This project aims to utilize predictive and generative Artificial Intelligence (AI) techniques to investigate metallic nanoclusters as promising candidates for catalysis. The primary goal is to enhance decision-making and optimize chemical processes through Machine Learning (ML) models, especially by Bayesian optimization. Our focus is on developing protocols that promote the digitalization of experiments, simulations, and material synthesis processes, with an emphasis on studying metallic nanoclusters, particularly those composed of transition metals (TMs). These atomic structures exhibit exceptional properties in terms of selectivity and chemical reactivity, though their study is highly complex due to morphological and compositional variability, as well as their high capacity for interaction with different chemical environments. The project seeks to simplify complex protocols that integrate computational and experimental data, transforming them into accessible and efficient digital workflows for industrial application. Our goal is to develop computational tools that facilitate the access and communication of high-performance quantum mechanical calculations, such as those based on Density Functional Theory (DFT), integrating them with experimental data and protocols through automation and digitalization processes. This will ensure greater predictability and reproducibility, while allowing for efficient planning and management of closed-loop workflows.
