Semiconductor Physics: Energy Levels & Doping
Uploaded by
soham sawantSemiconductor Physics: Energy Levels & Doping
Uploaded by
soham sawantCommon questions
Powered by AIThe energy density of states g(E) near the conduction band minimum is given by g(E) = (1/2π²)(2m*/ħ²)^(3/2)√(E-E_c), where E is the energy, E_c is the conduction band edge, and m* is the effective mass tensor expressed as a scalar. This stems from the parabolic band assumption near the extremum, translating to a square root dependency on energy above the band edge .
In silicon, germanium, and gallium arsenide, the intrinsic Fermi-level shifts closer to the valence band edge as temperature increases due to increasing intrinsic carrier concentrations. In Si, the Fermi-level is approximately mid-gap at room temperature but moves closer to the valence band at higher temperatures. In Ge and GaAs, similar behavior is observed, though the specific energy level shift is different due to each material's band gap and effective masses .
The electron concentration n(T) and the Fermi-level E_F are influenced by the doping densities and activation energies of the two donor types. By solving charge neutrality equations considering both ionized donor levels and conduction band electrons, one can derive expressions for n(T). The temperature dependence typically involves an exponential growth relation of free electron concentration with thermal energy, modifying E_F shifts, observable in plots of ln(n) versus T as carriers freeze out at low temperatures .
The total current through a semiconductor out of equilibrium is determined by the gradient of the quasi-Fermi levels: J_total = q(nabla E_Fn)σ_n + q(nabla E_Fp)σ_p, where E_Fn and E_Fp are the quasi-Fermi levels for electrons and holes, respectively, and σ_n, σ_p are their conductivities. This relationship derives from the generalized current continuity equations under non-equilibrium conditions, relating spatial changes in quasi-Fermi levels to carrier currents .
In a GaAs sample at room temperature, as the shallow acceptor doping concentration N_a increases from 10^13 cm^-3 to 10^17 cm^-3, the hole concentration p increases while the electron concentration n decreases due to compensation effects. The increased acceptor concentration provides more states for electrons from the valence band, boosting hole concentration, while excess doping significantly reduces free electron numbers due to increased recombination with holes .
For an intrinsic degenerate n-type semiconductor, the Fermi-level position E_F(T) can be determined by integrating the density of states with the Fermi-Dirac distribution to find number equality between charge carriers and intrinsic carrier density. Techniques such as using the full Fermi-Dirac integral rather than Boltzmann approximation are necessary due to high carrier densities, and temperature dependence is modeled by evaluating the position shift relative to the conduction band minimum .
The valence band effective density of states N_v in semiconductors is influenced by the effective masses of heavy (m_hh) and light holes (m_lh) through a composite effective mass m*_v calculated by the convoluted density-of-states effective mass formula √(m_hhm_lh). This reflects how heavier hole masses decrease DOS at higher energy states in the valence band, and effective integration over these energies adjusts the combined carrier statistics through an altered density state count at energy levels .
The minimum energy separation between levels in the valence band in a simple cubic lattice can be determined by evaluating the energy difference between two adjacent momentum states in the free electron model. For a 3D system, the energy E is given by E = ħ²k²/2m, where k is the wave vector. In a cubic lattice with lattice constant 'a' and dimensions L_x, L_y, and L_z, the allowed k values are quantized as k_x = (2πn_x)/L_x, k_y = (2πn_y)/L_y, k_z = (2πn_z)/L_z. The smallest non-zero k results in the minimum energy separation, found by taking the difference between energies at n_x, n_y, n_z = 1 and n_x, n_y, n_z = 0, leading to ΔE_min = ħ²(2π)²/2mL² .
The inverse effective mass tensor at the band extrema is derived from the second derivative of the E-k relationship (i.e., the band structure) and is expressed as (1/m*)_{ij} = (1/ħ²) d²E/dk_i dk_j, where i, j are the coordinate axes. This tensor describes how the curvature of the energy band at extremes (valence or conduction band) translates into the effective mass tensor, characterizing carrier dynamics in semiconductors .
The conduction band effective density of states N_c is calculated using the formula N_c = 2(2πm*_n k_BT/h²)^(3/2), where m*_n is the electron effective mass tensor at the conduction band minimum expressed as a scalar effective mass. This involves evaluating the determinant of the mass tensor matrix and correlating it to the density of states effective mass used to calculate the effective density of states at the conduction band .