Chlorine–nickel interactions in gas phase catalytic hydrodechlorination: catalyst deactivation and the nature of reactive hydrogen

Abstract

The gas phase hydrodechlorination of chlorobenzene and 3-chlorophenol (where 473 K⩽T⩽573 K) has been studied using a 1.5% w/w Ni/SiO2 catalyst which was also employed to promote the hydrogenation of benzene, cyclohexene and phenol. In the former two instances the catalyst was 100% selective in removing the chlorine substituent, leaving the aromatic ring intact. While the dechlorination of chlorobenzene readily attained steady state with no appreciable deactivation, the turnover of 3-chlorophenol to phenol was characterised by both a short and a long term loss of activity. Chlorine coverage of the catalyst surface under reaction conditions was probed indirectly by monitoring, via pH changes in an aqueous NaOH trap, HCl desorption after completion of the catalytic step. Contacting the catalyst with the chlorinated reactants was found to severely limit and, depending on the degree of contact, completely inhibit aromatic ring reduction although a high level of hydrodechlorination activity was maintained. Hydrogen temperature programmed desorption (TPD) reveals the existence of three forms of surface hydrogen which are tentatively assigned as: (i) hydrogen bound to the surface nickel; (ii) hydrogen at the nickel/silica interface; (iii) spillover hydrogen on the silica support. The effect of chlorine–nickel interactions on the resultant TPD profiles is presented and discussed. The (assigned) spillover hydrogen appears to be hydrogenolytic in nature and is responsible for promoting hydrodechlorination while the hydrogen that is taken to be chemisorbed on, and remains associated with, the surface nickel metal participates in aromatic hydrogenation. Hydrodechlorination proceeds via an electrophilic mechanism, possibly involving spillover hydronium ions. The experimental catalytic data are adequately represented by a kinetic model involving non-competitive adsorption between hydrogen and the chloroaromatic, where incoming chloroaromatic must displace the HCl that remains on the surface after the dechlorination step. Kinetic parameters extracted from the model reveal that chlorophenol has a higher affinity than chlorobenzene for the catalyst surface but the stronger interaction leads to a greater displacement of electron density at the metal site and this ultimately leads to catalyst deactivation. IntroductionThe reductive dehalogenation of organic halides is not only important as a synthetic rout but is now gaining increasing significance as a potential methodology for treating toxic halogenated waste In the latter application, the organic halide is converted to the corresponding hydrocarbon and the HCl that is produced can be readily separated while the hydrocarbon is recycled as a means of waste minimisation. Thermal hydrodechlorination processes only proceed to an appreciable degree at temperatures in excess of 973 K5 but the presence of a catalyst lowers the operating temperature significantly The catalytic hydrodechlorination of chlorobenzene and chlorophenol(s), known environmental hazards promoted using a solid silica supported nickel catalyst is considered in this paper. The catalytic hydrodechlorination of chlorobenzene(s) has been reported in both the ga–11 and liqui–14 phases using palladium–16 platinum rhodiu and nicke based catalysts. The treatment of chlorophenols has, by comparison, received less attention but catalytic data are available for the liquid phase reaction over carbon supported palladiu and gas phase transformations over nickel system as well as the electrochemical dechlorination on palladized electrodes Liquid phase hydrodehalogenations can proceed in the presence of both molecular hydrogen (at pressures up to 50 atm) and hydrogen donors such as metal hydrides, formic acid and its salts and alcohols The mechanism of C–Cl bond hydrogenolysis in heterogeneous systems is still far from understood and a number of kinetic models have been propose–11,13,15 to account for the observed catalytic trends. Moreover, the reaction has been viewed in terms of both electrophili and nucleophili substitution and attempts have been made to identify the possible reactant–catalyst interactions A marked drop in dechlorination activity with reaction time has been reported for supported palladium rhodiu and bulk nickel catalyst while time invariant reaction profiles have been generated for Ni/Al2O3 (ref. 8) and a Pt/zeolite The drop in dechlorination activity has been linked in one instance to a loss of the supported active phas and also to a surface poisoning due to the formation of stable surface chloride species The study of catalytic dechlorination is still in a formative stage and the published studies are, in essence, a compilation of rate data which characterise individual systems while such issues as the nature of the reactive adsorbed species, the catalytically active site(s) and the source of catalyst deactivation (when it occurs) are still not established. In this paper we provide kinetic data for the gas phase (molecular) hydrogen treatment of chlorobenzene and 3-chlorophenol and, where feasible, compare our results with the above cited reports. The effect of prolonged exposure of the nickel/silica catalyst to concentrated chlorinated aromatic gas streams is examined, particularly in terms of the changes to the nature of the surface reactive hydrogen.