Hybrid Functionals for Hybrid Materials

• Applicant: Professorin Dr. Hilke Bahmann
• Subject Area: Theoretical Chemistry: Electronic Structure, Dynamics, Simulation
• Term: since 2019
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 418140043

Density functional theory (DFT) is the major tool for quantum chemical calculations, extending to industrial applications in catalysis and material design. The central quantity in this method is the approximated exchange-correlation functional, which is often specifically tailored to a given system or property via empirical parameters. However, since these parameters are global and system-wide, this procedure is of limited use, especially for very heterogeneous systems, such as interfaces and molecules on surfaces. Another challenge in DFT are strong correlation effects, that are, e.g. necessary to correctly describe bond dissociation in molecules, bonds involving transition metals or heavy elements, as well as band gaps in semi-conductors or nanostructures.The aim of this project is to improve the predictive power of DFT calculations for hybrid materials. I will develop new schemes that adjust to the local electronic environment and novel nonlocal functionals for strong correlation. To ensure widespread applicability, I combine simple concepts with efficient algorithms for the implementation into an established quantum chemistry program. New methods are thus readily at hand for application to systems of chemical and physical interest.Addressing strong correlation, novel functionals for this regime that are based on the spherically-averaged charge density will be explored. This nonlocal quantity contains information on the electron density at different points in space and represents a powerful new ingredient for the construction of exchange-correlation functionals. Refined functional forms will be addressed along with their self-consistent implementation.Two complementary strategies, belonging to the larger category of hybrid exchange functionals, are pursued in this project. Both rely on a flexible combination of exact and approximate DFT exchange energies. The first strategy are local hybrid functionals, which become increasingly popular for molecular systems, and perform excellently for ground and excited states. Moreover, band offsets and local electronic states at interfaces are directly accessible with local hybrid functionals. They will be implemented for periodic boundary conditions to enable their long overdue assessment for solids and surfaces. The second strategy is an extension of the well-established long-range corrected exchange functionals to local range-separation. Focusing on ground state properties of finite systems first, suitable mathematical forms for the range-separation function will be derived from physical constraints. Ultimately, locally range-separated functionals will be applied to charge-transfer excitations in systems relevant to catalysis.Overall, the methods developed in this project are expected to improve the description of strongly heterogeneous systems like molecules adsorbed on surfaces or organic/inorganic interfaces and describe strong correlation effects in molecules and solids on the same footing.

• DFG Programme: Independent Junior Research Groups