Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69

Contents lists available at ScienceDirect

Journal of Photochemistry and Photobiology A:

Chemistry
journal homepage: www.elsevier.com/locate/jphotochem

Facile synthesis of carbon-rich g-C3N4 by copolymerization of urea and

tetracyanoethylene for photocatalytic degradation of Orange II
Zhuliang Wanga,b , Yaoxing Huoa , Yiping Fana , Rong Wua , Hao Wua,b,* , Fang Wanga,b ,
Xiaohong Xua,b,*
a
School of Chemistry and Materials Science of Shanxi Normal University & Key Laboratory of Magnetic Molecules and Magnetic Information Materials of
Ministry of Education, Linfen, 041004, China
b
Research Institute of Materials Science of Shanxi Normal University & Collaborative Innovation Center for Shanxi Advanced Permanent Magnetic Materials
and Techonology, Linfen, 041004, China

A R T I C L E I N F O A B S T R A C T

Article history:
Received 4 December 2017 A novel graphitic carbon nitride (g-C3N4) photocatalyst was synthesized by a facile one-pot thermo-
Received in revised form 4 February 2018 induced copolymerization method using urea and tetracyanoethylene (TCNE) for the first time. It was
Accepted 27 February 2018 revealed that the as-prepared samples retained the original framework of g-C3N4, showed an increased
Available online 10 March 2018 surface area and enhanced visible-light harvesting, and promoted photoexcited charge carrier
separation. The photocatalytic efficiency of the samples was investigated by degradation of Orange II
Keywords: solutions under visible-light irradiation (l > 420 nm). The sample CN/TCNE-40 (40 mg TCNE in 30 g urea)
g-C3N4 achieved an Orange II degradation rate of about 91% in 60 min under visible-light irradiation, which is
Copolymerization
significantly higher than that of pristine g-C3N4 (51%). The pseudo-first-order rate constant for Orange II
Urea
degradation on CN/TCNE-40 was 0.0317 min
1, 4.88 times as large as that of pristine g-C3N4. The CN/
TCNE
Photocatalytic TCNE-40 sample maintained its high photoactivity after five cycles under identical experimental
conditions and could degrade Orange II at various pH conditions under visible-light irradiation,
confirming the high durability of the photocatalyst. The mechanism of the enhanced photocatalytic
performance was also discussed. The results suggested that the CN/TCNE samples are potential
photocatalysts for environmental and energy applications.
© 2018 Elsevier B.V. All rights reserved.

1. Introduction attracted global attention and has become a new-generation

polymeric semiconductor for water splitting, contaminant remov-
Energy shortage and environmental pollution are global crises al, selective oxidation, CO2 reduction, solar cells, and sensing [6–
and threats to the long-term development of modern human 12]. Nevertheless, the photocatalytic efficiency of bulk g-C3N4 is
society [1,2]. Semiconductor-based photocatalysis has been limited because of its low surface area, fast recombination rate of
considered to be one of the most attractive methods for photogenerated electron–hole pairs, and lack of absorption in the
environmental pollutant elimination and energy transformation, visible-light region, which inevitably restricts its widespread
e.g., degradation of organic pollutants and splitting water to application.
produce hydrogen [3,4]. Recently, Wang et al. reported that To take better advantage of g-C3N4, it is important to explore
graphitic carbon nitride (g-C3N4) can be used as an attractive suitable strategies to improve the separation efficiency of
metal-free organic photocatalyst to promote water splitting for photoinduced carriers and to optimize photoreactivity. Various
hydrogen generation under visible light [5]. Since then, g-C3N4, methods have been implemented to improve the photocatalytic
which possesses a suitable band gap (2.7 eV), high thermal and performance of bulk g-C3N4, such as mesoporous structure
chemical stability, and good electrical and optical properties, has formation [13], doping [14,15], semiconductor coupling [16–20],
immobilization [21], and copolymerization [22]. Among these,
copolymerization with other organic compounds precursors is an
effective route to fabricate modified g-C3N4 polymers with
* Corresponding authors at: School of Chemistry and Materials Science of Shanxi enhanced photocatalytic activity; this method has attracted
Normal University & Key Laboratory of Magnetic Molecules and Magnetic increased attention because it retains the metal-free nature of
Information Materials of Ministry of Education, Linfen, 041004, China.
g-C3N4 [23–25], which has an important merit over metal-
E-mail addresses: [email protected] (H. Wu), [email protected] (X. Xu).

https://doi.org/10.1016/j.jphotochem.2018.02.036
1010-6030/© 2018 Elsevier B.V. All rights reserved.
62 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69

containing g-C3N4 in that metal-free g-C3N4 prevents potential 2.3. Synthesis of CN/TCNE
toxic metal leaching that usually causes secondary contamination
during water purification. However, the effect of copolymerization TCNE-copolymerized g-C3N4 (CN/TCNE) was also synthesized
on the photocatalytic performance of g-C3N4 is affected by the via the facile one-pot thermally induced copolymerization method
synthesis method and the selected precursor [26,27]. Therefore, mentioned above. In detail, 30 g urea and a certain amount of TCNE
seeking appropriate comonomers with a diverse chemical were mixed and ground evenly and collected in a crucible, and the
composition and structure to modify the band structure and same method as that used for the synthesis of pristine g-C3N4 was
optoelectronic properties of carbon nitride is a relevant and followed. The as-obtained materials were denoted as CN/TCNE-x,
important task. where x represents the amount of TCNE added (x = 10 mg, 20 mg,
It has been reported that carbon-rich g-C3N4exhibits a large 40 mg, 60 mg, 120 mg, 240 mg).
surface area, retarded photogenerated electron–hole recombina-
tion, increased visible-light absorbance, and enhanced photore- 2.4. Characterization of samples
duction activities[28,29]. In this study, tetracyanoethylene(TCNE),
a carbon-rich material containing only C and N (similar to g-C3N4) The crystal structure of the as-prepared samples was analysed
and four cyano groups that can directly integrate into the carbon by X-ray diffraction (XRD, Rigaku Ultima IV, Japan). The
nitride polymers, was investigated to improve the photocatalytic morphology and size were detected by scanning electron
activity of g-C3N4. A facile one-pot thermally induced copolymeri- microscopy(SEM, JSM-7500F, JEOL, Japan) and transmission
zation method was developed to synthesize TCNE-modified g-C3N4 electron microscopy (TEM, JEOL2100, Japan), and Fourier trans-
polymers (CN/TCNE) using urea as the g-C3N4 precursor and TCNE form infrared (FTIR) spectra were recorded on a Varian 640 FT-IR
as the modifier. The photocatalytic efficiency was investigated by spectrophotometer (Varian Inc., Palo Alto, CA, USA) using KBr
the photodegradation of Orange II in aqueous solutions under pellets dispersed with sample powders. The steady and time-
visible-light irradiation. One interesting finding was that the as- resolved fluorescence of samples were studied by photolumines-
fabricated CN/TCNE-40 samples exhibited remarkably enhanced cence (PL, FLS980, UK) spectra. The specific surface area and pore
visible-light-induced degradation performance and high stability, size distributions were measured using liquid nitrogen at 77 K
which demonstrated their versatile application as a photocatalyst. using a Micromeritics ASAP 2020 instrument via the Brunauer-
Emmett-Teller (BET) method (Quantachrome, USA). X-ray photo-
electron spectroscopy (XPS, Escalab 250Xi, Thermo Fisher Co., USA)
2. Experimental section
was performed to investigate the chemical compositions and
states of the materials. A UV–visible (UV–vis) spectrophotometer
2.1. Materials and chemicals
was used to record the UV–visible diffuse spectra of these samples
(UV-2700, Shimadzu Japan). The amperometric photocurrent–
Urea were purchased from Kermel (99%, Tianjin, China), and
time (i–t) responses were carried out by an IM6e electrochemical
TCNE (98%) and Orange II (99.9%) were provided by Acros Organics
workstation (ZAHNER, Germany). All nuclear magnetic resonance
(New Jersey, USA). The other reagents were of analytical grade and
(NMR) spectroscopy experiments were performed on an Agilent
used without further purification. Double distilled water was used
600 MHz NMR spectrometer DD2 (Agilent Technologies, USA).
throughout this study.

2.5. Photodegradation of Orange II

2.2. Synthesis of g-C3N4
Photocatalytic efficiencies of the CN/TCNE samples were
Graphitic carbon nitride was prepared by a thermally induced evaluated for the degradation of Orange II (10 mg/L) under
copolymerization method using urea as the precursor. In a typical visible-light irradiation derived from a Xe source (500 W) with a
run, 30 g urea was taken in a crucible with a loose cover and then cut-off filter (420 nm). In a typical experiment, 100 mL Orange II
heated at 550  C for 4 h in a muffle furnace at a heating rate of aqueous solution was added with 50 mg photocatalyst (0.5 g/L)
20  C/min. When the temperature was cooled down to room into a double-jacket reactor with cycling water pumped from a
temperature, solid pristine g-C3N4 was obtained, which was water bath at 20  C. A magnetic stirrer was used to ensure that the
ground to a powder for further use. photocatalyst dispersed homogeneously during the reaction. The

Fig. 1. (a) XRD patterns and (b) FT-IR spectra of g-C3N4 and CN/TCNE samples.
 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69 63

reaction system was kept in the dark for 30 min to achieve indicating that the copolymerization with TCNE led to an increase
adsorption–desorption equilibrium, and then 3 mL of the Orange II in the interplanar distance of the graphitic layered structure [26].
solution was withdrawn using a syringe at each time interval, and In addition to the peak shift, with increasing TCNE content, the
injected into a vial through a polytetrafluoroethylene filter peaks of the CN/TCNE samples became weaker and broader than
(0.45 mm). Finally, the solutions were investigated using a UV– those of pristine g-C3N4, which could be attributed to the
vis spectrophotometer (l = 484 nm). conversion of larger size graphitic-like structure to the small size
[32].This suggested that with increasing TCNE content, especially
3. Results and discussion when it exceeded 40 mg in the samples, the graphitic-like
structure was changed to small size, leading to more photocatalytic
3.1. Crystal phase and microstructure deficiency in g-C3N4.
FTIR spectroscopy was conducted to analyse the functional
The XRD patterns of pristine g-C3N4 and CN/TCNE are shown in groups of the as-prepared CN/TCNE samples, and the results are
Fig. 1a. For all samples, even though the TCNE content was given in Fig. 1b. The characteristic spectra of all the samples were
increased up to 240 mg, two peaks were observed at around 13.1 similar. The intense bands located in the 1200–1700 cm
1 region
and 27.4 , which was in accordance with the crystalline structures were assigned to the characteristic stretching vibration modes of
of g-C3N4 [30]. The high-intensity peak at 27.4 , corresponding to aromatic CN heterocycle repeating units [33]. The band located
an interlayer distance of 0.326 nm, reflected the interlayer stacking around 800 cm
1 is a typical feature of the breathing mode of
of aromatic segments and could be indexed as the (002) peak triazine (C3N3) units. Additionally, the peaks at 3100–3400 cm
1
observed for graphitic materials. The low-intensity peak at 13.1, were attributed to a small amount of residual amino (–NH2) group
corresponding to an interlayer distance of 0.676 nm, was indexed stretching vibration. This indicated that the TCNE copolymeriza-
as the (100) peak and could be associated with an in-plane tion did not alter the core chemical skeleton of g-C3N4, and that the
structural packing motif of tri-s-triazine units [31]. This implied main structural framework of the samples remained intact after
that the CN/TCNE samples could maintain the in-plane and the copolymerization reaction [34].
interlayer characters of g-C3N4. Further observation of the The morphology and microstructure of pristine g-C3N4 and CN/
magnified view of the (002) peak revealed that with increasing TCNE were investigated using SEM and TEM, shown in Fig. 2a–h.
TCNE content, the peaks of the CN/TCNE samples shifted slightly The SEM of pristine g-C3N4 in Fig. 2a shows a plate-like bulk
towards lower angles compared to those of pristine g-C3N4, structure with a considerably large size. With increasing TCNE

Fig. 2. SEM (a–d), TEM (e–h) images, and N2 adsorption-desorption isotherms (i) of g-C3N4, CN/TCNE-20, CN/TCNE-40 and CN/TCNE-120.
64 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69

content, the plate-like bulk structure of CN/TCNE became smaller analyses of the samples, indicating that new states of carbon atoms
and thinner, as shown in Fig. 2b–d. From the TEM image in Fig. 2e, were not observed in the CN/TCNE samples. This showed that the
it could be observed that pristine g-C3N4presented typical two- TCNE copolymerization reaction did not introduce a new carbon
dimensional, sheet-like large nanosheetswith a smooth and intact species into the structure of g-C3N4, but led to a decrease in the N
surface. However, as shown in the TEM images in Fig. 2f–h, when content. On the other hand, the relative content of different states
the TCNE content increased, the size of g-C3N4 nanosheets became of N in g-C3N4 and CN/TCNE-40 were calculated by N1 s spectra
smaller and more porous structure than that of pristine g-C3N4 was showed the four peaks at 398.5 eV, 399.8 eV, 401 eV and 404.3 eV
formed, indicating that the TCNE copolymerization reaction led to with peak area ratio of 1:0.208:0.111:0.042 and
nanosheet fragmentation and generated pores on the surface of g- 1:0.236:0.107:0.034 for g-C3N4 and CN/TCNE-40, respectively,
C3N4. This result was consistent with the XRD analysis. indicating that the content of

NH2 in CN/TCNE-40 was less
The N2 adsorption–desorption isotherms of pristine g-C3N4, CN/ than that ing-C3N4 [40]. This further confirmed that the CN/TCNE
TCNE-20, CN/TCNE-40, and CN/TCNE-120 are presented in Fig. 2i. surface was poor in nitrogen, that is, carbon-rich g-C3N4 could be
According to the IUPAC classification, the N2 adsorption–desorp- synthesized by thermal copolymerization of urea and TCNE.
tion isotherms of all samples are type IV with obvious hysteresis According to the above discussion, we proposed the possible
hoops of type H3, indicating the presence of mesopores associated copolymerization processes of urea with TCNE at high tempera-
with the sheet-like interlayer character of g-C3N4 [35]. As shown in ture. During the thermal condensation process of g-C3N4 forma-
Table 1, the specific surface areas and pore volumes of pristine g- tion, a transition unit containing three polycyclic aromatic
C3N4, CN/TCNE-20, CN/TCNE-40, and CN/TCNE-120 were 54.59, hydrocarbons with amino groups, tri-s-triazine, was produced.
57.84, 81.56, and 118.34 m2/g and 0.28, 0.31, 0.57, and 0.76 m3/g, As proposed in Fig. 4a, a tri-s-triazine ring with three amino groups
respectively. Thus, the surface areas and pore volumes of the CN/ was formed by urea without TCNE at high temperature. Because
TCNE samples were higher than those of pristine g-C3N4, cyano groups of TCNE could integrate directly into the carbon
suggesting that the porosity of the samples could be adjusted nitride polymers, the amino group of the tri-s-triazine ring was
by adding a small amount of TCNE, which was in accordance with replaced by TCNE when urea copolymerized with TCNE, leading to
the SEM and TEM analyses. the formation of a C

C bond attached to the tri-s-triazine ring. The
The chemical compositions of pristine g-C3N4 and CN/TCNE-40 C

C bond was less stable than the C

N bond because its bond
were further investigated by solid-state 13C-cross-polarisation length was greater (1.42 Å vs. 1.35 Å); therefore, the C

C bond
NMR and XPS. As illustrated in Fig. 3a, the 13C NMR spectra of g- attached to the tri-s-triazine ring might break at high temperature
C3N4 and CN/TCNE-40 displayed two distinct peaks at around 156 during copolymerization, leading to amino group vacancies
and 165 ppm, which could be assigned to the C

(N)3 and C

NHx (Fig. 4b). The absence of the amino group, which acted as bridging
in the g-C3N4 framework, respectively [36]. This indicated the links between the tri-s-triazine rings to form g-C3N4, decreased the
presence of a characteristic poly-structure of tri-s-triazine in g- connectivity of the tri-s-triazine ring in the polymerization process
C3N4 and CN/TCNE-40, which was in good agreement with the bulk (Fig. 4c), and the formation of nitrogen-poor and carbon-rich g-
g-C3N4 synthesized via a high-temperature copolymerization C3N4, resulted in the fracture of the g-C3N4network structure,
method. It could be seen from Fig. 3b that sharp photoelectron leading to some defects in the CN/TCNE samples, which in turn led
peaks at binding energies of about 285 eV and 399 eV could be to the generation of more pores and a larger surface area.
ascribed to C 1 s and N 1s, respectively, indicating that the samples
were primarily composed of carbon and nitrogen [31]. Three peaks 3.2. Optical absorption and charge separation
at 284.6, 286.5, and 288.1 eV could be found for C 1s (Fig. 3c): the
first peak was related to the C

C bonds of graphitic or Fig. 5a shows the UV–vis diffuse reflectance spectra (DRS) and
adventitious carbons, the second peak was attributed to C

(N)3 photos of pristine g-C3N4 and the CN/TCNE samples. g-C3N4pre-
bonds, and the last peak was a typical signal for sp2-hybridized sented an absorption threshold at 461 nm, associated with a band
carbon, determined as the C

N¼C backbone of g-C3N4 [32,37]. As gap energy of 2.69 eV, which was in good agreement with the
shown in Fig. 3d, the N1s spectrum could be fitted into four peaks. reported value of 2.7 eV [5]. The absorption thresholds of CN/
The main peaks centred at 398.8 eV and 400 eV could be assigned TCNE-10, CN/TCNE-20, CN/TCNE-40, CN/TCNE-60, CN/TCNE-120,
to the sp2-hybridized nitrogen (C

N¼C) in the triazine rings and and CN/TCNE-240 were 471, 495, 506, 523, 539, and 606 nm and
the bridging nitrogen atoms (C)3

N, respectively [38], whereas their band gap energies were then evaluated to be 2.63, 2.51, 2.45,
the additional weak peak at 404.3 eV could be attributed to the 2.37, 2.30, and 2.05 eV, respectively. As shown in the inset, pristine
amino functional groups with hydrogen (C

N

H) and might be g-C3N4 was bright yellow, and the copolymerization reaction
associated with incomplete condensation [39]. changed the colour to dark yellow and brown with increasing TCNE
The XPS analysis also revealed the molar ratio of C/N for g-C3N4 content. The redshift could be closely related with more nitrogen
and the CN/TCNE samples, as shown in Table 1. The C/N ratio deficiency in g-C3N4. The nitrogen deficiency in g-C3N4 left
gradually increased with increasing TCNE content, from 0.695 to additional electrons, which in turn led to the presence of abundant
0.697, 0.709, and 0.727 for g-C3N4, CN/TCNE-10, CN/TCNE-40, and nitrogen vacancy-related C3+ states in the band gap, thus
CN/TCNE-120, respectively. This increasing C/N ratio could be decreasing the intrinsic band gap from 2.69 to 2.05 eV [40,41].
explained by the increased C or decreased N in the CN/TCNE The results suggested that TCNE copolymerization could effective-
samples. However, peaks corresponding to the olefinic bond and ly narrow the band gap and increase the absorption in the visible-
acetylene bond were not observed in the FTIR, NMR, and XPS light region and was anticipated to improve the photoexcitation

Table 1
The BET surface area, pore volume, average pore size, and molar ration of pristine g-C3N4, CN/TCNE-20, CN/TCNE-40 and CN/TCNE-120 samples.

sample BET surface area (m2/g) Pore volume (cm3/g) Average pore size (nm) Molar ration of C/N
g-C3N4 54.59 0.28 20.55 0.695
CN/TCNE-20 57.84 0.31 20.98 0.697
CN/TCNE-40 81.56 0.57 25.79 0.709
CN/TCNE-120 118.34 0.76 27.89 0.727
 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69 65

Fig. 3. NMR (a) and XPS spectra of g-C3N4 and CN/TCNE-40, (b) survey (c) C1 s (d) N1s.

PL spectral emission was caused by the recombination of

photogenerated electron–hole pairs and was therefore useful to
probe the transfer and recombination processes of these photo-
generated carriers [45]. As shown in Fig. 5d, obviously, a broad
visible PL band could be observed for the CN/TCNE samples,
indicating that the electron–hole pairs were produced upon
absorption of photons, which is a crucial step to initiate the
photocatalytic reaction. The intensity of the PL signals gradually
decreased with increasing TCNE content, indicating a decrease in
the recombination rate of the photogenerated charge carriers. It
was suggested that the energy-wasting carrier recombination
would be suppressed by the copolymerization with TCNE under
visible-light irradiation. The improved charge separation efficiency
could be attributed to the increased pore volume and surface area,
and it facilitated the transfer and separation of photogenerated
charge carriers, prolonged their lifetime, and thus lowered the PL
Fig. 4. The possible existence forms in copolymerization processes of urea with
TCNE at high temperature: (a) only urea, (b) urea with TCNE, (c) g-C3N4 fragment.
intensity [46]. Moreover, the emission peak in the PL spectrum of
the CN/TCNE samples gradually shifted towards longer wave-
lengths with increasing TCNE content, which was consistent with
efficiency of the CN/TCNE samples, thereby enhancing the photo- the narrowing of the band gap in the UV–vis DRS analysis.
activity to a certain extent [42]. The inhibited recombination of active charge carries could be
Valence-band (VB) XPS was performed to analyse the VB further confirmed by the measurement of their lifetime. It could be
potential (Fig. 5b). For g-C3N4, CN/TCNE-40, and CN/TCNE-120, the seen from Fig. 5e that both spectra decayed exponentially and that
VB potentials were 1.74 eV, 1.67 ev, and 1.54 eV, respectively, which the fluorescent decay kinetics of CN/TCNE-40 was slightly slower
is consistent with the literature [43,44]. Combined with their band than that of g-C3N4. The medium lifetimes were determined to be
gap, the conduction band (CB) level could be deduced. As 3.49 ns for g-C3N4 and 6.81 ns for CN/TCNE-40. Clearly, more
illustrated in Fig. 5c, with the increase of TCNE content, the VB charge carriers with a longer lifetime were present in CN/TCNE-40
edges increase and the CB edges decrease, the determined relative than in g-C3N4, helping reduce the probability of recombination of
CB edges of the samples for g-C3N4, CN/TCNE-40, and CN/TCNE-120 the photoelectrons and holes and increase the accessibility of the
were
0.95 ev,
0.76,
0.51, respectively. It was clear that the charge carriers to the reactants [11]. To further investigate the
electronic structure of the CN/TCNE samples could be tailored by charge carrier transfer behaviour of the samples, the transient
adjusting the amount of TCNE. It was beneficial to reduce the photocurrent responses (I-t) was carried out, as shown in Fig. 5f.
excitation energy of photocatalysts, thereby considerably improv- The photocurrent reached its peak value and obtains almost
ing the light efficiency [36]. unchanged upon the light on and returns to the background level
66 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69

Fig. 5. (a) UV–vis DRS and photos (inset), (b) High resolution VB XPS spectra, (c) schematic band structures, (d) photoluminescence spectra emission, (e) time-resolved
fluorescence decay spectra, and (f) photocurrent spectra of the g-C3N4 and CN/TCNE samples.

once the light was switched off. The g-C3N4 shows a small where k is the degradation rate constant and c0 and c are the
photocurrent and the CN/TCNE-40 exhibits a higher photocurrent. initial concentration and the concentration at reaction time t,
These results suggested that the copolymerization of TCNE urea respectively. The k values of the different samples are shown in
and could efficiently inhibit the recombination of electron-hole Fig. 6b. The CN/TCNE-40 sample has the highest rate constant of
pairs and enhance photocatalytic activity [47]. k = 0.0317 min
1, which is almost 4.88 times as large as that of
pristine g-C3N4 (k = 0.0065 min
1). The enhanced photocatalytic
3.3. Photocatalytic activity and stability activities could be ascribed to the synergistic effect of the increased
surface area and pore volume, adjusted band structure, enhanced
The photocatalytic activity of g-C3N4 and the CN/TCNE samples visible-light absorption capacity, and increased electron–hole
was evaluated by decomposing Orange II under visible-light separation efficiency. However, on increasing the TCNE content
irradiation (l > 420 nm). As demonstrated in Fig. 6a, Orange II beyond 40 mg, the photocatalytic activity of the CN/TCNE samples
could be degraded in the presence of each sample; the degradation decreased. This may be because copolymerization with excessive
rates of g-C3N4, CN/TCNE-10, CN/TCNE-20, CN/TCNE-40, CN/TCNE- TCNE generated more defects in the carbon nitride network, which
60, CN/TCNE-120, and CN/TCNE-240 were 50.88%, 61.53%, 84.99%, changed its semiconductor character and offered trapping centres
90.16%, 80.01%, 78.16%, and 72.04%, respectively. Compared to for photogenerated electrons. Therefore, selecting an appropriate
pristine g-C3N4, all as-fabricated CN/TCNE samples exhibited amount of TCNE is very important to improve photocatalytic
enhanced photocatalytic activity, and the optimal sample was CN/ activity.
TCNE-40 and its degradation rate was much higher than that of The catalytic stability of a photocatalyst is essential to ensure its
pristine g-C3N4. The experimental data were analysed and the practical application, especially for degradation of dyes in
photocatalytic reaction kinetics could be expressed as follows: wastewater of different pH values. As shown in Fig. 6c, Orange II
could be degraded at various pH values by CN/TCNE-40 under
ln(c0/c) = kt
visible-light irradiation, and the efficiency of photocatalytic
 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69 67

Fig. 6. (a) photocatalytic degradation rate and (b) pseudo-first-order kinetic constant of Orange II by g-C3N4 and CN/TCNE samples, (c) photocatalytic degradation of Orange II
by CN/TCNE-40 at various pH, (d) cycling photocatalytic degradation efficiency of Orange II by CN/TCNE-40 under visible-light, (e) trapping experiments of active species in
the photocatalytic process, (f) photocatalytic degradation mechanism. (For interpretation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)

degradation strongly depended on the pH value of the solution. 3.4. Possible photocatalytic mechanism
The best degradation was achieved at pH 2.58, which might have
been due to the strongest adsorption of Orange II on the surface of To understand the photocatalytic mechanism of the CN/TCNE
g-C3N4. However, the reason was that Orange II is negatively composites, trapping experiments were performed to determine
charged when dissolved in an aqueous solution, whereas g-C3N4 is the main active species in the photocatalytic process (Fig. 6e) using
positively charged by protonation when the pH value is low 1,4-benzoquinone (BQ), tert-butyl alcohol (TBA), and disodium
(pH = 2.58) [48]; this electrostatic attraction played a key role in ethylenediaminetetraacetate (EDTA-2Na) to capture superoxide
favouring the adsorption. If the pH value of the solution increased, radicals (O2
), hydroxyl radicals (OH), and holes (h+), respectively
the surface positive charge decreased, which weakened the [49,50]. Once BQ was employed, the degradation rate was
electrostatic attraction to Orange II, leading to a decrease in significantly suppressed, indicating that O2
radicals were the
absorption. However, further increasing the pH to 11.3, the main reactive species in our experiment. It was found that the
efficiency of the photocatalytic degradation showed an increasing introduction of TBA could decrease the degradation rate to some
trend again, possibly caused by the increase in the hydroxyl extent, implying that a few OH radicals might also exist in the
radicals (OH) generated in the alkaline solution. Fig. 6d shows that system. When EDTA was introduced in the system to capture h+,
the CN/TCNE-40 sample maintained its high photoactivity for the the degradation rate of Orange II slightly increased. The hole
Orange II photodegradation after five cycles under identical trapping would inhibit the recombination of the photogenerated
experimental conditions, confirming the high durability of the electron–hole pairs, which is beneficial for the production of
photocatalyst. photogenerated electrons. The photogenerated electrons would
68 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69

interact with the dissolved oxygen to formO2
radicals, further [9] S. Xu, P. Zhou, Z. Zhang, C. Yang, B. Zhang, K. Deng, S. Bottle, H. Zhu, Selective
confirming that Orange II was mainly degraded by O2
radicals. oxidation of 5-hydroxymethylfurfural to 2,5-furandicarboxylic acid using O2
and a photocatalyst of Co-thioporphyrazine bonded to g-C3N4, J. Am. Chem.
Based on the above discussion, the possible mechanism of Soc. 139 (2017) 14775–14782.
Orange II photodegradation by CN/TCNE could be described as [10] Y. Zhang, Q. Zhang, Q. Shi, Z. Cai, Z. Yang, Acid-treated g-C3N4 with improved
follows (Fig. 6f). First, Orange II was absorbed on the CN/TCNE photocatalytic performance in the reduction of aqueous Cr(VI) under visible-
light, Sep. Purif. Technol. 142 (2015) 251–257.
composites by their enlarged surface area and improved porous [11] Y. Li, Y. Jiang, Z. Ruan, K. Lin, Z. Yu, Z. Zheng, X. Xu, Y. Yuan, Simulation-guided
structure. Second, under visible-light irradiation, the photogen- synthesis of graphitic carbon nitride beads with 3D interconnected and
erated holes of the CN/TCNE samples would oxidize water into continuous meso/macropore channels for enhanced light absorption and
photocatalytic performance, J. Mater. Chem. A 5 (2017) 21300–21312.
oxygen (O2), and the electrons on the VB became excited and [12] R. Sun, Q. Shi, M. Zhang, L. Xie, J. Chen, X. Yang, M. Chen, W. Zhao, Enhanced
transferred to the CB of the CN/TCNE samples, interacting with photocatalytic oxidation of toluene with a coral-like direct Z-scheme BiVO4/g-
dissolved O2 to form O2
radicals. Finally, the generated O2
C3N4 photocatalyst, J. Alloys Compd. 714 (2017) 619–626.
[13] Q. Han, B. Wang, J. Gao, Z. Cheng, Y. Zhao, Z. Zhang, L. Qu, Atomically thin
radicals with a powerful oxidation ability could react with Orange
mesoporous nanomesh of graphitic C3N4 for high-Efficiency photocatalytic
II and produce degradation products (CO2, H2O). Therefore, TCNE- hydrogen evolution, ACS Nano 10 (2016) 2745–2751.
copolymerized g-C3N4 with good stability could be used as an [14] Q. He, F. Zhou, S. Zhan, Y. Yang, Y. Liu, Y. Tian, N. Huang, Enhancement of
effective photocatalyst in solutions having different pH values for photocatalytic and photoelectrocatalytic activity of Ag modified Mpg-C3N4
composites, Appl. Surf. Sci. 391 (2017) 423–431.
organic compound degradation. [15] S. Tonda, S. Kumar, S. Kandula, V. Shanker, Fe-doped and –mediated graphitic
carbon nitride nanosheets for enhanced photocatalytic performance under
4. Conclusion natural sunlight, J. Mater. Chem. A 2 (2014) 6772.
[16] J. Yan, H. Wu, H. Chen, Y. Zhang, F. Zhang, S.F. Liu, Fabrication of TiO2/C3N4
heterostructure for enhanced photocatalytic Z-scheme overall water splitting,
In summary, for the first time, carbon-rich CN/TCNE photo- Appl. Catal. B: Environ. 191 (2016) 130–137.
catalysts were successfully synthesized via a facile one-pot [17] M. Mousavi, A. Habibi-Yangjeh, M. Abitorabi, Fabrication of novel magnetically
separable nanocomposites using graphitic carbon nitride, silver phosphate
thermally-induced copolymerization strategy by introducing a and silver chloride and their applications in photocatalytic removal of different
small amount of TCNE into urea. Compared to pristine g-C3N4, the pollutants using visible-light irradiation, J. Colloid Interface Sci. 480 (2016)
resultant photocatalyst exhibited superior photocatalytic activity 218–231.
[18] M. Mousavi, A. Habibi-Yangjeh, Magnetically recoverable highly efficient
for Orange II degradation under visible light, and the optimal TCNE
visible-light-active g-C3N4/Fe3O4/Ag2WO4/AgBr nanocomposites for
content was determined to be 40 mg in 30 g urea. This enhanced photocatalytic degradations of environmental pollutants, Adv. Powder
photocatalytic activity of the CN/TCNE samples could be attributed Technol. 29 (2018) 94–105.
[19] M. Mousavi, A. Habibi-Yangjeh, Magnetically separable ternary g-C3N4/Fe3O4/
to the synergistic effect of the increased surface area and pore
BiOI nanocomposites: novel visible-light-driven photocatalysts based on
volume, narrowed band structure, improved visible-light utiliza- graphitic carbon nitride, J. Colloid Interface Sci. 465 (2016) 83–92.
tion, and promoted separation of charge carriers. The CN/TCNE-40 [20] Mitra Mousavi, Aziz Habibi-Yangjeh, Shima Rahim Pouran, Review on
sample maintained its high photoactivity after five cycles under magnetically separable graphitic carbon nitride-based nanocomposites as
promising visible-light-driven photocatalysts, J. Mater. Sci.: Mater. Electron. 29
identical experimental conditions and could degrade Orange II at (2018) 1719–1747.
various pH values under visible-light irradiation, confirming the [21] F. Dong, Z. Wang, Y. Li, W.-K. Ho, S.C. Lee, Immobilization of polymeric g-C3N4
high durability of the photocatalyst. The study provides CN/TCNE on structured ceramic foam for efficient visible light photocatalytic air
purification with real indoor illumination, Environ. Sci. Technol. 48 (2014)
photocatalysts fabricated by a facile method for the degradation of 10345–10353.
organic pollutants, which is valuable for the preparation of novel [22] X. Fan, L. Zhang, R. Cheng, M. Wang, M. Li, Y. Zhou, J. Shi, Construction of
metal-free semiconductors for photocatalytic application. graphitic C3N4-based intramolecular donor–acceptor conjugated copolymers
for photocatalytic hydrogen evolution, ACS Catal. 5 (2015) 5008–5015.
[23] L. Song, C. Guo, T. Li, S. Zhang, C60/graphene/g-C3N4 composite photocatalyst
Competing interest and mutually- reinforcing synergy to improve hydrogen production in
splitting water under visible light radiation, Ceram. Int. 43 (2017) 7901–7907.
[24] J. Zhang, X. Chen, K. Takanabe, K. Maeda, K. Domen, J.D. Epping, X. Fu, M.
The authors declare no competing financial interest. Antonietti, X. Wang, Synthesis of a carbon nitride structure for visible-light
catalysis by copolymerization, Angew. Chem. Int. Ed. 49 (2010) 441–444.
Acknowledgments [25] J. Liu, Effect of phosphorus doping on electronic structure and photocatalytic
performance of g-C3N4: insights from hybrid density functional calculation, J.
Alloys Compd. 672 (2016) 271–276.
This work was supported by the National Science Foundation of [26] W. Ho, Z. Zhang, W. Lin, S. Huang, X. Zhang, X. Wang, Y. Huang,
China (Grant No. 61434002), the Special Funds of Sanjin Scholars Copolymerization with 2,4,6-Triaminopyrimidine for the rolling-up the
Program. layer structure, tunable electronic properties, and photocatalysis of g-C3N4,
ACS Appl. Mater. Interfaces 7 (2015) 5497–5505.
[27] J. Zhang, G. Zhang, X. Chen, S. Lin, L. Möhlmann, G. Dołe˛ga, G. Lipner, M.
Antonietti, S. Blechert, X. Wang, Co-monomer control of carbon nitride
References semiconductors to optimize hydrogen evolution with visible light, Angew.
Chem. Int. Ed. 51 (2012) 3183–3187.
[1] H. Sun, S. Wang, Research advances in the synthesis of nanocarbon-based [28] Z. Zhao, Y. Sun, F. Dong, Y. Zhang, H. Zhao, Template synthesis of carbon self-
photocatalysts and their applications for photocatalytic conversion of carbon doped g-C3N4 with enhanced visible to near-infrared absorption and
dioxide to hydrocarbon fuels, Energy & Fuels 28 (2014) 22–36. photocatalytic performance, RSC Adv. 5 (2015) 39549–39556.
[2] M.N. Chong, B. Jin, C.W.K. Chow, C. Saint, Recent developments in [29] G. Dong, K. Zhao, L. Zhang, Carbon self-doping induced high electronic
photocatalytic water treatment technology: a review, Water Res. 44 (2010) conductivity and photoreactivity of g-C3N4, Chem. Commun. 48 (2012) 6178.
2997–3027. [30] X. Wang, X. Chen, A. Thomas, X. Fu, M. Antonietti, Metal-containing carbon
[3] X. Lang, X. Chen, J. Zhao, Heterogeneous visible light photocatalysis for nitride compounds: a new functional organic-metal hybrid material, Adv.
selective organic transformations, Chem. Soc. Rev. 43 (2014) 473–486. Mater. 21 (2009) 1609–1612.
[4] X. Chen, S. Shen, L. Guo, S.S. Mao, Semiconductor-based photocatalytic [31] Q. Xiang, J. Yu, M. Jaroniec, Preparation and enhanced visible-light
hydrogen generation, Chem. Rev. 110 (2010) 6503–6570. photocatalytic H2-production activity of graphene/C3N4 composites, J. Phys.
[5] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M. Chem. C 115 (2011) 7355–7363.
Antonietti, A metal-free polymeric photocatalyst for hydrogen production [32] Y. Li, Y. Sun, F. Dong, W.-K. Ho, Enhancing the photocatalytic activity of bulk g-
from water under visible light, Nat. Mater. 8 (2009) 76–80. C3N4 by introducing mesoporous structure and hybridizing with graphene, J.
[6] Y. Wang, X. Wang, M. Antonietti, Polymeric graphitic carbon nitride as a Colloid Interface Sci. 436 (2014) 29–36.
heterogeneous organocatalyst: from photochemistry to multipurpose [33] M.J. Bojdys, J.-O. Müller, M. Antonietti, A. Thomas, Ionothermal synthesis of
catalysis to sustainable chemistry, Angew. Chem. Int. Ed. 51 (2012) 68–89. crystalline, condensed graphitic carbon nitride, Chem.—Eur. J. 14 (2008) 8177–
[7] Z. Zhao, Y. Sun, F. Dong, Graphitic carbon nitride based nanocomposites: a 8182.
review, Nanoscale 7 (2015) 15–37. [34] Y. Zheng, Z. Zhang, C. Li, S. Proulx, Surface hydroxylation of graphitic carbon
[8] S. Ye, R. Wang, M.-Z. Wu, Y.-P. Yuan, A review on g-C3N4 for photocatalytic nitride: enhanced visible light photocatalytic activity, Mater. Res. Bull. 84
water splitting and CO2 reduction, Appl. Surf. Sci. 358 (2015) 15–27. (2016) 46–56.
 Z. Wang et al. / Journal of Photochemistry and Photobiology A: Chemistry 358 (2018) 61–69 69

[35] D. Tang, G. Zhang, Fabrication of AgFeO2/g-C3N4 nanocatalyst with enhanced [43] F. Dong, Z. Ni, P. Li, Z. Wu, A general method for type I and type II g-C3N4/g-C3N4
and stable photocatalytic performance, Appl. Surf. Sci. 391 (2017) 415–422. metal-free isotype heterostructures with enhanced visible light
[36] Y. Wang, H. Wang, F. Chen, F. Cao, X. Zhao, S. Meng, Y. Cui, Facile synthesis of photocatalysis, New J. Chem. 39 (2015) 4737–4744.
oxygen doped carbon nitride hollow microsphere for photocatalysis, Appl. [44] B. Chai, X. Liao, F. Song, H. Zhou, Fullerene modified C3N4 composites with
Catal. B: Environ. 206 (2017) 417–425. enhanced photocatalytic activity under visible light irradiation, Dalton Trans.
[37] G. Zhang, S. Zang, X. Wang, Layered Co(OH)2 deposited polymeric carbon 43 (2014) 982–989.
nitrides for photocatalytic water oxidation, ACS Catal. 5 (2015) 941–947. [45] L. Shi, L. Liang, F. Wang, M. Liu, K. Chen, K. Sun, N. Zhang, J. Sun, Higher yield
[38] F. Dong, Y. Sun, L. Wu, M. Fu, Z. Wu, Facile transformation of low cost thiourea urea-derived polymeric graphitic carbon nitride with mesoporous structure
into nitrogen-rich graphitic carbon nitride nanocatalyst with high visible light and superior visible-light-responsive activity, ACS Sustain. Chem. Eng. 3 (2015)
photocatalytic performance, Catal. Sci. Technol. 2 (2012) 1332. 3412–3419.
[39] G. Dong, L. Zhang, Porous structure dependent photoreactivity of graphitic [46] Z. Wang, W. Guan, Y. Sun, F. Dong, Y. Zhou, W.-K. Ho, Water-assisted production
carbon nitride under visible light, J. Mater. Chem. 22 (2012) 1160–1166. of honeycomb-like g-C3N4 with ultralong carrier lifetime and outstanding
[40] P. Niu, L.-C. Yin, Y.-Q. Yang, G. Liu, H.-M. Cheng, Increasing the visible light photocatalytic activity, Nanoscale 7 (2015) 2471–2479.
absorption of graphitic carbon nitride (Melon) photocatalysts by [47] Y. Li, S. Wu, L. Huang, J. Wang, H. Xu, H. Li, Synthesis of carbon-doped g-C3N4
homogeneous self-modification with nitrogen vacancies, Adv. Mater. 26 (2014) composites with enhanced visible-light photocatalytic activity, Mater. Lett.
8046–8052. 137 (2014) 281–284.
[41] W.-J. Ong, L.-L. Tan, Y.H. Ng, S.-T. Yong, S.-P. Chai, Graphitic carbon nitride (g- [48] Y. Zhang, A. Thomas, M. Antonietti, X. Wang, Activation of carbon nitride solids
C3N4)-based photocatalysts for artificial photosynthesis and environmental by protonation: morphology changes enhanced ionic conductivity, and
remediation: are we a step closer to achieving sustainability? Chem. Rev. 116 photoconduction experiments, J. Am. Chem. Soc. 131 (2009) 50.
(2016) 7159–7329. [49] E. Cheng, S. Zhou, M. Li, Z. Li, Synthesis of g-C3N4-based NaYF4:Yb, Tm@TiO2
[42] W.-J. Ong, L.-L. Tan, S.-P. Chai, S.-T. Yong, A.R. Mohamed, Surface charge ternary composite with enhanced Vis/NIR-driven photocatalytic activities,
modification via protonation of graphitic carbon nitride (g-C3N4) for Appl. Surf. Sci. 410 (2017) 383–392.
electrostatic self-assembly construction of 2D/2D reduced graphene oxide [50] W.J. Kim, E. Jang, T.J. Park, Enhanced visible-light photocatalytic activity of ZnS/
(rGO)/g-C3N4 nanostructures toward enhanced photocatalytic reduction of g-C3N4 type-II heterojunction nanocomposites synthesized with atomic layer
carbon dioxide to methane, Nano Energy 13 (2015) 757–770. deposition, Appl. Surf. Sci. 419 (2017) 159–164.