Search

Detoxification of water containing chlorinated aromatic compounds is performed in a single fixed bed using a periodic sequence of three steps: transient adsorption, catalytic hydrogenation, and thermal regeneration. The solid is both the adsorbent (activated carbon) and the catalyst (2.6% ruthenium loading). For chlorophenols, the end products are cyclohexanol and sodium chloride diluted in less than 2% of the amount of processed water. Adsorption is the critical step. The whole breakthrough curve is easily modeled. Conversely, the model fails to describe the early beginning which governs the purity of the effluent. The reaction step is performed under mild conditions (T < 353 K, P < 0.4 MPa) and is governed by the transient desorption of the pollutant, the catalytic reaction, and the mass-transfer processes. Thermal regeneration under hydrogen flow restores the adsorbent capacity and catalytic activity.

Catalytic hydrodechlorination and hydrodearomatisation of 13 different mono-, di-, tri-, tetra- and pentachlorophenols were carried out simultaneously under mild reaction conditions (298–353 K, 0.3–0.5 MPa). Experimental conditions suitable for complete conversion of the pollutants into nontoxic cyclohexanol and sodium chloride were first determined using a commercial Ru/C catalyst. A new supported ruthenium catalyst was then designed that possesses a carbonaceous support suitable to concentrate chlorophenol through adsorption. Using parachlorophenol as a model molecule, the influence of the base concentration in solution and the kinetics of the reaction under chemical regime were studied. A p-chlorophenol→phenol→cyclohexanol consecutive kinetic scheme was considered that drove to numerous possible kinetic models. A dynamic simulation software was used both for helping in model discrimination and for kinetic parameter estimation. The estimated activation energy values amount to ca. 45.0±2.7 and 33.2±2.4 kJ mol−1 for the hydrodechloration and the hydrodearomatisation steps, respectively.

The hydrogenation of acetophenone in a slurry airlift reactor was conducted over a wide range of operating conditions. An important deactivation of the catalyst was experimentally observed and an empirical deactivation law was optimized. The deactivation rate constant increases with the inlet concentration and gas flow rate and decreases with the catalyst loading. The product selectivity measured in the slurry airlift reactor did not fit well with the computed one. The computation was performed using the results of the kinetic and hydrodynamic studies. A sensitivity analysis pointed out that the mass transfer is not limited. To perform a good fit with the experiments it was necessary to change some rate constants. This readjustment was justified by the difference in catalyst pretreatment between the continuous and the semi-batch reactor where significant catalyst pretreatment effects were shown.

In a first step the intrinsic kinetics of acetophenone hydrogenation on a Rh/C catalyst was studied in a semi-batch reactor. Tiny catalyst grains of 17 μm in average diameter were used in order to avoid any intraparticle diffusion limitation. Experiments were performed over a wide range of operating conditions and various Langmuir–Hinshelwood kinetic equations were discriminated over the complete conversion range. The model based on nondissociative adsorption of hydrogen and noncompetitive adsorption of the organic species with the gas molecules was found to fit better all the experimental data. The relevance of its optimized parameters was then discussed. In a second step the intraparticle diffusion limitations were also studied in a semi-batch reactor by varying the particle size. Two models based on the complex kinetics previously established were developed, taking into account the catalyst grain size and shape. After an additional adjustment of one of the adsorption constants ratios, they were found to provide a good representation of the data in terms of activity and selectivity.

The catalytic hydrogenation of adiponitrile into hexamethylenediamine (HMD) under mild conditions (20 bar H2, 120 to 180°C) using monometallic Ni, Co and Ru and bimetallic NiCr, NiTi, NiCo and CoFe Ziegler-type catalysts is reported. It is found that water exerts a strong negative influence on the selectivity to the primary amine HMD whereas it has a spectacular beneficial effect on the catalyst activity, especially for Ni based catalysts. Adding a second metal (Cr, Ti) as a dopant for nickel has a beneficial effect on the selectivity in HMD but results in lower activities. The ratio Ni/M (M Cr, Ti) does not appear to be a key parameter in these systems as the doping effect on the HMD selectivity already appears at low dopant loading (atom.-%: 3% for Cr, 10% for Ti) and is constant up to 60%. The temperature has an influence on the HMD selectivity: the NiCr and NiTi catalysts show a peculiar behaviour, maintaining a high selectivity level (ca. 90%) over all the temperature range. This is not the case for the Co based catalysts.

The hydrogenation of acetone was investigated in basic aqueous solution with (Co, Cr, Cu, Mo, Ti) doped and undoped Raney nickel catalysts. The reaction was carried out under pressure in an autoclave equipped with a reference electrode. The hydrogen consumption and the electrode potential were measured during the course of the reaction. Addition of Cr, Mo or Ti improves the activity in acetone hydrogenation while the activity of Co or Cu promoted catalysts remains similar to that of the undoped nickel. The kinetics obeys a Langmuir-Hinshelwood mechanism. A mathematical model was applied which fits well the experimental kinetic data. It allows the computation of the rate constant and the adsorption equilibrium constants. The metallic catalyst particles behave like a dispersed electrode and an electrochemical double layer is formed at their surface. In the presence of hydrogen alone, the metal potential is determined by the Nernst law for the hydrogen electrode. In the presence of acetone, the measured potential goes to the positive region for several tenths millivolts. The extent of the initial potential shift measured when acetone is introduced in the reactor is experimentally correlated with the physicochemical characteristics of the catalysts and their activities for acetone hydrogenation in the liquid phase.

The hydrogenation of acetone was investigated in basic aqueous solutions with undoped and chromium-doped catalysts. The reaction was carried out under pressure in an autoclave equipped with a reference electrode. The consumption of hydrogen and the electrode potential were measured during the course of the reaction. A mathematical model was applied which fits the experimental kinetic data well. It allows the computation of the rate constant and the adsorption equilibrium constants. The kinetics obey a Langmuir-Hinshelwood mechanism with competitive adsorption. The metallic catalyst particles behave like a dispersed electrode and an electrochemical double layer is formed at their surface. In the presence of hydrogen alone, the metal potential obeys the Nernst law for the hydrogen electrode. During acetone hydrogenation, the double layer is modified and the measured potential goes to the positive region for several tens of millivolts, depending on whether the catalyst is doped or not. In all cases an experimental correlation was found between this experimental potential rise and the reaction rate.

We prepared chromium promoted Raney nickel by two methods: catalysts of type I, by alkali leaching of Ni40-xCrx Al60 alloys (0.5 ⩽x ⩽ 2.2 ) and by soaking the catalyst obtained from Ni40Al60 alloy in an aqueous chromium chloride solution, catalysts of type II. The structure of these catalysts was investigated by measuring the total and metallic surface area, by micro-analysis at fine scale with analytical transmission electron microscopy, X-ray diffraction and Auger spectroscopy. For all the catalysts the STEM/EDX analysis showed essentially two types of agglomerates, one of them being detected more frequently. This component presented itself very similar whether the dope was introduced by a metallurgical process or by a chemical way, but the more rarely observed agglomerates were different. From Auger spectroscopy it appeared that the surface Cr/Ni atomic ratios were higher in the catalysts of type II than in the catalysts of type I.

The hydrogenation of acetophenone has been investigated in cyclohexane, with chromium promoted Raney nickel. The preparation and the physico-chemical properties of these catalysts have been described in part I. The reaction was carried out in a static reactor and its progress was followed by measuring the concentration of the components in the liquid phase. The ratios kibM/kjbN where ki, kj are the rate constants and bM, bN are the adsorption equilibrium constants for two competitive parallel reactions or two consecutive reactions, have been computed and used as selectivity criteria. These selectivity criteria allow to compare the activity of the different sites of hydrogenation. In the presence of chromium the hydrogenation of the C O bond is favoured with respect to the two side reactions: the hydrogenation of the aromatic ring and the hydrogenolysis of the C-OH bond. Catalysts prepared by soaking Raney nickel in an aqueous chromium chloride solution gave the highest selectivity in 1- phenylethanol.

This paper is devoted to the fundamental study of the doping effect of iron metal added to Ni-Al alloy precursor of Raney nickel catalysts. Their activities and selectivities in the liquid phase were determined taking into account the effect of the basicity of the solvent. The experiments were performed on catalysts coming from homogeneous NiAl 3 and Ni 2 Al 3 phases. A synergetic effect between the base in the solution and the iron dope in the catalyst was demonstrated.

The influence of catalyst preparation on the structure, texture and the metallic A1 content is reviewed. A correlation between the A1 content and the saturation magnetization indicates an electronic interaction with a d band of nickel. Studies of adsorbed hydrogen show that there is a reversible and an irreversible adsorption with an electronic interaction with d bands of nickel as revealed by magnetic methods. Neutron inelastic scattering detects bridged and linear adsorbed H species. Two examples of hydrogenation reactions, with liquid acetone and gaseous benzene, are described and the mechanisms are discussed in relation to the kinetics and the adsorbed hydrogen species.

Hydrogen adsorption on MgO-supported platinum was studied by thermal desorption and infrared spectroscopy at 300 and 800 K. For both temperatures, reversibly and irreversibly adsorbed species have been detected. At 300 K, reversible adsorption leads to the appearance of infrared bands at 2120 and 2060 cm −1 , attributed to terminal Pt-H species. Irreversibly adsorbed hydrogen has been detected by thermal desorption, whereas no infrared band was detected in the spectral range 4000–4750 cm −1 for Pt/MgO sample. For hydrogen adsorption at 800 K, reversibly adsorbed hydrogen gave the same picture as for the 300 K adsorption. An additional form of irreversibly adsorbed hydrogen has been evidenced both by thermal desorption and infrared spectroscopy. This form corresponds to hydrogen strongly adsorbed on platinum and gives an infrared band at 950 cm −1 which is characteristic of an hydrogen atom in interaction with more than one platinum atom (multicentered) species.

Magnetic and volumetric measurements associated with neutron inelastic spectroscopy demonstrate that water is dissociatively adsorbed on nickel. In the case of Raney nickel, for low coverages, the oxygen is fixed on surface aluminium atoms and hydrogen occupies two nickel sites. At high coverage, the water molecule is fixed via oxygen bonding to the hydrogen covered nickel. On nickel prepared from its hydroxide, adsorption is reversible, the water molecule occupying six metal sites.

IR emission spectra obtained with a Fourier transform spectrometer at 110/sup 0/C revealed that pyridine adsorbed on platinum Y zeolite by coordination to the Na ion and by formation of pyridinium ions with bands at 755 and 685 cm/sup -//sup 1/; the formation of pyridinium ions suggested the presence of acidic OH groups formed during platinum reduction with hydrogen. Bands observed at 465 and 2070 cm/sup -//sup 1/ in the spectra of adsorbed carbon monoxide were attributed to linearly adsorbed species, and a band at 1850 cm/sup -//sup 1/ to carbon monoxide bonded to two platinum atoms. The experimental technique and its scope are discussed.

The pore texture of Raney-nickel was determined by impedance measurements carried out over a wide frequency range. The impedance obtained could be characterized by three resistances, R, Re and Rt and one capacitor C. R is the high frequency limit of the electrode impedance, and is equal to the electrolyte resistance between the reference capillary-tip and the upper-surface of catalyst layer. Re is the electrolyte resistance through the cavity of catalyst layer. Rt is linked in parallel with the capacitor C.C value allows to evaluate the surface area of catalyst in contact with electrolyte provided that the double layer capacitance of nickel electrode per unit surface is known. R value, measured with and without catalyst layer allows to estimate its thickness, hence the total pore volume when the density of catalyst-metal is known. The Raney-nickel has a double pore-structure: the one related to micro-pores inside catalyst grain and the other to a heaping of grains. If the volumes of these two types of pore are known, the pore radius of micro-pore can be evaluated. Therefore, impedance measurements determine, in situ, the pore texture of Raney-nickel in liquid phase. Results were in good agreement with those determined in gas phase by current methods.

Bimetallic Pt-Co and Pt-Ru catalysts were prepared by liquid phase impregnation of activated charcoal. The pretreatment of the charcoal, the impregnation by metal salts, the drying of the solid and the reduction by hydrogen are made following a precise procedure. These catalysts are selective and very active in the liquid phase hydrogenation of cinnamaldehyde to cinnamyl alcohol. The influence of the reaction products on kinetics, which is important for pure platinum, becomes negligible for bimetallic catalysts.

Hydrogen adsorption isotherms and thermodesorption curves have been established for a large variety of unsupported and supported catalysts. Upon adsorption at 300 K, 2 species, β1 and β2, were detected during desorption. The formation of β2 is activated but not that of β1. On Raney platinum a determination of the population of these two species was possible and we found (0.85 ± 0.1)× 1015 and (1.25 ± 0.1)× 1015 H atoms cm–2 for β1 and β2, respectively. The total volume adsorbed at 300 K corresponds to 1.4–1.8 H per superficial Pt atom, depending on the assumed surface structure of the metal. At 900 K, this volume is increased by 10%. On supported catalysts the adsorbed amount depends on the interaction between metal and support. For a weak interaction, as on Pt—SiO2 Eurocat, the total adsorbed amount corresponds to 1.8 hydrogen per surface metal atom under the two forms β1 and β2. For a strong interaction between metal and support, as on Pt-zeolite, the volume of hydrogen adsorbed at 300 K corresponds to only 1.1 H per surface atom. A new species, β3, is formed upon adsorption at 900 K which produces an increase of 30% in the total adsorbed volume. β3 is attributed to H atoms in a bridged position between platinum and oxygen. When oxygen is admitted at 300 K, the surface coverage is 0.8–0.9. Subsequent hydrogen adsorption on the oxygen covered surface is not disturbed by the formation of water.Excellent agreement was generally found with recent results obtained on single crystals.

The impedance diagram of fuel-cell hydrogen electrodes has been established at different temperatures and current densities. An equivalent electric circuit simulating the a.c. behaviour of electrodes is proposed which accounts fully for the experimental results. The computed adjustment of its elements allows the evaluation of the wet surface of catalyst, of the part played by diffusion in the reaction and of the catalytic activity of the electrode in the electro-oxidation of hydrogen. Using this technique the evaluation of the specific surface area of Raney nickel is possible by in situ measurements, which is of great use in probing catalyst aging. The method also gives a value for the electrocatalytic activity, which cannot be determined by any other method. One thus has the means to compare the activities of various catalysts.