"Chemistry is a game that electrons play."
J. J. W. McDouall 

Tools

Our research group employs a broad range of state-of-the-art computational techniques to investigate materials at the electronic level. Chief among these are first-principles quantum-mechanical methods like Density Functional Theory (DFT) for ground-state properties and Time-Dependent DFT (TD-DFT) for excited-state phenomena. DFT is widely regarded as the method of choice for efficiently calculating ground-state electronic structures with reliable accuracy. However, for systems where standard DFT approximations are insufficient, we turn to beyond-DFT approaches grounded in many-body perturbation theory. In particular, the GW approximation and the GW-Bethe–Salpeter Equation (GW-BSE) framework allow us to capture electron-electron interactions and optical excitations with higher fidelity. By leveraging this comprehensive toolkit, we gain a deep fundamental understanding of the electronic structure of materials and uncover how atomic-scale structure influences macroscopic properties. These first-principles calculations provide critical insight into structure–property relationships, revealing how variations in composition, bonding, or defects can alter material properties.

We integrate our electronic structure methods into high-throughput screening workflows to accelerate materials discovery. This involves the automated execution of hundreds or thousands of DFT/TD-DFT calculations across broad chemical and structural spaces, allowing us to rapidly predict material properties and identify promising candidates for specific applications. Such high-throughput computational screening enables efficient exploration of structure–property relationships on a large scale. In parallel, we harness artificial intelligence (AI) and machine learning (ML) techniques alongside DFT to further speed up discovery through data-driven approaches. For example, we train ML models on data from DFT simulations so that they can quickly predict properties of new materials. This synergy between physics-based simulations and AI-driven models significantly accelerates the design and discovery of novel materials. By combining rigorous first-principles calculations with machine learning prediction, our group’s workflow can rapidly screen material libraries and guide experimental efforts toward the most promising material systems, all while maintaining a high level of accuracy and scientific insight.

See also Materials and Applications