Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction

Article abstract of DOI:10.1016/j.chempr.2020.10.018

Electrochemical CO₂ reduction (CO₂R) to fuels is a promising route to close the anthropogenic carbon cycle and store renewable energy. Cu is the only metal catalyst that produces C₂₊ fuels, yet challenges remain in the improvement of electrosynthesis pathways for highly selective fuel production. To achieve this, mechanistically understanding CO₂R on Cu, particularly identifying the product-specific active sites, is crucial. We rationally designed and fabricated nine large-area single-crystal Cu foils with various surface orientations as electrocatalysts and monitored their surface reconstructions using operando grazing incidence X-ray diffraction (GIXRD) and electron back-scattered diffraction (EBSD). We quantitatively established correlations between the Cu atomic configurations and the selectivities toward multiple products and provide a paradigm to understand the structure-function correlation in catalysis.Aqueous electrochemical CO₂ reduction (CO₂R) on Cu can generate a variety of valuable fuels, yet challenges remain in the improvement of electrosynthesis pathways for highly selective fuel production. Mechanistically, understanding CO₂R on Cu, particularly identifying the product-specific active sites, is crucial. Herein, we rationally designed and fabricated nine large-area single-crystal Cu foils with various surface orientations as electrocatalysts and identified the voltage- and facet-dependent CO₂R selectivities. Operando grazing incidence X-ray diffraction (GIXRD) and electron back-scattered diffraction (EBSD) were applied to track the top-surface reconstructions of Cu, and we correlate the structural evolution with the change of product selectivities. We extracted three distinct structural descriptors, including crystal facet, atomic coordination number, and step-terrace angle, to reveal the intrinsic structure-function relationships and uniquely identify the specific product-producing sites for CO₂R. Our work guides the rational design of Cu-based CO₂R electrocatalysts and, more importantly, establishes a paradigm to understand the structure-function correlation in catalysis.We rationally designed and fabricated nine large-area single-crystal Cu foils with various orientations as electrocatalysts and operandoly monitored their surface reconstructions. Quantitative correlations between Cu atomic configurations and the selectivities toward multiple products were established. We extracted three distinct structural descriptors, including crystal facet, atomic coordination number, and step-terrace angle, to reveal the intrinsic structure-function relationships and uniquely identify the specific product-producing sites of Cu for CO₂R.

Full text of DOI:10.1016/j.chempr.2020.10.018

ll
Article
Product-Specific Active Site Motifs of Cu for
Electrochemical CO2 Reduction
Chenyuan Zhu, Zhibin Zhang,
Lixiang Zhong, ..., Hao Ming
Chen, Kaihui Liu, Liming Zhang
lisz@ntu.edu.sg (S.L.)
haomingchen@ntu.edu.tw (H.M.C.)
khliu@pku.edu.cn (K.L.)
zhanglm@fudan.edu.cn (L.Z.)
HIGHLIGHTS
Large-format single-crystal Cu
foils were fabricated as
electrocatalysts for CO2R
Operando GIXRD and ex situ
EBSD monitored the surface
reconstruction of Cu
Specific product-producing sites
of Cu for CO2R were identified
We rationally designed and fabricated nine large-area single-crystal Cu foils with
various orientations as electrocatalysts and operandoly monitored their surface
reconstructions. Quantitative correlations between Cu atomic configurations and
the selectivities toward multiple products were established. We extracted three
distinct structural descriptors, including crystal facet, atomic coordination number,
and step-terrace angle, to reveal the intrinsic structure-function relationships and
uniquely identify the specific product-producing sites of Cu for CO2R.
Zhu et al., Chem 7, 1–15
February 11, 2021 ª 2020 Elsevier Inc.
https://doi.org/10.1016/j.chempr.2020.10.018

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
Product-Specific Active Site Motifs
of Cu for Electrochemical CO2 Reduction
Chenyuan Zhu,1,8 Zhibin Zhang,2,8 Lixiang Zhong,3,8 Chia-Shuo Hsu,4,8 Xiaozhi Xu,5 Yingzhou Li,6
Siwen Zhao,1 Shaohua Chen,1 Jiayi Yu,1 Shulin Chen,7 Mei Wu,7 Peng Gao,7 Shuzhou Li,3,
*
2,
1,9,
Hao Ming Chen,4, Kaihui Liu, and Liming Zhang
*
*
*
SUMMARY
The Bigger Picture
Electrochemical CO2 reduction
(CO2R) to fuels is a promising
route to close the anthropogenic
carbon cycle and store renewable
energy. Cu is the only metal
catalyst that produces C2+ fuels,
yet challenges remain in the
improvement of electrosynthesis
pathways for highly selective fuel
production. To achieve this,
mechanistically understanding
CO2R on Cu, particularly
Aqueous electrochemical CO2 reduction (CO2R) on Cu can generate
a variety of valuable fuels, yet challenges remain in the improvement
of electrosynthesis pathways for highly selective fuel production.
Mechanistically, understanding CO2R on Cu, particularly identifying
the product-specific active sites, is crucial. Herein, we rationally de-
signed and fabricated nine large-area single-crystal Cu foils with
various surface orientations as electrocatalysts and identified the
voltage- and facet-dependent CO2R selectivities. Operando grazing
incidence X-ray diffraction (GIXRD) and electron back-scattered
diffraction (EBSD) were applied to track the top-surface reconstruc-
tions of Cu, and we correlate the structural evolution with the
change of product selectivities. We extracted three distinct struc-
tural descriptors, including crystal facet, atomic coordination num-
ber, and step-terrace angle, to reveal the intrinsic structure-function
relationships and uniquely identify the specific product-producing
sites for CO2R. Our work guides the rational design of Cu-based
CO2R electrocatalysts and, more importantly, establishes a para-
digm to understand the structure-function correlation in catalysis.
identifying the product-specific
active sites, is crucial. We
rationally designed and fabricated
nine large-area single-crystal Cu
foils with various surface
orientations as electrocatalysts
and monitored their surface
reconstructions using operando
grazing incidence X-ray diffraction
(GIXRD) and electron back-
INTRODUCTION
The direct electrochemical transformation of CO2 to valuable carbon-based fuels is a
promising future technology with the potential to impact large-scale sustainable en-
ergy storage and reduce carbon emissions.1,2 Distinguished from electrochemical
water reduction to produce hydrogen, CO2 reduction (CO2R) has multiple reaction
pathways, leading to a range of reduced carbon fuels, such as C1 (CO, formate
[HCOOꢀ], methane [CH4], etc.), C2 (ethylene [C2H4], ethanol [C2H5OH], etc.), and
C3 (n-propanol [n-C3H7OH], etc.).3,4 To date, significant hurdles regarding the
poor selectivity still need to be overcome if electrochemical CO2R is to become a
viable option for storing renewable electricity.5 To achieve this goal, mechanistically
understanding CO2R, particularly identifying the product-specific active sites, is
crucial.6 Of the reported prevalent electrocatalysts effective for CO2R, Cu is the
only metal catalyst that produces C2+ fuels, which are more desirable as feedstocks
in the chemical industry, with substantial yields.3,7–11 Thus, significant attention is
devoted to unraveling the factors that determine the activity and selectivity of Cu-
based electrocatalysts.
scattered diffraction (EBSD). We
quantitatively established
correlations between the Cu
atomic configurations and the
selectivities toward multiple
products and provide a paradigm
to understand the structure-
function correlation in catalysis.
The complexity of CO2R originates from the multiple electron transfer steps involved
and the strong surface structure sensitivity of the process. A number of theoretical
works have shown that the activation energies of some key steps in CO2R, such as
Chem 7, 1–15, February 11, 2021 ª 2020 Elsevier Inc.
1

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
*CHO formation and *CO dimerization, etc., highly depend on the surface atomic
arrangement.12–14 Previous analysis demonstrated that the C–C coupling pathway
is much favored on Cu(100).15–17 The catalytic roles of Cu+ and the positive effect
of oxygenated surface species on alcohol production have also been proposed.18,19
However, the widely used nano-polycrystalline Cu and partially oxidized Cu catalysts
generally have very complex atomic geometries3,18 and pose extreme challenges to
directly correlate the atomic structure with the product selectivity. More importantly,
the reconstruction of Cu under CO2R operating conditions is vital to the product
generation;10 however, very few of the previous studies monitored the dynamic
change of Cu surface operandoly, which precludes a direct relevance between
real atomic configuration and catalytic performance.20 Recently, in situ electrochem-
ical scanning tunneling microscopy (EC-STM) demonstrated its remarkable ability in
tracking the surface reconstruction of Cu,21,22 yet it is quite time consuming to
obtain a representation of the entire catalyst surface at micro-scale, and severe bub-
ble generation under CO2R disables effective image collection.23
A well-defined surface structure with known composition is essential for the funda-
mental understanding of parameters that control CO2R selectivity. Single-crystal
Cu is a perfect model system to study structure-function correlations.24–27 The
CO2R electrocatalysis on single-crystal Cu was first performed by Frese, who
observed increasing CH4 production on Cu(100), Cu(110), and Cu(111) electrodes.28
Using a galvanostatic method, Hori et al. systematically studied the significant influ-
ence of Cu facet on the selectivities of CO2R products, including CH4, C2H4,
CH3COOH, CH3CHO, and C2H5OH.29 However, the dynamic surface reconstruction
of Cu and its relevance to the change in product selectivities were neglected. In
addition, voltage-dependent selectivity was not mentioned in these works, as Hori
et al. quantified product generation at a constant current density of 5 mA/cm2.
Recently, CO(2)R on single-crystal Cu with low-index facets—Cu(100) and
Cu(111)—has experimentally confirmed that Cu(100) is more active for C–C coupling
and Cu(111) is preferable to CH4 production.16,17,19,26 The Jaramillo group also
1Department of Chemistry, Fudan University,
Shanghai 200438, China
explored CO2R on high-index Cu(751) facet to show the benefit of a kinked surface
for oxygenate generation.26 Despite this progress, the understanding of product-
2State Key Laboratory for Mesoscopic Physics,
Frontiers Science Center for
Nano-Optoelectronics, School of Physics, Peking
specific active sites on Cu is still limited by voltage-dependent product generation
and the operando surface reconstruction of Cu. Distinct surface descriptors showing
the intrinsic structure-function relationships are still elusive. Additionally, the unifor-
mity of constructed monocrystalline facets and the extreme difficulty to synthesize
Cu with exposed high-index facets also pose challenges to understanding CO2R
comprehensively.
University, Beijing 100871, China
3School of Materials Science and Engineering,
Nanyang Technological University, Singapore
639798, Singapore
4Department of Chemistry, National Taiwan
University, Taipei 10617, Taiwan
5Guangdong Provincial Key Laboratory of
Quantum Engineering and Quantum Materials,
School of Physics and Telecommunication
Engineering, South China Normal University,
Guangzhou 510631, China
Here, we establish quantitative correlations between Cu atomic configurations and
the selectivities of multiple products across CO2R potential window. We rationally
designed and fabricated nine large-area single-crystal Cu foils (up to 100 cm2) as
electrocatalysts and monitored their surface reconstructions using operando graz-
ing incidence X-ray diffraction (GIXRD) and electron back-scattered diffraction
(EBSD), which show the diverse stabilities of different Cu facets. We correlated the
structural evolution with the generation of C1 products (e.g., CO, HCOOꢀ, and
CH4) and C2+ products, including C2H4 and oxygenate alcohols (e.g., C2H5OH
and n-C3H7OH), at multiple CO2R potentials. From the perspective of the atomic co-
ordination environment and topological structure, we established three distinct de-
scriptors, including crystal facet, atomic coordination number, and step-terrace
angle, to reveal the intrinsic structure-function correlations and uniquely identify
the specific product-producing sites of Cu for CO2R. Overall, our results demon-
strate a benchmark system to explicitly link Cu atomic configurations to CO2R
6Department of Mathematics, Duke University,
Durham, NC 27708, USA
7Electron Microscopy Laboratory and
International Center for Quantum Materials,
School of Physics, Peking University, Beijing
100871, China
8Those authors contributed equally
9Lead Contact
*Correspondence: lisz@ntu.edu.sg (S.L.),
haomingchen@ntu.edu.tw (H.M.C.),
khliu@pku.edu.cn (K.L.),
zhanglm@fudan.edu.cn (L.Z.)
https://doi.org/10.1016/j.chempr.2020.10.018
2
Chem 7, 1–15, February 11, 2021

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
Figure 1. Large-Area Single-Crystal Cu Foils
(A) A schematic illustration of high-index single-crystal Cu(320) facet having surface atoms with
distinct coordination environments. The side view is shown as an inset. The coordination numbers
of the unsaturated Cu atoms are: six (green), seven (red), nine (blue), and eleven (gray).
(B) An inverse pole figure showing the location of nine monocrystalline Cu facets.
(C) A digital photograph of a large-area single-crystal Cu(100) foil and the representative low-
energy electron diffraction (LEED) images sampled at three random locations.
(D and E) Atomically resolved STEM image (D) and large-scale EBSD mapping (E) of Cu(100) foil.
Inset of (E) shows the pole figure orientation map.
(F and G) Representative XRD 2 theta-scan of low- (F) and high-index (G) single-crystal Cu foils
measured by Cu and Ag targets, respectively.
catalytic performance and provide a paradigm to understand the correlation be-
tween atomic configuration and catalysis.
RESULTS AND DISCUSSIONS
Design and Preparation of Large-Area Single-Crystal Cu Foils
According to the Miller indices {hkl}, Cu facets can be classified into low-index facets
constituting the small library of {100}, {111}, and {110}, and high-index facets, which
constitute terraces, steps, and/or kinks with variable orientations. Using the microfa-
cet notation developed by Somorjai et al. in the 1980’s for cubic structures,30 each
high-index facet can be described as a combination of low-index facets with different
constitution ratios. For instance, Cu(320) can be described as Cu (S)-
[22(110)+11(100)], indicating that there are two (110) and one (100) unit cells in a
(320) unit cell. Thereby, Cu(320) can be visualized as a kinked surface with narrow
(110) terraces and a high density of (100) steps (Figure 1A). The angle between
the terrace and step is 74.5ꢁ. The presence of low-index facets suggests a high de-
gree of site heterogeneity, which can be clearly observed in the inset of Figure 1A,
where the kinks in Cu(320) are shown to have atomic sites with six- (green), seven-
(red), nine- (blue), and eleven-fold (gray) coordination. To elucidate the structure-
function correlation, we rationally designed nine single-crystal Cu facets with distinct
Chem 7, 1–15, February 11, 2021
3

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
coordination environments and site motifs, including three low-index facets, e.g.,
Cu(100), Cu(111), and Cu(110), and six high-index facets, e.g., Cu(331), Cu(311),
Cu(210), Cu(221), Cu(320), and Cu(211). These high-index facets can be denoted
as different combinations of (100), (111), and (110) (Figure S1) and are located on
the inverse pole figure as shown in Figure 1B.
Although previous works have fabricated single-crystal Cu electrode/nanopar-
ticles,17,27,29 the ultra-small single-crystal Cu (< 0.1 cm2) without a high degree of
uniformity and/or the short length scale of nanostructured Cu particles covered
with surfactants make it difficult to accurately quantify the atomic structure distribu-
tions and to correspondingly elucidate the correlations between the structure
configuration and product preference. To address these drawbacks, we prepared
large-format single-crystal Cu foils via our most recently developed ‘‘seeded
abnormal grain growth’’ technique (see more details in Experimental Procedures).31
Notably, this strategy can prepare uniform single-crystal electrodes with a super-
large geometric area (up to 100 cm2), thus offering a significant feasibility for the
electrochemical device integration, particularly on the in situ detection apparatus.
Figure 1C shows an optical image of a large-area single-crystal Cu foil with out-of-
plane textured along <100> direction. The identical rotation angle in the low energy
electron diffraction (LEED) patterns sampled at multiple random positions suggests
the absence of the in-plane misorientation on the surface, which can be further
corroborated by the atomic-resolved scanning transmission electron microscopy
(STEM) image (Figure 1D) and the uniform color contrast of the EBSD mapping (Fig-
ure 1E). X-ray diffraction (XRD) was used to track the crystallinity of the bulk Cu foils.
As shown in Figures 1F and 1G, XRD symmetric scans establish the single crystallinity
nature of fcc Cu foils textured in nine out-of-plane directions, consistent with the
LEED patterns, EBSD mapping, and the correlated TEM images summarized in Fig-
ures S2–S4. All these results demonstrate that we were able to explicitly control the
surface atomic structures of Cu on a decimeter scale.
Single-Crystal Cu Foils for Electrochemical CO2R
The electrocatalytic activities of single-crystal Cu foils were evaluated by performing
electrolysis in a 0.1 M KHCO3 aqueous solution under 1 atm of CO2 in a two-compart-
ment H-cell. Each Cu surface was tested using chronoamperometry at multiple poten-
tials, and gas and liquid products were quantified using gas chromatography (GC)
and nuclear magnetic resonance (NMR) spectroscopy, respectively (see more details
in Experimental Procedures; Figures S5–S7). The major products detected were CO,
HCOOꢀ, CH4, C2H4, C2H5OH, n-C3H7OH, and H2 (produced from the competing pro-
ton reduction). Acetate (CH3COOꢀ), ethylene glycol (HOCH2CH2OH), and propionald-
hyde (CH3CH2CHO) were also observed in intermediate amounts on some single-crystal
Cu electrodes, such as Cu(110), Cu(320), and Cu(211). The selectivity, time-dependent
current density, and partial current density of each product on both low-and high-index
Cu facets were plotted as a function of applied potential, as summarized in Figures S8–
S10. As a competitive reaction, proton reduction dominates at both low and high over-
potentials, e.g., > ꢀ0.7 and < ꢀ1.2 V versus RHE (reversible hydrogen electrode), due to
the weak CO2R activity at lower overpotentials and the mass transfer limitation of CO2
under the large electrochemical polarization, respectively. In the medium potential re-
gion, CO2R kinetically dominates the reaction and is not limited by the supply of
CO2.3,17 To exclude the influence from mass transport, we focused on the medium po-
tential region to understand CO2R kinetics in the following study.
To identify product distribution clearly, we grouped CO2R products as 2eꢀ products
(2 mol of eꢀ are consumed for the generation of product per mol, e.g., CO and
4
Chem 7, 1–15, February 11, 2021

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
Figure 2. Contour Mapping for the Tendency of Multiple Fuel Generation
(A–D) The F.E(s). of HCOOꢀ (A), > 2eꢀ products (B), C2+ (C), and oxygenates (D) at ꢀ0.95 G 0.01 V
(left), ꢀ1.05 G 0.02 V (middle), and ꢀ1.15 G 0.01 V (right) versus RHE on nine single-crystal Cu
facets.
HCOOꢀ) and > 2eꢀ products (> 2 mol of eꢀ are consumed for the generation of
product per mol), including CH4, C2H4, and oxygenates (e.g., C2H5OH, n-
C3H7OH, and CH3COOꢀ, etc.), and utilized the inverse pole figure to map the prod-
uct distribution tendency on nine single-crystal orientations under three selected
potentials (ꢀ0.95, ꢀ1.05, and ꢀ1.15 V versus RHE). Among CO2R products, HCOOꢀ
is considered to be a terminal 2eꢀ pathway formed through a different mechanism
apart from CO and > 2eꢀ products. As shown in Figure 2A, Cu(211) and Cu(311)
are observed to be more preferable for HCOOꢀ production across the potential win-
dow, indicating that Cu-n(111) 3 (100) is a HCOOꢀ-producing site in CO2R and is
impervious to the potential perturbation. Furthermore, comparing the activities be-
tween Cu(311) and (211), we can deduce that the narrower (111) terrace (the smaller
angle between terrace and step), the higher the selectivity for HCOOꢀ production.
To uncover the atomic configuration leading to >2eꢀ products, we then plotted the
total faradic efficiencies (F.Es.) of > 2eꢀ fuels at various potentials on nine Cu facets
in Figure 2B, which in contrast to HCOO-, exhibited a tremendous evolution with
changing voltage. At ꢀ0.95 V versus RHE, the Cu facets containing n(100) 3 (110)
show a higher selectivity toward > 2eꢀ products, consistent with our density func-
tional theory (DFT) calculations showing their lower energy barrier in *CO + H* /
*CHO (rate determining step, r.d.s.) for the production of > 2eꢀ fuels (Figure S11).32
With the negative potential increasing, the preference of > 2eꢀ products on Cu(210)
Chem 7, 1–15, February 11, 2021
5

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
and Cu(320) is diminishing, and n(111)3 (100) dominates > 2eꢀ fuels production at
ꢀ1.15 V versus RHE. In >2eꢀ products, CH4 production also demonstrates an
obvious voltage dependence. The line of n(100) 3 (110) first demonstrated the high-
est selectivity toward CH4, which was further surpassed by Cu(110) and Cu(111) with
the negative bias increasing (Figure S12). We attributed the initial favorable produc-
tion of CH4 on Cu(100) to a higher concentration of adsorbed *CO, as indicated by
the higher CO generation rate on Cu(100) before ꢀ0.85 V versus RHE (Figure S8).
Distinguished from CH4 products, the formation of C2+ in >2eꢀ products requires C–
C coupling of the adsorbed *CO intermediate,16 for instance, *CO + *CHO / *CO-
CHO32 has been identified as the r.d.s. We show that, Cu(100) displays a strong
preference toward C2+ production at different potentials, which can clearly be
seen from the contour mappings of the F.Es. of C2+ (Figure 2C). Theoretical calcula-
tions further corroborate the higher selectivity of C2+ on Cu(100) from the lowest for-
mation barrier of *COCHO (Figure S13). As one important component of C2+ fuels,
C2H4 production positively relates with the fraction ratio of Cu(100) lattice (Fig-
ure S14), consistent with the previous study of Hori et al.7 Last but not least, it is use-
ful to compare the selectivity of liquid oxygenate fuels. The contour mapping shown
in Figure 2D demonstrates that Cu(110) is most active toward oxygenate production
across the potential window, and there also exists a positive correlation between
alcohol selectivity and the fraction ratio of (110) in a high-index Cu unit cell
(Figure S15).
Overall, it is clear that in CO2R electrocatalysis, the selectivity toward multiple prod-
ucts is both voltage- and facet-dependent, and our work is the first report, as far as
we know, to show the two effects simultaneously. However, it is known that Cu sur-
face undergoes kinetic reconstructions at CO2R operating conditions.21,22 To illumi-
nate the product-specific active site motifs, it is desirable to track the time-induced
dynamic evolution of Cu surface and illuminate the surface reconstruction.
Operando Dynamic Surface Reconstruction of Cu
We customized a three-electrode liquid cell (Figure S16) for in situ GIXRD measure-
ments, in which a kapton polyimide film (DuPont) with a thickness of 0.06 mm served
as an X-ray window, and electrolyte thickness was precisely controlled below 1 mm
to suppress the interference from liquid electrolyte. The super-large area of single-
crystal Cu foils provides a significant feasibility for the electrochemical device inte-
gration. Comparing with the in situ EC-STM, which is generally used to characterize
the local structural evolution of surface, in situ GIXRD is capable of showing the
average structural evolution in a large area and providing a time-saving quantitative
analysis strategy to correlate the real surface atomic configuration with the
measured catalytic activity and selectivity toward multiple products. By controlling
the incident angle below 0.5ꢁ, GIXRD can probe the structural information of Cu sur-
face within a depth of 5 nm. We tracked the evolution of surface orientation on low-
index Cu(111), Cu(110), and Cu(100) during chronoamperometric CO2R electroca-
talysis (Figures 3A and S17). Cu(111) and Cu(110) were observed to be quite robust
in the time scale of 3 h under a negative potential of ꢀ1.15 V versus RHE, indicative
of the strong durability of Cu(111) and Cu(110) in CO2R. For Cu(100), however, we
observed a partial structural transformation. A new XRD peak assigned to Cu(111)
appeared on Cu(100) in less than 30 min, and stabilized quantificationally at ~15%
(I(111)/I(100)) after a 3-h electrolysis (Figure 3B). The increased ratio of the (111) unit
cell can well explain the increased CH4 production and decreased C2H4 generation
at a higher negative bias on Cu(100), as experimentally demonstrated in Figure S8,
as well as the second pathway for C2H4 production observed previously.25
6
Chem 7, 1–15, February 11, 2021

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
Figure 3. Dynamic Reconstruction of Cu Surface under CO2R
(A) Operando GIXRD images of single-crystal Cu(111), Cu(110), and Cu(100) (from top to bottom) at ꢀ1.15 V versus RHE during CO2R.
(B) Quantitative analysis of Cu(111), Cu(110), and Cu(100) (from top to bottom) surface evolution at ꢀ1.15 V versus RHE under CO2R.
(C) EBSD images (with facet decomposition) of six high-index Cu facets before and after CO2 electrocatalysis at ꢀ1.15V versus RHE.
We further applied EBSD to characterize the surface reconstruction of high-index Cu
facets after CO2R electrolysis at ꢀ1.15 V versus RHE ex situ, which shows that the
fraction ratio of Cu(111) increases on a majority of facets. As listed in Figure 3C,
the outer planes of Cu(331), Cu(221), Cu(210), and Cu(320) were transformed to
Cu(653), Cu(553), Cu(641) and Cu(661), respectively (see structural models in Fig-
ure S18). The increasing ratio of Cu(111) explicitly demonstrates that the closely
packed nine-coordinated Cu(111) surface is the most stable facet under CO2R.
This result is consistent with our calculation shown in Figure S19, which confirms
the lowest surface energy of Cu(111). The structural evolution can correlate well
with the changing of product selectivities under a higher negative bias, for instance,
the increased CH4 and decreased alcohol production on Cu(210), Cu(320), Cu(331),
and Cu(221) probably arise from the increased constitution ratio of Cu(111). In stark
contrast to these high-index facets undergoing reconstruction, Cu(211) and Cu(311)
were unreconstructed after electrocatalysis, consistent with their impervious
HCOOꢀ preference to the potential perturbation.
Product-Specific Active Site Motifs of Cu for CO2R
Based on the aforementioned results, we aim to extract the product-specific active
site motifs of Cu for CO2R. As the product distribution highly depends on the exact
nature of the Cu surface, the structural reconstruction under a higher negative bias
should be taken into account. We are aware that the catalysis micro-environment,
such as cations in electrolyte,33 pH values,34 reduction temperature,35 etc., also
Chem 7, 1–15, February 11, 2021
7

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
Figure 4. Pearson Correlation Map between Cu Atomic Configurations and Multiple Fuel
Generation
(A) Schematic illustration of the atomic coordination environment and topology on an ideal single-
crystal Cu surface.
(B–D) Pearson correlation coefficient (from ꢀ1 to 1) showing the correlation of lattice facet (B),
coordination number of surface atoms (C), and step-terrace angle (D) with the specific product
selectivity at ꢀ0.95 G 0.01 V (left column) and ꢀ1.15 G 0.01 V versus RHE (right column),
respectively.
influence the product selectivity; however, we aim to reveal the intrinsic structure-
function relationships of Cu for CO2R in this work. Here, we avoid studying the com-
plex reaction pathways of CO2R and focus instead on descriptors for the structure-
dependent product distribution. As depicted in Figure 4, we show the correlations
between atomic configurations and various product generation under both ꢀ0.95
(left column) and ꢀ1.15 V versus RHE (right column). At ꢀ1.15 V versus RHE, we uti-
lized the high-index facets Cu(311), Cu(211), Cu(653), Cu(553), Cu(641), and Cu(661)
formed after surface reconstruction to extract the surface descriptors. As schemati-
cally shown in Figure 4A, we established three distinct descriptors from the perspec-
tive of the atomic coordination environment and topological structure, including
crystal facet, atomic coordination number, and step-terrace angle, to reveal the
8
Chem 7, 1–15, February 11, 2021

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
intrinsic structure-function correlations and uniquely identify the specific product-
producing sites of Cu for CO2R. The correlation is calculated as the Pearson correla-
tion coefficient (giving values between ꢀ1 and 1) showing the correlation between
two variables, in which 1 is a perfect positive linear correlation, 0 is no correlation,
and ꢀ1 is a perfect negative linear correlation.36
As the high-index facet can be described as a combination of low-index facets with
different constitution ratios, we first calculated the fraction ratio of Cu(100), Cu(111),
and Cu(110), respectively, in all high-index facets before and after surface recon-
struction. Figure 4B demonstrates the correlation between the facets and the prod-
uct-specific selectivities. We first focus on the similarities in correlation under two
different potentials. Regardless of the applied voltage, Cu(111), Cu(100), and
Cu(110) show an obvious positive correlation with the production of HCOOꢀ,
C2H4, and C2H5OH, respectively. These results suggest that to design a highly active
Cu catalyst for HCOOꢀ, C2H4, and C2H5OH, the Cu surface should have more
Cu(111), (100), and (110) lattice, respectively. We also observed that the active facet
for CH4 production is potential dependent. Cu(100) is more active for CH4 produc-
tion at a lower negative bias, whereas Cu(111) is more active at a higher negative
bias. This transformation may be attributed to the changing interactions between
the intermediates and Cu atoms or variable reaction pathways under different po-
tentials. Beyond that, we found Cu(111) negatively correlates with both C2H4 and
C2H5OH across the potential window, Cu(100) inhibits HCOOꢀ and C2H5OH gener-
ation, and Cu(110) is unbeneficial for the production of CO.
These experimental results are consistent with our theoretical calculations on three
low-index Cu facets (Figures S20–S23). The selectivity between HCOOꢀ and CO
were explored on three low-index Cu facets using DFT calculations. As shown in Fig-
ure S20, the maximum energy barrier along the free-energy profiles for HCOOꢀ and
CO generation determines the dominating reaction pathway, and the point devi-
ating from the diagonal line indicates the preferable generation of either CO or
HCOOꢀ. These analyses show that in three low-index facets, Cu(111) is apt to pro-
duce HCOOꢀ, which is consistent with our experimental observations (Figure S8).
In addition, the reaction barrier for *CO-*CHO coupling is only 0.55 eV on
Cu(100), whereas it is much higher on Cu(111) (0.75 eV) and Cu(110) (0.81 eV) (Fig-
ure S21), confirming the preferable C2H4 production on Cu(100). Furthermore, the
key step for C2H5OH production, reducing *HCCOH to *HCCHOH,37 is highly favor-
able in energetics on Cu(110) with an energy drop of 0.66 eV (Figure S22)—much
more significant than those on other facets. Interestingly, in contrast to Hori
et al.’s results that Cu(111) is the most active facet for CH4, we only observed a pos-
itive correlation between Cu(111) and CH4 generation under a high negative bias.
Instead, at lower negative bias, we observed Cu(100) is preferable toward CH4 gen-
eration. The energy barrier for CH4 production on three low-index facets was ex-
tracted using DFT calculations (Figure S23), which demonstrates a smaller energy
barrier on Cu(100) for the key step of CH4 formation, *CO to *CHO, consistent
with our experimental observations (Figure S12).32
Next, we correlated the product-specific selectivities with the surface coordination
number (CN) of Cu atoms, which is defined as the number of nearest neighboring
atoms (Figure 4C). Atoms that have an 8-fold coordination show a strong positive
correlation with C2H4 generation across the full potential window, consistent with
the favorable C2H4 production on Cu(100). HCOOꢀ production was much more pref-
erable on Cu atoms that have a CN of seven and ten under both potentials but was
suppressed on the atoms having a CN of eight. Instead, C2H5OH production could
Chem 7, 1–15, February 11, 2021
9

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
be promoted on a more ‘‘open’’ surface, which consist of more Cu atoms having a
smaller CN, such as six and seven. In addition, voltage-dependent CH4 production
could also be observed, which displayed a higher activity on atoms that have a CN of
eight and nine at a lower and higher negative bias, respectively.
In addition to the atomic coordination environment, we also extracted the correla-
tion between the product tendency and step-terrace angle, from the perspective
of topological structure (Figure 4D). Across the full potential window, a larger
step-terrace angle is beneficial for C2H5OH generation, on the contrary, CH4 pro-
duction is preferable on the Cu surface having a smaller step-terrace angle. These
results demonstrate that both the atomic coordination environment and topological
structure are capable of influencing the product distribution of CO2R on the Cu
catalyst.
Conclusion
In conclusion, we unambiguously demonstrated a correlation between Cu atomic
configurations and the selectivities toward multiple fuels. We rationally designed
and fabricated large-area single-crystal Cu foils with various surface orientations,
tracked the dynamic top-surface reconstruction of Cu facets under CO2R, and re-
vealed the product-specific active sites of Cu toward CO2R by correlating the atomic
structures with the product selectivities. Our work demonstrates a benchmark sys-
tem for CO2R electrocatalysis and provides an important mechanistic insight that
could further guide the rational design and optimization of Cu-based CO2R cata-
lysts. This analysis method can be straightforwardly extended to other single-crystal
electrodes, as well as other chemistries, to unravel the active site motifs of catalysts.
EXPERIMENTAL PROCEDURES
Resource Availability
Lead Contact
Further information and requests for resources and reagents should be directed to
the Lead Contact, Liming Zhang (zhanglm@fudan.edu.cn).
Materials Availability
The materials generated in this study will be made available on request.
Data and Code Availability
All of the data are available from the corresponding author upon reasonable
request.
Seeded Growth of Single-Crystal Cu Foils
We followed our most recently designed ‘‘seeded abnormal grain growth’’ tech-
nique to successfully fabricate the large-area single-crystal Cu foils with nine low-
and high-index facets from commercially available polycrystalline Cu foils.31 Typi-
cally, we first pre-oxidized the polycrystalline Cu foil in air to generate a large grain
seed with a low- or high-index facet and further annealed it in a reducing atmosphere
to enable the abnormal grain growth of the seed, which subsequently turns the
whole foil into a single crystal. The detailed preparation procedure is as follows:
we first loaded the Cu foil (25 mm-thick, 99.8%, Sichuan Oriental Stars Trading Co.
Ltd., #Cu-1031) into a chemical vapor deposition (CVD) furnace (Tianjin Kaiheng
Co. Ltd.) using a quartz substrate. Through mild heating, the Cu foil was oxidized
at 150ꢁC–650ꢁC in air for 1–4 h. After the pre-oxidation, the system was heated to
1,020ꢁC in 1 h with 800 sccm Ar and 50 sccm H2, then the Cu foil was annealed at
1,020ꢁC for 3–10 h under the same atmosphere, and finally, the system was naturally
10
Chem 7, 1–15, February 11, 2021

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
cooled down to room temperature to obtain large-scale single-crystal Cu foils (the
system was kept at atmospheric pressure throughout the annealing process). By
repeating the typical annealing procedure, nine kinds of single-crystal Cu foils can
be produced.
Physical Characterization
LEED measurements were performed using Omicron LEED system in ultra-high vac-
uum with base pressure below 3 310ꢀ7 Pa. STEM experiments were carried out on
FEI Titan Themis G2 300 operated at 300 keV. EBSD characterization was performed
using PHI 710 Scanning Auger Nanoprobe. The phase purity of each sample was
examined by XRD 2q-scan measurements using a Bruker D8 Advanced system
˚
˚
with Cu (Cu Ka radiation, l = 1.54 A) and Ag (Ag Ka radiation, l = 0.56 A) targets
for low- and high-index facet characterization, respectively.
Electrochemical CO2 Reduction
Pretreatment of Cu Single-Crystal Electrodes
The single-crystal Cu foils were first physically polished with alumina slurries (0.3 mm,
CH instrument) to mirror-like finishes and were then electropolished in a H3PO4
aqueous electrolyte at a constant potential of 1.5 V versus Ag/AgCl for 90 s.
CO2R Electrochemical Measurements
The electrocatalytic characterizations were carried out in a customized H-cell, which
has two compartments separated by an anion exchange membrane (Fumatech com-
pany, Fumasep FAA-3-PK130). Each compartment contained 6 mL of electrolyte
(0.1 M KHCO3 aqueous solution, pH 6.8), and the compartment holding the working
electrode was sealed to measure gaseous products. The electrolyte was purged with
CO2 at a constant flow rate of 20 sccm for 20 min before and during each measure-
ment while stirring. A platinum wire and a Ag/AgCl (CHI, 3M KCl) were used as
counter and reference electrodes, respectively. The potentials measured were con-
verted to RHE scale by V (versus RHE) = V (versus Ag/AgCl, 3M KCl) + 0.210 V +
0.0591 3 pH. The solution resistance was compensated for 85% by the potentiostat
and the rest 15% was post-corrected.
All the electrochemical measurements were explored using a Biologic-SP300 poten-
tiostat. Before electrolysis, linear sweep voltammetry (at a scan rate of 50 mV/s) was
performed and then a set potential was applied for chronoamperometry (CA). Dur-
ing the constant potential electrolysis (1 h), the gas products were quantified using a
GC (Agilent 7890B; column HP-PLOT). Samples for GC were collected at 20-min in-
tervals, and the separated gas products were analyzed by both a thermal conductiv-
ity detector (for H2) and a flame ionization detector (for CO and hydrocarbons).
Liquid products were analyzed afterward by 500 MHz NMR (AVANCE III HD) using
dimethyl sulfoxide and phenol as an internal standard. Solvent presaturation tech-
nique was implemented to suppress the water peak. Faradic efficiencies were calcu-
lated from the amount of charge passed to produce each product, divided by the
total charge passed at a specific time (gas) or during the overall run (liquid).
Calculation of the faradic efficiency of gas products:
À
ꢀ
Á
F
flow 3 Cgas Vm 3 n 3 F
F:E:ðgasÞ =
3 100
I
total 3 60
Calculation of the faradic efficiency of liquid products:
Chem 7, 1–15, February 11, 2021
11

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
À
Á
C
liquid 3 V 3 n 3 F
F:E: liquid
=
3 100
Qtotal
F.E. (gas): faradic efficiency of gas product, %
F.E. (iquid): faradic efficiency of liquid product, %
F
flow: flow rate of CO2, mL minꢀ1
Cgas: volume ratio of gas product, determined by on-line GC
Vm: the molar volume of an ideal gas at 1 atmosphere of pressure (molar volume at
normal temperature and pressure is 0.0245 mL), mL molꢀ1
Cliquid: concentration of liquid product after 1 h of electrolysis, determined by NMR,
mol Lꢀ1
V: volume of the electrolyte in the working cell, L
I
total: steady-state cell current
total: total charge in 1 h of bulk electrolysis, C
Q
n: number of transferred electrons for certain product
F: Faradic constant, 96,485 C molꢀ1
Operando GIXRD Measurement
The GIXRD of the Cu foils were performed at the 12B beamline of Japan, in which
the ring was operated at energy 8 GeV with a typical current of 100 mA. Two pairs
of slits and one collimator were set up in experimental hutch to provide a colli-
mated beam with a beam size of 300 mm at the sample position. An incident angle
of 0.5ꢁ was employed in the present study, in which the X-ray irradiation can pass
through 5 nm upon surface of sample. The geometry allows us to significantly
enhance the X-ray scattering intensity from the surface of Cu foils. The grazing
angle was optimized to probe the interior of the sample surface and to achieve
the best S/N ratio for maximizing the scattering signal from the electrode surface.
˚
In the present study, the wavelength of the incident X-ray was 0.6909 A (16 keV),
delivered from the 5 T Superconducting Wavelength Shifter and a Si (111) trian-
gular crystal monochromator, which allowed an optimized condition for distin-
guishing scattering rings to be achieved. The diffraction pattern was collected
with a MX225-HE Rayonix CCD detector that was located ~195 mm behind the
sample, while the total pixel was 3,072 3 3,072 and the pixel size of image was
73 mm with a typical exposure duration of 1 s. To achieve a correct condition,
the diffraction angles were calibrated according to Bragg positions of CeO2
(SRM 674b) standards in desired geometry, and then a program of GSAS II was
employed to obtain the corresponding one-dimensional powder diffraction profile
with cake-type integration. In terms of the liquid cell, a kapton polyimide film (Du-
Pont) with a thickness of 0.06 mm served as X-ray window, and the thickness of
electrolyte was precisely controlled below 1 mm to suppress the interference
from liquid electrolyte.
12
Chem 7, 1–15, February 11, 2021

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
Computational Methods
All the calculations were performed using DFT as implemented in the Vienna ab ini-
tio simulation package (VASP).38 The ion-electron interactions were treated with the
projected augmented wave (PAW) pseudopotentials,39 and the plane-wave basis set
was cut off at 400 eV. The general gradient approximation (GGA) parameterized by
Perdew, Burke, and Ernzerhof (PBE) was adopted to describe the exchange-correla-
tion functional in structural relaxations.40 All structures were fully relaxed by a con-
jugate gradient method until the residual force component was less than 0.02 eV/
˚
A, and the convergence criterion of total energy in the self-consistent field method
was set to 10ꢀ5 eV. The optimized lattice constant of Cu is 3.627 A. The slab model
˚
was used to simulate Cu surfaces, where the thickness of the vacuum was larger than
˚
15 A to make sure that there was no superficial interaction between different layers.
The size of the super cells was larger than 10 A, and the k-point grids (~0.2 Aꢀ1) used
for the Brillouin-zone integration were sampled by the Monkhorst-Pack scheme.41
Larger cells or denser k-points yielded the same results in our test calculations.
The climbing-image nudged elastic band method was used to locate the transition
state and determine the activation energy.42 The Gibbs free energies (DG) of
different intermediates were calculated to characterize the CO2R performance on
Cu facets. Chemisorption energies of the reaction intermediates were calculated by:
˚
˚
DEadsorbate = Etotal – Esubstrate ‒ Eadsorbate
,
where Etotal and Esubstrate are the DFT energies for the system with and without adsor-
bate, respectively, and Eadsorbate is calculated relative to CO2, H2O, and H2. The cor-
responding free energy of chemisorption was then calculated by correcting for the
zero-point vibrational energy EZPE, entropy TS, and solvation energy Esol
.
DGadsorbate = DEadsorbate + D (EZPE – TS + Esol),
where DGadsorbate was used to construct the free-energy profiles. The solvation ef-
fects were simulated by using the Poisson-Boltzmann continuum-solvation model
in the VASPsol code.43
SUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.chempr.
2020.10.018.
ACKNOWLEDGMENTS
This research was supported by the National Natural Science Foundation of China
(grants 21872039, 51991340, and 51991342), Science and Technology Commission of
Shanghai Municipality (grant 18JC1411700), National Key Research and Development
Program of China (2016YFA0300903 and 2016YFA0300804), Beijing Natural Science
Foundation
(JQ19004),
Beijing
Excellent
Talents
Training
Support
(2017000026833ZK11), Beijing Municipal Science
&
Technology Commission
(Z191100007219005), Beijing Graphene Innovation Program (Z181100004818003),
and the Key Research and Development Program of Guangdong Province
(2020B010189001, 2019B010931001, and 2018B030327001). We sincerely thank Dr.
Bing Deng, Prof. Hailin Peng, and Prof. Zhongfan Liu for providing some low-index sin-
gle-crystal Cu electrodes when initiating the work.
AUTHOR CONTRIBUTIONS
L.Z. and C.Z. designed and conceived the experiment. C.Z., S.C., S.Z., and J.Y. fabri-
cated the electrodes and performed the electrochemical characterization and data
Chem 7, 1–15, February 11, 2021
13

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
analysis. B.Z. and K.L. carried out the synthesis of single-crystal Cu foils and LEED
characterization. C.S.H. and H.M.C. performed the operando GIXRD measure-
ments. L.Z. and S.L. worked on the DFT calculations. S.C., M.W., and P.G. carried
out the TEM characterization. Y.L. helped with the modeling and data fitting. All au-
thors discussed the results and participated in writing the manuscript.
DECLARATION OF INTERESTS
The authors declare no competing interests.
Received: June 23, 2020
Revised: July 24, 2020
Accepted: October 17, 2020
Published: November 11, 2020
REFERENCES
1. Seh, Z.W., Kibsgaard, J., Dickens, C.F.,
Chorkendorff, I.B., Nørskov, J.K., and
electroreduction of CO2 to C2-C3 products.
Proc. Natl. Acad. Sci. USA 114, 10560–10565.
20. Handoko, A.D., Wei, F., Jenndy, Yeo, B.S., and
Seh, Z.W. (2018). Understanding
Jaramillo, T.F. (2017). Combining theory and
experiment in electrocatalysis: insights into
materials design. Science 355, eaad4998.
heterogeneous electrocatalytic carbon dioxide
reduction through operando techniques. Nat.
Catal. 1, 922–934.
11. Zhuang, T.T., Pang, Y., Liang, Z.-Q., Wang, Z.,
Li, Y., Tan, C.-S., Li, J., Dinh, C.T., De Luna, P.D.,
Hsieh, P., et al. (2018). Copper nanocavities
confine intermediates for efficient
2. Ross, M.B., De Luna, P.D., Li, Y.F., Dinh, C.T.,
Kim, D., Yang, P., and Sargent, E.H. (2019).
Designing materials for electrochemical
carbon dioxide recycling. Nat. Catal. 2,
648–658.
21. Kim, Y.G., Baricuatro, J.H., Javier, A., Gregoire,
J.M., and Soriaga, M.P. (2014). The evolution of
the polycrystalline copper surface, first to
Cu(111) and then to Cu(100), at a fixed CO₂RR
potential: a study by operando EC-STM.
Langmuir 30, 15053–15056.
electrosynthesis of C3 alcohol fuels from
carbon monoxide. Nat. Catal. 1, 946–951.
12. Peterson, A.A., Abild-Pedersen, F., Studt, F.,
Rossmeisl, J., and Nørskov, J.K. (2010). How
copper catalyzes the electroreduction of
carbon dioxide into hydrocarbon fuels. Energy
Environ. Sci. 3, 1311–1315.
3. Kuhl, K.P., Cave, E.R., Abram, D.N., and
Jaramillo, T.F. (2012). New insights into the
electrochemical reduction of carbon dioxide
on metallic copper surfaces. Energy Environ.
Sci. 5, 7050–7059.
22. Kim, Y.G., Javier, A., Baricuatro, J.H., Torelli, D.,
Cummins, K.D., Tsang, C.F., Hemminger, J.C.,
and Soriaga, M.P. (2016). Surface
13. Garza, A.J., Bell, A.T., and Head-Gordon, M.
(2018). Mechanism of CO2 reduction at copper
surfaces: pathways to C2 products. ACS Catal.
8, 1490–1499.
reconstruction of pure-Cu single-crystal
electrodes under CO-reduction potentials in
alkaline solutions: a study by seriatim ECSTM-
DEMS. J. Electroanal. Chem. 780, 290–295.
4. Gao, D., Ara´n-Ais, R.M., Jeon, H.S., and Roldan
Cuenya, B. (2019). Rational catalyst and
electrolyte design for CO2 electroreduction
towards multicarbon products. Nat. Catal. 2,
198–210.
14. Montoya, J.H., Shi, C., Chan, K., and Nørskov,
J.K. (2015). Theoretical insights into a CO
dimerization mechanism in CO2
electroreduction. J. Phys. Chem. Lett. 6, 2032–
2037.
23. Pfisterer, J.H.K., Liang, Y., Schneider, O., and
Bandarenka, A.S. (2017). Direct instrumental
identification of catalytically active surface
sites. Nature 549, 74–77.
5. Bushuyev, O.S., De Luna, P.D., Dinh, C.T., Tao,
L., Saur, G., van de Lagemaat, J., Kelley, S.O.,
and Sargent, E.H. (2018). What should we make
with CO2 and how can we make it? Joule 2,
825–832.
24. Bagger, A., Ju, W., Varela, A.S., Strasser, P., and
Rossmeisl, J. (2019). Electrochemical CO2
reduction: classifying Cu facets. ACS Catal. 9,
7894–7899.
15. Wang, Y., Wang, Z., Dinh, C.-T., Li, J., Ozden,
A., Golam Kibria, M.G., Seifitokaldani, A., Tan,
C., Gabardo, C.M., Luo, M., et al. (2020).
6. Birdja, Y.Y., Pe´ rez-Gallent, E., Figueiredo,
M.C., Gottle, A.J., Calle-Vallejo, F., and Koper,
¨
M.T.M. (2019). Advances and challenges in
understanding the electrocatalytic conversion
of carbon dioxide to fuels. Nat. Energy 4,
732–745.
Catalyst synthesis under CO2 electroreduction
favours faceting and promotes renewable fuels
electrosynthesis. Nat. Catal. 3, 98–106.
25. Schouten, K.J., Qin, Z., Pe´ rez Gallent, E., and
Koper, M.T.M. (2012). Two pathways for the
formation of ethylene in CO reduction on
single-crystal copper electrodes. J. Am. Chem.
Soc. 134, 9864–9867.
16. Loiudice, A., Lobaccaro, P., Kamali, E.A., Thao,
T., Huang, B.-H., Ager, J.W., and Buonsanti, R.
(2016). Tailoring copper nanocrystals towards
C2 products in electrochemical CO2 reduction.
Angew. Chem. Int. Ed. 55, 5789–5792.
7. Kuhl, K.P., Hatsukade, T., Cave, E.R., Abram,
D.N., Kibsgaard, J., and Jaramillo, T.F. (2014).
Electrocatalytic conversion of carbon dioxide
to methane and methanol on transition metal
surfaces. J. Am. Chem. Soc. 136, 14107–14113.
26. Hahn, C., Hatsukade, T., Kim, Y.G., Vailionis, A.,
Baricuatro, J.H., Higgins, D.C., Nitopi, S.A.,
Soriaga, M.P., and Jaramillo, T.F. (2017).
Engineering Cu surfaces for the electrocatalytic
conversion of CO2: controlling selectivity
toward oxygenates and hydrocarbons. Proc.
Natl. Acad. Sci. USA 114, 5918–5923.
17. Huang, Y., Handoko, A.D., Hirunsit, P., and
Yeo, B.S. (2017). Electrochemical reduction of
CO2 using copper single-crystal surfaces:
effects of CO* coverage on the selective
formation of ethylene. ACS Catal. 7, 1749–
1756.
8. Kortlever, R., Shen, J., Schouten, K.J.P., Calle-
Vallejo, F., and Koper, M.T.M. (2015). Catalysts
and reaction pathways for the electrochemical
reduction of carbon dioxide. J. Phys. Chem.
Lett. 6, 4073–4082.
27. Suen, N., Kong, Z., Hsu, C., Chen, H., Tung, C.,
Lu, Y., Dong, C., Shen, C., Chung, J., and Chen,
H.M. (2019). Morphology manipulation of
copper nanocrystals and product selectivity in
the electrocatalytic reduction of carbon
dioxide. ACS Catal. 9, 5217–5222.
18. Li, C.W., and Kanan, M.W. (2012). CO2
reduction at low overpotential on Cu
9. Hori, Y., Takahashi, I., Koga, O., and Hoshi, N.
(2003). Electrochemical reduction of carbon
dioxide at various series of copper single
crystal electrodes. J. Mol. Catal. A Chem. 199,
39–47.
electrodes resulting from the reduction of thick
Cu2O films. J. Am. Chem. Soc. 134, 7231–7234.
19. Jiang, K., Sandberg, R.B., Akey, A.J., Liu, X.,
Bell, D.C., Nørskov, J.K., Chan, K., and Wang,
H. (2018). Metal ion cycling of Cu foil for
selective C-C coupling in electrochemical CO2
reduction. Nat. Catal. 1, 111–119.
28. Frese, K.W., Jr. (1993). In Electrochemical and
Electrocatalytic Reactions of Carbon Dioxide,
B.P. Sullivan, K. Krist, and H.E. Guard, eds.
(Elsevier).
10. Kim, D., Kley, C.S., Li, Y., and Yang, P. (2017).
Copper nanoparticle ensembles for selective
14
Chem 7, 1–15, February 11, 2021

Please cite this article in press as: Zhu et al., Product-Specific Active Site Motifs of Cu for Electrochemical CO2 Reduction, Chem (2020), https://
doi.org/10.1016/j.chempr.2020.10.018
ll
Article
29. Hori, Y., Takahashi, I., Koga, O., and Hoshi, N.
(2002). Selective formation of C2 compounds
from electrochemical reduction of CO2 at a
series of copper single crystal electrodes.
J. Phys. Chem. B 106, 15–17.
and concentration on electrochemical
reduction of CO2. ChemElectroChem 5, 1064–
1072.
calculations using a plane-wave basis set. Phys.
Rev. B Condens. Matter 54, 11169–11186.
39. Blochl, P.E. (1994). Projector augmented-wave
¨
method. Phys. Rev. B Condens. Matter 50,
17953–17979.
34. Lim, C.F.C., Harrington, D.A., and Marshall,
A.T. (2017). Effects of mass transfer on the
electrocatalytic CO2 reduction on Cu.
Electrochim. Acta 238, 56–63.
30. Van Hove, M.A., and Somorjai, G.A. (1980). A
new microfacet notation for high-Miller-index
surfaces of cubic materials with terrace, step
and kink structures. Surf. Sci. 92, 489–518.
40. Perdew, J.P., Burke, K., and Ernzerhof, M.
(1996). Generalized gradient approximation
made simple. Phys. Rev. Lett. 77, 3865–3868.
35. Ahn, S.T., Abu-Baker, I., and Palmore, G.T.R.
(2017). Electroreduction of CO2 on
41. Monkhorst, H.J., and Pack, J.D. (1976). Special
points for Brillouin-zone integrations. Phys.
Rev. B 13, 5188–5192.
31. Wu, M., Zhang, Z., Xu, X., Zhang, Z., Duan, Y.,
Dong, J., Qiao, R., You, S., Wang, L., Qi, J., et al.
(2020). Seeded growth of large single-crystal
copper foils with high-index facets. Nature 581,
406–410.
polycrystalline copper: effect of temperature
on product selectivity. Catal. Today 288, 24–29.
36. Pearson, K. (1920). Notes on the history of
42. Henkelman, G., Uberuaga, B.P., and Jo´ nsson,
H. (2000). A climbing image nudged elastic
band method for finding saddle points and
minimum energy paths. J. Chem. Phys. 113,
9901–9904.
correlation. Biometrika 13, 25–45.
32. Todorova, T.K., Schreiber, M.W., and
Fontecave, M. (2020). Mechanistic
understanding of CO2 reduction reaction
(CO2RR) toward multicarbon products by
heterogeneous copper-based catalysts. ACS
Catal. 10, 1754–1768.
37. Li, Y.C., Wang, Z., Yuan, T., Nam, D.-H., Luo, M.,
Wicks, J., Chen, B., Li, J., Li, F., de Arquer,
F.P.G., et al. (2019). Binding Site diversity
promotes CO2 electroreduction to ethanol.
J. Am. Chem. Soc. 141, 8584–8591.
43. Mathew, K., Sundararaman, R., Letchworth-
Weaver, K., Arias, T.A., and Hennig, R.G. (2014).
Implicit solvation model for density-functional
study of nanocrystal surfaces and reaction
pathways. J. Chem. Phys. 140, 084106.
38. Kresse, G., and Furthmuller, J. (1996). Efficient
¨
33. Resasco, J., Lum, Y., Clark, E., Zeledon, J.Z.,
and Bell, A.T. (2018). Effects of anion identity
iterative schemes for ab initio total-energy
Chem 7, 1–15, February 11, 2021
15