Mild Deoxygenation of Sulfoxides over Plasmonic Molybdenum Oxide Hybrid with Dramatic Activity Enhancement under Visible Light

Article abstract of DOI:10.1021/jacs.8b04711

Harvesting solar light to boost commercially important organic synthesis still remains a challenge. Coupling of conventional noble metal catalysts with plasmonic oxide materials which exhibit intense plasmon absorption in the visible light region is a promising option for efficient solar energy utilization in catalysis. Herein, we for the first time demonstrate that plasmonic hydrogen molybdenum bronze coupled with Pt nanoparticles (Pt/H_xMoO_3-y) shows a high catalytic performance in the deoxygenation of sulfoxides with 1 atm of H₂ at room temperature, with dramatic activity enhancement under visible light irradiation relative to dark conditions. The plasmonic molybdenum oxide hybrids with strong plasmon resonance peaks pinning at around 556 nm are obtained via a facile H-spillover process. Pt/H_xMoO_3-y hybrid provides excellent selectivity for the deoxygenation of various sulfoxides as well as pyridine N-oxides, in which drastically improved catalytic efficiencies are obtained under the irradiation of visible light. Comprehensive analyses reveal that oxygen vacancies massively introduced via a H-spillover process are the main active sites, and the reversible redox property of Mo atoms and strong plasmonic absorption play key roles in this reaction. The catalytic system works under extremely mild conditions and can boost the reaction by the assistance of visible light, offering an ultimately greener protocol to produce sulfides from sulfoxides. Our findings may open up a new strategy for designing plasmon-based catalytic systems that can harness visible light efficiently.

Full text of DOI:10.1021/jacs.8b04711

Subscriber access provided by NORTH CAROLINA A&T UNIV
Article
Mild Deoxygenation of Sulfoxides over Plasmonic Molybdenum
Oxide Hybrid with Dramatic Activity Enhancement under Visible Light
Yasutaka Kuwahara, Yukihiro Yoshimura, Kohei Haematsu, and Hiromi Yamashita
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b04711 • Publication Date (Web): 17 Jun 2018
Downloaded from http://pubs.acs.org on June 18, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
Mild Deoxygenation of Sulfoxides over Plasmonic Molybdenum Ox-
ide Hybrid with Dramatic Activity Enhancement under Visible Light
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Yasutaka Kuwahara,†,‡ Yukihiro Yoshimura, Kohei Haematsu, and Hiromi Yamashita*
†
†
,†,‡
†
Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, 2-1 Yamada-
oka, Suita, Osaka 565-0871, Japan
‡
Unit of Elements Strategy Initiative for Catalysts & Batteries (ESICB), Kyoto University, Katsura, Kyoto 615-8520,
Japan
KEYWORDS: deoxygenation, sulfoxides, hydrogen molybdenum bronze, surface plasmon resonances, heterogeneous
catalyst
ABSTRACT: Harvesting solar light to boost commercially important organic synthesis still remains a challenge. Coupling
of conventional noble metal catalysts with plasmonic oxide materials which exhibit intense plasmon absorption in the
visible light region is a promising option for efficient solar energy utilization in catalysis. Herein we for the first time demon-
strate that plasmonic hydrogen molybdenum bronze coupled with Pt nanoparticles (Pt/H MoO3-y) shows a high catalytic
x
performance in the deoxygenation of sulfoxides with 1 atm H2 at room temperature, with dramatic activity enhancement
under visible light irradiation relative to dark condition. The plasmonic molybdenum oxide hybrids with strong plasmon
resonance peaks pinning at around 556 nm are obtained via a facile H-spillover process. Pt/H MoO3-y hybrid provides ex-
x
cellent selectivity for the deoxygenation of various sulfoxides as well as pyridine N-oxides, in which drastically improved
catalytic efficiencies are obtained under the irradiation of visible light. Comprehensive analyses reveal that oxygen vacancies
massively introduced via a H-spillover process are the main active sites, and reversible redox property of Mo atoms and
strong plasmonic absorption play key roles in this reaction. The catalytic system works under extremely mild conditions
and can boost the reaction by the assist of visible light, offering an ultimately greener protocol to produce sulfides from
sulfoxides. Our findings may open up a new strategy for designing plasmon-based catalytic systems that can harness visible
light efficiently.
1
. INTRODUCTION
Harvesting and converting abundant sunlight to drive
Recently, heavily doped semiconductors such as trans-
14-15
parent conducting oxide (TCO) and transition metal ox-
1
6-17
ides/chalcogenides
have emerged as a new family of
chemical reactions is of immense interests in terms of solar
energy conversion and environmental consideration, espe-
cially for heterogeneous catalysis that has been industrially
implemented in a large-scale. Heterogeneous photo-catal-
ysis based on surface plasmonic resonance (SPR) opens up
a promising alternative technology in this field since the
plasmon resonance frequency predominantly appear in
plasmonic materials alternative to conventional noble
metals. Plasmonic features of these materials can be con-
trolled by tuning the concentration of free carriers (elec-
trons and holes) in the crystal structure, which can be ac-
complished by heavily doping aliovalent heteroatoms or
1
4, 18-20
lattice vacancies.
Hydrogen doping is a rational pro-
1-5
tocol to achieve plasmon resonances in semiconductors,
since it induces intrinsic defects and concurrently intro-
duces massive electrons to the given host structures with-
out extensive structural expansion, thus yielding sufficient
delocalized electrons to support the plasmon resonance in
visible- to near-infrared (NIR) light region. Recent stud-
ies have witnessed that the SPR of noble-metal-based plas-
monic materials (e.g. Au and Ag) enables to boost catalytic
performance in various important organic transformations,
including oxidation of aromatic alcohols, reduction of ke-
tones, epoxides, and azo compounds on Au nanoparticles
6
2
1-22
the spectrum of visible light region.
Among this mate-
MoO ),
7
4, 8-9
rial family, H-doped molybdenum oxides (H
x
3
(NPs), reduction of nitroaromatics on Ag or Au NPs,
1
0
namely hydrogen molybdenum bronze, can be regarded as
a promising plasmonic platforms for the design of light-
harvesting catalytic system because of i) the appearance of
intense plasmon absorption in the visible light to near-in-
frared (NIR) region due to high dopant concentration (>
and epoxidation of ethylene on Ag NPs. Bimetallic alloy
NPs composed of plasmonic NPs and other catalytically ac-
tive noble metals (e.g. Pd) have been found to be efficient
1
1-13
plasmon-catalysts to drive Suzuki coupling reactions
1
2
and oxidation of benzylamine by the assist of visible light,
mainly due to the plasmon-induced generation of ener-
getic electrons at the surface of active metals.
21
-3
1
0 cm ) and ii) the intercalated H atoms and concurrently
injected free electrons in the MoO matrices that are re-
versibly exchangeable during redox events.
3
23-24
Owing to
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
the high energy of the H
conductor SPRs, it has currently found diverse applications
in the areas of bioimaging, sensing, optical devices, and ca-
x
MoO
3
SPR relative to other semi-
Characterization of plasmonic molybdenum oxide
hybrid. The synthesis of plasmonic molybdenum oxide
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
coupled with Pt NPs (Pt/H
spillover process (Figure 1). α-MoO
x
MoO3-y) was performed via a H-
25-28
22
talysis.
plasmonic H
tion of ammonia borane (NH
ceeds even at ambient conditions.
denum oxide hybrids consisting of H
metal catalysts (e.g. Pd) can provide more extended cata-
Previous studies on the catalytic applications of
MoO have been limited to the dehydrogena-
BH ), which inherently pro-
3
containing Pt ions
x
3
was reduced in a flow of H at controlled temperature (R.T.,
2
3
3
100 °C, and 200 °C) to yield a dark blue powder. Here the
23, 29
+
Plasmonic molyb-
MoO and active
dopants (H ) and free electrons are injected into MoO
3
ma-
x
3
trices through in-situ formed Pt NPs, causing a rapid dis-
+
-
sociation of molecular H over Pt surface (H → 2H + 2e ),
2
2
22
+
lytic opportunities, such as p-nitrophenol reduction and
in which the H dopant levels and oxygen defect densities
can easily be modulated by the reduction temperature. The
H-spillover process was verified by hydrogen temperature-
3
0
Suzuki coupling reactions, in which reaction efficiencies
are significantly improved under the irradiation of visible
light; however, application to more practically important
organic transformations has not been accomplished yet.
The intercalated dopants and the defect sites (H and oxy-
gen defects) massively introduced in the MoO
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
programmed reduction results (H -TPR, Figure S1). MoO
2
3
could be reduced at below 100 °C when coupled with Pt
+
NPs, while pristine MoO was hardly reduced by molecular
3
3
framework
H up to 550 °C. The molar ratio of H atoms consumed per
2
may offer unique reactivities towards surrounding reac-
tants; for example, Mu et al. have claimed that the Mo−OH
mole of Pt atoms contained in the catalyst (H/Pt) was de-
termined to be 885, whereas Pt/SiO as a comparative sam-
2
x
group on the H MoO
3
surface play a key role in the selec-
ple gave H/Pt molar ratio of 4.2, demonstrating an interca-
lation of massive amount of hydrogen atoms into MoO3
matrices via a H-spillover process.
tive C−O bond cleavage of biomass-derived tetrahydro-
3
1
furfuryl alcohol, and other reports have demonstrated
that oxygen defects in MoO3-x suboxide are responsible for
3
2
the active sites in the hydrodeoxygenation of ketones,
while these reactions have been achieved under harsh re-
action conditions (~400 °C). Combination of plasmonic
property and unique catalytic properties of hydrogen mo-
lybdenum bronze is expected to offer unprecedented cata-
lytic opportunities under light irradiation conditions,
which may open up a new avenue for the design of plas-
mon-based heterogeneous catalysts that can harness visi-
ble light efficiently.
Herein we for the first time demonstrate that hydrogen-
doped MoO
performance in the deoxygenation of sulfoxides with 1 atm
at room temperature, with dramatic activity enhance-
3
coupled with Pt NPs shows a high catalytic
H
2
ment under visible light irradiation relative to dark condi-
tion. Deoxygenation of sulfoxides to sulfides is a funda-
mental and important reaction for organic synthesis and
Figure 1. Schematic illustration of the formation of hydrogen
33-34
biochemistry.
The deoxygenation of sulfoxides typically
molybdenum bronzes (H
Overview of the crystal structures of (left) orthorhombic
MoO and (right) monoclinic H MoO3-y (x ~ 0.9), showing OH
and OH groups predominantly located at terminal oxygen at-
oms (O ) and oxygen vacancies (O ) predominantly formed at
-bridging oxygen sites (μ -O).
x
MoO3-y) via a H-spillover process.
requires stoichiometric amounts of reducing agents such
3
5
36
37
38
as phosphines, boranes, silanes, and alchohols, which,
however, suffer from the use of toxic reagents and the pro-
duction of undesirable by-products as well as low atom-ef-
ficiency. Alternatively, the catalytic deoxygenation of sul-
3
x
2
t
v
μ
2
2
foxides using molecular H
2
as the most greener reducing
The intercalation of hydrogen atoms into MoO
is known to yield a class of H MoO (0 < x ≤ 2) along with
The peaks at 12.7, 25.7, and
8.92° for the untreated Pt/MoO correspond to the (020),
040), and (060) crystal planes of orthorhombic α-MoO
PDF#5-0508). XRD patterns of Pt/H MoO3-y reduced un-
condition are predominantly indexed to monoclinic
phase (PDF#53-1024), which contains H atoms at
3
matrices
3
3
39
agent has been examined using MoO
2
Cl
-Hol, but they require harsh
reaction conditions including high H pressure (> 7 atm).
Recently, Mitsudome et al. demonstrated catalytic deoxy-
genation of sulfoxides under atmospheric H pressure us-
at 100 °C; however, there has been no report
on the deoxygenation of sulfoxides using 1 atm H at room
2
,
Pd/C, Pt-
x
3
4
0
41
MoO
x
/TiO
2
,
and Pt/V(1−x)Cr
x
43-46
partial reduction of Mo ions.
2
3
3
(
(
3
2
x
4
2
ing Ru/TiO
2
der H
2
2
H0.9MoO
3
temperature. The present catalytic system works under ex-
tremely mild conditions and can boost the deoxygenation
of sulfoxide by the assist of visible light, offering an ulti-
mately greener protocol to produce sulfides from sulfox-
ides.
the interlayer positions of double layers of MoO octahe-
6
45-46
dra,
crystallographically evidencing the intercalation of
matrices upon H reduction (Figure
H atoms into MoO
3
2
2a). Shift of the position of (020) basal plane toward
smaller angle indicates the expansion of interlayer spac-
+
ings due to the intercalation of H ions at the interstitial
2. RESULTS AND DISCUSSION
sites. With the rise of reduction temperature, the relative
ACS Paragon Plus Environment

Page 3 of 10
proportion of MoO
Journal of the American Chemical Society
3
decreases, indicating a gradual trans-
phase at elevated tem-
fine structure (XAFS) measurement for Mo K-edge in a
flow of H , in which a gradual shift of the absorption edge
toward a lower energy region and a shrinkage of Mo–O
bond associated with the phase transition during H reduc-
Figure 2d
shows O 1s XPS spectra of the samples. Pristine MoO ex-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
formation of α-MoO
3
to H0.9MoO
3
2
peratures. It is also noteworthy that peak intensities of
decreases with the rise of reduction temperature.
This may be interpreted by the perturbation of H0.9MoO
crystal structure associated with the introduction of oxy-
H0.9MoO
3
2
22-23
3
tion were directly observed (Figure S4).
3
22
gen vacancies and distortion of MoO
reduction, Pt/H MoO3-y samples exhibit strong absorption
in the visible to NIR region with an intense SPR peak pin-
ning at around 556 nm, while the unreduced Pt/MoO only
shows an absorption below 400 nm owing to the semicon-
ductive nature (Figure 2b). As H reduction temperature
increases, a blue shift of the plasmonic wavelength from
64 to 556 nm is observed, together with an intensification
of the absorption peaks, which is due to the increased con-
6
octahedra. After H
2
hibits the peaks ascribed to lattice O atoms (Olatt) at 530.8
eV and small portion of –OH groups (O-OH) at 532.4 eV. The
peak for the O-OH species drastically increases in intensity
x
3
after H
Mo−OH bond. Furthermore, a small peak appeared at
533.7 eV can be assigned to −OH groups coordinated to
2
reduction at 200 °C, evidencing the emergence of
4
8
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
Mo atoms (OH2O). The proportion of (O-OH + OH2O)/Ototal
5
increased in the order of pristine MoO
3
(30 at%) <
Pt/H
Pt/H
x
MoO3-y(RT) (55 at%) < Pt/H
x
MoO3-y(100) (58 at%) <
1
8, 22
centration of free carriers in the MoO
3
host structure.
x
MoO3-y(200) (59 at%), again evidencing a higher de-
+
However, further rise of reduction temperature to 300 °C
leads to the quenching of plasmonic absorption due to the
structural deformation into amorphous MoO3-x suboxide
gree of H -intercalation at elevated reduction tempera-
tures (for detail, see Figure S3, Table S1). In addition, Pt
0
species are found to dominantly exist as metallic Pt NPs
(
Figure S2).
(average diameter = ca. 2.2 nm) after H
ure S5). These results evidence that, during H
2
reduction (see Fig-
reduction,
matrices through in-
2
(a)
(b)
Pt/H MoO3-y(200)
x
556 nm
Pt/H MoO3-y(200)
Pt/H MoO3-y(100)
the H atoms are injected into MoO
situ formed Pt NPs as windows, and are then inserted at
x
x
3
Pt/H MoO3-y(RT)
Pt/H MoO3-y(100)
x
x
0
Pt/H MoO3-y(RT)
x
interlayer positions of layered MoO framework by coordi-
3
5
+
Pt/MoO3
(unreduced)
nating with terminal oxygen atoms (O ) through Mo −OH
t
46, 49-50
bond (see Figure 1).
The formation of such –OH sites
H0.9MoO3 simulated
Pt/MoO3
was verified by temperature-programmed desorption
measurement using NH as a probe molecule (NH -TPD,
Figure S6). Reduced Pt/H MoO3-y hybrids showed desorp-
MoO3 simulated
(unreduced)
3
3
1
0
15
20
25
30
35
40
400
600
800
1000
1200
x
2
 / degree
Wavelength / nm
tion peaks of NH
reduced Pt/MoO
3
in the range of 200-600 °C, whereas un-
showed no peak in this range, proving
(
c)
(d)
Pt/H MoO3-y(200)
x
Pt/H MoO3-y(200)
x
3
the emergence of acid sites associated with –OH groups in
the MoO framework upon the H2 reduction treatment.
Such a structural alteration upon H -intercalation results
in a decreased intensity of Raman peaks of MoO (Figure
S7), suggesting the concomitant formation of oxygen va-
5
+
6+
Mo
O
H2
O
6
+
Mo
5+
Mo
O-OH
Olatt
Mo
3
Exp.
Exp.
+
6
+
Mo
Mo
Mo
O
latt
MoO3
5+
O-OH
3
4
+
Olatt
OH2
O
6
+
6
+
Mo
MoO3
Mo
fitted
fitted
O-OH
cancies (O
v
). In order to quantify the amounts of interca-
reduction,
2
40 238 236 234 232 230 228 226 538 536 534 532 530 528 526
Binding energy / eV
Binding energy / eV
+
lated H and oxygen vacancies formed upon H
2
thermogravimetric (TG) analysis was carried out in either
N
2
or air gas environment. Based on the weight changes
Figure 2. (a) Powder XRD patterns, (b) UV-vis-NIR diffuse re-
flectance spectra for the Pt/H MoO3-y hybrids reduced at var-
ied temperatures and for the unreduced Pt/MoO . (c) Mo 3d
XPS spectra and (d) O 1s XPS spectra for the Pt/H MoO3-y(200)
and for the pristine MoO
+
associated with the release of intercalated H (in N
2
envi-
x
ronment) and the filling of O
stoichiometric compositions of the series of the samples
were determined to be
v
(in air environment), the
3
x
Pt/H0.19MoO2.98(RT),
3
.
Pt/H0.50MoO2.97(100), and Pt/H0.82MoO2.49(200), respec-
X-ray photoelectron spectroscopy (XPS) is used to con-
firm the oxidation state of the samples. For the series of
tively (Table 1) (for detailed TG analysis, see Figure S8),
43,
according to the methods previously reported elsewhere.
Pt/H
x
MoO3-y samples, Mo species with lower oxidation
47, 51
This result ascertains the introduction of increased
5+
states appear in Mo 3d XPS spectra (Mo ; BE at 230.8 and
+
amounts of H and O
v
sites at elevated reduction tempera-
4+
6+
2
34.0 eV, Mo ; BE at 229.5 and 233.0 eV), while Mo spe-
cies (BE at 235.8 and 232.1 eV) are dominantly observed for
pristine MoO (Figure 2c), indicative of partial reduction
of Mo species upon H
tures, which are well coincided with the results shown
above. It is also worth noting that there was no significant
structural variation upon the H2 reduction treatment in
terms of surface area and pore volume, despite such con-
siderable changes in crystal structure and chemical com-
position (Table 1).
3
4
7
2
reduction. Based on fitting analy-
sis, the average oxidation number (Noxi) is found to gradu-
ally reduce from 5.89 to 5.42 along with the rise of reduc-
tion temperature from R.T. to 200 °C (for detail, see Figure
S3, Table S1). Such a change in the oxidation state of Mo
species was also corroborated by in-situ X-ray absorption
ACS Paragon Plus Environment

Journal of the American Chemical Society
Table 1. Textural and compositional parameters of Pt/H
Page 4 of 10
x
MoO3-y samples
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
N
2
physisorption
TG analysis
Concentration
of O (y)
Reduction Temp.
°C)
Samples
Pt/MoO
a
b
S
BET
2
V
total
3
Concentration of
intercalated H (x)
(
+
(m /g)
(cm /g)
v
3
None
7.5
0.052
0.0
0.0
(unreduced)
Pt/H
Pt/H
Pt/H
a
x
MoO3-y(RT)
MoO3-y(100)
MoO3-y(200)
RT
100
200
7.6
7.1
0.057
0.046
0.047
0.19
0.50
0.82
0.02
0.03
0.51
x
x
7.0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Determined by the BET method by using adsorption data ranging from p/p
0
= 0.05 to 0.30. b Reported at p/p
0
= 0.99.
Deoxygenation of sulfoxide. The catalytic perfor-
mance of the samples was initially assessed in the deoxy-
genation of diphenyl sulfoxide (1) as a model substrate at
catalyst
H
2
(1 atm), toluene
1
R.T.
2
room temperature under a flow of 1 atm H in the dark con-
2
dition (Scheme 1). Toluene was chosen as a suitable sol-
vent to achieve a high catalytic efficiency. A better catalytic
efficiency was obtained with non-polar solvents and the
use of polar solvent resulted in a poor catalytic efficiency
(for the effect of solvent, see Figure S9). Among Pt NP cat-
alyst supported on metal oxides (metal loading was fixed
1
00
80
100
(a)
(b)
80
Pt/H MoO
Pt/H MoO
x
3-y
x
3-y
Pt/WO
3
6
4
0
0
60
40
20
0
Pt/V O5
2
Pt/MoO2
Pt/SiO2
Pd/H MoO
x 3-y
Pt/Al O3
2
to be 0.5 wt%), Pt/H
x
MoO3-y exhibited the highest catalytic
20
0
Ru/H
x
MoO3-y
6
activity, affording diphenyl sulfide (2) in >99% yield in 7 h
of reaction (Figure 3a), corresponding to a turnover num-
ber (TON) of 160 with respect to the total number of Pt
atoms in the catalyst. Considering the reaction conditions
implemented in this study, this value is superior to those
obtained with previously reported catalysts requiring high
0
2
4
6
8
0
2
4
8
Reaction time / h
Reaction time / h
(
c) 100
(d)200
with H2
8
0
150
2
00 ℃
1
00 ℃
6
0
100
1
5
12 mol
temperature and high H
2
pressure (Pd/C: TON 50 (at 90 °C,
without H2
40
RT
3
9
41
10
5
79 atm); Pt/V(1−x)Cr -Hol: TON 230 (at 100 °C, 1 atm);
x
2
0
0
4
2
Ru/TiO
Pd) catalyst was totally inactive in this reaction under the
conditions implemented in this study. Ru/TiO catalyst re-
ported by Mitsudome et al. exhibited an activity under
the identical reaction conditions, giving 97% conversion
within 2 h of reaction; however, TON with respect to the
total number of Ru atoms was calculated to be 6.5 (TOF =
2
: TON 500 (at 100 °C, 1 atm) ). In fact, Pd/C (5%
unreduced
6
3
9
0
0
2
4
8
0
5
10
15
20
Reaction time / h
Reaction time / h
2
4
2
Figure 3. Time course in the deoxygenation of 1 over (a) vari-
ous Pt-loaded oxide catalysts (reduction Temp. = 200 °C), (b)
M/H
200 °C), and (c) Pt/H
perature. (d) Time course of the yield of 2 over Pt/H
(200) catalyst with or without H . Reaction conditions: cata-
lyst (0.05 g), 1 (0.2 mmol), toluene (10 mL), room temperature,
bubbling (10 mL/min), in the dark.
x
MoO3-y catalysts (M = Pt, Pd, Ru) (reduction Temp. =
MoO3-y catalysts reduced at varied tem-
MoO3-
x
-
1
3
.3 h ), while that afforded by Pt/H
x
MoO3-y(200) catalyst
was 160 (TOF = 23 h ), clearly demonstrating that the ac-
tivity of Pt/H MoO3-y(200) catalyst is superior to those ob-
tained with previously reported catalysts. Pt/SiO and
Pt/Al were inactive for this reaction, indicating that Pt
NPs are not the main active sites in this reaction. The ac-
tivity of Pt supported on H MoO3-y was higher than those
of Pd and Ru supported on H MoO3-y (Figure 3b). The or-
der of efficiency of the metals can be associated with the
ability of the metals to dissociate H , i.e., Pt and Pd NPs
dissociates H with lower activation energy than Ru NPs,
x
-
1
y
2
x
2
H
2
2
O
3
Pt supported on easily reducible metal oxides such as
Pt/WO and Pt/V exhibited moderate activities (Figure
3a), suggesting the significance of defect sites on oxide sup-
ports. It was found that the catalytic activity of the
Pt/H MoO3-y increases as the reduction temperature in-
3
2
O
5
x
x
x
2
creases (Figure 3c). Considering the correlation between
2
the observed catalytic activities and the stoichiometry of
thereby reducing the total activation energy required for
this reaction. Plasmonic molybdenum oxide nanosheet
prepared by solvothermal method exhibited a slight ac-
tivity, giving 2.3% conversion in 8 h, which is due to the
absence of Pt NPs acting as H2 dissociation sites.
+
H or O
v
determined by TG analysis, oxygen defect sites
23
would be the main active sites for this reaction (for the cor-
relations between the stoichiometry of H or O
lytic activities, see Figure S10). To verify the above hypoth-
esis, the deoxygenation reaction was carried out in the ab-
sence of H
after 24 h of reaction per 50 mg of catalyst even in the ab-
sence of H (Figure 3d). Pt/H MoO3-y(200) treated with 1
in toluene under an Ar atmosphere at room
+
v
and cata-
Scheme 1 Deoxygenation of diphenyl sulfoxide to di-
phenyl sulfide with 1 atm H2 at room temperature
2
gas. It was found that 12 μmol of 2 was yielded
2
x
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
-
1
(
a) Pt/H MoO3-y(200)
Exp.
(b)
(c)
819 cm
(Mo -O)
1
2
3
4
5
6
7
8
9
x
6+
Mo
2
treated with DPSO
5
+
+
Pt/H MoO3-y(200)
Mo
Mo
x
O
O
Mo6+
O
O
O
O
4
Pt/H MoO3-y(200)
treated with DPSO
x
9
94 cm
-1
Pt
O
Mo6+
O
fitted
treated with DPSO
O
O
-1
(Mo=O)
O
2
88 cm
-1
6
66 cm
g⊥=1.929
g⊥=1.925
(Mo=O)
(Mo -O)
3
Ph
Ph
Ph
S
Re-oxidation
Ph
O
S
Pt/H MoO3-y(200)
x
O
Pt/H MoO3-y(200)
O
O
O
x
O
Mo5+
O
Pt/H MoO3-y(200)
Mo5+
x
Pt
O
O
O
O
Pt/MoO3
(unreduced)
2
H O
g||=1.868
H
2
reduction
Pt/MoO3
H
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pt/MoO3
(unreduced)
O
O
Mo6+O
O
O
O
O
(
unreduced)
Pt2+
×1/40
O
Mo6+
O
O
O
2
40 238 236 234 232 230 228 226
3000 3200 3400 3600 3800 4000
Magnetic field / Gauss
200 400 600 800 1000 1200
-1
Binding energy / eV
Raman shift / cm
Figure 4. (a) Mo 3d XPS spectra, (b) electron spin resonance (ESR) spectra, and (c) Raman spectra of as-synthesized Pt/MoO
Pt/H MoO3-y(200) and the retrieved Pt/H MoO3-y(200) after the treatment with diphenyl sulfoxide (DPSO) in toluene under an
3
,
x
x
Ar gas atmosphere at room temperature for 24 h.
temperature for 24 h was further analyzed by XPS, electron
spin resonance (ESR), and Raman spectroscopy analyses.
As aforementioned, in Mo 3d XPS spectrum, Pt/H
(200) show peaks attributable to Mo species with lower
oxidation states, while as-synthesized Pt/MoO
nantly exhibits peaks attributable to Mo species. After
treatment of Pt/H MoO3-y(200) with 1 in an Ar environ-
ment for 24 h, peaks attributable to Mo species were
Based on these experimental results, the reaction path-
way for the deoxygenation of sulfoxide over Pt/H
hybrid is proposed as illustrated in Scheme 2; a heavily-
x
MoO3-y
x
MoO3-
y
doped molybdenum oxide maintains massive oxygen va-
3
domi-
cancies, which are generated in MoO
spillover process (step 1 and 2). The thus formed oxygen
vacancies, mainly involving μ -bridging oxygen sites, with
3
matrices via a H-
6+
x
2
6+
positive charges effectively trap the negatively-charged O
atom of sulfoxide (step 3) and act as the main active sites
to directly deoxygenate sulfoxides to sulfides, which results
in a filling of oxygen vacancies and a concomitant oxida-
tion of adjacent Mo species (step 4). Afterwards, the oxi-
dized species readily react with the active H atoms, which
are dissociated on Pt NPs, to regenerate the original active
species. Thus, massive oxygen vacancies and the reversible
again dominantly emerged (Figure 4a). In ESR spectrum
of Pt/H
x
MoO3-y(200), g values typical of five-coordinated
5
+
52-53
Mo species (g = 1.925 and g = 1.868) are observed,
⊥
‖
while such a peak is not observable for the as-synthesized
Pt/MoO having only Mo species. The ESR signal at-
tributable to five-coordinated Mo species was obviously
diminished after the treatment with 1 in Ar atmosphere,
evidencing the re-oxidation of the reduced Mo species to
6+
3
5
+
redox property of Mo atoms in H MoO3-y can provide an
effective catalytic system for this reaction under mild con-
ditions.
x
6
+
Mo (Figure 4b). In Raman spectrum of Pt/H
x
MoO3-
y
(200), the Raman peak assignable to doubly coordinated
2
8, 50, 54
-1
bridging oxygen (Mo
2
–O)
seen at 819 cm increased
Scheme 2. Possible reaction pathway of the deoxygen-
ation of sulfoxide over Pt/H MoO3-y hybrid catalyst us-
in intensity after the treatment with 1 in Ar atmosphere,
relative to other Raman peaks (Figure 4c), suggesting a
filling of oxygen vacancies at the µ -O position with O at-
x
ing molecular H
2
as a reductant
2
O
O
oms derived from 1. Upon this treatment, a red shift of
plasmonic wavelength was observed in UV-vis-NIR spec-
trum, while the crystal structure remained unchanged (see
Figure S11). These results indicate re-oxidation of the re-
duced Mo species and a decreased concentration of dopant
levels by the reaction with 1 due to a filling of oxygen va-
cancies with O atoms derived from 1. Furthermore, to track
the progress of the deoxygenation reaction, in-situ FTIR
measurement using dimethyl sulfoxide (DMSO) as a probe
molecule was performed at room temperature in a contin-
O
O
Mo6+
O
O
Pt
O
Mo6+
O
O
O
R2
R1
O
H
2
S
4
Deoxygenation
1
H
2
dissociation
R2
R1
S
δ+
H
H
O
O
O
δ-
O
Mo5+
O
O
O
O
Mo5+
O
Mo6+
O
O
Pt
Pt
O
Mo6+
O
O
O
O
O
O
O
O
O
oxygen vacancy
uous flow of H
to strongly chemisorbed DMSO molecule was observed on
Pt/H MoO3-y(200), while it was hardly observed on Pt/SiO
proving the existence of strong adsorption sites on
MoO3-y surface. This IR band was gradually decreased in
intensity after the exposure to H gas flow, clearly demon-
strating the deoxygenation of sulfoxide occurring on
MoO3-y surface (not on Pt NPs surface) using H mole-
cule as a reductant.
2
(Figure S12). An intense IR band assignable
R2
R1
S
O
H
O
2
O
O
O
O
Mo5+
O
x
2
,
Pt
Mo5+
O
3
Adsorption
of sulfoxide
O
O
2
H-spillover
Formation of
oxygen vacancy)
O
O
(
H
x
2
H
x
2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
Plasmon-driven activity enhancement. Intriguingly, of the reaction suspension was found to increase by only
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
under the irradiation of visible light (λ > 420 nm, I = 19.6
mW/cm ), a significant improvement of catalytic efficiency
ca. 0.5 °C under visible light irradiation, thereby the con-
tribution of photothermal effect can be disregarded in the
present catalytic system. The negligible temperature in-
crease during the light irradiation may be explained by the
high non-polarity of toluene used as a medium, which is
far less susceptible to SPR effect relative to other polar me-
dia. In the plasmonic hybrid system, hot electrons pro-
duced by plasmonic SPR are, in general, believed to be
transferred to the contiguous metal NPs to create nega-
tively charged NPs, which accelerates specific catalytic re-
2
was observed over Pt/H MoO3-y(200) catalyst, giving a two-
x
fold faster reaction rate relative to dark condition (Figure
5a). Such an activity enhancement was not observed on
non-plasmonic catalysts such as Pt/SiO
2
, Pt/WO
3
, and
Pt/MoO upon the exposure to visible light (Figure 5b).
2
The catalytic activity increased in proportion to the inten-
sity of the incident light (Figure 5c). The catalytic reac-
tions were also performed with LED lamp irradiation, in-
cluding blue (λmax = 470 nm), green (λmax = 530 nm), and
red LED (λmax = 650 nm). The degree of activity increase
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3, 22, 30
actions (interfacial charge transfer)
Pt/H MoO3-y catalytic system, Pt NPs anchored on plas-
monic molybdenum oxide act as dissociation sites for H
however, kinetic analysis revealed that the reaction rate is
independent from partial pressure of H gas both in the
. In the case of
x
(
normalized by each light intensity) was nearly consistent
2
;
with the absorption spectrum of Pt/H MoO3-y(200) meas-
x
ured by UV-vis-NIR spectroscopy (Figure 5d), demon-
strating that the activity enhancement is driven by plas-
monic effect of hydrogen molybdenum bronze. To the best
of our knowledge, this is the first report on plasmon-in-
duced activity enhancement in the deoxygenation reaction.
2
dark and under visible light irradiation conditions (reac-
tion order was close to zero, see Figure S13), thereby dis-
counting the positive involvement of charge transfer in the
present catalytic system. Therefore, we speculated that the
enhanced catalytic activity is attributed to the electric
near-field enhancement (ENFE) effect of hydrogen molyb-
denum bronze. To verify this hypothesis, in-situ FTIR
measurement using DMSO was performed at room tem-
perature under visible light irradiation (Figure S14). A
gradual shift of the IR band assignable to S=O bond of
chemisorbed DMSO toward lower wavenumbers was un-
ambiguously observed upon the irradiation of visible light.
This result suggests that the electric near-field induced
around the surface of hydrogen molybdenum bronze ow-
ing to the SPR effect upon visible light irradiation facili-
tates the activation of sulfoxide by stretching the S=O bond.
The stretched sulfoxide displays a larger molecular polarity
and can be deoxygenated more efficiently by active centers,
a)100
(b) 60
(
Visible
light
50
Visible light
Under dark
8
0
Under dark
4
3
2
1
0
0
0
0
0
60
4
2
0
0
0
No reaction
0
2
4
6
8
Pt/SiO2 Pt/WO3 Pt/MoO2 Pt/H MoO
x
3-y
Reaction time / h
Catalyst
(
c) 60
(d) 2.0
Green LED
5
0
1
1
0
.5
.0
.5
40
3
2
1
0
0
0
Red LED
Blue LED
5
5
thus accelerating the reaction.
0.0
0
5
10
15
20
400
600
800
1000
Table 2. Scope of substrates in the deoxygenation of
-2
Incident light / mW･cm
Wavelength / nm
sulfoxides and pyridine N-oxides with 1 atm H2 at
a
room temperature over Pt/H
x
MoO3-y(200) catalyst
Figure 5. (a) Time course in the deoxygenation of 1 over
Pt/H
x
MoO3-y(200) catalyst in the dark or under visible light (λ
Time Conv.b Sel.b
2
>
420 nm, I = 19.6 mW/cm ) irradiation conditions. (b) Com-
parison of catalytic activities over various catalysts in the dark
or under visible light irradiation conditions (reduction Temp.
200 °C). (c) Catalytic activity dependence on the intensity of
visible light over Pt/H MoO3-y(200) catalyst. (d) The increased
yield of 2 as a function of the wavelength by LED irradiation
over Pt/H MoO3-y(200) catalyst (left axis) and its Kubelka-
Munk function (right axis). Reaction conditions are the same
as those described in Figure 3.
Entry
1
Substrate
Product
(h) (%)
(%)
>99
>99
3
=
(67) (>99)
x
2
3
>
99
>99
2
4
(
54) (>99)
x
>99 >99
(
60) (>99)
The origin of the visible light-induced activity enhance-
ment in plasmonic semiconductor materials may involve i)
photothermal effect, ii) interfacial charge transfer mecha-
4
>99
(66) (>99)
>99
3
3
2
6
5
nism or iii) electric near-field enhancement (ENFE) effect.
5
6
7
>99
66) (>99)
>
(
99
To eliminate the contribution of the photothermal effect,
a control experiment at 40 °C was performed in the dark.
The yield of 2 after 1 h of reaction in the dark at 40 °C was
38%, which was higher than the yield obtained in the dark
at near room temperature (30% at 30 °C) but was appar-
ently lower than that attained under light irradiation (50%
at room temperature) (Figure 5c). In fact, the temperature
>99
(71)
>99
(>99)
>99
>99
(
72) (>99)
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
8
>99
>99
platform in heterogeneous catalysis, which can boost di-
verse selective organic transformations under solar light.
2
24
2
1
(
44) (>99)
2
9
3
>99
>99
4. EXPERIMENTAL SECTION
Materials. Hexaammonium heptamolybdate tetrahydrate
(
69) (>99)
>99 >99
47) (>99)
4
1
0c
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
((NH
drate (RuCl
chased from Wako Pure Chemical Industries, Ltd. Potassium
tetrachloroplatinate(II) (K PtCl ), disodium tetrachloropalla-
)
6
Mo
7 2
O24‧4H O; 99.0%), ruthenium (III) chloride hy-
4
(
3
‧xH
2
O), and tungsten(VI) oxide (WO ) were pur-
3
11c
>99
>99
3
(
85) (>99)
2
4
2d
2 4 2 5
date (II) (Na PdCl ), vanadium(V) oxide (V O ), urea (99.0%),
and other commercially available chemical reagents for cata-
lytic tests were purchased from Nacalai Tesque, Inc. Fumed
1
>99
>99
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(78) (>99)
>99 >99
75) (>99)
1
3d
silica (SiO
2 2
) and molybdenum(IV) oxide (MoO ) as catalyst
3
(
supports were purchased from Sigma-Aldrich Co. γ-Al
2 3
O
aReaction conditions: catalyst (Pt/H
foxide (0.2 mmol), toluene (10 mL), room temperature, H2
bubbling (10 mL/min), visible light (λ > 420 nm) emitted from
(JRC-ALO-4) as a catalyst support was supplied from the Ca-
talysis Society of Japan. All chemicals were used as received
without further purification.
x
MoO3-y(200), 0.05 g), sul-
b
Synthesis of Pt/H
gen molybdenum bronzes coupled with Pt NPs (Pt/H
) were prepared by a two-step synthetic process. Firstly, α-
x
MoO3-y hybrid. The plasmonic hydro-
MoO3-
5
00 W Hg-Xe short arc lamp using a cut-off filter. Values in
c
x
parentheses are those obtained in the dark condition. pyri-
22
y
dine N-oxide (1.0 mmol), acetonitrile was used as a solvent.
d
MoO
3
was prepared by thermal decomposition of
Mo O at 500 °C for 4 h in air with the ramping
was obtained by a precipitation
method using urea as a precipitant; α-MoO (1.0 g) was dis-
persed in 100 mL of distilled water containing 1.5 mL of K PtCl
pyridine N-oxide (0.5 mmol), acetonitrile was used as a sol-
vent.
(
NH
4
)
6
7 2
O24‧4H
rate of 1 °C/min. Pt/MoO
3
Pt/H MoO3-y hybrid provided excellent selectivity for the
x
3
deoxygenation of other sulfoxides (Table 2). In all cases,
drastically improved catalytic efficiencies were obtained
under visible light irradiation relative to dark condition. A
variety of aromatic and aliphatic sulfoxides were converted
to the corresponding sulfides with excellent yields (entries
2
4
solution (35 mM). To this suspension, 0.2 g of urea as a pre-
cipitation reagent was added, followed by vigorous stirring for
2+
6 h at 95 °C to precipitate Pt ions. The obtained slurry was
centrifuged, washed with deionized water and ethanol and
then vacuum-dried to yield as-prepared Pt/MoO
quently, hydrogen reduction of Pt/MoO was conducted in a
flow of H (20 mL/min) for 30 min at ambient pressure (1 bar)
with a heating rate of 5 °C/min to yield Pt/H MoO3-y(T) (T rep-
resents the reduction temperature (°C)) as a dark blue powder.
0 min of H reduction was enough to yield Pt/H
3
. Subse-
1
–9), and pyridine N-oxides were also successfully deoxy-
genated, providing the corresponding pyridines (entry 10–
3) under the irradiation of visible light. In addition, the
3
2
1
x
structural and optical stability upon the catalytic use was
examined by XRD and UV-vis-NIR measurements (Figure
S11). Although a plasmonic wavelength caused a red-shift
from 556 to 660 nm due to the decreased carrier concen-
3
2
x
MoO3-y(T)
and no further significant optical and compositional variation
was observed even though the reduction time was extended.
Pt loading was determined to be 0.49 wt% by inductively cou-
pled plasma (ICP) analysis. This is the Pt loading optimized in
the preliminary study (for the effect of Pt loading, see Figure
S16).
tration in the MoO
ing of oxygen vacancies, the crystalline structure
phase) remained unchanged during the reac-
tion. With specific treatment to regenerate the oxygen va-
cancies, Pt/H MoO3-y hybrid could be reused over multiple
3
matrix associated with the partial fill-
(H0.9MoO
3
x
Synthesis of reference catalyst. Reference catalysts used
for the deoxygenation of sulfoxides were prepared by a similar
catalytic runs without significant loss of catalytic perfor-
mances (Figure S15). These results demonstrate the poten-
tial application of the plasmonic molybdenum oxide hy-
brid as an efficient and stable deoxygenation catalyst.
procedure mentioned above. Pd/H
x x
MoO3-y and Ru/H MoO3-y
hybrids were prepared using Na PdCl and RuCl ‧xH O as
2
4
3
2
precursors instead of K PtCl , respectively, and by reducing in
2
4
2
a flow of H at 200 °C for 30 min. Pd and Ru loadings were
determined to be 0.59 and 0.66 wt%, respectively, by induc-
tively coupled plasma (ICP) analysis. Supported Pt NP cata-
3
. CONCLUSIONS
In conclusion, efficient catalytic deoxygenation of sulfox-
lysts (Pt/WO
pared by using the corresponding oxide as a catalyst support
instead of α-MoO , and by reducing in a flow of H at 200 °C
for 30 min. Pt loading was fixed to be 0.5 wt%. Specific surface
3 2 2 2 3
, Pt/MoO , Pt/SiO , and Pt/Al O ) were also pre-
ides was demonstrated using plasmonic H
with Pt NPs. Notably, this deoxygenation proceeds under
mild reaction conditions (using 1 atm H and at room tem-
x
MoO3-y coupled
3
2
2
perature) and is drastically accelerated under visible light
irradiation relative to dark condition. Comprehensive anal-
yses revealed that the massive oxygen vacancies, reversible
redox property of Mo atoms and strong plasmonic absorp-
tion of hydrogen-doped molybdenum oxides play crucial
roles in this reaction. Such an extraordinary performance
presents a great opportunity for utilizing degenerately
doped plasmonic molybdenum oxide as a light-harvesting
2
2
area (SBET) was calculated to be 4.2 m /g (Pt/WO
3
), 5.8 m /g
O
2 3
2
2
(
Pt/MoO
2
), 222 m /g (Pt/SiO
2
), and 193 m /g (Pt/Al ), re-
spectively.
Characterization. Powder X-ray diffraction (XRD) patterns
were recorded using a Rigaku Ultima IV diffractometer with
Cu Kα radiation (λ = 1.5406 Å, 40 kV−40 mA). UV-vis-NIR dif-
fuse reflectance spectra were collected on a Shimadzu UV-
2600 recording spectrophotometer equipped with an integrat-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
ing sphere at room temperature. BaSO
4
was used as a refer-
Experimental procedures for TPR, TPD, in-situ XAFS, in-
situ FTIR measurements, TPR profiles (Figure S1), XRD
patterns and UV-vis-NIR spectra (Figure S2), Mo 3d and O
1s XPS spectra (Figure S3, Table S1), in-situ XAFS spectra
(Figure S4), TEM image and Pt 4f XPS spectra (Figure S5),
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ence and the absorption spectra were obtained using the Ku-
belka-Munk function. X-ray photoelectron spectroscopy
(XPS) measurements were carried out on a Shimadzu ESCA-
3400 photoelectron spectrometer with a Mg Kα radiation
-
6
(
1253.6 eV) X-ray source at a pressure lower than 10 Pa. The
binding energy was calibrated with the adventitious carbon (C
s) peak at 284.5 eV. Transmission electron microscopy (TEM)
3
NH -TPD profiles (Figure S6), FT-Raman spectra (Figure
S7), TG profiles (Figure S8), effect of solvent (Figure S9),
relationship between activity and some compositional pa-
rameters of catalysts (Figure S10), XRD patterns and UV-
vis-NIR spectra of spent catalyst (Figure S11), in-situ FTIR
1
images were obtained with a Hitachi H-800 electron micro-
scope operated at 200 kV. Specific surface area (SBET) and total
pore volume (Vtotal) was determined using the N
isotherm data collected at −196 °C using a BELsorp-max sys-
tem (MicrotracBEL Corp.). SBET was determined by the BET
(Brunauer–Emmett–Teller) method by using adsorption data
2
adsorption
2
spectra in H2 flow (Figure S12), effect of H partial pressure
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
on the reaction rate (Figure S13), in-situ FTIR spectra un-
der visible light (Figure S14), reusability test (Figure S15),
effect of Pt loading (Figure S16) and summary of XPS anal-
ysis (Table S1) (PDF)
0
ranging from p/p = 0.05 to 0.30. Raman spectra were obtained
with a Laser Analyzer of RAMAN-11 (Nano Photo company)
with a narrow linewidth diode laser at 532 nm and laser inten-
sity of 2 mW. Electron spin resonance (ESR) spectra were ac-
quired on a JEOL RESONANCE JES-TE200 spectrometer at
room temperature with a Klystron frequency of 9.45 GHz at
AUTHOR INFORMATION
Corresponding Author
*
yamashita@mat.eng.osaka-u.ac.jp
1.2 mW power. For ESR spectrum taken under dehydrated
Notes
conditions, as-prepared Pt/MoO
00 °C for 30 min in a flow of H
3
sample was reduced at
2
and was transferred to a
The authors declare no competing financial interest.
2
quartz-made tube without exposure to the air prior to analysis.
The signal intensities were normalized with respect to a MnO
marker.
Thermogravimetric analysis. The concentrations of hy-
v 3
drogen atoms and oxygen vacancies (O ) introduced in MoO
matrices were quantified from the thermogravimetric (TG)
data measured on a Rigaku Thermo plus EVO2 under a flow
ACKNOWLEDGMENT
This work was supported by a Grant-in-Aid from the Frontier
Research Base for Global Young Researchers, Osaka Univer-
sity, and the Grant-in-Aid for Scientific Research (KAKENHI)
from JSPS (no. 26620194, 26220911). Part of this work was per-
formed under the management of “Elements Strategy Initia-
tive for Catalysts & Batteries (ESICB)” supported by MEXT, Ja-
pan. The synchrotron radiation experiments for XAFS meas-
urements were performed at the BL01B1 beam line in SPring-8
with the approval from JASRI (no. 2016A1057, 2017A1063).
of N
mounted on an alumina pan were reduced at desired temper-
ature (T = 40, 100, and 200 °C) in a flow of H for 30 min to
yield Pt/H MoO3-y(T), and subsequently cooled to 40 °C. TG
profiles were collected by heating the samples from 40 to
50 °C with a heating rate of 10 °C/min under a 100 mL/min
flow of either N or bone-dry air.
Catalytic test: In a typical procedure, as-synthesized
Pt/MoO catalyst (0.05 g) was placed in a Pyrex glass tube, fol-
lowed by H reduction at 200 °C for 30 min. After cooling to
2 3
and air, respectively. The as-prepared Pt/MoO (10 mg)
2
x
4
REFERENCES
Linic, S.; Christopher, P.; Ingram, D. B. Nat. Mater. 2011,
0 (12), 911-21.
Clavero, C. Nat. Photonics 2014, 8 (2), 95-103.
Peiris, S.; McMurtrie, J.; Zhu, H.-Y. Catal. Sci. Technol.
2016, 6 (2), 320-338.
He, R.; Wang, Y. C.; Wang, X.; Wang, Z.; Liu, G.; Zhou,
W.; Wen, L.; Li, Q.; Wang, X.; Chen, X.; Zeng, J.; Hou, J. G. Nat.
Commun. 2014, 5, 4327.
Cheng, H.; Fuku, K.; Kuwahara, Y.; Mori, K.; Yamashita,
H. J. Mater. Chem. A 2015, 3 (10), 5244-5258.
Zhang, X.; Ke, X.; Zhu, H. Chem. Eur. J. 2012, 18 (26),
048-8056.
2
1.
1
2.
3.
3
2
ambient temperature, 10 mL of toluene solution containing
sulfoxides (0.2 mmol) was injected into the reactor through a
rubber septum by using a syringe. The reaction was performed
4.
2
in a continuous flow of atmospheric pressure of H (10
5
.
mL/min) at room temperature (20±3 °C) with magnetic stir-
ring in the dark or under visible light irradiation (λ > 420 nm)
conditions. A 500 W Hg-Xe short arc lamp (SAN-EI ELECTRIC
XEF-501S) with a 420 nm cutoff filter was used to simulate vis-
6.
8
7
.
Ke, X.; Sarina, S.; Zhao, J.; Zhang, X.; Chang, J.; Zhu, H.
2
ible light (λ > 420 nm, I = 19.6 mW/cm ). The reaction was also
Chem. Commun. 2012, 48 (29), 3509-3511.
Zhu, H.; Ke, X.; Yang, X.; Sarina, S.; Liu, H. Angew. Chem.
Int. Ed. 2010, 49 (50), 9657-9661.
Mori, K.; Verma, P.; Hayashi, R.; Fuku, K.; Yamashita, H.
Chem. Eur. J. 2015, 21 (33), 11885-93.
carried out upon irradiation by LED light (blue light: λ = 470
8
.
2
nm, I = 33.3 mW/cm ; green light: λ = 530 nm, I = 17.0
2
2
mW/cm ; red light: λ = 650 nm, I = 23.5 mW/cm ). At given
time intervals, 500 μL of the suspension was retrieved, filtered
and analysed by gas chromatograph (Shimadzu GC-2010) with
a frame ionization detector (FID) equipped with a capillary
column (HiCap-CBP10; 0.22 mm × 25 m) using biphenyl as an
internal standard.
9
.
1
4
0.
67-472.
11.
Christopher, P.; Xin, H.; Linic, S. Nat. Chem. 2011, 3 (6),
Wang, F.; Li, C.; Chen, H.; Jiang, R.; Sun, L. D.; Li, Q.;
Wang, J.; Yu, J. C.; Yan, C. H. J. Am. Chem. Soc. 2013, 135 (15), 5588-
601.
12.
Chen, C.; Zhao, J. J. Am. Chem. Soc. 2013, 135 (15), 5793-5801.
13. Verma, P.; Kuwahara, Y.; Mori, K.; Yamashita, H. J.
Mater. Chem. A 2016, 4 (26), 10142-10150.
5
ASSOCIATED CONTENT
Supporting Information.
This Supporting Information is available free of charge via the
Internet at http://pubs.acs.org.
Sarina, S.; Zhu, H.; Jaatinen, E.; Xiao, Q.; Liu, H.; Jia, J.;
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
1
4.
Della Gaspera, E.; Chesman, A. S. R.; van Embden, J.;
32.
Prasomsri, T.; Nimmanwudipong, T.; Román-Leshkov,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Jasieniak, J. J. ACS Nano 2014, 8 (9), 9154-9163.
Y. Energy Environ. Sci. 2013, 6 (6), 1732-1738.
1
5.
Johns, R. W.; Bechtel, H. A.; Runnerstrom, E. L.;
33.
Sousa, S. C. A.; Fernandes, A. C. Coord. Chem. Rev. 2015,
Agrawal, A.; Lounis, S. D.; Milliron, D. J. Nat. Commun. 2016, 7,
284, 67-92.
34.
35.
1
1
1583.
6.
Madesclaire, M. Tetrahedron 1988, 44 (21), 6537-6580.
Arnáiz, F.; Sanz, R.; Escribano, J.; Aguado, R.; Pedrosa,
Comin, A.; Manna, L. Chem. Soc. Rev. 2014, 43, 3957-
Liu, X.; Swihart, M. T. Chem. Soc. Rev. 2014, 43 (11),
Li, Y.; Cheng, J.; Liu, Y.; Liu, P.; Cao, W.; He, T.; Chen,
3
975.
M. Synthesis 2004, 2004 (10), 1629-1632.
36. Fernandes, A. C.; Romão, C. C. Tetrahedron Lett. 2007,
48 (52), 9176-9179.
17.
3908-20.
18.
37.
Mikami, Y.; Noujima, A.; Mitsudome, T.; Mizugaki, T.;
R.; Tang, Z. J. Phys. Chem. C 2017, 121 (9), 5208-5214.
9. Luther, J. M.; Jain, P. K.; Ewers, T.; Alivisatos, A. P. Nat.
Mater. 2011, 10 (5), 361-6.
Jitsukawa, K.; Kaneda, K. Chem. Eur. J. 2011, 17 (6), 1768-72.
1
38.
K.; Kaneda, K. Chem. Lett. 2014, 43 (4), 420-422.
39. Ogura, K.; Yamashita, M.; Tsuchihashi, G. Synthesis
1975, 1975, 385-387.
Takahashi, Y.; Mitsudome, T.; Mizugaki, T.; Jitsukawa,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
21.
0.
017, 674, 1-52.
Hu, X. K.; Qian, Y. T.; Song, Z. T.; Huang, J. R.; Cao, R.;
Xiao, J. Q. Chem. Mater. 2008, 20, 1527-1533.
2. Cheng, H.; Wen, M.; Ma, X.; Kuwahara, Y.; Mori, K.; Dai,
Y.; Huang, B.; Yamashita, H. J. Am. Chem. Soc. 2016, 138 (29), 9316-
Kriegel, I.; Scotognella, F.; Manna, L. Physics Reports
40.
Touchy, A. S.; Hakim Siddiki, S. M. A.; Onodera, W.;
Kon, K.; Shimizu, K.-i. Green Chem. 2016, 18 (8), 2554-2560.
2
41.
Mizuno, N. Catal. Sci. Technol. 2017, 7 (9), 1912-1920.
42. Mitsudome, T.; Takahashi, Y.; Mizugaki, T.; Jitsukawa,
K.; Kaneda, K. Angew. Chem. Int. Ed. 2014, 53 (32), 8348-8351.
43.
44.
Chem. 1988, 92, 467-470.
45.
46.
5957-5969.
47.
Uematsu, T.; Ogasawara, Y.; Suzuki, K.; Yamaguchi, K.;
2
2
4.
3.
Cheng, H.; Kamegawa, T.; Mori, K.; Yamashita, H.
Angew. Chem. Int. Ed. 2014, 53 (11), 2910-29.
2
Medehkar, N. V.; Beane, G. A.; Kaner, R. B.; Russo, S. P.; Ou, J. Z.;
Kalantar-zadeh, K. Adv. Mater. 2014, 26, 3931-3937.
Birtill, J. J.; Dickens, P. G. Mat. Res. Bul. 1978, 13, 311-316.
Pichat, P.; Mozzanega, M.-N.; Hoang-Van, C. J. Phys.
4. Alsaif, M. M. Y. A.; Latham, K.; Field, M. R.; Yao, D. D.;
Adams, S. J. Solid State Chem. 2000, 149, 75-87.
Braïda, B.; Adams, S.; Canadell, E. Chem. Mater. 2005, 17,
2
5.
Alsaif, M. M. Y. A.; Chrimes, A. F.; Daeneke, T.;
Balendhran, S.; Bellisario, D. O.; Son, Y.; Field, M. R.; Zhang, W.;
Nili, H.; Nguyen, E. P.; Latham, K.; van Embden, J.; Strano, M. S.;
Ou, J. Z.; Kalantar-zadeh, K. Adv. Funct. Mater. 2016, 26 (1), 91-100.
Kim, H. S.; Cook, J. B.; Lin, H.; Ko, J. S.; Tolbert, S. H.;
Ozolins, V.; Dunn, B. Nat. Mater. 2017, 16 (4), 454-460.
2
6.
Balendhran, S.; Deng, J.; Ou, J. Z.; Walia, S.; Scott, J.;
48.
S.; Hany, R.; Nuesch, F. Sci. Rep. 2017, 7, 40761.
49. Lei, Y.-H.; Chen, Z.-X. J. Phys. Chem. C 2012, 116 (49),
25757-25764.
Borgschulte, A.; Sambalova, O.; Delmelle, R.; Jenatsch,
Tang, J.; Wang, K. L.; Field, M. R.; Russo, S.; Zhuiykov, S.; Strano,
M. S.; Medhekar, N.; Sriram, S.; Bhaskaran, M.; Kalantar-Zadeh, K.
Adv. Mater. 2013, 25, 109-114.
2
7.
ChemPhysChem 2016, 17 (17), 2630-9.
8. Liu, W.; Xu, Q.; Cui, W.; Zhu, C.; Qi, Y. Angew. Chem.
Int. Ed. 2017, 56 (6), 1600-1604.
Tan, X.; Melkersson, J.; Wu, S.; Wang, L.; Zhang, J.
50.
Ou, J. Z.; Campbell, J. L.; Yao, D.; Wlodarski, W.;
Kalantar-zadeh, K. J. Phys. Chem. C 2011, 115, 10757–10763.
51.
Soc. Jpn. 1989, 62, 903-907.
52.
53.
6930-6941.
54.
55.
2
Sotani, N.; Eda, K.; Sadamatu, M.; Takagi, S. Bull. Chem.
29.
Yin, H.; Kuwahara, Y.; Mori, K.; Cheng, H.; Wen, M.;
Seshadri, K. S.; Petrakis, I. J. Catal. 1973, 30, 195-203.
Watson, R. B.; Ozkan, U. S. J. Phys. Chem. B 2002, 106,
Yamashita, H. J. Mater. Chem. A 2017, 5 (19), 8946-8953.
3
0.
H. Adv. Mater. 2015, 27 (31), 4616-4621.
1. Guan, J.; Peng, G.; Cao, Q.; Mu, X. J. Phys. Chem. C 2014,
18 (44), 25555-25566.
Cheng, H.; Qian, X.; Kuwahara, Y.; Mori, K.; Yamashita,
Eda, K. J. Solid State Chem. 1992, 98, 350-357.
Zhang, X.; Ke, X.; Du, A.; Zhu, H. Sci. Rep. 2014, 4, 3805.
3
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table of Contents (TOC)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0
ACS Paragon Plus Environment