Smart design of crystal growth furnaces and processes

• Applicants: Dr. Natascha Dropka; Dr. David Linke
• Subject Area: Thermodynamics and Kinetics as well as Properties of Phases and Microstructure of Materials ; Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics ; Metallurgical, Thermal and Thermomechanical Treatment of Materials
• Term: from 2021 to 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 467401796

• Final Report Year: 2025

Final Report Abstract

The project-funded research in the co-applicant’s group at LIKAT Rostock was pursued in two directions. The main direction was supporting the research of the applicant’s group at the application of machine learning, statistical, and optimization methods. They were applied to the modelling of Czochralski crystal growth of Ge, Si and GaAs, as well as of vertical gradient freeze growth and floating zone growth. In the preliminary step, the machine learning methods were tuned for small data applications using existing CFD data from other crystal growth techniques, such as vertical gradient freeze and floating zone. As a complementary research direction, research into some machine learning, statistical, and optimization methods has been performed, in particular into some applications of artificial neural networks and classification methods, and into the landscape-analysis aspect of evolutionary optimization.

Publications

• Development and Optimization of VGF-GaAs Crystal Growth Process Using Data Mining and Machine Learning Techniques. Crystals, 11(10), 1218.
  Dropka, Natasha; Böttcher, Klaus & Holena, Martin
  
• Smart Design of Cz-Ge Crystal Growth Furnace and Process. Crystals, 12(12), 1764.
  Dropka, Natasha; Tang, Xia; Chappa, Gagan Kumar & Holena, Martin
  
• Text-to-ontology mapping via natural language processing models. In ITAT, pages 28–34, 2022
  U. Yorsh, A.S. Behr, N. Kockmann & M. Holeňa
  
• Using artificial neural networks to determine ontologies most relevant to scientific texts. In ITAT, pages 44–54, 2022
  L. Korel, A.S. Behr, N. Kockmann & M. Holeňa
  
• Decision Tree-Supported Analysis of Gallium Arsenide Growth Using the LEC Method. Crystals, 13(12), 1659.
  Tang, Xia; Chappa, Gagan Kumar; Vieira, Lucas; Holena, Martin & Dropka, Natasha
  
• Development of the VGF Crystal Growth Recipe: Intelligent Solutions of Ill‐Posed Inverse Problems using Images and Numerical Data. Crystal Research and Technology, 58(11).
  Dropka, Natasha; Holena, Martin; Thieme, Cornelia & Chou, Ta‐Shun
  
• Text-to-Ontology Mapping via Natural Language Processing with Application to Search for Relevant Ontologies in Catalysis. Computers, 12(1), 14.
  Korel, Lukáš; Yorsh, Uladzislau; Behr, Alexander S.; Kockmann, Norbert & Holeňa, Martin
  
• Using paraphrasers to detect duplicities in ontologies. In Proceedings of the 15th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management – Volume 2, pages 40–49, 2023
  L. Korel, A.S. Behr, N. Kockmann & M. Holeňa
  
• Data‐Driven Cz–Si Scale‐Up under Conditions of Partial Similarity. Crystal Research and Technology, 59(6).
  Dropka, Natasha; Böttcher, Klaus; Chappa, Gagan Kumar & Holena, Martin
  
• Suitability of modern neural networks for active and transfer learning in surrogate-assisted black-box optimization. In IAL´24, pages 47–67, 2024
  M. Holeňa & J. Koza
  
• Unraveling conditions for W-shaped interface and undercooled melts in Cz-Si growth: A smart approach. Journal of Crystal Growth, 648, 127897.
  Dropka, Natasha; Petkovic, Milena; Böttcher, Klaus & Holena, Martin
  
• An Analysis of Elusive Relationships in Floating Zone Growth Using Data Mining Techniques. Advanced Theory and Simulations, 8(5).
  Vieira, Lucas; Menzel, Robert; Holena, Martin & Dropka, Natasha
  
• Landscape Analysis for Surrogate Models in the Evolutionary Black-Box Context. Evolutionary Computation, 33(2), 249-277.
  Pitra, Zbyněk; Koza, Jan; Tumpach, Jiří & Holeňa, Martin