Advancing Atom Probe Tomography: Multiscale Modelling and Simulation

Atom Probe Tomography (APT) is a powerful technique for three-dimensional compositional imaging at near-atomic resolution. While extensively applied to metals and alloys, its extension to biological systems remains challenging due to difficulties in sample preparation, the insulating nature of biomaterials, and the limited understanding of laser-assisted field evaporation mechanisms. This thesis addresses these challenges through a multiscale computational framework developed within the European MIMOSA project, which aims to establish APT as a novel platform for four-dimensional microscopy of biological materials by combining structural and chemical information at the nanoscale. The first part of the work focuses on the sol–gel encapsulation process used to embed proteins in silica matrices suitable for APT analysis. A reactive coarse-grained molecular dynamics model based on the Martini 3 force field was developed to simulate silica polymerization under physiologically relevant conditions, enabling access to timescales inaccessible to atomistic simulations. Complementary all-atom molecular dynamics simulations investigated the interaction of silica oligomers with proteins of diverse structural classes, demonstrating the overall biocompatibility of the embedding process and providing molecular-level insight into protein–silica interactions, including the roles of hydrogen bonding, electrostatic interactions, and local structural rearrangements. The second part addresses the mechanisms governing laser-assisted field evaporation in silica-based systems. Using Density Functional Theory (DFT) and Time-Dependent DFT, the effects of structural defects, contaminants, strong electric fields, and terahertz (THz) irradiation were investigated. The results reveal that defects significantly enhance optical absorption in the UV and deep-UV spectral regions, providing a microscopic explanation for experimentally observed improvements in APT performance. Furthermore, first-principles simulations identified threshold THz field strengths for ion evaporation and clarified the role of electronic dynamics in non-thermal evaporation processes relevant to insulating and biological materials. As an additional application of molecular simulation methods, atomistic molecular dynamics studies of the Interleukin-22 receptor (IL-22Rα) were performed to investigate cholesterol-dependent signaling mechanisms. The simulations identified specific cholesterol-sensitive interactions within the transmembrane domain and provided a mechanistic framework linking membrane composition, local aromatic interactions, and receptor regulation. Overall, this thesis demonstrates how multiscale computational modelling can bridge atomistic, mesoscale, and electronic-structure descriptions to support the development of next-generation APT methodologies for biological applications. The results provide both methodological advances and fundamental physical insight, contributing to the long-term goal of achieving chemically resolved, four-dimensional microscopy of biomolecular systems.