Dealing with municipal solid waste has become a problematic issue in the United Kingdom (UK). With actions to mitigate the potentially adverse impacts of climate change debate and space for, and costs of, landfill becoming critical, a landfilldominated strategy is no longer acceptable. In this context, the attitudes and behaviour of young adults, particularly university students, who often have little experience of taking responsibility for waste management activities, have not been studied in great detail. Since the 1960s, the UK higher education system has expanded six fold to >2.4 million students. The overall production of waste at Higher Education institutions (HEIs) is therefore very large and presents significant challenges as the associated legislative, economic and environmental pressures can be difficult to control and manage. Therefore, a comprehensive research focusing on university students is urgently required. Changing the way HEIs deal with their waste is an important issue because of fast-changing legislation and increasing costs. The solution is a new approach to waste management: a revolutionary change in the way that HEIs think, the way HEIs act and the way HEIs handle their waste. This has massive implications for the Higher Education (HE) sector. It means developing extensive institution-wide infrastructure to provide greater flexibility and user-centric solutions to suit the need of students and staff. It also means that HEIs work together and potential collaboration between HEIs and Local Authorities (LAs) to maximise resource efficiency, meet future legislative requirements and achieve their corporate responsibilities and commitments. This thesis reports on a study of waste management practices at HEIs in the UK. The issue was approached from both a theoretical and a practical standpoint. The study used the University of Southampton (UoS) as a case study and examined how waste recycling projects can be developed effectively using infrastructure, service provision and behavioural change techniques as part of a wider research programme investigating waste management in medium- and high-density housing. The study clearly showed that there was potential for significantly improving reuse and recycling at university halls of residence (HoR) and that more convenient and higher quality infrastructure and service provision resulted in higher recycling rates. Furthermore, students have lifestyles that impact significantly on waste arisings and consequently on waste management operations at HoR (and probably at HEIs and student-dominated residential areas). For schemes to be successful at HEIs, they must be based on a thorough understanding of students’ recycling behaviour, and their perceptions of the barriers to recycling. The key to unlocking behaviour change lies in the provision of appropriate infrastructure and effective service provision alongside a targeted behaviour change programme. Mass media coverage especially the Internet has a rising influence on university students’ environmental knowledge while environmental education at school has become the secondary source of information. The results also revealed that university students possessed less knowledge than they believed which makes informative behavioural interventions a vital component of effective recycling schemes at HEIs

The role of voids in composite failures has not been well understood or been characterized. This study presents the result of an investigation aimed at examining the mechanical behaviour of unidirectional carbon fibre-reinforced polymer matrix materials as used by the wind turbine industry. The experimental program investigates the effect of the void content, different manufacturing methods, type of loading (the three-point bending vs. the four-point bending) on the static strength and fatigue life under a flexural load. The four ply unidirectional carbon/epoxy composites were manufactured using the SPRINT and prepreg manufacturing methods. Manufacturing by these methods has successfully produced the composite materials with varied void contents and the voids are found to concentrate primarily in the area where the adjacent plies meet. The SPRINT materials contain voids in the range of 1.63- 2.89% while the prepreg laminates have an average void content of 3.6% for the debaulked laminates and 6.8% for the non-debaulked laminates. The voids in the SPRINT laminate are small and they are distributed inhomogeneously. The majority of the voids are less than 0.04 mm2 in size with a shape between a circle and an ellipse. The three point bending tests show that the flexural strength exponentially decrease as much as 6.4% for the SPRINT and 6% for the prepreg for every 1% increase of the void content. The similar void effect for both materials may be due to their similar microstructures. There is a similar effect on the increase in the void content for the flexural strength and the flexural fatigue sensitivity for both the three and four point bending tests. This means that the voids do not play any large role in the three-point bending as compared to the four point bending. The mechanism of failure is observed using the optical microscopy and the X-ray tomography of the polished edge and cross section of the area near the failure. The voids are found to interact with the cracks in both the static and fatigue failures. The image of the cross section near the failure area of the static test samples seem to suggest that the voids coalesce to each other under loading. By using the acoustic emission, it can be confirmed that a major failure occurs due to the fibre breakage. In addition, the acoustic emission results also show that the failure behaviour of the low and high void content specimens is significantly different.

Bio-fuel cells are driven by diverse and abundant bio-fuels and biological catalysts. The production/consumption cycle of bio-fuels is considered to be carbon neutral and, in principle, more sustainable than that of conventional fuel cells. The cost benefits over traditional precious-metal catalysts, and the mild operating conditions represent further advantages. It is important that mathematical models are developed to reduce the burden on laboratory based testing and accelerate the development of practical systems. In this study, recent key developments in bio-fuel cell technology are reviewed and two different approaches to modelling biofuel cells are presented, a detailed physics-based approach, and a data-driven regression model. The current scientific and engineering challenges involved in developing practical bio-fuel cell systems are described, particularly in relation to a fundamental understanding of the reaction environment, the performance and stability requirements, modularity and scalability. New materials and methods for the immobilization of enzymes and mediators on electrodes are examined, in relation to performance characteristics and stability. Fuels, mediators and enzymes used (anode and cathode), as well as the cell configurations employed are discussed. New developments in microbial fuel cell technologies are reviewed in the context of fuel sources, electron transfer mechanisms, anode materials and enhanced O2 reduction. Multi-dimensional steady-state and dynamic models of two enzymatic glucose/air fuel cells are presented. Detailed mass and charge balances are combined with a model for the reaction mechanism in the electrodes. The models are validated against experimental results. The dynamic performance under different cell voltages is simulated and the evolution of the system is described. Parametric studies are performed to investigate the effect of various operating conditions. A data-driven model, based on a reduced-basis form of Gaussian process regression, is also presented and tested. The improved computational efficiency of data-driven models makes them better candidates for modelling large complex systems.

This research addresses gaps in current knowledge regarding process issues associated with long term semi-continuous digestion of food waste as a sole substrate, and the role of trace elements and biomass retention in digestion of food wastes. Source segregated food wastes were collected from a university catering facility and found, in characterisation studies, to have a total solids (TS) content of 28.1±0.25 %, a volatile solids (VS) content of 95.5±0.06% of TS and a chemical oxygen demand (COD) of 422±16 g kgwet weight -1. The total Kjeldahl nitrogen (TKN) and total lipid content were 22±1% and 3.8±0.24% of TS, respectively. The substrate was then processed during a number of digestion trials using mesophilic continuously-stirred tank reactors (CSTRs), to establish the suitability of this substrate for CSTR digestion. It was found that although good specific methane production of 0.36 l gVSadded -1 was obtained from the substrate, the process was unstable at a hydraulic retention time (HRT) of 25 days, with methanogenic failure occurring after 80 days or when the organic loading rate (OLR) was increased. Further digestion trials were initiated, therefore, to investigate the effects of trace element supplementation and extending HRT on process stability, areas for which there is little information in existing literature. Reactors with hydraulic retention times of 25, 30, 50, 100, and 180 days supplemented with a trace element solution showed stable digestion for longer periods than duplicate control digesters without supplementation. The time points of failure in the control digesters were shown to be related to washout time, as calculated using the HRT. Trace element supplementation allowed stable operation at an OLR up to 3.5 gVS l-1d-1, with specific methane production ranging from 0.41-0.47 l gVSadded -1 and VS destruction of 63-77%. Supplementation with trace elements did not, however, guarantee indefinite stable operation, as digesters at the shortest (25 days) and longest (180 days) retention time eventually showed methanogenic failure. A slow methanogenic biomass growth rate and accumulation of inhibitory substances, respectively, were hypothesised as possible reasons for these failures. Analysis of metal concentrations in the digestate showed that cobalt was the metal most likely to be responsible for the observed benefits of the mixed trace metal supplementation as the concentration of this increased in the supplemented digester whilst decreasing in its non-supplemented control. The relative importance of the liquid and solid fractions in maintaining stability were investigated in novel digestion trials in which solid and liquid retention times were uncoupled. Digesters with SRT of 25 days and HRT of over 150 days exhibited methanogenic failure after approximately 45 days. In contrast, reactors with SRT of over 150 days and HRT of 25 days maintained stable digestion, with specific methane production of 0.53 l gVSadded -1, and also showed recovery from a thermal shock applied during the experiment. Inhibitory compounds such as VFA were kept low by flushing through the system while alkalinity was regenerated by the action of biomass kept in the system. The retention of solids may also have facilitated the retention of trace metals.

The potential for using the crop maize (Zea mays) for the production of biogas in simple anaerobic leach beds was evaluated. The results showed that leach beds coupled to high-rate methanogenic reactors performed better than other systems on a specific methane yield per gram of substrate added basis while their performance on a volumetric gas yield basis was poorer. Initial experiments using a single stage digester showed rapid acidification due to the low buffering capacity of the system. To overcome this problem, leach beds were used as part of a two-phase system in which the intermediate metabolic products were flushed out and used as substrate for a second stage methanogenic reactor. Further experiments simulated the effect of a hydraulic flush in the leach bed using clean water as the flush liquid. Methane potential of the leachate was estimated based on the cumulative soluble chemical oxygen demand (SCOD) production. Under this operation mode, the effect of substrate to inoculum ratio, fresh substrate load (FSL), hydraulic retention time (HRT) and buffer and trace element addition was tested. The performance of the leach bed was found to be poor compared to conventional digesters where the methane yield is ~0.35 l CH4 g-1 VSadded and volatile solids (VS) destruction is ~85% and this was thought to be due to the low pH in the reactor. Increasing the FSL improved methane yield but the maximum value obtained was 0.12 l CH4 g-1 VS. Decreasing the HRT allowed the leach bed to operate at a slightly higher pH. In this case, a volatile solids destruction of ~50% and a methane yield of 0.17 l CH4 g-1 VS was achieved using a HRT of 2.6 days. The addition of buffer (NaHCO3) to maintain pH ~ 6.5 increased VS destruction to 89% and methane yield to 0.37 l CH4 g-1 VS at an HRT of 1.5 days. This performance was similar at a HRT of 28 days despite the high VFA concentrations. Acid production increased with the addition of buffer as 75-97% of SCOD was converted to this form. Buffering was also shown to increase the number of culturable anaerobic cellulolytic microorganisms. VS degradation and methane potential were further enhanced by the addition 0.5 mg l-1 of cobalt to the buffered flush medium giving an apparent VS destruction of 115% and a methane yield of 0.46 l CH4 g-1 VSadded, which indicated that a proportion of the inoculum was also degraded. In the final part of the research the leach beds were coupled to methanogenic reactors and trials were conducted using different feed cycle durations, in which all the digestate and leachate from the preceding run was used as inoculum, and only the solids destroyed were replaced with fresh feed material. The effect of the methanogenic reactor was multi-fold as it not only stripped out intermediate compounds, according to its primary design function, but also played an important role in stabilising pH, maintaining nutrients and retaining the microbial population in the system. The leach bed operated with a 7-day feed cycle showed higher substrate degradation and was able to receive a higher OLR of 2.4 gTS l-1reactor d-1 than the 14-day feed cycle at 1.7 gTS l-1reactor d-1 and the 28-day feed cycle at 1.3 gTS l-1reactor d-1. This provided a higher volumetric methane yield the shorter the feed cycle, 0.839 lCH4 l-1 d-1, 0.618 lCH4 l-1 d-1 and 0.482 lCH4 l-1 d-1 in the 7-day, 14-day and 28-day feed cycle, respectively. However, the specific methane yield obtained from the system was slightly higher in the 14-day and 28-day cycles at 0.434 l CH4 g-1 VSadded while in the 7-day cycle it was 0.418 l CH4 g-1 VSadded. The retention of the digestate and leachate over successive cycles for a period of ~160 days appeared, however, to cause an accumulation of suspended solids (SS) and total and soluble COD in the leachate. This was especially the case in the higher loaded 7-day feed cycle reactor and was probably the cause of the lower methane production. Initially the methanogenic reactors were responsible for most methane production but with progressive cycles the leach beds themselves became methanogenic and eventually accounted for more than 50% of the methane generated in the system. The methanogenic and cellulolytic bacteria were shown to be present in the leachate from both reactors and suggested a synergy between them in exchange of microbial consortia.

Wheat straw is a major potential source of waste biomass for renewable energy production, but its high salt content causes problems in combustion. Work was undertaken to evaluate straw washing as a means of reducing the alkali index of the straw by washing out light metal cations, primarily potassium. In addition to loss of salts, organic matter is also washed out of the straw and this is a potential source of energy through anaerobic biodegradation to produce methane as a fuel gas. The rate of washout of both potassium and organic carbon was dependent on the temperature of the washwater, although cold water washing could reduce the alkali index to a suitable level, after a long retention time. Using this technique an organically dilute washwater was produced with a chemical oxygen demand of around 2.0 g l-1, suitable for a short hydraulic retention time immobilised cell digester. An upflow anaerobic sludge blanket (UASB) was chosen for initial trials, but this was later compared with an anaerobic filter. As a control throughout the experiments digesters were also maintained on a synthetic wastewater which gave a performance baseline against which the activity and methane production potential of the UASB digesters could be judged. Initial trials showed an accumulation of potassium in the granular sludge bed and an initial apparent drop in the specific methane production and COD removal efficiency. This could be recovered and the potassium washed out of the bed by switching the feed from wheat straw washwater (WSW) to synthetic sewage. Repetitive cycling between these two substrates did not damage or disrupt the digestion process. When allowed to stabilise on WSW alone the COD removal was around 83% and the specific methane production was 0.216 l g-1 CODadded under pseudo steady state conditions. The accumulation of potassium also stabilised at around 11 mg g-1 granule wet weight. Under these conditions the organic loading rate could be increased to ~3 g COD l-1 day-1 without adversely affecting digester performance. Whilst operating on wheat straw washwater the conversion of COD to methane compared to the stoichiometric potential was less than that seen for the synthetic wastewater in the same digester with the same granular biomass. It is hypothesised that a proportion of the carbon converted is used in maintaining the osmotic integrity of the cells by a metabolically-linked potassium transport system. Evidence to show intracellular accumulation of potassium was provided by transmission electron microscopy coupled to EDX analysis of granular sections. Complementary studies were carried out to determine the resistance to salt toxicity of two different types of anaerobic inoculum, taken from a mesophilic municipal wastewater biosolids digester and a saline estuarine mud, for comparison with the UASB granules. Both of these inoculums had a higher tolerance to both Na and K than the granular material, and the wastewater plant digestate was used to further acclimate a dispersed growth inoculum to seed an anaerobic filter. In the acclimatisation, which was carried out in semi-continuous fed stirred tank digesters, the digestate successfully acclimated to 10 g l-1 of KCl, NaCl and a mixture of the two salts. When operated at the same loading on either synthetic wastewater or WSW there was no difference between performance of the UASB and anaerobic filter in either COD removal efficiency or specific methane production. A simple energy balance was conducted taking into account only the energy required for heating washwater to reduce the washing time necessary to meet the alkali index for the straw. This would, however, consume most of the energy produced by anaerobic digestion of the washwater even when other energy consuming activities were not considered.

This thesis reports on experimental and modelling work carried out in order to make quantitative estimates on the environmental impacts of home composting. The focus of the work was climate relevant gaseous emissions, and developing and utilising a methodology for quantifying them. Experiments using 220L open bottomed home compost bins, alongside purpose built 200L composting reactors with airflow control were performed. A variety of composting conditions were tested, using different compositions of garden and kitchen wastes. The experiments were monitored for headspace gas composition, including CO2, O2, NH3, N2O, CH4 and volatile organic compounds, as well as temperature, humidity, moisture and solids losses and pH. From the CO2 emission rates calculated from the reactor experiments, theoretical analysis and modelling and airflow pathway tests on home compost bins, it was concluded that molecular diffusion, rather than bulk convective flow, is the dominant gas transfer mechanism from home compost bins. There were no detected emissions of N2O but emissions of NH3 up to 16 g/T feed. Only a few cases of CH4 emission were detected, typically in the first 2-3 days following a feed addition, with the highest single concentration measured at 86 ppm within the headspace. The total anthropogenic greenhouse gas emissions from home composting were estimated as between 3 and 12 Kg CO2E/Tw with almost 90% coming from the lifecycle of the compost bin. This compares with between 20 and 56 Kg CO2E/Tw from centralised facilities, at least more than double that for home composting. Total anthropogenic CO2-equivalent emissions from home composting in the UK in 2008 were estimated to be in the region of 7 thousand tonnes CO2E.

The work investigates if fly ash from Ekibatuz Power Plant can stabilise low concentrations of mercury in the environment and prevent it from becoming soluble in water and in preventing it transforming into the methylated form. The work demonstrates that mercury bound to fly ash from the coal fired 4,000 Mwatt Ekibatuz Power Plant in Kazakhstan is fairly stable at pH levels that are found in most natural water bodies. The adsorption behavior followed the Freundlich adsorption model. The adsorption capacity of the fly ash for Hg (II) was found to be 3.0 mg.g-1 of dry ash, the adsorption equilibrium being reached after 96 hours. The adsorption kinetic and studied at pHs between 6 and 8. The study showed that between the pH range of 6.0 and 8.0 bound mercury on wet and air dried ash was fairly resistant to leaching with the maximum leaching being 0.292 mg.l-1 and 0.14 mg.l-1 for the wet and air dried fly ash, respectively, with leachate at pH 7.0. Laboratory studies of the stability of the adsorbed mercury on fly ash when mixed with organic rich sediments in an anaerobic environment at pH 7.0 showed that despite ideal conditions for methylation to take place after 8 weeks, the concentration in solution was less than 2 μg.l-1. The studies showed that unburnt carbon contained in raw fly ash was the key factor for adsorption reaction. The results indicated that fly ash from the 4,000 Mwatt Ekibatuz Power Plant in Kazakhstan fired with high ash medium volatile coal can be used to stabilise low concentration of mercury in the natural aquatic environment

Buses, the main form of public transport in many urban areas, are considered as an efficient and environmentally friendly transport mode because of their high passenger capacity. The concept of bus priority was originally proposed to protect buses from urban traffic congestion so that buses can be perceived as a faster mode than private cars. One such measure which is expanding in extent and sophistication is bus priority at traffic signals – or Bus Signal Priority (BSP) Strategies. The previous research on BSP has mostly focused on its effectiveness on improving bus efficiency and bus regularity/punctuality, as well as the impacts on general traffic. However the potential environmental impacts that could be caused by BSP have not been studied, particularly on emissions - despite this being an increasingly important criterion in transport assessments. For bus priority implementations this could be particularly important, if the small disbenefits to a large number of the nonpriority vehicles outweigh the benefits to buses. This lack of knowledge and potential concern has been the main motivation for this research. The thesis sets out a comprehensive review on the state-of-the-art BSP systems and evaluation approaches. It revealed that microscopic traffic simulation models are the most appropriate approach for this study with the ability to model different BSP strategies in various user-defined scenarios. The Aimsun model was selected after review and comparison. From the review on the measurement and modelling approaches for transport related emissions, instantaneous emission models were found to be able to estimate emission behaviour by relating emission rates to vehicle operation during a series of short time intervals (often one second) and for a small scale. This was required by this study as at junction areas emissions could be dominated by vehicle operational modes where the traditionally ‘average speed’ models were unable to accurately capture the emission variations. The dynamic and individual-oriented features of microsimulation models and instantaneous emission models enabled their integration at various spatial and temporal levels and at different levels of vehicle aggregation. After calibration and validation to some critical parameters in Aimsun, a signalised junction under VA control was set up, and two BSP strategies were modelled, - one including green extension and early green recall facilities and the other one including additional compensation and inhibition facilities. These strategies were applied to 18 typical scenarios, involving variables of ‘degree of saturation’, traffic flows and bus flows. The results illustrated the importance of strategy optimising in the more challenging conditions of junctions operating under high degrees of saturation and /or high bus flows. The worst-case scenario was in heavy traffic conditions with high bus flows and BSP on the minor road only, when emissions could increase by about 10%. Under a free flow condition implementing BSP on the main road is an environmentally friendly measure with emissions reductions of up to 6%.The thesis also describes a method to value emissions in monetary terms, so that operational and emissions impacts can be compared in common units. Application of this method indicated that the impact of emissions is much smaller than that for delay/journey time impacts, though some under-estimation in emissions valuation is suspected.

The research reports on the potential for CO2 pressurisation as a means of enhancing biogas production in the anaerobic digestion of co-settled sewage sludge, a technique reported in the commercial literature as showing great benefits. The possibility of using this technique to reduce the number of faecal indicator bacteria is also explored, as an alternative means of complying with the UK water industry's Safe Sludge Matrix. Initial research using pure cultures of both E. coli (a strain isolated from sewage sludge) and Salmonella enterica showed that the results from different methods for isolating, recovering, resuscitating and enumerating E. coli were comparable. Further testing using heat stressed and unstressed E. coli showed the advantages of resuscitation and the MPN technique with 2-stage incubation was therefore used in the experiments to maximise the recovery of damaged but viable cells. The results for pure cultures showed conclusively that under the conditions of time and pressure used CO2 pressurisation and rapid depressurisation could cause irreversible cell damage and loss of viability in both test strains. The effect was reproducible and a time-pressure relationship was established for the apparatus used. It was shown, however, that the sanitising effect was influenced by culture volume, probably as this affected mass transfer of CO2 and hence its penetration into the cells. This finding may limit the practical application of the approach and further work is needed to establish design parameters and develop reactor systems to overcome this issue. Even under the favourable conditions used, exposure times required for 6 log reduction were too long for commercial application. Optimisation of the pressurisation vessel design may improve this and should be a focus of any future investigations. Batch and semi-continuous anaerobic digestion tests with co-settled sewage sludge were carried out to ascertain the effect of pressurised CO2 treatment on biogas production. These showed conclusively that the treatment did not improve either biogas productivity or specific methane yield. Experimental work also showed that even treatment conditions which gave an 8 log reduction in E. coli in pure culture were ineffective in reducing the number of indigenous E. coli in the sludge, or of S. enterica when added to a sewage sludge matrix. These findings led to further investigation of the effects of the size of pressure vessel and sample used. The results showed that this was an important factor, but could not fully explain the lack of performance in comparison with pure culture. It was concluded that the sludge matrix itself must play a role in protecting the microbial consortia from the effects of pressurised CO2. The exact reasons for this were not discovered but may be due to the effect of dissolved compounds present in the sludge and/or the structure and nature of the sludge flocs themselves.