Essa tese visa o desenvolvimento de metodologias simples, eficazes, versáteis e ambientalmente amigáveis para se obter nanomateriais metálicos com controle fino sobre sua forma, composição e estrutura (interior sólido ou vazio) para aplicações em catálise. Em especial, temos interesse no desenvolvimento de nanoestruturas ocas esféricas (nanocascas) compostas por prata-ouro (AgAu), prata-paládio (AgPd) e prata platina (AgPt). Essas nanocascas foram obtidas através da reação de substituição galvânica entre esferas de Ag e íons AuCl4-, PdCl42- ou PtCl62-, respectivamente. Como a reação de substituição galvânica permite não apenas o controle sobre a composição destes sistemas, mas também a obtenção de interiores vazios, esta estratégia representa uma alternativa promissora para a obtenção de nanomateriais apresentando características controláveis e atrativas para aplicações catalíticas. Diante dessas qualidades, esse projeto focou em aplicações para reações orgânicas de redução e acoplamento. Foi realizada uma investigação, de maneira sistemática, como a estrutura e composição dos nanomateriais metálicos produzidos influenciam a sua atividade catalítica, mostando que as atividades foram fortemente dependentes da composição e estrutura, abrindo a possibilidade para o planejamento de nanocatalisadores com atividades catalíticas controladas para uma transformação de interesse. / This thesis aims at developing facile, efficient, versatile, and environmentally friendly methodologies to obtain metallic nanomaterials with controlled shapes, compositions and structure (solid or hollow interiors) for applications in catalysis. In particular, we focused on hollow nanospheres (nanoshells) composed of silver-gold (AgAu), silver-palladium (AgPd), and silver-platinum (AgPt). These nanoshells were obtained by galvanic replacement reaction between Ag nanosphere and AuCl4-, PdCl42- or PtCl62-, respectively. The galvanic replacement reaction not only allows control over the composition of these systems, but also to obtain hollow interiors. Therefore, this strategy is a very promising alternative for obtaining nanomaterials with controllable features attractive for catalytic applications. In this case, we investigated applications towards reduction and coupling transformations. A systematic investigation was carried out regarding how the structures and compositions of the produced nanoshells influenced their catalytic performance. Our results showed that the activities were strongly dependent on the composition and structure, opening a range of possibility for designing nanocatalysts with desired catalytic activities for a target transformation.

This work describes a novel technique to prepare ultra-thin composite Pd-porous metal membranes for H2 separation. This novel technique consists of the gradual smoothing of the Porous Metal (PM) support's surface with several layers of pre-activated alumina particles of different sizes. The deposition of coarse, fine and ultra-fine alumina particles resulted in the narrowing of the PM' surface pore size distribution. The excellent surface smoothness achieved after the grading of the PM 's surface support allowed for the preparation of gas tight Pd layers as thin as 5.6?m. The Pd layers were extremely uniform due to the presence of the grade layer and strongly attached to the support. Composite Pd membranes prepared on graded supports showed H2 permeance as high as 50 m3/(m2 h bar0.5) at 500ºC and ideal selectivities (H2/He) as high as 27000. Moreover, the H2 permeance and ideal selectivity were stable over 1100 hours at 500ºC in H2 atmosphere. Composite Pd-Cu membranes showed H2 permeance as high as 30 m3/(m2 h bar0.5) at 450ºC and ideal selectivities (H2/He) as high as 900. The H2 permeance and ideal selectivity of Pd-Cu membranes were stable over 500 hours at 450ºC in H2 atmosphere. The outstanding long-term H2 permeance and ideal selectivity stability of all composite Pd and Pd-Cu membranes represented a breakthrough in composite Pd membrane synthesis. The thermal stresses arising from the mismatch in the coefficient of thermal expansion between the Pd film and the support were determined by means of x-ray diffraction. The results indicated that the release of stresses began to occur at temperatures close to 400ºC. Also, the release of stresses took place with a visible sintering of Pd clusters within the thin Pd film. The stresses due to the absorption of H2 were also studied and modeled. It was estimated that the maximum compressive stress under which these composite Pd membranes were characterized was equal to 260 MPa.

This work provides a detailed characterization study on H2S poisoning of Pd and Pd/Au alloy composite membranes to obtain fundamental understandings of sulfur poisoning phenomena and preparation of sulfur tolerant membranes. The enhancement of the sulfur tolerance by alloying Pd with Au has been confirmed by both permeation test and microstructure analysis (SEM and XRD). While pure Pd membranes exhibited the permeance decline in the presence of H2S due to both sulfur adsorption and bulk Pd4S formation, Pd/Au alloy membranes showed the permeance loss merely resulted from the surface sulfur adsorption without bulk sulfide formation up to 55 ppm H2S. The XPS study confirmed that the H2S adsorption on the Pd/Au alloy surfaces was dissociative, and both surface Au and Pd sulfides were formed with the preferential Au-S bonding. The adsorption type of sulfur on the Pd/Au alloy surfaces was monolayer with a limited coverage, which increased with decreasing temperature. The permeance loss of Pd/Au membranes was essentially fully recoverable in H2, and the integrity of the membranes remained unaltered after the poisoning/recovery tests. Increasing Au composition in the Pd/Au membranes increased the sulfur tolerance. A Pd/Au alloy membrane of 16.7 wt% Au exhibited a permeance over 50% of its original value in the presence of 5 ppm H2S at 400°C, while a Pd membrane showed 85% permeance loss. The Pd/Au alloy membranes were fabricated by the Au displacement deposition, which had an empirical reaction order of 3.2 determined by the AAS. The HT-XRD study verified that the formed Pd/Au alloy layers were thermally stable up to 500°C.

ABSTRACT Recent global interests in developing hydrogen economy generate substantial research and development for hydrogen production worldwide. Pd membranes are especially suited for high temperature hydrogen separation and membrane reactor applications. Alloying Pd with Ag not only suppresses hydrogen embrittlement, but also increases the permeability of the alloy membrane. The main objective of this work was to carry out fundamental studies to understand the properties of the porous stainless steel (PSS), morphologies of Pd and Ag deposits on PSS, and the structural changes of the membrane layer upon heat treatment. Both coating and diffusion and co-plating techniques were employed in the study. The Pd-Ag membranes that had sandwiched Ag layers suffered from very low selectivity due to the voids formed because of high diffusion rate of Ag. Alloy membranes with high selectivity can be prepared by applying intermediate annealing after each Ag deposition. On the other hand, the homogeneity of the alloy depended very much on the thickness of the deposited layers and annealing temperature and time. A stable co-plating bath was developed to co-plate Pd and Ag simultaneously. Pd-Ag membranes were prepared from co-plating bath using ultrasound to accelerate the plating rate.

A oxidação seletiva de alcoóis é importante para a síntese de intermediários e produtos químicos utilizados na fabricação de inúmeros materiais. O interesse em novos métodos catalíticos para a oxidação de moléculas orgânicas tem aumentado juntamente com a busca por processos ecossustentáveis. Neste trabalho foram estudadas diferentes estratégias de deposição de nanopartículas bimetálicas de ouro e paládio sobre um suporte magnético, constituído por nanopartículas de magnetita revestidas por sílica. Dois métodos de síntese de nanopartículas bimetálicas de ouro e paládio foram estudados: (I) a síntese de nanopartículas bimetálicas formadas por ligas de ouro e paládio, seguida pela deposição dessas nanopartículas pré-formadas sobre o suporte catalítico e (II) a síntese de nanopartículas bimetálicas do tipo core-shell, por meio da redução sucessiva de ouro e paládio sobre o suporte catalítico. A morfologia dos catalisadores foi determinada por diferentes técnicas de microscopia eletrônica. O desempenho catalítico das nanopartículas bimetálicas foi estudado na reação modelo de oxidação de álcool benzílico com oxigênio. Os catalisadores mostraram-se ativos e seletivos para benzaldeído sem a necessidade do uso de solvente ou base e, após separação magnética, foram usados em sucessivas reações com desempenho satisfatório / The selective oxidation of alcohols is important for the synthesis of intermediates and chemical products used for the production of many materials. The interest in new catalytic methods for the oxidation of organic molecules has increased along with the search for eco-sustainable processes. In this thesis we studied different deposition strategies of gold and palladium bimetallic nanoparticles on a magnetic support, comprised by silica coated magnetite nanoparticles. Two synthetic methods for the preparation of supported gold and palladium bimetallic nanoparticles were investigated: (I) the synthesis of gold and palladium bimetallic nanoparticles, followed by deposition of the preformed nanoparticles on the catalytic support and (II) the synthesis of core-shell bimetallic nanoparticles, through the sequential reduction of gold and palladium on the catalytic support. The catalysts\' morphology was investigated by electronic microscopy techniques. The catalytic behavior of the bimetallic nanoparticles was investigated in the model reaction of the oxidation of benzyl alcohol with oxygen. The catalysts were active and selective for benzaldehyde without the use of solvent and base and, after magnetic separation, were able to be reused in successive runs and showed a good performance.

Palladium membranes have been identified as the membranes of choice in hydrogen separation and purification processes due to their infinite selectivity to hydrogen when defect free. Despite their potentials in hydrogen processes, palladium membranes pose challenges in terms of cost and embritllement which occurs when palladium comes in contact with hydrogen at temperatures below 573 K. The challenges posed by palladium membranes have encouraged research into nonpalladium based membranes such as Silica and Alumina. This thesis investigates hydrogen permeation and separation in palladium membranes and also the use of nonpalladium membranes, Silica and Alumina membranes in hydrogen permeation. In this study, hydrogen permeation behavior was investigated for 3 types of membranes, Palladium, Silica and Alumina. Thin palladium films were deposited onto a 30 nm porous ceramic alumina support using both conventional and modified electroless plating methods. The hydrogen separation and purification behavior of the membranes were investigated including the effect of annealing at higher temperatures. Gas permeation through Silica and Alumina membranes was investigated for 5 single gases including hydrogen. The Silica and Alumina membranes were fabricated using the dip coating method and their hydrogen permeation behavior of investigated at different coatings. A thin Palladium (Pd1) membrane with a thickness of 2 μm was prepared over porous ceramic alumina support using the electroless plating method and a maximum hydrogen flux of 80.4 cm3 cm-2 min-1 was observed at 873 K and 0.4 bar after annealing the membrane. The hydrogen flux increased to 94.5 cm3 cm-2 min-1 at same temperature and pressure for the Palladium membrane (Pd2) prepared using the modified electroless plating method. The hydrogen flux increased to 98.1 cm3 cm2 min-1 for the palladium/silver (Pd/Ag) membrane prepared using the codeposition electroless plating method and the PdAg membrane avoided the hydrogen embrittlement at low temperature. Hydrogen purity for the membrane was also investigated for a reformate gas mixture and a maximum hydrogen purity of 99.93% was observed at 873 K and 0.4 bar. The hydrogen purity was observed to increase as a result of the addition of sulphur which surpresses the inhibition effect of the carbon monoxide in the reformate gas mixture. The presence of CO and CO2 was observed to lead to an increase of the exponential factor n above 0.5 as a result of the inhibiting effect of these compounds on hydrogen permeation. The value of the exponential factor n depicting the rate limiting step to hydrogen permeation in the palladium and palladium-alloy membranes was also investigated. Deviations from Sievert’s law were observed from the Palladium membranes inverstigated in this work. In single gas hydrogen permeation investigation for the Pd1 membrane prepared using the conventional electroless plating method, the value of the exponential factor n = 0.5 in accordance with Sievert’s law. However, for the mixed gas hydrogen separation investigation n=0.62 at 573 K which decreased to 0.55 when the membrane was annealed at 873 K. For the Pd2 membrane prepared using the modified elctroless plating method, n=1 at 573 K but the value decreased to 0.76 for the mixed gas hydrogen separation investigation at same temperature which depicts a deviation from Sievert’s law. In all the investigations carried out for the Pd3 palladium alloy membrane prepared using the co-deposition Pd/Ag electroless plating method at same conditions with the Pd1 and Pd2 membranes, n=0.5 in accordance with Sievert’s law. For the Nonpalladium based Silica and ceramic Alumina membranes, investigations were carried out for hydrogen permeation and 5 other single gases; He, CO2, CH4, N2 and Ar. For the Silica membranes, a maximum hydrogen permeance of 3.12-7 x 10 mol m-2 s-1 Pa-1 at 573 K and 0.4 bar was observed which increased to 4.05 x 10-7 mol m-2 s-1 Pa-1 at 573 K and 0.4 when the membrane was modified with Boehmite sol prior to deposition of the Silica layer. The permeance for hydrogen and the 5 single gases was investigated for the alumina membrane at 5 successive coatings. It was observed that the commercial alumina membrane displayed a maximum hydrogen permeance of 9.72 x 10-7 mol m-2 s-1 Pa-1 at 573 K and 0.4 bar which increased to 9.85 x 10-7 mol m-2 s-1 Pa-1 at same temperature and pressure when the membrane was modified with Boehmite sol.

A catálise é a chave para o desenvolvimento de processos químicos sustentáveis e, portanto, o preparo de catalisadores que sejam mais ativos e seletivos é sempre uma questão atual. Utilizando as propriedades diferenciadas de nanopartículas metálicas e as vantagens de separação e estabilidade de um catalisador heterogêneo, este trabalho descreve novos catalisadores de níquel e de níquel-paládio em escala nanométrica, suportados em sólidos magnéticos, para hidrogenação de olefinas. Os catalisadores de níquel apresentados neste trabalho foram preparados pela metodologia de decomposição do precursor organometálico Ni(COD)2 (COD = ciclo-octadieno), que apresenta o metal em estado de oxidação zero e permite a formação de nanopartículas metálicas após a hidrogenação das ligações C=C do ligante. O catalisador de níquel aprimorado, composto por nanoagregados de níquel, apresentou atividade superior a qualquer outro reportado na literatura para o mesmo tipo de reação. Além disso, mesmo com a facilidade de oxidação do Ni(0), este novo catalisador teve apenas sua superfície oxidada a NiO ao ser armazenado em ar. O NiO formado foi facilmente reduzido por hidrogênio a Ni(0) em condições brandas (75°C) se comparado ao NiO bulk (270-520°C). Os catalisadores bimetálicos de níquel-paládio foram sintetizados por uma reação de substituição galvânica com Pd(OAc)2 em um catalisador de Ni(0) suportado e por decomposição simultânea dos precursores organometálicos Ni(COD)2 e Pd2(dba)3 (dba = dibenzilideno acetona) em diferentes frações molares. Os catalisadores bimetálicos preparados por substituição galvânica não resultaram em estruturas core(Ni)-shell(Pd) como o esperado, mas sim em nanopartículas de paládio depositadas sobre os nanoagregados de níquel. Os catalisadores preparados por decomposição simultânea dos complexos organometálicos, seja pela decomposição direta dos precursores sobre o suporte magnético ou pela impregnação de nanopartículas coloidais previamente formadas, resultaram tanto em nanoligas quanto em nanopartículas com segregação de níquel na superfície de acordo com as proporções empregadas dos dois metais. Todos os métodos explorados possibilitaram a obtenção de catalisadores bimetálicos ativos na hidrogenação do cicloexeno, sendo que o catalisador composto por 1,3% em massa de Ni e 0,017% em massa de Pd, obtido por uma reação de substituição galvânica, foi o catalisador que atingiu a maior atividade na reação de hidrogenação do cicloexeno. O uso de precursores organometálicos para a síntese de nanopartículas suportadas de níquel e níquel-paládio se mostrou um método eficiente para a obtenção de catalisadores com atividade diferenciada. A separação magnética, método empregado para a separação e recuperação dos catalisadores de níquel, permitiu o fácil manuseio e evitou a exposição ao ar e oxidação dos catalisadores, prolongando sua vida útil. / Catalysis is the key for the development of sustainable chemical processes, and consequently, the preparation of active and selective catalysts is always a current issue. Using the unique properties of metal nanoparticles and the advantages of separation and stability of heterogeneous catalysts, this Thesis describes new nanometric nickel and nickel-palladium catalysts, supported on magnetic solids, for hydrogenation of olefins. The nickel catalysts described in this Thesis were synthesized by the decomposition of the organometallic precursor Ni(COD)2 (COD = 1,5-cyclooctadiene), which contains zerovalent nickel, and allows the formation of metal nanoparticles after the hydrogenation of the C=C bonds of the ligand. The optimized nickel catalyst, composed by nickel nanoaggregates, showed superior activity when compared to any other catalyst reported in the literature for the same kind of reaction. Even with the propensity of oxidation of Ni(0), this new catalyst had only the surface oxidized when exposed to air. The fine NiO shell formed was easily reduced to Ni(0) with hydrogen under mild conditions (75°C) when compared to NiO bulk (270-520 °C). The bimetallic nickel-palladium catalysts were synthesized by the galvanic replacement reaction of Pd(OAc)2 and a supported Ni(0) catalyst and by the simultaneous decomposition of the organometallic precursors Ni(COD)2 and Pd2(dba)3 (dba = dibenzylidene acetone) in different molar ratios. The bimetallic catalysts obtained by the galvanic replacement reaction were not formed by core(Ni)-shell(Pd) structures as expected, but they were formed by palladium nanoparticles deposited over the nickel nanoaggregates. The catalysts obtained by simultaneous decomposition of the organometallic complexes, either by the decomposition of the precursors directly over the support or by the impregnation of pre-synthesized nanoparticles, resulted in both nanoalloys and nanoparticles with nickel segregation on the surface, depending of the ratio between the two metals. All methods of preparation of the bimetallic catalysts explored in this study allowed the formation of very active catalysts. On top of that is the catalyst with 1,3 wt% of Ni and 0,017 wt% of Pd, obtained by the galvanic replacement reaction, which achieved the highest activity in the hydrogenation of cyclohexene. The organometallic approach for the synthesis of supported nickel and nickel-palladium nanoparticles is an efficient method to obtain catalysts with outstanding activities. The magnetic separation method employed for separation and recovery of the catalysts containing nickel allows the easy handling and prevents exposure to air and undesirable oxidation of catalysts, extending their lifetimes.

O estudo da eletrorredução de nitrato teve início na década de 70 e o catalisador mais estudado para esta reação é a platina, embora, para fins ambientais, o paládio seja mais indicado por apresentar maior seletividade para a formação de N2. Neste trabalho a eletrorredução de nitrato foi estudada em eletrodos de Pd, Pt e Pt com camadas de Pd. Primeiramente foi feito um estudo com nanopartículas de Pt, Pd e Pt com uma monocamada de Pd dispersas em carbono, mas a dificuldade em sugerir um mecanismo nesse tipo de eletrodo levou à utilização de eletrodos monocristalinos, que tem a superfície conhecida. Foram feitos estudos eletroquímicos em eletrodos de Pd(hkl) que não apresentaram grande atividade frente a redução de nitrato, uma vez que o potencial inferior não foi menor que 0,2 V, para evitar excessiva absorção de hidrogênio. Em Pt(hkl) os resultados obtidos estão de acordo com os publicados na literatura e comprovam a sugestão de que o produto final é NH4 +. Para observar a atividade do paládio frente à reação, foram utilizados eletrodos monocristalinos de Pt (hkl) com multicamadas de paládio, que se comportam como monocristais de paládio, mas sem absorver hidrogênio. Nestes materiais foram feitos estudos eletroquímicos e de FTIR. Verificou-se que a atividade dos eletrodos depende da estrutura e do eletrólito, mas a definição de um mecanismo fica comprometida pelo fato do solvente (H2O ou D2O) favorecer a formação de um produto ou outro, embora tenha sido comprovada a formação de NO(ads) em todos os casos. / The study of nitrate electro-reduction started in the 70s and the most studied catalyst for this reaction is platinum, however because of environmental reasons, palladium is more indicated since it has higher selectivity for the formation of N2. In this work the nitrate electro-reduction was studied on Pd, Pt and Pt modified with Pd layers. First, a study involving carbon-supported nanoparticles of Pt, Pd and Pt with a Pd single layer was carried out, but the difficulty suggesting a reaction mechanism for this kind of electrode has driven to the use of single crystals, because of their defined surface structure. So, electrochemical studies were made on Pd(hkl) electrodes, but these materials did not present great activity for nitrate reduction, because the electrode potential was not decreased below 0,2 V, to avoid large hydrogen absorption. For Pt(hkl) the obtained results are in agreement with those published previously, confirming the suggestion that NH4 + is the final product. To characterize the palladium activity for this reaction, Pt(hkl) single crystals with palladium multilayer were employed, because these materials have a behavior similar to palladium single crystals, but without hydrogen absorption. Electrochemical and FTIR experiments were done to characterize the i vs E responses and to follow the reaction intermediates and products. It was seen that the activity depends on the electrode structure and the electrolyte, but the mechanism determination pointed to a uncertain situation where the solvent (H2O or D2O) drives to the formation of one product or another, although the production of NO(ads) has been observed in all cases.

A reação de hidrodescloração (HDC) tem recebido grande atenção como um método de tratamento de resíduos orgânicos mais eficiente que a incineração, especialmente para compostos aromáticos clorados, pois evita a formação de espécies potencialmente tóxicas, como furanos e dioxinas. O presente estudo tem como objetivo a preparação de nanocatalisadores de paládio suportados para a reação de HDC. Buscando facilitar a separação do catalisador do meio reacional e sua reutilização em sucessivas reações foram empregados suportes sólidos com propriedades magnéticas e desenvolvidas estratégias para a imobilização de nanopartículas metálicas nestes suportes. Para isso foi preparado um suporte composto por nanopartículas magnéticas revestidas por uma camada protetora de sílica densa. Os núcleos magnéticos foram preparados pelo método da coprecipitação e em seguida recobertos com sílica pelo método da microemulssão reversa. A estratégia usada para a preparação das nanopartículas de paládio suportadas foi a funcionalização da superfície do suporte, a imobilização do precursor de Pd(II) e a subseqüente redução do metal por hidrogênio em condições brandas. Como resultado foram obtidas nanopartículas de Pd bem dispersas no suporte funcionalizado, sendo que o tamanho de partícula foi dependente do grupo funcional presente no suporte. Suportes funcionalizados com amina e etilenodiamina resultaram em nanopartículas de Pd com diâmetro de 6,4 ± 1,4 nm e 1,3 ± 0,3 nm, respectivamente. Em testes catalíticos de hidrogenação de cicloexeno o catalisador amino-funcionalizado apresentou melhor desempenho com relação ao funcionalizado com etilenodiamina. Para a reação de HDC foram estudadas as melhores condições reacionais para obtenção de uma maior taxa de conversão do substrato, preservando a estrutura do catalisador. Clorobenzeno foi empregado como substrato modelo. Os testes preliminares em diferentes bases e solventes sugeriram que a trietilamina em isopropanol é o melhor meio por resultar em boa conversão de substrato sem alteração significativa na estrutura do suporte após a reação. O lixiviamento dos metais, nos casos estudados, não foi relevante em relação à quantidade de metal presente inicialmente no suporte. As reações em tampão acetato e carbonato também resultaram em boas taxas de conversão, sendo que no teste de reuso o catalisador começou a diminuir a atividade apenas no quarto reciclo. O meio tamponado é uma alternativa interessante por ser menos agressiva, mas ainda muito pouco explorada para a reação de HDC. / The hydrodechlorination reaction (HDC) has received great attention as an alternative treatment of organic residues, which is more efficient than incineration, especially for aromatic organic compounds, and avoids the formation of toxic species, such as furans and dioxins. The present study is focused on the preparation of supported palladium nanocatalysts for HDC. In order to facilitate the separation of the catalyst from the reaction medium and its reuse in successive reactions solid supports with magnetic properties were employed and strategies for the immobilization of metal nanoparticles on the surfaces of these supports were developed. For this purpose a catalyst support comprised of magnetic nanoparticles coated by a protective dense silica layer was developed. The magnetic nuclei were prepared by the co-precipitation method followed by silica coating by a reverse microemulsion. The strategy used for the preparation of supported Pd nanoparticles was first the functionalization of the support surface, immobilization of Pd (II) precursors and than metal reduction by hydrogen under mild conditions. As a result, Pd nanoparticles well-dispersed on the functionalized support were obtained, although the size of the Pd nanoparticles was tuned by the ligand grafted on the support surface. Amine and ethylenediamine functionalized supports formed Pd nanoparticles of 6,4 ± 1,4 nm and 1,3 ± 0,3 nm, respectively. In the catalytic tests of ciclohexene hydrogenation, the amino-functionalized catalyst showed the best performance compared to the ethylenodiamine functionalized support. For the HDC reaction, the most favorable reaction conditions to reach the highest substract conversion rates while preserving the catalyst structure were studied. Chlorobenzene was selected as substrate for the HDC experiments. Initial tests using different bases and solvents suggested that triethylamine in isopropanol are the best conditions for good conversion rates without meaningful change in the support structure after reaction. The metal leaching was negligible in all reactions studied with respect to the initial metal loading. The HDC reactions in acetate and carbonate buffer solutions also resulted in good conversion rates, while the catalyst activity began to decrease only in the fourth cycle. Buffer medium is an interesting less aggressive alternative for HDC reactions, but still very little exploited

Um dos problemas que contribui para a diminuição do desempenho de células a combustível de membrana trocadora de prótons é o cruzamento de combustível do compartimento anódico para o catódico. Um método de avaliar o contato do combustível anódico com o cátodo consiste da exposição do catalisador oxidado à molécula de interesse, em condições de circuito aberto. Do ponto de vista fundamental, a análise desses transientes de circuito aberto pode fornecer informações importantes acerca do mecanismo reacional associado. São apresentados nessa Dissertação, resultados da interação entre metanol ou etanol e superfícies oxidadas de paládio, em meio alcalino. Entre os parâmetros investigados, maior ênfase foi dada ao efeito da natureza do cátion alcalino presente no eletrólito. Observou-se que a presença de Li+ ou K+ no eletrólito influencia desde a quantidade de óxido formada, até o tempo requerido para a redução desses óxidos pela molécula orgânico dissolvida. A natureza do composto orgânico presente também exerce diferenças consideráveis no transiente de circuito aberto. Finalmente, foram propostos esquemas reacionais para representar a interação de circuito aberto com o metanol e etanol. Em ambos os casos, os mecanismos incluem a produção auto-catalítica de sítios livres de paládio. / One of the problems that contributes to the performance decreasing observed in proton exchange membranes is the fuel crossover from the anodic to the cathodic compartment. A method to evaluate the contact of the anodic fuel with the cathode consists of exposing the oxidized catalyst to the molecule of interest under open circuit conditions. From the fundamental point of view, the analysis of the open circuit transients may provide relevant information on the associated reaction mechanism. In this Dissertation, results of the open circuit interaction between methanol or ethanol with oxidized palladium surfaces, in alkaline media, are presented. Among the distinct investigated parameters, emphasis will be put on the effect of the nature of the alkali cation present in the electrolyte. It was observed that the presence of Li+ or K+ influences several aspects, from the amount of oxide formed to the time needed for the reduction of the oxide by the dissolved organic molecules. The identity of the dissolved organic molecule also exerts marked influence on the open circuit transients. Finally, two reaction schemes for the open circuit interaction with methanol or ethanol were proposed. In both cases, the mechanisms include the auto-catalytic production of free palladium sites.