A new end-to-end, machine-learning-guided workflow developed by the Gagliardi Group is helping bridge the gap between computational materials discovery and real-world application.
Working through the Center for Advanced Materials for Environmental Solutions (CAMES), the team used the approach to design and help synthesize two new zinc-based metal-organic frameworks, UCHI-1 and UCHI-2, for methane separation. The materials demonstrate high-performing gas adsorption and separation while offering a more efficient path from computational design to experimental validation and potential industrial use.
Led by postdoctoral scholar and first author Andrea Darù, the work brings together data mining, machine learning, materials design, synthesis, and experimental testing in a single discovery cycle—helping overcome a key challenge in materials research: promising computational discoveries often never make it to the laboratory or beyond.
Read the full story and learn more about the research in the Journal of the American Chemical Society: Machine learning smooths the road from idea to real-world climate impact