Historically, materials discovery has relied on trial-and-error, with researchers gradually expanding compositional and processing steps through manual, iterative testing. In recent years, this paradigm has shifted with the rise of data-driven materials discovery, where computational tools like density functional theory (DFT) help refine experimental search spaces, drastically reducing the need for exhaustive trial-and-error. These computational techniques are now being enhanced even further through the integration of artificial intelligence (AI) and machine learning (ML) methods.

As data-driven models become faster and more sophisticated, however, the traditional pace of experimentation has become a bottleneck that slows down the otherwise accelerated materials discovery pipeline. Conventional methods simply cannot keep up with the speed at which computational predictions are now being generated.

Our lab looks at how we can analyze data from high-throughput experimentation in an automated way, allowing us to analyze samples and directly make decisions on further experiments without direct researcher intervention.

Associated researchers:

                              Harrison (Hyun Sang) Park                                                                               Alex deJong