Gbmm Guy Marin

An investigation on the reaction mechanism for the partial oxidation of methane to synthesis gas over platinum

The partial oxidation of methane to synthesis gas has been investigated by admitting pulses of pure methane, pure oxygen and mixtures of methane and oxygen to platinum sponge at temperatures ranging from 973 to 1073 K. On reduced platinum the decomposition of methane results in the formation of surface carbon and hydrogen. No deposition of carbon occurs during the interaction of methane with a partly oxidised catalyst. Oxygen is present in three different forms under the conditions studied: platinum oxide, dissolved oxygen and chemisorbed oxygen species. Carbon monoxide and hydrogen are produced directly from methane via oxygen present as platinum oxide. Activation of methane involving dissolved oxygen provides a parallel route to carbon dioxide and water. Both platinum oxide and chemisorbed oxygen species are involved in the oxidation of carbon monoxide and hydrogen. In the presence of both methane and dioxygen at a stoichiometric feed ratio the dominant pathways are the direct formation of CO and H2 followed by their consecutive oxidation. A Mars-van Krevelen redox cycle is postulated for the partial oxidation of methane: the oxidation of methane is accompanied by the reduction of platinum oxide, which is reoxidised by incorporation of dioxygen into the catalyst.

Platinum deactivation: in situ EXAFS during aqueous alcohol oxidation reaction

With a new set‐up for in situ EXAFS spectroscopy the state of a carbon‐supported platinum catalyst during aqueous alcohol oxidation has been observed. The catalyst deactivation during platinum‐catalysed cyclohexanol oxidation is caused by platinum surface oxide formation. The detected Pt–O co‐ordination at 2.10 Å during exposure to nitrogen‐saturated cyclohexanol solution is different from what is observed for the pure oxidised platinum surface (2.06 Å).

Kinetic modeling for wet air oxidation of formic acid on a carbon supported platinum catalyst

A study of the intrinsic kinetics of the oxidation of formic acid over a 1% Pt/C catalyst has been performed, at oxygen concentrations in water varying from 0.6 to 2.8 mol m−3 and at formic acid concentrations varying from 30 to 400 mol m−3. Experiments were performed in a continuous-flow stirred slurry reactor. To avoid mass transfer limitation of oxygen, a temperature interval from 282 to 293 K and a total pressure in the reactor of 0.6 MPa has been used. The pH interval of the experiments ranged from 1.1 to 13.0. Single response non-linear regression has been used to fit the parameters for a kinetic model. The proposed model can predict the steady-state disappearance rate of formic acid, as a function of the oxygen concentration, the formic acid concentration, the temperature and of the pH. Electrochemical experiments have been carried out for a better understanding of the pH effect on the formic acid oxidation.

Selective oxidation of methyl alpha-D-glucoside on carbon supported platinum III. Catalyst deactivation

Abstract Platinum supported on activated carbon and platinum supported on graphite were characterized before, during and after oxidation of methyl α- d -glucoside in water. Ex situ information was obtained by X-ray photoelectron spectroscopy, cyclic voltammetry, scanning transmission electron microscopy and carbon monoxide chemisorption. In situ information was obtained by cyclic voltammetry. Furthermore, the open circuit potential of a platinized platinum foil was monitored during reaction. A distinction could be made between a reversible deactivation, occurring on a time scale of 10 ks and an irreversible deactivation, occurring on a time scale of 100 ks. The reversible deactivation is attributed to a slowly increasing oxygen surface coverage rather than to the formation of platinum oxide. Chemisorption takes place beyond the steady-state degree of oxygen coverage expected from the intrinsic initial reaction kinetics and causes a reversible decrease of the reaction rate. The irreversible deactivation is caused by the dissolution and subsequent redeposition of platinum. The preferential dissolution of small platinum particles and redeposition on larger ones leads to a decrease of the fraction of exposed platinum atoms from 0.7 to 0.5. The effect of the irreversible decrease of the platinum surface area on the specific reaction rate is strongly attenuated by an antipathetic structure sensitivity. An increase of the average platinum particle diameter results in an increase of the turnover frequency.

Fixed bed reactor design for gas phase chain reactions catalysed by solids: the oxidative coupling of methane

Abstract A multitubular catalytic fixed bed reactor for the oxidative coupling of methane is designed. A heterogeneous two-dimensional reactor model taking into account axial dispersion of mass and enthalpy is applied. A reaction network consisting of 36 gas phase radical reactions and 10 catalytic reactions, based upon experiments with an Sn/Li/MgO catalyst at atmospheric pressure, temperatures of 923–1,023 K and methane-to-oxygen ratios of 2–12, is used. Intrapellet concentration profiles of molecules and, even more so, of radicals are shown to affect the selectivity towards ethane and ethylene. The importance of homogeneous reactions in the void space between the catalyst pellets is highlighted. The reactor dimensions are determined by heat removal considerations.

Partial oxidation of methane to synthesis gas over Rh/alpha-Al2O3 at high temperatures

The partial oxidation of methane to synthesis gas has been studied in a continuous flow reactor using a Rh/α-Al2O3 catalyst under conditions as close as possible to those industrially relevant: pressures up to 800 kPa and temperatures higher than 1274 K in order to avoid the formation of carbon and to obtain high equilibrium selectivities to CO and H2. Intrinsic kinetic data were obtained when the feed was diluted with helium. Gas-phase reactions were found to occur at 500 kPa when the feed was not diluted. A reaction network has been derived from experimental results in which oxygen conversions range from 0 to 1. CO2, C2H6 and H2O are the primary products. C2H4 is formed by oxidative dehydrogenation of C2H6. CO and H2 are formed by reforming of CH4 by CO2 and H2O; an additional direct route to CO and H2 at low oxygen conversions cannot be excluded. The catalyst appears to be present in two states, the transition being at an oxygen conversion of 0.4 under the conditions used. The support probably enhances oxidation reactions by reverse spillover of oxygen or hydroxyl species onto rhodium. The support as such behaves similarly to the catalyst at low oxygen conversions, but shows no reforming activity.

On-line characterization by EXAFS of tin promoted platinum graphite catalysts in the aqueous phase

The structure of tin promoted graphite supported platinum catalysts has been studied with extended X-ray absorption fine structure spectroscopy (EXAFS). A newly developed EXAFS cell allows on-line characterization avoiding contact to ambient or drying. Hereto catalyst samples are transferred from a slurry reactor to the EXAFS cell forming a ‘‘bed’’ of catalyst particles in the EXAFS cell. The cell design was based on considerations concerning possible mass transport limitations while performing reactions in the liquid phase. The structures of the tin promoted platinum catalysts were investigated directly after preparation, drying, treatments with hydrogen (363 K) and oxygen (RT) in aqueous phase and a hydrogen gas treatment at 573 K at both the Pt LIII and the Sn K-edge. After preparation, under aqueous hydrogen, reduced platinum can be detected with three coordinations: Pt‐Pt, Pt‐C and Pt‐Sn. Tin appears to be partly oxidic showing a Sn‐O and a Sn‐Pt coordination. A treatment with aqueous oxygen or exposure to ambient leads to oxidized platinum and tin. At the Pt LIII-edge only a Pt‐Pt and Pt‐O coordination for platinum are detected. At the Sn K-edge tin has only a Sn‐O coordination. An aqueous treatment with hydrogen at 363 K reduces platinum showing, however, different coordination numbers for the Pt‐Pt and Pt‐Sn coordination. Tin only shows a Sn‐O coordination. A treatment with hydrogen at 573 K reduces both the platinum and the tin. Platinum shows a Pt‐Pt, Pt‐C and Pt‐Sn coordination. Tin shows a Sn‐Pt and Sn‐O coordination indicating tin deposition on the platinum, tin being bonded via oxygen to the graphite support. Reductive treatments in the aqueous phase appear to reduce platinum and only the tin deposited on the platinum. The effects of drying and consecutive reductive treatments could only be studied since the developed EXAFS cell allowed catalyst preparation and treatments avoiding contact to ambient. # 1998 Elsevier Science B.V.

A quantitative analysis of transient kinetic experiments: the oxidation of CO by O2 over Pt

Abstract Transient kinetic experiments allow to assess quantitatively the rate coefficients of individual elementary steps. Experiments were performed in the temperature range 300–400 K with temporal analysis of products (TAP) for CO oxidation over polycrystalline platinum, initially fully precovered with O adatoms, at varying degrees of CO and O coverage. Experimental evidence for the existence of islands was found. The results were described with an island model, that accounts for CO chemisorption from a precursor state, followed by either CO incorporation into CO islands or by a surface reaction with O adatoms from adjacent O islands. CO chemisorption was not activated and its rate coefficient was estimated as 40 s −1 , while the surface reaction was relatively fast. Indications for the size of the CO and O islands were found.

Competing reactions in three-way catalytic converters: modelling of the NOx conversion maximum in the light-off curves under net oxidising conditions.

Automobile exhaust gas conversion was simulated with a reactor model based on first principles. The monolithic reactor was modelled as adiabatically operating with a uniform flow distribution over the channels and with constant heat and mass transfer coefficients. The kinetic rate equations in the model were constructed from elementary step kinetics of the individual global reactions. The model predictions for light-off curves compare quite well with experimental data from the literature. Light-off is in the sequence hydrogen, carbon monoxide, propene, but the light-off temperatures do not differ very much. The nitric oxide conversion as function of the reactor feed temperature passes through a maximum at high hydrocarbon conversion, but does not reach the level of 50%, defined as light-off. Hydrogen is the major oxygen consumer in the front part of the reactor, while a slow reaction proceeds between CO and NO. At sufficiently high feed temperatures thermal reactor ignition occurs when the degree of CO surface coverage drops considerably. The corresponding increase of the O adatoms coverage causes increased reaction rates and a sudden temperature rise. Beyond the ignition point oxygen is mainly consumed by propene and by unconverted carbon monoxide. NO is reduced by unconverted hydrogen, while NO reduction by the hydrocarbon is not significant. Catalysts capable of increasing the NO surface coverage or the NO dissociation, or leading to a lower oxygen sticking coefficient would show a higher NO conversion maximum. The results indicate that well-known reaction mechanisms are capable to describe the behaviour of automotive exhaust gas converters, if mutual interactions of gaseous components and surface species are taken into account via elementary step kinetics.

Selective oxidation of methyl alpha-D-glucoside on carbon supported platinum II. Assessment of the Arrhenius and Langmuir parameters

Abstract In order to estimate the Arrhenius and Langmuir parameters of the selective oxidation of methyl α-D-glucoside on a platinum catalyst, the temperature was varied from 293 to 333 K. The initial reaction kinetics can be described adequately over this temperature range by a Langmuir-Hinshelwood rate equation which is based on two parallel reaction paths: one involving adsorbed methyl α-D-glucoside and dominating at low pH, the other involving the adsorbed methyl α-D-glucoside anion and dominating at high pH. Both reaction paths contain a rate-determining step consisting of a surface reaction involving the corresponding methyl α-D-glucoside species and chemisorbed oxygen. The apparent activation energies amount to 50.5±6.4 kJ mol −1 and 110.5± 10.6 kJ mol −1 . The assessment of the Arrhenius parameter estimates within the framework of the transition state theory showed that these rate-determining steps lead to physically meaningful estimates. Langmuir adsorption coefficients account for the reversible dissociative chemisorption of oxygen and for the reversible associative adsorption of methyl α-D-glucoside and methyl α-D-glucoside anion. A standard adsorption enthalpy for oxygen of −60.0 ± 10.5 kJ mol −1 , for methyl α-D-glucoside of −17.6±8.3 kJ mol −1 and for the methyl α-D-glucoside anion of − 61± 54 kJ mol −1 were estimated. During the oxidation at a pH of eight oxygen adatoms are the most abundant surface species, followed by adsorbed methyl α-D-glucoside. With increasing temperature the degree of oxygen coverage decreases from 0.8 to 0.6, while the degree of coverage with methyl α-D-glucoside increases from 0.08 to 0.13. At a pH of 10 the surface coverage with oxygen decreases from 0.6 at 293 K to 0.5 at 333 K and the surface coverage with methyl α-D-glucoside anion from 0.5 to 0.3.