139HIGHLIGHTS 2020

PROBING CATALYST ACTIVE SITES DURING CARBON DIOXIDE HYDROGENATION TO METHANOL

The use of carbon dioxide instead of carbon monoxide for the synthesis of methanol, which is one of the most important reactants in the chemical industry, is receiving great academic and industrial attention. Here, carbon dioxide hydrogenation mechanisms and copper-zinc active phases are revisited using high-pressure operando XAS and SSITKA-FTIR techniques.

Chemical valorisation of carbon dioxide via thermo-catalytic hydrogenation into methanol, which is one of the most important basic molecules of the chemical industry used for the synthesis of plastics, solvents and fuels, is currently receiving great attention due to the increased global energy demand and the need to decrease the carbon footprint on the environment. For this process, a commercial copper-zinc-alumina (CZA) catalyst possessing high activity is commonly studied. The copper- zinc oxide is a unique system and prototype heterogeneous material, which has been actively investigated for decades [1-3]. Despite these efforts, driven by both academia and industry, structure-performance relations remain elusive. Multiple research groups ascribe the active sites to completely different chemical and structural centres, which triggers a debate about the role of copper-zinc alloy [4-5]. Most studies are based on experiments under ex-situ and in-situ conditions (typically vacuum and low temperature) that are far away from the real catalytic experiment (>15 bar; 513-553 K).

Hence, even a simple comparison and systematisation of the existing experimental data is a rather complicated challenge.

To address this challenge, operando characterisation of the catalyst and the process under relevant catalytic conditions of carbon dioxide hydrogenation was performed. Using a state-of-the-art steady-state isotope labeling experiment coupled with infrared spectroscopy (SSITKA-FTIR), together with time-resolved X-ray absorption spectroscopy (XAS), X-ray diffraction (XRD) measured at beamline BM31, and ab- initio modelling, it was possible to identify the reaction intermediates, the pathways of their formation, and transformations of the active sites in the copper-zinc oxide catalyst.

By varying the reduction temperature and partial pressures of hydrogen and carbon dioxide during pre-treatment steps, it was possible to tune the content of copper-zinc alloy in the catalyst. CZA materials, containing significantly different amounts of reduced zinc, possess

Fig. 121: a) Zn K-edge XANES spectra of copper-zinc oxide catalyst collected after the switch from hydrogen to CO2/H2 mixture. b) Results of PCA analysis of XANES spectra as a function of time after the switch from hydrogen to CO2/H2 mixture and back at 533 K and 15 bar over CZA catalyst. c) Operando XRD patterns (l = 0.4975 A) of CZA catalyst.