This research aimed to investigate the bioavailability of 9 metals (Cr, Mn, Fe, Ni, Cu, Zn, Mo, Cd, and Pb) to vegetable crops (spinach, lettuce, carrot and radish) cultivated in compost soils at different levels of metal contamination. The uptake and accumulation of these metals by the plants were examined. The elemental speciation using SEC-UV-ICP-MS and Nanospray Mass Spectrometry had been performed to characterize the metal containing species induced in the plants exposed to metal stress. In order to evaluate potential health risks arising from ingestion of the metal contaminated plants, the oral bioaccessibility i.e. the use of an in vitro physiologically based extraction test (PBET) simulating the transition of the metal pollutants in the plants into the human gastrointestinal system was undertaken. It was found that, with the exception of Cr, metal concentrations (Mn, Fe, Ni, Cu, Zn, Mo and Cd) in lettuce, spinach, carrot and radish depended on the concentrations of the total metal in the soils in which the plants were grown. For Pb, the amounts accumulated in the leafy vegetables also depended on their levels of contamination in the soils while the root vegetables had rather low uptake and the uptake levels did not increase when higher levels of contamination were applied. Mn, Fe and Zn were relatively easily mobilised from soils to plants; they tended to accumulate in all plants studied at high concentrations. The elements which were more enriched in leaves included Mn and Zn (in all plant types), and Fe and Cd (only in the root vegetables). In contrast, Fe, Ni, Cu, Mo and Pb were accumulated more in roots of the leafy vegetables. Among all plants studied, it was observed that carrot had low uptake for all elements (Cr, Ni, Cu, Mo, Cd and Pb), except for Mn, Fe, Zn which were found in all plants. The metal mobilised from soil to plant as indicated by the metal contents accumulated in the plants decreased in the order Mn >> Zn > Fe > Cu > Mo > Ni > Cd > Pb Cr. The metal bioavailability to plants was assessed by measuring transfer factor (TF) values of the metals based on total metal contents in the soils. It was found that the order of TF values was Mn > Zn >> Cd > Ni > Cu > Mo Pb > Cr Fe. The mean TF values of each element irrespective of plant types were 1.93, 1.77, 0.485, 0.194, 0.111, 0.052, 0.045, 0.037 and 0.036 for Mn, Zn, Cd, Ni, Cu, Mo, Pb, Cr, and Fe, respectively. Hence, Mn and Zn were most bioavailable to plants i.e. they can be transferred from soils to plants more easily than Ni, Cu, Mo, Pb, Cr and Fe. Whereas, the bioavailability of Cd was relatively moderate. In addition, the results enabled the development of statistical regression models that are suited to predict metal uptake by plants. It indicated that the relationship between the TF values and the extractable soil metals followed the power regression curve. However, there were some cases in which it did not follow the power regression curve but a linear model, these are; Mn (for carrot leaves and radish roots), Mo (for spinach roots and carrot roots), and Cd (for lettuce leaves, spinach roots and leaves and carrot roots). In the multi-elemental speciation study, it was found that a common association of the metals (Cd, Cu, Mo, Pb, and Zn) to the high molecular weight (MW) fractions (8160 Da) was observed in all plant extracts. The lower MW fractions of approximately 1000 — 3000 Da of Cd, Cu, Mo, Ni, Pb, and Zn containing compounds were found to be present in all plant extracts. Iron was not detected in the roots of carrot and radish, but present as both high MW (8200 Da) and low MW (2500 Da) compounds in the leaves of spinach and lettuce. To characterize the individual metal containing species present in the plant samples, the Nanospray Mass Spectrometry was employed. Unfortunately, no evidence from this analysis can confirm that these compounds are related to the phytochelatin family.

A former industrial site now used for recreational activities was investigated for total PTE content, uptake of the PTEs by foraged fruits and mobility of the PTEs using single extraction such as HOAc and EDTA. In order to evaluate the health risks arising from ingestion of the PTE contaminated soil, the oral bioaccessibility using in vitro physiologically based extraction test (PBET) and tolerable daily intake (TDI) or mean daily intake (MDI) was used. The PBET simulates the transition of the PTE pollutants in the soil into human gastrointestinal system while the TDI or MDI is the mass of soil that a child would require to take without posing any health risk. In addition to the former industrial site, an investigation of the urban road dust from Newcastle city centre and its environs was undertaken with the view to looking into the PTE content, oral bioaccessibility and the platinum group elements (PGEs). Optimized microwave procedure was applied to 19 samples obtained from a former industrial site (St Anthony's lead works) in Newcastle upon Tyne. Of the range of PTEs potentially present at the site as a consequence of former industrial activity (As, Cd, Cr, Cu, Ni, Pb and Zn), the majority of top soil samples indicated elevated concentrations of one or more of these PTEs. In particular, data obtained using either inductively coupled plasma mass spectrometry (ICP-MS) or flame atomic absorption spectroscopy (FAAS) indicates the high and wide concentration of Pb on the site (174 to 33,306 mg/kg). Comparing the resulting PTEs data with UK Soil Guidelines Values (SGVs) suggests at least parts of the site represent areas of potential human health risk. It was found that Pb soil values exceeded the SGV on 17 out of the 19 sampling sites; similarly for As 7 out of 19 sampling sites exceeded the SGV. While for Cd and Ni the soil levels were below the stated SGVs. Samples of foraged fruits collected from the same site were also analysed for the same PTEs. The foraged fruit was gathered over two seasons along with samples of soil from the same sampling areas, acid digested using a microwave oven, and then analysed by ICP-MS. The foraged fruits samples included blackberries, rosehips and sloes which were readily available on the site. The concentration levels of the selected elements in foraged samples varied between not detectable limits and 24.6 mg/kg (Zn). Finally, the soil-to plant transfer factor was assessed for the 7 elements. In all cases, the transfer values obtained were below 1.00,except Cd in 2007 which is 1.00, indicating that the majority of the PTE remains in the soil and that the uptake of PTE from soil to plant at this site is not significant. The determination of total or pseudo total PTE content of soil is often insufficient to assess the risk to humans. A range of extraction protocols were applied to the 19 samples urban topsoils, and report on the correlations between pseudo total PTE content and results obtained following a physiologically-based extraction procedure (oral bioaccessibility), EDTA and HOAc extraction protocols (reagent-specific available fraction), for a broad range of PTEs (As, Cd, Cu, Cr, Ni, Pb, Zn). Results of the single-reagent extraction procedures did not, in general, provide a good indication of oral bioaccessibility but shows positive correlation with the pseudo total PTE content. The bioaccessibility data shows that considerable variation exists both spatially across the site, and between the different PTEs, but correlates well with the pseudo-total concentrations for all elements (r2 exceeding 0.8). One of the main objectives of this work is to show the role of bioaccessibility in generic risk assessment. Comparison of the pseudo-total PTE concentrations with SGV or generic assessment criteria (GAC) indicated that all of the PTEs investigated need further action, such as receptor exposure modelling.

A method has been developed to analyse PAHs in solid environmental matrices using an in-situ PFE-GC-MS method. The method involves the use of 2 g of alumina as the in-situ clean-up sorbent, in order to remove interferences and impurities in the soils that could contaminate the instrument. By using this method, samples from two sites have been analysed for PAHs content, specifically (i) soils from a contaminated former Tar Works site, and (ii) urban road dust from Newcastle upon Tyne, UK. It was found that particle size was a significant parameter in both cases, showing a higher concentration (from 9 to 1404 mg/kg in the Tar Works, and 0.5 to 95 mg/kg in the road dust site) in the smaller grain size (< 250 μm); this is important when considering the ingestion exposure pathway as smaller particles are more likely to be ingested by children via hand-to-mouth behaviour. In addition, the source of the PAHs was investigated in the anthropogenically contaminated areas; it was found that pyrogenic sources (higher molecular weight PAHs, 4-5-6 rings) of PAHs were significantly more abundant compared to petrogenic sources (lower molecular weight PAHs: 2-3 rings). Generally the lower molecular weight such as naphthalene, acenaphthene, acenaphthylene and fluorene were found in lower concentration than fluoranthene, pyrene and other higher molecular weight PAH. In the case of the Tar Works lower molecular weights PAH were showing individual PAH concentration below 50 mg/kg whereas higher molecular weights were showing individual concentrations up to 270 mg/kg. The same trend was observed in the road dust samples, and was clearly identified by using ratios of PAH concentration to demonstrate dominance of pyrogenic sources. In this latter case, the pyrogenic sources were clearly identified as vehicle exhaust. However, other sources were identified such as the road pavement and the tire debris as potential sources of PAHs in urban areas. In the former case the PAH distribution was attributed to the locations of the chemicals productions areas in the former industrial site. The mean daily oral intake was used as an estimate of the environmental health risk from the sites; values of PAH intake were determined based on the PAH individual concentration and compared against known values. Risk was often present for pyrogenic PAHs in road dust and soil samples. Further investigation of the environmental health risk was realized using a physiologically-based extraction test on soil samples from the former Tar works; the results, using a fed-version of the test, showed elevated bioaccessibilities of PAHs, mainly due to the presence of food and the lipophilicity of PAHs, however other PAH properties could influence their individual mobilizations such as the molecular weight, the ring number and the liquid-to-soil ratio. It was noticed that the risk can be evaluated differently and can show different conclusions depending on the risk assessment chosen. Overall, the determination of PAHs in environmental soil and urban dust samples has highlighted the necessity to assess the potential impact on human health of their presence. The use of the fed-version of the physiologically-based extraction test is one tool that could be used to assess the environmental health risk to humans. This tool was shown to be robust using an inter-laboratory study, as values for total PAH content and bioaccessible fractions were within the same acceptable range.

This thesis describes the development and application of a laboratory system to investigate the loss of peroxy radicals to aerosol. The laboratory system consisted of an aerosol flow tube coupled to a custom-built PEroxy Radical Chemical Amplifier (PERCA). Aerosol was generated using an atomiser and their distribution measured using a SMPS. New values of the HO2 uptake coefficient, γ , were obtained for wet and dry sodium chloride aerosol and dry ammonium sulphate aerosol. The mass accommodation coefficient, α, was also determined for NaCl. A box model was used to determine the implications of the new uptake coefficient of HO2 loss to NaCl aerosol for tropospheric chemistry in the marine boundary layer.

Engineered nanoparticles are defined as having a dimension that is between one and one hundred nanometres. With toxicology studies reporting various degrees of toxicity the need to investigate nanoparticle fate and behaviour is vital. Monodispersed engineered nanoparticles were synthesised in-house to produce suitable materials to examine such processes. Iron oxide nanoparticles (5 nm) and citrate coated silver nanoparticles (20 nm) were subjected to different conditions of pH, ionic strength and different types of commercially available natural organic matter. Changes in particle size and aggregation were examined using a multi-method approach. Results showed that the natural organic matter was able to absorb onto nanoparticle surfaces and improve their stability when subjected to changes in pH and ionic strength, where they would normally aggregate. The presence of higher concentrations of NOM in some cases promoted aggregation due to bridging. This work also concluded that silver nanoparticles could be produced in the presence of NOM without additional stabilisers and that they themselves were stable. This work has demonstrated that engineered nanoparticles could remain stable within a range of environmental conditions, and thus raise future pollution concerns.

Mathematical modelling approaches have been widely used to evaluate the capacity of an ecosystem to assimilate anthropogenic wastes. This is essential to develop sustainable management strategies and for the prevention of eutrophication. This project aimed to assess the importance of the benthic-pelagic interactions in Ria Formosa and to develop a simple biogeochemical model for the management of nutrient inputs. This was done by adapting the simple version of the CSTT model, for pelagic eutrophication, to the system and by adding a benthic primary producer, the microphytobenthos (MPB), which was previously indicated as one of the main components of the system. This research project has three main parts: 1) field work that provided context and data for model testing; 2) experimental work, which aimed at evaluating key parameters of the system to be used in the modelling approach; 3) model development work that was used to test hypotheses and provided a synthesis of ecological achievements. An initial assessment of the optimal methodology for MPB chlorophyll extraction was carried out to implement a strategy for an accurate chlorophyll measurement. The MPB temporal, spatial and vertical variability was investigated. The complex temporal pattern revealed a small influence of seasonality. However, phytoplankton was found to have a much more important seasonal component. The most important component of the MPB variability was found to be the small and large scale spatial variability, which explains around 61% of the total variance. MPB was also found to be the most important source of chlorophyll to the lagoon system, contributing around 99% of the total chlorophyll. The experimental approach carried out to investigate the yield of MPB chlorophyll from nutrients, which was previously considered to be one of the most important parameters of the CSTT model for phytoplankton, revealed interesting results. Estimates were found to be larger than the values used for phytoplankton. The development process of the biogeochemical dCSTT-MPB model allowed the investigation of the importance of several processes. Pore water nutrients were found to be essential to support the large MPB community. Moreover, MPB cells were also found to have a large impact on the pelagic chlorophyll concentrations by re-suspension. The model was able to predict concentrations in the appropriate range of values observed in the lagoon. However, the temporal pattern is still weakly simulated and improvements are still required.

The main objective of the work reported in this thesis was to investigate molecules of atmospheric importance and their reactions. This has been achieved using spectroscopic methods, notably photoelectron spectroscopy, u.v.-visible spectroscopy and infrared spectroscopy, as well as kinetics simulations, global modelling and quantum chemical calculations. Photoelectron spectroscopy (PES) was used to investigate the pyrolysis behaviour of two hydrofluorocarbon (HFC) fire suppression agents, pentafluoroethane (CF3CHF2) and 2-H heptafluoropropane (CF3CHFCF3) to determine their thermal decomposition pathways. The thermal decomposition of flowing pentafluoroethane (CF3CHF2) diluted in argon was studied over the temperature range 600-1600 ˚C. At lower temperatures, (< 1000 ˚C) there is evidence of production of the thermal decomposition products C2F4 and HF and at higher temperatures (over 1000 ˚C) the decomposition products CF2 and CF3H are observed. The adiabatic and vertical ionisation energy (AIE and VIE) have been measured for the first band of CF3CHF2 from the recorded photoelectron spectrum as (12.71 +- 0.05) eV and (13.76 +- 0.02) eV respectively. The main pathways for decomposition have been established over the temperature range investigated. Pyrolysis of 2-H heptafluoropropane (CF3CHFCF3), at low pressure, diluted in argon, has been studied over the temperature range 600-2000 ˚C. Comparison of the results obtained has been made with results of recent electronic structure calculations of possible decomposition pathways. The most favoured reaction thermodynamically, to produce CF3CF=CF2 + HF, is found to be the main decomposition reaction at lower temperatures, 600-900 ˚C. At higher temperatures, 900-1200 ˚C, decomposition gave C2F4 + CF3H and it was found that C3F6 decomposes to C2F4 + CF2, and C2F4 decomposes to CF2 at temperatures above 1400 ˚C. Three ozone-alkene reactions (O3 + ethene, O3 + 2-methylpropene (2MP) and O3 + 2, 3 dimethyl-2-butene (DMB)) were separately investigated, each at low pressure, using a flow-tube interfaced to a photoelectron spectrometer. Photoelectron spectra, recorded as a function of reaction time, have been used to estimate partial pressures of the reagents and products, using photoionisation cross-sections for selected photoelectron bands of the reagents and products, (which have been measured separately) for each reaction. The yields of all the main products have been determined for each reaction. For each reaction, oxygen was observed as a product for the first time and for the O3 + ethene reaction acetaldehyde was measured as a product for the first time. Kinetics simulations were performed using reaction schemes which were developed for these reactions in order to determine the main reactions for production of the observed products. A feasibility study was carried out on the first PE bands of four selected non-linear triatomic molecules, SF2, HO2, HOCl and Cl2O in order to optimise the experimental production conditions for a threshold PES study using synchrotron radiation. Reliable methods were developed to record threshold PE spectra of SF2, HOCl and Cl2O consistently for upwards of 4 hours. The work on HO2 is still ongoing and the method developed requires further optimisation. Absorption spectroscopy was used to determine the photolysis rate coefficient, in the atmosphere, of the atmospherically important molecule monochlorodimethyl sulphide (CH3SCH2Cl) from u.v. photoabsorption cross-section measurements in the gas-phase.

Uncertainty about the causes of glacial-interglacial CO2 variations demonstrates our incomplete grasp of fundamental processes that govern our climate and thus one of the foremost problems in palaeoceanography and Earth System Science regards the mechanism(s) responsible for natural changes in atmospheric CO2 concentration. It is becoming clear that the Southern Ocean overturning circulation plays an important role in the global carbon cycle because altered communication between the atmosphere and abyss in the Southern Ocean is relatively well documented and often implicated in explanations of past and future climate changes, but the ambiguity of the paleoceanographic record defies interpretation of the mechanisms involved. Using a coarse resolution ocean general circulation model and coupled biogeochemistry code, an ensemble of idealised perturbations to external forcing and internal physics of the Southern Ocean is examined to explain the processes that link ocean circulation, nutrient distributions and biological productivity, and determine the extent to which the Southern Ocean governs the partitioning of CO2. Strengthened or northward-shifted winds result in oceanic outgassing and increased atmospheric carbon dioxide levels, while weakened or southward-shifted winds cause oceanic carbon uptake and reduced atmospheric carbon dioxide concentration. Driven by the work done on the ocean by the winds, changes in the rate or spatial pattern of the Southern Ocean residual overturning circulation lead to alteration of upper ocean stratification and the rate and depth from which carbon and nutrient-rich deep waters are upwelled to the surface. These surface waters, imprinted with the pattern of air-sea gas exchange, are subducted to intermediate depths in the ocean interior, not the abyss as previous suggested. These results are robust to significant alterations to surface heat and freshwater boundary conditions, mesoscale eddy activity and rates of air-sea gas exchange and represent a significant proportion of the change in glacial-interglacial CO2 that can be currently generated by altered ocean circulation in a variety of models, revealing that the upper limb of the Southern Ocean overturning circulation is important in determining atmospheric CO2 levels.

The toxicity and ubiquitousness of PAHs within different terrestrial environments has been an increasing cause for concern amongst environmental scientists in the last decades, in particular regarding their transport within the soil. In an attempt to understand the role of pure inorganic phases in PAH-mobility; experiments exposing mineral soil components with low organic matter content to a PAH-representative were carried out. The systems consisted of four different mineral phases (quartz sand, hematite, iron coated quartz sand and montmorillonite) which were individually exposed to naphthalene in electrolyte solutions prepared at increasing ionic strengths (NaNO3: 0.001 M; 0.01 M; 0.1 M) and pH (4.0 and 5.5). All experiments were conducted over at 24 reaction intervals and at ambient temperature conditions.Mineral geosorbents are traditionally known to be poor PAH-scavengers; in particular when compared to organic, high surface area materials such as activated carbons. On this basis, a preliminary validation experiment (Proof of Concept Experiment) was conducted to test the sensitivity of the selected extraction method (SPME) under complete uptake (activated carbon) and very low uptake (quartz sand) conditions. By extracting and analysing the supernatant after 24 hr of exposure of both sorbents to naphthalene under identical conditions it was concluded that SPME was a feasible extraction technique, yielding good reproducibility (n=3, inter-day RSD%= 11.18% ) even at very low PAH concentrations (0.2 µg / L). The final concentration of naphthalene in the sample supernatant after 24 hours was determined by GC-FID. All samples were extracted using the Solid Phase Microextraction method developed during the Proof of Concept which allowed the rapid extraction of naphthalene in the headspace HS-SPME (extraction time = 3 minutes) using temperature control and ultrasonication as means of agitation. Each sample set included triplicates of blanks and samples as well as calibration standards (in duplicate where possible)Out of the four minerals, only quartz sand and hematite showed a slight tendency towards naphthalene removal from solution; a finding which correlated well with increasing ionic strength. The other two minerals did not show any such trend and the results were deemed inconclusive. In regards to the results for quartz and hematite; the detected uptake was found to be below the sensitivity of the current SPME extraction method according to the error analysis carried out by comparing the sample and blank means whilst accounting for error equal to 1σ. The overlapping of both means in the majority of the samples indicated that both averages were too close to be accurately resolved (due to very low naphthalene uptake). Modifications to the SPME method could improve the reproducibility and decrease the spread of the data; however, this measure would only guarantee higher statistical confidence (95 %) and not higher naphthalene uptake by these minerals. These observations lead to the conclusion that naphthalene was being salted out of solution rather than being removed by sorption; and under these experimental conditions it would not have been possible to detect any real PAH-mineral interaction. In view of this outcome, a different approach was attempted in order to detect surface reactions between the minerals and naphthalene. A series of preliminary (qualitative) surface analysis (AFM, XPS and ATR-FTIR) on pre-loaded mineral specimens were carried out in air at ambient temperature conditions. No naphthalene was positively identified on the surfaces of the studied sorbents. Factors such as molecular size, sorbents characteristics (i.e. roughness, surface charge) and loading conditions impeded the detection of the target molecules. Innovative sample preparation protocols as well as controlled analytical conditions would need to be implemented and evaluated before this kind of analytical tool can be used. The main outcome of this research work was the successful adaptation of SPME to the rapid extraction of naphthalene in electrolyte solutions at optimal and sub-optimal concentration levels; as the proof of concept preliminary experiment showed.

Aerosol effects are one of the major uncertainties in assessing global climate change, ecosystem processes and human health. This is because they critically change the balance between the radiation entering and leaving the atmosphere, as well as influencing cloud formation and having direct effects on biological systems e.g. through the respiratory system. It is the direct radiative effects of aerosol that are the focus of this work. The Aerosol Optical Depth (AOD) is a measure of the extinction of radiation by aerosol throughout the depth of the atmosphere. It is wavelength dependent and is traditionally measured at a number of visible wavelengths, but there is little AOD data available at UV wavelengths especially in the UVB. The Brewer spectrophotometer makes direct sun measurements in the UV spectral range, which can in principle be used to calculate AOD at those wavelengths using a form of Beer's law. This work explores the capabilities of the Brewer for UV AOD measurements and applies the results to data from the Tropics and temperate mid-latitudes. Instrument specific weighting functions were tested for their ability to improve the AOD retrieval, but while they changed the partitioning of absorption between ozone and SO2, they had negligible effect on the resulting AOD. After correcting the existing Brewer software for AOD retrieval, data from Manchester UK were compared with independent measurements of AOD, and measurements from the Manchester instrument were also evaluated against a standard Brewer on location in Spain. The inter Brewer comparisons were consistent with differences of the order 6-10%, while comparison with independent methods was qualitatively consistent, but absolute differences were of the order 10-30%. This might partially be attributed to wavelength mismatches between the different methods, and assumptions in the various methods of calculating the AOD.Following the validation exercise the AOD was retrieved from Brewer instruments, using standard weighting functions, in Manchester from 2000- 2008 and Reading from 2003-2008. Based on this work in the UK, the method was then applied to data from Malaysia (1998-2007), where the climate is totally different to that of the UK. The AOD data obtained from the UK and from the Tropics have been compared. Kuala Lumpur, Malaysia gave the highest average AOD values probably due to it being a developing city with high pollution levels indicating a human impact on climate change. Brewer AOD measurements obtained at a narrow range of wavelengths were then used to calculate Angstrom parameters by applying Volz Method. The results, often generating a negative alpha, were deemed unreliable at the UK sites. This was largely attributed to the high solar zenith angle and low signal to noise of the direct sun measurement, exacerbated by a limited number of clear sky measurements available for the work. However, calculation of Angstrom parameters was more successful in Malaysia due to a low solar zenith angle, high intensity, and greater frequency of truly could free sky. Results indicated that aerosol particles in the capital Kuala Lumpur show a clear domination by fine mode. This domination is probably caused by urban pollution, mainly from road traffic, industrial and anthropogenic activities, which is consistent with a large capital city undergoing rapid development.Thus the capabilities and limitations of the Brewer spectrophotometer to provide both AOD data at UV wavelengths, and from those further aerosol properties, has been tested and demonstrated in two contrasting climatic regions. Air mass, limiting the UV signal, and stray-light within the instrument are two of the factors that limit the success of the Brewer for these measurements, which proved more reliable when the sun was high in the sky as in the Tropics.