Engineering topics
Publications and source records attributed to Kavalsky, Lance.
We propose a multiobjective scoring scheme toward closed-loop computational discovery of electrocatalysts. To demonstrate, we autonomously search over single-atom alloys for active, stable, and cost-effective systems for nitrogen reduction.
Carbon-supported Ni–Mo composites catalyze alkaline hydrogen evolution and oxidation with negligible onset overpotential. Electrolyzers using Ni–Mo cathodes perform comparably to Pt–Ru cathodes, but oxidative instability limits fuel cell operation.
A combination of task automation, calculation runtime improvements, machine learning surrogatization, and sequential learning-guided candidate selection within a closed-loop computational workflow can accelerate materials discovery by up to 20×.