This thesis reports the results of investigations carried out into the composition of the saturate and aromatic unresolved complex mixtures (UCMs) in crude oils. It is divided into two sections. Section A reports on studies of the saturate UCM and Section B reports on studies of the aromatic UCM. UCMs are mixtures of very large numbers of compounds of low individual abundances, hence structural information on individual components is difficult or impossible to obtain using conventional GC or GC-MS techniques. The investigations reported in this thesis used a combination of GC-MS techniques (Section A) and oxidations of UCMs followed by GC-MS characterisation of the oxidation products (Section B) to develop a more detailed picture of the structures of components of the saturate and aromatic UCMs. UCMs are present in all crude oils, and may account for the vast majority of the material present in heavily weathered or biodegraded oils. An understanding of the types of compounds present may have a bearing on the refining processes an oil is subjected to, as well as assessing its potential environmental and toxicological effects.Single branched C(subscript)18 isomers were prepared to establish the chromatographic behaviour and mass spectral fragmentation patterns of open chain compounds with ethyl-, propyl-, butyl- and pentyl- substituents. All open chain structural isomers with a single n-alkyl branch larger than methyl in the range C(subscript)10 to C(subscript)20 were identified at each carbon number in a series of crude oils of varying ages, source types and depositional environments. Also, C(subscript)21 to C(subscript)25 structural isomers containing an ethyl branch were identified in all of these samples. This represents a total of 163 compounds. These monoalkylalkanes comprise approximately 3 % of the alkanes in these oils, with the n-alkanes (35-60 %) and ++ / methylalkanes (10 %) being the most abundant compound classes present. Isoprenoids, alkylcyclohexanes and other branched and/or cyclic alkanes make up the remainder of the material.Rock samples from a sedimentary sequence of Late Cretaceous age were analysed for ethylalkanes using GC-MS techniques. In the less mature samples, 3- and 5-ethylalkanes were in higher abundance relative to the other isomers at odd carbon numbers from C(subscript)17 to C(subscript)23. In the more mature samples, this odd preference was no longer apparent. Several other low maturity sediment samples were analysed and found to have a similar ethylalkane distribution to the shallow sample from the sedimentary sequence. A mature crude oil which was also analysed exhibited a distribution similar to the deeper sample from the sedimentary sequence. Tetralin pyrolysis of a low maturity lignite sample yielded only the 3- and 5-substituted ethylalkanes, providing supporting evidence that the initial odd preference is the result of defunctionalisation of specific natural product precursors containing a 3- or 5-ethylalkyl structural moiety. With increasing maturity this preference is diluted by input of ethylalkanes without preference. A mechanism involving acid-catalysed rearrangement of n-alkenes to form monoalkylalkanes is proposed to account for the loss of preference in ethylalkane relative abundances.Investigations into the composition of aromatic unresolved complex mixtures were conducted by oxidising the total aromatic fraction of a moderately biodegraded crude oil (biodegradation level 4) using potassium permanganate. This reagent cleaves the alkyl substituents attached to aromatic rings between the alpha and beta carbons and oxidises the alpha carbon to a carboxylate group. A biodegraded crude oil was chosen because the majority of the resolved components have been removed by biodegradation, ++ / leaving a complex mixture of compounds almost completely unresolved by gas chromatography. The oxidation product was separated into dichloromethane-soluble monocarboxylic acids (both aliphatic and aromatic) and water-soluble polycarboxylic acids (aromatic only). GC-MS analysis of these oxidation products gave the proportions of monosubstituted:disubstituted:trisubstituted:tetrasubstituted monoaromatic rings as 29:59:12:0.1, of which from one to three substituents were carboxylic acid groups with the remainder of the substituents being unoxidised methyl groups. Of the disubstituted monoaromatic oxidation products, 53 % were dicarboxylic acids with the most sterically hindered 1,2-substitution pattern. This observation was interpreted as evidence for the presence of significant amounts of naphthenoaromatic systems such as tetralins and indanes in the crude oil aromatic fraction. Analysis of the permanganate oxidation products also enabled a quantitative measure of the proportion of methyl substituents to be made. It was found that methyl groups accounted for a significant proportion of the alkyl substituents attached to aromatic systems. Of the disubstituted monoaromatic oxidation products, 59 % had a methyl group as one of the substituents, while of trisubstituted monoaromatic systems, 41% had one methyl and 37 % had two methyls. Compounds containing a biphenyl carbon skeleton comprised 3 % of the aromatic oxidation products, with isomers containing from one to four substituents of which one was a carboxylic acid group and the remainder were unoxidised methyls. This indicates that biphenyls containing more than one alkyl (>C(subscript)1) substituent were not present. Of the monosubstituted biphenylcarboxylic acids, the ratio of ortho:meta:para substituted isomers was 0:65:35, which correlates well with literature reports of the relative abundances of methylbiphenyl ++ / isomers, and suggests that the overall distribution of all monosubstituted biphenyls has not been significantly affected by biodegradation to level 4.The monoaromatic, diaromatic and triaromatic fractions of the same biodegraded crude oil (level 4) were separately treated with ruthenium tetroxide, which cleaves aromatic rings so that the ring carbon bearing the substituent is oxidised to become the carbonyl carbon of a carboxylic acid. These oxidation products represent the alkyl moieties that were attached to aromatic rings in the initial crude oil aromatic fractions. Identification of these alkyl side chains provides an insight into the nature of the components of the aromatic crude oil UCM. The oxidation products were separated into dichloromethane-soluble monocarboxylic acids, which were subsequently reduced to monodeuterated hydrocarbons for characterisation using gas chromatography-mass spectrometry (GC-MS) techniques, and water-soluble dicarboxylic acids which were analysed as dimethyl esters. n-Alkanes, methylalkanes, alkylalkanes, alkylcyclohexands, methylalkylcyclohexanes, isoprenoids and bicyclic alkanes were identified in the monodeuterated hydrocarbon samples derived from all three aromatic fractions. Most of these compounds had carbon skeletons strikingly similar to those observed in the saturate fractions of unbiodegraded crude oils, with the only differences being the addition of a carbon from the aromatic ring, and the presence of a deuterium atom attached to that carbon. Because the electron-withdrawing nature of carboxylic acid groups prevents further aromatic ring oxidation, numerous aromatic monocarboxylic acids were also identified in the acidic products of the oxidation of the crude oil diaromatic and triaromatic fractions. These included C(subscript)1 to C(subscript)3 alkyl-substituted benzoic acids derived from compounds containing a biphenyl ++ / or phenylnaphthalene structural moiety, as well as omega-phenylalkanoic acids with chain lengths up to C(subscript)11, derived from compounds in which two aromatic systems are connected by an alkyl chain. The main components of the dicarboxylic acid oxidation products of all three aromatic fractions were alpha, omega-dicarboxylic acids and alkylcyclopentane-dicarboxylic acids and alkylcyclohexane-dicarboxylic acids, with phthalic acids also present in the oxidation products of the diaromatic and triaromatic fractions. The observation that 1,5-pentanedicarboxylic acids and 1,6-hexanedicarboxylic acids were the only alpha, omega-dicarboxylic acids in the oxidation products of the crude oil monoaromatic fraction, and were present in high abundance relative to other alpha, omega-dicarboxylic acids in the oxidation products of the diaromatic and triaromatic fractions indicated that substituted indanes and/or tetralins were quantitatively important constituents of the overall crude oil aromatic fraction. This finding is supported by the results of the analysis of the mass spectra of the crude oil aromatic fractions.These studies of aromatic UCMs have provided new insights into the origins of the aromatic components of petroleum. The presence of a pronounced odd-over-even predominance in the C(subscript)25, C(subscript)27, and C(subscript)29 monodeuterated n-alkanes (CPI = 1.07), which corresponds to the odd-over-even predominance observed in the n-alkane components of unbiodegraded crude oils from the same basin, suggests that the n-alkyl side chains and the n-alkanes have a common source. Evidence is presented to support the hypothesis that the n-alkylaromatics are formed in part by geosynthetic processes involving alkylation of aromatic systems by electrophilic species such as carbocations and acylium ions formed from carboxylic acids. This hypothesis is then extended to ++ / explain the formation of other groups of compounds, including aromatic systems with isoprenoidal-, methylalkyl- and monoalkyl-branched side chains. Because isoprenoids, methylalkanes and alkylalkanes are well-known components of petroleum, these results suggest that these alkylaromatic components of petroleum may share a common source with the corresponding alkane components. It is suggested that aromatic unresolved complex mixtures arise due to the very large number of structurally related compounds present, which are formed by geosynthetic processes such as alkylation of aromatic rings.

yes / Heavy petroleum industries, including the fluid catalytic cracking (FCC) unit, are useful for producing fuels but they are among some of the biggest contributors to global greenhouse gas (GHG) emissions. The recent global push for mitigation efforts against climate change has resulted in increased legislation that affects the operations and future of these industries. In terms of the FCC unit, on the riser side, more legislation is pushing towards them switching from petroleum-driven energy sources to more renewable sources such as solar and wind, which threatens the profitability of the unit. On the regenerator side, there is more legislation aimed at reducing emissions of GHGs from such units. As a result, it is more important than ever to develop models that are accurate and reliable, that will help optimise the unit for maximisation of profits under new regulations and changing trends, and that predict emissions of various GHGs to keep up with new reporting guide-lines. This article, split over two parts, reviews traditional modelling methodologies used in modelling and simulation of the FCC unit. In Part I, hydrodynamics and kinetics of the riser are dis-cussed in terms of experimental data and modelling approaches. A brief review of the FCC feed is undertaken in terms of characterisations and cracking reaction chemistry, and how these factors have affected modelling approaches. A brief overview of how vaporisation and catalyst deactiva-tion are addressed in the FCC modelling literature is also undertaken. Modelling of constitutive parts that are important to the FCC riser unit such as gas-solid cyclones, disengaging and stripping vessels, is also considered. This review then identifies areas where current models for the riser can be improved for the future. In Part II, a similar review is presented for the FCC regenerator system.

Platinum complexes (cisplatin, carboplatin and oxaliplatin) are effective anticancer agents. However the major drawbacks of platinum chemotherapy are toxic side effects and resistance. The affinity of platinum complexes to sulfur donor ligands of side chains of methionine and cysteine amino acids was assumed to be responsible for toxicity and resistance. Recently, it was found that the reaction of platinum complex with proteins containing sulfur donor ligands could actually favor its anticancer activity. Copper transporter 1 (Ctr 1), a protein involved in the transport of copper into the cell, also helps in the influx of cisplatin by binding to N-terminal domain of Ctr 1 which is rich in methionine and histidine residues. A better understanding of how the size and shape of amine ligand, and leaving groups affect the reaction of platinum (II) complexes with methionine could give new ways to optimize its anticancer activity. This preliminary research focuses to answer this by HPLC-UV-VIS analysis of bulky platinum complexes including [Pt(dien)Cl]Cl, Pt(Me4en)(NO3)2 and Pt(en)(NO3)2 with two methionine containing small peptides that serve as models for protein interactions.

La RMN est un outil analytique extrêmement puissant pour l’analyse quantitative et structurale, et est très utilisée en biologie structurale. C’est dans ce contexte que nous avons choisi d’étudier les interactions par RMN. Le premier système choisi est un système de polymères, les PEO. Ils sont caractérisés par plusieurs grandeurs physiques dont l’indice de polydispersité. Cet indice représente la distribution de taille d’une population de polymères. Nous avons développé une nouvelle méthode de diffusion par RMN, la DOSY afin d’accéder à cet indice : J. Viéville et al. / Journal of Magnetic Resonance 212 (2011)169–173 Le deuxième système complexe étudié est un type de molécules chimiques mimétiques de l’ADN, les PNA, acides nucléiques peptidiques. Les PNA possèdent des bases nucléiques leur permettant de former des liaisons avec l’ADN ou l’ARN. Nous avons montré par des analyses de RMN que les PNA en solution peuvent former des complexes très stables avec des températures de fusion élevées. Ces analyses RMN sont complétées par des analyses de dichroïsme circulaire.Afin de compléter notre panel d’études sur les interactions moléculaires par RMN, nous avons étudié les interactions protéine-ligand avec greffage de la protéine sur phase solide : J.M.P. Viéville et al. / Journal of Pharmaceutical and Biomedical Analysis 89 (2014) 18–23 / NMR is a powerfull technique we decided to extend to follow interactions in complex mixtures. First part is about polydispersity index of polymer which is an important physical parameter when working with polymers. We developed here a new method, based on PEO analysis, using diffusion experiments (DOSY) by NMR to assess the polydispersity index: J. Viéville et al. / Journal of Magnetic Resonance 212 (2011) 169–173. In a second time, we worked on peptidic nucleic acids, PNA. These chemicals molecules are designed to bind DNA or RNA for clinical studies. We studied PNA in solution, by NMR to showwhich kind of interactions they form on themselves. We found very stable complexes with high fusion temperatures. Circular dichroism measurements were helpful for fusion temperature determination and structural studies. To complete this panel, we were interested in the study of protein-ligand interaction. We developed a new way to follow them, using a grafted protein on a solid phase based on HR-MAS NMR technology: J.M.P. Viéville et al. / Journal of Pharmaceutical and Biomedical Analysis 89 (2014) 18–23

Concern within the public health community is mounting regarding what some deem as "candy-flavored tobacco". A recent study by King et al. (2014) found that >40% of middle and high school student smokers use flavored cigarettes or flavored little cigars. This study investigated the validity of the "candy-flavored tobacco" designation by comparing flavor profiles of 18 flavored tobacco products with 15 candy and Kool-Aid products using gas chromatography mass spectrometry (GC/MS). Significant compositional overlap was found; nearly 75% of the compounds detected were present in at least one flavored tobacco and one candy or Kool-Aid product. Benzaldehyde and or benzyl alcohol were measured in cherry candies, Kool-Aid and tobacco. Similar levels of benzaldehyde were measured in the cherry Kool-Aid and wild cherry Cheyenne cigars at 3338 ± 623 and 3937 ± 251 μg/serving. Methyl anthranilate, 1-hexanol, [lowercase gamma]-decalactone, and raspberry ketone were found in all grape, apple, peach and berry products, respectively. Vanillin and or ethyl vanillin were constituents of all flavored tobacco products analyzed. Many flavorants, such as limonene, are also volatilizable biogenic organic compounds (VBOCs). A more comprehensive understanding of the identities and properties of VBOCs, precursors to secondary organic aerosol (SOA), would support air quality and climate change research and management. Our knowledge is limited by extreme compositional diversity within the VBOC class. Only recently have techniques such as two-dimensional gas chromatography time of flight mass spectrometry (GCxGC/TOFMS) been employed, and the complexity of the data poses analysis challenges. To address this, agglomerative hierarchical cluster analysis (HCA) was performed on data generated by GCxGC/TOFMS analysis of air samples collected during the Particle Investigations at a Northern Ozarks Tower: NOx, Oxidant, Isoprene Research (PINOT NOIR) study. HCA resulted in the assignment of 204 compounds into 27 clusters: these clusters were grouped into 4 distinct types making the data significantly more manageable. The assignment of a cluster to a type was mostly based on the frequency with which compounds appeared in samples. Type I clusters contained compounds that were present in only one sample, suggesting meteorological influence.

<p>The research described in this dissertation focuses on several areas: developing analytical methods to distinguish structural isomers, identifying the chemical compositions of aviation fuels and evaluating the effectiveness of organic dopants to swell aircraft o-ring seals. Chapter 2 discusses fundamental aspects of mass spectrometry, and ionization methods and the instrumentation used to complete this research. </p> <p>Chapter 3 discusses and compares two activation methods used to distinguish ionized structural isomers. Ionized naphthene-containing aromatic structural isomers were subjected to collision-activated dissociation (CAD) in an ion trap (ITCAD) and to medium-energy collision-activated dissociation (MCAD) in an octupole collision cell, both in the energy-resolved mass spectrometry mode (ERMS). MCAD was shown to be superior over ITCAD at the structural differentiation of the ionized isomers. </p> <p>Determination of the chemical compositions of petroleum-based jet and diesel fuels, potential alternative fuels and fuel blending components by using a GCxGC/(EI)TOF MS is discussed in chapter 4. The ability to determine the chemical compositions of fuels and to correlate the identified compounds and their concentrations to the physical and chemical properties and aircraft performance of the fuels is vital for the development of future resilient, alternative fuels. The chemical compositions of petroleum-based fuels were found to be different from potential alternative fuels.</p> <p>Chapter 5 discusses the effectiveness of aromatic and nonaromatic compounds in swelling air craft o-ring seals, which prevents leaks in the fuel circulation systems. The aim of this study was to identify aromatic and nonaromatic compounds that most effectively swell o-ring seals. Steric effects were shown to decrease the efficiency of the compounds to swell seals. Ethylbenzene and indane were found to swell o-ring seals more effectively than any other compounds studied, including a currently approved alternative fuel. </p>

Neste trabalho são apresentados dois modelos para o ensaio de destilação ASTM D-86, desenvolvidos no simulador EMSO e validados por dados experimentais de frações de petróleo disponíveis na literatura. Estes modelos apresentam índice algébrico-diferencial igual a um, diferentemente do primeiro modelo proposto na literatura que apresentava índice algébrico-diferencial igual a 2, o qual não possibilitava ser resolvido pelos simuladores atuais. As predições obtidas com os modelos propostos geraram resultados com desvio médio em torno de 0,6 % em relação aos dados experimentais da literatura. Em seguida, foi estudado o método proposto para determinação de mistura substituta de componentes reais e foram propostas algumas alterações. Esta metodologia pode ser dividida em duas etapas consecutivas. A primeira etapa determina um conjunto de componentes reais e a segunda calcula a composição da mistura substituta de componentes reais de modo a caracterizar misturas complexas de composição desconhecida, como frações de petróleo, substituindo o conceito de pseudo-componentes. Para isso, o algoritmo se baseia em dados experimentais tais como a curva PEV, o °API, a massa molar média da mistura e outras propriedades globais disponíveis. Foram obtidos resultados que se mostraram satisfatórios. Para finalizar este trabalho, foi proposto um novo método para determinação da mistura substituta que utiliza a primeira etapa do algoritmo de determinação de mistura substituta e modificado por este trabalho. Em seguida, é realizada uma estimação dinâmica de parâmetros de condição inicial utilizando um dos modelos desenvolvidos neste trabalho para o ensaio de destilação ASTM D-86. A estimação é baseada principalmente na curva experimental da destilação ASTM D-86 e na densidade à 60°F (ou °API). Os resultados desta nova metodologia foram comparados com outros métodos de determinação de mistura substituta e constatou-se uma melhora na aproximação das suas propriedades quando comparadas com a mistura original de composição desconhecida. / This work presents two models for the assay of distillation ASTM D-86, developed in the EMSO Simulator and validated using experimental data of petroleum fractions available in the literature. These models showed differential-algebraic index equal to one, differently from the first model proposed in the literature that showed differential-algebraic index equal to two. Problems with an index higher than one are usually impossible to solve by modern simulators. The predictions obtained with the proposed models generated results with average deviation of about 0,6% in relation to those in the literature. After this, a method proposed in the literature to determinate the substitute mixture of real compounds for petroleum fractions was studied and slightly modified. This methodology can be divided in the two following steps. The first one determines a group of real components and the second one calculates the composition of the substitute mixture of the real components in order to characterize a complex mixture of unknown composition, as petroleum fractions. This methodology aims at replace the pseudo-components concept. To make this, the algorithm is based in experimental data as the TBP curve, the °API, the average molar mass of the mixture and others global properties available. The obtained results showed to be suitable. Finally, this work proposes a new method to determine the substitute mixture that uses the first step of the algorithm of determination of substitute mixture and modified in this work. Next, it was used a dynamic estimation of parameters in initial conditions using one of the models proposed in this work for the distillation assay ASTM D-86. The estimation is mainly based on distillation experimental curve ASTM D-86 and the density at 60°F (or °API). The results of this new methodology were com pared to other methods of determination of substitute mixture and it was found an evolution in the similarity of its properties when compared to the original mixture with unknown composition.

Neste trabalho são apresentados dois modelos para o ensaio de destilação ASTM D-86, desenvolvidos no simulador EMSO e validados por dados experimentais de frações de petróleo disponíveis na literatura. Estes modelos apresentam índice algébrico-diferencial igual a um, diferentemente do primeiro modelo proposto na literatura que apresentava índice algébrico-diferencial igual a 2, o qual não possibilitava ser resolvido pelos simuladores atuais. As predições obtidas com os modelos propostos geraram resultados com desvio médio em torno de 0,6 % em relação aos dados experimentais da literatura. Em seguida, foi estudado o método proposto para determinação de mistura substituta de componentes reais e foram propostas algumas alterações. Esta metodologia pode ser dividida em duas etapas consecutivas. A primeira etapa determina um conjunto de componentes reais e a segunda calcula a composição da mistura substituta de componentes reais de modo a caracterizar misturas complexas de composição desconhecida, como frações de petróleo, substituindo o conceito de pseudo-componentes. Para isso, o algoritmo se baseia em dados experimentais tais como a curva PEV, o °API, a massa molar média da mistura e outras propriedades globais disponíveis. Foram obtidos resultados que se mostraram satisfatórios. Para finalizar este trabalho, foi proposto um novo método para determinação da mistura substituta que utiliza a primeira etapa do algoritmo de determinação de mistura substituta e modificado por este trabalho. Em seguida, é realizada uma estimação dinâmica de parâmetros de condição inicial utilizando um dos modelos desenvolvidos neste trabalho para o ensaio de destilação ASTM D-86. A estimação é baseada principalmente na curva experimental da destilação ASTM D-86 e na densidade à 60°F (ou °API). Os resultados desta nova metodologia foram comparados com outros métodos de determinação de mistura substituta e constatou-se uma melhora na aproximação das suas propriedades quando comparadas com a mistura original de composição desconhecida. / This work presents two models for the assay of distillation ASTM D-86, developed in the EMSO Simulator and validated using experimental data of petroleum fractions available in the literature. These models showed differential-algebraic index equal to one, differently from the first model proposed in the literature that showed differential-algebraic index equal to two. Problems with an index higher than one are usually impossible to solve by modern simulators. The predictions obtained with the proposed models generated results with average deviation of about 0,6% in relation to those in the literature. After this, a method proposed in the literature to determinate the substitute mixture of real compounds for petroleum fractions was studied and slightly modified. This methodology can be divided in the two following steps. The first one determines a group of real components and the second one calculates the composition of the substitute mixture of the real components in order to characterize a complex mixture of unknown composition, as petroleum fractions. This methodology aims at replace the pseudo-components concept. To make this, the algorithm is based in experimental data as the TBP curve, the °API, the average molar mass of the mixture and others global properties available. The obtained results showed to be suitable. Finally, this work proposes a new method to determine the substitute mixture that uses the first step of the algorithm of determination of substitute mixture and modified in this work. Next, it was used a dynamic estimation of parameters in initial conditions using one of the models proposed in this work for the distillation assay ASTM D-86. The estimation is mainly based on distillation experimental curve ASTM D-86 and the density at 60°F (or °API). The results of this new methodology were com pared to other methods of determination of substitute mixture and it was found an evolution in the similarity of its properties when compared to the original mixture with unknown composition.

Neste trabalho são apresentados dois modelos para o ensaio de destilação ASTM D-86, desenvolvidos no simulador EMSO e validados por dados experimentais de frações de petróleo disponíveis na literatura. Estes modelos apresentam índice algébrico-diferencial igual a um, diferentemente do primeiro modelo proposto na literatura que apresentava índice algébrico-diferencial igual a 2, o qual não possibilitava ser resolvido pelos simuladores atuais. As predições obtidas com os modelos propostos geraram resultados com desvio médio em torno de 0,6 % em relação aos dados experimentais da literatura. Em seguida, foi estudado o método proposto para determinação de mistura substituta de componentes reais e foram propostas algumas alterações. Esta metodologia pode ser dividida em duas etapas consecutivas. A primeira etapa determina um conjunto de componentes reais e a segunda calcula a composição da mistura substituta de componentes reais de modo a caracterizar misturas complexas de composição desconhecida, como frações de petróleo, substituindo o conceito de pseudo-componentes. Para isso, o algoritmo se baseia em dados experimentais tais como a curva PEV, o °API, a massa molar média da mistura e outras propriedades globais disponíveis. Foram obtidos resultados que se mostraram satisfatórios. Para finalizar este trabalho, foi proposto um novo método para determinação da mistura substituta que utiliza a primeira etapa do algoritmo de determinação de mistura substituta e modificado por este trabalho. Em seguida, é realizada uma estimação dinâmica de parâmetros de condição inicial utilizando um dos modelos desenvolvidos neste trabalho para o ensaio de destilação ASTM D-86. A estimação é baseada principalmente na curva experimental da destilação ASTM D-86 e na densidade à 60°F (ou °API). Os resultados desta nova metodologia foram comparados com outros métodos de determinação de mistura substituta e constatou-se uma melhora na aproximação das suas propriedades quando comparadas com a mistura original de composição desconhecida. / This work presents two models for the assay of distillation ASTM D-86, developed in the EMSO Simulator and validated using experimental data of petroleum fractions available in the literature. These models showed differential-algebraic index equal to one, differently from the first model proposed in the literature that showed differential-algebraic index equal to two. Problems with an index higher than one are usually impossible to solve by modern simulators. The predictions obtained with the proposed models generated results with average deviation of about 0,6% in relation to those in the literature. After this, a method proposed in the literature to determinate the substitute mixture of real compounds for petroleum fractions was studied and slightly modified. This methodology can be divided in the two following steps. The first one determines a group of real components and the second one calculates the composition of the substitute mixture of the real components in order to characterize a complex mixture of unknown composition, as petroleum fractions. This methodology aims at replace the pseudo-components concept. To make this, the algorithm is based in experimental data as the TBP curve, the °API, the average molar mass of the mixture and others global properties available. The obtained results showed to be suitable. Finally, this work proposes a new method to determine the substitute mixture that uses the first step of the algorithm of determination of substitute mixture and modified in this work. Next, it was used a dynamic estimation of parameters in initial conditions using one of the models proposed in this work for the distillation assay ASTM D-86. The estimation is mainly based on distillation experimental curve ASTM D-86 and the density at 60°F (or °API). The results of this new methodology were com pared to other methods of determination of substitute mixture and it was found an evolution in the similarity of its properties when compared to the original mixture with unknown composition.

L'objectif de cette thèse est d'explorer les applications potentielles d'un système de langue électronique basée sur des récepteurs combinatoires à réactivités croisées et l'imagerie par résonance de plasmons de surface, pour l'analyse et la discrimination de différents milieux complexes et de bactéries. L'étude de milieux complexes a été réalisée sur des échantillons de différentes natures comme le vin, la bière et le lait d'origines végétale et animale. Les expériences ont démontré que notre système de langue électronique est capable de répondre avec une bonne sélectivité à ces milieux complexes et qu'il génère ainsi des profils continus 2D et des images 3D, propres à chaque échantillon. La différentiation et la classification de ces divers types de boissons ont été réalisées grâce à ces signatures 2D et 3D. Le dispositif a également prouvé son efficacité pour le suivi du vieillissement du lait. Une seconde étude a été dédiée à l'application du système pour la détection de bactéries. Dans un premier temps, des paramètres fluidiques ont été optimisés comme la forme et la profondeur de la cuve ou le débit fluidique, en raison de la morphologie variable des bactéries, considérées ici comme des objets biologiques complexes et volumineux. Dans un second temps, le système s'est révélé performant pour l'analyse de bactéries et a montré la possibilité de quantifier ces analyses. En effet, la langue électronique a permis la discrimination de différentes bactéries selon leur genre, leur espèce et en fonction des souches grâce aux profils continus 2D et aux images 3D. / The objective of this PhD thesis is to explore the potential applications of the electronic tongue, based on combinatorial cross-reactive receptors and surface plasmon resonance imaging, for analysis and discrimination of different complex mixtures and bacteria. In this work, various complex mixtures of different nature such as wine, beer, and milk (either animal-based or plant-based) are used. It has been demonstrated that the electronic tongue is capable of responding differently to theses samples with good selectivity. For each of them, it can generate characteristic continuous 2D profile and 3D image, based on which the differentiation and classification of the complex mixtures have been carried out. Furthermore, it has been illustrated that the electronic tongue is efficient for monitoring the deterioration of milk. In the second part of this thesis, the electronic tongue has been applied for detection and analysis of bacteria. At first, some fluidic parameters have been optimized due to the variable morphology of these complexes and large biological objects. Under optimized experimental conditions, the electronic tongue is effective for analysis of bacteria with the possibility for quantification. Thereafter, the electronic tongue has allowed for the discrimination of different bacteria according to their genus, species and strains based on continuous 2D profiles and 3D images.