No description available.

This thesis explores various novel ways of treatment of wastewater contaminated by toxic organic pollutants using single and combined advanced oxidative wastewater treatment technologies in conjunction with a variety of acoustic and hydrodynamic cavitational reactors. There have been many reports in the literature on the use of hydroxyl radicals as the core part of AOPs and hence, as the first objective, the amount of hydroxyl radical generation from different acoustic and hydrodynamic cavitational reactors was studied using the potassium iodide dosimeter. The results reveal that optimum concentrations of less toxic chloroalkanes (chloroform and dichloromethane) could be efficient alternatives to carbon tetrachloride for enhancement of hydroxyl radical generation in cavitational reactors. Increasing ultrasonic amplitudes and operating hydrodynamic cavitational pressures lead to higher rates of hydroxyl radical production. Having explored the efficiency of generation of hydroxyl radicals the capacity of the reactors to degrade the model pollutant phenol, via a modified classic Fenton reaction which uses zero valent iron catalysts (instead of iron salts) and hydrogen peroxide under acidic conditions was studied. This process, named the advanced Fenton process (AFP), is the main foundation of the phenolic wastewater treatment reported in this thesis. Phenol degradation was assessed using different frequencies of ultrasound where a comparison between 20, 300 and 520 kHz ultrasonic reactors showed that 300 kHz was by far the most efficient US reactor resulting in 100% phenol removal and 37% total organic carbon (TOC) mineralization in 25 min. The concept of Latent Remediation (LR) was discovered during investigations into innovative approaches towards development of cost/energy-effective methods to treat phenolic wastewater. LR consists of inputting only 15 min of either ultrasound or stirring to the reaction medium, which contains optimised amounts of hydrogen peroxide and iron catalyst, and then the silent-dark AFP phenol degradation was studied over 24 h. The excellent results revealed that >80% TOC mineralization was achieved after this time. It was also found that zero valent copper catalysts were effective for phenol degradation and offered an excellent alternative to iron in the AFP, however toxicity analysis on the 24, 48 and 72 h samples showed that zero valent iron exhibited decreased toxicity when compared to zero valent copper. Conventional granular/powdered activated carbons were replaced with activated carbon cloth and investigations on the potential use of this material for phenol removal/decomposition was studied in detail at different operating pHs (3, 5.5 and 9), temperatures (20, 40 and 80 °C), oxidants (H2O2/O3) in various reactors (pump, shaker and US bath). Another aspect of the AOP application, disinfection of natural waters, was studied employing hydrodynamic cavitation and ozonation in a novel Liquid Whistle Reactor system. Model markers of faecal coliforms, <i>Escherichia coli</i>, were chosen for the study and the combined technologies of hydrodynamic cavitation and stepwise ozonation proved be highly beneficial, resulting in ~ 6 log bacterial reduction revealing 99.9999% disinfection efficiency of the process.

In recent years, there has been an increased interest in forward osmosis (FO) from academic research and industry with a rising number of FO academic publications in the last decade. The common perception of FO as a low energy process compared to reverse osmosis (RO) sparked interest in this area. Nevertheless, there are some major challenges that need to be addressed before FO can be successfully implemented as an effective technology. Some of these challenges are addressed in this dissertation, starting with the assessment of FO as a low energy process. A modelling approach was used to assess the energy consumption of various FO hybrid processes and provide a detailed comparison with RO for desalination, in an effort to answer the critical question: Is FO truly a low energy process compared to RO? Results showed that there was practically no difference in specific energy consumption (SEC) between standalone RO, and FO with nanofiltration (NF) draw solution (DS) recovery; this can be generalised for any pressure-driven membrane process used for the DS recovery stage in a hybrid FO process, such as UF or RO. It was also found that even if any or all of the membranes considered, FO, RO or NF, were perfect (i.e. had infinite permeance and 100% rejection), it would not improve the SEC significantly. Furthermore, in order to reduce the higher membrane footprint required by FO hybrid processes, internal concentration polarisation (ICP) within the support has to be greatly reduced or eliminated. Hence, any advantage possessed by the FO hybrid process derives from the lower fouling propensity of FO, lower pretreatment costs arising from reduced fouling, use of draw solutes which can be recovered with low cost thermal energy sources and specific applications where RO cannot compete. Inspired by this insight, subsequent work was performed to study the multifaceted interactions alongside membrane and process parameters involved in the fouling of FO membranes, specifically the HTI TFC and CTA membrane. The chapter on organic fouling behaviour of structurally and chemically different FO membranes revealed that fouling on the HTI TFC membrane was more significant compared to HTI CTA in both membrane orientations, arising from a variety of factors associated with surface chemistry, membrane morphology and structural properties. Interestingly, it was observed that in FO mode, membrane surface properties dominated over fouling layer properties in determining fouling behaviour, with some surface properties (e.g. surface roughness) having a greater effect on fouling than others (e.g. surface hydrophilicity). In PRO mode, structural properties of the support played a more dominant role whereby fouling mechanism was specific to the foulant size and aggregation as well as the support pore size relative to the foulant. Whilst pore clogging was observed in the TFC membrane due to its highly asymmetric and porous support structure, fouling occurred as a surface phenomenon on the CTA membrane support layer, indicating that the latter’s structure was more symmetric in relation to the foulant (alginate) studied. Besides pore clogging, the severe fouling observed on the TFC membrane in PRO mode was due to a high specific mass of foulant adsorbed in its porous support. A new method was successfully introduced to quantify the density of the fouling layer and correlate it with hydrodynamic conditions and fouling behaviour of the membranes studied. It was observed that a trade-off between enhanced membrane performance and fouling mitigation is apparent in these membranes, with both membranes providing improvement in one aspect at the expense of the other. Hence, significant development in their surface and structural properties are needed to achieve good anti-fouling properties without compromising flux performance. Measured fouling densities on the studied surfaces suggest that there is not a strong correlation between foulant-membrane interaction and fouling density. Cleaning results suggest that physical cleaning was more efficient on the CTA membrane compared to the TFC membrane. Further, they implied that despite different mechanisms of fouling and quantities of foulant adsorbed in FO membranes, FO is a resilient process with high cleaning efficiencies and fouling reversibility. Finally, to address the challenge of ICP, a novel method of fabricating FO membranes was developed by interfacially polymerising a free-standing, salt rejecting polyamide (PA) film using a floating technique and directly depositing this layer onto an open mesh fabric. By doing this, the need for a phase inversion support was entirely eliminated. The fabrication method resulted in the successful formation of a defect-free, salt-rejecting FO membrane with significantly reduced or eliminated ICP, attributed to large open mesh sizes and straight channels in the fabric support. Interestingly, it was observed that even in the absence of ICP, flux was limited by the support layer at lower effective open areas of the mesh fabric. At higher mesh sizes and effective open areas, the effect of the fabric support became less significant and FO performance was likely governed by diffusion through the PA film, limited by its structure and transport properties. A trade-off between surface roughness and thickness of the PA film was observed, which is linked to the mechanism of film formation at the bulk interface. It was proposed that the design of FO membranes with ideal supports should also include tailoring the PA film properties in order to achieve superior FO performance. Additionally, the use of supports with higher percentage open areas or porosities should be considered.

Ultraviolet-C light emitting diodes (UV-C LEDs) are a promising technology for water disinfection. Over the last decade UV-C LEDs have shown dramatic improvements in output power, efficiency, lifetime, and cost. The low-voltage power supply and effective point source geometry of LED light sources allows for new applications and designs. Mercury gas discharge lamps have been used in UV disinfection for several decades, and as a result the industry has developed around the limited specifications of these lamps, such as their size and shape, high power requirements, and mercury content. The increased design freedom offered by LED sources is expected to revolutionise the UV disinfection industry. A model for the optical (fluence rate) distribution within a UV photoreactor was developed to match the freedoms afforded by the use of LED sources. The model development focused on design flexibility with regard to the shape of the reactor body to allow for the accurate modelling of unusual designs. Particular attention was paid to the modelling of internal reflections within the reactor. A Monte Carlo method was selected, where general solutions may be implemented to specific conditions within an arbitrary reactor design. The model focussed on the effects of optical design and did not include fluid dynamics simulations. A series of batch reactors, from 100 – 7500 mL in volume, were modelled for a range of internal reflectivity conditions, from wholly absorptive to specular or diffuse reflectivity. The model produced a 3D representation of the fluence rate distribution within each reactor; this distribution could be evaluated qualitatively to inform future reactor designs. The specific required inactivation energy (SRIE) was defined to allow for the quantitative comparison of different reactors. It was determined that the reactor designs with the highest efficacy have diffusely reflective internal surfaces: the scattering of radiation by diffuse surfaces increases the fluence rate uniformity, which is important for efficient disinfection. The parameter space available for investigation is vast; it was therefore not possible to find a globally optimised solution. However, variation across three parameters showed a broad range of near-optimum designs which resulted in a comparable disinfection performance. This suggests that relative improvements that would result from a global optimisation may be small. A cylindrical reactor of internal depth 90 mm and diameter 119 mm, with Lambertian diffusely reflective internal surface, illuminated by a single LED source (of arbitrary optical power) could achieve an SRIE of 6.56 mJ mL-1 for a treatment water of 90 % transmittance. For an LED source of 100 mW and peak emission 254 nm, this corresponds to a treatment time of 66 s for the 1000 mL volume of the proposed reactor, under a static batch treatment regime. This predicted performance of this device is suitable for application in home-scale point of use water disinfection.

Reservoir managers use operational rule curves as guides for managing and operating reservoir systems. However, this approach saves no water for impending droughts, resulting in large shortages during such droughts. This problem can be tempered by integrating hedging with the rule curves to curtail the water releases during normal periods of operation and use the saved water to limit the amount and impact of water shortages during droughts. However, determining the timing and amount of hedging is a challenge. This thesis presents the application of genetic algorithms (GA) for the optimisation of hedging-integrated reservoir rule curves. However, due to the challenge of establishing the boundary of feasible region in standard GA (SGA), a new development of the GA i.e. the dynamic GA (DGA), is proposed. Both the new development and its hedging policies were tested through extensive simulations of the Ubonratana reservoir (Thailand). The first observation was that the new DGA was faster and more efficient than the SGA in arriving at an optimal solution. Additionally, the derived hedging policies produced significant changes in reservoir performance when compared to no-hedging policies. The performance indices analysed were reliability (time and volume), resilience, vulnerability and sustainability; the results showed that the vulnerability (i.e. average single periods shortage) in particular was significantly reduced with the optimised hedging rules as compared to using the no-hedging rule curves. This study also developed a monthly inflow forecasting model using artificial neural networks (ANN) to aid reservoir operational decision-making. Extensive testing of the model showed that it was able to provide inflow forecasts with reasonable accuracy. The simulated effect on reservoir performance of forecasted inflows vis-à-vis other assumed reservoir inflow knowledge situations showed that the ANN forecasts were superior, further reinforcing the importance of good inflow information for reservoir operation. The ability of hedging to harness the inherent buffering capacity of existing water resources systems for tempering water shortage (or vulnerability) without the need for expensive new-builds is a major outcome of this study. Although applied to Ubonratana, the study has utility for other regions of the world, where e.g. climate and other environmental changes are stressing the water availability situation.

No description available.

A water distribution system WDS is often divided into smaller isolated and independent zones called district metering areas (DMA). A DMA can have anywhere from a few hundred to a few thousand properties. Normally only three locations within a district metering area are actively monitored for pressure or flow readings. These are the supply point pressure and flow and the critical point pressure which is the point of the lowest pressure in the DMA. As leakage rates are typically directly proportional to average pressures in the DMA, keeping the network pressure as low as possible while maintaining desired serviceability is an effective and widely used method for leak reduction. With advancement in technology this network pressure reduction is now done in real-time, where the network pressure is increased or decreased based on the demand. However, such real-time optimisation changes the DMA dynamics making it different from traditional unoptimised DMAs. We consider the problem of detecting and quantifying leaks in pressure optimised DMA, using only these three DMA-level hydraulic measurements. The DMA-level measurements represent the current aggregate water demand/consumption within the DMA. Detecting leaks at this point is challenging, particularly small leaks, as they do not produce a significant increase in the aggregated DMA-level measurements. Furthermore, the DMA-level data exhibits input signal dependence whereby both noise and leaks are dependent on the flow and pressure being measured, making leak detection task more difficult. To address this, we first propose a Gaussian process (GP) based approach that uses only the DMA-level flow to detect leaks (NSGP). We devise an additive diagonal noise covariance for the GP that is able to handle the input dependant noise observed in this setting. A parameterised mean step change function is used to detect and approximate leaks. As accurate leak data is often not available due to poor record keeping, we develop a detailed simulated model of a pressure optimised DMA and use it for analysing proposed leak detection methods. We show that active pressure optimisation changes the dynamics of a DMA. In light of the change in DMA dynamics, we proposed a domain specific, data driven, multi output gaussian process model, to detect and quantify leaks in pressure optimised DMAs (SMOGP). The novelty of the model is, firstly its ability to use all available information from a DMA to detect leaks, secondly the ability to model the pressure dependant leak process mathematically within the GP framework. We compare the performance of the proposed methods with the current state of the art leak detection method. We show that our proposed method out perform other approaches considerably both in terms of the accuracy of leak detection and leak magnitude estimation.

Membrane filtration is a simple concept for water purification: water containing particulate contaminants is forced through a semi-permeable membrane that rejects the particulates leaving clean water to flow out. Nevertheless, there are many complex features of membrane filtration, the most important of which is the accumulation of the particulates at the membrane surface. This leads ultimately to fouling of the membrane and a reduction in the efficiency of the process. Concentration polarization is the precursor of fouling, that is, a high concentration of contaminants develops in front of the membrane without the contaminants attaching to each other or the membrane surface. However, several types of acute membrane fouling develop from the layer formed in concentration polarization, including internal fouling, pore blocking and caking. Addressing these and related problems has been at the forefront of membrane research since the process' inception. In this thesis we develop mathematical models of aspects of crossflow and directflow filtration operating at constant flux. We begin by addressing questions related to the initial stages of concentration polarization in crossflow systems. In particular, we study the influence of particulates on the viscosity of the filtrate, and show how the filtration efficiency may be improved by tailoring the wall permeability to reduce the effects of osmosis. We then address the development of membrane fouling and caking in directflow systems: the transmembrane pressure difference, the possibility of elastic deformations during filtration, and the influence of these on the development of fouling and caking are all considered. We show that even small elastic effects can worsen fouling and suggest how the process can be operated to avoid this. We then discuss further opportunities for mathematical modelling in this area.

A number of low-cost synthetic filtration media have been proposed for the removal of arsenic (As) from drinking water in areas such as Bangladesh, where exposure to environmental arsenic is a major human health issue. This PhD research project examines the application of recycled glass and waste stainless steel fragments as a practical medium for arsenic removal at a household scale. To assess the performance of recycled glass media as a practical filter bed, glass granules were differentiated by colour, size and mode of glass size reduction (imploded and ground). The selected glass granules were used as media for batch adsorption and column filtration experiments using a prepared As (III) test solution and using natural As-contaminated water in Bangladesh, where recycled glass in column filtration mode was used to treat arsenic contaminated natural water in the presence of other metalloids. Filter media made from recycled glass and waste stainless steel fragments were characterized via SEM and PXRF. SEM study also gave information about the mechanism of arsenic removal by glass granules. Sequential extraction experiments were also performed on used filtration media to assess arsenic removal and adsorption processes. Results indicate that glass granules associated with stainless steel fragments (sstl) can remove arsenic from drinking water at an efficiency suitable for household application. Arsenic removal effectiveness depends largely on the presence of stainless steel fragments with glass (introduced with the glass media during the recycling and preparation process). The glass particle size and mode of size reduction was also found to influence the removal of arsenic: ground glass performed better than imploded glass and smaller ground glass particles (s < 0.5 mm) performed better than imploded glass of the same size. Batch experiment results concluded that glass colour may have minor influences on arsenic removal although the differences were not significant. Further results also revealed that < 0.05 kg sstl can remove arsenic to below acceptable limits from a 0.50 ppm arsenic solution with an effectiveness > 0.168 mg/g sstl. It was found that 57 kg of small clear DSGF (dry sieved ground fresh) glass (s < 0.5 mm) can treat 132.5 l of water with 100% removal of arsenic from starting concentrations of 0.50 ppm, using a recycled glass filter column. There is a scope for improvement of the glass filter media by adding stainless steel fragments but the study did not determine the potential and further work is required to optimize the ratio of DSGF glass and stainless steel fragments. Considering the price and operational drawbacks of other existing filters in Bangladesh, recycled glass has potential to be used in more sustainable arsenic filter filtration units. The results, coupled with the low cost of waste glass, indicate that waste glass should be investigated further for use in domestic water filtration for arsenic removal.

The environmental impacts of oil pollution in the Niger Delta have been well-documented, with research centered on transnational environmental justice issues, institutional failings at government level, and the political strategies of high profile representatives of the Ogoni community. However, the responses of affected communities has elicited relatively limited attention from academic researchers, and in particular, little is understood about the complex factors that shape how households have sought to tackle the impact of environmental degradation. This thesis addresses this gap by analysing the ways in which rural households in this region have sought to adapt and rebuild their livelihoods. The thesis combines insights from vulnerability science with a sustainable livelihoods approach to develop a conceptual framework that draws attention to the social dynamics of capacity and resilience as expressed in households’ everyday livelihood practices, and the factors that support or impede these. The study deploys a mixed method case study approach in the Ogoni community of Bodo, where people continue to struggle with the aftermath of two incidents of spillage from a Shell Petroleum Development Company of Nigeria (SPDC) pipeline in 2008. Questionnaires, semi-structured interviews, focus group discussions and a desk-based survey of regional natural resource management policies are used to elicit data on land use, livelihood strategies and the impacts of oil spill, and to explore factors shaping livelihood responses to the oil spill and interventions that have followed in its wake. Findings show that social differences are critical in shaping vulnerability and capacity, and that patterns of disadvantage have become more entrenched not only as a result of oil spill, but through the unintended consequences of institutional and household responses. Gender emerges as a particularly salient factor that has contributed to variations in household resilience in the face of oil spill and the socio-political environment that followed. Environmental degradation had a particularly heavy impact on women’s farming and shell fish collecting activities, damaging household economies, but also dismantling many of the social networks that formed around women’s collective labour. Moreover, corporate social responsibility interventions had unintended gender consequences, as emphasis was given to reinforcing ‘traditions’ that limited women’s empowerment. The research findings advance an understanding of the complex gender dynamics that result in socially differentiated pathways towards greater vulnerability or resilience in the face of anthropogenic environmental hazards, in this case relating to oil pollution and its clean-up in Ogoniland.