\mathbf{G}=m'\mathbf{b}_1+n'\mathbf{b}_2 \tag{8}$$
To determine the components of $\mathbf{b}_1=(x,y)$ for graphene, substitute the components of $\mathbf{a}_1$ and $\mathbf{a}_2$ from Equation (4) into the first row of Relation (7):
$$\begin{aligned} a\left(\frac{\sqrt3}{2}x+\frac32y\right)&=2\pi\\ a\left(-\frac{\sqrt3}{2}x+\frac32y\right)&=0 \end{aligned} \tag{9}$$
Adding the two equations eliminates $x$, giving $3ay=2\pi$, or $y=\frac{2\pi}{3a}$. Substituting $y$ back yields $x=\frac{2\pi}{\sqrt3a}$. Performing a similar calculation for $\mathbf{b}_2$ gives:
$$\begin{aligned} \mathbf{b}_1&=\left(\frac{2\pi}{\sqrt3a},\frac{2\pi}{3a}\right)\\ \mathbf{b}_2&=\left(-\frac{2\pi}{\sqrt3a},\frac{2\pi}{3a}\right) \end{aligned} \tag{10}$$
Both vectors have equal lengths: $|\mathbf{b}_1|=|\mathbf{b}_2|=\frac{4\pi}{3a}$. The reciprocal lattice formed by these two primitive vectors is also a triangular lattice, matching the geometry of real space. Two wave vectors $\mathbf{k}$ and $\mathbf{k}+\mathbf{G}$ separated by a reciprocal lattice vector $\mathbf{G}$ produce identical phase factors under any lattice translation. Consequently, they can be grouped together as equivalent states exhibiting identical phase shifts.
Although wave vector space extends infinitely, wave vectors separated by a reciprocal lattice vector $\mathbf{G}$ represent physically equivalent states with identical phase changes. To avoid redundant counting, we want to collect representative wave vectors by retaining only those closest to the origin from each equivalence group. For example, subtracting $\mathbf{b}_1$ from a point $\mathbf{k}$ near $\mathbf{b}_1$ translates it closer to the origin while preserving its physical properties. The boundary where the distance to the origin equals the distance to $\mathbf{b}_1$ is the perpendicular bisector of the line segment connecting the origin and $\mathbf{b}_1$. Constructing perpendicular bisectors for the six reciprocal lattice points surrounding the origin forms a regular hexagon centered at the origin. Wave vectors inside this hexagon are closer to the origin than to any other reciprocal lattice point. This region is called the first Brillouin zone. Any wave vector outside the zone can be mapped into this hexagon by subtracting an appropriate reciprocal lattice vector $\mathbf{G}$. Along the boundaries and vertices, distances to multiple reciprocal lattice points become equal. Consider the six vertices of the regular hexagon in particular. Label the rightmost vertex as the K point and translate it by reciprocal lattice vectors: adding $\mathbf{b}_2$ moves it to the upper-left vertex, and subtracting $\mathbf{b}_1$ from there moves it to the lower-left vertex. Thus, three alternating vertices belong to an equivalent group interconnected by reciprocal lattice vectors. Starting instead from the opposite vertex, the K' point, adding $\mathbf{b}_1$ leads to the upper-right vertex, and subtracting $\mathbf{b}_2$ leads to the lower-right vertex. These remaining three vertices form a second equivalent group. No integer combination of $\mathbf{b}_1$ and $\mathbf{b}_2$ connects K to K', meaning these two groups remain distinct. The former group is designated as K points and the latter as K' points. In summary, the six vertices of the first Brillouin zone divide into two fundamentally distinct groups: K and K'. The zone center $\Gamma$, the edge midpoint M, and the vertices K and K' serve as key reference points determining electron wave properties. Let us calculate the coordinates of these points: The edge midpoint M lies halfway between the origin $\Gamma$ and $\mathbf{b}_1$:
The rightmost vertex K is equidistant from the origin $\mathbf{0}$, $\mathbf{b}_1$, and $-\mathbf{b}_2$, corresponding to the centroid of the equilateral triangle formed by these three points:
$$\mathbf{K} =\frac{\mathbf{0}+\mathbf{b}_1-\mathbf{b}_2}{3} =\left(\frac{4\pi}{3\sqrt3a},0\right) \tag{11}$$
The opposite vertex K' is obtained simply by reversing the sign:
$$\mathbf{K}'=-\mathbf{K} =\left(-\frac{4\pi}{3\sqrt3a},0\right) \tag{12}$$
Expressing the translation vector from K to K' in terms of $\mathbf{b}_1$ and $\mathbf{b}_2$ gives:
$$\mathbf{K}'-\mathbf{K} =-2\mathbf{K} =-\frac23\mathbf{b}_1+\frac23\mathbf{b}_2 \tag{13}$$
Because the coefficients $-\frac{2}{3}$ and $\frac{2}{3}$ are not integers, this displacement is not a reciprocal lattice vector. This proves that the six vertices split cleanly into two distinct equivalence groups, K and K'. Using the carbon–carbon bond length $a=0.142\text{ nm} = 1.42\text{ \AA}$, the area of the first Brillouin zone is calculated as:
| Segment | Exact value | nm⁻¹ | Å⁻¹ |
|---|---|---|---|
| Γ–M | $\frac{2\pi}{3a}$ | 14.75 | 1.475 |
| Γ–K | $\frac{4\pi}{3\sqrt3\,a}$ | 17.03 | 1.703 |
| M–K | $\frac{2\pi}{3\sqrt3\,a}$ | 8.52 | 0.852 |
| Adjacent K–K′ (one edge) | $\frac{4\pi}{3\sqrt3\,a}$ | 17.03 | 1.703 |
| K–K (same family) | $\frac{4\pi}{3a}$ | 29.50 | 2.950 |
| Opposite K–K′ | $\frac{8\pi}{3\sqrt3\,a}$ | 34.06 | 3.406 |
| Nearest reciprocal lattice points | $\frac{4\pi}{3a}$ | 29.50 | 2.950 |
| Between the tips of b₁ and b₂ | $\frac{4\pi}{\sqrt3\,a}$ | 51.09 | 5.109 |
| Opposite edges | $\frac{4\pi}{3a}$ | 29.50 | 2.950 |
The hexagon has area $\dfrac{8\pi^2}{3\sqrt3\,a^2}$, approximately $753.6\,\mathrm{nm}^{-2}=7.536\,\text{Å}^{-2}$.
| Point / vector | $(k_x,k_y)$ |
|---|---|
| Γ | $(0,0)$ |
| M | $\left(\frac{\pi}{\sqrt3\,a},\frac{\pi}{3a}\right)$ |
| K · 0° | $\left(\frac{4\pi}{3\sqrt3\,a},0\right)$ |
| K · 120°, 240° | $\left(-\frac{2\pi}{3\sqrt3\,a},\pm\frac{2\pi}{3a}\right)$ |
| K′ · 180° | $\left(-\frac{4\pi}{3\sqrt3\,a},0\right)$ |
| K′ · 60°, 300° | $\left(\frac{2\pi}{3\sqrt3\,a},\pm\frac{2\pi}{3a}\right)$ |
| $\mathbf b_1,\mathbf b_2$ | $\left(\pm\frac{2\pi}{\sqrt3\,a},\frac{2\pi}{3a}\right)$ |
Distances and coordinates follow from the basis in .
06 How Wave Behavior Changes on the Honeycomb Lattice across $\Gamma$, M, and KLet us vary the wave vector from the center $\Gamma$ of reciprocal space to the edge midpoint M, and then to the vertex K. Moving away from the origin $\Gamma$ increases the magnitude of the wave vector, shortening the corresponding wavelength. Changing the direction of the wave vector alters the propagation direction of the wave in real space. Superimposing these waves onto the honeycomb lattice highlights phase differences among individual carbon atoms. The contrast in wave behavior between K and K' is particularly significant. Switching the wave vector from K to K' maintains the overall spatial wavelength of the fringe pattern while exactly reversing the rotation direction of the phase along atomic rows. This property plays a key role in dictating the unique electronic states of graphene.
The crystal structure clarifies the relationship between the periodicity of the honeycomb lattice and the phase of electron waves. What energy, then, do electrons propagating through this lattice possess? Unraveling the phenomenon mentioned at the beginning: where electrons behave as massless particles, requires calculating the energy band structure to determine how energy evolves with the wave vector. An approximation method that starts from the atomic orbitals of individual carbon atoms and accounts for electron hopping between adjacent sites is called the tight-binding approximation. Applying this approach to the honeycomb lattice yields the energy as a function of wave vector. Examining the shape of the band structure near the Brillouin zone vertices K and K' shows why electrons in graphene behave as massless particles.
