Integrated process-property modeling of YBa₂Cu₃O₇ superconducting film for data and model driven process design

Area of Science:

• Materials Science
• Process Engineering
• Data Science

Background:

• Material properties, cost, and production are determined by process engineering.
• Current process design relies heavily on engineer experience, highlighting the need for advanced modeling.
• Optimizing Yttrium Barium Copper Oxide (YBa$_{2}$Cu$_{3}$O$_{7}$) film fabrication is crucial for coated conductor applications.

Purpose of the Study:

• To develop a data/model-driven strategy for optimizing YBa$_{2}$Cu$_{3}$O$_{7}$ film fabrication.
• To integrate process and property modeling for enhanced material design.
• To enable accurate prediction of critical current density in YBa$_{2}$Cu$_{3}$O$_{7}$ films.

Main Methods:

• Utilized Gaussian process regression augmented with transfer learning and physics knowledge.
• Incorporated non-numerical factors like chamber design and substrate material into transfer learning.
• Applied physics-aided techniques to extend models to varying magnetic fields.

Main Results:

• Successfully modeled substrate temperature dependence of critical current density using limited data.
• Accurately predicted magnetic field dependence of critical current density across different pulsed laser deposition systems.
• Demonstrated an integrated approach for process and property modeling in YBa$_{2}$Cu$_{3}$O$_{7}$ film fabrication.

Conclusions:

• The developed integrated modeling strategy enables data/model-driven process design for YBa$_{2}$Cu$_{3}$O$_{7}$ films.
• This approach offers new opportunities for optimizing material properties and production.
• The findings are directly applicable to the fabrication of coated conductors.