Hidden Mechanism Behind the Roughness-Enhanced Selectivity of Carbon Monoxide Electrocatalytic Reduction

Article abstract of DOI:10.1002/anie.202016332

High roughness has been proved to be an effective design strategy for electrocatalyst in many systems. Especially, high selectivity of carbon monoxide reduction (CORR) in competition with the hydrogen evolution reaction has been observed on high roughness electrocatalysts. However, the two well-known mechanisms, i.e., decreasing the energy barrier of CORR and increasing local pH, failed to understand the roughness-enhanced selectivity in a recent experiment. Herein we unravel the hidden mechanism by establishing a comprehensive kinetic model for CORR on catalysts with different roughness factors. We conclude that the roughness-enhanced CORR selectivity is actually kinetic controlled by local-electric-field-directed mass transfer of adsorbed species on the electrode surface. Several ways to optimize CORR selectivity are predicted. Our work highlights the kinetics in electrocatalysis on nanocatalysts, and provides a conceptually new principle for future catalyst design.

Full text of DOI:10.1002/anie.202016332

Angewandte
Communications
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 11133–11137
International Edition: doi.org/10.1002/anie.202016332
German Edition: doi.org/10.1002/ange.202016332
CO Reduction
Hot Paper
Hidden Mechanism Behind the Roughness-Enhanced Selectivity of
Carbon Monoxide Electrocatalytic Reduction
Yinghuan Liu, Huijun Jiang,* and Zhonghuai Hou*
Abstract: High roughness has been proved to be an effective
design strategy for electrocatalyst in many systems. Especially,
high selectivity of carbon monoxide reduction (CORR) in
competition with the hydrogen evolution reaction has been
observed on high roughness electrocatalysts. However, the two
well-known mechanisms, i.e., decreasing the energy barrier of
CORR and increasing local pH, failed to understand the
roughness-enhanced selectivity in a recent experiment. Herein
we unravel the hidden mechanism by establishing a compre-
hensive kinetic model for CORR on catalysts with different
roughness factors. We conclude that the roughness-enhanced
CORR selectivity is actually kinetic controlled by local-
electric-field-directed mass transfer of adsorbed species on
the electrode surface. Several ways to optimize CORR
selectivity are predicted. Our work highlights the kinetics in
electrocatalysis on nanocatalysts, and provides a conceptually
new principle for future catalyst design.
RF. The failure of both CORR-activity-promoting and local-
pH-elevating mechanisms implies a new mechanism under-
lying the RF-enhanced CORR selectivity.[24] Unraveling this
novel mechanism would provide insight into easy-taking
design principles for catalysts with enhanced selectivity.
Here, we address the mechanistic issue by establishing
a comprehensive reaction-diffusion model of CORR on
multisite catalysts with constant local pH. Key kinetic
processes are included, such as adsorption, desorption, sur-
face reactions, diffusion, and specifically the directional mass
transfer due to curvature-induced local electric field.[28,29]
Simulations successfully reproduce the key experimental
finding that catalysts with high RF (i.e., surface with large
number of high curvature sites (HC sites)) would efficiently
enhance the multi-carbon selectivity.[24] Detailed analysis
shows that distribution of surface OHꢀ is redistributed by
field-induced mass transfer, leading to high OHꢀ coverage at
HC sites for high RF catalysts. HER activity is therefore
suppressed and results in an enhanced CORR selectivity. In
other words, the RF-induced high CORR performance is
actually kinetic-controlled by directional field-induced mass
transfer on the rough surface. Serval approaches are sub-
sequently predicted to optimize the selectivity of carbon
feedstock by modifying the kinetics.
Converting CO2 into value-added multi-carbon feedstock by
electrochemical reactions is considered to be a promising way
of mitigating the growing energy demands and reducing CO2
in the atmosphere.[1–7] CO2 reduction reaction consists of
a 2eꢀ process from CO2 to CO and a following surface
reaction from CO to multi-carbon products.[8–10] As the CO2 to
CO process has been well established, unraveling the
mechanism underlying CO reduction reaction (CORR) with
high selectivity is then of particular importance.[11–15] Never-
theless, as the equilibrium potential of CORR is close to that
of hydrogen evolution reaction (HER), elevation of CORR
activity is inevitably accompanied by the unwanted promo-
tion of HER activity.[8,16] So far, there are two major proposed
mechanisms to enhance CORR selectivity, i.e., increasing
solely the intrinsic activity of CORR[17–19] or increasing local
pH to suppress HER activity.[1,16,20–22]
To reveal the underlying mechanism, we employ a com-
prehensive reaction-diffusion model for enhanced CORR on
multisite catalysts. The roughness in experiments is mimicked
by HC sites randomly distributed on a planar electrode
surface. As shown in Figure 1, adsorption, desorption, dif-
fusion, reduction of CO, as well as the accompanied HER are
included. Specially, since the curvature of HC sites on the
electrode surface can lead to an enhanced local electric field,
the local-field-induced directional mass transfer is also taken
Recently, an effective design strategy of simply increasing
the roughness factor (RF) of catalysts has been reported to
improve the selectivity of CORR in experiments.[23–27] How-
ever, both the electrochemical-active-surface-area (ECSA)
normalized CORR rate and the local pH near the electrode
surface were found to keep nearly constant with increasing
[*] Y. Liu, Dr. H. Jiang, Dr. Z. Hou
Department of Chemical Physics & Hefei National Laboratory for
Physical Sciences at the Microscale, iChEM, University of Science
and Technology of China
Hefei, Anhui 230026 (China)
E-mail: hjjiang3@ustc.edu.cn
hzhlj@ustc.edu.cn
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202016332.
Figure 1. Schematic of the kinetic model for CORR and HER on
roughness catalysts. Products of the main reaction CORR are hydro-
carbons while the one for sub-reaction HER is hydrogen.
Angew. Chem. Int. Ed. 2021, 60, 11133 –11137
ꢀ 2021 Wiley-VCH GmbH
11133

Angewandte
Communications
Chemie
into account.[28–30] The reaction-diffusion equations (RDE)
for CORR on multisite electrocatalysts surface are derived as
follows (Details in supplementary information, SI)
[Eqs. (1a)–(1b)].
equilibrium potential, the overpotential and the reducing
coefficient for HER, respectively.[31,32] It has been acknowl-
edged that for HER in alkaline conditions, OHꢀ would adsorb
on the electrode surface and further blocks the HC sites for
the dissociation of water and then suppresses the HER.[31] The
effect of competition between adsorption of OHꢀ and
dissociation of water in HER is taken into account by
1ꢀqOH which measures the remained space for water disso-
ciation.[32,33] As the reaction rates also depend on the local
electric potential which is highly enhanced on HC sites, the
range of grids on which reactions happen should be truncated
around HC sites. Accordingly, the reactions are considered to
occur only on HC sites.
@qCOðrÞ
@t
ꢀ
ꢁ
CO
ad
¼ k c0CO 1 ꢀ qCOðrÞ ꢀ kdCOqCOðrÞ
DCO
ꢂ
ꢀ
ꢁ
ꢀ
ꢁꢃ
þDCOr qCOðrÞ ꢀ
r ꢁ qCOðrÞ 1 ꢀ qCOðrÞ r @qCO VCOðrÞ
2
kBT
ꢀ
ꢁ
COR
h¼0
ꢀk exp lCORhCOR qCOðrÞ
ð1aÞ
@qOHðrÞ
ꢀ
ꢁ
OH
ad
¼ k c0OH 1 ꢀ qOHðrÞ ꢀ kdOHqOHðrÞ
@t
In simulations, Equation (1) is discretized on a L ꢀ L
rectangle lattice with n ꢀ n grids and is solved by Euler
difference methods with periodic boundary conditions. N HC
sites (proportional to the experimentally measured roughness
factor RF) are randomly distributed on the surface. Param-
eters are normalized by DOH-, L and kBT. The apparent rates
nCORR and nHER for CORR and HER are summations over the
total electrode surface. The Faradaic efficiency (FE) can be
calculated by e ¼ hð4nCORRÞ=ð4nCORR þ nHERÞi, and the ECSA-
DOH
ꢂ
ꢀ
ꢁ
ꢀ
ꢁꢃ
þDOHr qOHðrÞ ꢀ
r ꢁ qOHðrÞ 1 ꢀ qOHðrÞ r @qOH VOHðrÞ
2
kBT
ꢀ
ꢁ
COR
h¼0
þ4k exp lCORhCOR qCOðrÞ
ꢀ
ꢁ
ꢀ
ꢁ
HER
h¼0
þk
1 ꢀ qOH ðrÞ exp lHERhHER
ð1bÞ
In the RDEs, variables with superscript OH, CO, or H2O
denote the corresponding ones for species OHꢀ, CO, or H2O,
respectively. q(r) is the surface coverage of each chemical
species at location r.
~
normalized rates are nHERðCORRÞ ¼ hnHERðCORRÞ=Ni where hꢁi
denotes ensemble average over different distribution of HC
sites with the same N.
In the right side of Equation (1a) and (1b), the first-
(second) term describes the adsorption(desorption) of CO
and OHꢀ respectively, where kad(kd) is the adsorption(de-
sorption) rate constant. In the first term, c0 is the concen-
tration in solution near the electrode surface, and 1ꢀq(r)
measures the number of available space for adsorption. Here,
the effect of local pH is taken into account by the relationship
pH ¼ ꢀlogð10ꢀ14=cO0 HÞ. As experiments demonstrated that
local pH keeps nearly unchanged for CORR on catalysts with
different RF, c0OH is set to be constant.[24] Similarly, c0CO is also
set to be constant due to the constant CO flow in experi-
ments.[24] The second term describes the desorption with
constants kd.
CORR on multisite catalysts is simulated with parameters
listed in Table S1 (Determination of the parameters are
discussed in SI). To validate our theoretical model, FE of the
main product hydrocarbons (blue bar) and the side product
H2 (grey bar) for different RFare plotted in Figure 2(a). As N
increases, FE of hydrocarbons increases while FE for H2
decreases. This leads to an enhanced selectivity of hydro-
carbons from less than 40% at N = 5 to more than 80% at
N = 40. The simulated selectivity (dash line) and that in
experiments[24] (red star) are compared in the inset of
Figure 2(a). It can be observed that the RF-enhanced
selectivity in experiments[24] is well reproduced by our
simulations. For further validation, apparent rates of CORR
and HER are also calculated in our simulations. Similar to
that in the experiments,[24] nCORR (blue line) increases
significantly with RF, while nHER (gray line) changes slightly
as shown in the top panel in Figure 2(b), leading to the
observed enhancement of CORR selectivity.
The third and fourth terms describe the mass transfer on
the electrode surface due to normal diffusion with diffusion
constant D and curvature-enhanced local electric field,[28,29]
respectively. Here, kB is the Boltzmann constant and T the
temperature. The effective interaction potential between
local electric field and chemical species is chosen as Gaus-
ꢀ
ꢁ
P
2
sian-like VðrÞ ¼ site aqu0expðꢀjr ꢀ rsitej = r20Þ ,[28] where a is
the interaction strength, u0 denotes the intensity of electric
field on an isolated single HC site located at rsite, and r0 is
the characteristic length.[28,30] Since CO can be polarized by
the local electric field and be attracted to HC sites, we take
aCO < 0 to describe such attraction. On the contrary, the
adsorbed OHꢀ would be pushed away from HC sites, and
consequently aOH > 0.
The intrinsic activities for CORR and HER are explored
~
by ECSA-normalized reaction rate nHERðCORRÞ which are
presented in the bottom of Figure 2(b). It can be found that
~
nCORR diminutively changes with N, which indicates a nearly
~
unchanged CORR intrinsic activity. Nevertheless, nHER and
consequently the intrinsic activity of HER declines sharply
with increasing N. The above observations agree with
experimental ones well, implying that the CORR-activity-
promoting mechanism fails to explain the enhanced CORR
selectivity on multisite catalysts. Furthermore, as experimen-
tal observations demonstrated that local pH keeps nearly
unchanged for CORR on catalysts with different RF, the
concentration of OHꢀ in the solution near the electrode
surface cO0 H is set to be constant (Figure 2(c)). This rules out
the local-pH-elevating mechanism from the enhanced CORR
selectivity. One may argue that local PH may not be the OHꢀ
The fifth term shows the change of q due to the main
reaction CORR basing on Butler-Volmer equation, where
COR
h¼0
k
is the reaction rate constant at the equilibrium potential,
hCOR is the overpotential, and lCOR denotes the reducing
coefficient of the energy barrier by the applied overpoten-
tial.[8] Additionally, the last term in Equation (1b) describes
the change of OHꢀ coverage as a result of the side reaction
HER
h¼0
HER, where k , hHER and lHER are the rate constant at
11134 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 11133 –11137

Angewandte
Communications
Chemie
Figure 3. Mechanism of RF enhanced CORR selectivity. (a) and (b) are
the surface coverage of OHꢀ on the catalysts with N=5 and N=40,
respectively, in which the black particles are the location of HC sites.
(c) is the distribution of OH groups on the HC sites of catalysts with
different roughness with the interaction strength 10.0. (d) and (e) are
potential of OH groups in the electric field induced by catalysts with
N=5 and N=40, respectively.
plotted in Figure 3(c). For small RF catalysts (such as N =
5,10, blue bars in Figure 3(c)), the distribution is Gaussian-
like with a peak of low coverage. As the RF increases to be
medium (such as N = 20,30, green bars in Figure 3(c)), a new
peak of high coverage emerges, and the distribution changes
to be bimodal. The peak of low coverage then declines with
further increased RF (such as N = 40, red bar in Figure 3(c)),
while the peak of high coverage rises gradually. Since the
reaction rates of HER would decrease as the HC sites are
blocked by OHꢀ, the high-RF-induced high coverage of OHꢀ
on HC sites thus contributes to the suppression of HER in the
electrocatalytic process.
Figure 2. Selectivity and activity of CORR on roughness catalysts.
(a) The simulated results of Faradic efficiency (FE) of hydrocarbons
(HC) and hydrogen (H2). The inset shows our simulation results and
those in experiments in Ref. [24] (b) Dependence of the apparent rates
~
n (top) and ECSA-normalized rates n (bottom) for CORR and HER on
the sites number N. (c) The averaged OHꢀ coverage on the electrode
surface (red line) and the OHꢀ concentration near the electrode
surface (green line) of catalysts with different roughness factors.
concentration in the solution near the electrode surface but
OHꢀ coverage on the electrode surface. Thus, the averaged
ꢀ
ꢀ
coverage of OH , qOH, on the electrode surface is also plotted
Now, the key point turns to why the coverage of OHꢀ on
HC sites is low for low RF catalysts and increases as RF
increases. As OHꢀ is generated at the HC sites and can diffuse
to nearby area by normal diffusion according to the concen-
tration gradient, the coverage of OHꢀ would be higher on HC
sites than other regions of the surface (or nearly equal if
diffusion is much faster than reactions). As mentioned in the
kinetic modeling part, the curvature-enhanced local electric
field can also lead to an extra mass transfer besides the
normal diffusion. The interaction potentials between local
electric field and OHꢀ on the surface of electrode with low
(N = 5) and high (N = 40) RF are plotted in Figure 3(d,e),
respectively. It is observed that, low RF catalyst leads to
separated high potential islands located around each HC site,
resulting in directional mass transfer of OHꢀ from HC sites to
other regions of the electrode surface. Such mass transfer then
leads to the concentration distribution shown in Figure 3(a)
where OHꢀ coverage is low around HC sites, i.e., OHꢀ
ꢀ
in Figure 2(c). Clearly, qOH is also nearly unchanged when the
RF increases, demonstrating again that local-pH-elevating
mechanism is not the reason for enhanced selectivity for
CORR on multisite catalysts.
To investigate the very reason for the observed HER
suppression on multisite catalysts, we now focus on adsorbed
OHꢀ on the electrode surface during the electrocatalytic
process. Snapshots of stationary OHꢀ surface coverages, qOH
,
on the catalysts with different N are shown in Figure 3(a) and
Figure 3(b). For low roughness catalysts, there are several low
coverage regions (white area in Figure 3(a)) and all of the HC
sites (black particles in Figure 3(a)) are located in the center
of these regions. Remarkably, for high roughness catalysts,
only few of the HC sites locate in the low coverage area and
most of them locate in relatively high coverage area (Fig-
ure 3(b)). To quantitatively compare the observed difference,
the probability distribution of HC sites with given qOH is
Angew. Chem. Int. Ed. 2021, 60, 11133 –11137
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 11135

Angewandte
Communications
Chemie
coverage at HC sites is depressed by the local-field-induced
mass transfer.
field in relative to that of OHꢀ. For example, there may be
some anions acting as Lewis acid and would interact with the
Lewis base, CO, and then enhance the mass transfer within
local electric field.[29,34,35]
However, as shown in Figure 3(e), the islands of high
interaction potential around HC sites are overlapped for high
RF catalysts. Consequently, there are some closely
approached HC sites located in regions of high interaction
potential, while others have to distribute in regions of low
potential. As the local-field-induced mass transfer is always
from regions of the high interaction potential to ones with low
potential, the depression of OHꢀ coverage is then weakened,
and OHꢀ coverage at some HC sites increases as the
concentration distribution presented in Figure 3(b). In
short, the HER activity on high RF catalysts is suppressed
by local-field-induced directional mass transfer of OHꢀ.
Based on the above revealed kinetic-controlled mecha-
nism of multi-site enhanced high selectivity of CORR, it is
possible to further improve the selectivity of carbon products
by tuning kinetic factors.
One of the most significant kinetic factors is the inter-
action strength a between the local electric field and chemical
species. Large a leads to high local-field-induced mass
transfer while small a results in a weak one. As shown in
Figure 4(a), CORR selectivity increases with the relative
interaction strength aCO/aOH and saturates at large enough
aCO/aOH for all of the catalysts with RF ranging from N = 5 to
40. Interestingly, the saturated selectivity would reach to
nearly 100% for catalysts with large RF at a lower relative
interaction strength aCO/aOH in comparison with that for
catalysts with small RF. The above findings indicate that the
CORR selectivity can be well tuned in experiments by ways
of increasing the ability of CO interacting with local electric
The characteristic length of the local electric field is
another significant kinetic factor for CORR selectivity. As
shown in Figure 4(b), increasing the characteristic length r0
can also enhance the CORR selectivity for all of the catalysts
with different RF. The CORR selectivity on high RF catalyst
is enlarged obviously as r0 increases, while that on low RF
catalysts increases slightly with increasing r0. For large enough
r0 and RF, CORR selectivity reaches nearly 100%. As shown
in FigureS(3), large r0 prefers to form overlapped islands of
high interaction potential and enhances the suppression of
HER activity, while the activity of CORR changes little. As it
is well known that modifying the ion strength of the electro-
lyte would manipulate the characteristic length efficiently,[36]
CORR selectivity may be optimized by enlarging solely the
characteristic length of the local electric field.
Besides the local electric field, the mass transfer is also
controlled by the equilibrium of adsorption and desorption
OH
(k =kOd H) of OHꢀ from bulk to surface which determines the
ad
concentration of species on the electrode surface. It can be
seen in Figure 4(c) that the selectivity of CORR is enhanced
OH
ad
OH
ad
by an enlarged k =kdOH. For small k =kOd H, the selectivity is
rather small and the difference of selectivity on catalysts with
OH
different RF is minor. As k =kOd H increases, the selectivity of
ad
CORR on high RF catalyst, such as N = 40, is enhanced
obviously (from 40% to nearly 100%) while that on low RF
catalyst, such as N = 5, shows little difference. This finding
OH
shows that for high RF catalysts, raising the ratio of k =kdOH
ad
would efficiently optimize the selectivity of CORR. However,
OH
optimizing the k =kOd H does little to the FE for catalysts with
ad
low RF.
In summary,
a comprehensive kinetic model was
employed to investigate CORR selectivity in competition
with HER on catalysts of different roughness factors (RF).
The kinetic model successfully reproduced the experimentally
observed RF-enhanced CORR selectivity. It was revealed
that the high-curvature-induced local electric field directs the
mass transfer of OHꢀ during electrocatalytic process. High
RF catalysts lead to a weaker local-field-induced mass
transfer as a result of the highly overlapped local electric
field. This further results in the suppression of HER and the
enhancement of CORR selectivity. Based on the kinetic-
controlled mechanism, several approaches were predicted to
optimize CORR selectivity by modifying the kinetics of
CORR and HER, such as the local-field-induced mass
transfer, adsorption and desorption between bulk solution
and electrode surface, etc. Our work highlights the kinetics in
electrocatalysis on nanocatalysts, and provides a new princi-
ple for future catalyst design.
Figure 4. Prediction of the optimal CORR selectivity with optimized
kinetic factors. (a) FE (selectivity) of CORR on catalysts with different
roughness with the increasing interaction strength between adsorbed
OHꢀ and local electric field. (b) The FE of HC products on catalysts
with different roughness with the increasing characteristic length of
electric field r0. (c) the FE with the increasing equilibrium of adsorption
and desorption of OH from bulk to the electrode surface.
Acknowledgements
This work is supported by MOST (2018YFA0208702,
2016YFA0400904), NSFC (21973085, 21833007, 21790350,
21673212, 21521001), Anhui Initiative in Quantum Informa-
11136 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 11133 –11137

Angewandte
Communications
Chemie
[16] Y.-J. Zhang, V. Sethuraman, R. Michalsky, A. A. Peterson, ACS
tion Technologies (AHY090200), and the Fundamental
Research Funds for the Central Universities
(WK2340000104).
Catal. 2014, 4, 3742 – 3748.
[17] A. A. Peterson, J. K. Nørskov, J. Phys. Chem. Lett. 2012, 3, 251 –
258.
[18] C. W. Li, J. Ciston, M. W. Kanan, Nature 2014, 508, 504 – 507.
[19] S. H. Lee, I. Sullivan, D. M. Larson, G. Liu, F. M. Toma, C.
Xiang, W. S. Drisdell, ACS Catal. 2020, 10, 8000 – 8011.
[20] M. Ma, B. J. Trzesniewski, J. Xie, W. A. Smith, Angew. Chem. Int.
Ed. 2016, 55, 9748 – 9752; Angew. Chem. 2016, 128, 9900 – 9904.
[21] R. Kas, R. Kortlever, H. Yılmaz, M. T. M. Koper, G. Mul,
ChemElectroChem 2015, 2, 354 – 358.
Conflict of interest
The authors declare no conflict of interest.
Keywords: carbon monoxide reduction · kinetics ·
electrocatalysis · mechanistic studies · roughness
[22] X. Wang, C. Xu, M. Jaroniec, Y. Zheng, S. Z. Qiao, Nat.
Commun. 2019, 10, 4876.
[23] B. A. Zhang, T. Ozel, J. S. Elias, C. Costentin, D. G. Nocera, ACS
Cent. Sci. 2019, 5, 1097 – 1105.
[24] L. Wang, S. Nitopi, A. B. Wong, J. L. Snider, A. C. Nielander,
C. G. Morales-Guio, M. Orazov, D. C. Higgins, C. Hahn, T. F.
Jaramillo, Nat. Catal. 2019, 2, 702 – 708.
[1] Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K.
Norskov, T. F. Jaramillo, Science 2017, 355, eaad4998.
[2] O. S. Bushuyev, P. De Luna, C. T. Dinh, L. Tao, G. Saur, J.
van de Lagemaat, S. O. Kelley, E. H. Sargent, Joule 2018, 2, 825 –
832.
[3] S. Xu, E. A. Carter, Chem. Rev. 2019, 119, 6631 – 6669.
[4] D. D. Zhu, J. L. Liu, S. Z. Qiao, Adv. Mater. 2016, 28, 3423 – 3452.
[5] T. Zheng, K. Jiang, H. Wang, Adv. Mater. 2018, 30, 1802066.
[6] D. Gao, R. M. Arꢁn-Ais, H. S. Jeon, B. Roldan Cuenya, Nat.
Catal. 2019, 2, 198 – 210.
[7] M. B. Ross, P. De Luna, Y. Li, C.-T. Dinh, D. Kim, P. Yang, E. H.
Sargent, Nat. Catal. 2019, 2, 648 – 658.
[8] Y. Hori, R. Takahashi, Y. Yoshinami, A. Murata, J. Phys. Chem.
B 1997, 101, 7075 – 7081.
[9] J. Feng, S. Zeng, J. Feng, H. Dong, X. Zhang, Chin. J. Chem.
2018, 36, 961 – 970.
[10] A. Rendꢂn-Calle, S. Builes, F. Calle-Vallejo, Curr. Opin. Electro-
chem. 2018, 9, 158 – 165.
[11] W. Luc, X. Fu, J. Shi, J.-J. Lv, M. Jouny, B. H. Ko, Y. Xu, Q. Tu, X.
Hu, J. Wu, Q. Yue, Y. Liu, F. Jiao, Y. Kang, Nat. Catal. 2019, 2,
423 – 430.
[12] M. Jouny, W. Luc, F. Jiao, Nat. Catal. 2018, 1, 748 – 755.
[13] Y. Pang, J. Li, Z. Wang, C.-S. Tan, P.-L. Hsieh, T.-T. Zhuang, Z.-
Q. Liang, C. Zou, X. Wang, P. De Luna, J. P. Edwards, Y. Xu, F.
Li, C.-T. Dinh, M. Zhong, Y. Lou, D. Wu, L.-J. Chen, E. H.
Sargent, D. Sinton, Nat. Catal. 2019, 2, 251 – 258.
[25] A. Auer, M. Andersen, E.-M. Wernig, N. G. Hçrmann, N. Buller,
K. Reuter, J. Kunze-Liebhꢃuser, Nat. Catal. 2020, 3, 797—803.
[26] C. Choi, S. Kwon, T. Cheng, M. Xu, P. Tieu, C. Lee, J. Cai, H. M.
Lee, X. Pan, X. Duan, W. A. Goddard, Y. Huang, Nat. Catal.
2020, 3, 804—812 .
[27] T. Kim, G. T. R. Palmore, Nat. Commun. 2020, 11, 3622.
[28] H. Jiang, Z. Hou, Y. Luo, Angew. Chem. Int. Ed. 2017, 56,
15617 – 15621; Angew. Chem. 2017, 129, 15823 – 15827.
[29] M. Liu, Y. Pang, B. Zhang, P. De Luna, O. Voznyy, J. Xu, X.
Zheng, C. T. Dinh, F. Fan, C. Cao, F. P. de Arquer, T. S. Safaei, A.
Mepham, A. Klinkova, E. Kumacheva, T. Filleter, D. Sinton,
S. O. Kelley, E. H. Sargent, Nature 2016, 537, 382 – 386.
[30] Q.-X. Chen, Y.-H. Liu, X.-Z. Qi, J.-W. Liu, H.-J. Jiang, J.-L.
Wang, Z. He, X.-F. Ren, Z.-H. Hou, S.-H. Yu, J. Am. Chem. Soc.
2019, 141, 10729 – 10735.
[31] R. Subbaraman, D. Tripkovic, D. Strmcnik, K.-C. Chang, M.
Uchimura, A. P. Paulikas, V. Stamenkovic, N. M. Markovic,
Science 2011, 334, 1256 – 1260.
[32] S. Sharifi-Asl, D. D. Macdonald, J. Electrochem. Soc. 2013, 160,
H382 – H391.
[33] E. Liu, J. Li, L. Jiao, H. T. T. Doan, Z. Liu, Z. Zhao, Y. Huang,
K. M. Abraham, S. Mukerjee, Q. Jia, J. Am. Chem. Soc. 2019,
141, 3232 – 3239.
[14] T.-T. Zhuang, Y. Pang, Z.-Q. Liang, Z. Wang, Y. Li, C.-S. Tan, J.
Li, C. T. Dinh, P. De Luna, P.-L. Hsieh, T. Burdyny, H.-H. Li, M.
Liu, Y. Wang, F. Li, A. Proppe, A. Johnston, D.-H. Nam, Z.-Y.
Wu, Y.-R. Zheng, A. H. Ip, H. Tan, L.-J. Chen, S.-H. Yu, S. O.
Kelley, D. Sinton, E. H. Sargent, Nat. Catal. 2018, 1, 946 – 951.
[15] J. Li, F. Che, Y. Pang, C. Zou, J. Y. Howe, T. Burdyny, J. P.
Edwards, Y. Wang, F. Li, Z. Wang, P. De Luna, C. T. Dinh, T. T.
Zhuang, M. I. Saidaminov, S. Cheng, T. Wu, Y. Z. Finfrock, L.
Ma, S. H. Hsieh, Y. S. Liu, G. A. Botton, W. F. Pong, X. Du, J.
Guo, T. K. Sham, E. H. Sargent, D. Sinton, Nat. Commun. 2018,
9, 4614.
[34] J. Li, X. Li, C. M. Gunathunge, M. M. Waegele, Proc. Natl. Acad.
Sci. USA 2019, 116, 9220 – 9229.
[35] J. D. Roth, M. J. Weaver, Langmuir 1992, 8, 1451 – 1458.
[36] H. Daiguji, Chem. Soc. Rev. 2010, 39, 901 – 911.
Manuscript received: December 9, 2020
Accepted manuscript online: March 3, 2021
Version of record online: April 8, 2021
Angew. Chem. Int. Ed. 2021, 60, 11133 –11137
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 11137