O titanato de estrôncio (SrTiO3) é um óxido de estrutura perovskita e tem sido intensamente estudado para uso em diversas aplicações, entre elas, como suporte catalítico. Entretanto, sua utilização especificamente na reação de deslocamento gás-água ainda é pouco explorada. Esta reação é de interesse para a produção de hidrogênio livre de CO, necessário para aplicações como o abastecimento de células de combustível. Este trabalho de tese teve como objetivo o estudo dos compostos SrTi1-xCuxO3, CuO/SrTiO3 e NiO/SrTiO3 como catalisadores para a reação de deslocamento gás-água, uma vez que, dentre os metais de baixo custo, Cu e Ni são altamente ativos para esta reação. As amostras SrTi1-xCuxO3 foram sintetizadas pelo método dos precursores poliméricos com calcinação em N2 e O2, possibilitando a obtenção de partículas de maior área superficial em comparação com a calcinação convencional em atmosfera ambiente. Para as amostras CuO/SrTiO3 e NiO/SrTiO3, o suporte SrTiO3, foi sintetizado pelo método de sol-precipitação e a impregnação com cobre e níquel foi realizada por via úmida. As técnicas de absorção e difração de raios-X in situ em condições de reação mostraram a estabilidade da estrutura e do estado de oxidação após o tratamento de redução. Imagens de microscopia eletrônica de varredura (MEV) e de transmissão (TEM) em conjunto com a espectroscopia de raios-X de energia dispersiva (EDX) foram utilizadas a fim de estabelecer uma relação entre a atividade catalítica e o teor a dispersão de fase ativa sobre o suporte. Todas as composições estudadas se mostraram ativas entre 250 e 350°C, entretanto, a composição NiO/SrTiO3 com 10% de Ni apresentou o melhor resultado, com uma conversão de CO a 350°C, próxima ao equilíbrio e estável por um período mínimo de10 h. / Strontium titanate (SrTiO3) is an oxide of perovskite structure and has been extensively studied for use in several applications, including as catalytic support. However, its use specifically in the water-gas shift reaction is still little explored. This reaction is of interest for the production of CO-free hydrogen, required for applications such as in fuel cell. This work aimed to study SrTi1-xCuxO3, CuO/SrTiO3 and NiO/SrTiO3 compounds to be applied as catalysts for the water-gas shift reaction, since, among the low-cost metals, Cu and Ni are highly active for this reaction. The SrTi1-xCuxO3 samples were synthesized by the polymeric precursor method with the samples submitted to a N2 and O2 calcination, making possible to obtain particles with a larger surface area compared to conventional calcination in ambient atmosphere. For the CuO/SrTiO3 and NiO/SrTiO3 samples, the SrTiO3 support was synthesized by the sol-precipitation method and the impregnation with copper and nickel on the support was performed by a wet method. The in situ X-ray absorption and diffraction techniques under reaction conditions showed the stability of the structure and the oxidation state after the reduction treatment. Scanning electron microscopy (SEM) and transmission (TEM) images in conjunction with energy dispersive X-ray spectroscopy (EDX) were used in order to establish a relationship between the catalytic activity and the content and dispersion of the active phase on the support. All the compositions studied were active at 250 to 350 °C, however, the NiO/SrTiO3 sample with 10% of Ni presented the best result, with a CO conversion at 350 °C, close to equilibrium and stable for a minimum of 10 h.

A produção de hidrogênio está em ascensão no cenário das refinarias de petróleo devido à sua utilização no processo de hidrotratamento do diesel, ou seja, para a adequeação do teor de enxofre permitido nos combustíveis visando atender às regulamentações da Agência Nacional do Petróleo. Uma unidade de geração de hidrogêno (UGH) produz hidrogênio através do processo de reforma catalítica a vapor, sendo composta principalmente por dois reatores, um onde ocorre a reforma e o outro onde ocorre reação de deslocamento, além de uma etapa de purificação do produto final, sendo a PSA (pressure swing adsorption) a mais utilizada hoje em dia. O reator de reforma é composto por vários tubos alocados dentro de um forno onde ocorre a reforma do gás natural, que é carga da unidade, em contato com o vapor produzindo principalmente hidrogênio e dióxido de carbono. Já a reação de deslocamento visa produzir mais hidrogênio, através da reação do monóxido de carbono gerado na reforma com o vapor de água. Este trabalho visa desenvolver métricas de eficiência operacional para UGH para fins de avaliar a eficiência da unidade como um todo. Para que a métrica possa ser desenvolvida, os principais reatores da unidade foram modelados, o forno reformador e o reator da reação de deslocamento, utilizando-se uma modelagem matemática e modelos cinéticos mais adequados às características do processo e dos catalisadores utilizados encontrados na literatura. A modelagem foi implementada em linguagem Modelica sendo simulada com jmodelica e python Como estudo de caso se utilizou dados reais de uma Unidade de Geração de Hidrogênio de uma refinaria localizada no sul do Brasil. Através da simulação dos dados foi possível concluir que a modelagem representou adequadamente os dois reatores da unidade, sendo que no reator de reforma se conseguiu uma conversão simulada de 82% em média de metano e no reator de shift, 81% de conversão média de CO. O erro relativo médio entre as medidas de temperatura, simulados e real, foi de 2% e entre os percentuais de hidrogênio produzido, foi de 5% para a reação de reforma. Já para a reação de deslocamento, os erros médios relativos entre os valores simulados e reais de temperatura e de produção de hidrogênio foi de 1,2% e 1,3%, respectivamente. Com os valores obtidos da simulação do modelo em comparação com os reais de planta foi possível observar uma ineficiência na unidade em estudo que pode ser visualizada através da métrica proposta, que foi a razão de hidrogênio total produzido na saída do reator de deslocamento pela carga de gás natural do forno reformador. Também foi avaliada a eficiência da PSA em relação à variação da carga da unidade, apresentando uma eficiência média de 83%. / Hydrogen production is on the rise in the oil refinery scenario due to its use in the diesel hydrotreating process, ie to adjust the permitted sulfur content in fuels in order to comply with the regulations of the National Petroleum Agency. A hydrogen generating unit (UGH) produces hydrogen through the process of steam reforming, being composed mainly of a two reactors, one where the reform takes place and the other where a displacement reaction takes place, besides a step of purification of the product PSA is the most used today. The reforming reactor is composed of several tubes placed inside an oven where the reform of natural gas occurs, which is charge of the unit, in contact with the vapor producing mainly hydrogen and carbon dioxide. In the displacement reaction, a displacement reaction occurs producing more hydrogen with the carbon monoxide generated in the reform and not converted to CO2. This work aims to obtain operational efficiency metrics of a UGH in order to evaluate the efficiency of the unit as a whole. For the metric to be developed, the main reactors of the unit were modeled, the reforming furnace and the reactor of the displacement reaction, using a mathematical modeling and kinetic models more appropriate to the process characteristics and catalysts used in the literature. The modeling was implemented in Modelica language being simulated with jmodelica and python As a case study we used real data from a Hydrogen Generation Unit of a refinery located in southern Brazil. By means of the simulation of the data it was possible to conclude that the modeling adequately represented the two reactors of the unit, and in the reforming reactor, a simulated conversion of 82% in average methane and in the shift reactor was achieved, 81% average CO . The average relative error between simulated and real temperature measurements was 2% and among the percentages of hydrogen produced was 5% for the reform reaction. As for the displacement reaction, the mean relative errors between the simulated and actual values of temperature and hydrogen production were 1.2% and 1.3%, respectively. With the obtained values of the simulation of the model in comparison with the actual ones of plant, it was possible to observe a deficiency in the efficiency of the study unit that can be visualized through the proposed metric, which was the ratio of total hydrogen produced at the exit of the displacement reactor by the Natural gas charge of the reformer furnace. The efficiency of the PSA was also evaluated in relation to the variation of the unit load, presenting an average efficiency of 83%.

In this research project, adsorption is considered in conjunction with the reverse water gas shift reaction in order to convert CO2 to CO for synthetic fuel production. If the CO2 for this process can be captured from high emitting industries it can be a very good alternative for reduced fossil fuel consumption and GHG emission mitigation. CO as an active gas could be used in Fischer-Tropsch process to produce conventional fuels. Literature review and process simulation were carried out in order to determine the best operating conditions for reverse water gas shift (RWGS) reaction. Increasing CO2 conversion to CO requires CO2/CO separation downstream of the reactor and recycling unreacted CO2 and H2 back into the reactor. Adsorption as a viable and cost effective process for gas separation was chosen for the CO2/CO separation. This was started by a series of adsorbent screening experiments to select the best adsorbent for the application. Screening study was performed by comparing pure gas isotherms for CO2 and CO at different temperatures and pressures. Then experimental isotherm data were modeled by the Temperature-Dependent Toth isotherm model which provided satisfactory fits for these isotherms. Henry law’s constant, isosteric heat of adsorption and binary mixture prediction were determined as well as selectivity for each adsorbent. Finally, the expected working capacity was calculated in order to find the best candidate in terms of adsorption and desorption. Zeolite NaY was selected as the best candidate for CO2/CO separation in adsorption process for this project. In the last step breakthrough experiments were performed to evaluate operating condition and adsorption capacity for real multi component mixture of CO2, CO, H2 in both cases of saturated with water and dry gas basis. In multi components experiments zeolite NaY has shown very good performance to separate CO2/CO at low adsorption pressure and ambient temperature. Also desorption experiment was carried out in order to evaluate the working capacity of the adsorbent for using in industrial scale and eventually temperature swing adsorption (TSA) process worked very well for the regeneration step. Integrated adsorption system downstream of RWGS reactor can enhance the conversion of CO2 to CO in this process significantly resulting to provide synthetic gas for synthetic fuel production as well as GHG emission mitigation.

The production of clean fuel faces the challenges of high production cost and complying with stricter environmental regulations. In this research, the ability of using a novel technology of upgrading heavy oil to treat Light Gas Oil (LGO) will be investigated. The target of this project is to produce cleaner transportation fuel with much lower cost of production. Recently, a novel process for upgrading of heavy oil has been developed at University of Waterloo. It is combining the two essential processes in bitumen upgrading; emulsion breaking and hydroprocessing into one process. The water in the emulsion is used to generate in situ hydrogen from the Water Gas Shift Reaction (WGSR). This hydrogen can be used for the hydrogenation and hydrotreating reaction which includes sulfur removal instead of the expensive molecular hydrogen. This process can be carried out for the upgrading of the bitumen emulsion which would improve its quality. In this study, the hydrodesulphurization (HDS) of LGO was conducted using in situ hydrogen produced via the Water Gas Shift Reaction (WGSR). The main objective of this experimental study is to evaluate the possibility of producing clean LGO over dispersed molybdenum sulphide catalyst and to evaluate the effect of different promoters and syn-gas on the activity of the dispersed Mo catalyst. Experiments were carried out in a 300 ml Autoclave batch reactor under 600 psi (initially) at 391oC for 1 to 3 hours and different amounts of water. After the hydrotreating reaction, the gas samples were collected and the conversion of carbon monoxide to hydrogen via WGSR was determined using a refinery gas analyzer. The sulphur content in liquid sample was analyzed via X-Ray Fluorescence. Experimental results showed that using more water will enhance WGSR but at the same time inhibits the HDS reaction. It was also shown that the amount of sulfur removed depends on the reaction time. The plan is to investigate the effect of synthesis gas (syngas) molar ratio by varying CO to H2 ratio. It is also planned to use different catalysts promoters and compare them with the un-promoted Mo based catalysts to achieve the optimum reaction conditions for treating LGO. The results of this study showed that Ni and Co have a promoting effect over un-promoted Mo catalysts for both HDS and WGSR. Ni was found to be the best promoter for both reactions. Fe showed no significant effect for both WGSR and HDS. V and K have a good promoting effect in WGSR but they inhibited the HDS reaction. Potassium was found to be the strongest inhibitor for the HDS reaction since no sulfur was removed during the reaction

The production of clean fuel faces the challenges of high production cost and complying with stricter environmental regulations. In this research, the ability of using a novel technology of upgrading heavy oil to treat Light Gas Oil (LGO) will be investigated. The target of this project is to produce cleaner transportation fuel with much lower cost of production. Recently, a novel process for upgrading of heavy oil has been developed at University of Waterloo. It is combining the two essential processes in bitumen upgrading; emulsion breaking and hydroprocessing into one process. The water in the emulsion is used to generate in situ hydrogen from the Water Gas Shift Reaction (WGSR). This hydrogen can be used for the hydrogenation and hydrotreating reaction which includes sulfur removal instead of the expensive molecular hydrogen. This process can be carried out for the upgrading of the bitumen emulsion which would improve its quality. In this study, the hydrodesulphurization (HDS) of LGO was conducted using in situ hydrogen produced via the Water Gas Shift Reaction (WGSR). The main objective of this experimental study is to evaluate the possibility of producing clean LGO over dispersed molybdenum sulphide catalyst and to evaluate the effect of different promoters and syn-gas on the activity of the dispersed Mo catalyst. Experiments were carried out in a 300 ml Autoclave batch reactor under 600 psi (initially) at 391oC for 1 to 3 hours and different amounts of water. After the hydrotreating reaction, the gas samples were collected and the conversion of carbon monoxide to hydrogen via WGSR was determined using a refinery gas analyzer. The sulphur content in liquid sample was analyzed via X-Ray Fluorescence. Experimental results showed that using more water will enhance WGSR but at the same time inhibits the HDS reaction. It was also shown that the amount of sulfur removed depends on the reaction time. The plan is to investigate the effect of synthesis gas (syngas) molar ratio by varying CO to H2 ratio. It is also planned to use different catalysts promoters and compare them with the un-promoted Mo based catalysts to achieve the optimum reaction conditions for treating LGO. The results of this study showed that Ni and Co have a promoting effect over un-promoted Mo catalysts for both HDS and WGSR. Ni was found to be the best promoter for both reactions. Fe showed no significant effect for both WGSR and HDS. V and K have a good promoting effect in WGSR but they inhibited the HDS reaction. Potassium was found to be the strongest inhibitor for the HDS reaction since no sulfur was removed during the reaction

Heavy oil derived from oil sands is becoming an important resource of energy and transportation fuels due to the depletion of conventional oil resources. However, bitumen and heavy oils have a low hydrogen/carbon ratio and contain a large percentage of sulfur and nitrogen heterocyclic compounds. At the level of deep desulfurization, aromatic poly-nuclear molecules, especially nitrogen-containing heterocyclic compounds, exhibit strong inhibitive effect on hydrodesulfurization (HDS) due to competitive adsorption on catalytically active sites with sulfur-containing molecules. Therefore, it is necessary to study the HDS of refractory sulfur-containing compounds and also the effect of nitrogen-containing species on the deep HDS for achieving the ultra low sulfur specifications for transportation fuels. Additionally, the cost of H2 increased in recent years and a bitumen emulsion upgrading technique using an alternative in-situ H2 generated via the water gas shift (WGS) reaction during the hydro-treating was developed in our group. In the present study, a kind of nano-dispersed unsupported MoSx based catalyst was developed and used for hydrodesulfurization, hydrodenitrogenation (HDN) and upgrading bitumen emulsions. Objectives of this thesis were to (1) improve the catalytic activity of the nano-dispersed Mo based catalysts towards the HDS and HDN reactions of refractory sulfur-/nitrogen-containing compounds; and (2) compare the reactivity of in-situ hydrogen generated via the WGS reaction versus externally provided molecular hydrogen in HDS and HDN reactions to improve the efficiency of the bitumen emulsion upgrading technology developed by our group. In the present study, to stimulate the reaction system of bitumen emulsion, water was added into the organic reaction system, so there are different phases in this reaction system. To investigate the activity of the catalyst, the catalyst particles dispersed in different phases were characterized separatedly via HRTEM-EDX. After HRTEM-EDX study, all phases were mixed up and dried for further characterizations, BET, SEM, and XRD. The catalyst prepared in in-situ hydrogen was found to have higher surface area and smaller particle size than the one made in molecular hydrogen. The presence of sulfur-/nitrogen-containing compounds in the preparation system caused significant changes in the morphology of dispersed Mo sulfide catalyst according to HRTEM observations. Refractory sulfur-containing compounds of dibenzothiophene (DBT) and 4,6-dimethyldibenzothiophene (4,6-DMDBT) were used as model compounds in HDS studies. The simultaneous HDS of both model compounds was performed at different reaction temperatures from 330??C to 400??C. The effect of the reaction temperature on the WGS reaction in the presence of sulfur-containing model compounds was reported. A kinetic model for HDS reactions was proposed and used in discussing experiment results. The relative HDS reactivity of 4,6-DMDBT to DBT using dispersed Mo sulfide catalyst in in-situ hydrogen was found to be higher than the reported results which were obtained over supported catalysts. Nickel and potassium were introduced into Mo sulfide catalysts as promoters and their effect on the WGS reaction and the HDS reaction were discussed. The simultaneous HDS was carried out in the two different hydrogen sources. The in-situ hydrogen reaction system showed higher conversion and desulfurization results of both sulfur model compounds. This observation has been found to be mainly contributed by the higher activity of the Mo sulfide catalyst prepared in in-situ H2. Strong inhibitive effect of nitrogen-containing compounds, basic quinoline or non-basic carbazole, on the HDS of refractory sulfur model compounds was observed and discussed. Basic quinoline was a much stronger inhibitor than non-basic carbazole. The two HDS reaction pathways were affected by nitrogen-containing compounds to different extents. The HDN of quinoline over the dispersed Mo sulfide catalyst using in-situ hydrogen had been studied extensively by a previous member in our group. In this thesis, the HDN of carbazole was studied. From the identification of HDN products of carbazole, a HDN reaction network was proposed. The HDN of carbazole was processed at different reaction temperatures. The WGS reaction was not inhibited in the presence of carbazole. Comparable reactivity of the two hydrogen sources towards the HDN of carbazole was observed. The presence of 4,6-DMDBT caused significant effect on the HDN of carbazole due to the competitive adsorption on the catalyst surface.

In this research project, adsorption is considered in conjunction with the reverse water gas shift reaction in order to convert CO2 to CO for synthetic fuel production. If the CO2 for this process can be captured from high emitting industries it can be a very good alternative for reduced fossil fuel consumption and GHG emission mitigation. CO as an active gas could be used in Fischer-Tropsch process to produce conventional fuels. Literature review and process simulation were carried out in order to determine the best operating conditions for reverse water gas shift (RWGS) reaction. Increasing CO2 conversion to CO requires CO2/CO separation downstream of the reactor and recycling unreacted CO2 and H2 back into the reactor. Adsorption as a viable and cost effective process for gas separation was chosen for the CO2/CO separation. This was started by a series of adsorbent screening experiments to select the best adsorbent for the application. Screening study was performed by comparing pure gas isotherms for CO2 and CO at different temperatures and pressures. Then experimental isotherm data were modeled by the Temperature-Dependent Toth isotherm model which provided satisfactory fits for these isotherms. Henry law’s constant, isosteric heat of adsorption and binary mixture prediction were determined as well as selectivity for each adsorbent. Finally, the expected working capacity was calculated in order to find the best candidate in terms of adsorption and desorption. Zeolite NaY was selected as the best candidate for CO2/CO separation in adsorption process for this project. In the last step breakthrough experiments were performed to evaluate operating condition and adsorption capacity for real multi component mixture of CO2, CO, H2 in both cases of saturated with water and dry gas basis. In multi components experiments zeolite NaY has shown very good performance to separate CO2/CO at low adsorption pressure and ambient temperature. Also desorption experiment was carried out in order to evaluate the working capacity of the adsorbent for using in industrial scale and eventually temperature swing adsorption (TSA) process worked very well for the regeneration step. Integrated adsorption system downstream of RWGS reactor can enhance the conversion of CO2 to CO in this process significantly resulting to provide synthetic gas for synthetic fuel production as well as GHG emission mitigation.

A produção de hidrogênio está em ascensão no cenário das refinarias de petróleo devido à sua utilização no processo de hidrotratamento do diesel, ou seja, para a adequeação do teor de enxofre permitido nos combustíveis visando atender às regulamentações da Agência Nacional do Petróleo. Uma unidade de geração de hidrogêno (UGH) produz hidrogênio através do processo de reforma catalítica a vapor, sendo composta principalmente por dois reatores, um onde ocorre a reforma e o outro onde ocorre reação de deslocamento, além de uma etapa de purificação do produto final, sendo a PSA (pressure swing adsorption) a mais utilizada hoje em dia. O reator de reforma é composto por vários tubos alocados dentro de um forno onde ocorre a reforma do gás natural, que é carga da unidade, em contato com o vapor produzindo principalmente hidrogênio e dióxido de carbono. Já a reação de deslocamento visa produzir mais hidrogênio, através da reação do monóxido de carbono gerado na reforma com o vapor de água. Este trabalho visa desenvolver métricas de eficiência operacional para UGH para fins de avaliar a eficiência da unidade como um todo. Para que a métrica possa ser desenvolvida, os principais reatores da unidade foram modelados, o forno reformador e o reator da reação de deslocamento, utilizando-se uma modelagem matemática e modelos cinéticos mais adequados às características do processo e dos catalisadores utilizados encontrados na literatura. A modelagem foi implementada em linguagem Modelica sendo simulada com jmodelica e python Como estudo de caso se utilizou dados reais de uma Unidade de Geração de Hidrogênio de uma refinaria localizada no sul do Brasil. Através da simulação dos dados foi possível concluir que a modelagem representou adequadamente os dois reatores da unidade, sendo que no reator de reforma se conseguiu uma conversão simulada de 82% em média de metano e no reator de shift, 81% de conversão média de CO. O erro relativo médio entre as medidas de temperatura, simulados e real, foi de 2% e entre os percentuais de hidrogênio produzido, foi de 5% para a reação de reforma. Já para a reação de deslocamento, os erros médios relativos entre os valores simulados e reais de temperatura e de produção de hidrogênio foi de 1,2% e 1,3%, respectivamente. Com os valores obtidos da simulação do modelo em comparação com os reais de planta foi possível observar uma ineficiência na unidade em estudo que pode ser visualizada através da métrica proposta, que foi a razão de hidrogênio total produzido na saída do reator de deslocamento pela carga de gás natural do forno reformador. Também foi avaliada a eficiência da PSA em relação à variação da carga da unidade, apresentando uma eficiência média de 83%. / Hydrogen production is on the rise in the oil refinery scenario due to its use in the diesel hydrotreating process, ie to adjust the permitted sulfur content in fuels in order to comply with the regulations of the National Petroleum Agency. A hydrogen generating unit (UGH) produces hydrogen through the process of steam reforming, being composed mainly of a two reactors, one where the reform takes place and the other where a displacement reaction takes place, besides a step of purification of the product PSA is the most used today. The reforming reactor is composed of several tubes placed inside an oven where the reform of natural gas occurs, which is charge of the unit, in contact with the vapor producing mainly hydrogen and carbon dioxide. In the displacement reaction, a displacement reaction occurs producing more hydrogen with the carbon monoxide generated in the reform and not converted to CO2. This work aims to obtain operational efficiency metrics of a UGH in order to evaluate the efficiency of the unit as a whole. For the metric to be developed, the main reactors of the unit were modeled, the reforming furnace and the reactor of the displacement reaction, using a mathematical modeling and kinetic models more appropriate to the process characteristics and catalysts used in the literature. The modeling was implemented in Modelica language being simulated with jmodelica and python As a case study we used real data from a Hydrogen Generation Unit of a refinery located in southern Brazil. By means of the simulation of the data it was possible to conclude that the modeling adequately represented the two reactors of the unit, and in the reforming reactor, a simulated conversion of 82% in average methane and in the shift reactor was achieved, 81% average CO . The average relative error between simulated and real temperature measurements was 2% and among the percentages of hydrogen produced was 5% for the reform reaction. As for the displacement reaction, the mean relative errors between the simulated and actual values of temperature and hydrogen production were 1.2% and 1.3%, respectively. With the obtained values of the simulation of the model in comparison with the actual ones of plant, it was possible to observe a deficiency in the efficiency of the study unit that can be visualized through the proposed metric, which was the ratio of total hydrogen produced at the exit of the displacement reactor by the Natural gas charge of the reformer furnace. The efficiency of the PSA was also evaluated in relation to the variation of the unit load, presenting an average efficiency of 83%.

86 f. / Submitted by Ana Hilda Fonseca (anahilda@ufba.br) on 2013-04-11T16:26:04Z No. of bitstreams: 1 HilmaFonseca_dissertação_ 2011.pdf: 2091220 bytes, checksum: ab1c531b8b0822721d83a6c9c22c4001 (MD5) / Approved for entry into archive by Ana Hilda Fonseca(anahilda@ufba.br) on 2013-05-10T16:56:08Z (GMT) No. of bitstreams: 1 HilmaFonseca_dissertação_ 2011.pdf: 2091220 bytes, checksum: ab1c531b8b0822721d83a6c9c22c4001 (MD5) / Made available in DSpace on 2013-05-10T16:56:08Z (GMT). No. of bitstreams: 1 HilmaFonseca_dissertação_ 2011.pdf: 2091220 bytes, checksum: ab1c531b8b0822721d83a6c9c22c4001 (MD5) Previous issue date: 2011 / CNPq / A reação de deslocamento de monóxido de carbono com vapor d‟água (water gas shift reaction, WGSR) é um processo industrial amplamente utilizado, sendo uma etapa fundamental para a produção comercial de hidrogênio de alta pureza. A reação também é importante para a remoção de monóxido de carbono, a partir de vapores ricos em hidrogênio, uma vez que ele envenena a maioria dos catalisadores metálicos. O catalisador clássico da WGSR, conduzida na faixa de 350-420 °C, é a hematita dopada com oxido de cromo, que é toxico e perde área superficial específica, durante os processos industriais. Portanto, um esforço considerável tem sido feito nos últimos anos, a fim de obter catalisadores alternativos para a reação. Os catalisadores do tipo perovskita têm atraído muita atenção nos últimos tempos devido à alta flexibilidade da sua estrutura, às suas propriedades redox e à possibilidade de controlar as propriedades ácido-base. Desta forma foram estudados, neste trabalho, óxidos tipo perovskitas LaFe1-xCoxO3 (0 ≥ x ≤ 1), que foram empregados como precursores de catalisadores alternativos da WGSR. As amostras foram preparadas por decomposição térmica dos precursores, obtidos pelo método do citrato amorfo, seguida de calcinação a 600 ° C, por 4 h. As amostras foram caracterizadas por espectroscopia no infravermelho com transformada de Fourier, difração de raios X, fluorescência de raios X, medidas de área de superfície específica, redução à temperatura programada, espectroscopia de refletância difusa no ultravioleta e visível e microscopia eletrônica de varredura. Os catalisadores foram avaliados em WGSR, conduzida a 1 atm e distintas temperaturas, na faixa de 250 a 600 °C. Antes da reação, as amostras foram reduzidos sob fluxo de hidrogênio a 600 °C, por 1 h. Todas as amostras exibiram uma única fase de perovskita. A amostra isenta de ferro mostrou uma baixa área superficial específica (3,5 m2g-1), que aumentou com a introdução de ferro, sendo alcançados valores na faixa de 12 a 17 m2.g-1. A redução da perovskita LaCoO3 ocorreu em duas etapas, a primeira em torno de 300 °C, atribuída à redução da espécies Co3+ para Co2+ e a segundo em cerca de 500 °C, relacionada à redução de espécies Co2+ para Co0. Em todas as amostras, a adição de ferro dificultou a produção de espécies Co0 e este efeito aumentou com a quantidade de ferro em sólidos. Todos os catalisadores levaram a valores similares de conversão de monóxido de carbono em temperaturas até 300 °C. O catalisador LaCoO3 foi o mais ativo na faixa de 250-450 °C e a adição de ferro diminuiu a atividade neste intervalo de temperatura. Em temperaturas superiores a 450 °C, o efeito do ferro sobre a atividade catalítica foi dependente da sua quantidade nos sólidos. Em quantidades baixas (x= 0,1), altas (x= 0,9) ou iguais (x= 0,5) a atividade diminuiu, enquanto em quantidades intermediárias (x= 0,3) e (x= 0,7), houve um aumento. Estes resultados podem ser explicados pelo fato de que o cobalto ser facilmente reduzido na estrutura perovskita garantindo alta atividade na WGSR. A adição de quantidades elevadas de ferro (x = 0,9) gera um sólido com alta resistência à redução e, portanto, menos ativos na WGSR. Por outro lado, a adição de uma quantidade intermediária (x = 0,3) leva a um sólido capaz de ser reduzido em temperaturas superiores a 450 °C, aumentando a atividade catalítica. A partir desses resultados, pode-se concluir que óxidos com estrutura perovskita do tipo LaFe1-xCoxO3 são precursores promissores para catalisadores da WGSR em altas temperaturas (>350 °C); a adição de ferro é benéfica em quantidade suficiente para produzir uma perovskita tipo LaFe0,7Co0,3O3, obtém-se o catalisador mais ativo em altas temperaturas / Salvador

A produção de hidrogênio está em ascensão no cenário das refinarias de petróleo devido à sua utilização no processo de hidrotratamento do diesel, ou seja, para a adequeação do teor de enxofre permitido nos combustíveis visando atender às regulamentações da Agência Nacional do Petróleo. Uma unidade de geração de hidrogêno (UGH) produz hidrogênio através do processo de reforma catalítica a vapor, sendo composta principalmente por dois reatores, um onde ocorre a reforma e o outro onde ocorre reação de deslocamento, além de uma etapa de purificação do produto final, sendo a PSA (pressure swing adsorption) a mais utilizada hoje em dia. O reator de reforma é composto por vários tubos alocados dentro de um forno onde ocorre a reforma do gás natural, que é carga da unidade, em contato com o vapor produzindo principalmente hidrogênio e dióxido de carbono. Já a reação de deslocamento visa produzir mais hidrogênio, através da reação do monóxido de carbono gerado na reforma com o vapor de água. Este trabalho visa desenvolver métricas de eficiência operacional para UGH para fins de avaliar a eficiência da unidade como um todo. Para que a métrica possa ser desenvolvida, os principais reatores da unidade foram modelados, o forno reformador e o reator da reação de deslocamento, utilizando-se uma modelagem matemática e modelos cinéticos mais adequados às características do processo e dos catalisadores utilizados encontrados na literatura. A modelagem foi implementada em linguagem Modelica sendo simulada com jmodelica e python Como estudo de caso se utilizou dados reais de uma Unidade de Geração de Hidrogênio de uma refinaria localizada no sul do Brasil. Através da simulação dos dados foi possível concluir que a modelagem representou adequadamente os dois reatores da unidade, sendo que no reator de reforma se conseguiu uma conversão simulada de 82% em média de metano e no reator de shift, 81% de conversão média de CO. O erro relativo médio entre as medidas de temperatura, simulados e real, foi de 2% e entre os percentuais de hidrogênio produzido, foi de 5% para a reação de reforma. Já para a reação de deslocamento, os erros médios relativos entre os valores simulados e reais de temperatura e de produção de hidrogênio foi de 1,2% e 1,3%, respectivamente. Com os valores obtidos da simulação do modelo em comparação com os reais de planta foi possível observar uma ineficiência na unidade em estudo que pode ser visualizada através da métrica proposta, que foi a razão de hidrogênio total produzido na saída do reator de deslocamento pela carga de gás natural do forno reformador. Também foi avaliada a eficiência da PSA em relação à variação da carga da unidade, apresentando uma eficiência média de 83%. / Hydrogen production is on the rise in the oil refinery scenario due to its use in the diesel hydrotreating process, ie to adjust the permitted sulfur content in fuels in order to comply with the regulations of the National Petroleum Agency. A hydrogen generating unit (UGH) produces hydrogen through the process of steam reforming, being composed mainly of a two reactors, one where the reform takes place and the other where a displacement reaction takes place, besides a step of purification of the product PSA is the most used today. The reforming reactor is composed of several tubes placed inside an oven where the reform of natural gas occurs, which is charge of the unit, in contact with the vapor producing mainly hydrogen and carbon dioxide. In the displacement reaction, a displacement reaction occurs producing more hydrogen with the carbon monoxide generated in the reform and not converted to CO2. This work aims to obtain operational efficiency metrics of a UGH in order to evaluate the efficiency of the unit as a whole. For the metric to be developed, the main reactors of the unit were modeled, the reforming furnace and the reactor of the displacement reaction, using a mathematical modeling and kinetic models more appropriate to the process characteristics and catalysts used in the literature. The modeling was implemented in Modelica language being simulated with jmodelica and python As a case study we used real data from a Hydrogen Generation Unit of a refinery located in southern Brazil. By means of the simulation of the data it was possible to conclude that the modeling adequately represented the two reactors of the unit, and in the reforming reactor, a simulated conversion of 82% in average methane and in the shift reactor was achieved, 81% average CO . The average relative error between simulated and real temperature measurements was 2% and among the percentages of hydrogen produced was 5% for the reform reaction. As for the displacement reaction, the mean relative errors between the simulated and actual values of temperature and hydrogen production were 1.2% and 1.3%, respectively. With the obtained values of the simulation of the model in comparison with the actual ones of plant, it was possible to observe a deficiency in the efficiency of the study unit that can be visualized through the proposed metric, which was the ratio of total hydrogen produced at the exit of the displacement reactor by the Natural gas charge of the reformer furnace. The efficiency of the PSA was also evaluated in relation to the variation of the unit load, presenting an average efficiency of 83%.