Hydrogen, as an energy carrier, is of paramount importance in the energy transition. At the industrial level, it can be derived from various sources, including fossil fuels, biomass, or electrolysis. In Alkaline Water Electrolysis (AWE), the growth of hydrogen bubbles directly impacts system efficiency. Understanding and simulating this growth, attributed to the diffusion of dissolved hydrogen in the supersaturated electrolyte near nucleation sites via diffusive and convective mass transfer, is a crucial step towards advancing knowledge in this field and unlocking new possibilities.

This research focuses on simulating the growth of a single hydrogen bubble in a supersaturated domain, both far from and near the cathode, in a 30 wt% KOH solution. Bubble growth, a mesoscale phenomenon, is investigated using the Lattice Boltzmann Method (LBM). A comprehensive comparison of the Shan-Chen (SC), Colour Gradient (RK), and Interface Tracking Phase-Field (HZC) methods was conducted to measure the intricacies of the multiphase system accurately. The Laplace Law equation served as a benchmark, demonstrating that the HZC method produced the most accurate results.

A continuous species transfer method is employed to track hydrogen transport from the supersaturated electrolyte into the bubble, validated with Newman's analytical solution of mass transfer controlled by pure diffusion inside a sphere. Two cases are then analyzed: one of a single bubble far from an electrode in a supersaturated domain and another of a single bubble near an electrode with a constant hydrogen flux. For the first case, bubble growth follows a power law equivalent to R ∼ t^0.5, while in the second case, growth follows R ∼ t^0.7, matching results from previous studies. Finally, this method is extended to a 3D model; however, the results cannot be directly compared to the 2D model due to the shorter runtime resulting from computational cost.

Recently, a mechanism has been proposed to describe how salts do (not) contribute to bubble coalescence inhibition. This is done by considering how ions (re)distribute themselves along the G/L-interface, expressing this as a Gibbs-Marangoni pressure. The thesis’ goal is to investigate the potential and applicability of this mechanism to predict bubble coalescence inhibition for salts in biotechnological processes, specifically bubble columns. This goal is achieved by recreating the mechanism and evaluating whether the Gibbs- Marangoni pressure is a better predictor than the ionic strength. To make this comparison, experiments are performed to study which ionic strengths/concentrations induce bubble coalescence inhibition. These numbers are then modelled in terms of Gibbs-Marangoni pressures, from where it is assessed which of the two better predicts this inhibition. More- over, additional experimental results from the literature have been tested on the model to support this assessment. The outcome of this thesis is a combination of multiple interesting insights. First, the above mechanism is successfully reproduced into a model. This statement is supported by com- paring the model’s results with previous literature applying this mechanism, showing that results deviate less than 10% from each other. Secondly, from the performed experiments, bubble coalescence inhibition by salts is observable at an average ionic strength of 47.82 mM. Using these results in the model shows that this inhibition happens at an average Gibbs-Marangoni pressure of 1312.91 Pa. How- ever, the inhibiting salt concentrations from the experiments deviate by more than 168% compared to the literature. This difference can be caused by many factors, such as dissimilarities in the experimental setups used. It shows that the critical Gibbs-Marangoni pressure is dependent on these factors and therefore not entirely generalizable for all scenarios. Finally, the model’s results are validated by comparing whether the additionally calculated surface tensions deviate less than 10% from the literature. This showed resulting Gibbs- Marangoni pressures are valid for pure salts at concentrations from 0.02-1.50 M. However, the validity of the model for salt mixtures is questionable. To conclude, the proposed mechanism can describe bubble coalescence inhibition for salts in bubble columns. However, its accuracy is debatable as the model’s results cannot be validated for mixtures. This may require the inclusion of other ion contributions into the model, such as ion polarizability, solvation energy and ion-charge density. Hopefully, this brings us closer to better modelling the effect of salts, and other compounds, on bubble coalescence inhibition.

The sea and the shoreline form a complex ecosystem driven by tides. So far, studies often ignore the moving boundary caused by these tides. The focus of this thesis to incorporate this boundary by using a coordinate transformation and a time-explicit numerical method. To achieve this, first the one-dimensional shallow water equations are derived from the 3D Navier Stokes equations. Then these 1D equations are non-dimensionalized and the coordinate transformation is done. This results in a system of non-linear equations. The seabed is modelled as a straight line. At the seaward side there is a periodic forced wave and at the landward side the water depth is 0. The time-explicit numerical method of Lax-Friedrichs is used. This method is stable under a more restricted Courant-Friedrichs-Lewy condition and is convergent for refined grids. For the Ameland inlet system the water depth, velocity and length of the basin results are calculated and compared to a simplified model and complemented by a Fourier analysis. The results are realistic (constant in the beginning of the basin with visible non-linearities at the landward side). An analysis is done to understand how the model behaves for dierent physical parameters, such as: the amplitude of the periodically forced wave, the undisturbed water depth, the length of the basin and the resistance. The model remains stable and the results are realistic.

In recent years, computational fluid dynamics (CFD) has become an essential design tool across various industries, allowing engineers to tackle complex fluid dynamics problems that would otherwise require costly and time-consuming real-life experiments. For Formula 1 teams, who must experiment within strict time limits in the wind tunnel and on track, the ability to simulate airflow around their race cars under various conditions is crucial for maintaining competitiveness in the fast-paced world of Formula 1 racing. Reynolds-Averaged Navier-Stokes (RANS) simulations remain the industry standard for simulating turbulent flows, as they allow engineers to conduct simulations efficiently. However, this efficiency comes at the expense of accuracy, as RANS cannot resolve all turbulence scales, leading to uncertainties.

Recent advances in data-driven RANS turbulence modeling have enabled partial correction of these uncertainties. However, obtaining a correction that is generalizable under different geometries and flow conditions remains a challenge. Turbulence models are calibrated to fit specific flow regimes, so correcting these models across the entire domain can disturb these calibrations, worsening performance. A solution is to divide the domain into regions based on identifiable physical phenomena and apply local corrections without disturbing calibrated regions. In Formula 1 race car design, the most critical region is the shear layer, where RANS shows the largest discrepancies.

In this thesis, a classifier was developed to distinguish the shear layer from the rest of the domain based on the ratio between turbulent kinetic energy production and destruction, as well as turbulence intensity. Within this classifier region, corrections to the k-ω SST turbulence model are made by extracting model form errors from high-fidelity data using k-corrective-frozen RANS. These corrections include a residual term added to both the k and equations and a term for the anisotropy of the Reynolds stress tensor. The Spars Regression of Turbulent Stress Anisotropy (SpaRTA) framework, based on elastic-net regularization, was used to regress symbolic expressions for the corrections, enabling their application to simulations of unseen test cases.

The models discovered with the SpaRTA framework for the shear layer show promising results, improving the prediction of separation and reattachment positions. These models were tested on various geometries and simulations at different Reynolds numbers, demonstrating a certain level of generalizability. While there is room for further improvement, this thesis shows that integrating targeted model corrections into RANS simulations, informed by isolated shear layer data, can enhance the understanding and prediction of shear layer dynamics in 2D-separated flows.

In this study, a model is developed in COMSOL to investigate the nature of energy losses in ion exchange membranes used in electrochemical energy conversion. By use of electrochemical impedance spectroscopy, the nature of resistance and reactance, and therefore the nature of Ohmic resistances and charge transfer resistances, are evaluated. This is done by first obtaining a model of the setup and the resulting impedance spectrum and then making adjustments in the model to observe the impact on the impedance spectrum as well as the resulting ion concentration profiles. In the used system, the diffusion effects were dominant compared to the double layer effects. Therefore, the impedance spectrum is mainly shaped by the diffusion boundary layer, especially at low frequencies. The ion concentration, diffusion coefficients, and membrane properties were found to have a significant influence on the impedance spectrum. Insights from this study can be used in the optimization of ion exchange membranes and the efficiency of energy conversion systems.

This thesis focuses on the influence of tides on flow behaviour and salt intrusion in a semi-infinite channel ending at the sea, with a specific focus on the effects of adding a side channel. Tide causes variations in the flow behaviour, which in turn affects salt concentrations in the channel. Adding a side channel has an impact on the flow pattern, mainly because of reflections of the tidal wave, leading to changes in the salt intrusion. The effect investigated is the influence of varying the distance from the sea to the junction, and the length of the side channel. A 1-dimensional exploratory model based on the shallow water equations for the flow behaviour is developed, analysing only the bidaily M2 tide caused by the gravitational pull of the moon. The flow pattern is solved analytically, while the salt balance is solved analytically in time but numerically in space. The salt balance contains an advection term and an effective dispersion term. The effect of adding a branching channel is compared to the same situation without a branching channel. It is found that the addition of a side channel affects the salt intrusion in the order of 1 km compared to a single channel. Moreover, both positive and negative changes occur in the shift of the intrusion length. The results of this study can be used to understand and explain the behaviour in more complex geometries.

This study aims to relate the size, number density, and surface coverage of bubbles adhering to a vertical gas-evolving electrode to the current density and shear rate of alkaline electrolyte flow.
A force balance that includes the known forces on the bubbles is formulated. This force balance is made to predict the release radius of the adhered bubbles. Because the electric and Marangoni forces are still an open topic of research, they are approximated using a constant. Also, because the influence of the drag force on bubbles in alkaline water electrolysis is not exactly known, it is approximated using the Stokes drag force with a multiplicative fit factor. An equation for this force balance model that approximates the release radius of the bubbles is formulated. To know the exact value of the shear rate on the electrode, this study uses laminar, fully developed flow. Equations are formulated to relate the fully developed laminar flow to the shear rate.
The bubbles adhered to the electrode are visualized using a high-speed camera. The camera films through a PMMA rod that goes through the first electrode, allowing visualization of the bubble-covered surface of the second electrode. A long stainless steel rectangular duct develops the Laminar flow before it arrives at the electrode.
The experiment is to relate the bubble size, bubble number density, and surface coverage to the shear rate due to forced flow and the current density. Increasing the shear rate has the biggest effect on the largest bubbles adhered to the surface. The maximal radius reduces, while, for the range in shear rate used in this study, the median stays almost constant. The bigger bubbles will be replaced by smaller bubbles. This will reduce the surface coverage and increase the bubble number density. Increasing the current density, for the range in current density used in this study, means an increase in bubble interactions. This means an increase in bubble coalescence, collisions between bubbles, large bubbles being pushed off the surface by smaller ones behind them, and the increased shear rate from natural convection caused by moving bubbles. These increased bubble interactions will result in a reduction in bubble sizes, surface coverage and an increase in bubble number density.
The force balance model is related to the maximal radius of the bubbles. The constant of the electric and Marangoni forces is negative at the anode and positive at the cathode, which is expected because the forces should be attractive at the anode and repulsive at the cathode. The multiplicative fit factor of the drag force due to forced flow indicates a higher force than the Stokes drag force because the flow is not perfectly homogeneous and will flow through other bubbles, which can accelerate the local flow velocities. Also, bubbles will collide at a higher speed with increased forced flow. The multiplicative fit factor of the drag force due to natural convection indicates a higher force than the Stokes drag force because of the increased bubble interactions with increasing current density. Both the higher multiplicative fit factors are observed at the anode and cathode.

Fluidized bed reactors are catalytic multiphase reactors capable of processing large volumes with relatively high mass and heat transfer, which is appealing to a variety of industries. The presence of voids, in this case gas bubbles, within these systems can be detrimental to the efficiency and are desired to be minimized. Further fundamental understanding of the bubbles is necessary to model, design, and control them. However, the study of bubbles is difficult, as these beds are typically opaque and 3D. Current studies of bubbles within fluidized beds are often limited to use of quasi-2D set ups, specialized particles, or internal measuring devices. This research utilizes a newly constructed experimental setup consisting of three x-ray source and flat panel detector pairs that produce 2D projections. This setup allows for investigation into the bubble dynamics inside a 3D cylindrical column with industrial particles. The possibilities and limitations of the use of 2D projections to obtain fully 3D time-resolved reconstructions were explored through simple 3D reconstructions of injections of single bubbles. These 3D reconstructions showed the potential of the setup for 3D time-resolved studies of the bubble dynamics and laid the groundwork for future studies. Apart from the ability to obtain 3D reconstructions, the 2D projections can be used to study the dynamics of the bubbles. In this research, 2D projections of an injection of a single bubble were used to develop a deeper fundamental understanding of the shape, motion, and dynamics of an individual bubble traversing through a bed. The bubble shape and size were obtained through a horizontal slicing technique, which assumes axis-symmetry around the vertical axis. Different background fluidization levels, injection volumes, injection velocities, particles, and column diameters were investigated. An analysis of the individual bubble trajectories and the statistical averages provided insight into the interconnection of these factors. The two previously studied bubble stages, formation and rising, were observed. The rising stage was observed to have two regions, stable and unstable. The stable region, following the formation stage, occurs when the motion of the bubble is essentially rectilinear. The unstable region, following the stable region, occurs when the bubble moves sideways while traversing the column in an irregular motion. A simple analogy to gas bubbles in a liquid was drawn, suggesting the bubble cloud, or particles surrounding the bubble, plays a vital role, especially in the formation of the bubble. Standard correlations for the bubble rise velocity based on a constant bubble Froude number were found to be insufficient at accurately describing the bubble velocity and dynamics. The Froude number for individual bubbles appears to be non constant and dependent on parameters that have not been previously considered, like the level of background fluidization.

A phenomenon that compromises the health of city inhabitants is the urban heat island (UHI) effect. This effect is mainly driven by thermal radiation and designates the typically higher air temperature in cities compared to the surrounding rural areas. Over 55% of the world’s population lives in cities, and with this percentage set to rise to more than 65% by 2050, it is vital to model the urban climate to be able to mitigate the UHI-effect. In this study, the surface energy balance (SEB) is modeled in urban environments using a one-way coupled RANS 𝑘 − 𝜖 approach in order to gain fundamental insights into the influence of thermal radiation on the UHI-effect. The SEB model is built mechanism-by-mechanism, adding in different physical mechanisms one at a time and validating the model at each step with literature- or theoretical values. The SEB model showed a good agreement with experimentally found values for the effect of surface irregularity of an urban structure on the absorption and reflection of incident solar radiation. A good agreement was also found with theoretical values for the longwave radiation and conductive heat flux part of the SEB model. Subsequently, a study was done on a 2D street canyon to compare the model with literature results. This study showed a good agreement for the radiative fluxes, but a lesser agreement for the conductive- and sensible heat fluxes. A more complex scenario of an intersection between buildings was studied. It was shown that the longwave trapping effect is highly correlated with building height. Finally, the effect of wall heating on flow characteristics in a 2D street canyon was examined. For homogeneously heated walls, the SEB model showed a relatively good agreement with literature results. A study was also done with non-homogeneously heated walls. This study showed less realistic results and should be further investigated in further research.

In dit verslag is een eolisch sedimenttransport model geformuleerd en het model van De Vries geïmplementeerd en geanalyseerd. Na de afleiding van de eendimensionale advectie-diffusievergelijking vanuit de driedimensionale continuïteitsvergelijking is eerst het model van De Vries zonder diffusie onderzocht. In het model kan er onderscheid gemaakt worden tussen 3 verschillende situaties: Geen (toevoer van) sediment, Sediment-gelimiteerd en Erosie-gelimiteerd. In elke situatie zijn voor lange tijd\-schalen de toestanden van het model afgeleid vanuit de advectievergelijking. In de resultaten kwam naar boven dat de theoretische toestanden sterk overeenkwamen met de toestanden in de numerieke simulatie. Verder is het effect van numerieke diffusie van het model van De Vries geconstateerd en de grootte van dit effect afgeleid. Om het effect van numerieke diffusie te verminderen is diffusie in het model geïntroduceerd samen met het Lax-Wendroff numerieke schema. Dit schema gebruikt de gewogen upwind methode samen met de gewogen centrale Euler methode om de nauwkeurigheid van het model te verbeteren. In de resultaten is aangetoond dat numerieke diffusie kan worden gecompenseerd door het juist kiezen van dit gewicht. Uit een stabiliteitsanalyse blijkt dat de Lax-Wendorff methode een betere stabiliteitcriterium heeft dan de upwind methode. Hierdoor kan een grotere tijdstap gekozen in een numerieke simulatie. De maximale tijdstap is getest door het model te runnen met verschillende tijdstappen rondom het theoretische maximum. Uit de resultaten bleek dat het model op tijdstap die een fractie kleiner was dan de theoretische maximale tijdstap, stabiele oplossingen simuleerde. Zodra de tijdstap de maximale waarde behaalde werden de oplossingen instabiel. Door het toevoegen van diffusie aan het model is een extra voorgeschreven randvoorwaarde nodig. Door implementatie van verschillende randvoorwaarden aan het einde van het domein is gebleken dat de randvoorwaarde 'extrapoleren tweede orde' het best is. De dimensie-analyse van de advectie-diffusievergelijking wijst uit de eolischsedimenttransport gedomineerd wordt door advectief transport en het effect van diffusie verwaarloosd mag worden.

Photosensitizers play a key role in photodynamic therapy as a drug that is activated by light. In this thesis, the effect of ionizing radiation on the generation of singlet oxygen when using photosensitizers (Chlorin e6) and photocatalysts (titanium dioxide) was studied. By using radionuclides and photosensitizers (PS), cells can be killed with reactive oxygen species (ROS) which are produced by PS under influence of radiation that is emitted by radioactive decay. ROS are produced naturally in cells, but an excess of ROS can lead to damage to proteins, nucleic acids, lipids, membranes and organelles, which can lead to activation of cell death processes such as apoptosis. The focus of this thesis is the production of singlet oxygen which is a type of ROS.
Indium-111, lutetium-177 and yttrium-90 which decay by different types of radiation have been used as radiation sources. In this research, the influence of these different radiation sources have been tested on Ce6 and TiO2 and their ability to produce singlet oxygen. Results show that Ce6 and TiO2 have increased production of singlet oxygen when exposed to ionizing radiation. Ce6 generates singlet oxygen when irradiated by both low energy gamma sources
and high energy beta-minus radiation, while TiO2 generates singlet oxygen when irradiated by high energy beta-minus radiation.

Understanding the dynamics of bubbles in broth is crucial for various biotechnological processes conducted in bubble columns, such as the manufacturing of cultured meat. In this study, the drag coefficient of individual bubbles ascending in sodium chloride, sodium sulphate, albumin, and egg-white protein solutions was determined. This was accomplished by extracting bubble characteristics, such as the equivalent diameter and terminal velocity, from images captured by a camera system, utilizing bubble detection by a software suite. The obtained terminal velocities and computed drag coefficients were then compared to existing predictive models developed by Clift, Fan and Tsuchiya, Ishii and Zuber, and Peebles and Garber. This comparative analysis was conducted to assess the accuracy of these models in predicting bubble behavior in contaminated water. The results indicated that the drag coefficient of single bubbles rising in salt and protein solutions is higher compared to when ascending in demineralized water. The drag coefficient of single bubbles rising in protein solutions predicted by the drag model of Ishii and Zuber aligned well with the experimentally derived drag coefficients. However, both drag models tended to underpredict the drag coefficient of bubbles ascending in protein solutions. In the case of salt solutions, the drag coefficient predictions by Peebles and Garber approached the experimentally derived drag coefficients more closely although both models tended to overpredict the drag coefficient of bubbles ascending in salt solutions. Additional data acquisition of bubbles within the same size distribution rising in the examined salt and protein solutions is needed to validate and further refine these observations.

This thesis is aimed at developing a fundamental understanding of the key physical mechanisms associated with swirling gas-liquid flow in a vertical pipe, in order to construct a mechanistic model for predicting the flow behaviour. The model is based on quasi-1D reasoning, in analogy to the classical models for gas-liquid flow without swirl. As a benchmark case, it is used to investigate the onset of columnar flow patterns that are characteristic to swirling multiphase flow, for low gas input flows.

In this work neural networks are used to approximate the solutions of multiple differential equa- tions. Three main differential equations are investigated: the mass-spring equation, the heat equation and the Navier-Stokes equations coupled with the continuity equation. Each of these equations are approached differently with the current methods, therefore these equations test the applicability of the newly proposed method on a general differential equation. The results of various experiments performed in this work show that the method is able to approximate to great precision the mass-spring equation as well as the heat equation. However, the method does not perform well on the Navier-Stokes equations coupled with the continuity equation. The main advantage of this method is that meshes used by this method can often be very small compared to other approximation algorithms. When a slight decrease in performance is allowed, the method can be made completely mesh free. A huge drawback of the method is that there is no guarantee that the method will generate a reasonable approximation, for a given neural network configuration. This makes it necessary to look for a good configuration of the neural network. At this moment, this can only be done using trial and error and is therefore very time-consuming. Sometimes no good configuration is found at all. Another drawback is that the generated approximation does not satisfy the boundary and initial conditions. Multiple ways are proposed in this work to tackle this problem, but they introduce much more complexity and a performance decrease. The neural networks used in this work are generated using software called PyTorch, which comes with an automatic analytic differentiation tool. This part of PyTorch makes it possible to approximate differential equations using neural networks.