Molecules in realistic environments

This project focuses on modeling molecules in realistic environments subjected to various perturbations. Examples of such systems include molecules in condensed phases or on interfaces involving heavy elements (i.e., from the bottom of the periodic table), studied through their interactions with other molecules and/or external electromagnetic fields. Their modeling has become an essential step towards designing solutions for the sustainable future of modern society. Examples include applications in medicine (radiopharmaceuticals, nuclear imaging agents), various industries (catalysts, photonic devices, single molecular magnets), and energy sources (nuclear fuels), to name but a few. Computational studies of such systems are also invaluable for basic research. Analyzing interactions between molecules and between molecules and external fields enhances our understanding of the electronic structure and molecular properties.

Such computer simulations are not easy. At the quantum chemistry level, the appropriate method should be able to describe relativistic effects (scalar and spin-orbit) and electron correlation; it should also account for the presence of an environment that may significantly affect the geometry, electronic structure, and properties of the molecule of interest. However, the cost of accurate methods often prevents their application to studied systems. An interesting way around this is to divide a large system into small parts and treat each separately with the best cost-efficient model. This is the key point of subsystem-based methods in quantum chemistry. An additional argument for adopting partitioning-based approaches is that the systems of interest are often composed of an easily distinguishable active subsystem and its environment. The Frozen Density Embedding (FDE) - a central method developed in this project - is an example of such a technique. In principle, FDE employs partitioning the system’s electron density, and the effect of all other subsystems on the properties of the active one is accounted for through the embedding potential. The key contributions of this project include the adaptations and testing of FDE in the relativistic framework and the optimization of FDE for demanding applications (such as the minimization of errors due to an insufficient description of coupling between the subsystems).


Selected publications related to this project:


Project funded by the National Science Center, grant number 2016/23/D/ST4/03217 and 2020/38/E/ST4/00614