Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / In this study, an analytical method was developed for the quantification of statins Atorvastatin and Simvastatin, in hospital effluent (University Hospital of Santa Maria, using Solid phase extraction and Dispersive liquid-liquid microextraction, followed in liquid chromatography coupled to diode array detector. The principal variables of Dispersive liquid-liquid microextraction were investigated with aid of experimental central composite design. The recovery rates of Solid phase extraction varied from 95% for Atorvastatin and 83.2% for Simvastatin, in aqueous solution, and 92.0% for Atorvastatin and 81.5% for Simvastatin, in hospital effluent. As for the Dispersive liquid-liquid microextraction the recovery rates ranged from 85.5% for Atorvastatin and 89.4% for Simvastatin, in aqueous solution, and 83.0% for Atorvastatin and 83.1% for Simvastatin, in hospital effluent. The sampling was carried out in two sample points called Emergency Effluent and General Effluent . The concentration rates detected in the effluent were 30.1 μg L-1 in Emergency Effluent and 29.0 μg L-1 in General Effluent for Atorvastatin and 37.1 μg L-1 in Emergency Effluent and 36.1 μg L-1 in GE for Simvastatin. Ozonation and electrocoagulation were applied for the degradation/removal of statins. The principal variables of electrocoagulation were investigated with aid of experimental central composite design. The degradation rate obtained in ozonation was 100% for Atorvastatin, in 25 minutes reaction for aqueous solution and hospital effluent; and 100% for Simvastatin, in 4 minutes reaction for aqueous solution and 10 minutes reaction in hospital effluent. As for the electrocoagulation the removal rate was 75.6% in aqueous solution and 70.9% in hospital effluent for Atorvastatin; and 100% in aqueous solution and hospital effluent for Simvastatin in 60 minutes reaction. A preliminary risk evaluation revealed that the statins showed a low environmental risk, PEC/PNEC < 1. / Neste trabalho foi desenvolvida metodologia analítica para a quantificação de estatinas Atorvastatina e Sinvastatina, em efluente hospitalar (Hospital Universitário de Santa Maria), utilizando Extração em fase sólida e Microextração dispersiva líquido-líquido seguida de cromatografia líquida com detector de arranjo de diodos. As principais variáveis da Microextração dispersiva líquido-líquido foram investigadas utilizando a Metodologia de Superfície de Resposta através de um delineamento composto central rotacional. As taxas de recuperação da Extração em fase sólida foram de 95,8% para Atorvastatina e 83,5% para Sinvastatina, em solução aquosa e 92,0% para Atorvastatina e 81,5% para Sinvastatina em efluente hospitalar. Já para a Microextração dispersiva líquido-líquido as taxas de recuperação foram de 85,5% para Atorvastatina e 89,4% para Sinvastatina, em solução aquosa e 83,0% para Atorvastatina e 83,1% para Sinvastatina, em efluente hospitalar. A amostragem foi feita em dois pontos de coleta, denominados Efluente PA e Efluente Geral . A concentração média detectada no efluente foi de 30,1 μg L-1 no Efluente PA e 29,0 μg L-1 no Efluente Geral para Atorvastatina e 37,1 μg L-1 no Efluente PA e 36,1 μg L-1 no Efluente Geral para Sinvastatina. Ozonização e eletrocoagulação foram aplicados na degradação/remoção das estatinas. As principais variáveis da eletrocoagulação foram investigadas utilizando a Metodologia de Superfície de Resposta através de um delineamento composto central. As taxas de degradação obtidas na ozonização foram de 100% para Atorvastatina, em 25 min de reação, para solução aquosa e efluente hospitalar; e 100% para Sinvastatina em 4 min em solução aquosa, e 10 min em efluente hospitalar. Já para a eletrocoagulação as taxas de remoção foram de 75,9% em solução aquosa e 70,9% em efluente hospitalar para Atorvastatina; e 100% para solução aquosa e afluente hospitalar para Sinvastatina, em 60 min de reação. A análise preliminar de risco ambiental revelou que as estatinas apresentam baixo risco ambiental, com PEC/PNEC < 1.

A lamotrigina (LTG) é um fármaco pertencente à classe das feniltriazinas utilizado no tratamento de crises epilépticas generalizadas e focais e no tratamento adjunto da epilepsia refratária. Devido à alta variabilidade interindividual, às interações medicamentosas e aos efeitos adversos apresentados durante a administração da LTG, a monitorização terapêutica nos pacientes que fazem uso deste fármaco é necessária para ajuste de dose individual e evitar os efeitos adversos. Assim, o objetivo deste trabalho foi a avaliação de duas técnicas de microextração: a microextração em fase líquida com fibras ocas (HF-LPME) e a microextração líquido-líquido dispersiva (DLLME) para análise da lamotrigina em amostras de plasma de pacientes epilépticos. Primeiramente foram definidas as condições eletroforéticas: foi utilizado um capilar de sílica fundida de 75 ?m de diâmetro interno e 50 cm de comprimento efetivo. O eletrólito de corrida (BGE) foi composto por ácido 2-morfolinoetanosulfônico (MES), na concentração de 130 mmol L-1 e pH 5,0. As análises foram realizadas à temperatura de 20°C e tensão de 15 kV. A amostra foi injetada hidrodinamicamente (0,5 psi por 10 s) e a detecção foi feita em 214 nm. Nestas condições a LTG e o padrão interno (PI), lidocaína, puderam ser analisados em menos de 7 minutos. A HF-LPME foi avaliada no modo de 3 fases, usando 500 ?L de plasma e 3,5 mL de solução fosfato de sódio 50 mmol L-1 pH 9,0 como fase doadora. O solvente utilizado para impregnar a fibra foi o 1-octanol. Como fase aceptora foram utilizados 60 ?L de solução de ácido clorídrico pH 4,0. Para avaliação da DLLME, foi necessária uma etapa de pré-tratamento da amostra (500 ?L de plasma) com 1 mL de acetonitrila. Após isto, 1,3 mL do sobrenadante foram adicionados a 4 mL de solução fosfato de sódio 50 mmol L-1 pH 9,0 e 120 ?L de clorofórmio (solvente extrator) foram injetados nesta amostra aquosa e 165 ?L de fase sedimentada foram recuperados. As características de desempenho analítico para ambos os métodos foram avaliadas, sendo obtida linearidade na faixa de concentração plasmática de 1-20 ?g/mL e limite inferior de quantificação (LIQ) de 1 ?g mL-1. Os ensaios de precisão e exatidão apresentaram valores de acordo com os guias oficiais. Além disso, os métodos foram seletivos, não apresentaram efeito residual e as amostras foram estáveis. Os valores de recuperação foram de 54,3 e 23% para HF-LPME e DLLME, respectivamente. Os métodos validados foram aplicados com sucesso em amostras de plasma de pacientes epilépticos em tratamento com a LTG. Além disso, as duas técnicas foram comparadas e a HF-LPME apresentou vantagens em relação à DLLME, mostrando ser uma técnica promissora para análise de matrizes complexas, com reduzido consumo de solvente orgânico e possibilidade de automação. / Lamotrigine (LTG) is an antiepileptic drug, which belongs to the class of phenyltriazine that can be used in the treatment of new-onset and refractory epilepsy. Due to its high interindividual variability, drug interactions and the adverse effects presented during the LTG administration, therapeutic drug monitoring is very important to dose adjustment and to avoid toxicity effects. Thus, the goal of this study was to develop and validate two microextraction techniques: the hollow fiber liquid-phase microextraction (HF-LPME) and the dispersive liquid-liquid microextraction (DLLME) to analyze LTG in plasma samples of epileptic patients. First of all, the eletroforetic conditions were optimized. A fused-silica uncoated capillary with 75 ?m internal diameter, and 50 cm effective length was used. The 2-(N-morpholino)ethanesulfonic acid (MES) 130 mmol L-1 pH 5.0 was chosen as background electrolyte (BGE). The temperature and the voltage were kept constant at 20°C and 15 kV respectively. For sample injection, hydrodynamic injection mode was used, with a pressure of 0.5 psi applied for 10 s. The wavelength was set at 214 nm. Under final conditions, LTG and the internal standard (IS) lidocaine were analyzed in less than 7 minutes. HF-LPME was evaluated in the three phase mode. The analyte was extracted from 4.0 mL of a basic donor phase (composed of 500 ?L of plasma and 3.5 mL of sodium phosphate solution 50 mmol L-1 pH 9.0) into an organic phase composed of 1-octanol immobilized in the pores of the hollow fiber, and further into an acidic acceptor phase (hydrochloric acid solution pH 4.0) placed in the lumen of the fiber. To evaluate DLLME, the plasma samples were pretreated to remove the proteins, and 500 ?L of plasma sample was mixed with 1 mL of acetonitrile. After that, 1,3 mL of the upper layer was added to 4 mL of sodium phosphate solution 50 mmol L-1 pH 9.0, and 120 ?L of chloroform (extracting solvent) was rapidly injected in the aqueous sample and 165 ?L of the sedimented phase was collected. Under the optimized conditions, both methods were linear over the plasmatic concentration range of 1.0-20.0 ?g mL-1 and the lower limit of quantification (LLOQ) was 1.0 ?g mL-1. Both methods showed good precision, accuracy, selectivity to LTG, with no carryover and the samples were stable under the studied conditions. The recovery were 54,3 and 23% to HF-LPME and DLLME respectively. The validated methods were successfully applied for the quantification of LTG in plasma samples of epileptic patients. The techniques were compared and HF-LPME was more advantageous for being more suitable to analysis of complex matrices using small amount of organic solvent, and also can be automated.

Orientador: Prof. Dr. Hugo Barbosa Suffredini / Tese (doutorado) - Universidade Federal do ABC, Programa de Pós-Graduação em Ciência e Tecnologia/Química, Santo André, 2018. / Este trabalho descreve a realizacao de estudos eletroquimicos fundamentais em interfaces do tipo liquido/liquido. Primeiramente, pretendeu-se compreender as diferencas na oxidacao de uma molecula modelo (ferroceno) por meio de estudos eletroquimicos relacionados a cinetica e termodinamica em diferentes situacoes. A primeira situacao enfoca a oxidacao direta do ferroceno em um eletrolito de suporte convencional aquoso e a segunda e relacionada ao ferroceno sendo adicionado diretamente em oleo mineral que quando em contato com o eletrolito de suporte aquoso, forma uma interface do tipo oleo/agua. Experimentos adicionais tambem foram conduzidos em uma emulsao, que foi constituida por ferroceno em oleo misturados com eletrolitos convencionais e cloreto de tetrabutilamonio (CTBA). Uma placa de diamante dopado com boro (DDB) foi usada como eletrodo de trabalho. Observou-se um deslocamento de aproximadamente 0,07 V quando a oxidacao do ferroceno foi conduzida na interface oleo/agua em comparacao com a oxidacao direta na fase aquosa convencional. Esta mudanca de potencial tambem foi calculada usando a Teoria do Funcional da Densidade (DFT) como ferramenta teorica. Uma comparacao adicional foi realizada em relacao a interface e as emulsoes, constatando que a forca ionica e um fator determinante nesses tipos de estudos. Finalmente, as curvas de Tafel indicaram que as coordenadas da reacao do ferroceno sao diferentes quando o eletrodo e colocado na interface versus diretamente na fase aquosa, sendo que a energia de ativacao obtida para o ferroceno na agua foi de 129 kJ mol L-1 e na interface igual a 53,0 kJ mol L-1. Posteriormente, mostrou-se o uso da microextracao liquido-liquido dispersiva (DLLME) combinada com a eletroquimica de interface oleo/agua, utilizando ferroceno como molecula- modelo. Uma mistura de 150 ¿ÊL de etanol (solvente dispersante) e 50 ¿ÊL de 1-undecanol (solvente de extracao) foi injetada em 9 mL de uma solucao aquosa de ferroceno preparada em H2SO 4 1,0 x10 -3 mol L -1. Uma gota da fase organica contendo ferroceno foi inserida diretamente na superficie de um eletrodo de carbono impresso, e uma curva analitica foi obtida, ponto a ponto, com boa linearidade. O eletrodo de DDB foi utilizado para promover o calculo do coeficiente de particao, em que foi possivel calcular as concentracoes finais de ferroceno nas fases aquosa e organica como sendo iguais a 4,2 ~ 10 -4 mol L-1 e 0,15 mol L -1, respectivamente. Ao usar estes valores, o coeficiente de particao (log P) de 2,5 foi calculado. Por fim, a eletroquimica de interface oleo/agua foi aplicada para estudar a influencia do acido naftenico na corrosao do aco API-5L-X70 a temperatura ambiente a partir de um sistema bifasico constituido por petroleo/solucao aquosa. Um estudo in-situ baseou-se na espectroscopia de impedancia eletroquimica (EIS), enquanto medidas localizadas de impedancia eletroquimica (LEIS) e microscopia eletroquimica de varredura (SECM) foram utilizadas para estudar o efeito da pre-exposicao do aco ao sistema bifasico. Os resultados de LEIS e SECM mostraram que a pre-exposicao ao petroleo com acido naftenico produz uma superficie mais ativa do que apenas a solucao de sulfato. A adicao de acido naftenico ao oleo provoca corrosao. A morfologia da corrosao e caracterizada por tres regioes: i) na fase de oleo; ii) na fase aquosa e iii) em torno da interface. A corrosao foi mais evidente em torno da interface. Propoe-se um mecanismo no qual as moleculas de acidos naftenicos se auto-organizam na interface oleo/agua, gerando acidificacao local no contato trifasico entre oleo / fase aquosa / metal. / This work describes the realization of fundamental electrochemical studies at a liquid / liquid interface. Firstly, it was intended to understand the differences in the oxidation of a model molecule (ferrocene) through electrochemical studies related to kinetics and thermodynamics in different situations. The first situation focuses the direct oxidation of ferrocene in an aqueous conventional support electrolyte and the second one is when ferrocene is directly placed in mineral oil, subsequently placed in contact with an aqueous support electrolyte thereby forming an oil/water interface. Additional experiments were also conducted in an emulsion, which was constituted by ferrocene in oil and mixed with conventional electrolytes and tetrabutylammonium chloride (TBAC). A Boron-Doped Diamond Electrode (BDD) was used as the working electrode. We observed an approximately 0.07 V shift when the ferrocene oxidation was conducted at the oil/water interface compared with the direct oxidation in the conventional aqueous phase. This potential shift was also calculated using the Density Functional Theory (DFT) as a theoretical tool. An additional comparison was conducted related to the interface and emulsions, finding that the ionic strength is determining factor in these types of studies. Finally, Tafel plots indicated that reaction coordinates of the ferrocene are different when the electrode is placed at the interface versus directly in the aqueous phase, in which the activation energy obtained for ferrocene in water was 129 kJ mol L-1 and at the interface equal to 53.0 kJ mol L-1. Posteriorly, we showed the use of dispersive liquid-liquid microextraction (DLLME) combined with organic/water interface electrochemistry by using ferrocene as a model-molecule. A mixture of 150 ¿ÊL of ethanol (disperser solvent) and 50 ¿ÊL of 1-undecanol (extractant solvent) was injected into 9 mL of a ferrocene aqueous solution prepared in 1.0~10.3 mol L.1 H2SO4. Afterward, a drop of the organic phase containing ferrocene was directly inserted on a screen-printed carbon electrode and the analytical curve was obtained, point-to-point, with a good linearity. Besides that, a BDD electrode was used to promote the partition coefficient calculation, and it was possible to calculate the final concentrations of ferrocene in the aqueous and organic phases as being equal to 4.2~10.4 mol L.1 and 0.15 mol L.1, respectively. By using these values, the partition coefficient (log P) was calculated as 2.5. Finally, the oil/water interface was applied to study the role of naphthenic acid in the API.5L.X70 steel corrosion at room temperature from a biphasic system consisting of petroleum/aqueous solution. An in-situ study was based on electrochemical impedance spectroscopy, while localized measures of electrochemical impedance (LEIS) and scanning electrochemical microscopy (SECM) were used for the effect of pre-exposure from steel to biphasic system. The results LEIS and SECM showed that the pre-exposure to oil with acid produces a more active surface than just sulfate medium. Corrosion morphology is characterized by three regions: i) in oil phase; ii) in the aqueous phase, and iii) around the interface. Naphthenic acid corrosion was restricted to the area around the interface. A mechanism is proposed in which naphthenic acids self-assemble at the oil/water interface, generating local acidification at the oil/aqueous phase/metal three-phase contact.

Submitted by MICHAEL PÉREZ RODRÍGUEZ null (michaelpr@iq.unesp.br) on 2016-09-16T20:43:06Z No. of bitstreams: 1 TESE MICHAEL VERSÃO FINAL.pdf: 4495815 bytes, checksum: 5d4a4f8ffcd143bb08ad336b289fe7ab (MD5) / Approved for entry into archive by Ana Paula Grisoto (grisotoana@reitoria.unesp.br) on 2016-09-22T14:44:30Z (GMT) No. of bitstreams: 1 perezrodriguez_m_dr_araiq_par.pdf: 1340231 bytes, checksum: 9e639833f42ff1e6fc87e7f9f965f8fc (MD5) / Made available in DSpace on 2016-09-22T14:44:30Z (GMT). No. of bitstreams: 1 perezrodriguez_m_dr_araiq_par.pdf: 1340231 bytes, checksum: 9e639833f42ff1e6fc87e7f9f965f8fc (MD5) Previous issue date: 2016-08-26 / Pró-Reitoria de Pós-Graduação (PROPG UNESP) / Este trabalho descreve o desenvolvimento de métodos analíticos ambientalmente mais amigáveis utilizando sistemas de screening e procedimentos de análise por injeção em fluxo para a detecção e quantificação de antibióticos, da classe das tetraciclinas, em amostras de medicamentos veterinários e alimentos de origem animal como leite bovino e suplementos de ovos. Os métodos desenvolvidos foram baseados na reação de acoplamento diazo obtida entre as tetraciclinas e o ácido p-sulfanílico diazotado em meio básico, resultando na formação de azo compostos com máximos de absorção em torno de 435 nm, permitindo determinar esses antibióticos por espectrofotometria. As condições analíticas foram otimizadas através de análise multivariada utilizando planejamentos experimentais fatoriais e o desempenho dos métodos propostos foi avaliado através de parâmetros como linearidade, efeito matriz, precisão, exatidão, limite de detecção, limite de quantificação, recuperação e estudo de interferências. O primeiro trabalho desenvolvido fez uso de um procedimento de injeção em fluxo contínuo para a rápida determinação de tetraciclinas em formulações veterinárias comerciais, visando maior eficiência no controle de qualidade na indústria farmacêutica. Na sequência, foi desenvolvido um método de screening mediante análise por injeção em fluxo empregando uma cela capilar de longo caminho óptico (LWCC) para a detecção de resíduos de tetraciclinas em amostras de leite bovino, sem a necessidade de passos de extração dispendiosos e demorados ou complicados tratamentos de amostras. Finalmente, um novo método de microextração líquido-líquido dispersiva assistida por ultrassom (US-DLLME) foi desenvolvido para a simples, rápida e eficiente extração de resíduos de tetraciclinas a partir de amostras de suplemento de ovos e, posterior análise por injeção em fluxo baseada em uma cela capilar de longo caminho óptico (LWCC), empregando um banho de aquecimento com controle de temperatura. Os métodos descritos apresentaram importantes características como simplicidade operacional, possibilidade de automação, rapidez e baixo custo das análises. Além disso, estes métodos seguiram os princípios da química verde, de modo que permitiram realizar determinações ambientalmente mais limpas, através da redução considerável do consumo de reagentes e amostras, evitando ou minimizando o emprego de solventes tóxicos. Também, os métodos desenvolvidos para a análise de resíduos mostraram excelente desempenho para aplicação no controle de qualidade dos produtos alimentares estudados na indústria alimentícia, a fim de proteger os consumidores. / This work describes the development of environmentally friendly analytical methods using screening systems and flow injection procedures for detection and quantification of tetracycline antibiotics in veterinary pharmaceuticals and animal-derived foods such as bovine milk and egg-based protein supplements. The developed methods were based on the diazo coupling reaction between the tetracyclines and diazotized p-sulfanilic acid in basic medium, resulting in the formation of azo compounds that present maximum absorption around 435 nm, which allows determining these antibiotics by spectrophotometry. The analytical conditions were optimized by means of multivariate analysis using factorial experimental designs. Performance of the proposed methods was assessed through parameters such as linearity, selectivity, precision, accuracy, detection limit, quantification limit, matrix effect and recovery. The first developed work used a continuous flow injection procedure for the rapid determination of tetracyclines in commercial veterinary formulations, aiming greater efficiency in quality control in the pharmaceutical industry. Thereafter, it was developed a screening method by flow injection analysis using a liquid waveguide capillary cell (LWCC) for detecting tetracyclines residues in bovine milk samples, without the need for expensive extraction steps that are time-consuming or complicated sample treatments. Finally, a new ultrasound-assisted dispersive liquid-liquid microextraction (US-DLLME) method was developed for simple, rapid, and efficient extraction of tetracyclines residues from egg supplement samples, followed by flow injection analysis based on a liquid waveguide capillary cell (LWCC), using a controlled temperature heating bath. The described methods presented important features such as operational simplicity, possibility of automation, quickness and low cost of the analysis. Furthermore, these methods followed the principles of green chemistry, making possible to perform environmentally cleaner determinations, due to the considerable reduction of reagents and samples consumption, avoiding or minimizing the use of toxic solvents. Besides, the methods developed for residues analysis showed excellent performance for application in the food industry for quality control of the studied food products in order to protect consumers.

The distribution and potential sources of 15 polycyclic aromatic hydrocarbons (PAHs) in soils and Digitaria eriantha in the vicinity of three South African coal-fired power plants, Matla, Lethabo and Rooiwal were determined by gas chromatography–mass spectrometry. An ultrasonic assisted dispersive liquid-liquid microextraction (UA-DLLME) method was developed for the extraction of polycyclic aromatic hydrocarbon in soil, followed by determination using gas chromatography mass spectrometry. The study showed that an extraction protocol based on acetonitrile as dispersive solvent and C2H2Cl2 as extracting solvent, gave extraction efficiencies comparable to conventional soxhlet extraction for soil samples. The extraction time using ultrasonication and the volume of the extraction solvent was also investigated. Using a certified reference material soil (CRM), the extraction efficiency of UA-DLLME ranged from 64 to 86% in comparison with the Soxhlet result of 73 to 95%. In comparison with the real sample, the CRM result did not show a significant difference at 95% C.I. The UA-DLLME proved to be a convenient, rapid, cost-effective and greener sample preparation approach for the determination of PAHs in soil samples. PAH compound ratios such as phenanthrene/phenanthrene + anthracene (Phen/ Phen + Anth) were used to provide a reliable estimation of emission sources. The total PAH concentration in the soils around three power plants ranged from 9.73 to 61.24 μg g−1, a range above the Agency for Toxic Substances and Disease Registry levels of 1.0 μg g−1 for a significantly contaminated site. Calculated values of the Phen/Phen + Anth ratio were 0.48±0.08, 0.44±0.05, and 0.38+0.04 for Matla, Lethabo and Rooiwal, respectively. The flouranthene/fluoranthene + pyrene (Flan/ Flan + Pyr) levels were found to be 0.49±0.03 for Matla, 0.44±0.05 for Lethabo, and 0.53±0.08 for Rooiwal. Such values indicate a xx pyrolytic source of PAHs. Higher molecular weight PAHs (five to six rings) were predominant, suggesting coal combustion sources. The carcinogenic potency B[a]P equivalent concentration (B[a] Peq) at the three power plants ranged from 3.61 to 25.25, indicating a high carcinogenic burden. The highest (B[a] Peq) was found in samples collected around Matla power station. It can, therefore, be concluded that the soils were contaminated with PAHs originating from coal-fired power stations. Nine metals (Fe, Cu, Mn, Ni, Cd, Pb, Hg, Cr and Zn) were analysed in soil and the Digitaria eriantha plant around three coal power plants (Matla, Lethabo and Rooiwal), using ICP-OES and GFAAS. The total metal concentration in soil ranged from 0.05 ± 0.02 to 1835.70 ± 70 μg g-1, 0.08 ± 0.05 to 1743.90 ± 29 μg g-1 and 0.07 ± 0.04 to 1735.20 ± 91 μg g-1 at Matla, Lethabo and Rooiwal respectively. The total metal concentration in the plant (Digitaria eriantha) ranged from 0.005 ± 0.003 to 534.87 ± 43 μg g-1 at Matla, 0.002 ± 0.001 to 400.49 ± 269 μg g-1 at Lethabo and 0.002 ± 0.001 to 426.91 ± 201 μg g-1 at Rooiwal. The accumulation factor (A) of less than 1 (i.e. 0.003 to 0.37) at power plants indicates a low transfer of metal from soil to plant (excluder). The enrichment factor values obtained (2.4 – 5) indicate that the soils are moderately enriched, with the exception of Pb that had significant enrichment of 20. The Geo-accumulation Index values of metals indicate that the soils are moderately polluted (0.005 – 0.65), except for Pb that showed moderate to strong pollution (1.74 – 2.53). / Chemistry / D. Phil. (Chemistry)

The distribution and potential sources of 15 polycyclic aromatic hydrocarbons (PAHs) in soils and Digitaria eriantha in the vicinity of three South African coal-fired power plants, Matla, Lethabo and Rooiwal were determined by gas chromatography–mass spectrometry. An ultrasonic assisted dispersive liquid-liquid microextraction (UA-DLLME) method was developed for the extraction of polycyclic aromatic hydrocarbon in soil, followed by determination using gas chromatography mass spectrometry. The study showed that an extraction protocol based on acetonitrile as dispersive solvent and C2H2Cl2 as extracting solvent, gave extraction efficiencies comparable to conventional soxhlet extraction for soil samples. The extraction time using ultrasonication and the volume of the extraction solvent was also investigated. Using a certified reference material soil (CRM), the extraction efficiency of UA-DLLME ranged from 64 to 86% in comparison with the Soxhlet result of 73 to 95%. In comparison with the real sample, the CRM result did not show a significant difference at 95% C.I. The UA-DLLME proved to be a convenient, rapid, cost-effective and greener sample preparation approach for the determination of PAHs in soil samples. PAH compound ratios such as phenanthrene/phenanthrene + anthracene (Phen/ Phen + Anth) were used to provide a reliable estimation of emission sources. The total PAH concentration in the soils around three power plants ranged from 9.73 to 61.24 μg g−1, a range above the Agency for Toxic Substances and Disease Registry levels of 1.0 μg g−1 for a significantly contaminated site. Calculated values of the Phen/Phen + Anth ratio were 0.48±0.08, 0.44±0.05, and 0.38+0.04 for Matla, Lethabo and Rooiwal, respectively. The flouranthene/fluoranthene + pyrene (Flan/ Flan + Pyr) levels were found to be 0.49±0.03 for Matla, 0.44±0.05 for Lethabo, and 0.53±0.08 for Rooiwal. Such values indicate a xx pyrolytic source of PAHs. Higher molecular weight PAHs (five to six rings) were predominant, suggesting coal combustion sources. The carcinogenic potency B[a]P equivalent concentration (B[a] Peq) at the three power plants ranged from 3.61 to 25.25, indicating a high carcinogenic burden. The highest (B[a] Peq) was found in samples collected around Matla power station. It can, therefore, be concluded that the soils were contaminated with PAHs originating from coal-fired power stations. Nine metals (Fe, Cu, Mn, Ni, Cd, Pb, Hg, Cr and Zn) were analysed in soil and the Digitaria eriantha plant around three coal power plants (Matla, Lethabo and Rooiwal), using ICP-OES and GFAAS. The total metal concentration in soil ranged from 0.05 ± 0.02 to 1835.70 ± 70 μg g-1, 0.08 ± 0.05 to 1743.90 ± 29 μg g-1 and 0.07 ± 0.04 to 1735.20 ± 91 μg g-1 at Matla, Lethabo and Rooiwal respectively. The total metal concentration in the plant (Digitaria eriantha) ranged from 0.005 ± 0.003 to 534.87 ± 43 μg g-1 at Matla, 0.002 ± 0.001 to 400.49 ± 269 μg g-1 at Lethabo and 0.002 ± 0.001 to 426.91 ± 201 μg g-1 at Rooiwal. The accumulation factor (A) of less than 1 (i.e. 0.003 to 0.37) at power plants indicates a low transfer of metal from soil to plant (excluder). The enrichment factor values obtained (2.4 – 5) indicate that the soils are moderately enriched, with the exception of Pb that had significant enrichment of 20. The Geo-accumulation Index values of metals indicate that the soils are moderately polluted (0.005 – 0.65), except for Pb that showed moderate to strong pollution (1.74 – 2.53). / Chemistry / D. Phil. (Chemistry)

Koffein (1,3,7-Trimethylxanthin) und Coprostanol (5beta-cholestan-3beta-ol) wurden im Berliner Oberflächenwasser nachgewiesen. Ihre Konzentrationen korrelierten mit dem Verunreinigungsgrad der Proben, was nahelegt, dass sie sich als Marker für menschliche Aktivität eignen. Bemerkenswerterweise wurde Koffein in jeder einzelnen Oberflächenwasserprobe oberhalb der Bestimmungsgrenze von 0,025 µg/L gefunden. Um Oberflächenwasserproben in größeren Serien zu untersuchen, war die Entwicklung zweier neuer Methoden erforderlich: ein Immunoassay, basierend auf einem monoklonalen Antikörper für Koffein und eine dispersive flüssig-flüssig Mikroextraktionsmethode (DLLME), gefolgt von Flüssigkeitschromatographie gekoppelt mit Tandem-Massenspektrometrie (LC-MS/MS) für Coprostanol. Der entwickelte Koffein-Immunoassay zeigt die beste je erhaltene Nachweisgrenze für Koffein (0,001 µg/L), erlaubt Hochdurchsatz-Analysen und erfordert keine Probenvorbereitung. Der Assay wurde auch erfolgreich für die Messung von Koffein in Getränken, Haarwaschmitteln, Koffeintabletten und menschlichem Speichel angewendet. Antikörper gegen Coprostanol sind nicht kommerziell erhältlich. Eine neue Strategie Anti-Coprostanol-Antikörper zu generieren wurde erarbeitet, die eine analoge Verbindung – Isolithocholsäure (ILA) – als Hapten verwendet, mit der eine Gruppe von Mäusen immunisiert wurde. Ein polyklonales Anti-ILA-Serum wurde produziert, welches Coprostanol bindet, aber die niedrige Affinität erlaubte nicht den Aufbau eines Immunoassays, der die Messung von Umweltkonzentrationen des Anayten (im Bereich ng/L) zulässt. Spezifische Anti-ILA-Immunglobuline G wurden auch in den Faeces der Mäuse gefunden. Coprostanol wurde in den Wasserproben durch die Verwendung einer neuentwickelten LC-MS/MS-Methode unter APCI-Ionisation (atmospheric pressure chemical ionisation) gemessen. Konzentrationen oberhalb von 0,1 µg/L wurden nach Voranreicherung der Probe mittels DLLME bestimmt. / Caffeine (1,3,7-trimethylxanthine) and coprostanol (5beta-cholestan-3beta-ol) were detected in samples of Berlin’s surface water. Their concentrations correlated with the contamination status of the samples, suggesting their usefulness as markers of human activity. Remarkably, caffeine concentrations were always well above the limit of quantitation of 0.025 µg/L. In order to screen surface water samples in larger series, the development of two novel methods was required: a monoclonal antibody-based immunoassay for caffeine and a dispersive liquid-liquid microextraction (DLLME) method, followed by liquid chromatography tandem mass spectrometry (LC-MS/MS) for coprostanol. The caffeine immunoassay developed shows the best analytical limit of detection (LOD) obtained so far for caffeine (0.001 µg/L), allows high-throughput analysis, and does not require sample pre-treatment. The assay was also successfully employed to measure caffeine in beverages, shampoos, caffeine tab-lets, and human saliva. Antibodies to coprostanol are not commercially available. A new strategy to generate anti-coprostanol antibodies was elaborated using an analogous com-pound as hapten – isolithocholic acid (ILA) – and immunizing a group of mice. A polyclonal anti-ILA serum was produced, which binds coprostanol but the low affinity did not permit setting up an immunoassay to measure environmental concentrations of the analyte (in the range of ng/L). Specific anti-ILA immunoglobulin G were also found in the faeces of the immunized mice. Coprostanol was quantified in the water samples using a newly developed LC-MS/MS method using atmospheric pressure chemical ionisation (APCI). Concentrations above 0.1 µg/L were determined after sample preconcentration using DLLME. This extraction method also proved to be successful for enrichment of coprostanol-related compounds such as cholesterol, cholestanol, cholestanone, ergosterol, and stigmasterol.