2D simulation for the impact of edge effects on the performance of planar InGaAs/InP SPADs

InGaAs/InP SPADs are solid-state devices able to detect near-infrared single photons up to 1700 nm. The pn junction is defined by Zn diffusions in a lightly n-doped InP layer. If a simple Zn diffusion were employed, it would suffer edge effects; therefore a double diffusion is used in order to smooth the electric field at the periphery of the active area. However, since most of the main parameters of the SPAD depend on the electric field profile, it is of outmost importance to properly evaluate the electric field, both in the active area and at its periphery. Currently-available programs for 2D simulations require heavy and long computations and are not tailored for SPAD performance assessment, thus often 1D custom simulation is performed for qualitative evaluation of “detection” characteristics. We present two 1D and 2D device simulators designed for InGaAs/InP SPAD detectors and the models we implemented therein. We compare the differences in the results between the 1D and 2D approaches in terms of electric field profile, breakdown voltage, trigger efficiency and dark count rates. The 1D simulator overestimates breakdown voltage by few percent, while the 2D simulator matches the measured values. We show how trigger efficiency is not constant in the device and how the high electric field near the edges contributes to increase the dark count rate due to tunneling effects.