Chemosphere 239 (2020) 124713

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Photocatalytic behavior of biochar-modified carbon nitride with

enriched visible-light reactivity
Lirong Meng a, Wenhua Yin a, Shengsen Wang a, b, c, Xiaoge Wu a, c, Jianhua Hou a, c,
Weiqin Yin a, c, Ke Feng a, b, c, Yong Sik Ok d, *, Xiaozhi Wang a, b, c, **
a
College of Environmental Science and Engineering, Yangzhou University, Jiangsu, 225127, China
b
Joint International Research Laboratory of Agriculture and Agri-Product Safety of Ministry of Education of China, Yangzhou University, Yangzhou, 225127,
Jiangsu, China
c
Jiangsu Collaborative Innovation Center for Solid Organic Waste Resource Utilization, Nanjing, 210095, China
d
Korea Biochar Research Center & Division of Environmental Science and Ecological Engineering, Korea University, Seoul, 02841, Republic of Korea

h i g h l i g h t s

The p-conjugated electron systems from biochar (BC) elevate charge separation efficiency.

BC narrowed the band gap and strengthened the visible light response of g-C3N4.

BC, as an electron-acceptor, hinders charge recombination through a well-contacted interface.

Nano-layered structure of g-C3N4 decreases the recombination of the e-hþ pairs by shortening the diffusion pathway.

a r t i c l e i n f o a b s t r a c t :

Article history: Ultra-thin layered structures and modified bandgaps are two efficient strategies to increase the photo-
Received 17 July 2019 catalytic performance of various materials for the semiconductor industry. In the present study, we
Received in revised form combined both strategies in one material to form carbon-doped graphitic carbon nitride (g-C3N4) nano-
27 August 2019
layered structures by the method of melamine thermal condensation, in the presence of different mass
Accepted 28 August 2019
Available online 31 August 2019
ratios of biochar. The characterization showed that the composite with the best ratio retained the g-C3N4
polymeric framework and the bond with g-C3N4. The biochar was established via pep stacking in-
Handling Editor: J. Rinklebe teractions and ether bond bridges. The p-conjugated electron systems provided from biochar can elevate
charge separation efficiency. The ultra-thin structure also curtailed the distance of photogenerated
Keywords: electrons migrating to the surface and enlarge specific surface area of materials. The presence of carbon
Carbon-doped g-C3N4 narrowed the bandgap and increased light absorption at a wider range of wavelengths of g-C3N4. The
Biochar biochar/melamine ratio of 1:15 presented the best performance, 2.8 and 5 times faster than g-C3N4
Nanocomposite degradating Rhodamine and Methyl Orange, respectively. Moreover, the catalyst presented a good sta-
Photocatalysis
bility for 4 cycles. In addition to that, biochar from waste biomass can be considered a sustainable, cost-
Visible light
effective, and efficient option to modify geC3N4ebased photocatalysts.
© 2019 Elsevier Ltd. All rights reserved.

1. Introduction energy generation can be relieved by photocatalysis as it uses solar

energy to degrade pollutants, product hydrogen and photo-
The growing demand of purifying polluted water and clean reduction CO2, which is considered a sustainable process (Zhang
et al., 2018; Hao et al., 2019). A great deal of semiconductors such
as TiO2 and ZnO show a good photocatalytic performance under
* Corresponding author. Division of Environmental Science and Ecological Engi-
ultraviolet light, but this light source is not considered renewable
neering, Korea University, Seoul, 02841, Republic of Korea. (Grover et al., 2017). Recently, the low cost graphitic carbon nitride
** Corresponding author. College of Environmental Science and Engineering, (g-C3N4) has become attractive owing to visible light response and
Yangzhou University, Yangzhou, 225127, China. low bandgap energy (~2.8 eV) and low cost (Wang et al., 2012).
E-mail addresses: [email protected] (Y.S. Ok), [email protected]
Nevertheless, the practical photocatalytic applications of g-C3N4 are
(X. Wang).

https://doi.org/10.1016/j.chemosphere.2019.124713
0045-6535/© 2019 Elsevier Ltd. All rights reserved.
2 L. Meng et al. / Chemosphere 239 (2020) 124713

limited by the rapid recombination of photo-induced electron-hole analytical grade, and deionized (DI) water was also used.
pairs and the weak visible light response. Therefore, elevating
photo-generated electron transfer efficiency and more sensitive
visible-light response are two pivotal factors to optimize the optical 2.2. Synthesis
performance of g-C3N4.
One g-C3N4 modification method is preparing ultra-thin nano- BC was obtained via pyrolysis of rice straw under nitrogen gas.
sheets to increase the active sites for degradation of contaminants The rice straw was washed with DI water, dried at 80  C, and
and the efficiency of photo-induced charges reaching the surface of crushed. Next, the processed biomass was inserted in a tube
g-C3N4. (Yu et al., 2019). For example, Su and Zhang used melamine furnace and heated from 20 to 600  C at a 10  C min1 rate. After
(M) and oxalic acid as starting reagents to synthesize g-C3N4 porous that, the temperature was held constant for 2 h under inert con-
nanosheets by means of a two-step thermal process to enhance ditions. Then, the furnace was cooled to room temperature. After
photocatalytic capacity (Su and Zhang, 2017). As another example, which, the cooled BC was ground.
Huang et al. prepared g-C3N4 nanosheets with porous nanosheet The composites were prepared through thermal condensation.
structure which showed profoundly enhanced visible-light photo- BC and M was ground evenly at different mass ratios, denoted as 1/
catalytic performance (Huang et al., 2017a). Another approach to X BC/M (X ¼ 5, 10, 15, and 30). The mixture was transferred into the
improve photocatalytic capacity is to dope the catalyst with tube furnace in a ceramic boat and heated from 20 to 520  C at
nonmetallic elements, such as carbon (Hou et al., 2018). Elemental 2.3  C min1 and maintained for 4 h. The resultant agglomerates
doping is an effective strategy to enhance polarization, and thus to were ball milled and further heated from 20 to 620  C for 3 h at 5  C
increase the bulk or surface charge separation (Chen et al., 2019). min1. The furnace was cooled to room temperature and the
Due to the p-conjugation of graphene, the combination of g-C3N4 product was magnetic stirred for 6 h in 30% HCl, and then filtered
and carbon-based materials has been adopted to change its charge- by a vacuum pump and subsequently washed with DI water till the
separating and optical characteristics, resulting in significant ad- pH of filtrate was neutral. Afterwards, the product was oven dried
vances in photocatalytic performance (Xu et al., 2017). For instance, at 60  C for 12 h to obtain BC/M samples. Pure g-C3N4 was also
Yu et al. used graphene to modify g-C3N4 to enhance activity under synthesized with the same method using M as precursor.
visible light irradiation (Yu et al., 2014). However, the synthetic
route of graphene is complex and expensive for large-scale appli-
cations. In the study by Pi et al., a leaf-derived biochar (BC) was 2.3. Characterization
modified with g-C3N4. However, it exhibited a mild and slow
removal of the target pollutant, methyl orange (4 h) (Pi et al., 2015). Crystal structures were identified by X-ray diffraction (XRD)
That was probably because BC leads to limited g-C3N4/carbon (AXS D8 ADVANCE, Bruker) using Cu Ka radiation (1.5406 Å) at 6
contact interface. In addition, BC is not activated by light, and bulk min1 and 10e80 . Fourier transform infrared spectroscopy (FT-IR)
BC can cover the light irradiation. was conducted using a TENSOR 27 spectrophotometer (Bruker,
Herein, we used more readily available BC to provide p-conju- Germany). X-ray photoelectron spectroscopy (XPS) (ESCALAB
gation structure for geC3N4ebased composites to change the 250Xi, Thermo Fisher) was employed to determine surface element
charge-separating efficiency and optical character of g-C3N4. It is states. Microstructure was explored through S4800 scanning (SEM)
reported that combining g-C3N4 with carbonaceous materials such (Hitachi) and high-resolution transmission electron microscopy
as carbon nanotubes, graphene and carbon fibre could improve the (HRTEM) (Tecnai G2 F30 S-Twin, FEI). UVevis diffuse reflectance
conductivity and photoelectrochemical properties of g-C3N4, which spectra (DRS) (TU-1901, PERSEE) were obtained to study optical
promotes the more appropriate band gap structures with excellent characteristics. Photoluminescence (PL) spectra equipped with a Xe
charge carrier separation (Christoforidis et al., 2018; Chen et al., lamp at an excitation wavelength of 325 nm were performed on
2016; Mohamed et al., 2018). Meanwhile, two-step calcination of Hitachi F-4500 fluorescence spectrophotometer. The specific sur-
M mixed with BC was performed to reduce the nanometer size of face area was estimated by the BrunauereEmmetteTeller (BET)
the material by pyrolysis, so that the obtained material would have (ASAP 2460, Micromeritics) method.
more abundant well-contacted g-C3N4/carbon interface. In addi-
tion, carbon-rich materials can act as photo-generated electron
acceptors, effectively reducing the recombination rate of photo- 2.4. Photocatalytic activity evaluation
generated electron-hole pairs. In this study, BC was prepared by
pyrolysis of rice straw, which is a widely available material (Singh Photocatalysis activity was tested by degrading RhB and MO at
et al., 2017). In this case, modification of photocatalytic material room temperature under visible light in an XPA-system photo-
by using BC of rice straw provides a sustainability advantage with chemical reactor (XPA-7, Xujiang Instrument, Nanjing). The irradi-
respect to waste biomass, in addition to a p-conjugation structure ation setup included a 500 W Xe lamp, with continuously flowing
for the geC3N4ebased composites. The aim of this study was to find cooling water between the lamps so that there was no heat effect
the optimum BC doping ratio to obtain a photocatalyst with optimal produced during photocatalysis. For the experiments, 50 or 20 mg
performance, and investigate the influence of BC content on pho- photocatalysts were added to the 50 mL and 10 mg L1 RhB and MO
tocatalytic abilities for the degradation of organic dyes. solution. To attain adsorption-desorption equilibrium, the mixture
was first magnetically stirred under dark conditions for 1.5 h. After
2. Experimental section pre-absorption, the visible light source was switched on, and sub-
samples were taken out at time intervals (5, 10, 15, 30, 60 and
2.1. Materials 90 min) and immediately centrifuged at 8000 r min1 for 5 min to
obtain liquid samples without solid catalysts. The concentrations of
Melamine(C3N3(NH2)3), hydrochloric acid (HCl), p-benzoqui- RhB and MO in the remaining clear liquid was measured by a visible
none (BQ), isopropyl alcohol (IPA), and triethanolamine (TEOA), spectrophotometer (722G, JINGKE, Shanghai) at the wavelength of
rhodamine (RhB) and methyl orange (MO) were bought from 553 nm and 464 nm, respectively. The regeneration of the synthe-
Sinopharm Chemical Reagents Co., Ltd. Rice straw was obtained sized catalyst was studied using the same batch of catalyst over
from the Lianyungang countryside, Jiangsu. The reagents were of multiple cycles.
 L. Meng et al. / Chemosphere 239 (2020) 124713 3

3. Results and discussion XPS. Fig. 3a shows XPS survey spectra, highlighting C, N and O,
where 1/15 BC/M composite has higher content of O which is
3.1. X-ray diffraction patterns probably introduced by BC. C 1s high resolution XPS spectra was
deconvoluted into two peaks, displaying binding energies of 284.6
The crystalline structure of BC/M composites and interactions and 287.9 eV (Fig. 3b) (Hou et al., 2017). For g-C3N4, the peak at
between carbon and g-C3N4 were investigated by XRD pattern 287.9 eV corresponds to the sp2 carbon in the nitrogen-containing
(Fig. 1). There are two typical diffraction peaks of g-C3N4. One aromatic ring (NeC]N), which, in g-C3N4, was considered the
intensive peak at about 27.7 is a typical (002) plane peak due to the major carbon species (Guo et al., 2016). Compared with g-C3N4, the
aromatic interlayer stacking in g-C3N4 (Yan et al., 2009). The small sp2 hybridized carbon position of 1/15 BC/M composite showed a
diffraction peak near 12.8 represents (100) plane, which is slight positive shift from 287.9 to 288.0 eV, which suggested the
ascribed to the in-plane structure of the aromatic system (Liu et al., fact that in all probability, BC embedding in the g-C3N4 framework,
2011). Compared with pure g-C3N4 and BC, the peak near 21.7 of forming the interactions between carbon and g-C3N4 (Li et al.,
the 1/X BC/M composite is increasingly intense in line with 2014). The 284.6 eV peak is attributed to adventitious carbon spe-
increased carbon content, which means that the composite was cies and/or typically ascribed to the C bond (CC) of the sp2 plane
successfully doped by carbon (Hou et al., 2018). After further graphene on the surface of the samples (Lu et al., 2016). For 1/15 BC/
observation (Fig. 1 inset), the (002) diffraction peak shifted slightly M composite, the peak area at 284.6 eV was significantly larger than
from 27.7 of g-C3N4 to 28.1 of 1/15 BC/M, revealing a decrease in that of g-C3N4. Therefore, we may come to the conclusion that the
the inter-layer spacing, giving an improved visible light response introduced BC was partly in the form of amorphous carbon and/or
(Zuluaga et al., 2015). The reduction of the interplanar stacking is plane graphene carbon, which correlates with the XRD results.
beneficial to obtain a nanosheet structure, and the crystal structure Similarly, three peaks were observed after the deconvolution in the
tends to become more stable (Chai et al., 2017). N 1s region (Fig. 3c). The peak at 398.7 eV corresponded to the sp2 N
in triazines, while the peak at 400.3 eV was ascribed to the N(C)3
3.2. Morphology characterization groups (Liu et al., 2005). These two, along with sp2 C (NC]N,
287.9 eV), formed heptazine heterocyclic rings of g-C3N4, which are
The morphologies of BC/M composites samples obtained from the basic substructure of the g-C3N4 polymer (Yu et al., 2014). The
different BC/M ratios can be observed via SEM images in Fig. S1. Pure weak peak at 404.3 eV indicates a p-excitation (Chang et al., 2013).
g-C3N4 (Fig. S1d) exhibits aggregated, sheet, or layered structure, For 1/15 BC/M, the peak area at 404.3 eV is 1556.6, larger than that
which is typical for g-C3N4 synthesized from M (Zhang et al., 2014). of g-C3N4 (740.1) (Fig. 3c inset). These results indicate that, after
In addition, by introducing the carbon into the g-C3N4, their color carbon doping, the composite has more complete pp conjugation
dramatically changed from light yellow to grey and dark grey, as structure, which is conducive efficient electronic transfer. There is a
shown in the insets of Figs. S1aef. HRTEM was utilized to investigate new peak in the O 1s XPS spectrum of 1/15 BC/M composite indi-
the morphology and microstructure of the catalyst. The high- cating the creation of COC bonds (Li et al., 2013), signifying that
resolution images (Fig. 2) showed the nanosheet structure of g- the linkage of C with g-C3N4 is via ether bonding and pep stacking
C3N4 and 1/15 BC/M. The darkest part of the image indicates that BC interactions.
particles were being sandwiched between g-C3N4 sheets. Subse-
quently, the image (Fig. 2e) clearly shows that the BC particles were
embedded in g-C3N4 layers, which is also verified by EDX spectra of 3.4. UVevis diffuse reflectance spectrum
g-C3N4 and 1/15 BC/M. Compared to g-C3N4, EDX spectra of 1/15 BC/
M confirms the presence of K, Ca, Mg and Si elements, which be- The UVevis DRS provides information on the correlation be-
longs to the typical mineral components in BC (Yao et al., 2010). tween carbon and the light response of g-C3N4. As shown in Fig. 4a,
1/15 BC/M composite exhibited vast absorption near visible light
wavelengths, ascribed to the reduction of light reflection by the
3.3. Chemical states analysis
carbon. In this sense, increased light absorption at a wider range of
wavelengths should be advantageous for photocatalysts to absorb a
The elemental composition of the samples was measured by
more extensive range of photons wavelengths. This phenomenon
leads to increased production of electron-hole pairs during the
photoreaction, which are essential for oxidation and reduction
(Sudrajat, 2018; Li et al., 2016). As shown in Fig. 4b, the energy band
gaps of g-C3N4 and 1/15 BC/M are approximately 2.6 eV and 2.1 eV,
respectively. This is consistent with Mohamed et al.'s work that the
band gap easily tuned with kapok fibre (carbon sources) loading in
urea precursor which responsible for in-situ carbon doping
(Mohamed et al., 2018). The decreased band gap energy permits
more light with lower energy to be absorbed by the photocatalyst
(Li et al., 2018). This situation enables more light to be absorbed by
the photocatalyst to generate more charge carriers, consequently
enhancing photocatalytic capacity. The VB XP spectra were
employed to further investigate the band structure of g-C3N4 and 1/
15 BC/M. The VB edge was measured to be 1.6 eV for g-C3N4 and
1.38 eV for 1/15 BC/M. Combined with the UVeVis results, the op-
tical conduction band (CB) potentials of g-C3N4 and 1/15 BC/M
locate at 0.99 and 0.72 eV, respectively (Fig. 4d). Ma et al. also
confirmed this conclusion that the VB down-shift after construc-
tion of carbon sphere/g-C3N4 composite, thus affecting the band
Fig. 1. XRD patterns of g-C3N4 and 1/X BC/M composites. gap of the composite (Ma et al., 2017).
4 L. Meng et al. / Chemosphere 239 (2020) 124713

Fig. 2. HRTEM images of (a-b) pure g-C3N4 and (d-e) 1/15 BC/M; EDX spectra of (c) g-C3N4 and (f) 1/15 BC/M.

Fig. 3. XPS spectra of g-C3N4 and 1/15 BC/M: (a) survey spectra; (b) high resolution C 1s spectra; (c) N 1s spectra; (d) O 1s spectra.

3.5. Evaluation of photocatalytic performance adsorbed from the solution with g-C3N4 and the 1/X BC/M (X ¼ 5,
10, 15, 30) composites, respectively. 1/15 BC/M composite clearly
The photocatalytic degradation of as-prepared samples under showed a significantly better adsorption capability of RhB. How-
visible light (500 W Xe lamp, l > 420 nm) were evaluated with ever, 1/5 and 1/10 BC/M composites showed extremely low
cationic (RhB) and anionic dyes (MO). Fig. 5a displays the adsorp- adsorption performance probably resulted from g-C3N4 particles
tion of RhB in dark conditions until equilibrium. After 1.5 h from embedded in the pore structure of the BC, preventing BC compo-
equilibrium, 17.6%, 3.2%, 7.4%, 31.2%, and 23.9% of RhB were nents to exert their adsorption capacity. A similar trend was
 L. Meng et al. / Chemosphere 239 (2020) 124713 5

Fig. 4. (a,b) DRS UVevis spectra of all samples, (c) XPS valence band (VB) spectra and (d) electronic band structures of g-C3N4 and 1/15 BC/M.

observed for the adsorption of MO (Fig. 5d), for which the 1/15 BC/ C3N4 (0.0137 min1). As for the degradation of MO, the kinetic
M composites had the best adsorption performance. For nitrogen constant of the 1/15 BC/M composite (0.0263 min1) is 5 times the
sorption measurements (Fig. S2 and Table S1), the 1/15 BC/M rate of g-C3N4 (0.0053 min1). The 1/15 BC/M composite displayed
sample exhibited the highest value, with BET surface area of the maximum and fastest degradation of contaminants. This result
54 m2 g1, larger than g-C3N4 (40 m2 g1), which contributed to the corresponds with the structure characterizations and spectroscopic
high adsorption capacity. It can be concluded from the above test presented earlier. Moderate amounts of BC increased visible
analysis that BC content affected the adsorption behavior of the light adsorption, which excited more photo induced electron-hole
composites. pairs that are advantageous for photocatalysis. These results are
Fig. 5b-c, e-f displays the photocatalytic capacities of g-C3N4 and also consistent with the DRS analysis. More importantly, the
1/X BC/M composites for degradation of RhB and MO. The control enhanced absorptivity is supplemental to the high photocatalytic
with no catalyst was also performed under the same conditions. activity of the 1/15 BC/M composite.
The self-degradation of RhB and MO are negligible when there is no In commercial applications, reusability and stability of the
photocatalysts under visible light for 90 min. Photodegradation photocatalysts are essential parameters. Thus, having the best
efficiency of dyes by BC is about 5% which suggested the fact that performance (1/15 BC/M), stability studies were performed by
the photocatalytic activity of BC itself can be neglected. RhB and MO repeatedly performing RhB and MO degradation under identical
clearly went through decomposition in the presence of various experimental conditions. Separation of the photocatalyst particles
photocatalysts under visible light irradiation. The photo- from the slurry system is done by centrifugation. No auxiliary
degradation of RhB (Fig. 5b) after 90 min of irradiation was 71.5%, chemical is used in the regeneration process before subsequent
98.2%, and 89.2% for the g-C3N4, 1/15, and 1/30 BC/M composites, cycle. As Fig. 5g and h shows, the 1/15 BC/M composite was stable
respectively. However, 1/5 and 1/10 BC/M composites demon- up to four cycles. A minor decrease in activity was observed due to
strated low photocatalytic activity. That was probably due to the loss of catalyst in the experiment. In addition, the XRD patterns of
thick layer of BC, blocking the absorption of visible light by g-C3N4, the 1/15 BC/M composite (Fig. 5i) indicated a stable photocatalyst
resulting in decreased photocatalytic efficiency. However, the throughout the experiment. The 1/15 BC/M composite can be re-
appropriate amount of BC retained the g-C3N4 polymeric frame- generated by visible light and perform continuous decontamina-
work and enhanced photocatalytic performance of g-C3N4. Similar tion through visible light.
aspects were observed on the photo-degradation of MO (Fig. 5e),
where the 1/15 BC/M composite exhibited the maximum decolor-
ization efficiency of 92.4%. Pure g-C3N4 reduced the degradation 3.6. Mechanism discussion
efficiency of MO to 37.4% compared to RhB, suggesting that the
decolorization of MO was governed by the adsorption process. The active radicals take crucial role in the photodegradation
For degradation of RhB (Fig. 5c), the k rate constant for the 1/15 reactions. More reactive radical species generated on the photo-
BC/M composite is 0.0386 min1, which is 2.8 times faster than g- catalysts accounted for the high visible-light photochemical per-
formance (Huang et al., 2015). To investigate the decolorization
6 L. Meng et al. / Chemosphere 239 (2020) 124713

Fig. 5. RhB as pollutant: adsorption rates until equilibrium (a), photocatalytic capacities (b), and Pseudo-first-order kinetics curves (c) of BC, g-C3N4 and 1/X BC/M composites; MO
as pollutant: adsorption rates (d), photocatalytic capacities (e), and Pseudo-first-order kinetics curves (f) of g-C3N4 and 1/X BC/M composites. Stability study on photocatalytic
degradation of (g) RhB and (h) MO solution over 1/15 BC/M composite. XRD patterns of (i) 1/15 BC/M before and after photocatalysis.

mechanisms of RhB and MO on the 1/15 BC/M composite, trapping Fig. 6b, it can be seen that the photocurrent density of 1/15 BC/M is
experiments were performed to identify the active species. Scav- higher than that of g-C3N4. Photocurrent is formed mainly by
enging of superoxide species (
O

2 ), hydroxyl species ( OH), and transferring photogenerated electrons to the counter electrode; the
holes (hþ), was performed by BQ, IPA and TEOA, respectively higher photocurrent reveals more effective separation and mobility
(Huang et al., 2016). As shown in Fig. S4a, IPA only had little effect of photogenerated charge carriers. EIS measurement is a valid
on the degradation of RhB, whereas TEOA and BQ decreased the electrochemical approach to illustrate the electrons transfer effi-
degradation to 13.8% and 10.3%, respectively, compared to studies ciency at the interface of solid-electrolyte. In general, the smaller
without scavengers. Similar trends were observed for MO degra- diameter of semicircle arc in EIS Nyquist plot represents the lower
dation, implying that hþ and
O 2 are species involved in the interfacial electron transfer resistance (Zhu et al., 2015). In Fig. 6c,
decolorization of both RhB and MO. the diameter of EIS for the 1/15 BC/M is smaller than that of g-C3N4,
An efficient separation of photogenerated charge carriers is suggesting that carbon doping g-C3N4 has a relatively lower charge
essential for photocatalysis, and the PL spectrum reflects the charge transfer resistance at the contact interface. This may prove that the
separation efficiency (Jiang et al., 2018). It is well known that higher separation and transfer efficiency of photogenerated electron-hole
PL intensity results in fast recombination of charge carriers and pairs is improved by carbon doping g-C3N4.
thus lower photocatalytic activity. As shown in Fig. 6a, the fluo- The heightened performance of 1/15 BC/M composite could
rescence intensity of 1/15 BC/M is lower than that of g-C3N4, indi- come down to four major causes. Firstly, the nano-layered structure
cating that the addition of BC suppressed recombination of can obtain larger specific surface area, since larger surface area
photoinduced electrons and holes. Therefore, the interaction be- enhances dye adsorption and efficient mass transfer during pho-
tween reactive species and pollutants was promoted, which todegradation (Sun et al., 2013; Kibanova et al., 2012). Secondly, the
improved the photocatalytic efficiency of BC/M composites. introduced BC narrowed the band gap (2.1 eV) and increased light
Compared with 1/15 BC/M, 1/30 BC/M had lower BC content, which absorption at a wider range of wavelengths. Hence, more photo-
limited the g-C3N4/carbon contact interface. As a result, the sepa- induced electrons get excited, which, coupled with the lower
ration of electron-hole pairs and excited electron transfer efficiency band gap energy, make the photocatalytic degradation process
are lower than those of 1/15 BC/M. faster than g-C3N4. Lastly, after BC modification, impurities are
The separation efficiency of photoinduced charges on g-C3N4 formed in the bandgap and act as a trap for electrons. The trapped
and 1/15 BC/M composite was studied qualitatively by means of electrons cannot recombine spontaneously with holes, and conse-
transient photocurrent responses and EIS experiments. From quently promote enhanced charge carrier separation.
 L. Meng et al. / Chemosphere 239 (2020) 124713 7

Fig. 6. Photoluminescence (PL) spectra (a), transient photocurrent responses (b) and EIS Nyquist profiles of g-C3N4 and 1/15 BC/M composite (c); Proposed photocatalytic
mechanism of 1/15 BC/M composite (d).

The probable degradation mechanism of the 1/15 BC/M photo- enhancement of the photocatalytic performance of the 1/15 BC/M
catalyst is presented in Fig. 6d. Under visible light, the composite composite was ascribed to the higher optical sorption and BET
was rapidly excited because of its low band gap energy of about specific surface area (54.09 m2 g1), in addition to the synergism
2.1 eV. The photo-generated electrons were transferred to the CB, between BC and g-C3N4, which led to longer lifetime of excited
producing holes in VB. Next, the electrons were transferred to the electrons of semiconductor and hindered photo-induced electron-
BC particles, diffusing to the surface of photocatalyst to participate hole recombination. Nevertheless, excess addition of BC (higher
in a series of reactions. Consequently, the photo-induced electrons than 1:15) blocks the visible light, and low addition (lower than
and holes were separated effectively. According to previous ex- 1:15) leads to lower g-C3N4/carbon contact interface, which is un-
periments, hþ and
O þ
2 are the major active species, and h can favorable to the transfer of photo-induced electrons. The photo-


directly oxidize pollutants (Yuan et al., 2016). The radicals O2 were degradation activity was dominated by
O 2 and hole oxidation.
reduced from dissolved O2 upon reaction with photo-excited Moreover, the 1/15 BC/M nanocomposite showed good stability
electrons (Meng et al., 2016; Huang et al., 2017b). Due to the low over 4 cycles. In this study, the BC modification can not only realize
electron-hole pair recombination rate, photoelectrons lifetime was sustainable utilization of waste biomass but also develop the po-
prolonged, and more
O 2 was produced, leading to improved tential of BC as a conductor to transfer electrons. Overall, BC
photocatalytic activity (Huang et al., 2017b). Nevertheless, for 1/5 modifications can offer an efficient, economic, sustainable
and 1/10 BC/M composites, excessive BC can block the absorption of approach to enhance photocatalytic performance of g-C3N4.
light by g-C3N4 and hinder the photocatalytic process (Wang et al.,
2018). Therefore, excessive carbon loading is not conducive to
photocatalysis. In view of this, a 1:15 mass ratio of BC to M maxi- Acknowledgements
mized the ability of photocatalytic oxidation of the incorporated g-
C3N4. This work was financially supported by the Qing Lan Project, the
National Science Foundation of China [31772394, 41771349,
41977085, 51602281], the Modern Agriculture Project in Jiangsu
4. Conclusion Province [BE2018328], and the Six Talent Peaks Project in Jiangsu
Province [2018-TD-JNHB-012]. We would like to thank the Testing
In conclusion, BC-modified g-C3N4 nanocomposites were syn- Center of Yangzhou University for their help with sample
thetized through a solid-state reaction based on thermal poly- characterization.
condensation of M with different mass ratios of BC. The optimal BC/
M ratio was 1:15, which exhibited superior photocatalytic activity
for MO and RhB under visible light. The introduction of a small Appendix A. Supplementary data
amount of BC retained the g-C3N4 polymeric framework and pre-
sented narrower bandgap (2.1 eV). Carbon was combined with g- Supplementary data to this article can be found online at
C3N4 via pep stacking interaction and ether bond bridge. The https://doi.org/10.1016/j.chemosphere.2019.124713.
8 L. Meng et al. / Chemosphere 239 (2020) 124713

References Liu, G., Zhang, X., Xu, Y., Niu, X., Zheng, L., Ding, X., 2005. The preparation of Zn-
doped TiO2 nanoparticles by sol-gel and solid phase reaction methods
respectively and their photocatalytic activities. Chemosphere 59, 1367e1371.
Chai, B., Yan, J., Wang, C., Ren, Z., Zhu, Y., 2017. Enhanced visible light photocatalytic
Liu, J., Zhang, T., Wang, Z., Dawson, G., 2011. Simple pyrolysis of urea into graphitic
degradation of Rhodamine B over phosphorus doped graphitic carbon nitride.
carbon nitride with recyclable adsorption and photocatalytic activity. J. Mater.
Appl. Surf. Sci. 391, 376e383.
Chem. 21, 14398e14401.
Chang, C., Fu, Y., Hu, M., Wang, C., Shan, G., Zhu, L., 2013. Photodegradation of
Lu, W., Xu, T., Wang, Y., Hu, H., Li, N., Jiang, X., Chen, W., 2016. Synergistic photo-
bisphenol A by highly stable palladium-doped mesoporous graphite carbon
catalytic properties and mechanism of g-C3N4, coupled with zinc phthalocya-
nitride (Pd/mpg-C3N4) under simulated solar light irradiation. Appl. Catal. B
nine catalyst under visible light irradiation. Appl. Catal. B Environ. 180, 20e28.
Environ. 142e143, 553e560.
Ma, L., Fan, H., Fu, K., Lei, S., Hu, Q., Huang, H., He, G., 2017. Protonation of graphitic
Chen, F., Huang, H., Guo, L., Zhang, Y., Ma, T., 2019. The role of polarization in
carbon nitride (g-C3N4) for an electrostatically self-assembling carbon@g-C3N4
photocatalysis. Angew. Chem. Int. Ed. 58, 10061e10073.
core-shell nanostructure toward high hydrogen evolution. ACS Sustain. Chem.
Chen, K., Chai, Z., Li, C., Shi, L., Liu, M., Xie, Q., Zhang, Y., Xu, D., Manivannan, A.,
Eng. 5, 7093e7103.
Liu, Z., 2016. Catalyst-free growth of three-dimensional graphene flakes and
Meng, S., Ye, X., Ning, X., Xie, M., Fu, X., Chen, S., 2016. Selective oxidation of aro-
graphene/g-C3N4 composite for hydrocarbon oxidation. ACS Nano 10,
matic alcohols to aromatic aldehydes by BN/metal sulfide with enhanced
3665e3673.
photocatalytic activity. Appl. Catal. B Environ. 182, 356e368.
Christoforidis, K., Syrgiannis, Z., Parola, V., Montini, T., Petit, C., Stathatos, E.,
Mohamed, M., Zain, M., Minggu, L., Kassim, M., Amin, N., Salleh, W., Salehmin, M.,
Godin, R., Durrant, J., Prato, M., Fornasiero, P., 2018. Metal-free dual-phase full
Nasir, M., Hir, Z., 2018. Constructing bio-templated 3D porous microtubular C-
organic carbon nanotubes/g-C3N4 heteroarchitectures for photocatalytic
doped g-C3N4 with tunable band structure and enhanced charge carrier sepa-
hydrogen production. Nano Energy 50, 468e478.
ration. Appl. Catal. B Environ. 236, 265e279.
Guo, S., Deng, Z., Li, M., Jiang, B., Tian, C., Pan, Q., Fu, H., 2016. Phosphorus-doped
Pi, L., Jiang, R., Zhou, W., Zhu, H., Xiao, W., Wang, D., Mao, X., 2015. g-C3N4 modified
carbon nitride tubes with a layered micro-nanostructure for enhanced visible-
biochar as an adsorptive and photocatalytic material for decontamination of
light photocatalytic hydrogen evolution. Angew. Chem. Int. Ed. 55, 1830e1834.
aqueous organic pollutants. Appl. Surf. Sci. 358, 231e239.
Grover, I.S., Prajapat, R.C., Singh, S., Pal, B., 2017. Highly photoactive Au-TiO2
Singh, H., Chauhan, G., Jain, A.K., Sharma, S.K., 2017. Adsorptive potential of agri-
nanowires for improved photo-degradation of propiconazole fungicide under
cultural wastes for removal of dyes from aqueous solutions. J. Environ. Eng. 5,
UV/sunlight irradiation. Sol. Energy 144, 612e618.
122e135.
Hao, L., Kang, L., Huang, H., Ye, L., Han, K., Yang, S., Yu, H., Batmunkh, M., Zhang, Y.,
Su, F., Zhang, W., 2017. Carbonyl-grafted g-C3N4 porous nanosheets for efficient
Ma, T., 2019. Surface-halogenation-induced atomic-site activation and local
photocatalytic hydrogen evolution. Chem. Asian J. 12, 515e523.
charge separation for superb CO2 photoreduction. Adv. Mater. 31, 1900546.
Sudrajat, H., 2018. A one-pot, solid-state route for realizing highly visible light
Hou, J., Jiang, K., Wei, R., Tahir, M., Wu, X., Shen, M., Wang, X., Cao, C., 2017. Popcorn-
active Na-doped g-C3N4, photocatalysts. J. Solid State Chem. 257, 26e33.
derived porous carbon flakes with an ultrahigh specific surface area for superior
Sun, Z., Bai, C., Zheng, S., Yang, X., Frost, R.L., 2013. A comparative study of different
performance supercapacitors. ACS Appl. Mater. Interfaces 9, 30626e30634.
porous amorphous silica minerals supported TiO2 catalysts. Appl. Catal. A-Gen.
Hou, J., Wei, R., Wu, X., Tahir, M., Wang, X., Butt, F.K., Cao, C., 2018. Lantern-like
458, 103e110.
bismuth oxyiodides embedded typha-based carbon via in-situ self-template
Wang, X., Blechert, S., Antonietti, M., 2012. Polymeric graphitic carbon nitride for
and ion exchange-recrystallization for high-performance photocatalysis. Dalton
heterogeneous photocatalysis. ACS Catal. 2, 1596e1606.
Trans. 47, 6692e6701.
Wang, T., Liu, X., Ma, C., 2018. A two step hydrothermal process to prepare carbon
Huang, H., Li, X., Wang, J., Dong, F., Chu, P.K., Zhang, T., Zhang, Y., 2015. Anionic
spheres from bamboo for construction of core-shell non-metallic photo-
group self-doping as a promising strategy: band-gap engineering and multi-
catalysts. New J. Chem. 42, 6515e6524.
functional applications of high-performance CO2 3 -doped Bi2O2CO3. ACS Catal.
Xu, Q., Jiang, C., Cheng, B., Yu, J., 2017. Enhanced visible-light photocatalytic H2-
5, 4094e4103.
generation activity of carbon/g-C3N4 nanocomposites prepared by two-step
Huang, H., Xiao, K., He, Y., Zhang, T., Dong, F., Du, X., Zhang, Y., 2016. In situ assembly
thermal treatment. Dalton Trans. 46, 10611e10619.
of BiOI@Bi12O17Cl2 p-n junction: charge induced unique front-lateral surfaces
Yan, S., Li, Z., Zou, Z., 2009. Photodegradation performance of g-C3N4 fabricated by
coupling heterostructure with high exposure of BiOI {001} active facets for
directly heating melamine. Langmuir 25, 10397e10401.
robust and nonselective photocatalysis. Appl. Catal. B Environ. 199, 75e86.
Yao, F., Arbestain, M.C., Virgel, S., Blanco, F., Arostegui, J., Maci  , J.A.,
a-Agullo
Huang, H., Xiao, K., Tian, N., Dong, F., Zhang, T., Du, X., Zhang, Y., 2017a. Template-
Macías, F., 2010. Simulated geochemical weathering of a mineral ash-rich bio-
free precursor-surface-etching route to porous, thin g-C3N4 nanosheets for
char in a modified Soxhlet reactor. Chemosphere 80, 724e732.
enhancing photocatalytic reduction and oxidation activity. J. Mater. Chem. A. 5,
Yu, Q., Guo, S., Li, X., Zhang, M., 2014. Template free fabrication of porous g-C3N4/
17452e17463.
graphene hybrid with enhanced photocatalytic capability under visible light.
Huang, H., Tu, S., Zeng, C., Zhang, T., Reshak, A.H., Zhang, Y., 2017b. Macroscopic
Mater. Technol. 29, 172e178.
polarization enhancement promoting photo- and piezoelectric-induced charge
Yu, H., Li, J., Zhang, Y., Yang, S., Han, K., Dong, F., Ma, T., Huang, H., 2019. Three-in-
separation and molecular oxygen activation. Angew. Chem. Int. Ed. 56,
one oxygen vacancies: whole visible-spectrum absorption, efficient charge
11860e11864.
separation, and surface site activation for robust CO2 photoreduction. Angew.
Jiang, L., Yuan, X., Zeng, G., Wu, Z., Liang, J., Chen, X., Leng, L., Wang, H., Wang, H.,
Chem. Int. Ed. 58, 3880e3884.
2018. Metal-free efficient photocatalyst for stable visible-light photocatalytic
Yuan, X., Zhou, C., Jing, Q., Tang, Q., Mu, Y., Du, A., 2016. Facile synthesis of g-C3N4
degradation of refractory pollutant. Appl. Catal. B Environ. 221, 715e725.
nanosheets/ZnO nanocomposites with enhanced photocatalytic activity in
Kibanova, D., Sleiman, M., Cervini-Silva, J., Destaillats, H., 2012. Adsorption and
reduction of aqueous chromium (VI) under visible light. Nanomaterials 6,
photocatalytic oxidation of formaldehyde on a clay-TiO2 composite. J. Hazard
173e184.
Mater. 211e212, 233e239.
Zhang, F., Chai, B., Liao, X., Ren, M., Liu, B., 2014. Preparation and visible light
Li, Y., Zhang, H., Liu, P., Wang, D., Li, Y., Zhao, H., 2013. Cross-linked g-C3N4/rGO
photocatalytic properties of RGO/C3N4 composites. Chin. J. Inorg. Chem. 30,
nanocomposites with tunable band structure and enhanced visible light pho-
821e827.
tocatalytic activity. Small 9, 3336e3344.
Zhang, F., Zhuang, H., Song, J., Men, Y., Pan, Y., Yu, S., 2018. Coupling cobalt sulfide
Li, Y., Wu, S., Huang, L., Wang, J., Xu, H., Li, H., 2014. Synthesis of carbon-doped g-
nanosheets with cadmium sulfide nanoparticles for highly efficient visible-
C3N4, composites with enhanced visible-light photocatalytic activity. Mater.
light-driven photocatalysis. Appl. Catal. B Environ. 226, 103e110.
Lett. 137, 281e284.
Zhu, Y., Ren, T., Yuan, Z., 2015. Mesoporous phosphorus-doped g-C3N4 nano-
Li, C., Sun, Z., Huang, W., Zheng, S., 2016. Facile synthesis of g-C3N4/montmorillonite
structured flowers with superior photocatalytic hydrogen evolution perfor-
composite with enhanced visible light photodegradation of rhodamine B and
mance. ACS Appl. Mater. Interfaces 7, 16850e16856.
tetracycline. J. Taiwan Inst. Chem. E. 66, 363e371.
Zuluaga, S., Liu, L., Shafiq, N., Rupich, S.M., Veyan, J.F., Chabal, Y.J., Thonhauser, T.,
Li, C., Sun, Z., Zhang, W., Yu, C., Zheng, S., 2018. Highly efficient g-C3N4/TiO2/
2015. Structural band-gap tuning in g-C3N4. Phys. Chem. Chem. Phys. 17,
kaolinite composite with novel three-dimensional structure and enhanced
957e962.
visible light responding ability towards ciprofloxacin and S. aureus. Appl. Catal.
B Environ. 220, 272e282.