Radioisotope thermoelectric generators produce electricity for space exploration. These generators use thermoelectric material, usually a silicon-germanium alloy, to turn heat from radioactive decay into electricity. Recently, the thermoelectric performance (ZT) of Si80Ge20 was improved by reducing the grain size to the nanometer range, significantly lowering the thermal conductivity.
To optimize the properties of silicon-germanium we studied the effect of doping concentration and processing parameters on the microstructure and properties of boron-doped Si80Ge20 produced by arc melting, ball milling and spark plasma sintering. The thermal conductivity was estimated with a model. The electrical conductivity and Seebeck coefficient were measured.
Some conclusions from this work are that the current production process can be used to produce nanostructured Si80Ge20Bx with a crystallite size of 50-100 nm. This material reaches a maximum, but not necessarily optimal, doping concentration when x=1 due to limited solubility. The material suffers from grain growth when exposed to high temperatures for several days. The use of iron in the ball milling process significantly affected the microstructure and properties.

In this thesis, molecular simulations are performed to design and assess novel 2D materials for H2 storage applications (chapters 2-3) and to predict thermodynamic and transport properties of H2 in aqueous electrolyte solutions for storage and production of H2 (chapters 4-8). Both ab-initio and force field-based methods are used in this thesis.....

The detrimental effects of Cu contamination during steel recycling and production are primarily due to the segregation of Cu at Fe grain boundaries (GBs). A promising approach to mitigate these effects is the introduction of alloying elements that inhibit Cu segregation at Fe GBs, which can be investigated through atomistic simulations. Currently, popular simulation methods include density functional theory (DFT)-based simulations and empirical interatomic potentials (EIPs)-based simulations. DFT calculations can provide simulations with high accuracy, while the high computational cost limits their application to simpler GB structures. Moreover, simulations utilizing EIPs may lack reliable potentials, especially when predicting GB segregation tendencies across a broader range of alloy systems. To address these challenges, universal machine learning interatomic potentials (uMLIPs), which are trained on DFT data and applicable for most of elements, have emerged as a promising alternative. Although uMLIPs have shown potential in various materials simulation tasks, their reliability for out-of-distribution tasks, such as simulating GB segregation behavior, remains unproven.

This thesis project evaluates the performance of the best available uMLIPs, specifically MACE-MP-0, CHGNet, M3GNet, and SevenNet-0, in predicting single-solute GB segregation energies, GB energies for both body-centered cubic (BCC) Fe and face-centered cubic (FCC) Fe systems, and solution enthalpies for BCC Fe and cementite. The results were compared against existing studies conducted via DFT calculations to assess the accuracy and applicability of each uMLIP. Additionally, some EIPs were tested for comparison, serving as an extra reference. The findings reveal that MACE-MP-0 generally outperforms the other uMLIPs in both accuracy and stability of convergence. While all tested uMLIPs perform well in BCC Fe systems, CHGNet(v0.2.0) and SevenNet-0 show reduced accuracy in FCC Fe simulations. Although most of the simulations using uMLIPs converged well in BCC Fe GBs, many unconverged cases were reported in FCC Fe systems, particularly for uMLIPs other than MACE-MP-0 and CHGNet(v0.3.0). Furthermore, a consistent underprediction of segregation tendencies for highly segregating solute elements, such as Cu, is observed in the results of MACE-MP-0 and CHGNet. This suggests that while uMLIPs hold significant potential for atomistic simulations, fine-tuning pre-trained uMLIP models for out-of-distribution tasks, such as calculating GB segregation energy, is recommended to improve accuracy and convergence behavior. This work offers a valuable benchmark for using uMLIPs in future GB segregation studies.

For the first time, we present data on the actual growth rate, supersaturation, viscosity, and growth mechanism that lead to spherulitic growth. So far, fundamental questions regarding the growth conditions, mechanisms, and the necessity of a viscous growth medium have remained unresolved. Surprisingly, we observed a sol-gel transition in evaporating silica-free sulfate mixtures, followed by the growth
of sodium sulfate spherulites. We characterized the spherulites’ morphological evolution and chemical/structural composition using various microscopy techniques and Raman Spectroscopy. The study reveals that faceted crystals, during their morphological evolution, can transiently exhibit a spherulitic morphology before attaining their final shape. We demonstrate that adding bivalent ions to sulfate solutions can create the conditions required for the spherulitic growth of the crystal phase. We show how to obtain perfectly developed spherulites through an in-depth experimental investigation of ion concentrations, evaporation rate, and geometric constraints. Moreover, quantifying the growth conditions enables a precise understanding and facilitates a comprehensive discussion on a general approach for
cultivating spherulites through solvent evaporation that is imperative for innovative purposes.

In this work, the added value of machine learning (ML) molecular force fields (FF) for the community of molecular simulations is showcased by successfully calculating transport properties of aqueous potassium hydroxide (KOH (aq)). Classical FFs use relatively simple interatomic potentials to simulate the nano scale. These simulations can predict macroscopic properties, such as density, heat of evaporation, viscosity, and self-diffusivity of the modeled materials. However, these FFs struggle to model materials in which more complicated interactions are relevant for the macroscopic behavior. Examples of such interactions are three-body interactions and chemical reactions. Quantum scale simulation methods are able to compute properties of materials in which these challenging interactions occur, although these methods are limited in length and time scales that can be modeled with realistic computational costs. Transport properties, such as viscosity, self-diffusivity and electric conductivity need these larger length and time scales to be determined accurately. ML can be used for a multi scale approach, bridging the gap between the quantum and the nano scale by training coefficients of general interatomic potentials. This provides the possibility of reaching the time and length scales of traditional molecular simulations with the accuracy of quantum mechanic models. KOH (aq) is selected to highlight the prospects of these multi scale techniques, as the self-diffusion of OH- in this electrolyte is dominated by proton transfer reactions, which has not been modeled successfully with classical FFs.

Results of structure properties produced with ab initio molecular dynamics (AIMD, at quantum scale) simulations are compared with machine learning molecular dynamics (MLMD, at multi scale) simulations. There are no significant differences in the calculated shortest typical atomic distances and coordination numbers for both KOH (aq) and pure water systems. The determined transport properties are in the same order of magnitude as experimental results, although the calculated viscosity is overestimated and the self-diffusion of H2O and K+ are underestimated. This is because the system is simulated at a higher than experimental density and hydrogen bonding is overestimated with the selected quantum mechanics model. The proton transfer reactions are captured in the MLMD simulations, calculating the enhanced self-diffusion of OH- to be (6±2)e-9 m squared per second, which matches experimental results at infinite dilution.

In this thesis, the first part will be a review of the literature that includes as much of the commercially available information on carbon capture and mainly, Direct Air Capture (DAC). What are the difficulties surrounding capturing CO2 from gas mixtures and what are the possible solutions that have been investigated? What type of negative emission technologies have been proposed, what are the important parameters for the design of carbon removal technologies and what are the biggest obstacles that need to be overcome? Gathering sufficient data and comparing the different finding will be of importance, given that DAC technologies are still in developing stages and information is limited. The focus of this thesis will be on chemically amine based solid sorbents, as this can be argued that this is the most promising technology for a viable DAC system.

The modelling part of the thesis will be focused on the design of and development of a DAC model in the open source software ’Python’. The amine based sorbent that is investigated and used for the creation of the model is Lewatit VP OC 1065. Using experimental data gathered from the literature together with feasible assumptions, a model will be built to recreate the the whole DAC process and analyse the system. The main focus will be on acquiring a flexible model for both the adsorption and desorption parts of the DAC process to make further investigation of the system parameters possible. This model will be used after this thesis for further development and, for instance, analysing different possible sorbents.

Boron-based two-dimensional materials showcase promise in variety of fields like hydrogen storage, fabrication of electronic devices and catalytic applications. These materials have garnered interest owing to their unique properties such as high electron mobility, high gravimetric capacity for hydrogen especially after metal decoration, thermal conductivity and tensile strength amongst others. However, for sustained operations of the devices involving these materials, their chemical stability against oxygen is of paramount importance. Especially in the applications involving exposure of the material to air. Abundance of oxygen in the air and its high reactivity increases the likelihood of oxidation of the material. In this work, chemical stability of hydrogen passivated 2D Boron structures; Borophane and 2D Boron Hydride against oxygen were analysed. First-principles calculations reveal that Borophane and 2D Boron Hydride have a less negative binding energy thus indicating that the oxygen binds less strongly to compared to Borophene which does not possess surface passivation by hydrogen. Experimental studies in the literature involving synthesis of 2D Boron Hydride reported presence of vacancies. The effect of vacancy site towards the reaction with oxygen was therefore analysed to maintain consistency with the synthesized materials. The simulations were performed on Borophane and 2D Boron Hydride by introducing a Boron vacancy in the system. The DFT simulation of defect containing Borophane revealed that oxygen binds less strongly in the physisorbed state and more strongly in the chemisorbed state, relative to the values obtained for Borophane without any defects. In case of 2D Boron Hydride it was observed that the vacancies provide a stable site for oxygen, for both physisorbed and chemisorbed state. From these simulations we can conclude that presence of vacancies in Boron based 2D materials generally leads to a stable site for oxygen.

The formability of Advanced High Strength Steels is critical for their usability in automotive applications. It has been observed that the presence of hydrogen, even in concentrations of the order of 1 ppm, leads to a considerable drop in formability. Hydrogen atoms may get absorbed during steel-making and are known to get trapped at various sites in the lattice. When sufficient activation energy is made available, hydrogen atoms that are weakly trapped can diffuse towards critical regions in the microstructure, such as crack tips and voids, where one or more embrittlement mechanisms might be activated. On the other hand, a strongly trapped hydrogen atom remains immobile and plays no part in the embrittlement process. Precipitates of transition metals are known to be strong traps for hydrogen. It is speculated that by promoting the formation of strong traps in the microstructure, the amount of freely diffusible hydrogen can be limited, which would lead to an improvement in mechanical performance.

In this work, a combined ab-initio - experimental approach was used to study the absorption of hydrogen in dual-phase steel. Density Functional Theory (DFT) calculations were employed to study and compare the trapping of hydrogen by carbide and nitride of titanium and vanadium. A carbon or nitrogen vacancy in the bulk of the precipitate was found to be the most efficient trap site. When coupled with the vacancy formation energy, trapping was found to be more efficient in off-stoichiometric vanadium carbide and nitride than that in titanium carbide and nitride. To validate the theoretical findings, cyclic voltammetry experiments were conducted on two grades of DP800 steel with different concentrations of vanadium and titanium. The amount of diffusible hydrogen in the vanadium grade was found to be approximately 25 \% higher than that in the titanium grade. This was in contradiction to the theoretical results. Characterisation of the specimen post testing revealed that an oxide film had formed on the sample surface and while the film on vanadium grade was uniform and dense, that on titanium grade was sparse and irregular. It was evident that the oxide layer contributed to trapping of hydrogen, however the amount of hydrogen trapped by the oxide could not be specified. Overall, designing steels resistant to hydrogen embrittlement by promoting the formation of precipitates of a particular element is theoretically attainable, however, it was not possible to obtain experimental validation with the method employed.