Computational Study Of CO₂ Electroreduction

Electrochemical reduction of CO₂ to valuable hydrocarbons is one promising solution for environmental crisis and fossil fuel shortage.Copper is the only pure metal that capable of electrochemically transforming CO₂ into hydrocarbons,but its application is hindered by i)the high over-potential?1V vs RHE?during hydrocarbon transforming and ii)the competing of the hydrogen evolution reaction?HER?.Although a large amount of materials ranging from alloys,metal oxides,metal carbides,have been investigated for CO₂ reduction reaction?CRR?,their performances are yet not sufficient to meet industrial demand.Therefore,the demand for catalysts capable of efficiently produce desirable hydrocarbons at a much lower onset potential is still an urgent issue.Aimed for the goal,the article is mainly organized in terms of two aspects:On one hand,we focus on catalysts,including mechanism of CO₂ reduction reaction on pure metals,size effect and alloy modification and machine learning for high-throughput screening of desired catalysts.One the other hand,thinking out of the box,we attempt to highlight the reaction environment around the catalyst interface.i.e.chemical filed environment?CFE?.Original models are developed to explore the effect of CO₂ activation at what scales and at what molecules.More details are as follows:For catalysts,firstly,we investigate the adsorption properties of 26 intermediates during CO₂ reduction process on the Cu?111?,Cu?100?,Cu?211?facets,and their corresponding Gibbs free energies.For the former,we concluded the correlation between the adsorption of adsorbates,like scaling relationship between adsorbates with the same central atoms.For the latter,we analyzed the possible pathways of transforming CO₂ to products like H₂,HCOOH,CH₄,C₂H₄.The results demonstrated the reaction Gibbs free energy of the two basic steps of CO₂?*COOH and*CO?*CHO?*COH?are less than zero,especially the latter one,which is the rate-determining step.Through the comparison of CO₂?*COOH?*CO?*CHO?*COH?,we draw a conclusion for CRR that the adsorption energy of CO on catalysts qualitivly determine whether the catalyst is able to transform CO₂ to hydrocarbons,and the limiting potential of the two steps*CO?*CHO?*COH?and CO₂?*COOH quantitively decide how much over-potential is needed to make reaction happen,and*H adsorption energy influence the H₂ evolution reaction.Secondly,based on the realization above,we investigate the size effect of Cu13,Cu55,Cu147 and Cu309 clusters with cuboctahedron models in terms of geometry structure and electronic structure,and explore their CO₂ performance through calculating the adsorption properties of*CO,*COOH,*CHO,*COH and*H intermediates on them.Apart from that,we calculated the Gibss free energy profiles of CO₂?*COOH?*CO?*CHO?*COH?and*H over 48 kinds of Au and Ag alloys for desired catalysts capable of transforming CO₂ to hydrocarbons,it turns out the Au/Pt/Au?111?,Au/Rh/Au?111?and Au/Ni/Au?111?alloys are most promising catalyst at lower overpotential,higher selectivity of hydrocarbons and weak H₂ evolution reaction.Then,we attempt to establish predictive models of CO adsorption energy with d band as descriptor via the method of machine learning.LASSO method is used to reduction the dimensionality of descriptor.Various learning algorithms and various descriptors of d band are compared,the results illustrated the better the prediction,the more the details of d band.The research provide some basic realization for high-throughput screening of desired catalysts within short time.For chemical field,new models associated with frustrated geometry and electric field?EEF?were designed to qualitatively and quantitatively explore CO₂ activation through density functional calculations.We investigated the influence of the microenvironments of methylamine?CH₃NH₂?hanging at different height above Cu?111?surface on CO₂ in the presence and absence of electric field.The results demonstrate that at approximate 6±1?distance of CH₃NH₂ and metal surface,neither lower nor higher,under EEF over 0.4 V/?,it brings about remarkable synergistic effect of chemical interaction and EEF to activate CO₂,which also lowers demands,in contrast to separated factor or any pair of them to achieve it.Besides,various other chemical groups and substrates were taken into account.It demonstrates that constraining or facilitating CO₂ activation which strongly depends on distance between chemical group and substrates.The CFE constructed by the combinations of three factors substrate,chemical group and EEF provide new protocols to make CO₂activation easier and controllable.Based on the CFE,we report a good example which could illustrate the idea in practice.Functionalizing the imidazolium ion with a propanol substituent at the N site can significantly enhance the catalytic activity of IMILs,causing a positive shift of the onset potential for the CO₂ reduction by about 90mV in an acetonitrile electrolyte.Theoretical calculations indicated that the propanol hydroxyl could bridge a local hydrogen-bonding chain as shortcut for proton transfer,leading to a dramatic decrease of the activation barrier for the catalytic reduction of CO₂ in IMIL.