This thesis describes a fractionation column method of separating isobutane and isobutene to isolate isooctane.

Proceeding from the aim to reduce global pollution emissions from the continuous burning of hydrocarbons stimulated by increasing energy demand, more efficient and ultra-low emissions’ combustion concepts such as the homogenous charge compression ignition engines (HCCI) have been developed. These new engines rely on the low temperature chemistry (LTC) combustion concept. A detailed investigation of the properties of cool flames, governed by LTC, is essential for the design of these new engines. The primary goal of this work was to build a fundamental counterflow experiment for cool flames studies in a diffusive system, to better understand combustion in LTC engines. The project was intended to provide a basic understanding of the low-temperature reactivity and cool flames properties of butane isomers under atmospheric pressure conditions. This was achieved by establishing self-sustaining cool flames through a novel technique of ozone addition to an oxygen stream in a non-premixed counterflow model. The ignition and extinction limits of butane isomers’ cool flames have been investigated under a variety of strain rates. Results revealed that establishment of cool flames are favored at lower strain rates. Iso-butane was less reactive than n-butane by showing higher ignition and extinction limits. Ozone addition showed a significant influence on cool flame ignition and sustenance; it was found that increasing ozone concentration in the oxidizer stream dramatically increased the reactivity of both fuels. Results showed increased fuel reactivity as the temperature of the fuel stream outlet increased. 4 A numerical analysis was performed to simulate ignition and extinction of the cool flame in diffusive systems. The results revealed that ignition and extinction limits of cool flames are predominantly governed by LTC. The model qualitatively captured experimental trends for both fuels; however, it overpredicted both ignition and extinction limits under all strain rates and ozone concentrations. The discrepancies were within a factor of eight for the ignition limit and a factor of two for the extinction limit. Finally, sensitivity analyses were conducted to understand the reactions responsible for cool flames ignition. It was found that majority of the sensitive reactions are those that occur at low temperatures.

Nitrogen, isobutane, isobutylene and 2,2,3,3-tetramethylbutane are the only products found in the thermal decomposition of 2,2'-azoisobutane. They are the results of the direct free radical dimerization and disproportionation. Ratio of the rate constants of disproportionation to dimerization was found to be 1.9 at 268°C and 4.5% azobutane in helium. This compared with Blackham and Eatough's value of 1.4 at 243°C and 3.6 at 290°C with 4% azobutane in helium. The ratio of the disproportionation to dimerization increased with an increase in temperature. A change in contact time affected the ratio at lower reaction temperatures but had little effect on it at higher temperatures. The contact times used were between three and seven seconds. Research would be useful at lower contact times.

Aromatic sulfonation and some substitution reactions of chlorobenzene show deviations in the selectivity relation. A determination of chlorobenzene sulfonation provided the following rate ratio relative to benzene and partial rate factors: 1.0, [---] of 0.064 , mf 1.7 and pf 4. 2. Our work, Olah's nitration, Ferguson's bromination and Stock's chlorination define a new relation that can be express by adding an entropy term (A) to the selectivity relation. We determined the rate of bromine-cyclohexene addition in dichloro-methane at 0° to provide a basis for a systematic variation of solvents at low temperature. An unusual zero-order reaction was encountered. The reaction was light catalyzed and related to hydrogen bromide concentration. The reaction was independent of bromine, cyclohexene, hydroperoxide and oxygen concentration. The reaction order depended on a steady-state concentration of bromonium ion (Br+). We also observed a novel photo effect with mixtures of oxygen, nitrogen and air. The synthesis of Thorpe's caged triacid, 4-methyltricyclo[1.1.0.02-4] butane-1,2,3-tricarboxylic acid has never been repeated or verified. A key intermediate, 1,1,1-ethanetriacetic,was readily converted to 1,1,1-ethanetriacetyl bromide. This last compound could only be a-brominated in the presence of a trace of ether. This successful a-bromination represents an important breakthrough toward getting Thorpe's caged acid.

La mesure de la formation de HO2 et H2O2 lors de l’oxydation de carburant est très difficile. Par contre, elle est extrêmement importante pour déterminer l’importance relative des voies de terminaison de chaînes de R + O2 et de ramification des chaînes menant à la production des radicaux OH. Par ailleurs, ces informations sont essentielles pour améliorer les modèles cinétiques. Afin de répondre à cette demande, un nouveau dispositif expérimental a été développé dans notre laboratoire : un réacteur auto-agité par jets gazeux couplé à un détecteur cw-CRDS qui permet d’analyser en ligne des produits de combustion. Grâce à ce nouveau système, pour la première fois, HO2 a été mesuré directement lors de l’oxydation du n-butane et de l’éther di-méthylique dans un réacteur auto-agité par jets gazeux. L’échantillonnage est toujours à basse pression et les produits sont détectés dans le proche infrarouge. Toutes nos expériences ont été réalisées à pression atmosphérique dans le domaine de température 500-900 K. Les produits de combustion mesurés sont CH2O, H2O2, HO2, C2H4 et H2O. Nos résultats expérimentaux ont été utilisés pour tester des modèles cinétiques issus de la littérature que nous avons analysés en menant des analyses de sensibilité et de voies réactionnelles. / Measuring the formation of HO2 and H2O2 from the oxidation of fuels is challenging but extremely important for determining their tendency to follow chain-termination pathways from R+O2 compared to chain-branching leading to the production of OH radicals. Furthermore, such data are vital for improving existing detailed chemical kinetics models. In order to meet these requirements, a new experimental setup has been developed in our laboratory: a jet-stirred reactor coupled with the cw-CRDS, which allows analyzing online combustion products. Thanks to this new system, for the first time, HO2 was measured directly during the oxidation of n-butane and dimethylether in a jet-stirred reactor. The sampling is always in vacuum and the species were detected at near infrared. All of our experiments were carried out at atmospheric pressure and in the range of temperature between 500 K to 900 K. The combustion products measured were CH2O, H2O2, HO2, C2H4, and H2O. Our experimental results were used to test published kinetic models that were analyzed by performing sensitivity and reaction paths analyzes.

Depuis quelques années, le domaine de l’extraction végétale est en pleine mutation, avec à la fois un intérêt croissant des consommateurs pour des ingrédients d’origine naturelle, combiné à des préoccupations environnementales. Il apparaît dès lors indispensable de remplacer les procédés actuels utilisant des solvants pétrochimiques nocifs par de nouveaux procédés d’extraction réduisant le besoin énergétique, la toxicité du solvant et la quantité de déchets tout en s’assurant du rendement et de la qualité de l’extrait obtenu. L’objectif de cette thèse a donc consisté à développer un nouveau procédé d’extraction mettant en œuvre des gaz liquéfiés comme solvants. Ce manuscrit présentera tout d’abord l’état de l’art sur les gaz liquéfiés existants et leur mise en œuvre pour l’extraction des produits naturels. A l’issue de cette présentation, 3 gaz liquéfiés ont été sélectionnés (n-butane, HFO-1234ze et le DME) comme solvants pour mener des essais au laboratoire, grâce à un prototype dont la conception est détaillée dans le chapitre II. Dans un second temps, les essais réalisés à l’aide de ces gaz liquéfiés pour l’extraction de composés lipophiles ont été décrits. L’approche expérimentale a été couplée à une approche prédictive par l’utilisation d’outils d’aide à la décision : les paramètres de solubilité de Hansen et le modèle COSMO-RS. La prédiction théorique ainsi que les essais expérimentaux ont confirmé l’intérêt des gaz liquéfiés pour la solubilisation et l’extraction de composés lipophiles d’intérêt biologique et à haute valeur ajoutée. Parallèlement l’étude des impacts du procédé sur l’environnement, la qualité, la réglementation et la sécurité ont montré que l’extraction par gaz liquéfié était un procédé facilement transposable à l’échelle industrielle. / In recent years, the industrial sector of vegetable extraction has been evolving due to the growing interest of consumers for natural food ingredients combined with growing environmental concerns. Therefore, it seems essential to replace existing processes using toxic petroleum bases solvents with greener extraction processes with lower energy consumption, less wastes but higher extract quality. The objective of this thesis has consisted in the research and development of a new extraction process using liquefied gases as liquid solvents. First, this manuscript will outline the state of the art on the liquefied gases and how they are used for extraction of natural products. After this survey, 3 liquefied gases (n-butane, HFO-1234ze and DME) have been selected for laboratory scale experiments performed using a dedicated extraction unit whose design is detailed in chapter II. Then, the tests performed with these 3 gases for extraction of lipophilic compounds from several plant materials has been described. The experimental approach has been combined with a predictive one using decision tools: Hansen Solubility Parameters and COSMO-RS model. This survey demonstrated that liquefied gases are interesting solvents for solubilization and extraction of lipophilic compounds of interest. In parallel, the impacts of the process on environment, safety regulation and quality showed that liquefied gas extraction could be easily transposed at industrial scale.

Olefins have enjoyed many uses in a wide variety of industries, from car manufacturing to energy production. Energy consuming processes of catalytic dehydrogenation, turning paraffins into olefins, has been commercialised since the early 20th century, while catalytic oxydehydrogenation of paraffins to olefins is still in prototype stages. The conflict between kinetic and thermodynamic yield constraints, has delayed the commercialisation of this process. The solution to achieving the relevant process route is exploitation of the right catalyst at moderate temperatures and pressures. Co5MgAlO is studied under atmospheric pressure and 350°C temperature, to dehydrogenate propane and butane to olefins using oxygen as a reactant. Thermodynamic models showing how many reaction routes are possible under atmospheric pressure were explored. Experimental results for butane to air at ratio of 1:0.8 and 1:1.2 hydrocarbons to air gave better selectivity of 1-butene which was more than 12%. When compared with propane at similar reaction ratios the reaction favoured CO2 at selectivity of more than 95%. / Civil and Chemical Engineering / M.Tech. (Chemical Engineering)

No description available.

The availability of kinetic data for the oxidative dehydrogenation (ODH) of n-butane from Téllez et al. (1999a and 1999b) and Assabumrungrat et al. (2002) presented an opportunity to submit a chemical process of industrial significance to Attainable Region (AR) analysis. The process thermodynamics for the ODH of n-butane and 1-butene have been reviewed. The addition of oxygen in less than the stoichiometric ratios was found to be essential to prevent deep oxidation of hydrocarbon products {Milne et al. (2004 and 2006c)}. The AR concept has been used to determine the maximum product yields from the ODH of n-butane and 1-butene under two control régimes, one where the partial pressure of oxygen along the length of the reactor was maintained at a constant level and the second where the oxygen partial pressure was allowed to wane. Theoretical maxima under the first régime were associated with very large and impractical residence times. The Recursive Convex Control policy {Seodigeng (2006)} and the second régime were applied to confirm these maxima {Milne et al. (2008)}. Lower and more practical residence times ensued. A differential side-stream reactor was the preferred reactor configuration as was postulated by Feinberg (2000a). Abstract A.D. Milne Page 4 of 430 The maximum yield of hydrocarbon product, the associated residence time and the required reactor configuration as functions of oxygen partial pressure were investigated for the series combinations of an inert porous membrane reactor and a fixed-bed reactor. The range of oxygen partial pressures was from 85 kPa to 0.25 kPa. The geometric profile for hydrocarbon reactant and product influences the residence times for the series reactors. The concept of a residence time ratio is introduced to identify the operating circumstances under which it becomes advantageous to select an inert membrane reactor in preference to a continuously stirred tank reactor and vice versa from the perspective of minimising the overall residence time for a reaction {Milne et al. (2006b)}. A two-dimensional graphical analytical technique is advocated to examine and balance the interplay between feed conditions, required product yields and residence times in the design of a reactor {Milne et al. (2006a)}.. A simple graphical technique is demonstrated to identify the point in a reaction at which the selectivity of the feed relative to a product is a maximum {Milne et al. (2006a)}. Literature Cited Assabumrungrat, S. Rienchalanusarn, T. Praserthdam, P. and Goto, S. (2002) Theoretical study of the application of porous membrane reactor to Abstract A.D. Milne Page 5 of 430 oxidative dehydrogenation of n-butane, Chemical Engineering Journal, vol. 85, pp. 69-79. Feinberg, M. (2000a) Optimal reactor design from a geometric viewpoint – Part II. Critical side stream reactors, Chemical Engineering Science, vol. 55, pp. 2455-2479. Milne, D., Glasser, D., Hildebrandt, D., Hausberger, B., (2004), Application of the Attainable Region Concept to the Oxidative Dehydrogenation of 1- Butene in Inert Porous Membrane Reactors, Industrial and. Engineering Chemistry Research, vol. 43, pp. 1827-1831 with corrections subsequently published in Industrial and Engineering Chemistry Research, vol. 43, p. 7208. Milne, D., Glasser, D., Hildebrandt, D., Hausberger, B., (2006a), Graphical Technique for Assessing a Reactor’s Characteristics, Chemical Engineering Progress, vol. 102, no. 3, pp. 46-51. Milne, D., Glasser, D., Hildebrandt, D., Hausberger, B., (2006b), Reactor Selection : Plug Flow or Continuously Stirred Tank?, Chemical Engineering Progress. vol. 102, no. 4, pp. 34-37. Milne, D., Glasser, D., Hildebrandt, D., Hausberger, B., (2006c), The Oxidative Dehydrogenation of n-Butane in a Fixed Bed Reactor and in an Inert Porous Membrane Reactor - Maximising the Production of Butenes and Butadiene, Industrial and Engineering Chemistry Research vol. 45, pp. 2661-2671. Abstract A.D. Milne Page 6 of 430 Milne, D., Seodigeng, T., Glasser, D., Hildebrandt, D., Hausberger, B., (2008), The Application of the Recursive Convex Control (RCC) policy to the Oxidative Dehydrogenation of n-Butane and 1-Butene, Industrial and Engineering Chemistry Research, (submitted for publication). Seodigeng, T.G. (2006), Numerical Formulations for Attainable Region Analysis, Ph.D. thesis, University of the Witwatersrand, Johannesburg, South Africa. Téllez, C. Menéndez, M. Santamaría, J. (1999a) Kinetic study of the oxidative dehydrogenation of butane on V/MgO catalysts, Journal of Catalysis, vol. 183, pp. 210-221. Téllez, C. Menéndez, M. Santamaría, J. (1999b) Simulation of an inert membrane reactor for the oxidative dehydrogenation of butane, Chemical Engineering Science, vol. 54, pp. 2917-2925. __________________________________

No presente trabalho são apresentados resultados referentes ao primeiro coeficiente Townsend de ionização (α) no isobutano puro, para a faixa de campo elétrico reduzido (E/N) de 145 até194 Td. A configuração do aparato experimental consiste de uma configuração semelhante a uma RPC, com o anodo constituído por um vidro de elevada resistividade (2 x 1012 Ωcm) e um catodo metálico, ligado diretamente a um eletrômetro, onde fotoelétrons são produzidos pela incidência de um feixe de laser pulsado. O coeficiente α é determinado por meio da medição da corrente elétrica em regime de ionização primária e em regime de avalanche. Uma vez que, para o isobutano puro não há valores experimentais disponíveis na literatura, para a faixa de E/N analisada por este trabalho, os valores obtidos foram comparados com os resultados da simulação Magboltz 2. Os estudos incluíram a determinação do coeficiente α para diferentes taxas de repetição e intensidades do feixe de laser. Como a relação entre a carga rápida e a total relaciona-se com o primeiro coeficiente de Townsend, estudos relativos à contribuição iônica e eletrônica para a corrente média também foram realizados. Como existem poucos resultados disponíveis na literatura referentes às secções de choque de colisão e parâmetros de transporte para o isobutano é comum considerar os resultados de seu isômero estrutural: o n-butano. Assim, a fim de realizar uma análise comparativa, o coeficiente α foi determinado também para o n-butano. / In the present work, results concerning the first Townsend ionization coefficient (α) in pure isobutane within the density-normalized electric field (E/N) range of 145 to 194 Td are presented. The experimental setup consists of RPC-like cell with the anode made of a high resistivity glass (2 x 1012 Ωcm) and a metallic cathode, directly connected to an electrometer, on which photoelectrons are produced by the incidence of a pulsed laser beam. The α coefficient is determined by measuring the current under primary ionization and avalanche regime. Since for the E/N range covered by this work, there are no experimental values for pure isobutane available in the literature, the obtained values were compared with Magboltz 2 results. Our studies included the determination of α coefficient for different repetition rates and laser beam intensity. The ratio of the fast charge to the total charge is related to the first Townsend coefficient, so studies concerning the ionic and the electronic contribution to the average current were also performed. Since there are few results available in the literature for isobutane, concerning collisional cross section and electron transport parameters, is common to consider results from its structural isomer: n-butane. Thus, in order to perform a comparative analysis, the coefficient α was also determined for n-butane.