Machine-learning-assisted and real-time-feedback-controlled growth of InAs/GaAs quantum dots

Chao Shen; Wenkang Zhan; Kaiyao Xin; Manyang Li; Zhenyu Sun; Hui Cong; Chi Xu; Jian Tang; Zhaofeng Wu; Bo Xu; Zhongming Wei; Chunlai Xue; Chao Zhao; Zhanguo Wang

doi:10.1038/s41467-024-47087-w

Machine-learning-assisted and real-time-feedback-controlled growth of InAs/GaAs quantum dots

Nat Commun. 2024 Mar 29;15(1):2724. doi: 10.1038/s41467-024-47087-w.

Authors

Chao Shen^#^{1

2

3}, Wenkang Zhan^#^{1

2}, Kaiyao Xin^{2

4}, Manyang Li^{1

2}, Zhenyu Sun^{1

2}, Hui Cong^{2

5}, Chi Xu^{2

5}, Jian Tang⁶, Zhaofeng Wu³, Bo Xu^{1

2}, Zhongming Wei^{2

4}, Chunlai Xue^{2

5}, Chao Zhao^{7

8}, Zhanguo Wang^{1

2}

Affiliations

¹ Laboratory of Solid State Optoelectronics Information Technology, Institute of Semiconductors, Chinese Academy of Sciences, Beijing, 100083, China.
² College of Materials Science and Opto-Electronic Technology, University of Chinese Academy of Science, Beijing, 101804, China.
³ School of Physics Science and Technology, Xinjiang University, Urumqi, Xinjiang, 830046, China.
⁴ State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences, Beijing, 100083, China.
⁵ Key Laboratory of Optoelectronic Materials and Devices, Institute of Semiconductors, Chinese Academy of Sciences, Beijing, 100083, China.
⁶ School of Physical and Electronic Engineering, Yancheng Teachers University, Yancheng, 224002, China.
⁷ Laboratory of Solid State Optoelectronics Information Technology, Institute of Semiconductors, Chinese Academy of Sciences, Beijing, 100083, China. zhaochao@semi.ac.cn.
⁸ College of Materials Science and Opto-Electronic Technology, University of Chinese Academy of Science, Beijing, 101804, China. zhaochao@semi.ac.cn.

^# Contributed equally.

Abstract

The applications of self-assembled InAs/GaAs quantum dots (QDs) for lasers and single photon sources strongly rely on their density and quality. Establishing the process parameters in molecular beam epitaxy (MBE) for a specific density of QDs is a multidimensional optimization challenge, usually addressed through time-consuming and iterative trial-and-error. Here, we report a real-time feedback control method to realize the growth of QDs with arbitrary density, which is fully automated and intelligent. We develop a machine learning (ML) model named 3D ResNet 50 trained using reflection high-energy electron diffraction (RHEED) videos as input instead of static images and providing real-time feedback on surface morphologies for process control. As a result, we demonstrate that ML from previous growth could predict the post-growth density of QDs, by successfully tuning the QD densities in near-real time from 1.5 × 10¹⁰cm^-2 down to 3.8 × 10⁸cm^-2 or up to 1.4 × 10¹¹cm^-2. Compared to traditional methods, our approach can dramatically expedite the optimization process and improve the reproducibility of MBE. The concepts and methodologies proved feasible in this work are promising to be applied to a variety of material growth processes, which will revolutionize semiconductor manufacturing for optoelectronic and microelectronic industries.

Abstract

Grants and funding