Global warming and climate change are challenging the sustainable development of human society, and already have observable impacts on every aspect of human life. Carbon dioxide (CO2) emissions are regarded as the major driver of global warming. Enormous amounts of CO2 have been emitted in the past 70 years, as a result of fast development of human civilisation and heavy consumption of fossil fuels. To alleviate CO2 emissions, one promising way of carbon capture and recycling is to convert CO2 together with green/blue hydrogen (H2) to renewable methanol. Methanol is an important basic chemical feedstock, a greener transport fuel, and a good energy carrier of green/blue H2. The replacement of fuel-derived methanol by renewable methanol in its current applications, could thus significantly decarbonise the chemical and transport sectors to reach net carbon zero.
CO2 is chemically inert, to hydrogenate CO2 to methanol, active, selective, and stable catalysts are thus needed. In the present study, physicochemical properties, and catalytic performance of nanostructured CuO/CeO2 catalysts in CO2 hydrogenation to methanol synthesis were investigated.
Gram-scale ceria nanowires (NW) were successfully synthesised at a lab scale, and nanorods (NR) and nanocubes (NC) of desirable sizes were also prepared. Ceria NW and NR had a high specific BET surface area of 113 and 101 m2.g-1, respectively, as well as NC (55 m2.g-1). They were highly reducible and had a high surface Ce3+ content of 25-28% and surface oxygen vacancy concentration. 0.1-5wt% CuO/CeO2-NW, and 2 and 5wt% CuO/CeO2-NC and NR catalysts were synthesised. XPS, H2-TPR, and s-TPR together showed that all catalysts were highly reducible, and had one or two types of Cu species on the surface, i.e. well-dispersed and bulk-like Cu species. For CuO/CeO2-NW catalysts, from 0.1 to 5wt% CuO loading, ceria surface became fully covered by well-dispersed Cu species, and more bulk-like Cu species of larger sizes was formed. For 2 and 5wt% CuO/CeO2-NC and NR catalysts, both types of Cu species were observed. A higher CuO loading had a positive impact on catalyst reduction in H2-TPR and formation of surface Ce3+ ions and oxygen vacancies, as well as CO2 desorption in methanol TPD.
Methanol TPD suggested that ceria and catalyst surfaces were defective; type II methoxy was mainly oxidised to bidentate formate (to CO), bidentate/monodentate carbonate, further to CO2 ¬¬at high temperatures. The rate determining step of CO2 formation in methanol TPD was on the surface Cu sites. In CO2 hydrogenation to methanol synthesis, linear dependence of catalyst intrinsic activity on average Cu cluster size was observed. One correlation for NC catalysts, and one for NR and NW catalysts, with average Cu cluster size varying between 1.3 and 21.5 nm. These together indicated that the rate determining step of CO2 hydrogenation to methanol synthesis was on the surface Cu sites, and both surface Cu cluster size and ceria morphology had an impact on catalyst intrinsic activity.
The theoretical investigation of small Cun (n=1-4) adsorption on stoichiometric and defective CeO2(110) surfaces showed that both surfaces were readily reduced upon Cun adsorption. On both surfaces, Cu1 grew to Cu3 along the long bridge sites, forming strong Cu-O bonds at adjacent long bridge sites, which modelled a Cu monolayer growth mechanism, and provided theoretical insight into Cu dispersion on ceria nanowires with extremely low CuO loadings of 0.1 and 0.5wt%. For larger Cun (n=8,13,16,24) adsorption on stoichiometric CeO2(110) surface, Cun clusters had a polyhedral shape and interacted with CeO2(110) via anchoring Cu species without wetting the surface. Adsorbed larger Cun clusters had mainly Cu+ or Cuδ+ species as surface anchoring sites. The number of electrons retained in Cun increased with Cun size, which may enable electron-rich larger Cun to interact with surface intermediates more strongly in CO2 hydrogenation to methanol synthesis.