Browsing by study line "Laskennallinen materiaalifysiikka"

Halide perovskites are a promising materials class for solar energy production. The photovoltaic efficiency of halide perovskites is remarkable but their toxicity and instability have prevented commercialization. These problems could be addressed through compositional engineering in the halide perovskite materials space but the number of different materials that would need to be considered is too large for conventional experimental and computational methods. Machine learning can be used to accelerate computations to the level that is required for this task. In this thesis I present a machine learning approach for compositional exploration and apply it to the composite halide perovskite CsPb(Cl, Br)3 . I used data from density functional theory (DFT) calculations to train a machine learning model based on kernel ridge regression with the many-body tensor representation for the atomic structure. The trained model was then applied to predict the decomposition energies of CsPb(Cl, Br)3 materials from their atomic structure. The main part of my work was to derive and implement gradients for the machine learning model to facilitate efficient structure optimization. I tested the machine learning model by comparing its decomposition energy predictions to DFT calculations. The prediction accuracy was under 0.12 meV per atom and the prediction time was five orders of magnitude faster than DFT. I also used the model to optimize CsPb(Cl, Br)3 structures. Reasonable structures were obtained, but the accuracy was qualitative. Analysis on the results of the structural optimizations exposed shortcomings in the approach, providing important insight for future improvements. Overall, this project makes a successful step towards the discovery of novel perovskite materials with designer properties for future solar cell applications.

In this thesis, sputtering of several low- and high-index tungsten surface crystal directions are investigated. The molecular dynamics study is conducted using the primary knock-on atom method, which allows for an equal energy deposition for all surface orientations. The energy is introduced into the system on two different depths, on the surface and on a depth of 1 nm. Additionally to the sputtering yield of each surface orientation, the underlying sputtering process is investigated. Amorphous target materials are often used to compare sputtering yields of polycrystalline materials with simulations. Therefore, an amorphous surface is also investigated to compare it's sputtering yield and process with crystalline surface orientations. When the primary knock-on atom was placed on the surface all surface orientations had a cosine shaped angular distribution with little variation in the sputtering yield for most of the surface orientations. Linear collision sequences were observed to have a large impact on the sputtering yield when the energy was introduced deeper inside the material. In these linear collision sequences the recoils are traveling along the most close packed atom rows in the material. The distance from the origin of the collision cascade to the surface in the direction of the most close packed row is therefore crucial for the sputtering yield of the surface. Surface directions with high angles between this direction and the surface normal hence show a reduction in the sputtering yield. The amorphous material had a little lower sputtering yield than the crystalline materials when the primary knock-on atoms was placed on the surface whereas the difference rose into several orders of magnitude when the energy was given at 1 nm. It is impossible for linear collision sequences to propagate long distances in the amorphous material and therefore the angular distribution in both cases is cosine shaped. The amorphous material has no long range order and was therefore unable to reproduce the linear collision sequences, which are characteristic for the crystalline materials. The difference in the sputtering yield was hence up to several orders of magnitude as a result when the energy was introduced at 1 nm depth.

Ion beams have been the subject of significant industry interest since the 1950s. They have gained usage in many fields for their ability to modify material properties in a controlled manner. Most important has been the application to semiconductor devices such as diodes and transistors, where the necessary doping is commonly achieved by irradiation with appropriate ions, allowing the development of the technology that we see in everyday use. With the ongoing transition to ever smaller semiconductor devices, the precision required of the manufacturing process correspondingly increases. A strong suite of modeling tools is therefore needed to advance the understanding and application of ion beam methods. The binary collision approximation (BCA) as a simulation tool was first introduced in the 1950s. It allows the prediction of many radiation-related phenomena for single collision cascades, and has been adopted in many experimental laboratories and industries due to its efficiency. However, it fails to describe chemical and thermodynamic effects, limiting its usefulness where ballistic effects are not a sufficient description. Parallel to BCA, the molecular dynamics (MD) simulation algorithm was developed. It allows a more accurate and precise description of interatomic forces and therefore chemical effects. It is, however, orders of magnitude slower than the BCA method. In this work, a new variant of the MD algorithm is developed to combine the advantages of both the MD and the BCA methods. The activation and deactivation of atoms involved in atomic cascades is introduced as a way to save computational effort, concentrating the performed computations in the region of interest around the cascade and ignoring surrounding equilibrium regions. By combining this algorithm with a speedup scheme limiting the number of necessary relaxation simulations, a speedup of one order of magnitude is reached for covalent materials such as Si and Ge, for which the algorithm was validated. The developed algorithm is used to explain the behavior of Ge nanowires under Xe ion irradiation. The nanowires were shown experimentally to bend towards or away from the ion beam, and computational simulations might help with the understanding of the underlying physical processes. In this thesis, the high-fluence irradiation of a Ge nanowire is simulated and the resulting defect structure analyzed to study the bending, doubling as a second test of the developed algorithm.

The complex physical mechanisms involved in the formation of vacuum arcs have been of interest for many decades. Vacuum arcs are relevant in many engineering disciplines, but the physics behind them is not yet fully understood. In recent years, there have been many experimental and computational studies focused on understanding aspects of vacuum arcs. This thesis focuses on further development of a simulation model to describe the physical processes starting from electron emission and leading to the formation of an ionized plasma. The FEMOCS code is extended to include plasma simulation based on previous work on ArcPIC. Emission of electrons and heating of the cathode is simulated using the finite element method, while plasma simulation is performed using the particle-in-cell method. We add evaporation of neutral atoms from the cathode, as well as ionization processes for multiple species of ions. Monte Carlo collisions for elastic, Coulomb, impact ionization, charge exchange and recombination collisions between particles are added. Direct field ionization of neutrals is included to account for ionization at high electric fields. A dynamic weighting scheme is described for adjusting superparticle weights during the simulation. Ion bombardment effects such as bombardment heating and sputtering are added to account for additional supply of neutrals resulting from energetic ions accelerated by the electric field. Finally, we add a circuit model for coupling to an external circuit. A static nanotip is simulated with different parameters to study local field thresholds leading to thermal runaway. We find that our simulations agree with experimental results. The most significant interactions contributing to initial formation of vacuum arcs are identified. We find most neutrals are created via evaporation rather than sputtering. The most important collision for plasma formation is impact ionization of neutrals into Cu+ ions, while higher-order ions are found to play a lesser role. Direct field ionization of neutrals is also found to be significant at high fields on the order of 10 GV/m.

This thesis examines the implementation of general purpose graphics processing unit (GPGPU) acceleration to a non-equilibrium Green’s function (NEGF) equation solver in the context of a computational photoelectrochemical (PEC) cell model. The goal is to find out whether GPGPU acceleration of the NEGF equation solver is a viable option. The current model does not yet have electron-photon scattering, but from the results it is possible to assess the viability of GPGPU acceleration in the case of a complete PEC cell model. The viability of GPGPU acceleration was studied by comparing the performance difference of two graphics processing unit (GPU) solutions against a multi core central processing unit (CPU) solution. The difference between the two GPU solutions was in the used floating-point precision. The GPU solutions would use LU factorization to solve the NEGF equations, and the CPU solution a banded solver (Gauss tridiagonal) provided by Scipy Python package. The performance comparison was done on multiple different GPU and CPU hardware. The electrical transport properties of the PEC cell were modeled by a self-consistent process in which the NEGF and Poisson equations were solved iteratively. The PEC cell was described as a semiconductor device connected with a metal and electrolyte contacts. The device was assumed to be a simple one dimensional tight-binding atom chain made of GaAs, where the transverse modes in the y–z plane are treated with a logarithmic function. The computational model did lack electron-photon scattering, which would be implemented in the future. From the benchmark results, it can be concluded that the GPGPU acceleration via LU factorization is not a viable option in the current code or in the complete model with electron-photon scattering and the assumed approximations. The parallel multi-core CPU code generally outperformed the GPU codes. The key weakness of the GPU code was the usage of LU factorization. Despite of this, there could be an opportunity for GPGPU acceleration if a more complex lattice structure and more exact scattering terms would be used. Also, a GPU accelerated tridiagonal solver could be a possible solution.

The rapidly increasing global energy demand has led to the necessity of finding sustainable alternatives for energy production. Fusion power is seen as a promising candidate for efficient and environmentally friendly energy production. One of the main challenges in the development of fusion power plants is finding suitable materials for the plasma-facing components in the fusion reactor. The plasma-facing components must endure extreme environments with high heat fluxes and exposure to highly energetic ions and neutral particles. So far the most promising materials for the plasma-facing components are tungsten (W) and tungsten-based alloys. A promising class of materials for the plasma-facing components is high-entropy alloys. Many high-entropy alloys have been shown to exhibit high resistance to radiation and other wanted properties for many industrial and high-energy applications. In materials research, both experimental and computational methods can be used to study the materials’ properties and characteristics. Computational methods can be either quantum mechanical calculations, that produce accurate results while being computationally extremely heavy, or more efficient atomistic simulations such as classical molecular dynamics simulations. In molecular dynamics simulations, interatomic potentials are used to describe the interactions between particles and are often analytical functions that can be fitted to the properties of the material. Instead of fixed functional forms, interatomic potentials based on machine learning methods have also been developed. One such framework is the Gaussian approximation potential, which uses Gaussian process regression to estimate the energies of the simulation system. In this thesis, the current state of fusion reactor development and the research of high-entropy alloys is presented and an overview of the interatomic potentials is given. Gaussian approximation potentials for WMoTa concentrated alloys are developed using different number of sparse training points. A detailed description of the training database is given and the potentials are validated. The developed potentials are shown to give physically reasonable results in terms of certain bulk and surface properties and could be used in atomistic simulations.

This thesis focuses on the initial measurements and development of a 1.5K target temperature cryostat to contain all quantum standards in the quantum metrological triangle. Currently, the cryostat can reach 4.2K end temperature with a cryocooler to liquefy helium for cooling experiments related to the quantum standards of SI-units which require 1.5K end temperature and the 1.5K cooling circuit is one of the research questions of the thesis for future development of the cryostat. Measurements included cooling cycles in a helium environment for liquefaction, as well as no load conditions. Results of the experimentation conclude that liquefaction was unsuccessful at this stage, but could be achieved with improved temperature control. The thesis is based on experimental work carried out in VTT technical research center of Finland, during the course of which the cryostat was progressively developed towards the future utility of combining experiments that realize the quantum standards of Ampere, voltage and resistance under a single low temperature experimental platform, the unified quantum standard cryostat.

In the field of cryogenics and superconducting technology, the effect of eddy currents presents a significant challenge, as large inductive currents can affect the performance and stability of cryogenic components. In our work, we examine how the magnitude of eddy current effects varies with different samples when placed within a changing magnetic field. In high magnetic field applications, any deviation in the magnetic field induces currents within low-resistivity components, leading to eddy current heating and Lorentz forces. This thesis focuses on how to simulate the eddy current phenomena in an accurate way through methods of finite element analysis, utilizing the commercial software of COMSOL Multiphysics. To provide a comprehensive simulation of the effects of eddy currents, our work involves the coupling of three distinct physical fields: solid mechanics, heat transfer, and electromagnetic fields. To solve multiphysical problem in an efficient way, we explain different strategies on how the coupled fields can be solved in a simplified, but effective way. The simulation was examined for two different time scale scenarios: 1. Turning on the magnet, where the time scale of the phenomena is in the order of several hours, and 2. The more demanding scenario of the quench of a superconducting magnet, where the time scale is in the order of several seconds. We found that during the ramping of the magnet the electromagnetic heating of the sample can reach the scales of milliwatts, which is significant head load in a cryogenic setting. During magnet quench, we found that the Lorentz forces can reach up to scales of kilonewtons. The results indicate that the volume of the sample has significant impact to effects of eddy currents, but when considering the magnitude of the Lorentz force the length and spatial location of the sample has significant effect. Hence, it is crucial to pay attention to the appropriate design of the sample that is placed into the magnetic field.

Vacuum breakdown is a limiting factor in the design of powerful and cost-efficient particle accelerators. Modern models have suggested that the rate of breakdowns is driven by dislocation dynamics in the electrode materials suffering from breakdowns. In order to understand why specifically the copper-2wt%beryllium alloy outperforms other electrode materials in vacuum breakdown rate and maximum electric fields in breakdown experiments at CERN, a new machine-learning interatomic potential (ML-IAP) for the CuBe alloy was developed. Density functional theory (DFT) was used in calculating a dataset of atomic forces, energies, and virials for a set of CuBe structures. This dataset was performed a fit on with Gaussian process regression, producing an IAP with close-to-DFT accuracy in its intended use cases. With the developed IAP, the interactions between single interstitial beryllium atoms and edge dislocations in a face-centered cubic (FCC) copper matrix were studied with molecular dynamics (MD). It was found that beryllium atoms bind to the edge dislocations, inhibiting their mobility under shear stress. Furthermore, beryllium atoms were found to increase the intrinsic stacking fault energy of FCC copper, possibly leading to an increase in dislocation mobility. These two findings suggest that beryllium atoms could increase copper's resistance to vacuum breakdown mainly via trapping dislocations. Future studies could look at how precipitates of beryllium, or other alloys of copper, play a role in dislocation dynamics.

Energy is an essential input for any non-spontaneous mechanism. In biological organisms, the process of producing energy currency, adenosine triphosphate, is called cellular respiration. It is made of three smaller steps, out of which the last one is oxidative phosphorylation that is responsible for the largest production of adenosine triphosphate molecules in the whole process. Oxidative phosphorylation is performed by the electron transport chain made of five protein complexes, named respiratory complex I-V. Complex I is the first and largest protein complex in the electron transport chain, and it is the least understood. Its primary function is to transfer electrons from nicotinamide adenine dinucleotide to ubiquinone, which is coupled to the pumping of four protons across the mitochondrial inner membrane. Although the overall reaction of complex I is understood, the intricate detail of the mechanism is still largely unknown. There is significance in the details because there are numerous point mutations, which have been strongly correlated with neurogenerative diseases, such as Leigh’s syndrome, and aging. Therefore, a more thorough understanding of its mechanism can give insight into potential target drug development. Complex I is made of 14 highly conserved subunits that can be found in most species that use the electron transport chain. They create an L-shape, where seven subunits are embedded in the inner membrane, the membrane domain, and the others are floating in the mitochondrial matrix, the peripheral arm. In mitochondrial complex I, however, there are in addition around 30 accessory subunits. It has been previously thought that the main mechanism is conducted by the 14 subunits that are found in all species. However, in the past couple of years, it has been shown that accessory subunits can play an important role in the mechanism of mitochondrial complex I. The work presented in this thesis uses a multiscale computational approach to study the effect of three mutations, F89A, Y43A and L42A, from an accessory subunit LYRM6 on the function of complex I. Previous experiments demonstrated that the mutations decreased the overall activity of complex I by 76-86 %. The LYRM6 subunit is located at the pivot of the membrane and periplasmic domains. The results of this study show that the point mutations have a long-range effect on the conformations of three loops from three conserved subunits in this region. The shift in the loop dynamics causes a drop in water occupancy. The observed water pathway is tested for the capability of proton transfer. The findings are demonstrated with the help of molecular dynamics and quantum mechanics/molecular mechanics simulations.