We’re excited to share our latest work, Predicting and Accelerating Nanomaterials Synthesis Using Machine Learning Featurization, which uses AtomCloud’s AI/ML powered automation to accelerate key steps in the materials synthesis feedback loop, deliver insights faster and earlier, and save time while helping avoid doomed trials. Thanks to our collaborators Yansong Li, Guanyu Zhou, Rehan Younas, and Chris Hinkle (Hinkle Lab) on materials growth and feedback with AtomCloud.

Advanced Materials Bottleneck

The primary bottleneck in advanced materials R&D and scale-up is making high-quality materials. For novel material targets, including those identified by simulation, recipe design and process optimization is a high-dimensional, nonlinear, and empirical problem to solve. Processing times can take many hours and result verification adds additional tools, time, and steps to the workflow. Repeat this over the many trials required for process optimization and the time it takes to reach acceptable synthesis control can be quite long.

With AtomCloud

Using our tailored AI pipelines for characterization data, we show that we can unlock predictive insights on expected material quality and material composition from raw data already captured in the daily synthesis workflow. Parsing this data manually is traditionally time consuming and introduces bias across different samples when done by different operators.

Our automated approach efficiently captures both expected results and uncovers new trends from the original data that are missed even by expert practitioners. From small sets of labeled trials in the early stages of experiment design, we can avoid 60% of doomed trials before they start and predict key materials properties without additional measurement.

We’re just scratching the surface of what’s possible for AI in the world of atoms. By accelerating the way process data is collected, transformed, and used for recipe optimization, we are opening new frontiers in materials synthesis. Contact us to learn more or book a demo today!