Deep Reinforcement Learning for Identifying the Global Minima of Platinum Nanoclusters

Abstract Prediction of stable nanocluster structures remains a significant challenge in materials and nanocluster research due to the complex nature of potential energy surfaces (PES). To overcome this complexity, a novel deep reinforcement learning (DRL) framework was employed to efficiently scan the PES and identify the global minimum of the Pt13 nanocluster alongside other low-energy configurations. The DRL agent iteratively learns to generate energetically favorable configurations by adjusting atomic positions based on feedback from a reward function designed to promote structural stability and discourage unrealistic geometries, such as overlapping or dissociating atoms. Starting from randomized initial structures, the model successfully identifies the most stable configuration of Pt₁₃ with icosahedral (Ih) symmetry, and the framework reveals 25 distinct low-energy isomers. The successful identification of a stable structure verifies the effectiveness of the DRL framework. Additionally, Density Functional Theory (DFT) calculations confirm the stability of the Pt13 nanocluster by finding the cohesive energy. The negative cohesive energy confirms the stability, and thermodynamic stability was also assessed at 300 K. The charge, electron localization function, electron density, d-band center, and total density of states indicate that Pt13 nanoclusters exhibit the ideal electronic fingerprint of a highly active nano-catalyst. To further check the DRL framework's adaptability, we performed experiments on Pt10 and Pt18. This study highlights the efficacy of DRL in navigating complex energy landscapes, predicting stable nanocluster configurations, and providing a robust methodology for optimizing nanoclusters.