Computational-Experimental Synergy for Innovative Optoelectronic Molecule design – CESIOM

This project focuses on developing new methodologies to generate, synthesize and use new red and near infrared-absorbing or emitting molecules for diverse applications, including OLEDs, organic solar cells and for bio-related applications such as photodynamic therapy within the therapeutic window. Combining experimental chemistry, Quantum Mechanics (QM), and Artificial Intelligence, our approach facilitates a broader exploration, surpassing traditional fragment modifications around known scaffolds. De novo generation automates the construction of molecules meeting predefined specifications, extending beyond familiar fragments, potentially unveiling innovative molecules. Optimizing this process involves considering electronic properties and other factors. Developing an efficient objective function is a multidisciplinary task, striking a balance between experimental requirements, computational feasibility, and problem formalization. Evaluating electronic properties, an expensive process requiring QM calculations, is the focus of WP1, which designs a tailored quantum calculation method, favoring Density Functional Theory and Density Functional Tight-Binding for a cost effective yet interpretable score. Furthermore, this method could allow us to optimize the dynamics of the charge transfers. Simultaneously, WP2 focuses on formally describing these problems as constrained combinatorial optimization problems and the implementation of molecular optimization methods combining different fields of artificial intelligence, including combinatorial optimization and machine learning. WP3 handles the synthesis, characterization, and prototyping of promising candidates, and provides important feedback to the others WP1, embodying our consortium's unique and complementary approach, integrating chemical space exploration, multi-level QM calculations, and experimental prototyping.