Sintering kinetics and microstructure analysis of composite mixed ionic and electronic conducting electrodes

Mixed ionic-electronic conducting (MIEC) cathode materials are promising for application in intermediate-temperature solid oxide fuel cell due to their potentially low polarization resistance for oxygen reduction reaction. Sintering is a key step in fabricating MIEC composite cathode to form interconnected networks for gas transport and electron/ion conduction. However, the sintering process of MIEC composite electrodes has rarely been studied and thus not well understood. In this study, an enhanced kinetic Monte Carlo (KMC) model is proposed to study the microstructure evolution of the composite MIEC electrodes. The key microstructural properties such as average particle size, pore size, porosity, tortuosity, effective surface area and effective three-phase boundary (TPB) length are comprehensively extracted from the reconstructed electrode by a series of the estimation algorithm. A backpropagation artificial neural network (ANN) is established for the quick prediction of micro properties during the sintering process. Besides, a global sensitivity analysis approach based on the elementary effects method is applied to assess the sensitivity degree of structural and operating parameters while Kolmogorov-Smirnov test is conducted to ensure the reliability of the sampling strategy (P-value >> .05). By considering the time evolution within the ANN surrogate model, this study firstly explores the coupling parametric effects of the KMC model at different sintering stages. The initial sintering stage (KMC steps <1000) is found to be the dominant section throughout the sintering procedure due to higher initial system free energy. Tortuosity and porosity show the most significant influence on the formation of effective surface and TPB reaction sites. Large tortuosity and small porosity favor the formation of effective surface reaction area but adversely hinder the gas transport at the same time. Free energy and sintering temperature are the determining factors in the sintering process. Even though grain growth and pore migration frequency show almost the same importance, their influence on the different output factors is quite different. This study establishes a qualitative framework to gain an essential perception about how the sintering process influences the micro properties of the electrode thereby efficiently guiding the sintering control and microstructure design. Highlights A kinetic Monte Carlo model is established considering sintering kinetics. A BP-ANN method is proposed for quick prediction of micro-characteristics An elementary effects approach is employed to perform global sensitivity analysis. Decisive role of initial powder fabrication on the sintering process is emphasized Porosity and tortuosity play the most significant effects than other structural factors.

Automated summary

This study establishes an enhanced kinetic Monte Carlo model to study the microstructure evolution of mixed ionic-electronic conducting (MIEC) composite cathode materials. The backpropagation artificial neural network (BP-ANN) method is proposed for quick prediction of micro properties, and a global sensitivity analysis approach based on the elementary effects method is employed to assess the sensitivity degree of structural and operating parameters.