Abstract

In this paper we report direct measurements of carrier recombination at 1.58 µm via time-correlated single photon counting using a Ge single-photon avalanche photodiode with time resolution of 300 ps. Under high optical excitation, this provides a means for assessing the contribution of Auger processes. These are believed to play a role in the high temperature sensitivity of 1.5 µm lasers although to what extent remains unclear. The strained (1% compressive) and unstrained samples mimicked our laser structures (four and 5 30 Å InGaAs quantum wells with 150 Å InGaAsP barriers lattice matched to InP) but were nominally undoped throughout (background~5 × 10¹⁵ cm⁻³,n-type) Fits to the data were calculated using the initial photogenerated earner density (Δn(t=0)) and Auger coefficient (C_A) as fitting parameters and included radiative, extrinsic non-radiative and Auger recombination. For example, fitting the room temperature photoluminescence decay for a strained sample gave C_A= 1.0 × 10⁻²⁸ cm⁶ s⁻¹ and Δn(t=0) = 3.5 ×10¹⁷ cm⁻³, typical of the sample set. At this initial carrier density, Auger accounts for 13% of the total recombination. Fit quality is degraded if Auger recombination is omitted. This value for C_A is greater than previously measured (1.3 × 10^-29 cm⁶ s^-1 [1]), and has important implications for laser operation. In order to explore the relation between the time-resolved photoluminescence results (TRPL) and the laser devices in more detail, we have studied lasers with similar layer structures to the time-resolved photoluminescence samples and cavity lengths the range 300- 1160 µm. The temperature dependence of the threshold current shows two distinct regions; for T=230-340K threshold current increases exponentially with temperature, whilst above 340K it increases super-exponentially. By comparing these results with the TRPL studies as a function of temperature we are able to assess the relative contribution of Auger recombination in these samples.

© 1998 IEEE