Method for determining effective nonradiative lifetime and leakage losses in double-heterostructure lasers

Abstract

Carrier losses in double-heterostructure lasers are twofold: (i) nonradiative recombination through killers in the bulk of the active region and at all its boundaries (interfaces and surfaces), and (ii) leakage out of the active region. A simple theory shows the following. In the high-injection regime (p?n) all processes under (i) are directly proportional to n. Consequently their contributions can be lumped together in a single effective nonradiative carrier lifetime τnr ; this τnr is constant (i.e., independent of n) owing to the constant degree of occupation of all killers in the mentioned regime. On the other hand, the leakage losses (ii) are superlinear in n. This provides a well-grounded basis for disentangling the contributions of (i) and (ii) in a given sample. Further, a simple method is presented for accurately determining τnr from data of the external quantum efficiency ηext measured as a function of current I in the spontaneous high-injection regime below the laser threshold. Knowledge of the light-extraction factor (i.e., the ratio of external and internal quantum efficiencies) is essentially unnecessary with this method. However, optionally it can be determined easily from a slight extension of the method. For illustration the method of determining τnr, which is also applicable to double-hetero LED’s, has been applied to some thirty LPE and metal-organic VPE GaAs-(Ga,Al)As lasers of widely varying qualities. The values found vary between 0.8 and 55 ns. From the measured values of τnr it follows that the upper limit for the interface recombination velocity in the best samples is 270 cm/s. For most samples τnr cannot account for all electrical losses at laser threshold. The superlinear excess losses are ascribable to leakage.