Date of Award

8-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Materials Science and Engineering

Committee Chair/Advisor

Jianhua Tong

Committee Member

Feng Luo

Committee Member

Fei Peng

Committee Member

Luiz Jacobsohn

Committee Member

Dilpuneet Aidhy

Abstract

Environmental pollution and rapid energy consumption have become common problems in global development and will continue to grow with the world population. PCFCs use proton-conducting ceramics as electrolytes, with low activation energy and high ionic conductivity at intermediate temperatures, enabling them to operate at intermediate-temperature conditions, which can effectively solve the problems of poor stability and high cost of exotic materials of traditional solid oxide fuel cells. However, as the operating temperature decreases, the electrocatalytic activity of the cathode decreases significantly, seriously affecting PCFC’s performance. Therefore, developing high-performance cathode material suitable for working under intermediate-temperature conditions has become the key to commercializing PCFCs.

This Ph.D. dissertation combines the synthesis of nanocomposites with ML-assisted material discovery methods to explore new cathode materials. In chapter two, the perovskite nanocomposite cathode of Pr0.3(Ba0.5Sr0.5)0.7Co0.8Fe0.2O3−δ was designed, fabricated, and studied for its structure and electrochemical properties. Chapter three focuses on predicting new perovskite oxide materials with high hydrated proton concentration as a potential PCFC cathode material or a component of high-performance nanocomposite cathodes. The trained ML model predicted the HPC values of two unknown perovskite materials for verification. The predicted value is very close to the true value. Chapter four focuses on predicting perovskite oxide materials with low area-specific resistance to serve as another component of nanocomposite cathodes. With the help of the well-trained model, ASR values of unknown perovskite materials were predicted for verification. The overall results show that the model has an important reference value in trend judgment and preliminary material screening.

Available for download on Monday, August 31, 2026

Share

COinS