Valorization of chloromethanes by hydrodechlorination with metallic catalysts

The performance of Pd, Pt, Rh and Ru based catalysts in the hydrodechlorination of chloromethanes to obtain ethane and ethylene was evaluated by means of computational analysis and hydrodechlorination experiments. A computational analysis using density functional theory (DFT) was developed to obtain preliminary insight on the potential catalytic mechanisms for the reactions involved using palladium, platinum, rhodium and ruthenium metallic clusters. Stable catalytic intermediates were obtained by quantum-chemical calculations in the hydrodechlorination of dichloromethane on Pd6and Rh6clusters, presenting [rad][rad]CH2and [rad]CH3radicals and C2H4, C2H6and CH4products. On the contrary, it was not possible to obtain all these stable intermediates using Pt6and Ru6clusters. Theoretical analysis revealed lower desorption energies for ethane and ethylene products in Pd6than in Rh6clusters, what indicates a favorable selectivity of Pd-based catalyst for desired C2products. Then, carbon supported catalysts containing these four metals were prepared and experimentally evaluated in the hydrodechlorination of dichloromethane (DCM) and trichloromethane (TCM) at low H2excess and a reaction temperature range of 150-400 °C. In agreement with computational results, in experimental tests, the Pd based catalyst showed the best performance for the hydrodechlorination of chloromethanes to obtain C2products, followed by Rh, Ru and Pt have a poor performance, in special Pt based catalyst, which shows almost no selectivity to C2products. This computational and experimental study emphasizes, for the first time, the good performance (high activity and selectivity) of Pd carbon supported catalysts in the valorization of chloromethane compounds to obtain C2hydrocarbon productsThe authors gratefully acknowledge financial support from the Spanish Ministerio de Economía y Competitividad (MINECO) through the project CTM 2014-53008 and to the “Centro de Computación Científica de la Universidad Autónoma de Madrid” for computational facilitie