About this Abstract |
| Meeting |
MS&T25: Materials Science & Technology
|
| Symposium
|
Autonomous Platforms for Designing and Understanding Materials
|
| Presentation Title |
From deposition to degradation of thin films and devices through autonomous experimentation |
| Author(s) |
Davi Febba, Stephen Schaefer, Brooks Tellekamp, William Callahan, Andriy Zakutayev |
| On-Site Speaker (Planned) |
Davi Febba |
| Abstract Scope |
Autonomous experimentation is transforming materials science. By automating repetitive tasks and using artificial-intelligence-driven experiment planners, we can accelerate the materials-discovery pipeline while minimizing human intervention, freeing researchers to focus on higher-level questions. In this presentation, we will summarize recent progress at NREL on (i) autonomous sputtering and molecular beam epitaxy growth of thin films and (ii) long-term degradation studies of electronic devices under extreme environmental condition. These platforms employ genetic algorithms, computer-vision feedback, and multidimensional Bayesian optimization to identify the most informative experiments in real time, maximizing information gain per unit time. Finally, we will discuss the practical challenges of retrofitting existing laboratories for autonomy—including instrument-level automation, workflow orchestration, data-management infrastructure, and subnetworks that link remote servers to command-and-control computers—and outline solutions that have enabled seamless, closed-loop operation of scientific instruments in the context of autonomous experimentation at NREL. |