In an article in the Journal of Physical Chemistry Letters, researchers at Virginia Tech have reported a novel approach that should "significantly accelerate materials discovery". Using machine learning techniques, they have developed a unique model that permits the fast and accurate prediction of novel alloy materials for catalysis.

Two Virginia Tech chemical engineering faculty members, Luke Achenie and Hongliang Xin, along with Xianfeng Ma and Zheng Li from Xin's research group, authored the article. "This is the first example of learning from data in catalysis," said Xin. "We anticipate that this new approach will have a huge impact in future materials design."

Ninety percent of industrially important chemicals are made using catalysts. It is a major field in applied science; hence the importance of the new approach by the Virginia Tech chemical engineering members.

"This is the first example of learning from data in catalysis. We anticipate that this new approach will have a huge impact in future materials design."Hongliang Xin, Virginia Tech

The mixing of two or more metals with very precise atomic structures and compositions "has shown great promise for catalyzing many chemical and electrochemical reactions," said Xin. Developing such alloys with conventional techniques is far from simple, however. "The process is very time-consuming and costly to search for highly optimized alloys," said Achenie.

That is why Achenie and Xin decided to use existing data on catalytic metal alloys to train computer algorithms to make predictions for interesting new materials, a field called machine learning. This approach captures complex, nonlinear interactions of molecules on metal surfaces through artificial neural networks, thus allowing "large scale exploration alloy materials space", according to their article.

The researchers specifically concentrated on the electrochemical reduction of carbon dioxide on metal electrodes "because of the current interest in this process for sustainable production of fuels and value added chemicals," Xin explained. Conversion of carbon dioxide to something useful could dramatically reduce its emission into the atmosphere and thus help alleviate global warming.

Using their model and design approach, they have were able to identify a few promising copper multi-metallics with high energy conversion efficiency and selectivity for converting carbon dioxide to ethylene, an extremely useful chemical in industry for making plastics. "This study opens a new way for designing metal-based catalysts with complexities, for example, geometry and composition, promoters and poisons, defects, and nano-effects," concluded Xin.

This story is adapted from material from Virginia Tech, with editorial changes made by Materials Today. The views expressed in this article do not necessarily represent those of Elsevier. Link to original source.