The manufacture of polyvinyl chloride (PVC) relies on the production of vinyl chloride (VCM) via a two-step process comprising ethylene oxychlorination to ethylene dichloride (EDC) on intrinsically unstable cupric chloride catalysts and a subsequent energy-intensive thermal cracking. Recently, we demonstrated that certain oxides or oxychlorides (e.g., CeO₂) could serve as stable bifunctional catalysts, performing the two reactions in a single step and attaining VCM in up to 25% yield.^1,2 To gain deeper understanding of the oxychlorination mechanism, we investigate the complex reaction network by density functional theory (DFT), kinetic testing, and advanced characterization. In a first step, decoupling the complexity of the reaction network and CeO₂, DFT calculations were conducted on the simpler RuO₂, revealing the thermodynamic and kinetic control of oxidation and chlorination processes, respectively. Furthermore, the degree of chlorine coverage emerged as a key factor governing the selectivity, as it confines or destabilizes intermediates and determines the ability of the surface to abstract H atoms. This enabled simplification of the reaction network (Figure 1) for its investigation on a CeO₂ {111} surface in a second step. Preliminary results suggest that an ideal CeO₂ surface is unable to catalyze oxychlorination as the barriers for all important chlorination and dehydrogenation steps are 1.7 eV and higher. The introduction of defects, i.e., by replacing a surface O by a Cl atom which can react and form a vacancy, leads to at least 0.5 eV decreased barriers for most chlorination reactions. Inspired by these results, we are developing a microkinetic model and experimentally assessing the impact of vacancies formed by controlled reduction of CeO₂ on its oxychlorination performance. Previously, acidity was shown to be strongly related to VCM selectivity,^1,2 but the exact nature of the sites remained unknown. DFT indicates that ceria acts as a Brønsted and Lewis acid/base pair, encouraging the investigation of the acid sites by solid-state NMR spectroscopy after adsorption of the probe molecule trimethylphospine. The combination of these results will reveal the detailed mechanism of ethylene oxychlorination on oxide-based catalysts and guide the development of more selective VCM catalysts.

Figure 1 Simplified reaction network of ethylene oxychlorination, excluding oxidation steps which can start at any node.

[1] M. Scharfe, P. A. Lira-Parada, V. Paunović, M. Moser, A. P. Amrute, J. Pérez-Ramírez, Angew. Chem. Int. Ed. 2016, 55, 3068.
[2] M. Scharfe, P.A. Lira-Parada, A. P. Amrute, S. Mitchell, J. Pérez-Ramírez, J. Catal. 2016, 344, 524.