Low Resistance Asymmetric III-Nitride Tunnel Junctions Designed by Machine Learning

Abstract

The tunnel junction (TJ) is a crucial structure for numerous III-nitride devices. A fundamental challenge for TJ design is to minimize the TJ resistance at high current densities. In this work, we propose the asymmetric p-AlGaN/i-InGaN/n-AlGaN TJ structure for the first time. P-AlGaN/i-InGaN/n-AlGaN TJs were simulated with different Al or In compositions and different InGaN layer thicknesses using TCAD (Technology Computer-Aided Design) software. Trained by these data, we constructed a highly efficient model for TJ resistance prediction using machine learning. The model constructs a tool for real-time prediction of the TJ resistance, and the resistances for 22,254 different TJ structures were predicted. Based on our TJ predictions, the asymmetric TJ structure (p-Al0.7Ga0.3N/i-In0.2Ga0.8N/n-Al0.3Ga0.7N) with higher Al composition in p-layer has seven times lower TJ resistance compared to the prevailing symmetric p-Al0.3Ga0.7N/i-In0.2Ga0.8N/n-Al0.3Ga0.7N TJ. This study paves a new way in III-nitride TJ design for optical and electronic devices.

A machine learning study on superlattice electron blocking layer design for AlGaN deep ultraviolet light-emitting diodes using the stacked XGBoost/LightGBM algorithm

“…[35][36][37][38][39] Lin et al applied a tree-based ensemble XGBoost model to predict the tunnel junction resistance and designed a novel asymmetrical tunnel junction with low resistance. 40 In this paper, we propose a new ML framework for the SL-EBL design, which can predict the IQE with high accuracy. Instead of incurring a single model, we stacked two different existing methods in the training and testing process: XGBoost and LightGBM.…”

“…Recently, researchers used a neural network structure to predict the optical and electronic properties of semiconductor devices [35][36][37][38][39]. Lin et al applied a treebased ensemble XGBoost model to predict the tunnel junction resistance and designed a novel asymmetrical tunnel junction with low resistance 40. In this paper, we propose a new ML framework for the SL-EBL design, which can predict the IQE with high accuracy.…”

Advanced Design of a III‐Nitride Light‐Emitting Diode via Machine Learning

“…[13] Rongyu Lin et al leveraged XG-Boost to predict resistance of the p-AlGaN/i-InGaN/n-AlGaN tunnel junctions. [14] Nevertheless, most current works are limited to specific layers in the device, thus lacking adequate features to provide sufficient accuracy for the prediction space. At the same time, these results are based on the dataset simulated by the researchers themselves, which may result in homogenized data and introduce biased dataset.…”

Printed layers height calibration curve and porosity in laser melting deposition of Ti6Al4V combining experiments, mathematical modelling and deep neural network

“…Porosity is a common defect caused by a highly focused laser beam's rapid heating and cooling of a given material (Liao et al, 2020;Niu et al, 2022;Zhuang et al, 2022). Porosity can be analyzed using experiments, modeling and machine learning (Lin et al, 2020(Lin et al, , 2021(Lin et al, , 2022.…”