Directional oxygen activation by oxygen-vacancy-rich WO2 nanorods for superb hydrogen evolution via formaldehyde reforming

Article abstract of DOI:10.1039/c9ta03051d

Molecular oxygen activation (MOA) into reactive oxygen species (ROS) is extremely crucial in numerous catalytic processes, while precise control of ROS products remains difficult. Herein, we find that oxygen substoichiometric tungsten dioxide nanorods (WO

Full text of DOI:10.1039/c9ta03051d

View Article Online
View Journal
Journal of
M at erials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: K. Qian, L. Du, X.
Zhu, S. Liang, S. Chen, H. Kobayashi, X. Yan, M. Xu, Y. Dai and R. Li, J. Mater. Chem. A, 2019, DOI:
1
0.1039/C9TA03051D.
Volume
4
Number
1
7
January 2016 Pages 1–330
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry A
Accepted Manuscripts are published online shortly after
Materials for energy and sustainability
www.rsc.org/MaterialsA
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2050-7488
PAPER
Kun Chang, Zhaorong Chang et al.
Bubble-template-assisted synthesis of hollow fullerene-like
MoS nanocages as a lithium ion battery anode material
2
rsc.li/materials-a

Page 1 of 11
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C9TA03051D
Directional Oxygen Activation by Oxygen-Vacancy-Rich WO Nanorods for Superb Hydrogen
2
Evolution via Formaldehyde Reforming

J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 2 of 11
View Article Online
DOI: 10.1039/C9TA03051D
ARTICLE
Directional Oxygen Activation by Oxygen-Vacancy-Rich WO2
Nanorods for Superb Hydrogen Evolution via Formaldehyde
Reforming
Received 00th January 20xx,
Accepted 00th January 20xx
a
a
a
a
a
b
DOI: 10.1039/x0xx00000x
Kaicheng Qian, Leilei Du, Xiaohui Zhu, Shipan Liang, Shuang Chen, Hisayoshi Kobayashi,*
c
d
d
a
Xiaoqing Yan, Min Xu, Yihu Dai, and Renhong Li,*
Molecular oxygen activation (MOA) into reactive oxygen species (ROS) is extremely crucial in numerous catalytic
processes, while precise control of ROS products remains difficult. Herein, we find that oxygen substoichiometric tungsten
dioxide nanorods (WO
produce H . Rich oxygen vacancies (OVs) in WO
reaction, because they guarantee the favorable chemisorption of molecular O
channel for the delivery of electrons to O species. Importantly, MOA initiated by WO
systematically modulating the proton−coupled electron transfer (PCET) pathway to controllably obtain the desired ROS
and thus to directionally produce H , in which molecular O receives the electrons to form superoxide radicals in a
chemisorbed state (O
•−) and couple with protons provided by HCHO to convert into •OOH radicals, rather than •OH or
other ROS. The surface stabilized HOO•−WO NRs complex act as the catalytically active center in the dehydrogenation
reaction. As O continuously uptakes and releases protons by formation and consumption of •OOH radicals, O
persistently circulates on WO NRs surface, making itself an unconventional catalyst. This study develops a new technique
to clean energy supply and deepens our understanding of MOA process during heterogeneous catalysis.
2
NRs) are appropriate alternative to realize the highly efficient O
NRs play a fundamental role in MOA induced catalytic H
on catalyst surface as well as build up a
NRs catalyst can be achieved by
2
promoted HCHO reforming to
2
2
2
evolution
2
2
2
2
2
2
2
2
2
2
Introduction
Molecular oxygen activation (MOA) is of great importance to
the success of numerous catalytic processes. Usually, MOA
enables the feasible generation of reactive oxygen species
participate in the formaldehyde (HCHO) oxidation reaction to
7
form H
2
and CO
2
as the final products, respectively. It means
that the redox reaction pathways can be readily tuned by
selective controlling the types of ROS generated through MOA
process. However, the use of noble metal prevents MgO-based
catalysts from being applied on a large scale. Therefore,
exploring low-cost and oxygen-stable catalysts with
competitive activity relative to precious metal to produce
desirable ROS are of great importance to realize designated
catalytic processes, such as hydrogen evolution reaction.
It is well accepted that oxygen vacancies (OVs) present in
metal oxide catalysts could serve as shallow donors to improve
the electrical conductivity and donor density and finally
1
, 2
(ROS) via different reaction pathways, and ROS has been
shown to play an indispensable role in many catalytic oxidation
3
reactions. The fact that the oxidation ability of ROS varies
•
−
with its final chemical composition (e.g., •OH > •OOH > O
makes the selective MOA to generate desired ROS products
much valuable for specified catalytic reactions. Nevertheless,
the types of ROS products such as superoxide, peroxide,
singlet oxygen and hydroxyl radical are often difficult to be
precisely controlled in a given catalytic system.5 Recently, we
2
)
4
, 6
8, 9
2
enhance the adsorption of surface species (e.g., O ).
2
reported that molecular O can be selectively reduced on
Moreover, the abundant coordinatively unsaturated sites
provided by OVs could promote the strong electronic
interaction between molecular O and metal oxide catalysts
2
MgO-based catalyst surface to form various ROS in an external
energy-free system. The resulting •OOH species generated on
Ag/MgO surface and •OH radicals on Pt/MgO surface is able to
surface, thus the electron transfer can be spatially accelerated
from catalyst to oxygen species, leading to the implement of
a.Department of Materials Engineering, College of Material and Textiles, Zhejiang
Sci-Tech University, Hangzhou 310018, China. E-mail: lirenhong@zstu.edu.cn
10, 11
MOA process.
In this case, non-stoichiometric tungsten
oxide, which exhibits good performance in electrocatalysis and
b.Emeritus Professor of Department of Chemistry and Materials Technology, Kyoto photocatalysis, is an ideal candidate for MOA.12, 13 Herein, we
Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan. E-mail:
hisabbit@yahoo.co.jp
find that oxygen substoichiometric tungsten dioxide nanorods
2
(WO NRs) are appropriate alternative to realize the highly
c.
Department of Chemistry, College of Science, Zhejiang Sci-Tech University, efficient O
2
promoted HCHO reforming into H
play a fundamental role
evolution reaction, as they
2
as a clean
Hangzhou 310018, China
energy source, where rich OVs in WO
in this MOA induced catalytic H
2
d.Institute of Advanced Synthesis, School of Chemistry and Molecular Engineering,
2
Nanjing Tech University, Nanjing, 211816, China
†
Electronic
Supplementary
Information
(ESI)
available.
See
DOI: 10.1039/x0xx00000x
Please do not adjust margins

Page 3 of 11
J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
ARTICLE
ensure the favorable chemisorption of molecular O
Journal Name
2
on were recorded on a Bruker D8 diffractometer VuieswinAgrticCleuOnKlinαe
catalyst surface as well as build up a channel for the delivery of radiation. UV-vis diffuse reflectance specDtrOaI:(1U0.V10-v39is/CD9RTSA)03w0e51rDe
electrons to O
More importantly, at the molecular level, the MOA process Elmer) equipped with an integrating sphere. The UV-vis DRS of
can be controlled by proton uptake and release, electron solid samples were collected in 200-800 nm against BaSO
2
species.
recorded on a Lambda 750 UV-vis-NIR spectrometer (Perkin-
4
donate and accept to manipulate the formation of ROS. A reflectance standard. High-resolution X-ray photoelectron
typical approach is known as the proton−coupled electron spectroscopy (XPS) measurements were performed in a VG
transfer (PCET).1
4, 15
We show here that the MOA initiated by Scientific ESCALAB Mark II spectrometer equipped with two
NRs catalyst can be achieved by ultrahigh vacuum (UHV) chambers. All binding energies were
systematically modulating the PCET pathway to controllably referenced to the C 1s peak at 284.6 eV of the surface
obtain the desired ROS and thus to directionally produce H , in adventitious carbon. TPO measurements were carried out in a
which molecular O receives the electrons to form superoxide flowing O /He (20%) stream (30 mL/min). The sample was
radicals in a chemisorbed state (O
•−) and couple with protons heated up to 750 °C at a rate of 2 °C/min. The effluents were
provided by HCHO to convert into •OOH radicals. The surface monitored by an online TCD. The harmonics of a nanosecond
stabilized HOO•−WO NRs complex, having radical character Nd: YAG laser and their hydrogen Raman shifted laser lines
and originating from a two-electron reduction of O , acts as were used to generate the 416 nm excitation wavelength
the catalytically active center in the dehydrogenation reaction. employed in the resonance Raman experiments. The excitation
As O continuously uptakes and releases protons by formation laser beam was loosely focused to the quartz tube involving
and consumption of •OOH radicals, O persistently circulates formaldehyde solution and/or solid catalyst as a function of
on WO NRs surface, making itself an unconventional catalyst. reaction time. A heating equipment was used to maintain the
noble−metal free WO
2
2
2
2
2
2
2
2
2
2
o
The reaction mechanism uncovered here not only paves a way reaction mixture at 25 ± 0.5 C. A backscattering geometry was
to find low-cost catalyst for the purpose of clean energy supply used for sample excitation and for collection of the Raman
but also deepens our understanding of MOA process during scattered light by reflective optics that imaged the Raman
heterogeneous catalysis.
scattered light through a polarizer and entrance slit of a 0.5 m
spectrograph. The grating of the spectrograph dispersed the
light onto a liquid nitrogen cooled charge coupled device (CCD)
mounted on the exit of the spectrograph, and the CCD
acquired the Raman signal for about 90-150 s before being
read out to an interfaced personal computer. About 10-30 of
these scans were summed to obtain the resonance Raman
spectrum. The Raman shifts of the resonance Raman spectra
were calibrated using the known vibrational frequencies of the
dichloromethane solvent Raman bands. The spectra of an
intensity calibrated deuterium lamp were used to correct the
resonance Raman spectrum intensities for the variation in
detection efficiency as a function of wavelength, and portions
of the resonance Raman spectra were fitted to a baseline plus
a sum of Lorentzian bands to find the integrated areas of the
Raman bands. X-band EPR signals were recorded on a Bruker
EPR A-300 spectrometer. The settings for the EPR
spectrometer were as follows: center field, 3511.39 G; sweep
width, 100 G; microwave frequency, 9.86 G; modulation
frequency, 100 kHz; power, 101 mW; conversion time, 10 ms.
Experimental
Materials synthesis
Preparation of rod-like WO precursor
x
Commercial WO
3
(200 mg) was dispersed in 15 mL
ethylenediamine under vigorous stirring for 40 min in a 50 mL
Teflon-lined stainless-steel autoclave. Then the system was
o
treated at 180 C for 10 h and cooled down to room
temperature spontaneously. Afterwards the products (namely
WO -precursors) was collected by centrifuge, washed with
x
water and ethanol three times respectively, and dried under
vacuum at 60 °C overnight.
Preparation of WO
To fabricate WO NRs, WO
tube with diameter of 6 mm. The precursors were heated in a
flow under a three-stage heating ramp: from room
2
NRs
2
x
-precursors were put into a quartz
N
2
o
o
o
temperature to 350 C at 5 C/min, and then to 700 C at 1
o
C/min and holding at that temperature for 5 h. Then, the
The solid EPR analysis was carried out by packing WO
2
NRs
product is then naturally cooled to room temperature for
powder into a glass capillary tube, which was placed in the EPR
chamber for testing. The location and the intensity of the g
factors were determined by Bruker’s WINEPR program based
on hν = gβH, where h is Planck’s constant, H is the applied
magnetic field, and β is the Bohr magneton. The spin trapping
experiments were performed as follows: 5,5-Dimethyl-
pyrroline-N-oxide (DMPO) spin-trapping reagent and other
chemicals were purchased from Sigma-Aldrich Chemical Co.
and used without further purification. The formaldehyde
solution containing calculated catalyst and ice-cooled DMPO
solution (0.08 M) was quickly transferred into a glass capillary
tube and tested by EPR spectroscopy at room temperature.
2
further use (named as WO NRs).
By adjusting the calcination time at 700 C to 4 h, 3 h, 2 h
and 1h, a series of reference samples with different oxygen
o
2
vacancy concentrations was prepared (named as WO NRs-4h,
WO
Preparation of WO
WO
2
NRs-3h, WO
2
2
NRs-2h, WO NRs-1h).
3
3
was synthesized using the same method as that of WO
2
2
NRs except that it was prepared in air instead of under N .
Characterizations
TEM images were recorded on a JEOL JEM-1230 operated at
00 kV. HR-TEM was performed on a FEI TITAN Cs-corrected
ChemiSTEM, operating at 200 kV. Wide-angle XRD patterns
1
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 4 of 11
Journal Name
ARTICLE
Catalytic experiments
case, the rod-like morphology of WO
x
View Article Online
-precursors is well
Catalytic H
out with 40 mg of catalyst suspended in 5 mL of aqueous ethylenediamine are able to extract O atoms from WO
HCHO solution (1.0 M) in a 50 mL Pyrex test tube under stirring precursors to form NO , so that substoichiometric WO could
400 ± 10 rpm). A water bath was used to maintain the be formed successfully. HRTEM image displays clear lattice
reaction temperature at 25 ± 0.5 °C. The oxygen content in the fringes with a crystal plane spacing of 0.345 nm (Fig. 1d),
reaction tube was adjusted by bubbling with pure O gas for a corresponding to the (011) plane of monoclinic WO . XRD
calculated time, and the tubes were finally sealed with silica pattern shown in Fig. 1e demonstrates that the as-prepared
2
production from HCHO/H
2
O solution was carried preserved (Fig. 1c), and more impoDrtOaIn: t1l0y.1,03N9/Ca9tToAm03s051iDn
x
-
x
2
2
2
(
2
2
gel stoppers. Gas volumes of 400 μL were extracted from the WO
test tubes using a microliter syringe at regular intervals, and WO
2
NRs are consisted of high purity monoclinic structured
(PDF No. 32-1393), which agrees well with its SAED
2
GC-TCD was employed for evaluating the gas evolution pattern (inset of Fig. 1d). The small diffraction peak around 24°
amount, including H , O . The TOF reported here is an could be assigned to the in situ formed graphitized carbon,
apparent TOF value based on the surface number of WO in which serves as a rigid substrate not only benefiting the fast
the catalyst, which is calculated by the following equation: TOF electron transfer but also improving the catalyst stability.23 XPS
(nH2/nWO2 × t) × (d/1.3), where nH2 and nWO2 represent the of W 4f core level spectra of WO NRs (Fig. 1f) reveals the
molar amounts of evolved H within t h and WO catalyst, existence of typical +4 oxidation state of W atoms (32.85 and
respectively, and d is the average diameter of WO NPs (~150 34.25 eV) for WO as well as the +6 oxidation state (37.90 and
nm calculated using Scherrer’s equation based on the peak 35.62 eV) and +5 oxidation state (36.60 and 34.80 eV) of W
width of the WO (111) reflection) and d/1.3 is the empirical atoms due to the partial oxidation of the surface layers in air,
surface atom dispersion degree for nanoscale particles. The which is consistent with reported characteristic of tungsten
2
2
2
=
2
2
2
2
2
2
2
4-26
recycling experiment is carried out as follows: after 3 h of oxides.
Based on the above characterization results, we
NRs are obtained.
reaction (1 cycle), the test tube was opened and then purged can conclude that high purity of WO
2
with oxygen for 5 min to remove residual H
for the next cycle.
2
, and again sealed Importantly, the absence of O atoms at specific locations
results in the unsaturation of W atom coordination, inducing
10, 11
the lattice disorder and dislocation,
production of high concentration OVs in WO
which enables the
NRs.
DFT calculations
The plane wave based program, Castep was employed to carry
2
16,
out DFT calculations with the periodic boundary conditions.
1
7
The Perdew-Burke-Ernzerhof (PBE) functional was used
together with the ultrasoft-core potentials.1
8-20
The basis set
cutoff energy was set to 300 and 340 eV for geometry
optimization and post-energy calculation, respectively. The
1
2
2
electron configurations of the atoms were H: 1s , C: 2s 2p , O:
2
2
4
2
6
4
2
s 2p , and W: 5s 5p 5d 6s . (WO
main part of catalyst, and a graphitized carbon molecule of
14 was put down under the WO slab as the support, as
2
)20 slab was employed as a
C
30
H
2
shown in Fig. S21†. The unit cell has a diamond shape, and the
lattice parameters are a = c = 11.126 Å, b =30 Å (including
vacuum region), α = γ = 90°, and β = 120.47°.
Results and discussion
Tungsten dioxide nanorods (WO
synthesized by calcining the WO
temperatures in different atmosphere.21 WO
2
NRs) or WO
-precursors at high
-precursors
3
were
x
x
exhibits a rod-like morphology and good crystalline state (Fig.
S1†). Fig. S2† reveals two characteristic peaks in W 4f region
assigned to +6 oxidation state, and no other oxidation state of
W in the WO
WO NPs shows that a large number of NPs are randomly
formed (Fig. 1a), most likely due to the destruction of WO
x
-precursors is observed. The TEM image of the
3
x
-
precursors when calcined in air. HRTEM (Fig. 1b) reveals the
lattice spacing of ~0.37 nm, corresponding to the (100) plane
of WO
the products can be indexed to pure monoclinic phase WO
PDF No. 43-1035). However, WO NRs are produced by
3
. XRD and XPS results shown in Fig. S3† illustrate that
Fig. 1 (a) TEM and (b) HRTEM images of WO
images of WO NRs sample, (e) XRD (PDF No. 32-1393) and (f) XPS in the W 4f region
patterns of WO NRs.
3
NPs sample, (c) TEM and (d) HRTEM
3
2
(
2
2
2
changing the calcination atmosphere from air to N . In this
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Page 5 of 11
J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
ARTICLE
Journal Name
peaks at 37.89 and 35.61 eV are observed in the spectrum,
View Article Online
DOI: 10.1039/C9TA03051D
which is a typical feature of +6 W. The XPS spectra
demonstrate that after oxidizing with H
2
O
2
, the valence of W
has been changed into W . Remarkably, the XRD patterns (Fig.
S4c†) of WO NRs-H and WO NRs are very similar,
confirming that the WO NRs-H sample still maintains the
WO monoclinic phase. The above results clearly demonstrate
that the production of OVs is directly related to the
6+
2
2
O
2
2
2
2 2
O
2
6+
5+
4+
replacement of W /W by W .
Furthermore, determining the specific location of OVs is of
importance to elucidate the MOA and subsequent
dehydrogenation process. As shown in Fig. 1c and 1d, on
account of the rod-like morphology and the lattice spacing
2
(0.345 nm), it is found that the WO NRs are enclosed by (011)
plane, where OVs are mainly located. Indeed, OVs are created
by removing terminal O atoms of W−O bonds within
x
stoichiometric WO -precursors. The transformation process
Fig. 2 (a) Solid EPR spectra of WO
UV-vis diffuse reflectance spectra of WO
and WO , and (d) EPR detection of superoxide radicals formation using DMPO as the
spin-trapping agent in HCHO solution with WO NRs or WO
2
NRs and WO
3
carried out at 298 K in open air, (b))
from WO
x
-precursors to WO
2
is monitored by in-situ XRD (Fig.
2
NRs and WO
3
, (c) TPO studies of WO NRs
2
o
S5†). Upon heating to 250 C, there is no significant change in
the XRD peaks, while keeping heating to 350~600 C, WO2.92
forms (PDF No. 30-1387). Notably, new XRD peaks appear in
3
30
o
2
3
.
o
the case of 700 C calcination, which can be indexed to
Solid electron paramagnetic resonance (EPR) and UV/vis
absorption spectra further confirm the presence of OVs. As
illustrated in Fig. 2a, no apparent EPR signals are detected in
monoclinic WO
and ~37.2°, respectively. It indicates that
transformation from WO2.92 to WO with lattice parameters a
0.558, b = 0.490, and c = 0.556 nm is completed. Moreover,
2
with strong (011) and (002) peaks at ~28.7°
a
phase
2
3 2
WO samples. However, WO NRs exhibit a symmetrical ESR
=
signal at g = 2.002, which could be designated to the electrons
the peak of the (001) plane of WO
2
NRs gradually increases as
trapped by OVs.2
7, 28
Moreover, the signal at g = 2.002 is most
o
a function of calcination time at 700 C (Fig. S10b†), possibly
due to the lattice compression caused by the formation of
OVs. This transformation is accompanied by a loss of O atoms,
•
−
likely attributed to the surface localized O
2
, indicating that
2 2
WO NRs prefer to reduce O into superoxide. UV-vis diffuse
reflectance spectra is an effective method to prove the
presence of OVs in tungsten oxides, because their existence
can narrow the bandgap and increase the conductivity by a
large degree.29 As shown in Fig. 2b, a large absorption tail
present in the visible and near infrared regions gives clear
resulting in the vacancies with WO
6
octahedral units sharing
their corners, producing different length of W−O bonds.1
1, 21
In
comparison with WO2.92, the WO
2
with abundant OVs
possesses a reduced total coordination number. Specifically,
the reduction of coordination number mainly occurs at the
evidence that the WO
detail, WO without OVs only displays intrinsic absorption in
the narrow range of 350~480 nm. As compared with WO , an
additional broad absorption (480~800 nm) in WO NRs is
2
NRs contain a large number of OVs. In
11
longer W−O bonds, identifying the locations of OVs. The rod-
like structure of the WO NRs leads to a high intensity of the
00k) diffraction lines associated with the basal planes of WO
3
2
3
(
2
2
due to preferential orientation. A contraction of the c-axis
combined with a elongation of the b-axis is observed from
observed, which is attributed to the new discrete energy bands
below the conduction band created by the OVs. Thus, the
significant difference in UV-visible diffuse reflectance spectra
31, 32
WO
2
NRs-1h to WO
2
NRs.
Therefore, it is supposed that
OVs mainly replace bridging O atoms in the c-direction W−O
bonds. Additionally, in-situ Raman is used to further
between WO
photophysical properties of WO
evidence that WO NRs possess a large amount of OVs.
Moreover, it is deduced that the W4+ species produced by
the removal of O atoms in the WO -precursors during
calcination are the key oxidation states associated with the
generation of rich OVs. To prove this, WO NRs treated with
hydrogen peroxide (H ) for 1 h (denoted as WO NRs-H
3
and WO
2
NRs not only shows unusual
2
NRs, but also provides strong
o
investigate the phase transformation from 600 to 700 C (Fig.
2
−
1
S6†). The two strongest peaks at 710 and 802 cm are
attributed to stretch vibration mode of W−O bonds
x
−
1
(νas(W−O)), and peaks at 118 and 253 cm assigned to the
wagging vibration mode of W=O bonds (δ(W=O)). The increase
2
o
in peak intensity of W=O at 700 C is ascribed to the loss of O
2
O
2
2
2 2
O )
atoms from W−O bonds, resulting in the formation of a
are used for comparison. Its corresponding EPR spectrum (Fig.
S4a†) only shows a tiny peak at g = 2.002, indicating that the
33
fraction of W=O bonds. The Raman bands at 316 and 908
−
1
cm are assigned to the bend and stretch vibration of the
terminal W=O bonds along the a- and b-axes, respectively,34
whose bond distances are shorter than that of the W–O bond
OVs on the surface of WO
oxidizing with H . The change in the valence state of W
atoms before and after H treatment is characterized by
XPS (Fig. S4b†). It is found that the WO NRs-H sample
behaves inversely to the untreated WO NRs: only two sharp
2
NRs nearly disappear after
2
O
2
2
O
2
3
1, 35
distance along the c-axis.
The generation of OVs should
2
2 2
O
therefore lead to anion vacancies along the c-axi. Based on the
above results, it is concluded that OVs are mainly located on
2
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 6 of 11
Journal Name
011) faces of WO
ARTICLE
(
2
NRs and they predominantly replace OVs through electron transfer, which is consistent with the
View Article Online
DOI: 10.1039/C9TA03051D
bridging O atoms in the W−O bonds along the c-direction.
solid EPR results.
Since the generation of OVs is caused by the loss of O atoms
2 2
Since the strong electronic interaction between O and WO
in the W−O bonds, a large number of W atoms with a low NRs is able to induce the production of ROS, it provides a
oxygen coordination are generated, it is thus suggested that generic platform for MOA initiated catalytic reactions. As
OVs on WO
unsaturated W sites for O
catalytic system. The interaction between oxygen species and (pO2 = 1 atm). Nevertheless, for this particular system the
2
NRs surface can provide coordinatively shown in Fig. 3a and Fig. S8†, H
2
is produced steadily over
2
molecular adsorption in this WO NRs catalyst from HCHO solution in the presence of O
2
2
−
1
WO
2
NRs catalyst is studied by high resolution O 1s XPS reaction rate is rather slow (0 M NaOH, TOF = 2.7 h ). Based
analysis (Fig. S7†). The peaks at 530.1 and 530.5 eV are on previous studies, the reaction could be accelerated by
respectively attributed to the W-O bonds in WO
3
and WO
2
simply adding bases into the reaction system, while NaOH
11, 36
44, 45
NRs.
However, the O 1s region in WO
2
NRs reveals two appears to be the optimal candidate.
Indeed, the catalytic
extra peaks at 532.2 eV and 531.5 eV: the former one is activity is dramatically improved in the presence of NaOH. The
relative to hydroxyl groups (−OH) belonging to H
2
O adsorbed
2
H generation rate increases with the increased NaOH
on the catalyst surface;3
7, 38
the latter could be assigned to concentrations to an optimum level of 0.6 M (Fig. 3a), where
−
1
oxygen species chemisorbed at OV sites with a low oxygen the TOF value reaches 173 h (nearly 64 times higher than the
coordination.3
9, 40
Based on the DFT and XPS results (Fig. 5b case without base). As compared with commercial WO
is more preferentially nanoparticles (denoted as WO NPs, Fig. S9a†), WO NRs
and competitively adsorbed on the catalyst surface than that exhibit ~1.4 times higher H evolution rate than that of WO
of H O molecules. It is thus reasonable to deduce that WO NPs under the same reaction conditions (Fig. S9b†), which is
NRs have strong oxygen adsorption and activation property probably attributed to the faster electron transfer rate caused
over OVs. by the in-situ formed graphitized carbon. Additionally, we
2
and Fig. S23†), it is suggested that O
2
2
2
2
2
2
2
Transition metal oxides are favored by oxidizing agents conducted the catalytic reaction with different concentrations
where OVs are located.4
1, 42
Thus, it is possible to correlate O
of pure NaOH solution (WO
was produced in all cases (Table S3†),
programmed oxidation (TPO) process (Fig. 2c). In this case, indicating that HCHO is an indispensable hydrogen donors in
2
2
NRs + 0.2~5 M NaOH) as the
chemisorption performance with OVs by temperature- substrate, while no H
2
WO
3
gives no oxidation peak over the entire temperature this catalytic dehydrogenation system. Furthermore, the blank
o
range (50‒750 C). In comparison, WO
2
NRs exhibit two weak experiments using 1 M HCHO + 0.2~5 M NaOH mixture
peaks, with significantly lower initial oxidation temperature without catalysts were also carried out under otherwise
o
(
~60 C), indicating that O
2
can be readily activated into ROS identical reaction conditions. The negligible H
surface. However, given the weak oxidation peak indicates that NaOH acts as a promoter for H production from
intensity, WO NRs are actually highly resistant to oxidation HCHO solution (Table S3.† and Fig. S9b† ). We propose that
even O is easily activated on its surface. Benefit to the the addition of NaOH can facilitate the formation of hydrated
2
production
on WO
2
2
2
2
coordination unsaturated sites provided by OVs, an HCHO oligomer (the Cannizzaro intermediate). It is known that
unparalleled chemisorption effect is expected to be formed HCHO in water is in principal solely present in the form as
between O
accompanied by electron transfer from the coordination >1), which is the real ingredient for dehydrogenation.
unsaturated sites to the O species, while OVs-deficient WO Previous studies have shown that HCHO in alkaline solution
surface can hardly supply the same sites for the chemisorption tend to produces hydrated HCHO oligomer intermediates and
and electronic coupling of O . It is thus anticipated that the then produce formic acid and methanol (i.e., Cannizzaro
strong coupling between WO
their electron transfer, would act as a bridge to enable energy system, it is suggested that appropriate concentration of NaOH
transfer from WO NRs surface to O species, leading to the (0.2~0.6 M) contributes to the formation of hydrated HCHO
achievement of MOA and subsequent production of ROS. oligomer and thus accelerates the dehydrogenation reactions.
Liquid EPR analysis is adopted to detect possible ROS to This dehydrogenation pathway is further confirmed by in-situ
follow the evolution of O in our catalytic system. As shown in resonance Raman spectroscopy (Fig. 6). However, much
Fig. 2d, a group of signals labeled by black dots emerged in the concentrated NaOH (1~5 M) may enhance the competition of
suspension containing WO NRs, HCHO solution and DMPO Cannizzaro reaction, resulting in the decline of H production
rate (Fig. 3a and Fig. S8†). It should be noted that the present
NRs is the first noble-metal-free catalyst for HCHO
2 2 2 2 n
species and WO NRs. The chemisorption is hydrated HCHO oligomer (HCHO + nH O → HO(CH O) H, n
46, 47
2
3
2
48
2
2 2
NRs and O , characterized by reaction). Therefore, in the present catalytic H evolution
2
2
2
2
2
•
−
trapping agent, which can be assigned to the DMPO−O
2
adduct.7 No obvious EPR signal is detected when WO
, 43
NRs is WO
2 2
instead by WO
activate surface adsorbed O
3
. It means that WO
2
NRs can effectively dehydrogenation with such a high TOF value, which is even
species at abundant higher than most of heterogeneous noble metal catalysts
•
−
2
into O
2
(Table S1.†).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Page 7 of 11
J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
ARTICLE
Journal Name
the initial rate against pO2, it is found that the H generation
View Article Online
2
DOI: 10.1039/C9TA03051D
rate increases almost linearly with the increase of pO2 (Fig. 3d).
In contrast to conventional MOA catalytic reactions, although
molecular O
evolution process, it is not consumed during the entire
reaction regardless of the initial concentration of O (inset of
Fig. 3d and Fig. S12†). It means that O also acts a sort of
special “catalyst”, which cooperates with WO NRs to
accelerate the rate of H evolution by HCHO reforming.
A proper HCHO concentration is in favour of the interaction
between HCHO molecules and dissolving O molecules on
catalyst surface, leading to the favorable production of H
2 2
plays an indispensable role during this H
2
2
2
2
2
2
.
Thus, the catalytic hydrogen production experiments by using
different concentrations of HCHO (up to 10 M) as the substrate
were conducted. For the present reaction, the ideal HCHO
concentration for dehydrogenation is in the range of 1~4 M.
The H
2 2
evolution rate over WO NRs catalyst increases
Fig. 3 (a) The effect of NaOH concentrations on H
performance of WO NRs, WO NRs-4h, WO NRs-3h, WO
WO in dehydrogenation of HCHO solution; pO2 = 1 atm; NaOH concentration = 0.6 M,
and performance of WO NRs in N atmosphere, reaction time: 3h, (c) the effect of pO2
on the H
2
evolution at pO2 = 1 atm, (b)
significantly with increasing the HCHO concentration to a
maximum at 2 M corresponding to a TOF of 189 h (pO
atm), and gradually decreases to 98 h at a concentration as
2
2
2
2
NRs-2h, WO NRs-H and
2
2 2
O
–
1
2
=1.0
3
–
1
2
2
evolution rate as a function of reaction time; NaOH concentration = 0.6 M, high as 10 M (Fig. S13†). To achieve high energy density, high
2
and (d) the double logarithmic plots of the reaction rate against pO2 (the inset shows HCHO concentration is usually necessary. However, high HCHO
the O
temperature = 25 C.
2
content evolution as a function of reaction time at pO2 = 1 atm); reaction
concentration is also favorable for the undesired
disproportionation of the aldehyde (Cannizzaro reaction),
o
It is highly possible that the amount of OVs may affect the which follows the second order kinetics. Consequently, high
production performance because OVs play a fundamental HCHO concentration is disadvantageous for the H evolution in
H
2
2
role in oxygen activation in this catalytic system. To correlate these catalytic systems. It is detrimental to the parallel
catalytic performance with OVs, a series of reference samples dehydrogenation reaction pathway. Further, there is no
–
1
with different OV concentrations was prepared by adjusting significant improvement in activity at 2 M (TOF = 189 h ) as
o
–1
the calcination time (x h) at 700 C (denoted as WO
Solid EPR spectra (Fig. S10a†) show the OV concentrations of solution is utilized for the investigation of dehydrogenation
the samples gradually reduce as calcination time decrease. In reaction. Moreover, WO NRs catalyst still shows promising
addition, their crystal structures are not affected as evidenced hydrogen evolution activity even in highly concentrated HCHO
2
NRs-xh). compared to 1 M (TOF = 173 h ). Accordingly, the 1 M HCHO
2
-
1
by XRD (Fig. S10b†). Fig. 3b shows the H
performance as a function of WO
NRs with different OVs higher than most of previous reports (Table S1.†).
concentrations. Because the OVs follow the order of WO NRs
The effect of reaction temperature on the rate of H
NRs-2h, it turns out that generation is also studied by setting the concentration of
2
evolution solution (10 M) with corresponding TOF at 98 h , which is
2
2
2
>
WO
2
2
2 2
NRs-4h > WO NRs-3h > WO
the H
evolution rates obeys a strong first order correlation NaOH at 0.6 M (Fig. S14a†). The initiate rate of H
2
generation
o
with the OVs concentrations (Fig. S11†). It is worth noting that increases rapidly from 0 to 50 C with corresponding reaction
-
1
-1
only negligible H
WO or WO NRs-H
abundant OVs in this MOA induced catalytic H
reaction. Remarkably, the calculated kinetic rate constant for the ln(TOF) versus 1/T (Fig. S14b†). From the linear Arrhenius
2
is generated when WO
(Fig. 3b), indicating the importance of the H
evolution temperature range, the Arrhenius plot is obtained by plotting
2
NRs are replaced by rate increasing from 2.35 to 12 mmol h g catalyst. Given that
evolution rate is constant with time in the given
3
2
2
O
2
2
2
−
1
WO
NRs-H
boosted by increasing the number of OVs, manifesting their reaction. Further increasing the temperature leads to the
decisive role as catalytically active sites.
reduction of hydrogen production, possibly due to the
In addition, trace amount of H is generated using WO
as a catalyst in N atmosphere demonstrates that O
important role (Fig. 3b). Benefit from the rich OVs on the
surface of WO
NRs catalyst and the strong electronic free from the common concern that surface defects typically
interaction between O and OVs, the O
concentration is show low durability in catalytic reactions. Impressively, WO
obviously vital to the efficiency of H
evolution (Fig. 3c). NRs exhibit an excellent stability in 5 catalytic cycles (Fig.
Indeed, the H
evolution rate is significantly enhanced by S15a†), consistent with the good cyclic performance of
increasing oxygen partial pressure (pO2) from 0 to 1 atm, commercial WO NPs (Fig. S15b†). In detail, the catalytic
2
-NRs reaches 173 h , ~22 times higher than that of WO
2
-
behavior, the apparent activation energy (E
(7.9 h ). It means that the reaction kinetics can be be 21.3 kJ/mol for WO NRs catalyze HCHO reforming
a
) is calculated to
−
1
2
O
2
2
NRs instability of
plays an temperatures.
oxygen-centered radicals
at higher
NRs are
2
2
7
2
2
Another significant finding is that the OVs-rich WO
2
2
2
2
2
2
2
2
-1
corresponding to a TOF increasing from 0 to 173 h (Table activity is even slightly increased within 2-cycle reaction, and
S2.†). Importantly, by fitting the double logarithmic plots of then gradually degraded (recycle 3-4). The reason for the
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 8 of 11
Journal Name
ARTICLE
increase in activity in the first 2 cycles is supposed to be the proton and electron transfer are not concerted but rather
View Article Online
DOI: 10.1039/C9TA03051D
•−
residual products (such as NO
x
, C
x
H
x
) of ethylenediamine sequential. In detail, O
2
is preferentially reduced to O
2
adsorbed on the catalyst surface are cleared after reaction, through one-electron transfer process, which is coupled with
producing more active sites on catalyst surface. The slight free protons to form the key reactive oxygen species •OOH. In
degradation of activity may be derived from the consumption NaOH-free system, proton transfer to the formation of •OOH
of a small amount of OVs after several cycles. As Compared is the rate-determining step of PCET, leading to the slow rate
with the fresh catalyst, the XRD spectra of the used catalysts of H
2
production, while the addition of NaOH largely
show limited change in crystal phase after recycling (Fig. S16†). accelerates the proton coupling step by readily C−H bond
XPS analysis shows the peaks of W 4f and O 1s are almost cleavage in HCHO to form •H radicals as well as •OOH species.
unchanged before and after catalytic reaction (Fig. S17†), This can be explained by the fact that the nucleophilic addition
indicating that the valence state of WO
2
NRs has not been of a base to a carbonyl group enhances the electron repelling
altered yet. A small decrease in peaks intensity of +4 and +5 ability of C=O, resulting in the weakening of the chemical bond
51
oxidation state of W atoms is possibly due to the slight between carbon and hydrogen atoms, and finally facilitating
oxidation of part W atoms on WO surface. Besides, we the entire PCET process. Thus, the dehydrogenation of HCHO
2
carried out the TEM and HRTEM examinations to obtain the
structural information of used catalysts. The well maintained
rod-like structure and the 0.345 nm crystal plane spacing
corresponding to the (011) plane of WO
stability of structure (Fig. S18†). The high stability of WO
2
further confirm the
NRs
2
most likely benefits from in situ-formed carbon act as a rigid
substrate, which is able to improve the stability and resistance
toward oxidation.22 Moreover, solid EPR and UV/vis spectra
were employed to investigate the OVs of WO
reaction. The solid EPR spectra reveal that the signal at g =
.002 decline slightly (Fig. S19†). UV/vis spectra (Fig. S20†) also
2
NRs after
2
show that even there is a slight red shift caused by a small
amount of OV consumption, it still maintains effective
absorption intensity throughout the entire range (350~800
nm). Thus, we propose that WO
reaction, demonstrating that the OVs are stable on WO
2
NRs maintain their OVs after
NRs
2
catalyst surface. Obviously, the remarkable stability is of great
importance for future practical catalytic applications.
2 2
Fig. 3 (a) DMPO adducts of the WO NRs reaction system recorded in the O
Due to the fact that O
WO NRs, it is highly possible that O
electrons from OVs to form certain ROS during this
heterogeneous H evolution reaction. Thus, in-situ EPR spectra
and spin-trapping technique are utilized as a semi-quantitative solution is more favorable under alkaline condition.
analysis to detect potential oxygen-centered radicals. As
It is noted when the suspension is purged with N2 in
shown in Fig. 4a, a pronounced 6-fold signal emerged in a advance, no •OOH or •H radicals are found neither in WO2
suspension containing WO NRs catalyst, HCHO solution and NRs nor WO3 systems under otherwise identical reaction
DMPO in O atmosphere (blue line), which can be ascribed to conditions (Fig. 4b), indicating the decisive role of O in the
the DMPO−O
9.3, α = 16.5). It means that O
species through electron transfer on WO
However, after the addition of NaOH, the signal of O
2
interacts strongly with OVs within atmosphere without NaOH and after the addition of NaOH, (b) DMPO adducts
uptakes or releases recorded in the WO NRs or WO reaction systems containing HCHO, H O and NaOH in
open air and (c) in N atmosphere, (d) RDE voltammograms for oxygen reduction
catalyzed by WO NRs or WO in oxygen-saturated NaOH/HCHO solution.
2
2
2
3
2
2
2
3
2
2
2
2
•
−
2
adduct (hyperfine splitting constants: α
is initially activated into O
H
=
dehydrogenation kinetics. In addition, no obvious signal is
•
−
1
N
2
2
detected by EPR spectra in WO system (Fig. 4c), because such
3
2
NRs surface. efficient electron transfer (ET) can only be realized by the
•
−
2
2
strong electronic coupling between O species and OVs site.
disappears while some positive new signals appear Previous reports showed that the enhanced W−O bond
immediately (Fig. 4a, red line). Among them, a sextet signal covalency in defective tungsten oxides may promotes the
labeled by green squares can be attributed to DMPO−•OOH coupling between
W atoms and adsorbates through
H
H
7, 49
adduct (α
probably derived from the combination of O
radicals. The formation of •H radicals (α
N
= 14.5 G, α
β
= 14.6 G, and α
γ
= 1.5 G),
which is chemisorption, benefiting the electron transfer for the
and •H formation of O2 , indicating that OVs are essential for the
•
−
•− 9
2
H
=16.6 G; αH(1) = αH(2) success of PCET.
=
22.6 G) is determined by another group of characteristic 9-
The potential of WO2 NRs to form peroxide species by
7, 50
fold signals marked by blue triangles.
Moreover, the reduction of O2 is further demonstrated by rotating disk
intensity of •OOH and •H signals are almost not decayed after electrode (RDE) measurements. Fig. 4d shows the steady-state
0 min reaction (Fig. 4a, purple line). Consequently, it is voltammograms measured with two different electrocatalysts
3
suggested that the •OOH species is derived from a unique supported on a glassy carbon electrode in O saturated 0.1 M
2
proton-coupled electron transfer (PCET) process, where the NaOH electrolyte. The WO3 electrode exhibits a one-step
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Page 9 of 11
J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
ARTICLE
Journal Name
process over –0.3 V with a steady-state diffusion current (black rate-determining step, whereas the relative eVnieewrAgryticleisOnl9in4e
curve), suggesting a four-electron pathway for oxygen kJ/mol. The peroxide species (•OOH) iDsOmI: 1u0c.1h03s9t/aCb9iTliAz0e3d05b1Dy
reduction reaction (ORR). In contrast, the WO NRs electrode bonding to the W site [#10] (–315 kJ/mol) (Fig. S25†). At this
2
with abundant OVs displays a two-step process for ORR with stage, the formed HOOH is desorbed (i.e., removed from the
the onset potential of about –0.24 V and –0.63 V, respectively system for calculation), and •OOH and H remain adsorbed on
(
blue curve). The first sharp step over –0.24 V can be ascribed the catalyst. In the final step, H
2
and O
The two-step •OOH−(W) and H−(O), for the reaction H(−O) + HOO(−W) → H
(−W) + (O). All the elementary reactions (#11 through #13)
electrolyte (red curve), while the onset potentials positively proceed on the WO site (Fig. S26†). The stoichiometry for the
2
are generated from
–
52, 53
to the two-electron reduction of O
process maintains after adding HCHO solution to the + O
2
to HO
2
.
2
2
2
shift to –0.19 V and –0.49 V, respectively, with a concomitant overall dehydrogenation reactions over WO
increase in the current density. One possible reason is that the suggested to be as follows:
2
NRs is thus
hydrogen atom provided by HCHO contributes to produce HCHO + O
peroxide species. It indicates the crucial role of peroxide
2
+ H
2
O → •OOH + •H + HCOOH → O
2
+ H +HCOOH
2
In-situ resonance Raman spectroscopy is a powerful tool to
species (•OOH) serving as central reactive species toward monitor the dissociation of molecular bonds in organic
reactants during catalysis under real reaction conditions. The
spectra of WO
2 2
NRs catalyzed H production from alkaline
HCHO solution in air recorded in the frequency range of
−
1
4
00−2000 cm are shown in Fig. 6. It is well known that HCHO
molecules easily combine with water molecules to form
hydrated HCHO oligomer (HCHO + nH HO(CH O) H, n >
Thus, the C−O, C−H and O−H bond cleavage in
O) is essential to understand the real
2
O
→
2
n
4
6, 47
1
).
HO(CH
2
n
H
dehydrogenation process.
Due to the presence of the O−C−O groups within the
HO(CH
037 cm− from aqueous HCHO solution are assigned to the
C−O bending vibrational mode (δ(OCO)), symmetric mode
2 n
O) H oligomer chains, the bands around 538, 908 and
1
1
(
ν
s
(OCO)) and anti-symmetric mode (ν
α
(OCO)) respectively
5
4
(Fig. 6a, red line). The intensity of these bands hardly
changes after adding NaOH (blue line), while monotonically
2
Fig. 4 (a) Energy changes along the reactions occurring on a (WO )20 slab, and (b) the
structures of four transition states (TS’s) based on (WO )20 slab.
2
decreases or even disappears upon the addition of WO
evidencing a rapid break of C−O bond in oligomer on WO
2
NRs,
NRs
2
surface. Similar results are also obtained in the frequency
dehydrogenation of HCHO solution.
−1
range of 1200−2000 cm (Fig. 6b). Two bands observed at
DFT calculation is carried out to gain further insight into the
reaction pathway at the molecular level. The catalyst model is
shown in Fig. S21†, and the energy changes for the overall
−1
around 1260 and 1490 cm can be attributed to the δ(HCH)
deformation modes belonging to the C−H bond of
HO(CH
NRs catalyze H
that the H−C−H bonds in HO(CH
2
O)
n
H. Similar intensity decrease is observed as WO
production from HCHO solution, indicating
O) H are constantly disrupted
most likely corresponding to the proton coupling step: HCHO
2
reactions on (WO
2
)20 model catalysts are shown in Fig. 5a. The
2
whole reaction consists of 13 configurations, including 9 local
minima and 4 transition states (TS’s), and their optimized
structures are shown in the ESI with configuration numbers #1
to #13 (Fig. S22-S26†), and only TS’s are shown in Fig. 5b. The
2
n
(
+
•−
O
2
2 n
→ HCO• + •OOH) and HO(CH O) H oligomers are
completely decomposed. The bands observed at around 1411
reaction begins with the adsorption of O
site (#2), O is adsorbed on the W site with an end-on
coordination (Fig. S22†). This form of adsorption facilitates the
smooth transfer of electrons from catalysts to O species,
preparing for the formation of peroxides. Next, water is co-
adsorbed on the WO surface and dissociated into OH and H
#3 to #5) (Fig. S23†). Then HCHO is further adsorbed
2
molecular at the W
−1
and 1448 cm with or without NaOH (blue and red line) both
2
correspond to the COH in−plane bending modes connected
55
with the OH group. Similarly, the intensity of the two bands
decreases and almost disappears after the addition of WO
2
2
NRs as a function of reaction time, suggesting that the OH
groups are easily dissociated over the catalyst. Besides, the
2
(
−1
δ(OH) bending vibrational mode at 1632 cm corresponding
molecularly onto this surface complex, and the total
adsorption energy is calculated as −230 kJ/mol. The next step
to the water molecules adsorption on the catalyst and
interaction with HO(CH
2
O)
n
H
oligomer. The gradual
is H abstraction from HCHO by an O
which is schematically represented as (W)−O
W)−OOH + HCO•. The formed HCO• has a stable isolated
2
molecule [#6 to #7],
enhancement of the band intensity and slight red shift of the
band suggest that H−O bonds between water molecules and
oligomers are decomposed as the catalytic reaction proceed
2
+ HCHO →
(
configuration #8 (Fig. S24†). Restarting optimization from a
slightly different structure leads to HCOOH (•OH + HCO• →
HCOOH) as configuration #9, this step is considered to be the
(
H
corresponding to the H
2 2
O dissociation on WO NRs surface,
2
O → •OH + •H), suggesting that water acts as a hydrogen
donor in the catalytic reaction. Moreover, a new peak appears
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 10 of 11
Journal Name
ARTICLE
at 1350 cm−1 after NaOH is added, which is ascribed to the radicals, O
symmetric stretch vibration (ν
generated by the Cannizzaro reaction of trace HCHO.
can be regenerated at the end of the reaction,
(OCO)) of surface formate making itself an unconventional catalyst for the present H
2
2
View Article Online
DOI: 10.1039/C9TA03051D
s
51, 56
This evolution reaction. In addition, because the proton transfer to
peak gradually increases after the addition of catalyst, which is the formation of •OOH is the rate-determining step, the
caused by the formation of formate during catalytic addition of NaOH is able to accelerate the proton coupling step
by readily C−H bond cleavage in HCHO to form •H radicals as
well as •OOH species, thus finally leading to the enhanced H
2
production. The reaction mechanism uncovered here endows
a new insight into the molecular oxygen activation process
during heterogeneous catalysis and paves an avenue for clean
energy supply from biomass solution using specified surface
engineered catalyst.
Acknowledgements
We are grateful for financial support from the National Natural
Science Foundation of China (21872123, 21503189 and
Fig. 5 In-situ resonance Raman spectroscopy of WO
2
NRs catalyze HCHO solution (1 M)
in air at 25 C in wavenumber range of (a) 400−1200 cm and (b) 1200−2000 cm-1.
o
-1
2
1403197) and Zhejiang Provincial Natural Science Foundation
−
+
dehydrogenation (HCO• + HO• → HCOOH → HCOO + H ).
The exact reaction mechanism could be elucidated based on
the above reaction processes. First, the oxygen vacancies
of China (LY18B030007 and LY15B030009).
provide abundant surface sites for O
likely with an end–on configuration. Subsequently, the
2
chemisorption, most Notes and references
1
S. Hu, W. Xiao, W. Yang, J. Yang, Y. Fang, J. Xiong, Z. Luo, H.
Deng, Y. Guo, L. Zhang and J. Ding, ACS Appl. Mater. Interfaces,
2018, 10, 17167-17174.
W. H. Saputera, J. Scott, H. Tahini, G. K. C. Low, X. Tan, S. Smith,
D.-W. Wang and R. Amal, ACS Catal., 2017, 7, 3644-3653.
H. Zhang, L. H. Guo, D. Wang, L. Zhao and B. Wan, ACS Appl.
Mater. Interfaces, 2015, 7, 1816-1823.
K. Zhao, L. Zhang, J. Wang, Q. Li, W. He and J. J. Yin, J. Am.
Chem. Soc., 2013, 135, 15750-15753.
N. Waiskopf, Y. Ben-Shahar, M. Galchenko, I. Carmel, G.
Moshitzky, H. Soreq and U. Banin, Nano Lett., 2016, 16, 4266-
•
−
chemisorbed O
electron transfer over WO
the success of the overall H
2
is activated into O
NRs surface, which is a key step to
evolution reaction. Then, the
species at OVs could attack C−H bond in HCHO
2
species through one-
2
2
2
3
4
5
•
−
resulting O
2
molecule with the generation of •OOH radical by a PCET
process. NaOH facilitates the cleavage of C−H bonds in HCHO
by accelerating the proton coupling step to easily form •OOH
species, overcoming the difficult cleavage of C−H bond in most
heterogeneous catalytic reaction systems. Finally, the reactive
•
OOH species combine with H protons provided by H
2
O
4
273.
dissociation to release H and O , achieving the efficient
2
2
6
7
X. Shen, W. Liu, X. Gao, Z. Lu, X. Wu and X. Gao, J. Am. Chem.
Soc., 2015, 137, 15882-15891.
dehydrogenation of HCHO solution and thereby enabling O
regeneration.
2
R. Li, X. Zhu, X. Yan, H. Kobayashi, S. Yoshida, W. Chen, L. Du, K.
Qian, B. Wu, S. Zou, L. Lu, W. Yi, Y. Zhou and J. Fan, ACS Catal.,
2
017, 7, 1478-1484.
Conclusions
8
9
1
J. Song, Z.-F. Huang, L. Pan, J.-J. Zou, X. Zhang and L. Wang, ACS
Catal., 2015, 5, 6594-6599.
Z. F. Huang, J. Song, L. Pan, X. Zhang, L. Wang and J. J. Zou, Adv.
Mater., 2015, 27, 5309-5327.
0 Y. Sun, Q. Liu, S. Gao, H. Cheng, F. Lei, Z. Sun, Y. Jiang, H. Su, S.
Wei and Y. Xie, Nat. Commun., 2013, 4, 2899.
In summary, we find that WO
vacancies is a robust, stable and noble-metal-free catalyst for
HCHO reforming into at room temperature. The
dehydrogenation kinetic is driven by a new PCET-dependent
2
NRs with abundant oxygen
H
2
molecular oxygen activation process, where the proton and 11 N. Zhang, X. Li, H. Ye, S. Chen, H. Ju, D. Liu, Y. Lin, W. Ye, C.
electron transfer are not concerted but rather sequential. In
Wang, Q. Xu, J. Zhu, L. Song, J. Jiang and Y. Xiong, J. Am. Chem.
Soc., 2016, 138, 8928-8935.
2 C. Shu, S. Kang, Y. Jin, X. Yue and P. K. Shen, J. Mate. Chem. A,
•
−
detail, O
electron transfer process, and a subsequent protonation to
form •OOH (rather than •OH or other ROS) over the WO NRs
surface. It enables the controllable formation of the desired
ROS that serves as a crucial active species for H evolution
reaction. Importantly, abundant oxygen vacancies on the
surface of WO provide coordination unsaturated sites to
enhance O chemisorption and activation, resulting in the
reaction rate boosting linearly by increasing the O partial
pressure. Meanwhile, since O
2
is preferentially reduced to O
2
through one-
1
1
2
017, 5, 9655-9660.
2
3 M. R. Waller, T. K. Townsend, J. Zhao, E. M. Sabio, R. L.
Chamousis, N. D. Browning and F. E. Osterloh, Chem. Mater.,
2
2
012, 24, 698-704.
1
1
4 S. Visitsatthawong, P. Chenprakhon, P. Chaiyen and P.
Surawatanawong, J. Am. Chem. Soc., 2015, 137, 9363-9374.
5 T. Honda, T. Kojima and S. Fukuzumi, J. Am. Chem. Soc., 2012,
2
2
2
1
34, 4196-4206.
2
continuously uptakes and 16 M. C. Payne, M. P. Teter, D. C. Allan, T. A. Arias and J. D.
releases protons by formation and consumption of •OOH
Joannopoulos, Rev. Mod. Phys., 1992, 64, 1045-1097.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Page 11 of 11
J Po lue r an sael od fo Mn aot et rai ad l js u Cs ht emm ai rs gt ri yn sA
ARTICLE
Journal Name
1
7 V. Milman, B. Winkler, J. A. White, C. J. Pickard, M. C. Payne, E. 47 B. Chazallon, A. Oancea, B. Capoen and C. FocsaV,iePwhAyrsti.cleCOhnelmine.
V. Akhmatskaya and R. H. Nobes, In. J. Quantum Chem., 2000, Chem. Phys., 2008, 10, 702-712.
7, 895-910.
8 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996,
7, 3865-3868.
9 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1997,
8, 1396-1396.
DOI: 10.1039/C9TA03051D
7
48 L. E. Heim, N. E. Schlörer, J.-H. Choi and M. H. G. Prechtl, Nat.
Commun., 2014, 5.
49 R. Li, X. Zhu, X. Yan, D. Shou, X. Zhou and W. Chen, RSC Adv.,
2016, 6, 100103-100107.
50 K. Makino, M. M. Mossoba and P. Riesz, J. Am. Chem. Soc., 1982,
104, 3537-3539.
1
1
7
7
2
2
2
0 D. Vanderbilt, Phys. Rev. B, 1990, 41, 7892-7895.
1 X. Hu, Q. Ji, J. P. Hill and K. Ariga, CrystEngComm, 2011, 13.
2 R. Wu, J. Zhang, Y. Shi, D. Liu and B. Zhang, J. Am. Chem. Soc.,
51 E. C. Ashby, F. Doctorovich, C. L. Liotta, H. M. Neumann, E. K.
Barefield, A. Konda, K. Zhang, J. Hurley and D. D. Siemer, J. Am.
Chem. Soc., 1993, 115, 1171-1173.
2
015, 137, 6983-6986.
2
2
2
2
2
2
2
3
3 Y. Li, H. Wang, L. Xie, Y. Liang, G. Hong and H. Dai, J. Am. Chem. 52 K. Gong, P. Yu, L. Su, S. Xiong and L. Mao, J. Phys. Chem. C, 2007,
Soc., 2011, 133, 7296-7299. 111, 1882-1887.
4 A. F. Lamic, C. H. Shin, G. Djéga-Mariadassou and C. Potvin, 53 K. Gong, F. Du, Z. Xia, M. Durstock and L. Dai, Science, 2009,
Appl. Catal. A: Gen., 2006, 302, 5-13. 323, 760.
5 A. Katrib, F. Hemming, P. Wehrer, L. Hilaire and G. Maire, J. 54 B. Chazallon, N. Lebrun, P. Dhamelincourt, C. Toubin and C.
Electron Spectrosc. Relat. Phenom., 1995, 76, 195-200. Focsa, J. Phys. Chem. B, 2005, 109, 432-439.
6 A. Katrib, V. Logie, M. Peter, P. Wehrer, L. Hilaire and G. Maire, 55 G. R. Möhlmann, J. Raman Spectros., 1987, 18, 199-203.
J. Chim. Phys., 2017, 94, 1923-1937.
7 F. Lei, Y. Sun, K. Liu, S. Gao, L. Liang, B. Pan and Y. Xie, J. Am.
Chem. Soc., 2014, 136, 6826-6829.
56 M. Pohl, A. Pieck, C. Hanewinkel and A. Otto, J. Raman
Spectros., 1996, 27, 805-809.
8 I. Nakamura, N. Negishi, S. Kutsuna, T. Ihara, S. Sugihara and K.
Takeuchi, J. Mol. Catal. A: Chem., 2000, 161, 205-212.
9 J. Yan, T. Wang, G. Wu, W. Dai, N. Guan, L. Li and J. Gong, Adv.
Mater., 2015, 27, 1580-1586.
0 B. Ingham, S. V. Chong and J. L. Tallon, J. Phys. Chem. B, 2005,
1
09, 4936-4940.
3
3
1 L. J. A. F. K. Kihlborg, 1964, 21, 443-&.
2 M. Dieterle, G. Weinberg and G. Mestl, Phys. Chem. Chem. Phys.,
2
002, 4, 812-821.
3
3
3
3
3 S.-H. Lee, H. M. Cheong, J.-G. Zhang, A. Mascarenhas, D. K.
Benson and S. K. Deb, Appl. Phys. Lett., 1999, 74, 242-244.
4 J. A. Horsley, I. E. Wachs, J. M. Brown, G. H. Via and F. D.
Hardcastle, J. Phys. Chem., 1987, 91, 4014-4020.
5 G. Mestl, N. F. D. Verbruggen, E. Bosch and H. Knözinger,
Langmuir, 1996, 12, 2961-2968.
6 J. Gan, X. Lu, J. Wu, S. Xie, T. Zhai, M. Yu, Z. Zhang, Y. Mao, S. C.
Wang, Y. Shen and Y. Tong, Sci. Rep., 2013, 3, 1021.
3
3
7 S. R. Bathe and P. S. Patil, Solid State Ionics, 2008, 179, 314-323.
8 H. Wang, D. Yong, S. Chen, S. Jiang, X. Zhang, W. Shao, Q. Zhang,
W. Yan, B. Pan and Y. Xie, J. Am. Chem. Soc., 2018, 140, 1760-
1
766.
3
4
4
4
9 Y. Xu, M. Zhou, X. Wang, C. Wang, L. Liang, F. Grote, M. Wu, Y.
Mi and Y. Lei, Angew. Chem. Int. Ed., 2015, 54, 8768-8771.
0 K. K. Banger, Y. Yamashita, K. Mori, R. L. Peterson, T. Leedham,
J. Rickard and H. Sirringhaus, Nat. Mater., 2011, 10, 45-50.
1 L. Li, X. Feng, Y. Nie, S. Chen, F. Shi, K. Xiong, W. Ding, X. Qi, J.
Hu, Z. Wei, L.-J. Wan and M. Xia, ACS Catal., 2015, 5, 4825-4832.
2 M. Vasilopoulou, A. M. Douvas, D. G. Georgiadou, L. C. Palilis, S.
Kennou, L. Sygellou, A. Soultati, I. Kostis, G. Papadimitropoulos,
D. Davazoglou and P. Argitis, J. Am. Chem. Soc., 2012, 134,
1
6178-16187.
4
3 H. Li, H. Shang, X. Cao, Z. Yang, Z. Ai and L. Zhang, Environ. Sci.
Technol., 2018, 52, 8659-8665.
4
4
4 S. Li, H. Hu and Y. Bi, J. Mater. Chem. A, 2016, 4, 796-800.
5 F. Fu, C. Wang, Q. Wang, A. M. Martinez-Villacorta, A. Escobar,
H. Chong, X. Wang, S. Moya, L. Salmon, E. Fouquet, J. Ruiz and
D. Astruc, J. Am. Chem. Soc., 2018, 140, 10034-10042.
4
6 N. Lebrun, P. Dhamelincourt, C. Focsa, B. Chazallon, J. L.
Destombes and D. Prevost, J. Raman Spectrosc., 2003, 34, 459-
4
64.
1
0 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins