Enhanced Activity and Stability of Carbon-Decorated Cuprous Oxide Mesoporous Nanorods for CO2 Reduction in Artificial Photosynthesis

Article abstract of DOI:10.1021/acscatal.6b01455

The development of photocatalysts with superior activity and stability to produce organic fuels through CO₂ reduction under renewable sunlight is of great significance due to the depletion of fossil fuels and severe environmental problems. In this study, we presented a "hitting three birds with one stone" strategy to synthesize carbon layer coated cuprous oxide (Cu₂O) mesoporous nanorods on Cu foils via a facile chemical oxidation and subsequent carbonization method. The thin carbon layer not only works as a protective layer to quench the common photocorrosion problem of Cu₂O but also endows the sample a mesoporous and one-dimensional nanorod structure, which can facilitate reactant molecule adsorption and charge carrier transfer. Substantially, the coated samples exhibited remarkably improved stability as well as decent activity for CO₂ reduction under visible light irradiation. The optimized sample attained an apparent quantum efficiency of 2.07% for CH₄ and C₂H₄ at λ₀ 400 nm, and 93% activity remained after six photoreduction cycles under visible light. This work provides a facile strategy to address the stability and activity issues of Cu₂O under visible light irradiation, which is presumably suitable for other semiconductors as promising candidates for CO₂ reduction in artificial photosynthesis.

Full text of DOI:10.1021/acscatal.6b01455

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Enhanced Activity and Stability of Carbon-Decorated Cuprous Oxide
Mesoporous Nanorods for CO2 Reduction in Artificial Photosynthesis
Luo Yu, Guojian Li, Xiaoshu Zhang, Xin Ba, Guodong Shi,
Yong Li, Po Keung Wong, Jimmy Chai Mei Yu, and Ying Yu
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b01455 • Publication Date (Web): 22 Aug 2016
Downloaded from http://pubs.acs.org on August 22, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
Enhanced Activity and Stability of Carbon-Decorated
Cuprous Oxide Mesoporous Nanorods for CO2
Reduction in Artificial Photosynthesis
Luo Yua, Guojian Lia, Xiaoshu Zhanga, Xin Baa, Guodong Shia, Yong Lia, Po Keung
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Wongb, Jimmy C. Yuc and Ying Yua*
a College of Physical Science and Technology, Central China Normal University,
Wuhan 430079, China
b School of Life Science, The Chinese University of Hong Kong, Shatin, NT, Hong
Kong SAR, China
c Department of Chemistry, The Chinese University of Hong Kong, Shatin, NT, Hong
Kong SAR, China
* Corresponding author. Tel: 86ꢀ27ꢀ67867037; email: yuying01@mail.ccnu.edu.cn
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 41
1
2
3
4
ABSTRACT:
5
6
7
8
The development of photocatalysts with superior activity and stability to produce
organic fuels through CO2 reduction under renewable sunlight is of great significance
due to the depletion of fossil fuels and severe environmental problems. In this study,
we presented a “hitting three birds with one stone” strategy to synthesize carbon layer
coated cuprous oxide (Cu2O) mesoporous nanorods on Cu foils via a facile chemical
oxidation and subsequent carbonization method. The thin carbon layer not only works
as a protective layer to quench the common photocorrosion problem of Cu2O, but also
endows the sample a mesoporous and oneꢀdimensional nanorod structure, which can
facilitate reactant molecule adsorption and charge carrier transfer. Substantially, the
coated samples exhibited remarkably improved stability as well as decent activity for
CO2 reduction under visible light irradiation. The optimized sample attained an
apparent quantum efficiency of 2.07% for CH4 and C2H4 at λ0 = 400 nm, and 93%
activity remained after six photoreduction cycles under visible light. This work
provides a facile strategy to address the stability and activity issues of Cu2O under
visible light irradiation, which is presumably suitable for other semiconductors as
promising candidates for CO2 reduction in artificial photosynthesis.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS: Cu2O mesoporous nanorods, carbon coating, photocatalysis, CO2
reduction, stability, visible light
2
ACS Paragon Plus Environment

Page 3 of 41
ACS Catalysis
1
2
3
4
1. INTRODUCTION
5
6
7
8
9
The deepening crisis of greenhouse effect and depletion of fossil resources are
becoming enormous threats to sustainable development. To convert CO2 into fuel is
considered to be a winꢀwin strategy since it has the potential not only to generate
valuable fuels but also to mitigate the negative effect of anthropogenic CO2
overꢀemission.1ꢀ3 Noticeably, artificial photosynthesis has been regarded as a
promising way to achieve CO2 reduction for fuel production by using solar energy.4
Beyond doubt, photocatalysts play vital roles in the conversion process so that
developing such materials with good activity and stability is of great significance.
Since the pioneering work of photoreduction of CO2 to organic compounds by
Inoue et al.,5 as one of the earliest studied photocatalysts, titanium dioxide (TiO2) has
been proverbially investigated due to its lowꢀcost, nontoxicity and high stability.6,7
Notwithstanding, the narrow scope of light absorption (< 387 nm) and high
lightꢀinduced electronꢀhole recombination rate restrict its photocatalytic
performance.8 Therefore, extensive research efforts have been made to explore other
suitable semiconductor materials as efficient photocatalysts for CO2 conversion. To
date, multifarious materials including metal complexes,9,10 organic compounds,11
metal oxide,12ꢀ14 metal sulfides,15 graphitic compounds,16 and their composites17ꢀ20
have been reported. Among these materials, cuprous oxide (Cu2O) is considered to be
one of the most promising candidates for solar energy conversion. It is known that
Cu2O is a semiconductor with direct band gap of about 2.0 eV, which guarantees
efficient visible light absorption.21 Moreover, Cu2O possesses favorable energy band
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 41
1
2
3
4
5
6
7
8
9
positions with the conduction band lying up CO2 reduction potential, which facilitates
the reduction of CO2 to hydrocarbon fuels.22 Additionally, Cu2O is abundant, lowꢀcost
and environmentally friendly.23
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It is generally acknowledged that the photocatalytic reduction of CO2 mainly
involves charge carrier generation, charge carrier transfer, and finally, multielectron
chemical reduction on particular active sites.4 Since Cu2O is responsive to visible
light, the first step of the charge carrier generation is efficiently achievable under
visible light. Whereas, on account of poor conductivity, the process of charge carrier
transfer is severely confined for this photocatalyst, and vast electronꢀhole pairs
recombine during their migration to the surface.24 The last step of reduction on active
sites is not only compromised by the recombination of electronꢀhole pairs, but also
related to the structure of Cu2O. Besides, the photocorrosion problem is an inherent
drawback for Cu2Oꢀbased photocatalysts, in which the cuprous ion (Cu+) is highly
prone to be oxidized to cupric ion (Cu2+) or reduced to copper (Cu).25 The unstable
character for Cu2O hampers its practical application severely.
To ameliorate the performance of Cu2O in photocatalytic reactions, the strategies
including doping,26 structure engineering,27 exposed facet engineering,28 and
semiconductor compositing29 etc. have been employed. Substantially,
oneꢀdimensional (1ꢀD) nanostructures, including nanowire, nanorod and nanotube, are
highly attractive building blocks owing to their inherent anisotropies and electron
efficient transport within the smallest dimension.30,31 Cu2O nanowires has been
reported to have the enhanced activity for water splitting.32 In addition, protective
4
ACS Paragon Plus Environment

Page 5 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
layer coating is also one of the most effective and handy solutions to improve stability
as well as conductivity for Cu2O. As a promising material widely applied in energy
conversion field, carbon is so appropriate to serve as a coating layer for its unique
physicochemical properties, such as strong mechanical strength, nontoxicity, excellent
electronic and thermal conductivity, and solution processing feasibility.33 Recently, it
was reported that a sandwichꢀlike structure with NiFe nanoparticles encapsulated in
coupled graphene sheets (NiFe/C) showed oxygen evolution with high efficiency,34
since the coupled graphene sheets around active catalysts accelerated electron transfer
and prevented NiFe nanoparticles from aggregation during longꢀtime electrochemical
reactions. Additionally, CdS petalous nanostructures coated with a thin carbon layer
displayed remarkably improved photostability and activity due to that the carbon layer
not only significantly reduced the amount of surface traps, but also provided perfect
chemical protection for inner CdS nanoparticles.35 Furthermore, porous structure is
extraordinarily beneficial in a photocatalytic process for it can provide large specific
area and many active sites, which facilitates reactant molecule adsorption and
reaction.24 In particular, improving the CO2 adsorption ability of photocatalysts has
been regarded as an efficient approach to enhance CO2 conversion efficiency. To the
best of our knowledge, so far, there has been no report about the combination of 1ꢀD
nanostructure, carbon coating and mesoporous structure for Cu2O as a photocatalyst
for artificial photosynthesis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Herein, we presented a facile and economical approach to synthesize carbon layer
coated cuprous oxide mesoporous nanorods (CCMNRs) for CO2 reduction under
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 41
1
2
3
4
5
6
7
8
9
visible light. Firstly, Cu(OH)2 nanotubes (NTs) were synthesized directly on a Cu foil
through a handy method of chemical oxidation. Subsequently, the obtained Cu(OH)2
NTs were dipped into glucose solution, which was used as a green carbon source.
Then the glucose coated Cu(OH)2 NTs were calcined in N2 atmosphere to obtain
CCMNRs. This facial “hitting three birds with one stone” strategy endowed the
samples with three characteristics: coating protective carbon layer, mesoporous
structure and 1ꢀD nanorod structure, all of which brought about the favorably robust
performance and superior stability for asꢀprepared CCMNRs as the photocatalysts
used for CO2 reduction under visible light.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. EXPERIMENTAL SECTION
2.1. Materials synthesis. All of the chemicals used in this study and copper foil
with purity of 99.98% were purchased from Shanghai Guoyao Chemicals Ltd. Co.
The chemicals were analytical reagents and used without further purification.
Deionized water was used for the preparation of all aqueous solutions. In this study,
Cu(OH)2 NTs was synthesized through a modified chemical oxidation method,36 and a
typical process was as follows. A piece of Cu foil with a size of 2 × 5 cm2 and a
thickness of 0.2 mm was cleaned in hydrochloric acid (37%) and then ultrasonically
in acetone and ethanol for 5 min sequentially. The cleaned Cu foil was then immersed
into 80 mL aqueous solution containing 2.5 M NaOH and 0.125 M (NH4)2S2O8 for 20
min. Notably, a temperature of about 3℃ remained for the solution during the
reaction to prevent Cu(OH)2 NTs from agglomeration. Finally, the Cu foil with a light
6
ACS Paragon Plus Environment

Page 7 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
blue color was taken out from the solution, rinsed with deionized water, and dried in
air. Afterward, the prepared Cu(OH)2 NTs had immerged into an aqueous glucose
solution for about 12 h. Then, they were dried at ambient condition, and annealed at
500℃ in N2 atmosphere for 2 h to form CCMNRs. The concentration of the glucose
solution was varied to control the carbon layer thickness of the CCMNRs. For
comparison, without glucose, the asꢀprepared Cu(OH)2 NTs was directly annealed at
500℃ in N2 atmosphere for 2 h, which was labeled with Cu2O.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2. Materials characterization. The morphology of the prepared samples was
detected with scanning electron microscopy (SEM) with the model of JEOL
JSMꢀ6700, transmission electron microscopy (TEM) and darkꢀfield scanning
transmission electron microscopy (DFꢀSTEM) with the model of JEOL JEMꢀ2100F
coupled with an energy dispersive Xꢀray spectroscopy (EDS), operated with an
accelerating voltage of 200 kV. Xꢀray powder diffraction (XRD) patterns of the
samples were analyzed by using a PANalytical diffractometer (D/max 40kv) using Cu
Kα radiation (λ=0.154598 nm) for crystalline phase. The surface Xꢀray diffraction
(SXRD) measurements for thin film sample were performed in theta/theta mode with
a step width of 0.02° and a scan rate 100 s/step. The surface chemical analysis of the
samples was achieved with Xꢀray Photoelectron Spectroscopy (XPS) (VG
Multiabꢀ2000) using a PHI Quantum 2000 XPS system with a monochromatic Al Kα
source and charge neutralizer. All of the spectra were calibrated to C1s peak at 284.6
eV. The UVꢀvis diffuse reflectance spectra (UVꢀvis DRS) were recorded with
PerkinElmer Lambda 35 Spectrophotometer in the range of 400ꢀ800 nm using BaSO4
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 41
1
2
3
4
5
6
7
8
9
as a reference. Raman spectra were collected at room temperature using a
LabRAMHR Raman system under Ar+ (532 nm) laser excitation. Micromeritics
ASAP2020 nitrogen adsorption apparatus (USA) was used to characterize the
BrunauerꢀEmmettꢀTeller (BET) specific surface area of the samples. CO2 adsorption
isotherms were measured at room temperature using a BEL SORPꢀmini II (BEL
Japan). Pulsed CO chemisorption capacity for the different samples was tested using a
BELCATꢀBꢀ293 Catalyst Analyzer (BEL Japan), equipped with a thermal
conductivity detector (TCD) to measure CO consumption. Generally, each sample
(63.6 mg) was placed in a Uꢀtype quartz tube and pretreated at 573 K with a helium
flow (50 cm3 min−1) for 20 min and then maintained in the helium atmosphere at the
same temperature for 15 min. After pretreatment, the sample was cooled down to 323
K and maintained at the temperature for 10 min. Then CO (CO/He, 10%) pulses were
injected from a constant volume loop valve into the Uꢀtype quartz tube. CO
adsorption was considered to be completed after two or three peaks showed the same
peak areas, and the difference of first peak area and last peak area was sample
adsorption amount. Moreover, a PerkinꢀElemer 4500 fluorescence spectrophotometer
was used to analyze the photoluminescence (PL) properties of the samples with an
excitation wavelength of 310 nm at room temperature.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.3. Electrochemical measurements. Electrochemical measurements were
carried out on an electrochemical station (PARSTAT 2273, Princeton) in a standard
three electrode system with the prepared samples as working electrode, a platinum
plate as the counter and standard Ag/AgCl electrode as the reference. A 0.02 M
8
ACS Paragon Plus Environment

Page 9 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
Na2SO4 aqueous solution was used as the electrolyte. The photocurrent of the samples
was measured under a chopped irradiation from a 350W Xenon lamp (Lap Pu, XQ)
with a 420 nm cutoff filter right above the reactor (light on/off cycle: 20 s). The
output current for the lamp was controlled to be 15 A with an average intensity of
irradiation 97.3 mW/cm2 determined by a light meter (LIꢀCOR Biosciences, LIꢀ250A).
The scan rate for the linear sweep voltammetry (LSV) was 2 mV sꢀ1. The
MottꢀSchottky plots for the samples were obtained at a frequency of 1.0 KHz, and the
electrochemical impedance spectroscopy (EIS) was conducted at a frequency from 0.1
Hz to 100 KHz.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.4. Photocatalytic performance tests. The photocatalytic CO2 reduction was
conducted in a homemade 250 mL sealed cell with 170 mL of 0.1 M KHCO3 as
electrolyte solution.22 No wire was connected between electrode and sample, and no
closed circuit was formed during every photocatalytic CO2 reduction test. Before
reduction, high purity CO2 (99.99%) gas was passed through the electrolyte solution
for 30 min until the CO2 concentration reached saturation and dissolved oxygen was
removed completely. A 350 W Xenon lamp with a 420 nm cutoff filter was used as
light source, which was positioned 24 cm away from the reactor to maintain the
photocatalytic reaction temperature. The output current for the lamp was controlled to
be 8 A with an average intensity of irradiation 1.16 mW/cm2. All the experiments
were performed at room temperature and ambient pressure. The gaseous products
evolved during CO2 photocatalytic reduction were collected by a 1ꢁm syringe and
then immediately analyzed by two gas chromatograph instruments. One is GCꢀ2014
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 41
1
2
3
4
5
6
7
8
9
(Serial No. C11484403556 SA, SHIMADZU, Japan) equipped with a flame
ionization detector (FID) for organic compound detection, the other is GCꢀ2014
(Serial No. C11485013433 SA, SHIMADZU, Japan) with a thermal conductivity
detector (TCD) for analysis of inorganic gases. For GCꢀ2014 with FID, a packed
column (PROPAKꢀQ, 2 m × 3 mm) was used, and for GCꢀ2014 with TCD was a
different packed column (C36880ꢀ14, Restek, 2 m × 1 mm). The carrier gas was
nitrogen (N2). The amount of the evolved products was the average value of three
measurements. Control experiments were performed under identical conditions but in
the absence of CO2 (under N2 atmosphere) or light irradiation. Additionally, the
activity of a blank Cu foil and a carbon coated Cu foil (CꢀCu) annealed under the
same condition were measured in a similar way. For the stability measurements, after
each cycle of photoreduction for 3 h, the sample was taken out of the cell and then put
into it again for the next cycle reaction under the same reaction conditions. Before
each cycle, fresh CO2 and electrolyte periodically replaced the used ones. The gaseous
products collected for isotopic measurement with a gas chromatographyꢀmass
spectrometry (Pfeiffer OmniStar) were evolved with the same process mentioned
above except that the carbon source of the high purity CO2 was taken the place of by
isotope 13CO2 (13C 99%, SigmaꢀAldrich).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3. RESULTS AND DISCUSSION
Scheme 1 illustrates the schematic representation of the preparation process of
CCMNRs. Briefly, Cu(OH)2 NTs was synthesized through oneꢀstep chemical
10
ACS Paragon Plus Environment

Page 11 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
oxidation of a Cu foil, which was then followed by glucoseꢀcoating and thermal
reduction to form CCMNRs. The conversion of Cu(OH)2 to Cu2O was achieved by
dehydration of Cu(OH)2 to CuO, followed by oxygen removal from the lattice of CuO
to form Cu2O at high temperature in N2 atmosphere.37 Figure S1 (Supporting
Information) showed the Xꢀray diffraction (XRD) pattern of asꢀprepared Cu(OH)2,
which confirmed pure Cu(OH)2 crystal phase without identified Cu2O and CuO.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Schematic illustration of the formation of pure Cu2O and carbon layer
coated Cu2O mesoporous nanorods (CCMNRs).
A series of carbon coated samples were synthesized with different amount of
glucose. Specifically, 1, 2, 3, 4 and 5 mg mlꢀ1 of glucose were used, and the samples
were denoted with Cꢀx/Cu2O, in which x represented glucose solution concentration
(mg mLꢀ1). Typical XRD patterns recorded from Cꢀ2/Cu2O, along with pure Cu2O and
Cu foil (annealed under the same condition) are shown in Figure 1a. It can be clearly
found that besides three peaks for elemental Cu resulted from the Cu substrate, the
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 41
1
2
3
4
5
6
7
8
9
others were all assigned to Cu2O phase with a strong (111) orientation (PDF
No.78ꢀ2076), and no peaks attributed to CuO or Cu(OH)2 was identified. Figure 1b
presents the comparison of the XRD patterns in 15° ~ 29.5° range. Obviously, broad
peak around 23° was observed for Cꢀ2/Cu2O, which was ascribed to carbon species.24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (a) XRD patterns and (b) corresponding enlarged parts in the range of 15° ~
29.5° (2θ) for Cu foil, pure Cu2O and Cꢀ2/Cu2O. (c) Raman spectra of pure Cu2O and
Cꢀ2/Cu2O. High resolution XPS spectra of (d) C 1s, (e) Cu 2p, and (f) O 1s for
Cꢀ2/Cu2O.
Raman spectra were displayed in Figure 1c to further confirm the presence of carbon
in Cꢀ2/Cu2O. Distinct D and G band located at 1360 and 1590 cmꢀ1 was present,
corresponding to amorphous and graphite carbon respectively,38 which manifested
that carbon was successfully decorated on the surface of Cꢀ2/Cu2O. Additional peaks
at 148, 218, 412, and 626 cmꢀ1 were assigned to vibrational modes of Cu2O.39
12
ACS Paragon Plus Environment

Page 13 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
Highꢀresolution XPS spectra for C 1s, Cu 2p, and O 1s are depicted in Figures 1d,
1c and 1f, respectively to verify the elemental composition and chemical state in
Cꢀ2/Cu2O. In Figure 1d, the strong CꢀC peak at 285.0 eV implied the formation of a
carbon layer on the Cu2O surface. The other three peaks with lower intensity at 286.4,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2ꢀ
288.7, and 289.5 eV belonged to CꢀOH, OꢀC=O, and CO3 respectively, which was
due to the presence of the adsorbed oxygen functional groups on the carbon layer.40
From Figure 1e, the sharp and symmetrical peak centered at 932.5 and 952.5 eV was
found, corresponding to Cu+ 2p3/2 and Cu+ 2p1/2 respectively. No evident satellite peak
was observed, indicating the absence of Cu2+. From the O 1s core level spectrum in
Figure 1f, the same oxygen functional groups adsorbed on the carbon layer were
detected. The existence of hydroxyl groups on the surface indicated that H2O
molecules were easily accessible to the surface, which might lead to higher activity
for CO2 photoreduction.41 The XPS survey spectrum of Cꢀ2/Cu2O is shown in Figure
S2 (Supporting Information), from which no evidence was discovered for the
formation of a CuꢀC bond. So, the above results verified that carbon decorated Cu2O
samples were successfully prepared via the calcination of Cu(OH)2 precursors treated
with glucose.
Scanning electron microscopy (SEM) images in Figures 2a~2c reveal the
morphology of the prepared samples. Figure 2a showed that the Cu(OH)2 NTs covered
almost all surface of copper substrate uniformly and compactly. Pure Cu2O thin films
grown on Cu foil were prepared when the Cu(OH)2 NTs was directly annealed in N2
atmosphere without glucoseꢀcoating. As the SEM images with different magnification
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 41
1
2
3
4
5
6
7
8
9
shown in Figure 2b, a nanorodꢀlike morphology composed of aggregated and irregular
nanoparticles with a large size of about 200 nm was observed for pure Cu2O. The fine
1ꢀD nanotube structure disappeared compared with Cu(OH)2 precursor and the
composed particle size was large.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. SEM images for (a) Cu(OH)2 NTs, (b) pure Cu2O, and (c) Cꢀ2/Cu2O (the
insets are those with high magnifications). TEM images of Cꢀ2/Cu2O in (d) low
magnification (the inset is the detailed image of red dashed part) and (e) high
magnification. (f) HRTEM image of Cꢀ2/Cu2O. (g) DFꢀSTEM image of Cꢀ2/Cu2O
and corresponding elemental mapping images of C, Cu and O.
When calcined in N2 after glucoseꢀcoating, the Cu(OH)2 NTs were transformed
into CCMNRs. The SEM images in Figure 2c displayed that the 1ꢀD structure still
14
ACS Paragon Plus Environment

Page 15 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
remained for the prepared samples. More importantly, the Cꢀ2/Cu2O sample exhibited
a rough but dense surface, which might provide a large specific surface area to trap
the incident light efficiently by internal multiscattering.42 Additionally, the dense
carbon coated nanorods could be advantageous for enhanced photostability in
electrolyte solution. The SEM images of other Cꢀx/Cu2O with different glucose
concentration are presented in Figure S3 (Supporting Information), from which it can
be clearly seen that all the samples possessed 1ꢀD structure as well and the surface of
the samples became denser with the increase of glucose concentration. This was
attributed to the fact that the thickness of the carbon layer on the surface of CCMNRs
increased when increasing glucose concentration, which was validated by the
HRTEM images of Cꢀx/Cu2O with different glucose concentration (Figure S3 in
Supporting Information). The different morphology of pure Cu2O and Cꢀx/Cu2O
indicated that the carbon layer coated on the surface was paramount for the original
1ꢀD morphology retention of Cu(OH)2 NT precursors.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Transmission electron microscopy (TEM) images for Cꢀ2/Cu2O are presented in
Figures 2d~2f to explore the nanostructure and amorphous carbon on the surface. As
shown in Figure 2d, quintessential nanorod morphology with a length of about 2.2 ꢁm
and a width of about 300 nm was distinctly observed. Such a large aspect ratio can be
highly in favor of electron diffusion.43 The insert in Figure 2d was the close
observation to the end of a single nanorod, from which mesoporous structure was
revealed and a number of mesopores were distributed widely within Cꢀ2/Cu2O. From
Figure 2e, it can be seen that the nanorod was composed of nanoparticles with a
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 41
1
2
3
4
5
6
7
8
9
diameter of about 20 nm, which was much smaller than that of pure Cu2O.
Highꢀresolution TEM (HRTEM) (Figure 2f) displayed that an amorphous layer with a
thickness of about 5 nm was delicately coated on the surface of Cꢀ2/Cu2O, as shown
in the marked red line. Typical lattice fringes with an interplanar spacing of 0.242 nm
corresponding to the (111) plane for Cu2O were detected for Cꢀ2/Cu2O, which was in
conformity with the XRD result in Figure 1a. Figures 2g1~2g4 showed the darkꢀfield
scanning transmission electron microscopy (DFꢀSTEM) and corresponding elemental
mapping images of a single nanorod, confirming the presence and uniform
distribution of element Cu, O, and C. Figure S4 (Supporting Information) displays a
DFꢀSTEM image of another nanorod and corresponding elemental mapping images
for Cꢀ2/Cu2O, which further demonstrated that the nanorod was consisted of
nanoparticles and there was an uniform coverage of carbon on the surface of Cu2O
nanorods distinctly. Furthermore, pulsed CO chemisorption was performed to
measure CO consumption for estimating the coverage extent of carbon layer. Pure
Cu2O sample showed CO chemisorption with an amount of 0.113 cm3 gꢀ1, consistent
with the report that Cu2O is capable of CO adsorption.44 However, both Cꢀ1/Cu2O and
Cꢀ2/Cu2O showed no CO chemisorption ability, which indicated that the carbon layer
totally covered the surface of Cu2O nanorods, and thus the coverage extent was
almost 100%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To further validate the porous structure of the Cꢀ2/Cu2O sample, N2
adsorptionꢀdesorption isotherms and pore diameter distributions are particularly
presented in Figure 3a. A typical mesoporous nature was determined, which was in
16
ACS Paragon Plus Environment

Page 17 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. (a) N2 adsorptionꢀdesorption isotherm curves, (b) CO2 adsorption isotherm
curves and (c) UVꢀvis diffuse reflectance spectra for pure Cu2O and Cꢀ2/Cu2O. The
insert in (a) shows the corresponding pore size distribution, and that in (c) is the plots
of [F(R)]2 versus photo energy.
accordance with a typeꢀIV H2 hysteresis loop curve for Cꢀ2/Cu2O. In contrast, no
porous characteristic was observed for pure Cu2O with surface area around 2.0 m2 gꢀ1,
which was much smaller than Cꢀ2/Cu2O with surface area of 58.93 m2 gꢀ1. The reason
was largely attributed to the carbon layer, which was beneficial to retain the original
nanostructure of Cu(OH)2 NTs. In the process of thermal reduction, melted glucose
permeated into the voids on the surface of Cu(OH)2 NTs and might be partially
pyrolyzed to amorphous carbon.38 With the presence of carbon layer, Cꢀ2/Cu2O
nanorods maintained the mesoporous nanostructure. However, without the protection
of carbon, the growth and aggregation of nanograins was not suppressed, leading to
the formation of large Cu2O particles with the deterioration of porous structure. The
insert plots were pore size diameter distribution of the samples, and a pore size
distribution in the range of 2~20 nm was detected for Cꢀ2/Cu2O, which was in
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 41
1
2
3
4
5
6
7
8
9
agreement with the mesopores shown in the TEM image (Figure 2d). Nevertheless,
there was no obvious pore size distribution for pure Cu2O. The large specific surface
area and mesoporous structure for Cꢀ2/Cu2O facilitated the adsorption of reactant
molecules,45 which was confirmed by the CO2 adsorption behavior at room
temperature. It is found in Figure 3b that the maximum CO2 uptake for Cꢀ2/Cu2O was
about 13.4 cm3 gꢀ1, which was much higher than that of 2.2 cm3 gꢀ1 for pure Cu2O.
UVꢀvis diffuse reflectance spectrum (DRS) presented in Figure 3c was used to
probe the optical absorption property and band structure of the materials. The two
samples demonstrated excellent visible light absorption in the range of less than 640
nm. Compared with pure Cu2O, Cꢀ2/Cu2O showed an apparent decreased absorption
in visible light region, which implied that the carbon layer prevented the Cu2O NRs
from light harvesting to some extent. This phenomenon was also observed for other
Cꢀx/Cu2O samples as shown in Figure S5 (Supporting Information). Thus, with the
increase of carbon thickness, superior conductivity and high strength were achieved
whereas light absorption in visible light region was sacrificed. From a combined
viewpoint, there existed an optimal carbon thickness, which could bring supreme
active performance and stability for CCMNRs as photocatalysts in CO2 reduction. In
this study, the Cꢀ2/Cu2O sample with a thickness of about 5 nm was the optimized one,
which would be verified by electrochemical tests below. According to KubelkaꢀMunk
function, the DRS measurements were transformed and plotted against energy of light
(insert in Figure 3c),46 from which the same band gap values of 1.93 eV for pure
Cu2O and Cꢀ2/Cu2O were estimated, indicating that no band gap change was observed
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18
ACS Paragon Plus Environment

Page 19 of 41
ACS Catalysis
1
2
3
4
5
6
with the carbon layer on the surface of Cu2O. It further implied that there was no
carbon doped into the Cu2O lattice, which agreed well with the XPS results.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Photocurrent density of (a) pure Cu2O and (b) Cꢀ2/Cu2O with light on/off
cycles under visible light irradiation. (c) Enlarged photocurrent density plots of the first
200 s for pure Cu2O and Cꢀ2/Cu2O. (d) MottꢀSchottky plots of pure Cu2O and
Cꢀ2/Cu2O measured in an aqueous solution of Na2SO4 (0.02 M). (e) Linear sweep
voltammetry (LSV) curves and (f) electrochemical impedance spectra (EIS) measured
in an aqueous solution of Na2SO4 (0.02 M) under dark and illumination conditions for
pure Cu2O and Cꢀ2/Cu2O.
Figures 4a and 4b show the photocurrent density for pure Cu2O and Cꢀ2/Cu2O
respectively. It can be found that pure Cu2O and Cꢀ2/Cu2O owned a photocurrent
density of about 2.0 and 2.9 mA cmꢀ2 respectively under the first irradiation cycle. The
enhanced photocurrent for Cꢀ2/Cu2O was ascribed to the superior conductivity
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 41
1
2
3
4
5
6
7
8
9
contributed by the coated carbon layer and efficient charge carrier transfer offered by
the 1ꢀD nanostructure. Whereas, the photocurrent decreased rapidly for pure Cu2O,
while only a little decay was observed for Cꢀ2/Cu2O. Here, the photocurrent density at
the end of the last light cycle was used to compare with that at the end of the first light
cycle within 20 min measurement period for representing the photostability of the
samples.47,48 In comparison with pure Cu2O with a photostability of only 52%, the
photostablility for Cꢀ2/Cu2O increased to be 75%. Indeed, other Cꢀx/Cu2O samples all
exhibited highly improved photostability, as shown in Figure S6 (Supporting
Information), and the photostability increased with the increase of glucose
concentration during the treatment. Nevertheless, thicker carbon layer hindered the
absorption of visible light. So, the photocurrent density dwindled with the increase of
glucose concentration for Cꢀx/Cu2O samples (x > 1) (Figure S6 in Supporting
Information). Figure S7 (Supporting Information) showed the photocurrent density (at
the end of the last light cycle from the photocurrent density plots) and photostability
of the Cꢀx/Cu2O samples as a function of glucose concentration. Considering the two
factors, the Cꢀ2/Cu2O sample with a thickness of around 5 nm was regarded as the
best one. Figure 4c displays the enlarged photocurrent density plots of the first 200 s
to further compare the electrochemical property of pure Cu2O and Cꢀ2/Cu2O. It can be
seen that the instant photocurrent under irradiation was extremely large for pure Cu2O,
but it decreased rapidly on account of the recombination of photogenerated
electronꢀhole pairs, and thus a distinct spike was generated. However, the spikes were
small for Cꢀ2/Cu2O, and gradually disappeared with the increase of glucose
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment

Page 21 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
concentration for Cꢀx/Cu2O (Figure S6 in Supporting Information), indicating the low
recombination rate of photogenerated electronꢀhole pairs for the carbon coated
samples. Hence, the carbon layer coated on the surface of Cu2O NRs had a paramount
contribution to hinder the recombination of photogenerated electrons and holes, to
facilitate the fast electron transfer, as well as to enhance the photostability of Cu2O.
MottꢀSchottky (MS) measurements were performed in order to determine the
carrier concentrations and conduction band position of the samples, and the results are
shown in Figure 4d. The negative slope indicated that pure Cu2O and Cꢀ2/Cu2O were
pꢀtype semiconductors with hole conduction. The donor concentration of the samples
can be quantified by the MottꢀSchottky equation 49
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1/C2 = (2/eε0εNA)[(VꢀVFB)ꢀkT/e],
(1)
where C represents the capacitance of space charge region, e the electron charge, ε0
the vacuum permittivity, ε the dielectric constant of material (for Cu2O, ε = 6.3 50), V
the electrode applied potential, k the Boltzmann constant, T the absolute temperature
and NA the donor concentration. At room temperature, the term of kT/e is calculated to
be 25.693 meV, which is exceedingly small and can be neglected. Then the slope
determined from the analysis of MottꢀSchottky plot can be used to estimate the donor
concentration using the following equation 51
NA = (2/eε0ε)[d(1/C2)/dV]ꢀ1 ,
(2)
So, the hole concentrations were then calculated to be 6.30 × 1018 and 1.69 × 1019
cmꢀ3 for pure Cu2O and Cꢀ2/Cu2O respectively, which was higher than 9.25 × 1017
cmꢀ3 for the Cu2O prepared by electrodeposition method in our previous report.26
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 41
1
2
3
4
5
6
7
8
9
Additionally, the hole concentration for Cꢀ2/Cu2O was nearly three times as that of
pure Cu2O, which inferred that the carbon layer could accelerate electron transfer and
the standard 1ꢀD structure was in favor of charge carrier transfer. In addition, from the
MS plots, the flatꢀband potential (EFB) was also quantified to be 0.436 and 0.414 V
(vs.Ag/AgCl) for pure Cu2O and Cꢀ2/Cu2O respectively. The measured EFB can be
converted to the normal hydrogen electrode (NHE) scale via Nernst equation:18
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ENHE = EAg/AgCl + 0.05916 pH + E0
,
(3)
Ag/AgCl
where ENHE is the converted potential vs. NHE, EAg/AgCl the experimental potential
measured against Ag/AgCl reference electrode, and E0Ag/AgCl the standard potential of
Ag/AgCl at 298 K (0.1976 V). Then the conductionꢀband (ECB) edge was calculated
to be ꢀ0.734 and ꢀ0.756 V for pure Cu2O and Cꢀ2/Cu2O respectively according to the
following equation:
EFB ꢀ ECB = kBTln(NV/NA) ,
(4)
where NA is the carrier density and NV the effective density of states in the valence
band which can be expressed as
NV = 2(2πm*kBT/h2)3/2
,
(5)
where m* = 0.58m0 is taken as the effective hole mass.39 The calculated
conductionꢀband position was more negative than the reduction potential of CO2,52
which might provide a large driving force for CO2 photoreduction thermodynamically.
Figure 4e displays the linear sweep voltammetry (LSV) curves under dark and
illumination conditions for pure Cu2O and Cꢀ2/Cu2O. The dark current of pure Cu2O
was infinitesimally small due to its poor conductivity, while the dark current became
22
ACS Paragon Plus Environment

Page 23 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
much large for Cꢀ2/Cu2O with the help of the coating carbon layer. Under
illumination, the currents of the two samples increased greatly without doubt.
Interestingly, under this condition, the current of pure Cu2O was just comparable with
Cꢀ2/Cu2O, which was due to the reason that the carbon layer blocked the optical
absorption. The electrochemical impedance spectra (EIS) shown in Figure 4f were
used to analyze the interfacial charge transfer process for the different electrodes in
electrolyte. The measurement was conducted in the frequency range of 100 kHz to 0.1
Hz at a potential of 0 V vs Ag/AgCl under dark and illumination conditions. The
diameter of the semicircle is indicative of the interfacial charge transfer resistance, Rct,
across the electrode/electrolyte interface.53 Compared with pure Cu2O, Cꢀ2/Cu2O
showed a lower Rct value both in dark and under illumination, which was attributed to
the conductivity improvement resulted from the carbon layer on the surface of Cu2O
NRs. The carbon layer facilitated the electron transfer from Cu2O to the electrolyte,
resulting in an overall decrescent charge transfer resistance. The results were also
consistent with the LSV and photocurrent results.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Finally, the photocatalytic activity of pure Cu2O and Cꢀ2/Cu2O toward CO2
reduction for the formation of hydrocarbon under visible light irradiation was
evaluated. In this study, methane (CH4) and ethylene (C2H4) were found to be the
major products of CO2 photoreduction. Control experiments under identical
conditions but in the absence of CO2 (under N2 atmosphere) or light irradiation were
performed, in which no hydrocarbon product was detected. Additionally, the activity
of a blank Cu foil and a carbon coated Cu foil (CꢀCu) annealed under the same
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 41
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. (a) Yield of CH4 and C2H4 evolved from photocatalytic CO2 reduction for
Cꢀ2/Cu2O compared with that of controlled conditions under 6 h irradiation. (b) Gas
chromatogram of CH4 and C2H4 evolved from 12CO2 and 13CO2 as carbon source.
Mass spectra after 6 h photocatalytic reduction of (c) 12CO2 and (d) 13CO2.
condition were measured for comparison as well. The results for 6 h photocatalytic
reaction are displayed in Figure 5a. It can be seen that the annealed Cu foil and carbon
coated Cu foil just had negligible activity toward CO2 photoreduction but the sample
Cꢀ2/Cu2O produced a large amount of hydrocarbon. 13Cꢀlabeled isotopic experiment
was also carried out to provide solid evidence for validating the origin of the evolved
products, in which 13CO2 was used to replace 12CO2. In Figure 5b, two peaks at
retention time of 0.42 and 0.68 min were assigned to CH4 and C2H4 respectively.
Figure 5c and 5d showed the mass spectrum after 6 h photocatalytic conversion of
24
ACS Paragon Plus Environment

Page 25 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
12CO2 and 13CO2 respectively. When 13CO2 was the reactant, 13CH4 (m/z = 17) and
13C2H4 (m/z = 30) were detected, which was significantly different from the signal for
12CH4 (m/z = 16) and 16C2H4 (m/z = 28) when using 12CO2 as the reactant. This is the
direct evidence to prove that the carbon source for the evolved hydrocarbons came
from CO2 but not from carbon layer or any other carbon source.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Timeꢀdependent product evolution over (a) pure Cu2O and (b) Cꢀ2/Cu2O. (c)
Histogram of total yield and (d) stability study of photocatalytic evolution (CH4 and
C2H4) for pure Cu2O and Cꢀ2/Cu2O under visible light irradiation.
The yield of CH4 and C2H4 on the prepared photocatalysts as a function of
irradiation time was plotted in Figures 6a and 6b. As shown in Figure 6a, for pure
Cu2O sample, the total yield of CH4 and C2H4 increased rapidly within 5 h, but
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 41
1
2
3
4
5
6
7
8
9
thereafter the conversion rates diminished gradually for another 4 h, and finally the
concentration of the formed organic compounds remained unchanged in the last few
hours. This was presumably due to the ubiquitous issue of photocorrosion for Cu2O.
Substantially, the total yield of CH4 and C2H4 increased gradually within the whole
irradiation time for Cꢀ2/Cu2O sample (Figure 6b). The total amount of CH4 and C2H4
reached 0.385 ꢁmol after 12 h visible light irradiation, which was 2.5 times as high as
that for pure Cu2O. The histogram of the total yield of CH4 and C2H4 in Figure 6c
displayed a further comparison for pure Cu2O and Cꢀ2/Cu2O. The total yield of pure
Cu2O was barely comparable to that of Cꢀ2/Cu2O in the preliminary 4 h, but it then
fell behind with further irradiation. We checked the durability of Cꢀ2/Cu2O for CO2
reduction carefully. Figure S8a (Supporting Information) showed the total yield (CH4
and C2H4) evolution over Cꢀ2/Cu2O during prolonged irradiation up to 34 h. It can be
found that the photocatalyst still maintained most of its intrinsic photocatalytic
activity after 28 h, indicating that the durability of Cꢀ2/Cu2O was much better than
that of pure Cu2O. Additionally, the apparent quantum efficiency (AQE) of CH4 and
C2H4 was calculated to be 0.71% and 1.36% respectively, and the total AQE was thus
2.07% for Cꢀ2/Cu2O at λ0 = 400 nm. The details for the calculation process are shown
in Supporting Information. The AQE of 2.07% was higher than that for the published
Cu2Oꢀbased photocatalysts such as Cu2O/reduced graphene oxide composites (AQE =
0.34%)54 and Cu2O/RuOx composites (AQE = 1.6%)55. A detailed comparison with
the AQE values for other representative photocatalysts is presented in Table S1 in
Supporting Information, which showed the AQE of 2.07% was decent. Most
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
26
ACS Paragon Plus Environment

Page 27 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
importantly, neither catalyst hybridization nor cocatalyst was used in our engineered
material. Another effective strategy like photoelectrocatalysis (PEC) combining the
merits of both electrocatalysis (EC) and photocatalysis (PC) in our future work is
anticipated to further improve the CO2 conversion efficiency over Cꢀ2/Cu2O. The
results of other Cꢀx/Cu2O toward CO2 photoreduction under the same condition were
shown in Figure S8b (Supporting Information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To further testify the recycling photostability of asꢀprepared photocatalysts,
repeated CO2 photoreduction experiments were conducted for six cycles with every
cycle of 3 h under the same reaction conditions. The data are shown in Figure 6d. It
can be seen that significant deactivation was observed along with prolonged reaction
duration for pure Cu2O. The total yield of CH4 and C2H4 decreased to 0.037ꢁmol in
the sixth cycle, merely about 46% of that for the first run. While the photocatalytic
activity of Cꢀ2/Cu2O almost remained after six cycles with total yield of CH4 and
C2H4 0.079 ꢁmol in the sixth cycle, just about 7% loss compared to that for the first
run. So, Cꢀ2/Cu2O had superior stability to pure Cu2O for CO2 photoreduction.
In order to probe the specific structure change for pure Cu2O and Cꢀ2/Cu2O
during the CO2 reduction under irradiation, SEM images of the samples after three
and six reaction cycles were collected and presented in Figures 7a1 and 7b1, and
Figures 7a2 and 7b2, respectively. For pure Cu2O, the large particles were aggregated
to some lumps after three cycles, then totally agglomerated and fused after six cycles.
However, the 1ꢀD nanorod architecture remained for Cꢀ2/Cu2O, and the size for the
nanorods became just slightly larger after six cycles, validating the high stability of
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 41
1
2
3
4
the nanostructure for CCMNRs.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. SEM images of pure Cu2O (a1 and a2), and Cꢀ2/Cu2O (b1 and b2) after
three (a1 and b1) and six (a2 and b2) reaction cycles of CO2 reduction. High
resolution XPS spectra of Cu 2p for pure Cu2O (c1 and c2), and Cꢀ2/Cu2O (d1 and d2)
before (c1 and d1) and after (c2 and d2) six cycles of CO2 reduction. Surface XRD
patterns of pure Cu2O (e1) and Cꢀ2/Cu2O (e2) before and after six cycles of CO2
reduction under visible light irradiation. (f) XPS valenceꢀband spectra, and (g) band
structure alignments of pure Cu2O and Cꢀ2/Cu2O. (h) Time courses of photocatalytic
O2 production over pure Cu2O and Cꢀ2/Cu2O under visible light irradiation.
XPS detections were also performed to analyze the composition change of the two
samples after six reaction cycles. The Cu 2p core level spectra are shown in Figures
7c1, 7c2, 7d1 and 7d2, from which it can be seen that for pure Cu2O, a new peak at
28
ACS Paragon Plus Environment

Page 29 of 41
ACS Catalysis
1
2
3
4
935.6 eV emerged and the satellite peaks became stronger, indicating the presence of
5
6
7
56
CuO after six cycles (Figure 7c2). While the Cu 2p spectrum for Cꢀ2/Cu2O had
8
9
almost no change with only the presence of tiny satellite peaks (Figure 7d2). The
formation of Cu2+ after photoreduction was mainly originated from the photocorrosion
process, in which Cu+ was oxidized by holes.57 Yet, Cu+ and Cu0 are indistinguishable
from the XPS spectrum resulted from the binding energy overlap of Cu+ and Cu0 in
Cu 2p3/2 region.58 So, it is necessary to use other technique such as surface Xꢀray
diffraction (SXRD) with corrected background. As shown in Figure 7e2, all peaks in
the SXRD pattern of Cꢀ2/Cu2O after CO2 photoreduction well matched to those for
the fresh sample before reaction except a new tiny peak (50.4°), which belonged to
Cu0. Whereas, for pure Cu2O, two evident diffraction peaks of Cu0 with a strong
intensity were present (Figure 7e1), indicating that most Cu2O was reduced to Cu0
after photoreduction reaction. Unfortunately, Cu has no photocatalytic activity. No
diffraction peak corresponding to CuO was observed from the SXRD pattern, which
was presumably due to its tiny amount on the surface. All the above confirmed the
high stability of CCMNRs, which definitely accounted for the maintained
performance of CO2 photoreduction in prolonged duration.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The mechanism of CO2 photoreduction is complex, and multielectron transfer
process is generally considered.59 The evolved products closely depend on the number
of electrons involved in the reduction as well as the reduction potential of different
products, as shown in Equations (6) ~ (11) 52
CO2 + 2H+ + 2eꢀ → HCOOH, E0 = ꢀ0.61 V vs. NHE at pH 7
(6)
29
ACS Paragon Plus Environment