Understanding three-dimensionally interconnected porous oxide-derived copper electrocatalyst for selective carbon dioxide reduction

Article abstract of DOI:10.1039/c9ta10135g

In this work, we have investigated a hierarchical CuO-derived inverse opal (CuO-IO) catalyst with high CO selectivity up to 80-90% and minimal H₂ evolution at moderate potentials for CO₂ electroreduction. The three-dimensionally (3D) structured, porous catalyst was composed of small CuO nanoparticles and exhibited a peak CO faradaic efficiency (FE) of 72.5% (±1.8), complete suppression of H₂ formation, and good stability over 24 hours operation at -0.6 V versus the reversible hydrogen electrode (RHE). In situ Raman, X-ray absorption spectroscopy and X-ray diffraction measurements indicated reduction of the catalyst into metallic Cu⁰ oxidation state with dominant Cu(111) orientation under electrocatalytic conditions. We suggest that rapid depletion of CO₂ and protons at the highly roughened catalyst surface likely increased the local pH during the electrolysis. The combination of C₁ favoring Cu(111) surfaces and reduced local proton/CO₂ availability facilitated selective conversion of CO₂ into CO and reduced H₂ and C₂ products. Our work provides additional understanding of the structure-property relationships of 3D porous electrocatalysts for CO₂ reduction applications by evaluating the crystallographic orientation, oxidation state, and crystallite size of a CO-selective CuO-IO catalyst under realistic working conditions.

Full text of DOI:10.1039/c9ta10135g

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Understanding three-dimensionally
interconnected porous oxide-derived copper
electrocatalyst for selective carbon dioxide
reduction†
Cite this: J. Mater. Chem. A, 2019, 7,
7576
2
Thuy-Duong Nguyen-Phan, *ab Congjun Wang, Chris M. Marin, Yunyun Zhou,
ab
ab
a
c
a
a
d
a
Eli Stavitski, Eric J. Popczun, Yang Yu, Wenqian Xu, Bret H. Howard,
ab
c
a
Mengling Y. Stuckman, Iradwikanari Waluyo,
Paul R. Ohodnicki Jr
a
and Douglas R. Kauffman
*
In this work, we have investigated a hierarchical CuO-derived inverse opal (CuO-IO) catalyst with high CO
selectivity up to 80–90% and minimal H evolution at moderate potentials for CO electroreduction. The
three-dimensionally (3D) structured, porous catalyst was composed of small CuO nanoparticles and
formation, and
2
2
exhibited a peak CO faradaic efficiency (FE) of 72.5% (ꢀ1.8), complete suppression of H
2
good stability over 24 hours operation at ꢁ0.6 V versus the reversible hydrogen electrode (RHE). In situ
Raman, X-ray absorption spectroscopy and X-ray diffraction measurements indicated reduction of the
0
catalyst into metallic Cu oxidation state with dominant Cu(111) orientation under electrocatalytic
conditions. We suggest that rapid depletion of CO
likely increased the local pH during the electrolysis. The combination of C
reduced local proton/CO availability facilitated selective conversion of CO
and C products. Our work provides additional understanding of the structure–property relationships of
D porous electrocatalysts for CO reduction applications by evaluating the crystallographic orientation,
2
and protons at the highly roughened catalyst surface
favoring Cu(111) surfaces and
into CO and reduced H
1
Received 13th September 2019
Accepted 22nd November 2019
2
2
2
2
DOI: 10.1039/c9ta10135g
3
2
rsc.li/materials-a
oxidation state, and crystallite size of a CO-selective CuO-IO catalyst under realistic working conditions.
requires signicant overpotentials and can suffer from poor
product selectivity and competitive hydrogen evolution reaction
Introduction
Electrochemical CO
approach to convert CO
2
reduction (EC-CO
emissions into industrially-relevant CO
2
RR) is a promising (HER). The development of highly active, selective and robust
conversion catalysts is of vital interest to overcome these
and value-added chemicals and fuels. However, due to slow drawbacks. The activity and selectivity towards specic products
2
2
1–7
kinetics and multi-electron transfer pathway, EC-CO RR usually strongly depend on electrocatalyst morphology, surface rough-
2
ness, nature of electrochemically active sites, electronic
a
conguration, transport limitations, local pH environment at
the electrode surface, etc.
National Energy Technology Laboratory, 626 Cochrans Mill Road, P.O. Box 10940,
2
–18
Pittsburgh, PA 15236-0940, USA. E-mail: ThuyDuong.NguyenPhan@netl.doe.gov;
Douglas.Kauffman@netl.doe.gov
To date, numerous materials have been studied, including
b
Leidos Research Support Team, 626 Cochrans Mill Road, P.O. Box 10940, Pittsburgh, metals, oxides, and carbonaceous composites.1–10 Expensive
PA 15236-0940, USA
metals such as gold and silver can selectively convert CO
CO,
2
into
which is a commodity chemical used in a variety of
processes, including methanol and Fischer–Tropsch
c
Photon Sciences Division, National Synchrotron Light Source II, Brookhaven National
4,6,19,20
Laboratory, Upton, New York 11973, USA
d
X-ray Science Division, Advanced Photon Source, Argonne National Laboratory,
21,22
synthesis.
Copper-derived materials have also attracted
Argonne, Illinois 60439, USA
much attention due to their low cost, high abundance, and
†
Electronic supplementary information (ESI) available: Includes detailed
experimental methods; materials characterizations; calculation of faradaic ability to produce hydrocarbons or oxygenated hydrocar-
efficiency; determination of electrochemical surface area and roughness factor;
setup of in situ experiments; ex situ SEM, HR-TEM, EDX, XPS, SXRD, Raman,
so XAS data; chronoamperometric proles, faradaic efficiencies, selectivities,
total current densities, CO partial current densities, loading dependence
performance; post-reaction XPS data; Tafel plots; control experiments; double
layer capacitance plot; pore size comparison; in situ Raman, XANES, EXAFS, nanowires, prisms, dendrites, etc.
SXRD data; gas and ion chromatography spectra; and ve ESI tables. See DOI:
2,8–18,23–29
bons,
and efforts have recently focused on structural
control to improve their product selectivity. A number of
different structures and dimensions of copper-based catalysts
have been investigated, such as nanoparticles, nanofoams,
9
–18,23–30
CuO-derived hierar-
chical nanostructures composed of nanowires exhibited
1
0.1039/c9ta10135g
27576 | J. Mater. Chem. A, 2019, 7, 27576–27584
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
Journal of Materials Chemistry A
selective CO and HCOOH production with a total FE of 82.5% at opal materials were prepared by inltration of copper precursor
0.55 V vs. RHE that was attributed to the 3D porous structure solution with poly (methyl methacrylate) (PMMA) opal lm (see
ꢁ
29
of catalysts. Mesoporous Cu O-derived foams were also found ESI† for synthesis detail of PMMA latex and opal template). 20
2
to selectively produce C H and C H with a maximum C FE mL of copper precursor solution including 0.625 g of copper(II)
2
4
2
6
2
reaching 55% at ꢁ0.9 V vs. RHE due to the presence of domi- nitrate trihydrate (Cu(NO
3 2 2
) $3H O), 0.375 g of citric acid mon-
nant (100) surface sites for C–C coupling and temporal trapping ohydrate (C
6
H
8
O
7
$H
2
O), and 10 mL of absolute ethanol
30
ꢃ
of gaseous intermediates inside the mesopores. Despite this (C
2
H
5
OH, 200 proof) was penetrated slowly into 10 -tilted
progress, it is still challenging to fully understand the nature of PMMA opal and naturally evaporated overnight. The inltrated
ꢃ
electrochemically active sites because the product selectivity of lm was subsequently annealed in air at 400 C with ramping
ꢃ
ꢁ1
copper catalysts strongly depends on their structure, rate of 1 C min for 4 h to completely remove PMMA and
morphology, and oxidation state.
10–15,18,24,31
reassemble hierarchical CuO inverse opal (namely CuO-IO) as
Inverse opal (IO) materials have been widely studied for a negative replica of bare PMMA opal.
applications in catalysis, photonics, photovoltaic devices,
31
energy conversion, and energy storage. The three-dimensional
3D) interconnected, highly porous structure of IOs are
arranged in hexagonal close packed framework, offering large Electrochemical CO
Electrochemical CO
2
reduction measurement
(
2
reduction experiments were carried out in
surface area and better adsorbability of reactant mole- a gas-tight, two-compartment H-cell separated by a Naon 117
2
6,27,31,32
cules.
Despite these benets, only few studies on IO proton exchange membrane. Each compartment was lled with
4,6,26,27,31
catalysts for EC-CO RR have been reported. Porous 50 mL of aqueous 0.1 M KHCO3 electrolyte (99.99%, Sigma-
mesostructured Au and Ag IO catalysts have shown improved Aldrich) and contained 100 mL headspace. Here ultra-pure
CO selectivity due to the generation of pH gradients that deionized water (DIW) with 18.3 MU cm resistivity (Barn-
2
ꢁ1
reduced proton availability at the catalyst surface and sup- stead EASYpure LF) was used in all electrochemical experiments
4
,6
26
pressed competitive H
found improved CO selectivity (ꢂ45%) for cube-like Cu-IO, but on the EC-CO
the oxidation state and crystallographic orientation during EC- ously purged with CO
2
evolution. Zhang and coworkers
to minimize the effect of any trace metal impurities from water
33,34
2
RR performance.
The catholyte was continu-
2
(99.999%, Butler gas) at a ow rate of 20
ꢁ
1
CO RR were not investigated. However, larger IO pore size mL min (pH ꢂ6.8) during the experiments and stirred at
2
signicantly decreases CO FE while enhancing H and C2 200 rpm. The counter and reference electrodes were Pt wire and
2
26,27
formation.
Substantial efforts have been devoted to characterizing Cu- was prepared by dispersing 4 mg of as-prepared CuO-IO (scra-
based electrocatalysts during EC-CO RR working condi- ped down from the glass substrates) in 200 mL of methanol and
however, in situ investigations of C selective 10 mL of Naon® 117 solution binder (Sigma-Aldrich, 5%).
Ag/AgCl (saturated NaCl, BASi®), respectively. The catalyst ink
2
8,12,15,17,18,31
tions;
1
Cu-IO-based catalysts are still needed to understand their Working electrodes were fabricated by drop-casting the
enhanced selectivity. Here, we investigate the performance of prepared ink onto PTFE-coated carbon paper gas diffusion layer
a hierarchical oxide-derived copper inverse opal (CuO-IO) cata- (Toray paper 060, Alfa Aesar). The as prepared CuO-IO loading
ꢁ
2
lyst that is constructed from ꢂ15 nm Cu-oxide nanoparticles on carbon paper was kept at 2.8 ꢀ 0.1 mg cm
(based on
geo
arranged in a 3D porous framework. The catalyst demonstrated geometric area) unless otherwise noted.
impressive CO selectivity and strong suppression of H
2
evolu-
CO2 reduction experiments were performed at ambient
tion over a wide potential window, achieving a peak FE of 72.5% temperature and pressure using a SP-300 potentiostat (BioLogic
(
ꢀ1.8) at ꢁ0.6 V vs. RHE, and demonstrating good 24 hour Science Instrument). All potentials were referenced against the
durability. In situ characterization techniques, including reversible hydrogen electrode (RHE) and the uncompensated
Raman, X-ray absorption spectroscopy (XAS), and synchrotron resistance was automatically corrected at 85% (iR-correction)
9
,11,12,14
X-ray diffraction (SXRD) under working conditions showed that using the instrument soware.
Typical working elec-
0
the catalyst reduced to metallic Cu with a dominant (111) trode resistances were 30–40 U. Short-term chronoampero-
surface under electrocatalytic conditions. We attribute the high metric experiments were conducted for 30 min at each applied
C
1
selectivity and suppressed HER to a combination of the potential sequentially between ꢁ0.2 V and ꢁ1.2 V vs. RHE.
8,9,11
dominant presence of a Cu(111) surface
and a high local pH Long-term chronoamperometric experiments were conducted
depleting the local concentration of protons available for H
2
for 24 hours at ꢁ0.6 V vs. RHE. The total and partial current
production. This work provides additional insight into the densities were normalized to the exposed geometric area. Each
catalytic-activity of copper-based IO catalysts, and further data point is an average of at least three independent experi-
demonstrates that catalyst morphology can function as a cata- ments on different fresh electrodes. The evolved gas products
lyst design principle for selective CO conversion.
were quantied by PerkinElmer Clarus 600GC equipped with
both FID and TCD detectors, using ShinCarbon ST 80/100
Column and He as a carrier gas. The GC was calibrated regu-
larly using a calibration mixture of gases with known compo-
sition. The liquid products in the catholytes were determined by
2
Experimental
Synthesis of as-prepared hierarchical CuO inverse opal
All chemicals were purchased from Sigma-Aldrich and used as Dionex ICS-5000+ ion chromatography using ED50 conducto-
received without further purication. Hierarchical CuO inverse metric detector, ASRS suppressor in auto-generation mode,
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A, 2019, 7, 27576–27584 | 27577

View Article Online
Journal of Materials Chemistry A
Paper
AS11-HC column and KOH eluent with a gradient of 0.4–30 mM samples into Kapton capillaries, and two-dimensional diffrac-
in 45 min run. The calculation of faradaic efficiency for each tion patterns were collected by a PerkinElmer amorphous
product is described in the ESI.†
silicon detector. The in situ SXRD cell and experimental condi-
tions were identical to those used for in situ XAS experiments,
and SXRD patterns were collected during the application of
cathodic potentials. The data acquisition was performed with
Materials characterizations
Scanning electron microscopy (SEM) imaging was performed on QXRD and the diffraction ring was integrated using GSAS-II
36
a FEI Quanta 600F microscope operated at 10–20 kV equipped package. The instrument parameters were calibrated using
with an energy-dispersive X-ray (EDX) detector. High-resolution a Si standard; the phase determination and compositional
transmission electron microscopy (HR-TEM) and EDX analysis fraction were subsequently determined by Rietveld renement
were carried out on a FEI Titan G2 80–300 kV operated at using GSAS-II.
accelerating voltage of 300 kV. The CuO-IO sample was scraped
down from the glass substrates, suspended in ethanol, drop-
casted onto a holey carbon support Au grid, and naturally
Results
dried in air.
In situ Raman spectroscopy during EC-CO
Materials characterizations
2
RR was performed
The electron microscopy images in Fig. 1A and S2–S4 (ESI†)
show a three-dimensional interconnected backbone of as-
prepared CuO-IO with an average cavity size of 180 ꢀ 5 nm.
The HR-TEM micrograph in Fig. 1B reveals the CuO-IO structure
is composed of 15–20 nm CuO nanoparticles with lattice spac-
ings of 0.272 nm, 0.252 nm, and 0.232 nm that can be indexed
to (110), (002), and (200) planes of polycrystalline CuO,
respectively.
on a Horiba LabRam HR-Evolution spectrometer with 785 nm
laser as excitation source, a 10ꢄ long working distance objec-
tive, and a custom-made Teon cell with a quartz window. 5 mL
of the catalyst ink, composed of CuO-IO, Naon and methanol,
was drop-cast onto a glassy carbon working electrode. A Pt wire
and Ag/AgCl were used as counter and reference electrodes, and
iR-correction was applied in all measurements. 5 mL of 0.1 M
KHCO
3 2
was continuously purged with CO during the
The synchrotron XRD pattern of CuO-IO in Fig. 1C displays
measurements and sequential Raman spectra were collected at
various constant applied potentials. Ex situ measurement of
fresh and post-reaction electrodes were recorded with 633 nm
laser source and 100ꢄ working distance objective.
several diffraction peaks representative of monoclinic CuO
˚
(
3
space group C2/c) with lattice constants a ¼ 4.7119 A, b ¼
˚
˚
.4350 A, c ¼ 5.1164 A. Neither cuprite Cu
2
O nor metallic copper
phases were found, and the mean CuO crystallite size of ca.
5 nm closely matched particle sizes determined from HR-TEM
Hard X-ray absorption spectroscopy (XAS) characterization
was performed at the 8-ID (ISS) beamline of the National
Synchrotron Light Source II (NSLS-II) at Brookhaven National
1
imaging. Cu K-edge XANES spectra (Fig. 1D) reveal that the line
shape, and the positions of pre-edge (1s / 3d transition, at
Laboratory. In situ Cu K-edge XAS experiments during CO
2
8977 eV), shakedown feature (1s / 4p transition, at 8986 eV)
electrolysis were carried out using a custom-made poly-
carbonate cell which consisted of three interconnected cham-
bers for working, Ag/AgCl reference, and Pt wire counter
electrodes. A Kapton window in the working electrode chamber
allowed the passage of X-rays (Fig. S1, ESI†). The cell was lled
and white line (at 8998 eV) for CuO-IO resemble the CuO stan-
2+
dard, indicating Cu oxidation state. Two maxima centered at
2
˚
1
.53 and 2.52 A in corresponding Fourier transformed k -
weighted EXAFS spectrum (Fig. S5, ESI†) are ascribed to Cu–O
bond in the nearest neighbor shell and Cu–Cu in the next near
with 7 mL of aqueous 0.1 M KHCO electrolyte and saturated
14,15
3
neighbor coordination, respectively.
Additional XPS, Auger
ꢁ
1
with CO gas using a constant ow rate of 10 mL min . The
2
and Cu L-edge XAS data in Fig. S6 (ESI†) also conrmed the
fabrication of the CuO-IO working electrode was identical to the
electrochemical measurement in H-cell, although the in situ cell
was an open design and not gas tight. In situ Cu K-edge XAS
spectra were collected using a Passivated Implanted Planar
presence of CuO.
Electrochemical performance
2
Silicon detector during the application of various cathodic The CO electroreduction activity of as-prepared CuO-IO was
potentials (vs. RHE, iR-corrected). Cu foil was used for energy evaluated using chronoamperometry between ꢁ0.2 V and
calibration. Reference samples, including bulk Cu O and CuO ꢁ1.2 V vs. RHE as shown in Fig. 2A. Impressive C selectivity
2
1
powders, were diluted with boron nitride and pressed into was demonstrated over an extremely wide potential range,
pellets for ex situ measurement. These materials were also with only minor C
collected for energy normalization (at 8979 eV) and model ꢁ1.2 V and trace ethane ꢁ1.2 V (<0.1% FE) (Table S1, ESI†).
tting for X-ray absorption near-edge structure (XANES) and Total geometric current densities of CuO-HIO catalyst were
extended X-ray absorption ne structure (EXAFS) data process- comparable to other oxide-derived copper electrocatalysts
ing. All Cu K-edge data were collected in uorescence modes (Fig. S8, ESI†), but the C product selectivity was much higher
2
H
4
(<5%) and H
2
between ꢁ0.9 V and

1
35
8–10,12–14,23,26,27
and subsequently analyzed using IFEFFIT freeware package.
than expected for copper-based catalysts.
Synchrotron X-ray diffraction (SXRD) measurements were Methane and formic acid were the dominant products below
˚
conducted at beamline 17-BM-B (l ¼ 0.24136 A, 51.4 keV) of the ꢁ0.4 V, while CO production increased to a maximum faradaic
Advanced Photon Source at Argonne National Laboratory. Ex efficiency (FE) of 72.5% at ꢁ0.6 V and maximum CO selectivity
situ SXRD analysis of powder samples was conducted by loading up to nearly 90% was observed between ꢁ0.7 V and ꢁ0.8 V
27578 | J. Mater. Chem. A, 2019, 7, 27576–27584
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 1 (A) SEM image, (B) HR-TEM micrograph, (C) SXRD pattern (orange pattern is for simulated CuO), and (D) XANES Cu K-edge of as prepared
CuO-IO. Orange, gray and blue spectra in (D) are bulk CuO, bulk Cu O and Cu foil standards.
2
(
Fig. S9, ESI†). We point out that some potentials had less than
Long-term CO2 electrolysis demonstrated consistent CO
1
00% FE, which may stem from some fraction of the current selectivity for the CuO-IO catalyst. As shown in Fig. 2B, an
going to copper oxide reduction and/or the production of average 67 ꢀ 2% CO FE was found over 24 hours at ꢁ0.6 V with
ꢁ
2
liquid products that were either below the detection limit or a stable current density of ca. 2.5 mA cm and no detectable H
not detectable by ion chromatography (Fig. S10, ESI†). evolution (Fig. S15, ESI†). Pt crossover through Naon
Nevertheless, we observed an average C
FE of 78 ꢀ 2% was membrane or deposition of other trace metal impurities from
2
1
3
3,34
observed between ꢁ0.2 to ꢁ1.1 V vs. RHE (Fig. S11, ESI†), water can impact catalyst activity and stability.
XPS analysis
which decreased to approximately 60% at ꢁ1.2 V vs. RHE due of post-reaction electrodes aer long-term runs at ꢁ0.6 V using
to the increased HER at large overpotentials. The CO yield Pt wire and graphite counter electrodes ruled out signicant
ꢁ1
ꢁ1
strongly increased from ꢂ30 mmolCO
g
catalyst
h
at ꢁ0.2 V to deposition of trace Pt, Zn, Pb or Fe elements onto the electrode
ꢁ1
ꢁ1
6
0–70 mmolCO
g
catalyst
h
at potentials more negative than surface (Fig. S16, ESI†). Post reaction electron microscopy in
Fig. S17 (ESI†) revealed the catalyst preserved its general inverse
ꢁ
1.0 V (Table S1, ESI†).
Control experiments with different CuO-IO catalyst loadings, opal structure. The sustained CO selectivity and current density
varied catalyst layer thickness, and a graphite counter electrode over 24 hour operation indicates the CuO-IO catalyst is a robust
produced similarly high CO selectivities (Fig. S12 and S13, CO -to-CO conversion catalyst.
2
ESI†), which rules out a strong loading-dependence or unin-
For comparison, we also tested the EC-CO RR performance
2
tentional contamination from the Pt counter electrode of commercially-available bulk CuO powder (ꢂ1–5 mm) and
37
impacting the measured product distribution. Finally, the non-IO, ꢂ50 nm diameter CuO nanoparticles (NPs) with similar
bare carbon paper demonstrated almost exclusive H produc- catalyst loading. The morphology, crystallographic orientation
2
tion with only trace CO and CH
Fig. S14, ESI†).
4
detected from ꢁ0.7 V to ꢁ1.2 V and oxidation state of these oxide materials were determined by
(
SEM, SXRD and XAS measurements (Tables S2, S3 and Fig. S18,
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A, 2019, 7, 27576–27584 | 27579

View Article Online
Journal of Materials Chemistry A
Paper
2 2 3
Fig. 2 (A) Potential-dependent faradaic efficiencies for CO reduction products over CuO-IO catalyst in CO saturated 0.1 M KHCO . (B) Long-
term electrocatalytic performance of CuO-IO catalyst at ꢁ0.6 V vs. RHE. Comparisons of (C) CO faradaic efficiency (FE) and (D) CO partial
current density (jCO) at various negative potentials for CuO-IO, ꢂ50 nm CuO NPs, and bulk CuO powder.
ESI†). The potential-dependent faradaic efficiencies for all voltammogram (CV) of CuO-IO in CO
2 3
saturated 0.1 M KHCO
products in Fig. S19 and S20 (ESI†) show that unlike CuO-IO, (Fig. 3A). The oxidation state of Cu-based catalysts during
14,15,18,24,28,39,40
these more traditionally structured CuO catalysts produced CO RR is still debated in the literature,
and we
2
mostly H (FE ꢂ 50–70%) with small yields of CO (FE < 20%) at conducted in situ XAS, Raman spectroscopy, and XRD experi-
2
moderate negative potentials and ꢂ20% FE of ethylene at high ments to monitor the oxidation state, surface structure, crys-
overpotentials. The product distribution obtained over these tallite size, and crystallographic orientation of CuO-IO under
catalysts is similar to those of many copper catalysts reported electrochemical potential control.
1
1,12,14,24,25,28
before.
CuO-IO demonstrated substantially higher FEs and selectivities potentials in CO
towards CO, and CO partial current density (jCO) compared with Fig. 3B shows the CuO-IO was in the Cu oxidation state under
As shown in Fig. 2C, D and Fig. S21 (ESI†), the
Cu K-edge XANES and EXAFS were collected at various
2
saturated 0.1 M KHCO . The EXAFS spectra in
3
2
+
ꢁ
1
the CuO NPs and bulk CuO catalysts. The 141 mV dec Tafel open circuit. A reduction of the Cu–O and Cu–Cu scattering
˚
˚
slope for CO production at CuO-IO was close to the 120 mV peaks of CuO at 1.53 A and 2.52 A, and the emergence of the rst
ꢁ
1
˚
dec expected for a rate determining step involving the initial Cu–Cu coordination shell in metallic Cu at 2.21 A indicate the
10,13,26,29,31
electron transfer to CO
2
(Fig. S22, ESI†).
Tafel slopes onset of Cu-oxide reduction at an applied potential of ꢁ0.2 V vs.
ꢁ1
14,15
for the CuO NPs and bulk CuO were 178 and 184 mV dec
respectively.
,
RHE.
These changes became more apparent with increas-
ingly cathodic potentials, and comparison with the bulk Cu foil
reference indicates near complete reduction beyond ꢁ0.6 V vs.
RHE. These potential-dependent spectroscopic changes are
consistent with the Cu-oxide reduction peak centered at
approximately ꢁ0.45 V vs. RHE in Fig. 3A, and they agree with
In situ measurements
38
The Pourbaix diagram for the Cu–H O system indicates CuO
2
38
the Cu–H O Pourbaix diagram and similar behavior recently
should reduce to metallic Cu under EC-CO
2
RR at potentials
2
40
reported by Velasco-V ´e lez et al. The associated Cu K-edge
more negative than ꢁ0.5 V, which is consistent with the cyclic
27580 | J. Mater. Chem. A, 2019, 7, 27576–27584
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
Journal of Materials Chemistry A
2
Fig. 3 (A) Cyclic voltammetry of fresh CuO-IO electrode in CO
2
saturated 0.1 M KHCO
3
. (B) Potential-dependent k -weighted R-space EXAFS
analysis (no phase correction) from ꢁ0.2 V to ꢁ1.2 V vs. RHE (collected at 30 min at each potential). (C) In situ Raman spectra for tracking surface
structure of CuO-IO during EC-CO
RR at ꢁ0.6 V (using 785 nm laser source). (D) SXRD patterns of CuO-IO electrode under open circuit and
steady state at ꢁ0.6 V vs. RHE (*indicates residual carbon paper features and Kapton window from background subtraction).
2
XANES spectra collected at different potentials and while being Cu
held at ꢁ0.6 V are presented in Fig. S23 and S24 (ESI†).
2
O were found during the reaction in the region of 100–
250 cm and 330–600 cm . Similar results were obtained at
XAS is a bulk technique, and some studies have suggested ꢁ0.8 V and ꢁ1.0 V vs. RHE (Fig. S26, ESI†). Metallic copper is
ꢁ
1
ꢁ1
the presence of residual surface or sub-surface oxides during Raman-inactive, and the immediate decrease in intensity and
2 2
EC-CO RR that could impact EC-CO RR activity and selec- subsequent disappearance under electrocatalytic potentials
8,14,15,18,24
0
tivity.
To address this question, we also employed in situ strongly indicate the reduction of CuO into Cu on the elec-
Raman spectroscopy as a more surface sensitive technique to trocatalyst surface. These results are also in good agreement
8
,12,27,39
probe the surface structure changes of CuO-IO during the with other previous reports of Cu-oxide reduction,
28
39
application of electrochemical potentials. Ex situ Raman spec- specically, those of Ren and Wang et al. who attributed the
trum in Fig. S25 (ESI†) shows three indicative A , B1g, and B2g disappearance of the A mode during the application of
modes for fresh CuO-IO electrode. As shown in Fig. 3C, we cathodic potentials to copper-oxide reduction and Cu forma-
monitored the wavenumber region associated with the tion. Raman spectrum collected once the catalyst returned to
g
g
0
ꢁ1
g 2
predominant A feature of CuO at 294 cm to track oxidation open circuit showed the presence of mixed Cu O and CuO
state during CO RR. The spectrum collected before reaction is species (Fig. S27, ESI†), which reects the reversibility of the Cu
2
consistent with previous reports of CuO samples containing redox process.
41
ꢂ10 nm grain sizes recorded with a 782 nm laser. This feature
gradually disappeared during the application of at ꢁ0.6 V vs. ꢁ0.6 V vs. RHE to further probe the crystallographic orientation
RHE in CO -purged KHCO , and no other peaks associated with and crystallite size during EC-CO RR (Fig. 3D). CuO was
Finally, in situ XRD were collected under open circuit and at
2
3
2
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A, 2019, 7, 27576–27584 | 27581

View Article Online
Journal of Materials Chemistry A
Paper
identied before reaction, which is consistent with both Cu K- roughened, porous CuO-IO surface allowed rapid consumption
+
28,42
edge XAS/EXAFS and Raman measurements. Under steady of both CO2 and H .
The observed current density likely
state operating conditions at ꢁ0.6 V vs. RHE, we found the increased the local pH sufficiently to reduce the number of
4
2,51
presence of Cu (111), (200), (220), (311), and (222) peaks indic- protons available for HER,
while the Cu(111) orientation
9
,23,30,42,43,45,47,49,50
ꢀ
ative of face-centered cubic Cu (space group Fm3m), and we did favored C
1
production over C
2
formation.
not observe any signicant signatures associated with oxides
under working conditions (Fig. S24 and S28, ESI†). The results increase the local pH at copper electrodes to 13 or higher.
In the high overpotential range, large current densities can
42,43
identify metallic copper with lattice constant a ¼ b ¼ c ¼ 3.6102 These basic conditions can activate additional C
2
forming
˚
A and mean crystallite size of 10–11 nm during EC-CO
2
RR, reaction pathways and deplete the concentration of available
42,44,51,52
which is consistent with pre-reaction TEM analysis. The peak CO molecules.
The effect of these two phenomena was
2
intensity ratio of Cu(111) to Cu(200) for an ideal polycrystalline observed beyond ꢁ0.8 V vs. RHE with increased H evolution
2
30
Cu surface was reported to be ca. 3.03. We have estimated that and the emergence of C product formation at the CuO-IO
2
the average relative intensity ratio of Cu(111) and Cu(200) peaks catalyst. While we did observe hydrocarbon formation at large
at during EC-CO
2
RR at ꢁ0.6 V vs. RHE is 3.58 (Fig. S28, ESI†), overpotentials, methane was substantially favored over
implying the dominance of closed-packed Cu(111) surface or ethylene. These results are largely consistent with the expected
preferential (111) orientation of the catalyst during the reaction.
Importantly, we observed that the crystallographic orientation
9,23,30,42,43,45,47,49,50
C
1
preference of Cu(111) facets.
Previous works have proposed C2+ forming reaction path-
48,49
and crystallite size of operating catalyst did not substantially ways on Cu(111).
C H formation can share the *COH
2 4
change during ve hours of measurement at ꢁ0.6 V vs. RHE pathway with CH4 production, go through a *CHO pathway
(
Fig. S28, ESI†). Similar to Raman measurements, we also followed by C–C coupling, and/or undergo *CO dimeriza-
12,15,16,23,27,42,43,48
observed re-oxidation of the catalyst once it was returned to tion.
2
Strong suppression of C products along
open circuit aer the application of cathodic potential (Fig. S29, with favorable CH
4
formation at potentials beyond ꢁ0.9 V vs.
ESI†). Correlating these various in situ measurements strongly RHE indicate that CuO-derived Cu-IO promotes either the CHO
0
implies that Cu with a preferred Cu(111) orientation is or COH pathway over CO dimerization. Our results are consis-
a dominant species present in the CuO-IO catalyst during EC- tent with previous ndings of preferential methane formation
CO RR.
on Cu(111) at more elevated overpotentials over C2
2
42,43,45,47,49,50
products.
Finally, we can compare our results with others to identify
Discussion
26,27
the impact of Cu-IO pore size.
Cu-IO catalysts prepared by
The oxide-derived Cu-IO catalysts exhibited some of the highest electrochemical deposition formed with larger diameter pores
CO -to-CO selectivity among several oxide-derived copper elec- (340–600 nm) and contained larger, more bulk-like particles
2
26,27
trocatalysts at low to moderate overpotentials (Table S4, and crystallites.
ESI†).
Fig. S31 (ESI†) shows that increasing pore
For example, both Kanan, Ren and their diameter from 180 nm (current study) to 600 nm decreased CO
coworkers reported that oxide-derived copper catalysts with selectivity while increased both H and C production. Cu-IOs
9,10,13,17,26–29,42
10
28
2
2
a relatively high surface roughness could achieve selective CO with larger crystallite size and pore diameter demonstrated
and/or HCOOH production at low overpotentials. Wang and product distributions similar to that of bulk CuO powder
9,17,29
coworkers
also achieved CO FEs around 60% between (Fig. S20, ESI†), suggesting both pore diameter and crystallite
ꢁ
0.3 V and ꢁ0.5 V vs. RHE using Cu O-derived copper nano- size may be a design principle for tuning catalyst selectivity.
2
wires with a surface containing small (<10 nm) crystallites, as
well as oxide-derived, 3D copper nanostructures.
Taken together, our results suggest the CuO-IO catalyst
selectively produced CO owing to the synergy between the
2
In the present study, the selective EC-CO RR performance morphology and structure of the electrocatalyst. The 3D inter-
and strong HER inhibition demonstrated by the CuO-IO catalyst connected porous structure created pH gradients that
may be attributed to both its 3D morphology and crystallo- decreased proton availability and suppressed HER, while the
graphic surface orientation. The CuO-IO catalyst is composed of dominant Cu(111) orientation favored selective C formation.
1
small nanoparticles arranged in a 3D interconnected porous Our work provides better understanding of the structure–
structure that offers a large surface-to-volume ratio. The property relationships of porous copper-based catalysts, and
2
measured ECSA (2.96 cm ) and RF (30.8) of CuO-IO were provides insight into the selective, precious-metal free reduc-
considerably larger than the ꢂ50 nm diameter CuO NPs and tion of CO
2
into CO at CuO-IO.
2
bulk CuO powder (0.37–0.55 cm and RF ¼ 5.3–7.8; Fig. S30 and
Table S5, ESI†). The preferential Cu(111) surface/orientation is
Conclusions
1
also expected to demonstrate higher C selectivity due to weaker
binding of *CO and *COOH intermediates, whereas Cu(100) In summary, our hierarchical CuO-derived IO catalyst has shown
facets have favored C2+ production owing to a lower energetic impressive CO selectivity across a wide potential range with
8,9,11,23,30,42–50
barrier for intermediate hydrogenation.
In the low and moderate overpotential range (ꢁ0.2 to ꢁ0.8 V efficiency (ꢂ72.5%), selectivity (ꢂ90%) and CO current density
vs. RHE), the CuO-IO catalyst produced exclusive C products compared with other bulk and nanoparticulate copper oxide
and almost no H evolution. In this potential range, the highly catalysts (FE < 20%). Our results have demonstrated the benet
minor H evolution. This performance is superior in faradaic
2
1
2
27582 | J. Mater. Chem. A, 2019, 7, 27576–27584
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
of 3D interconnected porous structure that promotes EC-CO
by creating local pH gradients within the catalyst pores that
deplete the local concentration of protons available for HER. In
addition, the high surface roughness reduced copper surface,
Journal of Materials Chemistry A
2
RR
2 Y. Terunuma, A. Saitoh and Y. Momose, J. Electroanal. Chem.,
1997, 434, 69–75.
3 F. Studt, I. Sharafutdinov, F. Abild-Pedersen, C. F. Elkjær,
J. S. Hummelshøj, S. Dahl, I. Chorkendorff and
J. K. Nørskov, Nat. Chem., 2014, 6, 320–324.
1
and Cu(111) oriented-surface facilitate a C reaction path. In this
regard, our work may provide more understanding on structure–
property relation of oxide-derived copper catalyst for CO
4 A. S. Hall, Y. Yoon, A. Wuttig and Y. Surendranath, J. Am.
Chem. Soc., 2015, 137, 14834–14837.
production from fossil fuel-generated CO
2
emission.
5 N. Sreekanth, M. A. Nazrulla, T. V. Vineesh, K. Sailaja and
K. L. Phani, Chem. Commun., 2015, 51, 16061–16064.
6
Y. Yoon, A. S. Hall and Y. Surendranath, Angew. Chem., Int.
Ed., 2016, 55, 15282–15286.
Disclaimer
This work was funded by the Department of Energy, National
Energy Technology Laboratory, an agency of the United States
Government, through a support contract with Leidos Research
Support Team (LRST). Neither the United States Government
nor any agency thereof, nor any of their employees, nor LRST,
nor any of their employees, makes any warranty, expressed or
implied, or assumes any legal liability or responsibility for the
accuracy, completeness, or usefulness of any information,
7 J.-H. Kim, H. Woo, J. Choi, H.-W. Jung and Y.-T. Kim, ACS
Catal., 2017, 7, 2101–2106.
8 L. Mandal, K. R. Yang, M. R. Motapothula, D. Ren,
P. Lobaccaro, A. Patra, M. Sherburne, V. S. Batista,
B. S. Yeo, J. W. Ager, J. Martin and T. Venkatesan, ACS
Appl. Mater. Interfaces, 2018, 10, 8574–8584.
9 D. Raciti, K. J. Livi and C. Wang, Nano Lett., 2015, 15, 6829–
6835.
apparatus, product, or process disclosed, or represents that its 10 C. W. Li and M. W. Kanan, J. Am. Chem. Soc., 2012, 134,
use would not infringe privately owned rights. Reference herein 7231–7234.
to any specic commercial product, process, or service by trade 11 R. Reske, H. Mistry, F. Behafarid, B. R. Cuenya and
name, trademark, manufacturer, or otherwise, does not neces- P. Strasser, J. Am. Chem. Soc., 2014, 136, 6978–6986.
sarily constitute or imply its endorsement, recommendation, or 12 D. Ren, Y. Deng, A. D. Handoko, C. S. Chen, S. Malkhandi
favoring by the United States Government or any agency thereof. and B. S. Yeo, ACS Catal., 2015, 5, 2814–2821.
The views and opinions of authors expressed herein do not 13 M. Ma, K. Djanashvili and W. A. Smith, Phys. Chem. Chem.
necessarily state or reect those of the United States Govern-
Phys., 2015, 17, 20861–20867.
ment or any agency thereof.
14 H. Mistry, A. S. Varela, C. S. Bonifacio, I. Zegkinoglou,
I. Sinev, Y.-W. Choi, K. Kisslinger, E. A. Stach, J. C. Yang,
P. Strasser and B. R. Cuenya, Nat. Commun., 2016, 7, 12123.
Conflicts of interest
15 A. Eilert, F. S. Roberts, D. Friebel and A. Nilsson, J. Phys.
Chem. Lett., 2016, 7, 1466–1470.
There are no conicts to declare.
16 C. Reller, R. Krause, E. Volkova, B. Schmid, S. Neubauer,
A. Rucki, M. Schuster and G. Schmid, Adv. Energy Mater.,
Acknowledgements
2017, 7, 1602114.
This work was performed in support of the US Department of 17 L. Cao, D. Raciti, C. Li, K. J. T. Livi, P. F. Rottmann,
Energy's Fossil Energy Crosscutting Technology Research
Program. The Research was executed through the NETL Research
K. J. Hemker, T. Mueller and C. Wang, ACS Catal., 2017, 7,
8578–8587.
2
and Innovation Center's CO Utilization Technologies. Research 18 A. Eilert, F. Cavalca, F. S. Roberts, J. Osterwalder, C. Liu,
performed by Leidos Research Support Team staff was conducted
under the RSS contract 89243318CFE000003. This research used
the 8-ID (ISS) and 23-ID-2 (IOS) beamlines of the National
M. Favaro, E. J. Crumlin, H. Ogasawara, D. Friebel,
L. G. M. Pettersson and A. Nilsson, J. Phys. Chem. Lett.,
2017, 8, 285–290.
Synchrotron Light Source II, a U.S. Department of Energy (DOE) 19 W. Zhu, Y.-J. Zhang, H. Zhang, H. Lv, Q. Li, R. Michalsky,
Office of Science User Facility operated for the DOE Office of
Science by Brookhaven National Laboratory under Contract No.
A. A. Peterson and S. Sun, J. Am. Chem. Soc., 2014, 136,
16132–16135.
DE-SC0012704. This research used Beamline 17-BM of the 20 K. Jiang, P. Kharel, Y. Peng, M. K. Gangishetty, H.-Y. G. Lin,
Advanced Photon Source, a U.S. Department of Energy (DOE)
Office of Science User Facility operated for the DOE Office of
E. Stavitski, K. Attenkofer and H. Wang, ACS Sustainable
Chem. Eng., 2017, 5, 8529–8534.
Science by Argonne National Laboratory under Contract No. DE- 21 V. R. Calderone, N. R. Shiju, D. C. Ferr ´e and G. Rothenberg,
AC02-06CH11357. The authors acknowledge use of the Materials
Green Chem., 2011, 13, 1950–1959.
Characterization Facility at Carnegie Mellon University sup- 22 O. Martin, C. Mondelli, D. Curulla-Ferr ´e , C. Drouilly,
ported by grant MCF-677785. The authors also thank Dr William
A. Rogers and Dr Bryan Hughes for in situ cell printing.
R. Hauert and J. P ´e rez-Ram ´ı rez, ACS Catal., 2015, 5, 5607–
5616.
2
2
3 D. Raciti and C. Wang, ACS Energy Lett., 2018, 3, 1545–1556.
4 P. De Luna, R. Quintero-Bermudez, C.-T. Dinh, M. B. Ross,
O. S. Bushuyev, P. Todorovi ´c , T. Regier, S. O. Kelley,
P. Yang and E. H. Sargent, Nat. Catal., 2018, 1, 103–110.
References
1
Q. Lu and F. Jiao, Nano Energy, 2016, 29, 439–456.
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A, 2019, 7, 27576–27584 | 27583

View Article Online
Journal of Materials Chemistry A
Paper
2
2
2
5 H. S. Jeon, S. Kunze, F. Scholten and B. R. Cuenya, ACS 40 J.-J. Velasco-V ´e lez, T. Jones, D. Gao, E. Carbonio, R. Arrigo,
Catal., 2018, 8, 531–535.
6 X. Zheng, J. Han, Y. Fu, Y. Deng, Y. Liu, Y. Yang, T. Wang and
L. Zhang, Nano Energy, 2018, 48, 93–100.
7 H. Song, M. Im, J. T. Song, J.-A. Lim, B.-S. Kim, Y. Kwon,
S. Ryu and J. Oh, Appl. Catal., B, 2018, 232, 391–396.
8 D. Ren, J. Fong and B. S. Yeo, Nat. Commun., 2018, 9, 925.
C.-J. Hsu, Y.-C. Huang, C.-L. Dong, J.-M. Chen, J.-F. Lee,
P. Strasser, B. R. Cuenya, R. Schl ¨o gl, A. Knop-Gericke and
C.-H. Chuang, ACS Sustainable Chem. Eng., 2019, 7, 1485–
1492.
41 J. F. Xu, W. Ji, Z. X. Shen, W. S. Li, S. H. Tang, X. R. Ye,
D. Z. Jia and X. Q. Xi, J. Raman Spectrosc., 1999, 30, 413–415.
2
2
9 D. Raciti, Y. Wang, J. H. Park and C. Wang, ACS Appl. Energy 42 D. Raciti, M. Mao, J. H. Park and C. Wang, J. Electrochem.
Mater., 2018, 1, 2392–2398. Soc., 2018, 165, F799–F804.
0 A. Dutta, M. Rahaman, N. C. Luedi, M. Mohos and 43 K. J. P. Schouten, Z. Qin, E. P. Gallent and M. T. M. Koper, J.
P. Broekmann, ACS Catal., 2016, 6, 3804–3814. Am. Chem. Soc., 2012, 134, 9864–9867.
1 Y. Yang, L. Ohnoutek, S. Ajmal, X. Zheng, Y. Feng, K. Li, 44 Y. Hori, I. Takahashi, O. Koga and N. Hoshi, J. Mol. Catal. A:
3
3
T. Wang, Y. Deng, Y. Liu, D. Xu, V. K. Valev and L. Zhang,
J. Mater. Chem. A, 2019, 7, 11836–11846.
2 A. Stein, B. E. Wilson and S. G. Rudisill, Chem. Soc. Rev.,
Chem., 2003, 199, 39–47.
45 Y. Lum, B. Yue, P. Lobaccaro, A. T. Bell and J. W. Ager, J. Phys.
Chem. C, 2017, 121, 14191–14203.
3
3
3
2
013, 42, 2763–2803.
3 A. Wuttig and Y. Surendranath, ACS Catal., 2015, 5, 4479–
484.
4 D. H. Won, H. Shin, M. W. Chung, H. Jung, K. H. Chae,
46 A. Bagger, W. Ju, A. S. Varela, P. Strasser and J. Rossmeisl,
ACS Catal., 2019, 9, 7894–7899.
47 M. Gattrell, N. Gupta and A. Co, J. Electroanal. Chem., 2006,
594, 1–19.
4
H.-S. Oh, Y. J. Hwang and B. K. Min, Appl. Catal., B, 2019, 48 X. Nie, W. Luo, M. J. Janik and A. Asthagiri, J. Catal., 2014,
58, 117961. 312, 108–122.
5 B. Ravel and M. Newville, J. Synchrotron Radiat., 2005, 12, 49 C. Hahn, T. Hatsukade, Y.-G. Kim, A. Vailionis,
2
3
3
3
5
37–541.
6 B. H. Toby and R. B. Von Dreele, J. Appl. Crystallogr., 2013,
6, 544–549.
J. H. Baricuatro, D. C. Higgins, S. A. Nitopi, M. P. Soriaga
and T. F. Jaramillo, Proc. Natl. Acad. Sci. U. S. A., 2017, 114,
5918–5923.
4
7 E. L. Clark, J. Resasco, A. Landers, J. Lin, L.-T. Chung, 50 A. A. Peterson, F. Abild-Pedersen, F. Studt, J. Rossmeisl and
A. Walton, C. Hahn, T. F. Jaramillo and A. T. Bell, ACS
Catal., 2018, 8, 6560–6570.
8 B. Beverskog and I. Puigdomenech, J. Electrochem. Soc., 1997,
J. K. Norskov, Energy Environ. Sci., 2010, 3, 1311–1315.
51 M. R. Singh, E. L. Clark and A. T. Bell, Phys. Chem. Chem.
Phys., 2015, 17, 18924–18936.
3
3
1
44, 3476–3483.
52 A. S. Varela, M. Kroschel, T. Reier and P. Strasser, Catal.
Today, 2016, 260, 8–13.
9 L. Wang, K. Gupta, J. B. M. Goodall, J. A. Darr and K. B. Holt,
Faraday Discuss., 2017, 197, 517–532.
27584 | J. Mater. Chem. A, 2019, 7, 27576–27584
This journal is © The Royal Society of Chemistry 2019