Development of scientifc software and research in Aqueous Molecular Systems with environmental and energy impact

Resumen

The thesis has been dedicated to the development/production of new software tools that are easy to use and of the highest quality, offering solutions that integrate different approaches and computational methodologies, facilitating the current and future development and research in the field of molecular sciences. One of the main objectives of the thesis is to develop additional operability to the DENEB computational software, that will allow a full cycle of molecular modelling: from the design to the calculations, as well as the analysis of the results in a collaborative environment. The software is intended to deal with the technical details of the calculations (algorithms, parameter adjustment, convergence studies, error evaluation, ...). These tools have made it possible to approach scientific research in an efficient and integrated manner. Thus, through computational modelling/simulation, new leading protocols have been provided based on methodologies and calculations of first-principles, to solve relevant problems in future applications of environmental and energy interest. Specifically, the objectives have been framed according to the type of system to be treated: Pure aqueous homogeneous systems, heterogeneous water ion-molecule systems, and aggregates of different nature such as ion and He atoms. The thesis is organized into 3 chapters. The introduction presents the motivation and the current state of the topic and its interest; Chapter 2 entitled "Scientific Software and Methods in Computational Molecular Science" describes the methods, approaches and techniques developed in the field of electronic structure, as well as in molecular dynamics simulations, including a section dedicated completely to the development of various new modules and plugins incorporated in the DENEB code package in the philosophy of automating repetitive initiation, organization and analysis tasks during the investigation. Finally, chapter 3 shows several applications to actual challenging problems in molecular physics, such as benchmark electronic structure calculations in aqueous systems, modelling of underlying interactions and trends in the microsolvation process in aggregates of different nature, as well as nuclear quantum simulations for exploring evolution and stability of ion-water aggregates. Throughout the thesis, complementary approaches are used, such as performance and assessment of electronic structure computational methods, as well as nuclear molecular dynamics simulations, data-driven model development, etc, and a detailed analysis of various molecular properties (dynamic and static) is presented. The code has been developed mainly using the Java programming language within the Netbeans development environment, version control has been carried out with Git and other technologies such as SQLite have been used for data storage. For the calculation of properties, classical and quantum molecular dynamics (CMD, PIMD) methods have been employed, through the i-PI calculation packages, and our implementations. Evolutionary programming (genetic algorithm) methods have also been explored for the sampling and minimization of potential energy surfaces, as well as active learning techniques to efficiently and intelligently weigh the configurations during the process of potential energy parameterization, reducing in this way the size of the training set, and thus minimizing the computational cost. Likewise, electronic structure packages of codes, such as Molpro, Gaussian and Orca have been heavily used.