Dark current in monolithic extended-SWIR GeSn PIN photodetectors

TL;DR

This study investigates the dark current mechanisms in GeSn PIN photodetectors operating at 2.6 μm, revealing how diffusion, SRH, and TAT processes vary with temperature, bias, and device size, impacting device performance.

Contribution

It provides a detailed analysis of dark current mechanisms in GeSn PIN photodetectors, highlighting the influence of device size, bias, and temperature on leakage processes.

Findings

Diffusion and SRH dominate at low bias and small size.

TAT leakage becomes dominant at high bias and larger size.

Non-radiative lifetime reduces with increasing bias due to TAT leakage.

Abstract

The monolithic integration of extended short-wave infrared (e-SWIR) photodetectors (PDs) on silicon is highly sought-after to implement manufacturable, cost-effective sensing and imaging technologies. With this perspective, GeSn PIN PDs have been the subject of extensive investigations because of their bandgap tunability and silicon compatibility. However, due to growth defects, these PDs suffer a relatively high dark current density as compared to commercial III-V PDs. Herein, we elucidate the mechanisms governing the dark current in $2.6 \, \mu$m GeSn PDs at a Sn content of $10$ at.%. It was found that in the temperature range of $293 \, $K -- $363 \,$K and at low bias, the diffusion and Shockley-Read-Hall (SRH) leakage mechanisms dominate the dark current in small diameter ($20 \, \mu$m) devices, while combined SRH and trap assisted tunneling (TAT) leakage mechanisms are prominent in larger diameter ($160 \, \mu$m) devices. However, at high reverse bias, TAT leakage mechanism becomes dominant regardless of the operating temperature and device size. The effective non-radiative carrier lifetime in these devices was found to reach $\sim 300$ -- $400$ ps at low bias. Owing to TAT leakage current, however, this lifetime reduces progressively as the bias increases.