Complex reaction kinetics of the solid acid alkylation of isobutane with butenes over a proton-exchanged Y-zeolite has been modeled at the elementary step level. Starting with a computer algorithm that generated the reaction network based on the fundamentals of the carbenium ion chemistry, the formation of over 100+ product species has been modeled in order to gain understanding of the underlying phenomena leading to rapid catalyst deactivation and product selectivity shifts observed in experimental runs. An experimental investigation of the solid acid alkylation process was carried out in a fixed bed catalytic reactor operating with an excess of isobutane under isothermal conditions at moderate temperatures (353-393 K) in liquid phase. Experimental data varying with run-time for a set of butene space-times and reaction temperatures were collected for parameter estimation purposes. A kinetic model was formulated in terms of rate expressions at the elementary step level including a rigorous modeling of deactivation through site coverage. The single event concept was applied to each rate coefficient at the elementary step level to achieve a significant reduction in the number of model parameters. Based on the identification of structural changes leading to the creation or destruction of symmetry axes and chiral centers in an elementary step, formulae have been developed for the calculation of the number of single events. The Evans-Polanyi relationship and the concept of stabilization energy were introduced to account for energy levels in surface-bonded carbenium ions. A novel functional dependency of the stabilization energy with the nature of the carbenium ion and the carbon number was proposed to account for energy effects from the acid sites on the catalyst. Further reductions in the number of parameters and simplification of the equations for the transient pseudohomogeneous one-dimensional plug-flow model of the reactor were achieved by means of thermodynamic constraints. Altogether, the single event concept, the Evans-Polanyi relationship, the stabilization energy approach and the thermodynamic constraints led to a set of 14 parameters necessary for a complete description of solid acid alkylation at the elementary step level.

Catalytic combustion is an environmentally friendlytechnique to combust fuels in e.g. gas turbines. Introducing acatalyst into the combustion chamber of a gas turbine allowscombustion outside the normal flammability limits. Hence, theadiabatic flame temperature may be lowered below the thresholdtemperature for thermal NOXformation while maintaining a stable combustion.However, several challenges are connected to the application ofcatalytic combustion in gas turbines. The first part of thisthesis reviews the use of catalytic combustion in gas turbines.The influence of the fuel has been studied and compared overdifferent catalyst materials. The material section is divided into two parts. The firstconcerns bimetallic palladium catalysts. These catalysts showeda more stable activity compared to their pure palladiumcounterparts for methane combustion. This was verified both byusing an annular reactor at ambient pressure and a pilot-scalereactor at elevated pressures and flows closely resembling theones found in a gas turbine combustor. The second part concerns high-temperature materials, whichmay be used either as active or washcoat materials. A novelgroup of materials for catalysis, i.e. garnets, has beensynthesised and tested in combustion of methane, a low-heatingvalue gas and diesel fuel. The garnets showed some interestingabilities especially for combustion of low-heating value, LHV,gas. Two other materials were also studied, i.e. spinels andhexaaluminates, both showed very promising thermal stabilityand the substituted hexaaluminates also showed a good catalyticactivity. Finally, deactivation of the catalyst materials was studied.In this part the sulphur poisoning of palladium, platinum andthe above-mentioned complex metal oxides has been studied forcombustion of a LHV gas. Platinum and surprisingly the garnetwere least deactivated. Palladium was severely affected formethane combustion while the other washcoat materials were mostaffected for carbon monoxide and hydrogen. <b>Keywords:</b>catalytic combustion, catalyst materials,palladium, platinum, bimetallic, garnet, spinel, hexaaluminate,deactivation, sulphur, poisoning, diesel, methane,hydrocarbons

Voltage-gated ion channels play fundamental roles in neural excitability, they are for instance responsible for every single heart beat in our bodies, and dysfunctional channels cause disease that can be even lethal. Understanding how the voltage sensor of these channels function is critical for drug design of compounds targeting neuronal excitability. The opening and closing of the pore in voltage-gated potassium (Kv) channels is caused by the arginine-rich S4 helix of the voltage sensor domain (VSD) moving in response to an external potential. In fact, VSDs are remarkably efficient at turning membrane potential into conformational changes, which likely makes them the smallest existing biological engines. Exactly how this is accomplished is not yet fully known and an area of hot debate, especially due to the lack of structures of the resting and intermediate states along the activation pathway. In this thesis I study how the VSD activation works and show how toxic compounds modulate channel gating through direct interaction with these quite unexplored drug targets. First, I show that a secondary structure transition from alpha- to 3(10)-helix in the S4 helix is an important part of the gating as this helix type is significantly more favorable compared to the -helix in terms of a lower free energy barrier. Second, I present new models for intermediate states along the whole voltage sensor cycle from closed to open and suggest a new gating model for S4, where it moves as a sliding 3(10)-helix. Interestingly, this 3(10)-helix is formed in the region of the single most conserved residue in Kv channels, the phenylalanine F233. Located in the hydrophobic core, it directly faces S4 and creates a structural barrier for the gating charges. Substituting this residue alters the deactivation free energy barrier and can either facilitate the relaxation of the voltage sensor or increase the free energy barrier, depending on the size of the mutant. These results are confirmed by new experimental data that supports that a rigid ring at the phenylalanine position is the rate-limiting factor for the deactivation gating process, while the activation is unaffected. Finally, we study how the activation can be modulated for pharmaceutical reasons. Neurotoxins such as hanatoxin and stromatoxin push S3b towards S4 helix limiting S4's flexibility. This makes it harder for the VSD to activate and might explain the stronger binding affinities in resting state. All these results are highly important both for the general topic of biological macromolecules undergoing functionally critical conformational transitions, as well as the particular case of voltage-gated ion channels where understanding of the gating process is probably the key step to explain the effects of mutations or drug interactions. / <p>QC 20121115</p>

Complex reaction kinetics of the solid acid alkylation of isobutane with butenes over a proton-exchanged Y-zeolite has been modeled at the elementary step level. Starting with a computer algorithm that generated the reaction network based on the fundamentals of the carbenium ion chemistry, the formation of over 100+ product species has been modeled in order to gain understanding of the underlying phenomena leading to rapid catalyst deactivation and product selectivity shifts observed in experimental runs. An experimental investigation of the solid acid alkylation process was carried out in a fixed bed catalytic reactor operating with an excess of isobutane under isothermal conditions at moderate temperatures (353-393 K) in liquid phase. Experimental data varying with run-time for a set of butene space-times and reaction temperatures were collected for parameter estimation purposes. A kinetic model was formulated in terms of rate expressions at the elementary step level including a rigorous modeling of deactivation through site coverage. The single event concept was applied to each rate coefficient at the elementary step level to achieve a significant reduction in the number of model parameters. Based on the identification of structural changes leading to the creation or destruction of symmetry axes and chiral centers in an elementary step, formulae have been developed for the calculation of the number of single events. The Evans-Polanyi relationship and the concept of stabilization energy were introduced to account for energy levels in surface-bonded carbenium ions. A novel functional dependency of the stabilization energy with the nature of the carbenium ion and the carbon number was proposed to account for energy effects from the acid sites on the catalyst. Further reductions in the number of parameters and simplification of the equations for the transient pseudohomogeneous one-dimensional plug-flow model of the reactor were achieved by means of thermodynamic constraints. Altogether, the single event concept, the Evans-Polanyi relationship, the stabilization energy approach and the thermodynamic constraints led to a set of 14 parameters necessary for a complete description of solid acid alkylation at the elementary step level.

<p>Catalytic combustion is an environmentally friendlytechnique to combust fuels in e.g. gas turbines. Introducing acatalyst into the combustion chamber of a gas turbine allowscombustion outside the normal flammability limits. Hence, theadiabatic flame temperature may be lowered below the thresholdtemperature for thermal NO_Xformation while maintaining a stable combustion.However, several challenges are connected to the application ofcatalytic combustion in gas turbines. The first part of thisthesis reviews the use of catalytic combustion in gas turbines.The influence of the fuel has been studied and compared overdifferent catalyst materials.</p><p>The material section is divided into two parts. The firstconcerns bimetallic palladium catalysts. These catalysts showeda more stable activity compared to their pure palladiumcounterparts for methane combustion. This was verified both byusing an annular reactor at ambient pressure and a pilot-scalereactor at elevated pressures and flows closely resembling theones found in a gas turbine combustor.</p><p>The second part concerns high-temperature materials, whichmay be used either as active or washcoat materials. A novelgroup of materials for catalysis, i.e. garnets, has beensynthesised and tested in combustion of methane, a low-heatingvalue gas and diesel fuel. The garnets showed some interestingabilities especially for combustion of low-heating value, LHV,gas. Two other materials were also studied, i.e. spinels andhexaaluminates, both showed very promising thermal stabilityand the substituted hexaaluminates also showed a good catalyticactivity.</p><p>Finally, deactivation of the catalyst materials was studied.In this part the sulphur poisoning of palladium, platinum andthe above-mentioned complex metal oxides has been studied forcombustion of a LHV gas. Platinum and surprisingly the garnetwere least deactivated. Palladium was severely affected formethane combustion while the other washcoat materials were mostaffected for carbon monoxide and hydrogen.</p><p><b>Keywords:</b>catalytic combustion, catalyst materials,palladium, platinum, bimetallic, garnet, spinel, hexaaluminate,deactivation, sulphur, poisoning, diesel, methane,hydrocarbons</p>

CO hydrogenation and CO oxidation are two important processes addressing the energy and environmental issues of great interest. Both processes are carried out using metallic catalysts. The objective of this dissertation is to study the catalytic processes that govern these two reactions from a molecular perspective using quantum mechanical calculations. Density Functional Theory (DFT) has proven to be a valuable tool to study adsorption, dissociation, chain growth, reaction pathways etc., on well-defined surfaces. DFT was used to study the CO reduction reactions on promoted cobalt catalyst surfaces and CO oxidation mechanisms on cobalt surfaces. CO hydrogenation via Fischer-Tropsch Synthesis (FTS) is a process used to produce liquid fuels from synthesis gas. The economics of the Fischer-Tropsch process strongly depends on the performance of the catalyst used. The desired properties of a catalyst include selectivity towards middle distillate products such as diesel and jet fuel, higher activity and longer catalyst life. Catalysts are often modified by adding promoters to obtain these desirable properties. Promoters can influence the reaction pathways, reducibility, dispersion, activity and selectivity. In FTS, understanding the effect of promoters in the molecular scale would help in tailoring catalysts with higher activity and desired selectivity. Preventing deactivation of catalyst is important in FTS to increase the catalyst life. Deactivation of Co catalyst can occur by reoxidation, C deposition, sintering, formation of cobalt-support compounds etc. Designing catalyst with resistance to deactivation by the use of promoters is explored in this dissertation. The influence of promoters on the initiation pathways of CO hydrogenation is also explored as a first step towards determining the selectivity of promoted catalyst. The influence of Pt promoter on O removal from the surface of Co catalyst showed that Pt promoter reduced the activation barrier for the removal of O on both flat and stepped Co surfaces. An approximate kinetic model was developed and a volcano plot was established. The turn-over frequency (TOF) calculated based on the activation barriers showed that Pt promoted Co surface had a higher rate than unpromoted Co surface. The effect of Pt and Ru promoters on various pathways of C deposition on Co catalyst was studied to gain a mechanistic understanding. The promoters did not affect the subsurface C formation but they increased the barriers for C-C and C-C-C formation and also decreased the barriers for C-H formation. The promoters also influence the stabilities of C compounds on the Co surface suggesting that Pt and Ru promoters would decrease C deposition on Co catalysts. The effect of Pt promoter on unassisted and H-assisted CO activation pathways on Co catalyst was studied. Pt promoted Co surface followed H-assisted CO activation. Pt promoter decreased the activation barriers for CO activation pathways on Co catalyst thereby increasing the activity of Co catalyst. CO oxidation is a process used to prevent poisoning of fuel cell catalysts and reduce pollution of the atmosphere through exhaust gases containing CO. Expensive catalysts like Pt are widely used for CO oxidation which significantly increases the cost of the process and hence it is necessary to search for alternative lower cost catalysts. Understanding the mechanism of a reaction is the first step towards designing better and efficient catalyst. DFT is helpful in determining the basic mechanism and intermediates of reactions. The mechanism of CO oxidation on CoO catalyst was explored. Four possible mechanisms for CO oxidation on CoO catalyst were studied to determine the most likely mechanism. The mechanism was found to be a two-step process with activation barrier for formation of CO2 larger than the barrier for formation of the intermediate species.

The slags from iron-steel and zinc industries are rich in metal oxide contents like FeO, MnO, CaO. However, these slags are not used extensively, except some usage in the cement industry. These slags can be used in removing H2S from waste gases from different industrial sources. The purpose of this research is to study the effect of initial concentration of H2S on the capacity and sorbent efficiency of waste materials from iron-steel and zinc industries. Experiments were conducted in a 25 mm-quartz reactor with simulated gases containing H2S as reactive gas. Breakthrough curves for sulfidation reactions were obtained for 3000 ppmv, 4000 ppmv and 5000 ppmv initial H2S concentrations at the reaction temperature range of 500&deg / C&ndash / 700&deg / C. According to the results obtained from the experiments, the H2S removal capacity of both slags increased with increasing reaction temperature, however, the H2S removal capacity of the slags decreases as the initial H2S concentration increases. Cyclic sulfidation and regeneration tests were applied to both steel and zinc slags in order to determine the regenerability of the slags. In cyclic tests, zinc slag gave better results than steel slag. A &ldquo / Deactivation Model&rdquo / was used in order to fit the breakthrough curves obtained experimentally to the breakthrough curves predicted from the deactivation model. A very good fit was obtained for both steel and zinc slags. Zinc slag was shown to be more suitable for gas cleanup than steel slag taking into account its high H2S removal efficiency, regenerability and low cost (almost free of charge).

The objective of this study was to investigate the possibility of recycling enzymes during saccharification of cellulose for the production of ethanol from woodchips. To make enzyme recycling feasible and economical when woodchips are processed for ethanol, the lignin in the wood is to be removed before the enzymes are added. Since enzymes constitute a major part of the input costs, second only to the feedstock, the ability to reuse the enzymes could lead to a considerable decrease in the production cost of ethanol. Tulip poplar woodchips were selected as the feedstock. Different delignification methods with recovery of byproducts were investigated. Alkali extraction, using dilute NaOH for the removal of lignin after steam pretreatment, was used as the base case against which all other processes were compared. Recovery of furfural and methanol, produced during the pretreatment of the woodchips, for sale as byproducts was one modification to the alkali extraction process that was investigated. The conversion of xylose to furfural and the recovery of the furfural as a byproduct was the third case explored. Solvent extraction using a 50:50 ethanol-water mixture instead of extraction with NaOH was the fourth case examined. Process flow sheets were then developed to recycle the enzymes during the hydrolysis and fermentation of this prehyrolyzed and delignified wood. Two reactor setup schemes were examined for enzyme recycling. One scheme involved a single train of reactors, with the whole pretreated slurry flowing from one reactor to the next, whereas, in the other scheme, the slurry was split among parallel trains of reactors. The activity loss of the enzymes was modeled such that a part of the enzymes entering the reactor lost all their activity. The loss of activity in multiple steps, with enzymes losing only some of their activity, was also modeled. Here the enzymes entering the reactor constituted a mixture with different activities instead of all the enzymes having the same activity like in the previous single step model. Recovering methanol and furfural reduced the minimum ethanol selling price. High temperature ethanol water pretreatment and lignin extraction reduced the minimum ethanol selling price compared to the base case of steam pretreatment followed by alkali extraction. Enzyme recycling also reduces the minimum ethanol selling price. The magnitude of the price reduction depends on the recycling scheme selected and the rate of enzyme deactivation, which has not been measured.

Catalyst deactivation in single catalyst pellets and in an isothermal catalytic fixed bed reactor have been analytically studied. The work reported here is unlike the vast majority of previous theoretical analyses which are numerical. This thesis deals with two types of deactivation - parallel and series mechanisms in which respectively, reactant and product are directly responsible for poisoning. For the single particle studies, the principal analytical tools used are based on singular perturbation theory. Use of these techniques in the temporal domain depends crucially on the smallness of the ratio of the deactivation rate constant to that of the main reaction. Depending on the range of Thiele modulus, whether small, intermediate or large, three different techniques are used in the spatial domain. First, when the Thiele modulus is small, a lumping technique due to Frank-Kamentskii (1955) is used to replace the Laplacian operator by a suitable constant. This reduces the coupled partial differential equations to ordinary ones. Main chemical kinetics of n-th order and of Michaelis-Menten type are investigated. Second, when the Thiele modulus is very large, matched asymptotic expansions in the spatial domain are used. The analysis is based on the smallness of the inverse Thiele modulus, 1/phi2. A moving reaction zone of thickness 0(1/phi) is found to separate the dead shell from the active core of the catalyst pellet. The catalyst activity profile exhibits a sharp change within the reaction zone and the structure of this profile is found to be self-preserving during the period of its propagation. Solutions are obtained for three different geometries - planar, cylindrical and spherical. The large Thiele modulus results obtained here are found to be more accurate than the shell-model solutions of Masamune and Smith (1966) (except for a slab geometry, when they are identical). Finally, for an intermediate range of Thiele modulus, finite Sturm-Liouville integral transforms along with the concept of an effective average are successfully applied. The approach taken is novel, and although not rigorously justifiable, it leads to results of suprising accuracy. The versatility of the technique is demonstrated by application to various non-linear problems which posess exact solutions and remarkable agreement is found. The finite-cylindrical catalyst pellet is also investigated using a double-integral transform in the spatial domain and it is shown that for small Thiele modulus, the infinite cylinder and slab results are good approximations to finite length cylinders with small and large ratio, R/L, respectively. The analytical solutions reported in this thesis agree well with the known numerical results of others (Masamune and Smith, 1966; Khang and Levenspiel, 1973 and Lamba and Dudukovic, 1978). The parametric - dependence of these solutions is explicit and numerical results can be easily obtained from them by hand calculation. All the single pellet results are brought together in the final chapter and used to analyze the performance of isothermal fixed-bed reactors undergoing poisoning. Such effects as external mass transfer resistance, pellet shape and chemical kinetic type are included in the analysis, which embraces the entire range of Thiele modulus.

Submitted by Edileide Reis (leyde-landy@hotmail.com) on 2013-04-23T12:46:56Z No. of bitstreams: 1 Luciene Carvalho.pdf: 878421 bytes, checksum: bd74bd7628de44a85a4f2776c0680a24 (MD5) / Made available in DSpace on 2013-04-23T12:46:56Z (GMT). No. of bitstreams: 1 Luciene Carvalho.pdf: 878421 bytes, checksum: bd74bd7628de44a85a4f2776c0680a24 (MD5) Previous issue date: 2003 / Catalisadores Pt-Re-Sn/Al2O3-Cl foram avaliados na reforma de n-octano para produção de compostos aromáticos, especialmente xilenos, e foram comparados com o catalisador monometálico (Pt/Al2O3) e catalisadores bimetálicos (Pt-Re e Pt-Sn/Al2O3) correspondentes. As amostras foram preparadas por coimpregnação e impregnações sucessivas da alumina, com as soluções dos precursores metálicos. Os sólidos foram, então, secos (120oC), calcinados (500oC, 4h, ar) e reduzidos (500oC, 4h, H2). No caso das impregnações sucessivas, os sólidos foram calcinados e reduzidos após cada adição de metal. Os catalisadores contendo rênio foram previamente sulfetados com dissulfeto de carbono. As seguintes amostras foram preparadas (0,3% p/p de cada metal): Pt/Al2O3, Re/Al2O3 e Sn/Al2O3; (Pt+Re)/Al2O3, (Pt+Sn)/Al2O3, (Re+Sn)/Al2O3 e (Pt+Re+Sn)/Al2O3, preparados por coimpregnação; Pt-Re/Al2O3, Pt-Sn/Al2O3, Pt-Re-Sn/Al2O3, Pt-Sn-Re/Al2O3, Re-Pt-Sn/Al2O3, Re-Sn-Pt/Al2O3, Sn-Pt-Re/Al2O3, Sn-Re-Pt/Al2O3, preparados por impregnações sucessivas. As funções ácida e metálica dos catalisadores foram avaliadas através das reações modelo de isomerização de n-pentano e desidrogenação de ciclohexano, respectivamente. A tiotolerância dos catalisadores foi avaliada na desidrogenação de ciclo-hexano. As amostras também foram caracterizadas por análise elementar, redução à temperatura programada (TPR), espectroscopia de infravermelho com transformadas de Fourier (FTIR) de CO adsorvido, quimissorção de H2, microscopia eletrônica de transmissão (TEM) e difração de elétrons (ED). Após os testes de isomerização de n-pentano e reforma de n-octano, as amostras foram analisadas por oxidação à temperatura programada (TPO). Observou-se que a atividade de desidrogenação da platina foi reduzida por rênio e/ou estanho, e este efeito foi mais forte com o estanho. As energias de ativação na desidrogenação de ciclo-hexano foram mais afetadas pelo estanho, que modificou os sítios ativos, produzindo outros de atividade mais baixa. Ambos os metais podem formar ligas com a platina, como mostrado pelos experimentos de TPR e ED. Estes metais tornam a platina mais rica em elétrons, diminuindo sua capacidade de quimissorção. A existência deste efeito eletrônico foi confirmada por FTIR de CO adsorvido. O enriquecimento eletrônico da platina também afetou a tiotolerância, uma vez que a ligação Pt-S tornou-se mais forte. Por isso, os catalisadores bimetálicos apresentaram uma tiotolerância mais baixa do que o catalisador monometálico, e os trimetálicos foram ainda mais susceptíveis ao envenenamento por enxofre, devido à ação conjunta dos dois metais sobre a platina. Nas amostras preparadas por coimpregnação, o envenenamento foi mais severo, por causa do melhor contato entre os metais. A função ácida também foi afetada pela presença de estanho e rênio, principalmente na forma de óxidos. Nos catalisadores trimetálicos, efeitos positivos foram encontrados quando o estanho foi adicionado primeiro, porque o óxido de estanho reduziu a quantidade de sítios ácidos mais fortes da alumina, produzindo uma acidez mais conveniente às reações de isomerização e ciclização. Dessa forma, o catalisador Sn-Pt-Re-S/Al2O3 foi o mais promissor em reforma de n-octano, com a mais alta seletividade a xilenos, maior estabilidade e produção de gases mais baixa. O catalisador trimetálico coimpregnado mostrou um desempenho similar e tem a vantagem adicional de não precisar ser previamente sulfetado. / Salvador