Machine Learning Quantum Chemistry for Reactivity
Understanding how quantum properties influence chemical reactivity is a key challenge for designing next-generation catalysts and energy conversion materials. Traditional first-principles approaches, such as density functional theory (DFT), provide rigorous insights but are computationally demanding, especially when exploring broad chemical and structural parameter spaces. Machine learning (ML) offers a new pathway. Once trained, ML models can rapidly predict key quantum descriptors (e.g., spin polarization, Berry curvature) and correlate them with catalytic activity. In our earlier work, we developed first-principles-based Green’s function methods to study spin-dependent reactivity, such as the oxygen evolution reaction in chiral crystals [1–3].
Building on this foundation, the project now aims to integrate ML with quantum-chemistry-informed models to uncover hidden and nonlinear correlations between quantum descriptors and reaction kinetics. By identifying such relationships, the project seeks to establish new design principles for catalysts that explicitly harness quantum effects.
We are seeking a highly qualified and motivated PhD candidate with a background in machine learning, theoretical chemistry/materials science, or related fields. This project offers the opportunity to work collaboratively across leading research institutes, including TU Dresden, the Max Planck Institute for Chemical Physics of Solids (MPI-CPfS) in Dresden, and the Max Planck Institute of Microstructure Physics (MPI-Halle) in Halle.
