Wide Bandgap Semiconductor Device Design via Machine Learning

  • Rongyu Lin

Student thesis: Doctoral Thesis


The research of III-nitride wide-bandgap semiconductor devices, such as laser diodes (LDs), ultra-violet (UV) light-emitting diodes (LEDs), and high electron mobility transistors (HEMTs), has recently increased. Numerous opportunities exist for performance improvement in the wide bandgap semiconductor device structure, including material selection, compound compositions, polarization effects, and layer thicknesses. On the other hand, they can make optimization more challenging. It still takes a lot of resources to analyze and test complicated structures in a systematic manner. This dissertation creates a new path for device design by using TCAD and machine learning to deliver quick forecasts of III-nitride semiconductor device performance. The dissertation includes three major components. In Chapter 2, the TCAD-assisted HEMT device design is discussed. We demonstrate the performance improvement of using the new material BAlN as an interlayer in GaN/AlGaN HEMT devices and compare the various interlayer design alternatives for HEMTs. In chapter 3, we propose asymmetrical p-AlGaN/i-InGaN/n-AlGaN tunnel junctions (TJs) by combining machine learning (ML) with TCAD calculations. The resistances for 22254 various TJ structures were predicted by the model, which creates a tool for real-time TJ resistance prediction. Based on our TJ predictions, we proposed asymmetric TJ with higher Al content in the p-layer and lower TJ resistance. In Chapter 4, using the stacked XGBoost/LightGBM algorithm, we thoroughly examined the superlattice (SL) electron blocking layer (EBL) for AlGaN deep ultra-violet (DUV) LEDs. Based on the ML model, we suggest a low Al-content SL-EBL (1 nm/5 nm Al0.7Ga0.3N/Al0.58Ga0.42N) that is simpler, experimentally realizable and can greatly improve carrier transport. Additionally, we examine the prediction data and show how the composition and thickness affect the improvement of the IQE. The work contributes to the advancement of using SL-EBLs for high-efficiency DUV LEDs by providing methodical research on SL-EBLs. This dissertation presents novel approaches to the design of electrical and optical wide bandgap semiconductor devices, which opens up a new avenue for future research. It is possible that it might be used in a broad variety of fields, including illumination, sensing, disinfection, and power devices.
Date of AwardNov 2 2022
Original languageEnglish (US)
Awarding Institution
  • Computer, Electrical and Mathematical Sciences and Engineering
SupervisorXiaohang Li (Supervisor)


  • wide bandgap
  • semiconductor
  • machine learning
  • LED
  • HEMT
  • GaN
  • AlN
  • tunnel junction
  • superlattice
  • EBL

Cite this