On the catalytic behaviors of Cu/SiO2 and Cu/γ-Al2O3 for dimethyl oxalate hydrogenation from microkinetic analysis including a plug flow reactor model

Cu-based catalysts have been widely used in dimethyl oxalate (DMO) hydrogenation due to their ability to activate C-O/C=O bonds without breaking C–C bonds. In this work, the electronic structures of Cu/SiO₂ and Cu/γ-Al₂O₃ as well as their catalytic performance have been studied by the machine-learning-based stochastic surface walking-global neural network potential (SSW-NN) method, density functional theory calculation, and microkinetic analysis including a plug flow reactor (PFR) model. Among SiO₂- and γ-Al₂O₃-supported Cu_n (n = 1–9), the Cu₅ and Cu₇ clusters are held most tightly on SiO₂(111) and γ-Al₂O₃(110), respectively. The electron transfer from Cu₅ to SiO₂(111) leads to the formation of Cu^δ+ and Cu⁰, which are responsible for the stabilization of unsaturated C and O atoms in the intermediates, respectively, while on γ-Al₂O₃(110) an electron-rich Cu⁰-Al_3c site is most active. Both the Cu^δ+-Cu⁰ and the Cu⁰-Al_3c sites synergistically catalyze the dissociation of gas-phase species and hydrogenation of intermediates. Under the typical operating conditions, although the selectivity towards methyl glycolate (MG) is invariably highest at the reactor inlet, Cu₅/SiO₂(111) and Cu₇/γ-Al₂O₃(110) are actually selective for the production of ethylene glycol (EG) and ethanol (EtOH), respectively, if the overall selectivity is taken into consideration, signifying the importance of including a reactor model to probe the kinetics of series of consecutive reactions. The dissociation of DMO is found to be the rate-determining step, and the high energy barrier for EG dissociation on Cu₅/SiO₂(111) hinders its deep hydrogenation while the relatively low barrier on Cu₇/γ-Al₂O₃(110) is beneficial to the formation of EtOH.