Energy &

Environmental
Science
View Article Online
PAPER View Journal

Understanding cation effects in electrochemical

Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

CO2 reduction†
Cite this: DOI: 10.1039/c9ee01341e
Stefan Ringe, ‡*ab Ezra L. Clark, ‡cd Joaquin Resasco,e Amber Walton,c
d
Brian Seger, Alexis T. Bell c and Karen Chan *f

Solid–liquid interface engineering has recently emerged as a promising technique to optimize the
activity and product selectivity of the electrochemical reduction of CO2. In particular, the cation identity
and the interfacial electric field have been shown to have a particularly significant impact on the activity
of desired products. Using a combination of theoretical and experimental investigations, we show the
cation size and its resultant impact on the interfacial electric field to be the critical factor behind the
ion specificity of electrochemical CO2 reduction. We present a multi-scale modeling approach that
combines size-modified Poisson–Boltzmann theory with ab initio simulations of field effects on critical
reaction intermediates. The model shows an unprecedented quantitative agreement with experimental
trends in cation effects on CO production on Ag, C2 production on Cu, CO vibrational signatures on Pt
Received 26th April 2019, and Cu as well as Au(111) single crystal experimental double layer capacitances. The insights obtained
Accepted 3rd July 2019 represent quantitative evidence for the impact of cations on the interfacial electric field. Finally, we
DOI: 10.1039/c9ee01341e present design principles to increase the activity and selectivity of any field-sensitive electrochemical
process based on the surface charging properties: the potential of zero charge, the ion size, and the
rsc.li/ees double layer capacitance.

Broader context
The electrochemical reduction of CO2 has the potential to reduce greenhouse gas emissions while producing valuable fuels and industrially relevant chemicals.
A central challenge for commercialization is the need for increased activity and selectivity of catalysts towards desired products. In recent years, electrolyte
optimization has emerged as a new strategy for electrocatalyst design. In particular, the cation identity has been shown to have a significant effect on CO2
reduction (CO2R) towards valuable products. In this joint experimental-theoretical work, we develop a multi-scale, continuum/ab initio modeling approach that
shows the cation specificity to arise from differences in cation–cation repulsion and corresponding differences in the interfacial field. This model shows an
unprecedented quantitative agreement with a wide range of experimental observations. The findings also show the hitherto neglected interfacial charging
properties in determining the electrocatalytic activity for CO2R and beyond and pave the way for electrolyte- and interface- engineering for general field-
sensitive electrochemical processes.

a
SUNCAT Center for Interface Science and Catalysis, Department of Chemical Introduction
Engineering, Stanford University, Stanford, California 94305, USA.
E-mail: sringe@stanford.edu Solid–liquid interface engineering has emerged in recent years
b
SUNCAT Center for Interface Science and Catalysis, SLAC National Accelerator as a promising technique to optimize the reactivity and selec-
Laboratory, Menlo Park, California, 94025, USA tivity of electrochemical reactions. In recent years, a number of
c
Joint Center for Artificial Photosynthesis, Lawrence Berkeley National Laboratory,
engineering approaches have been explored, going beyond the
Berkeley, CA 94720, USA
d
Surface Physics & Catalysis (SurfCat), Department of Physics Technical University
design of catalyst’s electronic structure towards a full optimiza-
of Denmark, Denmark tion of the reaction environment.1 For instance, researchers
e
Department of Chemical Engineering, University of California, Santa Barbara, have investigated the effect of interfacial electric field,2–6 pH7–11
California 93117, USA mass transport,12–16 catalyst nano-structuring,15,17–25 and the
f
CatTheory Center, Department of Physics, Technical University of Denmark,
electrolyte composition, such as the solvent,26–28 the buffer,29–32
Kongens Lyngby 2800, Denmark. E-mail: kchan@fysik.dtu.dk
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
or the cations.4,33–47 In particular, the identity of cations in the
c9ee01341e electrolyte has been shown to drastically affect the catalytic
‡ Contributed equally to this work. conversion rate in a number of critical electrochemical processes,

This journal is © The Royal Society of Chemistry 2019 Energy Environ. Sci.
 View Article Online

Paper Energy & Environmental Science

such as the oxidation of water,34,48 CO,35 formate,36 chloride,37 Model formulation

hydrogen or alcohols,38,47 and in the reduction of oxygen38,39,48 Modeling cation effects – the ai1c approach
and CO2 (CO2R).4,40–45 CO2R has been shown to be particularly
sensitive to cation identity, with 1–3 orders of magnitude differ- Fig. 1 outlines our combined ab initio/continuum approach to
ence in activity between Li+ and Cs+ containing electrolytes. This modeling ion specific electrocatalytic activity. This approach is
effect has been demonstrated on Ag for the conversion to CO and motivated by the previously noted strong double layer electric
on Cu towards the formation of hydrocarbons and alcohols field-dependence of critical reaction intermediates in CO2R.3–5,57
containing multiple carbon atoms (‘‘C2+ products’’).4,44 The interfacial electric field is generated by the electrode surface
charge density s that is present at applied electrode potentials
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

The present joint theoretical-experimental study focusses on

the mechanism behind cation specificity in the electrochemical different from the potential of zero charge. As illustrated in the
reduction of CO2R, as a case study for general ion-specific electro- right panel of Fig. 1, we determine the surface-charge density
chemical processes. The origin of cation effects on CO2R has been dependence of the energetics of critical reaction intermediates via
contentious in recent literature. Hydrated cations were suggested to charged implicit solvent corrected DFT calculations using a planar
act as proton donors49 and to modify the local electrode pH countercharge.58 These energies are related to reaction rates
through shifts in local potential at the outer Helmholtz plane41,44 through a simple rate-limiting step approximation. As later
or through acting as a buffer close to the electrode.45 Cation- described in more detail, we find the changes in the reaction
adsorbate interactions have been further suggested to occur chemical potential Dm(s) to be largely independent of the counter
via non-covalent chemical interactions.38,39,48 It has also been charge distribution. This important observation suggests that
suggested that alkali cation effects in CO2R arise from chemi- cations affect catalytic activity by a change of the surface charge
sorbed ions.50,51 However, we note that alkali ion adsorption density at a fixed electrode potential.
including charge transfer is unlikely due to their very negative In order to describe the variation of surface charge density s
reduction potentials (E3 V vs. Standard Hydrogen Electrode as a function of cation type and applied electrode potential, we
(SHE)52).33,37,38 apply a continuum electrolyte model, the size-modified Poisson–
The plethora of hypotheses for cation effects in CO2R high- Boltzmann (MPB) approach (cf. left panel in Fig. 1).56 As
lights a pressing need for microscopic insight using computa- illustrated in Fig. 2, the model predicts that cations with a
tional simulations. Ions at the electrochemical interface are smaller hydration shell like Cs+ are more concentrated at the
particularly challenging to treat from an ab initio perspective, as electrode compared to larger cations like Li+. This difference in
their slow motion compared to water and large solvation shells
introduce many degrees of freedom and associated sampling
challenges. On the other hand, continuum models of the
electric double layer have a long history,53–55 and have yet to
be exploited in recent theoretical investigations of ion effects.
Recent studies from our groups demonstrated that cations
affect the CO2R activity through their electrostatic interactions
with the electric dipole of specific adsorbates.3,4 Here, we build
upon these studies, taking a mean-field electrostatic approach
and modeling the electrolyte distribution with a modified
Poisson–Boltzmann model.56 We find that the surface charge
density and electric field is reduced by ion-size dependent
hydrated cation repulsions at the outer Helmholtz plane. We
perform surface charge-dependent density functional theory
calculations of reaction intermediates in order to relate the
ion-specific surface charge differences to differences in electro-
catalytic activity. The resulting multi-scale approach is evalu-
ated with a wide range of ion-specific experimental data:
CO production on various epitaxial surface facets of Ag, C2
formation on epitaxial Cu, the vibrational stretching mode of
Fig. 1 Schematic illustration of our multi-scale modeling approach to
*CO on polycrystalline Pt and Cu as well as impedance data on
model cation effects on field-driven electrocatalysis. The process of sur-
single crystal Au(111) electrodes. Using a single set of experi- face charge generation as a function of potential (left panel) is simulated by
mentally motivated cation size parameters, we obtain unprece- a 1D-continuum electrostatic description of the electrolyte. The ion-size
dented quantitative agreement with all experiments, which modified Poisson–Boltzmann approach (MPB) enables us to model the
suggests the validity of the simple picture presented. Finally, effect of ion size on the generated surface charge at a fixed potential.
Surface charge density dependent reaction energetics are obtained from
we present cation and system design principles based on the
charge-dependent DFT calculations of the rate-limiting species (right
optimization of the surface charge density and electric field, panel). Combining the results via interpolation, we obtain the catalytic
which we envision to have general applicability to electrocata- activity or current density as a function of cation size and potential of zero
lytic processes beyond CO2R. charge at fixed applied potential.

Energy Environ. Sci. This journal is © The Royal Society of Chemistry 2019
 View Article Online

Energy & Environmental Science Paper

In what follows, we lay out the details of the model.

Continuum surface charging model

The electrostatic potential f in a solution is described by the
one-dimensional Poisson equation (given in SI units)

d2 f X
eb ¼ zi ci ½f; (1)
dx2 i
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

where eb is the bulk dielectric permittivity (in units of the

vacuum permittivity) and zi and ci refer to electrolyte species
changes and concentrations. We apply the finite ion-size
modified version of the Poisson–Boltzman model, which for a
z:z electrolyte is:56
zi Ff
Fig. 2 Illustration of the origin of cation effects in field-driven electro-
e RT
catalysis as suggested by this work. Repulsive interactions between c ½f ¼ cb  ; (2)
hydrated cations at the outer Helmholtz plane reduce the local concen- zi Ff
1  w0 þ w0  cosh
tration of cations, the surface charge density s (depicted by the red- RT
colored region) and the electric double layer field. The diffuse layer that is
explicitly modeled by the MPB model is depicted as well as the Helmholtz with the Faraday and ideal gas constants F and R, the tempera-
gap capacitance region and the interfacial ion diameter determined in this ture T, the ionic charges zi, the ionic bulk concentration cb and
work. the ion-occupied volume fraction w0 = 2cba3NA, with the Avogadro
constant NA and the lattice cell length a. This expression arises
from a statistical lattice model in which each anion and cation is
concentration increases the surface charge density and the field only allowed to occupy a single cell (cf. Fig. 1). This restriction
at a given potential relative to the potential of zero charge (PZC) effectively leads to a maximum possible ion concentration deter-
fM  fM,PZC. The continuum model provides us with the mined by w0. a denotes a lattice cell length in the statistical model
dependence of the surface charge density on potential and is equivalent to an effective ion diameter if ions are considered as
ion size a, s(fM  fM,PZC,a). Coupling the Dm(s) determined hard particles. In this picture, the distance of the cation from the
through DFT with s(fM  fM,PZC,a) we obtain the dependence electrode will directly correspond to a/2 (cf. Fig. 2) which has been
of catalytic activity on ion size. determined experimentally by X-ray scattering and diffraction
The present model differs from previous work in several methods39,77–81
significant ways. The primary hypothesis here is that cation Interestingly, a comparison of experimentally tabulated
effects in CO2R are mean-field in nature, in contrast to previous electrode-cation distances suggests these to be only slightly
studies that consider cation effects to arise from explicit inter- dependent on the applied potential and metal electrode. In the
actions of cations with key reaction intermediates.3,4,46,51 case of Cs+, a distance of around 3.5 Å has been determined
Furthermore, we model the impact of ion size on the double at both the Au(111)81 and Pt(111)79 electrodes. K+, which
layer structure independently from the ab initio simulations of is a harder cation with a larger hydration shell in the bulk
the reaction energetics. electrolyte,82 was found to be located at larger distances of 4.1 Å
There are several advantages to this modelling approach. to the Ag(111) electrode.77 These results suggest that ion sizes
Since the reaction energetics are determined independently of are relatively invariant on different metal surfaces. In the
the double layer model, a given Dm(s) from a single ab initio following, we use these two experimentally determined values
simulation can be coupled with a variety of detailed continuum while values for other cations are inter- and extrapolated from
double layer models, and even with experimental capacitance experimental measurements. All cation sizes are listed in Table 2.
data. This flexibility is not possible with constant potential The MPB equation is solved with the following boundary
grand canonical DFT simulations which include complex conditions (cf. also Fig. 1). At the bulk electrolyte boundary
descriptions of the electrolyte directly in the DFT setup6,59–69 (x = xN, 80 mm), we apply the Dirichlet boundary condition f = 0.
which functional form can not be ambiguously chosen and is At the electrode side x = 0, we apply the Robin boundary
difficult to parametrize.65,70 Furthermore, constant potential condition:83,84
grand canonical DFT simulations require a reference to relate
df  
the simulated work function to potential. Typically the reference s ¼ eb  ðx ¼ 0Þ ¼ Cgap fM  fM;PZC  fðx ¼ 0Þ ; (3)
dx
that is applied is the work function of the standard hydrogen
electrode, which has been reported by different experiments and where fM is the metal electrode potential relative to the bulk
modeling approaches to have a range from 4.3 to 5.3 eV.71–76 Here electrolyte and Cgap is an interfacial Helmholtz-like capacitance
our approach allows us to use the PZC of a given metal|solution which acts in series with the diffuse layer capacitance. This
interface as a reference, which can have a higher accuracy than mixed boundary condition allows us to correctly describe
the SHE reference in well-characterized metals. surface charging at potentials far from the PZC, where the

This journal is © The Royal Society of Chemistry 2019 Energy Environ. Sci.
 View Article Online

Paper Energy & Environmental Science

gap capacitance Cgap dominates the surface charging response. The accuracy of this approach is demonstrated in Fig. S2–S9
Physically, Cgap arises mostly from Pauli repulsion of the in the ESI,† where the parabolic function nearly perfectly fits
electrons, which creates a vacuum-like gap between solid and the calculated DFT data points as a function of surface charge
liquid (cf. Fig. 2).85–87 This boundary condition also has the density.
practical advantage of offering a direct link between surface In this work, we investigate cation effects on CO2R at Ag and
charge density and electrode potential, which is determined by Cu as well as CO adsorption on Pt. In the following, we will
the macroscopic properties of the electrode as the PZC and the consider the partial current density normalized to a particular
Helmholtz gap capacitance (cf. Fig. 1). We note that x = 0 cation, which represents the cation effect on the turn over
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

corresponds here approximately to the outer part of the Helm- frequency removing the need to specify an active site density.
holtz gap, as illustrated in Fig. 2. For each of these cases, the current density is expressed in our
Experimentally, the value of the double layer capacitance Cdl model as determined by the formation energy of the rate-
at negative potentials far from the PZC has been found to be limiting species. For CO2 at Ag, recent literature suggested
approximately facet-independent on Ag,88 and is very often CO2 adsorption with concomitant electron transfer to limit
independent of metal identity.89,90 Cdl was measured to be the CO production rate.57,99–102 Spectroscopic studies on
around 20–25 mF cm2 for Ag surfaces,88,91–93 Cu(111)94 and Ag103 and Cu104 have also identified carboxylate intermediates.
Pt(111).95 At potentials far from the PZC, the gap capacitance As shown in Fig. S2 (ESI†), *CO2 is strongly stabilized by electric
dominates, which is why we apply a value of Cgap = 25 mF cm2 double layer fields. This stabilization is also clear from the
throughout the paper. Finally, experimental PZC’s have been field-corrected free energy diagram given in Fig. S10 (ESI†) that
used to parametrize eqn (3) as listed in Table 1 with the also suggests that CO2 adsorption limits the CO production rate
exception of the pc-Ag surface, where we used a theoretical in accordance with the literature. Furthermore, recent theore-
estimate due to the diversity of the experimental data (cf. ESI†). tical results have shown that the transition state of the CO2
Solving the MPB equation then gives the relation s(fM,PZC,a), adsorption process is close to the final state.105 The CO partial
which is shown in Fig. S1 of the ESI.† current density jCO can thus be expressed as a function of the
*CO2 adsorption energy at standard conditions Dm*CO2:
Ab initio derived field dependent electrocatalysis  
Dm CO2 ðsÞ
As mentioned before, dipolar and polarizable intermediates, jCO / exp  : (5)
RT
such as *CO2 or *OCCO, interact significantly with electric
double layer fields.3,4 The dependence can be derived from In the case of CO2 reduction on Cu, the most significant
surface charge-dependent DFT calculations. The electrolyte cation effects were observed for C2 product formation. From
counter charge can be represented by either explicit cations3,96,97 previous studies, the CO–CO coupling step has been discussed
or a continuum representation as implemented into various DFT to limit the corresponding production rate.3,7,30,106–108 Similar
program packages.61–64,67–69 Here, we applied a mean-field to CO2 adsorption, CO–CO coupling is a chemical step which is
formulation using both a planar counter charge (PCC) as well driven by interfacial field stabilization of the dipolar adsorbate
as a linearized PB (LPB) representation. Fig. S2 and S3 of the (*OCCO).3 This dipole-field interaction results in a large surface
ESI† show the dependence of the free energy change Dm for CO2 charge dependence as shown in Fig. S7 (ESI†). We assume a
adsorption on Ag as a function of s. Dm is dependent on neither similar dipole for the coupling transition state as in the final
the electrolyte model (i.e. LPB or PCC) nor the location of *OCCO state, which results in a C2 production rate expression
the counter-charge in the PCC model (cf. Fig. S3, ESI†). This as a function of the *OCCO energy relative to the bare surface,
observation is critical since it shows that, at least in a mean- Dm*OCCO:
field approximation, cations affect reaction kinetics via a  
Dm OCCO ðsÞ
change of the surface charge density. jC2 / exp  : (6)
RT
The Dm(s) function obtained from DFT is nearly parabolic,
which in the case of a field-dependent expression arises from
both the first order dipole interaction and the 2nd order Results & discussion
polarizability.4,98 Therefore we can obtain an analytic expression
by interpolation the surface charge density dependent formation Cation effects on CO2R
energies with the parabolic function: In what follows, we evaluate the model against the experimen-
tally observed ion-specificity of CO2R on two surfaces, Ag and
Dm(s) = Dm(s = 0) + ass + bss2. (4)
Cu. Fig. 3 shows the theoretical (lines) and experimental (dots)
shifts in activity towards CO at 1 V vs. RHE for Ag(111),
Ag(110), and pc-Ag, where the first two surfaces are epitaxial
Table 1 Experimental potential of zero charges (PZC) vs. SHE for the
thin films.4,5 The activity data is normalized to that of Li+
different surfaces that were used in this work.109,110 Only the pc-Ag value
was estimated from theoretical considerations (cf. ESI) (cf. Fig. S19 in the ESI† for full polarization curves).
As seen from the comparison, the model gives essentially
Ag(111) Ag(110) pc-Ag Cu(111) Cu(100) CO@Pt(111) pc-Cu quantitative agreement with experiment. Cations such as Cs+
0.45 0.734 0.584 0.2 0.54 1.1 0.09 have the smallest hydrated cation radius and therefore show

Energy Environ. Sci. This journal is © The Royal Society of Chemistry 2019
 View Article Online

Energy & Environmental Science Paper

Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

Fig. 3 CO partial current density at Ag(110), Ag(111) and poly-crystalline Fig. 4 Partial current density of C2 products (ethanol and ethylene) at
Ag at 1 V vs. RHE for different cations normalized to the CO current Cu(111) and Cu(100) at 1 V vs. RHE for different cations normalized to the
density in the Li+ case. Filled circles represent the experimental data points, C2 current density in the Li+ case. Filled circles represent the experimental
solid lines the theoretical prediction. The Ag(110) data is plotted relative to data points, solid lines the theoretical prediction.
the Na+ cation due to possible impurities in the Li+ measurement (cf. ESI†).

evolution rate (cf. Fig. S4 in the ESI†) suggests that the dipole of
the smallest repulsion close to the electrode. The resulting the corresponding transition state is also significantly smaller.
higher concentrations of cations lead to a larger surface charge Resasco et al. have furthermore shown that formate production
density and stronger interfacial electric field, which drives the varies with cation identify by a similar magnitude as CO
adsorption of CO2, as illustrated in Fig. 2 and 1. production, making a similar mechanism and rate-limiting
The validity of our mean field approach is further supported step likely.
by the correct prediction of the facet dependence of cation We now turn to the effect of cations on the activity on Cu
effects. Recent work has suggested that carefully grown epitaxial epitaxial thin films which has been studied by Resasco et al.4
thin films exhibit a small number of step defects,5,132 which Similar to Ag, no cation effects were observed for hydrogen
predominate in the activity of Ag electrodes.5 Recent Pb deposition evolution reaction, which we again attributed to the small
studies on Au single crystals have further discovered direct dipole and polarizability of adsorbed H atoms. In contrast,
evidence for the predominant activity of steps.111 Since the dipole strong cation effects on C2 formation were observed, which can
moments of the adsorbates involved are essentially facet inde- be attributed to the large dipole of the critical *OCCO inter-
pendent (cf. Fig. S2 and S7 in the ESI†), the actual binding mediate. Fig. 4 shows a comparison of the theoretical and
energies of the adsorbates do not come into play in determining experimental relative activities, and again a surprisingly good
the relative activities amongst the cations. Instead, it is the agreement of the theoretical prediction with the experimental
charging properties (PZC and capacitance) that determine ion results is obtained, with slight deviations for Cs+. Dynamic
specificity. As shown above, although step defects could exhibit interactions with *OCCO may give an additional stabilization of
specific charging properties,2 activity trends still follow the char- *OCCO as seen from explicit DFT calculations,46 leading to the
ging properties of the dominating surface facet. The stronger direct impact of cations on formation energies, which has not
cation effects at the Ag(111) facet can be rationalized by its more been considered here. The stronger cation effects on the
positive PZC and consequently its higher surface charge density Cu(111) surface can be again rationalized by the more positive
and sensitivity to a change of cations (cf. also Fig. S1 in the ESI†). PZC compared to the Cu(100) facet. We stress that these results
The observed cation dependence of the CO production rate were obtained using exactly the same ion sizes as in the other
indicates the efficiency of cation modulation for process two cases.
optimization. In contrast, the small dipole of adsorbed *H
Cation effects on electrochemical CO Stark shift
atoms has been found before to result in a negligible electric
field dependence.5 The relative independence of the hydrogen Next, we considered CO adsorption on Pt surfaces, a system
that has been well studied for electrochemistry.35,87,110,112–120
Among the plethora of experimental insights that has been
Table 2 Obtained effective interfacial cation radii (a/2 in Å) from reported over years, it was found that the CO stretching vibra-
experiment77,79,81 (K+ and Cs+) and inter-/extrapolation to fit the experi-
tion varies significantly with an applied field under ultra-high
mental data of CO2R to CO at Ag surfaces and CO electrochemical Stark
shift at Pt and Cu vacuum conditions.115 This change in the CO vibration
frequency with electric field has been understood as a result
Cs Rb K Na Li TMA TEA TPA TBA of the Stark effect, i.e. the interaction of applied electric fields
3.5 3.9 4.1 5.2 5.8 7.8 8.0 8.1 8.4 with the CO vibrational mode.115 Under electrochemical

This journal is © The Royal Society of Chemistry 2019 Energy Environ. Sci.
 View Article Online

Paper Energy & Environmental Science

Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

Fig. 5 (a) *CO stretching vibrational mode shift on polycrystalline Pt as a function of surface charge density leading to the experimentally observed
electrochemical Stark effect. The upper three lines refer to adsorption on top, the lower ones to adsorption on bridge sites. Different symbols and colors
distinguish different CO coverages. (b) Cation effect on the *CO stretching frequency. Filled circles represent absorbance maxima from the ATR-SEIRAS
spectra of ref. 121 at 1.1 V vs. SHE, solid lines depict the theoretical prediction using the ai1c approach. Orange refers to CO adsorption at the bridge site,
blue to adsorption at the top site. (c) *CO stretching vibrational mode shift on polycrystalline Cu as a function of surface charge density. Filled circles
represent the ATR-SEIRAS data of ref. 118 at 1.39 V vs. SHE, solid lines depict the theoretical prediction using the ai1c approach (using Cu(100) model
surface). The orange line refers to CO adsorption at the bridge site, the blue one to adsorption at the top site. The experimental results could not be
attributed to one of the two sites.

conditions, the interfacial field is controlled by the metal from correlating all cation sizes with experimental radii
potential fM. The Stark tuning rate here, though, may be (cf. Fig. 7), as discussed below. Using this strategy, we generally
complicated by two factors: CO may bind to different sites at found good agreement with the experimental trends.
different potentials, and also at high coverages depolarization A similar experimental ATR-SEIRAS study has also consid-
may occur. ered the cation effect on the CO stretching vibration on a
The frequency shift with electrode potential can be polycrystalline Cu electrode.118 Assuming a coverage of around
expressed as: 50%, which has been found by ab inito based micro-kinetic
modeling,7 we applied the same strategy as for Pt. Fig. 5(c)
d~n n
@~ ds n
@~
¼ ¼ Cdl : (7) shows again excellent agreement with the experimentally
dðfM  fM;PZC Þ @s dðfM  fM;PZC Þ @s
observed cation effect. The consistency between experimental
n
@~ and theoretical Stark shifts further supports the generality of

The first part, , represents the sensitivity of the frequency v to our model and the developed understanding of cation effects.
@s
a change of the surface charge density s, which is roughly
Cation effects on surface charging
constant as a function of s, as depicted in Fig. 5(a). More
n
@~ So far, we discussed that cation repulsion leads to a decrease of
importantly, is nearly the same for *CO at bridge and top the surface charge density and corresponding double layer
@s
sites at a fixed *CO coverage making it a function of the electric field. Fig. 6(a) shows that for a Au(111) surface that
coverage alone (cf. Fig. S15 in the ESI†). Experimental analysis this effect leads to the double layer capacitance decreasing with
of the *CO coverage has been performed in the past110,112 cation size at potentials away from the PZC (0.56 V vs. SHE122).
suggesting it to adopt a fixed, saturated value of around 65% We performed impedance spectroscopy on Au(111) single
at negative potentials below 1 V vs. SHE.110 Considering these crystal electrodes using 0.05 M KClO4 and NaClO4 solutions
results, the experimentally measured change of the CO stretch- to confirm this behavior. Independent of the circuit used for
ing frequency reflects the pure Stark tuning rate without the fitting, we found the double layer capacitance to decrease from
effects of site-redistribution or coverage-dependent depolariza- Na+ to K+ as depicted in Fig. 6(b) (cf. also Fig. S12 and S13 in the
tion, in agreement with previous studies.112,118 ESI†). Indeed, the capacitance increase has been also observed
Fig. 5(b) shows the measured effect of cations on the *CO in Monte Carlo simulations of the electric double layer123 as
stretching frequency on Pt at 1.1 V vs. SHE (dots)121 and the well as impedance studies on single crystal electrodes124 and
theoretical curve (lines) from the ai1c model. In order to supercapacitors.125
simulate the trends with our model, we took the same ion radii Finally, we note that previous studies suggested that cations
as before for the alkali cations and additionally obtained the also affect the PZC as well as the capacitance close to the
radii of TBA and TPA from a fit of our model to the experi- PZC.124 In our data, we noticed a frequency dispersion close to
mental data in Fig. 5. In the case of TEA and TMA, a direct fit the PZC, making a direct interpretation difficult. We also note
would give ion sizes that are not following the expected size that under the commonly applied highly negative potentials for
ordering of the organic cations (TMA o TEA o TPA o TBA). CO2R, the variation in Helmholtz capacitance with ion size
A possible explanation for this could be inaccuracies in the likely dominates the overall surface charge variation (cf. Fig. 6(b)).
experimental determination of the stretching frequency. In Finally, recent studies have found that cation effects do not
order to correct for this, we obtained the TEA and TMA sizes depend on the cation concentration120 which can be seen as

Energy Environ. Sci. This journal is © The Royal Society of Chemistry 2019
 View Article Online

Energy & Environmental Science Paper

Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

Fig. 7 Correlation of the effective interfacial cation radius obtained in this

study with the crystal radius from ref. 82 The analytic expressions 2.2x +
7.6 and 0.42x + 6.3 have been found to fit the cation correlation for
hydrated and hydration-free cations, respectively.

the ESI,† we have simulated model predictions of the relative

CO production rate resulting from both single-valent and
multivalent hydrated inorganic ions using the obtained corre-
lation. The net impact of multivalent ions arise from two
competing effects; their larger charge would increase the inter-
facial field, while their larger sizes tend to decrease the field.
Fig. 6 Double layer capacitance at the Au(111) single-crystal electrode
using a 0.05 M KHCO3 or NaHCO3 electrolyte. (a) Potential-dependence
We found several cations to exhibit up to two orders of
of the double layer capacitance obtained from fitting a RC circuit to the magnitude higher activity than Cs. Ion-correlation effects126
impedance data. Filled circles denote the data points, the solid gray line the or chemical ion-adsorbate interactions could, however, lead to
difference in surface charge density between both experiments under deviations of the estimations using our simplified model. One
the assumption of the same PZC of 0.97 V vs. RHE.122 (b) Solid black line:
experimental study does suggest that lanthanides could exhibit
MPB model predicted double layer capacitance as a function of effective
interfacial cation radius at 0 V vs. RHE. Values at the here determined
a high CO production activiy.127 This study also suggested a
interfacial radii are depicted by filled circles, the measured double layer possible correlation of activity with the surface charge density,
capacitance for K+ and Na+ by the dashed lines. which our theory confirms. From an experimental perspective,
multi-valent cations such as lanthanides or actinides would be
interesting candidates, since their low standard reduction
further evidence that the diffuse layer region does not significantly potential prevents them from adsorption under reducing
contribute to the observed effects. CO2R conditions, and they are furthermore soluble enough to
provide sufficient ionic conductivity.
A second way to increase the conversion rate of electric field
Implications for electrochemical driven reactions is through modification of surface charging
system design properties as the PZC or the gap capacitance. Experimental
studies on CO adsorption on Pt for example have shown that
The sensitivity of CO2R to the interfacial field suggests that the adsorption of CO leads to a PZC increase from 0.23 to 1.1 V vs.
electrolyte and the charging properties of the interface can be SHE.110 By performing DFT-based PZC calculations using
used as design parameters towards electrocatalyst optimiza- implicit solvation, we found a similar increase but only for
tion. In order to estimate the limits of a cation screening bridge bound *CO, while top site bound *CO was less sensitive
approach, we plotted in Fig. 7 the effective interfacial cation (cf. ESI,† Fig. S11). The sensitivity of the PZC to surface
radii as a function of the experimentally measured ionic crystal adsorbate coverage was also shown for other systems128 sug-
radii.82 We find two separate nearly perfect correlations, one for gesting the possible use of co-adsorbates during CO2R. Besides
the organic cations and one for the hydrated ones. The size of this, the PZC can be also modified by altering the surface
the solvated organic cations at the interface generally increases composition. Previous studies electrodeposited a Ag monolayer
with their crystal ion radius.112 Inorganic cations, on the other on Pt or Au and found the PZC to be up to 0.25 V more positive
hand, have relatively rigid solvation shells which scale inversely compared to Ag(111). Moreover, a recent study that deposited
with the crystal radius, leading to a negative slope. In Fig. S17 of Au nanoparticles with a highly positive PZC on a Cu electrode

This journal is © The Royal Society of Chemistry 2019 Energy Environ. Sci.
 View Article Online

Paper Energy & Environmental Science

has found increased C2 selectivity129 that we suggest to be also radiation (40 kV, 40 mA). Symmetric out-of-plane y/2y scans
likely attributed to the more positive PZC compared to Cu. We were conducted to identify the out-of-plane growth orientation
anticipate that the outlined strategies are fully generalizable to of the crystallites in the thin films. Symmetric in-plane j scans
other field-sensitive electrochemical processes, highlighting at Bragg reflections corresponding to both Si and the metallic
their significant impact on the electrocatalyst design. thin film were conducted to determine the orientation of the
thin film crystallites with respect to the Si substrate. Symmetric
out-of-plane O scans were conducted to determine the average
Conclusions degree of misorientation of the thin film crystallites with
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

respect to the surface normal. X-ray pole figures of the thin

In this work, we developed a combined ab initio/continuum
films were acquired using a PANanalytical X’Pert diffractometer
model of cation and electric double layer field effects in electro-
using Cu Ka radiation. The near-surface composition of the
catalysis that was validated against a wide range of experimental
thin films was measured before and after electrolysis with a
data. With the continuum, modified Poisson Boltzmann
Kratos Axis Ultra DLD X-ray photoelectron spectrometer (XPS).
approach we find that the surface charge density and the
All spectra were acquired using monochromatized Al Ka radia-
associated electric field are essentially altered by repulsive
tion (15 kV, 15 mA). The kinetic energy scale of the measured
interactions amongst hydrated cations in the Helmholtz layer.
spectra was calibrated by setting the C 1s binding energy to
This effect, combined with surface-charge dependent reaction
284.8 eV. The same instrument was also used to measure the
energetics determined using DFT, allowed for the theoretical
surface composition of the thin films before and after electro-
prediction of ion-specific catalytic activities.
lysis by ion scattering spectroscopy (ISS).
Using a single set of cation sizes derived from experimental
data, the model showed quantitative agreement with experi- Electrochemical characterization
mentally observed cation effects on CO production on Ag, and
All electrochemical activity measurements were conducted in a
C2 production on Cu, as well as the respective surface facet
custom gas-tight electrochemical cell machined from PEEK.1
dependence. The model also correctly predicted cation effects
The cell was sonicated in 20 wt% nitric acid and thoroughly
on the vibrational stretching mode of *CO on Pt and Cu, as well
rinsed with DI water prior to all experimentation. The working
as on the double layer capacitance of Au(111) single crystal
and counter electrodes were parallel and separated by an anion
electrodes determined by impedance spectroscopy.
exchange membrane (Selemion AMV AGC Inc.). Gas dispersion
The unprecedented agreement with diverse experimental
frits were incorporated into both electrode chambers to provide
data sets demonstrated the generality of the developed under-
ample electrolyte mixing. The exposed geometric surface area
standing of cation and field effects. Finally, we present some
of each electrode was 1 cm2 and the electrolyte volume of each
universal design principles to optimize the conversion
electrode chamber was 1.8 mL. The counter electrode was a
efficiency of field-sensitive electrochemical processes. These
glassy carbon plate (Type 2 Alfa Aesar) that was also sonicated
comprise the use of high valent cations with a small hydration
in 20 wt% nitric acid prior to all experimentation. Platinum was
radius, but also the increase of the potential of zero charge or
not used as the anode due to the possibility of contaminating
capacitance in order to maximize surface charge density and
the cathode.31,130 The working electrode potential was refer-
the corresponding interfacial electric fields. These general
enced against a miniature Ag/AgCl electrode (Innovative Instru-
design principles represent a major step forward for solid–
ments Inc.) that was calibrated against a homemade standard
liquid interface engineering for electrocatalysis.
hydrogen electrode. 0.05 M M2CO3 (M = Li, Na, K, and Cs
99.995% Sigma Aldrich) solutions prepared using 18.2 MO cm
Experimental DI water were used as the electrolyte. Metallic impurities in the
as-prepared electrolytes were removed before electrolysis by
Electrode preparation
chelating them with Chelex 100 (Na form Sigma Aldrich) except
Cu and Ag thin films were deposited onto polished single in the case of Li, which was not purified by additional
crystal Si wafers (1–10 O cm Virginia Semiconductor) with means.130,131 Both electrode chambers were sparged with CO2
(111), (100), and (110) orientations using an AJA ATC Orion-5 (99.999% Praxair Inc.) at a rate of 10 sccm for 30 min prior to
magnetron sputtering system. The Si wafers were etched imme- and throughout the duration of all electrochemical measure-
diately before deposition using 5 wt% HF. Cu thin films were ments unless explicitly stated otherwise. Upon saturation with
deposited at room temperature and Ag thin films were depos- CO2 the pH of the electrolyte was 6.8, which was maintained
ited at 300 1C using an IR lamp. Cu (99.999% Kurt J. Lesker) throughout the duration of all electrocatalytic measurements.
and Ag (99.999% Kurt J. Lesker) were sputtered using Ar ions The hydrodynamic boundary layer thickness at the cathode
onto the etched Si wafers at a rate of 1 Å s1 to obtain a thin surface was determined to be 50 mm by measuring the diffusion
film with a thickness of 100 nm. limited current of ferricyanide reduction.
Double layer capacitance measurements were performed in
Electrode characterization a glass cell that was cleaned with aqua regia prior to all
The crystal structures of the Cu and Ag thin films were analyzed experimentation. The working electrode was a Au(111) single
with a Rigaku Smartlab X-ray diffractometer (XRD) using Cu Ka crystal that was annealed prior to each measurement and

Energy Environ. Sci. This journal is © The Royal Society of Chemistry 2019
 View Article Online

Energy & Environmental Science Paper

utilized in the hanging meniscus configuration. The counter Conflicts of interest

electrode was a Au wire and the reference electrode was a Ag/
AgCl electrode (Pine Research). 0.05 M MClO4 (M = Na and K There are no conflicts of interest to declare.
99.99% Sigma Aldrich) solutions prepared using 18.2 MO cm DI
water were used as the electrolyte. Metallic impurities in the
as-prepared electrolytes were removed before electrolysis by Acknowledgements
chelating them with Chelex 100 (Na form Sigma Aldrich).130,131 This material is based in part on work performed by the Joint
The electrolyte was sparged with Ar (99.999% Praxair Inc.) for Center for Artificial Photosynthesis, a DOE Energy Innovation
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

30 min prior to all electrochemical measurements. The headspace Hub, supported through the Office of Science of the U.S.
of the electrochemical cell was swept with Ar during all measure- Department of Energy, under Award No. DE-SC0004993. This
ments to prevent oxygenation while minimizing measurement research used resources of the National Energy Research Scien-
artifacts arising from electrolyte agitation. tific Computing Center, a DOE Office of Science User Facility
Electrochemistry was performed using a Biologic VSP-300 supported by the Office of Science of the U.S. Department of
potentiostat. All electrochemical measurements were recorded Energy under Contract No. DE-AC02-05CH11231. B. S., and
versus the reference electrode and converted to the RHE scale. K. C. also acknowledge a research grant (9455) from ‘‘VILLUM
Potentiostatic electrochemical impedance spectroscopy (PEIS) FONDEN’’. We also thank Bingjun Xu and Aliaksandr Bandarenka
was used to determine the uncompensated resistance (Ru) of for insightful discussions.
the electrochemical cell and the double layer capacitance (Cdl)
by applying voltage waveforms with an amplitude of 20 mV and
frequencies ranging from 5 Hz to 500 kHz. The potentiostat References
compensated for 85% of Ru in situ and the last 15% was post-
corrected to arrive at accurate potentials. The electrocatalytic 1 Q. Lu and F. Jiao, Electrochemical CO2 reduction: Electro-
activity of the thin films was assessed by conducting chrono- catalyst, reaction mechanism, and process engineering,
amperometry staircases from 0.5 to 1.5 V vs. RHE with a step Nano Energy, 2016, 29, 439–456.
size of 100 mV and a step length of 15 min. Each thin film 2 M. Liu, et al., Enhanced electrocatalytic CO2 reduction via
orientation was tested at least three separate times to ensure fieldinduced reagent concentration, Nature, 2016, 537,
the statistical relevance of the observed trends. 382–386.
3 R. B. Sandberg, J. H. Montoya, K. Chan and J. K. Nørskov,
A. Product analysis COCO coupling on Cu facets: Coverage, strain and field
The effluent from the electrochemical cell was introduced effects, Surf. Sci., 2016, 654, 56–62.
directly into the sampling loop of an Agilent 7890B gas 4 J. Resasco, L. D. Chen, E. Clark, C. Tsai, C. Hahn,
chromatograph (GC) equipped with a pulsed-discharge helium T. F. Jaramillo, K. Chan and A. T. Bell, Promoter Effects of
ionization detector (PDHID). The effluent was sampled at least Alkali Metal Cations on the Electrochemical Reduction of
10 min after each chronoamperometry potential step. The Carbon Dioxide, J. Am. Chem. Soc., 2017, 139, 11277–11287.
constituents of the gaseous sample were separated using a 5 E. L. Clark, S. Ringe, M. Tang, A. L. Walton, C. Hahn,
Hayesep-Q capillary column (Agilent) in series with a packed T. F. Jaramillo, K. Chan and A. T. Bell, Influence of Atomic
ShinCarbon ST column (Restek Co.). He (99.9999% Praxair Inc.) Surface Structure on the Activity of Ag for the Electroche-
was used as the carrier gas. After sampling the effluent of the mical Reduction of CO2 to CO, ACS Catal., 2019, 9(5),
electrochemical cell, the column oven was maintained at 50 for 4006–4014.
1 min followed by a temperature ramp at 30 1C min1 to 250 1C, 6 J. A. Gauthier, S. Ringe, C. F. Dickens, A. J. Garza, A. T. Bell,
which was then maintained for the duration of the analysis. M. Head-Gordon, J. K. Nørskov and K. Chan, Challenges in
The signal response of the PDHID was calibrated by analyzing a Modeling Electrochemical Reaction Energetics with Polar-
series of NIST-traceable standard gas mixtures (Airgas Inc.). izable Continuum Models, ACS Catal., 2019, 9, 920–931.
The electrolyte from both electrode chambers was collected 7 X. Liu, P. Schlexer, J. Xiao, Y. Ji, L. Wang, R. B. Sandberg,
after electrolysis and analyzed using a Thermo Scientific M. Tang, K. S. Brown, H. Peng, S. Ringe, C. Hahn, T. F.
UltiMate 3000 liquid chromatograph (HPLC) equipped with a Jaramillo, J. K. Nørskov and K. Chan, pH effects on the
refractive index detector (RID). The electrolyte samples were electrochemical reduction of CO(2) towards C2 products
stored in a refrigerated autosampler until analyzed to minimize on stepped copper, Nat. Commun., 2019, 10, 32.
the evaporation of volatile liquid-phase reaction products. The 8 I. Katsounaros, J. C. Meier, S. O. Klemm and A. A. Topalov,
liquid-phase products contained in a 10 mL aliquot were sepa- et al., The effective surface pH during reactions at the
rated using a series of two Aminex HPX 87-H columns (Bio-Rad solidliquid interface, Electrochemistry, 2011, 13(6), 634–637.
Inc.) and a 1 mM sulfuric acid eluent (99.999% Sigma Aldrich). 9 I. Ledezma-Yanez, W. D. Z. Wallace, P. Sebastián-Pascual,
The column oven was maintained at 60 1C for the duration of V. Climent, J. M. Feliu and M. T. M. Koper, Interfacial
the analysis. The signal response of the RID was calibrated by water reorganization as a pH-dependent descriptor of the
analyzing standard solutions of each product at a concentration hydrogen evolution rate on platinum electrodes, Nat.
of 1, 10, and 50 mM (see SI-6, ESI†). Energy, 2017, 2, 17031.

This journal is © The Royal Society of Chemistry 2019 Energy Environ. Sci.
 View Article Online

Paper Energy & Environmental Science

10 L. Giordano, B. Han, M. Risch, W. T. Hong, R. R. Rao, Silver Catalysts, Angew. Chem., Int. Ed., 2017, 56,
K. A. Stoerzinger and Y. Shao-Horn, pH dependence of 11394–11398.
OER activity of oxides: Current and future perspectives, 23 F. S. Roberts, K. P. Kuhl and A. Nilsson, High Selectivity for
Catal. Today, 2016, 262, 2–10. Ethylene from Carbon Dioxide Reduction over Copper
11 H. Xiao, T. Cheng, W. A. Goddard 3rd and R. Sundararaman, Nanocube Electrocatalysts, Angew. Chem., Int. Ed., 2015,
Mechanistic Explanation of the pH Dependence and Onset 54, 5179–5182.
Potentials for Hydrocarbon Products from Electrochemical 24 A. Eilert, F. S. Roberts, D. Friebel and A. Nilsson, For-
Reduction of CO on Cu (111), J. Am. Chem. Soc., 2016, 138, mation of Copper Catalysts for CO2 Reduction with High
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

483–486. Ethylene/Methane Product Ratio Investigated with In Situ

12 M. R. Singh, J. D. Goodpaster, A. Z. Weber, M. Head- X-ray Absorption Spectroscopy, J. Phys. Chem. Lett., 2016, 7,
Gordon and A. T. Bell, Mechanistic insights into electro- 1466–1470.
chemical reduction of CO2 over Ag using density func- 25 Q. Lu, J. Rosen and F. Jiao, Nanostructured Metallic
tional theory and transport models, Proc. Natl. Acad. Sci. Electrocatalysts for Carbon Dioxide Reduction, Chem-
U. S. A., 2017, 114, E8812–E8821. CatChem, 2015, 7, 38–47.
13 N. Gupta, M. Gattrell and B. MacDougall, Calculation for 26 M. A. Mellmer, C. Sanpitakseree, B. Demir, P. Bai, K. Ma,
the cathode surface concentrations in the electrochemical M. Neurock and J. A. Dumesic, Solvent-enabled control of
reduction of CO2 in KHCO3 solutions, J. Appl. Electrochem., reactivity for liquid-phase reactions of biomass-derived
2006, 36, 161–172. compounds, Nat. Catal., 2018, 1, 199.
14 M. Dunwell, X. Yang, B. P. Setzler, J. Anibal, Y. Yan and 27 Z. Zhao, R. Bababrik, W. Xue, Y. Li, N. M. Briggs, D.-T.
B. Xu, Examination of Near-Electrode Concentration Gra- Nguyen, U. Nguyen, S. P. Crossley, S. Wang, B. Wang and
dients and Kinetic Impacts on the Electrochemical D. E. Resasco, Solvent-mediated charge separation drives
Reduction of CO2 using Surface-Enhanced Infrared alternative hydrogenation path of furanics in liquid water,
Spectroscopy, ACS Catal., 2018, 3999–4008. Nat. Catal., 2019, 2, 431–436.
15 S. Suter and S. Haussener, Optimizing Mesostructured 28 H.-K. Lim, Y. Kwon, H. S. Kim, J. Jeon, Y.-H. Kim, J.-A. Lim,
Silver Catalysts for Selective Carbon Dioxide Conversion B.-S. Kim, J. Choi and H. Kim, Insight into the Microenvir-
into Fuels, Energy Environ. Sci., 2019, 12, 1668–1678. onments of the Metal-Ionic Liquid Interface during
16 T. Burdyny and W. A. Smith, CO2 reduction on gas- Electrochemical CO2 Reduction, ACS Catal., 2018, 8,
diffusion electrodes and why catalytic performance must 2420–2427.
be assessed at commercially-relevant conditions, Energy 29 K. Yang, R. Kas and W. A. Smith, Tenacious Mass Transfer
Environ. Sci., 2019, 12, 1442–1453. Limitations Drive Catalytic Selectivity during Electro-
17 E. E. Benn, B. Gaskey and J. D. Erlebacher, Suppression of chemical Carbon Dioxide Reduction, 2019, https://chemr
Hydrogen Evolution by Oxygen Reduction in Nanoporous xiv.org/articles/Tenacious_Mass_Transfer_Limitations_Dri
Electrocatalysts, J. Am. Chem. Soc., 2017, 139, 3663–3668. ve_Catalytic_Selectivity_during_Electrochemical_Carbon_
18 A. Verdaguer-Casadevall, C. W. Li, T. P. Johansson, Dioxide_Reduction/7770338.
S. B. Scott, J. T. McKeown, M. Kumar, I. E. L. Stephens, 30 A. S. Varela, M. Kroschel, T. Reier and P. Strasser, Control-
M. W. Kanan and I. Chorkendorff, Probing the Active ling the selectivity of CO2 electroreduction on copper: The
Surface Sites for CO Reduction on Oxide-Derived Copper effect of the electrolyte concentration and the importance
Electrocatalysts, J. Am. Chem. Soc., 2015, 137, 9808–9811. of the local pH, Catal. Today, 2016, 260, 8–13.
19 D. Gao, I. T. McCrum, S. Deo, Y.-W. Choi, F. Scholten, 31 M. Dunwell, Q. Lu, J. M. Heyes, J. Rosen, J. G. Chen, Y. Yan,
W. Wan, J. G. Chen, M. J. Janik and B. Roldan Cuenya, F. Jiao and B. Xu, The Central Role of Bicarbonate in the
Activity and Selectivity Control in CO2 Electroreduction to Electrochemical Reduction of Carbon Dioxide on Gold,
Multicarbon Products over CuOx Catalysts via Electrolyte J. Am. Chem. Soc., 2017, 139, 3774–3783.
Design, ACS Catal., 2018, 10012–10020. 32 J. Resasco, Y. Lum, E. Clark, J. Z. Zeledon and A. T. Bell,
20 Q. Lu, J. Rosen, Y. Zhou, G. S. Hutchings, Y. C. Kimmel, Effects of Anion Identity and Concentration on Electro-
J. G. Chen and F. Jiao, A selective and efficient electro- chemical Reduction of CO2, ChemElectroChem, 2018, 5,
catalyst for carbon dioxide reduction, Nat. Commun., 2014, 1064–1072.
5, 3242. 33 G. Garca, C. Stoffelsma, P. Rodriguez and M. T. M. Koper,
21 J. Rosen, G. S. Hutchings, Q. Lu, S. Rivera, Y. Zhou, In-fluence of beryllium cations on the electrochemical
D. G. Vlachos and F. Jiao, Mechanistic Insights into the oxidation of methanol on stepped platinum surfaces in
Electrochemical Reduction of CO2 to CO on Nanostruc- alkaline solution, Surf. Sci., 2015, 631, 267–271.
tured Ag Surfaces, ACS Catal., 2015, 5, 4293–4299. 34 C. McDonnell-Worth and D. R. MacFarlane, Ion effects in
22 H. Mistry, Y.-W. Choi, A. Bagger, F. Scholten, C. S. water oxidation to hydrogen peroxide, RSC Adv., 2014, 4,
Bonifacio, I. Sinev, N. J. Divins, I. Zegkinoglou, H. S. 30551–30557.
Jeon, K. Kisslinger, E. A. Stach, J. C. Yang, J. Rossmeisl and 35 C. Stoffelsma, P. Rodriguez, G. Garcia, N. Garcia-Araez,
B. Roldan Cuenya, Enhanced Carbon Dioxide Electroreduc- D. Strmcnik, N. M. Marković and M. T. M. Koper, Promo-
tion to Carbon Monoxide over Defect-Rich Plasma-Activated tion of the oxidation of carbon monoxide at stepped

Energy Environ. Sci. This journal is © The Royal Society of Chemistry 2019
 View Article Online

Energy & Environmental Science Paper

platinum singlecrystal electrodes in alkaline media by Reaction Kinetics in Acidic and Alkaline Solutions, J. Phys.
lithium and beryllium cations, J. Am. Chem. Soc., 2010, Chem. C, 2016, 120, 27452–27461.
132, 16127–16133. 50 E. Pérez-Gallent, G. Marcandalli, M. C. Figueiredo, F. Calle-
36 B. A. F. Previdello, E. G. Machado and H. Varela, The effect Vallejo and M. T. M. Koper, Structure- and Potential-
of the alkali metal cation on the electrocatalytic oxidation Dependent Cation Effects on CO Reduction at Copper
of formate on platinum, RSC Adv., 2014, 4, 15271–15275. Single-Crystal Electrodes, J. Am. Chem. Soc., 2017, 139,
37 G. A. Ragoisha, T. A. Auchynnikava, E. A. Streltsov and 16412–16419, PMID: 29064691.
S. M. Rabchynski, Electrochemical impedance of platinum 51 S. A. Akhade, I. T. McCrum and M. J. Janik, The Impact
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

in concentrated chloride solutions under potentiodynamic of Specifically Adsorbed Ions on the Copper-Catalyzed
anodic polarization: Effect of alkali metal cations, Electro- Electroreduction of CO2, J. Electrochem. Soc., 2016, 163,
chim. Acta, 2014, 122, 218–223. F477–F484.
38 D. Strmcnik, K. Kodama, D. van der Vliet, J. Greeley, 52 P. Vanysek, in CRC Handbook of Chemistry and Physics, ed.
V. R. Stamenkovic and N. M. Marković, The role of non- Lide, D. R., CRC Press LLC, 2005, pp. 8–30.
covalent interactions in electrocatalytic fuel-cell reactions 53 O. Z. Stern, Theory of the electrolytic double layer, Elek-
on platinum, Nat. Chem., 2009, 1, 466–472. trochem. Angew. Phys. Chem., 1924, 30, 508–516.
39 D. Strmcnik, D. F. van der Vliet, K.-C. Chang, 54 D. L. Chapman, Theory of electrocapillarity, Philos. Mag.,
V. Komanicky, K. Kodama, H. You, V. R. Stamenkovic 1913, 25, 475–481.
and N. M. Marković, Effects of Li+, K+, and Ba2+ Cations 55 G. Gouy, Constitution of the electric charge at the surface
on the ORR at Model and High Surface Area Pt and Au of an electrolyte, J. Phys., 1910, 9, 457–468.
Surfaces in Alkaline Solutions, J. Phys. Chem. Lett., 2011, 2, 56 I. Borukhov, D. Andelman and H. Orland, Adsorption of
2733–2736. large ions from an electrolyte solution: a modified Pois-
40 A. N. Frumkin, Influence of cation adsorption on the son–Boltzmann equation, Electrochim. Acta, 2000, 46,
kinetics of electrode processes, Trans. Faraday Soc., 1959, 221–229.
55, 156–167. 57 L. D. Chen, M. Urushihara, K. Chan and J. K. Nørskov,
41 A. Murata and Y. Hori, Product Selectivity Affected by Electric Field Effects in Electrochemical CO2 Reduction,
Cationic Species in Electrochemical Reduction of CO2 ACS Catal., 2016, 6, 7133–7139.
and CO at a Cu Electrode, Bull. Chem. Soc. Jpn., 1991, 64, 58 O. Andreussi, I. Dabo and N. Marzari, Revised self-
123–127. consistent continuum solvation in electronic-structure cal-
42 W. Paik, T. N. Andersen and H. Eyring, Kinetic studies of culations, J. Chem. Phys., 2012, 136, 064102.
the electrolytic reduction of carbon dioxide on the mercury 59 R. Sundararaman and K. Schwarz, Evaluating continuum
electrode, Electrochim. Acta, 1969, 14, 1217–1232. solvation models for the electrode-electrolyte interface:
43 S. Kaneco, K. Iiba, H. Katsumata, T. Suzuki and K. Ohta, Challenges and strategies for improvement, J. Chem. Phys.,
Effect of sodium cation on the electrochemical reduction 2017, 146, 084111.
of CO2 at a copper electrode in methanol, J. Solid State 60 F. Nattino, M. Truscott, N. Marzari and O. Andreussi, Con-
Electrochem., 2007, 11, 490–495. tinuum models of the electrochemical diffuse layer in elec-
44 M. R. Thorson, K. I. Siil and P. J. A. Kenis, Effect of Cations tronicstructure calculations, J. Chem. Phys., 2019, 041722.
on the Electrochemical Conversion of CO2 to CO, 61 R. Jinnouchi and A. B. Anderson, Electronic structure
J. Electrochem. Soc., 2013, 160, F69–F74. calculations of liquid-solid interfaces: Combination of
45 M. R. Singh, Y. Kwon, Y. Lum, J. W. Ager 3rd and A. T. Bell, density functional theory and modified Poisson–Boltz-
Hydrolysis of Electrolyte Cations Enhances the Electroche- mann theory, Phys. Rev. B: Condens. Matter Mater. Phys.,
mical Reduction of CO2 over Ag and Cu, J. Am. Chem. Soc., 2008, 77, 245417.
2016, 138, 13006–13012. 62 S. Ringe, H. Oberhofer, C. Hille, S. Matera and K. Reuter,
46 E. Pérez-Gallent, G. Marcandalli, M. C. Figueiredo, F. Calle- Function-Space-Based Solution Scheme for the Size-
Vallejo and M. T. M. Koper, Structure- and Potential- Modified Poisson–Boltzmann Equation in Full-Potential
Dependent Cation Effects on CO Reduction at Copper DFT, J. Chem. Theory Comput., 2016, 12, 4052–4066.
Single-Crystal Electrodes, J. Am. Chem. Soc., 2017, 139, 63 S. Ringe, H. Oberhofer and K. Reuter, Transferable ionic
16412–16419. parameters for first-principles Poisson–Boltzmann solva-
47 E. Sitta, B. C. Batista and H. Varela, The impact of the alkali tion calculations: Neutral solutes in aqueous monovalent
cation on the mechanism of the electro-oxidation of ethylene salt solutions, J. Chem. Phys., 2017, 146, 134103.
glycol on Pt, Chem. Commun., 2011, 47, 3775–3777. 64 K. Letchworth-Weaver and T. A. Arias, Joint density func-
48 J. Suntivich, E. E. Perry, H. A. Gasteiger and Y. Shao-Horn, tional theory of the electrode-electrolyte interface: Applica-
The Influence of the Cation on the Oxygen Reduction and tion to fixed electrode potentials, interfacial capacitances,
Evolution Activities of Oxide Surfaces in Alkaline Electro- and potentials of zero charge, Phys. Rev. B: Condens. Matter
lyte, Electrocatalysis, 2013, 4, 49–55. Mater. Phys., 2012, 86, 075140.
49 S. Zhu, X. Hu, L. Zhang and M. Shao, Impacts of Perchloric 65 N. G. Hörmann, O. Andreussi and N. Marzari, Grand
Acid, Nafion, and Alkali Metal Ions on Oxygen Reduction canonical simulations of electrochemical interfaces in

This journal is © The Royal Society of Chemistry 2019 Energy Environ. Sci.
 View Article Online

Paper Energy & Environmental Science

implicit solvation models, J. Chem. Phys., 2019, Layer during Capacitive Charging/Discharging Processes,
150, 041730. J. Phys. Chem. C, 2014, 118, 22136–22140.
66 Y.-H. Fang, G.-F. Wei and Z.-P. Liu, Theoretical modeling 81 M. Nakamura, Y. Nakajima, K. Kato, O. Sakata and
of electrode/electrolyte interface from first-principles per- N. Hoshi, Surface Oxidation of Au(111) Electrode in
iodic continuum solvation method, Catal. Today, 2013, Alkaline Media Studied by Using X-ray Diffraction and
202, 98–104. Infrared Spectroscopy: Effect of Alkali Metal Cation on
67 K. Mathew and R. G. Hennig, Implicit self-consistent descrip- the Alcohol Oxidation Reactions, J. Phys. Chem. C, 2015,
tion of electrolyte in plane-wave density-functional theory, 119, 23586–23591.
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

2016, arXiv:1601.03346. 82 E. R. Nightingale, Phenomenological Theory of Ion Solva-

68 M. Otani and O. Sugino, First-principles calculations of tion. Effective Radii of Hydrated Ions, J. Phys. Chem., 1959,
charged surfaces and interfaces: A plane-wave nonrepeated 63, 1381–1387.
slab approach, Phys. Rev. B: Condens. Matter Mater. Phys., 83 K. Chan and M. Eikerling, A Pore-Scale Model of Oxygen
2006, 73, 115407. Reduction in Ionomer-Free Catalyst Layers of PEFCs,
69 I. Dabo, Y. Li, N. Bonnet and N. Marzari in Fuel Cell J. Electrochem. Soc., 2011, 158, B18–B28.
Science, ed. A. Wieckowski and J. K. Nørskov, Electrodics 84 J. Huang, A. Malek, J. Zhang and M. H. Eikerling, Non-
2A, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2010, vol. monotonic Surface Charging Behavior of Platinum:
10, pp 415–431. A Paradigm Change, J. Phys. Chem. C, 2016, 120,
70 R. Sundararaman, K. Letchworth-Weaver and K. A. Schwarz, 13587–13595.
Improving accuracy of electrochemical capacitance and 85 Y. Ando, Y. Gohda and S. Tsuneyuki, Ab initio molecular
solvation energetics in first-principles calculations, J. Chem. dynamics study of the Helmholtz layer formed on solid-
Phys., 2018, 148, 144105. liquid interfaces and its capacitance, Chem. Phys. Lett.,
71 R. Gomer and G. Tryson, An experimental determination 2013, 556, 9–12.
of absolute half-cell emf’s and single ion free energies of 86 A. Ruzanov, K. Karu, V. Ivaništšev, R. R. Nazmutdinov and
solvation, J. Chem. Phys., 1977, 66, 4413–4424. E. Lust, Interplay between the hydrophilicity of metal
72 E. R. Kötz, H. Neff and K. A. Müller, UPS, XPS and work electrodes and their interfacial capacitance, Electrochim.
function study of emersed silver, platinum and gold elec- Acta, 2016, 210, 615–621.
trodes, J. Electroanal. Chem. Interfacial Electrochem., 1986, 87 R. Sundararaman, M. C. Figueiredo, M. T. M. Koper and
215, 331–344. K. A. Schwarz, Electrochemical Capacitance of CO-
73 W. N. Hansen and G. Hansen, Absolute half-cell potential: Terminated Pt(111) Dominated by the CO-Solvent Gap,
A simple direct measurement, J. Phys. Rev. A Gen. Phys., J. Phys. Chem. Lett., 2017, 8, 5344–5348.
1987, 36, 1396–1402. 88 A. Hamelin, T. Vitanov, E. Sevastyanov and A. Popov, The
74 W. R. Fawcett, The ionic work function and its role in electrochemical double layer on sp metal single crystals:
estimating absolute electrode potentials, Langmuir, 2008, The current status of data, J. Electroanal. Chem. Interfacial
24, 9868–9875. Electrochem., 1983, 145, 225–264.
75 A. A. Isse and A. Gennaro, Absolute potential of the 89 A. A. Kornyshev, E. Spohr and M. A. Vorotyntse, in Ency-
standard hydrogen electrode and the problem of intercon- clopedia of Electrochemistry, ed. A. J. Bard, Techn. Report no
version of potentials in different solvents, J. Phys. Chem. B, 14 to the office of Naval Research of Febr. 18, 1954, Wiley-
2010, 114, 7894–7899. VCH Verlag GmbH & Co. KgaA, Weinheim, Germany, 2007,
76 J. Haruyama, T. Ikeshoji and M. Otani, Electrode potential vol. 41, p 2307.
from density functional theory calculations combined 90 E. Lust, in Encyclopedia of Electrochemistry, ed. Bard, A. J.,
with implicit solvation theory, Phys. Rev. Mater., 2018, Wiley-VCH Verlag GmbH & Co. KgaA, Weinheim, Germany,
2, 095801. 2007, vol. 16, p. 1.
77 C. A. Lucas, P. Thompson, Y. Gründer and N. M. Markovic, 91 G. Valette, Double layer on silver single-crystal electrodes in
The structure of the electrochemical double layer: Ag(111) contact with electrolytes having anions which present a slight
in alkaline electrolyte, Electrochem. Commun., 2011, 13, specific adsorption: Part I. The (110) face, J. Electroanal.
1205–1208. Chem. Interfacial Electrochem., 1981, 122, 285–297.
78 M. Nakamura, N. Sato, N. Hoshi and O. Sakata, Outer 92 G. Valette, Double layer on silver single crystal electrodes
Helmholtz plane of the electrical double layer formed at in contact with electrolytes having anions which are
the solid electrodeliquid interface, ChemPhysChem, 2011, slightly specifically adsorbed: Part II. The (100) face,
12, 1430–1434. J. Electroanal. Chem. Interfacial Electrochem., 1982, 138,
79 M. Nakamura, Y. Nakajima, N. Hoshi, H. Tajiri and 37–54.
O. Sakata, Effect of Non-Specifically Adsorbed Ions on 93 G. Valette, Double layer on silver single crystal electrodes
the Surface Oxidation of Pt(111), ChemPhysChem, 2013, in contact with electrolytes having anions which are
14, 2426–2431. slightly specifically adsorbed: Part III. The (111) face,
80 M. Nakamura, H. Kaminaga, O. Endo, H. Tajiri, O. Sakata J. Electroanal. Chem. Interfacial Electrochem., 1989, 269,
and N. Hoshi, Structural Dynamics of the Electrical Double 191–203.

Energy Environ. Sci. This journal is © The Royal Society of Chemistry 2019
 View Article Online

Energy & Environmental Science Paper

94 M. Weidenauer and K. G. Weil, Studies of the Differential and Its Low-Overpotential Reduction to CH3CH2OH on
Double Layer Capacitance of the Copper (111) Face in Cu(511) Surface Reconstructed from Cu(pc): Operando
Aqueous Solutions, Berichte der Bunsengesellschaft für phy- Studies by Seriatim STM-EQCN-DEMS, J. Electrochem.
sikalische Chemie, 1988, 92, 1368–1372. Soc., 2018, 165, J3350–J3354.
95 T. Pajkossy and D. M. Kolb, On the origin of the double 109 S. Trasatti and E. Lust, Modern aspects of electrochemistry,
layer capacitance maximum of Pt(111) single crystal elec- Springer, 2002, pp 1–215.
trodes, Electrochem. Commun., 2003, 5, 283–285. 110 A. Cuesta, Measurement of the surface charge density of
96 J. Rossmeisl, E. Skúlason, M. E. Björketun, V. Tripkovic CO saturated Pt(111) electrodes as a function of potential:
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

and J. K. Nørskov, Modeling the electrified solid-liquid the potential of zero charge of Pt(111), Surf. Sci., 2004, 572,
interface, Chem. Phys. Lett., 2008, 466, 68–71. 11–22.
97 K. Chan and J. K. Nørskov, Electrochemical Barriers Made 111 S. Mezzavilla, S. Horch, I. E. L. Stephens, B. Seger and
Simple, J. Phys. Chem. Lett., 2015, 6, 2663–2668. I. Chorkendorff, Structure Sensitivity in the Electrocatalytic
98 S. E. Weitzner and I. Dabo, Voltage-dependent cluster Reduction of CO2 with Gold Catalysts, Angew. Chem., Int.
expansion for electrified solid-liquid interfaces: Applica- Ed., 2019, 58, 3774–3778.
tion to the electrochemical deposition of transition metals, 112 M. R. Anderson and J. Huang, The influence of cation size
Phys. Rev. B, 2017, 96, 205134. upon the infrared spectrum of carbon monoxide adsorbed
99 B. A. Rosen, A. Salehi-Khojin, M. R. Thorson, W. Zhu, on platinum electrodes, J. Electroanal. Chem. Interfacial
D. T. Whipple, P. J. A. Kenis and R. I. Masel, Ionic Electrochem., 1991, 318, 335–347.
liquidmediated selective conversion of CO2 to CO at low 113 F. Tao, S. Dag, L.-W. Wang, Z. Liu, D. R. Butcher,
overpotentials, Science, 2011, 334, 643–644. H. Bluhm, M. Salmeron and G. A. Somorjai, Break-up of
100 A. Wuttig, Y. Yoon, J. Ryu and Y. Surendranath, Bicarbo- stepped platinum catalyst surfaces by high CO coverage,
nate Is Not a General Acid in Au-Catalyzed CO2 Electro- Science, 2010, 327, 850–853.
reduction, J. Am. Chem. Soc., 2017, 139(47), 17109–17113. 114 I. Grinberg, Y. Yourdshahyan and A. M. Rappe, CO on
101 A. Wuttig, M. Yaguchi, K. Motobayashi, M. Osawa and Pt(111) puzzle: A possible solution, J. Chem. Phys., 2002,
Y. Surendranath, Inhibited proton transfer enhances Au- 117, 2264–2270.
catalyzed CO2-to-fuels selectivity, Proc. Natl. Acad. Sci. U. S. 115 M. Mamatkulov and J.-S. Filhol, An ab initio study of
A., 2016, 113, E4585–E4593. electrochemical vs. electromechanical properties: the case
102 M. Dunwell, W. Luc, Y. Yan, F. Jiao and B. Xu, Under- of CO adsorbed on a Pt(111) surface, Phys. Chem. Chem.
standing Surface-Mediated Electrochemical Reactions: Phys., 2011, 13, 7675–7684.
CO2 Reduction and Beyond, ACS Catal., 2018, 8, 116 M. J. S. Farias, E. Herrero and J. M. Feliu, Site Selectivity for
8121–8129. CO Adsorption and Stripping on Stepped and Kinked
103 N. J. Firet and W. A. Smith, Probing the Reaction Mecha- Platinum Surfaces in Alkaline Medium, J. Phys. Chem. C,
nism of CO2 Electroreduction over Ag Films via Operando 2013, 117, 2903–2913.
Infrared Spectroscopy, ACS Catal., 2017, 7, 606–612. 117 N. Dimakis, N. E. Navarro, T. Mion and E. S. Smotkin,
104 I. V. Chernyshova, P. Somasundaran and S. Ponnurangam, Carbon Monoxide Adsorption Coverage Study on Platinum
On the origin of the elusive first intermediate of CO2 and Ruthenium Surfaces, J. Phys. Chem. C, 2014, 118,
electroreduction, Proc. Natl. Acad. Sci. U. S. A., 2018, 11711–11722.
115(40), E9261–E9270. 118 C. M. Gunathunge, V. J. Ovalle and M. M. Waegele, Probing
105 M. Bajdich, M. Fields, L. D. Chen, R. B. Sandberg, K. Chan promoting effects of alkali cations on the reduction of CO
and J. K. Nørskov, Electron Transfer to CO2 during Adsorp- at the aqueous electrolyte/copper interface, Phys. Chem.
tion at the Metal – Solution Interface, J. Phys. Chem. Lett., Chem. Phys., 2017, 19, 30166–30172.
2019, https://chemrxiv.org/articles/Electron_Transfer_to_ 119 G. T. K. Kalhara Gunasooriya and M. Saeys, CO Adsorption
CO2_during_Adsorption_at_the_Metal_Solution_Interface/ Site Preference on Platinum: Charge Is the Essence, ACS
8187629. Catal., 2018, 8, 3770–3774.
106 T. Cheng, H. Xiao and W. A. Goddard 3rd, Full atomistic 120 M. C. Figueiredo, D. Hiltrop, R. Sundararaman, K. A. Schwarz
reaction mechanism with kinetics for CO reduction on and M. T. M. Koper, Absence of diffuse double layer effect on
Cu(100) from ab initio molecular dynamics free-energy the vibrational properties and oxidation of chemisorbed
calculations at 298 K, Proc. Natl. Acad. Sci. U. S. A., 2017, carbon monoxide on a Pt(111) electrode, Electrochim. Acta,
114, 1795–1800. 2018, 281, 127–132.
107 Y. Huang, A. D. Handoko, P. Hirunsit and B. S. Yeo, 121 M. Dunwell, J. Wang, Y. Yan and B. Xu, Surface enhanced
Electrochemical Reduction of CO2 Using Copper Single- spectroscopic investigations of adsorption of cations on
Crystal Surfaces: Effects of CO* Coverage on the Selective electrochemical interfaces, Phys. Chem. Chem. Phys., 2017,
Formation of Ethylene, ACS Catal., 2017, 7, 1749–1756. 19, 971–975.
108 C. F. Tsang, A. C. Javier, Y.-G. Kim, J. H. Baricuatro, 122 Modern Aspects of Electrochemistry, ed. R. E. White, et al.,
K. D. Cummins, J. Kim, G. Jerkiewicz, J. C. Hemminger KluwerAcademic/PlenumPublishers, New York, vol. 33,
and M. P. Soriaga, Potential-Dependent Adsorption of CO 1999.

This journal is © The Royal Society of Chemistry 2019 Energy Environ. Sci.
 View Article Online

Paper Energy & Environmental Science

123 R. Górniak and S. Lamperski, On the infuence of physical 128 P. Ramrez, R. Andreu, A. Cuesta, C. J. Calzado and
parameters on the properties of the electric double layer J. Calvente, Determination of the potential of zero charge
modelled by soft potentials. A Monte Carlo study, Electro- of Au(111) modified with thiol monolayers, J. Anal. Chem.,
chim. Acta, 2018, 286, 279–286. 2007, 79, 6473–6479.
124 B. Garlyyev, S. Xue, S. Watzele, D. Scieszka and 129 C. G. Morales-Guio, E. R. Cave, S. A. Nitopi, J. T. Feaster,
A. S. Bandarenka, Infuence of the Nature of the Alkali L. Wang, K. P. Kuhl, A. Jackson, N. C. Johnson, D. N. Abram,
Metal Cations on the Electrical Double-Layer Capacitance T. Hatsukade, C. Hahn and T. F. Jaramillo, Improved CO2
of Model Pt(111) and Au(111) Electrodes, J. Phys. Chem. reduction activity towards C2+ alcohols on a tandem gold on
Published on 03 July 2019. Downloaded by Nottingham Trent University on 7/17/2019 11:10:43 AM.

Lett., 2018, 9, 1927–1930. copper electrocatalyst, Nat. Catal., 2018, 1, 764–771.

125 J. Zhu, Y. Xu, J. Wang, J. Lin, X. Sun and S. Mao, The effect 130 E. L. Clark, J. Resasco, A. Landers, J. Lin, L.-T. Chung,
of various electrolyte cations on electrochemical perfor- A. Walton, C. Hahn, T. F. Jaramillo and A. T. Bell, Data
mance of polypyrrole/RGO based supercapacitors, Phys. Acquisition Protocols and Reporting Standards for Studies
Chem. Chem. Phys., 2015, 17, 28666–28673. of the Electrochemical Reduction of Carbon Dioxide, ACS
126 E. Wernersson, R. Kjellander and J. Lyklema, Charge Catal., 2018, 8, 6560–6570.
Inversion and Ion-Ion Correlation Effects at the Mercury/ 131 A. Wuttig and Y. Surendranath, Impurity Ion Complexa-
Aqueous MgSO4 Interface: Toward the Solution of a Long- tion Enhances Carbon Dioxide Reduction Catalysis, ACS
Standing Issue, J. Phys. Chem. C, 2010, 114, 1849–1866. Catal., 2015, 5, 4479–4484.
127 A. Schizodimou and G. Kyriacou, Acceleration of the 132 C. Hahn, T. Hatsukade, Y.-G. Kim, A. Vailionis, J. H. Baricuatro,
reduction of carbon dioxide in the presence of multivalent D. C. Higgins, S. A. Nitopi, M. P. Soriaga and T. F. Jaramillo,
cations, Electrochim. Acta, 2012, 78, 171–176. Proc. Nat. Acad. Sci. U. S. A., 2017, 114, 5918–5923.

Energy Environ. Sci. This journal is © The Royal Society of Chemistry 2019