We present a comparison of theoretical calculations and experimental measurements of the Auger recombination rate in a narrow-gap semiconductor superlattice with a complex band structure. The calculations and measurements indicate that the rate depends on density as n 2 for low density, and changes to an n dependence when the electrons and holes become degenerate. The calculations are the first to incorporate superlattice umklapp processes, which contribute about half of the total rate and substantially improve the agreement with experiment.

A comparison is performed between measured and calculated Auger recombination rates for four different narrow-gap superlattices based on the InAs/GaSb/AlSb material system. The structures are designed for optical or electrical injection for mid-infrared laser applications, with wavelengths ranging from 3.4 to 4.1 m. The electronic band structures are computed employing an accurate 14-band restricted basis set ͑superlattice K"p) methodology that utilizes experimental information about the low-energy electronic structure of the bulk constituents. The superlattice band structures and their associated matrix elements are directly employed to compute Auger recombination rates. Varying amounts of Auger recombination suppression are displayed by the various superlattices as compared to bulk mid-infrared systems. The greatest disagreement between theory and experiment is shown for the structure predicted to have the most Auger suppression, suggesting the suppression is sensitive either to theoretical or growth uncertainties.

We present calculations of the differential gain and threshold current densities for a 3.7 m multiple quantum well structure consisting of a ''well'' composed of several periods of an InAs/InGaSb superlattice alternating with a quinternary alloy ''barrier.'' We find serious limitations to the optical properties of active regions composed of these multiple quantum wells, and propose a four-layer superlattice structure which corrects these problems.

We have used the 830 nm, subpicosecond output of a mode-locked Ti:sapphire laser, together with subpicosecond 3.55 m pulses from a synchronously pumped optical parametric oscillator, to perform room-temperature, time-resolved, differential transmission measurements on a multiple quantum well structure with AlGaSb barriers and GaInSb/InAs superlattice wells. From these measurements, we have determined a Shockley-Read-Hall rate of 2.4ϫ10 8 s Ϫ1 and an Auger coefficient of 7ϫ10 Ϫ27 cm 6 /s. In addition, we estimate the carrier capture efficiency into the wells to be ϳ52% and have demonstrated that carrier cooling, cross-well transport, and capture are complete within ϳ10 ps after excitation.

The ideal performance of bulk, quantum well, and superlattice active regions for III–V interband midinfrared lasers are compared according to the maximum net gain per unit current density. Based on this figure of merit, which is appropriate for high-power as well as near-threshold operation, InAsSb quantum well active regions should have an order of magnitude lower threshold current than bulk InAs at room temperature. Optimized four-layer superlattices based on the InAs/GaInSb material system, however, should have two to ten times lower threshold currents than the quantum well active regions. Optimal thicknesses for these active regions were evaluated assuming a separate confinement region design. For the four-layer superlattices the optimal thickness is substantially thinner than has been commonly grown: 3 periods rather than 40 periods.