This computational-chemistry project concerns subjects that are of relevance for environmental issues and for energy-resource utilization applications. For example, we are studying chemical vapour deposition (CVD) and the atomic layer deposition (ALD) processes, which are very important in the development of more efficient and energy-effective materials. Currently, the intricate molecular reaction mechanisms that underlie the epitaxial growth on these materials are being studied. We are also developing modelling techniques for CVD growth, that could replace in-vitro experiments by in-silico simulations, thus avoiding the use of sometimes hazardous chemicals. The aim is to depict a scheme that is based on the results for basic molecule-molecule or molecule-surface interactions but yet directly and simply tells us the consequences for the macroscopic thermodynamics, phase transitions, phase diagram as well as dynamical quantities such as time-dependent stabilities, reaction rates or sensor responses. To meet this end, we perform molecular quantum-chemical computations followed by thermodynamic and kinetic modelling, and recently also computational fluid dynamics. Other research subjects relate to the debate on the rising carbon dioxide levels in the atmosphere: CO2 to CH3OH conversion, chemical sensors, natural gas hydrates and CO2 storage. The focus is on understanding molecular processes that occur at the interface between adsorbed species and inorganic framework materials, such as the surface reactions in heterogeneous catalysis and chemical sensors, and to model the phase equilibria and the reaction rates ab initio. For example, methane ice clathrate structures are modelled in order to calculate their phase diagram, which is of interest when considering the stabilities of natural gas deposits in the tundra and under the ocean floor if global warming occurs. Many different and versatile theoretical-chemistry methods are applied to carry out these studies, for which supercomputing resources are essential. The main methods used are quantum-chemical computations (QC) of both the molecular and periodic types, molecular dynamics (MD) and currently foremost Monte Carlo (MC) simulations, topology analysis, ab initio MD and CPMD simulations. The codes used are public domain, commercial or developed by us.