Multiscale modeling of Xe NMR biosensors

The dissertation examines a physical phenomenon called nuclear magnetic resonance (NMR), in which an atomic nucleus, behaving like a tiny magnet, interacts with an external magnetic field. The strength of this interaction depends strongly on the microscopic environment of the atom or molecule carrying the nucleus. In NMR spectroscopy, this dependence is exploited to study substances and materials – for example, the magnetic resonance imaging technique familiar from hospitals is based on the NMR phenomenon.

The dissertation focuses particularly on the NMR of the noble gas xenon and its use in xenon biosensors, which are small molecular-level "devices" employed to investigate, for example, chemical processes at extremely low concentrations, such as the metabolism of cancer cells. A large number of new computational and theoretical methods were developed in this work for studying xenon biosensors and, more broadly, in the field of NMR spectroscopy. One major achievement was the integration of three existing computational approaches into a multiscale modeling framework, enabling more accurate comparisons between experiments and simulations. When theoretical predictions align with experimental measurements, a direct link can be established between molecular-level properties and the observable properties of materials. This interplay between theory and experiment forms the foundation for the development of new technologies and medical applications.