One-pot synthesis of secondary amine via photoalkylation of nitroarenes with benzyl alcohol over Pd/monolayer H1.07Ti1.73O4·H2O nanosheets

Article abstract of DOI:10.1016/j.jcat.2018.02.005

The photoalkylation of nitroarenes with benzyl alcohols in one pot at room temperature and 1 atm N₂ was achieved over the Pd/H_1.07Ti_1.73O₄·H₂O nanosheets. The sample shows efficient photocatalytic activity with high conversion of nitrobenzene (99%) and selectivity of secondary amine (85%). This flexible photocatalytic system is also applicable to other nitroarenes with high efficiency. Results of in situ FTIR, DRS, and in situ ESR revealed that the benzyl alcohol and nitrobenzene molecules can bind with the surface Lewis and Br?nsted acid sites in the catalyst via the H–O?Ti and NO₂?H–O–Ti species. The formation of surface coordination species results in not only the activation of reactant molecules via surface electron transfer, but also the expanded visible light absorption of the catalyst. Moreover, in situ ESR suggested that the surface coordination can also facilitate the formation of oxygen vacancies in catalysts, which can greatly promote the exposure of Lewis sites and enhance the activation of reactant molecules. Finally, a possible hydrogen transfer strategy over the sample is proposed on a molecular level.

Full text of DOI:10.1016/j.jcat.2018.02.005

Journal of Catalysis 361 (2018) 105–115
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
One-pot synthesis of secondary amine via photoalkylation of nitroarenes
with benzyl alcohol over Pd/monolayer H1.07Ti1.73O4ꢀH2O nanosheets
Yujie Song a, Hao Wang a, Shijing Liang a,b, Yan Yu c, Liuyi Li c, Ling Wu a,
⇑
a State Key Laboratory of Photocatalysis on Energy and Environment, Fuzhou University, Fuzhou 350116, People’s Republic of China
b National Engineering Research Center of Chemical Fertilizer Catalyst, College of Chemical Engineering, Yishan Campus, Fuzhou University, Fuzhou 350002, People’s Republic of China
c Key Laboratory of Ecomaterials Advanced Technology, Fuzhou University, Fuzhou 350008, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
The photoalkylation of nitroarenes with benzyl alcohols in one pot at room temperature and 1 atm N2
was achieved over the Pd/H1.07Ti1.73O4ꢀH2O nanosheets. The sample shows efficient photocatalytic activ-
ity with high conversion of nitrobenzene (99%) and selectivity of secondary amine (85%). This flexible
photocatalytic system is also applicable to other nitroarenes with high efficiency. Results of in situ
FTIR, DRS, and in situ ESR revealed that the benzyl alcohol and nitrobenzene molecules can bind with
the surface Lewis and Brønsted acid sites in the catalyst via the HAOꢀ ꢀ ꢀTi and NO2ꢀ ꢀ ꢀHAOATi species.
The formation of surface coordination species results in not only the activation of reactant molecules
via surface electron transfer, but also the expanded visible light absorption of the catalyst. Moreover,
in situ ESR suggested that the surface coordination can also facilitate the formation of oxygen vacancies
in catalysts, which can greatly promote the exposure of Lewis sites and enhance the activation of reactant
molecules. Finally, a possible hydrogen transfer strategy over the sample is proposed on a molecular
level.
Received 11 December 2017
Revised 6 February 2018
Accepted 7 February 2018
Keywords:
Surface coordination
Heterogeneous photocatalysts
Secondary amines
Cascade reactions
Hydrogen transfer
Multifunctional catalysis
Ó 2018 Elsevier Inc. All rights reserved.
1. Introduction
catalyst is needed for the synthesis of secondary amines from
nitroarenes with alcohols in one-pot multistep reactions under
Tandem catalysis that enables multistep reactions in one pot
has attracted great attention because it avoids time-consuming,
tedious intermediate separation processes, energy consumption,
and waste of production [1,2]. Tandem reactions are widely used
in various organic transformations [3], especially the production
of secondary amines, because secondary amines and their deriva-
tives are extremely important intermediates for synthesis of phar-
maceuticals, agrochemicals, and polymers [2–5]. An attractive
route for synthesis of secondary amines is the tandem alkylation
of amines in one pot with an alcohol as the alkylating agent [6–
8]. Many heterogeneous catalysts have been developed for the
alkylation of amines with alcohols, such as CuAMo/TiO2 [9], Pd-
ligand [10], Pd/MgO [11], IrAZr MOFs [12], Pd/C [13], and Ru
(OH)x/Al2O3 [14]. However, as typical anilines precursors, the syn-
thesis of secondary amines from nitroarenes still faces great chal-
lenges. Very recently, PdAg@MIL-101 [15], Au/TiO2, Pd/TiO2, and
Pt/TiO2 [16] have been used as catalysts for the tandem alkylation
of amines from nitroarenes. But most of these catalyses were used
under hydrogen or at elevated temperature. Therefore, an efficient
mild conditions. This is desirable for green chemistry.
The hydrogenation of nitroarenes to primary amines is the chal-
lenging step in such a one- pot multistep reaction under mild con-
ditions. Several reductive methods have been developed to
hydrogenate nitroarenes to amines, such as photocatalysis [17]
and electrolytic reduction [18]. Photocatalysis has been widely
used in the hydrogenation of nitroarenes. In our previous studies,
we have developed a series of photocatalysts for the hydrogenation
of nitrobenzene and its derivatives to corresponding anilines in
water at room temperature and under 1 atm N2 [19]. In these pho-
tocatalytic systems, (NH4)2C2O4 or HCOONH4 was added as a hole
scavenger to generate the hydrogen for the reaction. If the photo-
generated holes can be consumed by alcohols, producing the corre-
sponding aldehydes, the produced anilines and aldehydes may
react to form the secondary amines or imines. How can an appro-
priate photocatalyst with the strong redox capacities be rationally
designed to catalyze coupled alcohol oxidation and nitroarene
reduction in such a one-pot multistep reaction?
Recently, Higashimoto and co-workers reported the photosyn-
thesis of imines from benzyl alcohol and nitrobenzene in one pot
over TiO2-based materials [20]. This would provide the possibility
of photoalkylation of nitroarenes with benzyl alcohols to produce
⇑
Corresponding author.
E-mail address: wuling@fzu.edu.cn (L. Wu).
https://doi.org/10.1016/j.jcat.2018.02.005
0021-9517/Ó 2018 Elsevier Inc. All rights reserved.

106
Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
secondary amines. It is reported that the Lewis acid sites in the cat-
alyst can facilitate the reaction between the formed aldehyde and
the primary amine to form a secondary amine [20,21]. Further-
more, the Lewis acid sites have been proved to promote the con-
version of benzyl alcohol to benzaldehyde [22]. Thus, efficiently
exposed Lewis acid sites could be considered in the design of cat-
alysts. Recently, we prepared nanosheets of the lepidocrocite-type
monolayer H1.07Ti1.73O4ꢀH2O (Ti-NS). Ti-NS, with a band gap of 3.4
eV, shows ae strong redox capacity. The valence (VB) and conduc-
tion (CB) band of Ti-NS are about ꢁ0.88 and 2.42 eV vs. NHE
respectively [23]. Moreover, Ti-NS possesses abundant surface
Lewis and Brønsted acid sites, resulting in the highly efficient pho-
tocatalytic hydrogenation of phenol to cyclohexanone with the
assistance of Pd [24]. Therefore, suitable band structure and abun-
dant surface acid sites make Ti-NS a promising candidate for pho-
tocatalysis of the synthesis of secondary amines from nitroarenes
with alcohols. Based on this, we developed Pd/H1.07Ti1.73O4ꢀH2O
(Pd/Ti-NS) as a photocatalyst for the photo-alkylation of nitroare-
nes with benzyl alcohol in one pot at room temperature and under
1 atm N2 (Scheme 1). High conversion of nitroarenes (99%) and
selectivity of secondary amines (85%) were achieved over the Pd/
NS photocatalyst. Influences on photocatalytic activity have been
discussed in detail on in relation to structures, morphologies,
superficial properties, and optical properties through XRD, SEM,
TEM, XPS, and UV-DRS characterization. The coordination behavior
of reagent molecules on superficial Lewis and Brønsted acid sites of
catalysts has been revealed by in situ FTIR, in situ ESR, and N2
adsorption techniques. Furthermore, interfacial electron transfer
from reagent molecules to catalysts via surface coordination was
also elucidated by UV-DRS, in situ FTIR, and in situ ESR. Finally, a
possible photocatalytic mechanism was proposed on a molecular
level, which is distinct from that of classic catalysis. This work
not only highlights the photocatalytic synthesis of secondary ami-
nes from nitroarenes and alcohols, but also may spark the design
and construction of efficient photocatalysts for organic synthesis
in industry.
Inc., 99%), 4-nitrophenol (Aladdin Industrial Inc., AR), 3-
nitrophenol (Aladdin Industrial Inc., 99%), 2-nitrophenol (Aladdin
Industrial Inc., 99%), 4-nitroanisole (Aladdin Industrial Inc., 98%),
1-chloro-4-nitrobenzene (Aladdin Industrial Inc., CP, 98%), 1-
bromo-4-nitrobenzene (Aladdin Industrial Inc., GC, >99%), and ben-
zotrifluoride (Aladdin Industrial Inc., 99%) were also used. Deion-
ized (DI) water (ꢂ18.2
MX cm) was used throughout the
experiments.
2.2. Preparation of layered H1.07Ti1.73O4ꢀH2O (Ti-L) and
H1.07Ti1.73O4ꢀH2O nanosheets (Ti-NS)
The preparation of Ti-L and Ti-NS was carried out via a high
temperature solid state reaction [23,24]. The Ti-L was synthesized
by the proton exchange process from K0.8Ti1.73Li0.27O4. The Ti-NS
was obtained through the intercalation and exfoliation process
with TBAOH as the intercalation agent. A mole ratio of Ti-L and
TBAOH of 1:1.2 was stirred for more than 4 days at ambient tem-
perature. A white Ti-NS colloidal solution was obtained. Next, 1
M HCl solution was added to the colloidal solution for the floccula-
tion of Ti-NS, followed by centrifugal separation and washing with
deionized water. After a drying process at 60 °C, the white powder
of Ti-NS was successfully prepared.
2.3. Preparation of Pd/Ti-NS
The metal particles loaded on the surfaces of samples were pre-
pared through photodeposition. An aqueous solution of Pd(NO3)2-
ꢀ2H2O (10 mg/mL) was injected into a colloidal solution of Ti-NS,
and 5 mL methanol was added as a sacrificial agent. After being
strongly stirred for 30 min with N2 as protective atmosphere, the
suspensions were irradiated with a Xenon lamp for 4 h. For com-
parison, the mixed suspensions were irradiated 6 h to obtain a dif-
ferent size of Pd nanoparticles. The products were collected and
washed with water several times, and then dried at 60 °C for fur-
ther characterization and use.
2.4. Materials characterization
2. Experimental
X-ray diffraction (XRD) patterns were collected on a Bruker D8
2.1. Reagents and chemicals
Advance X-ray diffractometer with CuKa radiation (k = 0.15406
nm). The UV–vis diffuse reflectance spectra (UV–vis DRS) were
obtained for the dry-pressed disk samples using a Cary 500 Scan
UV–vis spectrophotometer (Varian). The Brunauer–Emmett–Teller
(BET) specific surface areas of the samples were measured by nitro-
gen adsorption/desorption isotherms at 77 K on an Autosorb-1C-
TCD physical adsorption instrument (American Quantachrome)
using a Micrometrics ASAP 2020 system. Before being analyzed,
the samples were degassed under vacuum at 180 °C for 10 h.
Transmission electron micrographs (TEM) were obtained using a
JEOL JEM-1010 electron microscope operated at an accelerating
voltage of 200 kV. Scanning electron micrographs (SEM) were
obtained using an FEI Quanta 200F electron microscope. X-ray pho-
toelectron spectroscopy (XPS) analysis was measured on an ESCA-
LAB 250 photoelectron spectroscopy (Thermo Fisher Scientific Inc.)
Reagents such as TiO2, metal carbonates, and tetrabutylammo-
nium hydroxide (TBAOH; 40 wt% solution) used in this work were
of analytical grade and without further purification. Methanol
(CH3OH, Sinopharm Chemical Reagent Co., >99%), nitrobenzene
(Sinopharm Chemical Reagent Co., CP), benzyl alcohol (Aladdin
Industrial Inc., AR), 4-chlorobenzyl alcohol (Alfa Aesar, 99%), 4-
bromobenzyl alcohol (Aladdin Industrial Inc., 99%), 4-
methylbenzyl alcohol (Alfa Aesar, 99%), 4-methoxybenzyl alcohol
(Aladdin Industrial Inc., 98%), 4-nitrotoluene (Aladdin Industrial
at 3.0 ꢃ 10ꢁ10 mbar with monochromatic AlK
a Radiation (E =
1486.2 eV). A tapping-mode atomic force microscope (AFM, Agli-
ent 5500) with Si-tip cantilever was used to evaluate the morphol-
ogy of the prepared nanosheets on the mica substrate. A 300 W Xe
lamp (Beijing Trustech, PLS-SXE300c) was used as a light source.
2.5. Simulated in situ FTIR measurement
The in situ FTIR spectra of nitrobenzene and benzyl alcohol
adsorbed onto the samples were determined on a Nicolet Nexus
670 Fourier transform infrared (FT-IR) spectrometer at a resolution
Scheme 1. Sketch map of photocatalytic N-alkylation from nitrobenzene and
benzyl alcohol.

Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
107
of 4 cmꢁ1. A total of 64 scans were performed to acquire each spec-
Conversion ð%Þ ¼ ½ðC0 ꢁ CnitroarenesÞ=C0ꢄ ꢃ 100;
Selectivityð%Þ ¼ ½Camines=ðC0 ꢁ CnitroarenesÞꢄ ꢃ 100:
trum. First the powder samples were pressed into a self-supporting
IR disk (18 mm diameter, 15 mg); then the disk was placed in the
sample holder which could be moved vertically along the cell’s
tube by a magnet. Before the in situ FTIR measurements were ini-
tiated, the disk was treated under dynamic vacuum (6 ꢃ 10ꢁ4 Torr)
at 150 °C for 3 h to remove surface contaminants. After the disk
C0 represents the initial concentration of nitroarenes, and Cnitroarenes
and Camines are the concentrations of the substrate nitroarenes and
the corresponding amines, respectively.
cooled to ambient temperature, 10 ll of nitrobenzene or benzyl
alcohol was spiked into the cell with a syringe via the septum.
After 30 min, when adsorption equilibrium had been reached, the
in situ FTIR spectra of the samples were collected. The physisorbed
benzyl alcohol and nitrobenzene were removed by further evacua-
tion under 6 ꢃ 10ꢁ4 Torr at 150 °C for 3 min. Another in situ FTIR
spectrum of the sample was then taken.
3. Results and discussion
With our designed photocatalyst and expected experimental
process, we began our investigation to explore the alkylation of
nitroarenes with benzyl alcohols in one pot. The dehydrogenation
of benzyl alcohol to benzaldehyde at room temperature and 1 atm
O2 was chosen as the first step reaction (Table 1). No product was
detected in the absence of catalyst under light irradiation and con-
trolled temperature at 298 K (Table 1, entry 1). The Ti-NS shows
the high activity, chemoselectivity, and versatility for the dehydro-
genation of benzyl alcohol (Table 1, entries 2–7) under the same
conditions. The conversion of benzyl alcohol reaches 40% under
light irradiation (k ꢅ 700 nm) with 99% selectivity of benzaldehyde
(Table 1, entry 2). It should be noted that 24% conversion of benzyl
alcohol with 99% selectivity of benzaldehyde was achieved over
the Ti-NS under visible light irradiation (Table 1, entry 3), despite
Ti-NS and alcohols having no visible light adsorption. It would be
proposed that some intermediates with visible light adsorption
may be formed, as stated in the previous reports on TiO2 [21]. Thus,
the interactions between benzyl alcohol and prepared samples
would be further revealed in subsequent experiments. Moreover,
the Ti-NS can also catalyze the derivatives of benzyl alcohol to
form the corresponding benzyl aldehydes under visible light irradi-
ation and room temperature (Table 1, entries 4–7). Additionally,
the thermocatalysis of Ti-NS for selective oxidation of benzyl alco-
hol has been also explored. To elucidate the catalysis more clearly,
a set of thermocatalysis experiments were carried out. As shown in
Table 1, entries 8 and 9, the conversion of benzyl alcohol was neg-
ligible even when it was heated to 353 K under darkness. These
experimental results suggested that the oxidation of benzyl alcohol
over the Ti-NS is mainly driven by light energy. These encouraging
results indicated that the Ti-NS photocatalyst has great possibili-
ties in photocatalytic one-pot tandem reaction of alcohols and
nitroarenes.
However, the layered counterpart, Ti-L, shows a much lower
conversion rate for the photocatalytic dehydrogenation of benzyl
alcohols to corresponding aromatic aldehydes (Table S1 in the Sup-
porting Information). For example, only 2% of benzyl alcohol was
transformed to benzaldehyde under UV light irradiation, and the
conversion rate under visible light was lower (Table S1, entry 2).
Moreover, the catalytic activities of Ti-L for the derivatives of ben-
zyl alcohol dehydrogenation are also very low (Table S1, Entries 3–
7). In addition, the conversion of benzyl alcohol is 2.1% over the
TiO2 photocatalyst under UV light irradiation (Table S1, entry
10). These experimental results suggest that the photocatalytic
activities for alcohols dehydrogenation over Ti-NS are greatly pro-
moted. It is feasible to design a photocatalyst using Ti-NS as the
basic material for the one-pot tandem reaction from alcohols and
nitrobenzenes.
2.6. Simulated in situ ESR measurement
Electron spin resonance (in situ ESR) signals were recorded with
a Brucker A300 spectrometer. First, 50 mg powder samples were
pressed into a self-supporting in situ ESR quartz tube. Before the
in situ ESR measurements were initiated, the tube was treated
under dynamic vacuum (6 ꢃ 10ꢁ4 Torr) to remove surface contam-
inants. Then, 10 ll of nitrobenzene or benzyl alcohol was spiked
into the tube with a syringe via the septum. After 30 min, when
adsorption equilibrium was reached, the physisorbed benzyl
alcohol or nitrobenzene was removed by further evacuation under
6 ꢃ 10ꢁ4 Torr. The in situ ESR spectra of the samples were
recorded.
2.7. Temperature-programmed ammonia desorption measurement
Temperature-programmed desorption of ammonia was carried
out on a Micromeritics Auto Chem 2920 instrument. About 50
mg of a sample was first pretreated for 1 h in a quartz tube at
180 °C, and then the sample was cooled to 40 °C to allow the
adsorption of ammonia (3.0% NH3 in N2, 30 mL/min) for 1 h. The
desorption of ammonia was conducted from 40 to 500 °C at a heat-
ing rate of 10 °C/min in a helium flow (40 mL/min), and the desorp-
tion amount was monitored by a thermal conductivity detector
(TCD).
2.8. Photocatalytic activity for N-alkylation of nitroarenes in benzyl
alcohols
A quantity of 20 mg of catalyst was placed in a 20 mL Schrunk
glass bottle (1) which was filled with N2 at a pressure of 1 bar. Ben-
zyl alcohols (5 mmol) and nitroarenes (0.5 mmol) were mixed fully
in a Schrunk glass bottle (2) and nitrogen was bubbled to remove
dissolved oxygen molecules. The mixed solution was transferred
to bottle (1) under 1 atm N2 and stirred to make the catalyst blend
evenly in the solution. The suspensions were irradiated by a 300 W
Xe arc lamp (PLS-SXE300, Beijing Perfect Light Co.), and an IR-cut
filter was used to remove all wavelengths longer than 800 nm.
After the reaction, the mixture was centrifuged to completely
remove the catalyst particles. The remaining solution was deter-
mined by an Agilent online gas chromatograph (GC 6890, FID) with
an HP-5973 mass spectrometer. An HP-5 column (length 30 m;
inner diameter 320 mm; film thickness 0.25 mm) was applied for
separation of product. Helium (purity 99.999%) was used as the
carrier gas at a constant flow rate of 20 mL minꢁ1. The tempera-
tures of the injector and detector were maintained at 280 and
300 °C, respectively. The pressure of injection was set at 8.363
psi. The column temperature was programmed from 40 to 180 °C
at 15 °C/min, and then up to 280 °C at 15 °C/min, and held 5 min.
A set of experiments were carried out to better understand the
catalytic behavior of Pd/Ti-NS for the solvent-free photoalkylation
of benzyl alcohol and nitrobenzene under the simulated sunlight
irradiation and room temperature (Table 2). It is obvious that the
poor yield of desired product was obtained in the absence of Pd
nanoparticles (Table 2, entry 1), suggesting that the Pd plays the
key role in this reaction. Considering the dispensable role of Pd
in the oxidation of benzyl alcohol, they may participate in the
hydrogenation of nitrobenzene and imine. Moreover, Pd nanopar-
The injection volume was 5
ll. Conversion of nitroarenes and
selectivity for secondary amines were as follows:

108
Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
Table 1
Photocatalytic activity for the dehydrogenation of benzyl alcohols over Ti-NS.
Entry
Catalyst
R
Temp. (K)
Light (k)
Conv. (%)
Sel. (%)
1
2
3
4
5
6
7
8
9
–
H
H
H
OCH3
CH3
Cl
F
H
298
298
298
298
298
298
298
333
353
ꢅ700 nm
ꢅ700 nm
ꢆ400 nm
ꢆ400 nm
ꢆ400 nm
ꢆ400 nm
ꢆ400 nm
–
–
–
Ti-NS
Ti-NS
Ti-NS
Ti-NS
Ti-NS
Ti-NS
Ti-NS
Ti-NS
56
24
71
58
32
17
0
>99
>99
90
>99
>99
>99
0
H
–
0
0
Note: Conditions: benzotrifluoride 2 mL, alcohols 0.2 mmol.
Table 2
Photoalkylation of nitrobenzene in benzyl alcohol over Pd/Ti-NS.
Entry
Cat.
Particle size (nm)
Temp. (K)
Light
Con. (%)
Sel. (%) a
1
2
3
4
5
6
7
8
9
Ti-NS
–
7
3
3
–
3
3
3
3
298
298
298
298
298
333
353
298
298
+
+
+
45
91.7
99
–
–
0
0
14
9
7
Pd/Ti-NS
Pd/Ti-NS
Pd/Ti-NS
–
Pd/Ti-NS
Pd/Ti-NS
Pd/Ti-L
Pd/TiO2
77
85
–
–
0
0
5
3
ꢁ
+
ꢁ
ꢁ
+
+
Reaction time: 8 h.
ticles with different sizes were also loaded onto the Ti-NS to study
the influence of Pd size on this reaction. The TEM images suggested
that the Pd nanoparticles with size 7 nm are densely deposited on
the ultrathin Ti-NS (Fig. S2). In addition, the XPS spectra were used
to examine the surface chemical state of Pd. As shown in Fig. S3,
the spectra of Pd3d indicated the metallic Pd of the Ti-NS. It is
observed that the conversion of 99% for nitrobenzene and the yield
of 85% for secondary amines were achieved over the Pd/Ti-NS with
particle size about 3 nm. The gas chromatogram and mass spectro-
gram identified the synthesis of secondary amines (Fig. S1). The
turnover frequency (TOF) and turnover number (TON) here are cal-
culated by the moles of substrate converted and secondary amines
produced per mole of Pd in the whole catalyst per hour. Thus, the
TOF of the Ti-NS for the conversion of nitrobenzene is 33.3 hꢁ1, and
the TON of secondary amines has been calculated as 226.1 in 8 h.
The Pd/Ti-NS with 7 nm shows decreased activity compared with
that with 3 nm, suggesting that the Pd particle has a great effect
on the conversion efficiency of secondary amines. The reason
may be ascribed to the size effects of Pd that closely relate to elec-
tronic properties, exposed active sites, and facets [25]. Considering
that the influence of Pd nanoparticles on reactivity is very compli-
cated, the research on geometric effects of Pd is expected to be
revealed in further study. To gain more information on the one-
pot photoalkylation reaction, the thermocatalysis experiments
were studied. Under dark conditions or in the absence of photocat-
alyst, no desired product was detected (Table 2, entries 4 and 5).
Additionally, even when the reaction temperature was increased
to 353 K, there were no secondary amines in the thermocatalytic
system, indicating that the one-pot coupling reaction of benzyl
alcohol and nitrobenzene over the Pd/Ti-NS was a photocatalytic
process. To elucidate the roles of Ti-NS, different supports were
used in this one-pot reaction. Compared with Pd/Ti-NS, Pd/Ti-L
showed much lower photocatalytic activity for the photoalkylation
of benzyl alcohol with nitrobenzene under the same conditions.
Moreover, when TiO2 was used as the support, greatly decreased
photocatalytic activities were observed over the Pd/TiO2. The
results showed that the properties of supports can greatly affect
the catalytic activities. These experimental results revealed that
Ti-NS would greatly improve the adsorption and transformation
of benzyl alcohol, while Pd nanoparticles play indispensable roles
in the following hydrogenation reactions. Thus, the Ti-NS and Pd
nanoparticles jointly catalyze the reductive N-alkylation of
nitrobenzene with benzyl alcohol under mild conditions.
To further understand the conversion of nitrobenzene in benzyl
alcohol, kinetics experiments were carried out based on the first 8
h. Fig. 1 shows the average reaction rates for the conversion of
nitrobenzene and generation of secondary amines based on their
respective amounts. As shown in Fig. 1a, the plot of the concentra-
tion of nitrobenzene against the reaction time is a straight line. The
average conversion rate of nitrobenzene over Pd/Ti-NS is 67.5
l
mol hꢁ1. Moreover, the number of secondary amines reached
430 mol after 8 h (Fig. 1b). The average generation rate of sec-
ondary amines over Pd/Ti-NS is 57.5
mol hꢁ1
l
l
.
The reuse of the photocatalyst was also studied (Figs. S4 and
S5). The recycling of the photocatalyst for five runs showed no
obvious decrease in photocatalytic activity. After the reaction, the
heterogeneous photocatalyst was separated easily from the photo-
catalytic system via a centrifugation and the photocatalyst can be

Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
109
Fig. 1. The kinetic data for the conversion of nitrobenzene (a) and yield of secondary amines (b) over Pd/Ti-NS.
reused directly. The XRD patterns and XPS spectra of sample are
almost unchanged after the several recycles demonstrating the
high stability of the Pd/Ti-NS (Fig. S5).
Why do the Pd/Ti-based materials have a huge difference for the
photocatalytic N-alkylation of nitrobenzene with benzyl alcohol?
Why can the tandem reaction be driven by visible light over the
Pd/Ti-NS and unavailable for Pd/TiO2? The detailed photocatalytic
behavior for the one-pot tandem reaction should be revealed in
depth. The superficial and electronic structures of Ti-NS distin-
guishing them from their layered precursor and TiO2 may tremen-
dously affect the photocatalysis, which may result in special
photocatalysis different from the other reported.
The preparation of Ti-L, Ti-NS, and Pd/Ti-NS followed a previ-
ously reported procedure [23,24]. The XRD patterns (Fig. S6) of
the prepared samples confirm that the Ti-L, Ti-NS, and Pd/Ti-NS
were successfully synthesized with high crystallinity. The Ti-L
show a characteristic peak (0 2 0) at 2h = 9.9° assigned to the
reflections of the lamellar structure, due to the periodic layers
stacking in a b-axis direction. It is obvious that the diffraction peak
(0 2 0) disappears in the Ti-NS due to the exfoliation of stacked lay-
ers into one layer in the b-axis direction. Moreover, the introduc-
tion of Pd particles has no impact on the original structure of the
Ti-NS. The structural representation of exfoliation is also shown
in Fig. S6.
The aforementioned experimental results confirm that photo-
catalytic N-alkylation of nitrobenzene with benzyl alcohol can be
achieved over the Pd/Ti-NS via a green photocatalysis. In order to
confer the high flexibility of the Pd/Ti photocatalyst, photo-
alkylation should be able to expand to the nitroarenes. The exper-
imental results have been summarized in Table 3. The Pd/Ti-NS can
efficiently catalyze the alkylation of nitrobenzene in benzyl alcohol
in one pot tandem reaction via a photocatalysis process. It is note-
worthy that the substituent groups have a great effect on the reac-
tion rates. Nitrobenzene derivatives with electron-withdrawing
substituents such as -Cl and -Br show a slight increasing in conver-
sion rate. On the contrary, electron-donating group like AOH and
ACH3 exhibit a decreased conversion rate. Taking the catalysts’
surface properties into account, in the photocatalytic reduction of
nitrobenzene and their derivatives, the adsorbed nitro group is
activated and then easily reduced. The functional groups of
electron-withdrawing and -donating in the aromatic rings would
enhance and weaken the interface charge transfer from Ti-NS to
the adsorbed nitro group, which greatly influences the reduction
of the nitro group [26]. Moreover, the reaction rate can also be
affected by the steric hindrance effects. The nitrophenol molecules
show different reaction rates when the hydroxyl group (AOH) is at
different positions on the benzyl ring. The reaction rate is in the
order ortho- < meta- < para-, which is similar to the previous study
on Au-TiO2 [21].
Representative scanning electron microscopy (SEM) images for
Ti-L and Ti-NS are displayed in Figs. S7a–S7b. It is obvious that the
Ti-L and Ti-NS show a periodic layered and a 2D sheetlike mor-
phology, respectively. Further detailed structural information was
obtained by transmission electron microscopy (TEM). As shown
in Fig. S7c, the ultrathin sheetlike morphology of the Ti-NS photo-
catalyst was obvious, and a lateral scale had a range from submi-
Table 3
Photocatalytic activity for the photoalkylation of nitroarenes in benzyl alcohol over Pd/Ti-NS.
Entry
Substrate
Time (h)
Con. (%)
Yield (%) a
Yield (%) a,
1
2
3
4
5
6
7
Cl
Br
8
8
8
10
10
10
12
100
100
98
99
94
70
77
77
82
75
65
56
7.9
11.4
5.6
5.2
5
OCH3
OH
m-OH
o-OH
CH3
88
90
3.7
3.4
Note: R at the para position of benzene rings, except for m-OH and o-OH.

110
Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
crometer to several micrometers. Figs. S7d and S7e show the TEM
image and EDX of Pd/Ti-NS. It is obvious that highly dispersed Pd
with particle size about 3 nm on the surface of ultrathin Ti-NS
has been obtained. Moreover, the chemical states of Pd nanoparti-
cles were determined by X-ray photoemission spectroscopy (XPS;
Fig. S7f). The peaks at 335.1 and 340.2 eV can be attributed to
the Pd3d5/2 and Pd3d3/2 of metallic Pd0 [24,25]. The thickness of
the ultrathin nanosheets was measured by atomic force micro-
scopy (AFM). As shown in Fig. S8, the prepared Ti-NS has only
about 0.7 nm thickness. According to the crystal structure informa-
tion of Ti-NS (cif document), the theoretical thickness of a single
layer is about 0.55 nm. Thus, the prepared Ti-NS with thickness
0.7 nm may be considered as the monolayer thickness. Therefore,
the photocatalysts with discrepant morphologies possess special
structural characters and present significantly different photocat-
alytic activities toward the tandem reaction of alcohols and
nitroarenes.
Fig. 3. Temperature-programmed desorption profiles of Ti-NS and Ti-L.
The chemisorption of reactant molecules in catalysts has a large
effect on the catalysis. Most of the catalysis usually occurs on the
catalysts’ surface. Therefore, the catalytic activity, selectivity, and
stability of certain catalysts largely depend on their surface proper-
ties. In our previous studies [24], we found that the Ti-NS contains
the abundant bridging hydroxyl groups Ti (OH) Ti, strongly acidic
OH groups, and TiAOH groups. These would form the distinctive
Lewis and Brønsted acid sites. The catalytic behavior of Ti-NS
may be greatly influenced by these surface acid sites. The interac-
tions between surface properties of the prepared samples and
regent molecules have been studied by in situ FTIR. As shown in
Fig. 2A, after the adsorption of benzyl, three peaks are observed
at about 1206, 1367, and 1452 cmꢁ1 on the Ti-NS attributed to
However, no obvious peaks of NH3 were observed for Ti-L, suggest-
ing much fewer exposed acid sites on Ti-L than on Ti-NS. Thus, the
distinct morphology of Ti-NS would provide a more open structure,
which is beneficial for the exposure of acid sites and absorption of
NH3 in comparison with the compact structure of Ti-L. The Ti-NS
shows a high ability for the absorption of NH3 and benzyl alcohol.
These results are consistent with the in situ FTIR for benzyl alcohol
adsorption (Fig. 3). Benzyl alcohols coordinated on the surface of
Ti-NS would result in the redistribution of electrons via the coordi-
nation bonds, which would efficiently activate the benzyl alcohol
[27,28]. These can be also corroborated by the catalytic experimen-
tal results that the benzyl alcohol can be catalyzed by white Ti-NS
under the visible light irradiation.
mH-O, d(OH), and d(CH3)2 respectively [22]. These peaks also
The roles of surface charge transfer in coordination species were
further analyzed by UV-DRS (Fig. 4). Fig. 4A shows that the single
Ti-NS and Ti-L would not respond to visible light. After the benzyl
alcohol (BA) was absorbed, the Ti-NS showed an obvious visible
light absorption property. Simultaneously, the color changed from
the white of Ti-NS to the yellow of benzyl alcohol/Ti-NS (BA/Ti-NS)
(Fig. 4B). Considering that the alcohols could be adsorbed onto the
surface Lewis acid sites via Lewis acid–base interaction [28], the
formation of surface complexes would be responsible for the visi-
ble light activity. This has also been witnessed in previous reports
onr Nb2O5, HNb3O8, and TiO2, and this absorption in the visible
light region was assigned to the ligand-to-metal charge transfer
of the surface complex [22,28]. Moreover, XPS spectra for Ti-NS
with adsorbed benzyl alcohol were also measured to elucidate
the chemical states of surface Ti atoms. As shown in Fig. 5, com-
pared with the single Ti-NS, the Ti2p spectrum of Ti-NS with
adsorbed benzyl alcohol exhibits an additional peak at low binding
remained even after further evacuation at 6 ꢃ 10ꢁ4, Torr suggest-
ing that benzyl alcohol can strongly chemisorb on the monolayer
Ti-NS. Moreover, the peaks ds(OH) at 1367 cmꢁ1 in absorbed ben-
zyl alcohol molecules exhibit a red shift compared with those in
free benzyl alcohol molecules (1363 cmꢁ1), which may result from
the coordination effect of the surface unsaturated surface Ti atoms
for benzyl alcohol via the ACAOꢀ ꢀ ꢀTiAspecies [22]. However, the
peaks on Ti-L are much weaker (Fig. 2B), indicating the lower con-
centration of exposed Lewis acid sites in Ti-L. To further reveal the
effect of surface acid sites on catalytic activity, the numbers of sur-
face acid sites in Ti-NS and Ti-L have been studied by NH3
temperature-programmed desorption (NH3-TPD). As shown in
Fig. 3, there are two strong peaks at 100 and 275 °C for Ti-NS,
which can be attributed to the weak and strong surface acid sites,
respectively. The number of surface acid sites can be calculated as
the summation of weak and strong sites, which is 53.94 mmol/g.
Fig. 2. In situ FTIR of benzyl alcohol absorption on Ti-NS (A) and Ti-L (B): (a) After degassing at 150 °C for 4 h. (b) Adsorption for 30 min at RT (physisorption + chemisorption).
(c) Further evacuation of excess probe molecules at 150 °C for 5 min (chemisorption).

Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
111
Fig. 4. UV–vis DRS of Ti-NS, Ti-L, and Ti-NS with benzyl adsorbed (BA/Ti-NS) and Ti-L with benzyl adsorbed (BA/Ti-L).
Fig. 5. The Ti2p XPS spectra of Ti-NS before and after the adsorption of benzyl
alcohol.
energy, suggesting that the electron cloud density of surface Ti
atoms is increased. Thus, the new additional low-valence compo-
nent would indicate that benzyl alcohols adsorbed onto the surface
of Ti-NS result in the redistribution of electrons from benzyl alco-
hol to Ti. Correspondingly, the color of Ti-NS changes from white to
yellow after the adsorption of benzyl alcohol (Fig. 4). Based on
these experimental results, it is rational to believe that surface
electron transfer in the surficial complexes takes place from alco-
hols to Ti-NS via the CAOꢀ ꢀ ꢀTiA species. The direct charge transfer
results not only in the activation of benzyl alcohol molecules, but
also in the expanded visible light absorption of Ti-NS. Because of
the few exposed Lewis acid sites, small changes were observed
on Ti-L after the benzyl alcohol absorbed (Fig. 4C). All of the results
elucidate that the surface properties of catalysts have a deep influ-
ence on catalytic activities for benzyl alcohol by forming surface
coordination species and direct interface charge transfer.
Considering that the surface chemical state of Ti would have an
influence on the adsorption of benzyl alcohol, XPS spectra of Ti2p
in Ti-NS, Ti-L, and TiO2 should be investigated to better understand
their difference in catalytic activity. As shown in Fig. 6, the Ti2p
XPS spectrum of TiO2 exhibited symmetrical profiles, indicating
that crystalline TiO2 possess only a single oxidation state of the
Ti atoms. The two Ti2p binding energies are 458.1 (2p3/2) and
464.1 eV (2p1/2), respectively [29]. Compared with Ti2p in TiO2, a
shift toward the high-binding-energy side was observed in Ti-L
(458.5 and 464.2 eV) and Ti-NS (458.7 and 464.5 eV), suggesting
the enhanced strength of TiAO bands and decreased electron cloud
Fig. 6. The XPS spectra of Ti2p in TiO2, Ti-L, and Ti-NS.
density of Ti atoms in Ti-L and Ti-NS. These results indicated that
the AOH in benzyl alcohol would more easily coordinate with sur-
face Ti atoms in Ti-L and Ti-NS than those in TiO2. Moreover, it
should be noted that the Ti2p1/2 and Ti2p3/2 in Ti-NS show the
higher Binding energy than that in Ti-L. These may be attributed
to the structure of Ti-NS, which influences the electron cloud den-
sity of surface Ti atoms [23]. In our previous study, the results of
EXAFS suggested that elongated TiAO and TiATi bonds were
observed in Ti-NS, with Ti-L as comparison, revealing a disordered
and warped TiO6 structure [23]. These would result in an open
structure to efficiently expose the active sites, such as dangling
bonds and defects. Thus, the steric effect may be the main factor
that affects the coordination of benzyl alcohol in Ti-NS and Ti-L.
The absorption behavior of nitrobenzene on the samples was
also studied by in situ FTIR spectra (Fig. S9). Fig. S9A shows the
in situ FTIR spectra of nitrobenzene absorption on Ti-NS. Two
strong peaks located at 1350 and 1527 cmꢁ1 attributed to the sym-
metrical telescopic vibration ms NO2 and asymmetrical stretching

112
Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
vibration mas NO2 can be also observed after an evacuation process,
suggesting strong interactions between Ti-NS and nitrobenzene
molecules. Moreover, these peaks show a red shift compared with
the ANO2 at 1345 and 1523 cmꢁ1 in free nitrobenzene molecules
(Fig. S9), while the other characteristic bands show no change
compared with those of free nitrobenzene (Fig. S9). These results
indicate that the nitrobenzene would adsorb onto Ti-NS via a
TiAOAHꢀ ꢀ ꢀO@N or TiAOAHꢀ ꢀ ꢀOAN coordination bond forming
the surface complex. According to our previous studies, the surface
AOH and bridging hydroxyl groups would serve as Brønsted acid
sites [24]. Thus, these surface Brønsted acid sites may be responsi-
ble for the adsorption of nitrobenzene. The absorbed nitrobenzene
molecules could be activated by the exposed Brønsted acid via the
sharing of electrons. However, the adsorption of nitrobenzene onto
Ti-L is much weaker due to the lower concentration of exposed
Brønsted acid sites in Ti-L (Fig. S9B). These results demonstrate
that the Brønsted acid sites play important roles in the activation
and further hydrogenation of nitrobenzene molecules. These
results can be confirmed by UV-DRS (Fig. S10). The light absorption
properties of the Ti-NS and Ti-L samples were not changed after
nitrobenzene absorption, indicating that the Brønsted acid sites
in catalysts could affect the electronic distribution of the nitroben-
zene molecules, but not result in direct surface electron transfer
under visible light irradiation.
Fig. 8. Polyhedral representation of the crystal structure for lepidocrocite titanates
Ti-L. Solid lines show the orthorhombic unit cell.
enhance the mass transfer. The photocatalytic activity would also
be increased (Fig. 9).
The exposed Lewis and Brønsted acid sites in catalysts are clo-
sely related to the photoalkylation of nitrobenzene with benzyl
alcohol. To elucidate the underlying factors that affect the surface
acid site concentration and photocatalytic activity, the microstruc-
ture of the samples was analyzed by N2 adsorption–desorption
(Fig. 7). The BET surface area of Ti-NS is about 198 m2gꢁ1, which
is more 20 times that of the Ti-L (8.4 m2 gꢁ1). Structurally, lepi-
docrocite titanates Ti-L usually crystallize in an orthorhombic
structure, consisting of two-dimensional (2D) corrugated layers
of edge- and corner-sharing octahedral and interlayer H3O+, which
compensates for the negative charge [30], as shown schematically
in Fig. 8. Thus, every layer in Ti-L contains a TiO6 host framework
and nearby H+ in the form of AOH group, which act as the Brønsted
acid sites. After the exfoliation process, the TiO6 host framework
with the AOH group would form Ti-NS. Because of the open 2D
structure, the AOH group is exposed on the surface of Ti-NS. How-
ever, the interlayer distance is very small, restricting the exposure
of acid sites. Therefore, the measured BET surface area of Ti-NS is
much greater than that of Ti-L. Furthermore, the small distance
of adjacent layers results in a steric inhibition effect limiting the
utilization of the interlayer acid sites on Ti-L. That is, the concen-
tration of the acid sites of LT may be similar to that of Ti-NS, but
more acid sites are available in Ti-NS than in Ti-L. The large num-
ber of exposed acid sites could absorb more reactant molecules to
Electron spin resonance spectroscopy (ESR) was used to explore
the deep photocatalytic process (Fig. 10). First, we conducted the
in situ ESR measurements for benzyl alcohol adsorbed on the Ti-
NS and Ti-L under vacuum. As shown in Fig. 10, no ESR signals
were observed on single Ti-NS and Ti-L. However, when benzyl
alcohol adsorbed onto the surface of the samples, an obvious ESR
signal at the value of g = 2.003 could be observed, which may be
assigned to the oxygen vacancies (VO) [22,24]. Moreover, the ESR
signal would be further enhanced under light irradiation. The
in situ FTIR, DRS, and XPS suggested the formation of surface coor-
dination species ACAOꢀ ꢀ ꢀTiA via coordination bonds, resulting in
charge redistribution from alcohol to Ti-NS. Thus, the electron
structure of the TiAO molecular orbital in Ti-NS would be dis-
turbed, giving an ESR signal. Moreover, under light irradiation,
the coordination species ACAOꢀ ꢀ ꢀTiA would be excited, enhancing
the surface charge redistribution from alcohol to Ti-NS. Thus, the
ESR signal was enhanced under light irradiation (Fig. 10). These
VO not only contribute to the exposure of unsaturated Ti atoms,
but also strengthen the coordination bond of CAOꢀ ꢀ ꢀTiA. Consider-
ing the lower concentrations of the exposed unsaturated surface Ti
atoms, Ti-L shows a much weaker ESR signal of VO.
The absorption of nitrobenzene onto photocatalysts has also
been studied. As shown in Fig. 11, the peaks at about g = 2.003
and 2.008 were assigned to the oxygen vacancies and coordinated
Fig. 7. N2 adsorption–desorption of Ti-NS (A) and Ti-L (B).

Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
113
Fig. 9. Adsorption behavior of target molecules on Ti-L and Ti-NS.
Additionally, the strong interactions between Ti-NS and nitroben-
zene molecules with electron-withdrawing groups could result in
directional charge transfer, which contributes to the activation of
nitrobenzene molecules. Under light irradiation, all of the ESR sig-
nals can be greatly enhanced, suggesting that the interaction
between the photocatalysts and nitrobenzene molecules can be
intensified by light irradiation.
Fig. 12 shows the in situ ESR spectra of Pd/Ti-NS for the photo-
catalytic N-alkylation of nitroarenes with benzyl alcohol at 1 atm
N2 and room temperature. The ESR signal of VO and coordinated
nitric species can be observed, when the nitrobenzene and benzyl
alcohols are added to the system. These signals can be enhanced
under light irradiation. The results suggest that the benzyl alcohol
and nitrobenzene molecules can be activated by the catalysts via
the surface coordination species. Moreover, the VO species derived
from the surface coordination process can further enhance the effi-
ciency of absorption and mass transfer, which greatly promotes the
photocatalytic efficiency.
Fig. 10. In situ ESR spectra of Ti-L- and Ti-NS-absorbed benzyl alcohol (BA/Ti-L and
BA/Ti-NS).
Based on the experimental results, we propose a possible
scheme of photocatalytic N-alkylation of nitroarenes with benzyl
alcohol over Pd/Ti-NS (Scheme 2). (i) In the initial phase of the
reaction, alcohols and nitroarenes were chemisorbed and activated
on the surface Lewis and Brønsted acid sites via the surface specie
Fig. 11. In situ ESR spectra of Ti-L- and Ti-NS-absorbed nitrobenzene (Nitr-/Ti-L and
Nitr-/Ti-NS).
ANOꢁ species, respectively [31]. These results confirm that
nitrobenzene molecules can coordinate with the Brønsted acid
sites in Ti-NS via the surface TiAOAHꢀ ꢀ ꢀO@N and TiAOAHꢀ ꢀ ꢀOAN
species, introducing a large number of VO. These can further
promote the absorption and activation of regent molecules.
Fig. 12. In situ ESR of Pd/Ti-NS for photocatalytic N-alkylation of nitroarenes with
benzyl alcohol: (a) Pd/Ti-NS. (b) Pd/Ti-NS + nitroarenes + benzyl alcohol under
darkness. (c) Pd/Ti-NS + nitroarenes + benzyl alcohol under light irradiation.

114
Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
Scheme 2. Mechanism for photocatalytic N-alkylation of nitroarenes with benzyl alcohol over Pd/Ti-NS.
HAOꢀ ꢀ ꢀTi and NO2ꢀ ꢀ ꢀHAOATis. The formation of VO in this process
contributes to the further activation of alcohols and nitrobenzene
molecules. (ii) Under light irradiation, direct surface charge trans-
fer takes place from alcohols to Ti-NS catalyst (ligand-to-metal)
through the surface complexes (Scheme 3). (iii) Subsequently,
these charges would be recombined by the photogenerated holes
in Ti-NS. The benzyl alcohol would be induced to deprotonate an
H+ forming aldehyde (Scheme 3). (iv) Since Pd has a higher work
function than Ti-NS, the Pd nanoparticles can act as electron-
acceptor sites in accepting the photogenerated electrons [32],
resulting in the reduction of H+ to active H atoms (Scheme 3).
The active H atoms transfer easily to the activated nitrobenzene
bound to the Ti-NS, which subsequently undergoes hydrogenation
to form the primary amines. (v) The coupling of primary amines
and aldehydes produces the imines, which are further hydroami-
nated to the secondary amines. The successful release of
intermediate products from Pd/Ti-NS by the photocatalytic
one-pot tandem reaction completes the cycles and permits
efficient reuse.
4. Conclusions
In summary, the photocatalytic N-alkylation of nitroarenes in
benzyl alcohol over Pd/Ti-NS has been achieved in a one-pot tan-
dem reaction under light irradiation with high conversion (>99%)
of nitrobenzene and high selectivity (85%) for secondary amines.
Moreover, Pd/Ti-NS shows high flexibility and universality for the
redox coupling reaction. We achieved thorough understanding of
the surface coordination behavior of reactant molecules on photo-
catalysts and further revealed the surface catalysis process of
nitroarenes and benzyl alcohol at a molecular level. In situ FTIR,
UV-DRS, and in situ ESR techniques confirm that the exposed sur-
face unsaturated Ti atoms (Lewis acid sites) and Brønsted acid sites
can strongly bind with the benzyl alcohol and nitrobenzene mole-
cules via the surface species ArACAOꢀ ꢀ ꢀTi and ArANO2ꢀ ꢀ ꢀHAOATi,
forming surface complexes. The interface charge transfer on the
surface ArACAOꢀ ꢀ ꢀTi coordination induces the activation and
transformation of the alcohol molecules. Moreover, the oxygen
vacancies driving the surface coordination reaction can greatly
enhance the chemisorption and activation of reagent molecules.
The hydrogen transfer from benzyl alcohols to the nearby nitroare-
nes takes place with the cooperation of Pd nanoparticles under
light irradiation. Based on the experimental results, a probable sur-
face coordination photocatalytic mechanism has been proposed.
Considering the versatile coordination chemistry on the metal
atoms, we believe that this work has not only developed a more
efficient photocatalyst for the photoalkylation of secondary amines
from nitroarenes in alcohols, but also provided great possibilities
for the flexible design of ultrathin metal oxide nanosheet photocat-
alysts in green organic synthesis.
Acknowledgments
This work was supported by the National Natural Science Founda-
tion of China (51672048, 51672046, and 21677036) and the major
science and technology projects of Fujian Province (2015YZ0001-
1). S. Liang also thanks the Natural Science Funds for Distinguished
Young Scholars of Fujian Province (201620162016).
Scheme 3. The dehydrogenation of benzyl over Pd/Ti-NS.

Y. Song et al. / Journal of Catalysis 361 (2018) 105–115
115
(d) N. Karikalan, S. Kubendhiran, S. Chen, P. Sundaresan, R. Karthik, J. Catal.
356 (2017) 43–52.
Appendix A. Supplementary material
[19] (a) J. Xiong, Y. Liu, C. Cao, L. Shen, W. Wu, S. Liang, R. Liang, L. Wu, J. Mater.
Chem. A 3 (2015) 6935–6942;
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.jcat.2018.02.005.
(b) W. Wu, R. Lin, L. Shen, R. Liang, R. Yuan, L. Wu, RSC Adv. 3 (2013) 10894–
10899.
[20] (a) Y. Nakai, M. Azuma, M. Muraoka, H. Kobayashi, S. Higashimoto, Mol. Catal.
443 (2017) 203–208;
References
(b) S. Higashimoto, Y. Nakai, M. Azuma, M. Takahashi, Y. Sakata, RSC Adv. 4
(2014) 37662–37668.
[21] (a) Y. Shiraishi, M. Ikeda, D. Tsukamoto, S. Tanaka, T. Hirai, Chem. Commun. 47
(2011) 4811–4813;
(b) Y. Shiraishi, K. Fujiwara, Y. Sugano, S. Ichikawa, T. Hirai, ACS Catal. 3 (2013)
312–320.
[22] (a) N. Li, X. Lang, W. Ma, H. Ji, C. Chen, J. Zhao, Chem. Commun. 49 (2013)
5034–5036;
[1] M.J. Climent, A. Corma, S. Iborra, M. Sabater, ACS Catal. 4 (2014) 870–891.
[2] Y. Shiraishi, K. Fujiwara, Y. Sugano, S. Ichikawa, T. Hirai, ACS Catal. 3 (2013)
312–320.
[3] (a) B. Chen, L. Wang, W. Dai, S. Shang, Y. Lv, S. Gao, ACS Catal. 5 (2015) 2788–
2794;
(b) M.J. Climent, A. Corma, S. Iborra, M.J. Sabater, ACS Catal. 4 (2014) 870–891;
(c) A. Behr, A.J. Vorholt, K.A. Ostrowski, T. Seidensticker, Green Chem. 16
(2013) 982–1006;
(b) X. Lang, H. Ji, C. Chen, W. Ma, J. Zhao, Angew. Chem. Int. Ed. 17 (2011)
3934–3937;
(c) S. Liang, L. Wen, S. Lin, J. Bi, P. Feng, X. Fu, L. Wu, Angew. Chem. Int. Ed. 53
(2014) 1–6.
(d) Q. Yang, Y. Chen, Z.U. Wang, Q. Xu, H. Jiang, Chem. Commun. 51 (2015)
10419–10422.
[23] (a) T. Sasaki, F. Kooli, M. Iida, Y. Michiue, S. Takenouchi, Y. Yajima, Fujio Izumi,
B.C. Chakoumakos, M. Watanabe, Chem. Mater. 10 (1998) 4123–4128;
(b) T. Sasaki, M. Watanabe, J. Am. Chem. Soc. 120 (1998) 4682–4689;
(c) Y. Song, H. Wang, J. Xiong, B. Guo, S. Liang, L. Wu, Appl. Catal. B Environ.
221 (2018) 473–481.
[24] Y. Song, H. Wang, X. Gao, Y. Feng, S. Liang, J. Bi, S. Lin, X. Fu, L. Wu, ACS Catal. 7
(2017) 8664–8674.
[25] (a) O.M. Wilson, M.R. Knecht, J.C. Garcia-Martinez, R.M. Crooks, J. Am. Chem.
Soc. 128 (2006) 4510–4511;
[4] L. Greb, J.-M. Lehn, J. Am. Chem. Soc. 136 (2014) 13114–13117.
[5] (a) B. Chen, J. Li, W. Dai, L. Wang, S. Gao, Green Chem. 16 (2014) 3328–3334;
(b) S. Kobayashi, Y. Mori, J.S. Fossey, S M. M. alter, Chem. Rev. 111 (2011)
2626–2704;
(c) M. Tamura, K. Tomishige, Angew. Chem., Int. Ed. 54 (2015) 864–867;
(d) T. Sonobe, K. Oisaki, M. Kanai, Chem. Sci. 3 (2012) 3249–3255.
[6] (a) O. Saidi, A. John Blacker, M.-M. Farah, S.-P. Marsdenb, J.-M.J. Williams,
Chem. Commun. 46 (2010) 1541–1543;
(b) A.J. Rawlings, L.J. Diorazio, M. Wills, Org. Lett. 17 (2015) 1086–1089;
(c) S. Ruch, T. Irrgang, R. Kempe, Chem. Eur. J. 20 (2014) 13279–13285.
[7] (a) Y. Zhao, S. - W. Foo, S. Saito, Angew. Chem. Int. Ed. 50 (2011) 3006–3009;
(b) A. Corma, T. Rodenas, M.J. Sabater, Chem. Eur. J. 16 (2010) 254–260;
(c) T. Thanh Dang, B. Ramalingam, S.P. Shan, A.M. Seayad, ACS Catal. 3 (2013)
2536–2540.
[8] T. Yan, B.L. Feringal, K. Barta, Nat. Commun. 5 (2014) 5602.
[9] L. Zhang, Y. Zhang, Y. Deng, F. Shi, Catal. Sci. Technol. 5 (2015) 3226.
[10] C.P. Xu, Z.H. Xiao, B.Q. Zhuo, Y.H. Wang, P.Q. Huang, Chem. Commun. 46 (2010)
7834–7836.
(b) F. Li, Q. Zhang, Y. Wang, Appl. Catal. A Gen. 334 (2008) 217–226;
(c) T. Schalow, B. Brandt, D.E. Starr, M. Laurin, S.K. Shaikhutdinov, S.
Schauermann, J. Libuda, H.J. Freund, Angew. Chem. Int. Ed. 45 (2006) 3693–
3697;
(d) R. Guo, R.W. Murray, J. Am. Chem. Soc. 127 (2005) 12140–12143;
(e) H. Tsunoyama, H. Sakurai, Y. Negishi, T. Tsukuda, J. Am. Chem. Soc. 127
(2005) 9374–9375.
[26] (a) P. Eskandari, F. Kazemi, Z. Zand, J. Photochem. Photobiol. A Chem. 274
(2014) 7–12;
(b) J.L. Ferry, W.H. Glaze, Langmuir 14 (1998) 3551–3555;
(c) M. Warrier, M.K.F. Lo, H. Monbouquette, M.A. Garcia-Garibay, Photochem.
Photobiol. Sci. 3 (2004) 859–863.
[11] A. Corma, T. Rodenas, M.J. Sabater, Chem. Eur. J. 16 (2010) 254–260.
[12] A.M. Rasero-Almansa, A. Corma, M. Iglesias, F. Sanches, ChemCatChem
(2014) 1794–1800.
6
[27] S. Higashimoto, N. Kitao, N. Yoshida, T. Sakura, M. Azuma, H. Ohue, Y. Sakata, J.
Catal. 266 (2009) 279–285.
[28] (a) H. Kobayashi, S. Higashimoto, Appl. Catal. B Environ. 170–171 (2015) 135–
143;
[13] M. Rueping, V.B. Phapale, Green Chem. 14 (2012) 55–57.
[14] (a) R. Kawahara, K. i. Fujita, R. Yamaguchi, J. Am. Chem. Soc. 132 (2010)
15108–15111;
(b) C.S. Cho, B.T. Kim, T.J. Kim, S.C. Shim, J. Org. Chem. 66 (2001) 9020–9022.
[15] Y.Z. Chen, Y.X. Zhou, H.W. Wang, J.L. Lu, T.I. Uchida, Q. Xu, S.H. Yu, H.L. Jiang,
ACS Catal. 5 (2015) 2062–2069.
[16] (a) M. Sankar, Q. He, S. Dawson, E. Nowicka, L. Lu, C.A. Pieter, A.M. Bruijnincx,
C.J. Beale Kielyc, B.M. Weckhuysen, Catal. Sci. Technol. 14 (2016) 5473–5482;
(b) C.H. Tang, L. He, Y.M. Liu, Y. Cao, H.Y. He, K.N. Fan, Chem. Eur. J. 17 (2011)
7172–7177;
(c) D. Stıbal, J. Sa, J.A.V. Bokhoven, Catal. Sci. Technol. 3 (2013) 94–98.
[17] (a) X. Lang, X. Chen, J. Zhao, Chem. Soc. Rev. 43 (2014) 473–486;
(b) A. Tanaka, Y. Nishino, S. Sakaguchi, T. Yoshikawa, K. Imamura, K.
Hashimoto, H. Kominami, Chem. Commun. 49 (2013) 2551–2553;
(c) X. Ke, X. Zhang, J. Zhao, S. Sarina, J. Barry, H. Zhu, Green Chem. 15 (2013)
236–244;
(d) X. Yang, B. Chen, L. Zheng, L. Wu, C. Tung, Green Chem. 16 (2014) 1082–
1086.
[18] (a) C. Liu, A. Zhang, D. Pei, H.Yu. Environ, Sci. Technol. 50 (2016) 5234–5242;
(b) A. Wang, H. Cheng, B. Liang, N. Ren, D. Cui, N. Lin, B.H. Kim, K. Rabaey,
Environ. Sci. Technol. 45 (2011) 10186–10193;
(b) S. Higashimoto, K. Okada, T. Morisugi, M. Azuma, H. Ohue, T.H. Kim, M.
Matsuoka, M. Anpo, Top. Catal. 53 (2010) 578–583;
(c) W. Feng, G. Wu, L. Li, N. Guan, Green Chem. 13 (2011) 3265–3272;
(d) S. Higashimoto, N. Suetsugu, M. Azuma, H. Ohue, Y. Sakata, J. Catal. 274
(2010) 76–83.
[29] P. Weerachawanasak, G.J. Hutchings, J.K. Edwards, S.A. Kondrat, P.J. Miedziak,
P. Prasertham, J. Panpranot. Catal. Today 250 (2015) 218–225.
[30] (a) T. Sasaki, F. Kooli, M. Iida, Y. Michiue, S. Takenouchi, Y. Yajima, F. Izumi, B.
C. Chakoumakos, M. Watanabe, Chem. Mater. 10 (1998) 4123–4128;
(b) M. Hervieu, B. Raveau, Rev. Chim. Miner. 18 (1981) 642–649;
(c) I.E. Grey, C. Li, I.C. Madsen, J.A. Watts, J. Solid State Chem. 66 (1987) 7–19;
(d) T. Gao, H. Fjellvag, P. Norby, Chem. Mater. 21 (2009) 3503–3513.
[31] (a) T. Yoshimura, Inorg. Chim. Acta. 125 (1986) 27–29;
(b) C. Napoli, G. Aldini, J.L. Wallace, F. Nigris, R. Maffei, P. Abete, D. Bonaduce,
G. Condorelli, F. Rengo, V. Sica, F.P. D’Armiento, C. Mignogna, G. Rosa, M.
Condorelli, L.O. Lerman, L.J. Ignarro, Proc. Natl. Acad. Sci. USA 3 (2002) 1689–
1694;
(c) F. Taketa, W.E. Antholine, J.Y. Chen, J. Biol. Chem. 15 (1978) 5448–5451.
[32] H.L. Skriver, N.M. Rosengaard, Phys. Rev. B 46 (1992) 7157–7168.
(c) Y. Mu, R.A. Rozendal, K. Rabaey, J. Keller, Environ. Sci. Technol. 43 (2009)
8690–8695;