A.G. Gudkov1, D.D. Petreev2, R.R. Galimov3, E.A. Tertyshnaya4, V.G. Tikhomirov5, A.E. Sinev6

1,2,3 Bauman Moscow State Technical University (BMSTU) (Moscow, Russia)

4,5,6 Saint Petersburg Electrotechnical University «LETI» (ETU «LETI») (St. Petersburg, Russia)

1 profgudkov@gmail.com

This article is devoted to analysing the physical and technological barriers to modern growth processes and justifying the need to apply machine learning methods, as well as identifying the current limitations of machine learning technologies in this field and determining the prospects for their development.

To comprehensively analyse and systematise the artificial intelligence methods used for predictive control, parameter optimisation and in-situ data analysis in modern growth technologies in instrument engineering.

This paper presents the latest results on the implementation of machine learning methods in growth technologies and provides a classification of neural network architectures and optimisation algorithms relevant to crystal and thin-film growth tasks. Practical case studies of the implementation of machine learning methods for in-situ defect detection and real-time control of process parameters are examined.

The analysis and systematisation of artificial intelligence methods used for preliminary monitoring, parameter optimisation and the analysis of in-situ data in modern growth technologies for instrument engineering makes it possible to overcome the barrier of computational complexity, ensure continuous atomic-level monitoring and implement predictive control.

Gudkov A.G., Petreev D.D., Galimov R.R., Tertyshnaya E.A., Tikhomirov V.G., Sinev A.E. Physical and technological barriers in growth technologies and the prerequisites for the implementation of artificial intelligence // Nanotechnology development and applications-XXI Ce. 2026. V. 18. № 3. P. 22−30. DOI: https://doi.org/10.18127/j22250980-202603-02

1. Hespeler S., Dehghan-Niri E., Juhasz M. et al. Deep learning for in-situ layer quality monitoring during laser-based directed energy deposition (LB-DED) additive manufacturing process. Applied Sciences. 2022. V. 12. № 18. P. 8974. DOI: 10.3390/app12188974.
2. Liang G., Cheng L., Zha J. et al. In-situ quantification of the surface roughness for facile fabrications of atomically smooth thin films. Nano Research. 2022. V. 15. № 2. P. 1654–1659. DOI: 10.1007/s12274-021-3720-5.
3. Fowlie J., Georgescu A. B., Mundet B. et al. Machines for materials and materials for machines: Metal-insulator transitions and artificial intelligence. Frontiers in Physics. 2021. V. 9. P. 725853. DOI: 10.3389/fphy.2021.725853.
4. Podkaminer J. P., Patzner J. J., Davidson B. A., Eom C. B. Real-time and in situ monitoring of sputter deposition with RHEED for atomic layer-controlled growth. APL Materials. 2016. V. 4. № 8. P. 086111. DOI: 10.1063/1.4961503.
5. Choi W., Huang H.-C., Fan S. et al. Selective area heteroepitaxy of p-i-n junction GaP nanopillar arrays on Si (111) by MOCVD. IEEE Journal of Quantum Electronics. 2022. V. 58. № 4. P. 1–6. DOI: 10.1109/JQE.2022.3151971.
6. Tairov Y. M., Tsvetkov V. F. Investigation of growth processes of ingots of silicon carbide single crystals. Journal of Crystal Growth. 1978. V. 43. № 2. P. 209–212. DOI: 10.1016/0022-0248(78)90169-0.
7. Yu W., Zhu C., Tsunooka Y. et al. Geometrical design of a crystal growth system guided by a machine learning algorithm. CrystEngComm. 2021. V. 23. № 14. P. 2695–2702. DOI: 10.1039/D1CE00106J.
8. Yadav P., Rigo O., Arvieu C. et al. In situ monitoring systems of the SLM process: On the need to develop machine learning models for data processing. Crystals. 2020. V. 10. № 6. P. 524. DOI: 10.3390/cryst10060524.
9. Dropka N., Holena M. Application of artificial neural networks in crystal growth of electronic and opto-electronic materials. Crystals. 2020. V. 10. № 8. P. 663. DOI: 10.3390/cryst10080663.
10. Noh Y. K., Noh H. K., Park B. G. et al. Hybrid NH3/N2 molecular beam epitaxy with artificial-intelligence-assisted reflection high-energy electron diffraction analysis. International Conference on Compound Semiconductor Manufacturing Technology. 2021. P. 24–27.