Self-consistent scattering model of carrier dynamics in GaAs-AlGaAs terahertz quantum-cascade lasers

Towards automated design of quantum cascade lasers

“…The charge transport through a quantum cascade laser is thought to be mainly caused by incoherent electron-longitudinal optical phonon ͑LO͒ and electron-electron scattering, 23 and the inclusion of all mechanisms that take place in both the active and injector/collector regions of the structure is essential for an accurate description of the carrier dynamics. This is particularly significant in view of the fact that the gain values calculated from the simple three-level model under the unity injection approximation considerably depart from those obtained experimentally, [24][25][26][27] which is believed to be due to thermal activation of carriers directly into the continuum 13,28 and cannot be described within a simple model. The existence of this parasitic current path is one of the main obstacles in extending the working temperature range of GaAsbased QCLs.…”

“…[24][25][26][27] Consider a biased QCL structure with a large number of periods, each consisting of multiple quantum wells. The energy spectrum is formally continuous, but to a very good approximation can be considered to consist of quasidiscrete states ͑resonances͒.…”

“…For example, in the GaAs-based THz QCLs the energy separation between most of the subbands is smaller than the LO phonon energy and electron-electron scattering becomes an important scattering mechanism, hence necessitating a large number of possibly relevant scattering processes to be accounted for. 26 The modal gain is proportional to the population inversion ⌬n ij , which can be obtained from the self-consistent solution under the steady-state condition:…”

Self-consistent energy balance simulations of hole dynamics in SiGe∕SiTHz quantum cascade structures

“…Simulations of actual cascade structures which include electron-electron scattering calculated in this manner show quite good agreement with experiments. [10][11][12] Although the validity of this screening model may seem questionable in view of the results of calculations which use the dynamic RPA dielectric function, 40 the latter is computationally too demanding to be manageable for structures of the complexity encountered in present QCLs. For holes, with their nonparabolicity and anisotropy, the dynamic screening approach would be even slower.…”

“…One approach relies on self-consistent solutions of rate equations. [8][9][10][11][12][13] Another approach uses the microscopic, and computationally more demanding Monte Carlo technique. [14][15][16] Although the latter does not make the assumption of equilibriumlike carrier distributions over states within any single subband, and therefore gives a deeper insight into the electron dynamics, the former are much faster while still giving quite good estimates of device characteristics.…”