The main factor to improve the performance of CoSe2 for photocatalytic CO2 reduction: Element doping or phase transformation

Article abstract of DOI:10.1039/c9ta11595a

In this work, orthorhombic CoSe₂ is fabricated by nitrogen doping into a cubic CoSe₂ structure, in which the crystalline phase undergoes transformation. The doped atoms lead to the rotation of the Se₂^2- group, and then the faults of phase change can be observed. Thus, the reverse phase transformation provides an opportunity to study the aspects of element doping and phase change on photocatalytic CO₂ reduction simultaneously. The phase change material of N-doped orthorhombic CoSe₂ shows high activity toward the photoconversion of CO₂ to CO, and the average production rate of CO is up to 1.66 × 10⁴ μmol h^-1 g^-1 in the first 3 hours, being twice as efficient as that of cubic CoSe₂. Intermediate COOH? is detected by in situ infrared spectroscopy and theoretical calculations are also conducted to reveal the intrinsic impact of the doping element on the photosplitting of CO₂

Full text of DOI:10.1039/c9ta11595a

View Article Online
View Journal
Journal of
Materials Chemistry A
Materials for energy and sustainability
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Y. Xu, J. Mo, G.
Xie, X. Wang and S. Ding, J. Mater. Chem. A, 2020, DOI: 10.1039/C9TA11595A.
Volume
Number 12
28 March 2018
Pages 4883-5230
6
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
Materials for energy and sustainability
rsc.li/materials-a
Accepted Manuscripts are published online shortly after 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
Information for Authors.
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 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
COMMUNICATION
Zhenhai Wen et al.
An electrochemically neutralized energy-assisted low-cost
acid-alkaline electrolyzer for energy-saving electrolysis
hydrogen generation
rsc.li/materials-a

Please do not adjust margins
Journal of Materials Chemistry A
Page 1 of 7
View Article Online
DOI: 10.1039/C9TA11595A
ARTICLE
The main factor to improve the performance of CoSe2 for
photocatalytic CO2 reduction: element doping or phase
transformation†
Received 00th January 20xx,
Accepted 00th January 20xx
Yong Xu,ab Jiang Mo,a Guanqun Xie,a Xiaoxia Wang,*a and Shujiang Ding*b
DOI: 10.1039/x0xx00000x
In this work, orthorhombic CoSe2 is fabricated by nitrogen doping into cubic CoSe2 structure, in which the crystalline phase
2-
undergoes transformation. The doped atoms lead to the rotation of Se2 group, and then the faults of phase change can be
observed. Thus, the reverse phase transformation provides an opportunity to study in the respects of element doping and phase
change on photocatalytic CO2 reduction simultaneously. The phase change material of N-doped orthorhombic CoSe2 shows
high activity toward photoconversion of CO2 to CO, and the average production rate of CO is up to 1.66 × 104 μmol h-1 g-1 in
the first 3 hours, being twice as efficient as that of cubic CoSe2. Intermediate COOH* is detected by in situ infrared
spectroscopy and theoretical calculations are also conducted to reveal the intrinsic impact of doping element on photosplitting
of CO2.
conductivity and is deemed to expedite the CO2-adduct bonding as
well as the desorption of products.13
As promising alternatives to precious metal catalysts, 3d metal
1. Introduction
More than 80 % of global energy consumption is based on fossil fuels
in today’s world, while it may result in energy depletion, climate
warming and environmental pollutions.1,2 Photocatalytic conversion
of CO2 is a promising strategy to get two birds with one stone by
directly converting atmospheric CO2 and solar energy into valuable
chemicals. To achieve this carbon-neutral shift, one of the most
desirable approaches is to develop catalysts/co-catalysts with the
combination of high activity and high stability.3 Since the critical step
to accelerate CO2 reduction relates closely to the performance of
catalysts by reducing the kinetic energy barrier to avoid the breakage
of the high bond energy of C=O (750 kJ mol-1).4 So far, most catalytic
systems undergo the procedure of proton assisted multi-electron
transfer, while it remains a challenge to design the catalysts due to the
difficult in controlling the multi-electron transfer process and the
over-complication of the CO2 reduction mechanism.5,6 In 1979, CO2
gas was successfully transformed into organic compounds, as firstly
reported by Inoue.7 From then on, a variety of cobalt-containing
inorganic compound materials such as nitride, oxide, phosphide and
sulfide of cobalt, have been developed as heterogeneous catalysts for
catalytic conversion of CO2 into energy-rich chemicals.8-11 The reason
for the choice of cobalt as the metallic element lies in the fact that CO2
is more prone to bond to the cobalt centre and can thus be effectively
activated, which facilitates the CO2 photofixation.12 In addition, metal
cobalt has loosely bonded d-electrons, which leads to good electron
chalcogenides have been studied in various structures and
morphologies because of their low cost as well as environmental
benignity. For instance, cobalt diselenide (CoSe2) with multiple
3d electronic structure and low resistivity, has attracted wide
1
interest in H2 production.14 Nevertheless, with t2g6eg electronic
structure and metallic characteristics, CoSe2 has rarely been
explored in chemical conversion of CO2 under light
illumination.15-17 Furthermore, an additional advantage of
heterogeneous catalytic materials is that they provide
opportunities in investigating the catalytic reaction by modifying
internal structure, crystal phase, particle size and surface
properties of catalysts.4,18-20 In particular, structural phase
change is one of the most basic physical phenomena, however,
few studies could be found concerning the impact of the natural
phenomenon on photocatalytic CO2 reduction (PCR). MoS2 and
WS2 could undergo phase transformation by chemically
exfoliating their layered materials, and this transformation
resulted in an increase of H2 evolution activity due to the strained
metallic octahedral phase.21,22 Besides, heat treatment is a
common strategy to induce phase change for metal
chalcogenides such as Cu2Se, CoSe2 and MoSe2.23-26 Recently,
Yao used an Ar/O2 plasma method to simultaneously induce
exfoliation, surface recombination and phase change on CoSe2
to improve the performance of oxygen evolution.27 It should be
mentioned the plasma-assisted phase transformation process
calls for sophisticated apparatus and is accompanied by high-cost.
Until recently, there are still great challenges to fabricate phase
change materials (PCMs), let alone to synthesize crystalline
phase materials with low-stability from the ones with high-
stability. As has been reported recently, non-metallic doping has
been demonstrated to be effective in modifying the electronic
a. School of Materials Science and Engineering, Dongguan University of Technology,
Dongguan 523808, China.
b. Department of Applied Chemistry, School of Science, Xi´an Jiaotong University,
Xi´an, 710049, P.R. China.
†
Electronic Supplementary Information (ESI) available. See DOI:
10.1039/x0xx00000x
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 2 of 7
ARTICLE
Journal Name
configuration and properties of catalytic active sites, which may be an important driving force for Se-Se(N) bonds rotation. Hence
View Article Online
dramatically improve the catalytic ability.14,28,29 Yu et al. orthorhombic CoSe2, an inverse phase change material, was
DOI: 10.1039/C9TA11595A
reported that phosphorus-doping process could induce the prepared.17,31 The results of thermogravimetric analysis (TGA)
rotation of Se22- groups, which led to the transformation of cubic of c-CoSe2 and o-CoSe2/N are shown in Fig. S1.† As can be seen,
pyrite into orthorhombic marcasite.30 Inspired by this, the 9.0 wt% of c-CoSe2 was lost when the temperature reached 480
mechanism of the enhancement of PCR can be studied either oC, which was ascribed to the evaporation of adsorbed water and
respectively or synergistically from the respects of phase change thermal decomposition of DETA intercalated into the layered
and element doping.
structure. Besides, the -NH-, -NH2-, -CH2- and -CH- vibration
In this work, orthorhombic CoSe2 (o-CoSe2) was prepared bands showing up on the FTIR spectrum indicates the existence
from cubic CoSe2 (c-CoSe2) via elemental nitrogen doping of DETA molecules in the studied c-CoSe2 material (Fig. S2†).
method with urea as a nitrogen source. Among numerous doping
elements, nitrogen is regarded as an attractive option because of
its relatively small size, high electron affinity, good conductivity
and ion permeability.2,8,24 Since the electronegativity of selenium
is smaller than that of nitrogen, which may have influence on the
distribution of d electrons of element Co, thus causing rotating
of Se22- group. It is doping-induced reverse phase transformation
that provides an opportunity to investigate the effects of both
element doping and phase change of CoSe2 material on catalytic
CO2 reduction. The obtained N-doped o-CoSe2 (denoted as o-
CoSe2/N) is one of the best catalytic materials for
photoconversion of CO2 to CO, and its average generation rate
is up to 16,600 μmol h-1 g-1 in the first 3 hours. By comparison
Fig. 1 (a) XRD patterns of c-CoSe2 and o-CoSe2/N. (b,c) SEM images of
of o-CoSe2/N and o-CoSe2/P (P-doped o-CoSe2), it can be
c-CoSe2 nanosheets and o-CoSe2/N nanoflowers. (d) EDX spectra of c-
concluded that doping has a greater influence on catalytic
properties of CoSe2 than on phase change. In situ Fourier
transform infrared (in situ FTIR) spectroscopy discloses that
COOH* is the crucial intermediate. Furthermore, density
functional theory (DFT) calculations were carried out to study
the Gibbs free energy on the dopants of CoSe2 in photocatalytic
CO2 reduction. The nitrogen with high electronegativity can
enhance the bonding effect of Co and COOH*, and thus
improving its catalytic efficiency.
CoSe2 and o-CoSe2/N. C, O and Cu signals were originated from the
sample holder. (e) TEM image of synthesized c-CoSe2 nanosheets. Inset:
the corresponding HRTEM image and SAED pattern. (f) TEM image of
o-CoSe2/N product. (g,h) HRTEM images of o-CoSe2/N. Inset: the
corresponding SAED pattern. (i) HAADF-STEM image of o-CoSe2/N.
(j-l) Corresponding elemental mapping images.
In order to probe into the structural information of as-obtained
materials, X-ray diffraction (XRD) analysis was performed and
the results were shown in Fig. 1a, which presents a distinct
transformation of characteristic diffraction peaks from
thermodynamically stable pyrite CoSe2 with cubic phase
(JCPDS Card No. 09-0234) to marcasite CoSe2 with pure
orthorhombic phase (JCPDS Card No. 53-0449). The crystal
planes of (111) and (120) shift to the direction of large diffraction
angle in o-CoSe2/N compared with pristine o-CoSe2, which
should be caused by taking the place of certain big Se atoms with
N doping (Fig. S3†). The morphology of the nanosheets is
revealed in scanning electron microscopy (SEM) image of c-
CoSe2 sample (Fig. 1b). o-CoSe2/N, however, exposes the
flower-like morphology composed of bended and folded
nanosheets after nitrogen doping in Fig. 1c. The Brunaure-
Emmett-Teller (BET) specific surface area of c-CoSe2 and o-
CoSe2/N are 13.0 m2 g-1 and 19.7 m2 g-1, respectively, as
determined by N2 sorption test (Fig. S4 and S5†). In addition, o-
CoSe2/N possesses obvious mesoporous mainly centered at 3.8
nm, which endows o-CoSe2/N with both large surface area and
rich catalytic sites. Energy dispersive X-ray (EDX) spectra (Fig.
1d) of the prepared materials display that elemental N can
actually be doped into CoSe2 structure while urea acted as the
precursor. Transmission electron microscope (TEM) image in
Fig. 1e shows two-dimensional sheet-like structure for c-CoSe2
material, and high resolution TEM (HRTEM) displays the lattice
fringes with a d-spacing of 0.268 nm, corresponding to the
reflection of (210) plane. Selected area electron diffraction
2. Results and discussion
Scheme 1 Schematic illustration of hydrothermal method to synthesize
c-CoSe2, and N-doping-induced phase change process from c-CoSe2 to
o-CoSe2/N.
Cubic phase CoSe2 nanosheets were synthesized in binary
solution of diethylenetriamine (DETA) and water under
solvothermal conditions. The obtained c-CoSe2 contains
protonated DETA molecules that are adjacent to the CoSe2
nanolayers and coordinate with Se.17 The PCM of o-CoSe2/N
was fabricated through annealing urea and the as-prepared c-
CoSe2 nanosheets in the atmosphere of inert gas, during which
NH3 species was generated in situ by thermal decomposition of
urea (Scheme 1). Se vacancies were formed in heat-treating step,
and afterwards N generated from ammonia quickly took over
these vacancy defects. The doped N results in a closer Co-N
interaction due to the more electronegative N compared with Se.
The construction of new bonds (Co-N) brings about re-
distribution of d-orbital electrons of Co, which was believed to
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 7
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
(SAED) pattern (inset in Fig. 1e) exhibits single-crystalline 3p and SeOx.24 What’s more, the high resolution XPS peaks of
View Article Online
DOI: 10.1039/C9TA11595A
diffraction spots, while the pattern of o-CoSe2/N shows a quasi- Co 2p and Se 3d for o-CoSe2/N show marked shift toward the
single-crystal character and the diffraction spots are slightly direction of high binding energy compared with that of c-CoSe2
distorted because of the crystal lattice mismatch after the material, which may be caused by the partial replacement of Se
introduction of N (inset in Fig. 1g). TEM image of the three- with N in crystal lattice, thus leading to the increase of electron
dimension interlaced o-CoSe2/N is shown in Fig. 1f, the density. An obvious N 1s signal can be detected after N-doping
corresponding HRTEM image in Fig. 1g reveals that the phase with the high resolution spectrum deconvolved into five peaks
change faults arise in the annealing process. The decomposition (Fig. S6†). The peaks of 397.4, 398.3 and 400.0 eV can be
of DETA in the nanolayer provides space for the rotation of Se- assigned to Co-N feature bonds, while peaks located at 399.0 and
N bond, which also facilitates the formation of these faults. 400.3 eV are related to Se-N bonds.33,34 The Raman spectra with
Resolved lattice fringes are clearly shown in Fig. 1h, and the the characteristic peak at 670 cm-1 for both materials can be
spacing is 0.288 nm, which is ascribed to (101) plane. The high- attributed to the A1g stretching mode of CoSe2 (Fig. 2d).
angle annular dark-field scanning TEM (HAADF-STEM) image Furthermore, an obvious Raman peak at 190 cm-1 for c-CoSe2,
(Fig. 1i) and corresponding element mapping images (Fig. 1j-l) ascribed to the Se-Se stretching mode, disappeared after thermal
o
demonstrate uniform distributions of N, Co and Se throughout o- annealing at 400 C and a new peak showed up at 172 cm-1 as
CoSe2/N material.
Raman active peak in the spectrum of o-CoSe2/N, suggesting the
structural change from pyrite c-CoSe2 to marcasite o-CoSe2/N.
The composition of o-CoSe2/N was further determined by
inductively coupled plasma-optical emission spectrometer (ICP-
OES) with the content of N being 3.56 wt% and the mass
fractions of Co and Se were 31.10 and 65.34 wt% respectively.
A molar ratio of 2.05 for (Se + N) to Co can thus be deduced.
Fig. 2 (a) XPS survey spectra of the prepared c-CoSe2 and o-CoSe2/N.
(b,c) High resolution XPS spectra of Co 2p and Se 3d for c-CoSe2 and o-
CoSe2/N. (d) Raman spectra of c-CoSe2 and o-CoSe2/N samples.
The chemical constituents and surface electronic states of the
fabricated samples were measured by X-ray photoelectron
spectroscopy (XPS). As expected, peaks attributed to Co and Se
elements are clearly observed in both survey scans (Fig. 2a). The
existence of C species is generally inevitable and O is attributed
to the adsorbed H2O, CO2 and oxides on the surface of the
samples. The content of N in o-CoSe2/N is much higher than that
in initial c-CoSe2, which further verifies that element N has been
successfully doped into the cobalt diselenide material. Fig. 2b
displays high resolution XPS spectra of Co 2p, in which the two
split peaks at ~781 eV and ~796 eV are ascribed to Co 2p3/2 and
Co 2p1/2 of the Co2+ cations in CoSe2, as fitted by Lorentzian-
Gaussian function. The peaks centered at 793.0 and 796.7 eV are
associated with Co 2p1/2 state of o-CoSe2/N. The Co 2p3/2 peak
can be separated into three peaks at the binding energies of 778.6,
780.7 and 783.1 eV, and these chemical states are assigned to
Co-Se, Co-N and Co-O (thin oxidation layer on the surface)
structure, followed by a Co satellite peak at 785.8 eV.14,32 The
peaks located at 54.1 and 55.0 eV are associated with Se 3d5/2
and Se 3d3/2, respectively, corresponding to Co-Se bond (Fig. 2c).
The peaks at ~60 eV are related to the bonding structures of Co
Fig. 3 (a) MS analysis of CO generated from 13CO2 isotope experiments.
(b) Production of CO and H2 as a function of irradiation time in the
photocatalytic systems of c-CoSe2 and o-CoSe2/N. Reaction conditions:
catalysts (1 mg), [Ru(bpy)3]Cl2·6H2O (5 mg), solvent
(acetonitrile/H2O/TEOA = 4 mL : 1 mL : 1.5 mL), CO2 (1 atm),
irradiation with white LEDs lamp at room temperature. (c) The
comparison of catalytic properties of the four materials (c-CoSe2, c-
CoSe2(400), o-CoSe2/P and o-CoSe2/N) studied in this work. (d) Average
yields of CO and H2 in the first 3 hours with o-CoSe2/N catalyst in
recycling tests.
Photocatalytic reduction of CO2 has been carried out in a
visible light-driven catalyzing system by using the prepared
materials as catalysts, complexe of [Ru(bpy)3]Cl2·6H2O (2,2´-
bipyridine abbreviated as bpy) as photosensitizer and
triethanolamine (TEOA) as sacrificial agent in the mixed solvent
of acetonitrile and water. When the reduction reaction was
triggered with 200 W LEDs lamp at 420 ≤ λ ≤ 800 nm, only CO
and H2 were detected as the reduction products during all the
tests by gas chromatography (GC), with Ar as the carrier gas.
Further, no soluble reduction products, such as methanol or
1
formate, were measured by H nuclear magnetic resonance (1H
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 4 of 7
ARTICLE
Journal Name
NMR) spectroscopy (Fig. S7†), liquid chromatography and ion cubic phase unchanged with a great many of Se vacancies (Fig.
View Article Online
chromatography.
S9†), which is named as c-CoSe2(400D).OIS: 1u0r.1e03e9n/Co9uTgAh1,159th5Ae
Control experiments demonstrated that CO could not be catalytic ability has been slightly improved, as illustrated in Fig.
generated at all in the absence of any of the following 3c, and meanwhile the selectivity of CO is increased to 49.4%.
components: catalyst, photosensitizer, donor and light (Table Additionally, to clarify the catalytic performance of o-CoSe2/P
S1†). Nevertheless, when the catalytic system was degassed with and o-CoSe2/N, the following information can be involved: if the
Ar, large amounts of H2 was released from liquid phase. And CO average production rate of both materials are somewhat on a par,
gas is the sole carbon-containing reduction product in our PCR it should indicate that phase change is the dominating reason to
experiments. [Ru(bpy)3]Cl2 is the most frequently-used absorber, affect the catalytic activity; on the other hand, if catalytic ability
which is ready to accept one photon and get excited to of the two is quite different, it suggests doping is the key factor
2+*
Ru(bpy)3
,
then the obtained redox potential of to enhance catalytic performance. The experimental results show
E(Ru(bpy)33+/Ru(bpy)32+*) (-0.92 V vs NHE at pH 7) is negative that the catalytic ability of o-CoSe2/N is evidently higher than
enough to initiate the conversion of CO2 to CO (E(CO/CO2) = - that of o-CoSe2/P, so it is element doping rather than phase
0.53
V
vs NHE) and the production of H2 in change that is the predominant reason for promoting the catalytic
thermodynamics.35,36 To acquire solid proof for the source of activity of CoSe2. The XPS survey spectra certify that the target
elemental carbon, isotopic 13CO2 was utilized as feedstock gas in product of o-CoSe2/P was successfully synthesized (Fig. S10†)
the heterogeneous photocatalytic reaction. The released products with the doping amount of elemental P being 7.89 wt%. To
were identified by gas mass spectrometry (MS) with multi-ion evaluate the reusable ability of o-CoSe2/N material, recycling
channel mode. As shown in Fig. 3a, the peak appearing with a tests were carried out. The recovered o-CoSe2/N and fresh
mass/charge ratio (m/z) of 29 can be ascribed to the generated photosensitizer were redispersed into a fresh mixed solution for
13CO, and yet another peak at m/z = 45 is the unreacted 13CO2. further photoreaction. It was found that the used samples, even
The results firmly validate that the evolution of CO is derived after being stored on shelf for 3 months, still kept 97.6% of the
from the splitting of CO2 molecules rather than from substances initial efficiency for two subsequent recycling as demonstrated
such as TEOA. With the view of getting an in-depth in Fig. 3d. Besides, XRD and TEM characterizations were
understanding of photoreaction, Fig. 3b depicts the yields of CO performed on the recycled catalyst, only to find good structural
and H2 as a function of irradiation time for the catalysts of c- integrity with that of freshly prepared o-CoSe2/N (Fig. S11 and
CoSe2 and o-CoSe2/N. In the first 1 h, the evolution of both gases S12†).
increased rapidly, especially with o-CoSe2/N, nearly linear
increase of the amount of CO with the illumination time was
observed. However, with c-CoSe2, the generation of H2
displayed a rapid quantitative increase, particularly in the first
hour, while the release amount of CO significantly decreased on
the whole. The result suggests that c-CoSe2 material is more
inclined to produce H2 rapidly and continuously for a long time,
which is consistent with the recent report.26 As for o-CoSe2/N,
the average rate of CO generation is as high as 1.66 × 104 μmol
h-1 g-1 in the first 3 h with 76.9% selectivity over competing H2
production. The gases were analyzed by GC instrument with a
TCD detector, and the volume was calculated by comparing the
Fig. 4 (a) CO temperature-programmed desorption tests for c-CoSe2 and
o-CoSe2/N catalysts. (b) In situ FTIR spectra for visible light-driven CO2
reduction on o-CoSe2/N.
To gain better understanding of the enhanced CO2
peak area of obtained products with calibration curves. When the
photoreduction process driven by visible light, CO temperature-
irradiation time was prolonged beyond 3 hours, the evolution rate
of gases slowed down due mainly to the photobleaching of the
photosensitizer, which is a common issue in dye-sensitized
photochemical systems.9,37 Towards the end of the 3rd hour, an
programmed desorption (TPD) and in situ FTIR analysis were
conducted in this work. As depicted in Fig. 4a, with c-CoSe2 as
o
catalyst a high CO onset desorption temperature at 350 C was
required, while with o-CoSe2/N, a lower onset desorption
equal amount of fresh [Ru(bpy)3]Cl2·6H2O was replenished into
o
temperature of 300 C was observed. The results of CO-TPD
the reaction mixture, then the yield of CO and H2 progressively
manifest that the surfaces of o-CoSe2/N material are favourable
increased (Fig. S8†), thus proving the above conjecture. A series
to release CO, which benefits the photoinduced conversion of
of extraordinary experiments have been carried out to investigate
CO2 to CO. In situ FTIR measurement was performed to detect
the effect of both phase change and elemental doping on the
the generated intermediates during the reaction process. The
catalytic activity of CoSe2 for CO2 reduction. According to the
experiment with the same photoreaction system except in the
above results, c-CoSe2 shows a moderate catalytic activity with
the evolution rate of 0.84 × 104 μmol h-1 g-1, and the selectivity
dark as the background and the entire operation was carried out
under the atmosphere of CO2. The reaction system was saturated
for CO is 45.6%. As reported, vacancies usually act as active
with CO2 before ultrasonic dispersion, and then a small amount
sites during catalytic reaction, therefore the residual Se holes
of the mixture was added dropwise onto KBr for further infrared
may play a considerable role in photoreduction of CO2.16,38 In
measurement. In addition, all the FTIR spectra were presented
view of this, urea or sodium hypophosphite hydrate was not
by subtracting the background. Two obvious bands appeared
involved in the synthesis of CoSe2, and the final CoSe2 kept the
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 7
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
with the extension of illumination time from 0 to 9 minutes as
shown in Fig. 4b. Both the new peaks, located at 1356 and 1546
cm-1, gradually strengthened with the prolongation of exposure
time, which could be ascribed to COOH* radical as is consistent
with previous reports.39-42
3. Conclusions
View Article Online
DOI: 10.1039/C9TA11595A
In conclusion, orthorhombic CoSe2 was successfully prepared by
the method of a nitrogen-doping-induced phase transformation
from cubic CoSe2 material. The inverse phase change material of
o-CoSe2/N demonstrated high catalytic performance toward
photoinduced reduction of CO2, while c-CoSe2 displayed high
activity for H2 generation under the same reaction conditions.
The measurement of CO-TPD indicated that the catalytic
surfaces of o-CoSe2/N catalyst were in favor of releasing CO.
Moreover, the comparison of catalytic ability between o-
CoSe2/N and o-CoSe2/P suggested that elemental doping instead
of phase change was the dominant factor for promoting the
catalytic activity of CoSe2. In addition, the key intermediate of
COOH* was detected by in situ FTIR, and theoretical calculation
was carried out to investigate the barrier energies of o-CoSe2/N
as well as o-CoSe2/P catalyst. Nitrogen with large
electronegativity enables more electrons to locate around cobalt
active centres so that the bonding power between Co and
COOH* can be enhanced, which favors the reduction reaction.
This work provides not only an in-depth insight into the effect of
doping and phase transition on the CO2 photoreduction, but also
a novel promising way to design advanced catalysts.
Fig. 5 Gibbs free energy diagrams for the conversion of CO2 into CO on
o-CoSe2/N and o-CoSe2/P.
The Gibbs free energy (ΔG) diagrams were illustrated in Fig.
5 for stepwise calculated transformation of CO2 into CO. The
activation of adsorbed carbon dioxide (step II: CO2* + H+ + e-
→ COOH*) and the desorption of carbon monoxide (step IV:
CO* → CO) for the two catalysts were both endothermic
processes. The C-O bond breaking process (or activation of
adsorbed CO2) was regarded as the rate-determining step.12,39 In
this case, with o-CoSe2/N as catalyst, the reaction requires lower
energy barrier of 0.382 eV, which was relatively mild for CO2 4. Experimental section
reduction reaction, suggesting that the catalytic processes can
proceed at room temperature.41 The subsequent protonation of
4.1. Synthesis of c-CoSe2
All the chemicals were used as obtained without further
the COOH* intermediate, which would result in the formation of
CO*, was an exothermic process (ΔG < 0) and could occur
spontaneously.39 On the other hand, the Co-O bond (2.18 Å) on
o-CoSe2/N is clearly shorter than that on o-CoSe2/P (Fig. 6a),
which suggests a stronger binding interaction between CO2 and
CoSe2 doping with nitrogen. In addition, compared with P, which
has relatively lower electronegativity, N doped CoSe2 exhibited
higher catalytic activity for the PCR, indicating the large
electronegativity of doped element may lead to the effective
decrease in the formation energy of COOH*. And the reason
may be that N empowers a mass of electrons to locate around Co
atoms, thus improving the bonding effect between Co and
COOH*.43 As we can see in Fig. 6b, the differential charge
density of nitrogen is more obvious and more electrons are
transferred from N to CO2 (Fig. S13†). Finally, o-CoSe2/N does
favor to reach a more stable configuration, thus promoting CO2
photoreduction.
purification. c-CoSe2 was synthesized through solvothermal
method. Thus 1 mmol Co(OAc)2·4H2O (249 mg) and 1 mmol
Na2SeO3 (173 mg) were added into the solution of
tetraethylenepentamine with the volume of 24 mL. After being
stirred for 20 minutes, the mixture was diluted with 12 mL of
deionzed water and stirred for additional 20 minutes. Then the
wine solution was transferred into 50 mL of Teflon hydrothermal
autoclave, and kept at 180 oC for 15 hours. After cooling down
to room temperature naturally, the resulting brownish-black
floccules were collected, washed with ethanol three times and
dried under vacuum at 60 oC for 10 h.
4.2. Preparation of o-CoSe2/N
100 mg of the prepared c-CoSe2 and 3.0 g urea were placed at
two separated positions in an alumina boat with c-CoSe2 at the
o
downstream side. The samples were heated to 400 C with the
o
heating rate of 5 C/min, and maintained for 30 minutes under
Ar atmosphere. For o-CoSe2/P, the urea was replaced with 1.0 g
of NaH2PO2·H2O, and other conditions remained unchanged.
For the preparation of c-CoSe2(400), neither urea nor
NaH2PO2·H2O was used during the heating process.
4.3. Photocatalytic test
The photocatalytic experiments were carried out in a typical
process. 5 mg photosensitizer of [Ru(bpy)3]Cl2·6H2O and 1 mg
catalyst were uniformly dispersed in
a
solution of
acetonitrile/TEOA/H2O (4 mL/1.5 mL/1 mL) using an 55 mL
quartz tube, which was sealed with a silica gel plug. The
suspension was purged with CO2 to drive away the air, after 30
min, the suspension was divided into two layers, the upper fluid
was clear and the lower suspension was turbid. Then the tube
Fig. 6 (a) DFT-derived CO2 binding structures of o-CoSe2/N and o-
CoSe2/P, (b) differential charge density maps of CO2 adsorbed on o-
CoSe2/N and o-CoSe2/P. Charge accumulation is in blue and depletion in
yellow electronic location function.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 6 of 7
ARTICLE
Journal Name
14 H. Zhang, T. Wang, A. Sumboja, W. Zang, J. Xie, D. Gao, VSi.eJw. APretincnleyOconloinke,
was filled with of CO2 and the quartz tube was sealed up with
sealing film and adhesive tape. The sealed quartz tube was
irradiated immediately. During the potocatalytic reduction, the
sealed tube was irradiated with a 200 W homemade white LEDs
lamp stirring with a magnetic stirrer. The positions of the tubes
were changed every 30 min. After the reaction, the produced
gases were analyzed and quantified by gas chromatography. The
isotopic 13CO2 experiment was carried out under the same
conditions and analyzed by gas mass spectroscopy using an
offline injection method. To evaluate the durability of the sample,
the partition strategy was employed to compensate for filtration
and transferred loss of o-CoSe2/N among the cycles. In each
cycle, 1 mg of o-CoSe2/N from a total of 8 mg was used for
photoreaction, then after each 1 mg cycle, the parallel
experiment was carried out for the rest of catalyst under the same
conditions. Subsequently, all the catalysts were collected
together, after washing and drying, another 1 mg of catalyst was
taken out for the following run. The selectivity of CO was
calculated using the equation below, M(gas) is the molar quantity
of produced gas, which is obtained from peak area.
Z. Liu, C. Guan and J. Wang, Adv. Funct. Mater., 2018, 28, 1804846.
DOI: 10.1039/C9TA11595A
15 L. Liang, H. Cheng, F. Lei, J. Han, S. Gao, C. Wang, Y. Sun, S. Qamar, S.
Wei and Y. Xie, Angew. Chem. Int. Ed., 2015, 54, 1200412008.
16 Y. Liu, H. Cheng, M. Lyu, S. Fan, Q. Liu, W. Zhang, Y. Zhi, C. Wang, C.
Xiao, S. Wei, B. Ye and Y. Xie, J. Am. Chem. Soc., 2014, 136,
1567015675.
17 M. Gao, W. Yao, H. Yao and S. Yu, J. Am. Chem. Soc., 2009, 131,
74867487.
18 H. Cheng, N. Yang, Q. Lu, Z. Zhang and H. Zhang, Adv. Mater., 2018, 30,
1707189.
19 C. Dong, C. Lian, S. Hu, Z. Deng, J. Gong, M. Li, H. Liu, M. Xing and J.
Zhang, Nat. Commun., 2018, 9, 1252.
20 M. G. Kibria, C. Dinh, A. Seifitokaldani, P. D. Luna, T. Burdyny, R.
Quintero-Bermudez, M. B. Ross, O. S. Bushuyev, F. P. Arquer, P. Yang,
D. Sinton and E. H. Sargent. Adv. Mater., 2018, 30, 1804867.
21 M. A. Lukowski, A. S. Daniel, F. Meng, A. Forticaux, L. Li and S. Jin, J.
Am. Chem. Soc., 2013, 135, 1027410277.
22 D. Voiry, H. Yamaguchi, J. Li, R. Silva, D. C. B. Alves, T. Fujita, M. Chen,
T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nat. Mater., 2013, 12,
850855.
23 L. Chen, J. Liu, C. Jiang, K. Zhao, H. Chen, X. Shi, L. Chen, C. Sun, S.
Zhang, Y. Wang and Z. Zhang, Adv. Mater., 2019, 31, 1804919.
24 Y. Liu, Z. Chen, H. Jia, H. Xu, M. Liu and R. Wu, ACS Nano, 2019, 13,
61136124.
25 K. Leng, Z Chen, X. Zhao, W. Tang, B. Tian, C. T. Nai, W. Zhou and K.
P. Loh, ACS Nano, 2016, 10, 92089215.
26 P. Chen, K. Xu, S. Tao, T. Zhou, Y. Tong, H. Ding, L. Zhang, W. Chu, C.
Wu and Y. Xie, Adv. Mater., 2016, 28, 75277532.
Conflicts of interest
There are no conflicts to declare.
27 X. Wang, L. Zhuang, Y. Jia, H. Liu, X. Yan, L. Zhang, D. Yang, Z. Zhu
and X. Yao, Angew. Chem., 2018, 130, 1665916663.
28 J. Fu, B. Zhu, C. Jiang, B. Cheng, W. You and J. Yu, Small, 2017, 13,
1603938.
Acknowledgement
This work was financially supported by Project funded by China
Postdoctoral Science Foundation (3115200108), Science
Foundation for Distinguished Scholars of Dongguan University
of Technology (No. GC200906-10 and G200906-44), National
Natural Science Foundation of China (No. 51773165) and
Young Talent Support Plan of Xi´an Jiaotong University.
29 X. Hu, H. H. Hval, E. T. Bjerglund, K. J. Dalgaard, M. R. Madsen, M. Pohl,
E. Welter, P. Lamagni, K. B. Buhl, M. Bremholm, M. Beller, S. U.
Pedersen, T. Skrydstrup and K. Daasbjerg, ACS Catal., 2018, 8, 62556264.
30 Y. Zhang, P. Wu, M. Gao, X. Zhang, F. Gao, H. Ju, R. Wu, Q. Gao, R.
You, W. Huang, S. Liu, S. Hu, J. Zhu, Z. Li and S. Yu, Nat. Commun.,
2018, 9, 2533.
31 J. A. Tossell, D. J. Vaughan and J. K. Burdett, Phys. Chem. Minerals, 1981,
7, 177184.
32 C. Ray, S. C. Lee, B. Jin, A. Kunda, J. H. Park and S. C. Jun, J. Mater.
Chem. A, 2018, 6, 44664476.
33 P. Chen, T. Zhou, M. Zhang, Y. Tong, C. Zhong, N. Zhang, L. Zhang, C.
Wu and Y. Xie, Adv. Mater., 2017, 29, 1701584.
34 Z. Li, Y. Fang, J. Zhang and X. W. Lou, Adv. Mater., 2018, 30, 1800525.
35 M. D. Kärkäs, O. Verho, E. V. Johnston and B. Åkermark, Chem. Rev.,
2014, 114, 1186312001.
36 G. Sahara and O. Ishitani, Inorg. Chem., 2015, 54, 50965104.
37 J. Qin, S. Wang and X. Wang, Appl. Catal. B: Environ., 2017, 209,
476482.
38 X. Jiao, Z. Chen, X. Li, Y. Sun, S. Gao, W. Yan, C. Wang, Q. Zhang, Y.
Lin, Y. Luo and Y. Xie, J. Am. Chem. Soc., 2017, 139, 75867594.
39 X. Li, L. Liang, Y. Sun, J. Xu, X. Jiao, X. Xu, H. Ju, Y. Pan, J. Zhu and Y.
Xie, J. Am. Chem. Soc., 2019, 141, 423430.
40 Z. Jiang, H. Sun, T. Wang, B. Wang, W. Wei, H. Li, S. Yuan, T. An, H.
Zhao, J. Yu and P. K. Wong, Energy Environ. Sci., 2018, 11, 23822389.
41 B. Han, X. Ou, Z. Deng, Y. Song, C. Tian, H. Deng, Y. Xu and Z. Lin,
Angew. Chem. Int. Ed., 2018, 57, 1681116815.
Notes and references
1 S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Angew.
Chem. Int. Ed., 2013, 52, 73727408.
2 J. Ran, M. Jaroniec and S. Qiao, Adv. Mater., 2018, 30, 1704694.
3 Y. Chen, G. Jia, Y. Hu, G. Fan, Y. H. Tsang, Z. Li and Z. Zou, Sustain.
Energ. Fuels, 2017, 1, 18751898.
4 Q. Mu, W. Zhu, G. Yan, Y. Lian, Y. Yao, Q. Li, Y. Tian, P. Zhang, Z. Deng
and Y. Peng, J. Mater. Chem. A, 2018, 6, 2111021119.
5 J. Wu, X. Li, W. Shi, P. Ling, Y. Sun, X. Jiao, S. Gao, L. Liang, J. Xu, W.
Yan, C. Wang and Y. Xie, Angew. Chem., 2018, 130, 88558859.
6 A. Li, T. Wang, X. Chang, Z. Zhao, C. Li, Z. Huang, P. Yang, G. Zhou and
J. Gong, Chem. Sci., 2018, 9, 53345340.
7 T. Inoue, A. Fujishima, S. Konishi and K. Honda, Nature, 1979, 277,
637638.
8 B. Weng, W. Wei, Yiliguma, H. Wu, A. M. Alenizi and G. Zheng, J. Mater.
Chem. A, 2016, 4, 1535315360.
9 S. Wang, Z. Ding and X. Wang, Chem. Commun., 2015, 51, 15171519.
10 Z. Fu, R. Xu, J. T. Moore, F. Liang, X. Nie, C. Mi, J. Mo, Y. Xu, Q. Xu, Z.
Yang, Z. Lin and W. Fu, Chem. Eur. J., 2018, 24, 42734278.
11 Y. Yan, K. Li, X. Chen, Y. Yang and J. Lee, Small, 2017, 13, 1701724.
12 C. Gao, Q. Meng, K. Zhao, H. Yin, D. Wang, J. Guo, S. Zhao, L. Chang,
M. He, Q. Li, H. Zhao, X. Huang, Y. Gao and Z. Tang, Adv. Mater., 2016,
28, 64856490.
42 F. Xu, J. Zhang, B. Zhu, J. Yu and J. Xu, Appl. Catal. B: Environ., 2018,
230, 194202.
43 X. Jiao, X. Li, X. Jin, Y. Sun, J. Xu, L. Liang, H. Ju, J. Zhu, Y. Pan, W.
Yan, Y. Lin and Y. Xie, J. Am. Chem. Soc., 2017, 139, 1804418051.
13 S. Gao, Y. Lin, X. Jiao, Y. Sun, Q. Luo, W. Zhang, D. Li, J. Yang and Y.
Xie, Nature, 2016, 529, 6871.
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 7
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
Graphical abstract
View Article Online
DOI: 10.1039/C9TA11595A
Nitrogen doping-induced phase change material of cobalt diselenide
was first utilized for photocatalytic reduction of CO2. The doping of
N can enhance electron density of cobalt active sites, thus improving
the bonding ability of Co and COOH*.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins