Facilitating a high-performance photocatalyst for Suzuki reaction: Palladium nanoparticles immobilized on reduced graphene oxide-doped graphitic carbon nitride

Article abstract of DOI:10.1002/aoc.4623

We prepared a non-covalently coupled hybrid of reduced graphene oxide (rGO)-doped graphitic carbon nitride (g-C₃N₄) by freezing-assisted assembly and calcination. Fourier transform infrared, Raman and X-ray photoelectron spectroscopies and transmission electron microscopy confirmed that rGO was incorporated into the bulk g-C₃N₄, which was an ideal support for loading Pd nanoparticles. The Pd nanoparticles with an average size of 4.57?nm were uniformly dispersed on the rGO-doped g-C₃N₄ surface. The layered structure provided large contact area of g-C₃N₄ with rGO, further accelerating the electron transfer rate and inhibiting electron–hole recombination. Consequently, compared with Pd/rGO/g-C₃N₄ and Pd/g-C₃N₄, the Pd/rGO-doped g-C₃N₄ showed a prominent catalytic activity for visible-light-driven photocatalytic Suzuki–Miyaura coupling at ambient temperature. The Pd/rGO-doped g-C₃N₄ exhibited very high stability after six consecutive cycles with minimal loss of catalytic activity.

Full text of DOI:10.1002/aoc.4623

Received: 18 June 2018
DOI: 10.1002/aoc.4623
Revised: 28 August 2018
Accepted: 11 September 2018
F U L L P A P E R
Facilitating a high‐performance photocatalyst for Suzuki
reaction: Palladium nanoparticles immobilized on reduced
graphene oxide‐doped graphitic carbon nitride
1
1
1
2,3
Xiaohua Zhao | Jiateng Xie | Xin Liu | Xiang Liu
1
School of Materials Science and
We prepared a non‐covalently coupled hybrid of reduced graphene oxide
rGO)‐doped graphitic carbon nitride (g‐C N ) by freezing‐assisted assembly
Engineering, Jiangsu University,
Zhenjiang 212013, China
(
3
4
2
Zhenjiang Key Laboratory of Functional
and calcination. Fourier transform infrared, Raman and X‐ray photoelectron
spectroscopies and transmission electron microscopy confirmed that rGO was
incorporated into the bulk g‐C N , which was an ideal support for loading Pd
Chemistry and Institute of Medicine and
Chemical Engineering, Zhenjiang College,
Zhenjiang 212000, China
3
4
3
Biofuels Institute of Jiangsu University,
nanoparticles. The Pd nanoparticles with an average size of 4.57 nm were uni-
Zhenjiang 212013, China
formly dispersed on the rGO‐doped g‐C N surface. The layered structure pro-
3
4
vided large contact area of g‐C N with rGO, further accelerating the electron
Correspondence
3 4
Xiang Liu, School of Materials Science and
Engineering, Jiangsu University,
Zhenjiang 212013, China.
transfer rate and inhibiting electron–hole recombination. Consequently, com-
pared with Pd/rGO/g‐C N and Pd/g‐C N , the Pd/rGO‐doped g‐C N showed
3
4
3
4
3 4
Email: liuxiang0222@126.com
a prominent catalytic activity for visible‐light‐driven photocatalytic Suzuki–
Miyaura coupling at ambient temperature. The Pd/rGO‐doped g‐C N exhib-
3
4
Funding information
ited very high stability after six consecutive cycles with minimal loss of cata-
lytic activity.
Qing Lan Project; Modern Agricultural
Project of Zhenjiang City, Grant/Award
Number: NY2016022; Natural Science
Foundation of the Jiangsu Higher Educa-
tion Institutions of China, Grant/Award
Number: 17KJB430034; National Natural
Science Foundation of China, Grant/
Award Number: NSFC 21706105
KEYWORDS
3 4
Pd/rGO‐doped g‐C N , photocatalytic Suzuki–Miyaura, visible‐light‐driven
1
| INTRODUCTION
pursuing environmentally benign, green synthetic and
[
9–14]
sustainable processes.
Many studies centered on the
The Pd‐catalyzed Suzuki coupling reaction is one of the
most crucial transformations to achieve C─C bond for-
development of efficient catalysts for visible‐light‐driven
Suzuki coupling reactions, including the GO‐Pd@Ag‐
[1–4]
[15,16]
mation.
However, the traditional thermocatalytic‐
AgBr and Pd@NH ‐UiO‐66 catalysts.
Graphitic car-
2
driven process suffers from some disadvantages such as
high energy consumption and high pollution.
Photocatalysis is one of the current technologies with
high efficiency and energy economy. Until now, the main
research and development trend of the photocatalytic
process has focused on environmental remediation or
bon nitride (g‐C N ) has attracted enormous attention
due to its promising photocatalytic performance.
3
4
[17–20]
Recently, Li et al. prepared a Pd/g‐C N photocatalyst
3
4
for Suzuki coupling reaction under UV irradiation for
the first time. The results indicated that Pd nanoparticles
(NPs) could lead to photogenerated electrons via Mott–
Schottky heterojunction between g‐C N and Pd, which
[5–8]
energy conversion and storage.
3
4
With the development of photocatalysis technology,
light‐driven chemical reactions are becoming increasingly
noticeable and have been shown as a crucial way for
plays a crucial role in activating aromatic halides and
completing the Suzuki coupling reaction. Unfortunately,
the visible light photocatalytic performance and the
Appl Organometal Chem. 2018;e4623.
wileyonlinelibrary.com/journal/aoc
© 2018 John Wiley & Sons, Ltd.
1 of 10
https://doi.org/10.1002/aoc.4623

2
of 10
ZHAO ET AL.
stability of Pd/g‐C N were limited. On filtering out the
2.2 | Preparation of rGO‐doped g‐C N
3
4
3
4
UV radiation (<400 nm), the yield of the product
[21]
decreased to 49% after reacting for 12 h.
What is worse,
The GO precursor was synthesized from natural graphite
g‐C N is not considered to be an excellent support for
flake using a modified Hummers method. The as‐prepared
3
4
−1
heterogeneous catalysts because of the low specific sur-
face area. Hence, it is a great challenge to explore efficient
visible light photocatalysts for expediting the coupling
reaction under mild conditions. Graphene and reduced
graphene oxide (rGO) have been used as co‐catalysts for
improving the lifetime and transfer efficiency of charge
carriers photogenerated from semiconductors upon visi-
ble light irradiation, thereby enhancing the photoactivity
GO aqueous solution (1 mg ml , 12 ml) was mixed with
DCDA (60 mg), and then was placed into an ultrasonic bath
for 50 min. The solution was kept in an ultralow‐temperature
freezer (−72°C) to induce freezing‐assisted assembly. Next,
the obtained polymer was freeze‐dried. Finally, the target
sample was polymerized thermally from the freeze‐dried
−1
polymer by heating at 550°C for 2 h (ramp rate: 5°C min )
under nitrogen flow (Supplementary Figure 1).
[22–24]
of semiconductors.
In the work reported herein, we prepared a non‐
2
.3 | Preparation of Pd/rGO‐doped g‐C N
covalently coupled hybrid of rGO‐doped g‐C N4 for
3
4
3
supporting Pd NPs. Schematically illustrated in Figure 1,
GO was exfoliated into single sheets firstly and mixed with
Briefly, rGO‐doped g‐C N (200 mg) was dispersed in
3
4
100 ml of ethylene glycol under ultrasonication to form
the precursor monomer of g‐C N4 (dicyandiamide,
3
a uniform suspension. Then Pd(OAc) (5 mg) was dis-
2
DCDA) in water via electrostatic interaction. Then an
interconnected complex network was formed by a freeze‐
drying assembly, which was lastly converted to rGO‐doped
g‐C N by calcination. We used rGO‐doped g‐C N as a
solved in the suspension and magnetic stirring continued
for 120 min. After that, 300 μl of hydrazine hydrate solu-
tion (80 wt% in water) was added and stirred for another
3
4
3 4
120 min under ambient conditions. The final product,
support for Pd NPs based on its unusual features: (1)
nitrogen‐rich structural units can promote the fixation of
Pd NPs with high dispersion and (2) rGO is a general dop-
ant for g‐C N in band‐structure engineering, which leads
denoted as Pd/rGO‐doped g‐C N , was centrifuged,
3
4
washed with deionized water and dried in vacuum at
room temperature overnight. For comparison, Pd/rGO/
g‐C N and Pd/g‐C N were also synthesized.
3
4
3
4
3 4
to improved light harvesting, enhanced electronic conduc-
tivity and good electron transport properties. As expected,
the obtained Pd/rGO‐doped g‐C N was employed as a
3
4
2
.4 | Preparation of Pd/rGO/g‐C N and
3
4
photocatalyst for the Suzuki–Miyaura coupling reaction,
which exhibited excellent activity under visible light
irradiation at controlled room temperature.
Pd/g‐C N
3
4
The rGO/g‐C N composite was fabricated according to a
3
4
hydrothermal co‐assembly method. Typically, bulk g‐
was first obtained by directly heating DCDA. And
then, g‐C N nanosheets were prepared via chemical oxi-
2
2
| EXPERIMENTAL
.1 | Materials
C N
3 4
3
4
dation with concentrated H
SO
. Then, 80 mg of g‐C N
2
4
3 4
−
1
nanosheets was added to aqueous GO (2 mg ml ,
10 ml) and ultrasonicated for 15 min. Subsequently, the
mixture solution was transferred and sealed into a
All the chemicals were of analytical grade and used with-
out further purification.
3 4
FIGURE 1 Synthetic route to Pd/rGO‐doped g‐C N

ZHAO ET AL.
3 of 10
Teflon‐lined autoclave. Next, it was heated at 180°C over-
night and then cooled to 25°C naturally. Black rGO/g‐
C N was obtained by filtration and rinsed extensively
(JEOL, Japan). High‐resolution TEM (HRTEM) images
were obtained with a Tecnai G2 F30 S‐TWIN microscope
(FEI, USA). The specific surface area was acquired from
nitrogen adsorption–desorption isotherm at 77 K mea-
sured with a Quantachrome NOVA 3000e analyzer
(USA). Diffuse reflectance spectra were obtained with a
UV‐2550 spectrophotometer (Shimadzu, Japan) using
BaSO4 as compressed slices. Photoluminescence (PL)
emission spectra were obtained using a Quanta Master™
3
4
with deionized water and absolute alcohol.
Pd/rGO/g‐C N and Pd/g‐C N were prepared using
3
4
3 4
the same procedure as that for Pd/rGO‐doped g‐C N .
3
4
2.5 | Photocatalytic Suzuki coupling
reaction
4
0 instrument (PTI, USA) with an excitation wavelength
of 390 nm. Electrochemical impedance spectroscopy
EIS) was conducted using a CHI 660E (China) electro-
Typically, 0.1 mmol of aryl halide, 0.15 mmol of
(
arylboronic acid, 0.2 mmol of K CO and Pd/rGO‐doped
2
3
chemical workstation. The exact amounts of Pd were mea-
sured using inductively coupled plasma optical emission
spectrometry (ICP‐OES; ICP‐5000, China).
g‐C N (3 mg, 1.0 wt%) were added to a mixture of etha-
3
4
nol and water (4 ml, 1/1 v/v). The photocatalytic experi-
ment was conducted under a 300 W xenon lamp with a
UV‐cutoff filter to block UV radiation. To exclude the
photoheating effect, the temperature during the reaction
was maintained at 25°C via a water bath.
3
| RESULTS AND DISCUSSION
The composition and chemical bonding of GO, Pd/g‐C N ,
3
4
2
.6 | Characterization
Pd/rGO/g‐C
3
N
4
and Pd/rGO‐doped g‐C N were investi-
3 4
gated using FT‐IR spectra (Figure 2a). In the spectrum of
Pd/g‐C N , the broad peak located in the range 3000–
Fourier transform infrared (FT‐IR) spectra were obtained
using a Nicolet iS50 spectrometer (Thermofisher, USA)
3
4
−
1
3600 cm was indicative of the N─H group stretching
−
1
in the wavenumber range 400–4000 cm . Before mea-
surements, the prepared samples were dispersed into
KBr powder and compressed into pellets. Raman spectra
were obtained with a DXR spectrometer (Thermofisher,
USA) using 532 nm argon ion laser irradiation. Powder
X‐ray diffraction (XRD) measurements were conducted
with a D/MAX 2500PC diffractometer (Rigaku, Japan)
with a scanning angle (2θ) ranging from 10° to 70° and
scanning speed of 8° min . X‐ray photoelectron spectros-
copy (XPS) was performed using a PHI‐5702 instrument
with Al Kα radiation (150 eV). Field‐emission scanning
electron microscopy (SEM) was conducted with a Nova
Nano SEM450 microscope (FEI, USA) with an accelerat-
ing potential of 200 kV. Transmission electron microscopy
vibration. The characteristic peaks at 1636, 1559 and
−1
1418 cm were mainly due to typical stretching vibration
modes of C═N in the aromatic ring. Meanwhile, the peaks
at 1246 and 1328 cm were attributed to aromatic C─N
−1
[
25]
vibration mode.
The characteristic absorption band at
−1
810 cm was atributed as the breathing mode of triazine
[
26]
repeat units.
For GO, a broad absorption band at
−1
3404 cm was ascribed to O─H vibration mode of COOH
−
1
−1
groups. The peaks at 1739 and 1070 cm were associated
with C═O and C─O stretching vibrations of oxygen‐
containing groups, respectively.
between 1300 and 1700 cm corresponded to the typical
C═C stretching modes in the GO skeleton.
the O─H vibration bands of Pd/rGO/g‐C N slightly shifted
to lower wavenumber. The absorption peak at 1572 cm
[
27]
The series of peaks
−1
[28]
Whereas,
3
4
−1
(TEM) was performed with a JEOL‐2010F microscope
3 4 3 4 3 4
FIGURE 2 (a) FT‐IR spectra and (b) Raman spectra of GO, Pd/g‐C N , Pd/rGO/g‐C N and Pd/rGO‐doped g‐C N

4
of 10
ZHAO ET AL.
[
29]
was assigned to the stretching vibration of graphene.
67.92° were assigned to the (111), (200) and (220) crystal
2
[41,42]
Meanwhile, the disappearance of non‐sp carbon bands
C═O and C─O stretching bands at 1739 and 1070 cm
planes of Pd, respectively.
Unlike Pd/g‐C N ,
3
4
(
Pd/rGO/g‐C N showed a broad characteristic peak at
3
4
−
1
)
provided a preliminary proof for the successful reduc-
24.2°, which was evidence for loosely stacked rGO
[
30]
[42–44]
tion of GO to rGO after the reaction at 550°C.
The FT‐
sheets.
For the Pd/rGO‐doped g‐C N composite,
3
4
IR spectra of Pd/rGO/g‐C N and Pd/rGO‐doped g‐C N
the (002) peak corresponding to g‐C N was broadened
3
4
3
4
3 4
looked similar, confirming that the non‐covalent doping
technique did not affect the chemical structure of rGO
and g‐C N .
and shifted slightly from 27.38° to 26.62° due to the dop-
ing of rGO, suggesting the intercalation of rGO sheets
[
45,46]
into g‐C N .
The interlayer distance was larger than
3
4
3 4
Furthermore, the structural defects of the samples
were investigated using Raman spectroscopy (Figure 2
b). The spectrum of GO displayed three prominent peaks
that of g‐C N , but smaller than that of rGO (2θ = 24.2°).
3
4
This was indicative of tightly packed layers in Pd/rGO‐
doped g‐C N , not simply through π–π stacking interac-
3
4
−
1
at around 1346, 1575 and 2682 cm , assigned to the D, G
and 2D bands, respectively. Compared with GO, the G
bands of Pd/rGO/g‐C N4 and Pd/rGO‐doped g‐C N
tion, suggesting the successful production of rGO‐doped
g‐C N . The disappearance of diffraction peaks related
3
4
to C (002) also confirmed this conclusion.
3
3 4
−
1
shifted markedly from 1574 to 1587 and 1588 cm as a
result of π‐conjugated structure restoration during the
pyrolysis process, which was consistent with results
We used XPS to characterize the elemental composition
and the valent state of the elements in Pd/rGO‐doped
g‐C N . Figure 4a presents the survey spectrum of the
3
4
[28,31,32]
reported for g‐C N –rGO systems.
Apart from
XPS analysis, revealing the presence of C, N, O and
Pd elements. The high‐resolution N 1s spectrum could
be resolved into two peaks centered at 398.3 and
3
4
the shift of the G band, the I /I ratio was increased from
D
G
1
.13 for GO to 1.24 and 1.26, respectively, for Pd/rGO/g‐
C N and Pd/rGO‐doped g‐C N , which are similar to
400.0 eV (Figure 4b), corresponding to C─N═C and
3
4
3 4
33,34]
[
[47,48]
results of previous studies.
The increase of I /IG
N─(C) species, respectively.
The C 1s curve could
D
3
ratio resulted from the reduction of GO to rGO and the
be deconvoluted into two peaks, corresponding to the
presence of the sp ‐bonded carbon in nitrogen‐
2
effect of layered g‐C N sheets.
3
4
XRD patterns were utilized to verify the chemical
composition of the as‐prepared samples as shown in
Figure 3. The XRD pattern of Pd/g‐C N clearly showed
containing aromatic rings (N─C═N, 288.6 eV), C─O─C
bond (286.8 eV) and C─C bond (284.6 eV). The peaks
2
at 284.6 and 288.6 eV, together with the sp ‐hybridized
3
4
five well‐resolved peaks. The well‐defined broad peak at
7.38° was indexed to graphitic materials as the (002)
peak, suggesting the successful synthesis of g‐C N with
aromatic bonded nitrogen (C─N═C) and the tertiary
2
nitrogen groups (N─(C) ), confirmed the existence of
3
[
33,44]
g‐C N .
The peak at 286.8 eV was attributed to
3
4
3
4
[35–38]
ultrathin sheet structure.
The small peak at 13.01°
the residual C─O─C bond from GO after the hydro-
[
39,40]
[49]
was attributed to the (100) plane of layered C N .
thermal process.
The XPS spectrum of Pd 3d
3
4
Another three diffraction peaks at 40.10°, 46.69° and
consisted of two doublets (336.7 and 341.5 eV), which
was in good agreement with previous reports of metal-
lic Pd.
[
50,51]
The proposed packed layered structure of Pd/rGO‐
doped g‐C N was further supported by SEM and TEM
3
4
images. The morphologies of Pd/rGO/g‐C N and
3
4
Pd/rGO‐doped g‐C N4 are shown in Figure 5. The
3
Pd/rGO/g‐C N composite showed an aggregated struc-
3
4
ture with wrinkled and corrugated morphology, which
was similar to that of pristine g‐C N . The Pd NPs sup-
3
4
ported on rGO/g‐C N hybrid were uneven with much
3
4
aggregation, suffering from the weak interactions of Pd
NPs and rGO/g‐C N support (Figure 5a,b). In contrast,
3
4
for the Pd/rGO‐doped g‐C N , large amounts of irregular
3
4
two‐dimensional nanoflakes with many crumples
appeared, forming rGO‐doped g‐C N nanosheets with
3
4
lateral sizes of 3–10 μm (Figure S2). Furthermore, no
phase separation between rGO and g‐C N was observed,
3
4
FIGURE 3 XRD patterns of Pd/g‐C
rGO‐doped g‐C
3
N
4
, rGO/g‐C
3
N
4
/Pd and Pd/
indicating that rGO nanosheets were highly dispersed in
3
N
4
the Pd/rGO‐doped g‐C N composite. Notably, an even
3
4

ZHAO ET AL.
5 of 10
3 4
FIGURE 4 (a) XPS survey spectrum, and (b) N 1s, (c) C 1s and (d) Pd 3d spectra of Pd/rGO‐doped g‐C N
−
1
distribution of Pd NPs decorated on the two‐dimensional
multilayered nanosheets without agglomeration was
formed (Figure 5c,d). Based on the statistical size distri-
bution histograms obtained by measuring 50 randomly
chosen particles (Figure 5d), the average particle size
was estimated to be about 4.57 nm with a narrow size dis-
tribution (Figure 5e). HRTEM image (Figure 5f) clearly
revealed the good crystalline nature of the Pd NPs with
a lattice spacing of 0.226 nm, corresponding to the
interplanar spacing of (111) planes in face‐centered cubic
, which was higher than that of Pd/rGO/g‐C N
3
4
2
−1
2
−1
(133.70 m g ) and Pd/g‐C N (58.26 m g ), suggesting
that rGO was beneficial for increasing the specific surface
area of the composite. A large surface area showed that
rGO‐doped g‐C N4 was an ideal support for Pd NPs,
which could provide more reactive sites and reactant
adsorption for photocatalytic reactions.
3
4
3
The optical properties of Pd/g‐C N , Pd/rGO/g‐C N
3
4
3 4
and Pd/rGO‐doped g‐C N were investigated using UV–
visible diffuse reflectance spectra. As demonstrated in
Figure 6b, Pd/rGO/g‐C N4 and Pd/rGO‐doped g‐C N
3
4
[33,50]
Pd.
Further investigation with elemental mapping
3
3 4
disclosed that the Pd/rGO‐doped g‐C N4 composite
displayed increased visible light absorption compared
3
contained C, N and Pd elements as the main components,
and the elements were homogeneously distributed over
the entire area of the catalysts (Figure 5g). These results
further suggested that the rGO‐doped g‐C N facilitated
with Pd/g‐C N , indicating that the optical properties of
3
4
g‐C N sheets were successfully improved by modifying
3
4
[
52]
with rGO.
In addition, Pd/rGO/g‐C N retained an
3
4
absorption edge similar to that of g‐C N /Pd, revealing
3
4
3 4
the formation of small‐sized Pd NPs with high dispersity.
Therefore, all the above analyses confirmed that rGO was
successfully incorporated into g‐C N to form an interca-
no change in the basic structure of g‐C N after the incor-
3
4
poration of rGO. But for Pd/rGO‐doped g‐C N , the
3
4
absorption edge showed a slight red shift to a longer
3
4
lated layered structure. The intercalated layered structure
effectively stabilized the Pd NPs with high dispersion and
prevented their agglomeration.
wavelength, revealing the bandgap narrowing of the
Pd/rGO‐doped g‐C N4 composite, which could extend
3
the light absorption edge while simultaneously improving
the utilization of visible light. Apart from the light
absorption, we further analyzed the PL emission spectra.
Figure 6c shows the PL spectra of Pd/g‐C N , Pd/rGO/g‐
Figure 6a shows the nitrogen adsorption–desorption
isotherms of three samples. Brunauer–Emmett–Teller
2
surface area of Pd/rGO‐doped g‐C N was 168.63 m g
3
4
3 4

6
of 10
ZHAO ET AL.
FIGURE 5 (a, c) SEM and (b, d) TEM images of Pd/rGO/g‐C
g) C, N and Pd elemental mapping of Pd/rGO‐doped g‐C
3 4 3 4
N . (e) Pd NPs size distribution. (f) HRTEM image of Pd/rGO‐doped g‐C N .
(
3 4
N
[
44]
C N and Pd/rGO‐doped g‐C N . All the samples exhib-
transitions in g‐C N
The PL intensity of samples
3
4
3
4
3 4.
ited similar emission trends, with the principal peak at
diminished in the following order: Pd/g‐C N > Pd/
3
4
around 438 nm, resulting from the n–π* electron
rGO/g‐C N > Pd/rGO‐doped g‐C N . The emission peak
3
4
3 4

ZHAO ET AL.
7 of 10
FIGURE 6 (a) Nitrogen sorption isotherms, (b) UV–visible diffuse reflectance spectra, (c) PL spectra and (d) electrochemical impedance
spectra of Pd/g‐C , Pd/rGO/g‐C and Pd/rGO‐doped g‐C
3
N
4
3
N
4
3 4
N
intensity of Pd/rGO/g‐C N and Pd/rGO‐doped g‐C N
attributed to the excellent electron capturing and
transporting properties of rGO, which exhibited an
increased electrical conductivity and then reduced mark-
edly the electron–hole recombination. Meanwhile, the lay-
ered structure endowed the Pd/rGO‐doped g‐C N
3
4
3 4
was obviously weakened compared with g‐C N /Pd,
3
4
implying a delayed rate of charge carrier recombination.
Benefiting from excellent electrical conductivity, rGO
could capture and separate photoelectrons from g‐C N ,
3
4
3 4
resulting in
a
decreased recombination between
nanocomposite with a large surface area and abundant con-
tact interfaces, which in turn initiate photogenerated
charge separation.
photogenerated electrons and holes, which was consis-
tent with previous literature.
[53–55]
Moreover, the layered
structure of Pd/rGO‐doped g‐C N provided a large con-
3
4
tact area of g‐C N with rGO, further accelerating the
3
4
electron transfer rate and inhibiting electron–hole pair
recombination. Hence, the Pd/rGO‐doped g‐C N com-
3
.1 | Photocatalytic activity of
3
4
Pd/rGO‐doped g‐C N4
3
posite had the least recombination losses of
photogenerated charge carriers. To sum up, both
enhanced visible light absorption and improved separa-
Initially, the coupling of iodobenzene with phenylboronic
acid was selected as a model Suzuki reaction to evaluate
the photocatalytic ability of Pd/rGO‐doped g‐C N under
visible light irradiation. To exclude the photoheating
effect, the temperature during the reaction was main-
tained at 25°C via a water bath. The results of screening
experiments are summarized in Table 1. The Pd/rGO‐
doped g‐C N catalyst gave an excellent yield (98%) under
−
+
tion of e /h pairs endowed the Pd/rGO‐doped g‐C N4
3
3 4
composite with excellent photocatalytic activity.
To further confirm the photogenerated charge separa-
tion efficiency, EIS was carried out for the synthesized
photocatalysts, as shown in Figure 6d. Pd/rGO‐doped g‐
C N exhibited a markedly lower charge transfer resistance
3
4
3 4
(
36.11 Ω) than Pd/rGO/g‐C N (74.18 Ω) and Pd/g‐C N
the optimal conditions (Table 1, entry 3), while the reac-
tion did not occur when using only bare rGO‐doped g‐
C N as catalyst or without any catalyst under visible
3
4
3 4
(181.03 Ω), calculated from the diameter of the semicircle,
leading to enhanced separation efficiency of
3
4
−
+
photogenerated e –h pairs. This phenomenon could be
light irradiation (Table 1, entries 1 and 2). These results

8
of 10
ZHAO ET AL.
a
TABLE 1 Performance of catalysts for Suzuki coupling reaction of iodobenzene and phenylboronic acid
b
Entry
Changed parameters
hν
Yield (%)
1
2
3
4
5
6
7
No catalyst
+
—
—
98
39
68
57
76
rGO‐doped g‐C
3
N
4
+
Pd/rGO‐doped g‐C
Pd/rGO‐doped g‐C
Pd/rGO‐doped g‐C
3
3
3
N
N
N
4
4
4
+
−
+ −
+
Pd/g‐C
3 4
N
Pd/rGO/g‐C
3
N
4
+
a
Reaction conditions: iodobenzene (0.1 mmol), K
2
CO
3
(0.2 mmol), phenylboronic acid (0.15 mmol), catalyst (3 mg), EtOH–H
2
O (4 ml, 1:1) at 25°C, visible light
irradiation, reaction time of 40 min.
b
Gas yield.
suggested that Pd NPs were the main active centers in the
cross‐coupling reaction. Noticeably, the control reaction
without light irradiation (dark reaction) gave a yield of
coupling product of only 39% under the same conditions
(Table 2, entries 1–4), which made this reaction extremely
appealing. For the less reactive aryl bromides, the coupling
reactions could also proceed completely using irradiated
Pd/rGO‐doped g‐C N . We observed that aryl bromides
3
4
(Table 1, entry 4). This indicated the superiority of visible
containing electron‐deficient substituents (NO and CHO)
2
light for the Pd/rGO‐doped g‐C N ‐catalyzed Suzuki cou-
showed a far higher conversion rate, while for those with
3
4
pling reaction. The model reaction was exposed to visible
light for 20 min and then in the dark for another 20 min.
The reaction yield was only 68%, indicating that visible
light was the motive power and photocatalysis played a
crucial rule in the coupling reaction (Table 1, entry 5).
electron‐donating groups (OCH and CH ) the reaction
3
3
took longer for completion (Table 2, entries 5–8). This result
was reasonable because the electron‐deficient substituents
lead to an electron‐donating phenyl ring, which is easier
for nucleophilic attack by the photogenerated electrons on
electron‐rich Pd NPs.
In order to ascertain the advantage of rGO, Pd/g‐C N4
3
was used as the photocatalyst for comparison under visi-
ble light irradiation, giving a yield of only 57% (Table 1,
entry 6). A plausible explanation would seem to be that
rGO is a general dopant for g‐C N in band‐structure
To evaluate the reusability of the Pd/rGO‐doped g‐
C N , Pd/rGO/g‐C N and Pd/g‐C N4 catalysts, the
3
4
3
4
3
photocatalytic coupling of phenylboronic acid and
bromobenzene was performed under the same reaction
conditions as used previously. As shown in Figure 7,
the Pd/rGO‐doped g‐C N catalysts achieved a conver-
3
4
engineering, which leads to improved light harvesting,
enhanced electronic conductivity and good electron
transport properties. As is known, the catalytic activities
of rGO and g‐C N hybrids strongly depend on their
3
4
sion above 93% after six consecutive runs, indicating
excellent reusability. In contrast, the Pd/g‐C N and
3
4
3 4
structure and interaction, which are mainly determined
by the synthesis processes. Herein, rGO was intercalated
into bulk g‐C N to form a layered organic semiconduc-
Pd/rGO/g‐C N photocatalysts exhibited significant loss
of catalytic activity after six runs. These results demon-
3
4
strated that the Pd/rGO‐doped g‐C N4 photocatalyst
3
4
3
tor, which also provided a non‐covalent way to modulate
the electronic structure of g‐C N . Consequently, the pho-
had outstanding stability compared to Pd/g‐C N and
3
4
Pd/rGO/g‐C N , which was also quantitatively supported
3
4
3 4
tocatalytic efficiency of Pd/rGO‐doped g‐C N was better
by ICP‐OES analysis. The analysis showed that after six
recycling runs of the catalyst, only 0.16 wt% of Pd was
leached. As a comparison, the leaching of Pd from
Pd/g‐C N4 and Pd/rGO/g‐C N4 was higher than that
3
4
than that of Pd/rGO/g‐C N (76%; Table 1, entry 7).
3
4
Under the optimized reaction conditions, a series of aryl
halides and phenylboronic acids were investigated to study
the generality of the Pd/rGO‐doped g‐C N photocatalyst
3
3
from Pd/rGO‐doped g‐C N , at 1.8 and 1.12 wt%, respec-
3
4
3 4
for Suzuki reaction under mild conditions via visible light
irradiation. Most of the iodobenzenes with electron‐rich
or electron‐deficient substituents were applicable to the
photocatalytic coupling reaction, resulting in desired prod-
ucts with yields of 92–98% in no more than 40 min
tively. The high stability of the Pd/rGO‐doped g‐C N
3
4
photocatalyst could be attributed to the unique layered
structure and the nitrogen‐containing anchor sites for
Pd atoms, which were favorable for dispersing and stabi-
lizing the Pd NPs.

ZHAO ET AL.
9 of 10
a
3 4
TABLE 2 Scope of Suzuki reactions catalyzed by Pd/rGO‐doped g‐C N photocatalyst
b
Entry
R
X
Time (min)
Yield (%)
1
2
3
4
5
6
7
8
4‐OCH
3
I
40
40
30
40
90
90
50
50
96
4‐CH
3
I
95
4‐NO
2
I
98
4‐F
I
92
4‐OCH
4‐CH
4‐NO
4‐CHO
3
Br
Br
Br
Br
88
3
85.6
91.5
90.2
2
a
2 3 3 4 2
Reaction conditions: aryl halide (0.1 mmol), K CO (0.2 mmol), phenylboronic acid (0.15 mmol), Pd/rGO‐doped g‐C N (3 mg), EtOH–H O (4 ml, 1/1), visible
light irradiation, 25°C.
b
Gas yield.
absorption. Moreover, the Pd/rGO‐doped g‐C N
3
4
photocatalyst is sustainable due to uniformly dispersed
Pd NPs. In conclusion, the presented Pd/rGO‐doped g‐
C N layered composite can be regarded as an ideal
3
4
photocatalyst for Suzuki reactions.
ACKNOWLEDGMENTS
This work was partially supported by National Natural
Science Foundation of China (NSFC 21706105), Natural
Science Foundation of the Jiangsu Higher Education
Institutions of China (17KJB430034), Modern Agricul-
tural Project of Zhenjiang City (NY2016022), and ‘Qing
Lan’ Project Foundation of Jiangsu Province (2018).
3 4 3 4
FIGURE 7 Reusability of Pd/g‐C N , Pd/rGO/g‐C N and Pd/
ORCID
rGO‐doped g‐C
3 4
N photocatalysts for coupling of phenylboronic
acid and iodobenzene
Xiang Liu http://orcid.org/0000-0003-4437-2570
4
| CONCLUSIONS
REFERENCES
[
1] G. Li, S. Shi, P. Lei, M. Szostak, Adv. Synth. Catal. 2018, 360,
538.
2] J. Li, S. Yang, W. Wu, H. Jiang, Eur. J. Org. Chem. 2018, 2018,
284. https://doi.org/10.1002/ejoc.201701509
We have demonstrated the preparation of a Pd/rGO‐
doped g‐C N photocatalyst with packed layered struc-
1
3
4
[
ture and large surface area. Compared with Pd/g‐C N4
3
1
and Pd/rGO/g‐C N , the Pd/rGO‐doped g‐C N compos-
3
4
3 4
ite offered higher photocatalytic activity for the Suzuki
reaction under mild conditions. The excellent photocata-
lytic performance may be attributed to the impregnation
of rGO inside the layered structure of g‐C N . The rGO,
[3] P. Lei, G. Meng, M. Szostak, ACS Catal. 2017, 7, 1960.
[4] T. Ben Halima, W. Zhang, I. Yalaoui, X. Hong, Y.‐F. Yang, K.
N. Houk, S. G. Newman, J. Am. Chem. Soc. 2017, 139, 1311.
3
4
[5] L. Jiang, X. Yuan, Y. Pan, J. Liang, G. Zeng, Z. Wu, H. Wang,
emerging as an intrinsically superior electrical conductor,
Appl. Catal. B 2017, 217, 388.
3
rapidly captured the photoinduced electrons from g‐C N4
[6] W. Cui, J. Li, F. Dong, Y. Sun, G. Jiang, W. Cen, S. C. Lee, Z.
and then transferred them to the ultrafine Pd NPs, which
Wu, Environ. Sci. Technol. 2017, 51, 10682.
inhibited electron–hole recombination. In addition, dop-
ing rGO in layered g‐C N enhanced the visible light
[
7] X. Liu, X. Zhao, Y.‐Y. Yu, Y.‐Z. Wang, Y.‐T. Shi, Q.‐W. Cheng,
Z. Fang, Y.‐C. Yong, Electrochim. Acta 2017, 255, 41.
3
4

10 of 10
ZHAO ET AL.
[
8] M. Q. Wen, T. Xiong, Z. G. Zang, W. Wei, X. T. Tang, F. Dong,
[36] S. Yang, Y. Gong, J. Zhang, L. Zhan, L. Ma, Z. Fang, R. Vajtai,
Opt. Express 2016, 24.
X. Wang, P. M. Ajayan, Adv. Mater. 2013, 25, 2452.
[
9] J. C. Colmenares, R. Luque, Chem. Soc. Rev. 2014, 43, 765.
[37] S. Cao, Y. Li, B. Zhu, M. Jaroniec, J. Yu, J. Catal. 2017, 349, 208.
[38] P. Niu, L. Zhang, G. Liu, H.‐M. Cheng, Adv. Funct. Mater. 2012,
[
[
10] D. Wang, Z. Li, J. Catal. 2016, 342, 151.
2
2, 4763.
39] S. Martha, A. Nashim, K. M. Parida, J. Mater. Chem. A 2013, 1,
816.
11] M. Fernanda Galvez‐Lopez, M. J. Munoz‐Batista, C. Guadalupe
Alvarado‐Beltran, J. Luis Almaral‐Sanchez, B. Bachiller‐Baeza,
A. Kubacka, M. Fernandez‐Garcia, Appl. Catal. B 2018, 228, 130.
[
[
7
40] L. Zhang, X. He, X. Xu, C. Liu, Y. Duan, L. Hou, Q. Zhou, C.
Ma, X. Yang, R. Liu, F. Yang, L. Cui, C. Xu, Y. Li, Appl. Catal.
B 2017, 203, 1.
[
[
[
12] X. Hu, G. Zhang, F. Bu, A. Lei, Angew. Chem. Int. Ed. 2018, 57,
1
286.
13] G. Zhao, W. Zhou, Y. Sun, X. Wang, H. Liu, X. Meng, K.
[41] X. Zhao, X. Liu, RSC Adv. 2015, 5, 79548.
Chang, J. Ye, Appl. Catal. B 2018, 226, 252.
[42] M. Atarod, M. Nasrollahzadeh, S. M. Sajadi, J. Colloid Interface
Sci. 2016, 465, 249.
14] X.‐H. Jiang, Q.‐J. Xing, X.‐B. Luo, F. Li, J.‐P. Zou, S.‐S. Liu, X.
Li, X.‐K. Wang, Appl. Catal. B 2018, 228, 29.
[
[
[
[
[
43] K. Zhang, L. L. Zhang, X. S. Zhao, J. S. Wu, Chem. Mater. 2010,
[
15] D. Sun, Z. Li, J. Phys. Chem. C 2016, 120, 19744.
2
2, 1392.
44] Y. Li, H. Zhang, P. Liu, D. Wang, Y. Li, H. Zhao, Small 2013, 9,
336.
[
16] S. Gao, N. Shang, C. Feng, C. Wang, Z. Wang, RSC Adv. 2014,
4, 39242.
3
[17] Q. Xu, B. Cheng, J. Yu, G. Liu, Carbon 2017, 118, 241.
[18] D. Masih, Y. Ma, S. Rohani, Appl. Catal. B 2017, 206, 556.
[19] H. Zhao, S. Sun, P. Jiang, Z. J. Xu, Chem. Eng. J. 2017, 315, 296.
[20] P. Yang, H. Ou, Y. Fang, X. Wang, Angew. Chem. Int. Ed. 2017,
45] Y. Q. Sun, C. Li, Y. X. Xu, H. Bai, Z. Y. Yao, G. Q. Shi, Chem.
Commun. 2010, 46, 4740.
46] Y. J. Zhang, T. Mori, L. Niu, J. H. Ye, Energy Environ. Sci. 2011,
4, 4517.
5
6, 3992.
47] G. H. Dong, L. Z. Zhang, J. Mater. Chem. 2012, 22, 1160.
[
21] X.‐H. Li, M. Baar, S. Blechert, M. Antonietti, Sci. Rep. 2013, 3, 1.
22] Y. Xu, Y. Mo, J. Tian, P. Wang, H. Yu, J. Yu, Appl. Catal. B
[48] P. Kundu, C. Nethravathi, P. A. Deshpande, M. Rajamathi, G.
Madras, N. Ravishankar, Chem. Mater. 2011, 23, 2772.
[
2
016, 181, 810.
[49] Z. W. Tong, D. Yang, J. F. Shi, Y. H. Nan, Y. Y. Sun, Z. Y. Jiang,
ACS Appl. Mater. Interfaces 2015, 7, 25693.
[
23] W.‐C. Hou, I. Chowdhury, D. G. Goodwin Jr., W. M. Hender-
son, D. H. Fairbrother, D. Bouchard, R. G. Zepp, Environ. Sci.
Technol. 2015, 49, 3435.
[50] J. Sun, Y. Fu, G. He, X. Sun, X. Wang, Appl. Catal. B 2015, 165,
661.
[
[
24] K. Chang, Z. Mei, T. Wang, Q. Kang, S. Ouyang, J. Ye, ACS
Nano 2014, 8, 7078.
[51] J. Sun, Y. Fu, G. He, X. Sun, X. Wang, Catal. Sci. Technol. 2014,
4, 1742.
25] F. Chen, Q. Yang, S. Wang, F. Yao, J. Sun, Y. Wang, C. Zhang,
[52] W. Wang, J. C. Yu, D. Xia, P. K. Wong, Y. Li, Environ. Sci.
Technol. 2013, 47, 8724.
X. Li, C. Niu, D. Wang, G. Zeng, Appl. Catal. B 2017, 209, 493.
[
26] S. C. Yan, Z. S. Li, Z. G. Zou, Langmuir 2009, 25, 10397.
[53] R. C. Pawar, V. Khare, C. S. Lee, Dalton Trans. 2014, 43, 12514.
[
54] B. Ai, X. Duan, H. Sun, X. Qiu, S. Wang, Catal. Today 2015,
58, 668.
[
27] Z. Li, R. Lin, Z. Liu, D. Li, H. Wang, Q. Li, Electrochim. Acta
2
2
016, 191, 606.
[55] W.‐J. Ong, L.‐L. Tan, S.‐P. Chai, S.‐T. Yong, A. R. Mohamed,
Nano Energy 2015, 13, 757.
[
28] S.‐Q. Liu, B. Xiao, L.‐R. Feng, S.‐S. Zhou, Z.‐G. Chen, C.‐B. Liu,
F. Chen, Z.‐Y. Wu, N. Xu, W.‐C. Oh, Z.‐D. Meng, Carbon 2013,
6
4, 197.
29] Y. Fu, J. Zhu, C. Hu, X. Wu, X. Wang, Nanoscale 2014, 6,
2555.
[
[
[
[
[
[
[
SUPPORTING INFORMATION
1
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
30] A. Al Nafiey, A. Addad, B. Sieber, G. Chastanet, A. Barras, S.
Szunerits, R. Boukherroub, Chem. Eng. J. 2017, 322, 375.
31] J. Oh, S. Lee, K. Zhang, J. O. Hwang, J. Han, G. Park, S. O.
Kim, J. H. Park, S. Park, Carbon 2014, 66, 119.
32] Y. H. Li, Y. J. Sun, F. Dong, W. K. Ho, J. Colloid Interface Sci.
How to cite this article: Zhao X, Xie J, Liu X,
Liu X. Facilitating a high‐performance
photocatalyst for Suzuki reaction: Palladium
nanoparticles immobilized on reduced graphene
oxide‐doped graphitic carbon nitride. Appl
Organometal Chem. 2018;e4623. https://doi.org/
10.1002/aoc.4623
2
014, 436, 29.
33] W. Zhang, Q. Yao, X. Wu, Y. Fu, K. Deng, X. Wang,
Electrochim. Acta 2016, 200, 131.
34] J. Zhang, J. Ding, C. Li, B. Li, D. Li, Z. Liu, Q. Cai, J. Zhang, Y.
Liu, ACS Sustain. Chem. Eng. 2017, 5, 4982.
35] F. Goettmann, A. Fischer, M. Antonietti, A. Thomas, Angew.
Chem. Int. Ed. 2006, 45, 4467.