Promoter Effects of Alkali Metal Cations on the Electrochemical Reduction of Carbon Dioxide

Article abstract of DOI:10.1021/jacs.7b06765

The electrochemical reduction of CO₂ is known to be influenced by the identity of the alkali metal cation in the electrolyte; however, a satisfactory explanation for this phenomenon has not been developed. Here we present the results of experimental and theoretical studies aimed at elucidating the effects of electrolyte cation size on the intrinsic activity and selectivity of metal catalysts for the reduction of CO₂. Experiments were conducted under conditions where the influence of electrolyte polarization is minimal in order to show that cation size affects the intrinsic rates of formation of certain reaction products, most notably for HCOO^-, C₂H₄, and C₂H₅OH over Cu(100)- and Cu(111)-oriented thin films, and for CO and HCOO^- over polycrystalline Ag and Sn. Interpretation of the findings for CO₂ reduction was informed by studies of the reduction of glyoxal and CO, key intermediates along the reaction pathway to final products. Density functional theory calculations show that the alkali metal cations influence the distribution of products formed as a consequence of electrostatic interactions between solvated cations present at the outer Helmholtz plane and adsorbed species having large dipole moments. The observed trends in activity with cation size are attributed to an increase in the concentration of cations at the outer Helmholtz plane with increasing cation size.

Full text of DOI:10.1021/jacs.7b06765

Subscriber access provided by UNIV OF NEWCASTLE
Article
Promoter effects of alkali metal cations on the
electrochemical reduction of carbon dioxide
Joaquin Resasco, Leanne D. Chen, Ezra Clark, Charlie Tsai,
Christopher Hahn, Thomas F. Jaramillo, Karen Chan, and Alexis T. Bell
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b06765 • Publication Date (Web): 24 Jul 2017
Downloaded from http://pubs.acs.org on July 25, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 28
Journal of the American Chemical Society
1
2
3
4
Promoter effects of alkali metal cations on the electrochemical reduction of carbon dioxide
5
6
7
8
9
Joaquin Resasco1,2, Leanne D. Chen3,4,5, Ezra Clark1,2, Charlie Tsai3,4 Christopher Hahn3,4, Thomas F.
Jaramillo3,4, Karen Chan3,4, and Alexis T. Bell1,2*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1. Department of Chemical Engineering, University of California, Berkeley, CA 94720
2. Joint Center for Artificial Photosynthesis, Material Science Division, Lawrence Berkeley National
Laboratory, Berkeley, CA 94720
3. SUNCAT Center for Interface Science and Catalysis, Department of Chemical Engineering, Stanford
University, Stanford, CA 94305
4. SUNCAT Center for Interface Science and Catalysis, SLAC National Accelerator Laboratory, Menlo
Park, CA 94025
5. Present Address: Division of Chemistry and Chemical Engineering, California Institute of Technology,
Pasadena, CA 91125
Submitted to
Journal of the American Chemical Society
March 7, 2017
*To whom correspondence should be addressed: alexbell@berkeley.edu
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 28
1
2
3
4
5
6
Abstract
7
8
9
The electrochemical reduction of CO2 is known to be influenced by the identity of the alkali
metal cation in the electrolyte; however, a satisfactory explanation for this phenomenon has not been
developed. Here we present the results of experimental and theoretical studies aimed at elucidating the
effects of electrolyte cation size on the intrinsic activity and selectivity of metal catalysts for the reduction
of CO2. Experiments were conducted under conditions where the influence of electrolyte polarization is
minimal in order to show that cation size affects the intrinsic rates of formation of certain reaction
products, most notably for HCOOꢀ, C2H4, and C2H5OH over Cu(100)ꢀ and Cu(111)ꢀoriented thin films,
and for CO and HCOOꢀ over polycrystalline Ag and Sn. Interpretation of the findings for CO2 reduction
was informed by studies of the reduction of glyoxal and CO, key intermediates along the reaction
pathway to final products. Density functional theory calculations show that the alkali metal cations
influence the distribution of products formed as a consequence of electrostatic interactions between
solvated cations present at the outer Helmholtz plane and adsorbed species having large dipole moments.
The observed trends in activity with cation size are attributed to an increase in the concentration of
cations at the outer Helmholtz plane with increasing cation size.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Keywords: Electrochemical reduction, CO2, copper, silver
2
ACS Paragon Plus Environment

Page 3 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
Introduction
8
9
The electrochemical reduction of CO2 offers a means for storing electrical energy produced by
renewable but intermittent resources, such as wind and solar radiation.1, 2 Hydrocarbons and alcohols,
rather than carbon monoxide and formic acid, are the preferred products of CO2 reduction because of their
high energy density. Despite extensive efforts aimed at identifying electrocatalysts that can produce these
products, copper (Cu) remains the only material capable of doing so with significant yields.3ꢀ5
Nevertheless, Cu requires a high overpotential (~−1 V versus RHE) and produces a broad spectrum of
products.6,7 Clearly, the discovery of novel means for reducing CO2 to desirable products with higher
efficiency and selectivity is needed. To do so requires understanding the fundamental processes
occurring at the electrode surface. In addition to metalꢀadsorbate interactions, a full description of
electrochemical reactions must include the effects of solvation and interfacial electric fields. A number of
studies have shown that one way of influencing the product distribution is through changes in the
electrolyte cation.8ꢀ13
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The role of cations is particularly interesting because a number of studies have shown that both
the activity and selectivity of Ag, Hg, and Cu for the CO2 reduction reaction (CO2RR) are influenced
significantly by the size of the alkali metal cation in the electrolyte.8ꢀ13 Similar effects have also been
observed for the electrochemical reduction of oxygen, the oxidation of hydrogen, and the oxidation of low
molecular weight alcohols.14ꢀ18 For Ag and Hg, which are selective for CO2 reduction to carbon monoxide
(CO) and formate anions (HCOOꢀ), respectively, increasing the alkali metal cation size increases the rates
of formation of these products.8, 10, 11 For Cu, increasing alkali metal size leads to higher selectivities to C2
products, e.g., ethylene and ethanol.9, 12, 13 Explanations for these phenomena have been attributed to
differences in local pH, or differences in kinetic overpotentials due to the electrochemical potential in the
outer Helmholtz plane being affected by cation size.8, 10, 11, 13
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 28
1
2
3
4
5
6
7
8
9
One of the challenges to finding a consistent interpretation for the effects of cation size on the
intrinsic kinetics of CO2 reduction is that many studies are carried out under conditions where mass
transport also contributes to the observed activity and product distribution. In recent study, we reported
the effects of cation size on the rate of the CO2RR under strong massꢀtransport limitations.13 We showed
that the increase in pH and the decrease in CO2 concentration near the cathode, caused by inadequate
mass transfer through the hydrodynamic boundary layer in front of the cathode, could be offset by the
hydrolysis of solvated cations. We showed that the increased buffering effect of solvated cations with
cation size could be captured theoretically and suggested that this effect is responsible for the observed
changes in the overall current density and the distribution of product current densities with cation size.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In this work, we present a combined experimental and theoretical study of the effects of
electrolyte cation size on the intrinsic activity and selectivity of metal catalysts for the reduction of CO2.
In contrast to earlier studies, experiments were conducted at sufficiently low applied voltages to avoid the
effects of concentration polarization of the electrolyte, so that the influence of cation size on the intrinsic
activity and selectivity of each catalyst could be observed. Most of our work was conducted using
epitaxially grown Cu(100) or Cu(111) thin films; a more limited study was carried out using
polycrystalline Ag and Sn films. Density functional theory (DFT) calculations were used to investigate
how cations and cation size affect the stabilization of reaction intermediates involved in the reduction of
CO2. These calculations demonstrate that adsorbed species with large dipole moments oriented away from
the solvent are stabilized by electrostatic interactions with solvated cations, while trends in activity with
cation size arise from differences in ion concentration at the Helmholtz plane.
Methods
Electrode Preparation
4
ACS Paragon Plus Environment

Page 5 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Singleꢀside polished Si(100) or (110) wafers (Virginia semiconductor, 1ꢀ10 ꢁ cm) were diced
into ~4 cm2 sized pieces that were then used as electrode substrates. Prior to Cu deposition, the native
oxide was removed from the Si substrates by submerging them in a 10% HF solution for 5 min.
Immediately after HF etching, the Si pieces were transferred to a vacuum chamber for sputter deposition
of Cu in an AJA ATC Orionꢀ5 sputtering system. The base pressure of the sputtering system prior to
deposition was ~2x10ꢀ7 torr. The flow rate of the sputtering gas (Ar) was 25 sccm and the sputtering
pressure was adjusted to 2x10ꢀ3 torr by controlling the speed at which the chamber was pumped, using a
variable butterfly valve. Cu (99.999% Kurt Lesker) was deposited at a rate of 1 Å/s, as determined by a
calibrated quartz crystal monitor, at ambient temperature. The total film thickness deposited was 100 nm.
Polished, polycrystalline foils of Ag and Sn (99.99% Alfa Aesar) were also used as electrodes without
further preparation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Electrode Characterization
The structure of the Cu thin films was characterized by Xꢀray diffraction. The orientation and
epitaxial quality of the films were determined using symmetric θꢀ2θ scans, in plane ϕ scans, ω scans or
rocking curves, and pole figures. XRD patterns were taken with a PANanalytical X'Pert diffractometer,
which uses a Cu Kα (λ = 1.54056 Å) Xꢀray source. Symmetric θꢀ2θ scans were collected on samples
fixed onto a flat glass slide in lockedꢀcoupled mode with a goniometer resolution of 0.001°. Measured
diffraction patterns were compared to known standards taken from the International Center for Diffraction
Data (ICDD) PDF4 database (card #71ꢀ4610 for Cu).
Electrochemical Measurements
All electrochemical experiments were conducted in a gasꢀtight electrochemical cell machined
from polyether ether ketone (PEEK).19 The cell was cleaned with 20 wt. % nitric acid and oxidized in
UVꢀgenerated ozone for 15 min prior to the initiation of an experiment. The working and counter
electrodes were parallel and separated by an anionꢀconducting membrane (Selemion AMV AGC Inc.). A
gas dispersion frit was incorporated into the cathode chamber to provide vigorous electrolyte mixing. The
exposed geometric surface area of each electrode was 1 cm2 and the electrolyte volume of each electrode
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 28
1
2
3
4
5
6
7
8
9
chamber was 1.8 mL. The counter electrode was a Pt foil (99.9% Sigma Aldrich) that was flame annealed
prior to each experiment. The working electrode potential was referenced against an Ag/AgCl electrode
(Innovative Instruments Inc.) that was calibrated against a homemade standard hydrogen electrode (SHE).
0.05 M M2CO3 (M referring to an alkali metal cation) solutions were prepared by mixing ultraꢀpure salts
(Sigma Aldrich 99.995%) and 18.2 Mꢁ DI, and were used as the electrolyte without further purification.
The cathode chamber was sparged with CO2 (99.999% Praxair) at a rate of 5 sccm for 20 min prior to and
throughout the duration of all electrocatalytic measurements. Upon saturation with CO2 the pH of the
electrolyte was 6.8, which was maintained throughout the duration of electrolysis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Electrochemical measurements were performed using a Biologic VSPꢀ300 potentiostat. All
electrochemical data were recorded versus the reference electrode and converted to the RHE scale.
Potentiostatic electrochemical impedance spectroscopy (PEIS) was used to determine the uncompensated
resistance (Ru) of the electrochemical cell by applying voltage waveforms about the open circuit potential
with an amplitude of 20 mV and frequencies ranging from 50 Hz to 500 kHz. The potentiostat
compensated for 85% of Ru in situ and the last 15% was postꢀcorrected to arrive at accurate potentials.
The electrocatalytic activity was assessed by conducting chronoamperometry at each fixed applied
potential for 70 min.
Product Analysis
The effluent from the electrochemical cell was passed through the sampling loop (250 ꢂL) of an
Agilent 7890B gas chromatograph equipped with a pulsedꢀdischarge helium ionization detector (PDHID).
He (99.9999% Praxair) was used as the carrier gas. The effluent of the electrochemical cell was sampled
every 14 min. The gaseous products were separated using a HayesepꢀQ capillary column (Agilent)
connected in series with a packed ShinCarbon ST column (Restek Co.). The column oven was maintained
at 50 °C for 1 min followed by a temperature ramp at 30 °C/min to 250 °C, which was maintained for the
duration of the analysis. The signal response of the PDHID to each gaseous product was calibrated by
analyzing a series of NISTꢀtraceable standard gas mixtures (Air Gas).
6
ACS Paragon Plus Environment

Page 7 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
The electrolyte from both electrode chambers was collected after electrolysis and analyzed using
a Thermo Scientific UltiMate 3000 liquid chromatograph equipped with a refractive index detector (RID).
The electrolyte aliquots were stored in a refrigerated autosampler until analyzed in order to minimize the
evaporation of volatile products. The liquidꢀphase products contained in a 10 ꢂL sample were separated
using a series of two Aminex HPX 87ꢀH columns (BioꢀRad) and a 1 mM sulfuric acid eluent (99.999%
Sigma Aldrich). The column oven was maintained at 60 °C for the duration of the analysis. The signal
response of the RID to each liquidꢀphase product was calibrated by analyzing standard solutions of each
product at a concentration of 1, 10, and 50 mM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Theoretical Calculations
Planeꢀwave density functional theory (DFT) employing periodic boundary conditions was used
for all calculations in this study. The Quantum ESPRESSO20 code and the Atomic Simulation
Environment (ASE)21 were used, along with ultrasoft pseudopotentials and the Bayesian error estimation
exchangeꢀcorrelation functional with van der Waals interactions (BEEFꢀvdW).22 This functional has been
optimized for both chemisorption energies and longꢀrange van der Waals interactions. For all
calculations, a planeꢀwave cutoff of 500 eV and a density cutoff of 5000 eV were used, based on
convergence tests from previous studies.23 A MonkhorstꢀPack24 k-point grid of (4×4×1) was used for
(3×3) cells while a (2×2×1) grid was used for (6×6) cells. All systems were modeled using a periodic
Cu(111) slab, two iceꢀlike water bilayers, the solvated ion and at least 12 Å of vacuum perpendicular to
the surface. The field effect on various CO2 adsorbates without ions or solvating waters was also
investigated by applying a sawtooth potential in each simulation cell, for moderate field strengths where
the vacuum energy remains above the Fermi level.25 As an example, Figure S3 in the Supporting
Information (SI) shows the average potential energy along the z-direction for CH3 on Cu (111), with an
applied sawtooth potential.
Constrained minima hopping26 (CMH) was used for the global optimization of water structures
surrounding each solvated ion. CMH enforces molecular identity by constraining the atomic distances
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 28
1
2
3
4
5
6
7
8
9
within each molecular entity through Hookean constraints. The atoms are thermalized in a Maxwellꢀ
Boltzmann distribution at an initial temperature of 1000 K, then evolved through molecular dynamics and
relaxed to locate new local minima. To save computational cost, CMH was performed on two layers of
Cu. The Cu atoms were fixed, while all others were allowed to relax. The convergence criterion for the
CMH was satisfied when three successive lowest energy minima were found within 0.05 eV of each
other. Thereafter, the system was relaxed until the forces on all atoms were less than 0.05 eV/Å. Finally,
the third layer of Cu was added back into the system to perform a single point calculation and obtain a
more accurate value for the energy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Results
Cation effects on Cu(100) and Cu(111) oriented films
8
ACS Paragon Plus Environment

Page 9 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
Figure 1: Structural characterization of oriented Cu thin films. a) Symmetric θꢀ2θ, in plane ϕ, and ω
rocking curve scans for Cu deposited on Si(100). b) Symmetric θꢀ2θ, in plane ϕ, and ω rocking curve
scans for Cu deposited on Si(110). c) Cu(111) pole figure for Cu deposited on Si(100). d) Cu(200) pole
figure for Cu deposited on Si(110).
7
8
9
Epitaxially grown Cu thin films were used as electrocatalysts in order to have wellꢀdefined
catalytic surfaces.27 These films were prepared by radioꢀfrequency sputtering of Cu onto silicon (Si)
single crystal substrates, taking advantage of the epitaxial relationship between Cu and Si substrates of
different orientations.27ꢀ28 Most of our studies were conducted on the Cu(100) surface because it has been
shown that polycrystalline Cu consists mostly of (100) oriented crystallites and the Cu(100) surface has a
higher selectivity to more desirable products such as C2+ hydrocarbons and oxygenates (e.g., ethylene and
ethanol).29ꢀ32 Symmetric θꢀ2θ XRD patterns of 100 nm thick Cu films deposited on Si(100) and Si(110)
substrates are shown in Figure 1. The observation of a single Cu(200) diffraction peak for Cu deposited
on Si(100) indicates that the film is oriented with the (100) crystallographic direction normal to the film
plane. Similarly, for Cu deposited on Si(110) only the diffraction peak associated with Cu(111) is
observed.27ꢀ28 While symmetric XRD scans establish the outꢀofꢀplane texture relationships, both outꢀofꢀ
plane and inꢀplane texture analyses are necessary to determine whether the Cu thin films grow epitaxially
on Si, as a film with strong fiber texture would show a similar diffraction pattern. To this end, in planeꢀ
texture analysis was conducted using Xꢀray pole figures. Pole figures represent the texture of a material
and show the distribution of particular crystallographic directions in a stereographic projection. The
Cu(111) Xꢀray pole figure for Cu(100) on Si(100) exhibits discrete Bragg reflections, indicating cubeꢀonꢀ
cube epitaxial growth of Cu on the Si(100) substrate. Fourꢀfold symmetry is observed for the Cu(111)
Bragg reflections with an azimuthal angle of 90° apart, which is expected for a Cu (100) singleꢀcrystal.
For Cu (111) on Si (110), sixꢀfold symmetry is observed for the Cu (200) Bragg reflections, indicating
both strong outꢀofꢀplane and inꢀplane texture and thus epitaxial growth on the Si(110) substrate (Figure
1a). Six diffraction spots in the pole figure are seen instead of three because there are two discrete sets of
crystallites from twinning with an azimuthal angle of 60° apart. These results demonstrate that Cu thin
films can be grown epitaxially with both (100) and (111) orientations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2: Effect of alkali metal cations on the total activity over Cu(100). a) Linear sweep
voltammograms for CO2 reduction on Cu(100) in CO2 saturated 0.1 M bicarbonate electrolytes containing
different metal cations. b) Average current densities obtained during bulk electrolysis as a function of
metal cation at different potentials.
The steady state activity and selectivity of the Cu(100) surface was investigated in CO2ꢀsaturated
bicarbonate electrolytes containing different alkali metal cations, by using potentiometric electrolysis at
potentials from −0.7 to −1.1 V vs the reversible hydrogen electrode (RHE). Figure 2 shows the total
current density as a function of electrolyte cation at various potentials. From both steady state
potentiostatic and potential sweep measurements, it is clear that the total current density increases with
cation size, and this trend is observed at all applied potentials. To identify the contributions to the increase
in the total current density, we examined the effects of cation size on the product distribution. As large
differences in total current density exist for electrolytes containing different cations, partial currents
provide a better representation of trends in product formation rates than faradaic efficiencies. Complete
information regarding faradaic efficiencies can be found in the supplementary information (Figure S1,
Table S1).
10
ACS Paragon Plus Environment

Page 11 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3: Cation effect on the rates of formation of major products of CO2 reduction over Cu(100).
Partial current densities for each of the major products as a function of the electrolyte metal cation on
Cu(100). Data are presented at potential between −0.7 to −1.1 V vs RHE.
Figure 3 shows the partial current density for each of the major products as a function of
electrolyte cation. Formation rates of minor liquid products can be found in the supplementary
information (Table S1). A key finding is that the production rates of hydrogen and carbon monoxide are
relatively unaffected by increasing cation size, whereas the rates of formate, ethylene, and ethanol
formation increase monotonically with increasing cation size. The rate of methane production reaches a
maximum for Na+ꢀ and K+ꢀ based electrolytes and is lower in all other electrolytes. The trends seen in Fig.
3 are consistent with previous amperometric measurements.9 It is notable that these trends are
independent of the applied overpotential.
To probe the effects of mass transport limitations, the CO2 flow rate into the electrochemical cell
was varied. As the CO2 flow rate is increased, the hydrodynamic boundary layer thickness will decrease,
resulting in a larger maximum diffusive flux of CO2 to the electrode surface. At low reaction rates (i.e.,
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 28
1
2
3
4
5
6
7
8
9
low current densities), for which diffusion of CO2 is sufficiently rapid to prevent depletion at the cathode
surface, increasing the flowrate of CO2 should have no effect on catalyst activity. Whereas under mass
transport limited conditions, increased reactant availability will affect the product distribution
measured.13, 32 Figure 4 shows that doubling the CO2 flow rate from 5 to 10 sccm has no impact on the
measured selectivity or activity of Cu (100) at −0.9 V vs RHE and −1.0 V vs RHE. However, at −1.1 V vs
RHE, doubling the CO2 flowrate decreases the faradic efficiency to H2 and CH4 and increases the faradaic
efficiency for C2+ products. As shown previously, these trends are due to the increased rate of mass
transfer at higher gas flowrates, which improves the supply of CO2 to the cathode.32 Therefore, the effects
of cation size on the partial current densities observed in Figure 3 at potentials more positive than ~ −1.0
V vs RHE are ascribable to the influence of cation size on the intrinsic kinetics of Cu(100), particularly
for the formation of formate, ethylene, and ethanol, and to a lesser degree, methane.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: CO2 flow rate studies for probing external mass transport limitations: Current efficiency
(bars) and total current density (dots) for CO2 reduction over Cu(100) at different flow rates of CO2.
12
ACS Paragon Plus Environment

Page 13 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5: Cation effect on the rates of formation of major products of CO2 reduction over Cu(111).
Partial current densities for each of the major products as a function of the electrolyte metal cation on
Cu(111). Data are presented for the potential of ꢀ1.0 V vs RHE.
The effects of cation size were also investigated for CO2 reduction over a Cu(111) surface and the
results are presented in Fig. 5. The trends with increasing cation size for the closeꢀpacked Cu(111) surface
are consistent with those observed for the Cu(100) surface. Partial currents to HCOOꢀ, C2H4, and C2H5OH
increase with increasing cation size while rates of formation of H2, CO, and CH4 are less affected by
cation size. However, the effect of cation size is more pronounced on the closeꢀpacked surface, resulting
in high selectivity (~40% FE) to C2H4 in CsHCO3 even on a Cu(111) surface, which is known to be more
selective for producing CH4 than C2H4 in K+ꢀcontaining electrolytes.29, 31, 33ꢀ36
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6: Relative effects of alkali metal cations. Partial currents for ethylene, ethanol, and formate
formation as a function of the electrolyte composition on Cu(100). Data are presented for a potential of
−1.0 V vs RHE. A mixture of LiHCO3 and XHCO3 electrolyte is used with X being Na, K, Cs (second
cation). A fixed total salinity of 0.1 M was used in all experiments.
Relative effects of alkali metal cations
To further investigate the effects of the alkali metal cations, electrolysis was performed with
mixtures of small (Li+) and larger (Na+, K+, Cs+) cations in the electrolyte, maintaining the total salinity at
0.1 M. The results are shown in Figure 6. It is evident that the composition of the electrolyte has a
noticeable effect on the partial currents for HCOOꢀ, C2H4, and C2H5OH. No significant effect on the
partial currents of H2, CO, and CH4, was observed. For the latter three products, the effect of the larger
cation dominates that of the Li+ cation, and even at relatively low percentages of the larger cation.
Conversely, an electrolyte mixture that is dilute in the smaller cation results in identical selectivity to a
solution that does not contain the small cation. These results suggest that the concentration of larger
cations near the cathode surface is larger and/or that the larger cations have a much more significant
promotional effect on the production of C2H4, C2H5OH, and HCOOꢀ.
14
ACS Paragon Plus Environment

Page 15 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7: Cation effect on two electron products. Partial currents for each of the major products as a
function of the electrolyte metal cation on polycrystalline Ag and Sn. Data are presented for a potential of
ꢀ1.0 V vs RHE.
Cation effect on polycrystalline Ag and Sn
As carbon monoxide is a reaction intermediate which undergoes further reduction on Cu, the
production rate of CO itself is less straightforward to interpret than that of the other major products.37
Therefore, to investigate the effect of cation size on CO production, and to provide further insights into
the trends observed for Cu(100) and Cu(111), experiments were conducted on polycrystalline Ag and Sn
foils—metals that selectively produce CO and HCOOꢀ, respectively.3, 38 As seen in Fig. 7, the current
densities for CO and HCOOꢀ production on both metals increases with cation size. This suggests that
increasing cation size facilitates activation of CO2. Here again, we see that as with Cu(100) and Cu(111),
the rate of H2 evolution is unaffected by cation size. These trends are also consistent with previous
literature reports.8ꢀ11, 13, 39
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8: Cation effects of the reduction of reaction intermediates. aꢀc) Partial currents for hydrogen,
ethanol, and acetaldehyde formation from the reduction of glyoxal on Cu(100). Data is presented at a
potential of ꢀ0.6 V vs RHE. dꢀf) Partial currents for hydrogen, ethylene, and ethanol formation from the
reduction of CO on Cu(100). Data is presented at a potential of ꢀ0.8 V vs RHE.
Cation effects on elementary reaction steps: reduction of intermediates
As shown above, the size of the electrolyte cation influences the formation of C2 products. Recent
theoretical studies suggest that CꢀC bonds are formed via the reaction of two molecules of adsorbed CO
or by the reaction of adsorbed CO with a formyl group, HCO.40 A recent study has shown that the latter
process dominates at applied potentials near –1 V and leads to the formation of adsorbed HC(O)ꢀC(O)
species, which then undergoes hydrogenation to form C2H4 or CH3CHO and C2H5OH.40 Experimental
studies suggest that the precursor to CH3CHO and C2H5OH is glyoxal (OHCꢀCHO).41 Therefore, to
investigate which steps in the formation of C2 products are affected by the electrolyte cation size, we
carried out experiments in electrolyte containing glyoxal. As shown in Figure 8, cation size has no effect
on the partial currents of CH3CHO and C2H5OH formation, which in this case require no CꢀC bond
16
ACS Paragon Plus Environment

Page 17 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
formation as glyoxal is the starting reagent. Consistent with previous observations, no C2H4 was observed
during the hydrogenation of glyoxal.41 These findings strongly suggest that the size of the electrolyte
cation affects the rate of CꢀC bond formation, the rate increasing with increasing cation size, but has a
negligible effect on the hydrogenation of glyoxal to CH3CHO and C2H5OH.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As larger cations facilitate CO2 activation, evidenced by the increase in rates to CO and HCOOꢀ
(see Figure 3), experiments were carried out in electrolyte solutions saturated with CO but free of
dissolved CO2, in order to assess whether an increase in the surface coverage of adsorbed CO relative to
adsorbed H favors CꢀC coupling. Consistent with the results for CO2 reduction over Cu(100), no change
in methane activity with cation size was observed from CO reduction at −1 V vs RHE. Figure 8b clearly
shows that the rates of C2H4 and C2H5OH formation increase with increasing cation size. We note that the
magnitude of this effect is comparable to that observed during the reduction of CO2, which shows that the
effect of cation size on the formation of multicarbon products cannot be ascribed solely to more facile
CO2 activation.
Discussion
Cation promoter effects
The data presented in this study demonstrate that the size of the alkali metal cations in the
electrolyte affects the intrinsic rates of formation of certain reaction products. On Cu (100) and Cu (111),
these effects are most notable for HCOOꢀ, C2H4, and C2H5OH; over polycrystalline Ag and Sn, the effects
of cation size are observed for the formation of CO and HCOOꢀ. It is also shown that larger cations have a
stronger effect than smaller cations, and that only a small fraction of the larger cation is needed in the
electrolyte to observe its influence. Trends in the partial current densities derived from the studies of the
reduction of key reaction intermediates (OHCCHO and CO) suggest that the elementary steps promoted
by the presence of metal cations are the activation of CO2 and the initial CꢀC coupling, whether through
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 28
1
2
3
4
5
6
CO dimerization or COꢀCHO coupling, and not the hydrogenation of C2 intermediates leading to ethylene
and ethanol.
7
We propose that the energetics for the formation of HCOOꢀ and C2 products are influenced by the
presence of hydrated alkali ions located at the edge of the Helmholtz plane, and that the observed effects
of cation size are attributable to differences in the cation concentration in the outer Helmholtz plane. This
explanation is supported by DFT calculations performed using minimum energy structures of solvated
cations at the interface. Since both Cu(100) and Cu(111) surfaces show consistent trends in activity with
respect to alkali cation size, the closeꢀpacked Cu(111) surface was chosen for the DFT simulations, which
has also been frequently used in previous theoretical CO2R studies42ꢀ45 and where the water structure is
more wellꢀdefined.46
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Because the potentials applied during CO2 reduction are much more negative than the potential of
zero charge (PZC) of the lowꢀindex facets of Cu ~ −0.7VSHE 47, solvated cations should accumulate near
the surface of the electrode during reaction.40, 48 Given the very negative reduction potentials of alkali
ions, they do not chemisorb under CO2 reduction conditions. There is also evidence from studies of
oxygen reduction that the cations remain partially solvated at the interface.49 The presence of the cation at
the metal surface gives rise to high electric fields of ~ −1 V/Å in the vicinity of the ion.25, 50 In contrast to
classical continuum models of the electrochemical interface, the fields are highly localized when the
structure of the cation and solvating waters is modeled explicitly. Refs. 25, 50 show that such fields extend
roughly 5 Å from the center of the solvated cations and outside of this region decay to zero.
The strength of the adsorbateꢀfield interaction can be probed qualitatively by the application of a
uniform field in vacuum. This approach circumvents the challenge of finding a global minimum in the
water and solvated cation structure in the presence of various adsorbates on the surface, which introduces
many more degrees of freedom than standard atomistic computations in heterogeneous catalysis. It has
also been shown previously that the effect of alkali promoters in heterogeneous catalysis can be modeled
18
ACS Paragon Plus Environment

Page 19 of 28
Journal of the American Chemical Society
1
2
3
4
5
approximately by the effect of a uniform field.51 The interaction energy between an adsorbate and a
uniform electric field at the interface is given by52
6
7
8
9
ꢀꢁ = ꢂꢃ − ꢄꢅ ꢆꢃꢅ + ⋯
,
(1)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
where
ꢀ and α are the dipole moment and the polarizability (respectively) of the adsorbate, and ε
represents the electric field. Adsorbate dipoles oriented in the opposite direction of the field will be
stabilized, and vice versa.
Figure 9 shows the field stabilization for various adsorbates that have been proposed to be
involved in the electrochemical reduction of CO2 on Cu(111).43, 49, 52 The change in adsorption free
energy for each species was determined by applying a uniform field oriented perpendicular to the surface
in vacuum.43 The corresponding values of
ꢀ
and
α, determined by fitting Eq. (1) the calculated data, are
given in Table 1. We note that H* does not have a significant value of
ꢀ
and is not affected by the electric
field, which is consistent with previous experimental studies that have shown that hydrogen adsorption is
unaffected by the identity of the cation in solution.14ꢀ16 Since the energy for hydrogen adsorption is a
descriptor for the hydrogen evolution activity on metals, the rate for hydrogen production should also be
relatively insensitive to the identity of the cation, consistent with the experimental data presented here.53
Recent calculations of the electrochemical barriers for CO2 reduction using an explicit solvent model have
shown that reduction of *CO to form *CHO is the limiting step for CH4 formation.43 Since the value of
ꢀ
for *CHO is zero, the cationꢀinduced field would stabilize *CO and therefore increase the energy
barrier for hydrogenation to *HCO, the rateꢀlimiting step on the path to CH4 formation. Therefore, CH4
formation should occur in regions of the catalyst that are not strongly influenced by the cations located at
the outer Helmholtz plane, and correspondingly the partial current density for CH4 should not be strongly
affected by cation identity, which is consistent with the trend in the experimental data shown above.
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9: Field effects on various CO2R intermediates on Cu(111). The energy of each adsorbate is
plotted as a function of field strength, which is obtained by applying a uniform electric field oriented
perpendicular to the slab. The curves are fit to Equation (1) and the corresponding dipole moment
polarizability of the adsorbates are given in Table 1. Solid curves highlight the adsorbates that are most
strongly affected by an electric field. Adsorbates with positive dipole moments oriented in the direction
(aligned with the electric
ꢀ and
α
ꢀ
(
opposite to the electric field) are stabilized, while adsorbates with negative
field) are destabilized.
ꢀ
Table 1 Dipole moments and polarizabilities of adsorbates on Cu(111), determined with Eq. (1)
Adsorbate
*CO2
ꢀ
(eÅ)
α
(eÅ2/V)
0.40
0.76
*COOH
*OCHO
−
0.19
0.60
0.42
0.04
*CO
0.23
0.00
0.66
0.75
0.22
0.36
0.54
0.74
0.52
*CHO
*OCCO
*OCCHO
*CHOH
−
0.05
*CH
*CH2
*CH3
*OH
*H
−
−
−
−
0.02
0.08
0.03
0.26
0.18
0.26
0.44
0.14
0.00
0.00
20
ACS Paragon Plus Environment

Page 21 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
For adsorbates containing C=O bonds oriented perpendicular to the surface—in particular, *CO2,
*OCCO, and *OCCHO—the field stabilization should be large, ~ 1 eV for fields of ~ −1 V/Å in the
vicinity of the cations. As shown in Table 1, these species have the largest dipole moments. Previous
theoretical calculations have shown that on weakly binding metals such as Ag, the fieldꢀinduced
stabilization of *CO2 is crucial for the activation of CO2 to form CO.25 The barrier for the hydrogenation
of *CO2 to *COOH is generally very small,54 so stabilization of *CO2 should increase the coverage of
*COOH. The *COOH species can undergo further hydrogenation to form HCOOH or *CO. The
chemical hydrogenation barriers for *COOH to form HCOOH are surmountable at 0.69 eV and 0.75 eV
on Cu(111) and Cu(211) respectively, and 0.46eV on Ag(211),23 consistent with increased HCOOH
production observed on all of the surfaces investigated. For weakly binding metals such as Ag and Sn,
the field effect would promote CO(g) production, whereas for moderate to strongly binding metals such as
Cu and Pt, *CO produced in the vicinity of a cation would also be field stabilized, thereby decreasing its
desorption rate and increasing its *CO coverage there.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
On Cu surfaces, formation of the initial intermediates to C2 products are also fieldꢀassisted. The
greater stabilization of the *OCCO relative to 2*CO in the presence of a cation is due to the differences in
the dipole moments of the product and reactant species,36 which has the effect of lowering the barrier to
CꢀC bond formation. In the absence of the cationꢀcreated electrostatic field and effects of solvation by
water, the barrier for this process is not surmountable at room temperature.55 At higher overpotentials,
where *CHO can be produced, *COꢀ*CHO coupling would also be a feasible route towards C2 products,
and the corresponding barrier should be similarly fieldꢀstabilized due to the large dipole on *OCCHO, as
shown in Figure 9.40 Since the formation of glyoxal via addition of a hydrogen atom to *OCCHO is the
first step along the path to forming ethylene and ethanol, the observation (see Figure 8) that the selectivity
of products derived from glyoxal is independent of cation size suggests that the rateꢀdetermining step in
formation of C2 products occurs at the point of CꢀC bond formation and prior to formation of glyoxal.54
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 28
1
2
3
4
5
6
The net effect of increased CO production, CO coverage and CꢀC coupling on Cu on fieldꢀstabilized sites
results in a net nonꢀmonotonic trend in the partial current density for CO with cation size.
7
8
9
While the influence of the cationꢀinduced field at the catalyst surface can explain the role of
cations in promoting elementary steps in the reduction of CO2 that involve strong dipole moments, it does
not explain the observed trend with cation size. We therefore explored the possibilities that this effect is
due to disparities in the magnitude of the electrostatic field induced by cations of different size and/or an
increase in the concentration of solvated cations at the outer Helmholtz plane as a function of size.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Constrained minima hopping26 (CMH) was used to obtain the optimized structure of water
surrounding each solvated ion on Cu(111). The structures of the solvated ions are shown in Figure S5
and the distances of the solvated cation from the Cu(111) surface are given in Table S3. Figure 10 shows
that the calculated fields in the vicinity of adsorbed CO (*CO) do not show a systematic variation with
cation size. Moreover, any differences in the binding energy of intermediates induced by cations would
give rise to exponential differences in current density, which we do not observe.
Figure 10. Electric field distribution near the center of the adsorbate plotted as a function of the zꢀ
coordinate of the simulation cell for the 2 *CO initial state. The geometry of an example structure is
overlaid to clearly illustrate the position of the adsorbate. The cell is viewed from the side and rotated
22
ACS Paragon Plus Environment

Page 23 of 28
Journal of the American Chemical Society
1
2
3
4
5
90°. The minimum z value is taken at the Cu surface and the zꢀposition of the solvated cations is roughly
at 20 Å and not included in the overlay.
6
7
8
9
We hypothesize, therefore, that the ion specificity arises from an increase in the concentration of
cations at the outer Helmholtz plane with increasing cation size. To determine the driving force for each
cation toward the Helmholtz plane from bulk solution, we used these globally optimized solvation
structures for each cation (Table S3) and calculated the energy required to move a solvated cation from
the bulk of the electrolyte to a fixed distance of 6.25 Å above the Cu(111) surface the (see the SI for
details). Figure 11 shows this energy change with the cathode potential relative to that for Li+, ∆∆E. The
potential U vs SHE is determined through:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢇ = ꢈꢉ ꢈ
(2)
ꢊꢋꢌ
ꢍ
where ꢎ is the work function of the interface and ꢎꢏꢐꢑ is that corresponding to the standard hydrogen
electrode, determined experimentally to be ~4.4eV.56 The value of ∆∆E becomes more negative with
increasing cation size, its magnitude increasing in the order Li+ < Na+ < K+ < Rb+ < Cs+. Therefore, the
driving force for each cation to be at the Helmholtz plane relative to the bulk electrolyte increases with
cation size. This trend suggests that the concentration of cations at the outer Helmholtz plane should
increase with increasing cation size, leading to increased product current densities. Consistent with this
reasoning, we observed (see Figure 6) that at a constant cation concentration, replacement of even a small
fraction of Li+ by Cs+ results in a shift in the product distribution from one characteristic for Li+ to one
characteristic for Cs+. This supports the hypothesis that there is a stronger driving force for Cs+ to be at
the Helmholtz plane compared to Li+. Without this stronger driving force on Cs+ to be in the Helmholtz
plane, the trends in Figure 6 would be linear.
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 11: The change in energy for bringing a solvated cation from bulk electrolyte to the outer
Helmholtz plane at the Cu (111) facet for different cations shown as a function of the standard
hydrogen electrode (SHE) potential. All energies are referenced to that for the Li+ cation. The
dependence of these energies on cathode potential results from the partial electron transfer to the ion upon
migration from the bulk to the Helmholtz plane, and is reflected in Eq. 5. This energy becomes more
negative with increasing ion size, suggesting a greater concentration of larger cations in the outer
Helmholtz plane and giving rise to increased product current densities. The vertical dotted line indicates
the potential for the standard hydrogen electrode.
Conclusions
The present study investigated the effects of electrolyte cation size on the activity and selectivity
of metal catalysts used for the electrochemical reduction of CO2. The experimental work was conducted
under conditions where the influence of electrolyte concentration polarization was negligible in order to
avoid the additional influence of changes in pH and the consequent effects on the concentration of CO2
near the cathode surface. For polycrystalline Ag and Sn, increasing the size of the alkali metal cation in
the electrolyte increases the partial current densities for HCOOꢀ and CO, but has no effect on the partial
current density for H2. Similar experiments conducted using Cu(100) and Cu(111) oriented films
24
ACS Paragon Plus Environment

Page 25 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
deposited on crystalline Si(100) and(111) surfaces show an increase in the current densities for HCOOꢀ,
C2H4, and C2H5OH but little to no effect on the partial current densities for H2, CO, and CH4 as the size of
the electrolyte cation is increased.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The observed effects are interpreted on the basis of DFT calculations. Hydrated alkali metal
cations in the outer Helmholtz plane create a dipole field of the order 1 V/Å, which can stabilize the
adsorption of surface intermediates having significant dipole moments (e.g., *CO2, *CO, *OCCO). This
field stabilization decreases the energy for the reduction of *CO2 adsorption, the precursor to twoꢀelectron
products, and CꢀC coupling to form *OCCO or *OCCHO, the precursor to C2H4 and C2H5OH.
Calculations of the stability of the solvated cation in the Helmholtz plane indicate that larger hydrated
cations are more energetically favored at the outer Helmholtz plane than smaller ones, which suggest a
larger coverage of cations as cation size increases. This result is consistent with the experimental
observation that a small percentage of larger cations has a large effect on product distribution.
These findings are consistent with the trends in partial current for C1 products observed on
polycrystalline Ag and Sn and on Cu(100) and Cu(111) surfaces, and for C2 products observed on
Cu(100) and Cu(111) surfaces. Also consistent with the conclusions drawn from the theoretical analysis is
that cation size affects the partial currents for C2H4 and C2H5OH produced during the electrochemical
reduction of CO over Cu(100) but has no influence on formation of these products upon reduction of
glyoxal. The absence of cation size effect on the partial current for H2 produced on Ag, Sn, Cu(100), and
Cu(111) is attributed to the absence of a dipole for *H. Similarly, no effect of cation size on the formation
of CH4 on Cu(100) and Cu(111) surfaces is expected since the field of a cation would tend to stabilize
*CO more than *CHO, thereby increasing the energy for *CO hydrogenation to *CHO, the rateꢀlimiting
step leading to CH4.
Finally, we note that differences in the effects of electrolyte cation size on the formation of CO on
Ag and Sn vs. Cu(100) and Cu(111) are attributable to the following considerations. For Ag and Sn, CO
is a final product, and therefore only influenced by its formation by hydrogenation of *COOH which
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 28
1
2
3
4
5
6
7
8
9
should have an increased coverage from the fieldꢀstabilization of *CO2. In contrast, for Cu(100) and
Cu(111), *CO is an intermediate in the formation of CH4. Since the dipole electrostatic field created by
hydrated cations accelerates both the formation and consumption of *CO, there is no observable effect of
cation size on the partial current for CO on Cu(100) and Cu(111).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgements
This material is based upon work performed by the Joint Center for Artificial
Photosynthesis, a DOE Energy Innovation Hub, supported through the Office of Science of the
U.S. Department of Energy under Award Number DEꢀSC0004993. J.R. gratefully acknowledges
support of the National Science Foundation Graduate Research Fellowship Program (NSF
GRFP) under Grant No. DGEꢀ0802270. The authors would also like to acknowledge the
allocation of computer time at the National Energy Research Scientific Computing Center, a
DOE Office of Science User Facility supported by the Office of Science of the U.S. Department
of Energy under Contract No. DEꢀAC02ꢀ05CH11231.
References
1.
2.
3.
Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. 2006, 103, 15729ꢀ15735.
Chu, S.; Majumdar, A. Nature 2012, 488, 294ꢀ303.
Hori, Y., Electrochemical CO2 Reduction on Metal Electrodes. In Modern Aspects of
Electrochemistry, Vayenas, C. G.; White, R. E.; GamboaꢀAldeco, M. E., Eds. Springer New York: New
York, NY, 2008; 89ꢀ189.
4.
Kuhl, K. P.; Hatsukade, T.; Cave, E. R.; Abram, D. N.; Kibsgaard, J.; Jaramillo, T. F. J. Am.
Chem. Soc. 2014, 136, 14107ꢀ14113.
5.
6.
7.
8.
Hori, Y.; Wakebe, H.; Tsukamoto, T.; Koga, O. Electrochim. Acta 1994, 39, 1833ꢀ1839.
Hori, Y.; Murata, A.; Takahashi, R. J. Chem. Soc. Faraday Trans. 1989, 85, 2309ꢀ2326.
Kuhl, K. P.; Cave, E. R.; Abram, D. N.; Jaramillo, T. F. Energy Environ. Sci. 2012, 5, 7050ꢀ7059.
Frumkin, A. N. Trans. Faraday Soc. 1959, 55, 156ꢀ167.
9.
Akira, M.; Yoshio, H. Bull. Chem. Soc. Jpn 1991, 64, 123ꢀ127.
10.
11.
12.
Thorson, M. R.; Siil, K. I.; Kenis, P. J. A. J. Electrochem. Soc. 2013, 160, 69ꢀ74.
Paik, W.; Andersen, T. N.; Eyring, H. Electrochim. Acta 1969, 14, 1217ꢀ1232.
Kaneco, S.; Iiba, K.; Katsumata, H.; Suzuki, T.; Ohta, K. J. Solid State Electrochem. 2007, 11,
490ꢀ495.
26
ACS Paragon Plus Environment

Page 27 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
13.
13012.
14.
Singh, M. R.; Kwon, Y.; Lum, Y.; Ager, J. W.; Bell, A. T. J. Am. Chem. Soc. 2016, 138, 13006ꢀ
Subbaraman, R.; Tripkovic, D.; Strmcnik, D.; Chang, K.ꢀC.; Uchimura, M.; Paulikas, A. P.;
Stamenkovic, V.; Markovic, N. M. Science 2011, 334, 1256ꢀ1260.
15. Stoffelsma, C.; Rodriguez, P.; Garcia, G.; GarciaꢀAraez, N.; Strmcnik, D.; Marković, N. M.;
Koper, M. T. M. J. Am. Chem. Soc. 2010, 132, 16127ꢀ16133.
16. Strmcnik, D.; Kodama K.; van der Vliet, D.; Greeley, J.; Stamenkovic, V. R.; Marković, N. M.
Nat. Chem. 2009, 1, 466ꢀ472.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
17.
18.
10ꢀ13.
19.
Suntivich, J.; Perry, E. E.; Gasteiger, H. A.; ShaoꢀHorn, Y. Electrocatalysis 2013, 4, 49ꢀ55.
Angelucci, C. A.; Varela, H.; TremiliosiꢀFilho, G.; Gomes, J. F. Electrochem. Comm. 2013, 33,
Lobaccaro, P.; Singh, M. R.; Clark, E. L.; Kwon, Y.; Bell, A. T.; Ager, J. W. Phys. Chem. Chem.
Phys. 2016, 18, 26777ꢀ26785.
20.
Giannozzi, P.; Baroni, S.; Bonini, N.; Calandra, M.; Car, R.; Cavazzoni, C.; Ceresoli, D.;
Chiarotti, G.; Cococcioni, M.; Dabo, I.; Dal Corso, A.; de Gironcoli, S.; Fabris, S.; Fratesi, G.; Gebauer,
R.; Gerstmann, U.; Gougoussis, C.; Kokalj, A.; Lazzeri, M.; MartinꢀSamos, L.; Marzari, N.; Mauri, F.;
Mazzarello, R.; Paolini, S.; Pasquarello, A.; Paulatto, L.; Sbraccia, C.; Scandolo, S.; Sclauzero, G.;
Seitsonen, A.; Smogunov, A.; Umari, P.; Wentzcovitch, R. J. Phys. Condens. Mat. 2009, 21.
21.
22.
Bahn, S. R.; Jacobsen, K. W. Comput. Sci. Eng. 2002, 4, 56ꢀ66.
Wellendorff, J.; Lundgaard, K. T.; Møgelhøj, A.; Petzold, V.; Landis, D. D.; Nørskov, J. K.;
Bligaard, T.; Jacobsen, K. W. Phys. Rev. B 2012, 85, 235149.
23.
24.
25.
26.
27.
Yoo, J. S.; AbildꢀPedersen, F.; Nørskov, J. K.; Studt, F. ACS Catal. 2014, 4, 1226ꢀ1233.
Monkhorst, H.; Pack, J. Phys. Rev. B 1976, 13, 5188ꢀ5192.
Chen, L. D.; Urushihara, M.; Chan, K.; Nørskov, J. ACS Catal. 2016, 6, 7133ꢀ7139.
Peterson, A. A. Top. Catal. 2013, 57, 40ꢀ53.
Hahn, C.; Hatsukade, T.; Kim, Y.ꢀG.; Vailionis, A.; Baricuatro, J. H.; Higgins, D. C.; Nitopi, S.
A.; Soriaga, M. P.; Jaramillo, T. F. Proc. Natl. Acad. Sci. 2017, 114, 5918ꢀ5923.
28.
3383.
29.
30.
9867.
31.
32.
15053ꢀ15056.
33.
Jiang, H.; Klemmer, T. J.; Barnard, J. A.; Payzant, E. A. J. Vac. Sci. Tech. A 1998, 16, 3376ꢀ
Hori, Y.; Takahashi, I.; Koga, O.; Hoshi, N. J. Phys. Chem. B 2002, 106, 15ꢀ17.
Schouten, K. J. P.; Qin, Z.; Gallent, E. P.; Koper, M. T. M. J. Am. Chem. Soc. 2012, 134, 9864ꢀ
Roberts, F. S.; Kuhl, K. P.; Nilsson, A. ChemCatChem 2016, 8, 1119ꢀ1124.
Kim, Y.ꢀG.; Baricuatro, J. H.; Javier, A.; Gregoire, J. M.; Soriaga, M. P. Langmuir 2014, 30,
Durand, W. J.; Peterson, A. A.; Studt, F.; AbildꢀPedersen, F.; Nørskov, J. K. Surf. Sci. 2011, 605,
1354ꢀ1359.
34.
35.
36.
37.
38.
Schouten, K. J. P.; Pérez Gallent, E.; Koper, M. T. M. ACS Catal. 2013, 3, 1292ꢀ1295.
Montoya, J. H.; Shi, C.; Chan, K.; Nørskov, J. K. J. Phys. Chem. Lett. 2015, 6, 2032ꢀ2037.
Sandberg, R. B.; Montoya, J. H.; Chan, K.; Nørskov, J. K. Surf. Sci. 2016, 654, 56ꢀ62.
Hori, Y.; Murata, A.; Yoshinami, Y. J. Chem. Soc. Faraday Trans. 1991, 87, 125ꢀ128.
Hatsukade, T.; Kuhl, K. P.; Cave, E. R.; Abram, D. N.; Jaramillo, T. F. Phys. Chem. Chem. Phys.
2014, 16, 13814ꢀ13819.
39.
40.
41.
Kyriacou, G. Z.; Anagnostopoulos, A. K. J. Appl. Electrochem. 1993, 23, 483ꢀ486.
Goodpaster, J. D.; Bell, A. T.; HeadꢀGordon, M. J. Phys. Chem. Lett. 2016, 7, 1471ꢀ1477.
Schouten, K. J. P.; Kwon, Y.; van der Ham, C. J. M.; Qin, Z.; Koper, M. T. M. Chem. Sci. 2011,
2, 1902ꢀ1909.
42.
43.
15444.
Akhade, S. A.; McCrum, I. T.; Janik, M. J. J. Electrochem. Soc. 2016, 163, 477ꢀ484.
Liu, X.; Xiao, J.; Peng, H.; Hong, X.; Chan, K.; Nørskov, J. K. Nat. Comm. 2017, 8, 15438ꢀ
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 28
1
2
3
4
5
6
7
8
9
44.
4727.
45.
Shi, C.; Hansen, H. A.; Lausche, A. C.; Nørskov, J. K. Phys. Chem. Chem. Phys. 2014, 16, 4720ꢀ
Xiao, H.; Cheng, T.; Goddard, W. A.; Sundararaman, R. J. Am. Chem. Soc. 2016, 138, 483ꢀ486.
Ogasawara, H.; Brena, B.; Nordlund, D.; Nyberg, M.; Pelmenschikov, A.; Pettersson, L. G. M.;
46.
Nilsson, A. Phys. Rev. Lett. 2002, 89, 276102.
47.
48.
49.
Lukomska, A.; Sobkowski, J. J. Electroanal. Chem. 2004, 567, 95ꢀ102.
Cheng, T.; Xiao, H.; Goddard, W. A. J. Am. Chem. Soc. 2016, 138, 13802ꢀ13805.
Strmcnik, D.; van der Vliet, D. F.; Chang, K. C.; Komanicky, V.; Kodama, K.; You, H.;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Stamenkovic, V. R.; Marković, N. M. J. Phys. Chem. Lett. 2011, 2, 2733ꢀ2736.
50.
51.
52.
Montoya, J. H.; Shi, C.; Chan, K.; Norskov, J. K. J. Phys. Chem. Lett. 2015, 6, 2032ꢀ2037.
Mortensen, J. J.; Hammer, B.; Nørskov, J. K. Phys. Rev. Lett. 1998, 80, 4333ꢀ4336.
Nørskov, J. K. N.; Studt, F.; AbildꢀPedersen, F.; Bligaard, T., Fundamental Concepts in
Heterogeneous Catalysis. Wiley, Hoboken, New Jersey: 2014.
53. Nørskov, J. K.; Bligaard, T.; Logadottir, A.; Kitchin, J. R.; Chen, J. G.; Pandelov, S.; Stimming,
U. J. Electrochem. Soc. 2005, 152, 23ꢀ26.
54. Shi, C.; O'Grady, C. P.; Peterson, A. A.; Hansen, H. A.; Nørskov, J. K. Phys. Chem. Chem. Phys.
2013, 15, 7114.
55.
56.
Montoya, J. H.; Peterson, A. A.; Nørskov, J. K. ChemCatChem 2013, 5, 737ꢀ742.
Trasatti, S. Pure Appl. Chem. 1986, 58.
Table of contents figure
28
ACS Paragon Plus Environment