Show similarFrom chemRxiv
Discovering high entropy alloy electrocatalysts in vast composition spaces with multi-objective optimization
High entropy alloys (HEAs) are a highly promising class of materials for electrocatalysis, as their unique active site distributions break the scaling relations that limit the activity of conventional transition metal catalysts. Existing Bayesian optimization (BO) based virtual screening approaches focus on catalytic activity as sole objective and correspondingly tend to identify promising materials that are unlikely to be entropically stabilized. Here, we overcome this limitation with a multi-objective BO framework for HEAs that simultaneously targets activity, cost-effectiveness and entropic stabilization. With a diversity-guided batch selection further boosting the data efficiency, the framework readily identifies numerous promising candidates for the oxygen reduction reaction that strike the balance between all three objectives in hitherto unchartered HEA design spaces comprising up to ten elements.
Simplify
Published on November 17, 2023
Copy BibTeX 
Loading...
This is an AI-generated summary
Key points
Created a framework to find good, cheap, stable alloy catalysts.
Used machine learning to predict alloy performance.
Found alloys for oxygen reactions that perform well and cost less.
Our approach also helps explore larger alloy combinations.
We used different data sets, models, and tests to balance effectiveness and efficiency.
Discovering high entropy alloy electrocatalysts in vast composition spaces with multi-objective optimization
Page 1
We've created a multi-objective Bayesian optimization framework for high entropy alloys (HEAs) that targets activity, cost-effectiveness, and entropic stabilization. Our framework, further enhanced by a diversity-guided batch selection, successfully identifies many potential candidates for oxygen reduction reactions in unexplored HEA design spaces containing up to ten elements.
Multi-Objective Bayesian Optimization Framework
Page 2
Our multi-objective Bayesian optimization framework operates on evaluated sample batches to strengthen the Pareto front, successfully discovering high entropy alloy (HEA) electrocatalysts for the oxygen reduction reaction (ORR) with diversified requirements in a composition space of up to 10 elements.