Stabilizing Copper for CO2 Reduction in Low-Grade Electrolyte

Article abstract of DOI:10.1021/acs.inorgchem.8b02311

We demonstrate herein a CO₂ reduction electrocatalyst regeneration strategy based on the manipulation of the Cu(0)/Cu²⁺ equilibrium with high concentrations of ethylenediaminetetraacetic acid (EDTA). This strategy enables the sustained performance of copper catalysts in distilled and tap water electrolytes for over 12 h. The deposition of common electrolyte impurities such as iron, nickel, and zinc is blocked because EDTA can effectively bind the metal ions and negatively shift the electrode potential of M/Mⁿ⁺. The Cu/Cu²⁺ redox couple is >600 mV more positive than the other metal ions and therefore participates in an equilibrium of dissolution and redeposition from and to the electrode in high concentrations of EDTA. These dynamic equilibria serve to further regenerate the surface copper catalyst to prevent the deactivation of catalytic sites. On the basis of this strategy, we show that >95% of initial hydrocarbon production activity can be maintained for 12 h in KHCO₃ (99% purity) enriched distilled water and 6 h in KHCO₃ (99% purity) enriched tap water.

Full text of DOI:10.1021/acs.inorgchem.8b02311

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Stabilizing Copper for CO2 Reduction in Low-Grade Electrolyte
Jingfu He,† Aoxue Huang,† Noah J. J. Johnson,† Kevan E. Dettelbach,† David M. Weekes,†
Yang Cao,† and Curtis P. Berlinguette*,†,‡,§
†Department of Chemistry, The University of British Columbia, 2036 Main Mall, Vancouver, British Columbia V6T 1Z1, Canada
‡Department of Chemical & Biological Engineering, The University of British Columbia, 2360 East Mall, Vancouver, British
Columbia V6T 1Z3, Canada
§Stewart Blusson Quantum Matter Institute, The University of British Columbia, 2355 East Mall, Vancouver, British Columbia V6T
1Z4, Canada
S
* Supporting Information
ABSTRACT: We demonstrate herein a CO2 reduction electrocatalyst
regeneration strategy based on the manipulation of the Cu(0)/Cu2+
equilibrium with high concentrations of ethylenediaminetetraacetic acid
(EDTA). This strategy enables the sustained performance of copper catalysts
in distilled and tap water electrolytes for over 12 h. The deposition of common
electrolyte impurities such as iron, nickel, and zinc is blocked because EDTA
can effectively bind the metal ions and negatively shift the electrode potential
of M/Mn+. The Cu/Cu2+ redox couple is >600 mV more positive than the
other metal ions and therefore participates in an equilibrium of dissolution and
redeposition from and to the electrode in high concentrations of EDTA. These
dynamic equilibria serve to further regenerate the surface copper catalyst to
prevent the deactivation of catalytic sites. On the basis of this strategy, we show
that >95% of initial hydrocarbon production activity can be maintained for 12
h in KHCO3 (99% purity) enriched distilled water and 6 h in KHCO3 (99% purity) enriched tap water.
but this strategy has yet to be applied to electrochemical CO2
reduction systems.
INTRODUCTION
■
Electrochemical CO2 reduction technologies require electro-
catalysts that can operate continuously for extended time
periods.1−4 CO2 reduction reaction (CO2RR) electrocatalysts
are therefore typically tested in ultrapure electrolytes in an
effort to avoid catalyst poisoning from metal impurities.5−11
Copper catalysts are an appealing class of CO2RR electro-
catalysts owing to their ability to produce hydrocarbons, but
they are particularly susceptible to poisoning. Indeed, copper
systems significantly lose catalytic activity in <1 h when
electrolysis is performed in an untreated electrolyte constitut-
ing Milli-Q water and ACS standard >99% purity KHCO3.3,12
Trace amounts of impurities in the electrolyte and buildup of
carbonaceous byproducts on the catalyst surface are purported
to be responsible for catalyst deactivation.12−14 This situation
presents a conflict for achieving long-term stability while using
an inexpensive electrolyte for practical electrolysis.
There are many ways to address catalyst deactivation for
electrolytic and photolytic processes beyond CO2RR (e.g.,
oxygen evolution,15−17 oxygen reduction,18 hydrogen evolu-
tion19). The strategy of electrolyte tuning/modification to
recover and maintain sustained electrolysis has resulted in
manganese-, cobalt-, and nickel-oxide-based water oxidation
catalysts that corrode and redeposit under catalytic con-
ditions.16,17,20−26 This reactivity has proven effective for
providing prolonged activity for oxygen evolution catalysis,
Efforts to maintain sustained CO2RR activity by copper
catalysts typically focus on preventing electrodes from impurity
poisoning from electrolyte. For example, pretreatment of an
electrolyte consisting of Milli-Q water and ACS standard >99%
purity KHCO3 by predeposition of electrolyte impurities onto
a sacrificial electrode or by Chelex ion exchange can enhance
the effective lifetime of a catalyst several-fold; however, the
reagent and time costs are high.3,12,27−29 Oxide-derived copper
and gold electrocatalysts reported by Kanan and co-workers in
2012 were demonstrated to be less susceptible to poisoning
from electrolyte impurities compared with copper and gold
planar electrodes due to their large surface area and can stably
produce CO and HCOOH for extended periods (∼7 h).7,30 A
self-cleaning strategy that blocks the deposition of impurities
and carbon byproducts was proposed by Wang and co-workers
in order to maintain the stability of a copper catalyst.4 Therein,
palladium atoms predeposited on a copper surface appeared to
induce a morphological and compositional restructuring of the
copper surface, serving to continually refresh the catalyst and
maintain electrochemical selectivity. The faradaic efficiency
(FE) for hydrocarbon products in that case was stable over 5 h
in pre-electrolyzed KHCO3 electrolyte (Milli-Q water and
Received: August 15, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
99.7% KHCO3). Despite these collective advances, there
remains no reported case of a copper catalyst that can mediate
hydrocarbon evolution in low- or medium-grade electrolytes
(e.g., distilled water or tap water containing up to 99% purity
electrolyte salt).
We show herein a catalyst regeneration strategy that blocks
the deposition of certain metallic impurities (e.g., Ni, Fe, Zn)
and prevents byproduct accumulation which inhibits electro-
catalytic performance by simultaneously re-forming the copper
catalyst surface layer during CO2 electrolysis conditions
(Scheme 1). This result is achieved by the addition of a high
equilibria of metals and metal−organic complexes to
regenerate CO2 reduction catalysts.
EXPERIMENTAL SECTION
■
CO2 reduction experiments were performed in a gastight two-
compartment glass H-cell with a proton exchange membrane (Nafion
117; Fuel Cell Store) as a separator. Each compartment contained 50
mL of 0.5 M KHCO3 electrolyte with various concentratations of
EDTA. The quality of the electrolyte is denoted by the type of water
used in each experiment, which is either Milli-Q (MQ) water, distilled
(DI) water, or tap (TAP) water. The H-cell was rinsed before each
series of experiments with a 1:1 mixture of concentrated HNO3/
H2SO4 in order to regulate the impurity levels of the experiments.
Prior to starting the experiment, the H-cell was rinsed with MQ water
three times and with electrolyte once. Long-term stability testing was
performed at −1.3 V vs RHE without iR compensation, as are all
voltages reported in this research. The resistance is 8 0.5 Ω for all
systems. This potential was chosen as the optimal CO2-to-hydro-
carbon potential region.32,33 The copper catalyst was prepared by
electrochemical deposition of copper onto a carbon cloth (see
Supporting Information).34−36 The dendritic morphology of copper
prepared by this method was reported by Reller to be an effective
hydrocarbon production catalyst.34 The total loading amount of as-
prepared copper on carbon cloth was ∼0.003 mmol/cm2 based on
UV−vis absorption titration experiments. The active area of the
copper catalyst/carbon samples was 1 cm2 for all the electrochemical
tests in this study. The experiments monitoring copper equilibrium
were halted every half hour in the first 2 h and subsequently every 1 h
to measure the copper loading on the carbon cloth by X-ray
fluorescence (XRF) and copper dissolution in electrolyte by UV−vis
spectroscopy.
Scheme 1. Proposed Mechanism for the Prolonged Stability
of Copper Electrocatalysts towards CO2-to-Hydrocarbon
Reduction as a Result of the Addition of
Ethylenediaminetetraacetic Acid (EDTA) in the Electrolyte
a
Solution
RESULTS AND DISCUSSION
■
a
EDTA binds all four metal ions (Cu2+, Ni2+, Fe2+/3+, Zn2+) with high
Electrochemical reduction of CO2 to CO, CH4, and C2H4 on
dendritic copper electrocatalysts was tested in electrolytes with
three different impurity levels (MQ, DI or TAP electrolyte)
and four different additive choices (without EDTA, 1 mg/L
EDTA, 1 g/L EDTA, or 1 g/L EDTA + 0.025 mM CuCl2)
(Figure 1). A high-purity electrolyte was prepared from MQ
water and (99.997%) K2CO3. A medium-purity electrolyte was
prepared from DI water and (99%) KHCO3, and a low-purity
electrolyte was prepared from untreated TAP and (99%)
KHCO3. The electrolytes all contained K+ concentrations of
0.5 M and were all presaturated with gaseous CO2 to reach a
pH of 6.9 prior to electrolysis. The metal-ion concentrations
for each of the three electrolytes were quantified by inductively
coupled plasma optical emission spectroscopy (ICP-OES and
ICP-MS; see Supporting Information).
The electrocatalytic performance in MQ electrolyte was in
accordance with previous literature reports (∼20−30% FE for
each of methane and ethylene; ∼10% FE for CO).32 The
overall current density (j) was stable at 38 mA cm−2 during 6 h
of operation; however, the total FE for CO2 reduction
products gradually decreased by about one-third during this
time (Figure 1a). Electrochemical CO2 reduction in DI
electrolyte displayed an initial performance similar to that of
the MQ electrolyte, but the performance rapidly declined over
the course of the experiment (FE = <10% for CH4 and <3% for
C2H4 and CO after 6 h of continuous operation; Figure 1b). In
the TAP electrolyte, the electrocatalyst was immediately
deactivated and no CO2 reduction products were detected
after merely 1 h of electrolysis (Figure 1c). These experiments
demonstrate the widely documented vulnerability of copper
toward electrocatalyst poisoning and deactivation while
operating in non-ultrapure electrolytes. Adding low concen-
binding affinity (log βM‑EDTA ∼ 16−18) and negatively shifts the
deposition potential; thus, only copper has a sufficiently positive
reduction potential to enable redeposition of the metal on the
electrode surface under an applied voltage and continuously
regenerate the surface catalytic sites. The deposition of impurity
metals (Fe, Ni, Zn) is blocked by the presence of EDTA, thereby
preventing catalyst poisoning. Reduction potentials are vs NHE.
concentration (>1000 mg/L) of the common metal chelator
ethylenediaminetetraacetic acid (EDTA)31 in the electrolyte
solution. Lower concentrations of EDTA (∼1 mg/L) in the
electrolyte have been reported to extend the lifetime of gold
catalysts in ultrapure electrolyte, but this strategy was reported
not to improve the lifetime of copper catalysts.27 EDTA has a
strong binding affinity (log βM‑EDTA ∼ 16−18) for four metal
ions Cu2+, Fe3+, Ni2+, and Zn2+ present in our local water
supply and readily forms the metal complexes in solution.27
The standard electrode potential of M2+/M0 for Cu is at least
600 mV positive compared to the relevant Fe, Ni, and Zn
redox couples, so it will be less likely to totally dissolve in the
EDTA-containing electrolyte during CO2RR but rather will
form a dynamic equilibrium of Cu/Cu2+/EDTA-Cu. The
leaching and redeposition of copper can be encouraged by a
high concentration of EDTA in the electrolyte, so that the
active surface of the electrocatalyst is constantly refreshed.
Using this strategy, we are able to maintain >95% of the
activity of a copper catalyst toward CO2-to-hydrocarbon
reduction for 12 h in distilled water with ACS grade KHCO3
(99% purity), and up to 6 h stability in untreated tap water.
This discovery showcases the possibility of manipulating the
B
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Time-resolved current densities (j, black line) and faradaic efficiencies (FEs, dots) for CO2 reduction products CO, CH4, and C2H4 on
dendritic copper catalysts supported on carbon cloth in a CO2-saturated 0.5 M KHCO3 electrolyte either (a−c) without EDTA, (d−f) with 1 mg/L
EDTA, (g−i) with 1000 mg/L EDTA in solution, or (j−l) with 1000 mg/L EDTA and 0.025 mM CuCl2 in solution and composed of (a, d, g, and
j) Milli-Q (MQ) water and 99.997% K2CO3, (b, e, h, and k) distilled (DI) water and 99% KHCO3, and (c, f, i, and l) tap (TAP) water and 99%
KHCO3. Electrolytes with 1000 mg/L EDTA are labeled as XEDTA electrolyte (X = MQ, DI, or TAP) throughout this paper. The liquid products of
1
formic acid, methanol, and ethanol were quantified by H NMR, but the amounts produced were <3%; we discontinued monitoring for liquid
products as shown in Figure S4. The resistances are 8 0.5 Ω for all systems and not compensated for in our system.
trations of EDTA (1 mg/L) did not influence the lifetime of
copper catalyst in the three types of electrolyte (Figures 1d−f),
and the decline in electrocatalytic stability is consistent with
those samples in which EDTA was not present. Considering
that the highest impurity concentration detected in MQ
electrolyte is 3.7 nM for Zn, 3 orders of magnitude lower than
the EDTA concentration, these data indicate that impurity
poisoning is not the only reason for the observed deactivation
process.
stability and the competitive reactions between HER and
CO2RR. Stable FEs of 40% and 10% were also recorded for
CH4 and C2H4, respectively, over 6 h in TAPEDTA and declined
only slightly after this time (Figure 1i). This result is in stark
contrast to the almost immediate decline in electrocatalytic
performance witnessed in the TAP electrolyte (Figure 1c,f).
The observations in TAPEDTA are the first demonstration of
copper-catalyzed CO2 reduction with sustained electrocatalytic
performance in a low-grade electrolyte (TAP).
Electrocatalytic CO2 reduction stability was enhanced in all
three electrolyte compositions by the addition of 1000 mg/L
EDTA to the electrolyte solution while holding all other
parameters at parity (Figure 1g−i; see Figures S1, S2, and S3
for performance and stability data). The initial FEs for CH4,
C2H4, and CO production were ∼50%, ∼15%, and ∼10%,
respectively, for both the MQEDTA and DIEDTA electrolytes.
These values remained virtually constant throughout 12 h of
electrolysis at j ∼ 40 mA cm−2 (Figure 1g,h). The current
densities for all the different electrolytes and EDTA
concentrations were approximately the same, and the changes
in current densities over 12 h electrolysis were less than 10%,
thereby providing a useful platform for understanding catalyst
The X-ray diffraction (XRD) pattern and X-ray photo-
electron spectroscopy (XPS) for the copper electrocatalyst did
not reveal apparent changes in peak intensities and positions
after 12 h of continuous operation (Figures S4a and S5). The
electrolytes were assessed postelectrolysis by 1H NMR
spectroscopy (Figure S4b) to reveal an intensity change of
<1% and no chemical shift of the EDTA hydrogen peaks (at
3.2 and 3.6 ppm) after 12 h of continuous operation. In a
separate control experiment, copper electrolysis in 0.5 M KCl
electrolyte containing 1000 mg/L EDTA was tested at the
same voltage under N2 flow (Figures S6 and S7). H2 was the
only product detected by GC and NMR analysis. The
negligible change in the EDTA concentration of the catholyte
C
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. (a) XRF spectra of a carbon cloth electrode after 4 h of electrodeposition at −1.3 V vs RHE in DIEDTA electrolyte containing 0.1 mM
Fe(NO3)3, Ni(NO3)2, Cu(NO3)2, and Zn(NO3)2 with EDTA at various concentrations (from top to bottom: 0, 1, 10, 100, 1000, 10000 mg/L).
(b) XRF spectra of a carbon cloth electrode after 4 h of electrodeposition at −0.9 V vs RHE in DIEDTA electrolyte containing 0.1 mM Fe(NO3)3,
Ni(NO3)2, Cu(NO3)2, and Zn(NO3)2 with EDTA at various concentrations (from top to bottom: 0, 1, 10, 100, 1000, and 10 000 mg/L).
(<2%) and the lack of any EDTA on the anolyte side rule out
the possibility of EDTA acting as a sacrificial agent during
electrocatalysis.
sample, even in 1 mg/L EDTA. Although metal-ion impurities
cannot be deposited on the copper catalyst regardless of the
low concentration of EDTA, the FTIR spectrum of copper foil
after 12 h of operation with 1 mg/L EDTA shows new peaks at
1400 and 1700 cm−1 (Figure S9), which are not observed for
copper operating in 1000 mg/L EDTA. The peak at ∼1400
cm−1 likely corresponds to CO3 and HCO3; however, this peak
remains when the foil is rinsed with water. This indicates that
there might be some chemical bond between copper and
carbonate. The peak at around ∼1700 cm−1 is probably similar
to some intermediate products of CO2 reduction, such as
CHO*.37 As mentioned in previous research studies, carbon-
based compounds can block the surface and deactivate the
catalyst, indicating that an important contribution of the large
concentration of EDTA is to prevent surface blocking.
Metal deposition experiments were performed with different
concentrations of EDTA to quantitatively determine the
influence of EDTA in the dynamics of metal-ion leaching/
deposition. The metal impurity concentration in our electro-
lytes is below our limits of detection (0.2 μmol/cm2 for XRF).
Therefore, 0.1 mM Fe(NO3)3, Ni(NO3)2, Cu(NO3)2, and
Zn(NO3)2 were purposely added to a DI electrolyte to
artificially induce much greater levels of inhibitive impurities
compared with the electrolytes used thus far. These impurities
are known to immediately deactivate electrocatalytic copper
surfaces under applied bias (Figure S10).12,27 The deposition
of impurity metals (e.g., Fe, Ni, Zn) on an electrode surface is
blocked in electrolyte containing different concentrations of
EDTA (0, 1, 10, 100, 1000, or 10 000 mg/L) as shown in
Figure 2. An electrodeposition potential of −1.3 V vs RHE was
applied, and XRF was used to monitor the quantity of metal
deposited onto the electrode surface (Figure 2a).38,39 All four
metals were readily deposited on the electrode when EDTA
was absent from the electrolyte. The deposition of Zn is
hindered when the EDTA concentration is increased to 10
mg/L. No Zn is deposited, and the deposition of Fe and Ni is
significantly reduced with an EDTA concentration of 100 mg/
L. Only copper is deposited when the EDTA concentration is
1000 mg/L or greater. This is reasonable, because in EDTA-
One significant effect of adding >1000 mg/L EDTA to the
electrolyte is a change in selectivity of the copper catalyst. The
copper catalyst operating in ultrapure electrolyte produces
similar amounts of C2H4 and CH4, while the CH4/C2H4
product ratio obtained in MQEDTA, DIEDTA, and TAPEDTA
electrolyte containing 1000 mg/L EDTA was increased to 5.
We rationalize that this change in selectivity was due to the
dissolution of copper atoms from the electrode to the
electrolyte (vide infra). It was found that the addition of
0.025 mM CuCl2 to the electrolyte solution containing EDTA
was sufficient to prevent the selectivity change of copper
electrolysis. The product selectivity for electrochemical CO2
reduction under these conditions (MQEDTA + Cu2+, DIEDTA
+
Cu2+ and TAPEDTA + Cu2+) is similar to reports of catalysts
working in ultrapure electrolytes and is consistent over 12 h
(Figure 1j−l).
As listed in Table S2, the concentrations of Fe, Ni, and Zn
impurities are 40−800 nM in TAP water and 0.05−3.7 nM in
MQ water. The EDTA concentration of 1000 mg/L (3.4 mM)
is 3 magnitudes higher than impurities in TAP water, while 1
mg/L (3.4 μM) EDTA is 3 magnitudes higher than impurities
in MQ water. The stability of copper is nonetheless
dramatically different in the two environments. It is notable
that, in Surendanth’s report, 1 mg/L of EDTA in the
electrolyte sufficiently binds metal-ion impurities in the
electrolyte but does not improve the lifetime of copper
catalysts for CO2 reduction.27 We have determined that
preventing metal ions from depositing onto the catalysts does
not account for the increased catalyst stability in our EDTA
electrolytes. Electrodeposition at −1.3 V was performed on a
copper foil (99.99%) electrode for 12 h in untreated TAP and
TAPEDTA electrolytes. The electrode was tested postelectrolysis
for surface contamination by X-ray photoelectron spectroscopy
(XPS; Figure S8). The resulting spectrum for the TAP
experiment showed clear peaks corresponding to Ni and Zn on
the electrode surface. These peaks are absent from the copper
D
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. (a) Relative concentration of Cu at the electrode surface (determined by normalized XRF) and (b) the concentration of Cu2+-EDTA
complex in the solution (determined by UV−vis spectroscopy) as measured over time during CO2 reduction electrocatalysis at −1.3 V vs RHE in
DIEDTA electrolyte with different concentrations of EDTA. (c) The relative concentration of Cu at the electrode surface and (d) the concentration
of Cu2+-EDTA complex in the solution as measured over time during CO2 reduction electrocatalysis at −0.9 V vs RHE in DIEDTA electrolyte with
different concentrations of EDTA.
containing electrolyte, the electrodeposition potentials of all
metal ions are negatively shifted with lower free metal-ion
concentrations. These experiments were repeated at an
electrodeposition potential of −0.9 V (the lower voltage
limit at which hydrocarbon production from CO2RR is
observed),40 with similar deposition blocking order of Zn >
Fe > Ni > Cu with increasing EDTA concentration. It is worth
noting that the deposition of transition metal ions still occurs
with a low concentration of EDTA (100 mg/L) at more
positive operating voltage.
where E° is the standard electrode potential of Mn+/M, R is the
universal gas constant, T is the temperature, F is the Faraday
constant, Z is the number of electrons transferred in the half-
reaction, and β is the formation constant. The electro-
deposition potential of metal ions is related to the free
metal-ion concentration and the standard electrode potential.
In a standard CO2RR experiment without EDTA, all metal
ions have more positive electrode potentials than CO2RR,
thereby resulting in deposition of all metal impurities on the
surface of the electrode. The EDTA binding ratio with the
transition metal ions studied in our research is almost 100%;
thus, the concentration of free metal ions falls as EDTA is
added to negatively shift the M/Mn+ electrode potential. This
effect is not apparent when the EDTA concentration is lower
than the total concentration of metal ions in the electrolyte.
Indeed, Figure 2a,b shows that electrodeposition occurs in
electrolytes containing 0−10 mg/L EDTA. When the EDTA
concentration surpasses the initial concentration of metal ions,
The leaching and redeposition equilibrium of metal ions can
be understood with the two equations listed below:
RT
ZF
n+
)
E = E° +
log(CM
(1)
(2)
[EDTA − M]
β =
n−
[EDTA ][Mn+]
E
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
the concentration of free metal ions would be decreased by
magnitudes. On the basis of the rough calculation of a 1:1
EDTA to metal ions ratio in eqs 1 and 2, the shifted electrode
potential would still follow the same order of Zn, Fe, Ni, Cu,
from negative to positive, which is consistent with the blocking
order observed in Figure 2. When the EDTA concentration is
increased to one magnitude higher than the metal ions in
electrolyte (in our case 1000 mg/L EDTA vs 0.1 mM Fe3+,
Ni2+, Zn2+), the metals are no longer deposited at the surface.
This can be rationalized as the electrode potential of M/Mn+
being negatively shifted by >0.5 V in accordance with eqs 1
and 2. The applied potential will also influence the equilibrium
of M/Mn+. In the electrodeposition experiment, Fe and Ni
deposition at the surface of the electrode is effectively
suppressed at −0.9 V with 100 mg/L EDTA, while at −1.3
V the ions can still be quickly deposited on the electrode. The
voltage range that can prohibit the deposition of metal is not
exactly the same with the prediction from eqs 1 and 2, because
CO2RR and HER act as competing reactions and the
resistance is not compensated in our system.
Figure 4. SEM micrographs of Cu electrocatalysts on carbon cloth:
(a) as-prepared, (b) after operating in DI electrolyte, (c) after
operating in DIEDTA electrolyte, and (d) after operating in DIEDTA
with additional 0.025 mM CuCl2. SEM images were collected after 12
h operating at −1.3 V vs RHE.
Copper is a unique metal species in this environment in that
it cannot be fully leached nor is completely insoluble with high
concentrations of EDTA present in the electrolyte. Instead, it
forms a Cu/Cu2+/EDTA-Cu equilibrium, which we consider
to be crucial for the regeneration of catalyst. The effect of
EDTA concentration on this Cu/Cu2+/EDTA-Cu equilibrium
was experimentally established in a DI electrolyte to confirm
the theoretical prediction by monitoring relative copper levels
on the electrode and in the electrolyte by XRF and UV−vis
spectroscopy (Figures S11 and S12), respectively, under
CO2RR conditions (applied bias of −1.3 V vs RHE) for 5 h
(Figure 3a,b). The XRF signal was stable during electrolysis in
electrolytes containing 100 mg/L EDTA or less, indicating that
no copper was lost from the electrode surface to the
electrolyte. The UV−vis spectra also showed no detectable
Cu2+-EDTA in the electrolyte at these concentrations of
EDTA. A detectable dissolution of copper occurs when the
concentration of EDTA is 1000 mg/L or greater. Increasing
EDTA concentrations to 10 000 mg/L results in a ∼40% loss
of copper from the electrode surface after 2 h, as indicated by
complementary signals in the XRF and the UV−vis spectra.
These signals stabilize after 3 h inferring the establishment of
an equilibrium between copper on the electrode surface and
copper(II) in the electrolyte. To further support this copper
dissolution and redeposition process, we performed electrolysis
at −1.3 V vs RHE for 6 h at an electrode where a blank carbon
cloth electrode (1 cm2) was placed adjacent to the working
copper/carbon cloth electrode (1 cm2). This experiment led to
the dissolution of copper from the working electrode and a
subsequent redeposition on the carbon cloth electrode (see
SEM images in Figure S13). An equilibrium is also reached
within ∼3 h at lower applied potentials (−0.9 V); however, the
position of equilibrium is shifted toward the electrolyte at this
less negative voltage (Figure 3c,d). This is also consistent with
our prediction based on eqs 1 and 2.
clusters with branch-like assemblies consistent with dendritic
electrodeposited copper (Figure 4a). The smooth surface
morphology and thick branches of the catalyst did not change
significantly after operating for 12 h in DI electrolyte (Figure
4b). By contrast, electrolysis in DIEDTA electrolyte resulted in
noticeable corrosion of both trunk and branches of the
dendritic structures (Figure 4c). Also apparent were small
additional clusters on the substrate indicating simultaneous
copper redeposition and dissolution (Figure S14). This
observation is consistent with our XRF data which shows the
copper signal decreasing for the first 2 h by 20% and then
remaining constant. This change in morphology of the copper
catalyst surface also correlates to a difference in CO2 reduction
product selectivity when using either DI or DIEDTA electrolyte
(Figure 1b,h). After 12 h of electrolysis, the copper catalyst in
DI electrolyte produced CH4 and C2H4 with FEs of 30% and
25%, respectively, whereas the catalyst in DIEDTA electrolyte
produced CH4 and C2H4 with FEs of 60% and >10%,
respectively. The addition of 0.025 mM CuCl2 to the
electrolyte solution containing EDTA was sufficient to shift
the equilibrium position toward the electrode surface such that
the copper loading on the electrocatalyst surface remained
constant from the outset of electrolysis (Figures S15 and S16).
Furthermore, the additional copper ions added to the
electrolyte prior to electrolysis had the effect of maintaining
a more consistent catalyst morphology (Figure 4d). The
product selectivity for electrochemical CO2 reduction under
these conditions (DIEDTA + Cu2+ and TAPEDTA + Cu2+) is
similar to reports of catalysts working in ultrapure electrolytes
and is consistent over 12 h (Figure 1k,l).
A threshold EDTA concentration of ∼1000 mg/L is
required to achieve long-term stability for Cu catalyst, as
indicated by the huge stability difference in low- and high-
concentration EDTA electrolytes regardless of the impurity
level (Figure 1). A very similar threshold EDTA concentration
of 100−1000 mg/L is observed by the introduction of a
leaching−redeposition process of copper, as shown in Figures
2−4. These results suggest that this newly established Cu/Cu2+
equilibrium created by high concentrations of EDTA is crucial
for the long-term stability of the Cu catalyst. For low EDTA
concentrations (1 mg/L) that cannot shift the equilibrium,
there is little mass exchange between Cu2+ in electrolyte and
The dissolution and redeposition of copper on electrode
introduced by EDTA in the electrolyte can also be confirmed
by the morphological changes upon the copper catalyst surface
in scanning electron microscopy (SEM) images. We sought to
qualify any surface changes caused by EDTA by examining a
typical copper electrocatalyst by SEM before and after 12 h
CO2 electrolysis in either DI or DIEDTA (Figure 4a−c). The as-
prepared copper on carbon cloth sample exhibited large
F
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Cu metal on electrode; therefore, no surface structure
regeneration occurs. For high EDTA concentrations (1 g/L),
a large amount of copper first dissolves into the electrolyte and
then forms a dissolution/deposition equilibrium to regenerate
the surface active sites. We can estimate from these results that
∼0.3% of total copper is leached from the catalyst surface per
minute at the onset of electrolysis at −1.3 V with 1000 mg/L
EDTA in the electrolyte. Assuming the dissolution kinetics do
not change significantly at equilibrium, this estimate suggests
that 0.3% of the copper catalyst is refreshed every minute
during electrocatalysis. This rapid mass exchange of copper
between the electrode and electrolyte allows for the
maintenance of a poison-free catalyst surface.
REFERENCES
■
́
(1) Costentin, C.; Robert, M.; Saveant, J.-M. Catalysis of the
electrochemical reduction of carbon dioxide. Chem. Soc. Rev. 2013, 42,
2423−2436.
(2) Qiao, J.; Liu, Y.; Hong, F.; Zhang, J. A review of catalysts for the
electroreduction of carbon dioxide to produce low-carbon fuels.
Chem. Soc. Rev. 2014, 43, 631−675.
(3) Hori, Y. Electrochemical CO2 reduction on metal electrodes. In
Modern Aspects of Electrochemistry;Springer: New York, 2008; pp 89−
189.
(4) Weng, Z.; Zhang, X.; Wu, Y.; Huo, S.; Jiang, J.; Liu, W.; He, G.;
Liang, Y.; Wang, H. Self-cleaning catalyst electrodes for stabilized
CO2 reduction to hydrocarbons. Angew. Chem., Int. Ed. 2017, 56,
13135−13139.
(5) Asadi, M.; Kim, K.; Liu, C.; Addepalli, A. V.; Abbasi, P.; Yasaei,
P.; Phillips, P.; Behranginia, A.; Cerrato, J. M.; Haasch, R.; et al.
Nanostructured transition metal dichalcogenide electrocatalysts for
CO2 reduction in ionic liquid. Science 2016, 353, 467−470.
(6) He, J.; Dettelbach, K. E.; Salvatore, D. A.; Li, T.; Berlinguette, C.
P. High-throughput synthesis of mixed-metal electrocatalysts for CO2
reduction. Angew. Chem., Int. Ed. 2017, 56, 6068−6072.
(7) Chen, Y.; Li, C. W.; Kanan, M. W. Aqueous CO2 reduction at
very low overpotential on oxide-derived Au nanoparticles. J. Am.
Chem. Soc. 2012, 134, 19969−19972.
CONCLUSIONS
■
This study presents an electrocatalyst regeneration strategy
that prolongs the lifetime of copper on carbon cloth
electrocatalysts toward CO2 electroreduction to hydrocarbon
products while operating in non-ultrapure electrolytes (e.g.,
distilled or tap water with reagent grade bicarbonate salt). The
addition of 1000 mg/L EDTA into the electrolyte solution has
the effect of scavenging impurities and inhibiting catalyst
blocking from byproducts while constantly regenerating the
copper catalyst surface (as demonstrated by XRF and UV−vis
data). This effect can primarily be attributed to the differences
in reduction potential across a series of transition metal ions
(Cu, Ni, Fe, Zn) that are known to bind EDTA. This work
demonstrates a tangible strategy to repair electrocatalysts by
manipulating the equilibrium of metal/metal ions with a
common chelation agent.
(8) Roberts, F. S.; Kuhl, K. P.; Nilsson, A. High selectivity for
ethylene from carbon dioxide reduction over copper nanocube
electrocatalysts. Angew. Chem., Int. Ed. 2015, 54, 5179−5182.
́
(9) Ma, M.; Trzesniewski, B. J.; Xie, J.; Smith, W. A. Selective and
efficient reduction of carbon dioxide to carbon monoxide on oxide-
derived nanostructured silver electrocatalysts. Angew. Chem., Int. Ed.
2016, 55, 9748−9752.
(10) Clark, E. L.; Hahn, C.; Jaramillo, T. F.; Bell, A. T.
Electrochemical CO2 reduction over compressively strained CuAg
surface alloys with enhanced multi-carbon oxygenate selectivity. J. Am.
Chem. Soc. 2017, 139, 15848−15857.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.8b02311.
■
S
(11) He, J.; Johnson, N. J. J.; Huang, A.; Berlinguette, C. P.
Electrocatalytic alloys for CO2 reduction. ChemSusChem 2018, 11,
48−57.
(12) Hori, Y.; Konishi, H.; Futamura, T.; Murata, A.; Koga, O.;
Sakurai, H.; Oguma, K. Deactivation of copper electrode” in
electrochemical reduction of CO2. Electrochim. Acta 2005, 50,
5354−5369.
XRD patterns, SEM images, NMR spectra, and CO2
electrochemical reduction performances (PDF)
(13) Xie, J.-F.; Huang, Y.-X.; Li, W.-W.; Song, X.-N.; Xiong, L.; Yu,
H.-Q. Efficient electrochemical CO2 reduction on a unique
chrysanthemum-like Cu nanoflower electrode and direct observation
of carbon deposite. Electrochim. Acta 2014, 139, 137−144.
(14) Jermann, B.; Augustynski, J. Long-term activation of the copper
cathode in the course of CO2 reduction. Electrochim. Acta 1994, 39,
1891−1896.
(15) Najafpour, M. M.; Fekete, M.; Sedigh, D. J.; Aro, E.-M.;
Carpentier, R.; Eaton-Rye, J. J.; Nishihara, H.; Shen, J.-R.;
Allakhverdiev, S. I.; Spiccia, L. Damage management in water-
oxidizing catalysts: from photosystem II to nanosized metal oxides.
ACS Catal. 2015, 5, 1499−1512.
(16) Kanan, M. W.; Nocera, D. G. In situ formation of an oxygen-
evolving catalyst in neutral water containing phosphate and Co2+.
Science 2008, 321, 1072−1075.
(17) Huynh, M.; Bediako, D. K.; Nocera, D. G. A functionally stable
manganese oxide oxygen evolution catalyst in acid. J. Am. Chem. Soc.
2014, 136, 6002−6010.
(18) Gong, H.; Cao, X.; Li, F.; Gong, Y.; Gu, L.; Mendes, R. G.;
Rummeli, M. H.; Strasser, P.; Yang, R. PdAuCu nanobranch as self-
repairing electrocatalyst for oxygen reduction reaction. ChemSusChem
2017, 10, 1469−1474.
(19) Liu, J.; Zhang, Y.; Lu, L.; Wu, G.; Chen, W. Self-regenerated
solar-driven photocatalytic water-splitting by urea derived graphitic
carbon nitride with platinum nanoparticles. Chem. Commun. 2012, 48,
8826−8828.
AUTHOR INFORMATION
Corresponding Author
*E-mail: cberling@chem.ubc.ca.
ORCID
Jingfu He: 0000-0002-7639-0151
Aoxue Huang: 0000-0003-2507-0198
Noah J. J. Johnson: 0000-0002-0721-3186
Kevan E. Dettelbach: 0000-0002-7789-2129
Yang Cao: 0000-0001-8636-7919
Curtis P. Berlinguette: 0000-0001-6875-849X
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors are grateful to the Canadian Natural Science and
Engineering Research Council (RGPIN 337345-13), Canadian
Foundation for Innovation (229288), Canadian Institute for
Advanced Research (BSE-BERL-162173), and Canada Re-
search Chairs for financial support. K.E.D. was supported by an
NSERC PGS-D scholarship. A.H. was supported by the
University of British Columbia with a Four Year Doctoral
Fellowship (4YF).
G
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
̌
(20) Dinca, M.; Surendranath, Y.; Nocera, D. G. Nickel-borate
dioxide into hydrocarbon fuels. Energy Environ. Sci. 2010, 3, 1311−
1315.
oxygen-evolving catalyst that functions under benign conditions. Proc.
Natl. Acad. Sci. U. S. A. 2010, 107, 10337−10341.
(21) Lutterman, D. A.; Surendranath, Y.; Nocera, D. G. A self-
healing oxygen-evolving catalyst. J. Am. Chem. Soc. 2009, 131, 3838−
3839.
(22) Cobo, S.; Heidkamp, J.; Jacques, P.-A.; Fize, J.; Fourmond, V.;
Guetaz, L.; Jousselme, B.; Ivanova, V.; Dau, H.; Palacin, S.; et al. A
janus cobalt-based catalytic material for electro-splitting of water. Nat.
Mater. 2012, 11, 802−807.
(23) Barroso, M.; Pendlebury, S. R.; Cowan, A. J.; Durrant, J. R.
Charge carrier trapping, recombination and transfer in hematite (α-
Fe2O3) water splitting photoanodes. Chem. Sci. 2013, 4, 2724−2734.
(24) Klahr, B.; Gimenez, S.; Fabregat-Santiago, F.; Bisquert, J.;
Hamann, T. W. Photoelectrochemical and impedance spectroscopic
investigation of water oxidation with “Co–Pi”-coated hematite
electrodes. J. Am. Chem. Soc. 2012, 134, 16693−16700.
(25) Kwon, G.; Kokhan, O.; Han, A.; Chapman, K. W.; Chupas, P.
J.; Du, P.; Tiede, D. M. Oxyanion induced variations in domain
structure for amorphous cobalt oxide oxygen evolving catalysts,
resolved by X-ray pair distribution function analysis. Acta Crystallogr.,
Sect. B: Struct. Sci., Cryst. Eng. Mater. 2015, 71, 713−721.
(26) Nocera, D. G. Chemistry of Personalized Solar Energy. Inorg.
Chem. 2009, 48, 10001−10017.
(27) Wuttig, A.; Surendranath, Y. Impurity ion complexation
enhances carbon dioxide reduction catalysis. ACS Catal. 2015, 5,
4479−4484.
(28) Jeon, H.; Kunze, S.; Scholten, F.; Cuenya, B. Prism-Shaped Cu
Nanocatalysts for Electrochemical CO2 Reduction to Ethylene. ACS
Catal. 2018, 8, 531−535.
́
(29) Schreier, M.; Heroguel, F.; Steier, L.; Ahmad, S.; Luterbacher,
J.; Mayer, M.; Luo, J.; Gratzel, M. Solar conversion of CO2 to CO
using Earth-abundant electrocatalysts prepared by atomic layer
modification of CuO. Nat. Energy 2017, 2, 17087.
(30) Li, C. W.; Kanan, M. W. CO2 reduction at low overpotential on
Cu electrodes resulting from the reduction of thick Cu2O films. J. Am.
Chem. Soc. 2012, 134, 7231−7234.
(31) Tarasov, K. A.; O’Hare, D.; Isupov, V. P. Solid-State Chelation
of Metal Ions by Ethylenediaminetetraacetate Intercalated in a
Layered Double Hydroxide. Inorg. Chem. 2003, 42, 1919−1927.
(32) Kuhl, K. P.; Cave, E. R.; Abram, D. N.; Jaramillo, T. F. New
insights into the electrochemical reduction of carbon dioxide on
metallic copper surfaces. Energy Environ. Sci. 2012, 5, 7050−7059.
(33) Kuhl, K. P.; Hatsukade, T.; Cave, E. R.; Abram, D. N.;
Kibsgaard, J.; Jaramillo, T. F. Electrocatalytic conversion of carbon
dioxide to methane and methanol on transition metal surfaces. J. Am.
Chem. Soc. 2014, 136, 14107−14113.
(34) Reller, C.; Krause, R.; Volkova, E.; Schmid, B.; Neubauer, S.;
Rucki, A.; Schuster, M.; Schmid, G. Selective electroreduction of CO2
toward ethylene on nano dendritic copper catalysts at high current
density. Adv. Energy Mater. 2017, 7, 1602114.
(35) Dutta, A.; Rahaman, M.; Mohos, M.; Zanetti, A.; Broekmann,
P. Electrochemical CO2 conversion using skeleton (Sponge) type of
Cu catalysts. ACS Catal. 2017, 7, 5431−5437.
(36) Sen, S.; Liu, D.; Palmore, G. T. R. Electrochemical reduction of
CO2 at copper nanofoams. ACS Catal. 2014, 4, 3091−3095.
(37) Zhu, S.; Jiang, B.; Cai, W.-B.; Shao, M. Direct observation on
reaction intermediates and the role of bicarbonate anions in CO2
electrochemical reduction reaction on Cu surfaces. J. Am. Chem. Soc.
2017, 139, 15664−15667.
(38) Dettelbach, K. E.; Kolbeck, M.; Huang, A.; He, J.; Berlinguette,
C. P. Rapid quantification of film thickness and metal loading for
electrocatalytic metal oxide films. Chem. Mater. 2017, 29, 7272−7277.
(39) Speck, F. D.; Dettelbach, K. E.; Sherbo, R. S.; Salvatore, D. A.;
Huang, A.; Berlinguette, C. P. On the electrolytic stability of iron-
nickel oxides. Chem. 2017, 2, 590−597.
(40) Peterson, A. A.; Abild-Pedersen, F.; Studt, F.; Rossmeisl, J.;
Nørskov, J. K. How copper catalyzes the electroreduction of carbon
H
DOI: 10.1021/acs.inorgchem.8b02311
Inorg. Chem. XXXX, XXX, XXX−XXX