Contents lists available at ScienceDirect

Journal of Hazardous Materials

journal homepage: www.elsevier.com/locate/jhazmat

Enhanced photocatalytic reduction for the dechlorination of 2-chlorodibenzo-p-dioxin by high-performance g-C3N4/NiO heterojunctioncomposites under ultraviolet-visible light illumination

Jiafeng Ding, Shihuan Lu, Lilai Shen, Ruopeng Yan, Yinan Zhang, Hangjun Zhang⁎

College of Life and Environmental Sciences, Hangzhou Normal University, 310018, Hangzhou, Zhejiang, China

A R T I C L E I N F O

Editor: Navid B Saleh

Keywords:2-CDDg-C3N4/NiOReductive dechlorinationDetoxification

A B S T R A C T

Polychlorinated dibenzo-p-dioxins (PCDDs), characterized by their high persistency and bioaccumulation, arewidely detected in the environment. In this study, high-performance g-C3N4/NiO heterojunctions were fabri-cated to degrade 2-chlorodibenzo-p-dioxin (2-CDD) under ultraviolet-visible (UV–vis) light illumination.Experiments revealed that the pure g-C3N4 and range of g-C3N4/NiO heterojunctions were synthesized by themixing and heating method, and then were characterized by XRD, TEM, XPS and PL etc. The composites ex-hibited enhanced dechlorination activities under anoxic conditions. After comparison, the g-C3N4/NiO (4:6)showed optimal dechlorination performance such that 70.4% of 2-CDD was removed within 8 h and 52.3% of 2-CDD was transformed to dibenzo-p-dioxin (DD), about fourfold higher than the pristine g-C3N4. The transfor-mation of 2-CDD was accompanied by the resale of Cl ion, and the additional oxygen was proven to be able toconsume electrons and hydrogen ions, thus greatly inhibiting the degradation of PCDD in systems. The g-C3N4/NiO (4:6) can be reused at least seven times, and the mechanism was proposed in detail to promote photo-induced electrohole separation and provide active sites. This study extends the use range of g-C3N4/NiO het-erojunctions and develops a new technology to degrade PCDDs with striking activity and stability.

1. Introduction

Polychlorinated dibenzo-p-dioxins (PCDDs) have been regarded aspriority environmental pollutants because of their carcinogenic, ter-atogenic and mutagenic effects by the United States EnvironmentalProtection Agency (U.S. EPA) (Mumbo et al., 2016). Serious concernhas been raised that PCDDs could be frequently detected in water,aquatic organisms, soils, sediments, or even human blood (Mumboet al., 2016; Arfaeinia et al., 2017; Pagano et al., 2018; Li et al., 2019;Ssebugere et al., 2019). Due to their properties of persistence, bioac-cumulation and high toxicity, as well as the increasing concentration inthe environment, substantial research has been conducted to developeffective strategies to remove PCDDs (Guemiza et al., 2017), such ashydrothermal processes (Qiu et al., 2019), photocatalyst reduction(Ding et al., 2018), catalytic oxidation (Ji et al., 2018), metal complex(Liu et al., 2016), and hydroxyl radical (%OH) utilization (Rashidianet al., 2019).

In response to the environmental risks, many researchers turn theirattention toward photocatalyst reduction. Due to the temperate reac-tion conditions and low energy consumption, the photocatalytic

reduction dechlorination process is considered a promising technologyto control water pollution. The reduction of PCDDs has led to breakageof the C-Cl bond (Lechner, 2009), thus decreasing the toxicity of PCDDsafter the release of chlorine (Wu and Ng, 2008). For example, Liu et al.(2016) showed that PCDDs could be converted into less chlorinatedPCDDs from condensed water with Pd/g-Al2O3. The g-C3N4 is a pro-mising nonmetallic semiconductor material, which has been widelyused in various photocatalytic reactions. However, the application ofpristine g-C3N4 still has problems of the relatively fast recombinationrate of photogenerated electron-hole, inefficient conductivity, low solarenergy availability rate and low quantum efficiency (Niu et al., 2012;Mishra et al., 2019; Zhang et al., 2019a).

Therefore, it is of great significance to develop catalysts withabundant resources and lower cost as substitutes to remove organicpollutants effectively. Wang et al. (2019a) systematically reviewed thepromise of using g-C3N4-metal/metal oxide nanohybrids for the de-gradation and removal of various contaminants. It illustrated the de-gradation mechanism of their performances at the water–environmentnexus including energy harvesting, water treatment, environmentalsensing and in situ nanoremediation. Metals (e.g., Fe, Ni, and Ag) and

https://doi.org/10.1016/j.jhazmat.2019.121255Received 7 August 2019; Received in revised form 11 September 2019; Accepted 17 September 2019

⁎ Corresponding author at: College of Life and Environmental Sciences, Hangzhou Normal University, Hangzhou, 310036, China.E-mail address: 13819172516@163.com (H. Zhang).

Journal of Hazardous Materials 384 (2020) 121255

Available online 28 September 20190304-3894/ © 2019 Elsevier B.V. All rights reserved.

T

nonmetallic material (e.g., S, P, O, dual-oxygen and Oxamide) havebeen intensively reported as effective cocatalysts to solve the problem(Xu et al., 2019; Tang et al., 2019; Liu et al., 2019; Zhang et al., 2019b;Tang et al., 2018a; Feng et al., 2018). For example, Tang et al. (2019)reported that oxamide were used as cocatalysts to modify g-C3N4 toadvance photocatalytic performance. Tang et al. (2018a) reported thatthe ternary Ag/g-C3N4/NaTaO3 photocatalysts exhibited effective ac-tivity to remove organic pollutants by their strong wide-spectrum lightabsorbance. Nickel (II) oxide (NiO) has gained wide attention due to itsexcellent electrochromism, electrochemical capacitance and heat sen-sitivity (Sahoo et al., 2012). Therefore, many studies have tried tocombine NiO with synergistic dopants to improve the photoelectricitylifetime and prolong the absorption of light. It has been used as a co-catalyst to effectively promote photocatalytic degradation (Chen et al.,2014; Thiyagarajan et al., 2018), hydrogen evolution (Liu et al., 2018),reduction of CO2 (Tang et al., 2018b) and water splitting (Fu et al.,2018). For example, Tang et al. (2018b) successfully synthesized NiO/g-C3N4 heterojunctions by the hydrothermal deposition method and theheterojunctions could provide more efficient electrons for the reductionof CO2. NiO has been considered an appropriate photocatalyst for theformation of the p-n junction (Zhang et al., 2010), and the TiO2-NiO p-njunction has been widely used in the photocatalytic evolution of H2

(Sreethawong et al., 2005). Therefore, the combination of g-C3N4 andNiO could serve as an efficient photocatalytic for PCDD degradation.

Herein, we aimed to achieve NiO-modified g-C3N4 into a hetero-structure to investigate their photocatalytic reductive performance to-ward the dechlorination of 2-CDD under UV–vis light illumination. Thestructures and morphologies of the pristine g-C3N4 and g-C3N4/NiOheterojunctions were characterized by XRD and SEM. Additionally, theexperiments were conducted to evaluate the effect of the contents,synthesis temperature, and environment impact as well as to proposethe possible photocatalytic reaction mechanism in detail. Thus, wemight extend the application range of g-C3N4/NiO heterojunctions andmay develop a new approach for the degradation of PCDDs withstriking activity and stability.

2. Materials and methods

2.1. Chemicals

The standard model pollutant (2-CDD, purity< 99%) and product(DD, purity< 99%) were purchased from Accustandard Company (NewHaven, CT, USA). Nickel(II) chloride hexahydrate (NiCl2·6H2O, analy-tical grade) and dicyandiamide (C2H4N4, analytical grade) were pur-chased from Aladdin Industrial Corporation (Shanghai, China). Organicsolvents, including isopropanol (C3H8O), ethanol (C2H6O), n-hexane(C6H14) and toluene (C7H9), were purchased from Sinopharm ChemicalReagent Co., Ltd (Shanghai, China). The water used in the experimentwas deionized water (18.2 MΩ cm resistivity), which was obtainedusing a Millipore water purification system.

2.2. Synthesis

Pure g-C3N4 was fabricated using a direct thermal polymerizationmethod (Ding et al., 2018; Tang et al., 2018b). Briefly, dicyandiamide(10 g) was put into an alumina crucible with a cover and then wasburned at 500 °C for 2 h at a rate of 10 °Cmin−1), followed by calciningat 520 °C for another 2 h at a rate of 5 °Cmin−1. The resultant yellowproduct was obtained after cooling. Next, the obtained powder wasadded to 50mL of ultrapure water under vigorous stirring and sustainedultrasonic dispersal for 2 h. The g-C3N4 was obtained after filtration anddrying at 100 °C in a vacuum oven overnight for further use.

The g-C3N4/NiO composites were fabricated by the followingmethod. First, the as-prepared g-C3N4 (0.5 g) was dispersed in ethanol(50mL) and then ultrasonic dispersal was sustained for 30min.Subsequently, a certain amount (0.318 g, 0.636 g, 0.792 g and 0.950 g)

of NiCl2·6H2O were added and ultrasonically dispersed for 6 h, followedby drying and grinding. The final samples were obtained through cal-cination at 400 °C for 3 h and grinded them into powders after coolingto room temperature. The as-prepared g-C3N4/NiO samples with massratios of NiO at 20%, 40%, 50% and 60% were denoted as g-C3N4/NiO(8:2), g-C3N4/NiO (6:4), g-C3N4/NiO (5:5) and g-C3N4/NiO (4:6), re-spectively. Additionally, the obtained g-C3N4/NiO (4:6) composite wasannealed at different temperatures (200 °C, 300 °C, 400 °C and 500 °C)to select the optimum synthesis temperature. The pure NiO was fabri-cated by the same method without the g-C3N4 addition.

2.3. Characterization

X-ray diffraction (XRD; XRD-6000, Shimadzu, Japan), using Cu Kα(λ =0.1546 nm) values with accelerating voltage and applying currentat 40 kV and 40mA, respectively. The diffractograms were recorded atthe 2θ range from 10° to 80°. Scanning electron microscopy (SEM andEDX; Supra55; Sapphire, Germany; accelerating voltage=0.02–30 kV).Transmission electron microscopy (TEM; TecnaiG2 F20S-TWIN, USA;accelerating voltage: 200 kV). High-resolution transmission electronicmicroscopy (HR-TEM), equipped with an energy-dispersive X-rayspectroscopy (EDX) detector. The specific surface area of compositeswas measured by N2 adsorption-desorption isotherms (77 K; 3H-2000PS2; Beishide, China), using the Brunauer–Emmett–Teller (BET)method and the pore size distribution was obtained from the desorptionbranch of the isotherms. The optical properties were analyzed byUV–vis adsorption spectra (UV–vis DRS; TU-1901, Pgeneral, China).Electrochemical experiments (i-t and EIS) were analyzed using anelectrochemical workstation (CH Instruments 650D, China) in a con-ventional three-electrode configuration. Na2SO4 aqueous solution(120mL, 0.1mol L−1) was used as the electrolyte, and argon gas wasdegassed prior to electrochemical measurements for 30 min. X-rayphoto-electron spectroscopy (XPS, Escalab 250, Thermo, England) withan Al Kα anode radiation as source (1486.71 eV) was employed toidentify surface elemental composition and chemical states of compo-sites. All the binding energies were referenced to the C 1s peak at 284.6eV. Photoluminescence (PL) is equipped with a fluorescence spectro-photometer (Fluorolog-3-Tau, France) and its excitation wavelengthwas 325 nm. The widths of the excitation and emission slits were 5.0nm. Electron spin resonance (ESR, JES FA200 spectrometer, Germany)with 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as spin-trapping agent,were used to detect the signals of radicals (%OH and %O2

−). The lightwavelength was in the range of 350–780 nm. The experimental opera-tion was as follows: photocatalyst (2.5 mg) was dispersed in deionizedwater (0.5 ml, for ·OH) or methanol (0.5 ml, for %O2

−) and then addedwith DMPO (45 μL, as spin-trapping agent).

2.4. Photocatalytic dechlorination setup

Details experiment processes in this study are provided inSupporting Information.

2.5. Chemical analysis

Details chemical analytic methods used in this study are provided inSupporting Information.

3. Results and discussion

3.1. Physical characterization of photocatalysts

The crystalline structures of pristine g-C3N4, NiO and g-C3N4/NiOheterojunctions were characterized by XRD (Fig. 1). The g-C3N4 ex-hibits two distinct characteristic peaks at 27.40° and 13.04°, corre-sponding to the characteristic (002) and (100) planes, arising from theinterlayer stacking of the conjugated aromatic system (Xu et al., 2013;

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

2

Wang et al., 2009). For NiO, intense diffraction peaks at 2θ=37.09°,43.10°, 62.58°, 75.04° and 79.01° could readily be assigned to (111),(200), (220), (311) and (222) planes, respectively. All the diffractionpeaks are consistent with the orthorhombic phase of NiO (Tong et al.,2015a). The peaks gradually became obvious with the increase in theNiO ratio in g-C3N4/NiO composites after coupling with each other.Furthermore, the clear phenomenon of g-C3N4 in g-C3N4 /NiO (4:6)suggests the successful combination of g-C3N4 and NiO because themain peak of g-C3N4 at 27.40° was not overlapped with that of NiO.(Tang et al., 2018b) The synthesis of catalysts presented no additionaldiffraction peaks, indicating that the products are of high purity.

The microstructure and morphology of pure g-C3N4 (Fig. 2a), NiO(Fig. 2b) and g-C3N4/NiO composites (Fig. 2c and d) were investigatedby SEM as mentioned. Our previous paper has reported the g-C3N4

comprises stacked bulks with a porous structure (Ding et al., 2018).After hybridization, NiO was adhered to the surface of g-C3N4 ran-domly. The g-C3N4/NiO samples presented a light, white ball-flowercurly structure. However, excessive NiO in g-C3N4/NiO samples wouldlead to agglomeration, which is not conducive to constitute a closely

interface (Tang et al., 2018b; Fu et al., 2018). The existence of NiO canbe further proven by TEM shown in Fig. 3. It displays that NiO (circle) isdispersed on the surface of g-C3N4, corresponding to the SEM results.The results showed that NiO was successful in close contact with g-C3N4

and formed an actual interface. Moreover, from the image shown inFig. 3d, the lattice fringe with a d spacing of 0.209 nm was clearlyobserved, which was corresponding to the crystal plane (200) of cubic-structured NiO (Yu et al., 2015). The attachment of NiO on the g-C3N4

surface is further affirmed by HR-TEM and EDX analyses (Fig. 3e). Theresults of EDX mapping show that the C, N, O and Ni elements are welldispersed on g-C3N4/NiO heterojunctions.

The FT-IR spectra of g-C3N4, NiO and g-C3N4/NiO composites areindicated in Fig. S1. The spectrum of g-C3N4 contains skeletal adsorp-tion bands between 1200 cm−1 and 1700 cm−1, which present thestretching modes of C–N heterocycles (Shi et al., 2014) and the otherpeak (810 cm−1) resulted to the breathing mode of C–N heterocycles(Lin et al., 2016). For pure NiO, the peaks around 430 cm−1 are as-cribed to the Ni-O vibration adsorption (Selvarajan et al., 2018). For g-C3N4/NiO composites, the characteristic peaks of g-C3N4 and NiO werealso found in the spectrum. Additionally, the peak at 430 cm−1 in NiOspectra shifts in g-C3N4/NiO heterojunctions, which suggests the tria-zine rings and hydroxyl groups of g-C3N4 are successfully adhering withNiO. The results further confirmed that g-C3N4/NiO composites con-sisted of g-C3N4 and NiO.

The chemical composition of Ni species on g-C3N4 was carried outby XPS. The high-resolution XPS spectra of g-C3N4/NiO (4:6) are shownin Fig. 4 in which the C 1s (a), N 1s (b), O 1s (c) and Ni 2p (d) char-acteristic peaks can be found. Two sub-bands (284.8 and 288.1 eV) in C1s of the g-C3N4 can be corresponding to CeC bond and NeC]N groupin the s-triazine ring (Fig. 4a), respectively (Li et al., 2013). The peak ofN 1s located at 398.4 ascribes to sp2 hybridized aromatic N connectedof C]NeC in Fig. 4b (Zhang et al., 2013), while the other peaks(399.9 eV and 404.2 eV) are belonged to the N–(C)3 groups and π ex-citations. (Thaweesak et al., 2017) The peak of O 1s (529.8 eV) is de-rived from Ni–O bone, while the other O species peak (531.6 eV) pre-sents the adsorption of H2O in Fig. 4c (Zhang et al., 2016). The spectraof high-resolution Ni 2p XPS are the typical characteristic peaks of Ni inNi–O bonds (Dai et al., 2014). From Fig. 4d, the binding energy at 855.7

Fig. 1. XRD patterns of NiO, g-C3N4 and g-C3N4/NiO composites.

Fig. 2. SEM micrographs of g-C3N4 (a), NiO (b) and g-C3N4/NiO composites (c, d).

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

3

Fig. 3. TEM micrographs of g-C3N4 (a), NiO (b), g-C3N4/NiO composites (c, d) and the Elements Mapping of g-C3N4/NiO (C–K, N-K, O-K,Ni-K) (e).

Fig. 4. XPS survey spectra of g-C3N4/NiO (4:6) for C 1s (a), N 1s (b), O 1s (c), Ni 2p (d).

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

4

and 873.1 eV corresponded to Ni 2p3/2 and Ni 2p1/2, respectively, in-dicated the existence of Ni2+ in NiO (Zhang et al., 2016). The other twopeaks at around 861.2 eV and 879.2 eV belong to the shakeup satellites(Tong et al., 2015b). Further, according to the results of EDX spectrum,the g-C3N4/NiO (4:6) is made up of C, N, O and Ni elements in Fig. 5b.

The results of N2 adsorption-desorption isotherms are presented inFig. 5a to analyze the textural properties of g-C3N4/NiO composites. Allthe samples exhibit type IV isotherms with H3 hysteresis loops, im-plying the existence of a mesoporous structure in the composites (Dongand Zhang, 2012). The isotherms of g-C3N4/NiO (4:6) and g-C3N4/NiO(6:4) shifted upwards, indicating a larger surface area and pore volumein contrast to g-C3N4. The specific surface area (SBET), average pore size(daverage) and total pore volume (Vtotal) of pristine g-C3N4 and g-C3N4/NiO heterojunctions are listed in Table 1. The SBET value was raisedfrom 6.29 to 62.21m2/g with the NiO contents from 0 to 60%, and thedaverage and Vtotal values were also increased accordingly. Generally, thelarger surface area of catalysts exposes more active sites and causeshigher reaction activity (Jiang et al., 2015). The optical properties ofpure g-C3N4 and a series of NiO/g-C3N4 composites were analyzed byUV–vis diffuse reflectance spectroscopy (DRS) in Fig. 5c. For g-C3N4/NiO composites, the adsorption of light in the visible region graduallyincreased with the increased NiO content. It may be caused by the sy-nergistic effect between g-C3N4 and NiO. Compared with g-C3N4, itsuggests that g-C3N4/NiO composites exhibit stronger adsorption valuesin the vertical coordinates, which is conducive to the solar energy

absorption and accelerating photocatalytic reaction (Tang et al.,2018b). The PL spectra of the g-C3N4/NiO composites in the wave-length range of 350–600 nm are shown in Fig. 5d. The main peak at450 nm is due to the emission corresponding to the band gap transitionof g-C3N4, meaning the faster recombination of electrons and holes. Theweaker emission peak of NiO/g-C3N4 composites in PL spectrum in-dicate that electron–hole pairs can be effectively separated due to theformation of NiO/g-C3N4 heterojunction. The results also lead to se-parating the photo-induced charges effectively. The bandgap of semi-conductor can be calculated by the equation (Tang et al., 2018b):

= −αhν A hν E( )gn

where a, h, v and A refers to the absorption coefficient, Planck constant,light frequency and proportionality, respectively. n is decided from theproperties of the transition in a semiconductor.

According to the previous description and the calculation of equa-tion, the bandgap of NiO and g-C3N4 are 3.4 and 2.7 eV, respectively(Fig. 6a). Moreover, the band gap position was determined directly byMott-Schottky plots and the results were shown in Fig. 6b. Based on thelinear C−2 potential curves and the intersection of potential, the flatband potentials of g-C3N4 and NiO can be determined at -0.88 and-0.6 V (vs. Ag/AgCl), respectively, corresponding to -0.68 and -0.4 V (vs.NHE). The conduction band bottom potential of n-type semiconductoris usually 0-0.2 eV more negative than the flat band potential (Tanget al., 2018b). Thus, the ECB of g-C3N4 and NiO are at -0.78 and -0.5 eV,after the potential difference is set to 0.1 eV. Finally, the EVB was cal-culated by the following equation:

EVB=Eg+ ECB

The ESR experiments of g-C3N4 and g-C3N4/NiO (4:6) were con-ducted to analyze the heterojunction mechanism by investigating thepresence of superoxide radicals (%O2

−) and hydroxyl radical (%OH)(Tang et al., 2018b; Zhang et al., 2014). It shows the detection of %O2

−

using DMPO−CH3OH as trapping agent. The signal of %O2− are not

detected in dark for g-C3N4 and g-C3N4/NiO (4:6), while %O2− signal

could be detected for both g-C3N4 and g-C3N4/NiO (4:6) under UV–vis

Fig. 5. Adsorption and desorption curves (a), UV–vis DRS (c) and PL (d) spectra of g-C3N4 and g-C3N4/NiO composites, EDX element spectrum of g-C3N4/NiO (4:6)(b).

Table 1Specific surface area average pore diameter and total pore volume of differentcatalysts measured by the adsorption/desorption isotherms of nitrogen.

Sample Sbet (m2 g−1) daverage (nm) Vtotal (cm3 g−1)

g-C3N4 6.29 36.04 0.06g-C3N4/NiO (4:6) 62.21 15.77 0.25g-C3N4/NiO (5:5) 56.45 16.76 0.24g-C3N4/NiO (6:4) 54.80 17.50 0.24g-C3N4/NiO (8:2) 53.47 18.35 0.25

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

5

illumination. It showed that the characteristic peak of %O2− for g-C3N4/

NiO (4:6) are higher than that of g-C3N4, which suggests that more%O2

− (EΘ (%O2-/O2)=−0.33 eV vs. NHE) are generated from the g-C3N4/NiO (4:6) than that of g-C3N4. The electrons in the CB of g-C3N4

and holes in the VB of NiO are efficiently separated. Consequence, moreelectrons in the CB of g-C3N4 can react with O2 to generate more %O2

−

which is consistent with the experiment result.The excitation and transfer of charge carriers in g-C3N4 and g-C3N4/

NiO heterojunction composites were carried out by photoelec-trochemical measurements. We all know that the photocatalytic activityis closely related to the transfer and separation efficiency of photo-induced electron-hole pairs. The higher electron-hole separation effi-ciency always causes better properties of the photocatalytic system(Long et al., 2018). As shown in Fig. 7, the responses of the photo-current in two on-off cycles under UV–vis irradiation were studied, andthen the images of i–t curves and electrochemical impedance spectra(EIS) were acquired. The results of transient photocurrents provideclearer evidence of electron transfer within the heterojunction forma-tion. The light on-off test presented the uniform and fast photocurrent

response, indicating stable photocatalytic reactivity. The pristine g-C3N4 shows the weakest photocurrent response (0.10 μA cm−2) inFig. 7a. It shows a rapid upward trend once the light is on and a sharpdownward trend once the light is off, which may be caused by theelectron capture and release mediated by surface defects (Liu et al.,2018). When g-C3N4 was combined with NiO, the photoresponse ca-pacity was significantly improved. The photocurrent density of g-C3N4/NiO composites follows the order of g-C3N4/NiO (4:6)> g-C3N4/NiO(5:5)> g-C3N4/NiO (6:4). As expected, the g-C3N4/NiO (4:6) compo-site exhibits the highest photocurrent intensity, indicating the chargetransfer with high efficiency. The elevated photocurrent density couldaccelerate the migration of charge carriers.

Fig. 7b shows the Nyquist plots of electrochemical impedancespectra of g-C3N4/NiO heterojunction composites. The semicircle radiiincrease in the order of g-C3N4/NiO (4:6)< g-C3N4/NiO (5:5)< g-C3N4/NiO (6:4)< g-C3N4. In general, the semicircle diameter of Ny-quist plots is equal to the interfacial charge transfer resistance (Rct)(Randviir and Banks, 2013). As shown in Fig. 7b, the smallest semicirclediameter was g-C3N4/NiO (4:6), suggesting that electron transfer is

Fig. 6. The plot of (αhv)1/2 and (αhv)2 versus energy (hv) for the band gap energies of g-C3N4 and NiO (a), Mott-Schottky plots of g-C3N4 and NiO (b), ESR signals ofthe DMPO-%O2

− of g-C3N4 (c) and g-C3N4/NiO (4:6) (d) irradiation of UV–vis light for 30min.

Fig. 7. Transient photocurrent (a) and Nyquist plots (b) of the electrochemical impedance spectrum of g-C3N4/NiO composites.

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

6

accelerated in this heterostructure composite. It is consistent with itsphotocatalytic activity. Both the i–t curves and Nyquist plots demon-strate that the heterojunction structure could improve the separationand transfer efficiency of charge carriers, thus increasing the photo-catalytic activity.

3.2. Dechlorination of 2-CDD by samples

Figs. 8 and 9 shows the typical chromatogram for the dechlorinationof 2-CDD by the g-C3N4/NiO (4:6) composite under UV–vis irradiation.The main product was identified by matching the mass spectra, whichcoincided well with their authentic standards. A previous study showedthat 2-CDD and DD presented intense precursor ions at 218 and 186(Ding et al., 2018). In the absence of catalysis or without light irra-diation, 2-CDD was fairly stable in the reaction solution, no measurable2-CDD loss was observed and no DD was detected over the reactioncourse. However, the 2-CDD concentration steadily decreased with theincreased concentration of DD accordingly when irradiated with UV–vislight (Fig. 9), indicating that 2-CDD was transformed to the final pro-duct (DD). Previous studies have reported that NiO combined with g-C3N4 are efficiently used in the reduction of CO2 (Tang et al., 2018b),water splitting (Fu et al., 2018), methylene blue (Chen et al., 2014) andCongo red and malachite green degradation (Estrellan and Iino, 2010).Fig. 8a and b reveals the photocatalytic dechlorination activities of 2-CDD over g-C3N4, g-C3N4/NiO (4:6) and g-C3N4/NiO (6:4), and thecarbon mass balances are also mentioned. For example, approximately22.5% of 2-CDD disappeared in 8 h by pure g-C3N4 and 10.2% of 2-CDDwas converted to DD. Similarly, approximately 70.4% of 2-CDD over g-

C3N4/NiO (4:6) with 52.2% was converted to DD under the samecondition (Fig. 8b). Approximately 8% carbon mass loss was observed,likely resulting from the existence of intermediate products. It wasproven that the concentration of 2-CDD decreased rapidly within thefirst 1 h, while the amount of generated product was less than the re-duced 2-CDD, which was the same as in a previous study (Ding et al.,2018).

Fig. 10a displays the effects of the NiO content on the photocatalyticdechlorination efficiency of 2-CDD. Obviously, the efficiency increasedwith the NiO content from 0 to 60%. The activity improvement may bedue to the increased amount of heterojunctions. The higher proportionof NiO could accelerate the reaction due to the interaction with 2-CDDbeing enabled more sufficiently. A maximum 2-CDD production amountwas achieved on the g-C3N4/NiO (4:6) catalyst during the 8 -h reaction,about fourfold as high as that of g-C3N4. NiO may play a crucial part inpromoting electron transport, delaying electron hole recombinationand enhancing reaction activity. The oxide-support interaction of NiOcan be also found on other photocatalysts, such as NiO supported on theFe2O3 (Singh and Sarkar, 2018), TiO2 (Li et al., 2015) and InNbO4

catalysts (Lee et al., 2012).Fig. 10b compares the transformation of 2-CDD at 8 h by g-C3N4/

NiO (4:6) obtained at different temperatures, indicating that NiOcombined with g-C3N4 at 400 °C showed the best photocatalytic ac-tivity, while crystalline NiO-modified g-C3N4 at 500 °C showed slightlypoor photocatalytic activity. Fig. 10c shows the dosage effect of the g-C3N4/NiO (4:6) catalyst on the degradation efficiency toward 2-CDD.For example, the dechlorination reaction increased with the g-C3N4/NiO (4:6) dosage varying from 0.1 g/L to 1.5 g/L because the largerdose of the g-C3N4/NiO (4:6) catalyst can provide more reactive sitesfor the dechlorination of 2-CDD. Fig. 10d shows that the additionaloxygen can greatly inhibit the degradation of 2-CDD in systems. Pre-vious studies have confirmed that the g-C3N4/NiO catalyst mediates atwo-electron pathway to realize water splitting, where the electrons inthe NiO conduction band reduce H2, and the holes in the g-C3N4 va-lence band oxidize to produce protons (Liu et al., 2018; Fu et al., 2018).In this study, the holes were rapidly quenched by isopropanol, leadingto electron accumulation in the CB of g-C3N4. However, the additionaloxygen may greatly consume electrons and H+ such that the efficiencyof the reaction was severely inhibited.

A continuous cycle test was conducted to observe the reusability ofthe g-C3N4/NiO composites, as shown in Fig. 11. It was obvious that noprominent deactivation was observed after seven cycling runs, de-monstrating that the g-C3N4/NiO (4:6) heterojunction has superiorstability. However, the activity was a slight decrease after each cycle,possibly due to the loss of catalysts during washing and filtration.

Fig. 8. Dechlorination of 2-CDD (a) and formation of DD (b) by g-C3N4 and g-C3N4/NiO (4:6) in isopropanol/water (v/v= 3:7) solution at 25 ± 1 °C. The reaction ofsolutions initially contained 4.59 μmol/L of 2-CDD and 1 g/L of photocatalysts.

Fig. 9. Typical chromatogram spectrum for DD from the transformation of 2-CDD by g-C3N4/NiO (4:6) under the irradiation of UV–vis light.

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

7

3.3. Photocatalytic mechanism

Overall, the possible reaction mechanism of photocatalytic reduc-tion for the dechlorination of 2-CDD by g-C3N4/NiO heterojunctioncomposites is presented in Scheme 1. The band alignment is an im-portant factor to determine the flow field of photogenerated carriers inheterogeneous structures (Wang et al., 2018). According to theabovementioned findings, the CB and VB potentials of g-C3N4 (-0.78 eVvs. NHE and 1.92 eV vs. NHE, respectively) are more negative thanthose of NiO (-0.5 eV vs. NHE and 2.9 eV vs. NHE). In this case, themigration paths of charge carriers may show that photogeneratedelectrons in the CB of g-C3N4 move to the CB of NiO and holes in the VBof NiO migrate to the VB of g-C3N4, called the “heterojunction”. Whenthe electrons and holes accumulate to equilibrium, the internal electric

field is formed. Thus, when the holes were quickly quenched by iso-propanol, the electrons on the CB of g-C3N4 react with 2-CDD convertedto DD selectively. Obviously, this pattern can isolate charge carriersspatially by forming a junction and maintaining a high redox capacity.Moreover, g-C3N4/NiO catalysts enlarged the surface area of structures,providing massive active sites, another main reason for superior pho-tocatalytic activity (Wang et al., 2019b). Furthermore, a previous studyshowed that the existence of Ni3+ on the NiO surface could enhance theabsorption of light and reduce the electric resistance of NiO (Guo et al.,2010). Additionally, the presence of oxygen can greatly inhibit thedegradation of 2-CDD in systems, which may be due to consumingmany electrons and hydrogen ions.

According to the discussion above, the mechanism of the de-chlorination of 2-CDD could be described as follows. The g-C3N4/NiOheterojunctions can be excited to produce photogenerated electron-holepairs under UV–vis light illumination. Owing to the intimate interfacecontact and well-matched band structure, the electrons in the CB of g-C3N4 are effectively migrated to the CB of NiO. Meanwhile, the holes inthe VB of g-C3N4 are rapidly quenched by isopropanol to accumulatephotogenerated electrons in the CB of g-C3N4. Next, the accumulatedelectrons with strong reducing ability react with 2-CDD to reduce intothe final product.

4. Conclusion

In summary, a series of NiO-modified g-C3N4 catalysts with differentNiO contents were fabricated by the mixing and heating method andthe liquid phase photocatalytic reductive dechlorination of 2-CDD wasevaluated. The results of XRD and TEM demonstrated the successfulcombination of g-C3N4 and NiO. Meanwhile, 2-CDD could be effectivelyreduced to DD by g-C3N4/NiO heterojunctions under UV–vis light il-lumination. The optimal activity was g-C3N4/NiO (4:6) with thermaltreatment at 400 °C, whose dechlorination efficiency was about fourfoldhigher than that of the pure g-C3N4. The enhancement of activity wasdue to the extension of the absorption of light and improvement of e−-h+ separation efficiency by NiO, as illuminated in UV–vis DRS, PL, XPS,

Fig. 10. Effects of the NiO content (a), g-C3N4/NiO (4:6) dosage (b), g-C3N4/NiO (4:6) obtained temperatures (c), and additional O2 (d) on the dechlorination of 2-CDD, C0= 4.59 μmol/L.

Fig. 11. Reusability of g-C3N4/NiO (4:6) for 2-CDD degradation.

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

8

i–t, and EIS analyses. g-C3N4/NiO (4:6) showed superior stability afterseven consecutive cycles, while the existence of oxygen could greatlyinhibit the reactions. This study extends the use range of g-C3N4/NiOcomposites and develops a new technology for the degradation ofPCDDs with striking activity and stability.

Acknowledgment

This work was supported by the Key Research & Developmentproject of Science Technology Department of Zhejiang Province (GrantNo. 2017C02026), the Program for 151 Talents in Zhejiang Province(Grant No. 4108Z061700303), the Program for 131 Talents inHangzhou City (Grant No. 4105F5061700104). The authors also highlyappreciate the assistance of instruments and expert analysis by Prof.Weirong Zhao in Zhejiang University.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in theonline version, at doi:https://doi.org/10.1016/j.jhazmat.2019.121255.

References

Arfaeinia, H., Asadgol, Z., Ahmadi, E., Seifi, M., Moradi, M., Dobaradaran, S., 2017.Characteristics, distribution and sources of polychlorinated biphenyls (PCBs) incoastal sediments from the heavily industrialized area of Asalouyeh, Iran. Water Sci.Technol. 76, 3340–3350.

Chen, H.Y., Qiu, L.G., Xiao, J.D., Ye, S., Jiang, X., Yuan, Y.P., 2014. Inorganic-organichybrid NiO-g-C3N4 photocatalyst for efficient methylene blue degradation usingvisible light. RSC Adv. 4, 22491–22496.

Dai, W., Pan, X.H., Chen, S.S., Chen, C., Wen, Z., Zhang, H.H., Ye, Z.Z., 2014. Honeycomb-like NiO/ZnO heterostructured nanorods: photochemical synthesis, characterization,and enhanced UV detection performance. J. Mater. Chem. C 2, 4606–4614.

Ding, J., Long, G., Luo, Y., Sun, R., Chen, M., Li, Y., Zhou, Y., Xu, X., Zhao, W., 2018.Photocatalytic reductive dechlorination of 2-chlorodibenzo-p-dioxin by Pd modifiedg-C3N4 photocatalysts under UV–vis irradiation: efficacy, kinetics and mechanism. J.Hazard. Mater. 355, 74–81.

Dong, G., Zhang, L., 2012. Porous structure dependent photoreactivity of graphitic carbonnitride under visible light. J. Mater. Chem. 22, 1160–1166.

Estrellan, C.R., Iino, F., 2010. Toxic emissions from open burning. Chemosphere 80,

193–207.Feng, C.Y., Deng, Y.C., Tang, L., Zeng, G.M., Wang, J.J., Yu, J.F., Liu, Y.N., Peng, B., Feng,

H.P., Wang, J.J., 2018. Core-shell Ag2CrO4/N-GQDs@g-C3N4 composites with anti-photocorrosion performance for enhanced full-spectrum-light photocatalytic activ-ities. Appl. Catal. B 239, 525–536.

Fu, Y., Liu, Ca., Zhu, C., Wang, H., Dou, Y., Shi, W., Shao, M., Huang, H., Liu, Y., Kang, Z.,2018. High-performance NiO/g-C3N4 composites for visible-light-driven photo-catalytic overall water splitting. Inorg. Chem. Front. 5, 1646–1652.

Guemiza, K., Coudert, L., Metahni, S., Mercier, G., Besner, S., Blais, J.-F., 2017. Treatmenttechnologies used for the removal of As, Cr, Cu, PCP and/or PCDD/F from con-taminated soil: a review. J. Hazard. Mater. 333, 194–214.

Guo, J., Fu, W., Yang, H., Yu, Q., Zhao, W., Zhou, X., Sui, Y., Ding, J., Li, Y., Cheng, S., Li,M., 2010. A NiO/TiO2 junction electrode constructed using self-organized TiO2 na-notube arrays for highly efficient photoelectrocatalytic visible light activations. J.Phys. D: Appl. Phys. 43, 24 245202.

Ji, L., Cao, X., Lu, S., Du, C., Li, X., Chen, T., Buekens, A., Yan, J., 2018. Catalytic oxi-dation of PCDD/F on a V2O5-WO3/TiO2 catalyst: effect of chlorinated benzenes andchlorinated phenols. J. Hazard. Mater. 342, 220–230.

Jiang, F., Tan, W., Chen, H., Tan, L., Liu, J., 2015. Effective catalytic hydrodechlorinationof chlorophenoxyacetic acids over Pd/graphitic carbon nitride. RSC Adv. 5,51841–51851.

Lechner, M.B., 2009. Anaerobic reductive dehalogenation of polychlorinated dioxins.Appl. Microbiol. Biotechnol. 84, 429–444.

Lee, D.S., Chen, H.J., Chen, Y.W., 2012. Photocatalytic reduction of carbon dioxide withwater using InNbO4 catalyst with NiO and Co3O4 cocatalysts. J. Phys. Chem. Solids73, 661–669.

Li, Y., Zhang, H., Liu, P., Wang, D., Zhao, H., 2013. Cross-linked g-C3N4/rGO nano-composites with tunable band structure and enhanced visible light photocatalyticactivity. Small 9, 3336–3344.

Li, L., Cheng, B., Wang, Y., Yu, J., 2015. Enhanced photocatalytic H-2-production activityof bicomponent NiO/TiO2 composite nanofibers. J. Colloid Interface Sci. 449,115–121.

Li, C., Yang, L., Shi, M., Liu, G., 2019. Persistent organic pollutants in typical lake eco-systems. Ecotoxicol. Environ. Saf. 180, 668–678.

Lin, X., Wei, B., Zhang, X., Song, M., Yu, S., Gao, Z., Zhai, H., Zhao, L., Che, G., 2016.Construction of heterostructured g-C3N4/Ag/Bi3.64Mo0.36O6.55 nanospheres with en-hanced visible-light photocatalytic activity. Sep. Purif. Technol. 169, 9–16.

Liu, M.C., Chang, S.H., Chang, M.B., 2016. Catalytic hydrodechlorination of PCDD/Fsfrom condensed water with Pd/g-Al2O3. Chemosphere 154, 583–589.

Liu, J., Jia, Q., Long, J., Wang, X., Gao, Z., Gu, Q., 2018. Amorphous NiO as co-catalyst forenhanced visible-light-driven hydrogen generation over g-C3N4 photocatalyst. Appl.Catal. B 222, 35–43.

Liu, Q.Q., Shen, J.Y., Yu, X.H., Yang, X.F., Liu, W., Yang, J., Tang, H., Xu, H., Li, H.M., Li,Y.Y., Xu, J.S., 2019. Unveiling the origin of boosted photocatalytic hydrogen evo-lution in simultaneously (S, P, O)-codoped and exfoliated ultrathin g-C3N4 na-nosheets. Appl. Catal. B 248, 84–94.

Long, G., Ding, J., Xie, L., Sun, R., Chen, M., Zhou, Y., Huang, X., Han, G., Li, Y., Zhao, W.,

Scheme 1. Schematic illustration of the 2-CDD dechlorination process over g-C3N4/NiO composites under irradiation of UV–vis light.

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

9

2018. Fabrication of mediator-free g-C3N4/Bi2WO6 Z-scheme with enhanced photo-catalytic reduction dechlorination performance of 2,4-DCP. Appl. Surf. Sci. 455,1010–1018.

Mishra, A., Mehta, A., Basu, S., Shetti, N.P., Reddy, K.R., Aminabhavi, T.M., 2019.Graphitic carbon nitride (g-C3N4)-based metal-free photocatalysts for water splitting:a review. Carbon 149, 693–721.

Mumbo, J., Pandelova, M., Mertes, F., Henkelmann, B., Bussian, B.M., Schramm, K.-W.,2016. The fingerprints of dioxin-like bromocarbazoles and chlorocarbazoles in se-lected forest soils in Germany. Chemosphere 162, 64–72.

Niu, P., Zhang, L., Liu, G., Cheng, H.-M., 2012. Graphene-like carbon nitride nanosheetsfor improved photocatalytic activities. Adv. Funct. Mater. 22, 4763–4770.

Pagano, J.J., Garner, A.J., McGoldrick, D.J., Crimmins, B.S., Hopke, P.K., Milligan, M.S.,Holsen, T.M., 2018. Age-corrected trends and toxic equivalence of PCDD/F and CP-PCBs in Lake Trout and walleye from the Great Lakes: 2004–2014. Environ. Sci.Technol. 52, 712–721.

Qiu, Q.L., Chen, Q., Jiang, X.G., Lv, G.J., Chen, Z.L., Lu, S.Y., Ni, M.J., Yan, J.H., Lin, X.L.,Song, H.B., Cao, J.J., 2019. Improving microwave-assisted hydrothermal degradationof PCDD/Fs in fly ash with added Na2HPO4 and water-washing pretreatment.Chemosphere 220, 1118–1125.

Randviir, E.P., Banks, C.E., 2013. Electrochemical impedance spectroscopy: an overviewof bioanalytical applications. Anal. Methods 5, 1098–1115.

Rashidian, N., Zahedi, E., Shiroudi, A., 2019. Kinetic and mechanistic insight into the OHcenter dot-initiated atmospheric oxidation of 2,3,7,8-tetrachlorodibenzo-p-dioxin viaOH center dot-addition and hydrogen abstraction pathways: a theoretical investiga-tion. Sci. Total Environ. 679, 106–114.

Sahoo, P., Misra, D.K., Salvador, J., Makongo, J.P.A., Chaubey, G.S., Takas, N.J., Wiley,J.B., Poudeu, P.F.P., 2012. Microstructure and thermal conductivity of surfactant-freeNiO nanostructures. J. Solid State Chem. 190, 29–35.

Selvarajan, S., Suganthi, A., Rajarajanc, M., 2018. Fabrication of g-C3N4/NiO hetero-structured nanocomposite modified glassycarbon electrode for quercetin biosensor.Ultrason. Sonochem. 41, 651–660.

Shi, W., Guo, F., Chen, J., Che, G., Lin, X., 2014. Hydrothermal synthesis of InVO4/gra-phitic carbon nitride heterojunctions and excellent visible-light-driven photocatalyticperformance for rhodamine B. J. Alloys Compd. 612, 143–148.

Singh, A.K., Sarkar, D., 2018. A facile approach for preparing densely-packed individualp-NiO/n-Fe2O3 heterojunction nano-wires for photoelectrochemical water splitting.Nanoscale 10, 13130–13139.

Sreethawong, T., Suzuki, Y., Yoshikawa, S., 2005. Photocatalytic evolution of hydrogenover mesoporous TiO2 supported NiO photocatalyst prepared by single-step sol-gelprocess with surfactant template. Int. J. Hydrogen Energy 30, 1053–1062.

Ssebugere, P., Sillanpaa, M., Matovu, H., Mubiru, E., 2019. Human and environmentalexposure to PCDD/Fs and dioxin-like PCBs in Africa: a review. Chemosphere 223,483–493.

Tang, L., Feng, C.Y., Deng, Y.C., Zeng, G.M., Wang, J.J., Liu, Y.N., Feng, H.P., Wang, J.J.,2018a. Enhanced photocatalytic activity of ternary Ag/g-C3N4/NaTaO3 photo-catalysts under wide spectrum light radiation: the high potential band protectionmechanism. Appl. Catal. B 230, 102–114.

Tang, J.Y., Guo, R.T., Zhou, W.G., Huang, C.Y., Pan, W.G., 2018b. Ball-flower like NiO/g-C3N4 heterojunction for efficient visible light photocatalytic CO2 reduction. Appl.Catal. B 237, 802–810.

Tang, H., Wang, R., Zhao, C.X., Chen, Z.P., Yang, X.F., Bukhvalov, D., Lin, Z.X., Liu, Q.Q.,2019. Oxamide-modifified g-C3N4 nanostructures: tailoring surface topography forhigh-performance visible light photocatalysis. Chem. Eng. J. 374, 1064–1075.

Thaweesak, S., Lyu, M., Peerakiatkhajohn, P., Butburee, T., Luo, B., Chen, H.J., Wang,L.Z., 2017. Two-dimensional g-C3N4/Ca2Nb2TaO10 nanosheet composites for efficient

visible light photocatalytic hydrogen evolution. Appl. Catal. B 202, 184–190.Thiyagarajan, K., Samuel, S., Babu, S.G., 2018. NiO/C3N4: hybrid photocatalyst for the

enhanced photodegradation of organic pollutant under visible light. Mater. Res.Express 5.

Tong, S., Zheng, M., Lu, Y., Lin, Z., Li, J., Zhang, X., Shi, Y., He, P., Zhou, H., 2015a.Mesoporous NiO with a single-crystalline structure utilized as a noble metal-freecatalyst for non-aqueous Li-O-2 batteries. J. Mater. Chem. A 3, 16177–16182.

Tong, S.F., Zheng, M.B., Lu, Y., Lin, Z.X., Li, J., Zhang, X.P., Shi, Y., He, P., Zhou, H.S.,2015b. Mesoporous NiO with a single-crystalline structure utilized as a noble metal-free catalyst for non-aqueous Li–O2 batteries. J. Mater. Chem. A 3, 16177–16182.

Wang, X., Maeda, K., Thomas, A., Takanabe, K., Xin, G., Carlsson, J.M., Domen, K.,Antonietti, M., 2009. A metal-free polymeric photocatalyst for hydrogen productionfrom water under visible light. Nat. Mater. 8, 76–80.

Wang, W., Li, G., An, T., Chan, D.K.L., Yu, J.C., Wong, P.K., 2018. Photocatalytic hy-drogen evolution and bacterial inactivation utilizing sonochemical-synthesized g-C3N4/red phosphorus hybrid nanosheets as a wide-spectral-responsive photocatalyst:the role of type I band alignment. Appl. Catal. B 238, 126–135.

Wang, D.J., Saleh, N.B., Sun, W.J., Park, C.M., Shen, C.Y., Aich, N.A., Peijnenburg,W.J.G.M., Zhang, W., Jin, Y., Su, C.M., 2019a. Next-generation multifunctional car-bon–metal nanohybrids for energy and environmental applications. Environ. Sci.Technol. 53, 7265–7287.

Wang, D., Li, J., Xu, Z., Zhu, Y., Chen, G., Cui, Z., 2019b. Synthesis of g-C3N4/NiO p-nheterojunction materials with ball-flower morphology and enhanced photocatalyticperformance for the removal of tetracycline and Cr6+. J. Mater. Chem. 54,11417–11434.

Wu, C.H., Ng, H.-Y., 2008. Photodegradation of polychlorinated dibenzo-p-dioxins andpolychlorinated dibenzofurans: direct photolysis and photocatalysis processes. J.Hazard. Mater. 151, 507–514.

Xu, M., Han, L., Dong, S., 2013. Facile fabrication of highly efficient g-C3N4/Ag2O het-erostructured photocatalysts with enhanced visible-light photocatalytic activity. ACSAppl. Mater. Interfaces 5, 12533–12540.

Xu, J.X., Qi, Y.H., Wang, C., Wang, L., 2019. NH2-MIL-101(Fe)/Ni(OH)(2)-derived C,N-codoped Fe2P/Ni2P cocatalyst modified g-C3N4 for enhanced photocatalytic hy-drogen evolution from water splitting. Appl. Catal. B 241, 178–186.

Yu, Y., Guo, L., Cao, H., Lv, Y., Wang, E., Cao, Y., 2015. A new Ni/Ni-3(BO3)(2)/NiOheterostructured photocatalyst with efficient reduction of CO2 into CH4. Sep. Purif.Technol. 142, 14–17.

Zhang, Z., Shao, C., Li, X., Wang, C., Zhang, M., Liu, Y., 2010. Electrospun nanofibers of p-type NiO/n-type ZnO heterojunctions with enhanced photocatalytic activity. ACSAppl. Mater. Interfaces 2, 2915–2923.

Zhang, C., Hao, R., Liao, H., Hou, Y., 2013. Object-based benthic habitat mapping in theFlorida keys from hyperspectral imagery. Nano Energy 2, 88–97.

Zhang, J.F., Hu, Y.F., Jiang, X.L., Chen, S.F., Meng, S.G., Fu, X.L., 2014. Design of a directZ-scheme photocatalyst: preparation and characterization of Bi2O3/g-C3N4 with highvisible light activity. J. Hazard. Mater. 280, 713–722.

Zhang, Y.Y., Wang, J.X., Wan, P.P., Ye, J.H., Hussain, S., Wei, H.M., Hou, D.W., 2016.Ag2O loaded NiO ball-flowers for high performance supercapacitors. Mater. Lett. 177,71–75.

Zhang, C., Li, Y., Shuai, D., Shen, Y., Xiong, W., Wang, L., 2019a. Graphitic carbon nitride(g-C3N4)-based photocatalysts for water disinfection and microbial control: a review.Chemosphere 214, 462–479.

Zhang, S., Liu, Y., Gu, P.C., Ma, R., Wen, T., Zhao, G.X., Li, L., Ai, Y.J., Hu, C., Wang, X.K.,2019b. Enhanced photodegradation of toxic organic pollutants using dual-oxygen-doped porous g-C3N4: mechanism exploration from both experimental and DFT stu-dies. Appl. Catal. B 248, 1–10.

J. Ding, et al. Journal of Hazardous Materials 384 (2020) 121255

10