Priyanka Shaw

• 21/12/2021
  
• 1 p.m.
• Venue: Campus Drie Eiken, O.02
• ​Online PhD defence
• You must present a valid Covid Safe Ticket and ID to attend this defence
• Supervisors: Annemie Bogaerts, Evelien Smits & Angela Privat-Maldonado
• Department of Chemistry

Abstract

Reactive oxygen and nitrogen species (RONS) generated by cold atmospheric plasma (CAP) can activate discrete signaling transduction pathways or disrupt redox cellular homeostasis, depending on their concentration. This makes that CAP possesses the therapeutic potential for wound healing, cancer, and other diseases. In order to effectively use CAP in the clinic, a clear understanding of the interaction of RONS with biomolecules (lipids, proteins, and nucleic acids) from the atomic to the macro scale, and their biological significance, is needed.

In this work, I have therefore studied the dual role of CAP-derived RONS, i.e., (i) in the signaling pathways involved in wound healing, and (ii) in their reaction with biomolecules to cause oxidation-mediated damage. I performed computer simulations to provide fundamental insight about the occurring processes that are difficult or even impossible to obtain experimentally. Furthermore, next to computational studies, I used both 2D and 3D tissue cultures. The 3D model allows proliferation in a more physiologically relevant geometry that stimulates the production of extracellular matrix proteins.

I investigated the treatment of human gingival fibroblasts with low doses of CAP-generated RONS. This treatment demonstrated that it can inhibit colony formation but does not induce cell death, induce the expression of metalloprotease proteins, induce extracellular matrix degradation, and promote cell migration, which could result in enhanced wound healing.

In contrast, at high concentrations, RONS can disrupt the cell membrane integrity and induce cancer cell death through oxidative stress-mediated pathways. I discovered how oxidation of the cell membrane (lipid-peroxidation) can facilitate the access of a drug (Melittin) into cancer cells, and in this way, reduce the required therapeutic dose of Melittin in melanoma and breast cancer cells (demonstrated using in vitro, in ovo and in silico approaches). Furthermore, I studied how excessive lipid-oxidation in chemoresistant pancreatic cancer cells promotes ferroptotic cell death. This was due to the stimulation of the iron-dependent Fenton reaction by targeting a redox-specific signaling network. However, upon oxidative stress, cells protect themselves via a sophisticated intracellular antioxidant system that involves the regulation of glutathione/glutathione peroxidase 4 (lipid repair enzyme). Cancer cells exhibited increased levels of intracellular RONS due to their hypermetabolism, leading to high expression of anti-oxidant systems. I, therefore, focus on the effect of reactive species on the intracellular anti-oxidant system and corresponding DNA damages in both temozolomide-sensitive as well as temozolomide-resistant glioblastoma spheroids, in a 3-dimensional tumor model with a more complex tumor microenvironment than cell monolayers.

Myrthe Van Hal

• 17/12/2021
  
• 10 a.m.
• Venue: Stadscampus, Hof van Liere - F. De Tassiszaal, Prinsstraat 13, 2000 Antwerpen
• You must present a valid Covid Safe Ticket and ID to attend this defence
• Supervisors: Silvia Lenaerts & Sammy Verbruggen
• Department of Bioscience Engineering

Abstract

Today’s society is increasingly challenged by a range of urgent environmental problems. Air pollution is one of these pressing topics. This thesis will mainly focus on the degradation of volatile organic compounds (VOCs) and particulate matter (PM) - more specifically soot. A second globally urging topic is the quest for sustainable energy production. To simultaneously target both environmental problems, a photoelectrochemical (PEC) cell will be studied in this thesis, combining air purification and sustainable energy production in a single device. Photocatalysis is used at the anode of the PEC cell to drive the air purification process, while the energy contained in the degraded compounds is (partially) recovered at the cathode, either as H2 gas or electricity. The first two experimental chapters focus on the proof of concept of such an unbiased all-gas phase PEC cell targeting VOC degradation, using both TiO2- and WO3-based photocatalysts. In the two following experimental chapters the photocatalytic soot oxidation capacity of these TiO2- and WO3-based photocatalysts was studied. In the final experimental chapter the previously obtained results were combined, striving towards an efficient, sunlight-driven and soot-degrading waste gas-to-energy PEC cell.

Weiqiang Zhu

• 16/12/2021
  
• 3 p.m.
• Venue: Campus Drie Eiken, R3
• ​Online PhD defence
• You must present a valid Covid Safe Ticket and ID to attend this defence
  
• Supervisor: Siegfried Vlaeminck
• Department of Bioscience Engineering

In Chapter 4, a return-sludge nursery concept, applying the sidestream nitritation and blending the resulting effluent with mainstream effluent to achieve an intermediate temperature and nitrogen concentrations, is proposed. That led to a 33–36% increase in nitrogen removal efficiency. Arrhenius’ expectations (10 ℃ higher temperature, θ=1.09) could only explain 49-51% of the activity increase in the nursery reactor, pointing to the role of other factors, e.g., the ~400% elevated electrical conductivity (15-16%), the 56-335% higher effluent nitrogen concentrations (12-14%), and the synergy and unknown factors (20-23%).

In Chapter 5, the N2O emissions, linked to three typical nitrite-oxidizing bacteria (NOB) suppression strategies (low dissolved oxygen (DO) level, free ammonia (FA), and free nitrous acids (FNA) treatments) were tested in a biofilm system. A low emerged DO level (~0.60 mg O2 L-1) was effective to suppress NOB activity and decrease N2O emissions, but NOB adaptation gradually appeared after 200 days. Further NOB inhibition was successfully achieved by periodical (3 hours per week) FA (~30 mg NH3-N L-1) or FNA (~3 mg HNO2-N L-1) treatments. The FA treatment promoted N2O production, while the FNA treatment had no effect. Thus, PN/A systems should be operated at relatively low DO levels with periodical FNA treatment.

In Chapter 6, the main conclusions drawn in this thesis are outlined. Beyond that, the possible design of a mainstream PN/A configuration that combined all described three technologies is demonstrated.

Samira Muhammad

• 25/11/2021
• 2 p.m.
• ​Online PhD defence
  
• Supervisors: Roeland Samson & Karen Wuyts
  
• Department of Bioscience Engineering
  

Abstract

Foliage of plants accumulate particulate matter (PM: a mixture of solid particles and liquid droplets suspended in the air) on their surfaces. Accordingly, plants can improve ambient air quality and lower the threat to human health. However, due to the differences in leaf macro and micro-morphology, PM accumulation and PM immobilization on leaf surfaces may vary between plant species. Therefore, effective plant species alongside effective leaf traits that contribute in PM reduction from ambient air need to be identified. To perform an impartial comparison between plant species, a common garden was set up at a background location away from a direct pollution source.

Ninety-six plant species each with 5 replicates were bought from one nursery and were potted and placed in the common garden. The ferro-magnetic and magnetizable component of PM was assessed using Saturation isothermal remanent magnetization (SIRM) while the morphological and anatomical leaf traits such as leaf area, leaf shape, leaf wettability, trichome and stomatal density were examined quantitatively. The outcomes of this research illustrated that plant species with high leaf trichome density and high leaf wettability were effective in PM accumulation and PM immobilization. The PM immobilization on leaf surfaces was in proportion to the net accumulated PM. Particle density by size fraction was analyzed using scanning electron microscopy (SEM) on leaves of investigated plant species. A high particle density was observed on leaves of plant species with a high trichome density. The effects of leaf traits were observed throughout this research but the assessment of magnetic PM accumulation at temporal scale indicated the effect of meteorological conditions. The PM accumulation was influenced by precipitation, wind speed (negatively) and by relative humidity, ambient PM2.5 concentrations (positively). Thus, concluding that PM accumulation on leaf surfaces is influenced by both leaf traits and meteorological conditions. The investigated plant species were also analyzed for their ecosystem services and ecosystem disservices using preference ranking organization method for the enrichments of evaluations (PROMETHEE). Based on the results we suggest that plant species such as T. cordata, T. platyphyllos, A. incana, A. campestre and P. abies as effective plant species for PM mitigation in urban environments considering both their ecosystem services and ecosystem disservices.

Niel Verbrigghe

• 23/11/2021
  
• 4 p.m.
• Venue: Campus Drie Eiken, O.01
• You must present a valid Covid Safe Ticket and ID to attend this defence
• ​Online PhD defence
• Supervisors: Sara Vicca & Ivan Janssens
• Department of Biology

Abstract

Stopping climate change is one of the greatest challenges of our generation, and soil, though overlooked by many, can be a substantial part of both the solution and the problem. More than a third of the world's carbon (C) stocks are stored in soils in the form of soil organic matter (SOM), which is more than in vegetation and the atmosphere combined. Climate warming may on the one hand accelerate the release of C from soils in the form of CO2, but on the other hand increase the productivity of autotrophs, mainly plants, leading to increased CO2 uptake. The position of this balance is crucial and determines whether soil finally releases or absorbs C and thus amplifies or mitigates climate warming.

Warming influences SOM decomposition rate and plant productivity, but the warming effect can be modified by (changes in) nutrient availability. Nitrogen has been shown to inhibit plant growth when deficient, thus negating the expected productivity increase. Although N is released during the decomposition of SOM, it has been shown that soil warming reduces the N retention capacity of soil microbes, resulting in a large part being lost. This can lead to N limitation in plants, and a smaller than expected increase in productivity. Northern ecosystems contain a disproportionate amount of soil C, but also warm up disproportionately fast, making them a crucial link in soil research. In addition, a geothermal temperature gradient on a grassland in Iceland provided a unique opportunity to analyse the effect of warming, warming duration and N availability on ecosystem dynamics and soil C storage. ​

In this PhD thesis, the effects of warming and increased N availability on plant productivity and soil C inputs, microbial community dynamics and SOM stabilisation, storage and degradation were analysed. Warming accelerated C-fluxes through the ecosystem, led to increased C inputs into the soil by plants, but reduced soil C storage. Although warming also depleted soil nitrogen, presumably resulting in increased N limitation in plants, N addition did not change most microbial dynamics, but led to an increase in plant productivity in warmed soils and C allocation to the roots. Finally, warming led to substantial C losses, caused by higher respiration that was not compensated for by higher productivity.

Daen JANNIS

• 08/11/2021
  
• 4 p.m.
• Venue: Campus Groenenborger, G.T.105
• ​Online PhD defence​
• You must present a valid Covid Safe Ticket and ID to attend this defence
• Supervisor: Johan Verbeeck
• Department of Physics

Abstract

Electron microscopy is an excellent tool which provides resolution down to the atomic scale with up to pm precision in locating atoms. The characterization of materials in these length scales is of utmost importance to answer questions in biology, chemistry and material science. The successful implementation of aberration-corrected microscopes made atomic resolution imaging relatively easy, this could give the impression that the development of novel electron microscopy techniques would stagnate and only the application of these instruments as giant magnifying tools would continue. This is of course not true and a multitude of problems still exist in electron microscopy. Two of such issues are discussed below.

One of the biggest problems in electron microscopy is the presence of beam damage which occurs due the fact that the highly energetic incoming electrons have sufficient kinetic energy to change the structure of the material. The amount of damage induced depends on the dose, hence minimizing this dose during an experiment is beneficial. This minimizing of the total dose comes at the expense of more noise due to the counting nature of the electrons. For this reason, the implementation of four dimensional scanning transmission electron microscopy (4D STEM) experiments has reduced the total dose needed per acquisition. However, the current cameras used to measure the diffraction patterns are still two orders of magnitude slower than to the conventional STEM methods. Improving the acquisition speed would make the 4D STEM technique more feasible and is of utmost importance for the beam sensitive materials since less dose is used during the acquisition.

In TEM there is not only the possibility to perform imaging experiments but also spectroscopic measurements. There are two frequently used methods: electron energy-loss spectroscopy (EELS) and energy dispersive x-ray spectroscopy (EDX). EELS measures the energy-loss spectrum of the incoming electron which gives information on the available excitations in the material providing elemental sensitivity. In EDX, the characteristic x-rays, arising from the decay of an atom which is initially excited due to the incoming electrons, are detected providing similar elemental analysis. Both methods are able to provide comparable elemental information where in certain circumstances one outperforms the other. However, both methods have a detection limit of approximately 100-1000 ppm which is not sufficient for some materials.

In this thesis, two novel techniques which can make significant progress for the two problems discussed above.

Jonathan Bogaerts

• 04/11/2021
  
• 4 p.m.
• Venue: Campus Middelheim, G0.10
• You must present a valid Covid Safe Ticket and ID to attend this defence
• ​Online PhD defence
• Supervisor: Christian Johannessen
• Department of Chemistry

Abstract

The importance of molecular structure determination is well recognized by scientist within a wide range of disciplines, including organic and physical chemists as well as pharmacists. The reason hereof is that information on the molecular structure can for example provide insight in the mode of action, create a better understanding on reaction mechanisms or can help in the rational design of new drug scaffolds. Fortunately, today many different techniques are available to help scientists in this quest. X-ray crystallography and Nuclear magnetic resonance (NMR) spectroscopy are probably the most wide-spread techniques for this purpose as they can provide structural information at atomic resolution. However, as one technique seldom permits to obtain all pieces of the structural puzzle, in most cases a combination of techniques are employed. Therefore, it is of utmost importance to understand what information can be extracted using a specific technique and understand its complementarity with regards to other methods.

This thesis focusses on exploring the application of the vibrational spectroscopic technique Raman optical activity (ROA) to determine the molecular structure of a molecule in solution. ROA is a so-called chiroptical method and is measured as the difference in right- and left circular polarized components of the Raman scattered light by a chiral molecule. In other words, ROA spectroscopy combines the wealth of structure sensitive bands present in a Raman vibrational spectrum with the sensitivity towards chirality. In the past, this unique combination has proven to be particularly useful in the study of the conformational behaviour of biomacromolecules such as proteins and carbohydrates. However, the application of ROA for molecular structure determination beyond these is rather limitedly examined and, consequently, not fully understood yet.

To start bridging this gap, in this thesis ROA was applied to a wide variety of chiral molecular to create a better understanding of both opportunities/strengths and weaknesses/limitations of ROA spectroscopy towards its applicability for structure determination.

Ahmed Hosni Ali Mohamed

• 03/11/2021
  
• 3 p.m.
• ​Online PhD defence
• Supervisor: Gerrit Beemster
• Department of Biology

Abstract

Gestational diabetes mellitus (GDM) is the most frequent health issue facing pregnant women. It is characterized by maternal glucose intolerance that first recognized during pregnancy. GDM results in aberrant placental function, severe consequences for fetal growth, and programming the biological systems of offspring to develop chronic diseases at adulthood. Following the global expansion of the diabetes epidemic, the prevalence of GDM increased to affect almost 16% of pregnancies worldwide.

Current treatment strategies for GDM include lifestyle change (healthy diet and physical activity) and/or insulin injections. If those fail to sustain normoglycemia, the oral antidiabetic drugs (as glyburide, metformin or their combination) are prescribed. The use of these medications during pregnancy is still a matter of great debate due to several adverse effects, and even less is known about their impact on the long-term health of mothers and offspring, proposing further investigations. Therefore, it is desirable to find effective new alternatives. In the context of ongoing investigations for safe and effective medication for GDM, my PhD study addressed the potential therapeutic effects of cinnamaldehyde (Ci; a promising antidiabetic agent) on the short-term impact of GDM, as well as the long-term transgenerational diabetic risk of GDM on the adult offspring, in comparison to the commonly prescribed oral hypoglycemic medication for GDM; glyburide/metformin-HCl (Gly/Met), using the fat-sucrose diet/streptozotocin rat model of GDM that simulates many of the clinical characteristics of the human disease.

First, I obtained a mechanistic understanding of the therapeutic effect of Ci in the management of placental vascular dysfunction caused by GDM through genome wide transcriptional study and targeted biochemical and histopathological analyses. Next, I investigated the efficacy of Ci in protecting the maternal/fetal hepatic and pancreatic tissues from the severe injuries induced by the diabetic stress. Finally, I established the importance of the early maternal treatment with Ci (during times of pregnancy and lactation) for the long-term health of the offspring: Ci increased the insulin sensitivity, improved the pancreatic β-cells functionality, protected from the liver metabolic derangements, and diminished the susceptibility to oxidative stress in the adult progeny. In contrast, Gly/Met intake achieves glycemic control, but does not prevent the impaired placental vascular performance, and does not provide sufficient hepatic redox balance at maternal, fetal and adult offspring levels. Future clinical research would provide further understanding of Ci efficacy as a promising alternative safe medicine for GDM.

Viveksharma Prabhakara

• 20/10/2021
  
• 4 p.m.
• ​Online PhD defence
• Supervisors: Sara Bals, Johan Verbeeck & Wilfried Vandervorst
• Department of Physics

Abstract

Scaling down the size of transistors has been a trend for several decades which has led to improved transistor performance, increased transistor density and hence the overall computation power of IC chips. The trend slowed in recent years due to reliability and power consumption issues at the nanoscale. Hence strain is introduced into transistor channels that has beneficial effects on improving the mobility of charge carriers, providing an alternative pathway for enhancing transistor performance. Therefore, monitoring strain is vital for the semiconductor industry. With the recent trend of decreasing device dimensions (FinFETS ~ 10-20nm) and strain modulation being used throughout, industry needs a reliable and fast method as quality control or defect characterisation. Such a universal strain measurement method does not exist, and one relies on a combination of quantitative in-line methods and complex off-line approaches.

In this thesis, I investigated TEM and Raman spectroscopy-based methodologies for strain measurement. In terms of TEM methodologies, advancements are made for the STEM moiré imaging, targeting strain spatial resolution enhancement. I introduce advanced quadrature demodulation and phase stepping interferometry applied to STEM moiré that greatly enhances the spatial resolution while providing enhanced field of view and sensitivity for strain measurement. I introduce ways to reduce scan distortions in strain maps using an alternative scan strategy called “Block scanning” and the non-linear regression applied for strain extraction. Prospects for 3D strain analysis using high-resolution tomography is also investigated which gives direct access for the full second order strain tensors calculation.

Finally, we compare strain measurements from TEM techniques with inline techniques like Raman spectroscopy. Raman stress measurement involves sensitive identification of the TO and LO phonon peaks. Raman spectrum of strained Ge transistor channel consists of strongly overlapping peaks within the spectral resolution of the spectrometer. Hence, the process of deconvolution of the two peaks is rather challenging. Hence, we explore new polarisation geometries like radially polarised incoming light which was shown to ease the deconvolution problem resulting in improved precision for Raman stress–strain measurements.

Yankai Xie

• 19/10/2021
  
• 3 p.m.
• Venue: Stadscampus, Klooster van de Grauwzusters, gebouw S, Lange Sint-Annastraat 7, 2000 Antwerpen
• You must present a valid Covid Safe Ticket and ID to attend this defence
• ​Online PhD defence
• Supervisor: Siegfried Vlaeminck
• Department of Bioscience Engineering

Abstract

The reactive nitrogen (Nr) has surpassed the planetary boundary three times in the past decades. The severe transgression of the safety level is not conducive to planetary sustainable development. As human activities keep generating Nr pollution, it is vital to sustainably manage the Nr on the Earth and restore humanity to the Nr boundary. In this thesis, both "Nr recovery (conversion)" and "Nr removal" were investigated to manage the point-source Nr in the nitrogen cycle.

In Chapter 2, a concept of a novel controlled mineralization and nitrification system was proposed, which can be integrated into hydroponics systems. This system is suitable for the NO₃⁻-N recovery from solid organic fertilizers. The economic assessment of nutrient production from the bioreactor systems was compared to commercially available inorganic fertilizers. Chapter 3 investigated Nr recovery from diluted urine (around 670 mg N L⁻¹) by controlled nitrification in a packed-bed trickling filter. In this concept, the TF, as a passively aerated configuration for nitrification, was proposed to produce nutrient solutions on-site for hydroponic or fertigation systems. Calcium hydroxide was used as a urine alkalization, nitrification pH control, and nutrient supplementation chemical.

In Chapter 4, packed-bed trickling filters were employed to explore the feasibility of Nr removal via the PN/A process. Synthetic wastewater containing 100-250 mg NH₄⁺-N L⁻¹ was tested to simulate household waste streams after carbon removal Interestingly, the cheap carrier based on expanded clay achieved similar rates as commercially used plastic carrier materials. The top passive ventilation combined with an optimum hydraulic loading rate of 1.8 m³ m⁻² h⁻¹ could reach approximately 60% total nitrogen (TN) removal at a rate of 300 mg N L⁻¹ d⁻¹. A relatively low NO₃⁻-N production (13%) via PN/A was achieved in TFs.

In Chapter 5, a secondary effluent polishing concept via sulfur-driven denitratation/anammox was proposed. The goal of this research was to investigate the feasibility and stability of sulfur-driven denitratation in long-term operation. Alternating the pH setpoints between 7.0 and 8.5 could temporarily stimulate the NO₂⁻-N accumulation. Both the residual NO₃⁻-N and BSNLR showed highly positive correlations with the NO₂⁻-N accumulation efficiency. Thiobacillus members may play a crucial role in managing the NO₂⁻-N accumulation, but the reduction of abundance and possible adaptation significantly impaired the efficacy of control strategies in the long run.

Miro Demol

• ​07/10/2021
  
• 4 p.m.
  
• Venue: Filmzaal Paddenhoek, Paddenhoek 1-3, Gent
  
• Wearing a face mask is recommended during the defence
  
• Supervisors: Bert Gielen, Ivan Janssens, Kim Calders & Hans Verbeeck
  
• Department of Biology (UAntwerpen) & Bio-science engineering (UGent)
  

Abstract

Improving the global monitoring of aboveground biomass (AGB) is crucial for forest management to be effective in global change mitigation and for advancing our understanding of the carbon cycle. Terrestrial laser scanning (TLS) is used to collect extremely precise and detailed 3D point clouds in forest environments. In the last decade, a range of methods have been developed to estimate AGB from TLS data. As such, TLS can address several drawbacks and uncertainties associated with conventional allometric and Earth observation methods that quantify AGB. The objective of this thesis was to evaluate and improve estimates of AGB using TLS. For this, a large suite of TLS and empirical measurements of tree size and mass were collected. Various point cloud processing and tree reconstruction modelling approaches were tested.

Wind effects, coregistration inaccuracies and reflectance scattering were major drivers for noise and point cloud inaccuracies, resulting in smaller branches (< 7 cm) to be overestimated in Quantitative Structure Models (QSMs). Stem volumes of broadleaved and coniferous temperate trees were accurately modelled using QSM, but crown volumes were usually overestimated. Several mitigation strategies were developed to correct this overestimation. Inter-tree and stump-to-tip variations in wood basic density (ρ) caused a consistent bias in TLS-derived AGB when measuring ρ at breast height or sourcing it from databases. QSM models were a helpful tool to calculate volume-weighted average ρ and as such constrain inaccuracies. Increment coring and improved knowledge of species-specific stump-to-tip ρ trends are required to obtain accurate AGB from TLS volume. A global synthesis from 10 TLS AGB validation studies featuring 393 trees ranging 13 – 43,000 kg AGB elucidated that TLS-derived AGB of smaller trees (< 1000 kg) was usually overestimated due to scattering and misalignment errors. TLS-derived AGB of big trees (> 1000 kg) on the other hand was nearly unbiased. Conversely, conventional allometric estimates of AGB were more biased than TLS.

This thesis contributed to our understanding of the capabilities and current limitations of TLS for acquiring accurate forest AGB estimates. These limitations are mainly technical and computational in nature. Therefore, larger-scale collection of ground-truth data for algorithm benchmarking is needed. TLS will be essential for calibration and validation of several forest biomass Earth observation missions and allometric scaling models.

Amir Sabzalipour

• ​05/10/2021
  
• 4 p.m.
  
• ​Online PhD defence
• Venue: Campus Drie Eiken, O.01
  
• We kindly ask you to respect social distancing during this defence
  
• Supervisor: Bart Partoens
  
• Department of Physics
  

Abstract

Novel quantum phases of matter and developing practical control over their characteristics is one of the primary aims of current condensed matter physics. It offers the potential for a new generation of energy, electronic and photonic technologies. Among all the newly found phases of matter, topological insulators are novel phases of quantum matter with fascinating bulk band topology and surface states protected by specific symmetries . For example, at the boundary of a strong topological insulator and a trivial insulator, metallic surface states appear that are protected by time-reversal symmetry. As a result, the bulk continues to be insulating, while the surface can support exotic high-mobility spin-polarized electronic states.

Since there is no such thing as a clean system, impurities and other disorders are always present in materials. Even while impurities appear to be unfavourable to a system at first look, doping the host system with impurities allows us to engineer different electronic properties of systems, such as the Fermi level or electron density. Because of the symmetry protected metallic states in topological insulators, charge transport responds distinctively to magnetic and non-magnetic impurities.

This doctoral dissertation explores how the longitudinal charge transport in magnetic topological thin films and the anomalous Hall effect on the surface of 3D magnetic topological insulators is influenced by point-like and randomly distributed dilute magnetic dopants. We are interested in how charge transport in these systems responds to the orientation of the magnetization orientation and how this response evolves based on the system's main characteristics, such as the magnitude of the Fermi level or gate voltage.

Because topological insulators have a strong spin-orbit coupling, the interaction between conducting electrons and local magnetic impurities is very anisotropic. We will show that this anisotropy even enhances when magnetic topological thin films are exposed to a substrate or gate voltage. Therefor, to properly capture this anisotropy in charge transport calculations, we rely on a generalized Boltzmann formalism together with a modified relaxation time scheme.

Maja Verstraeten

• ​04/10/2021
  
• 4.30 p.m.
  
• Venue: Stadscampus gebouw R aula R.014
  
• We kindly ask you to respect social distancing during this defence
  
• Supervisor: Nick Van Remortel
  
• Department of Physics
  

Abstract

The discovery of neutrino oscillations exposed the need for physics beyond the Standard Model. The neutrinos' behavior could be explained with additional neutrino states. Several anomalies in neutrino experiments gave rise to the hypothesis of a new sterile neutrino mass state with Δm² ~ 1 eV². The SoLid experiment will investigate flavor oscillations to the new state by measuring the dependence of the νe flux to a distance -and energy range, with a 1.6 ton, highly segmented detector at very short baseline of 6-9 m from the compact core of the 60 MW BR2 reactor of the Belgian Nuclear Research Center SCK-CEN. In addition, we will investigate the sensitivity of the SoLid detector to probe heavier sterile neutrinos (HNL) in the mass range of 1-10 MeV, by detecting their decay products. To perform the challenging measurements in a high radiation environment - close to the nuclear reactor and at Earth's surface - an innovative hybrid scintillator technology was developed, combining polyvinyltoluene and 6LiF:ZnS(Ag) scintillators into a unit cell of 5 × 5 × 5 cm³ to reach the necessary particle identification and energy reconstruction. 12800 of these unit cells make up the detector, which is read by a network of fibers and MPPCs. To validate the detector operation and to understand and predict the neutrino signal and the various background sources, a simulation framework was established. The present thesis describes how the simulation of the detector readout is developed and fine-tuned. The simulation models the energy response of the detector, converting energy deposits of passing particles to data-like output signals. This covers the photon generation and transport, the sensor response, the construction of digital waveform signals, the trigger operation and the signal readout with zero suppression. The implemented models are based on calibration measurements, detector data taken during reactor on and off periods, dedicated testbenches and specifications from manufacturers. The simulated neutrino and background events are used to construct selection requirements and train the Machine Learning models. For the oscillation analysis, a signal excess of 90 ± 22.8 IBD events per day can be reached, with a signal-to-background ratio of 1:5. The full oscillation analysis is expected this year. For the HNL analysis, the sensitivity study indicates that we have the potential to set stricter exclusion limits on the HNL parameter space in the low mass range than the previous lab experiments, namely Bugey and Triumf.

Kevin Van Sundert

• ​01/10/2021
  
• 2 p.m.
  
• ​Online PhD defence
• Venue: Campus Drie Eiken, Q002
  
• You must present a valid Covid Safe Ticket and ID to attend this defence
  
• Supervisors: Sara Vicca & Ivan Nijs
  
• Department of Biology
  

Abstract

Nutrients and water are resources vital to life on Earth. Unsurprisingly, therefore, current global atmospheric and climatic changes modifying the nutrient and water cycles have further cascading effects on the functioning of terrestrial ecosystems and services they provide to society, including biomass production and carbon storage. Despite this recognized importance, significant uncertainties regarding the full impact of global changes on terrestrial carbon cycling persist. Droughts, for example, can substantially reduce biomass production, but impacts vary widely. This variation is partly due to characteristics of the drought, but also ecosystem characteristics play an important role. Possibly, one such ecosystem characteristic, nutrient availability, plays a substantial, but overlooked, role in determining variation in drought responses. For example, nutrient rich ecosystems may be more sensitive to drought than nutrient poor ones because of reduced relative belowground investment. Such nutrient effects are overlooked, at least partly, because of the difficulty in comparing nutrient availability across scales. Nutrients may modify drought effects, but interactions in the opposite direction can also occur, as changing precipitation patterns can alter nutrient cycling. This has especially been shown for nitrogen (N), but modifications of other plant-relevant nutrient cycles are not well understood.

In this thesis, we combined database and field work to evaluate the role of nutrient availability and addition in determining drought effects on European grassland biomass production, and initiate an effort to make nutrient availability comparable across scales. We compiled a database of global change experiments that allows further disentangling the role of nutrient availability in drought and other global change responses of grasslands, forests and other ecosystems. Finally, post-drought rewetting effects on nutrients beyond N were also investigated in two subalpine mountain grasslands.

We found that addition of limiting nutrients increases drought sensitivity of graminoids in European grasslands. Elucidating the role of nutrient availability gradients in determining C cycle responses to global change is difficult because no standardized way of comparing the nutrient status exists. We therefore took important steps forward in developing a nutrient availability metric that allows comparisons of nutrient availability across sites. It was also shown that post-drought rewetting promotes release of not only N, but also other nutrients - especially K - in the soil. Databases of global change manipulation experiments can be used to further unravel the role of nutrients in carbon cycle responses to drought and global change, and should contribute to improved projections of the land carbon sink.

Yuan Hou

• ​27/09/2021
  
• 2 p.m.
  
• ​Online PhD defence
• Venue: Campus Middelheim, G.010
  
• You must present a valid Covid Safe Ticket and ID to attend this defence
  
• Supervisors: Annie Cuyt & Wen-shin Lee
  
• Department of Computer Science
  

Abstract

Latest advances in exponential analysis include a multivariate method and a Validated EXPonential Analysis (VEXPA) add on. The multivariate method can recover an n-term, d-dimensional exponential model from merely O((d+1)n) regularly collected samples, which is substantially fewer than other Prony-based multivariate methods would require. This method breaks the curse of dimensionality hence decreases the computation cost dramatically. The VEXPA method is built on a sub-Nyquist version of Prony's method. The validation technique can automatically deduce the sparsity n in the model and validate the estimated results, which is very useful under low signal-to-noise ratio conditions.

We expect that these developments can be used to overcome some of the current computational obstacles in classical exponential analysis. These computational problems are behind some of today's industrial challenges, especially when the data is sparse in one or other sense. The goal of this thesis is to investigate the new possibilities in certain signal and image processing applications.

We explore real world applications, from one-dimensional (1-D) to three-dimensional (3-D) space. In 1-D space, we deal with non-stationary signals, which are encountered widely in many engineering applications. We present a validated exponential analysis add-on for a variety of modulated signals. Our experiment results demonstrate the advantages in a number of practical applications. As for the problem of two-dimensional image analysis, we introduce an exponential analysis based decomposition method for texture decomposition. The usefulness of the method is investigated in two vision applications, namely texture classification and defect detection. In the 3-D space, exponential analysis plays a fundamental role in inverse synthetic aperture radar (ISAR) imaging. Our method can efficiently recover a large set of scattering centers from a set of noisy data.

Rani Moons

• ​23/09/2021
  
• 4 p.m.
  
• ​Online PhD defence
• Venue: CGB, U.024
  
• You must present a valid Covid Safe Ticket and ID to attend this defence
  
• Supervisors: Frank Sobott & Stuart Maudsley
  
• Department of Chemistry
  

Abstract

Over the last years the use of mass spectrometry (MS) in the structural biology field has significantly increased. Native MS approaches, structure-sensitive digestion and fragmentation, crosslinking and labeling techniques coupled to MS gained their position in the structural MS field. The information obtained includes the protein mass, subunit stoichiometry of protein complexes, which protein regions are solvent exposed or buried inside the structure, ligands interacting with the protein, ligand stoichiometry, ligand binding sites, general shape and conformational changes of the protein, protein-protein interaction sites and the protein sequence with eventual mutations or post translational modifications (PTMs). An important class of proteins are intrinsically disordered proteins (IDPs), that account for over 30% of all eukaryotic proteins. With a (partial) natively disordered structure these proteins are very challenging to characterize, due to their dynamic and heterogeneous conformational ensemble. Development and use of structural MS is important to further elucidate IDP structure since MS methods can cope with their flexible and dynamic nature.

In this thesis the IDP alpha synuclein (α-syn) is investigated using various MS approaches, to further characterize this protein and show the possibilities of MS as a structural technique in the challenging field of IDP characterisation. α-syn consists of 140 amino acids, is mainly expressed in presynaptic nerve terminals and plays a major role in the development of Parkinson’s disease (PD). α-syn monomers can aggregate and form intermediate structures such as oligomers, which then further aggregate and form mature α-syn fibrils. Various factors, e.g. mutations, PTMs, ligands, pH, presence of biological membranes, can affect this aggregation pathway. It is important to know how these changes occur at the molecular level to gain more understanding about aggregate formation and how this might be tackled.

Using native and ion mobility (IM) MS it was investigated if we can characterize interactions and conformational changes of α-syn monomers, and we show how instrumental advances in MS contribute to the structural IDP field. Native IM-MS was also used to determine possible structural effects of disease related mutations and relevant PTMs of α-syn. Finally we show that MS-based techniques can bridge the gap between molecular events that determine monomer conformations and the resulting aggregate structures. In general the value, relevance and importance of using MS-based techniques in the structural biology field to study challenging systems such as IDPs to characterize their full conformational ensemble and aggregation pathway is highlighted.

Olivier Jeunen

• ​22/09/2021
  
• 4 p.m.
  
• ​Online PhD defence
• Venue: Campus Groenenborger, G.T.129
  
• ​Registration required​
  
• You must present a valid Covid Safe Ticket and ID to attend this defence
• Supervisor: Bart Goethals
  
• Department of Computer Science
  

Abstract

Recommender systems are information retrieval applications that provide users with algorithmic recommendations, in order to assist decision-making when sufficient knowledge about the various options is lacking.

These systems have known widespread adoption in recent years and are extensively used by digital platforms to suggest restaurants, books, musical artists, retail products and even romantic partners — much like the recommendations you could provide for a friend.

Modern approaches to recommendation typically follow the "collaborative filtering" paradigm, making use of a large dataset of user behaviour to infer preferences, and to subsequently predict your preferences based on your historical behaviour.

Impressive advances in machine learning in recent years have found their way to the recommendation field, and these systems are becoming ever more accurate when it comes to learning and predicting user preferences.

There is, however, a gap between the recommendation use-case that is often posed in academic research and the use-case that practitioners typically face in industry. The work in this dissertation focuses on fundamental advances in three areas.

First, the research literature typically deals with a single model trained on a static dataset. In contrast, recommendation models in the real world are often part of a dynamic ecosystem where new data is constantly coming in and models need to be kept up-to-date to remain competitive. In such settings, a clear need arises for models that can be computed efficiently and incrementally. 

Second, newly proposed methods in the research literature are often evaluated using offline procedures on datasets containing user-item interactions. While this is common practice in the broader machine learning field, recommendation datasets often lack true “labels”.

As a result, currently existing evaluation procedures are notoriously uncorrelated with those obtained via the golden standard in industry of using online experiments such as randomised control trials — also known as A/B-tests. Third, most approaches to recommendation learn from observational datasets consisting of user-item interactions.

As they do not take any information regarding previously shown recommendations and their outcomes into account — they take a purely passive stance that focuses on prediction, which contrasts with the interventionist nature of the problem in practice. Indeed, we wish to show recommendations to users in hopes of encouraging engagement and satisfaction. As such data is being generated by every online platform with a recommendation component — it comes naturally that the value of such datasets needs to be explored.

Jelle Roegiers

• ​22/09/2021
  
• 3 p.m.
• Stadscampus, Klooster van de Grauwzusters, Gebouw S, Lange Sint-Annastraat 7, 2000 Antwerpen
  
• You must present a valid Covid Safe Ticket and ID to attend this defence
• Supervisor: Siegfried Denys
  
• Department of Bioscience Engineering
  

Abstract

Exposure to volatile organic compounds (VOCs) remains a major public health concern. Indoor VOC concentrations typically far exceed outdoor levels due to a variety of emission sources and the stringent insulation measures that are imposed today. Many attempts have been made to use photocatalysis for indoor air purification. In an ideal situation, photocatalysis is capable of complete mineralization of VOCs to H2O and CO2, without any byproduct formation. Moreover, the process can take place at standard atmospheric conditions, i.e. ambient temperature and atmospheric pressure. However, successful exploitation is still impeded due to low conversion efficiency, significant pressure loss (and hence a high energy consumption) and byproduct formation. In the first part of this thesis an attempt was made to tackles these problems by designing a novel type of photocatalytic (PCO) reactor. The PCO device consists of a cylindrical vessel filled with TiO2-coated glass tubes and equipped with UV fluorescence lamps. It was investigated in terms of fluid dynamics, coating properties, UV-light distribution and photocatalytic activity. Experimental data was later used to develop and calibrate a Multiphysics model. The model proved to be a useful tool for designing and upscaling the PCO reactor. Consequently, an optimized prototype reactor was constructed and tested according the CEN-EN-16846-1 standard for VOC removal. Although the prototype showed promising results for lab-scale conditions, it struggled with byproduct formation when purifying ppb-level VOCs. In the second part of this thesis, activated carbon adsorption was investigated in order to combine it with photocatalysis. Activated carbon fiber was opted for its fast kinetics, high adsorption capacity and thermo-electrical regeneration. The filter was studied in detail regarding the adsorption of polar and apolar VOCs at indoor air concentration levels and regeneration capabilities. Experimental data was used to develop a Multiphysics model for activated carbon adsorption as well. Consequently, a novel type of ACF filter was developed using the Multiphysics model, which was equipped with electrodes in the tips of the pleats for effective thermal regeneration. In the last part, the combination of both ACF and PCO was studied using a realistic case study. Based on the Multiphysics model, the feasibility of a so-called hybrid air purification device could be investigated. The Multiphysics model shows promising results for this hybrid PCO-ACF system and hence, a demo setup was constructed for future research.

Maksiem Erkens

• ​20/09/2021
  
• 3 p.m.
  
• ​Online PhD defence​
  
• Registration required via https://forms.gle/oUmdoR62GuQz2CKJ8
• Supervisors: Sofie Cambré & Wim Wenseleers
  
• Department of Physics
  

Abstract

Carbon nanotubes (CNTs) are hollow carbon cylinders with a diameter in the order of one nanometer, while easily reaching macroscopic lengths without any structural defects. This results in a uniquely large aspect-ratio that renders CNTs practically one-dimensional, reflecting in extraordinary electronic, vibrational and optical properties that depend critically on the exact atomic structure of the CNT, so-called chirality. These chirality-dependent characteristics make CNTs promising candidates for many optoelectronic applications. Interestingly, since every carbon atom resides at the surface, these properties are strongly influenced by both the outer and the inner environment, allowing for tuning these peculiar properties even further. However, before CNTs can be implemented in commercially viable devices, first these varying properties must be fully understood.

Although the commonly used spectroscopic techniques are incredibly useful for studying these varying properties, they cannot provide information on individual CNTs, let alone along the nanotube's length. Therefore, I developed a hyperspectral fluorescence microscopy setup that is capable of both spatially and spectrally resolving the fluorescence along the length of semiconducting single-wall CNTs (SWCNTs). With this setup, I probed the fluorescence from individual SWCNTs within a chirality-sorted sample using dedicated sample preparation, image correction and image analysis protocols in order to study the inhomogeneous linewidth broadening in the ensemble fluorescence. The resulting statistical dataset contains four distinct emission peaks that can be unambiguously assigned to fluorescence from empty and water-filled versions of the two enantiomer SWCNTs.

In a double-wall CNT (DWCNT), which consists of two coaxially aligned SWCNTs, the inter-layer interaction strongly modulates the already peculiar properties of the composing SWCNTs. Together with the many possible inner and outer layer chirality combinations, this makes the macroscopic characterisation of their structure exceptionally complex. In addition, often SWCNTs are unwittingly present in the DWCNT sample that can easily obscure the latter’s characterisation. I demonstrate this with the rapid ultrasonication-induced extraction of inner SWCNT shells from purified DWCNTs. Furthermore, I thoroughly characterised the radial breathing mode vibrations of and the fluorescence from these purified DWCNTs. Through detailed data analysis, employing two-dimensional fit models, I reveal the earlier postulated energy transfer from the inner to the outer shell of the DWCNT, explaining the inner shell fluorescence quenching. Lastly, based on these results, I developed a method with which several inner and outer layer chirality combinations in the macroscopic DWCNT sample can be identified.

Nick Sleegers

• ​20/09/2021
  
• 2 p.m.
  
• ​Online PhD defence
• Supervisor: Karolien De Wael
  
• Department of Chemistry
  

Abstract

Unnecessary exposure of bacteria to antibiotics accelerates antimicrobial resistance and thereby comprising the effectiveness of antibiotic treatments. Therefore, there is a need for better surveillance of antibiotics when released in the environment. Monitoring of antibiotic levels discharged in waste waters is one of many necessary measures. The use of electrochemistry is an inviting approach to address this surveillance need and, in general, it allows a fast, on-site and sensitive detection of low concentrations. The aim of this PhD thesis is to gain comprehensive insight in the voltammetric behavior of the cephalosporin antibiotics. Therefore, four cephalosporin antibiotics and the two main intermediates, were subjected to an electrochemical fingerprinting study. Oxidation signals of the core structure and the different side chains were elucidated. Their respective signals were further exploited to allow simultaneous detection of different cephalosporins in a single analysis below one minute. In parallel, investigations into the oxidation products were undertaken for all the redox centers of the cephalosporins. A combination of small scale electrolysis and liquid chromatography coupled to quadrupole-time-of-flight mass spectrometry was used to produce and identify the products formed during oxidation of the redox centers, which were subsequently fitted into the electrochemical frameworks provided by voltammetric analysis. This combined strategy allowed the identification of the mechanisms behind the oxidation of the three redox centers in cephalosporins. In order to develop a monitoring strategy for cephalosporins, the influence of the degradation on the electrochemical fingerprint had to be taken into account as degradation is employed to reduce antibiotic levels in industrial effluents before their release into the environment. Therefore, the changes of the voltammetric fingerprints upon degradation were investigated. It was shown that the characteristic voltammetric signals of the cephalosporin core structure disappeared upon hydrolysis. Additional oxidation signals that appeared after degradation were elucidated and linked to different inactive degradation products. Lastly, the research activity was directed towards the coupling of electrochemistry at screen printed electrodes to HPLC analysis. Essentially, the knowledge gained in the electrochemical fingerprints of cephalosporins was exploited and applied into the field of chromatographic analysis. Hereby, generating an extra dimension of separation to electrochemical analysis which will aid in the analysis of complex mixtures and matrices. The approach gave promising results as 15 antibiotics from four different classes were detected at low ppb-level.

Abner Velazco Torrejón

• ​10/09/2021
  
• 4 p.m.
  
• ​Online PhD defence​
  
• Supervisor: Johan Verbeeck
  
• Department of Physics
  

Abstract

Currently, a large variety of materials are studied by transmission electron microscopy (TEM) as it offers the possibility to perform structural and elemental analysis at a local scale. Relatively recent advances in aberration correctors and electron sources allow the instrument to achieve atomic resolution. Along with these advances, a state-of-the-art technology has been reached in TEM. However, the instrument is far from being perfect and imperfections or external sources can make the interpretation of information troublesome. Environmental factors such as acoustic and mechanical vibrations, temperature fluctuations, etc., can induce sample drift and create image distortions. These distortions are enhanced in scanning operation because of the serial acquisition of the information, which are more apparent at atomic resolution as small field of views are imaged. In addition, scanning distortions are induced due to the finite time response of the scan coils. These types of distortions would reduce precision in atomic-scale strain analysis, for instance, in semiconductors. Most of the efforts to correct these distortions are focused on data processing techniques post-acquisition. Another limitation in TEM is beam damage effects. Beam damage arises because of the energy transferred to the sample in electron-sample interactions. In scanning TEM, at atomic resolution, the increased electron charge density (electron dose) carried on a sub-Å size electron probe may aggravate beam damage effects. Soft materials such as zeolites, organic, biological materials, etc., can be destroyed under irradiation limiting the amount of information that can be acquired. Current efforts to circumvent beam damage are mostly based on low electron dose acquisitions and data processing methods to maximize the signal at low dose conditions.

In this thesis, a different approach is given to address drift and scanning distortions, as well as beam damage effects. Novel scan strategies are proposed for that purpose, which are shown to substantially overcome these issues compared to the standard scan method in TEM. ​ ​

Lore Verryckt

• ​07/09/2021
  
• 4 p.m.
  
• ​Online PhD defence
• Registration required via https://forms.gle/1CjPrPT4RbWdmbou6
• You must present a valid Covid Safe Ticket and ID to attend this defence
  
• Supervisors: Sara Vicca & Ivan Janssens
• Department of Biology

Abstract

Tropical forest are among the most productive ecosystems on earth, accounting for more than one third of global terrestrial gross primary productivity (GPP), and thereby play an important role in the global carbon cycle. GPP of tropical forests is shown to be constrained by nutrients. The two nutrients most often limiting plant growth are nitrogen and phosphorus, and the old and severely weathered soils in tropical forests are particularly low in phosphorus availability. Moreover, increases in carbon and nitrogen availability due to rising atmospheric CO2 levels and human induced N inputs to ecosystems, respectively, are not paralleled by a similar increase in phosphorus inputs, leading to stoichiometric imbalances with especially low phosphorus concentrations. It is yet unclear how foliar nutrients, and especially foliar phosphorus, affect photosynthesis in tropical forests, but that insight is essential for our understanding of terrestrial carbon cycling and to improve global models that aim to project future climate.

In this thesis, we assessed how to best measure photosynthesis in a tropical rainforest, which is challenging because of the very tall trees and strong light gradients within the dense canopies, among others. Branch excision prior to measuring photosynthesis is a commonly used method in tropical forest and we showed this method to be appropriate, and thus not affecting photosynthesis measurements. Additionally, we studied at which light level photosynthesis becomes light saturated, which is critical to avoid over- or underestimation of photosynthesis.

Foliar nitrogen has been widely identified as the major driver of net photosynthetic capacity and although foliar phosphorus is deemed to be important in tropical forest growing on phosphorus-poor soils, it has not been well-understood how phosphorus regulates the photosynthetic capacity. We found that foliar phosphorus is of major importance for leaf-level photosynthesis at the spatial scale, but surprisingly not at the vertical scale. Lastly, fertilisation experiments have been carried out to identify to what extent nutrient availability affects tropical rainforests and foliar nutrient concentrations and leaf-level photosynthesis have been shown to increase in response to fertilisation. However, three years of nitrogen and phosphorus addition did not affect foliar nutrient concentrations nor the photosynthetic capacity in French Guiana.

To conclude, we provide a framework for future measurements of leaf-level photosynthesis in tropical rainforests, guiding future tropical research, and we provide a comprehensive dataset of photosynthesis and potential drivers of carbon uptake in tropical rainforests to further improve model representations of photosynthesis. ​ ​

Adrián Pedrazo Tardajos

• ​03/09/2021
  
• 4 p.m.
  
• ​Online PhD defence​
  
• Supervisor: Sara Bals
  
• Department of Physics
  

Abstract

Transmission electron microscopy (TEM) is an ideal tool to investigate nanomaterials. The information from TEM experiments allows us to link the structure and composition of nanomaterials to their intrinsic physical properties. However, despite the significant evolution of the TEM field during the last two decades, major progress is still possible through the development of optimal TEM techniques and supports. The results presented in this thesis focus on the optimization of sample supports and their application. Among the different options, graphene has previously been reported as useful sample support for electron microscopy due to its unparalleled properties, for example, it is the thinnest known support and provides a protective effect to the sample under investigation. Unfortunately, commercial graphene grids show poor quality, in terms of intactness and cleanness, inhibiting their wide application within the field.

Therefore, this thesis focuses on the application of optimized graphene TEM grids, obtained by transferring high quality graphene using an advanced procedure. This improvement on the transfer has enabled the visualization of materials with low contrast and high sensitivity towards the electron beam, such as surface ligands capping gold nanoparticles or metal halide perovskites. Furthermore, the implemented protocol is not only of interest for conventional TEM grids but also a major benefit for in situ TEM studies, where the sample is investigated in real time under certain stimuli. Hence, the same graphene transfer technology can be also applied to advanced in situ MEMS holders dedicated for both heating and gas experiments, where the thickness and insulating nature of the silicon nitride (Si3N4) support may hamper some applications. By engineering periodic arrays of holes in their Si3N4 membrane by focused ion beam, onto which the graphene is transferred, it has been possible to get proof-of-concept 3D in situ investigations of heat-induced morphological and compositional transformations of complex nanosystems. As an example, it has enabled the investigation of the possible phase-transition of metal halide perovskites upon heating using 2D and 3D structural characterization. Moreover, it has allowed the study of in situ three-dimensional nanoparticle dynamics during gas phase catalysis as well as the first steps that would lead towards the design and creation of the first Graphene Gas Cell. Consequently, implementation of the advanced graphene transfer technology described in this thesis is envisaged to impact a broad range of future experiments.

Maaike Griffioen

• ​02/09/2021
  
• 3 p.m.
  
• ​Online PhD defence
• Supervisors: Wendt Müller & Arne Iserbyt
  
• Department of Biology
  

Abstract

Biparental care requires that two unrelated individuals raise their offspring together, which increases offspring survival and therefore parental fitness. Each parent has to invest in care, which comes with an individual cost and thus both parents benefit by investing less to retain resource for future reproductive events. Thus, parents are in a sexual conflict about investment, and they have to agree upon the level of investment into the current reproductive event. Recently, the suggestion was made that the strategy of conditional cooperation could provide a resolution for this conflict between parents. It implies that parents will follow up on each other’s provisioning visits and that a visit of the partner motivates the other to increase its own provisioning rate. This indicates that parents will have an equal and higher visit rate. However, some important aspects are still unknown, while vital for our understanding of the adaptive significance of this strategy. Therefore, the aim of this thesis is to investigate whether conditional cooperation, via turn taking of provisioning visits, indeed bears the potential for conflict resolution. I studied a wild population of blue tits, a species with biparental care, on which I performed multiple manipulation experiments in the field. Parents had stable alternation levels, which they maintained regardless of any disturbance. Yet, parents were expected to change their alternation level in accordance with the amount of conflict. Furthermore, parents were not responding as predicted by changing their provisioning rate to the rate of their partner.

However, parents were not using prey to exploit their partner, by investing less into bigger profitable prey items. Blue tits, therefore, seem to be honest in their investment in terms of prey sizes. Regardless of the honesty in prey sizes, the offspring did not benefit but also were not limited by the conditional cooperation strategy. From the experiment in which females’ brooding duration was manipulated, females were not restricted by this other care task in their potential to take turns. Taken all together, blue tit parents did not seem to be using turn taking rules as predicted by the initial theoretical concept. Yet, they seem to maintain a fixed level of alternated visits. A specific fraction of monitoring the partner’s feeding behaviour might provide a sufficient estimate of its investment and could serve as a signal to avoid substantial exploitation. Hence, partners may use different information streams to co-adjust their behaviour.

Inge Dox

• ​02/09/2021
  
• 3.30 p.m.
• ​Online PhD defence
• Registration required via https://docs.google.com/forms/d/e/1FAIpQLSdjbz5O3eIpETUDGGJaED_YK_-KMw4MIUeBKJ_IaeQy4ianEg/viewform?usp=sf_link
• Supervisor: Matteo Campioli
  
• Department of Biology
  

Abstract

Most phenological studies of deciduous forest ecosystems have investigated the onset of the growing season, mainly by characterizing the timing of canopy development and cambium reactivation, while autumn phenological events have been largely overlooked.

The end of cambial activity, cessation of xylem formation (or wood growth) and onset of foliar senescence are key autumn phenological events in temperate deciduous trees. Their timing is fundamental for the development and survival of trees, ecosystem nutrient cycling and the seasonal exchange of matter and energy between the biosphere and atmosphere, which all affect the impact and feedback of forests to global change.

In this work, we advanced the knowledge about autumn leaf- and wood formation phenology with a large-scale experimental effort and improved observational methods. We observed the end of cambial activity, cessation of wood growth, onset of foliar senescence and onset of foliar decoloration for different deciduous tree species (silver birch, European beech, pedunculate oak and common aspen) at different locations over Europe with different climates (Southern Norway, Belgium and Northern Spain) and over three growing seasons with different climatic conditions (2017, 2018 and 2019).

The main findings comprised: (i) the description of wood growth cessation dynamics in autumn which showed advances in years with very dry conditions but showing inter-species variability, (ii) the definition of an autumn phenology timeline, with wood growth cessation occurring before or concurrently with the onset of foliar senescence, except during a year following a very severe drought, and (iii) the report that the wood growing season length and the timing of crucial wood phenological phases varied more across locations than over years and were related to temperature and climate zone but did show only few general trends across species.

In summary, we filled important gaps in knowledge about autumn phenology of temperate deciduous forests and provided new unique datasets about these dynamics. Together, these new insights allow to better understand the functioning of deciduous trees and their possible future response to global change.

Annelies De wael

• ​30/08/2021
  
• 10 a.m.
  
• ​Online PhD defence
• Supervisors: Sandra Van Aert & Annick De Backer
  
• Department of Physics
  

Abstract

Nanomaterials can exhibit highly interesting properties for a variety of promising applications, ranging from sunscreen to batteries in electrical cars. A nanometer is a billion times shorter than a meter. At this scale, a material’s properties can be radically different as compared to bulk materials at larger scales. Furthermore, the properties of nanomaterials are strongly size and shape dependent. Small differences in the atomic positions, of the order of a picometer (yet a thousand times smaller than a nanometer), can already drastically change physical properties. Therefore, a reliable quantification of the atomic structure is of crucial importance in order to evolve towards materials design and to understand the structure-properties relationship of nanomaterials. Additionally, the atomic structure of nanomaterials can change over time, as a result of various physical processes. 

The research presented in this thesis enables the reliable quantification of dynamic structural changes of nanomaterials at the atomic scale using scanning transmission electron microscopy (STEM). I could realise this by developing methods to count the number of atoms “behind” each other in the atomic columns of the nanomaterial from electron microscopy images. An important difference with single frame counting procedures is that the probability for atomic columns to lose or gain atoms from one frame to the next during the time series is estimated. This quantitative method opens up new possibilities for unravelling the time dependent structure-properties relation of a nanomaterial, which can ultimately lead to the more efficient design and production of nanomaterials with innovative applications.

Bert Thys

• ​30/08/2021
  
• 5 p.m.
  
• ​Online PhD defence
• Supervisors: Marcel Eens & Rianne Pinxten
  
• Department of Biology
  

Abstract

Within animal populations, competition is often most intense among individuals of the same sex since such individuals require the same limited resources to maximize reproductive success. Although competition between females is widespread – including in species with traditional sex-roles – whether and how selection is acting on individual differences in female competitive trait expression remains poorly understood. Crucially, many traits with a potential role in female competition (e.g. same-sex aggression, plumage traits in birds) are typically repeatedly expressed across an individual’s life and hence (co)vary both among and within individuals. Moreover, behavioural traits involved in female competition should not be studied in isolation since they may covary with other ecologically relevant behavioural traits, forming a behavioural syndrome. In this thesis, using a free-living songbird and repeated measurement designs on the same individuals within and across years, I integrated female competitive trait expression into a multivariate framework with other behaviours (antipredator boldness as manifested by hissing behaviour, exploratory behaviour in a novel environment). Subsequently, revolving around the notion of life history trade-offs, I aimed at identifying the short-term and lifetime fitness consequences associated with individual differences in female behavioural phenotypes.

Females were found to consistently differ in their average level of same-sex aggression, hissing behaviour and exploratory behaviour (i.e. personality). Moreover, female aggression formed a behavioural syndrome with exploratory behaviour, but not with hissing behaviour. At the same time, substantial amounts of the total phenotypic variation in behaviours occurred on the within-individual level (i.e. behavioural plasticity). Using a behavioural reaction norm approach, I demonstrated that a part of the within-year plasticity in hissing behaviour was explained by the reproductive value of the offspring; while individual plasticity in female aggression was instead related to female age, where aggression decreased within females across their lifespan, potentially reflecting senescence. Female aggression also covaried with variation in a melanin-based black plumage trait, although depending on female age. Proximately, individual differences in female aggression and hissing behaviour were not associated with genomic sequence variation in a candidate gene, the serotonin transporter gene. Ultimately, and contrasting predictions, female aggression and hissing behaviour did not mediate the trade-off between current reproduction and future survival. Instead, results point towards heterogeneous forms of selection as a likely mechanism maintaining the observed individual differences in female behavioural (including competitive) phenotypes in the wild.

Wouter Van Den Haute

• ​25/08/2021
  
• 5 p.m.
  
• ​Online PhD defence​
  
• Supervisors: Wendy Lowen (UA) and Mark Sioen (VUB) - Joint PhD with VUB
  
• Department of Mathematics
  

Abstract

The study of frames is also known as pointfree topology. As the name suggests, this is a way of studying spaces without (mentioning) points. This idea is more natural than one might initially think: when drawing a point on paper, we do not draw an actual point, but rather a “spot”, which can be reduced in size if that would be required to serve our purposes. In topology, spots of varying sizes are modelled by open sets, which become the building blocks of so-called frames. In many cases, it is possible to reverse this process by defining points as being frame homomorphisms to 2={0,1}.

Approach spaces on the other hand can be used to combine the advantages of both topological and metric spaces. Topological spaces behave better with respect to metric spaces in the sense that we can take arbitrary products of topological spaces which still behave well. On the other hand, topological spaces are more of the all-or-nothing-type. Two points can be either the same, or not. A point can be inside a subspace, or not. Here, metric spaces have the advantage: we can say exactly how far two points are apart from each other. Approach spaces remedy this by defining the distances between a point and a set, not between points. One of the characterizations of approach spaces is the lower regular function frame L, consisting of contractions to P=[0,∞], which is in fact a frame.

In this thesis, we first remark that whereas point-set distances and lower regular function frames L have been extensively studied, their ‘interior’ counterparts ι and U were left mostly ignored. We begin by defining and studying ι and also consider upper regular function frame U. We then consider another property of approach spaces which was not extensively studied yet, namely normality. Normality is an important tool in order to prove existence of extensions of function, such as Urysohn maps, Tietze's extension and Katetov-Tong's insertion.

We then remark that, as for contractions, the set of semicontinuous maps to P also defines a frame. Moreover, a frame of opens can be described equivalently by the set of continuous maps to 2. Given that the points of a frame are defined by the frame homomorphisms to 2, we can replace both instances of 2 by another space/frame F. We investigate the transitions between spaces and F-frames and find a connection with free distributive lattices.

Dolf Huybrechts

• ​08/07/2021
  
• 10 a.m.
  
• ​Online PhD defence
• Supervisor: Michiel Wouters
  
• Department of Physics
  

Abstract

Open quantum systems have become subject of intense research in the latest years due to technological and experimental advances and their potential for quantum information applications. An open quantum system is subject to an interaction with its environment with which it can exchange e.g. particles or energy. Usually this type of interaction results in a dissipation of the system’s energy into the environment and a drive is needed to compensate for this loss. The competition of these driving and dissipation processes can result in very interesting physical phenomena, such as phase transitions, that are markedly distinct from their equilibrium counterparts. Subsequently, the theoretical interest in these systems has burgeoned and a plethora of theoretical techniques have been developed. Due to the scarcity of analytical solutions, these are mainly based on numerical simulations. A crucial obstacle to be overcome is the exponential growth in computational resources that is required in a numerically exact approach. As a result, there is a clear need for the development of approximative methods and methods that exploit the symmetries that are present in these systems to allow for a more efficient numerical study. ​ ​

In this thesis the properties of driven-dissipative quantum systems are studied by resorting to approximative methods based on a factorisation of the system’s state as well as exact simulation methods based on the exploitation of permutational symmetries. Additionally, an efficient method to extract the properties of these systems in the long time limit has been introduced. Our techniques have been applied to the study of the properties of the dissipative XYZ Heisenberg model as well as those of the driven-dissipative Bose Hubbard model, shedding light on the importance of quantum and classical correlations in these systems.

Pieter Moris

• ​05/07/2021
  
• 5 p.m.
  
• ​Online PhD defence
• Supervisors: Kris Laukens & Pieter Meysman
  
• Department of Computer Science
  

Abstract

In this thesis we study the application of data mining and machine learning techniques in the broad context of biomolecular networks. We cover three main topics, each related to a different type of network data, analysis method and underlying research question.

In the first section, we explore a new framework, based on frequent itemset mining and association rule mining, to distil biologically relevant information from host-pathogen protein-protein interaction networks and their annotation data, into a more interpretable format. The technique offers a translation of expert knowledge into a rule-based summary, which also lends itself nicely to visualizations, although it remains challenging to find the appropriate level of granularity for the specific taxonomic subject of interest.

The second module focuses on the well-studied problem of finding subgraph patterns in a bigger graph. Here we build upon earlier work that aims to uncover those subgraphs that are associated with a specific set of nodes of interest. It provides a unique extension to the widely used enrichment analysis methodologies by integrating network structure and functional annotations in order to discern novel biological subgraphs which are enriched in the targets of interest. We present a software package, termed MILES, which adds additional functionality and visualization capabilities to the original work.

The final part of this work is situated in the field of immunoinformatics. The molecular interactions between epitopes and T-cell receptors (TCRs) play an important role in the adaptive immune system. These interactions can also be represented as a network, and we showcase a novel technique for predicting new edges within it. Namely, we employ convolutional neural networks and a feature representation method inspired by image classification to create a generic classification model that can operate directly on the amino acid sequence of the two molecular partners. In addition, we compare validation strategies to assess the generalization performance for both seen and unseen epitopes, as well as discuss various challenges that are inherent to TCR-epitope data. While our method shows promise, it is clear that the open problem of predicting TCR-epitope interaction for unseen epitopes still requires further improvements, especially in terms of the diversity of the available data.

Tim Hellemans

• ​30/06/2021
  
• 5 p.m.
  
• ​Online PhD defence
• Supervisor: Benny Van Houdt
  
• Department of Computer Science
  

Abstract

In the load balancing literature, there has been a lot of interest in policies which balance load based on some information of d randomly selected servers. The analysis of these policies for a finite system is intractable in most cases, therefore one often relies on mean field methods. That is, you assume that all queues are i.i.d. which allows to describe the system's behaviour through the behaviour of a single queue.

Most prior work has focused on load balancing methods which balance the load based on the queue length. While reducing response times, small jobs may still get stuck behind long jobs for these policies. In contrast, our main focus is on policies which employ the actual work present at the servers. While causing additional overhead, these policies do allow to detect all large jobs. This includes policies which either favour servers with a low load, use some form of redundancy, have a memory at the dispatcher, assign jobs in batches,... We additionally consider policies which combine the queue length information with the runtime of the job currently receiving service. This allows us to detect large jobs if they have been receiving service for some time.

For all policies considered we obtain a numerical method to study its performance. We make use of simulations to illustrate the accuracy of the mean field approximation. Moreover, assuming exponential job sizes often allows us to compute performance metrics in closed form. In particular we noticed that by using a proper scaling, we can often obtain simple closed form formulas for the expected waiting time in case the system load is either close to instability or close to zero.

Alexander Skorikov

• ​25/06/2021
  
• 3 p.m.
• ​Online PhD defence
• Supervisor: Sara Bals
  
• Department of Physics
  

Abstract

Precise determination of 3D distribution of chemical elements is a problem of key importance for a range of advanced nanomaterials. Examples of such materials are catalysts, semiconductor electronics and nanoparticles with advanced optical properties, where even minute changes in the elemental distribution drastically alter the relevant properties of the material.

Currently, among the best-suited methods to analyze the elemental distribution in materials at the nanoscale are techniques based on a combination of transmission electron microscopy, spectroscopy and tomography, which offer excellent spatial resolution, high elemental sensitivity and the ability to retrieve the full 3D structure of the studied object. Unfortunately, these methods require very long acquisition times, which does not allow to apply them for high-throughput studies, such as statistical analysis of nanomaterials, in industrial settings or for investigating dynamic processes in materials. Moreover, the long exposure to the electron beam damages the materials under investigation, further limiting the applicability of such techniques.

In this thesis, the problem of high acquisition time and electron irradiation dose requirements for 3D analysis of elemental distribution in nanomaterials is approached by developing new improved methods for this task and optimizing the existing methodology for data acquisition and analysis. The utility of the proposed approaches for answering relevant materials science questions is demonstrated and the outlook on the future developments in this field is outlined.

Miles Martinati

• ​22/06/2021
  
• 10 a.m.
  
• ​Online PhD defence
• Supervisors: Sofie Cambré & Wim Wenseleers
  
• Department of Physics

Abstract

Carbon nanotubes (CNTs) represent the most ideal system to confine molecules in a 1D nanospace, due to their hollow structure, their smooth and impermeable sidewalls and the precise tunability of their diameter. These unique characteristics can be exploited to study the behavior of atoms and molecules confined in 1D, or to synthesize and stabilize new 1D nanostructures, overcoming the problem of chemical instability in free space. In this thesis, three 1D structures encapsulated inside CNTs are studied, i.e. graphene nanoribbons (GNR@SWCNTs), linear carbon chains (LCC@DWCNTs) and chains of water molecules (water@CNTs), by means of photoluminescence excitation (PLE) spectroscopy and/or wavelength-dependent resonant Raman spectroscopy. In particular, by analyzing the typical Raman modes of the GNRs together with their corresponding resonant Raman profiles (RRP), both the vibrational and the electronic properties of the encapsulated GNRs can be revealed. This enables us to assign the observed Raman modes to two specific structures of GNRs, namely the 6-armchair GNR and the 7-armchair GNR with widths of 0.61 and 0.74 nm and electronic band gaps of 1.83 and 2.18 eV, respectively. A similar wavelength-dependent analysis is used to study the excited vibronic states of two samples of LCC encapsulated inside DWCNTs. In this case, the analysis of the RRP enables to observe multiple Raman resonances and consequently to estimate the energies of the excited vibronic states. Lastly, the properties of 1D chains of water molecules are studied by probing the variation of the optical properties of the surrounding CNTs as a function of temperature with respect to a reference empty CNT sample. The phase transitions of water encapsulated inside four different CNT chiralities, i.e. (6,5), (7,5), (9,4) and (8,6), with diameters smaller than 1 nm, are observed at temperatures between 110 and 150 K, in agreement with the only observation reported so far in this diameter range.

Giovanni Castaldo

• ​18/06/2021
  
• 4 p.m.
  
• ​Online PhD defence
• Supervisors: Gudrun De Boeck & Ronny Blust
  
• Department of Biology
  

Abstract

Worldwide, aquatic ecosystems are under threat from metal pollution. Natural environments receive a wide variety of compounds and the prediction of mixture toxicity based on the toxicity of single compounds is difficult and shows a certain degree of uncertainty. Moreover, the effects of metal mixtures together with the effect of the temperature on toxicological processes remain very poorly documented. Considering that the temperature is one of the most important driving factors in organismal physiology and a crucial ecological factor, this is surprising.

In this work, common carp (Cyprinus carpio) were exposed via water to copper (Cu (II), zinc (Zn (II)) and cadmium (Cd (II)) in single or mixture exposure scenarios. Differences in the accumulation of these metals were observed, with Cu and Cd accumulating to a greater extent compared to Zn in the analysed tissues. When present together, different metals can interact with each other influencing the uptake, bioaccumulation and toxicity. For instance in our experiments, we observed reduced Cd levels in the gills of common carp simultaneously exposed to Cu and Cd.

Copper was the only metal found to cause an impairment in ion-homeostasis in the single exposure scenarios. In addition, when it is present in mixture scenarios (e.g Cu plus Cd) more marked effects on electrolyte levels (e.g. sodium) can be observed. In the final part of this work we focused on understanding to which extent different temperatures can affect metal mixture toxicity in common carp. In fish exposed to either a low (10 ºC) or high (20 ºC) temperature, both Cu and Cd accumulated in the gills, whereas Zn levels remained stable. However in fish kept at 10 ºC, Cu metal levels in the gills were higher compared to fish exposed at 20 ºC, in contrast to what was observed for Cd. Moreover at 20 ºC, after one week of exposure fish started to eliminate excess Cu. The obtained results suggest that at 20 ºC, fish had more efficient depuration processes for the essential elements Cu and Zn.

Overall, the present thesis provides new insights into the effects of metal ion toxicity when present in mixture scenarios. Moreover, these findings highlight the importance of considering the effect of environmental parameters in designing standard toxicology tests to derive water quality guidelines that are protective for environmentally realistic conditions.

Nicolaas Wolmarans

• ​15/06/2021
  
• 2 p.m.
  
• ​Online PhD defence
• Supervisors: Victor Wepener, Lieven Bervoets & Patrick Meire
  
• Department of Biology
  

Abstract

The Phongolo River floodplain in South Africa hosts the highest floodplain biodiversity in the country, while also being highly utilised for commercial and subsistence agriculture. The floodplain falls within the malaria risk region where vector control in the form of indoor residual spraying is still practised with dichlorodiphenyltrichloroethane (DDT). This region operates on a fragile socio-ecological balance. Previous studies identified a gap in available knowledge on malaria vector control pesticides and the associated toxicity to amphibians. This study used a tiered assessment approach to assess the risk to amphibian well-being in the Phongolo River floodplain. Ecosystem services were incorporated into the assessment alongside the risk to amphibian well-being to assess the relationship between these aspects. The first tier to this study involved identifying the data requirements, which included sub-lethal effects data regarding amphibians and malaria vector control pesticides along with a lack of current field monitoring data of these pesticide in amphibians. The second tier involved generating field monitoring data. This study showed amphibians from Ndumo Game Reserve in the floodplain actively accumulate DDT at sub-lethal levels. The next tiers in the assessment involved sub-lethal toxicity data generation. This was done through laboratory and simulated field exposures of Xenopus laevis to vector control pesticides. Behaviour, metabolomics and pesticide accumulation was measured as effect outcomes. Behavioural changes were seen in frogs exposed to a mixture of DDT and deltamethrin. Metabolomic changes were mostly attributed to a general stress response affected in all exposures compared to control. Metabolomic responses in the simulated field exposure had low overlap with those found in laboratory exposures, which was attributed to the addition of food sources in the simulated field environment. The mixture exposure of DDT and deltamethrin resulted in significant loss of invertebrate diversity. The final risk assessment incorporated data generated in this study to determine risk levels to amphibians using a relative risk model. Overall amphibian well-being was at moderate risk, driven by the likelihood of chronic or sub-lethal effects from pesticides and high amphibian biodiversity in the region. The aquatic habitats in the floodplain (river, temporary pans, and permanent pans) were identified as priority habitats for conservation in order to benefit socio-ecological functioning and maximise amphibian wellbeing in the process. The outcome of this study supports the use of amphibian well-being as a monitoring tool for the floodplain through amphibian well-being and biodiversity monitoring serving as sensitive indicators of ecological change.

Elise Daems

• 03/06/2021
  
• 4 p.m.
• ​Online PhD defence
• Supervisors: Karolien De Wael & Frank Sobott
• Department of Chemistry

Abstract

Aptamers are short, synthetic DNA or RNA molecules that are characterized by a specific 3D conformation which enables specific target recognition. Aptamers are promising tools in many application fields from sensing to therapeutics. One of the major challenges in the aptamer field is understanding the relationship between the sequence and what determines the higher-order structure and specific interactions with targets. Therefore, this PhD thesis focuses on the use of different mass spectrometry (MS) based approaches to characterize aptamers and their interactions. Several of these approaches are already widely applied to study other biomolecules, such as proteins, but are still largely unexplored for aptamers and oligonucleotides in general. ​ ​

A first focus was put on obtaining information on the higher-order structure and conformational stability of aptamers using a combination of MS and with ion mobility (IM) spectrometry by performing collision-induced unfolding (CIU) experiments. CIU was shown to hold great promise to analyze the conformational dynamics and gas-phase stabilities of aptamers.

Next, the capabilities and limitations of native IM-MS for the analysis of noncovalent interactions of aptamers were demonstrated. The conformational behavior and interactions of cocaine-binding aptamers were studied and it was found that relative binding affinities of aptamers that only differ slightly in sequence and structure can be determined using native MS. Moreover, native IM-MS allowed the detection of small conformational changes upon binding of a target, which were found to be dependent on the binding mode of the aptamer. An adaptive binding mechanism was suggested for flexible aptamers that require more reorganization upon binding. ​ ​

In the final part of this thesis, the importance of thoroughly characterizing and validating aptamer-target interactions before using them in an application was emphasized. Moreover, the gathered insights were applied in our own development of a proof-of-concept aptamer-based sensor. This was shown by investigating the interactions of ampicillin aptamers which were found to not bind the target they were selected for in the first place. A multi-analytical approach combining complementary techniques was used for this purpose since no single technique is generally applicable to characterize all aptamers and their interactions and to obtain a comprehensive picture of the aptamer-target interactions. Furthermore, such multi-analytical approach was used to characterize a testosterone-binding aptamer while developing an aptamer-based electrochemiluminescent sensing strategy for this target. This shows the importance of native MS, in combination with other techniques, to thoroughly understand the aptamer-target interactions in the development of a designed application.

Wannes Van Beeck​

• 12/05/2021
  
• 5 p.m.
• ​Online PhD defence
• Supervisor: Sarah Lebeer
• Department of Bioscience engineering

Abstract

Food fermentation is often considered the oldest biotechnology within the world, because it has been used by the Egyptians to preserve fresh food for an extended time. Recently, artisanal (plant-based) fermentations have regained interests by the broad public and haute cuisine chefs, because of their increased organoleptic properties. Fermented carrot juice is an example of such an artisanal vegetable fermentation. Spontaneous fermentations are generally considered as safe, but it was not yet well studied whether pathogens could persist when the fresh produce is highly contaminated. Therefore, we explored the food safety of the spontaneous carrot juice by doing a challenge test with three common food pathogens: Listeria monocytogenes, Salmonella Typhimurium and Escherichia coli O157:H7. These pathogens could survive and persist within the fermentation up to 8 days of fermentation.

A dedicated starter culture could be used to guide and speed up the fermentation and reduce potential pathogen load as early as possible. Traditionally, starter cultures are chosen from isolates from the food product itself (autochthonous), but isolates from others sources (allochthonous) and with additional functionalities could have benefits for the consumer. How and when such allopatric starter cultures could also guide vegetable fermentations such as fermented carrot juice was also not yet widely explored. Starter cultures were chosen from three important genera of the carrot juice fermentation: Leuconostoc, Lactiplantibacillus, and Lacticaseibacillus. Leuconostoc starter cultures had the largest impact on the acidification of the fermentation and resulted in a pH lower than 4.6 after one day of fermentation. This pH of 4.6 is an important food safety threshold. In general, the autochthonous starter cultures belonging to the Lacticaseibacillus and Lactiplantibacillus were able to better guide the fermentation towards a uniform end community than their allochthonous counterparts. There were some exceptions with allochthonous probiotic L. rhamnosus GR-1 being able to persist within the fermentation. Therefore, we subsequently aimed to explore characteristics important for persisting and guiding the carrot juice fermentation using an ecological framework. Biotic factors such as microbial competitions, appeared to be important for persistence in the fermentation whereas abiotic factors such as salt stress also played a role, but their effect was minor.

The results obtained during this PhD have shown that carrot juice fermentation is safe to consume when fermented for at least eight days, but that vegetable fermentations can be further enhanced using dedicated functional starter cultures.

Andrea Marchetti​

• 29/04/2021
  
• 1 p.m.
• ​Online PhD defence
• Supervisor: Karolien De Wael
• Department of Bioscience Engineering

Abstract

Cultural heritage represents the vehicle of our cultural identity, handed over from past to future generations throughout human history. As a repository of fundamental cultural and social values, the preservation of all forms of cultural heritage is a responsibility of every society and of humankind as a whole. When it comes to tangible cultural heritage, preservation of heritage translates into preservation of objects and, therefore, of the materials they are constituted of. This crucial task relies heavily on the application of scientific analytical methods to answer material and conservation-related questions.

The fundamental contribution of this analytical approach led, in the past decades, to an ever-deepening understanding of the factors governing the degradation of cultural heritage. However, the extreme complexity of the heritage object-environment system results in a massive research field, which inevitably presents relevant open questions. This is where the present PhD work comes into play, attempting to fill knowledge gaps in literature by starting from specific case studies and un-answered research questions. 

The multianalytical research conducted during this PhD unraveled fundamental information on the properties governing the reactivity and long-term behavior of different classes of materials, from α-brass in an indoor environment to artists’ pigments in the presence of light, moisture and soluble particulate matter (PM). The paramount importance of the synthesis conditions on the composition, physical properties and reactivity of heritage materials was also demonstrated, in particular for stable lead pyroantimonate and unstable Geranium lake artists’ pigments. Moreover, the study and characterization of specific heritage objects, namely a series of 16th century reliquary altarpieces and the painting L’Arlesienne, by Vincent Van Gogh, allowed to obtain relevant insights into their composition and on potential risks for their conservation. The challenging nature of the samples considered, created the perfect opportunity to test an innovative spectroscopic technique, optical photo-thermal IR (O-PTIR), for the characterization of heritage materials. Striking results were obtained, highlighting a great potential for the application of this non-destructive sub-micron molecular spectroscopy to the analysis of cultural heritage. Finally, in the last section of this work, strategies to implement the continuous monitoring of PM levels in indoor environmental quality studies were also considered, with a particular focus on the identification of environmental hazards for the collections housed in specific conservation environments (War Heritage Institute in Brussels and St. Martin’s church in Aalst, BE).  ​​

Patrick Bosch​

• 29/04/2021
  
• 4 p.m.
• ​Online PhD defence
• Supervisor: Steven Latré
• Department of Computer Science

Abstract

The last two decades brought a phenomenal increase in communication devices, mobile and stationary. The overall number of connected devices went past 18 billion with many available technologies. Two wireless technologies dominate the market space: LTE/5G and Wi-Fi. Other new wireless technologies also rose in popularity. This situation gave rise to new applications that require high bandwidth and low latency.

However, technologies share use cases, devices can not fully use their technologies, and technologies can interfere with each other. Avoiding interference and enabling technology cooperation are major cornerstones of improving user experience. Nevertheless, cooperation can not overcome all obstacles. Estimation and modeling of performance within certain environments are necessary.

Current solutions for integrated technology management are limited. They focus on specific technologies, specific use cases, or have limited management capability. Performance modeling is more common but focuses on interference from other communication technologies, not interference from generic electronic devices. The increasing use of electrical devices multiplies the problem and affects many devices and technologies. This dissertation provides three significant contributions to address these challenges.

The first contribution explores and models IEEE 802.11 systems' performance when an interfering source is present that is not a communication technology. We first explore the performance of IEEE 802.11 in a challenging environment via a wireless mesh network and further in a controlled setup and simulation. We provide two models. The first is based on base performance when no interference is present and is computationally fast. The second is an analytical model that models the entire system's behavior but is computationally expensive.

The second contribution consists of the ORCHESTRA framework, enabling inter-technology management seamlessly to the user and operator. This framework offers interference mitigation by using multiple technologies. It uses technology abstraction through a virtual layer and advanced packet-level functionalities, such as handovers, load balancing, and duplication. A central controller maintains a global view of the network and makes intelligent decisions to improve performance.

 As a third contribution, we present a load balancing solution that normalizes latency for links with different latency properties. Different technologies exhibit different latency properties that cause problems when using packet-level load balancing. We provide a machine learning based normalization method that smooths and reduces latency on a flow. Instead of sending out bursts of packets after reordering, packets are sent with a short time in between to avoid burst behavior.

Glenn Daneels​

• 26/04/2021
  
• 4 p.m.
• ​Online PhD defence
• Supervisors: Jeroen Famaey & Steven Latré
• Department of Computer Science

Abstract

The Internet of Things (IoT) paradigm is the shift towards a world where all things are connected to the Internet. While nowadays IoT has an impact on every aspect of society, it is being applied particularly efficiently to further revolutionize the automation and control of traditional manufacturing and industrial processes, leading to the term Industrial Internet of Things (IIoT). To fulfill the high-end requirements of IIoT applications, interconnecting all sensing and actuating devices was initially typically done through wiring. While the reliability advantage of wiring is obvious, it is costly and can be impractical in hard to reach locations or mobile machinery. Therefore, the transition to wireless communication seemed unavoidable and is becoming more and more ubiquitous. However, for this transition to be successful, wireless communication should show wire-like reliability in harsh industrial environments that suffer from external interference and multi-path fading effects that may easily disrupt the wireless signal. Additionally, to avoid having to frequently replace batteries in inconvenient places, the wireless devices should be able to run on limited battery capacity for years. Therefore, while required to be highly reliable, they should also exhibit low-power operations. A relatively recent wireless technique that has proven to be successful in fulfilling these requirement, is IEEE 802.15.4e Time-Slotted Channel Hopping (TSCH), that combines frequency diversity with strict time-synchronization, achieving wired-like reliability of more than 99.999% while having ultra-low power consumption. ​ ​

In this PhD thesis, I study the TSCH Medium Access Control (MAC) layer and how it can be further improved to deploy it successfully in industrial networks. More specifically, I focus on 3 research questions related to the energy consumption, latency and reliability of TSCH networks. First, I investigate how the TSCH power consumption can be precisely characterized. Second, I aim at minimizing the communication delay of recurrent monitoring data that is typical for IIoT applications. Finally, I aim at further improving the industrial network’s overall reliability by introducing a technique to allows multiple physical (PHY) layers simultaneously in a single TSCH network. As such, each device is able to adapt its PHY layer to the link’s propagation characteristics. Additionally, a heuristic is also proposed that helps a device in making an appropriate parent and PHY selection in multi-PHY TSCH networks.

 In summary, this PhD thesis targets an energy-efficient, responsive and reliable TSCH network, thereby contributing to a wireless network deployment that is ready for the IIoT.

Stijn Legrand

• 19/03/2021
  
• 2 p.m.
• ​Online PhD defence
• Supervisor: Koen Janssens
• Faculty of Science

Abstract

Last century the field of heritage sciences expanded beyond imagination. The inventions of X-ray radiography and infrared reflectography allowed experts to investigate paintings below the surface as well. More recent developments led to the advent of the field of hyperspectral imaging, to which the advanced chemical imaging methods, used in this thesis work, belong. These techniques not only allow to identify the components present in artworks, but also to visualize their distribution over these objects. The resulting distribution maps permit a broader public to interpret the scientific data and to relate these results with the artwork itself.

During this thesis work a range of flat artworks were investigated in a non-destructive manner using mainly two macroscopic imaging techniques: macroscopic X-ray fluorescence scanning and macroscopic Fourier transformed mid-infrared scanning in reflectance mode. The resulting images were sometimes supplemented with microscopic techniques on a minute selection of samples to fully understand the layer build-up, composition and distribution of these materials over the stratigraphy.

Illuminated manuscripts pushed the interpretation of the macroscopic imaging techniques: due to the impossibility of sampling, all answers had to be obtained non-destructively. Documenting masterpieces such as the Ghent Altarpiece by means of chemical imaging techniques, helped the restoration team, assisted by the international commission to make the daring decision of manually removing the non-original paint layers. Scanning stained-glass windows allowed experts to document the panels, create situation reports, identify later infills and guide the restoration process in a more efficient manner.

By initially applying non-destructive imaging techniques, many of the research/conservation questions could already be answered. Based on the resulting distribution maps, only a very limited amount of sampling was required to obtain a representative set to answer the remaining questions. In most cases the combination of multiple methods was necessary to fully understand the situation. A similar trend could be seen in the research field: the collaboration between divergent disciples was often required in order to explain all observations.

In order to completely break through, the scanning speed of these techniques has to increase even more in order to cover an acceptable surface in one workday. Parallel with the operational speed, the (basic) data treatment should also be streamlined more in order to allow a broader user group to access the results. Once these two improvements are carried out, these techniques become accessible to a larger public. ​ ​

Stijn Wittouck

• 15/03/2021
  
• 5 p.m.
• ​Online PhD defence
• Supervisors: Sarah Lebeer & Vera van Noort
• Department of Bioscience Engineering

Abstract

Lactic Acid Bacteria (LAB) are responsible for many types of fermented foods and are part of our natural microbiota. The goal of this PhD was to leverage publicly available genomes of LAB to gain new insights into the evolutionary history and habitat-adaptation of these bacteria. To make this possible, important taxonomic and computational challenges were solved.

Three groups of LAB were studied in the thesis. The first was the Lacticaseibacillus casei group: a cluster of closely related species with many applications as oral probiotics and in dairy fermentations, but with much confusion surrounding the classification of strains of the species L. casei, L. paracasei and L. zeae. Based on a comparison of all publicly available genomes from this group, the taxonomic confusion was cleared up, and a number of potentially habitat-relevant properties were identified that could discriminate between the species. For example, genes encoding catalases and putative epithelial adhesins were detected in L. casei genomes, and superoxide dismutase genes were found in L. paracasei genomes. The former were particularly relevant, because an L. casei strain with probiotic potential had previously been isolated from the upper respiratory tract of a healthy individual. Next, the family Lactobacillaceae was studied. For this purpose, a novel computational tool was developed to identify the core genes of a set of genomes in linear time. This tool was used to correct many species-level misclassifications of strains belonging to the family and to suggest mergers and splits of published species. For instance, a merger of the species Weissella thailandensis and Weissella jogaejeotgali was proposed, as well as a split of Ligilactobacillus aviarius. In addition, the genus Lactobacillus was split into 25 smaller genera and the families Leuconostocaceae and Lactobacillaceae were merged based on an analysis that included the use of signature genes to find biologically relevant clades. Finally, a novel tool was developed that could infer a pangenome (the collection of all gene families in a set of genomes) in near-linear time. This tool was then applied to create a pangenome database for the order Lactobacillales, which was subsequently explored to identify some trends in the evolution of these bacteria. For example, it was found that the number of core genes of species changes relatively slowly, and that genes encoding amino acid transporters experienced many duplications in the evolutionary history of the order.

Simone Messina

• 15/03/2021
  
• 2 p.m.
• ​Online PhD defence
• Supervisors: Marcel Eens, David Costantini & David P. Edwards
• Department of Biology

Abstract

Land-use changes are one main cause of biodiversity loss. Selective logging is the most common technique of timber extraction applied to tropical forests, driving species loss and population abundance changes. One main question to understand species’ responses to selective logging is which proximate mechanisms underlie species abundance changes. In this Ph.D. project I have used a cross-sectional approach to investigate the effects of selective logging on the stress physiology of understorey birds, and correlative analyses to investigate the effects of physiological changes on population abundance, across unlogged and selectively logged forest of Borneo.

The first goal of the project was to determine which physiological endpoints of vertebrates are affected by forest disturbance. To this end, I reviewed all available literature and used meta-analytical techniques to quantify the size of the effects of different forest disturbances, including selective logging, on physiological and immunological parameters. ​ ​

I have then investigated the effect of selective logging on the activity of the hypothalamic-pituitary-adrenal (HPA) axis in 10 understorey bird species. I used as marker of HPA axis activity the concentrations of corticosterone, the avian glucocorticoid hormone, deposited in feathers.

Another important physiological mechanism for maintaining homeostasis is the regulation of cellular oxidative status. Thus, I measured eight different markers of oxidative status in 15 understorey bird species living either in unlogged and selectively logged forests. I also investigated differences in the oxidative status between feeding guilds (i.e. insectivores and omnivores) and how they are affected by selective logging.

Last, I tested for differences in body size and body condition of more than 50 bird species across unlogged and selectively logged forests. Changes in body size and body condition can be sub-lethal effects of habitat degradation that may act as early signals to predict future population responses. This hypothesis was tested correlating changes in body size and body condition with changes in population abundance between the two types of forest. ​ ​

Results point to feather corticosterone as a promising tool for monitoring the impacts of sylvicultural practises on understorey birds. There is little long-term effect of logging on the oxidative status of understorey bird species. Last, frugivores and omnivores have reduced body size in the logged forest compared to unlogged, pointing to potential functional consequences related to seed dispersal.

Rosanne Reitsema​

• 26/02/2021
  
• 9 a.m.
• ​Online PhD defence
• Supervisors: Jonas Schoelynck & Patrick Meire
• Department of Biology

Abstract

Freshwater ecosystems are one of the most diverse, but also one of the most threatened ecosystems in the world. Aquatic macrophytes are highly affected by consequences of climate change like increased concentrations of dissolved organic carbon (DOC) and carbon dioxide (CO2), but also changes in flow dynamics and eutrophication. Knowledge on the effects of DOC and CO2 on macrophytes, and especially their interaction effects with other effects of climate change, is relatively limited. Therefore, the aim of this thesis was to study effects of climate change, like increases in carbon concentrations, using a holistic approach that also focused on their interaction effects with other environmental variables, rather than only studying effects separately.

The main macrophyte species studied in this thesis was Berula erecta (lesser water parsnip). Under natural conditions in a temperate lowland stream, this species was found to be highly variable in its biomass, morphology and nutrient content throughout the growing season, and there were interactions between plant growth (biomass and morphology) and environmental parameters like flow velocity and fine sediment depth.

The effects of climate change were tested in a greenhouse experiment by exposing two macrophytes species, B. erecta and Myriophyllum spicatum to different concentrations of CO2 and DOC in a wide range. The macrophytes responded to both treatments, with the strongest effects in the highest doses. There were large differences between the two species, with regard to growth and morphological responses. Finally, the interaction effects among different aspects of climate change were tested in further greenhouse experiments. In these experiments, B. erecta was exposed to varying combinations of DOC, CO2, flow velocity and nutrients. Those stressors sometimes had opposing effects: CO2 highly stimulated growth, DOC turned the water brown and limited macrophyte growth by shading effects, high nutrient concentrations indirectly limited growth by stimulating the growth of epiphytic algae that shaded the macrophytes, and increased flow velocity led to a more compact growth form.

From this thesis it can be concluded that climate change can have a large effect on macrophytes. Different aspects of climate change often have opposing effects, with many interaction effects occurring among them. Taking all aspects of climate change together, the results from this thesis indicate that submerged macrophytes in temperate lowland streams and rivers will decrease in biomass quantity and quality under continuing climate change. This will in turn have negative consequences for ecosystem processes and organisms that depend on the macrophytes.

Nick Hoeven

• 18/02/2021
  
• 2 p.m.
• ​Online PhD defence
• Supervisor: Pegie Cool
• Department of Chemistry

Abstract

The earth has been warming up at an unprecedented pace during the last decades, which is majorly caused by increasing greenhouse gases. High CO2 concentrations in the atmosphere led to worldwide awareness of environmentally conscious thinking and acting to reduce this compound and other greenhouse gases. CO2 conversion makes valorization possible through valuable chemicals and fuels. This cradle-to-cradle philosophy is necessary in our current society, both reducing atmospheric greenhouse gases and simultaneously partly responding to the need for alternative fuels. This is fundamental as continuous increase of anthropogenic greenhouse gases are one of the most important issues of this and future generations.

In this thesis, the photocatalytic reduction of CO2 with H2 in the gas phase over modified Ti-Beta zeolites is studied. The goal of this thesis is the development of improved photocatalytic materials for CO2 applications in the gas phase and to overcome limitations posed by the use of TiO2 and classical semiconductors.

Different methods for CO2 utilization and conversion have been discussed. Furthermore, an overview on the mechanism of photocatalysis and the limitations of the use of TiO2, as well as the strategies to overcome those limitations have been described. In particular, the superior photocatalytic activity of isolated tetrahedrally coordinated Ti-species in combination with the high surface area of zeolites has been highlighted. The importance of a well-designed photocatalytic reactor and its influence on the turnover frequencies (TOFs) of the reaction products are also described. The experimental work focusses on the optimization of the synthesis method and the Ti loading of the Ti-Beta zeolites. Next, the synthesized zeolites are tested in a photocatalytic reactor and the influences of the material properties on the product TOFs are discussed. In order to further enhance the photocatalytic properties and the product selectivities of the catalysts, noble metal nanoparticles (Pt and Pd) are deposited onto the Ti-Beta zeolites. Finally, alternative catalysts (e.g. 3D printed structures and Z-scheme catalysts) are tested in the photocatalytic reactor and compared to the highest performing Ti-Beta catalysts.

In conclusion, this PhD has put a step forward in the development of novel and highly active photocatalytic materials, with improved performance compared to classical pure TiO2, for the photocatalytic reduction of CO2 in the gas phase.

Sumit Srivastava

• 15/02/2021
  
• 1 p.m.
• ​Online PhD defence
• Supervisor: Pegie Cool
• Department of Chemistry

Abstract

The main objective of this work was to contribute towards the co-utilization of CO2(g) and residues from metallurgical industries to produce Fe-carbonate bonded monoliths that are free from cement clinker. While excessive CO2 is considered a significant problem due to the increased global warming and its associated effects, Fe-rich metallurgical wastes are still used for low-value applications or are landfilled. Moreover, due to the volume of their use, construction materials production accounts for 7-8% of total CO2-emissions. Therefore, the co-utilization of slag and CO2 to produce construction materials has significant potential to contribute towards achieving future sustainability goals. In this study, Fe(0) is initially chosen as a model system to understand the feasibility of producing FeCO3-bonded monoliths under the desired reaction conditions (<100 °C, and <25 bar CO2-pressure). In addition to demonstrating the feasibility of FeCO3-cementation, the underlying reaction mechanisms are also discussed. Since the dissolution of Fe-sources is usually known to be the rate-limiting step, Fe-dissolution in dilute conditions is studied as a function of temperature, CO2-pressure, and time. Similar to the dissolution studies on Fe(0), dissolution studies in dilute solutions are also extended to the Fe-Si rich non-ferrous slags as a function of temperature, CO2-pressure, and time. In both the studies, it is shown that high temperature and CO2-pressure are conducive towards the dissolution of Fe(0) and Fe-rich slags. To transfer the knowledge of FeCO3-cementation from the model Fe(0) system to the sources in which Fe co-exists with Ca, FeCO3-cementation in CO2-H2O-Fe(0)-Ca(OH)2 systems is also studied. The importance of microstructures of the products, and the formation of mixed (Ca, Fe)-carbonates is pointed in this study. Finally, it is shown that the non-ferrous slags can be co-utilized with ferrous metallurgical slags rich in Ca to produce carbonate-bonded monoliths with high mechanical strength. It is shown that the carbonation of the non-ferrous slags as mixes with ferrous slags can lead to a significant decrease in environmental leaching. With more than 575 million tonnes of metallurgical slags produced every year, they offer an opportunity for significant CO2-mineralization as well as to produce low-carbon construction materials.

Ruben Mennes

• 08/02/2021
  
• 4 p.m.
• ​Online PhD defence
• Supervisor: Steven Latré
• Department of Computer Science

Abstract

Wireless communication technologies became a part of our modern society. Every year the number of wireless devices and wireless technologies increases. Cisco expects that around 25.4 and 42.6 billion wireless devices will be connected to the Internet in 2022. This growth introduces some major challenges. One of these challenges is to use the wireless spectrum, used by all of these wireless devices, more efficient, especially within the radio bands itself. ​ ​

To meet the demand of more wireless devices and higher throughput, new techniques are necessary to optimise the use of the wireless spectrum. Based on literature, it was expected that collaboration between neighboring wireless networks (from all kind of technologies) can improve the efficiency of the use of the wireless spectrum. Increasing spectrum efficiency can be accomplished in two ways: (i) the improvement of physical transmission, (ii) the use Artificial Intelligence (AI) to improve the decisions made by the wireless nodes. The improvement of the physical transmissions has a direct effect on the efficiency of the use of the spectrum. It is clear that the more efficient data can be transmitted, to improve the bits per Hertz, the less spectrum will be used for the same data. AI, on the other hand, gives us the opportunity make smarter decisions based on the physical limitations of the wireless system and behavior of the environment. The use of AI can also enable the possibility to start and maintain collaboration with other neighboring technologies to improve the efficiency of the wireless spectrum. This dissertation focuses on the contributions made for the AI decision engine for wireless network technologies. Within the context of this dissertation we focus on the use of AI to improve the decisions made by the wireless nodes.

This dissertation provides multiple improvements to enable collaboration for wireless networks, which will lead to a more efficient use of the wireless spectrum. First, we describe a decision-making framework designed to enable AI-enabled algorithms within wireless radio stacks. All other algorithms described in this dissertation are implemented within the framework. Secondly, we present a spectrum prediction algorithm. This prediction algorithm is able to predict the behavior of neighboring wireless networks, even if insufficient information is available. This ability provides us to select better transmission moments. Finally, we introduce the policy-based flow selection algorithm. This algorithm is able to collaborate to improve the Quality of Service and optimize the spectrum footprint.  ​​

Bossissi Nkuba

• 02/02/2021
  
• 4 p.m.
• ​Online PhD defence​
• Supervisors: Lieven Bervoets, Sara Geenen & Landry Cizungu (Catholic University of Bukavu)
• Department of Biology

Abstract

This thesis investigates the use of mercury (Hg) in artisanal small-scale gold mining (ASGM), pollution of aquatic ecosystem by Hg and other metals, and human intoxication by these metals in Kamituga (Eastern Congo).

The first part assesses perceptions on mercury, making use of qualitative data as well as a quantitative survey. It  found that despite existing laws banning Hg use in ASGM, a lack of enforcement leads to widespread use of Hg. Miners use Hg on ore concentrates, thereby limiting Hg pollution. But they use Hg in residential areas and within streams’ watersheds, thus exposing communities and aquatic ecosystems. People are poorly informed on effects of Hg on environment and health. However, the findings include some promising signs, since our respondents prioritize environment and health protection more than economic profit from ASGM.

The second part analyses samples collected on a tributary of the Congo river (Zalya) and its network. It investigates the concentration of  metals in water, sediment, indigenous labeo fish captured in selected streams and in experimentally-exposed tilapia. It found that metal concentration in sediments, unlike in water, often exceeded environmental standards. Indigenous fish from mining-affected streams had higher metal concentration. For experimentally-exposed fish, high mortality was observed but no significant differences in terms of metal accumulation in surviving fish. Water consumers are safe, but fish consumers may be at risk of Hg, Cd and Cr intoxication if their daily consumption exceeds respectively 77, 145 g and 138 g.

The third part analyses the diet, health status and symptoms relatable to metal poisoning of miners and non-miners and compared to metal levels in their blood, urine, nails and hair. It found that many people had Hg levels in their blood, urine, nail and hair as well as concentration of other metals in different tissues above reference value. Miners had higher nail Hg but similar blood, urine and hair had similar Hg levels to non-miners. Prevalence of potential Hg poisoning was equally spread in the community but was higher for people with tilapia-rich diet. Symptoms were not correlated to higher Hg levels, but with other factors such as age and undernourishment.

The thesis recommends raising awareness in the community about dietary and occupational exposure, risks of mercury and quantities of fish that are safe for consumption; increasing formalization of the ASGM sector and capacity of law-enforcement agencies that supervise Hg use in mines; and monitoring of potentially polluted streams.

Jeroen Bomon

• 02/02/2021
  
• 4 p.m.
• ​Online PhD defence​
• Supervisor: Bert Maes
• Department of Chemistry

Abstract

With the decline of petroleum feedstock, rise of crude oil price and the necessity to reduce CO2 emission, society must innovate to discover new, more sustainable means to meet the needs of an ever expanding world population. The manufacture of products based upon (bio)renewable resources is one of the ways to address this. Particularly, (hemi)cellulose derived products have already been investigated intensively and even found mature applications in industry. However, these bio-polymers do not contain aromatic moieties, preferentially requiring other parts of plant tissue to access these key entities for chemical industry.

In this PhD Thesis, the use of different biorenewable substrates containing aromatic entities were considered for chemical conversion. For example, bio-polymer lignin, Earth’s largest resource of bioaromatic compounds, could serve as suitable feedstock. Several routes are known to depolymerize lignin in low molecular weight monomers, which are described in Chapter 1. In Chapter 2-4, several lignin-derived substrates were transformed into valuable products by defunctionaliztion and functionalization reactions. Next to lignin, also ferulic acid and eugenol, extracted from rice bran and clove, respectively, served as useful substrates.

In Chapter 2, a cheap methodology, only requiring strong acid and hot pressurized water, is presented to defunctionalize these monomers on carbon and oxygen atoms of the alkoxyarene moieties, delivering a hydroxybenzene (phenol, catechol, pyrogallol and derived isomers) as product. Similar conditions, based on the employment of the same strong acid or a heterogeneous alternative, are described in Chapter 3, dealing with selective defunctionalization on oxygen atoms in these monomers, forming C-alkylated hydroxybenzenes. Next to these defunctionalization approaches, functionalization of the considered starting material with nitrogen atoms is described in Chapter 4, since incorporation of this element is highly desired for the production of compounds with pharmaceutical or agrochemical application.

The use of a biorenewable substrate is one of the 12 Principles of Green Chemistry. In order to analyze if the developed reactions with these subtrates can also be considered as “green”, the CHEM21 Green Metrics Toolkit was applied on both our newly developed approaches and on literature procedures delivering the same reaction or product, allowing us to make a useful comparison. The result of this analysis is described in Chapter 5.

Jonas Bertels

• 01/02/2021
  
• 2 p.m.
• ​Online PhD defence​
• Supervisor: Gerrit Beemster
• Department of Biology

Abstract

Much is known about the impact of cadmium (Cd) stress on plants and the plant’s response to this form of abiotic stress. However, it is remarkable that the impact of Cd in the growth zone of monocotyledonous leaves remained largely unstudied. This growth zone hosts the two cellular processes driving growth, i.e. cell division and cell elongation. The aim of my PhD study was to assess the impact of Cd in this maize leaf growth zone at several biological levels.

We have found that Cd inhibited leaf growth mainly because it results in a significant reduction of cell production. Cells were halted at the G1-S transition of the cell cycle, which increased the cell cycle duration. In addition, when exposed to Cd, growing leaves had a lower number of meristematic cells and therefore less cells are contributing to cell division. In addition, we have found that Cd accumulated highest in the meristematic tissue, indicating that it could impact processes therein directly. To reveal these processes, we have performed a transcriptome study. This resulted in a broad range of Cd affected processes, which led me to perform biochemical analyses of several phytohormones, minerals, two oxidative stress related parameters and carbohydrates. We showed that Cd caused an increase in stress hormone levels (i.e. salicylic acid, abscisic acid and 1-aminocyclopropane 1-carboxylic acid (ACC, an ethylene precursor)) and a decrease of growth promoting hormones (i.e. gibberellin 1 and trans-zeatin riboside). For gibberellin 1, we were able to directly link changes in the spatial distribution of this phytohormone to changes in transcript levels of key gibberellin synthesis and degradation genes. Regarding the measured minerals, we mainly found manganese to be the most strongly and consistently Cd affected nutrient. Lipid peroxidation and antioxidant potential were increased throughout the entire maize leaf growth zone, demonstrating that Cd resulted in oxidative stress in all developmental stages. Lastly, we found that carbohydrates were increased under Cd stress, perhaps in response to oxidative or osmotic stress. 

During my PhD study, we have also published leafkin, an R package that contains four functions which allow the user to perform all calculations in a kinematic analysis of monocot leaf growth. In addition, it allows cell length profiles and leaf elongation rates to be easily extracted, which in turn can be used in separate analyses.

Bikram Koirala

• 18/01/2021
  
• 4 p.m.
• ​Online PhD defence​
• Supervisor: Paul Scheunders
• Department of Physics

Abstract

Hyperspectral cameras collect the reflected light of materials in hundreds of narrow, contiguous spectral bands in the visible, near and shortwave infrared wavelengths to provide a continuous reflectance spectrum for each pixel. Due to the complex interaction of light with materials, these spectra are highly nonlinear mixtures of the reflectances of the material constituents. The general goal of this thesis is to estimate the composition of materials from reflectance spectra.

Mixing models describe the reflectance spectrum of a material as a (nonlinear) mixture of the constituent materials. The main disadvantage of these models is that the model parameters are not properly interpretable in terms of the fractions. Moreover, not all spectra necessarily follow the same particular mixing model.

Alternatively, the complex mixing effects can be learned using supervised machine learning methods. This requires ground truth training data, in the form of the actual compositions (i.e., the spectra and fractions of the constituents). One major drawback of these strategies is that the estimated fractions do not comply with their physical constraints, leading to a loss of the physical meaning of the estimated parameters. Another disadvantage of the learned models is that they cannot perform well in case training and test spectra are obtained under different environmental conditions or by different sensors, causing spectral variability of the acquired spectra.

In this thesis, a hybrid framework was developed that combines the physical interpretability of a model and the flexibility of data-driven approaches. The general idea is to learn the complex relation between the nonlinear spectra and spectra that follow a particular mixing model by utilizing advanced machine learning regression algorithms. Based on this strategy, a number of different nonlinear unmixing methods were developed:

1) A supervised method that learns a mapping between the nonlinear spectra and the linear mixing model. 

2) A strategy for the estimation of leaf biochemical parameters from leaf reflectance and transmittance spectra, by learning a mapping to a leaf biochemical model (PROSPECT).

3) A semi-supervised method to reduce the number of training samples required to learn the nonlinearities, and additionally does not require the availability of pure pixels. 

4) A robust supervised method for nonlinear spectral unmixing that is invariant to endmember variability.

Steven Jillings

• 14/01/2021
  
• 5 p.m.
• ​Online PhD defence​
• Supervisors: Floris Wuyts, Angelique Van Ombergen, Ben Jeurissen & Athena Demertzi
• Department of Physics

Abstract

In over half a century of crewed missions to space, many different effects of spaceflight on the human body have been uncovered so far. However, little focus has been directed to investigating how space stressors affect the human brain. The largest body of work in this dissertation describes pioneering findings on brain structural and functional changes after spaceflight in Roscosmos cosmonauts by means of multi-modal magnetic resonance imaging (MRI) in a longitudinal and prospective design. Structural MRI modalities, such as T1-weighted and diffusion MRI, were used to unravel macroscopic volume and microstructural brain tissue composition changes. We found a widespread redistribution of the cerebrospinal fluid (CSF) with secondary mechanistic effects on the grey matter (GM) tissue. We also revealed increased neural tissue volume in motor regions of the brain that suggest evidence for structural brain adaptations, also known as neuroplasticity, associated with altered motor strategies in space. Most CSF changes after spaceflight were still detectable more than half a year after return to Earth, while the GM changes after spaceflight partially reversed in the long term. In addition, functional MRI data was acquired in these cosmonauts to study functional reorganisation of the brain after spaceflight, showing numerous functional connectivity (FC) alterations after spaceflight. Some of these changes persisted in the longer-term, whereas other changes returned back to pre-flight levels. Furthermore, this work also describes the experimental work and preliminary analyses of several Earth-based models. One is a longitudinal MRI pilot study in hindlimb-unloaded (HLU) mice, inducing fluid shifts to the head region, in order to better understand the consequence of these fluid shifts on the brain. A second study was performed in fighter pilots as a model for exposure to high g-levels and sensory conflicts, in which FC was compared to that in a control group. This work rendered a vast increase in available information on structural and functional brain changes after spaceflight compared to several years ago. In the future, the underlying mechanisms of the observed findings need to be understood in more detail. Ultimately, we aim to characterise the effects space stressors have on the brain, to then attempt to mitigate these changes through countermeasures and characterise beneficial coping mechanisms that we can enhance, in order to be fully prepared for future exploration missions into deep space.

Yannick Engelmann

• 12/01/2021
• 3 p.m.
  
• ​Online PhD defence​
• Supervisors: Annemie Bogaerts & Erik Neyts
• Department of Chemistry

Abstract

CO2 conversion, CH4 conversion and NH3 synthesis are three essential processes that can help to reduce greenhouse gas emissions. However, these processes typically require harsh reaction conditions when performed thermally, because of the strong chemical bonds of the reactants. Plasma catalysis can provide alternative methods to activating chemical bonds at ambient conditions. Due to the complexity of plasma-catalytic systems, fundamental understanding of the underlying mechanisms is still lacking, impeding the optimization of the technology and holding back its full potential. The aim of this dissertation is to provide fundamental understanding, needed to unlock the full potential of plasma catalysis.

As a tool to acquire the fundamental understanding, we introduced microkinetic modelling to provide detailed information on reaction mechanisms, kinetics and thermodynamics of the processes. In this way, we identified the limitations of thermal processes, but also unraveled if and how plasma-catalytic processes can overcome these limitations. The main difficulty of CO2 hydrogenation is to selectively produce CH3OH at sufficient rates. In plasma catalysis, the contribution of the plasma is twofold: excitation of the reactant molecules, lowering the barrier of dissociation and increasing the conversion rates of the thermal pathways, and generation of reactive radicals and intermediates, allowing new, unique pathways that potentially lead to CH3OH (often in a much faster way).

In the study on the conversion of CH4, we showed the limitations of transition metal catalysts to produce C2-hydrocarbons under thermal conditions. Thermally, the more noble catalysts are not able to dissociate the strong chemical bonds of the CH4 molecule, while the less noble catalysts suffer from cokes formation. In plasma catalysis, dissociation rates on noble catalysts can be increased by vibrationally exciting the reactants, or catalytic dissociation can be avoided by adsorption of plasma-generated radicals. Whether the adsorbed species couple directly to C2-hydrocarbons or undergo further dehydrogenation before coupling, can be controlled by the catalyst binding strength.

Lastly, the potential of the plasma-catalytic NH3 synthesis is locked in the enhanced catalytic rates, caused by plasma-induced excitation and plasma-generated radicals. Again, both vibrationally excited species and plasma-generated radicals are found to improve the NH3 synthesis rates. Due to the contribution of ER reactions, rates are not only increased on noble catalysts, but also on more strongly binding catalysts, making the choice of the catalyst material much less impactful.