Highly selective electrochemical CO2 reduction to CO using a redox-active couple on low-crystallinity mesoporous ZnGa2O4 catalyst

Article abstract of DOI:10.1039/c9ta00562e

The substantial overpotential of CO₂ activation, the complex CO₂ reduction pathway, and the competitive H₂ evolution reaction (HER) limit the practical applications of CO₂-based electrochemical energy conversion and storage technologies due to their low energy efficiency and product selectivity. Here, we proposed a strategy that combines mesopores and redox-active couple to effectively capture CO₂ and selectively generate CO product. As an example, we construct a redox-active couple, Zn²⁺/Zn⁺, on the low-crystallinity mesoporous ZnGa₂O₄ electrocatalyst. The mesopores not only help to capture CO₂ effectively but also inhibit the H₂ evolution. The weak lattice constraint of low crystallinity benefits the formation of a Zn²⁺/Zn⁺ redox couple to strongly interact with CO₂ molecule for subsequent activation and catalytic conversion, thus effectively decreasing the activation energy of CO₂ to the active species CO₂^- and accelerating the proton transfer to form the crucial COOH? intermediate. Consequently, this catalyst exhibited a high faradaic efficiency of 96% among Zn-based electrodes for CO generation at the relatively low applied potential of -1.4 V vs. Ag/AgCl (-0.8 V vs. RHE), and excellent stability during 10 h operation in a 0.1 M KHCO₃ solution.

Full text of DOI:10.1039/c9ta00562e

View Article Online
View Journal
Journal of
M at erials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Zhao, Y. Gu, P.
Chen, Z. Xin, H. Zhu, B. Wang, K. Zhu, S. C. Yan and Z. Zou, J. Mater. Chem. A, 2019, DOI:
1
0.1039/C9TA00562E.
Volume
4
Number
1
7
January 2016 Pages 1–330
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry A
Accepted Manuscripts are published online shortly after
Materials for energy and sustainability
www.rsc.org/MaterialsA
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2050-7488
PAPER
Kun Chang, Zhaorong Chang et al.
Bubble-template-assisted synthesis of hollow fullerene-like
MoS nanocages as a lithium ion battery anode material
2
rsc.li/materials-a

Page 1 of 8
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
View Article Online
DOI: 10.1039/C9TA00562E
Journal Name
ARTICLE
Highly Selective Electrochemical CO Reduction to CO Using a
2
Redox-Active Couple on Low-Crystallinity Mesoporous ZnGa O
2
4
Catalyst
a
a
a
a
a
a
b
Meiming Zhao , Yaliu Gu , Ping Chen , Zhenyu Xin , Heng Zhu , Bing Wang* , Kai Zhu* , Shicheng
a
ac
Yan* and Zhigang Zou
Substantial overpotential of CO
limit the practical applications of CO
energy efficiency and product selectivity. Here, we proposed a strategy that combines mesopores and redox-active couple
2
activation, complex CO
2 2
reduction pathway, and the competitive H evolution reaction (HER)
2
-based electrochemical energy conversion and storage technology due to the low
2
+
+
to effectively capture CO
on the low-crystallinity mesoporous ZnGa
also inhibit the H evolution. The weak lattice constraint of low crystallinity benefits to formation of Zn2+/Zn
2
and selectively generate CO product. As an example, we construct a redox-active couple, Zn /Zn ,
electrocatalyst. The mesopores not only help to capture CO effectively but
redox couple
molecule for subsequent activation and catalytic conversion, thus decreasing the activation
Received 00th January 20xx,
Accepted 00th January 20xx
2
O
4
2
+
2
to strongly interact with CO
2
energy of CO to active species CO effectively and accelerating the proton transfer to form crucial COOH* intermediate.
2
DOI: 10.1039/x0xx00000x
-
2
www.rsc.org/
Consequently, this catalyst exhibited highly faradaic efficiency of 96% among Zn-based electrodes for CO generation at the
relatively low applied potential of -1.4 V vs. Ag/AgCl (-0.8 V vs. RHE) and excellent stability during 10 h operation in 0.1 M
3
KHCO solution.
great potential to generate liquid fuels and synthetic petroleum
through Fischer-Tropsch methods.[ It is well established that
6, 7]
Introduction
The unprecedented consumption of fossil fuels induced the
energy shortage and potential global environmental crisis.[ In
this scenario, CO
converting chemical inert CO
the CO product from electrochemical CO
2
reduction mainly
[
8]
includes the following steps: the activation of CO by one
2
1-3]
-
-
electron via CO
2
+ e → CO
2
, proton reduction to generate
2
electrochemical reduction reaction (CO
2
RR),
-
+
COOH* (* is an indicative of adsorption site) by CO
COOH*, converting COOH* to CO* by COOH* + H + e → CO* +
2
+ * + H →
2
molecule to fuels for a
+
-
sustainable supply of electric energy, seems to be a promising
2
H O, and CO generation via CO* → CO + *. In such reaction
solution to support environmental stability and to balance the
pathway, a challenge is to stably generate the important species
use of natural resources.[
4, 5]
CO, as one significant component
2 2
of syngas (CO/H ), is a common CO RR product exhibiting a
-
CO
2
and COOH* for selectively producing CO.
Several precious metals, such as Au,[ Ag,[9] and Pd,[10, 11]
8]
were demonstrated to exhibit the tightly binding with COOH*
intermediate and weakly binding with CO*, thus affording high
a. Jiangsu Key Laboratory of Artificial Functional Materials, Eco-materials and
Renewable Energy Research Center (ERERC), National Laboratory of Solid State
Microstructures, Collaborative Innovation Center of Advanced Microstructures,
College of Engineering and Applied Sciences, Nanjing University, No. 22 Hankou
Road, Nanjing, Jiangsu 210093, P. R. China. Email: yscfei@nju.edu.cn,
bingwang@nju.edu.cn
[12]
faradaic efficiency (generally > 90% ） in CO generation.
[
13-
Similarly, metal Zn is also able to directly reduce CO
2
to CO.
1
5]
However, the faradaic efficiencies of Zn electrodes are still
limited at 70~85%, even in the nanostructures such as
b.
School of Information Science and Engineering, Nanjing University Jinling College,
nanoparticle, nanosheet, and nanodendrite.[
13-15]
A few of Zn
No. 8, Xuefu Road, Nanjing, Jiangsu 210089, People’s Republic of China.Email:
zhukai@jlxy.nju.edu.cn
Jiangsu Key Laboratory for Nano Technology, National Laboratory of Solid State
atoms or ions-containing catalysts have been supposed to
construct more effective active sites. For example, a zinc-
c.
Microstructures, School of Physics, Nanjing University, No. 22, Hankou Road,
Nanjing, Jiangsu 210093, People’s Republic of China
x
nitrogen (Zn-N ) ligand was constructed by coordinating the
†
Electronic Supplementary Information (ESI) available: [XRD pattern of NaGaO
XRD patterns, N adsorption-desorption isotherms, pore diameter, CO adsorption-
desorption isotherms, H
of H-ZnGa and L-ZnGa
saturated 0.1 M KHCO solution. Oxidation peak of commercial ZnO, L-ZnGa
commercial Ga in CO -saturated 0.1 M KHCO solution. Photograph of ZnO
electrode and ZnGa electrode after CV scanning. XPS spectra of H-ZnGa
.03V and -1.2 V. The cyclic voltammogram of the L-ZnGa in CO -saturated KCl,
solution (vs. RHE). Ultraviolet-visible (UV-vis) absorption spectrum of
2
.
single Zn atom with N-doped graphene and achieved a high
2
2
faradaic efficiency, partial current density and EPR signals faradaic efficiency for CO up to 91~95% due to that the
2
2
O
4
2
O
4
.Polarization curves on bare titanium sheet in N
2
or CO
2 4
2
-
synergistic effect of Zn and N atoms could decrease the free
energy barrier of COOH* formation and subsequently facilitate
3
O and
2
O
3
2
3
16, 17]
2
O
4
2
O
4
at - the desorption of CO*.[
The Zn-N ligand of organometallic
gas and
reduction
1
2
O
4
2
complex was found as a Lewis base to activate inert CO
2
KHCO
3
commercial ZnO, L-ZnGa
2
O
4
after and before reaction. High-magnification TEM successfully lowered the overpotential for CO
2
images and SAED patterns of L-ZnGa
DOI: 10.1039/x0xx00000x
2
O
4
and H-ZnGa
2
O
4
after reaction. See
[18]
approximately 0.6 V. CO
2
reduction on the Zn complex is
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 2 of 8
ARTICLE
expected to be achieved by a Zn redox couple, possibly Zn /Zn
Journal Name
2
+
+
Compared with other electrocatalyst support, titanium is of
View Article Online
2
+
0
+
10
1
DOI: 10.1039/C9TA00562E
or Zn /Zn . Monovalent Zn (3d 4s ) has been proved to be a high electrical conductivity and negligible ability in CO
2
functional species in many cases due to the coordinatively reduction. Because of its excellent properties, the as-prepared
unsaturated characteristic.[ For instance, the Zn -V
vacancy) complexes in Zn-containing layered double hydroxide thickness, 99.5 % purity) via electrophoretic deposition (EPD).
nanosheets are effective trapping sites for the CO adsorption Before the preparation of working electrode, titanium sheets
and significantly improve the activity of photocatalytic need to be chemically polished in a solution containing HF,
19]
+
o
2 4
(oxygen ZnGa O powders were deposited on a titanium sheet (0.2 mm
2
[
20]
reduction of CO
2
.
3 2
However, to construct abundant and stable HNO , H O (1:2:7 in volume) for 30s and cleaned with acetone,
Zn redox couple on the inorganic catalysts is still a challenge due ethanol, deionized water under ultrasonic concussion in order
to their strong lattice constraint and limited specific surface to remove the compact oxidation layer. Just prior to EPD, 40 mg
area.
2 4
of ZnGa O particles and 10 mg iodine were dispersed in 50 mL
Here, we proposed a strategy that constructs surface acetone solution by sonication for 30 min. Then, two titanium
2
+
+
abundant Zn /Zn redox-active couple on low-crystalline sheets were placed vertically at a 10 mm distance and
inorganic material with high specific area. The abundant redox- immersed in the suspension followed by adding 20 V bias
active couple easily formed on the surface of the low- between them for 5 min. During the EPD process, insulating
2
crystallinity mesoporous ZnGa
O
2 4
due to the weak lattice tapes were used in order to ensure the 1 cm deposition area of
constraint and inherently exhibited strong interactions with the electrode.
carbon-based species. Moreover, the mesoporous structure of
ZnGa
2
O
4
may generate diffusion gradients through the local Characterization of Mesoporous ZnGa
2
O
4
Powders
-
consumption of proton donor such as HCO
3
to inhibit H
2
2 4
The crystalline phases of the as-prepared ZnGa O powders
evolution, offers high specific area to generate abundant were determined by XRD (Rigaku Ultima III, CuKa radiation). The
2
+
+
Zn /Zn redox-active couple and captures CO
2 2 4
molecule specific surface area of the ZnGa O catalysts were obtained by
2
+
+
efficiently. The Zn /Zn redox couple on ZnGa
2
O
4
catalyst nitrogen absorption at 77K on a surface area and porosity
-
effectively activates the CO to the active species CO
2
2
and analyzer (Micromeritics ASAP 2020., USA) and calculated by BET
adsorption capacity was detected at 273 K based
on the Brunauer–Emmett–Teller (BET) method at P/P = 0.03.
and Zn /Zn accompanied with the intrinsic advantage of High-solution images and selected area electron diffraction
mesoporous structure effectively suppressed the competitive (SAED) patterns of the ZnGa catalysts, were collected by
evolution reaction, compared with other Zn-based transmission electron microscopy (TEM, FEI Tecnai G2 F30 S-
electrodes, the L-ZnGa achieved the highest faradaic Twin, USA) operated at 200 kV. UV-vis diffuse reflectance
strongly coupled the electron and proton to form crucial COOH* method. CO
intermediate. The inherently strong interaction between CO
2
2
0
2
+
+
2 4
O
H
2
2 4
O
efficiency (96%) at the moderate applied potential of -1.4 V vs spectra (DRS) were obtained using a UV-vis spectrophotometer
Ag/AgCl (-0.8 V vs. RHE), intuitively indicating the high (UV-LAMBDA 950, PerkinElmer, USA) and were transformed
selectivity and superior performance of L-ZnGa
2 4
O .
into the absorption spectra according to the Kubelka–Munk
relationship. The chemical state of ZnGa catalysts were
investigated by X-ray photoelectron spectroscopy (XPS) on a
PHI5000 Versa Probe (ULVAC-PHI, Japan) with
monochromatized Al Kα X-ray radiation (1486.6 eV). The
2 4
O solid powders were prepared by binding energy was corrected by reference to the C 1s line at
2 4
O
Experimental section
Synthesis of Mesoporous ZnGa
Briefly, the porous ZnGa
ion exchange between NaGaO
2 4
O Powders
2
colloidal suspension and an 284.6 eV. The electron paramagnetic resonance (EPR) spectra
2
COO) . At first, in order to obtain the were obtained using a Bruker (Model EMX-10/12 X-band,
aqueous solution of Zn(CH
NaGaO powders, Na CO
3
2
2
3
(1.07 g) and Ga
O
2 3
(1.87 g) (molar Bruker, Germany) electron paramagnetic resonance
ratio 1.01:1) were carefully ground in the mortar, then the spectrometer at room temperature (298 K).
mixed powders were transferred to an alumina crucible and
heated at 1123K for 12h. Next, NaGaO
M, 10 mL) was prepared though dissolving the NaGaO
in deionized water. The pre-made suspension was added to prepared electrode, the measurements of CO
Zn(CH COO) aqueous solution(0.05 M, 20 mL) and the mixture reduction were conducted in a N gas-tight H-type cell separated
was stirred for 3 h at room temperature. The mesoporous by a piece of proton exchange membrane (Nafion 117) with a
ZnGa was formed via a reaction of 2NaGaO + Zn(CH COO) three-electrode system. The counter electrodes was a Pt sheet
ZnGa + 2CH COONa. After the reaction finished, the white and an Ag/AgCl electrode filled with saturated KCl solution was
sediment was separated by centrifugation and washed by used as the reference electrode. 0.1 M KHCO or 0.1 M KCl
deionized water for several times, then dried in air atmosphere solution was adopted as the electrolyte. Before the performing
at 343 K for 24 h. The high-crystallinity ZnGa powder was of 1 h controlled potential electrolysis, high purity CO gas was
prepared by the same method with a heat treatment of the ion- purged through the catholyte and anolyte for 30 min to achieve
2
colloidal suspension (0.2 Electrochemical Measurements and Products Analysis
powder In order to evaluate the catalytic performance of the as-
electrochemical
2
2
3
2
O
2 4
2
3
2
→
O
2 4
3
3
O
2 4
2
exchange sample under air in muffle furnace at 973K for 8h.
saturate state. During the electrochemical measurements, the
solution was stirred by a magnetic stirring apparatus in order to
Preparation of Working Electrode
2
guarantee the CO mass transfer process.
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 8
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Journal Name
After the reaction, 1mL gas was taken from the cathodic
ARTICLE
Transmission electron microscope (TEM) observations show
View Article Online
DOI: 10.1039/C9TA00562E
reaction cell for CO analysis by using a gas chromatograph (GC- that the L-ZnGa
014, Shimadzu Corp., Japan) and H
chromatograph (GC-8A, Shimadzu Corp., Japan). The solution with the BET results. Obviously, the 20-50 nm nanoplates were
2 4
O has a typical wormholes resulting from the
2
2
analysis by another gas aggregation of 5-10 nm crystals (Figure 1a), well corresponding
was collected by a sealed bottle in order to detect the liquid observed in H-ZnGa
product through a liquid chromatograph (CBM-20A, Shimadzu growth of L-ZnGa
2
O
4
(Figure 1b), suggesting that the particle
nanocrystals occurs during the heat
2
O
4
Corp., Japan).
treatment. The selected area electron diffraction (SAED)
pattern presents single crystal characteristics, indicating that
the ZnGa
a
O
2 4
nanoplate is a single crystal.
Results and discussion
b
The NaGaO
mixture of Na
123K for 12h. The 1% excess amount of Na
compensate for the volatilization loss. The X-ray diffraction
XRD) indicated that single-phase NaGaO was successfully
prepared (Figure S1). Low-crystallinity ZnGa catalyst
denoted as L-ZnGa ) was synthesized by an ion-exchange
reaction between NaGaO and Zn(CH COO) at room
temperature. As well demonstrated in our previous report,
the NaGaO powders can be dissolved into water to form a
colloidal suspension. The NaGaO colloid tends to form a
mesoporous framework by flocculation due to a strong
2
precursor was first synthesized by heating the
CO and Ga with a mole ratio of 1.01:1 at
CO was added to
2
3
2 3
O
1
2
3
(
2
2 4
O
(
2 4
O
2
3
2
[
21]
d100
c 0
L-ZnGa2O4
H-ZnGa2O4
.0
2
80
-
-
0.5
1.0
2
60
2
+
H-ZnGa2O4 in N2
L-ZnGa2O4 in N2
L-ZnGa2O4 in CO2
40
20
0
electrostatic attraction. After introducing Zn , the ion exchange
-1.5
2
+
+
reaction between Zn and Na occurs and the mesoporous
-
-
2.0
H-ZnGa O4 in CO2
2
structure of NaGaO
inherited into ZnGa
2 4 2 4
ZnGa O (denoted as H-ZnGa O ) was prepared by heating the
2
colloidal template was copied and
2.5
-
0.6 -0.8 -1.0 -1.2 -1.4 -1.6 -1.8
Potential (V) vs. Ag/AgCl
-1.30 -1.35 -1.40 -1.45 -1.50 -1.55 -1.60
Potential (V) vs. Ag/AgCl
2 4
O . For comparison, a high-crystallinity
Figure 1. a) TEM image of L-ZnGa
HRTEM) lattice image. White dashed line shows the profile of the
nanocrystal. b) TEM image of H-ZnGa . Inset shows the SAED pattern. c)
Polarization curves on L-ZnGa (solid line) and H-ZnGa electrodes
dashed line) in N - or CO -saturated 0.1 M KHCO solution. d) Faradaic
efficiency for CO generation on ZnGa electrodes.
2 4
O . Inset shows the high-resolution TEM
low-crystallinity one at 973 K in air furnace. As illustrated in
Figure S2, the X-ray diffraction (XRD) patterns of both as-
prepared samples can be well indexed to the cubic spinel
(
2 4
O
O
2 4
2 4
O
ZnGa
low intensity indicated the low crystallinity of the L-ZnGa
contrast, the sharp XRD peaks for H-ZnGa
demonstrate that the crystallinity of the L-ZnGa
evidently improved after heat treatment. Nitrogen adsorption-
desorption studies on L-ZnGa (Figure S3) show a type-IV
O
2 4
(JCPDS 38-1240). The broadening of XRD peaks with
. In
would
was
(
2
2
3
2 4
O
2 4
O
2 4
O
2 4
O
Due to the electrochemically inert CO
titanium sheet (Figure S5), the as-prepared ZnGa
were electrophoretically deposited on a titanium sheet as a
working electrode. CO reduction on ZnGa electrodes was
carried out in CO -saturated 0.1 M KHCO solution in a gas-tight
two-compartment electrolytic cell with a three-electrode
system. Unless otherwise stated, all potentials were quoted
versus a silver chloride electrode (Ag/AgCl). As shown in Figure
2
reduction on
2 4
O
powders
2 4
O
isotherm, a typical mesoporous structure. The pore diameter
calculated by the Barrett-Joyner-Halenda (BJH) method is 3.6
nm, specific surface area calculated from the Brunauer-
2
2 4
O
2
3
Emmett-Teller (BET) plot ranging from P/P
0
= 0.05 to 0.29 is
exhibited a specific surface area of 26.8
m g without obvious porous structure. The pore diameter for
L-ZnGa calculated from the nitrogen adsorption isotherm by
the Barrett-Joyner-Halenda (BJH) method was 3.8 nm. The L-
2
-1
1
2 4
53.7 m g . H-ZnGa O
2
-1
1
c, the polarization curves of both H-ZnGa
CO -saturated 0.1 M KHCO solution exhibited a cathodic
current onset at approximately -1.3 V.Replacing CO with N will
induce a 0.3 V negative shift of current onset potential for L-
ZnGa and no obvious changes in potential-current
relationship for H-ZnGa 4, respectively.Only gaseous products,
CO and H , were detected. The faradaic efficiency of CO and H
2 4 2 4
O and L-ZnGa O in
2 4
O
2
3
-1
2
2
ZnGa
which is about 3 times as high as 5.92 mg g for H-ZnGa
Figure S4). The kinetic diameter of CO is 0.33 nm, suggests
2 4 2
O possess a CO adsorption capacity of 17.64 mg g ,
-1
2 4
O
2 4
O
(
2
2
O
that the ideal pore size seem to be 0.5–0.6 nm for the single
molecule adsorption. In our case, the 3.8 nm pore would allow
multimolecular adsorption to occur, probably due to the more
2
2
at various applied potentials were summarized in Figure 1d and
Figure S6. We found that CO initially generated at -1.3 V,
corresponding to an overpotential of 570 mV (theoretical redox
potential at -0.73 V for CO
ZnGa , the faradaic efficiency (FE) significantly increased
from 56% at -1.3 V to a peak value 96% at -1.4 V, and then
maintained at 70-80% when potentials varied from -1.4 to -1.6
2 4
V. However, the FE of CO on H-ZnGa O slightly varied at 28-37%
basic sites (metal sites) on the surface of L-ZnGa
lattice constraint. Indeed, it has been demonstrated that the
mesoporous SiO molecular sieves with 3.5 nm pore after
modification by basic groups such as hydroxyl or amino groups
2 4
O with weak
2
/CO). For CO generation on L-
2
2 4
O
exhibited excellent CO
2
adsorption and followed
a
multimolecular adsorption model. [
22]
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 4 of 8
ARTICLE
Journal Name
a
1
0
1
2
3
4
b
from -1.3 to -1.6 V. As shown in Table S1, the performance of L-
Zn 2p3/2
N2
View Article Online
DOI: 10.1039/C9TA00562E
CO2
L-ZnGa2O4
ZnGa
2
O
4
surpasses the most Zn-based electrodes. The CO
catalysts were
0
0
0
0
.45
.30
.15
.00
-
-
-
-
partial current densities on two ZnGa
O
2 4
L-ZnGa2O4
after ion milling
calculated through multiplying their faradaic efficiencies by the
polarization curves and then were plotted against the applied
L-ZnGa2O4 at -1.03 V
L-ZnGa2O4 at -1.2 V
-0.15
-0.30
potentials. Figure S7 clearly depicted that the L-ZnGa
O
2 4
-5
-6
-
1.4-1.3-1.2-1.1-1.0-0.9-0.8-0.7
Potential (V) vs. Ag/AgCl
exhibited a higher partial current density over all the applied
-
2.0
-1.5
-1.0
-0.5
0.0
1026
1024
1022
1020
1018
1016
Potential (V) vs. Ag/AgCl
Binding Energy (eV)
potentials, proving the high activity of L-ZnGa
reduction.
2 4 2
O in CO
c
C 1s
Ga 3d
d
The high activity of the L-ZnGa
investigated by cyclic voltammogram scanning. Under N
saturated 0.1 M KHCO electrolyte, a board oxidation wave
2
O
4
catalyst was further
L-ZnGa2O4
L-ZnGa2O4
2
-
L-ZnGa2O4
after ion milling
L-ZnGa2O4
after ion milling
3
around -1.2 V during positive-going potential scanning and its
corresponding reduction wave presented in the CV of the L-
L-ZnGa2O4 at -1.03 V
L-ZnGa2O4 at -1.03 V
L-ZnGa2O4 at -1.2 V
L-ZnGa2O4 at -1.2 V
2 4
ZnGa O electrode (Figure 2a). This evidence would originate
2
+
24
22
20
18
16
290
288
286
284
282
280
from the redox reaction of Zn , as demonstrated in the
previous reports.[ However, the oxidation wave narrowed
Binding Energy (eV)
Binding Energy (eV)
14]
Figure 2. a) Cyclic voltammogram of the L-ZnGa
.1 M KHCO solution. Inset shows oxidation wave of L-ZnGa
saturated 0.1 M KHCO solution. b) Zn 2p3/2, c) Ga 3d, and d) C1s XPS spectra
of L-ZnGa with and without electrochemical treatment.
2
O
4
in N
2
- or CO
2
-saturated
and positively shifted to the potential -1.03 V when used CO
saturated 0.1 M KHCO
indicated that there are strong interactions between CO
2
-
0
3
O
2 4
in CO
2
-
3
as electrolyte (Figure S8). This fact
3
2
2 4
O
2
+
molecules and the Zn species on L-ZnGa
2 4
O . Indeed, such a
strong interaction was only observed at -0.7 V on the ZnO
electrode in CO -saturated 0.1 M KHCO electrolyte (Figure S9a)
but not on the Ga electrode (Figure S9b), indicating that the
Zn species are mainly responsible for the CO reduction. The
difference in oxidation-wave potential would result from the
different chemical environments for Zn in ZnO and ZnGa . As
The surface chemical states of catalyst during the CO
reduction were investigated by X-ray photoelectron
spectroscopy (XPS). Clearly, as shown in Figure 2b, for the as-
2
2
3
2 3
O
2
prepared L-ZnGa
decomposed into two peaks: 1021.3 and 1019.5 eV. After
constant potential electrolysis on L-ZnGa at -1.2 V, a
reduction-current peak potential, in CO -satuared 0.1 M KHCO
2 4
O , the Zn 2p3/2 XPS spectrum was
2 4
O
2 4
O
depicted in Figure S10, after cyclic voltammetry scanning the
white ZnO electrode turned to black because of the formation
2
3
solution for 1 h, the Zn 2p3/2 XPS spectrum was deconvoluted
into two peaks at 1021.5 and 1019.2 eV. After 3 nm-depth ion
2 4
of metallic zinc while the ZnGa O electrode still remained the
2
+
initial colour, proving that the Zn was not reduced to the
metallic state but still the ionic state. Actually, the oxidation
milling for the as-prepared L-ZnGa
splitting disappeared and binding energy is 1021.9 eV, which is
.9 eV higher than the binding energy of Zn-O bonds in H-
ZnGa (1021.0 eV) due to the weak lattice constraint in low-
crystallinity L-ZnGa . Therefore, the similar Zn 2p3/2 XPS peak
splitting for L-ZnGa in air or CO -containing electrolyte is
probably a result of the strong interactions between CO and L-
ZnGa . The obvious binding energy splitting means that the
Zn species exhibited a mixed valence states. The peak at 1021.3
eV for L-ZnGa or 1021.5 eV for L-ZnGa at -1.2 V would be
assigned to the binding energy of Zn-O in ZnGa
Correspondingly, the lower binding energy at 1019.5 eV for L-
ZnGa or 1019.2 eV for L-ZnGa at -1.2 V may indicate that
a low-valence Zn species was stabilized by carbon-containing
2 4
O , the Zn 2p3/2 peak
2 4
wave at -1.03 V and the reduction wave at -1.2 V of L-ZnGa O
0
2
+
+
were perfectly matched the redox behaviour of Zn /Zn couple
in ZIF-8 catalyst in 0.5 M NaCl electrolyte, indicating the
2 4
O
2 4
O
2
+
+
[23]
existence of Zn /Zn redox couple in this system. Noting that
the oxidation wave is obviously bigger than the reduction wave
O
2 4
2
2
2 4
on L-ZnGa O . Similar irreversible reduction responses were
2 4
O
reported for Zn complexes, which are attributed to the
+
[24,25]
generation of rare, unstable, and low valent Zn species.
As
O
2 4
2 4
O
+
the generation of Zn requires an electron to enter the 4s orbital
[
27]
2 4
O .
with high potentials since the 3d shell is already filled. In our
+
case, the interactions between Zn and CO
2
may further enlarge
O
2 4
2 4
O
this irreversibility due to the delayed reaction kinetics. In
+
2+
addition to the oxidation of Zn to Zn , the big anodic wave may
species from CO
2
activation via an interaction between Zn
Indeed, the generation of low-valence Zn
include a contribution from carbon-containing species, such as
[
28]
species and CO
2
.
-
CO
2
, CO*, and COOH*, which are produced/adsorbed on the
+
ions such as Zn requires to capture an electron into its 4s orbital
because the 3d shell is fully filled. This evidence confirmed that
electrode surface. A similar anodic wave generation from
adsorption of carbon-containing species was also observed on
2
+
under a negative potential Zn on L-ZnGa
electrons to stay in a reduction state. The unstable low-valence
Zn ions is able to transfer electrons to CO molecules, thus
decreasing the activation barrier of CO . The strong interactions
between CO and Zn ions were also demonstrated at -1.03 V, an
oxidation potential. At -1.03 V, the low Zn 2p3/2 binding energy
2 4
O tends to gain
3
+
4+
[26]
Ti /Ti redox couple.
2
2
2
at 1019.5 eV confirmed the strong interactions between CO
and Zn ions even at a relatively positive potential.
2
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 8
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Journal Name
Ga 3d XPS spectrum of L-ZnGa
ARTICLE
2
+
+
O
2 4
was split into 19.7 eV for positive potential, probably implied that the Zn /Zn redox
View Article Online
3
+
DOI: 10.1039/C9TA00562E
-
2 4 2 4 2
Ga in ZnGa O and a low-intensity peak at 17.7 eV (Figure 2c). couple on ZnGa O did not only facilitate the generation of CO
After 3 nm-depth ion milling, a single-peak binding energy was but also accelerated the capture of proton significantly.
observed at 19.9 eV. Therefore, the peak at 17.7 eV may
originate from interactions between the Ga species and CO
from air. Indeed, the binding energy of 17.0 eV was only
detected in L-ZnGa treated at -1.03 V, not at -1.2 V. This
evidence supports that the Ga species on surface L-ZnGa
seems to interact with CO molecule due to the relatively high
applied potential which is not enough to drive CO conversion,
as demonstrated in CV scanning. The binding energy at 18.4 eV
occurs in both L-ZnGa treated at -1.03 and -1.2 V and is
2
a
0.45
-1.02
1.05
b
0.45
.30
-
0.98
-
-1.00
-1.02
-1.08
-1.11
0
0.30
0.15
0.00
-1.04
-1.14
0
.3
0.6
O
2 4
0.3
Concentration (moL
0.6 -1
0.15
0.00
0.15
Concentration (moL L-1)
L
)
2 4
O
-
2
-
0.15
0.30
0.45
0
.1 M KHCO3
.3 M KHCO3
-0.30
-0.45
-0.60
0.1 M KCl
0.3 M KCl
0.5 M KCl
0
2
-
-
0.5 M KHCO3
.7 M KHCO3
0.7 M KCl
0
-
1.4 -1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7
-
1.4 -1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7
2 4
O
Potential (V) vs. Ag/AgCl
Potential (V) vs. Ag/AgCl
assigned to the Ga in metallic state from reduction of surface
Ga ions, which probably suggests that the reduction of Ga
species is an irreversible process.[ Although the Ga species is
c
1
0
d 100
EDTA-2Na L-ZnGa2O4
80
29]
-1
60
able to interact with CO
the standpoint that the Zn species on ZnGa
contributed to the activation and reduction of CO
overpotentials.
2
, the CV scanning definitely supported
mainly
with the low
-2
-3
EDTA-2Na L-ZnGa2O4
L-ZnGa2O4
O
2 4
40
-4
-5
-6
2
20
0
-
1.30 -1.35 -1.40 -1.45 -1.50 -1.55 -1.60
The C 1s XPS spectrum for L-ZnGa
O
2 4
exhibited a peak at
-2.0
-1.5
-1.0
-0.5
0.0
Potential (V) vs. Ag/AgCl
Potential (V) vs Ag/AgCl
2
84.6 eV for adventitious standard reference carbon and a peak
-
[30, 31]
Figure 3. a) Cyclic voltammogram of the L-ZnGa
O
2 4
in CO
2
-saturated KHCO
3
at 288.4 eV for CO
2
(Figure 2d).
No peak at 288.4 eV was
solution. The inset shows the linear relationship between the peak potential
and the concentration of KHCO solution. b) Cyclic voltammogram of the L-
ZnGa in CO -saturated KCl solution. The inset shows the relationship
between the peak potential and the concentration of KCl solution. c) Cyclic
voltammogram of L-ZnGa in CO -saturated KHCO + EDTA-2Na solution. d)
in CO -saturated KHCO + EDTA-2Na
observed for L-ZnGa
verifying that the CO
surface of L-ZnGa . Two peaks at 283.4 and 285.5 eV occur in
both L-ZnGa treated at -1.03 and -1.2 V, were respectively
2 4
O
after 3 nm-depth ion milling, further
is from the activation of CO on the
2
-
3
2
O
2 4
2
2 4
O
2 4
O
O
2 4
2
3
assigned to the Zn-C bond in carbon-containing species and the
C-O bonds.[
32-35]
CO faradaic efficiency of the L-ZnGa
solution.
O
2 4
2
3
A peak at 288.8 eV on L-ZnGa
2
O
4
at -1.03 V is
-
assigned to the COOH*, resulting from the CO
2
species
[
36]
conversion by capturing proton. This evidence means that
-
-
Usually, HCO -containing electrolyte was considered as a
the CO
2
species (288.5 eV) on L-ZnGa
O
2 4
at -1.2 V is able to
3
1]
key proton transfer route for CO
2
reduction.[ Indeed, we
capture proton to form COOH* with assistance of Zn ions at a
potential as low as -1.03 V. Although a low-intensity Ga-C-
discovery that the peak potential for oxidation wave on L-
ZnGa
2
O
4
electrode negatively shifted with increasing the
containing species at 282 eV was observed in L-ZnGa
V, the existence of high-intensity Zn-C-containing species at
both -1.03 and -1.2 V would mean that the CO molecule is
mainly activated by Zn species to CO
-, to COOH*, and to CO*
for CO generation. Nevertheless, in the case of H-ZnGa
catalyst (Figure S11), the low-content Zn-C-containing species
1019.0 eV) in Zn 2p XPS spectrum indicated that the low-
crystallinity L-ZnGa with high specific surface area could
release Zn redox couple easily than high-crystallinity H-ZnGa
with small specific surface area, and hence enhancing the ability
of CO conversion.
On the basis of the experimental evidences from the XPS
and CV scanning, the CO activation is expected to be achieved
2 4
O at -1.03
-
concentration of HCO
depends linearly on the concentration of HCO
follows a formula of E (V) = -1.026 - 0.17 CHCO (mol L ). This
linear relationship indicated that the HCO
3
(Figure 3a). The peak potential E
p
(V)
-
-) and
3
(CHCO
3
2
-
-1
p
3
2
-
ion has a strong
2 4
O
3
influence on the formation of COOH* by proton transfer. It is
-
well known that the HCO
3
ion could dissociate in the aqueous
(
+
solution and release the H more easily than H
2
O molecule due
2 4
O
-
2-
+
to the smaller dissociation constant for HCO
3
↔ CO
3
+ H (pKa
2 4
O
o
+
-
o
[37]
=
10.33, 25 C) than H
2
O ↔ H + OH (pKa = 14, 25 C). In
-
order to further confirm that the HCO
source instead of H
voltammogram of the L-ZnGa
3
ion was the proton
2
2
O molecule, we performed the cyclic
catalyst in CO -saturated KCl
O
2 4
2
2
2
+
+
2+
-
+
solution (pH = 6.55) with the same ion concentration as KHCO3
by Zn capturing electrons to form Zn , Zn + e → Zn , and then
+
-
+
electrolyte. The almost unchanged peak potential clearly
Zn transfers electrons to activate CO
2
into CO
. This means that the surface lattice Zn on L-
2 4
O catalyst is first reduced into low-valence Zn species,
2 2
, Zn + CO →
2
+
+
2
+
-
2+
demonstrated that the Zn /Zn redox couple nearly captured
Zn + CO
ZnGa
2
-
no proton from KCl solution, namely that Cl or H
2
O nearly had
-
+
+
no effect on the proton transfer process to the CO
intermediate (Figure 3b). To further confirm the HCO is the
Zn . Subsequently, the Zn species interacted with CO
to form adsorbed CO
2
molecule
2
-
-
2+
2
and thereby Zn regenerated. It is well
3
source of protons, we converted the reference electrode from
Ag/AgCl electrode to the pH-sensitive RHE electrode (Figure
established that COOH* is a critical intermediate during the
generation of CO product. Theoretically, the formation of
3
S12a, Figure S12b). The pH of KHCO , KCl electrolyte with
COOH* intermediate during CO
2
electrochemical reduction was
different concentrations shown in Table S2 and the redox
a result of coupled electron and proton transfer to the inert CO
2
behavior of L-ZnGa
O
2 4
were almost no change compare to the
molecule. The COOH* species can be detected at a relatively
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 6 of 8
ARTICLE
Journal Name
2Na, and the FEs of CO are as low as 10-20 % (FigVuierweAr3ticdle).OTnlhinee
2 3
Ag/AgCl electrode. By adding HCl to 0.1 M CO -saturated KHCO
solution to adjust the pH value, the peak potential for oxidation ultraviolet-visible absorption spectra (FiDgOurI:e10S.11033)9/sCh9oTwA00t5h6a2tE,
wave on L-ZnGa
2 4 2 2 4
O electrode negatively shifted with decreasing after the electrochemical reduction of CO , the L-ZnGa O
the pH value (increasing the proton concentration)(Figure S12c), catalyst displayed an obvious absorption step of ZnO, indicating
namely that increasing the concentration of proton produced the presence of the high-activity zinc ions on L-ZnGa
2
2+
O
4
during
-, CO
electrochemical reduction. The high-activity Zn /Zn will
+
the same phenomenon as increasing the concentration of HCO
3
2
-
in other words, during the reaction the role of HCO
3
can be be immediately oxidized after electrolysis like the behavior of
2
+ [40]
analogized to proton. Thus, we can make a conclusion that the Cu .
In addition, for the sake of excluding the catalytic
+
reduced Zn could not only facilitate the one-electron activation influence of oxygen vacancies that maybe present in the
-
of inert linear CO
2
molecule but also interact with HCO
3
ion to ZnGa
O
2 4
catalysts, electron paramagnetic resonance (EPR)
specimens,
acquire proton for accelerating the CO
2
conversion.
experiments were performed on both ZnGa O
2 4
however, it showed no signal of oxygen vacancy (Figure S14).
These results further confirmed that the Zn ion was the main
active species and responsible for the selective generation of
Power
V
HCO -
3
2
CO from CO .
+H+
(c)
+e
(d)
-
+H
+
e-
+CO
(b)
2
+
Zn2+
Zn+
CO -
CO + H O
2
COOH*
2
(
a)
a
40
30
20
b4
00
00
00
CO2 L-ZnGa2O4
CO2 H-ZnGa2O4
a
)
Z
n
2
+
+
e
-
→
Z
n
+
(
3
b) Zn+ + CO
→ Zn2+ + CO
-
(
(
2
2
-
→ H+ + CO
2-
c) HCO
3
3
2
-
+
N2 L-ZnGa2O4
N2 H-ZnGa2O4
CO
2
+ * + H → COOH*
d) COOH* + H + e → CO* + H
*, adsorbed sites)
+
-
(
2
O
10
100
0
(
Figure 4. Proposed CO
of Zn2+ driven by external circuit to form Zn . (b) Oxidation of Zn by CO
2
reduction mechanism over L-ZnGa
2
O
4
. (a) Reduction
0
20
40
60
ZRe ( )
80
100
0
100 200 300 400 500 600 700
+
+
ZRe ()
2
to
Zn2+ and formation of CO
-
-
c
0.80
0.75
2
radical. (c) CO
2
radical captures proton releasing
d
1.5
1.0
0.5
0.0
100
H-ZnGa2O4
L-ZnGa2O4
by HCO -
3
to form COOH*. (d) COOH* captures the electron from external
8
0
+
-
Cycle 1
Cycle 2
3
circuit and the H from HCO to generate CO.
0
0
0
0
.70
.65
.60
4
8 mV/dec
60
40
4
6 mV/dec
Based on the above mentioned procedures, we proposed a
brief schematic diagram for illustrating the reaction mechanism
2
0
0
.55
-
of L-ZnGa
O
2 4
during CO
2
conversion catalysis (Figure 4). Firstly,
1.4 -1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6
-2
0
2
4
6
8
10
Log (jCO / mA cm
)
the Zn2+ was reduced to Zn with applied potential, when the
+
Time (h)
+
CO
2
molecule reached the surface of electrode, the Zn was Figure 5. Nyquist impedance plots of as-prepared ZnGa O catalysts in a)
2
4
2
+
oxidized to Zn and transferred the electron to CO
2
molecule to CO -purged 0.1 M KHCO solution and b) N -purged 0.1 M KHCO solution.
2
3
2
3
-
-
form and stabilize the CO
radical captured the proton which dissociated from HCO
2
radical. Consequently, the CO
2
2 4 2 4
c) Tafel plots of as-prepared ZnGa O catalysts. d) Stability test of L-ZnGa O
-
3
ion to catalyst for current density measurement and faradaic efficiency of CO at -
facilitate the formation of crucial COOH* intermediate, then the 1.4 V.
COOH* intermediate coupled with proton and electron
continuously and produced CO molecule. Notably, the cyclic
It is well known that both courses of CO RR and HER require
2
2
+
+
reduction and oxidation course of Zn /Zn redox couple keep the consumption of proton, thus a pH gradient would be
-
the steady formation of CO
2
radical.
created near the surface of electrode during the reaction. The
2
+
+
To fully understand the key role of Zn /Zn on mesoporous structure has been confirmed that exhibiting the
electrochemical reduction over L-ZnGa catalyst, we special ability to increase alkalinity which produce within the
performed poisoning test by adding 10 mM porous network, therefore directly suppress the H evolution
2 4
O
a
2
ethylenediaminetetraacetic acid disodium salt (EDTA-2Na) into through the depletion of proton donors during CO RR
2
solution (pH = 6.33). EDTA-2Na, as an excellent process.[ Aiming to check the performance of L-ZnGa O
41]
0
.1 M KHCO
3
2 4
chelator for transition metal ions, is able to capture the zinc ions mesoporous structure on the alkalinity increase to inhibit the H2
[
38, 39]
on L-ZnGa
2 4
O .
As shown in Figure 3c, the oxidation peak at evolution, we conducted electrochemical impedance
-
1.03 V disappeared at all, indicating that the existence of EDTA- spectroscopy (EIS) on both L-ZnGa O and H-ZnGa O
2
4
2 4
2
CO
Na can obviously suppress the interactions among L-ZnGa
2
O
4
,
2 3
electrodes at -1.4 V. In the CO -saturated KHCO electrolyte, the
L-ZnGa O exhibited a slightly smaller radius of the Nyquist plot,
2 4
-
2
, and HCO
3
. Moreover, the onset potential of L-ZnGa
2
O
4
catalyst in EDTA-2Na contained electrolyte shifted about 250 indicative of a slightly lower charge transfer resistance for L-
mV toward more negative potential. The FE of CO product ZnGa O than H-ZnGa O (Figure 5a). Interestingly, in N -
2
4
2
4
2
provided further evidence to determine the critical function of saturated KHCO electrolyte, the L-ZnGa O catalyst exhibited a
3
2 4
Zn ion on ZnGa
of CO on L-ZnGa
decreased compared with that in the electrolyte without EDTA- hydrogen evolution reaction (HER) (Figure 5b). These facts
2 4
2 4
O catalyst. Over the applied potentials, the FE much bigger impedance arc than the H-ZnGa O , meaning that
2 4
2 4
O in EDTA-2Na contained electrolyte strikingly the L-ZnGa O catalyst has an enormous resistance for the
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 8
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Journal Name
ARTICLE
indicated that the L-ZnGa
interface reaction kinetics and the L-ZnGa
2
O
4
and H-ZnGa
2
O
4
follow the similar study demonstrated that the low-crystallinity materials may be
View Article Online
DOI: 10.1039/C9TA00562E
2
O
4
suppressed HER a family of great potential materials for selective CO
2
reduction
effectively not only due to the strong interactions between CO
2
4
4
following a redox couple catalytic mechanism.
2
+
+
and the easily released Zn /Zn on low-crystallinity L-ZnGa
but also on account of the mesoporous structure of ZnGa
2
O
O
2
Conflicts of interest
There are no conflicts to declare.
generates diffusional gradients through the local consumption
-
of proton donor such as HCO
3
to inhibit H
2
evolution. Tafel
slope is also an effective parameter in evaluating the catalytic
ability of catalyst in CO
2
electrochemical reduction.
-
Theoretically, for CO
stands for that the rate-limiting step is initial one-electron
2
reduction, the Tafel slope of 118 mV dec Acknowledgements
1
-1
2
reduction of CO , while the Tafel slope of 59 mV dec implies
This work was supported primarily by the National Natural
Science Foundation of China (51872135, 51572121, 21603098,
and 21633004), the Natural Science Foundation of Jiangsu
Province (BK20151265, BK20151383, and BK20150580) and
the Fundamental Research Funds for the Central Universities
that the rate-determining step is a chemical reaction step after
molecule.[ In our case,
12]
the one-electron activation of inert CO
the calculated Tafel slopes of the L-ZnGa
2
O
2 4
and H-ZnGa
2 4
O
-1
-1
catalysts were 46 mV dec and 48 mV dec , respectively (Figure
5
1
c). The Tafel slopes of both catalysts were close to 59 mV dec-
(021314380133 and 021314380084).
,
pointing to a mechanistic pathway in which the coupled
proton and electron transfer after the fast initial electron
transfer step is rate-determining step.[ This proves that the
12]
Notes and references
Zn ions on L-ZnGa
2
O
4
catalyst are able to activate CO
2
to the
1
2
3
4
5
Y. H. Chen, C. W. Li and M. W. Kanan, J. Am. Chem. Soc., 2012,
34, 19969.
J. L. Qiao, Y. Y. Liu, F. Hong and J. J. Zhang, Chem. Soc. Rev.,
2014, 43, 631.
Y. H. Wang, J. L. Liu, Y. F. Wang, A. M. Al-Enizi and G. F. Zheng,
Small, 2017, 13, 1701809.
K. P. Kuhl, E. R. Cave, D. N. Abram and T. F. Jaramillo, Energy
Environ. Sci., 2012, 5, 7050.
-
CO
2
intermediate with a low energy barrier. Therefore, a more
1
likely scenario was that the rate-determining step is the
-
protonation of CO
2
intermediate because of the first-order
-
kinetic dependence of the concentration of HCO
3
on oxidation
peak potential. Meanwhile, in the case of same overpotential,
the L-ZnGa
than H-ZnGa
2
O
4
exhibited a higher partial current density of CO
. These experiments successfully demonstrated
2
O
4
Z. Y. Zhang, M. F. Chi, G. M. Veith, P. F. Zhang, D. A. Lutterman,
J. Rosenthal, S. H. Overbury, S. Dai and H. Y. Zhu, ACS Catal.,
the intrinsic advantage of the low-crystallinity material with
porous structure on HER inhibition.
2
016, 6, 6255.
6
7
T. Takeshita and K. Yamaji, Energy Policy, 2008, 36, 2773.
J. L. DiMeglio and J. Rosenthal, J. Am. Chem. Soc., 2013, 135,
The stability of the L-ZnGa
.1 M KHCO electrolyte. For the current density measurement,
the electrolyte was purged with CO gas to maintain the
constant pH value during the reaction. As shown in Figure 5d,
2 4
O electrodes was evaluated in
0
3
8
798.
2
8
W. L. Zhu, R. Michalsky, O. Metin, H. F. Lv, S. J. Guo, C. J.
Wright, X. L. Sun, A. A. Peterson and S. H. Sun, J. Am. Chem.
Soc., 2013, 135, 16833.
-2
the current density nearly stayed at 0.5 mA cm at -1.4 V for 10
h, suggesting the long-term stability of L-ZnGa catalyst. The
FE was determined in a closed electrolytic cell with adding of
CO in a given interval. The L-ZnGa catalyst showed the high
9
1
M. Ma, K. Liu, J. Shen, R. Kas and W. A. Smith, ACS Energy Lett.,
2 4
O
2
018, 3, 1301.
0 D. F. Gao, H. Zhou, J. Wang, S. Miao, F. Yang, G. X. Wang, J. G.
2
O
2 4
Wang and X. H. Bao, J. Am. Chem. Soc., 2015, 137, 4288.
FEs at 70-96%. Although the FE of CO decreased with the 11 W. J. Zhu, L. Zhang, P. P. Yang, C. L. Hu, Z. B. Luo, X. X. Chang,
Z. J. Zhao and J. L. Gong, Angew. Chem. Int. Ed., 2018, 57,
consumption of CO
FE at the beginning of the second cycle confirmed the excellent
durability of L-ZnGa catalyst. Moreover, the TEM image and
SAED pattern (Figure S15) for L-ZnGa after 10 h
2
at the first cycle, the good repeatability of
1
1544.
1
1
2 D. D. Zhu, J. L. Liu and S. Z. Qiao, Adv. Mater., 2016, 28, 3423.
3 B. H. Qin, Y. H. Li, H. Q. Fu, H. J. Wang, S. Z. Chen, Z. L. Liu and
F. Peng, ACS Appl. Mater. Interfaces., 2018, 10, 20530.
2 4
O
2 4
O
electrochemical reaction showed no obvious change, further 14 J. Rosen, G. S. Hutchings, Q. Lu, R. V. Forest, A. Moore and F.
confirming the stability of the catalyst.
Jiao, ACS Catal., 2015, 5, 4586.
1
1
1
1
5 T. T. Zhang, X. F. Li, Y. L. Qiu, P. P. Su, W. B. Xu, H. X. Zhong and
H. M. Zhang, J. Catal., 2018, 357, 154.
6 Z. P. Chen, K. W. Mou, S. Y. Yao and L. C. Liu, ChemSusChem,
Conclusions
2
018, 11, 2944.
7 F. Yang, P. Song, X. Z. Liu, B. B. Mei, W. Xing, Z. Jiang, L. Gu and
W. L. Xu, Angew. Chem. Int. Ed., 2018, 130, 12483.
2 4
In summary, low-crystallinity mesoporous ZnGa O was
proposed as an effective electrocatalyst for selective reduction
8 E. S. Donovan, B. M. Barry, C. A. Larsen, M. N. Wirtz, W. E.
Geiger and R. A. Kemp, Chem. Commun., 2016, 52, 1685.
of CO
2
to CO. The low crystallinity makes the release of active
2
+
+
Zn /Zn redox couple more easily than high-crystallinity one 19 G. B. Chan, Y. F. Zhao, L. Shang, G. I. N. Waterhouse, X. F. Kang,
L. Z. Wu, C. H. Tung and T. R. Zhang, Adv. Sci., 2016, 3, 1500424.
0 Y. F. Zhao, G. B. Chen, T. Bian, C. Zhao, G. I. N. Waterhouse, L.
Z. Wu, C. H. Tung, L. J. Smith, D. O’Hare and T. R. Zhang, Adv.
Mater., 2015, 27, 7824.
due to the weak lattice constraint. Mesoporous structure with
high specific area is beneficial to the CO adsorption, HER
2
2
+
2
+
inhibition and offers more active Zn /Zn redox couple to
activate CO . As a result, CO product with 96 % faradaic
2
2
1 S. C. Yan, S. X. Ouyang, J. Gao, M. Yang, J. Y. Feng, X. X. Fan, L.
efficiency was selectively produced at -1.4 V vs. Ag/AgCl. This
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 8 of 8
ARTICLE
Journal Name
J. Wan, Z. S. Li, J. H. Ye, Y. Zhou and Z. G. Zou, Angew. Chem.
Int. Ed., 2010, 49, 6400.
View Article Online
DOI: 10.1039/C9TA00562E
2
2
2 A. L. Chaffee, Fuel Process. Technol., 2005, 86, 1473.
3 Y. L. Wang, P. F. Hou, Z. Wang and P. Kang, ChemPhysChem
2
017, 18, 3142.
2
2
2
4 E. S. Donovan, B. M. Barry, C. A. Larsen, M. N. Wirtz, W. E.
Geiger and R. A. Kemp, Chem. Commun., 2016, 52, 1685.
5 A. K. Gupta, A. Dhir and C. P. Pradeep, Inorg. Chem., 2016, 55,
7
492.
6 D. B. Chu, G. X. Qin, X. M. Yuan, M. Xu, P. Zheng and J. Lu,
ChemSusChem, 2008, 1, 205.
2
2
7 L. W. Lai and C. T. Lee, Mater. Chem. Phys., 2008, 110, 393.
8 A. S. Alshammari, L. N. Chi, X. P. Chen, A. Bagabas, D. Kramer,
A. Alromaeh and Z. Jiang, RSC Adv., 2015, 5, 27690.
9 D. S. Li, M. Sumiya, S. Fuke, D. R. Yang, D. L. Que, Y. Suzuki and
Y. Fukuda, J. Appl. Phys., 2001, 90, 4219.
0 X. Y. Deng, A. Verdaguer, T. Herranz, C. Weis, H. Bluhm and M.
Salmeron, Langmuir, 2008, 24, 9474.
2
3
3
3
3
1 Y. Li, P. F. Zhu and R. X. Zhou, Appl. Surf. Sci., 2008, 254, 2609.
2 J. M. Vohs and M. A. Barteau, Langmuir, 1989, 5, 965.
3 L. Jappinen, T. Jalkanen, B. Sieber, A. Addad, M. Heinonen, E.
Kukk, I. Radevici, P. Paturi, M. Peurla, M. A. Shahbazi, H. A.
Santos, R. Boukherroub, H. Santos, M. Lastusaari and J.
Salonen, Nanoscale Res. Lett., 2016, 11, 413.
3
3
4 J. Q. Qin, X. Y. Zhang, C. W. Yang, M. Cao, M. Z. Ma and R. P.
Liu, Appl. Surf. Sci., 2017, 392, 196.
5 D. P. Dubal, N. R. Chodankar, Z. Caban-Huertas, F. Wolfart, M.
Vidotti, R. Holze, C. D. Lokhande and P. Gomez-Romero, J.
Power Sources., 2016, 308, 158.
3
6 M. T. Martinez, M. A. Callejas, A. M. Benito, M. Cochet, T.
Seeger, A. Ansón, J. Schreiber, C. Gordon, C. Marhic, O.
Chauvet, J. L. G. Fierro and W. K. Maser, Carbon, 2003, 41,
2
247.
3
3
7 Y. Hori, Modern Aspects of Electrochemistry, 2008, 42, 89.
8 H. Y. Jia, Z. Li, X. L. Wang and Z. Zheng, J. Mater. Chem. A.,
2
015, 3, 1158.
3
4
4
9 C. S. He, W. P. Zhu, Y. F. Xu, Y. Zhong, J. Zhou and X. H. Qian, J.
Mater. Chem., 2010, 20, 10755.
0 D. Kim, C. S. Kley, Y. F. Li and P. D. Yang, Proc. Natl. Acad. Sci.,
2
017, 114, 10560.
1 A. S. Hall, Y. Yoon, A. Wuttig and Y. Surendranath, J. Am.
Chem. Soc., 2015, 137, 14834.
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins