Autonomous Materials Cultivation: The Self-Driving Lab Innovation

Scientists traditionally spend months experimenting with temperature, composition, and timing to create thin metal films for various technologies. However, researchers at the University of Chicago Pritzker School of Molecular Engineering have introduced a revolutionary “self-driving” lab system that utilizes robotics and AI to make real-time decisions without human oversight.

The system, as described by Yuanlong Bill Zheng, an undergraduate and now a Ph.D. student at UChicago PME, automates the entire experimental process, from conducting experiments to analyzing results and using machine learning to guide subsequent attempts.

Published in npj Computational Materials, the work by Asst. Prof. Shuolong Yang and the team marks a significant step towards a futuristic mode of manufacturing using autonomous systems.

Enhancing the Physical Vapor Deposition Process

The self-driving system centers around the physical vapor deposition (PVD) process, where materials are vaporized and condensed into thin layers on surfaces. This process is highly sensitive to various parameters, making outcome prediction challenging.

The team, in collaboration with UChicago’s Computer Science Department, developed a machine learning algorithm to predict the parameters required for specific thin films. This algorithm guides the system through a series of experiments until the desired outcome is achieved.

To address unpredictable factors, the system starts each experiment with a calibration layer to account for variations, enhancing the model’s accuracy and reproducibility.

Advancing Material Synthesis

Testing the system with silver films, the researchers achieved the desired results in an average of 2.3 attempts, significantly reducing the time and effort required compared to traditional methods. The cost-effective setup, built by undergraduate students, showcases the potential for AI and robotics in material synthesis.

The team aims to expand this approach to complex materials, paving the way for advancements in electronics and quantum devices. As Yang emphasizes, this prototype demonstrates the transformative power of AI and robotics in materials discovery.