Subscriber access provided by UNIV OF LOUISIANA
Perspective
Understanding the Roadmap for Electrochemical Reduction of CO2 to Multi-Carbon Oxygenates and Hydrocarbons on Copper-based Catalysts Yao Zheng, anthony vasileff, xianlong zhou, Yan Jiao, Mietek Jaroniec, and Shi-Zhang Qiao J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b02124 • Publication Date (Web): 15 Apr 2019 Downloaded from http://pubs.acs.org on April 15, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
Understanding the Roadmap for Electrochemical Reduction of CO2 to Multi-Carbon Oxygenates and Hydrocarbons on Copperbased Catalysts Yao Zheng,1 Anthony Vasileff,1 Xianlong Zhou,1 Yan Jiao,1 Mietek Jaroniec,2 Shi-Zhang Qiao1,* 1 School of Chemical Engineering, The University of Adelaide, Adelaide, SA 5005, Australia 2 Department of Chemistry and Biochemistry, Kent State University, Kent, Ohio 44242, United States Keywords: CO2 conversion; Electrocatalysis; Reaction mechanism; DFT Calculations; Oxygenates; Hydrocarbons ABSTRACT: Electrochemical reduction of CO2 to highenergy-density oxygenates and hydrocarbons beyond CO is important for long-term and large-scale renewable energy storage. However, the key step of the C–C bond formation needed for the generation of C2 products induces an additional barrier on the reaction. This inevitably creates larger overpotentials and greater variety of products as compared to the conversion of CO2 to C1 products. Therefore, an in-depth understanding of the catalytic mechanism is required for advancing the design of efficient electrocatalysts to control the reaction pathway to the desired products. Herein, we present a critical appraisal of reduction of CO2 to C2 products focusing on the connection between the fundamentals of reaction and efficient electrocatalysts. An in-depth discussion of the mechanistic aspects of various C2 reaction pathways on copperbased catalysts is presented together with consideration of practical factors under electrocatalytic operating conditions. By providing some typical examples illustrating the benefit of merging theoretical calculations, surface characterization, and electrochemical measurements, we try to address the key issues of the ongoing debate toward better understanding electrochemical reduction of CO2 at the atomic level and envisioning the roadmap for C2 products generation. INTRODUCTION From giant progress in the sector of renewables, the cost of renewable electricity (from solar, wind, and hydropower, etc.) is now at a level competitive with the electricity generated from coal.1-2 However, to address the current global energy demand, the long-term, large-scale, and seasonal energy storage of renewable energy remains a big challenge. Storing energy in chemical bonds has many benefits over battery storage, including significantly higher energy density, enhanced safety and transportability.3 (Photo)electrochemical water splitting to gaseous hydrogen is a successful strategy, which integrates renewable energy and fuel production economy. Further, highenergy-density liquid fuels are better suited for the existing infrastructure and transportation system, as well as provide more complex feedstocks for chemicals production.2,4 Electrochemical transformation of the captured carbon dioxide (CO2) to various hydrocarbons via the CO2 reduction reaction (CRR) is one of the most effective ways for the utilization and storage of renewable electricity. Importantly, as compared to C1 products (CO, CH4, HCOO- and CH3OH), some C2 and C2+
products (C2H4, C2H5OH, CH3COO-, n-C3H7OH, etc.) have higher volumetric energy densities and can directly serve as building blocks for the synthesis of long-chain hydrocarbon fuels and oxygenates.5,6 However, the C–C bond formation competes with the formation of C–H and C–O bonds under CRR conditions, which makes the generation of multi-carbon products difficult.6 To date, the highest reported Faradaic efficiency (F.E.) for the simplest C2 product (C2H4) is ~60 %,7 while that of C3 product (n-C3H7OH) is 21 %.8 These two values are much lower than those for generation of CO and HCOO(close to 100 %).9-10 Contrary to the simpler and betterestablished C1 reaction pathways, those for C2 products are more complex and more strongly dependent on the catalyst surface and electrocatalytic operating conditions. Therefore, the mechanistic aspects of various C2 reaction pathways are not fully settled. More importantly, due to the much higher kinetic barrier of the C–C coupling step, the energy efficiency (E.E., collectively determined by overpotential and F.E.) and partial reduction current density of C2 products are much lower than those for C1 products. This ultimately limits the practical application of CO2 conversion to C2 products in commercial electrolysers. Suitable electrocatalysts can lower the energetic barriers for CRR by stabilizing reaction intermediates and transition states in the multistep proton-assisted electron-transfer process.11 Since Hori’s landmark discovery regarding the unique ability of copper (Cu) toward CRR to generate C2 hydrocarbons in comparison to other metals,12 a substantial effort has been invested to understand the generation of products beyond C1. Unfortunately, Cu is not a selective catalyst for specific carbon products because at least 16 different C1-C3 hydrocarbon/oxygenate products have been detected on Cu surfaces under CRR conditions (0.1 M KHCO3 solution and potential in the range of -0.7 ~ -1.2 V vs reversible hydrogen electrode, RHE).13 Therefore, it is important to modify Cu to achieve the desired reduction products. It is widely accepted that CO is the key intermediate in the formation of multi-carbon products via dimerization (in the form of CO–CO) or coupling (in the form of CO–COH) pathways.14,15 Clearly, each of these two steps needs the precursor, namely the steady-state adsorbed CO (*CO), to be present on the surface in close proximity to one another at a high coverage. This requires the specific morphological and electronic characteristics of the catalyst.16 For example, the well-designed morphology can affect the local pH or proton concentration on the catalyst surface to enhance
ACS Paragon Plus Environment
Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
C2 products’ F.E.17-20 For the chemical composition, a wide variety of oxide-derived Cu (OD-Cu) catalysts features enhanced relative selectivity toward C2H4 over CH4.9,21 While the origin of the high C2 F.E. observed for the OD-Cu catalysts is still debated, it is reasonable to attribute it to either physical or chemical properties of OD-Cu or both.9,21-22 At the atomic level, theoretical computations based on the density functional theory (DFT) have been widely used to investigate CRR mechanism. These studies include predicting the adsorption energy of key reaction intermediates to build up the reaction free energy diagram, which can be used to obtain the activity of one catalyst.11 In addition, the reaction barriers between different adsorption states are sometimes considered, mainly for deriving the selectivity of one certain product.23 In the last years, the theoretical studies of CRR progressed by adding layers of complexity gradually: the first DFT studies considered only reaction thermodynamics and suggested some key reaction intermediates including *CHO and *CO dimer.11, 16 Subsequently, some reaction kinetics inconsistencies in the thermodynamic models and other intermediates (*COH and *CH2, etc.) have been addressed by considering the reaction barriers.24-26 Importantly, the research community increasingly recognizes the significance of electrocatalytic operating conditions in DFT computation. The driving force for the socalled operando computations is based on the observation that the performance of a catalyst is influenced by actual reaction environment; factors like local pH and electrode potentials play fundamental roles in determining the activity and selectivity of catalysts.27 At the same time, direct observation of the surfacebound species and their interaction with electrode surfaces during the reaction by advanced operando spectroscopy provides a crucial evidence to support or refine the proposed mechanisms. For example, the most widely accepted C–C bond formation mechanism (through a negatively charged CO–CO-
CO insertion
5H+/5eH
C
H+/e-
H
C
OH-
CH3COO-
H
H C
H
H+/e-
C2H6
SDI
5H+/5eH
H
O
C
H C
3H+/3e-H2O
CH4
C
H O
O
species formed via *CO dimerization, which is further protonated to *CO–COH) was confirmed by in-situ Fourier transform infrared spectroscopy (FTIR), online products detection and DFT computations (discussed in detail later).16,28 At this stage, a combination of quantum chemistry computations, electrochemistry, and materials science has significantly accelerated the development of electrocatalytic CO2 reduction. The aim of this Perspective is to offer a critical understanding of the roadmap toward the CO2 reduction to multi-carbon hydrocarbons and oxygenates by collective assessing these three fields. First, we provide the initial framework for C2 and C3 pathways, starting from *CO on the Cu surfaces. Then, we further discuss some physical and chemical factors that can change or modify these pathways to achieve one certain product under given CRR operating conditions. Finally, we try to build up the structure-property and composition-property relationships for a wide variety of real electrocatalysts to establish the design principles for catalysts specific for C2 products. Based on a series of examples illustrating the benefit of combining the operando computations, operando spectroscopy and experimental observations, the main aim is to provide a brief but insightful account of the electrochemical reduction of CO2 to C2 to better understand this important process. ATOMISTIC MECHANISM OF VARIOUS C2 AND C3 PRODUCTS The overall roadmaps for C2 and C3 products from CO2 or CO feedstock on various Cu surfaces are constructed by identifying some key reaction intermediates for specific products. Similar to other electrocatalytic processes like CO2 reduction to CO, oxygen reduction reaction (ORR) and hydrogen evolution reaction (HER) etc.,29-30 the descriptor-based activity/selectivity trends for the C2 and C3 pathway on various Cu-based electrodes have also been established. H
O
CH4
Page 2 of 14
C
H+/e-
O
O
CO insertion
H
C
C
C O
H+/e-
C2H2O2
H C O
4H+/4e-
C2H6O2
H
H O
e-
O
O C
RDS
C
C
H+
C
H
H
O
-H2O
H
C
C
4H+/4e-
H
H
H
O
H+/e-
H C
H
C
H+/e-
H
H
H
O
H+/e-
O H
C2H4
H
H
SDI H+/e-
C
C
CO insertion 4H+/4e- -H2O
C H
C2H5OH H
C
C
H
O 2H+/2e-
n-C3H7OH
Figure 1. The most possible C2 and C3 pathways starting from *CO on Cu surfaces. Green, blue, and red routes are for trace, minor and major C2 products. Grey routes are for C1 product. Dashed arrows indicate the multiple consecutive electron/proton transfer steps.
Overall Roadmap. As summarized in Figure 1, there are three routes for the formation of overall seven different C2 and C3 products. Trace products (F.E. - 0.6 v
b
pH-dependent mechanism
C
while at high overpotentials, the coupling of *CO–COH was dominant.23,50 This can explain the experimentally observed potential-dependence of C2H4, C2H5OH and CH4 selectivity for the Cu surfaces.13,40,47At the atomic level, the overpotential dependent C2 pathways, including thermodynamic free energies and kinetic reaction barriers, have been obtained by DFT computations.27,51,52 Although not perfectly accurate, the conceptual computational hydrogen electrode (CHE) model provides a simple and feasible method to account for the chemical potentials of a hydrogen molecule on the RHE scale.53 As a result, the potential-dependent reaction free energy for elementary CRR steps can be obtained. Taking into account the potential-dependent kinetic barrier associated with the elementary electrochemical steps, many approximation approaches have been proposed giving a good agreement with experimental observations (Figure 4a).27 For example, on the Cu(100) surface, when the potential (U) is more positive than – 0.6 V, the C–C coupling through *CO–CO, then *CO–COH, and finally C2H4 is favored over CH4 formation via *CHO due to the lower energy barrier (0.69 vs. 1.0 eV).27 Therefore, C2H4 is the major product in this range (Figure 4b). From -0.6 to -0.8 V, adsorption of *H and *CO competes for surface sites, leading to a decrease in the CO surface coverage, which causes a decrease in the rate of CO dimerization to C2H4. As a result, the F.E. value for C2H4 decreases while that of H2 increases (Figure 4b). At potentials more negative than -0.8 V, this site blocking does not affect *CHO formation, which can be formed via an Eley–Rideal (E-R) mechanism from non-adsorbed CO to form *CO–CHO. Consequently, C2H4 and CH4 formations share one intermediate, which also explains the subsequent appearance of C2H4 and CH4 in this range (Figure 4b).44-45
Intensity (a.u.)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 14
C
C2H4
Figure 3. (a) Calculated the pH dependent reaction pathways for C1 (blue routs) and C2 (red routes) products on the Cu(111) surface. (b) The OLEMS signals of the reduction CO in 0.1 M phosphoric acid (pH = 1) and 0.2 M phosphate buffers (pH = 7 and 12) on the Cu(111) and Cu(100) surfaces. Reproduced from ref. 45. Copyright 2014 ELSVIE.
Overpotential Dependence. Like the case of electrolyte’s pH, the offset C1/C2 pathways also occur at the different ranges of the applied overpotentials. At low overpotentials, the *CO–CO dimerization pathway was found to be energetically favorable,
Surface Adsorbates (H+, OH-, cation, anion, H2O). The acidbase equilibria between bicarbonate and CO2 reactant include H+ (CO2(g) + H2O(l) ↔ H2CO3(aq) ↔ HCO3(aq) + H+ (aq)) while the reduction process comprises a series of protonation steps. Therefore, the H+ concentration on the surface of catalyst is important factors in deterring its product selectivity. For example, it was found that C2H4 is favored at low concentration of KHCO3 electrolyte (0.1 M), whereas CH4 and H2 are favored at concentrated electrolytes.47-48 Theoretical works show that OH- anions play a complex role, which possibly enhances the C2 selectivity by suppressing HER competition, similar to the aforementioned pH dependent C1 and C2 pathways.35,45 Following this concept, with the help of a flow cell electrolyser (discussed in details later), some well-designed electrode configurations with high surface pH can deliver extremely high C2 selectivity (F.E. > 60 %).54-56
ACS Paragon Plus Environment
Page 5 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
Besides H+ and OH-, the anions and cations in the bulk electrolyte can also influence C2 product selectivity by stabilizing reaction intermediates (i.e., increasing the *CO coverage).38,57-59 It was found that F.E. for H2 and CH4 is reduced with increasing size of alkali cation from Li+ to Cs+, while the values of F.E. for CO, C2H4, and C2H5OH increase.60 Further calculations explain this behavior as a consequence of electrostatic interactions between solvated cations in the outer Helmholtz plane (OHP) and adsorbed species with large dipole moments. Specifically, because the dipole moment of *CO is larger than that of *CHO, *CO is more stabilized by the electric field of the hydrated cations as compared to *CHO, yielding more C2 products.59 Halides also have the ability to stabilize the *CO intermediate through a covalent Cu-halide interaction.61 Of the three potassium halides (Cl−, Br−, and I−), KI electrolyte showed the highest C2 selectivity because it induced the highest *CO population on the catalyst surface.57 As expected, an electrolyte containing both Cs+ and I- (e.g., CsHCO3+CsI) leads to a record C2 F.E. of 69 % with a high C2/C1 ratio close to 40 on a CuOx surface.58 While water is conventionally assumed to be just a solvent in the electrolyte, it was found that the surface bound H2O molecules on Cu(111) play a critical role in promoting dehydration process that is involved in the formation of hydrocarbon products only.23 Therefore, tuning water adsorption may be used as a way to control the competition between hydrocarbon vs oxygenate production in the late reduction steps. Experimental proof showed that when performing C16O reduction in isotope labelled H218O electrolyte, ~90 % of CH3COO- and 60-70 % of the alcohols (C2H5OH and n-C3H7OH) contained 18O from the solvent, that indicating water plays a dominant role in their formation.62 However, a very recent study suggests that these experimental observations are likely the result of isotopic scrambling between transiently produced carbonyl-containing intermediate reaction products, such as acetaldehyde, and water (solvent).63 This assumption is supported by theoretical calculations indicating that the reversible hydration of these carbonylcontaining species is feasible in the vicinity of the Cu surface.63 The effect of water on the C2 product selectivity was also intensively investigated by computations, both within explicit and implicit solvation models.64 A direct strategy to improve the original CHE model is to introduce an explicit solvation model with several or few layers of water molecules on the electrode surface to reproduce the dielectric response of the liquid environment.64 Recently, solvent and electrolyte effects were both considered to describe the reaction under real conditions (for example five layers of the explicit solvent, a potential of −0.59 V vs RHE, pH = 7 of the electrolyte).27 With help of quantum-mechanical calculations, the full atomistic reaction mechanism with kinetics of various C2/C3 products was revealed.27 Considering the extensive computational demands of the explicit model, an implicit solvation model was developed to economically simulate the constant-potential electrochemical reactions with solution and electrolyte effects.65 A nice illustration of these effects is a quantitative comparison of the reaction energies and kinetic barriers for complete CRR pathways under three protonation mechanisms, i.e., standard surface *H in Langmuir-Hinshelwood (L-H) adsorption, solvated H+ in E-R adsorption, and a new surface *H2O in L-H adsorption model.23 This study theoretically explained numerous experimental facts such as why the C1 and C2 products are generally CH4, C2H4, C2H5OH but rarely
CH3OH, C2H2, and C2H6.13,38 Additionally, these method including implicit/explicit water layers and a dielectric continuum model are considered as the most reasonable models to describe real CRR conditions, and the overall diagram serves as a benchmark for the following investigation.23,27 Facets Sensitivity. Apart from physical factors, the intrinsic properties of catalysts also have a big influence on their activity/selectivity. While it is very clear from experiments that the CRR product selectivity shows a strong facets dependence, e.g., Cu(100) favors C2H4, Cu(111) favors CH4 and HCOO-, and Cu(110) favors secondary C2 products like CH3COO-, CH3CHO, and C2H5OH,66 the origin of this selectivity trend is still inconclusive. For example, one study claims the introduction of steps on Cu(100) terraces enhances C2H4 evolution and suppresses CH4 formation,66 while others attribute the selective formation of C2H4 to pristine Cu(100) terraces.44 Moreover, when the aforementioned pH and overpotential dependencies are integrated, we arrive at a more comprehensive explanation. For example, C2H4 can be formed via two different pathways: (i) on Cu(100), CO is reduced to C2H4 only via the CO dimer pathway (at low overpotentials); and (ii) on both Cu(100) and Cu(111), CH4 and C2H4 form simultaneously from the same *COH intermediate (at high overpotentials).35 These proposed facet dependent mechanisms agree well with electrochemical and spectroscopic observations (Figure 2b,3b).28,35 The C–C mechanism on another step surface, Cu(211) was investigated under an explicit solvent and cation-induced field conditions.26 It was found that the energy barriers for the formation of the *CO–CO dimer and its hydrogenation to *CO–COH are similar to those on Cu(100) but lower than those on Cu(111).52 Additionally, the operando scanning tunneling microscopy observations showed that polycrystalline Cu undergoes reconstruction to Cu(111) and then to Cu(100) under CRR conditions.67 Therefore, the real time monitoring of the surface active sites is also important for determining the mechanism. DESIGN PRINCIPLES OF C2 ELECTROCATALYSTS Even though theoretical studies indicate *CO dimerization or coupling to be the key step for the C–C bond formation, the optimization of the adsorption energetics to an appropriate level is still challenging. This difficulty arises from the linear scaling between the activation energy (kinetic barrier) of the reaction and the binding energies of reaction intermediates in a Bell-Evans-Polanyi (BEP) relationship.68 That is to say, a catalytic site needs to bind *CO intermediates strongly enough to build up a sufficient coverage for further C–C formation, but the associated activation barriers also increase with stronger *CO binding.6 At the same time, fundamental studies also provide some guide for the design principles of real CRR catalysts, which either optimize the composition-property relationships or the structure-property relationships to break or balance this BEP relationship. Consequently, a variety of highly active and selective C2 electrocatalysts have been developed, such as OD-Cu, nanostructured Cu, and bimetallic Cu, etc. Many factors like residual Cu+, subsurface oxygen, grain boundaries (GB), and inter-/intra-molecular synergy work in a complex way to influence the real dependence of C2 product selectivity. For example, the origin of the high C2 F.E. observed for the OD-Cu catalysts is still debated, it is reasonable to attribute it to the following points:7,18,19 (i) the roughened morphology of OD-Cu induces a high local pH at the electrolyte/electrocatalyst interface, favoring the pH-dependent
ACS Paragon Plus Environment
Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
(on RHE scale) C–C coupling pathway for C2H4 formation; (ii) high density of GBs, defects, and/or uncoordinated sites, which optimize the binding energy of the key *CO reaction intermediates for the C–C bond formation; and (iii) the presence of residual oxides and/or subsurface oxygen atoms on OD-Cu significantly affects the kinetics and thermodynamics of CO2 activation and CO dimerization processes, which promote the generation of C2 products. Chemical States. It was found that OD-Cu catalysts, prepared by successive electrochemical redox processes, yielded higher activity and selectivity for C2 products (especially C2H5OH) as compared to poly- and single-crystal Cu.69 The origin of such significant enhancement is still unclear; a major factor is the oxidation state of the surface Cu sites during CRR conditions.9 Thermodynamically, oxidized Cu2O should be reduced to Cu with an E0 of -0.36 V vs. NHE, which is more positive than the operating potential for most CRR catalysts. However, if the reduction process is sufficiently slow, some Cu+ and O2- species can still be present on the surface or subsurface of Cu.9 More likely, the high local pH formed during CRR can help stabilize Cu+ by shifting the overpotential for Cu2O reduction in negative direction.7 For example, by in-situ X-ray adsorption spectroscopy (XAS), it was demonstrated that 23 at. % of Cu+ can be stabilized in electro-deposited Cu until a potential of 1.2 V vs RHE (the potential required for C2H4 generation).70 Further, a correlation between hydrocarbon selectivity and Cu+ was found on the plasma-treated Cu where Cu+ remained on the Cu surface as confirmed by in-situ XAS.7 Theoretical studies indicated the Cu+ may work synergistically with Cu to promote CO2 activation and *CO dimerization steps (Figure 5a).71,72 Specifically, in the initial step, the Cu+ provides a strong H2O adsorption site via hydrogen bonds to stabilize CO2 molecule.71 In the next step, the Mulliken charge analysis indicates that *CO on Cu+ and Cu has opposite change (+0.11 on Cu+ and -0.31 on Cu). Therefore, the attractive electrostatics between two carbons assists C–C bond formation.71 Similarly, in an atompair catalyst system with two adjacent Cu atoms, it was found that Cu1x+ can adsorb H2O and the neighboring Cu10 can adsorb
CO2, which thereby promotes the CO2 activation process.73 As an experimental validation, a high F.E. toward C2 products of ~80 % was reported on the Cu-based electrocatalysts with an average Cu valence state of +0.35 tuned by boron doping (Figure 5b).70 However, opposite results are reported. For example, the in-situ selected ion flow tube mass spectrometry (SIFT-MS) measurements indicate that gaseous C2 products are not formed until Cu2O reduction to metallic Cu is complete (Figure 5c).22 This was also supported by in-situ Raman spectroscopy, which showed that *CO vibrations only occur after the Cu-O vibrations from Cu2O disappear.74 These results both suggest that the CRR occurs on the metallic Cu sites instead of Cu+. More interestingly, ambient-pressure X-ray photoelectron spectroscopy (APXPS) and quasi in-situ electron energy loss spectroscopy analysis confirmed that there is no residual Cu+ on Cu surface after CRR operation.75 However, it was claimed that residual subsurface oxygen alters the electronic structure of Cu and creates sites with higher *CO binding energy, serving as the active sites for C–C coupling.75 Although some theoretical studies show that subsurface oxygen indeed enhances Cu’s selectivity toward C2 products by increasing the coverage of *CO,72 it is argued that it is unstable under CRR conditions. Analysis of 18O labelled OD-Cu showed that only a small fraction (
