In solid-state physics, the band gap (also known as the energy gap or forbidden gap) is the range of energy levels within a crystalline solid where no electron states exist, representing the difference between the top of the valence band—where electrons are bound to atoms—and the bottom of the conduction band, where electrons can move freely to conduct electricity.[1] This gap arises from the quantum mechanical interactions of electrons with the periodic lattice potential in the material.[2]The magnitude of the band gap fundamentally classifies materials into conductors, semiconductors, and insulators: in metals (conductors), the valence and conduction bands overlap, allowing free electron movement with no significant gap; semiconductors feature a narrow band gap typically ranging from 0.1 to 3 electron volts (eV), enabling thermal or optical excitation of electrons across the gap for controlled conductivity; and insulators possess a wide band gap exceeding about 3–5 eV, preventing electron excitation under normal conditions.[1][3] This distinction is key to understanding a material's electrical and thermal properties, as the band gap size correlates with the energy required to promote electrons, directly influencing behaviors like resistivity and charge carrier density.[4]Beyond classification, the band gap plays a pivotal role in technological applications, particularly in optoelectronics and photovoltaics, where it determines the wavelengths of light a material can absorb or emit—for instance, narrower gaps in materials like silicon (1.1 eV at room temperature) allow absorption of visible and near-infrared light for solar cells, while wider gaps in materials like diamond (5.5 eV) enable use in high-power electronics.[5] Experimental measurement of band gaps often involves techniques such as optical spectroscopy or photoemission, with values varying by temperature, pressure, and material composition due to lattice vibrations and structural changes.[6]
Basic Concepts
Definition and formation
In solid-state physics, the band gap, denoted as E_g, represents the minimum energy difference required to excite an electron from the highest occupied energy state in the valence band to the lowest unoccupied state in the conduction band.[7][8] This energy separation determines the electrical and optical properties of materials, distinguishing insulators, semiconductors, and conductors based on the size of E_g.[1] Mathematically, it is expressed asE_g = E_c - E_v,where E_c is the energy at the bottom of the conduction band and E_v is the energy at the top of the valence band.[8][9]The band gap emerges from the quantum mechanical treatment of electrons moving in the periodic potential of a crystal lattice. According to Bloch's theorem, the eigenfunctions of the Schrödinger equation in such a potential take the form of plane waves modulated by periodic functions, known as Bloch waves.[10][11] This periodicity causes the discrete atomic energy levels to broaden into continuous bands of allowed energies, separated by regions of forbidden energies where no electron states exist—forming the band gaps.[12][11] The valence band consists of fully occupied states at absolute zero temperature, while the conduction band remains empty, with the gap preventing easy electron promotion in the absence of sufficient energy input.[1]Felix Bloch introduced these principles in his 1928 doctoral thesis, applying quantum mechanics to electron behavior in crystal lattices and laying the groundwork for band theory.[13][10] The ideas were further refined throughout the 1930s by contributions from physicists like Alan H. Wilson, who extended the theory to explain the distinctions between metals, insulators, and semiconductors.[14][15]Band diagrams illustrate this structure visually: the valence band appears as a shaded, filled region below the Fermi energy level, the conduction band as an unshaded, empty region above it, and the intervening band gap as a clear vertical separation, often plotted against electronmomentum or energy.[1][16]
Relation to energy bands
In periodic solids, energy bands form through the overlap and interaction of atomic orbitals as atoms are brought together to create a crystal lattice. The discrete energy levels of isolated atoms broaden into continuous bands due to the periodic potential experienced by electrons, allowing a large number of electrons to occupy a range of energies rather than fixed levels. The valence band arises from the filled or partially filled atomic orbitals that participate in bonding, while the conduction band forms from higher-energy orbitals that can accept electrons for conduction when excited. This band formation is described by Bloch's theorem, which accounts for the wave-like behavior of electrons in a periodic lattice.[17]Materials are classified based on the filling of these energy bands and the presence of a band gap E_g between the valence band (VB) and conduction band (CB). In metals, the valence and conduction bands overlap, or the valence band is partially filled, enabling free electron movement and high conductivity without thermal excitation. Insulators possess a large band gap, typically E_g > 3 eV, such that at room temperature, the thermal energy kT \approx 0.025 eV is insufficient to excite electrons from the filled valence band to the empty conduction band, resulting in negligible conduction. Semiconductors feature a smaller band gap, usually $0.1 < E_g < 3 eV, allowing some thermal excitation of electrons across the gap, which leads to temperature-dependent conductivity where carrier concentration increases exponentially with temperature./06%3A_Structures_and_Energetics_of_Metallic_and_Ionic_solids/6.08%3A_Bonding_in_Metals_and_Semicondoctors/6.8B%3A_Band_Theory_of_Metals_and_Insulators)The Fermi level E_F, defined as the chemical potential of electrons at absolute zero, plays a central role in determining conductivity by indicating the highest occupied energy state. In metals, E_F lies within the overlapping bands or partially filled valence band, facilitating easy access to conduction states. For insulators and semiconductors, E_F resides within the band gap; in intrinsic semiconductors, it is positioned near the midpoint between the VB and CB edges, balancing electron and hole concentrations. The position of E_F relative to the bands governs the availability of charge carriers: when E_F is close to the CB edge, electron concentration in the CB increases, enhancing n-type conductivity, while proximity to the VB edge promotes p-type behavior.[1]The density of states g(E), which quantifies the number of available electron states per unit energy interval per unit volume, is particularly important near the band edges as it influences carrier concentration. In three-dimensional semiconductors, assuming parabolic band approximations, the density of states near the conduction band edge is given byg_c(E) = \frac{1}{2\pi^2} \left( \frac{2m_e^*}{\hbar^2} \right)^{3/2} (E - E_c)^{1/2}for E > E_c, where m_e^* is the electron effective mass, \hbar is the reduced Planck's constant, and E_c is the conduction band minimum; a similar form applies to the valence band with effective hole mass m_h^*. This square-root dependence means that the number of states—and thus potential carriers—grows gradually from the band edge, affecting properties like carrier statistics via the Fermi-Dirac distribution.[18]The band structure, including the size and shape of the band gap, is influenced by intrinsic factors such as crystalsymmetry and lattice constant, which define the periodic potential. Higher crystalsymmetry often leads to simpler band dispersions and wider gaps due to stronger orbital overlaps, while deviations in symmetry can narrow or split bands. The lattice constant determines interatomic distances, affecting orbital hybridization: compression (smaller lattice constant) typically widens the band gap by increasing bonding-antibonding separation, whereas expansion narrows it. Temperature effects arise from lattice vibrations (phonons) and thermal expansion, which can shift band edges; for instance, increasing temperature often reduces E_g in semiconductors due to electron-phonon interactions and lattice dilation, with reported decreases of about 0.3–0.5 meV/K in materials like silicon.[19][20]
Types of Band Gaps
Direct versus indirect
In semiconductors, the nature of the band gap is classified as direct or indirect based on the relative positions of the valence band maximum and conduction band minimum in the Brillouin zone with respect to the electron wave vector \mathbf{k}.[21] A direct band gap occurs when these extrema are located at the same \mathbf{k}-point, typically at the center of the Brillouin zone (\mathbf{k} = [0](/page/0)), allowing vertical transitions in the energy-momentum diagram without a change in momentum.[7] This configuration enables efficient optical transitions, as electrons can absorb or emit photons directly, conserving both energy and crystal momentum since photons carry negligible momentum (\mathbf{q} \approx 0).[22]In contrast, an indirect band gap arises when the valence band maximum and conduction band minimum occur at different \mathbf{k}-points, requiring a net change in electron momentum for transitions between these bands.[23] Such transitions cannot proceed solely via photonabsorption or emission, as momentumconservation would be violated; instead, they necessitate interaction with a phonon (lattice vibration) to supply or absorb the required momentum difference.[21] This phonon-assisted process introduces additional scattering mechanisms and reduces the probability of optical transitions, leading to lower radiative efficiency compared to direct band gaps.[24]The distinction is fundamentally captured by Fermi's golden rule for the transition rate W between initial state |i\rangle and final state |f\rangle:W \propto \left| \langle f | H' | i \rangle \right|^2 \delta(E_f - E_i - \hbar \omega),where H' is the perturbation Hamiltonian (e.g., from the electromagnetic field), and the delta function enforces energy conservation with photon energy \hbar \omega.[22] Momentum conservation further requires \mathbf{k}_f = \mathbf{k}_i + \mathbf{q}, where \mathbf{q} \approx 0 for direct transitions, making the matrix element \langle f | H' | i \rangle non-zero without lattice involvement; for indirect cases, the phonon provides \mathbf{q}, but this suppresses the rate by factors related to phonon occupation and coupling strengths.[24]Representative examples illustrate this classification: gallium arsenide (GaAs) exhibits a direct band gap of approximately 1.42 eV at the \Gamma-point, while silicon (Si) has an indirect band gap of about 1.12 eV, with the conduction band minimum at the X-point.[23] These differences profoundly impact optical properties, as direct band gap materials support stronger light-matter interactions, making them preferable for applications requiring efficient photon emission or absorption, whereas indirect materials favor non-radiative recombination pathways.[7]
Electronic versus optical
The electronic band gap, often denoted as E_g, represents the fundamental energy difference between the top of the valence band and the bottom of the conduction band in terms of quasiparticle energies, accounting for many-body interactions such as electron-electron correlations through the self-energy operator \Sigma.[25] This true band gap is typically measured using ultraviolet photoelectron spectroscopy (UPS) to determine the valence band maximum and inverse photoemission spectroscopy (IPES) to probe the conduction band minimum, providing direct access to the unoccupied and occupied electronic states, respectively.[26] These techniques reveal the quasiparticle gap, which includes corrections from many-body effects beyond simple single-particle approximations like density functional theory.In contrast, the optical band gap is determined from the onset of optical absorption in spectroscopy, marking the threshold energy for photon-induced electronic transitions across the band gap.[27] This apparent gap is generally smaller than the electronic band gap due to excitonic effects, where the absorbed photon creates a bound electron-hole pair (exciton) rather than free carriers, reducing the effective transition energy by the exciton binding energy E_b.[25] Excitons form because of attractive Coulomb interactions between the electron and hole, stabilized by the material's dielectric screening, leading to a measurable discrepancy between the two gaps that highlights the role of many-body physics in optical processes.[27]The exciton binding energy E_b can be approximated using a hydrogen-like model adapted for semiconductors, given by the Rydberg equation:E_b = \frac{\mu}{m_0} \cdot \frac{13.6 \, \mathrm{eV}}{\varepsilon_r^2}where \mu is the reduced mass of the electron-hole pair, m_0 is the free electron mass, and \varepsilon_r is the relative permittivity of the material.[28] This binding energy typically ranges from tens to hundreds of meV in semiconductors, causing the optical gap to be E_g - E_b, with stronger binding in materials of lower dimensionality or reduced screening.[29]The distinction between electronic and optical band gaps was first noted in semiconductor studies during the 1950s, following the experimental observation of excitons by E. F. Gross and N. A. Karryev in cuprous oxide (Cu₂O) in 1952, which provided evidence for bound electron-hole states influencing optical absorption edges.[30] Subsequent work in the early 1950s confirmed Wannier-Mott excitons in semiconductors, establishing the excitonic origin of the gap discrepancy and laying the foundation for understanding many-body corrections in solid-state spectroscopy.[31]
Applications in Devices
Optoelectronics (LEDs and lasers)
In optoelectronics, particularly light-emitting diodes (LEDs), the direct band gap of semiconductors facilitates efficient light emission via radiative recombination, where electrons from the conduction band recombine with holes in the valence band, releasing photons with energy approximately equal to the band gap energy E_g. This process is highly efficient in direct band gap materials because momentum conservation is satisfied without phonon assistance, unlike in indirect band gap semiconductors where non-radiative paths dominate. The internal quantum efficiency \eta of an LED is defined as the ratio of the radiative recombination rate R_{\text{rad}} to the total recombination rate, \eta = R_{\text{rad}} / (R_{\text{rad}} + R_{\text{non-rad}}, where non-radiative processes include Shockley-Read-Hall (SRH) and Auger recombination; high \eta values exceeding 80% are achievable in optimized direct band gap structures at low current densities.[32]/Electronic_Properties/Electron-Hole_Recombination)A key milestone in LED development was the invention of the first visible-spectrum LED in 1962 by Nick Holonyak Jr. at General Electric, using a gallium arsenide phosphide (GaAsP) alloy with a direct band gap tuned to emit red light around 650 nm. Common materials for LEDs include III-V semiconductors such as gallium arsenide (GaAs), which has a direct band gap of approximately 1.42 eV at room temperature, suitable for near-infrared emission but alloyed (e.g., as GaAsP) for visible red wavelengths. For shorter wavelengths like blue and ultraviolet, indium gallium nitride (InGaN) alloys are employed, offering direct band gaps tunable from about 2.7 eV (blue) to 3.4 eV (UV) through composition control, enabling white-light generation via phosphor conversion.[33][34]In semiconductor lasers, the direct band gap is essential for achieving population inversion, where the occupation of the upper lasing level exceeds that of the lower level (N_2 > N_1), enabling stimulated emission that amplifies light coherently. The optical gain g(\omega) arises from this inversion and is proportional to the transition dipole moment squared |\mu|^2, the joint density of states \rho(\omega), and the inversion factor, given by g(\omega) \propto |\mu|^2 \rho(\omega) (N_2 - N_1); direct band gap materials like III-V compounds provide the necessary strong momentum matrix elements for appreciable gain near the band edge. These lasers operate on the same recombination principles as LEDs but require optical feedback (e.g., via Fabry-Pérot cavities) to sustain lasing above threshold.Challenges in these devices include reduced efficiency in indirect band gap materials, where Auger recombination—a non-radiative process involving three carriers—further suppresses photon emission by converting energy to heat rather than light. Additionally, efficiency droop in high-power LEDs, particularly nitride-based ones, stems from enhanced Auger recombination at elevated carrier densities, causing external quantum efficiency to decline by up to 50% or more beyond 100 A/cm² current injection.[35][36]
Photovoltaics
In photovoltaics, the band gap plays a crucial role in determining the absorption of photons in solar cells, where only photons with energy greater than the band gap energy E_g (i.e., \hbar \omega > E_g) can excite electrons from the valence band to the conduction band, generating electron-hole pairs for current production.[37] Photons with lower energy are transmitted through the material without contributing to photocurrent, while those with higher energy lose excess energy as heat through thermalization. This fundamental process limits the short-circuit current density J_{sc}, which is proportional to the integral of the absorption coefficient \alpha(\lambda) weighted by the solar irradiance spectrum I_{sun}(\lambda):J_{sc} \propto \int \alpha(\lambda) I_{sun}(\lambda) \, d\lambdawhere \alpha(\lambda) depends directly on E_g, dropping sharply below the absorption edge corresponding to E_g.[38]The theoretical maximum efficiency of single-junction solar cells is governed by the Shockley-Queisser limit, derived from detailed balance principles considering absorption, radiative recombination, and the solar spectrum; for a band gap of approximately 1.1 eV under the AM1.5 spectrum, this limit is about 33%.[39] Efficiency is further influenced by band gap mismatch with the terrestrial solar spectrum: a band gap that is too wide (e.g., >1.5 eV) misses a significant portion of infrared photons, reducing J_{sc}, while a too-narrow gap (e.g., <1.0 eV) increases thermalization losses and lowers the open-circuit voltage V_{oc}, as V_{oc} scales roughly with E_g / q. The overall power conversion efficiency \eta is given by \eta = V_{oc} \cdot J_{sc} \cdot FF, where FF is the fill factor, typically 0.8–0.9 in high-performance cells.[40]Common materials illustrate these trade-offs; crystalline silicon, with an indirect band gap of 1.1 eV, is widely used due to its maturity and abundance but suffers from weaker absorption near the band edge compared to direct-gap materials, requiring thicker absorbers (typically 100–300 \mum) to capture sufficient light.[41] In contrast, metal halide perovskites offer direct band gaps that are tunable from about 1.5 to 2.3 eV through compositional engineering (e.g., varying halide ratios in MAPb(I_{1-x}Br_{x})3), enabling better absorption and higher efficiencies in thin-film cells reaching up to 26.7% for single-junction and over 33% for tandems as of 2025.[42][43][44]To surpass single-junction limits, multi-junction solar cells stack materials with decreasing band gaps to sequentially absorb high-, mid-, and low-energy photons, minimizing spectrum mismatch; for example, GaInP (1.9 eV)/GaAs (1.4 eV)/Ge (0.7 eV) triple-junction cells have achieved efficiencies over 47% under concentrated illumination as of 2025 by optimizing current matching across junctions.[45][46]
Other semiconductor applications
In metal-oxide-semiconductor field-effect transistors (MOSFETs), the band gap determines the energy barrier height between the source and channel regions, influencing carrier injection and overall device performance.[47] Wide band gap materials such as silicon carbide (SiC, E_g \approx 3.2 eV) and gallium nitride (GaN, E_g \approx 3.4 eV) enable high-temperature operation up to 500°C and high-power handling due to reduced thermal generation of carriers and higher breakdown voltages compared to silicon (E_g \approx 1.1 eV).[48] These properties make SiC and GaN suitable for power electronics in electric vehicles and industrial inverters, where efficiency and reliability under extreme conditions are critical.[49]In diodes, Zener breakdown occurs via quantum tunneling across the band gap under high reverse bias, predominant in devices with narrow effective band gaps achieved through heavy doping.[50] For infrared detectors, mercury cadmium telluride (HgCdTe) alloys offer a tunable band gap from approximately 0.1 eV to 1.5 eV by varying the cadmium composition, allowing detection across mid- to long-wave infrared wavelengths (3–12 μm).[51] This tunability arises from the alloy's pseudo-binary nature, blending semimetallic HgTe (E_g \approx 0 eV) with semiconducting CdTe (E_g \approx 1.5 eV), enabling customizable cutoff wavelengths for focal plane arrays in thermal imaging.[52]Thermoelectric devices convert heat to electricity via the Seebeck effect, where the band gap optimizes the power factor S^2 \sigma (with S as the Seebeck coefficient and \sigma as electrical conductivity). The Seebeck coefficient follows the Mott-Jones relation:S \propto \frac{\pi^2 k_B^2 T}{3e} \left( \frac{d \ln \sigma}{dE} \right)_{E_F}where k_B is Boltzmann's constant, T is temperature, e is the electron charge, and the derivative is evaluated at the Fermi level E_F.[53] An optimal band gap of around 0.2–0.4 eV maximizes this by balancing carrier concentration and entropy transport, enhancing efficiency in materials like bismuth telluride for waste heat recovery.[54]Challenges in these applications include defect states within the band gap that act as electron or hole traps, reducing carrier mobility and increasing recombination losses.[55] Doping introduces controlled impurity levels near the band edges to modulate conductivity—donors near the conduction band for n-type and acceptors near the valence band for p-type—while avoiding deep traps that degrade performance.[56] For instance, wide band gap semiconductors with E_g > 2 eV, such as GaN, are used in radio-frequency (RF) amplifiers for high-power, high-frequency operation in radar and communication systems, benefiting from low noise and high electron mobility.[57]
Extensions to Other Systems
Quasi-particle band gaps
In solid-state physics, quasi-particle band gaps refer to energy ranges where the propagation of collective excitations, such as phonons, magnons, and excitons, is forbidden within periodic structures. These gaps arise from the interplay of the quasi-particles' dispersion relations and the lattice periodicity, analogous to electronic band gaps but for bosonic or composite excitations. Unlike single-particle electronic bands, quasi-particle spectra often emerge from diagonalizing quadratic Hamiltonians that couple creation and annihilation operators, leading to hybridized modes with gapped structures.[58]Phonon band gaps occur in phononic crystals, which are periodic composites designed to manipulate acoustic waves, creating forbidden frequency ranges where phonons cannot propagate. These gaps typically result from Bragg scattering, where the periodic modulation of elastic properties imposes boundary conditions on the wave equation, opening gaps in the ω(k) dispersion relation at the Brillouin zone edges. For instance, in one-dimensional phononic crystals composed of alternating layers with different sound speeds, the dispersion can be described by\cos(ka) = \cos(k_1 d_1) \cos(k_2 d_2) - \frac{Z_1 + Z_2}{2\sqrt{Z_1 Z_2}} \sin(k_1 d_1) \sin(k_2 d_2),where k_i, d_i, and Z_i are the wave number, layer thickness, and acoustic impedance of each material, respectively; band gaps appear where |\cos(ka)| > 1. This mechanism was first theoretically demonstrated for elastic waves in periodic media, enabling applications like acoustic insulators. Complete three-dimensional phonon band gaps, spanning all polarizations, have been achieved in structures like opal-inspired lattices with face-centered cubic symmetry.[58][59]Magnon band gaps manifest in magnetic materials as energy separations in the spin-wave spectrum, often induced by magnetic anisotropy that breaks rotational symmetry and lifts the degeneracy at zero wave vector. In ferromagnets, the magnon dispersion typically takes the form \omega(\mathbf{k}) = \Delta + D k^2, where \Delta is the gap energy arising from uniaxial anisotropy K via \Delta \approx 2K / M_s (with M_s the saturation magnetization), and D is the spinstiffness; an applied field adds a Zeeman term g \mu_B B. This gap prevents low-energy spin excitations, stabilizing the ordered state against thermal fluctuations. Seminal studies on anisotropic Heisenberg models showed that such gaps enhance magnon stability in thin films and superlattices, with experimental confirmation in materials like yttrium iron garnet. In antiferromagnets, anisotropy similarly opens gaps, though the dispersion involves two branches with linear terms at low k.[60][61]Exciton band gaps in insulators represent effective energy separations for bound electron-hole pairs, smaller than the bare electronic band gap E_g due to Coulomb attraction that forms excitons as composite quasi-particles. In narrow-gap semiconductors or semimetals approaching the excitonic instability, the exciton binding energy E_b can reduce the effective gap to E_g - E_b, potentially leading to spontaneous exciton condensation and an excitonic insulator phase where the chemical potential lies within this renormalized gap. The theoretical framework involves solving the Bethe-Salpeter equation for the two-particle Green's function, yielding exciton dispersion with a gap modulated by screening and lattice effects; for Wannier excitons in three dimensions, E_b \approx 13.6 \, \mathrm{eV} \times (\mu / m_e) / \epsilon_r^2, where \mu is the reduced mass and \epsilon_r the dielectric constant. This concept was proposed in the context of electron-hole liquids, with gaps observed in materials like layered transition-metal dichalcogenides under strain.[62]In superlattice structures, quasi-particle band gaps can be engineered for hybrid excitations like polaritons, which couple photons and phonons or excitons, forming minibands separated by gaps due to the periodic potential. For example, in semiconductorsuperlattices, the electronic minibands split into polariton branches with gaps tunable by layer thickness and coupling strength, enabling forbidden zones for light-matter quasiparticles. Such structures demonstrate miniband transport with gaps arising from avoided crossings in the dispersion, as seen in GaAs/AlGaAs systems where polariton gaps support terahertz emission control.The underlying theoretical framework for these quasi-particle band gaps relies on Bogoliubov quasiparticles, obtained by diagonalizing quadratic Hamiltonians of the form H = \sum_{\mathbf{k}} \begin{pmatrix} a_{\mathbf{k}}^\dagger & a_{-\mathbf{k}} \end{pmatrix} \begin{pmatrix} A(\mathbf{k}) & B(\mathbf{k}) \\ B^*(\mathbf{k}) & A(-\mathbf{k}) \end{pmatrix} \begin{pmatrix} a_{\mathbf{k}} \\ a_{-\mathbf{k}}^\dagger \end{pmatrix}, where a_{\mathbf{k}} are bosonic operators for phonons, magnons, or excitons, and the off-diagonal B(\mathbf{k}) captures pairing or anomalous terms from interactions. The Bogoliubov transformation b_{\mathbf{k}} = u_{\mathbf{k}} a_{\mathbf{k}} + v_{\mathbf{k}} a_{-\mathbf{k}}^\dagger (with |u_{\mathbf{k}}|^2 - |v_{\mathbf{k}}|^2 = 1) diagonalizes H into free quasi-particle modes with energies \epsilon_{\mathbf{k}} = \sqrt{A(\mathbf{k})^2 - |B(\mathbf{k})|^2}, revealing gaps when \min \epsilon_{\mathbf{k}} > 0. This approach, originally for superfluids, applies to solids by incorporating lattice periodicity, yielding band structures with gaps from the resulting \epsilon(\mathbf{k}).
Photonic and other band gaps
Photonic band gaps arise in periodic dielectric structures, where the periodic variation in refractive index creates ranges of frequencies in which electromagnetic wave propagation is forbidden, analogous to electronic band gaps in solids. These structures, known as photonic crystals, were first proposed by Eli Yablonovitch in 1987 as a means to inhibit spontaneous emission in solid-state devices by embedding emitters in a medium that prevents light of certain frequencies from propagating.[63] In one dimension, such as in multilayer dielectric stacks, the relative photonic band gap width is approximately given by \Delta \omega / \omega \approx \Delta \epsilon / \epsilon, where \Delta \epsilon / \epsilon represents the contrast in the dielectric constant between layers; this simple scaling holds for Bragg reflectors but becomes more intricate in higher dimensions due to the need for complete three-dimensional periodicity to achieve a full photonic band gap.[64] In three dimensions, realizing a true photonic band gap requires careful design of the lattice geometry, as the gap size and frequency range depend on the dielectric contrast and structural symmetry, often resulting in narrower gaps compared to one-dimensional cases unless high-contrast materials are used.[65]Topological band gaps extend the concept to systems where the band structure exhibits nontrivial topology, particularly in Chern insulators, which are two-dimensional materials with a bulk energy gap hosting protected edge states that conduct without backscattering. These edge states emerge due to the nonzero Chern number, a topological invariant calculated as the integral of the Berry curvature F(k) over the Brillouin zone, leading to quantized Hall conductance \sigma_{xy} = n e^2 / h, where n is the Chern number. The Berry curvature, analogous to a magnetic field in momentum space, ensures robustness against perturbations, making these gaps ideal for dissipationless transport in quantum devices.[66]Band gap analogs appear in other wave systems beyond electromagnetics, such as acoustic metamaterials, where periodic arrangements of materials with differing acoustic impedances create frequency ranges forbidding sound propagation. In seismic applications, these metamaterials generate band gaps that attenuate ground vibrations, protecting structures from earthquakes by reflecting low-frequency waves below the material's characteristic scale.[67] For instance, arrays of resonators or layered media can produce wide seismic band gaps in the sub-Hertz to tens-of-Hertz range, leveraging local resonances rather than Bragg scattering alone.[68]Applications of photonic band gaps include hollow-core photonic bandgap fibers, which guide light primarily through air rather than solid material, achieving ultralow propagation losses below 0.1 dB/km by confining modes within the band gap of the periodic cladding.[69] These fibers enable high-power delivery with minimal nonlinear effects and facilitate slow light propagation, where the group velocity is reduced near band edges, enhancing light-matter interactions for sensing and nonlinear optics.[70]
Materials and Data
Band gaps in common materials
Semiconductors are characterized by narrow band gaps typically ranging from 0.1 to 3 eV at room temperature, which allow partial thermalexcitation of electrons across the gap, enabling controlled electrical conductivity.[71] Elemental semiconductors like silicon (Si, 1.12 eV) and germanium (Ge, 0.66 eV) exemplify this range, while compound semiconductors such as gallium arsenide (GaAs, 1.42 eV) and cadmium telluride (CdTe, 1.5 eV) offer tunable properties for optoelectronic applications.[72] In alloyed semiconductors, band gap energy varies approximately linearly with composition according to Vegard's law, which relates the lattice parameter to the alloy fraction and predicts a corresponding linear shift in E_g, facilitating band gap engineering in ternary systems like AlGaAs.[73]Insulators possess wide band gaps exceeding 5 eV, resulting in negligible electron excitation at room temperature and high electrical resistivity. For instance, diamond exhibits a band gap of 5.5 eV, and aluminum oxide (Al₂O₃) has a value around 8.8 eV. These wide gaps contribute to exceptional dielectric strength, as the large energy barrier prevents carrier generation and breakdown under high electric fields, making such materials ideal for capacitors and insulators in high-voltage devices.[74]Metals lack a true band gap at the Fermi level, leading to high conductivity from overlapping valence and conduction bands; however, transition elements like those in the d-block display pseudogaps—regions of reduced density of states—in their d-bands due to crystal field splitting and hybridization effects.[75]The band gap in semiconductors decreases with increasing temperature, primarily due to lattice expansion and electron-phonon interactions that broaden the bands. This dependence is commonly modeled by the Varshni equation:E_g(T) = E_g(0) - \frac{\alpha T^2}{T + \beta}where E_g(0) is the band gap at 0 K, and \alpha and \beta are material-specific constants reflecting the strength and characteristic temperature of the thermal effects.[76]Under hydrostatic pressure, band gaps in many semiconductors narrow because compression reduces interatomic distances, enhancing orbital overlap and shifting band edges closer together, which can transition indirect gaps to direct or even induce metallization at high pressures.[77]
Tabulated band gap values
The band gap energies of semiconductors and insulators vary widely depending on the material's composition, structure, and external factors, with values typically measured at room temperature (300 K) using techniques such as optical absorption spectroscopy, which determines the onset of light absorption corresponding to electronic transitions, and spectroscopic ellipsometry, which analyzes changes in light polarization to derive the dielectric function and extrapolate the band edge.[78][79] These methods often yield the optical band gap, which can exceed the fundamental electronic band gap by the excitonbinding energy (typically 10–100 meV in inorganic semiconductors due to electron-hole pair formation).[80]The following table compiles representative band gap values for selected materials, drawn from experimental compilations; direct band gaps allow momentum-conserving optical transitions, while indirect ones require phonon assistance.[5][81][82]
Band gap values can exhibit variability due to external modifications; for instance, biaxial tensile strain in silicon nanowires can reduce the band gap by up to 100 meV per 1% strain through deformation potential effects on the conduction band minima.[83] Heavy doping in semiconductors like silicon leads to band gap narrowing of 10–50 meV at concentrations above $10^{19} cm^{-3}, arising from many-body interactions and impurity band formation that effectively broaden the valence and conduction bands.[84]In emerging two-dimensional materials, graphene possesses a zero band gap in its monolayer form but can be tuned to up to 0.25 eV in bilayers via perpendicular electric fields that break symmetry and open a gap at the Dirac points.[85] Similarly, monolayer molybdenum disulfide (MoS_2) exhibits a direct band gap of 1.8 eV, contrasting with the indirect 1.2 eV gap in its bulk form, enabling efficient optoelectronic applications; this value was determined via photoluminescence spectroscopy.[86]