​Miša Anđelković​

• 19 December 2019
  
• Supervisors: François Peeters and Lucian Covaci
• Department of Physics

Abstract

Natan Blommaerts​

• 18 December 2019
  
• Supervisors: Silvia Lenaerts and Sammy Verbruggen
• Department of Bioscience Engineering

Abstract

The use of plasmonic nanoparticles has attracted a great deal of interest in the last 10 years among researchers in various fields of application such as photocatalysis or surface enhanced Raman spectroscopy. However, there is a large limiting factor when using precious metal nanoparticles such as gold and silver, and that is their stability. They tend to oxidize and aggregate easily, certainly in an oxidative environment such as in air. An interesting approach for stabilization of plasmonic nanoparticles is to encapsulate them in a shell, in other words to form a core-shell nanoparticle.

There are many different ways in which core-shell nanoparticles can be synthesized. In the first instance, metal nanoparticles were surrounded by a (thin) TiO2 layer. Depending on the amount of TiO2 precursor, the thickness of the layer could be controlled to a few nanometers thick. The samples were tested for the photocatalytic degradation of a solid layer of stearic acid with the addition of 2 wt% metal @ TiO2 on P25 leading to a significant improvement in degradation efficiency compared to pure P25.
Another way to stabilize metal nanoparticles is to surround them with a polymer shell. In this way, the layer thickness could be controlled with sub-nanometer control, which is a very important factor for the amount of near-field enhancement that can penetrate outside the polymer shell. An XTT test was performed to determine the oxygen activation rate of gold and silver (and gold-silver bimetallic) nanoparticles, whether or not surrounded by a (non-) conductive polymer layer. When the samples were coated with four non-conductive polymer layers, the oxygen activation dropped practically to zero. On the other hand, when gold nanoparticles were surrounded by a conductive shell, there was still oxygen activation, although lower than in the case of gold without a layer.

The final part of this thesis focused more on possible applications in air purification. In this work, a glass tube coated on the inside with (Ag@polymer-modified) TiO2 was spiraled around a UVA lamp. The optimized spiral reactor was then compared to a conventional cylindrical photoreactor, with the same dimensions and total catalyst loading, over a wide range of experimental conditions. The results showed that the spiral reactor exhibited significantly better degradation efficiencies compared to the conventional cylindrical reactor over a large range of flows. An adsorption step in combination with the optimized spiral reactor could lead to a very powerful air purification Technology

Lynn Boen​

• 17 December 2019
  
• Supervisor: Karel in 't Hout
• Department of Mathematics

Abstract

Gunnar Lumbeeck​

• 12 December 2019
  
• Supervisor: Dominique Schryvers
• Department of Physics

Abstract

Xiayang YAO​

• 10 December 2019
  
• Supervisor: Dominique Schryvers

Abstract

Shape memory alloys (SMAs) are fascinating materials that have the ability to remember their original shape under well-defined temperature or stress conditions. One of the unique shape memory effect called two-way shape memory effect (TWSME) enables the alloys having very promising applications such as an artificial anal sphincter. One of the methods to obtain the TWSME is by constrained aging which introduces aligned Ni4Ti3 precipitates in the alloy.

During constrained aging, different stresses are applied in different regions of the bulk material which eventually lead to the different distribution of precipitates. This work focuses on the study of Ni4Ti3 precipitates in low temperature aged Ni-Ti shape memory alloy with TWSME. A technique called “ACOM-TEM” is used in this study to help quantify the precipitates under different constrained aging conditions which will induce special two-way shape memory effects.

Firstly, an optimization process of the ACOM-TEM is applied to the Ni-Ti alloy. The optimization process improved the quality of results from both data collection and template matching. After the optimization, the best settings for observing the Ni4Ti3 precipitates by ACOM-TEM is obtained and is applied to detailed quantitative studies.

Secondly, the size, morphology, and distribution of Ni4Ti3 precipitates in a constrained aged Ni-Ti alloy with TWSME are studied. The results show that in the constrained-aged Ni51Ti49 alloy, a higher Ni4Ti3 precipitates density leads to a higher recovery ratio in TWSME. During constrained aging, the influence of the aging time and temperature on the volume fraction is mainly due to the changes in the interparticle spacings rather than due to the changes in the size or the shape of the precipitates. The relation between the external stress and Ni4Ti3 precipitates is also revealed.

Thirdly, a special Ni-Ti alloy which is obtained by low temperature constrained aging and shows an abnormal TWSME is studied via both conventional TEM and HRTEM. The results show that Ni4Ti3 precipitates already form in the alloy which shows an abnormal TWSME, and the average size of the precipitates is approximately 3.7 nm, with an aspect ratio of 1.6. As the aging time increased, a normal TWSME appeared. The size and aspect ratio all increased as the aging time increased, and for the critical condition in which a normal TWSME becomes an abnormal TWSME, the size of the precipitates is approximately 6 nm.

​Radu-George Ciocârlan

• 9 December 2019
  
• Supervisor: Pegie Cool
• Department of Chemistry

Abstract

Nowadays water pollution is one of the most important social issues in many countries, mostly because of the presence of organic compounds in wastewaters after their use in industrial purposes. The textile industry is well-known for the use of considerable amounts of water for the big scale processes together with dyes. On the other hand, the climate change issue is becoming a more and more delicate topic, due to the continuous increase of “anthropogenic” greenhouse gases. With a contribution of around three quarters to the total greenhouse effect, carbon dioxide (CO2) is the most important compound in terms of abundance.

We  focused on developing materials able to sustain photocatalytic processes in liquid and gas phases, in order to overcome a part of the limitations related to the use of pure TiO2. During this search, we used a lab-scale approach and model-dye solutions, as finally to test our materials on real streams of wastewater, provided by authentic textile factory. The photocatalytic process occurs when the electrons (e−) from the valence band (VB) of a semiconductor are excited by an energy that matches or exceeds the band-gap energy of the semiconductor, leading to promoted electrons to the conduction band (CB). To assist this process, we used a series of different types of materials e.g. TiO2-anatase, ferrites, freudenbergite, layered double hydroxides (LDH) and noble metals (Au and Ag).

Titanium dioxide is generally regarded as one of the most promising materials for the use in photocatalytic field, due to its strong oxidation power, its high photocatalytic activity, excellent stability, non-toxicity and low cost production. However, some limitations prevent the use of pure TiO2, related to the difficulty to recover the material after the use, the electron-hole recombination and the wide band-gap energy. In order to solve a part of those problems, we selected a second material with semiconductive properties to be used in combination with TiO2, namely ferrite systems. In the first step, simple ferrites consisting of spinel-type MFe2O4 (M = Zn, Co, Ni and Cu) were used in combination with TiO2-anatase, in order to determine the influence of each cation on the photocatalytic process. Furthermore, complex mixed ferrites were used in aggregate nanocomposites together with TiO2. An immediate advantage of using ferrite-TiO2 systems proved to be the magnetic properties induced in the nanocomposites which give the opportunity to separate the catalyst after the reaction.

​Arne Ven​

• 6 December 2019
  
• Supervisors: Sara Vicca and Ivan Janssens
• Department of Biology

Abstract

Plants take up CO2 from the atmosphere and allocate this carbon to different plant functions, including growth and respiration. The fraction of carbon allocated to different processes, i.e., carbon partitioning (expressed as a fraction of gross primary production (GPP)), can vary substantially. Despite a central role in terrestrial carbon cycling, carbon partitioning and its relationship with nutrient availability remains poorly understood and seems to be very variable. Evidence is growing that nutrient availability plays a key role in this variation, with a variable carbon cost of plant-mycorrhizal symbiosis as the hypothesized underlying mechanism.

We performed two consecutive mesocosm nutrient manipulation experiments. The central aim of our experiments was to quantitatively determine the relationship between nutrient availability and carbon partitioning to all carbon pools and carbon-consuming processes, including the role of mycorrhizal fungi therein.

In the first experiment, the biomass production efficiency (BPE) increased with increasing nutrient availability. We found strong indications that this was linked to a reduced carbon partitioning to arbuscular mycorrhizal fungi (AMF). Our data indicate that autotrophic respiration relative to GPP (Raut:GPP) remained relatively constant under increasing nutrient availability. However, the second experiment showed an opposite response: BPE was relatively constant and even tended to decrease with increasing phosphorus availability. AMF abundance was much lower than in the first experiment; hence, when phosphorus conditions were low, plants likely invested in alternative pathways for phosphorus uptake, and this was associated with a surprising reduction of BPE with increasing phosphorus fertilization, while Raut:GPP increased.

Our experimental results indicated that the reason for these fundamental differences in carbon partitioning response to fertilization between both experiments was related to the abundance of AMF, which was substantially lower in the second than in the first experiment. Also the difference between pasteurized (no AMF) and AMF-inoculated treatments indicated AMF as one of the underlying mechanisms causing this variation in carbon partitioning. AMF were very important for plants, especially in low phosphorus conditions where symbiosis appeared critical for plant survival. They were essential for efficient phosphorus uptake by the plants, even when phosphorus was added.
 

To the best of our knowledge, we presented the first experiments where carbon balance closure was pursued and achieved, strengthening the robustness of the observed patterns. Our research emphasizes the need to take into account not only nutrient availability, but also mycorrhizal symbionts when studying and modelling carbon partitioning.

​Abbas Alloul​

• 5 December 2019
  
• Supervisos: Siegfried Vlaeminck and Gilbert Van Stappen
• Department of Bioscience Engineering

Abstract

Thimo Groffen​

• 22 November 2019
  
• Supervisors: Prof. Lieven Bervoets and Prof. Marcel Eens
• Department of Biology

Abstract

Environmental pollution is coeval with the appearance of humans and the global impact of environmental pollution has been reported for multiple contaminants. However, still little is known about the recently produced and detected perfluoroalkylated substances (PFAS). PFAS are chemicals which have been detected globally as a result of their production and use in multiple consumer products. They are highly persistent and bioaccumulative and hence end up in all compartments of the ecosystem in which they may cause detrimental effects.

During my PhD, we investigated the exposure of PFAS on terrestrial invertebrates and songbirds along a distance gradient from a well-known fluorochemical hotspot near Antwerp. In addition, we determined their accumulation and potential effects on reproduction and oxidative status in songbirds. Furthermore, the transfer throughout the food chain was investigated in order to understand the exposure pathways of PFAS. Finally, we examined the role of soil physicochemical properties on the sorption of PFAS to the soil and hence the bioavailability to biota.

Werner Jacobs​

• 19 November 2019
  
• Supervisors: Kris De Baere and Dirk Lamoen
• Joint defence UAntwerp - Antwerp Maritime Academy

Abstract

Chemical tankers transport a wide variety of chemical products over sea. On the modern parcel tankers, up to 50 different cargoes can be transported on the same voyage and request special attention towards compatibility and segregation of the products. But the characteristics of the various products might also be a danger for the ship and its crew, related to flammability and toxicity. Prevention of contamination of the cargo is another major concern of a ship's officer. It is very unusual that the same cargo is loaded consecutively in the same tank, this in contrast to the transportation of crude oil. Considering the number of tanks and parcels on board modern chemical tankers, tank cleaning and gas-freeing operations are frequently occurring events, both in port and at sea. The International Maritime Organisation (IMO) states verbatim that during cargo loading, tank cleaning and gas-freeing operations on board a tanker carrying liquid bulk containing benzene, the crew is exposed to the largest risk of exposure to vapours from the products carried. Experience has shown that it has proven impossible to keep the measured vapour concentrations below acceptable levels. The presence of cargo vapours in the accommodation gives particular cause for concern and is something we investigated in more detail.

From our results we found confirmation of the IMO statement related to the tank cleaning and gas-freeing operations of tanks. We compared our results with the European Union and IMO threshold-limit values, time-weighted average. For benzene we exceeded these threshold during tank cleaning and tank ventilating. In addition, we stated some important remarks on the use of these thresholds on board ships. With these measurements we looked for the optimum position of the ventilation inlets regarding possible pollution with cargo vapours. Furthermore, the pollutant concentrations inside the engine room were relatively high during all investigations, despite the powerful ventilation.

We also performed wind tunnel simulations and concluded that the use of the tank high velocity pressure-vacuum valve to evacuate cargo vapours gave the best results for all different wind conditions, except for the situations with a wind direction from the bow. These results were confirmed by CFD-simulations, where we further focused on the on the streamlines around the ship’s superstructure.

Ilke De Boeck​

• 4 November 2019
  
• Supervisors: Sarah Lebeer and Olivier Vanderveken
• Department of Bioscience Engineering

Abstract

Bulelani Sizani​

• 30 October 2019
  
• Supervisor: Gerrit Beemster
• Department of Biology

Abstract

Jiachen Sun​

• 25 October 2019
  
• Supervisors: Marcel Eens and Igor Eulaers
• Department of Biology

Abstract

The release of a large number of persistent contaminants by human activities remains of paramount environmental concern. Long-term and large-scale spatiotemporal trend monitoring is crucial for the adequate assessment of the effectiveness of legal regulations, as well as for identifying hotspots of continued or emerging contamination. The present thesis aimed to reconstruct spatiotemporal trends of three major groups of persistent contaminants: mercury (Hg), legacy organochlorines (OCs), and perfluoroalkyl substances (PFAS).

We sampled archived body feathers of an established sentinel species – the white-tailed eagle (Haliaeetus albicilla) from the West Greenland coast (n = 124; 1884 - 2013), the Norwegian coast (n = 102; 1866 - 2015), and the central Swedish Baltic coast (n = 152; 1967-2011). We observed generally highest concentrations of Hg, OCs, and PFASs in Swedish feathers, except for chlordanes, hexachlorobenzene and perfluorooctanesulfonamide (FOSA), which were elevated in Greenland feathers. These results likely indicate the relative difference between local contaminant influx in Sweden versus the long-range transport to Greenland. Overall decreasing temporal trends of Hg and OCs were observed across all three studied subpopulations, except for Hg and PCBs in Greenland. Such declining trends suggest that the measures taken to reduce Hg and OC emissions have been generally effective. Comparably, decreasing trends of perfluorooctane sulfonate (PFOS) and FOSA in both Greenland and Norwegian feathers aligned well with the phase-out of their precursor compound. However, PFOS levels in the Swedish feathers remained elevated and are not decreasing, suggesting prolonged contamination in the Baltic Sea region. Perfluoroalkyl carboxylates (PFCAs) by contrast are still being used, and were consequently observed to continuously increase in all three subpopulations. In addition, feather corticosterone remained constant over time in the Swedish feathers and therefore, may not be a suitable physiological proxy for temporal changes in contaminant exposure.

The present PhD work underscores the suitability of archived museum feathers in reconstructing long-term spatiotemporal trends of contaminants. Since temporal trends have not been decreasing consistently across the studied subpopulations, we encourage continued monitoring. There is also a need for further regulatory efforts on reducing the emission of contaminants of legacy concern such as Hg, PCBs and PFOS, as well as legislations on the use of contaminants, such as PFCAs, which mostly are currently not restricted in production.

Jan-Willem Wolters​

• 17 October 2019
  
• Supervisors: Ralf Verdonschot, Jonas Schoelynck and Patrick Meire
• Department of Biology

Abstract

Through their form and functioning, aquatic macrophytes can have a great impact on the macroinvertebrate community in temperate lowland streams. Effects of macrophytes range from abiotic, such as the impact of growth form on water flow, to biotic, such as a macroinvertebrate food source. Although certain interactions between macrophytes and macroinvertebrate have been studied before, much is still unknown about the role of macrophytes within the aquatic food web. This thesis aims to further elucidate the role of trophic and non-trophic interactions between macrophytes and macroinvertebrates in temperate lowland streams, paying special attention to the macrophytes’ direct (e.g. direct consumption of living macrophytes and macrophyte-derived organic matter) and indirect (e.g. influence on other food sources, such as epiphytic algae and bacteria) role in the aquatic food web.

Macrophytes were observed to influence the macroinvertebrate community through a variety of mechanisms. Macrophyte physical structure was found to influence habitat complexity and water flow velocity, both of which have an effect on the macroinvertebrate community. Furthermore, an increase in macrophyte complexity led to an increased epiphyton cover, an important food source for many herbivorous macroinvertebrates. However, the net effect of macrophytes on their epiphytic biofilm was mixed, due to the combined excretion of allelochemicals and nutrients.

Stable isotope measurements revealed that certain macroinvertebrate taxa, expected to feed on epiphytic biofilm, were observed to assimilate macrophyte tissue. The fact that taxa of the scraper functional group, feeding close to the macrophyte leaf surface, were calculated to assimilate more macrophyte derived compounds than taxa classified as gatherers, feeding further away from the leaf surface, led to the hypothesis that accidental leaf erosion during grazing was the cause for the observed macrophyte consumption. Additionally, certain macroinvertebrate shredders were observed to directly consume macrophyte tissues, Potentially leading to a drastic reduction in macrophyte populations, especially when these are already subjected to other stressors. In addition to the consumption of living macrophytes, macroinvertebrate filter-feeders were also observed to consume macrophyte-derived organic matter, after its breakdown to fine particulate organic matter (FPOM). This consumption pattern was especially prevalent at the end of the growing season, when large amounts of macrophyte tissue entered the detrital food web.

The observations from thesis indicate that macrophytes have a significant effect on the functioning of the aquatic systems in which they occur, and that this effect is broader than the purely structural role that is often focused upon.

Yixing Sui​

• 16 October 2019
  
• Supervisors: Silvia Lenaerts and Siegfried Vlaeminck
• Department of Bioscience Engineering

Abstract

Katrien Sprangers​

• 14 October 2019
  
• Supervisor: Gerrit Beemster
• Department of Biology

Abstract

Understanding plant growth regulatory processes is a longstanding research topic. Different model systems have been used over the years to study growth. We use the maize (Zea mays) leaf growth zone to study growth. Its spatial developmental gradient of dividing and elongating cells, combined with its large size, enables the study of cell division and elongation within the same leaf at high spatial resolution. Since leaf growth is a three-dimensional process wherein three axes are involved, the ratio between growth in these axes defines the degree of anisotropic growth and finally leaf shape. We found that growth in the division zone is highly anisotropic and that the degree of anisotropy decreases in the elongation zone.

To reveal new molecular regulators of leaf growth, we used a forward genetic approach to characterize two newly isolated dwarf maize mutants; dil1-2 and dpl. The dil1-2 mutation was mapped to the DIL1 gene, encoding an AP2-type transcription factor. Besides its reduced growth phenotype, dil1-2 is also characterized by aberrant stomatal development, specifically affected in asymmetric divisions. The reduced growth is partially explained by stress and starvation, revealed by a transcriptome analysis, probably caused by reduced photosynthesis in response to the impaired stomatal functioning. Additionally, the transcriptome data and an interactome analysis revealed two potential direct growth inhibiting targets of DIL1, related to microtubule organization and auxin transport.

The dwarf & pale leaves (dpl) mutant is characterized by a pale leaf color which is caused by a reduction in chloroplast number and size, correlating with a reduced chlorophyll content. This has a negative effect on photosynthetic rates, which ultimately limits sugar availability, demonstrated by a severe drop in soluble and insoluble sugars. Exogenous sucrose application at the tip of the leaf restores growth rates of dpl to approximately wild type rates, proving that limited sugar availability can fully explain the growth phenotype of dpl. The dpl mutation could not be linked to a causal gene yet, but is located in an 11.58 Mb genomic region at the tail of chromosome 7.
Thus, by using the maize leaf growth zone model system to study growth inhibition in dwarf maize mutants, we have contributed to a more comprehensive understanding of leaf growth regulation. In addition, new cellular and molecular regulators that control stomatal development and chloroplast numbers were revealed.

​Cláudio Ângelo Gonçalves Gomes​

• 9 October 2019
  
• Supervisors: Hans Vangheluwe and Paul De Meulenaere
• Department of Mathematics and Computer Science

Abstract

Georgi Trenchev​

• 9 October 2019
  
• Supervisor: Annemie Bogaerts

Abstract

In this thesis, DC plasma discharges for CO2 conversion are studied via computer modelling and experiments. Two main discharge types are considered –gliding arc (GA) and atmospheric pressure glow discharge (APGD). For the GA, a 3D computational plasma model is developed, involving the innovative reverse-vortex plasma stabilization, using simplified argon chemistry and a conceptual geometry. The model is subsequently extended into a 1:1 geometry, matching the experiment, and also including a 2D CO2 model. Furthermore, turbulent heat transfer is included in the model.

The APGD reactor is studied by the means of optical emission spectroscopy (OES) and plasma modelling, showing a good agreement between the model and the experiment. The APGD is then used for CO2 conversion, and is gradually upgraded by the means of a model-experiment analysis, reaching a high conversion performance. Finally, a novel, dual-vortex reactor is developed from scratch, featuring innovative concepts for improved energy efficiency. The reactor is experimentally tested, confirming the expectations. A 3D plasma model for the reactor is developed as well.

Anneleen Rutten​

• 3 October 2019
  
• Supervisors: Herwig Leirs and Jim Casaer
• Department of Biology

Abstract

The Eurasian wild boar (Sus scrofa L.) is one of the most widespread mammals of the world. In Flanders (northern Belgium), wild boar re-emerged in 2006 after almost half a century of absence due to local extinction. During their absence, the Flemish landscape changed dramatically due to economic growth, urbanisation and agricultural intensification. This results in Flanders currently being one of the most densely human populated regions in Europe characterised by a severely fragmented and anthropogenic landscape. Due to Human-Wildlife Impacts involving wild boar, stakeholder acceptance of wild boar being part of the Flemish wildlife is challenged. Especially crop damage is raising concerns because the extent of crop damage was unknown at the onset of this PhD study. Moreover, a better understanding of drivers of wild boar expansion can give us more information on potential future distribution.

Gaining this information is the essential basis of a risk assessment with future wild boar distribution and can lead to a more efficient implementation of management tools to prevent these Human-Wildlife Impacts. Using a drone, Geographic Object-Based Image Analysis and Random Forest models, we developed a time-efficient, standardized and accurate method to assess the damaged area within a field with high resolution. By assessing both the landscape around the field taking into account habitat fragmentation, as well as field-specific characteristics, we could make predictive models based on these characteristics to assess damage risk throughout Flanders. The gained information can now help taking preventive measures more efficiently or can guide agro-technical adjustments to optimally limit damage risk.

Furthermore, a landscape genetic analysis revealed the importance of forest in explaining spatial genetic patterns while habitat fragmentation was found not to have an influence on wild boar expansion in this highly fragmented landscape. Moreover, a species distribution model allowed us to get a better understanding of the future expansion of wild boar in Flanders. With forest, maize, scrub and other low natural cover being the main factors determining habitat suitability, remaining suitable habitat in Flanders could be mapped. The gained knowledge can now be used as an essential basis to conduct a risk assessment with current en future expansion of wild boar in Flanders.

Jonas Van der Paal​

• 26 September 2019
  
• Supervisors: Annemie Bogaerts and Erik Neyts
• Department of Chemistry

Abstract

In the last decades, the field of plasma medicine, in which cold atmospheric pressure plasmas (CAPs) are used for medical applications, has grown significantly. The applications investigated in this field cover many topics, including, e.g., wound decontamination, transdermal drug and anticancer treatment. Although very promising results are being obtained from in vitro to in vivo studies, and even some first clinical trials, the need for more fundamental insight in the underlying mechanisms remains at large.
In the first chapters of this thesis, the chemistry occurring inside plasma sources, the effluent or a liquid substrate is investigated by means of chemical kinetics models. Afterwards, molecular dynamics (MD) simulations are employed to study the interactions of CAP-generated reactive oxygen and nitrogen species (RONS) as well as electric fields with biological substrates (cellular membranes or skin layers). In the last chapter, the permeation of RONS and intracellular DNA damage into synthetic cells is assessed by means of lipid vesicle experiments, which allows for a very controlled environment. These experiments are linked with the previously reported MD simulations, as we the results of these MD simulations can be used to explain our experimental observations.

Taken together, this thesis provides insight in the generation mechanisms of RONS in different sources often used in the field of plasma medicine. Furthermore, multiple hypotheses have been put forward on how these species interact with biomolecules present in substrates of interest in the use of CAPs as an anticancer therapy, as well as in the enhancement of transdermal drug delivery, which can be used as possible explanations for experimental observations.

Matthias Van der Donck​

• 25 September 2019
  
• Supervisor: François Peeters
• Department of Physics

Abstract

​Jyotsna Shrivastava​

• 23 September 2019
  
• Supervisor: Gudrun De Boeck
• Department of Biology

Abstract

The dynamics of biological components in aquatic environments and fish culture systems are constantly being challenged due to undesirable anthropogenic influences as well as intensive culture practices. As such, aquatic systems in the environment are especially at risk because they often act as the ultimate sinks for a broad spectrum of waste streams and toxicants. Fish in these systems can be confronted simultaneously with different biotic and abiotic stressors which can act independently or in concert to influence production and overall performance of fish. Therefore, it is important to consider the effects of mixture toxicity, commonly referred to as cocktail effects, in terms of both mechanisms of action and the risk assessment. In addition, variations in mechanistic adaptations are expected among various groups of fish species which determines their sensitivity (or resistivity) to challenging environments. With this background in mind, this doctoral thesis was designed to investigate a set of physiological, biochemical, metabolic, and acid-base and ion-regulatory responses that occur as compensatory strategies in a variety of piscine group to deal with challenging environments, to help getting insights into underlying adaptive strategies. We have tested the dynamics of ammonia elimination and ion-osmoregulation homeostasis along with acid-base balance in freshwater as well as marine teleosts and cartilaginous fish when challenged with a suite of discrete but closely related water borne stressers including high environmental ammonia, periods of fasting, induced swimming (exercise), reduced seawater salinity (hypo-osmotic environment) and elevated CO2. Common carp (Cyprinus carpio) was used as model a species typical of freshwater teleosts, whereas European sea bass (Dicentrarchus labrax) and spotted ratfish (Hydrolagus colliei) were employed as representative of marine teleosts and marine cartilaginous fish group respectively.

For both freshwater and marine fish, gills are the primary site for nitrogenous waste excretion, gas exchange, ion regulation, and acid–base balance, and are also the first sites that come in direct contact with the environmental stressors/pollutants. Gill cells contain a number of transporters, pumps and exchangers that are involved in maintenance of ammonia homeostasis, and are also closely linked with ion-regulation and acid-base balance.

In brief, this thesis provided an insight in how a variety of fish species modulate ammonia transporters and ion-exchangers to compensate for the possible disruption or imbalance induced by the stressors. Of particular importance, we also reported the potential role of Rhesus (Rh) glycoproteins in the gills of experimental fish as the crucial mechanism to regulate ammonia homeostasis when challenged with multiple stressors.

Michiel Wouters​

• 19 September 2019
  
• Supervisor: Wim Vanroose
• Department of Mathematics and Computer Science 

Abstract

Dynamical systems form an important class of mathematical models, they allow us to study many phenomena in physics, chemistry, biology and even economics. Typically, the study of a dynamical system involves simulating the evolution of certain states over time. Often these states converge to equilibria: states of the system that remain constant in time.

Not only are the states with this property themselves interesting to study, their dependence towards physical parameters, like the temperature or initial concentrations of certain substances, forms an important research domain as well.

In practice, the dependency between state and physical parameter is investigated with a mathematical technique called numerical continuation. Hereto different continuous, connected, curves of equilibria are approximated using other numerical methods. The collection of such curves is also referred to as a (connected) solution landscape of the dynamical system.

The thesis focuses on this numerical continuation technique, more specifically the part that consists of automatic exploration. The automatic generation of a complete solution landscape, instead of just a single curve, is meant by this. It is achieved in two steps: bifurcation points (equilibria that induce a transition in properties of the system) are approximated, and tangent directions to new curves that emanate from these points are constructed.

Current techniques for performing numerical continuation often fall short. For example, a large part of the available software is based on dense linear algebra, which does not allow their use for large-scale systems. Furthermore, there are problems in the approximation of bifurcation points, due to a property of these equilibria: the Jacobian of the system, evaluated in a bifurcation point, is possibly singular. This causes problems when, for example, the Newton method is applied in an algorithm used to approximate the point. There are also problems with the construction of tangent directions to new curves. Although they can be determined efficiently for systems without symmetry, this is not the case if certain symmetries are contained in the system.

The thesis attempts to find a solution for the described problems. For this purpose we will develop the numerical algorithms required to automatically explore (connected) solution landscapes, entirely based on sparse linear algebra.

Jarmo Fatermans​

• 18 September 2019
  
• Supervisors: Sandra Van Aert and Arjan den Dekker
• Department of Physics

Abstract

Nanomaterials have attracted increasing scientific interest, because their exact atomic structure may lead to interesting and unexpected physical and chemical properties. Due to several important developments in aberration correction technology, transmission electron microscopy imaging has become an excellent technique to visualise nanomaterials down to sub-angstrom resolution in order to understand their properties.

However, a merely visual interpretation of electron microscopy images is inadequate to obtain precise structure information. Therefore, a quantitative approach is required. Hereby, an important assumption is that the number of atomic columns in the image is known. This number can be determined visually from atomic-resolution images of beam-stable materials for which a high incoming electron dose can be used resulting in a sufficiently high signal-to-noise ratio. However, beam-sensitive and light-element materials should be imaged with a sufficiently low electron dose to avoid beam damage, which causes images of such materials to exhibit low signal-to-noise ratio and low contrast. In these cases, a visual determination of the number of atomic columns is not straightforward and may lead to biased structure information.

To overcome this problem, an alternative, quantitative method is proposed which determines the number of atomic columns for which there is most evidence in the image data. This method allows detecting atomic columns and even single atoms from atomic-resolution electron microscopy images in an automatic and objective manner. The validity and usefulness of this method have been demonstrated by analysing images of samples of different shape, size, and atom type. Moreover, besides detecting atomic columns from electron microscopy images, the proposed methodology also offers a way to evaluate the relation between image-quality measures. In addition, a superior performance to detect the correct number of atomic columns is observed as compared to alternative detection techniques.

In conclusion, the development of a new quantitative method in this thesis has pushed quantitative electron microscopy towards a more objective interpretation. The method generalises the characterisation of nanomaterials at the atomic scale in electron microscopy in order to obtain accurate and precise structure information about a material.

Kilian Gladiné​

• 17 September 2019
  
• Supervisor: Joris Dirckx
• Department of Physics

Abstract

The middle ear plays a crucial role in our ability to hear and performs very well under various circumstances. However, some pathologies or situations prohibit the middle ear from functioning as intended. In this work, the mechanical functioning of the mammalian middle ear under different circumstances was investigated using optical measurement techniques and finite-element modelling. The nonlinear characteristics of the acoustic vibrations of the middle ear at high intensity sounds were evaluated using laser Doppler vibrometry. We further investigated the source of this nonlinear behavior by developing finite element models.

Using similar models, we investigated the effect on sound transmission when the first hearing ossicle, the malleus, is fractured. We found that the limited transmission did not simply occur due to a lack of contact between the broken parts. The shape and functionality of the eardrum was affected by the malleus fracture as well, due to eardrum prestrain. However, the current models do not incorporate such prestrain of the eardrum and no experimental data was available. Therefore, we built a digital image correlation setup to investigate and measure eardrum (pre-)strain. The measured data allow for further model validation.

To facilitate model validation, we also developed a data processing technique allowing for frequency response averaging without losing individual curve characteristics. This provides modelers more accurate data to compare their simulations with. In a last project, we developed a technique to diagnose several middle ear pathologies which immobilize the ossicles. The current methods to evaluate which ossicles are immobilized depend on the surgeon’s experience. Using laser Doppler vibrometry, we developed a technique which is minimally invasive for the patient and returns quantitative data about the mobility of the ossicles. This facilitates decision making and can prevent unnecessary surgeries.  In summary, the insights in the mechanical principles of the mechanical ear obtained in this work are a step towards improved finite element models and clinical evaluation of ossicular fixations.

Tom De Schepper​

• 16 September 2019
  
• Supervisors: Steven Latré and Jeroen Famaey
• Department of Mathematics and Computer Science 

Abstract

Over the last decades, we have witnessed a tremendous increase in the utilization and availability of wireless networks and devices. This growth is largely founded by the introduction of mobile devices and the Internet of Things paradigm. In parallel, the demands and expectations of users, in terms of connectivity, bandwidth, quality, and services, have never been higher. These evolutions have led to a complex and heterogeneous situation where different devices, applications, and technologies consist next to each other at identical or overlapping physical locations, often competing for the same wireless resources. Existing management approaches typically operate in a static manner as no coordination exists across different communication technologies and devices. Despite being equipped with multiple network interfaces, modern devices tend to statically select one of the available technologies (e.g., Wi-Fi or LTE) or connection points (e.g., access points or base stations) based on predefined priorities. As such, decisions are left to the applications or, even worse, the user and it is impossible to automatically react to the inevitable dynamic disruptions or changes in the wireless context. Furthermore, different technologies tend to compete against each other in the same frequency bands.

In contrast, in this dissertation, we try to tackle the aforementioned challenges in a more transparent and fundamental manner. The main idea is that a user is only aware of the fact that its device is connected to the Internet, while the network takes care of all the underlying decision-making and the provisioning of resources. To this extent, we present the ORCHESTRA framework for seamless inter-technology network management. The framework consists of two components: a centralized controller that maintains a global network view and a virtual MAC layer that offers a single connection point to the upper layers, while transparently bonding over the underlying network technologies. As such, it introduces packet-level inter-technology handovers, load balancing, and duplication across the different underlying links by using packet matching rules. On top of the introduced ORCHESTRA framework, intelligence is needed that can utilize the features of the framework to actually optimize the network. To this extent, we present a series of algorithms that significantly increase the network-wide throughput based purely on real-time monitoring information. Finally, we explore the option of detecting traffic patterns in the wireless spectrum. This information can be used by management algorithms to further optimize the network, especially in areas with multiple overlapping networks.

​Jaume Alonso i Fernandez​

• 11 September 2019
  
• Supervisor: Sonja Hohloch
• Department of Mathematics and Computer Science

Abstract 

Completely integrable systems are mathematical models that describe systems with many conserved quantities. Each possible state of the system corresponds to a point in the phase space. The time evolution of the system draws a curve in this space that must be confined in the common level sets of the conserved quantities. This allows us to use geometric tools to obtain dynamical results.

In this work we focus on semitoric systems, a specific class of completely integrable systems in four dimensions, where one of the conserved quantities is a proper map that induces a circular action. We require further that all singularities are non-degenerate and have no hyperbolic components. These systems appear in different areas of science, such as quantum chemistry and quantum optics, and can exhibit interesting phenomena like monodromy, an obstruction to the existence of global action-angle coordinates related to the presence of focus-focus points.

Semitoric systems were classified a few years ago using five symplectic invariants. However, it was unclear how to compute these invariants in practise.  In this work we have addressed this situation by studying three families of semitoric systems: the coupled spin-oscillator, the coupled angular momenta and a special family of systems that can have two focus-focus points at the same time. By using mathematical software and exploiting the properties of elliptic integrals, we have developed different methods to compute their symplectic invariants and completed their symplectic classification. We have also discovered some extra properties of these families, such as superintegrability in a specific energy level set with vanishing twist, diverging coefficients in the Taylor series invariant and certain symmetries between the different components of the invariants.

​Niels Cautaerts​

• 4 September 2019
  
• Supervisors: Nick Schryvers and Rémi Delville
• Department of Physics

Abstract

This work is about understanding the microstructural evolution that takes place during ageing heat treatment and separately ion irradiation of the alloy DIN 1.4970, a Ti-stabilized austenitic stainless steel. The material will serve as nuclear fuel cladding in the MYRRHA reactor under development at the Belgian Nuclear Research Center SCK-CEN. Of particular interest to this thesis are nano-scale titanium carbide (TiC) particles that can form in the material. Detailing their structure, mode of formation, stability under irradiation, and interaction with other precipitating phases are the main themes of this work and were studied primarily with transmission electron microscopy (TEM) and atom probe tomography (APT).

First, the microstructural evolution of the material after heat treatment was investigated. The as-received material was characterized with respect to composition, phase composition, element partitioning and defect structure. Then, material that was heat treated in the 500-1000 °C temperature range was characterized. Between 600 and 800 °C, TiC and M23C6 precipitates form inter- and intragranularly. The size, number densities and volume fractions were estimated using a range of techniques. These results were related to the partitioning of alloying elements over the different phases as measured by APT. In addition, a detailed nanoscale study on the intragranular TiC precipitates is presented. Above 800 °C, the dominant change in the microstructure is recrystallization, which was studied by electron backscatter diffraction (EBSD). Finally, some tensile and hardness test results are presented and compared to the microstructure.

Second, an APT and TEM study of defects and precipitates in the material after ion irradiation is presented. Heat treated and pristine DIN 1.4970 material were irradiated with iron ions at 300, 450 and 600 °C up to 40 dpa, and these materials were characterized. First, the trends in Frank dislocation loop populations (size and number density) are discussed. The results were compared to literature and a cluster dynamics model. Segregation and precipitation were also studied in the different samples. TiC and G-phase precipitates form after irradiation at 450 and 600 °C conditions, and their size and number densities were compared. Differences in these quantities were explained on the basis of models from the literature, and this lead to a better understanding of how the different precipitates form under irradiation.

​Ali Parsai​

• 22 August 2019
  
• Supervisor: Serge Demeyer
• Department of Mathematics and Computer Science

Abstract

Charlie Beirnaert​

• 9 July 2019
  
• Supervisors: Kris Laukens and Adrian Covaci

Abstract

The field of metabolomics studies the end products of cellular processes, referred to as metabolites. These metabolites are small molecules that perform a vast number of functions in living organisms. Experiments are performed to study these metabolites and their behaviour in specific environments. Nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry coupled to liquid chromatography (LC-MS) are most commonly used. These techniques result in enormous amounts of data, this is especially the case for LC-MS experiments. This dissertation focusses on the data processing and analyses part following a metabolomics experiment.

In the first part of this dissertation a novel tool, speaq 2.0, was developed for the pre-processing of NMR spectra. Analogous to the methods used for the pre-processing of LC-MS data, wavelets are used to perform peak-picking on the spectra. The use of peak-picking results in a substantial decrease in data size without a large loss of information. This is contrary to the common binning or bucketing pre-processing method for NMR. The software contains additional pre-processing steps such as grouping, peak filling, etc. to convert the spectra into a data matrix ready for statistical analysis or machine learning. A main advantage of the speaq software is its broad applicability. Firstly, it is possible to augment existing metabolomics workflows by providing a better pre-processing approach for data reduction and secondly, the method is applicable to other experiments (i.e. non metabolomics) that produce one dimensional spectra. The software is freely distributed in the form of the speaq R package.

The second part of the dissertation focusses on a specific type of metabolomics experiment called dynamic metabolomics. In these experiments, data of the same sample are collected multiple times over the course of the experiment. This results in an increase in both data size, compared to standard metabolomics experiments, and in data analysis complexity, as the data are now of a longitudinal nature. Data analysis for dynamic or longitudinal metabolomics is not commonplace. In this thesis a novel method was developed that couples an expert label collection procedure, called tinderesting, to machine learning. This allows the fast screening of large datasets in search of signals relevant to the specific experimental setup. To validate this method a comparison was made to the state of the art, both for the analysis of simulated data and experimental data. We show that our method performs equal to the state-of-the-art on the simulated data but has specific benefits and performance gains when using it for analyzing experimental data. Finally, the process of combining experimental and simulated data to train a machine learning model is used to find data features with specific patterns in dynamic metabolomics data. The initial results indicate that this method can be used to find specific signals in complex data.

​Istvan David​

• 1 July 2019
  
• Supervisor: Hans Vangheluwe
• Department of Mathematics and Computer Science

Abstract

The complexity of engineered systems has increased drastically over the past decades. Pertinent examples are mechatronic and cyber-physical systems, characterized by an enormous complexity, stemming from the number and heterogeneity of the components and concerns involved in their engineering.

Due to this complexity, ensuring the correctness of the system is a challenging task.

Model-based systems engineering (MBSE) proposes modeling the system, before it gets realized. Multi-paradigm modeling (MPM), specifically, advocates modeling every aspect of the system at the most appropriate level(s) of abstraction, using the most appropriate formalism(s). In such settings, the system components are developed in parallel, enabling a more efficient engineering. Parallelism, however, gives rise to inconsistencies between the design artifacts, compromising the ultimate correctness of the system.

In this work, we argue, that managing inconsistencies in an MBSE setting is an effective heuristic for managing the ultimate correctness of the system.

First, we formulate an appropriate definition for correctness and consistency in terms of semantic properties. Then, we define a process-based approach to provide formal foundations for the detection and management of inconsistencies. The process model is transformed so that the potential inconsistencies are managed, and its transit time is minimal. This is achieved by searching through the set of potential process candidates by the techniques of multi-objective design space exploration. The exploration is guided by the transit time performance metric of the process, obtained via a DEVS-based simulation. A general mapping between process models and DEVS provided.

The utility of the approach is shown through a demonstrator, using the prototype tooling developed to support the approach.

​Hossein Ghorbanfekr Kalashami​

• 26 June 2019
  
• Supervisors: François Peeters and Mehdi Neek-Amal
• Department of Physics

Abstract 

The primary aim of the thesis is the exploration of structural and dynamical properties of graphene-based membranes and nanoconfined water within the framework of large-scale molecular dynamics (MD) simulations. A membrane can be defined as a selective barrier that can form nanochannels able to efficiently separate specific molecules and ions. Controlled transport of water molecules through membranes and capillaries has important applications in water purification and ion sieving. MD simulation is a powerful tool that helps us understand the underlying physics at the atomic scale of the interactions between water and graphene surfaces, that can have drastic effects. This thesis predicts, on the basis of intensive MD simulations, coupled with experimental measurements, that the shape of the graphene nanobubble can depend critically on the properties of the trapped substance such as the water bubble forms amorphous but layered structure due to the extreme confinement. It provides insights into the effects of the specific material and the van der Waals pressure on the microscopic details of graphene nanobubbles.

For example, nanobubbles filled with small hydrocarbons (water) have a circular (non-circular) shape. In addition, it investigates water permeation through graphene oxide membranes, predicting a significant change in the slippage dynamics of confined water in the presence of surface functional groups. This explains the extremely strong impact of commensurability induced by nanoconfinement, on the intrinsic dynamical properties of water. We study various organic solvents intercalated in a montmorillonite clay membrane, an ionic nanochannel with interlayer metal cations. We are able to explain the solvents’ observed swelling properties and anomalous permeation, using polarity and aromaticity as fundamental mechanisms at the molecular scale. The thesis also studies nitrogen-doped monolayer graphene, predicting ripples and weaker mechanical strength due to the substitutional doping. In addition, it proposes an alternative approach for controlling the stochastic motion of a graphene flake over a graphene substrate, by engineering topological defects in the substrate.

​Linyang LI​

• 25 June 2019
  
• Supervisor: François Peeters
• Department of Physics

Abstract

Graphene is the most famous 2D Dirac materials since 2004. In the past ten years, the study of graphene make large contributions to theory, experiment, and applications. Its influence is not only in physics, but also in many other fields. From the point of view of the Dirac band structure of graphene, this thesis address one main question: Can we find other elements forming a stable 2D lattice with Dirac points?

For the question, we proposed three different kinds of 2D Dirac materials. Using first-principles calculations, we predicted a new stable carbon monolayer, H4,4,4-graphyne. Adding the adjacent element of C in the periodic table, i.e. the N element, two kinds of dumbbell C4N structures can be obtained. The three monolayers all can be called C-based materials. They all show Dirac band structures with a linear relationship between E and k and the SOC effect is too weak much that we can neglect it. These C-based monolayers show high Fermi velocities, making them become promising materials for future high-speed electronic devices. Remarkable, the Fermi velocities of H4,4,4-graphyne are slightly higher than that of graphene. To our knowledge, the Fermi velocities of H4,4,4-graphyne are the highest Fermi velocities among all of the 2D carbon structures.

On the other hand, we tried to find some materials exhibiting a large SOC effect. Since it is difficult to see the SOC effect in materials containing only light elements, we focused on materials having heavy elements, such as Bi. We proposed several Bi-based 2D topological insulators. The large nontrivial band gaps make it possible to realize the QSH effect at room temperature. Remarkable, we found Dirac points when changing the SOC strength. Although large SOC effect can be found in those Bi-based 2D topological insulators, they do not show any magnetic properties. Considering magnetic proprieties, we tried transition metals and we predict a spinpolarized Dirac point in Y-based materials. Due to the transition metal Y, ferromagnetic properties can be found. The bands of spin-up and spin-down are separated and the Dirac point formed by the spin-down bands is at the Fermi level. Including SOC, a large nontrivial band gap is opened. The proposed Chern insulator 1T-YN2 can be used to realized QAH effect.

​Camille Allonsius​

• 21 June 2019
  
• Supervisors: Sarah Lebeer and Peter Delputte
• Department of Bioscience Engineering

Abstract

The role of lactobacillary glycoconjugates in keystone host-microbe and microbe-microbe interactions was explored in this PhD thesis. Lactobacilli occur in a wide variety of ecosystems, ranging from various human niches towards various animals, plants and food-related habitats, where they can have a beneficial role as “keystone taxa”. This keystone role can be conceptualized as immune modulation of the host and antimicrobial effects on surrounding micro-organisms. The molecules conferring these modulatory and antimicrobial activities to Lactobacillus are however still largely unknown. Glycoconjugates are currently underexplored, but their large structural variety and prominent position on the cell wall or in the supernatant make them interesting components to contribute to specific modulatory and antimicrobial effects.

In the PhD thesis, first an overview of the Lactobacillus cell wall structures is provided, with a special focus on the already described functionalities of glycoconjugates in relation to microbe-microbe and microbe-host interactions. Next, the genetic abilities to produce exopolysaccharides (EPS) as main glycoconjugates of interest by lactobacilli were systematically explored. This analysis showed that certain EPS characteristics are more species- or phylogroup-specific than strain-specific as generally believed. A few of these more ‘general’ characteristics could be linked to the lifestyle associated with the phylogroups, indicating that certain structural features of EPS are indeed of importance for the role Lactobacillus plays in that specific environment. As a case study, we explored a possible association between EPS structure and lifestyle adaptation for Lactobacillus casei AMBR2, one of the first Lactobacillus isolates from the upper respiratory tract.

This EPS structure showed very unique structural properties and a special role in adhesion and cytokine induction by L. casei AMBR2. Subsequently, the functional activity of EPS and glycoproteins from the model gastro-intestinal probiotic strain Lactobacillus rhamnosus GG was explored against Candida albicans as key microbe-microbe interaction of interest for multiple applications. The EPS molecules of L. rhamnosus GG appeared to be indeed, involved in the direct competition of L. rhamnosus GG with C. albicans during epithelial adhesion. However, a more pronounced role was found for the glycoprotein and cell wall hydrolase Msp1 because Msp1 was able to significantly block hyphae formation by this opportunistic pathogen. To translate these observations to an in vivo setting, the effects of L. rhamnosus GG and Msp1 on C. albicans infection in Galleria mellonella larvae was investigated.

​Inne Michielsen​

• 18 June 2019
  
• Supervisors: Annemie Bogaerts and Vera Meynen
• Department of Chemistry

Abstract

This thesis investigated dry reforming of CH4 (DRM) in a packed bed DBD reactor. The objective of this research was to improve the current understanding of plasma-catalytic interactions for CO2 splitting and DRM.

Chapter 1 not only explains why there is a need for the conversion of these greenhouse gases, but also describes the current state of the art in CO2 splitting and DRM. It indicates the need for an energy-efficient technology that selectively produces value-added chemicals, and illustrates how plasma-catalysis can be a suitable technique.

Chapter 2 further specifies the reactor configuration, analytics and materials used in this work, for plasma catalysis. All experiments had a detailed analysis of the plasma characteristics and the catalytic materials, and the results of these are shown in the Appendix.

Chapter 3 focusses on the splitting of CO2, using four different spherical packing materials in three different sizes, as well as glass wool and quartz wool. The effect of packing material and size and gap size was studied, as well as the effect of glass wool, quartz wool and the interactions between the packing and the dielectric barrier material.

Chapter 4 reveals the conversion of CO2 and CH4 for five different spherical packing materials in three different bead sizes. A detailed comparison is made with Chapter 3, and the influence of the applied packing materials on the product fractions is discussed.

Chapter 5 investigates the peculiar case of BaTiO3, performing best for pure CO2 splitting (Chapter 3) and worst in DRM (Chapter 4). This is achieved by comparing a BaTiO3 packing with different bead sizes, in different gap sizes, at multiple residence times. Moreover, a comparison is made between the conversion for CO2 and CH4 in DRM and as individual gas components.

Finally, Chapter 6 gives a general conclusion.

​Kevin Hellemans​

• 14 June 2019
  
• Supervisor: Koen Janssens
• Department of Chemistry

Abstract

This investigation presents an overview of the capabilities of LA-ICP-MS as a tool for quantitative analysis, isotopic ratios and elemental mapping. To that end, an overview of the technique is given, together with a summary on the most important parameters that govern the quality of a LA-ICP-MS measurement.

For quantitative analysis, an overview is given of the current calibration strategies applied to LA-ICP-MS, both towards selection and/or preparation of standards and data treatment. Additionally, a novel calibration strategy based on linear combination of standards was put forward. A case study was presented where the default calibration strategy from literature was applied and compared to data from SEM-EDX to provide a baseline for the performance of the technique. In a second case study, our novel approach was employed and only LA-ICP-MS data was used, yielding excellent quantitative data.

In the field of isotopic ratios, we investigated the limitations of the use of a low resolution quadrupole mass spectrometer for the determination of isotopic ratios and were able to estimate the expected precision and accuracy of the technique. The performance proved sufficient to distinguish weapon grade plutonium from global fall-out, as demonstrated by the case study in that chapter.

Finally, we presented the capabilities of the technique in elemental imaging, together with the most popular multivariate data analysis tools in imaging experiments. To compare it against other techniques as a tool for elemental imaging, a paint sample previously analysed with SEM-EDX and μ-XRD was reanalysed to provide an accurate representation of the analytical performance of LA-ICP-MS.

​Jeroen Van Walsem​

• 4 June 2019
  
• Supervisors: Siegfried Denys, Silvia Lenaerts and Bart Modde
• Department of Bioscience Engineering

Abstract

Photocatalysis has been labeled for decades as a promising technique for air purification. The principle seems straightforward and requires a photocatalyst that is immobilized on a substrate, and UV sources to activate the photocatalyst. Yet it seems that the commercialization of photocatalytic systems does not break through on the global market. The aim of this thesis is to identify and tackle the bottlenecks that impede commercialization from an application-oriented approach.

The problem of indoor air pollution is enhanced by the fact that people spend more and more time indoors and that ventilation is kept to a minimum as an energy-saving measure. This inevitably leads to an accumulation of volatile organic compounds (VOCs). Human exposure to VOCs is related to the sick building syndrome leading to complaints such as headache, fatigue and lack of concentration. In addition, exposure to VOCs is related to serious long-term health effects such as cancer or respiratory diseases. Therefore, the integration or retrofitting of a photocatalytic air purifying unit into heating, ventilation and air conditioning (HVAC) equipment has been chosen as an interesting approach.

These ventilation systems are characterized by high flow rates and the necessity of minimal pressure losses. Therefore, the permeability and the available exposed surface were selected as main selection criteria for photocatalytic substrates. After a quantitative analysis of potential substrates, borosilicate glass tubes were selected. Glass tubes can be stacked to constitute a transparent monolithic multi-tube reactor. Moreover, borosilicate glass is relatively inexpensive and has excellent UV-A light transmitting properties.

Since the operation of photocatalytic reactors is based on a complex interaction of physical and chemical processes, mathematical models were developed, supported by experimental data, that include all these phenomena as a tool for reactor design and optimization. Intrinsic kinetic parameters provide the fundamentals for these models as they describe the photocatalytic reaction rate, independent of fluid dynamics, reactor geometry and radiation field. In this work they were estimated by means of a Computational Fluid Dynamics (CFD) study, based on FTIR  experiments with a lab scale multi-tube reactor.

Finally, the aforementioned CFD approach was used to obtain insights for the light source configuration in upscaled multi-tube reactors. After taking all these insights and some practical implications into account, a final upscaled multi-tube reactor design was proposed and converted into a first built prototype. Subsequently, it was evaluated according the CEN-EN-16486-1 standard for VOC removal by the external scientific research center ‘CERTECH’.

​Mohd Shahrimie MOHD ASAARI​

• 22 May 2019
  
• Supervisor: Paul Scheunders
• Department of Physics

Abstract

Close-range hyperspectral imaging (HSI) has recently gained popularity for its application in high-throughput phenotyping platforms (HTPP), which allow for rapid and non-destructive monitoring and assessment of growth dynamics of plants throughout their vegetative stages. The integration of close-range HSI in HTPPs for the extraction of phenotypic traits is still under development and  requires input from the biological as well as the technological perspective.

The acquisition of hyperspectral data in close-range setups is challenging and requires a detailed inspection of factors related to the imaging setup, that influence the obtained spectra. Particularly, the reflectance spectra from leaves in close-range imaging are highly influenced by plant geometry and its specific alignment towards the imaging system. This induces high uninformative external variability in the recorded signals, whereas the spectral signature informing on plant biological traits remains undisclosed. In this thesis, a method to deal with these effects is developed. In the proposed method, the standard normal variate (SNV) normalization method was applied to remove linear effects and a clustering approach is employed to remove pixels that exhibit nonlinear multiple scattering effects.

Once the uninformative variability due to the external factors has been normalized, the desired information was extracted from the spectral reflectance. In this thesis, two major strategies for the interpretation of reflectance spectra with respect to phenotypic traits were developed, unsupervised and supervised data-driven approaches. In the unsupervised data-driven approach, the plant characteristics were inferred entirely from the spectral fingerprint without any prior information about the plant’s parameters. For this, a spectral similarity measure and a band selection strategy were employed to quantify subtle biological information and characterize the plant growth dynamics. In the supervised data-driven approach, machine learning regression algorithms were employed to model the relationship between the spectral variables and the plant physiological parameters. The developed models were tested for their ability to estimate four plant physiological parameters which highly correlate with the water-deficit stress factors as the proxy for drought stress responses.

The proposed methodologies were demonstrated by the study of water-stress and recovery of maize plants in a high-throughput plant phenotyping platform. The results showed that the analysis methods allow for an early detection of drought stress responses and of recovery effects shortly after re-watering. This demonstrates that close-range HSI has a high potential to become a rapid and non-destructive novel technology for high throughput phenotype studies.

Maarten Muys​

• 22 May 2019
  
• Supervisors: Siegfried Vlaeminck and Pieter Vermeir
• Department of Bioscience Engineering

Abstract

Indirect resource recovery was explored by application of struvite as phosphorus and nitrogen source for MP production. Struvite, a recovery product in wastewater treatment, was found to be a high-purity product, making it a selective barrier between waste stream and MP production. Dissolution kinetics and MP growth experiments demonstrated that struvite can be easily dosed in function of the microbial nutrient needs, while problems related to turbidity are avoided, making phototrophic MP production possible.

In contrast, direct resource recovery was considered by studying nutritional value and safety of aerobic heterotrophic bacteria (AHB) produced on processing effluents from 25 companies in the food and beverage sector. Protein and amino acid (AA) content varied notably between companies and within one company through time, while protein content was found to be correlated with nitrogen loading rate and sludge retention time, indicating that process design tailored towards protein production can increase biomass quality. Additionally, feed safety was guaranteed for most contaminants.

In the subsequent chapter, the variability in nutritional quality and safety was studied of commercial microalgae products, cultivated on primary resources. The high nutritional variability observed necessitates further optimization of cultivation conditions. It was also observed that a high protein content does not imply a high overall nutritional quality since digestibility and protein quality can still be unfavorable. Based on measured contaminants a safe consumption dose indicates that microalgae can perfectly be consumed as protein source rather than as food supplement, their current main application. In a final experimental chapter it was studied how the determined variability in microalgal biomass quality can be reduced, and how a stable, high-value biomass production can be established. The influence of harvesting time, operation mode and photoperiod was determined on the biomass, protein and essential AA productivity and optimal production parameters were identified.

In summary, MP production by means of indirect and direct nutrient recovery is promising, while further technological developments and increasing social awareness are expected to facilitate the introduction of novel MP in feed and food markets. As such, they can help to sustainably feed the growing world population.

​Jamoliddin Razzokov​

• 17 May 2019
  
• Supervisor: Annemie Bogaerts
• Department of Chemistry

Abstract

Biomedical applications of cold atmospheric plasma (CAP) are gaining increasing interest. In particular, CAP seems very promising for various applications, such as bacterial decontamination, wound healing, drug delivery and even cancer treatment. However, the underlying mechanisms at the atomic scale are not yet fully understood. One of the reasons is that the exact mechanisms are difficult to explore experimentally.

CAP generates a rich mixture of reactive oxygen and nitrogen species (RONS), which interact with living cells, inducing molecular level modifications to their components (e.g., lipids, proteins and DNA) upon oxidation. This will influence the intra- and/or intercellular signaling pathways, leading to various alterations in the cellular metabolism, which are stated to cause apoptosis, necrosis or immunogenic cell death. To better understand the effect of plasma on the cellular processes, a fundamental insight in the RONS-cell interactions and in the effect of plasma-induced oxidation, is crucial.

Complementary to experiments, computer simulations allow us to study the underlying processes with nanoscale precision. Thus, in my PhD project, I elucidated the mechanisms of RONS and glucose permeation and PS flip-flop across the native and oxidized cell membrane by means of MD simulations. Furthermore, I studied the impact of plasma oxidation to globular, fiber-like and signaling proteins in the context of wound healing, Alzheimer’s disease and cancer treatment, applying docking and MD simulations.

These findings help to reveal the complex interaction of CAP with the cell constituents in a qualitative way, and serve as an investment in order to develop CAP systems for clinical translation.

​Wim Torfs​

• 13 May 2019
  
• Supervisor: Chris Blondia
• Department of Mathematics and Computer Science 

Abstract

Timeliness and more specifically low latency and time efficient operation of networks becomes more and more important in state of the art of wireless communications. This work considers the efficiency and latency of three different networks: Wireless Sensor Network, Wireless Local Area Network and Cognitive Radio Network.

In WSNs, the timeliness presents a challenge in amongst others synchronization protocols. This work studies the hardware parts that provide an influence on the clock stability of two different platforms, and evaluates the related work by mapping it to the hardware model of both platforms. As a result of these findings, a novel synchronization protocol is proposed, which provides an abstraction of the concept timestamp.

One of the challenges in TDMA protocols is the heterogeneous character of the network, where nodes might have a different bandwidth requirement. This work discusses a novel TDMA protocol that ensures a reliable operation in such a network, while making sure that nodes with small bandwidth requirements still get regularly access to the medium. Moreover, thanks to the allocation scheme of the protocol, nodes can join or leave the network, without causing any disruption of the network operation due to schedule updates. The protocol results in a deterministic behavior in terms of buffer size and latency.

COTS hardware based on the IEEE Std 802.11n standard provides usually only support for CSMA type of medium access. However, in certain use cases, a more deterministic access to the wireless medium is required. This work elaborates on a method to modify the Linux kernel drivers in order to deliver packets according to a precise timing interval.

The final contribution of this work can be found in the area of Cognitive Radio Networks and Neighborhood Discovery protocols, where it was found that making an attempt to a Rendezvous asynchronously offers a higher performance than otherwise. By proposing an extension that induces this asynchronism, every protocol can benefit from this.

​Bart Spinnewyn​

• 30 April 2019
  
• Supervisors: Steven Latré and Juan Felipe Botero Vega
• Department of Mathematics and Computer Science

Abstract

The next generation of Internet services, e.g., self-driving cars, augmented reality and cloud robotics, requires ultra-low latency wireless communications, produces vast quantities of data and requires the speedy deployment of real-time collaborations between a wide variety of devices. While traditionally cloud services are hosted on infrastructure that is located within a single data center, these novel services have throughput and response time requirements that dictate that at least some computational tasks are executed near the location of the end user.

Centralized and geo-distributed cloud environments are worlds apart. Compared to centralized clouds, geo-distributed cloud environments are much more heterogeneous. These environments incorporate both infrastructure in data centers and infrastructure at the network edge with very limited capabilities and that is much more failure-prone. This spread on capability; reliability; connectivity; and proximity to the end-user, severely complicates the management. This thesis investigates the challenges related to the orchestration of network services across heterogeneous cloud environments and proposes novel management approaches that address these challenges.

First, we investigate how to effectively replicate data across storage nodes in these environments. We approach this problem as a runtime revenue problem, that considers both Service Level Agreements (SLAs) regarding durability and the cloud characteristics. This approach builds on a dynamic availability model that considers both the impact of failure distribution and recovery times on data loss. Second, we investigate how to protect stateless network services against a combination of node and link failure in these environments. We approach the problem of placing applications while guaranteeing a minimum availability for each application and minimizing the placement cost as a resource allocation problem. To deal with the scarcity of resources at the edge and the reliability spread, our availability-aware approach introduces protections only where they are needed. Third, we investigate how to orchestrate network services in a Network Functions Virtualization (NFV) environment. We propose orchestration algorithms that can improve the acceptance ratio and placement quality through coordination of the service composition and embedding. Not only do we develop the required orchestration algorithms for an existing service model that can generate Virtual Network Function (VNF)-Forwarding Graphs (FGs) with a tree topology, we also develop a novel service model with improved applicability and develop the required orchestration algorithms.

​Frederik Vanmeert​

• 29 April 2019
  
• Supervisors: Koen Janssens and Karolien De Wael
• Department of Chemistry

Abstract

At or below the surface of painted works of art, valuable information is present that provides insights into an object’s past, such as the artist’s technique and the creative process that was followed or its conservation history, but also on its current state of preservation. Typically, a (very) limited set of small paint samples is taken which provide direct access to the individual paint layers. The chemical build-up of these layers can then be investigated in great detail using various microscopic analytical methods. However, in recent years a new trend towards both elemental and chemical imaging techniques has been set which are capable of visualizing the (often) heterogeneous composition of painted objects on a macroscopic scale.

In this dissertation, various forms of specificity attainable with X‑ray powder diffraction (XRPD) imaging are explored: at the chemical, material and spatial level. This high specificity is illustrated throughout several applications stemming from the field of cultural heritage, both at the macroscopic (MA) and microscopic (µ) scale.

As a first step, XRPD imaging was transformed to a transportable instrument that can be employed for the in situ investigation of artworks, e.g., inside museums and conservation workshops. With this unique instrument large‑scale maps (cm2 – dm2) reflecting the distribution of crystalline phases on/below the surface of flat painted artefacts can be visualized in a noninvasive manner. In this way compound-specific information was attained which can be related to original pigments or materials that have been added in a later stage and even degradation/secondary products that have formed spontaneously inside the paint layers.

Additionally, with MA‑XRPD imaging it was possible to link quantitative information of pigment compositions and preferred orientation effects to the 2D compound‑specific distribution images, allowing for a further distinction between very similar artists’ materials. Furthermore, promising results for the limited depth-selectivity of this technique, obtained by exploiting the small shift in the position of the diffraction signals originating from the layered sequence of the pigments, are shown.

Finally, a minute paint sample from Wheat stack under a cloudy sky by Van Gogh was investigated at a synchrotron radiation facility with tomographic µ‑XRPD imaging at the microscopic scale. The high chemical and spatial specificity of this imaging method was exploited to further elucidate the degradation pathway of the red lead pigment.

​Le Tian​

• 26 April 2019
  
• Supervisors: Jeroen Famaey and Steven Latré
• Department of Mathematics and Computer Science

Abstract

The Internet of Things (IoT) introduces a novel dimension to the world of information and communication technology where connectivity is available anytime, anywhere for anything. To make this into reality, a large number of battery powered smart things need to be connected to the Internet in an energy efficient manner. The newly released IEEE 802.11ah Wi-Fi standard is considered as a very promising technology for the IoT. One of the new features of IEEE 802.11ah, named Restricted Access Window (RAW), aims to increase efficiency in face of a large number of densely deployed and energy constrained stations. It divides stations into groups, limiting simultaneous channel access to one group, therefore reducing the collision probability and increasing scalability. The IEEE 802.11ah standard, however, does not specify how to configure the RAW grouping parameters. Therefore, this thesis aims to dynamically optimize RAW configurations in real time to adapt to the network conditions.

Firstly, the implementation of IEEE 802.11ah is detailed, along with experimental results to validate it.  Subsequently, the Traffic-Aware RAW Optimization Algorithm (TAROA) is proposed. TAROA introduces a traffic estimation method to predict the packet transmission interval of each station. By using the simulation results under saturated state, TAROA assigns stations to RAW groups according to the estimated traffic conditions, in order to maximize the throughput.  A further step is made by applying surrogate modelling. A surrogate model is an efficient mathematical representation of a black box system, it is based on supervised learning, and can be accurately trained with a few labeled sample data points. This research trains models for homogeneous networks to estimate RAW performance under a wide range of network and traffic conditions.

Based on the trained models, the Model-Based RAW Optimization Algorithm (MoROA) is proposed. MoROA inherits the traffic estimation method of TAROA, using the trained model to determine the optimal RAW configuration in real time through multi-objective optimization.  Finally, an advanced surrogate model is presented that can predict performance of heterogeneous networks. As heterogeneous networks have more parameters leading to an enormous design space, the training methodology is well designed, to speed up the training process and maintain relative high model accuracy.

In summary, the research develops solutions for real-time RAW optimization to support large scale IoT networks. An open-source IEEE 802.11ah simulator, a patent, multiple journal and conference papers have been produced as a result.

​Jens Hemelaer​

• 26 April 2019
  
• Supervisor: Lieven Le Bruyn
• Department of Mathematics and Computer Science

Abstract

​Jamie MacLaren​

• 25 April 2019
  
• Supervisors: Sandra Nauwelaerts and Peter Aerts
• Department of Biology

Abstract

The shape of the mammalian forelimb is inexorably tied to the functional demands of the species in question. Mammals which utilise their forelimbs for walking and running on the ground possess a multitude of forelimb shapes and postures adapted to achieve the basic functions of gravitational support, directional change, shock absorption, and propulsion. The shape of the forelimb has undergone no greater changes in terrestrial mammals than that of the perissodactyl family Equidae (horses, zebras and asses). The earliest equids were small-bodied, four-toed (tetradactyl) forest-dwellers living approximately 56 million years ago. Through time, equids have adapted their limbs though digit reduction (from four to three to one), distal element elongation, and tendonisation of distal muscles, ultimately leading to the evolution of the modern, single-toed (monodactyl) genus Equus. To provide insights into this presumably adaptive locomotor transition, this thesis takes an alternative approach by investigating the forelimb functional morphology of another extant perissodactyl family – the Tapiridae (tapirs) – which have remained both forest-dwelling and tetradactyl in their forelimb digit condition. By quantifying the anatomy of the tapir forelimb bones and muscles, I am able to gain an understanding of the functional similarities between the forelimbs of early tetradactyl perissodactyls and their modern counterparts. Furthermore, important shifts in osteology and muscular arrangement (with known functional outcomes) can be highlighted during the transition between four and three functional forelimb digits in equids and their kin. To quantify morphology, I used a three-dimensional geometric morphometric approach, based on laser surface scans. In addition, I also utilised linear ratios such as speed proxies, lever-arm measurements; body mass estimates; phylogenetic comparative methods; and muscular architecture quantification, in order to achieve a holistic understanding of tapir forelimb functional anatomy.

This thesis has revealed the interspecific variation present in the forelimb of the crown tapir genus Tapirus, across both extant and extinct species, revealing hitherto unrecognised interspecific variation which suggests: a) differential load application across the four toes during movement, and b) the postcranial skeleton of tapirs exhibits some morphological variation independent of both size and phylogenetic relatedness. I also demonstrate that tapirs share a series of morphological features indicating functional shifts in common with Eocene equids and palaeotheres (centrally placed scapular spine; extended volar process of the magnum; reduced flexion / increased adduction and abduction of the phalanges). These features indicate that early three-toed (tridactyl) equids and palaeotheres interacted with their underfoot substrate in a different manner, with equids exhibiting highly reduced limb stability early in their shift from four to three functional digits.

The inclusion of both osteological and muscular quantification of the tapir forelimb has made it possible, and justifiable, to draw inferences on the evolution of locomotion in equids. This thesis represents the first geometric quantification of such data in a tetradactyl perissodactyl, and can act as a springboard for further study of locomotor functional morphology across the Perissodactyla.

​Julie Cautereels​

• 16 April 2019
  
• Supervisor: Frank Blockhuys
  
• Department of Chemistry

Abstract

Spectroscopic and spectrometric techniques, like MS, IR and NMR spectroscopy, have been an important tool for the physicochemical characterisation of compounds for many decades. While quantum chemical calculations are routinely used to reproduce, predict and assign IR and NMR spectra, MS is one of the few remaining techniques for which no quantum chemical methods are available for the systematic reproduction or prediction of spectra. The current methods to predict mass spectra all display some disadvantages and, therefore, we developed a new quantum-chemical based method to predict mass spectrometric fragmentation pathways, named Quantum Chemical Mass Spectrometry for Materials Science (QCMS2).

QCMS2 takes no energy supply into account in the prediction resulting in the energetics of the fragmentations to be positive: QCMS2, therefore, follows the fragmentation pathway(s) with the least resistance from precursor ion to fragments through all fragment generations. The method is based on DFT/B3LYP/6-311+G* calculations of bond orders, reaction energies for bond cleavages and activation energies for rearrangements.

During this doctoral thesis, QCMS2 was applied to predict the electron ionisation fragmentation pathways of four small organic molecules and the electrospray collision-induced dissociation fragmentation pathways of melphalan and benzocaine analogues, sphingosylphosphorylcholine and 38 X-His-Z tripeptides whereby X and Z were chosen based on their ability to form strong hydrogen bonds. In all cases, QCMS2 was able to reproduce the main features in the mass spectra of these compounds and, more importantly, QCMS2 predicts new fragmentation pathways. The final chapter of this thesis presents an evaluation of the achievements, points of improvements and future perspectives of QCMS2.

​Marleen Ramakers​

• 9 April 2019
  
• Supervisor: Annemie Bogaerts
• Department of Chemistry

Abstract

The problems that arise from climate change can no longer be ignored by our society. They are gaining more and more attention in the media and various researchers are investigating techniques to limit climate change. The latter is caused by the increasing concentration of CO2 in our atmosphere.

It is therefore necessary to limit the emission of CO2 as much as possible. A technology that is extensively investigated for the conversion of CO2 into useful chemicals and fuels is plasma technology. In this thesis the CO2 conversion in a gliding arc plasmatron (GAP) is studied. This is a plasma reactor that uses a tangential gas inlet to obtain a vortex. This configuration has several advantages over a classical gliding arc reactor.

During this doctoral research, the influence of the reactor configuration and the addition of various other gases on the CO2 conversion and energy efficiency were studied. The plasma arc has also been visualized with a high-speed camera. The research shows that the GAP is very promising for CO2 conversion. The final chapter of this thesis presents the results of a techno-economic study and proposes possible points of improvement, which are necessary for potential industrialization of the GAP.

​Eline Oerlemans​

• 25 March 2019
  
• Supervisors: Sarah Lebeer and Gilbert Donders
• Department of Bioscience Engineering

Abstract

On and in our bodies, we carry a large collection of micro-organisms, often referred to as the microbiota. Although many associate bacteria and yeasts often with infection, the majority of the members of these microbial communities are harmless or even have a positive effect on our health. In various pathologies, the balance between the beneficial and the potentially harmful microorganisms is disturbed. In this thesis, we studied two vaginal and one dermatological pathologies and investigated whether supplementation of Lactobacillus strains, often associated with health promoting effects, could alleviate the studied conditions.

In a first experimental chapter, the composition and abundance of the microbiome of aerobic vaginitis is described. This underexplored vaginal condition is associated with severe symptoms and general health treats. Although the microbiota of its microbial counterpart, bacterial vaginosis, has been studied extensively, molecular analysis of the aerobic vaginitis microbiome has been lagging behind. In these two vaginal conditions, the microbiome deviates from the normally encountered abundance of healthy lactobacilli. In contrast, the microbiota in vulvovaginal candidosis, the focus of next experimental chapter, was previously suggested to remain relatively unchanged. However, because we hypothesized that the activity and functionality of the endogenous lactobacilli could be (temporarily) inhibited by the Candida infections, here, exogenous lactobacilli other than typical vaginal taxa were screened and selected for the development of a probiotic gel for vaginal use. The effect of this gel on the vaginal bacteriome and mycobiome was evaluated in a proof-of-concept clinical trial in patients with acute vulvovaginal candidosis.

As lactobacilli have been shown to hold probiotic potential in the vaginal conditions mentioned above and since we realized that our collection contained not enough endogenous Lactobacillus isolates, various lactobacilli were isolated from vaginal samples of the previous chapters. We evaluated relevant characteristics for application, such as growth speed and lack of antibiotic resistance genes, and possible beneficial functions, such as an antipathogenic effect on Streptococcus agalacticae, a vaginal pathobiont and important cause of neonatal infections.

In the last experimental chapter, we could show a clear association between vaginal lactobacilli and the Lactobacillus taxa found on the skin. In addition, a facial cream was developed containing live lactobacilli, which was developed with a reduction of acne symptoms as purpose. This chapter describes the selection of the Lactobacillus strains, the formulation of the cream and the microbiome of healthy controls and acne patients, before, during and after treatment with the cream.

Marjolein Van Ginneken​

• 5 March 2019
  
• Supervisors: Lieven Bervoets and Ronny Blust
• Department of Biology

Abstract

Metals are posing a worldwide threat to aquatic ecosystems. In these contaminated environments, trace metals most often occur in different mixtures, in which metals can interfere with each other, producing antagonistic, synergistic or additive toxic effects. Besides pollutants, natural stressors, such as fluctuating temperature, food shortages and predators, are present as well, which could alter metal toxicity. Yet, current environmental quality standards (EQS) are mainly based on laboratory tests in which test organisms are exposed to single metals under constant, favorable conditions. Moreover, in the setting of environmental standards, behavioral endpoints, which are more sensitive than mortality, have never been used. As a result, the current EQS for metals might result in under- or overprotection of the environment.

By combining metal mixtures with natural stressors and assessing the effects on different levels of biological organization, the present thesis aimed to contribute to the development of environmentally-relevant risk assessment. Firstly, we investigated the effects of metal mixtures and the natural stressors, temperature and predation pressure, on (sublethal) endpoints of the aquatic invertebrate Asellus aquaticus. This isopod is an important decomposer in freshwater ecosystems in the northern hemisphere. We linked the effects on metal accumulation, mortality, respiration, growth and, as behavioral endpoints, feeding rate and activity to metal water and body concentrations. Secondly, metal mixture effects on a whole aquatic community were assessed in small artificial ecosystems.

We determined that natural stressors, both biotic and abiotic factors, can change the toxicity of metals. Additionally, other metals can interact as well. These interaction patterns are complex and it is difficult to predict their combined effects. Depending on the exposure concentrations, the metals in the mixture and the endpoint, stressor interactions and effects can change. The Water Framework Directive does not take these additional stressors into account and, thus, does not accurately assess the toxicity of metals in freshwater ecosystems. Therefore, we advise complementing chemical-based monitoring with effect-based tools in situ to better identify ecological risks.

​Femke Danckaers​

• 18 February 2019
  
• Supervisors: Jan Sijbers and Toon Huysmans

Abstract

The human body appears in many shapes and sizes. For product developers, it is useful to have a virtual 3D mannequin available to generate and validate their designs. Such anthropometric tools are widely available, but often provide only a simplified representation of the body, based on 1D measurements. Modification of the shape is often done in an univariate way, so 3D shape variation is not incorporated in such models. A statistical shape model (SSM) can be used as digital mannequin, because it describes the main variations of shape inside the population.

To perform shape analysis, the shapes within the population are brought into correspondence with each other by applying elastic surface registration. From a population of corresponded shapes, an SSM can be built. This model consists of the average 3D shape of the object class and the main shape variations that occur inside the shape population. The shape of an SSM can be changed by adapting the shape parameters. Those parameters are typically not linked with specific shape characteristics. Therefore, body shape modeling based on intuitive parameters is discussed in this work. Shape variation captured by an SSM is often polluted by variations in posture, which may incorrectly correlate with features and negatively affect the compactness of those models. Therefore, a framework that has low computational complexity to build a posture-invariant SSM, by capturing and correcting the posture of an instance, is shown. SSMs are typically a static representation of a population. A movement acquired by a motion capturing system is integrated in the SSM, allowing to modify its pose in a realistic way and to add pre-recorded motion to different body shapes in a realistic way.

This PhD thesis presents methods for an improved shape analysis. The methodology is not restricted to body shapes and is applicable to almost every object class that contains natural variance.

​Sofie Moyson​

• 5 February 2019
  
• Supervisors: R. Blust, S. Husson and G. Baggerman
• Department of Biology

Abstract

Industrial and natural sources of heavy metals cause an increase in metal accumulation, which can lead to serious health hazards for diverse animals including humans, resulting in a persistent (eco)toxicological concern. Although metals (mainly) occur in mixtures in natural environments, metal effects are usually studied for each metal separately. Metal mixture studies are therefore necessary to provide a solid scientific basis for setting standards for environmental protection and risk assessment. Furthermore, metal effects can be studied from the molecular to the population level and beyond. Combining effects at different organisational levels is necessary to better understand the observed toxicity.

Therefore, the aim of this study was to gain insights into the sensitivity to the selected single metals (Cu, Cd, Zn), and to investigate whether and how toxicity changes in mixtures. To assess different endpoints (mortality, locomotion, chemosensation, body length and population size), we fully exploited the benefits of the nematode Caenorhabditis elegans to investigate the toxic metal effects at different organisational levels (molecular, individual and population). Finally, the bioavailability and accumulation of these metals in the body of C. elegans were linked to the observed toxic effects.

Our study showed that most mixtures had a stronger toxic effect than the constituent single metals, resulting in additive or synergistic effects for the ZnCu, CuCd and ZnCuCd mixtures, while metals in the mixture ZnCd acted antagonistically depending on the Zn concentrations used.

The toxic results were better understood when effects on different endpoints at different levels were combined. For example, similar trends were observed for different endpoints but they were more pronounced at one level than at another. Similarly, no metal effect was observed at the molecular level, although clear effects were noticed at the individual and population level. Some parameters also seemed to be more sensitive than other parameters measured under the same conditions. For example, metal toxicity was more pronounced in terms of locomotion than of mortality and chemosensation for short-term studies at the individual level, while it was more evident on population size than body length for long-term studies at the population level.

Our findings showed that free metal ions and hence the waterborne uptake route was identified to be the best predictor for body burdens of both single metals and mixtures and for toxicity of single metals.

​Nick Verhelst​

• 4 February 2019
  
• Supervisor: Jacques Tempere
• Department of Physics

Abstract

Trapped clouds of atoms can be cooled down to nanokelvin temperatures and reach the quantum mechanical state known as 'superfluidity'. A superfluid can flow coherently (like a laser), and without friction (like a superconductor). Many flow phenomena that are known for classical fluids, like solitary waves and vortices, have their quantum counterparts in superfluids. The aim of this project is to study vortices and collections of these vortices, in dilute atomic superfluids consisting of fermionic atoms. Fermionic atoms have to pair up in order to become superfluid, so the interplay of the coherence and the interatomic interaction is paramount. This is in contrast to bosonic atoms, which can become superfluid without pairing up. The fermionic system only became experimentally available a few years ago. These experiments also reveal that the properties in dilute fermionic superfluids differ from what is known in superfluid helium and in superconductors. To do investigate these properties, we developed tailored extensions of the techniques used to described superconductors, and combine them with recent theoretical models for fermionic superfluids obtained in the TQC lab where this research will take place. With our approach, we aim to understand and explain how the differences in superflow between bosonic and fermionic superfluids arise.

​Dirk Vrebos

• 29 January 2019
  
• Supervisors: Patrick Meire and Jan Staes
• Department of Biology

Abstract

Human developments, such as increasing urbanization, have a wide range of effects on catchments and river systems. This results in profound changes in the hydrological regime and a deterioration of the water quality, which threaten the sustainable use of these systems and bring considerable costs. To manage these changing catchment characteristics sustainably, both natural and anthropogenic processes need to be understood and evaluated in an integrated approach. Over the past decades, related management concepts, such as integrated catchment management, have been translated into policy and legislation. But their actual implementation remains a challenge. Integrating the concept of ecosystem services might improve the effectiveness of these management frameworks.

The overall objective of this work is the development of methodologies that allow for the implementation of the ecosystem services concept in integrated catchment and natural resources management. But to develop these, first a good understanding of the catchments processes is required. Therefore, in the first part of this work, the impact of human development and land use patterns on flow pathways, water quantity and quality is investigated in the Nete catchment, Belgium. Different modelling and statistical analysis are used to assess spatial and temporal relationships over different scales. Special attention goes to the impact of wastewater treatment infrastructure on catchment functioning. The results signify the fundamental changes that have taken place in river dynamics and reveal a number of specific challenges a complex catchment system poses. In the second part this system knowledge is used to develop methodologies which aim to integrate the ecosystem service concept in ICM and INRM in both data rich and data scarce catchments. Indicators are developed to evaluate the supply and demand of several ecosystem services in an upstream-downstream analysis. The developed methodologies illustrate the opportunities of such an integration, but also the remaining challenges and serious limitations of such an approach are discussed.

This thesis provides evidence of the human impact on the hydrological regime of rivers, with an emphasis on the importance of the sewer system. The temporal and spatial scales at which these changes have taken place, makes it difficult to investigate the relevant processes and flow paths. This hampers the actual implementation of such integrated management concepts. Overall this thesis illustrates how human development has affected natural systems to a point where management from an integrated, system perspective has become almost impossible.

​Joachim Mariën​

• 28 January 2019
  
• Supervisors: Herwig Leirs and Jonas Reijniers
• Department of Biology

Abstract

In West Africa, the multimammate mouse (Mastomys natalensis) is the primary reservoir of Lassa virus, an arenavirus that causes severe haemorrhagic fever in humans. The disease affects between 200.000 and 300.000 people yearly with a fatality rate of 1-2%. Humans get infected by close direct or indirect contact with the rodent or its excretions, which can be through contaminated food or water, direct consumption of the rodent, or inhalation of excretion particles. Because no vaccine for use in humans exists and therapeutic options are limited, rodent control is currently assumed to be the only feasible option to control Lassa fever. However, no solid information exists on how effective rodent control really is or which control strategies would be most effective to reduce spillover. Moreover, Mastomys natalensis also hosts several other arenaviruses such as Morogoro virus, which occurs in rodent populations in Tanzania. Because Morogoro virus is not pathogenic for humans, it is considered a safe substitute for studying closely related but pathogenic arenaviruses like Lassa virus.

In this thesis, we present new insights into how arenaviruses can persist in populations of their reservoir hosts. First, by using Morogoro virus as a model, we show that arenaviruses exhibit several characteristics that can prevent extinction at low rodent densities, such as their avirulent residence and ability to initiate chronic infections in the host. The importance of other characteristics for viral persistence was also suggested (e.g. host manipulation and indirect, sexual or vertical transmission), although they were not investigated in detail. Then, based on the insights derived from Morogoro virus and additional field experiments in Guinea, we developed a mathematical model to predict the effectiveness of rodent control techniques to manage Lassa virus spillover risk to humans. Both field experiments and model simulations suggest that annual rodent elimination is pointless, as M. natalensis will reinvade houses quickly and Lassa virus prevalence will return easily to levels before control. In contrast, the models suggest that continuous elimination or vaccination of rodents are more sustainable approaches, and should be considered by policymakers in combination with other prevention strategies, such as rodent proofing of houses.

​Stein van Bezouw​

• 25 January 2019
  
• Supervisors: Wim Wenseleers and Jochen Campo
• Department of Physics

Abstract

Organic molecules with a large second order nonlinear optical (SO NLO) response have previously found interest e.g. as active components in electro-optical modulators enabling ultra-fast modulation of light and for wavelength conversion of laser beams. In this research, a hyper-Rayleigh scattering (HRS) setup is used that is unique in the world for its wavelength tunability, accuracy, and sensitivity, to determine the SO NLO response of a wide range of organic molecules.

The HRS setup is further developed with a computer-controlled wavelength tunable optical parametric amplifier to allow automated wavelength dependent HRS measurements on a series of organic SO NLO compounds. Moreover, polarized HRS measurements, which can yield valuable information on the different tensor components of the SO NLO response of a molecule, are introduced in the setup, which includes a rigorous analysis of mixing of the polarization components of the scattering center.

The SO NLO response of several classes of molecules is measured, including compounds with a response that can be switched “on” and “off” by a change in pH of the environment, and molecules that have a larger response than what is expected based on the classical view of their ground state symmetry. Finally, molecules are encapsulated into the hollow core of carbon nanotubes (CNTs) to impose macroscopic ordering of the encapsulated molecules, in an attempt to generate nanohybrids with a huge SO NLO response. A dependence of the excitation wavelength of the encapsulated molecules on the diameter of the CNT they are encapsulated in is observed, which is explained by differences in stacking geometries of the molecules inside CNTs with different diameters, and which is an important observation for the future development of such SO NLO nanohybrids.

​Jeroen Mulkers​

• 11 January 2019
  
• Supervisors: Milorad Milosevic and Bartel Van Waeyenberge
• Department of Physics

Abstract

The demand for data storage and computational power of ICT devices increases enormously. Moreover, these devices should have a low power consumption. The study of material properties is of utmost importance to meet this increasing demand. This is also true for research on new magnetic materials because they could lead to more efficient magnetic storage systems and logic devices. Over the last decade, chiral magnetic systems have been studied extensively. In this thesis, we contributed to this field of research by examining how lateral boundaries and material interfaces affect the magnetization configuration.

The ground state magnetization of thin ferromagnetic films with a perpendicular magnetic anisotropy is uniform at nanoscale; only at larger length scales, domains of ‘up’ and ‘down’ magnetization can appear. The magnetization in chiral ferromagnetic films is subjected to the Dzyaloshinskii-Moriya interaction (DMI). This chiral interaction introduces a chirality in the magnetization on a small length scale, and when strong enough, stabilizes magnetic configurations which consist out of chiral structures such as cycloids, helices, and skyrmions.

In this thesis, we showed that it is possible to confine magnetic structures such as magnetic domain walls and skyrmions in small magnetic strips and platelets. In heterochiral films — films which have a non-homogeneous DMI strength — we find a similar confinement of magnetic textures. In both cases, the existence of boundaries lead to multiple stable states. Furthermore, we derived, based on symmetry arguments, that the upper and bottom surface of the ferromagnetic film can lead to an additional DMI. This boundary-induced DMI causes a 3D deformation of the magnetization configuration.

Alongside the static magnetization configurations, we also studied the non-reciprocal behavior of spin waves in heterochiral magnets. A non-trivial refraction of spin waves occurs at interfaces between regions with a different DMI strength. By generalizing the well known Snell's law, we provide a theoretical description to describe this refraction.