Toward Precise n-Type Doping Control in MOVPE-Grown β-Ga2O3 Thin Films by Deep-Learning Approach

In this work, we train a hybrid deep-learning model (fDNN, Forest Deep Neural Network) to predict the doping level measured from the Hall Effect measurement at room temperature and to investigate the doping behavior of Si dopant in both (100) and (010) beta-Ga2O3 thin film grown by the metalorganic vapor phase epitaxy (MOVPE). The model reveals that a hidden parameter, the Si supplied per nm (mol/nm), has a dominant influence on the doping process compared with other process parameters. An empirical relation is concluded from this model to estimate the doping level of the grown film with the Si supplied per nm (mol/nm) as the primary variable for both (100) and (010) beta-Ga2O3 thin film. The outcome of the work indicates the similarity between the doping behavior of (100) and (010) beta-Ga2O3 thin film via MOVPE and the generality of the results to different deposition systems.

Automated summary

In this research, a hybrid deep-learning model (fDNN) was trained to predict the doping level in (100) and (010) beta-Ga2O3 thin films grown by MOVPE and investigate the doping behavior of Si dopant. The model revealed that the Si supplied per nm (mol/nm) had a dominant influence on the doping process. An empirical relation was concluded to estimate the doping level using the Si supplied per nm (mol/nm) as the primary variable. The results indicated the similarity of doping behavior between (100) and (010) beta-Ga2O3 thin films through MOVPE and the generality of the findings to different deposition systems.