K-State engineer leading machine learning study to make chemical synthesis, advanced material discoveries more efficient

By Grant Guggisberg

Carl R. Ice College of Engineering

 

A Kansas State University engineer is leading a study aiming to harness the power of machine learning to revolutionize how chemical synthesis and composite material discoveries are made.

Davood B. Pourkargar, assistant professor in the Tim Taylor Department of Chemical Engineering, received more than $578,000 in funding from the National Science Foundation to lead the three-year project, which is focused on making the chemical synthesis process more efficient for use in a variety of renewable energy applications.

"Traditionally, scientists have relied on one-variable-at-a-time experimentation, which is time-consuming and often overlooks the complex interactions among various components and processing conditions," Pourkargar said. "In contrast, this project introduces an innovative approach called adaptive design of experiments, which allows for the simultaneous evaluation and optimization of multiple variables in dynamic processes. This method can significantly enhance the quality and efficiency of chemical synthesis, leading to better product properties while saving time and resources in the laboratory."

Working alongside Pourkargar are Bin Liu, William H. Honstead Professor in Chemical Engineering and associate professor at K-State, and Chuancheng Duan, a former member of the K-State faculty and current associate professor of chemical engineering at the University of Utah.

The study is focused on enhancing the design and production of perovskite oxides, which have unique properties for energy use and are often used in energy-related applications such as fuel cells, electrolyzers, catalysis and electrochemical reactors.

While the project specifically targets renewable energy applications, Pourkargar said the methods developed through this machine learning study will have an even broader reach.

"Beyond energy, the developed framework will provide a systematic method for optimizing manufacturing processes across industries such as chemicals, polymers, pharmaceuticals, agrochemicals and biological probes," he said. "The anticipated outcomes will accelerate innovation, reduce costs and improve product quality while fostering economic growth.

Pourkargar said industry partners in energy, advanced materials and manufacturing — including chemical, aerospace, automotive and defense sectors — can leverage the technology to enhance their efficiency and competitiveness.

###