1.2 Reciprocal lattice and Brillouin zones

Reciprocal Lattice

The reciprocal lattice gives you a way to describe a crystal's periodic structure in momentum space (also called k-space) rather than in real space. It's the Fourier transform of the real-space lattice, and it turns out to be the natural setting for understanding diffraction patterns, electronic band structures, and phonon dispersion relations.

Definition of Reciprocal Lattice

Every set of lattice planes in real space corresponds to a point in the reciprocal lattice. The reciprocal lattice vectors $\vec{b_1}$, $\vec{b_2}$, and $\vec{b_3}$ are constructed from the real-space lattice vectors $\vec{a_1}$, $\vec{a_2}$, and $\vec{a_3}$ using these relations:

• $\vec{b_1} = 2\pi \frac{\vec{a_2} \times \vec{a_3}}{\vec{a_1} \cdot (\vec{a_2} \times \vec{a_3})}$
• $\vec{b_2} = 2\pi \frac{\vec{a_3} \times \vec{a_1}}{\vec{a_1} \cdot (\vec{a_2} \times \vec{a_3})}$
• $\vec{b_3} = 2\pi \frac{\vec{a_1} \times \vec{a_2}}{\vec{a_1} \cdot (\vec{a_2} \times \vec{a_3})}$

The denominator $\vec{a_1} \cdot (\vec{a_2} \times \vec{a_3})$ is the volume of the real-space unit cell, $V_{\text{cell}}$. Notice that each reciprocal vector is built from the cross product of the other two real-space vectors.

Reciprocal Lattice Vectors

The defining property of these vectors is the orthogonality condition:

$\vec{a_i} \cdot \vec{b_j} = 2\pi \delta_{ij}$

Here $\delta_{ij}$ is the Kronecker delta (equals 1 when $i = j$, and 0 otherwise). This means $\vec{b_1}$ is perpendicular to both $\vec{a_2}$ and $\vec{a_3}$, and so on.

Two other useful facts:

• The magnitude of a reciprocal lattice vector is inversely proportional to the spacing between the corresponding lattice planes in real space: $|\vec{G}_{hkl}| = 2\pi / d_{hkl}$.
• The direction of a reciprocal lattice vector $\vec{G}_{hkl}$ is the normal to the $(hkl)$ planes.

Relationship to Real-Space Lattice

Because the reciprocal lattice is the Fourier transform of the real-space lattice, large spacings in real space map to small spacings in reciprocal space, and vice versa. The volume of the reciprocal-space unit cell is:

$V_{\text{recip}} = \frac{(2\pi)^3}{V_{\text{cell}}}$

This inverse relationship is why materials with large unit cells produce closely spaced diffraction spots, and materials with small unit cells produce widely spaced spots.

Applications in Crystallography

X-ray, electron, and neutron diffraction experiments produce patterns that are direct maps of the reciprocal lattice. Specifically:

• The positions of diffraction peaks tell you the lattice parameters and symmetry of the crystal.
• The intensities of those peaks depend on the structure factor, which encodes atomic positions and scattering strengths within the unit cell.
• Analyzing diffraction data in reciprocal space lets you reconstruct the electron density distribution and bonding characteristics of the material.

Brillouin Zones

Brillouin zones partition reciprocal space into regions that capture all the distinct wavevectors needed to describe electrons and phonons in a crystal. They're the reciprocal-space analog of the Wigner-Seitz cell in real space.

Definition of Brillouin Zones

A Brillouin zone is the Wigner-Seitz primitive cell of the reciprocal lattice. You construct it by drawing perpendicular bisector planes (called Bragg planes) between a reciprocal lattice point and all its neighbors. The enclosed region closest to that lattice point is the Brillouin zone.

The Bragg planes have direct physical meaning: they're the wavevector values where Bragg diffraction occurs, meaning incoming waves are strongly scattered by the crystal periodicity.

First Brillouin Zone

The first Brillouin zone (1st BZ) is centered at the origin of reciprocal space (the $\Gamma$ point). It contains every unique wavevector $\vec{k}$ you need to fully describe the electronic and phononic properties of the crystal. Any wavevector outside the 1st BZ can be mapped back into it by subtracting an appropriate reciprocal lattice vector $\vec{G}$.

For common semiconductor structures:

• FCC real-space lattice (Si, GaAs): the reciprocal lattice is BCC, so the 1st BZ is a truncated octahedron. High-symmetry points include $\Gamma$, $X$, $L$, $K$, and $W$.
• BCC real-space lattice: the reciprocal lattice is FCC, so the 1st BZ is a rhombic dodecahedron.

Band structure diagrams plot energy vs. $\vec{k}$ along paths connecting these high-symmetry points.

Higher-Order Brillouin Zones

Higher-order Brillouin zones are the successive shells of reciprocal space beyond the first zone. The $n$th Brillouin zone is the set of points reached from the origin by crossing exactly $n - 1$ Bragg planes.

Each higher zone has the same volume as the first zone, and when "folded back" by reciprocal lattice translations, it maps exactly onto the 1st BZ. This folding concept becomes especially important in superlattices and nanostructures, where an artificially enlarged real-space period creates a smaller Brillouin zone.

Brillouin Zone Boundaries

At the zone boundaries, the wavevector satisfies the Bragg condition, and the electron wavefunction forms standing waves rather than traveling waves. This has major physical consequences:

• Energy gaps open at zone boundaries because the two standing-wave solutions (one with charge density peaked on the ions, one peaked between them) have different energies.
• Van Hove singularities in the density of states often occur at or near zone boundaries, where the band dispersion flattens and $\nabla_k E(\vec{k}) \to 0$.
• These features directly shape the electronic and optical behavior of the semiconductor.

Relationship to Reciprocal Lattice

The size and shape of the Brillouin zones are entirely determined by the reciprocal lattice, which in turn depends on the real-space crystal symmetry. Higher crystal symmetry produces Brillouin zones with more symmetry operations, which reduces the number of independent $\vec{k}$-points you need to compute (the irreducible Brillouin zone). This symmetry reduction is what makes band structure calculations computationally tractable.

Wave Propagation in Periodic Structures

Electrons in a crystal don't behave like free particles. The periodic potential of the ion cores fundamentally changes how wavefunctions look and how energy depends on momentum. Bloch's theorem is the key result that makes this tractable.

Bloch Theorem

Bloch's theorem states that the eigenstates of an electron in a periodic potential take the form:

$\psi_{n\vec{k}}(\vec{r}) = e^{i\vec{k}\cdot\vec{r}} \, u_{n\vec{k}}(\vec{r})$

where $e^{i\vec{k}\cdot\vec{r}}$ is a plane wave and $u_{n\vec{k}}(\vec{r})$ is a function with the same periodicity as the lattice. The index $n$ labels the band, and $\vec{k}$ is the crystal momentum (restricted to the 1st BZ).

The physical picture: the electron still has wave-like character (the plane wave part), but its amplitude is modulated by the periodic environment (the $u$ part). This is why you can label every electronic state by just two things: which band it's in and what its wavevector is.

Bloch Wave Functions

Bloch wavefunctions satisfy a useful periodicity condition:

$\psi_{n\vec{k}}(\vec{r} + \vec{R}) = e^{i\vec{k}\cdot\vec{R}} \, \psi_{n\vec{k}}(\vec{r})$

where $\vec{R}$ is any real-space lattice vector. The wavefunction doesn't repeat exactly from cell to cell; instead, it picks up a phase factor $e^{i\vec{k}\cdot\vec{R}}$. Only at $\vec{k} = 0$ (the $\Gamma$ point) does the wavefunction have the full periodicity of the lattice.

Dispersion Relations

The dispersion relation $E_n(\vec{k})$ gives the energy of band $n$ as a function of wavevector. You get it by solving the Schrödinger equation with the periodic potential for each $\vec{k}$ in the 1st BZ.

Several important quantities come directly from the dispersion relation:

• Effective mass: $\frac{1}{m^*} = \frac{1}{\hbar^2} \frac{d^2 E}{dk^2}$, determined by the curvature of the band near an extremum.
• Group velocity: $v_g = \frac{1}{\hbar} \nabla_k E(\vec{k})$, which tells you how fast a wavepacket moves.
• Density of states: derived from how many $\vec{k}$-states fall within a given energy range, which depends on the shape of $E(\vec{k})$.

Band Structure of Semiconductors

The band structure is the collection of all dispersion relations $E_n(\vec{k})$ plotted along high-symmetry directions in the 1st BZ. For semiconductors, the critical features are:

• The valence band (highest occupied band at $T = 0$) and the conduction band (lowest unoccupied band), separated by the band gap $E_g$.
• Whether the gap is direct (valence band maximum and conduction band minimum at the same $\vec{k}$, as in GaAs) or indirect (at different $\vec{k}$-points, as in Si). This distinction controls whether optical transitions are efficient.
• The curvature at the band edges, which determines the effective masses of electrons and holes.

Significance in Semiconductor Physics

Electronic Properties of Semiconductors

Carrier concentration, mobility, and conductivity all trace back to the band structure. The reciprocal lattice and Brillouin zone framework lets you calculate how these properties change under:

• Strain: shifts band edges and modifies effective masses (used deliberately in strained-Si technology).
• Doping: moves the Fermi level within the gap, changing carrier populations.
• Quantum confinement: restricts allowed $\vec{k}$-values in one or more dimensions, creating subbands.

The effective mass $m^*$ is particularly important for device modeling. It's not a single number in general; near a band extremum, it's a tensor reflecting the anisotropy of $E(\vec{k})$. For example, silicon's conduction band has different longitudinal and transverse effective masses ($m_l^* \approx 0.98 \, m_0$, $m_t^* \approx 0.19 \, m_0$).

Optical Properties of Semiconductors

Optical absorption and emission depend on transitions between bands, governed by:

• Energy conservation: the photon energy must match the energy difference between initial and final states.
• Momentum conservation: since photons carry negligible crystal momentum, direct transitions ($\Delta \vec{k} \approx 0$) dominate. Indirect transitions require a phonon to supply the momentum difference.
• Selection rules: determined by the symmetry of the Bloch states at the relevant $\vec{k}$-points, encoded in the optical matrix elements.

These principles are why GaAs (direct gap) is excellent for LEDs and lasers, while Si (indirect gap) is not efficient for light emission but works well for absorption in solar cells.

Fermi Surfaces and Energy Bands

The Fermi surface is the constant-energy surface in $\vec{k}$-space at the Fermi energy $E_F$. It separates occupied from unoccupied states at $T = 0$.

• In metals, the Fermi surface has complex shapes that determine electrical conductivity, the Hall coefficient, and magnetoresistance.
• In intrinsic semiconductors at $T = 0$, the Fermi level sits in the gap, so there's no Fermi surface. But in heavily doped semiconductors, the Fermi level can enter a band, and Fermi surface analysis becomes relevant.

Effective Mass of Charge Carriers

The effective mass captures how an electron or hole accelerates in response to external fields. Near a band extremum at $\vec{k_0}$, you can expand the energy as:

$E(\vec{k}) \approx E(\vec{k_0}) + \frac{\hbar^2}{2} \sum_{ij} \frac{(k_i - k_{0i})(k_j - k_{0j})}{m^*_{ij}}$

The effective mass tensor $m^*_{ij}$ enters into carrier transport equations, density of states calculations, and optical transition rates. A small effective mass means high mobility (carriers accelerate easily), which is why materials like InSb ($m^* \approx 0.014 \, m_0$) have extremely high electron mobilities.

Experimental Techniques

X-ray Diffraction and Reciprocal Space

X-ray diffraction (XRD) is the most direct probe of the reciprocal lattice. When X-rays scatter off a crystal, constructive interference occurs when the Laue condition is satisfied:

$\vec{k'} - \vec{k} = \vec{G}$

where $\vec{k}$ and $\vec{k'}$ are the incident and scattered wavevectors, and $\vec{G}$ is a reciprocal lattice vector. This is equivalent to Bragg's law $2d\sin\theta = n\lambda$.

The diffraction pattern is a map of the reciprocal lattice. Peak positions give lattice parameters and symmetry; peak intensities encode the structure factor and atomic positions. High-resolution XRD is routinely used to measure strain and composition in semiconductor heterostructures.

Electron Diffraction and Brillouin Zones

Electron diffraction techniques provide complementary structural information:

• LEED (low-energy electron diffraction): probes surface structure and reconstruction. The diffraction pattern reflects the 2D surface reciprocal lattice.
• RHEED (reflection high-energy electron diffraction): monitors surface structure in real time during epitaxial growth (e.g., MBE). Oscillations in RHEED intensity track layer-by-layer growth.

Both techniques produce patterns directly related to the reciprocal lattice and Brillouin zones of the surface.

Angle-Resolved Photoemission Spectroscopy (ARPES)

ARPES directly maps the occupied electronic band structure $E(\vec{k})$. The technique works by:

1. Shining UV or soft X-ray photons onto the sample surface to eject electrons via the photoelectric effect.
2. Measuring the kinetic energy and emission angle of each photoelectron.
3. Converting the emission angle to the in-plane crystal momentum $\vec{k}_\parallel$ using $k_\parallel = \frac{\sqrt{2mE_{\text{kin}}}}{\hbar} \sin\theta$.

The result is a direct image of the band dispersion and Fermi surface in reciprocal space. ARPES has been essential for studying surface states, topological insulators, and correlated electron systems.

Scanning Tunneling Microscopy (STM)

STM images surfaces with atomic resolution by measuring the quantum tunneling current between a sharp tip and the sample. While STM operates in real space, its data connect to reciprocal-space concepts:

• The tunneling current at a given bias voltage is proportional to the local density of states (LDOS) at that energy.
• Fourier transforms of STM images reveal the wavevectors of surface electronic states, providing information about the Brillouin zone structure.
• STM can map out the spatial variation of the band gap and identify surface defects and reconstructions.

Applications in Device Modeling

Brillouin Zone Folding in Nanostructures

When you create a nanostructure with a larger period than the bulk crystal (a superlattice, for example), the Brillouin zone shrinks in proportion. Bulk bands get "folded" into this smaller zone, creating subbands with modified dispersion.

This folding has real consequences: it can turn an indirect-gap material into one that behaves more like a direct-gap material (the basis for some Si/Ge superlattice designs), and it creates quantized energy levels in quantum wells, wires, and dots.

Superlattices and Mini-Bands

A superlattice is a periodic stack of alternating semiconductor layers (e.g., GaAs/AlGaAs with a period of ~5-20 nm). The new periodicity creates:

• Mini-bands: narrow allowed energy bands formed by the coupling of quantum well states across thin barriers.
• Mini-gaps: energy gaps between mini-bands, tunable by adjusting layer thicknesses and compositions.

The width of the mini-bands depends on the barrier thickness and height. Thin barriers give wide mini-bands (strong coupling); thick barriers give narrow mini-bands (weak coupling). These structures are the basis for quantum cascade lasers, infrared photodetectors, and Bloch oscillator concepts.

Phonon Dispersion and Thermal Properties

Phonon dispersion relations $\omega(\vec{k})$ are calculated by solving the dynamical matrix eigenvalue problem across the Brillouin zone. From the full phonon dispersion, you can derive:

• Specific heat: from the phonon density of states (Debye and Einstein models are approximations to the full dispersion).
• Thermal conductivity: depends on phonon group velocities, scattering rates, and mean free paths.
• Thermal expansion: related to anharmonic terms in the interatomic potential.

Accurate phonon modeling is critical for thermal management in high-power devices, where heat dissipation limits performance.

Reciprocal Space in Device Simulations

Modern semiconductor device simulations rely heavily on reciprocal-space methods to compute electronic structure:

• k·p method: expands the band structure around a high-symmetry point using perturbation theory. Fast and accurate near band edges; widely used for III-V heterostructures.
• Tight-binding method: builds the Hamiltonian from atomic orbitals. Good for capturing the full Brillouin zone and for modeling nanostructures with atomistic detail.
• Pseudopotential method: replaces the true ionic potential with a smoother effective potential. Used for accurate bulk band structure calculations.

These methods compute band structures, carrier densities, and current-voltage characteristics, enabling the design and optimization of transistors, solar cells, LEDs, and lasers from first principles.