Dopability limits in Al-rich AlGaN alloys for far-UVC LEDs

TL;DR

This study investigates the limits of doping in Al-rich AlGaN alloys for far-UVC LEDs, revealing how defects and impurities affect carrier injection and device performance, with implications for improving solid-state UV lighting technology.

Contribution

It provides a systematic analysis of intrinsic and extrinsic defects in high-Al-content AlGaN alloys, emphasizing the importance of alloy modeling and impurity effects on doping efficiency.

Findings

Si dopants form compensating DX centers in Al-rich environments.

Carbon is identified as the most detrimental unintentional impurity.

Explicit alloy modeling aligns calculations with experimental carrier concentrations.

Abstract

Transitioning to solid-state ultraviolet (UV) lighting is critical for reducing global energy utilization to meet net-zero targets. AlGaN-based far-UVC LEDs offer a mercury-free, energy-efficient alternative to conventional mercury lamps, yet their performance is severely bottlenecked by poor carrier injection at Al compositions exceeding 80\%. Point defects are known to significantly affect carrier concentrations and radiative recombination efficiency, however, systematic studies of point defects in AlGaN alloys remain scarce. In this work, we investigate intrinsic and extrinsic defects in high-Al-content Al$_{1-x}$Ga$_x$N alloys ($x$ = 1/6, 1/4, and 1/3). We reveal that explicit alloy modeling and proper treatment of the temperature dependence of the band gap are essential to bring calculated carrier concentrations in line with experimental observations. We uncover that Si dopants preferentially substitute minority Ga atoms, forming compensating negative-\textit{U} \textit{DX} centers in Al-rich environments that severely limit n-type conductivity. We identify carbon as the most detrimental unintentional impurity, while the impact of oxygen and hydrogen is negligible in Si-doped samples typically used for devices. These findings highlight the significance of explicit alloy modeling and provide valuable insights into the design of AlN-based alloys.