Breaking Scaling Relationships in CO
2
Reduction on Copper Alloys
with Organic Additives
Yungchieh Lai,
†
Nicholas B. Watkins,
†
Alonso Rosas-Hern
á
ndez, Arnaud Thevenon, Gavin P. Heim,
Lan Zhou, Yueshen Wu, Jonas C. Peters,
*
John M. Gregoire,
*
and Theodor Agapie
*
Cite This:
ACS Cent. Sci.
2021, 7, 1756
−
1762
Read Online
ACCESS
Metrics & More
Article Recommendations
*
s
ı
Supporting Information
ABSTRACT:
Boundary conditions for catalyst performance in the
conversion of common precursors such as N
2
,O
2
,H
2
O, and CO
2
are governed by linear free energy and scaling relationships.
Knowledge of these limits o
ff
ers an impetus for designing strategies
to alter reaction mechanisms to improve performance. Typically,
experimental demonstrations of linear trends and deviations from
them are composed of a small number of data points constrained
by inherent experimental limitations. Herein, high-throughput
experimentation on 14 bulk copper bimetallic alloys allowed for
data-driven identi
fi
cation of a scaling relationship between the
partial current densities of methane and C
2+
products. This strict
dependence represents an intrinsic limit to the Faradaic e
ffi
ciency for C
−
C coupling. We have furthermore demonstrated that
coating the electrodes with a molecular
fi
lm breaks the scaling relationship to promote C
2+
product formation.
■
INTRODUCTION
The development of high-performing catalysts for sustainable
and economically viable transformations remains a central goal
of the chemical industry.
1
Chemical transformations are
controlled by thermodynamic and kinetic rate laws that
manifest as linear scaling relationships. Such relationships
relating structure, activity, and reaction conditions are
established for a range of reactions, including H
2
O oxidation
and N
2
,O
2
,CO
2
, and H
2
O reduction performed on both
heterogeneous and homogeneous catalysts.
1
−
7
Because they
provide theoretical or empirical trends for a particular chemical
process, these scaling relationships not only help explain
chemical reactivity but also guide the rational design of new
and improved catalysts. Determining the underlying con-
nections in chemical processes is particularly desirable toward
deconvoluting fundamental selectivity limitations and targeting
speci
fi
c products.
2
,
8
Typically, the experimental establishment
and breaking of scaling relationships, including mapping of
volcano plots, deals with a small set of data points, a limitation
that is sometimes compensated for by expansion of data sets
through computation. Our development of high-throughput
electrochemistry coupled to an automated product distribution
analysis provides new opportunities for identifying scaling
relationships.
9
,
10
Herein, we demonstrate a combination of
catalyst design, high-throughput experimentation, and data
science as a paradigm shift in both the identi
fi
cation of scaling
relationships and the discovery of strategies for breaking them.
We focus on applying this approach to CO
2
reduction (CO
2
R)
on Cu-based electrodes, an area where mechanistic complexity
has obscured the identi
fi
cation of scaling relationships, has
hindered catalyst optimization, and warrants further inves-
tigation.
11
As strategies to transform CO
2
at scale are considered for a
more sustainable carbon economy, exploiting the unique ability
of Cu to reduce CO
2
to C
2+
hydrocarbons and oxygenates
makes it an attractive catalyst for optimization. The complex
pathways toward myriad reduced products of CO
2
RonCu
stymie e
ff
orts for producing carbon-coupled products with
high selectivity and have prompted investigation into the
mechanism of the transformation.
11
Systematic trends a
ff
ecting
selectivity have been shown with respect to adsorption energy
scaling relationships and pH variation at the electrode
surface.
7
,
12
−
15
Promising strategies for improving CO
2
R
selectivity for C
2+
products include changing catalyst
morphology
16
−
19
and electrolyte composition,
14
,
20
employing
bimetallic systems and alloys,
21
−
23
and adding organic
modi
fi
ers.
24
−
29
While these techniques may facilitate altered
product distributions, they are not amenable to identifying
empirical scaling relationships due to substantial variation in
catalyst preparation and electrochemical testing conditions
across independent studies. It is consequently pertinent to
Received:
July 16, 2021
Published:
October 14, 2021
Research Article
http://pubs.acs.org/journal/acscii
© 2021 The Authors. Published by
American Chemical Society
1756
https://doi.org/10.1021/acscentsci.1c00860
ACS Cent. Sci.
2021, 7, 1756
−
1762
Downloaded via CALIFORNIA INST OF TECHNOLOGY on November 3, 2021 at 18:15:31 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
conduct studies that systematically and broadly vary select
parameters. We have identi
fi
ed bulk alloying of Cu as an
underdeveloped, though promising, strategy for catalyst
optimization, with a large parameter space available based on
the metal identity and composition,
30
,
31
well suited for
investigation using our high-throughput screening system.
9
,
10
Additionally, organic additives represent an attractive orthog-
onal parameter of catalyst design. They can impact perform-
ance in a manner that has seldom been achieved by tailoring
inorganic electrocatalyst composition or morphology alone.
32
Inspired by recent success using molecular
fi
lms to enhance
the selectivity of catalysts for CO
2
R,
29
,
33
−
35
herein, we describe
the generation of a uniquely broad and systematic CO
2
R
catalyst database by combining a Cu bimetallic alloying
strategy with the use of organic additives. Selectivity analysis
highlights the impact of integrating high-throughput exper-
imentation and data science to discover a power-law scaling
relationship between partial current densities of CH
4
and C
2+
that is broken upon coating with an organic additive,
demonstrating a fundamental limitation of CO
2
R on Cu and
a strategy to overcome it through hybrid inorganic
−
organic
interfaces.
■
RESULTS AND DISCUSSION
To elucidate correlations in CO
2
R, experiments were designed
to observe a large dynamic range of catalyst properties while
mitigating con
fl
ation with experiment parameters such as
electrolyte composition and mass transport conditions. For the
present work, we varied catalyst composition, applied potential,
and molecular additive presence. The choice of Cu alloys was
guided by our previous discovery that the alloying elements In,
Co, Mn, and Zn alter the activity and selectivity of Cu in
di
ff
erent ways, although that study was limited to the detection
of H
2
,CH
4
, and C
2
H
4
.
10
Studying Cu alloys with each of these
elements and with di
ff
erent concentrations that span face-
centered cubic (fcc) alloys and intermetallic phases (XRD of
homogeneous alloys shown in
Figures S12
−
S16
), we sought to
obtain a more comprehensive map of the reactivity of Cu-
based alloy catalysts and to identify any systematic trends. The
molecular additive,
N
,
N
′
-ethylene-phenanthrolinium dibro-
mide (
1-Br
2
), was selected based on its ability to enhance
Faradaic e
ffi
ciency (FE) and geometric partial current densities
for C
2+
products upon forming a well-de
fi
ned
fi
lm on
polycrystalline Cu, primarily composed of a
para
-
para
isomer
of the one-electron reduced and dimerized phenanthrolinium
(
Figure 1
a).
33
The catalyst performance with or without the additive was
evaluated by chronoamperometry (CA) at a series of up to 6
potentials with subsequent product analysis using the batch
reactor
fl
ow system illustrated in
Figure 1
. This system uses
rapid electrolyte
fl
ow, as opposed to vigorous CO
2
bubbling, to
generate suitable and reproducible mass transport conditions.
The rapid concentration of reaction products enhances
measurement throughput by enabling shorter electrolysis and
faster chromatography compared to traditional methods.
Hybrid metal
−
organic electrodes were prepared via electro-
deposition of organic
fi
lms on the polycrystalline metal
electrode from an aqueous 0.1 M KHCO
3
bu
ff
ered electrolyte
containing 0.1 mM
1-Br
2
. In total, experiments with 14 alloy
catalysts and pure Cu provide electrochemical and partial
current densities for 137 unique combinations of catalyst
composition, additive presence, and applied potential, as
shown for select Mn-doped catalysts in
Figure 2
a,b and for
all catalysts in
Figures S1 and S2
.
Pairwise relationships of the geometric partial current
density and the FE for representative products (
Figure S5
)
highlight the e
ff
ect of the combined strategy of alloying and
organic
fi
lms. The intrinsic modi
fi
cation of catalyst selectivity
can be detected through analysis of the Pearson correlation
coe
ffi
cient of the logarithm of partial current densities. A close-
to-unity positive correlation indicates that selectivity between
the two products cannot be tuned with the parameters under
consideration, which is indicative of a free-energy scaling
relationship. A substantially negative correlation indicates a
trade-o
ff
in selectivity, wherein enhanced formation of one
product occurs at the expense of the other, which is indicative
of kinetic competition for a shared reaction intermediate.
The large data set provided here via high-throughput
experimentation enables the study of correlation coe
ffi
cients
and their modi
fi
cation (
Figure 2
). Previous work on
polycrystalline Cu indicates that the kinetic regimes that
govern the CO
2
R product distribution di
ff
er with applied
potential due to modulation of the energy landscape as a
function of overpotential as well as second-order e
ff
ects such as
CO
2
mass transport and changes to the pH at the catalyst
surface.
36
,
37
To facilitate the observation of how the catalyst
itself a
ff
ects selectivity, we aim to mitigate the in
fl
uences from
the extrinsic e
ff
ects by limiting the overpotential range (
−
0.84
to
−
1.1 V vs reversible hydrogen electrode, RHE) and using
rapid electrolyte
fl
ow over
fl
at catalyst
fi
lms with a maximum
current density of 15 mA cm
−
2
, which promotes uniform mass
Figure 1.
(a) The electrochemical reductive coupling of two
1-Br
2
molecules results in a mixture of two products; (b) The high
throughput catalyst screening system where a select catalyst is
positioned under a recirculating electrochemical batch reactor. After
electrocatalyst operation, a robot sample handler (RSH) uses a
syringe (orange) to extract aliquots from the headspace and then
catholyte, with each aliquot injected into the respective analytical
instrument (green, syringe positions in translucent orange) for gas or
high-pressure liquid chromatography (GC, HPLC). The reference
electrode (RE) is placed in the electrolyte inlet to the working
electrode (WE) chamber, which is separated from the counter
electrode (CE) chamber by a bipolar membrane (BPM).
ACS Central Science
http://pubs.acs.org/journal/acscii
Research Article
https://doi.org/10.1021/acscentsci.1c00860
ACS Cent. Sci.
2021, 7, 1756
−
1762
1757
transport and limits pH gradients in the electrochemical
reactor. This potential range includes the onset of substantial
partial current density for highly reduced products, making
alteration of correlation coe
ffi
cients in this range a prime target
for controlling product selectivity with catalyst modi
fi
cation.
We
fi
rst demonstrate a Pearson correlation analysis to ascertain
the extent by which high correlation coe
ffi
cients can be
lowered via variation in catalyst composition. For example, the
box in
Figure 2
d with A = CH
4
and B = C
2+
shows a high
correlation coe
ffi
cient of 0.99 for these products when
considering a series of 7 electrolysis experiments with a
polycrystalline Cu catalyst in which partial current densities for
both CH
4
and C
2+
products varied from approximately 1
μ
A
cm
−
2
to 3 mA cm
−
2
. The analogous analysis for polycrystalline
Figure 2.
Illustration of acquired data and correlation analysis. The electrochemical and geometric partial current densities are shown for 5
electrolysis experiments with Cu
0.98
Mn
0.02
and 6 electrolysis experiments with Cu
0.84
Mn
0.16
catalysts, both (a) without additive and (b) with
1-Br
2
.
Select products or product categories were considered for correlation analysis. For A = CH
4
and B = C
2+
, part a contains 8 electrolysis experiments
with geometric partial current densities for both A and B above 1
μ
Acm
−
2
. The corresponding 8 points are shown in part c and used to calculate
the Pearson correlation coe
ffi
cient to represent additive-free Cu
−
Mn alloys. This analysis was applied to all 6 pairwise combinations of the products
HCOOH, CO, CH
4
, and C
2+
and repeated for pure Cu and each Cu
−
M alloy system. The resulting set of correlation coe
ffi
cients is shown in part
d. The printed numbers in each cell indicate the number of electrolysis experiments used in the calculation, for example, 8 for the A = C
2+
,B=
CH
4
, and M = Mn cell corresponding to the plot in part c. This analysis was also applied to electrolysis experiments from all compositions,
fi
rst with
and then without
1-Br
2
additive, to assess the impact of the additive on the 6 pairwise correlation coe
ffi
cients, as shown in part e. For A = C
2+
and B
=CH
4
in part e, the data underlying the correlation analysis are shown (f) without additive and (g) with
1-Br
2
, where points are colored according
to their composition. The data in part f follow the power-law relationship indicated by the dashed line, which is also depicted in part g to show the
extent by which the data with
1-Br
2
deviate from this power-law relationship. The horizontal error bars in part a estimate the variation in potential
in each electrolysis, and the quanti
fi
ed uncertainty for each partial current density is smaller than the point size (see the Experimental Uncertainty
section in the
SI
).
ACS Central Science
http://pubs.acs.org/journal/acscii
Research Article
https://doi.org/10.1021/acscentsci.1c00860
ACS Cent. Sci.
2021, 7, 1756
−
1762
1758
Cu
−
M alloys is summarized by the boxes with A = C
2+
and B
=CH
4
, where the correlation coe
ffi
cient was calculated for
each alloying element using various combinations of alloy
composition and applied potential. The source data and their
utilization of a correlation analysis are illustrated for the Cu
−
Mn system in
Figure 2
a,c.
The total number of electrolysis conditions and range of
alloy compositions (
x
in Cu
1
−
x
M
x
) are as follows: 7 conditions
with M = Co and
x
= 0.02 or 0.16; 10 conditions with M = In
and
x
= 0.02 or 0.17; 16 conditions with M = Zn and
x
= 0.04,
0.13, 0.21, or 0.51; and 8 conditions with M = Mn and
x
= 0.02
or 0.16. Despite the variation in composition and potential
within each of these Cu
−
M systems, each correlation
coe
ffi
cient remains in excess of 0.98, and in total, the
correlation coe
ffi
cient for all Cu
−
M alloys is not meaningfully
changed from that observed with pure Cu.
Figure 2
f shows the
aggregation of data for Cu and its alloys, demonstrating that a
power-law relationship is closely followed over a broad range
of composition and applied potential. This striking relation-
ship, over 3 orders of magnitude, in partial current densities
strongly suggests that on these bulk alloy catalysts there is a
common branching point, or combination of branching points,
that consistently partitions between CH
4
and C
2+
products
(
Figure 3
a). Preservation of the CH
4
/C
2+
ratio as observed
here represents a newly discovered fundamental limitation for
e
ff
orts to improve selectivity through bulk bimetallic alloying
alone.
A simple rationale for the near 1:1 ratio observed for CH
4
and C
2+
products is challenged by the complexity of the
mechanism of CO
2
R.
38
Under di
ff
erent reaction conditions,
such as using an alternate bicarbonate concentration, a similar
power-law trend is observed, but with a slightly di
ff
erent slope
(
Figure S4
). The up-shifted CH
4
to C
2+
ratio agrees with the
in
fl
uence of KHCO
3
concentration where 0.1 M was
considered to be the optimal environment for C
2+
products.
13
,
14
,
39
,
40
The preservation of the scaling relationship
is therefore supportive of an intrinsic mechanistic limitation for
the production of methane and C
2+
products for the set of
experimental conditions used herein. Also, the alloying
elements substantially alter other aspects of the product
distribution, making this collection of catalyst electrodes
particularly well-suited for inferring intrinsic reactivity trends;
the catalyst morphology is kept relatively constant with respect
to the compendium of results in the litera-
ture.
11
,
14
,
16
,
17
,
20
−
22
,
24
−
26
,
30
,
41
For example, through study of
well-de
fi
ned Cu surfaces, Hori and others identi
fi
ed that the
relative production of CH
4
and C
2
H
4
is highly facet-
dependent.
15
The distribution of exposed facets of a
polycrystalline fcc-phase metal electrode could be altered via
alloying due to changes in growth kinetics and/or relative
surface energies upon addition of the alloying element, which
would in principle provide a method to break the CH
4
/C
2+
scaling relationship by tuning catalyst composition. However,
the observation that the scaling relationship holds over a broad
range of alloy compositions with distinct product distributions
indicates that facet selectivity is not the main contributor in the
observed product selectivity.
CO
2
R to highly reduced products such as CH
4
and C
2+
products proceeds via a common
*
CO intermediate.
42
,
43
Methane synthesis is proposed to proceed via a Langmuir
−
Hinshelwood pathway, where a surface
*
H couples with
*
CO
to form a
*
CHO or
*
COH intermediate that is further
hydrogenated toward methane.
42
,
44
,
45
Meanwhile, the produc-
tion of C
2+
products occurs via the coupling of two precursor
*
CO molecules, potentially involving intermediate
*
CHO or
*
COH adsorbates.
42
,
43
The absence of CO
−
CH
4
or CO
−
C
2+
power-law relationships without the additive and the
observation of a negative correlation coe
ffi
cient between
CO
−
-ethylene in the presence of the additive are consistent
with kinetic competition for a common intermediate (
Figure
S5
). Despite variation in free CO produced (
Figure S19
), for
the conditions tested, the assumed variation of CO
concentration near the surface of the electrode is incon-
sequential with respect to the CH
4
−
C
2+
relationship. There-
fore, although the speci
fi
c mechanism or mechanisms remain
debated and may involve multiple pathways depending on the
Figure 3.
(a) Possible reaction mechanisms (selected from a multitude of variations previously proposed),
11
where pathways are highlighted in
color with respect to their products in part b. There are two branching points between CH
4
and C
2+
products that could be responsible for the
relationship observed in
Figure 2
. The strong relationship between the gray and green pathways is broken with the addition of molecular additives,
implying a potential change in mechanism. (b) Summary of molar selectivity for reduction of the CO
*
intermediate. Measured partial current
densities for CO, CH
4
, and C
2+
products are converted to molar
fl
ux of CO
*
required to produce the respective products, whose normalization
provides the ternary composition for inclusion in this
fi
gure. Each electrolysis experiment produces 1 data point that indicates the catalyst
’
s
selectivity with respect to the three reaction pathways highlighted in part a that start from the common CO
*
intermediate.
ACS Central Science
http://pubs.acs.org/journal/acscii
Research Article
https://doi.org/10.1021/acscentsci.1c00860
ACS Cent. Sci.
2021, 7, 1756
−
1762
1759
morphology or crystal facet, the observed scaling relationship
between CH
4
and C
2+
indicates that the relative kinetics at the
branching point(s), remarkably, remain rigorously locked at
the same ratio over the many catalysts and applied potentials
tested herein. Breaking this dependence is highly desirable for
improved selectivity for C
2+
products.
The Pearson correlation analysis was extended to the impact
of the
1-Br
2
additive (
Figure 2
e), where the correlation
coe
ffi
cient for each set of conditions includes the aggregation
of all catalyst compositions and potentials. Coating the
catalysts using
1-Br
2
lowers the correlation coe
ffi
cient for
CH
4
and C
2+
from 0.99 to 0.74, a striking alteration whose
implication is that, within the range of catalyst compositions
considered in the present work, tuning the selectivity between
CH
4
and C
2+
is only achieved in the presence of the additive,
underscoring the importance of multimodal catalyst develop-
ment.
46
The basis by which the additive disrupts the scaling
relationship between CH
4
/C
2+
(
Figure 2
f) by increasing C
2+
production and suppressing CH
4
formation (
Figure 2
g) is of
particular interest.
Figure 3
a illustrates the portion of the
CO
2
R reaction network wherein branching ratios dictate
whether the common CO
*
intermediate results in the
generation of CO, CH
4
,orC
2+
products.
Figure 3
b highlights
how catalyst modi
fi
cation with the organic additive moves
product distribution almost completely away from CH
4
, to the
CO
−
C
2+
vector of the graph. While accessing the CO-rich
portion of the graph is commonplace in CO
2
R electrocatalysis,
the C
2+
-rich portion of the graph is only accessed in the
presence of
1-Br
2
.
11
The maximal selectivity was obtained with
aCu
0.85
Zn
0.15
catalyst where 96% of the CO
*
intermediate was
reduced to carbon-coupled products.
The breaking of the scaling relationship in the presence of
1-
Br
2
cannot be explained by morphological changes or alloy
segregation, as no nanostructuring was observed (
Figures S7
−
S10
, XRD in
Figures S12
−
S16
), therefore suggesting that the
molecular additive improves selectivity via changes in the
microkinetic pathway(s) in this system. The organic additive
may a
ff
ect CH
4
−
C
2+
branching point(s) by (i) alleviating a
rate limitation of the formation of the bound
*
CHO/
*
COH
intermediate and lowering the barrier toward C
−
C coupling or
(ii) promoting dimerization of the bound
*
CO relative to
hydrogenation toward CH
4
. In either case, kinetic competition
for the
*
CO would be enhanced in the presence of the
additive, which is consistent with the observation of a large and
negative Pearson correlation between C
2+
and CO (
Figure 2
e).
We additionally note that neither of these explanations for the
mechanism underlying the sc
aling law disruption has
implications for the selectivity within the set of C
2+
products.
As shown in
Figure S6
, additional correlations among these
products are observed both in the absence and in the presence
of the molecular additive, motivating future tuning of the
catalyst system to tackle other branching points in the reaction
network for enhanced control over product selectivity.
■
CONCLUSION
High-throughput screening of the CO
2
R activity and selectivity
of Cu alloys with Co, In, Mn, and Zn revealed the propensity
of organic additive
1-Br
2
to enable the development of hybrid
electrocatalysts that can reduce CO
2
to high-order products
with improved activity and selectivity. The large data set led to
the observation of a CH
4
−
C
2+
scaling relationship that
demonstrates a particularly robust link between these products
over a large range of conditions. The CH
4
−
C
2+
relationship
represents an intrinsic limitation of selectivity tuning through
alloying. However, it can be disrupted to favor C
2+
products by
the presence of the organic additive, highlighting the potential
of hybrid organic
−
inorganic catalysts to tune branching ratios
in the CO
2
R reaction network. These observations highlight
the importance of data-driven identi
fi
cation of relationships
that provide mechanistic insights to guide the study of complex
reactions and catalyst development. Disentangling the possible
explanations of the combined mechanistic in
fl
uence of the
additive and alloying elements will require substantial further
investigation that will be guided by the observed data
relationships elucidated in this study.
■
ASSOCIATED CONTENT
*
s
ı
Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acscentsci.1c00860
.
All material preparation and characterization, synthetic
and electrochemical procedures, and raw data (
PDF
)
■
AUTHOR INFORMATION
Corresponding Authors
Jonas C. Peters
−
Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena,
California 91125, United States
; Email:
jpeters@
caltech.edu
John M. Gregoire
−
Division of Engineering and Applied
Science, Liquid Sunlight Alliance, California Institute of
Technology, Pasadena, California 91125, United States;
orcid.org/0000-0002-2863-5265
; Email:
gregoire@
caltech.edu
Theodor Agapie
−
Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena,
California 91125, United States;
orcid.org/0000-0002-
9692-7614
; Email:
agapie@caltech.edu
Authors
Yungchieh Lai
−
Division of Engineering and Applied Science,
Liquid Sunlight Alliance, California Institute of Technology,
Pasadena, California 91125, United States
Nicholas B. Watkins
−
Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena,
California 91125, United States
Alonso Rosas-Hernández
−
Division of Chemistry and
Chemical Engineering, California Institute of Technology,
Pasadena, California 91125, United States
Arnaud Thevenon
−
Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena,
California 91125, United States
Gavin P. Heim
−
Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena,
California 91125, United States
Lan Zhou
−
Division of Engineering and Applied Science,
Liquid Sunlight Alliance, California Institute of Technology,
Pasadena, California 91125, United States
Yueshen Wu
−
Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena,
California 91125, United States;
orcid.org/0000-0002-
3784-0594
Complete contact information is available at:
https://pubs.acs.org/10.1021/acscentsci.1c00860
ACS Central Science
http://pubs.acs.org/journal/acscii
Research Article
https://doi.org/10.1021/acscentsci.1c00860
ACS Cent. Sci.
2021, 7, 1756
−
1762
1760
Author Contributions
†
Y.L. and N.B.W. contributed equally to this work.
Notes
The authors declare no competing
fi
nancial interest.
■
ACKNOWLEDGMENTS
This material is based on work performed by the Liquid
Sunlight Alliance, which is supported by the U.S. Department
of Energy, O
ffi
ce of Science, O
ffi
ce of Basic Energy Sciences,
Fuels from Sunlight Hub under Award DE-SC0021266. A.T.
acknowledges Marie Sk
ł
odowska-Curie Fellowship H2020-
MSCA-IF-2017 (793471). The Resnick Sustainability Institute
at Caltech is acknowledged for support of the laboratory
facilities in which this research was conducted.
■
REFERENCES
(1) Seh, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.;
Nørskov, J. K.; Jaramillo, T. F. Combining Theory and Experiment in
Electrocatalysis: Insights into Materials Design.
Science
2017
,
355
(6321), No. eaad4998.
(2) Pérez-Ramírez, J.; López, N. Strategies to Break Linear Scaling
Relationships.
Nat. Catal.
2019
,
2
(11), 971
−
976.
(3) Greeley, J. Theoretical Heterogeneous Catalysis: Scaling
Relationships and Computational Catalyst Design.
Annu. Rev. Chem.
Biomol. Eng.
2016
,
7
(1), 605
−
635.
(4) Liu, X.; Xiao, J.; Peng, H.; Hong, X.; Chan, K.; Nørskov, J. K.
Understanding Trends in Electrochemical Carbon Dioxide Reduction
Rates.
Nat. Commun.
2017
,
8
(1), 15438.
(5) Waldie, K. M.; Ostericher, A. L.; Reineke, M. H.; Sasayama, A.
F.; Kubiak, C. P. Hydricity of Transition-Metal Hydrides:
Thermodynamic Considerations for CO
2
Reduction.
ACS Catal.
2018
,
8
(2), 1313
−
1324.
(6) Ostericher, A. L.; Waldie, K. M.; Kubiak, C. P. Utilization of
Thermodynamic Scaling Relationships in Hydricity To Develop
Nickel Hydrogen Evolution Reaction Electrocatalysts with Weak
Acids and Low Overpotentials.
ACS Catal.
2018
,
8
(10), 9596
−
9603.
(7) Martin, D. J.; Wise, C. F.; Pegis, M. L.; Mayer, J. M. Developing
Scaling Relationships for Molecular Electrocatalysis through Studies
of Fe-Porphyrin-Catalyzed O2 Reduction.
Acc. Chem. Res.
2020
,
53
(5), 1056
−
1065.
(8) Vojvodic, A.; Nørskov, J. K. New Design Paradigm for
Heterogeneous Catalysts.
Natl. Sci. Rev.
2015
,
2
(2), 140
−
143.
(9) Jones, R. J. R.; Wang, Y.; Lai, Y.; Shinde, A.; Gregoire, J. M.
Reactor Design and Integration with Product Detection to Accelerate
Screening of Electrocatalysts for Carbon Dioxide Reduction.
Rev. Sci.
Instrum.
2018
,
89
(12), 124102.
(10) Lai, Y.; Jones, R. J. R.; Wang, Y.; Zhou, L.; Richter, M. H.;
Gregoire, J. The Sensitivity of Cu for Electrochemical Carbon Dioxide
Reduction to Hydrocarbons as Revealed by High Throughput
Experiments.
J. Mater. Chem. A
2019
,
7
(47), 26785
−
26790.
(11) Nitopi, S.; Bertheussen, E.; Scott, S. B.; Liu, X.; Engstfeld, A. K.;
Horch, S.; Seger, B.; Stephens, I. E. L.; Chan, K.; Hahn, C.; Nørskov,
J. K.; Jaramillo, T. F.; Chorkendorff, I. Progress and Perspectives of
Electrochemical CO
2
Reduction on Copper in Aqueous Electrolyte.
Chem. Rev.
2019
,
119
(12), 7610
−
7672.
(12) Peterson, A. A.; Nørskov, J. K. Activity Descriptors for CO
2
Electroreduction to Methane on Transition-Metal Catalysts.
J. Phys.
Chem. Lett.
2012
,
3
(2), 251
−
258.
(13) Liu, X.; Schlexer, P.; Xiao, J.; Ji, Y.; Wang, L.; Sandberg, R. B.;
Tang, M.; Brown, K. S.; Peng, H.; Ringe, S.; Hahn, C.; Jaramillo, T.
F.; Nørskov, J. K.; Chan, K. PH Effects on the Electrochemical
Reduction of CO(2) towards C2 Products on Stepped Copper.
Nat.
Commun.
2019
,
10
(1), 32.
(14) Resasco, J.; Chen, L. D.; Clark, E.; Tsai, C.; Hahn, C.; Jaramillo,
T. F.; Chan, K.; Bell, A. T. Promoter Effects of Alkali Metal Cations
on the Electrochemical Reduction of Carbon Dioxide.
J. Am. Chem.
Soc.
2017
,
139
(32), 11277
−
11287.
(15) Hori, Y.; Takahashi, I.; Koga, O.; Hoshi, N. Electrochemical
Reduction of Carbon Dioxide at Various Series of Copper Single
Crystal Electrodes.
J. Mol. Catal. A: Chem.
2003
,
199
(1), 39
−
47.
(16) Jeon, H. S.; Kunze, S.; Scholten, F.; Roldan Cuenya, B. Prism-
Shaped Cu Nanocatalysts for Electrochemical CO2 Reduction to
Ethylene.
ACS Catal.
2018
,
8
(1), 531
−
535.
(17) Hahn, C.; Hatsukade, T.; Kim, Y.-G.; Vailionis, A.; Baricuatro,
J. H.; Higgins, D. C.; Nitopi, S. A.; Soriaga, M. P.; Jaramillo, T. F.
Engineering Cu Surfaces for the Electrocatalytic Conversion of CO
2
:
Controlling Selectivity toward Oxygenates and Hydrocarbons.
Proc.
Natl. Acad. Sci. U. S. A.
2017
,
114
(23), 5918
−
5923.
(18) Zhang, B.; Zhang, J.; Hua, M.; Wan, Q.; Su, Z.; Tan, X.; Liu, L.;
Zhang, F.; Chen, G.; Tan, D.; Cheng, X.; Han, B.; Zheng, L.; Mo, G.
Highly Electrocatalytic Ethylene Production from CO2 on Nano-
defective Cu Nanosheets.
J. Am. Chem. Soc.
2020
,
142
(31), 13606
−
13613.
(19) Wang, Y.; Wang, Z.; Dinh, C.-T.; Li, J.; Ozden, A.; Golam
Kibria, M.; Seifitokaldani, A.; Tan, C.-S.; Gabardo, C. M.; Luo, M.;
Zhou, H.; Li, F.; Lum, Y.; McCallum, C.; Xu, Y.; Liu, M.; Proppe, A.;
Johnston, A.; Todorovic, P.; Zhuang, T.-T.; Sinton, D.; Kelley, S. O.;
Sargent, E. H. Catalyst Synthesis under CO2 Electroreduction
Favours Faceting and Promotes Renewable Fuels Electrosynthesis.
Nat. Catal.
2020
,
3
(2), 98
−
106.
(20) Singh, M. R.; Kwon, Y.; Lum, Y.; Ager, J. W.; Bell, A. T.
Hydrolysis of Electrolyte Cations Enhances the Electrochemical
Reduction of CO2 over Ag and Cu.
J. Am. Chem. Soc.
2016
,
138
(39),
13006
−
13012.
(21) Lum, Y.; Ager, J. W. Sequential Catalysis Controls Selectivity in
Electrochemical CO
2
Reduction on Cu.
Energy Environ. Sci.
2018
,
11
(10), 2935
−
2944.
(22) Morales-Guio, C. G.; Cave, E. R.; Nitopi, S. A.; Feaster, J. T.;
Wang, L.; Kuhl, K. P.; Jackson, A.; Johnson, N. C.; Abram, D. N.;
Hatsukade, T.; Hahn, C.; Jaramillo, T. F. Improved CO 2 Reduction
Activity towards C 2+ Alcohols on a Tandem Gold on Copper
Electrocatalyst.
Nat. Catal.
2018
,
1
(10), 764
−
771.
(23) Zhong, M.; Tran, K.; Min, Y.; Wang, C.; Wang, Z.; Dinh, C.-T.;
De Luna, P.; Yu, Z.; Rasouli, A. S.; Brodersen, P.; Sun, S.; Voznyy, O.;
Tan, C.-S.; Askerka, M.; Che, F.; Liu, M.; Seifitokaldani, A.; Pang, Y.;
Lo, S.-C.; Ip, A.; Ulissi, Z.; Sargent, E. H. Accelerated Discovery of
CO 2 Electrocatalysts Using Active Machine Learning.
Nature
2020
,
581
(7807), 178
−
183.
(24) Wang, J.; Cheng, T.; Fenwick, A. Q.; Baroud, T. N.; Rosas-
Hernández, A.; Ko, J. H.; Gan, Q.; Goddard III, W. A.; Grubbs, R. H.
Selective CO
2
Electrochemical Reduction Enabled by a Tricompo-
nent Copolymer Modifier on a Copper Surface.
J. Am. Chem. Soc.
2021
,
143
(7), 2857
−
2865.
(25) Wei, X.; Yin, Z.; Lyu, K.; Li, Z.; Gong, J.; Wang, G.; Xiao, L.;
Lu, J.; Zhuang, L. Highly Selective Reduction of CO2 to C2+
Hydrocarbons at Copper/Polyaniline Interfaces.
ACS Catal.
2020
,
10
(7), 4103
−
4111.
(26) Chen, X.; Chen, J.; Alghoraibi, N. M.; Henckel, D. A.; Zhang,
R.; Nwabara, U. O.; Madsen, K. E.; Kenis, P. J. A.; Zimmerman, S. C.;
Gewirth, A. A. Electrochemical CO 2 -to-Ethylene Conversion on
Polyamine-Incorporated Cu Electrodes.
Nat. Catal.
2021
,
4
(1), 20
−
27.
(27) Buckley, A. K.; Lee, M.; Cheng, T.; Kazantsev, R. V.; Larson, D.
M.; Goddard III, W. A.; Toste, F. D.; Toma, F. M. Electrocatalysis at
Organic
−
Metal Interfaces: Identification of Structure
−
Reactivity
Relationships for CO
2
Reduction at Modified Cu Surfaces.
J. Am.
Chem. Soc.
2019
,
141
(18), 7355
−
7364.
(28) Banerjee, S.; Han, X.; Thoi, V. S. Modulating the Electrode
−
Electrolyte Interface with Cationic Surfactants in Carbon Dioxide
Reduction.
ACS Catal.
2019
,
9
(6), 5631
−
5637.
(29) Han, Z.; Kortlever, R.; Chen, H.-Y.; Peters, J. C.; Agapie, T.
CO2 Reduction Selective for C
≥
2 Products on Polycrystalline
Copper with N-Substituted Pyridinium Additives.
ACS Cent. Sci.
2017
,
3
(8), 853
−
859.
ACS Central Science
http://pubs.acs.org/journal/acscii
Research Article
https://doi.org/10.1021/acscentsci.1c00860
ACS Cent. Sci.
2021, 7, 1756
−
1762
1761
(30) Gao, D.; Arán-Ais, R. M.; Jeon, H. S.; Roldan Cuenya, B.
Rational Catalyst and Electrolyte Design for CO2 Electroreduction
towards Multicarbon Products.
Nat. Catal.
2019
,
2
(3), 198
−
210.
(31) He, J.; Dettelbach, K. E.; Salvatore, D. A.; Li, T.; Berlinguette,
C. P. High-Throughput Synthesis of Mixed-Metal Electrocatalysts for
CO2 Reduction.
Angew. Chem., Int. Ed.
2017
,
56
(22), 6068
−
6072.
(32) Nam, D.-H.; De Luna, P.; Rosas-Hernández, A.; Thevenon, A.;
Li, F.; Agapie, T.; Peters, J. C.; Shekhah, O.; Eddaoudi, M.; Sargent, E.
H. Molecular Enhancement of Heterogeneous CO 2 Reduction.
Nat.
Mater.
2020
,
19
(3), 266
−
276.
(33) Thevenon, A.; Rosas-Hernández, A.; Peters, J. C.; Agapie, T.
In-Situ Nanostructuring and Stabilization of Polycrystalline Copper
by an Organic Salt Additive Pr
omotes Electrocatalytic CO
2
Reduction to Ethylene.
Angew. Chem., Int. Ed.
2019
,
58
(47),
16952
−
16958.
(34) Li, F.; Thevenon, A.; Rosas-Hernández, A.; Wang, Z.; Li, Y.;
Gabardo, C. M.; Ozden, A.; Dinh, C. T.; Li, J.; Wang, Y.; Edwards, J.
P.; Xu, Y.; McCallum, C.; Tao, L.; Liang, Z.-Q.; Luo, M.; Wang, X.;
Li, H.; O
’
Brien, C. P.; Tan, C.-S.; Nam, D.-H.; Quintero-Bermudez,
R.; Zhuang, T.-T.; Li, Y. C.; Han, Z.; Britt, R. D.; Sinton, D.; Agapie,
T.; Peters, J. C.; Sargent, E. H. Molecular Tuning of CO 2 -to-
Ethylene Conversion.
Nature
2020
,
577
(7791), 509
−
513.
(35) Thevenon, A.; Rosas-Hernández, A.; Fontani Herreros, A. M.;
Agapie, T.; Peters, J. C. Dramatic HER Suppression on Ag Electrodes
via Molecular Films for Highly Selective CO2 to CO Reduction.
ACS
Catal.
2021
,
11
(8), 4530
−
4537.
(36) Ren, D.; Fong, J.; Yeo, B. S. The Effects of Currents and
Potentials on the Selectivities of Copper toward Carbon Dioxide
Electroreduction.
Nat. Commun.
2018
,
9
(1), 925.
(37) Hori, Y. Electrochemical CO
2
Reduction on Metal Electrodes.
In
Modern Aspects of Electrochemistry
; Vayenas, C. G., White, R. E.,
Gamboa-Aldeco, M. E., Eds.; Springer: New York, 2008; Vol.
42
,pp
89
−
189.
(38) Kortlever, R.; Shen, J.; Schouten, K. J. P.; Calle-Vallejo, F.;
Koper, M. T. M. Catalysts and Reaction Pathways for the
Electrochemical Reduction of Carbon Dioxide.
J. Phys. Chem. Lett.
2015
,
6
(20), 4073
−
4082.
(39) Hori, Y.; Murata, A.; Takahashi, R. Formation of Hydrocarbons
in the Electrochemical Reduction of Carbon Dioxide at a Copper
Electrode in Aqueous Solution.
J. Chem. Soc., Faraday Trans. 1
1989
,
85
(8), 2309.
(40) Murata, A.; Hori, Y. Product Selectivity Affected by Cationic
Species in Electrochemical Reduction of CO
2
and CO at a Cu
Electrode.
Bull. Chem. Soc. Jpn.
1991
,
64
(1), 123
−
127.
(41) Saberi Safaei, T.; Mepham, A.; Zheng, X.; Pang, Y.; Dinh, C.-T.;
Liu, M.; Sinton, D.; Kelley, S. O.; Sargent, E. H. High-Density
Nanosharp Microstructures Enable Efficient CO
2
Electroreduction.
Nano Lett.
2016
,
16
(11), 7224
−
7228.
(42) Cheng, T.; Xiao, H.; Goddard, W. A. Full Atomistic Reaction
Mechanism with Kinetics for CO Reduction on Cu(100) from Ab
Initio Molecular Dynamics Free-Energy Calculations at 298 K.
Proc.
Natl. Acad. Sci. U. S. A.
2017
,
114
(8), 1795
−
1800.
(43) Garza, A. J.; Bell, A. T.; Head-Gordon, M. Mechanism of CO2
Reduction at Copper Surfaces: Pathways to C2 Products.
ACS Catal.
2018
,
8
(2), 1490
−
1499.
(44) Schreier, M.; Yoon, Y.; Jackson, M. N.; Surendranath, Y.
Competition between H and CO for Active Sites Governs Copper-
Mediated Electrosynthesis of Hydrocarbon Fuels.
Angew. Chem., Int.
Ed.
2018
,
57
(32), 10221
−
10225.
(45) Xiao, H.; Cheng, T.; Goddard, W. A. Atomistic Mechanisms
Underlying Selectivities in C
1
and C
2
Products from Electrochemical
Reduction of CO on Cu(111).
J. Am. Chem. Soc.
2017
,
139
(1), 130
−
136.
(46) Other potential ways to disrupt the scaling relationship
observed here include nanostructuring,
16
−
18
higher overpotentials and
current densities resulting in pH gradients,
12
−
15
,
20
and bimetallic
system engineering.
21
−
23
ACS Central Science
http://pubs.acs.org/journal/acscii
Research Article
https://doi.org/10.1021/acscentsci.1c00860
ACS Cent. Sci.
2021, 7, 1756
−
1762
1762