The preparation and photocatalytic activity of Ag-Pd/g-C3N4 for the coupling reaction between benzyl alcohol and aniline

Article abstract of DOI:10.1016/j.mcat.2019.110533

In this study, the carrier g-C₃N₄ was prepared by melamine, and the Ag-Pd/g-C₃N₄ catalyst was synthesized by the NaBH₄ reduction method. Different characterization techniques, including SEM, TEM, XRD, UV–vis DRS, XPS, photoluminescence spectra (PL) and BET, were employed to investigate the morphology and optical properties of the as-prepared samples. The Ag-Pd/g-C₃N₄ catalyst was used for the synthesis of imine from a benzyl alcohol and aniline. The results show that when the total loading of Ag and Pd is 2 wt%, and the mass ratio of Ag and Pd is 1:1, the activity of the catalyst is the highest (The highest conversion of aniline is 86.7% and the product selectivity is >99%.). The reaction is optimized by changing the type of solvent, the type and amount of base, the type of catalyst, and the amount of reactants. The optimal reaction conditions are 6 ml of n-hexane, 1.4 mmol of Cs₂CO₃, 50 mg of the Ag-Pd/g-C₃N₄ (2 wt%, 1:1), and 2:1 mol ratio of benzyl alcohol and aniline. Under optimal reaction conditions, alcohol derivatives and amine derivatives were investigated to determine the suitable range of the catalyst for alcohols and amines. Then, the effects of different light intensities and wavelengths on the reaction were explored. Additionally, the catalyst's recycling ability was tested, and it was found to be relatively stable. The effect of reactive groups on the mechanism shows that the reaction is mainly achieved by the synergy between h⁺, e^? and ·O₂^?.

Full text of DOI:10.1016/j.mcat.2019.110533

MolecularCatalysis476(2019)110533
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
The preparation and photocatalytic activity of Ag-Pd/g-C3N4 for the
coupling reaction between benzyl alcohol and aniline
Jingjing Maa, Xiujuan Yua,b, Xiaoling Liua, Haiying Lia, Xueli Haoa, Jingyi Lia,⁎
a College of Chemistry and Chemical Engineering, Inner Mongolia University, Hohhot, Inner Mongolia Autonomous Region, 010021 People’s Republic of China
b Hebei Key Laboratory of Neuropharmacology, Hebei North University, Zhangjiakou, Hebei Province, 075000, People’s Republic of China
A R T I C L E I N F O
A B S T R A C T
Keywords:
In this study, the carrier g-C3N4 was prepared by melamine, and the Ag-Pd/g-C3N4 catalyst was synthesized by
the NaBH4 reduction method. Different characterization techniques, including SEM, TEM, XRD, UV–vis DRS,
XPS, photoluminescence spectra (PL) and BET, were employed to investigate the morphology and optical
properties of the as-prepared samples. The Ag-Pd/g-C3N4 catalyst was used for the synthesis of imine from a
benzyl alcohol and aniline. The results show that when the total loading of Ag and Pd is 2 wt%, and the mass
ratio of Ag and Pd is 1:1, the activity of the catalyst is the highest (The highest conversion of aniline is 86.7% and
the product selectivity is > 99%.). The reaction is optimized by changing the type of solvent, the type and
amount of base, the type of catalyst, and the amount of reactants. The optimal reaction conditions are 6 ml of n-
hexane, 1.4 mmol of Cs2CO3, 50 mg of the Ag-Pd/g-C3N4 (2 wt%, 1:1), and 2:1 mol ratio of benzyl alcohol and
aniline. Under optimal reaction conditions, alcohol derivatives and amine derivatives were investigated to de-
termine the suitable range of the catalyst for alcohols and amines. Then, the effects of different light intensities
and wavelengths on the reaction were explored. Additionally, the catalyst's recycling ability was tested, and it
Ag-Pd/g-C3N4
Coupling reaction
Benzyl alcohol
Aniline
Visible light irradiation
was found to be relatively stable. The effect of reactive groups on the mechanism shows that the reaction is
−
mainly achieved by the synergy between h+, e− and ·O2
.
1. Introduction
synthesis of imines. In the reaction process of aniline self-oxidative
coupling to a synthetic imine, no solvent is needed, and the activity of
the imine can reach 95% under an air atmosphere at 373 K. (3) The
reaction of nitrobenzene with alcohol to synthesize imines: Nakaia et al.
[4] showed a synthesis of imines from benzylic alcohols with sub-
stituted groups and nitrobenzene over a CdS-TiO2 photocatalyst under
visible light irradiation at room temperature. The photocatalytic ac-
tivity can be explained by the photocurrent efficiency and a shift in the
flatband potentials of the CdS photoelectrode. In this paper, the first
method is used to obtain an imine by the photocatalytic coupling of a
benzyl alcohol and aniline.
The traditional photocatalyst is mainly composed of the semi-
conductor TiO2, a wide-bandgap semiconductor, whose forbidden band
width is 3.2 eV, therefore, TiO2 can only use 3%˜5% of the solar spec-
trum [6–10], which greatly limits its range of practical applications.
Carbon nitride has been studied [11–14], and findings reveal that both
the carbon and nitrogen atoms in g-C3N4 are sp2 hybridized, which is a
thermodynamically stable arrangement. It has the most stable structure
and semiconductor properties at room temperature. Moreover, g-C3N4
possesses band gap energy of approximately 2.7 eV (which allows the
Imines are organic compounds containing the C]N bond. They are
a class of important intermediates in organic synthesis reactions such as
reduction, addition, condensation and cyclization. They are used in
pharmaceuticals, perfumes, fungicides, dyes and pesticides. Imines are
also a class of widely used intermediates [1–5]. In addition, imines are
also common ligands in coordination chemistry. Conventional methods
for synthesizing imines include the coupling of amine compounds with
aldehydes or ketones. However, in recent years, a growing number of
methods for the synthesis of imines under mild reaction conditions from
the perspective of reactant species have been developed. Imine synth-
esis is mainly divided into the following three categories: (1) The re-
action of an alcohol and amine to synthesize an imine: Chen et al. [1,2]
found that CeO2 can oxidize and catalyze this reaction at a low tem-
perature (303 K). The reaction principle is based on the fact that the
alcohol is oxidized to the corresponding aldehyde or ketone under the
action of a catalyst and then coupled with aniline to form an imine. (2)
The self-oxidative coupling of aromatic amines to synthetic imines:
Yuan et al. [3] loaded RuO2 on graphene oxide (RuO2/GO) for the
⁎ Corresponding author.
E-mail address: lijingyicn@163.com (J. Li).
https://doi.org/10.1016/j.mcat.2019.110533
Received 20 April 2019; Received in revised form 7 July 2019; Accepted 23 July 2019
2468-8231/©2019ElsevierB.V.Allrightsreserved.

J. Ma, et al.
MolecularCatalysis476(2019)110533
absorption of certain visible light), an excellent photocatalytic perfor-
mance, chemical stability, and a good resistance to corrosion by acid
and alkali, making it a favorite in the field of photocatalysis. However,
some flaws have also been exposed, such as a small specific surface
area, a low degree of separation of electrons and holes, and a limited
range of visible light for reactions. In this regard, researchers all over
the world have developed various methods of modifying g-C3N4, and
have made some progress in this pursuit. Among the most popular
methods to increase photocatalytic performance is doping with a metal,
forming a precious bimetal from Ag and Pd to improve the catalytic
activity. Ag nanoparticles exhibit surface plasmon resonance (SPR) ef-
fect. Under visible light irradiation, Ag nanoparticles undergo intra-
band transitions in the 6sp orbital. Under ultraviolet light, inter-band
transitions from the 5d or 6sp orbit occur [15]. Researchers have suc-
cessfully loaded Ag nanoparticles onto g-C3N4. Pd, as a member of the
platinum group, has strong hydrogen absorbing ability, but exhibits
poor selective oxidation ability toward reactant molecules [16–19].
When Pd nanoparticles are diluted by Ag nanoparticles, as the iso-
merization active center, the isomerization activity promotes the oxi-
dation of benzyl alcohols. The hydroxy group concentration decreases
due to the decrease in the concentration of Pd, while on the hydro-
genation active center, the synergistic effect with the Ag SPR effect
promotes the dissociation of hydrogen and Ag [19,20]. The overflow of
hydrogen atoms accelerates the reaction rate.
In this work, Pd/g-C3N4, Ag/g-C3N4, and Ag-Pd/g-C3N4 (with dif-
ferent loadings of Ag and Pd and different mass ratios of Ag and Pd)
catalysts were prepared, and the catalysts were exmined by SEM, TEM,
XRD, UV–vis DRS, XPS, PL, and BET techniques. The Ag-Pd/g-C3N4
catalyst was used to catalyze the reaction of a benzyl alcohol with
aniline under visible light to form an imine. The solvent type, the type
and amount of alkali, the mass ratio and the bimetallic loading, the
amount of reactants, alcohol and amine derivatives, and the effect of
the light intensity and wavelength on the reaction were discussed.
Finally, the stability of the catalyst was determined by recycling, and
the photocatalytic principle of the reaction was explored by adding a
capture agent, which provided a basis for research into the catalyst and
the design of an industrial device.
was added dropwise, finally stirred for 1 h and aged for 24 h.
(5) The resulting catalysts was washed three times with deionized
water and ethanol, and dried at 60 ℃ for 12 h.
2.2. Characterization methods
Scanning electron microscopy (SEM, model S-4800, Japan Electron
Optics Laboratory Limited Company, Hitachi) was used to characterize
the morphology of the obtained composites. The morphology and
structure of the samples were examined by transmission electron mi-
croscopy (TEM) using an FEI Tecnai G2 F20 S-Twin electron micro-
scope, which was operated at an acceleration voltage of 200 kV. The
crystal phases of the samples were analyzed by X-ray diffraction (XRD)
with a Rigaku D/MAX-2500 X-ray diffractometer using Cu Kα radiation
(λ = 1.5405 Å) at 40 kV and 100 mA. The UV–vis diffuse reflectance
spectra (UV–vis DRS) of the solid samples were obtained using a
scanning UV–vis spectrophotometer (U-3900, Hitachi) equipped with
an integrated sphere assembly with 100% BaSO4 as the reflectance
sample. X-ray photoelectron spectroscopy (XPS) with Al Kα X-ray
(hν = 1486.6 eV) radiation operating at 150 W (Escalab 250xi,
ThermoFisher Scientific USA) was applied to investigate the surface
properties. The photoluminescence spectra (PL) of the catalyst was
determined using an Edinburgh Instruments FLS920 photo-
luminescence spectrometer with a pulsed xenon lamp (450 W) that
possessed an excitation wavelength of 325 nm and a wavelength range
of 450–650 nm. A multipoint BET method was applied to determine the
Brunauer-Emmett-Teller (BET) specific surface areas. The reaction
products were quantitatively analyzed with a gas chromatograph (GC-
2014C). The identity of the product was confirmed with a Trace DSQ II
gas chromatograph-mass spectrometer (GC–MS) at Inner Mongolia
University.
2.3. Photocatalytic activity test
In this experiment, the reaction substrate was 1 mmol of benzyl
alcohol and 1 mmol of aniline, 1.4 mmol of weakly basic cesium car-
bonate (Cs2CO3), 50 mg of catalyst, and 6 ml of solvent in a circular
bottom flask under an air atmosphere. The substrate was stirred, and
the light reaction was conducted for 10 h. The reaction temperature was
2. Materials and methods
30
3 ℃.
2.1. Preparation of Ag-Pd/g-C3N4 photocatalyst
2.4. Recycling test
2.1.1. Preparation of g-C3N4
Melamine was quantitatively placed in a crucible using an analytical
balance, then placed in a box-type electric resistance furnace, slowly
heated to 550 ℃ at 2 ℃/min, and calcined for 4 h to obtain g-C3N4.
About 50 mg of the catalysts is used in each experiment at the first
time, 40 experiments are done in succession, there are some losses after
each experiment (The catalysts are washed with deionized water for 3
times and with ethanol for 1 time, and then dried at 60 ℃ for 8 h.).
Finally, the catalysts are not enough for the sixth recycling. Therefore,
the catalysts are recycled for five times.
2.1.2. Preparation of Ag-Pd/g-C3N4
(1) In all, 1.5 g of the supporter was dissolved in 200 ml of deionized
water, and sonicated for 30 min, 5 wt% of PEG2000 was added,
followed by stirring for 10 min, and sonicated for 10 min.
(2) A quantitative amount of AgNO3 (0.1 mol·L−1) was pipetted into a
volumetric flask to prepare 100 ml of an AgNO3 liquid. The pre-
pared solution was slowly added dropwise to the above solution for
approximately 20 min. Then, lysine (0.53 mol·L−1) was taken with
a pipette and added dropwise for 10 min. After the completion of
the dropwise addition, the mixture was magnetically stirred for
30 min.
3. Results and discussion
3.1. Scanning electron microscopy (SEM) analysis
As seen from Fig. 1(a), the synthesized g-C3N4 is a stacked irregular
sheet-like structure with partially stacked pores. As shown in
Fig. 1(b–d), when modified with Ag and Pd, the surface of g-C3N4 be-
comes rough due to the presence of the noble metals, and it is also
found that the precious metal particles are also deposited in the pores of
g-C3N4. Fig. 1(e) shows the elemental surface distribution of Ag-Pd/g-
C3N4 (2 wt%, 1:1). It can be seen that the Ag and Pd nanoparticles re-
duced by NaBH4 are uniformly dispersed on g-C3N4. From the results in
Fig. 1(f), the mass fractions of Ag and Pd are 0.87% and 0.74%, re-
spectively, which are consistent with the theoretical additions. The
metal mass ratio is also consistent, which indicates that we successfully
prepared a catalyst of uniform dispersion.
(3) A quantitative amount of PdCl2 was pipetted into a volumetric flask
to prepare 100 ml of a PdCl2 (0.01 mol·L−1) liquid, and the pre-
pared solution was added dropwise to the above solution for ap-
proximately 20 min. The lysine was taken with a pipette, added
dropwise for 10 min, and magnetically stirred for 30 min.
(4) NaBH4 (0.135 g) was weighed and dissolved in 10 ml of water,
added dropwise to the solution, and then 10 ml of HCl (0.3 mol·L−1
)
2

J. Ma, et al.
MolecularCatalysis476(2019)110533
Fig. 1. SEM images of the different catalysts (a) g-C3N4; (b)-(d) Ag-Pd/g-C3N4 (2 wt%, 1:1); (e) The element mapping of Ag-Pd/g-C3N4 (2 wt%, 1:1) ;(f) The EDX
image of Ag-Pd/g-C3N4 (2 wt%, 1:1).
3

J. Ma, et al.
MolecularCatalysis476(2019)110533
Fig. 2. TEM images of the different catalysts (a) g-C3N4; (b) Pd/g-C3N4 (1 wt %); (c) Ag/g-C3N4 (1 wt %); (d) and (e) Ag-Pd/g-C3N4 (2 wt %, 1:1); (f) TEM images of
the recycled Ag-Pd/g-C3N4 (2 wt %, 1:1) catalyst.
3.2. Transmission electron microscopy (TEM) analysis
of Ag is not found after loading Ag, which may be because the loading
of Ag nanoparticles is too low. The characteristic Pd(111) diffraction at
a 2θ of 40.2° indicates that Pd is loaded onto g-C3N4 [21]. It can be seen
from Fig. 3(b) that as the loading of bimetallic Ag and Pt nanoparticles
increases, the intensity of the characteristic diffraction peak of Pd is
continuously strengthened. Due to the partial formation of the alloy,
the characteristic diffraction peak of Pd is shifted to a lower angle. From
the XRD spectrum of the catalyst after the cycle, we can conclude that
the structure of the catalyst does not change significantly after 5 cycles,
and the structure of the catalyst is stable.
We can see from Fig. 2(a) that g-C3N4 possesses an irregular sheet
structure. Seen from Fig. 2(b), (c), and (d), when modified with Ag and
Pd, the Ag and Pd nanoparticles are present on the surface of g-C3N4,
and the structure of g-C3N4 does not change. The Ag and Pd nano-
particles supported on g-C3N4 are well dispersed. By determining the
particle size from the TEM image, the inset of Fig. 1(b1), (c1), and (d1)
is obtained. When only the single metal Pd nanoparticles are supported,
the particle size distribution of the nanoparticles is between 3–5 nm,
and the measured average particle size of Pd is 4.24 nm. When only
single metal Ag is supported, the Ag particles are easily agglomerated,
and the measured particle size distribution is 8 nm. When Pd and Ag are
simultaneously loaded on g-C3N4, the average particle diameter is
4.17 nm, which is smaller than the single metal particle sizes, and the
dispersion of Ag particles is also better, which may be due to the sy-
nergy between the two metals. Fig. 2(e) is an HRTEM image of Ag-Pd/g-
C3N4 (2 wt%, 1:1). The lattice fringes of two kinds of nanoparticles Ag
and Pd can be seen in Fig. 2(e). The average lattice spacings are
0.237 nm and 0.219 nm, corresponding to the lattice fringes of the Ag
(111) and the Pd(111) crystal planes [21]. Fig. 2(f) is a TEM image of
the catalyst recovered after recycling. It can be seen that the g-C3N4
sheet structure remains intact, but a portion of the loaded Ag and Pd
nanoparticles fell off.
3.4. UV–vis diffuse reflectance spectroscopy (DRS) analysis
Fig. 4 shows the UV–vis DRS spectra of the different catalysts. As
seen from Fig. 4(a), g-C3N4 exhibits an absorption at 460 nm [23] (the
band gap is approximately 2.7 eV [24]), while Ag/g-C3N4 has an ab-
sorption tail band at 500 nm, which is due to the SPR effect of the Ag
nanoparticles. The absorption strength of the Pd/g-C3N4 catalyst is
significantly enhanced over that of g-C3N4. The three-way catalyst with
Ag and Pd loading has a red shift in its absorption wavelength and a
stronger intensity than g-C3N4. When combined with data from
Fig. 4(a) and (b), the strength and width of Ag-Pd/g-C3N4 (2 wt%, 1:1)
are the largest, that is to say, its ability to utilize visible light is the
strongest, and its reactivity is the best.
3.3. X-ray diffraction (XRD) analysis
3.5. X-ray photoelectron spectroscopy (XPS) analysis
Fig. 3 shows the XRD patterns of the different catalysts. It can be
seen from Fig. 3(a) that there are clear diffraction peaks at 2θ values of
12.8° and 27.4°, corresponding to the typical tri-triazine structural units
of g-C3N4. The (100) crystal plane and the conjugated aromatic system
are stacked to form a (002) crystal plane, thereby indicating that the
structure of g-C3N4 is stable [22]; after loading the Ag and Pd nano-
particles, the structure is not changed. However, the characteristic peak
Fig. 5(a) shows the full spectrum of Ag-Pd/g-C3N4 (2 wt%, 1:1), and
the characteristic peaks of C1 s, N1 s, Ag3d, and Pd3d can be seen.
Fig. 5(b) is a 1 s high-resolution XPS spectrum of the C element in the
Ag-Pd/g-C3N4 (2 wt%, 1:1) catalyst. As seen from Fig. 5(b), 284.6 eV is
the binding energy of the CeC bond, and 287.8 eV corresponds to the
binding energy of NeC = N in g-C3N4 [23]. Fig. 5(c) is a 1 s high-re-
solution XPS spectrum of the N element in the catalyst Ag/Pd/g-C3N4
4

J. Ma, et al.
MolecularCatalysis476(2019)110533
Fig. 3. XRD of the different catalysts.
(2 wt%, 1:1). It can be seen from Fig. 5(c) that there are three char-
acteristic peaks at 398.7 eV, 400.5 eV and 404.0 eV. The peak at
398.7 eV corresponds to the nitrogen atom in the triazine ring, which is
hybridized by sp2 orbitals; the characteristic peak at 400.5 eV corre-
sponds to the nitrogen atom of the group N-(C) 3; and the weak char-
acteristic peak at 404.0 eV is caused by positive charges in the carbon
and nitrogen ring [25]. Fig. 5(d) is a high-resolution XPS spectrum of
the Ag3d orbital in the Ag-Pd/g-C3N4 (2 wt%, 1:1) catalyst with two
characteristic peaks at 367.8 eV and 373.8 eV, corresponding to Ag3d5/
2 (Ag0) and Ag3d3/2 (Ag0) [21], indicating that the Ag nanoparticles are
present in the metallic state. Fig. 5(e) is a high-resolution XPS spectrum
of Pd3d orbital in the Ag-Pd/g-C3N4 (2 wt%, 1:1) catalyst, with two
characteristic peaks at 339.9 eV and 334.4 eV [21], corresponding to
the combined energies of the metallic Pd0. In summary, g-C3N4 was
successfully prepared, and the noble metals were supported on the
carrier in their metallic states.
3.7. N2 adsorption/desorption isotherm (BET) analysis
Fig. 7(a) is the N2 adsorption/desorption isotherm curves of pure g-
C3N4, Ag/g-C3N4, Pd/g-C3N4, and Ag-Pd/g-C3N4 (2 wt, 1:1). As shown
in Fig. 7(a), the isotherms of pure g-C3N4 and the four catalysts loading
a single metal or the bimetal are type IV, and the hysteresis loops are all
H3 type, indicating that all four substances exhibit mesoporous struc-
tures. The H3 type hysteresis loop indicates that the pore shape is
formed by the gap between two plates, which is consistent with the
layered structure of g-C3N4 [23]. The total amount of loading is 2 wt%
in Fig. 7(b), which shows the surface area vs the conversion for the
catalysts with different mass ratios of Ag and Pd. The specific surface
area of the supported bimetallic catalyst is significantly larger than that
of the supported single metal. When the mass ratio of the Ag and Pd
loading is 2.5:1.5, the specific surface area is the largest. However,
when the loading is 1:1, the specific surface area is not substantially
different from that obtained with 2.5:1.5. In combination with the
conversion and economic effect, we chose the loading ratio of 1:1 as the
optimal loading ratio. Fig. 7(c) is a graph showing the relationship
between the specific surface area and conversion of catalysts with an Ag
and Pd mass ratio of 1:1 but different total loadings. With the increase
in the loading of Ag and Pd, the specific surface of the catalyst increases
and then decreases in Fig. 7(c). The conversion is also the largest for
2 wt% loading. In summary, the catalyst possesses a pore structure, and
the structure of g-C3N4 remains unchanged after loading the noble
metal. When the total loading is 2 wt%, the mass ratio of Ag and Pd is
1:1, the specific surface area of the catalyst is large, and the conversion
is the largest. The specific surface area is a factor that affects the re-
action.
3.6. Photoluminescence spectra (PL) analysis
As seen from Fig. 6, the prepared g-C3N4, Ag/g-C3N4, Pd/g-C3N4,
and Ag-Pd/g-C3N4 (2 wt%, 1:1) catalysts possess a broad fluorescence at
460 nm. The signal peak of pure g-C3N4 is the strongest, while the
multi-catalyst loaded with a single precious metal shows a weaker
fluorescent signal peak, and Ag-Pd/g-C3N4 presents the lowest in-
tensity. This result indicates that the loading of two precious metals can
promote the photogeneration and separation of electrons and holes,
while the synergy of Ag and Pd bimetals also inhibits the recombination
of photogenerated electrons and holes. Based on the comprehensive PL
and UV–vis DRS spectra, we know that the Ag-Pd/g-C3N4 catalyst not
only increases the response range in the visible light, but also effec-
tively suppresses the recombination of photogenerated electrons and
holes.
3.8. Catalyst activity test
The activity test for the amination reaction of an alcohol and amine
Fig. 4. UV–vis DRS of the different catalysts.
5

J. Ma, et al.
MolecularCatalysis476(2019)110533
Fig. 5. (a)Ag-Pd/g-C3N4 (2wt%,1:1) full spectrum; (b) The XPS C 1s spectrum of Ag-Pd/g-C3N4 (2wt%,1:1); (c)The XPS N 1s spectrum of Ag-Pd/g-C3N4 (2 wt%, 1:1);
(d) The XPS Ag 3d spectrum of Ag-Pd/g-C3N4 (2 wt%,1:1); (e) The XPS Pd 3d spectrum of Ag-Pd/g-C3N4 (2 wt%,1:1).
with Ag-Pd/g-C3N4 photocatalyst was performed.
3.8.1. Effect of different solvents on the reaction
In this experiment, the reaction substrate was selected as 1 mmol of
benzyl alcohol, 1 mmol of aniline, 1 mmol of the weak base Cs2CO3,
50 mg of the Ag-Pd/g-C3N4 (4 wt%, 1:1) catalyst, and 6 ml of solvent in
a round bottom flask. The results are shown in Table 1.
The conversion of the reactant is related to the polarity of the sol-
vent under visible light irradiation as revealed in Table 1. The smaller
the polarity index of the solvent is, the greater the conversion is. The
higher the polarity index of the solvent is, the larger the interaction
with the reaction substrate is, which weakens the interaction between
the reaction substrate and the catalyst, and further weakens their re-
activity. Therefore, the conversion of the reactant in the solvent n-
hexane is the highest. Additionally, the conversions of all photoreac-
tions are significantly greater than those of the dark reactions. The
solvents methanol, DMSO and isopropanol normally quenches the
Fig. 6. PL spectra of the different catalysts.
6

J. Ma, et al.
MolecularCatalysis476(2019)110533
Fig. 7. (a) N2 adsorption isotherms of the different catalysts; (b)The correlation between the BET surface area and the conversion for different mass ratios of Ag and
Pd in the catalyst; (c) The correlation between the BET surface area and the conversion for different loading amounts of the bimetal catalyst.
Table 1
Table 2
Effect of the different solvents on the reaction of benzyl alcohol and aniline.
Effect of the different bases on the reaction of benzyl alcohol and aniline.
No. Solvent
Polar index Light irradiation
In dark
No.
Base
Light irradiation
Conv. (%)
In dark
Conv. (%) Sel. (%) Conv. (%) Sel. (%)
Sel. (%)
Conv. (%)
Sel. (%)
1
2
3
4
5
6
7
8
DMSO
DMF
Methanol
1,4-Dioxane
Isopropanol
Toluene
Mesitylene
Petroleum ether 0.01
n-Hexane
Cyclohexane
7.2
6.4
6.6
4.8
4.3
2.4
1.9
0
0
2.2
0
4.5
26.5
18.9
41.2
43.5
42.0
> 0
> 0
78.1
> 0
66.7
> 99
> 99
> 99
> 99
> 99
0
0
0
0
2.9
2
1.3
12.9
14.1
9.0
> 0
> 0
> 0
1
2
3
4
5
6
7
8
NaOH
KOH
LiOH
Na2CO3
K2CO3
CH3ONa
Cs2CO3
Blank
20.6
13.1
2.2
21.2
13.8
1.9
> 99
> 99
> 99
> 0
> 99
> 99
> 99
> 0
11.9
0.4
0.6
0
> 99
> 99
> 99
> 0
> 0
> 0
> 0
67.2
> 99
> 99
> 99
> 99
> 99
0
4.0
14.1
0
43.5
0.3
> 99
> 0
9
10
0.06
0.1
explore the effect of alkali type on the experiment. According to
Table 2, a weak base is more favorable for the coupling reaction of an
alcohol and amine, with the conversion of aniline being 43.5%, and the
value is the highest using Cs2CO3. Additionally, it is known from the
blank experiment that the alkali is necessary.
photogenerated holes directly/indirectly (serving as sacrificial electron
donors) and therefore conversion efficiency is lowered with these sys-
tems. The conversions of isopropanol and methanol are slightly higher
than that of 1,4-dioxane. In summary, the coupling reaction of benzyl
alcohol with aniline presents the highest conversion in n-hexane sol-
vent, which is 43.5%. Furthermore, light is a necessary condition for
catalyzing this reaction.
3.8.3. Effect of different catalysts on the reaction
The optimum base was 1 mmol of Cs2CO3, and the other conditions
were unchanged. The effect of different catalysts on the coupling re-
action of an alcohol and amine was investigated. The results are shown
in Table 3.
It can be seen from the experimental data 1–8 of Table 3 that when
g-C3N4 or a single metal catalyst is used, the reaction is far less effective
than that using the catalyst supporting the bimetal, indicating that the
synergistic effect between Ag and Pd plays a significant catalytic role. It
3.8.2. Effect of different bases on the reaction
Using the above-selected 6 ml of n-hexane as the optimal solvent,
the effect of the type of base on the coupling reaction of an alcohol and
amine was investigated under the above conditions. The results are
shown in Table 2.
We chose 7 kinds of strong bases, weak bases and organic bases to
7

J. Ma, et al.
MolecularCatalysis476(2019)110533
Table 5
Effect of the amounts of reactant on the reaction of benzyl alcohol and aniline.
Table 3
Effect of the different catalysts on the reaction of benzyl alcohol and aniline.
No. Catalysts
Light irradiation
In dark
No. Benzyl alcohol
(mmol)
Aniline
(mmol)
Light irradiation
In dark
Conv. (%) Sel. (%) Conv. (%) Sel. (%)
Conv.(%) Sel.(%) Conv.(%) Sel.(%)
1
2
3
4
5
6
7
8
9
g-C3N4
5.7
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
0.6
7.2
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
1
2
3
4
5
6
7
4.0
3.0
2.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
3.0
4.0
62.5
77.4
86.7
76.8
69.5
54.9
54.7
> 99
> 99
> 99
> 99
> 99
> 99
> 99
7.3
> 99
> 99
> 99
> 99
> 99
> 99
> 99
Ag/g-C3N4(1 wt%)
Pd/g-C3N4 (1 wt%)
Ag-Pd/g-C3N4 (1 wt%)
Ag-Pd/g-C3N4 (2 wt%)
Ag-Pd/g-C3N4 (3 wt%)
Ag-Pd/g-C3N4 (4 wt%)
Ag-Pd/g-C3N4 (5 wt%)
Ag-Pd/g-C3N4 (Ag:Pd = 3:1)
Ag-Pd/g-C3N4
(Ag:Pd = 2.5:1.5)
Ag-Pd/g-C3N4
(Ag:Pd = 1.5:2.5)
Ag-Pd/g-C3N44 (Ag:Pd = 1:3) 32.9
Blank 2.3
12.3
31.7
31.4
58.9
25.6
43.5
32.8
22.5
49.2
28.4
30.2
32.4
22.9
23.2
21.3
18.5
11.3
23.7
14.9
14.1
9.7
5.4
13.6
10
the reactant reaches the maximum, but compared with the conversion
of 1.4 mmol, it does not increase significantly. Therefore, 1.4 mmol is
used as the optimum alkali amount. When the amount of Cs2CO3 is
continuously increased, the conversion of the reactant decreases, which
may be because the Cs2CO3 molecule and the benzyl alcohol preempt
the active site of the catalyst. In summary, 1.4 mmol is the optimum
alkali amount for the reaction, and the conversion rate is 76.8%.
11
54.0
> 99
16.1
> 99
12
13
> 99
> 99
13.7
0.5
> 99
> 99
Note: 4–8 catalyst: the Ag and Pd mass ratio is 1:1; 9–12 catalyst: the total
loading is 2 wt%.
can be seen from the comparison of data 9–12 and 5 that when the
Ag:Pd ratio is 1:1, the total loading of Ag and Pd is 2 wt%, and the
catalytic effect is the best, which may be related to the morphology and
the interaction between the bimetal. These results are consistent with
the previous N2 adsorption-desorption data and the specific surface area
characterization, which may be due to the increased loading, the ag-
glomeration of the supported Ag and Pd nanoparticles, and the
blockage of g-C3N4 due to the increased loading (Therefore, the surface
area is smaller). Through the blank experiment, the catalyst has a clear
catalytic effect on the reaction. In this experiment, 50 mg of Ag-Pd/g-
C3N4 (2 wt%, 1:1) was selected as the catalyst for the coupling reaction
of an alcohol and amine, and its conversion is 58.9%.
3.8.5. Effect of the reactant amounts on the reaction
Taking 6 ml of n-heptane, 1.4 mmol of Cs2CO3, 50 mg of the Ag-Pd/
g-C3N4 (2 wt%, 1:1) catalyst in a round bottom flask, and the other
reaction conditions remain unchanged. We then explore the effect of
the reactant amounts on the reaction of a benzyl alcohol and aniline,
and the results are shown in Table 5.
When the amount of aniline is fixed at 1 mmol, and the amount of
benzyl alcohol is gradually increased, the conversion of the reactants
increases at first and then decreases under illumination (Table 5); while
with the amount of benzyl alcohol is fixed at 1 mmol, and with gradual
increase in aniline, the conversion of the reactants is gradually reduced.
This result may be because that the reaction is carried out in two steps
in this amination process. First, benzyl alcohol is easily oxidized into
benzaldehyde under the photocatalytic action of Ag-Pd/g-C3N4 (2 wt%,
1:1), and then, the aldehyde group in the benzaldehyde is affinity-
substituted with the amine group in the aniline. When the aniline
molecules are increased in the reaction, the active sites on the catalyst
will adsorb more aniline, and because the surface of the catalyst adsorbs
too little benzyl alcohol, the first step of amination will be inhibited,
that is, the formation of the intermediate benzaldehyde is less, which is
not conducive to the amination reaction; the excess of benzyl alcohol
will also reduce the adsorption of aniline on the surface of the catalyst,
and the second part of the amination reaction will then be inhibited,
which is not conducive to this reaction. In summary, when the amount
of benzyl alcohol is 2 mmol, and the amount of aniline is 1 mmol, the
conversion is maximized, and the maximum conversion is 86.7%.
3.8.4. Effect of the amount of alkali on the reaction
Through the above qualitative analysis, the best selected solvent is
6 ml of n-hexane, the best base is 1 mmol of Cs2CO3, the best catalyst is
50 mg of Ag-Pd/g-C3N4 (2 wt%, 1:1), and the other reaction conditions
remain unchanged. The effect of the amount of base on the coupling
reaction of the alcohol with the amine is shown in Table 4.
When Cs2CO3 is gradually increased from 0.4 mmol to 1.6 mmol
under light illumination, the conversion of the reactant is also gradually
increased; however, when Cs2CO3 is continuously increased from
1.6 mmol to 1.8 mmol, the conversion of the reactant declines slightly.
Under black conditions, when Cs2CO3 is gradually increased from
0.4 mmol to 1.8 mmol, the conversion of the reactant increases con-
tinuously. As the amount of alkali increases, the base plays an in-
creasingly important catalytic role. However, the conversion of the
reactant under visible light irradiation is substantially larger than that
in the dark. When the amount of Cs2CO3 is 1.6 mmol, the conversion of
3.8.6. Effect of alcohol derivatives on the reaction
After the above experiment, the optimal conditions of the reaction
were finally determined, namely 6 ml of n-hexane as the solvent, 50 mg
of Ag-Pd/g-C3N4 (2 wt%, 1:1) as the catalyst, 1.4 mmol of Cs2CO3,
1 mmol of aniline, and 2 mmol of various alcohols as the reactants. The
effect of different alcohol derivatives on the reaction was investigated,
and the experimental results are given in Table 6.
Table 4
Effect of the different amounts of base on the reaction of benzyl alcohol and
aniline.
编 号 No.
Base amount (mmol)
Light irradiation
In dark
According to datum 4 in Table 6, the catalyst exhibits a poor cata-
lytic performance for the alcohol having an electron-donating group in
the para position; as seen from data 5, 6, 11, 12, and 13, the catalyst
with an electron-withdrawing group in the para position, or an alcohol
that easily forms a large conjugated π bond, presents a good catalytic
effect. This result may be because the reaction depends on the ben-
zaldehyde formed, and the electron donating group causes the H on the
hydroxyl group to fall off with difficultly, so that benzaldehyde is not
easily formed; instead, when using a catalyst with an electron-with-
drawing group or a large conjugated π bond, benzaldehyde is easily
Conv. (%)
Sel. (%)
Conv. (%)
Sel. (%)
1
2
3
4
5
6
7
8
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
15.6
25.8
37.4
58.9
67.3
76.8
78.1
69.5
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
6.4
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
11.4
16.7
23.7
27.5
32.4
32.0
33.5
8

J. Ma, et al.
MolecularCatalysis476(2019)110533
Table 7
Effect of the amine derivative on the reaction of alcohol and aniline.
Table 6
Effect of the alcohol derivative on the reaction of alcohol and aniline.
No.
Light irradiation
In dark
Conv. (%)
30.2
编编号 号
NH2R
Light irradiation
In dark
Conv.(%)
30.2
Conv. (%)
86.7
Sel. (%)
Sel. (%)
> 99
> 0
Conv.(%)
86.7
Sel.(%)
Sel.(%)
> 99
1
2
> 99
78.7
1
> 99
3.8
> 0
2
3
58.7
65.5
> 99
> 99
12.8
22.3
> 99
> 99
3
4
21.9
48.6
91.3
97.1
3.7
97.5
94.7
4
5
89.2
81.2
> 99
> 99
33.3
11.5
> 99
> 99
17.9
5
98.3
> 99
30.4
> 99
6
7
89.1
6.7
> 99
> 99
35.7
0.3
> 99
> 99
6
95.4
> 99
37.7
> 99
7
8
29.6
< 0.1
> 99
< 0.1
1.5
< 0.1
> 99
< 0.1
8
9
10
23.5
3.9
2.6
> 99
> 99
75.8
1.9
> 0
> 0
> 99
> 0
> 0
9
95.0
> 99
3.3
> 99
11
12
62.4
87.3
> 99
> 99
10.9
23.7
> 99
> 99
3.9. Effect of light intensity on the reaction
13
70.9
> 99
21.3
> 99
To explore the influence of different light intensities on the reaction,
five different sets of light intensities were selected for the reaction
under the above optimized conditions. The effect of different light in-
tensities on the reaction of a benzyl alcohol and aniline to synthesize an
imine is shown in Fig. 8.
It can be seen from Fig. 8 that under the illumination, as the light
intensity increases, the conversion of the reactant also increases, and
the contribution of light to the whole reaction is large, which indicates
that the reaction is excited by light. The above-mentioned possible
causes are that the light intensity is increased, then more photo-
generated electrons and holes are excited on the Ag nanoparticles and
g-C3N4, and the transfer of photogenerated carriers is accelerated.
formed. The H on the hydroxyl group of the alcohol easily falls off so
that benzaldehyde is easily formed. Therefore, this catalyst is suitable
for alcohols that have electron-withdrawing groups.
The catalyst is not conducive to the reaction of a secondary alcohol
and aniline, as given from data 2, 7, and 10. The steric hindrance
presented to aniline in attacking the secondary alcohol is relatively
large, and the secondary alcohol is oxidized to a ketone; ketones are
more stable than aldehydes, so they do not readily react with aniline.
It can be seen from data 3, 8, and 9 that the catalytic effects of the
fatty alcohol and the branched C-chain aromatic alcohol are not ap-
parent, because such alcohols do not easily drop H to form an aldehyde.
In summary, the prepared Ag-Pd/g-C3N4 (2 wt%, 1:1) catalyst has a
particularly clear catalytic effect on alcohols having an electron-with-
drawing group or capable of forming conjugated bonds.
3.10. Effect of wavelength on the reaction
To investigate the effect of different wavelengths on the selective
synthesis of imines, this experiment selected filters of different wave-
length ranges for use under the above optimized conditions.
As shown in Fig. 9, the conversion to the target product decreases as
the wavelength is gradually reduced from 400 to 800 nm to
650–800 nm. It can be found that the conversion with light possessing a
3.8.7. Effect of amine derivatives on the imine reaction
In this experiment, 1 mmol of different kinds of amines were se-
lected, the other experimental conditions were the same as above, and
the reaction was carried out. The effect of amine derivatives on the
reaction is shown in Table 7.
It can be seen from data 4, 5, and 6 in Table 7 that the catalytic
performance of the catalyst for the electron-donating group in the para
position is better than that for the electron-withdrawing group, which
may be due to the electron-donating group being capable of forming an
amine group on the aniline. The catalyst with an electron cloud of a
high density easily carries out nucleophilic reactions.
From datum 3, the conversion using the catalyst with the meta-
substituent is low, which may be due to steric hindrance. From data 2
and 7, the effects of the catalyst on the aliphatic amine and the bran-
ched C-chain aromatic amine are not apparent, because the amine is not
strong in nucleophilicity. It can be seen from data 8 and 9 that the effect
of the catalyst on the secondary aliphatic amine is not clear, and that
the secondary aromatic amine can be catalyzed.
In summary, the Ag-Pd/g-C3N4 (2 wt%, 1:1) catalyst has poor cat-
alytic activity for meta-substituted anilines, while for secondary fatty
amines, and branched carbon-containing amines, the catalytic activity
is slight. However, the catalyst is suitable for amines with electron-
donating groups and secondary aromatic amines.
Fig. 8. Effect of different light intensities on the reaction of benzyl alcohol and
aniline.
9

J. Ma, et al.
MolecularCatalysis476(2019)110533
3.11. Cycle capability test of the catalyst
The catalyst was subjected to a cycle ability test using the above
reaction conditions, and the results are shown in Fig. 10.
As shown in Fig. 10, after the Ag-Pd/g-C3N4 (2 wt%, 1:1) catalyst
was recycled 5 times, it still maintains a good selectivity, but the con-
version began to decrease after the third cycle. According to the pre-
vious XRD (Fig. 3(b)) and TEM characterization (Fig. 2(f)), the structure
of g-C3N4 is not destroyed during the cycling, but the Ag and Pd na-
noparticles partially fell off, which reduces the conversion. In general,
the catalyst has a good catalytic stability.
3.12. Effect of reactive groups on the mechanism
To study the role of superoxide radicals (·O2−), photogenerated
holes (h+) and photogenerated electrons (e−) in the catalytic reaction,
the corresponding trapping agents benzoquinone (BQ), ethylenedia-
minetetraacetic acid (EDTA), and CCl4 were added to the reaction
under the above optimal conditions, and the effect of the reactive
groups on the mechanism was explored. The results are shown in
Fig. 11.
Fig. 9. Effect of different wavelengths on the reaction of benzyl alcohol and
aniline.
It can be seen from Fig. 11 that after the addition of EDTA, the
reduction of the reactant conversion indicates that the photogenerated
hole (h+) is a substance directly involved in the reaction, and the
conversion is decreased after the addition of CCl4 and BQ, indicating
−
that the photogenerated electron and ·O2 directly participates in the
−
reaction. However, O2 can be converted into ·O2−, and then ·O2 is
involved in the coupling reaction between the benzyl alcohol and ani-
line. The results show that the reaction is mainly achieved by the sy-
nergy between h+, e− and ·O2
.
−
4. Conclusion
Ag-Pd/g-C3N4 was successfully prepared. The catalyst possesses a
remarkable enhanced visible light absorption and good electron-hole
separation efficiency and it can conduct the coupling reaction of alco-
hols and amines to imines. Compared with the g-C3N4, Ag/g-C3N4, Pd/
g-C3N4 and Ag-Pd/g-C3N4 systems, the Ag-Pd/g-C3N4(2 wt%, 1:1)
system exhibits optimal photocatalytic activities, which consist of a fast
electron and hole separation and a slow charge recombination. The
optimal reaction conditions are 6 ml of n-hexane, 1.4 mmol of Cs2CO3,
50 mg of the Ag-Pd/g-C3N4 (2 wt%, 1:1), and 2:1 mol ratio of benzyl
alcohol and aniline. The catalyst's recycling ability shows that it is re-
Fig. 10. The recycle ability test of Ag-Pd/g-C3N4(2 wt%,1:1) in the reaction of
benzyl alcohol and aniline.
latively stable. The effect of reactive groups on the reaction shows that
it is mainly controlled by the synergy between h+, e− and ·O2
.
−
Acknowledgments
This work was financially supported by the National Natural Science
Foundation of China (NSFC, Nos. 20567002, and Nos. 21067007), the
Natural Science Fund of Inner Mongolia (2010MS0203 and
2014MS0201), the 2013 Annual Grassland Talents Project of Inner
Mongolia Autonomous Region, and the 2013 Annual Inner Mongolia
Talent Development Fund.
Appendix A. Supplementary data
Fig. 11. Effects of scavengers on the reaction of benzyl alcohol and aniline.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.110533.
wavelength within 400–550 nm is the largest, and the Ag-Pd/g-C3N4
(2 wt%, 1:1) catalyst is mainly responsive within 400–550 nm and the
result is consistent with that of the UV–vis diffuse reflectance mea-
surements. These experimental results show that under visible light il-
lumination, the light intensity and wavelength range play an important
role in the catalytic reaction.
References
[1] B. Chen, L.Y. Wang, S. Gao, Recent advances in aerobic oxidation of alcohols and
amines to imines, ACS Catal. 5 (2015) 5851–5876, https://doi.org/10.1021/
acscatal.5b01479.
[2] J. Yu, J.Y. Li, H.L. Wei, J.W. Zheng, H.Q. Su, X.J. Wang, Hydrotalcite-supported
10

J. Ma, et al.
MolecularCatalysis476(2019)110533
[14] Y.J. Cui, Z.X. Ding, X.Z. Fu, X.C. Wang, Construction of conjugated carbon nitride
nanoarchitectures in solution at low temperatures for photoredox catalysis, Angew.
Chem. Int. Ed. 51 (2012) 11814–11818, https://doi.org/10.1002/anie.201206534.
[15] X.Y. Pan, Y.J. Xu, Defect-mediated growth of noble-metal (Ag, Pt, and Pd) nano-
particles on TiO2 with oxygen vacancies for photocatalytic redox reactions under
visible light, J. Phys. Chem. C. 117 (2013) 17996–18005, https://doi.org/10.1021/
jp4064802.
gold catalysts for a selective aerobic oxidation of benzyl alcohol driven by visible
light, J. Mol. Catal. A Chem. 395 (2014) 128–136, https://doi.org/10.1016/j.
molcata.2014.08.023.
[3] G.H. Yuan, M. Gopiraman, H.J. Cha, H.D. Soo, I.M. Chung, I.S. Kim, Interconnected
ruthenium dioxide nanoparticles anchored on graphite oxide: highly efficient can-
didate for solvent-free oxidative synthesis of imines, J. Ind. Engin. Chem. 46 (2017)
279–288, https://doi.org/10.1016/j.jiec.2016.10.040.
[4] Y. Nakaia, M. Azumaa, M. Muraokaa, H. Kobayashib, S. Higashimotoa, One-pot
imine synthesis from benzylic alcohols and nitrobenzene on CdS-sensitized TiO2
photocatalysts: effects of the electric nature of the substituent and solvents on the
photocatalytic activity, Mol. Catal. 443 (2017) 203–208, https://doi.org/10.1016/
j.mcat.2017.09.018.
[5] B. Chen, L.Y. Wang, W. Dai, S.S. Shang, Y. Lv, S. Gao, Metal-free and solvent-free
oxidative coupling of amines to imines with mesoporous carbon from macrocyclic
compounds, ACS Catal. 5 (2015) 2788–2794, https://doi.org/10.1021/acscatal.
5b00244.
[6] M. Zhang, Q.G. Shang, Y.Q. Wan, Q.R. Cheng, G.Y. Liao, Z.Q. Pan, Self-template
synthesis of double-shell TiO2@ZIF-8 hollow nanospheres via sonocrystallization
with enhanced photocatalytic activities in hydrogen generation, Appl. Catal. B 241
(2019) 149–158, https://doi.org/10.1016/j.apcatb.2018.09.036.
[7] A. Fujishima, K. Honda, Electrochemical photolysis of water at a semiconductor
electrode, Nature 238 (1972) 37–38, https://doi.org/10.1038/238037a0.
[8] S. Ahamed, D.F. Ollis, Solar photoassisted catalytic decomposition of the chlori-
nated hydrocar bons trichloroetihylene and trichloromane, Sol. Energy 32 (1984)
597–601, https://doi.org/10.1016/0038-092X(84)90135-X.
[16] D. Seeburg, D.J. Liu, J. Radnik, H. Atia, M.M. Pohl, M. Schneider, A. Martin,
S. Wohlrab, Structural changes of highly active Pd/MeOx (Me = Fe, Co, Ni) during
catalytic methane combustion, Catalysts. 8 (2018) 42, https://doi.org/10.3390/
catal8020042.
[17] Z.F. Jiao, Z.Y. Zhai, X.N. Guo, X.Y. Guo, Visible-light-driven photocatalytic Suzuki-
miyaura coupling reaction on mott-schottky-type Pd/SiC catalyst, J. Phys. Chem. C.
119 (2015) 3238–3243, https://doi.org/10.1021/jp512567h.
[18] R. Sahoo, A. Roy, C. Ray, C. Mondal, Y. Negishi, S.M. Yusuf, A. Pal, T. Pal,
Decoration of Fe3O4 base material with Pd loaded CdS nanoparticle for superior
photocatalytic efficiency, J. Phys. Chem. C. 118 (2014) 11485–11494, https://doi.
org/10.1021/jp503393x.
[19] P. Pandey, S. Kunwar, M. Sui, S. Bastola, J. Lee, Investigation on the morphological
and optical evolution of bimetallic Pd-Ag nanoparticles on sapphire (0001) by the
systematic control of composition, annealing temperature and time, PLoS One 12
(2017) 1–18, https://doi.org/10.1371/journal.pone.0189823.
[20] Z.C. Sun, M.S. Zhu, M. Fujitsuka, A.J. Wang, C. Shi, T. Majima, Phase effect of NixPy
hybridized with g-C3N4 for photocatalytic hydrogen generation, ACS Appl. Mater.
Interfaces 9 (2017) 30583–30590, https://doi.org/10.1021/acsami.7b06386.
[21] Z.L. Zhao, Q. Wang, L.Y. Zhang, H.M. An, Z. Li, C.M. Li, Galvanic exchange-formed
ultra-low Pt loading on synthesized unique porous Ag-Pd nanotubes for increased
active sites toward oxygen reduction reaction, Electrochim. Acta 263 (2018)
209–216, https://doi.org/10.1016/j.electacta.2017.12.036.
[22] Z.Y. Mao, J.J. Chen, Y.F. Yang, D.J. Wang, L.J. Bie, B.D. Fahlman, Novel g-C3N4/
CoO nanocomposites with significantly enhanced visible-light photocatalytic ac-
tivity for H2 evolution, ACS Appl. Mater. Interfaces 9 (2017) 12427–12435, https://
doi.org/10.1021/acsami.7b00370.
[23] J.N. Liu, Q.H. Jia, J.L. Long, X.X. Wang, Z.W. Gao, Q. Gu, Amorphous NiO as co-
catalyst for enhanced visible-light-driven hydrogen generation over g-C3N4 photo-
catalyst, Appl. Catal. B 222 (2018) 35–43, https://doi.org/10.1016/j.apcatb.2017.
09.073.
[24] F. He, G. Chen, Y.G. Yu, S. Hao, Y.S. Zhou, Y. Zheng, Approach to synthesize g-PAN/
g-C3N4 composites with enhanced photocatalytic H2 evolution activity, ACS Appl.
Mater. Interfaces 6 (2014) 7171–7179, https://doi.org/10.1021/am500198y.
[25] J.Q. Wen, J. Xie, H.D. Zhang, A.P. Zhang, Y.J. Liu, X.B. Chen, X. Li, Constructing
multifunctional metallic Ni interface layers in the g-C3N4nanosheets/amorphous
NiS heterojunctions for efficient photocatalytic H2 generation, ACS Appl. Mater.
Interfaces 9 (2017) 14031–14042, https://doi.org/10.1021/acsami.7b02701.
[9] Q.F. Zhang, C.S. Dandeneau, X.Y. Zhou, G.Z. Cao, ZnO nanostructures for dye-
sensitized solar cells, Adv. Mater. 21 (2009) 4087–4108, https://doi.org/10.1002/
adma.200803827.
[10] H.J. Liang, B. Zhang, H.B. Ge, X.M. Gu, S.F. Zhang, Y. Qin, Porous TiO2/Pt/TiO2
sandwich catalyst for highly selective semihydrogenation of alkyne to olefin, ACS
Catal. 7 (2017) 6222–6257, https://doi.org/10.1021/acscatal.7b02032.
[11] K.X. Li, Z.X. Zeng, L.S. Yan, M.X. Huo, Y.H. Guo, S.L. Luo, X.B. Luo, Fabrication of
C/X-TiO2@C3N4 NTs (X = N, F, Cl) composites by using phenolic organic pollutants
as raw materials and their visible-light photocatalytic performance in different
photocatalytic systems, Appl. Catal. B 187 (2016) 269–280, https://doi.org/10.
1016/j.apcatb.2016.01.046.
[12] H. Wei, W.A. McMaster, J.Z.Y. Tan, L. Cao, D.H. Chen, R.A. Caruso, Mesoporous
TiO2/g-C3N4 microspheres with enhanced visible-light photocatalytic activity, J.
Phys. Chem. C 121 (2017) 22114–22122, https://doi.org/10.1021/acs.jpcc.
7b06493.
[13] B. Jürqens, E. Irran, J. Senker, P. Kroll, H. Mu¨ller, W. Schnick, Melem (2,5,8-
triamino-tri-s-triazine), an important intermediate during condensation of mela-
mine rings to graphitic carbon nitride: synthesis, structure determination by x-ray
powder diffractometry, solid-state NMR, and theoretical studies, J. Am. Chem. Soc.
125 (2013) 10288–10300, https://doi.org/10.1021/ja0357689.
11