Artificial Intelligence Multiscale Modeling for Highly Ordered Pt@PtCo Core-Shell Catalysts as Durable Electro-Catalysts in the PEM Fuel Cell

Development of highly active electrocatalysts that are cost competitive takes the center stage in research fields for next-generation electrochemical energy conversion and storage systems like fuel cell and metal-air battery. Regarding the systems for commercialization, there are various challengeable issues, which should overcome sluggish kinetics and stability on the electrocatalyst for the aimed reactions such as oxygen reduction reaction (ORR) occurred in the high over-potential and harsh conditions. Although there was huge technological advancement as minimizing the usage of a noble metal or developing non-precious owing to many attempts to reach the purpose in the past decade, it still needs further improvement in order to gain acceptance in the market. To find new material or optimize it, the efficient way would be initialized from the knowledge of the underlying physical and chemical mechanisms in a material system. For this reason, therefore, fundamental studies using quantum mechanics (QM) along with experimental confirmation have been commonly conducted to explain the activity and stability of oxygen reactions. These approaches are very successful and provide fundamental insight to scientists and researchers. Nevertheless, calculative simulations are still the restrictions to obtaining trustable data and predicting correct consequences, especially on large-scale modeling. This talk will present recent efforts for accurate modeling with experimental data. We predicted and validated the ORR activity and stability for PtCo alloy and Pt@Co core-shell structure in density functional theory (DFT). The force field, which is an efficient way of describing the dynamics with large-scale simulations, was developed to simulate large-scale models via machine-learning Force Fields. This synergetic approach using both computational models and experiments with physicochemical analyses would accelerate to find new materials and optimize catalyst performance. This work was supported by the NRF of Korea Grant and by the Technology Innovation Program of MOTIE [NRF- 2022R1A2C2093090 & 20019175]