There is growing interest in developing mesoporous metallic alloys for electrochemical applications such as catalysts in fuel cells and batteries. As is well known, the chemical compositions of alloys can significantly affect their electrochemical properties. Although tuning the chemical compositions of mesoporous metallic alloys for enhancing the electrochemical activity has been reported, they have mostly been limited to binary components partly because experimental exploration over possible multi-compositional spaces is a time-consuming process. Here, we describe, for the first time, the application of the active learning scheme using Bayesian optimization for the exploratory search of the chemical compositions of mesoporous trimetallic PtPdAu alloys with optimum catalytic activity in the electrocatalytic oxidation of methanol. Unexpectedly, it was found that the PtPdAu alloys yielding the highest catalytic activity contain only a small percentage of Au. These compositions were discovered by performing only 47 experiments, less than 1% of all possible compositions in our experimental design. Our current approach is highly efficient and would be applicable to any system to accelerate the discovery of novel materials.
ASJC Scopus subject areas
- Renewable Energy, Sustainability and the Environment
- Materials Science(all)