Photocatalytic selective aerobic oxidation of amines to nitriles over Ru/γ-Al2O3: The role of the support surface and the strong imine intermediate adsorption

Article abstract of DOI:10.1039/c9cy01916b

Hydroxyl coordinated ruthenium dispersed on the surface of γ-Al₂O₃ can be applied to the selective oxidation of amines with light irradiation and an atmospheric pressure of O₂ at room temperature. Sunlight is also an effective light source for the selective aerobic oxidation of primary amines to corresponding nitriles. The high photocatalytic activity and selectivity over Ru/γ-Al₂O₃ originate from the adsorption of amines and imine intermediates on the abundant surface OH groups of the photocatalyst and further formation of Ru-amide species by ligand exchange of adsorbed amines and imine intermediates with adjacent exposed active Ru sites. Light is introduced to the system successfully via the formation of Ru-amide species, which are used as the light absorption sites of the photocatalytic selective oxidation of amines. Primary amines are directly converted to corresponding nitriles via a two-step oxidative dehydrogenation process.

Full text of DOI:10.1039/c9cy01916b

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Photocatalytic selective aerobic oxidation of
amines to nitriles over Ru/γ-Al2O3: the role of the
support surface and the strong imine intermediate
adsorption†
Cite this: DOI: 10.1039/c9cy01916b
ab
a
Pengqi Zhu,
Jin Zhang,ab Jie Wang, ab Peng Kong,
Yunwei Wang a and Zhanfeng Zheng
ab
*
*
Hydroxyl coordinated ruthenium dispersed on the surface of γ-Al2O3 can be applied to the selective oxida-
tion of amines with light irradiation and an atmospheric pressure of O2 at room temperature. Sunlight is
also an effective light source for the selective aerobic oxidation of primary amines to corresponding nitriles.
The high photocatalytic activity and selectivity over Ru/γ-Al2O3 originate from the adsorption of amines
and imine intermediates on the abundant surface OH groups of the photocatalyst and further formation of
Ru–amide species by ligand exchange of adsorbed amines and imine intermediates with adjacent exposed
active Ru sites. Light is introduced to the system successfully via the formation of Ru–amide species, which
are used as the light absorption sites of the photocatalytic selective oxidation of amines. Primary amines
are directly converted to corresponding nitriles via a two-step oxidative dehydrogenation process.
Received 22nd September 2019,
Accepted 5th December 2019
DOI: 10.1039/c9cy01916b
rsc.li/catalysis
amounts of I2, TBHP, and PIDA as oxidants. From the per-
spective of sustainable chemistry, molecular oxygen is a
Introduction
Nitrogen containing compounds, such as imines, nitriles and
amides, are essential industrial intermediates for the synthe-
sis of pharmaceuticals, natural products, and functional
materials.1–3 Conventional synthetic methods for nitrile com-
pounds include the Sandmeyer reaction,4 the Rosenmund–
von Braun reaction of aromatic halides,5 and the transition-
metal-catalyzed cyanation of aromatic halides.6 However,
these methods suffer from some drawbacks, such as the us-
age of toxic cyanide salts as the source of cyano, and the gen-
eration of stoichiometric amounts of metal and inorganic
wastes in the process. Therefore, in the past decades some
methods have been developed to avoid the use of toxic cya-
nide, such as oxidative conversion of alcohols, aldehydes, and
amines to nitriles,7,8 the oxidation of primary azides,9 and the
direct transformation of methyl arenes to aryl nitriles.10 It is
noteworthy that those reactions need to use stoichiometric
green, cheap, and efficient oxidant. McWhinnie et al. firstly
found that ruthenium catalysts may have the ability to oxidize
amines under an oxygen atmosphere.11 Despite all that, it
was not until the pioneering study of Tang et al. that homoge-
neous ruthenium complex catalytic oxidation of amines with
oxygen was reported.12 Since then, a number of ruthenium
complexes have been used for the oxidation of amines with
oxygen.13–18 Parvulescu et al. discovered that the ruthenium
terpyridyl complex functionalized with the azulenyl moiety at
the 4 position of the central pyridine core exhibited a much
higher reactivity for the oxidation of both aliphatic and aro-
matic amines with oxygen or air than other ruthenium
terpyridine-based ligand complexes reported. In addition, the
alkyl chain length of the reactant has a significant influence
on the conversion.17 Lahiri et al. reported that ruthenium
hydrido complexes can be used for the simultaneous genera-
tion of nitriles and imines by varying the catalyst. They
showed that the ligand played a vital role in altering the selec-
tivity in the oxidative dehydrogenation of primary amines.
However, this system also made use of TEMPO as an electron
transfer mediator.18 Meanwhile, Szymczak's group19 and Bera
et al.20 realized the double dehydrogenation of primary
amines to nitriles under oxidant-free and acceptorless condi-
tions by using a Ru/pyrazole system and an amide-derived Ru
complex, respectively. In order to reuse the noble metal cata-
lysts, a series of heterogeneous ruthenium catalysts have been
a State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese
Academy of Sciences, Taiyuan 030001, China. E-mail: zfzheng@sxicc.ac.cn,
wangyunwei@sxicc.ac.cn
b Center of Materials Science and Optoelectronics Engineering, University of
Chinese Academy of Sciences, Beijing 100049, China
† Electronic supplementary information (ESI) available: Experimental section,
catalyst characterisation, photocatalytic performances, study of hydroxyl groups
on the surface of Ru/γ-Al2O3, reaction kinetics, and identification of the electron
transfer between Ru and amine in the Ru–amide complex under light irradia-
tion. See DOI: 10.1039/c9cy01916b
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
reported for the oxidation of amines to corresponding nitriles
with oxygen as the sole oxidant.21–26 Kaneda et al. reported
that a truly heterogeneous Ru/HAP catalyst (Ru loading: 17
wt%) can be used for the aerobic oxidation of both aromatic
and aliphatic amines to corresponding nitriles.21 Later,
Mizuno's group synthesized an alumina supported ruthenium
hydroxide catalyst, which exhibited a much higher activity for
the oxidation of both aromatic and aliphatic amines than Ru/
HAP.22 Although a slightly decreased nitrile selectivity was ob-
served during benzylamine oxidation (82%), the catalyst also
benefitted from significantly decreased metal loadings with
respect to Ru/HAP, with only 1.4 wt% ruthenium being suffi-
cient for catalytic activity. Given the efficacy of the alumina-
supported ruthenium catalysts for this reaction, much work
has since been performed on supported ruthenium on vari-
ous metal oxides.23–25 Wang et al. reported that hydrated ru-
thenium oxide (RuO2·xH2O) dispersed on Co3O4 showed high
efficiency towards the aerobic oxidation of amines to ni-
triles.24 They discovered that the content of ruthenium and
the size of Co3O4 particles played crucial roles in determining
the catalytic performance and RuO2·xH2O was the active spe-
cies. Additionally, Pârvulescu's group immobilized the
Ru2Cl4IJaz-tpy)2 complexes on ionic liquid dispersed silica for
the oxidation of long chain linear primary amines to the cor-
responding nitriles.26 They concluded that the conversion was
strongly influenced by the alkyl chain length of the amine
substrate and the choice of oxidant. However, controlling the
selectivity of the oxidation of amines with molecular oxygen
as the oxidant and the activation of molecular oxygen under
mild reaction conditions are still an enormous challenge.27 It
is highly desirable to prepare these chemicals in a green
manner.10
(>99%).43 They suggested that the reaction proceeded via the
localized surface plasmon resonance-excited electron transfer
from Au nanoparticles to TiO2. When a supported Ru catalyst
was used for the photocatalytic oxidation of amines, the high
selectivity products were nitriles.44 Golovko et al. demon-
strated the first example of the photocatalytic aerobic oxida-
tion of amines to nitriles by using hydrous ruthenium oxide
supported on TiO2 under visible light irradiation.44 They
proposed that the photons were absorbed by hydrous ruthe-
nium oxide through interband transitions and/or the local-
ized surface plasmon resonance and the photoexcited
electrons were transferred into the TiO2 conduction band to
activate molecular oxygen.
However, the detailed reaction mechanism of the oxida-
tion of amines to corresponding nitrile products over
supported Ru catalysts with light irradiation is still unclear.
Therefore, it is highly desirable to understand the surface ad-
sorption and interaction between Ru species and amines,
how the reaction is initialised, and what is the surface cata-
lytic pathway for the high selectivity to nitriles.
Herein, we report that the highly dispersed and hydroxyl
coordinated ruthenium photocatalyst, Ru/γ-Al2O3, is efficient
for the oxidation of amines to nitriles with molecular oxygen
under mild reaction conditions (33 °C, 1 atm) under UV or
visible light irradiation. The abundant surface OH groups
and highly dispersed electron-deficient Ru active sites on the
Ru/γ-Al2O3 photocatalyst are responsible for the high activity
and selectivity of photocatalytic aerobic oxidation of amines.
The abundant surface OH groups of the photocatalyst play an
important role for the adsorption of amines and imine inter-
mediates, which are favourable for the formation of Ru–am-
ide species by ligand exchange with the adjacent exposed
highly dispersed electron-deficient Ru active sites. The sur-
face adsorption of reactants and intermediates and formation
of Ru–amide species ensure that the reaction proceeds by
two-step oxidative dehydrogenation to form nitriles instead
of condensation. Both aromatic amines and aliphatic amines
can be converted to their corresponding nitriles efficiently.
Recently, the photocatalytic process has attracted great at-
tention for the preparation of nitrogen containing com-
pounds starting from amines. It is found that the condensa-
tion compound imine (e.g. N-benzylidenebenzylamine from
benzylamine) is generally obtained on a large number of
solid surfaces.
A number of semiconductors, such as
TiO2,28,29 Nb2O5,30,31 CdS,32 mpg-C3N4,33 LDH,34 BiOCl,35
BiOV4,36 and a recently reported TiO2-complex,37,38 have been Experimental
successfully used for the photocatalytic oxidation of amines
Materials
to imines. Zhao et al.29 and Tanaka et al.30 reported the effi-
All chemical reagents were used as received without any fur-
cient catalytic oxidation of amines to imines under visible
light and oxygen conditions by using TiO2 and Nb2O5, respec-
tively. They found that the visible-light response of such sys-
tems was caused by the formation of a surface complex
through the adsorption of benzylamine on the surface of
TiO2 or Nb2O5. Novel material MOFs NH2-MIL-125IJTi)39 and
2D Por-sp2c-COF40 were also effective catalysts for the photo-
catalytic oxidation of amines to imines. Several transition
metal based catalytic systems have been established for this
transformation under light irradiation (e.g. Ag/AgI supported
on titanate nanotubes,41 Au–Pd/ZrO2,42 and Au/TiO2 43).
Tada's group reported that a Au/TiO2 photocatalyst exhibited
a high visible-light activity for aerobic oxidation of amines to
yield the corresponding imines with high selectivity
ther purification. RuCl3 (99%) and benzotrifluoride (98%)
were purchased from Energy Chemical. γ-Al2O3 (10 nm,
99.99%), γ-Al2O3 (20 nm, 99.99%), AlIJOH)3 (99.6%), SiO2
(99%), ZrO2 (99.99%), 4-fluorobenzylamine (99%), and
n-amylamine (98%) were purchased from Aladdin.
N-Benzylidenebenzylamine (99%) and n-hexylamine (99%)
were purchased from Sigma-Aldrich. All the other chemicals
were from Sinopharm Chemical Reagent Co. Ltd.
Synthesis of photocatalysts
Ru/γ-Al2O3 was prepared by an impregnation method. RuCl3
(0.3099 g) was placed in a beaker and dissolved in deionized
water (60 mL). γ-Al2O3 (10 nm, 2.0 g) was dispersed in the
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
resulting precursor solution and stirred at room temperature
for 2 h. The resulting slurry was evaporated to dryness at 80
°C. Then, the solid was added to deionized water (30 mL),
and the pH value of the suspension was adjusted to 13.2 by
the addition of NaOH aqueous solution (1 mol L−1). After stir-
ring at room temperature for 24 h, the solid was separated by
filtration, washed with deionized water and dried at 60 °C in
a vacuum oven for 12 h. Various ruthenium-based photo-
catalysts were prepared by the same method with different
supports.
EXAFS data were extracted and processed according to the
standard procedures using the ATHENA module imple-
mented in the IFEFFIT software packages. The structural pa-
rameter fitting was performed using the ARTEMIS module
with a fitting range of 1.10–2.20 Å.
Photocatalytic performance evaluation
The photocatalytic activity for selective oxidation of amines
to corresponding nitriles was evaluated at 33 °C in an oxygen
atmosphere (1 atm) for 3 h. Typically, the reaction system
consisted of
1 mL of benzotrifluoride, 0.2 mmol of
Synthesis of RuIJOH)x
benzylamine and 40 mg of photocatalyst. A 300 W high-
pressure xenon lamp (CEL-HXF 300, Beijing Aulight Co. Ltd.,
320–700 nm) was used and the output spectra are shown in
Fig. S1a.† The light source was mounted on the top of the re-
actor. The intensity of irradiation light was changed by
adjusting the distance between the light source and the reac-
tion liquid. The light intensity was measured at the cross-
section of the light and the upper liquid surface using an op-
tical power meter (Model 1918-R, Newport). Single coloured
LED lights (100 W COB LED lamp) include a purple LED
(375–430 nm, peaked at 400 nm), blue LED-430 (400–475 nm,
peaked at 430 nm), blue LED-455 (425–510 nm, peaked at
455 nm), green LED (475–600 nm, peaked at 525 nm), and
red LED (580–665 nm, peaked at 625 nm) (output spectra
shown in Fig. S1b†).
RuCl3 (0.3099 g) was placed in a beaker and dissolved in de-
ionized water (60 mL). Then, the pH value of the solution
was adjusted to 13.2 by the addition of NaOH aqueous solu-
tion (1 mol L−1). After stirring at room temperature for 24 h,
the solid was separated by filtration, washed with deionized
water and dried at 60 °C in a vacuum oven for 12 h.
γ-Al2O3 treated with water
γ-Al2O3 (10 nm, 2.0 g) was dispersed in deionized water (30
mL) and stirred at room temperature for 24 h. The resulting
slurry was separated by filtration and dried at 60 °C in a vac-
uum oven for 12 h.
Chemical titration of hydroxyl groups on Ru/γ-Al2O3
2 g benzoic anhydride and 50 mg Ru/γ-Al2O3 were dissolved
in 50 mL of CHCl3. After stirring under N2 protection at 60
°C for 24 h, the solid was separated by centrifugation,
washed with ethanol and dried at 60 °C in a vacuum oven for
12 h.
Procedure for the light-promoted electron transfer from
1,4-bisIJdimethylamino)benzene to molecular oxygen
40 mg of Ru/γ-Al2O3 and 0.1 mmol of 1,4-
bisIJdimethylamino)benzene were dispersed in 1 mL benzo-
trifluoride and irradiated with the xenon lamp in an O2 atmo-
sphere (1 atm) at 33 °C for 1 h. The resulting reaction mix-
ture was analyzed qualitatively by UV-vis.
Characterization
The X-ray diffraction (XRD) patterns were recorded on a
MiniFlex II diffractometer with Cu Kα radiation (λ = 1.5418
Å). The photocatalysts were analyzed by transmission electron
microscopy (TEM, Tecnai G2 F20 S-Twin) and diffuse reflec-
tance UV-vis (DR UV-vis) spectroscopy (Hitachi UV-3900). In-
ductively coupled plasma optical emission spectroscopy (ICP-
Photocatalytic performance evaluation using the Sun as a
light source
The photocatalytic oxidation of benzylamine using the Sun as
a light source was conducted at 33 °C in an oxygen atmo-
sphere (1 atm) for 6 h. The reaction system consisted of 1 mL
benzotrifluoride, 0.2 mmol benzylamine and 40 mg photo-
catalyst. The photocatalytic setup was placed outside the lab-
oratory (latitude 37°51′ north, longitude 112°33′ east) so that
the reactor would be directly irradiated by the Sun for the du-
ration of the experiment (10 a.m.–4 p.m. 18/05/2017, summer
in the Northern hemisphere).
OES) measurements were performed using
a Thermo
iCAP6300 spectrometer. The X-ray photoelectron spectra
(XPS) were recorded using a Thermo ESCALAB 250 spectro-
meter with an Al Kα X-ray source (hν = 1486.6 eV). The bind-
ing energy was calibrated using the C1s peak at 284.6 eV. In
situ Fourier transform infrared spectra (FT-IR) of the samples
before and after light irradiation were recorded on a Bruker
Tensor II IR spectrometer equipped with a MCT detector. The
photocatalytic products were analyzed using a Bruker SCION
SQ GC-MS spectrometer and a Shimadzu 2014C GC with a
WondaCap 5 column. Ru K-edge XAFS spectra were recorded
in fluorescence mode at the beam line 14W1 station in the
Shanghai Synchrotron Radiation Facility with an attached Si
(311) monochromator. The spectra of Ru foil and RuO2 refer-
ences were recorded in transmission mode. The acquired
Photoelectrochemical test
Transient photocurrent measurements were conducted on an
electrochemical workstation (CHI 760D, CH Instruments,
Shanghai, China) using a 300 W high-pressure xenon lamp
(CEL-HXF 300, Beijing Aulight Co. Ltd., 320–700 nm) as the
light source. A conventional three-electrode with an applied
bias of 0.9 V including the work electrode, a Ag/AgCl
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
electrode (saturated KCl) was used as a reference electrode
and a Pt wire was used as the counter electrode. The working
electrode was prepared on fluorine-doped tin oxide (FTO)
glass. The catalyst (5 mg) was dispersed in 1 mL of mixture
solution of water and ethanol (water : ethanol = 2 : 3) and 10
μl Nafion (DuPont, 5 wt%) to form a slurry solution by
ultrasonication. 20 μL of the slurry was coated on the FTO
glass and dried at ambient temperature for 12 h. The tran-
sient photocurrent measurements were performed under illu-
mination conditions and the working electrodes were im-
mersed in 0.1 mol L−1 Na2SO4 aqueous solution with the pH
maintained at 6.8.
Results and discussion
Characterization of Ru/γ-Al2O3
The structure of supported Ru/γ-Al2O3 was studied in detail to
understand the structure of Ru active sites. The Ru content
of Ru/γ-Al2O3 was 3.9 wt%, as confirmed by inductively
coupled plasma (ICP) analysis. The X-ray diffraction (XRD)
patterns (Fig. S2, ESI†) and transmission electron microscopy
(TEM) images of Ru/γ-Al2O3 (Fig. S3, ESI†) showed no Ru
structures and Ru-containing particles, indicating the high
dispersion nature of Ru species on the supports.
X-ray photoelectron spectroscopy (XPS), Ru K-edge X-ray ab-
sorption near-edge structure (XANES) and extended X-ray ab-
sorption fine structure (EXAFS) measurements were conducted
to further verify the electronic properties and local coordina-
tion environment of the Ru species in Ru/γ-Al2O3. The Ru
K-edge XANES spectra (Fig. 1a) show that the energy absorption
threshold value of Ru/γ-Al2O3 is higher than that of Ru foil and
lower than that of RuO2. Additionally, as shown in Fig. 3a and
S4,† the binding energies of Ru 3p3/2 and Ru 3p1/2 in Ru/γ-
Al2O3 are 463.6 eV and 485.8 eV, respectively, which are similar
Fig.
2
The IR analysis of surface OH groups over different
photocatalysts. (a) IR spectra of Ru/γ-Al2O3 and Ru/γ-Al2O3 treated
with different amounts of benzoic anhydride. (b) IR spectra of
ruthenium-based photocatalysts with different supports.
the existence of RuO2. Curve-fitting analysis for Ru/γ-Al2O3
was performed, and the fitting results are shown in Fig. S5
and Table S1.† The first shell was well-fitted by the use of one
longer Ru–O (2.09 Å) and five short Ru–O (2.04 Å). These facts
reveal that a monomeric Ru species with five hydroxyl groups
was anchored onto an oxygen atom originating from the hy-
droxyl group on the surface of γ-Al2O3.
to that of RuCl3IJIII) while higher than the value of 461.8 eV
45
for Ru(0) 3p3/2 and 463.1 eV for RuO2IJIV) 3p3/2
.
These facts
reveal that the Ru species is in the +3 oxidation state.
The Ru K-edge EXAFS Fourier transformed (FT) K3-
weighted χ(k) function spectra are shown in Fig. 1b. The main
peak of Ru/γ-Al2O3 appears at around 1.60 Å attributed to the
Ru–O bond. The weak peak observed at 2.20 Å is assigned to
the Ru–O–Al bonds.46,47 Compared with Ru foil, the absence
of a peak at around 2.35 Å means that there is no Ru–Ru
bond. The absence of the Ru–O–Ru bond (3.15 Å) excludes
Photocatalytic performance of Ru/γ-Al2O3
The effect of metal loading for the Ru/γ-Al2O3 photocatalyst in
the photocatalytic oxidation of benzylamine was studied (Ta-
ble S2†). The Ru/γ-Al2O3 photocatalyst with 3.9 wt% Ru con-
tent exhibited the highest activity for the photocatalytic oxi-
dation of benzylamine with O2 (1 atm) at 33 °C. 99% of
benzylamine was consumed in 3 h with 86% selectivity for
benzonitrile under light irradiation (entry 1, Table 1).
For comparison, the control reaction was conducted in the
dark and all the other reaction conditions were kept identi-
cal. The conversion of benzylamine was 39% with 84% selec-
tivity to benzonitrile. The catalytic activity and selectivity un-
der light irradiation were much higher than those in the dark
reaction, indicating the influence of light on the reaction.
Fig. 1 Structure characterization of Ru/γ-Al2O3. (a) Ru K-edge XANES
spectra and (b) Ru K-edge EXAFS Fourier transformed (FT) K3-weighted
χ(k) function spectra of Ru/γ-Al2O3 and reference samples.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Fig. 3 Identification of the presence of Ru–amide complexes and reaction intermediates. (a) XPS spectra of RuIJOH)x, Ru/γ-Al2O3, and Ru/γ-Al2O3
with adsorbed benzylamine. (b) DR UV-vis spectra of Ru/γ-Al2O3 before and after benzylamine adsorption. (c) Time profile of oxidation of
benzylamine under light irradiation. Reaction conditions: benzylamine (0.2 mmol), Ru/γ-Al2O3 (40 mg), benzotrifluoride (1 mL), O2 (1 atm), 33 °C,
xenon lamp and light intensity 0.87 W cm−2. (d) In situ FT-IR spectra obtained when Ru/γ-Al2O3 was exposed to vaporous benzylamine (BA) for 15,
35, and 55 min; N2 was introduced to the system for 6, 15, and 30 min; O2 was introduced to the system in the dark and under light irradiation.
Table 1 Catalytic activity for the selective oxidation of benzylamine to nitriles over different catalystsa
Entry
1
Catalyst
Incident
Conv.b (%)
Sel.b (%)
Ru/γ-Al2O3
Visible
Dark
Visible
Dark
Visible
Dark
Visible
Dark
99
39
8
86
84
0
—
11
5
61
24
—
—
2
γ-Al2O3
No detection
2
1
5
3c
4d
5
RuIJOH)x
Physical mixture of RuIJOH)x and γ-Al2O3
2
No catalyst
Visible
Dark
No detection
No detection
a
Reaction conditions: benzylamine (0.2 mmol), catalyst (40 mg), benzotrifluoride (1 mL), O2 (1 atm), 33 °C, 3 h, xenon lamp and light intensity
0.87 W cm−2
.
Selectivity to nitriles. Conversion and selectivity were determined by gas chromatography (GC). RuIJOH)x (8 mg, five times the
b
c
d
amount of Ru in the supported Ru/γ-Al2O3 photocatalyst). RuIJOH)x (8 mg), γ-Al2O3 (40 mg).
The influence of light intensity on the photocatalytic oxida-
tion of benzylamine was studied. The yield increased as the
light intensity was raised, which demonstrates that the reac-
tions are driven by light (Fig. S7†).48,49 The light contribution
was calculated as the contribution of the irradiation effect in
the reaction, and defined as [(YLight − YDark)/YLight] × 100%,
where YLight and YDark are the product yields under irradia-
tion and dark conditions, respectively.42 With the increase of
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
light intensity, the light contribution gradually increases. The
higher the incident light intensity, the more electrons in the
photocatalyst can be excited to facilitate the improvement of
photocatalytic activity.
lectivity of the reaction. Additionally, the XPS binding energy
of various Ru photocatalysts is similar (Fig. S6†), which indi-
cates that the coordination and electronic environment of ru-
thenium is similar in different Ru photocatalysts, and the
achievement of high activity relies on the adsorption of
amines on the surface OH groups and the highly dispersed ac-
tive Ru sites on the surface of the photocatalyst. Compared
with the other supports, γ-Al2O3 (10 nm) ensures high activity
and selectivity owing to the abundant OH groups and high dis-
persion of active Ru sites on its surface.
When the reaction was conducted in the presence of ar-
gon instead of O2, 7% benzylamine was transformed with
44% selectivity to benzonitrile, indicating that the reaction
may undergo an oxidative dehydrogenation pathway. In addi-
tion, there is no benzonitrile either in the presence of γ-Al2O3
or in the absence of the Ru/γ-Al2O3 photocatalyst (entries 2
and 5, Table 1), suggesting that Ru species are vital active
sites in the reaction. The catalytic activities of RuIJOH)x as
well as a mixture of RuIJOH)x and γ-Al2O3 are extremely lower
than that of the Ru/γ-Al2O3 photocatalyst (entries 3 and 4,
Table 1). These facts reveal that Ru species anchored on the
surface of γ-Al2O3 are indispensable in achieving high photo-
catalytic activity and selectivity.
In contrast to reported studies,22,44 our work reveals that
the high photocatalytic activity and selectivity over Ru/γ-Al2O3
(10 nm) presumably originate from the adsorption of amines
and imine intermediates over the abundant surface OH
groups of the photocatalyst and further formation of Ru–am-
ide species by ligand exchange of adsorbed amines and imine
intermediates with adjacent exposed active Ru sites.
Photocatalytic mechanism of Ru/γ-Al2O3
The role of the γ-Al2O3 support
To verify our hypothesis and further understand the reaction,
we studied the mechanism of photocatalytic oxidation of
benzylamine over the Ru/γ-Al2O3 photocatalyst. Previous re-
ports of photocatalytic oxidation of primary amines are differ-
ent from our system.28,51 Nitriles were obtained in the reac-
tion rather than secondary imines. There are two major
processes for the formation of secondary imines: an oxygena-
tion pathway [eqn (a)] and a dehydrogenation pathway [eqn (b)].
The absence of aldehydes and no accumulation of secondary
imines indicate that the reaction in our system may not be car-
ried out through pathways (a) and (b). Inspired by the above
mentioned pathways of the oxidation of primary amines, we
can hypothesise a reasonable two-step oxidative dehydrogena-
tion process for the production of nitriles [eqn (c)].
The electronic environment of Ru in Ru/γ-Al2O3 was investi-
gated by XPS analysis. The binding energy of Ru 3p3/2 (463.6
eV) in Ru/γ-Al2O3 is higher than the value of 463.0 eV for the
unsupported RuIJOH)x (Fig. 3a), which indicates the formation
of electron-deficient Ru species due to the strong interaction
of Ru and γ-Al2O3.
The role of surface OH groups on Ru/γ-Al2O3 was explored
according to a reported method.50 Benzoic anhydride was cho-
sen as the titrant to deactivate the OH groups on Ru/γ-Al2O3.
The infrared spectra (IR) of Ru/γ-Al2O3 (benzoic anhydride)
exhibit a decrease of the O–H signal at about 3450 cm−1 with
the increase of benzoic anhydride amount (Fig. 2a). The bands
appeared at 3063 and 2953 cm−1, combined with the vibration
at 1604, 1561, 1494, and 1433 cm−1 owing to the presence of a
benzene ring after titration and the strength gradually in-
creases with the increase of benzoic anhydride amount, indi-
cating the successful reaction between benzoic anhydride and
hydroxyl groups on Ru/γ-Al2O3. The conversion and selectivity
gradually decreased to 15% and 25% with the increase of
benzoic anhydride amount (Table S5†), demonstrating that the
surface OH groups on Ru/γ-Al2O3 facilitate the high activity
and selectivity of photocatalytic oxidation of amines.
(a)
(b)
(c)
To further explore the reasons for the high activity and se-
lectivity of photocatalytic oxidation of amines over the Ru/γ-
Al2O3 photocatalyst, the photocatalytic activities of other Ru-
based photocatalysts with different supports were tested. Apart
from Ru/SiO2 and Ru/ZrO2, Ru/γ-Al2O3 (10 nm), Ru/γ-Al2O3 (20
nm), and Ru/AlIJOH)3 exhibit higher selectivity (Table S3†). As
shown in Fig. 2b, the band appearing at 3450 cm−1 can be at-
tributed to the stretching vibration of OH groups. The band
intensity of the OH groups in Ru/γ-Al2O3 (10 nm), Ru/γ-Al2O3
(20 nm), and Ru/AlIJOH)3 is significantly greater than that of
Ru/SiO2 and Ru/ZrO2, indicating that there are more OH
groups on the surface of Ru/γ-Al2O3 (10 nm), Ru/γ-Al2O3 (20
nm), and Ru/AlIJOH)3. The surface OH groups are conducive to
the adsorption of imine intermediates to achieve the high se-
The interaction of Ru/γ-Al2O3 and benzylamine was investi-
gated by XPS analysis and diffuse reflectance UV-vis (DR UV-
vis) spectroscopy. The binding energy of Ru 3p3/2 and Ru
3p1/2 decreased to 463.0 eV and 485.3 eV, respectively, after
adsorption of benzylamine, indicating the formation of Ru–
amide species and the increase of the electron density of Ru
because of the electrons partly transferring from nitrogen to
Ru (Fig. 3a). γ-Al2O3 has no absorption in the visible light re-
gion (Fig. 3b).52 The absorption in the visible range is due to
the Ru species, and the two main absorption bands at 405
and 460 nm can be assigned to the metal-to-ligand charge
transfer transition (MLCT, d–π*) of Ru species.53 The differ-
ential spectrum before and after benzylamine adsorption
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
exhibits a broad absorption band centred at 430 nm. Because
benzylamine does not absorb light with a wavelength longer
than 300 nm,30 the new absorption band can be attributed to
the formation of a Ru–amide surface complex. As shown in
Table S4,† different single coloured LED lights can drive the
reaction, indicating that the reaction can be conducted not
only under UV, but also under visible light irradiation. Mean-
while, the photocatalytic activity under different single
coloured LED lights coincided with the differential spectrum
before and after benzylamine adsorption, further demonstrat-
ing that the Ru–amide species are the light absorption sites.
Ru/γ-Al2O3 can also promote the oxidation of benzylamine un-
der simulated sunlight irradiation. The experiment was
conducted under sunlight irradiation for 6 h, where the con-
version reached 83% with 88% benzonitrile selectivity. The
photocatalyst could be used for the photocatalytic oxidation
of amines without obvious activity loss after three times.
The reaction kinetic data of the photocatalytic oxidation
of benzylamine and N-benzylidenebenzylamine (N-BBA) were
obtained to demonstrate that N-BBA is not an intermediate
in the oxidation of benzylamine under light irradiation (see
the ESI† for details). If N-BBA was an intermediate, the accu-
mulation of N-BBA would exceed 37% because the rate con-
stant of benzylamine oxidation (k1 = 1.3812 h−1, Fig. S8†) is
larger than that of the N-BBA oxidation (k2 = 0.2681 h−1, Fig.
S9†). As shown in Fig. 3c, the yield of benzonitrile increases
gradually when the reaction time is prolonged and the reac-
tion profile exhibits a pseudo first-order reaction feature.
Meanwhile, no such accumulation of N-BBA is observed, indi-
cating that N-BBA is not an intermediate in the reaction.
To understand the surface reaction pathway, we investi-
gated the surface adsorption of the reactant and the reaction
process by in situ Fourier transform infrared (FT-IR) spectra
(Fig. 3d). When Ru/γ-Al2O3 was exposed to vaporous
benzylamine at room temperature, the intensities of bands at
strating the formation and adsorption of primary imine inter-
mediates on the surface of Ru/γ-Al2O3. The formation of Ru–
amide on the surface of Ru/γ-Al2O3 and the production of the
primary imines under light illumination were observed by in
situ FT-IR spectroscopy, which verify our hypothesis and pro-
vide sufficient evidence for demonstrating the mechanism.
To gain more insight into the reaction mechanism of
the photocatalytic system, we studied the compatibility of
Ru/γ-Al2O3 with different substituted aromatic and aliphatic
amines under the optimised reaction conditions. Such a
photocatalytic system can be successfully applied to aro-
matic and aliphatic amines with good to excellent conver-
sions (Table 2). Electron-drawing substrates such as
4-fluorobenzylamine gave 4-fluorobenzonitrile in 95% con-
version and 4-chlorobenzylamine in 97% conversion, both
of which were suitable substrates (entries 2 and 3, Table 2).
Electron-donating substrates were also studied under identi-
cal reaction conditions; 4-methylbenzylamine and 4-methoxy-
benzylamine were converted to the corresponding nitriles
with 98% and 99% conversions, respectively (entries 4 and 5,
Table 2). Not only aromatic amines, but also less reactive ali-
phatic amines could be converted to the corresponding nitriles
(entries 6 and 7, Table 2). Interestingly, the oxidation of pri-
mary aliphatic amines can be achieved with a better conver-
sion rate and selectivity than aromatic substrates, which
differs from the other reaction systems. This phenomenon in-
dicates that the formation of Ru–amide complexes is benefi-
cial to the activation of aliphatic amines. Moreover, it has
been reported that N radicals are generated during the
reaction process,31 and the Ru–amide complexes may play an
important role in stabilizing the adjacent
N
radical.
N-Benzylidenebenzylamine was converted to benzonitrile as
the main product (entry 8, Table 2). Secondary amines are suitable
for our system (entries
9
and 10, Table 2), and
N-methylbenzylamine can be transformed to corresponding
N-benzylidenemethanamine as the main product. Dibenzylamine
afforded the corresponding N-benzylidenebenzylamine as the
main product, together with benzaldehyde (32%), benzo-
nitrile (14%), and benzylamine (2%). This demonstrates that
the reactant will not remain as imine and can be converted
to benzonitrile over the Ru/γ-Al2O3 photocatalyst owing to the
adsorption of imine and the formation of Ru–amide com-
plexes. Because the secondary dehydrogenation of these im-
ines is more difficult, the reaction rate is slower than that of
primary amines.
It is reported that radicals were not involved in the oxida-
tion of primary amines.22 In this study, 1,4-benzoquinone as
a superoxide scavenger was added to the reaction to confirm
the presence of O2˙− radicals (Fig. 4a).55 The yield of benzo-
nitrile gradually decreased to 0.8% with the increase of the
amount of 1,4-benzoquinone, indicating the presence of O2˙−
radicals. In order to investigate whether the Ru/γ-Al2O3 photo-
catalyst can mediate electron transfer from electron-donating
amines to oxygen to generate O2˙− radicals under light irradia-
tion, 1,4-bisIJdimethylamino)benzene was employed as the indi-
cator.51,56 As shown in Fig. 4b, blue coloured cationic radical
3333 and 3256 cm−1 (NH2 stretching vibration, νNH ), 1391
2
cm−1 (CH2 bending vibration, ωCH ), and 1271 cm−1 (C–N
2
stretching vibration, νC–N) were observed and gradually in-
creased (Fig. 3d, BA adsorption 15, 35, and 55 min).54 After
flowing N2 to the system (Fig. 3d, flowing N2 6, 15, and 30
min), the bands related to νNH (3333 and 3256 cm−1) gradu-
2
ally disappeared with time, while a peak appeared at 1649
cm−1 in the region of the N–H bending vibration (δN–H).
These results indicate the dissociative adsorption of
benzylamine on the surface of the photocatalyst to form Ru–
amide species, which coincided with the results of the XPS
and DR UV-vis spectra (Fig. 3a and b). After flowing O2 to
the system in the dark, no obvious changes were observed
(Fig. 3, dark). When light was introduced to the system with
flowing O2 (Fig. 3, light), the intensity of the C–N stretching
vibration (νC–N) at 1271 cm−1 gradually decreased until it
disappeared, implying the consumption of the reactant.
Meanwhile, the band at 2272 cm−1 indicates the formation
of benzonitrile. And two bands appeared at 3277 cm−1 (N–H
stretching vibration of primary imine, νN–H) and 1650 cm−1
(CN stretching vibration of primary imine, νCN),54 demon-
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 2 Photocatalytic oxidation of various aminesa
Entry
1
Substrate
Product
t (h)
Conv.b (%)
99 (39)
Sel.b (%)
86 (84)
3
2
3
4
5
6
6
6
4
95 (52)
97 (58)
98 (51)
99 (42)
86 (83)
72 (45)
80 (75)
82 (75)
6c
7
8d
6
3
88 (38)
91 (24)
96 (33)
93 (93)
93 (93)
63 (47)
24
9e
24
36
89 (19)
48 (7)
45 (30)
85 (71)
10f
a
Reaction conditions: amine (0.2 mmol), Ru/γ-Al2O3 (40 mg), benzotrifluoride (1 mL), O2 (1 atm), 33 °C, xenon lamp and light intensity 0.87 W
cm−2
.
Conversion and selectivity were determined by gas chromatography (GC); the values in parentheses are the results in the dark. Solvent
b
c
d
e
was acetonitrile. N-Benzylidenebenzylamine (0.1 mmol), other products: benzaldehyde (28%) and benzylamine (8%). Dibenzylamine (0.1
mmol), other products: benzaldehyde (32%), benzonitrile (14%) and benzylamine (2%). f Other products: benzaldehyde (4%), benzonitrile (3%).
species of 1,4-bisIJdimethylamino)benzene were obtained in the
presence of the Ru/γ-Al2O3 photocatalyst and the obvious ab-
sorption bands can be observed in the UV-vis spectra. These re-
sults reveal the presence of O2˙− radicals in the photocatalytic
oxidation of primary amines and the Ru/γ-Al2O3 photocatalyst
can mediate the electron transfer from amines to oxygen under
light irradiation.
The photocurrent response of Ru/γ-Al2O3 before and after
benzylamine adsorption was compared under intermittent
light irradiation to evaluate the separation ability of photo-
electron–hole pairs. As shown in Fig. S10,† a photocurrent re-
sponse was generated immediately when light irradiation was
switched on, and then decreased quickly to zero when light
irradiation was switched off. The Ru/γ-Al2O3 adsorbed with
benzylamine displayed higher photocurrent density than
Ru/γ-Al2O3, indicating the formation of a surface Ru–amide
complex can effectively separate photoelectron–hole pairs
and extend the electron–hole distances by electron transfer
from nitrogen to Ru thereby restraining charge
recombination.57,58
different conditions with xenon light irradiation. When the
reaction was conducted in the presence of Ru/γ-Al2O3 and
oxygen, a dark blue solution was obtained (Fig. S11a†) and
the obvious absorption bands can be observed in the UV-vis
spectra (Fig. S11b†). A light blue solution was obtained in
the absence of oxygen (the reaction system was degassed by
three freeze–pump–thaw cycles) and the intensity of the ab-
sorption bands in the UV-vis spectra was much weaker than
that in the oxygen atmosphere, indicating that more blue
coloured cationic radical species of 1,4-bisIJdimethylamino)-
benzene were produced in the presence of oxygen. For com-
parison, control experiments were carried out in the ab-
sence of Ru/γ-Al2O3 under different atmospheres. No
obvious change of the solution and no corresponding ab-
sorption bands in the UV-vis spectra were observed. These
facts reveal the electron transfer from nitrogen to Ru in the
Ru–amide complex and the formation of the corresponding
nitrogen radicals. When the reaction was conducted in the
presence of oxygen under light illumination, the electrons
derived from nitrogen were trapped by RuIJIII) to generate
RuIJII), then the oxygen can be converted to O2˙− by RuIJII).
The reaction can proceed continually and produce a large
number of nitrogen radicals in the presence of oxygen.
However, no electron transferred from RuIJII) to oxygen in
the argon atmosphere, which leads to the formation of the
electron-rich Ru species and no sufficient RuIJIII) can trap
the electrons derived from nitrogen so only a small amount
of nitrogen radicals was obtained.
To further prove the electron transfer between Ru and
amines in the Ru–amide complex under light irradiation,
1,4-bisIJdimethylamino)benzene was employed as the sub-
strate because the blue coloured cationic radical species of
1,4-bisIJdimethylamino)benzene can indicate the occurrence
of the electron transfer.51,56 Fig. S11† shows the photo-
graphs and UV-vis absorption spectra of the cationic radical
species of 1,4-bisIJdimethylamino)benzene generated under
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
allows the high dispersion of hydroxyl coordinated Ru spe-
cies, which results in a strong interaction between the Ru
centre and the support. The abundant surface OH groups
and active Ru sites subsequently influence the reaction path-
way of surface adsorbed amines. The adsorption of reactants
and intermediates and the formation of Ru–amide complexes
on the surface of the photocatalyst under light irradiation
were identified by the detailed investigation of the reaction
mechanism, which ensure that the primary amines can be di-
rectly converted to corresponding nitriles with good to excel-
lent yields via a two-step oxidative dehydrogenation process.
It is expected that this method can be applied for designing a
new generation of heterogeneous catalysts.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
Fig. 4 Reaction mechanism over the Ru/γ-Al2O3 photocatalyst. (a)
Influence of 1,4-benzoquinone on the photocatalytic oxidation of
benzylamine. (b) UV-vis absorption spectra and photograph of the cat-
ionic radical species of 1,4-bisIJdimethylamino)benzene generated by
Ru/γ-Al2O3 using a xenon lamp in the presence of oxygen. (c) Pro-
posed mechanism for the photocatalytic oxidation of benzylamine on
the Ru/γ-Al2O3 photocatalyst.
We appreciate the financial support from the National Natu-
ral Science Foundation of China (No. 21773284, 21603259)
and the Hundred Talents Program of the Chinese Academy of
Sciences and Shanxi Province.
Notes and references
1 F. F. Fleming, L. Yao, P. C. Ravikumar, L. Funk and B. C.
Shook, J. Med. Chem., 2010, 53, 7902–7917.
2 F. F. Fleming, Nat. Prod. Rep., 1999, 16, 597–606.
3 J. S. Miller and J. L. Manson, Acc. Chem. Res., 2001, 34,
563–570.
4 T. Sandmeyer, Ber. Dtsch. Chem. Ges., 1884, 17, 1633–1635.
5 K. W. Rosenmund and E. Struck, Ber. Dtsch. Chem. Ges.,
1919, 2, 1749–1756.
6 P. Anbarasan, T. Schareina and M. Beller, Chem. Soc. Rev.,
2011, 40, 5049–5067.
7 K. Rajender Reddy, C. Uma Maheswari, M. Venkateshwar, S.
Prashanthi and M. Lakshmi Kantam, Tetrahedron Lett.,
2009, 50, 2050–2053.
8 S. Iida and H. Togo, Tetrahedron, 2007, 63, 8274–8281.
9 M. Lamani and K. R. Prabhu, Angew. Chem., Int. Ed.,
2010, 49, 6622–6625.
10 W. Zhou, L. Zhang and N. Jiao, Angew. Chem., Int. Ed.,
2009, 48, 7094–7097.
11 W. R. McWhinnie, J. D. Miller, J. B. Watts and D. Y. Waddan,
J. Chem. Soc. D, 1971, 2.
12 R. Tang, S. E. Diamond, N. Neary and F. Mares, J. Chem.
Soc., Chem. Commun., 1978, 562–562.
Based on the above analysis, we can suggest a reasonable
mechanism for this reaction (Fig. 4c). At first, the primary
amine is dissociatively adsorbed on the surface of Ru/γ-Al2O3
to form the Ru–amide surface complex (step 1). When the
surface complex is irradiated under light, the photogenerated
electrons derived from the N 2p orbital of the adsorbed
amine species can be trapped by RuIJIII) to generate RuIJII)
(step 2). The adsorbed O2 can be activated by RuIJII) to
generate O2˙−,59 which can be adsorbed on the surface of the
photocatalyst to form an intermediate (intermediate A)
possessing a six-member-ring structure (step 3). The cleavage
of the C–H bond and the Ru–N bond of intermediate A oc-
curs to form the corresponding primary imines and ruthe-
nium peroxide species (step 4). The primary imines adsorb
on the surface OH groups by the hydrogen bond and can fur-
ther be oxidised to nitriles in the same way (steps 5–8). How-
ever, some primary imines can escape from the surface of the
photocatalyst and react with primary amines to generate the
side product secondary imines (Scheme S1†). Finally, the
photocatalytic cycle is completed by the regeneration of Ru
sites on the surface of the photocatalyst through desorption
of oxygen (step 9).
13 S. Murahashi, Y. Okano, H. Sato, T. Nakae and N. Komiya,
Synlett, 2007, 11, 1675–1678.
14 A. J. Bailey and B. R. James, Chem. Commun.,
1996, 2343–2344.
Conclusions
In summary, we have developed a light-induced green path-
way for the synthesis of nitriles from oxidative dehydrogena-
tion of primary amines under mild conditions over a γ-Al2O3
supported Ru photocatalyst. The large surface area of γ-Al2O3
15 F. Porta, C. Crotti, S. Cenini and G. Palmisano, J. Mol. Catal.,
1989, 50, 333–341.
16 S. Cenini, F. Porta and M. Pizzotti, J. Mol. Catal., 1982, 15,
297–308.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
17 L. Cristian, S. Nica, O. D. Pavel, C. Mihailciuc, V. Almasan,
S. M. Coman, C. Hardacre and V. I. Parvulescu, Catal. Sci.
Technol., 2013, 3, 2646.
18 R. Ray, S. Chandra, V. Yadav, P. Mondal, D. Maiti and G. K.
Lahiri, Chem. Commun., 2017, 53, 4006–4009.
19 K. N. Tseng, A. M. Rizzi and N. K. Szymczak, J. Am. Chem.
Soc., 2013, 135, 16352–16355.
20 I. Dutta, S. Yadav, A. Sarbajna, S. De, M. Holscher, W. Leitner
and J. K. Bera, J. Am. Chem. Soc., 2018, 140, 8662–8666.
21 K. Mori, K. Yamaguchi, T. Mizugaki, K. Ebitani and K.
Kaneda, Chem. Commun., 2001, 461–462.
22 K. Yamaguchi and N. Mizuno, Angew. Chem., Int. Ed.,
2003, 42, 1479–1483.
38 J. Shi, H. Hao and X. Lang, Sustainable Energy Fuels, 2019, 3,
488–498.
39 D. Sun, L. Ye and Z. Li, Appl. Catal., B, 2015, 164, 428–432.
40 R. Chen, J. Shi, Y. Ma, G. Lin, X. Lang and C. Wang, Angew.
Chem., Int. Ed., 2019, 58, 6430–6434.
41 Z. Zheng, C. Chen, A. Bo, F. S. Zavahir, E. R. Waclawik, J.
Zhao, D. Yang and H. Zhu, ChemCatChem, 2014, 6,
1210–1214.
42 S. Sarina, H. Zhu, E. Jaatinen, Q. Xiao, H. Liu, J. Jia, C. Chen
and J. Zhao, J. Am. Chem. Soc., 2013, 135, 5793–5801.
43 S. Naya, K. Kimura and H. Tada, ACS Catal., 2012, 3, 10–13.
44 D. S. Ovoshchnikov, B. G. Donoeva and V. B. Golovko, ACS
Catal., 2015, 5, 34–38.
23 M. Kotani, T. Koike, K. Yamaguchi and N. Mizuno, Green
Chem., 2006, 8, 735.
24 F. Li, J. Chen, Q. Zhang and Y. Wang, Green Chem., 2008, 10,
553.
25 Y. Zhang, K. Xu, X. Chen, T. Hu, Y. Yu, J. Zhang and J.
Huang, Catal. Commun., 2010, 11, 951–954.
26 O. D. Pavel, P. Goodrich, L. Cristian, S. M. Coman, V. I.
Pârvulescu and C. Hardacre, Catal. Sci. Technol., 2015, 5,
2696–2704.
45 D. J. Morgan, Surf. Interface Anal., 2015, 47, 1072–1079.
46 Z. Wang, S. M. Xu, Y. Xu, L. Tan, X. Wang, Y. Zhao, H. Duan
and Y. F. Song, Chem. Sci., 2019, 10, 378–384.
47 K. Mori, T. Taga and H. Yamashita, ACS Catal., 2017, 7,
3147–3151.
48 Y. Wang, P. Ren, X. Gu, X. Wen, Y. Wang, X. Guo, E. R.
Waclawik, H. Zhu and Z. Zheng, Green Chem., 2016, 18,
1482–1487.
49 F. W. Bubb, Phys. Rev., 1924, 23, 137–143.
50 W. Qi, W. Liu, B. Zhang, X. Gu, X. Guo and D. Su, Angew.
Chem., Int. Ed., 2013, 52, 14224–14228.
27 Z. Shi, C. Zhang, C. Tang and N. Jiao, Chem. Soc. Rev.,
2012, 41, 3381–3430.
28 X. Lang, H. Ji, C. Chen, W. Ma and J. Zhao, Angew. Chem.,
Int. Ed., 2011, 50, 3934–3937.
51 Z. Wang, S. Ghasimi, K. Landfester and K. A. Zhang, Adv.
Mater., 2015, 27, 6265–6270.
29 X. Lang, W. Ma, Y. Zhao, C. Chen, H. Ji and J. Zhao, Chem. –
Eur. J., 2012, 18, 2624–2631.
30 S. Furukawa, Y. Ohno, T. Shishido, K. Teramura and T.
Tanaka, ACS Catal., 2011, 1, 1150–1153.
31 S. Furukawa, Y. Ohno, T. Shishido, K. Teramura and T.
Tanaka, J. Phys. Chem. C, 2013, 117, 442–450.
32 W. Zhao, C. Liu, L. Cao, X. Yin, H. Xu and B. Zhang, RSC
Adv., 2013, 3, 22944.
52 S. Zavahir, Q. Xiao, S. Sarina, J. Zhao, S. Bottle, M. Wellard,
J. Jia, L. Jing, Y. Huang, J. P. Blinco, H. Wu and H. Zhu, ACS
Catal., 2016, 6, 3580–3588.
53 N. H. Damrauer and J. K. McCusker, J. Phys. Chem. A,
1999, 103, 8440–8446.
54 R. M. Silverstein, F. X. Webster, D. Kiemle and D. L. Bryce,
Spectrometric identification of organic compounds, John Wiley
& Sons, 7th edn, 2005.
33 F. Su, S. C. Mathew, L. Mohlmann, M. Antonietti, X. Wang
and S. Blechert, Angew. Chem., Int. Ed., 2011, 50, 657–660.
34 X. Yang, B. Chen, X. Li, L. Zheng, L. Wu and C. Tung, Chem.
Commun., 2014, 50, 6664–6667.
55 J. Kou, Z. Li, Y. Yuan, H. Zhang, Y. Wang and Z. Zou,
Environ. Sci. Technol., 2009, 43, 2919–2924.
56 C. Su, R. Tandiana, B. Tian, A. Sengupta, W. Tang, J. Su and
K. P. Loh, ACS Catal., 2016, 6, 3594–3599.
35 Y. Wu, B. Yuan, M. Li, W. Zhang, Y. Liu and C. Li, Chem.
Sci., 2015, 6, 1873–1878.
36 B. Yuan, R. Chong, B. Zhang, J. Li, Y. Liu and C. Li, Chem.
Commun., 2014, 50, 15593–15596.
37 J. Shi, H. Hao, X. Li and X. Lang, Catal. Sci. Technol., 2018, 8,
3910–3917.
57 V. Sharma, M. Prasad, P. Ilaiyaraja, C. Sudakar and S. Jadkar,
Int. J. Hydrogen Energy, 2019, 44, 11459–11471.
58 B. Shan, S. Vanka, T. Li, T. G. Ludovic, M. K. Brennaman, Z.
Mi and T. J. Meyer, Nat. Energy, 2019, 4, 290–299.
59 S. E. Diamond, G. M. Tom and H. Taube, J. Am. Chem. Soc.,
1975, 97, 2661–2664.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019