Abstract: Photocatalysis has been labeled for decades as a promising technique for air purification. The principle seems straightforward and requires a photocatalyst that is immobilized on a substrate, and one or more UV sources to activate the photocatalyst. No waste products are produced, the reactions occur in mild conditions and the supplies are relatively cheap. Yet it seems that the commercialization of photocatalytic systems does not break through on the global market. The aim of this thesis is to identify and tackle the bottlenecks that impede commercialization from an application-oriented approach. The problem of indoor air pollution is enhanced by the fact that people spend more and more time indoors and that ventilation is kept to a minimum as an energy-saving measure. This inevitably leads to an accumulation of volatile organic compounds (VOCs) that are emitted by e.g. building materials, paint and furniture. Human exposure to VOCs is directly related to the sick building syndrome leading to complaints such as headache, fatigue, dizziness and lack of concentration. In addition, exposure to VOCs is related to serious long-term health effects such as cancer or respiratory diseases. Therefore, significant research efforts are focused on advanced indoor air purification methods. Integration or retrofitting of a photocatalytic (PCO) air purifying unit into heating, ventilation and air conditioning (HVAC) equipment has been chosen as an interesting approach. As a starting point of this thesis, the operational conditions of a ventilation system were mapped. These systems are characterized by high flow rates and the necessity of minimal pressure losses. Pressure losses increase the energy demand and can lead to failure of the ventilation fan and thereby undermine the proper functioning of the ventilation system. A suitable substrate must allow the contaminated air to pass through with a minimal pressure drop, allow sufficient contact time between VOC and photocatalyst, have a large surface area available for coating with excellent adhesion, and be transparent to UV light. Therefore, the permeability and the available exposed surface were selected as main selection criteria. After a thorough quantitative analysis of potential substrates, borosilicate glass tubes were selected. Glass tubes can easily be stacked to constitute a transparent monolithic multi-tube reactor, with their length parallel to the air flow in order to minimize the pressure drop. Moreover, borosilicate glass is relatively inexpensive and has excellent UV-A light transmitting properties. Based on a literature study, a sol-gel coating procedure was selected that is extremely suitable for coating glass substrates. The next step was to optimize the amount of P25 (commercial titanium dioxide) in the photocatalytic sol-gel coating for its application. More P25 in the sol-gel coating results in a higher adsorption capacity and consequently a higher photocatalytic activity, but greatly reduces the transparency of the coating. After an in-depth study, the concentration of 10 g L-1 P25 was selected as the most feasible for multi-tube reactors. Since the operation of photocatalytic reactors is based on a complex interaction of physical and chemical processes, mathematical models were developed, supported by experimental data, that include all these phenomena as a tool for reactor design and optimization. By making use of such models, time-consuming and expensive experimental research can be minimized. However, the experimental validation of models is of utmost importance to prove its reliability and accuracy. Intrinsic kinetic parameters provide the fundamentals for these models as they describe the photocatalytic reaction rate, independent of fluid dynamics, reactor geometry and radiation field. In this work they were estimated by means of a Computational Fluid Dynamics (CFD) study, based on FTIR (Fourier-transform infrared spectroscopy) experiments with a lab scale multi-tube reactor. The kinetic parameters were validated by an alternative analytic approach, emphasizing the accuracy and reliability of the simulations. Finally, the aforementioned CFD approach, based on the simultaneously modelling of airflow, mass transfer, UV light irradiation and photocatalytic reactions, was used to obtain insights for the light source configuration in upscaled multi-tube reactors. After taking all these insights and some practical implications into account, a final upscaled multi-tube reactor design was proposed and converted into a first built prototype. Subsequently, it was evaluated according the CEN-EN-16486-1 standard for VOC removal by the external scientific research center ‘CERTECH’. The scientific results, regarding the mineralization of the VOCs and photocatalytic efficiency of the reactor, demonstrated the feasibility for indoor air purification by the upscaled multi-tube reactor and the possible implementation in ventilation systems.

Abstract: This work is focused on an in-depth experimental characterization of multi-tube reactors for indoor air purification integrated in ventilation systems. Glass tubes were selected as an excellent photocatalyst substrate to meet the challenging requirements of the operating conditions in a ventilation system in which high flow rates are typical. Glass tubes show a low-pressure drop which reduces the energy demand of the ventilator, and additionally, they provide a large exposed surface area to allow interaction between indoor air contaminants and the photocatalyst. Furthermore, the performance of a range of P25-loaded sol-gel coatings was investigated, based on their adhesion properties and photocatalytic activities. Moreover, the UV light transmission and photocatalytic reactor performance under various operating conditions were studied. These results provide vital insights for the further development and scaling up of multi-tube reactors in ventilation systems which can provide a better comfort, improved air quality in indoor environments, and reduced human exposure to harmful pollutants.

Abstract: A novel multi-tube photoreactor is presented with a high efficiency (over 90% conversion) toward the degradation of acetaldehyde in air under UV conditions with an incident intensity of 2.1 mW cm−2. A CFD model was developed to simulate the transient adsorption and photocatalytic degradation processes of acetaldehyde in this reactor design and to estimate the corresponding kinetic parameters through an optimization routine using the experimentally determined outlet concentration profiles. The CFD model takes into account the entire reactor geometry and all relevant flow parameters, in contrast to analytical methods that often oversimplify the physical and chemical process characteristics. Using CFD, we show that both adsorption and desorption rate constants increase by respectively one and two orders of magnitude when the UV light is switched on, which clearly affects the transient behavior. The agreement of the experimental and modelled concentration profiles is excellent as evidenced by a coefficient of determination of at least 0.965. To demonstrate the reliability and accuracy of all parameters obtained from the modelling approach, an ultimate validation test was performed using other conditions than the ones used for estimating the kinetic parameters. The model was able to accurately simulate simultaneous adsorption, desorption and photocatalytic degradation.

Abstract: The combination of a non-thermal plasma with catalysis is considered as a sustainable indoor air purification technology to achieve complete oxidation at reduced energy cost with a longer electrode lifetime. An optimal window of operation for plasma assisted catalysis is found by varying the polarity, the applied voltage, the relative humidity of the gas phase and the configuration of the plasma reactor. The results show that, in general, negative corona discharge can obtain higher nitric oxide (NO) conversion efficiencies compared to positive corona. It is also clear that at higher applied voltages, higher conversion efficiency can be reached. The effect of relative humidity, however, is not found to be significant in the range (0 20.3 %) tested in this work. Additionally, the configuration of the plasma reactor is changed by varying the amount of pins that are attached at the collector electrode. The results show that there is an optimum at 10 pairs of pins to obtain a high conversion efficiency of NO. By applying a coating on the collector electrode of the plasma reactor, it is possible to see the influence of the coating on the performance of the plasma system, which was operating in the previously found optimal window. It stands clear that the use of a plasma assisted catalysis system has high potential as an integrated and sustainable indoor air purification technology.

Abstract: High-rate activated sludge (HRAS) is an essential cornerstone of the pursuit towards energy positive sewage treatment through maximizing capture of organics. The capture efficiency heavily relies on the degree of solid separation achieved in the clarifiers. Limitations in the floc formation process commonly emerge in HRAS systems, with detrimental consequences for the capture of organics. This study pinpointed and overcame floc formation limitations present in full-scale HRAS reactors. Orthokinetic flocculation tests were performed with varying shear, sludge concentration, and coagulant or flocculant addition. These were analyzed with traditional and novel settling parameters and extracellular polymeric substances (EPS) measurements. HRAS was limited by insufficient collision efficiency and occurred because the solids retention time (SRT) was short and colloid loading was high. The limitation was predominantly caused by impaired flocculation rather than coagulation. In addition, the collision efficiency limitation was driven by EPS composition (low protein over polysaccharide ratio) instead of total EPS amount. Collision efficiency limitation was successfully overcome by bio-augmenting sludge from a biological nutrient removal reactor operating at long SRT which did not show any floc formation limitations. However, this action brought up a floc strength limitation. The latter was not correlated with EPS composition, but rather EPS amount and hindered settling parameters, which determined floc morphology. With this, an analysis toolkit was proposed that will enable design engineers and operators to tackle activated solid separation challenges found in HRAS systems and maximize the recovery potential of the process. (C) 2018 Elsevier Ltd. All rights reserved.

Abstract: Deammonification systems are being implemented as cost- and resource-efficient nitrogen removal processes. However, their complexity is a major hurdle towards successful transposition from side- to mainstream application. Merely out-selecting nitrite oxidizing bacteria (NOB) or retaining anammox bacteria (AnAOB) does not guarantee efficient mainstream deammonification. This paper presents for the first time the interactions and synergies between kinetic selection, through management of residual substrates, and physical selection, through separation of solid retention times (SRTs). This allowed the formulation of tangible operational recommendations for successful deammonification. Activity measurements were used to establish retention efficiencies (η) for AnAOB for full-scale cyclones and rotating drum screens installed at a sidestream and mainstream deammonification reactor (Strass, Austria). In the sidestream reactor, using a screen (η = 91%) instead of a cyclone (η = 88%) may increase the capacity by up to 29%. For the mainstream reactor, higher AnAOB retention efficiencies achieved by the screen (η = 72%) compared to the cyclone (η = 42%) induced a prospective increase in capacity by 80–90%. In addition, the switch in combination with bioaugmentation from the sidestream made the process less dependent on nitrite availability, thus aiding in the outselection of NOB. This allowed for a more flexible (intermittent) aeration strategy and a reduced need for tight SRT control for NOB washout. A sensitivity analysis explored expected trends to provide possible operational windows for further calibration. In essence, characterization of the physical selectors at full scale allowed a deeper understanding of operational windows of the process and quantification of capacity, ultimately leading to a more space and energy conservation process.

Abstract: Thermophilic nitrification has been proven in lab-scale bioreactors at 50 °C. The challenge is now to develop a solution for thermophilic nitrogen removal, integrating nitrification with denitrification and aerobic carbon removal. This pioneering study aimed at a single-sludge nitrification/denitrification process at 50 °C, through exposing nitrification in a step by step approach to anoxia and/or organics. Firstly, recurrent anoxia was tolerated by a nitrifying community during long-term membrane bioreactor (MBR) operation (85 days), with high ammonium oxidation efficiencies (>98%). Secondly, five organic carbon sources did not affect thermophilic ammonium and nitrite oxidation rates in three-day aerobic batch flask incubations. Moving to long-term tests with sequencing batch reactors (SBR) and MBR (>250 days), good nitrification performance was obtained at increasing COD/Ninfluent ratios (0, 0.5, 1, 2 and 3). Thirdly, combining nitrification, recurrent anoxia and presence of organic carbon resulted in a nitrogen removal efficiency of 92–100%, with a COD/Nremoved of 4.8 ± 0.6 and a nitrogen removal rate of 50 ± 14 mg N g−1 VSS d−1. Overall, this is the first proof of principle thermophilic nitrifiers can cope with redox fluctuations (aerobic/anoxic) and the aerobic or anoxic presence of organic carbon, can functionally co-exist with heterotrophs and that single-sludge nitrification/denitrification can be achieved.

Abstract: In specific municipal and industrial cases, thermophilic wastewater treatment (>45 °C) might bring cost advantages over commonly applied mesophilic processes (1035 °C). To develop such a novel process, one needs sound parameters on kinetics, sludge yield and sludge settleability of three heterotrophic conversions: aerobic carbon removal, denitritation and denitrification. These features were evaluated in acetate-fed sequencing batch reactors (30, 40, 50 and 60 °C). Higher temperatures were accompanied by lower sludge production and maximum specific removal rates, resulting mainly from lower maximum growth rates. Thermophilic denitritation was demonstrated for the first time, with lower sludge production (1826%), higher nitrogen removal rates (2492%) and lower carbon requirement (40%) compared to denitrification. Acceptable settling of thermophilic aerobic (60 °C) and anoxic biomass (50 and 60 °C) was obtained. Overall, this parameter set may catalyze the establishment of thermophilic nitrogen removal, once nitritation and nitratation are characterized. Furthermore, waters with low COD/N ratio might benefit from thermophilic nitritation/denitritation.

Abstract: Anammox, the oxidation of ammonium with nitrite, is a key microbial process in the nitrogen cycle. Under mesophilic conditions (below 40 °C), it is widely implemented to remove nitrogen from wastewaters lacking organic carbon. Despite evidence of the presence of anammox bacteria in high-temperature environments, reports on the cultivation of thermophilic anammox bacteria are limited to a short-term experiment of 2 weeks. This study showcases the adaptation of a mesophilic inoculum to thermophilic conditions, and its characterization. First, an attached growth technology was chosen to obtain the process. In an anoxic fixed-bed biofilm bioreactor (FBBR), a slow linear temperature increase from 38 to over 48 °C (0.05–0.07 °C d−1) was imposed to the community over 220 days, after which the reactor was operated at 48 °C for over 200 days. Maximum total nitrogen removal rates reached up to 0.62 g N L−1 d−1. Given this promising performance, a suspended growth system was tested. The obtained enrichment culture served as inoculum for membrane bioreactors (MBR) operated at 50 °C, reaching a maximum total nitrogen removal rate of 1.7 g N L−1 d−1 after 35 days. The biomass in the MBR had a maximum specific anammox activity of 1.1 ± 0.1 g NH4+-N g−1 VSS d−1, and the growth rate was estimated at 0.075–0.19 d−1. The thermophilic cultures displayed nitrogen stoichiometry ratios typical for mesophilic anammox: 0.93–1.42 g NO2--Nremoved g−1 NH4+-Nremoved and 0.16–0.35 g NO3--Nproduced g−1 NH4+-Nremoved. Amplicon and Sanger sequencing of the 16S rRNA genes revealed a disappearance of the original “Ca. Brocadia” and “Ca. Jettenia” taxa, yielding Planctomycetes members with only 94–95% similarity to “Ca. Brocadia anammoxidans” and “Ca. B. caroliniensis”, accounting for 45% of the bacterial FBBR community. The long-term operation of thermophilic anammox reactors and snapshot views on the nitrogen stoichiometry, kinetics and microbial community open up the development path of thermophilic partial nitritation/anammox. A first economic assessment highlighted that treatment of sludge reject water from thermophilic anaerobic digestion of sewage sludge may become attractive.

Abstract: The use of temperature-time series measured in streambed sediments as input to coupled water flow and heat transport models has become standard when quantifying vertical groundwater-surface water exchange fluxes. We develop a novel methodology, called LPML, to estimate the parameters for 1-D water flow and heat transport by combining a local polynomial (LP) signal processing technique with a maximum likelihood (ML) estimator. The LP method is used to estimate the frequency response functions (FRFs) and their uncertainties between the streambed top and several locations within the streambed from measured temperature-time series data. Additionally, we obtain the analytical expression of the FRFs assuming a pure sinusoidal input. The estimated and analytical FRFs are used in an ML estimator to deduce vertical groundwater-surface water exchange flux and its uncertainty as well as information regarding model quality. The LPML method is tested and verified with the heat transport models STRIVE and VFLUX. We demonstrate that the LPML method can correctly reproduce a priori known fluxes and thermal conductivities and also show that the LPML method can estimate averaged and time-variable fluxes from periodic and nonperiodic temperature records. The LPML method allows for a fast computation of exchange fluxes as well as model and parameter uncertainties from many temperature sensors. Moreover, it can utilize a broad frequency spectrum beyond the diel signal commonly used for flux calculations.