Photocatalytic Activities of g-C3N4 (CN) Treated with Nitric Acid Vapor for the Degradation of Pollutants in Wastewater

In this article, we reported a novel setup treatment using nitric acid vapor to treat g-C3N4 (CN). By treatment with nitric acid vapour, the basic structure of the CN has not been destroyed. These adoptive treatments enhanced the photocatalytic performance of CN and were reflected in the elimination of rhodamine B (RhB) as well as tetracycline hydrochloride (TC). In comparison to CN, CN-6 demonstrated the highest photocatalytic yield for the breakdown of RhB (99%, in 20 min). Moreover, the excellent reuse of CN-6 for breaking down RhB was also demonstrated. This clearly demonstrated that treatment with nitric acid vapor promoted a blue shift, positively extended its valence band position, and increased the oxidizability of the holes. This also caused CN to disperse better into the aqueous phase, introducing more oxygen-containing functional groups. Thus, treatment with nitric acid vapor has the potential to be applied to delaminate the CN in order to enhance photocatalytic activity.


Introduction
Due to industrialization and population growth, many countries have critical environmental issues, such as energy crises and pollution of the environment [1]. Nowadays, the development of state-of-the-art technologies to address these issues at a global scale is of great importance [2,3]. Among various proposed technologies, photocatalytic technology has tremendously gained global attention to alleviate environmental contamination because of its ability to make direct use of inexhaustible solar power [4][5][6]. Although one of the earliest investigated photocatalysts is titanium dioxide (TiO 2 ), it exhibits significant limitations due to the large band gap and responds only to ultraviolet light [7,8]. As a visible light active photocatalyst, metal-free graphitic carbon nitride (g-C 3 N 4 ) has recently attracted a lot of attention. It has distinctive properties for photocatalysis, such as a tunable band gap (2.7 eV), single optical properties to harvest sunlight, high stability to resist photo-corrosion, suitable electronic properties, environmental friendliness, and low cost [9][10][11][12][13]. Though the g-C 3 N 4 exhibited immense potential in the photocatalysis field, it still suffers from poor active sites, a low redox potential of photogenerated charge carriers, a low quantum yield, and a poor dispersion of the aqueous phase [14,15].
Several modifications have been developed to ameliorate the photocatalytic performance of the bulk g-C 3 N 4 , such as element doping [16][17][18], heterojunction building [19,20], and morphology controlling [21,22]. Motivated by the pioneering work on graphene [23], many researchers have made efforts to delaminate the bulk g-C 3 N 4 into 2D thinner counterparts and explore their novel properties. The ultrathin 2D g-C 3 N 4 nanosheet has distinct physicochemical properties that set it apart from bulk g-C 3 N 4 , such as a broader bandgap, a higher redox ability of charge carriers, an open-up surface, and an extended life for charge carriers. These properties give g-C 3 N 4 nanosheets improved photocatalytic performance in real-world applications [24][25][26].
Niu and his coworkers first developed g-C 3 N 4 nanosheets by long-time thermal oxidation exfoliation that increased bandgap and ameliorated the capacity of electron transport. The super photocatalytic activities of g-C 3 N 4 nanosheets as photocatalysts were established via H 2 evolution [27]. Yan et al. successfully exfoliated g-C 3 N 4 in hot water to obtain ultrathin g-C 3 N 4 . The as-prepared g-C 3 N 4 ultrathin has a thickness of 1.2 nm after 12 h of treatments, with a wide lateral dimension and improved photoactivity for the degradation of organic dye when exposed to visible light [28]. Miao et al. synthesized 2D g-C 3 N 4 nanosheets with acid treatment that demonstrated an excellent photocurrent response close to 0.5 mA/cm 2 and lower charge transfer resistance [29]. Although the majority of the currently used techniques (such as thermal exfoliation, liquid etching, and chemical exfoliation using potent oxidizing acids) can exfoliate bulk g-C 3 N 4 into a single layer, they are also ineffective because the tri-s-triazine units are destroyed, the processing time is prolonged, and the yield of nanosheets is subpar [30]. Therefore, it is advisable to develop an advanced strategy with high efficiency and low energy consumption for delaminating bulk g-C 3 N 4 to ultrathin g-C 3 N 4 nanosheets.
Herein, we present a simple process to prepare g-C 3 N 4 nanosheets through nitric acid vapor treatment using a new practical carbon material processor invented by our research group. Compared with aqueous exfoliation, the process has the merits of a controllable treatment time, avoiding over-oxidation, easy collection of products, and high yield. The obtained nanosheets were studied according to the crystal structure, surface group, composition, and physicochemical properties. The g-C 3 N 4 nanosheets were found to exhibit a stronger oxidizing capacity and disperse well into an aqueous solution. As anticipated, when exposed to visible light, g-C 3 N 4 nanosheets demonstrated outstanding photocatalytic activity and stability for the destruction of RhB and TC. Moreover, a proper mechanism and the active species for the degradation of RhB were also studied through scavenger experiments.

Materials
Urea was bought from Bodi (Tianjin, China). Tetracycline hydrochloride (TC) was obtained from Shanghai Ruiyong Biotechnology Co., Ltd (Shanghai, China). Triethanolamine (TEOA), concentrated nitric acid and absolute ethanol were purchased from Fuyu (Tianjin, China). Rhodamine B (RhB) was purchased from Kemiou (Tianjin, China). p-Benzoquinone (p-BQ) was purchased from Shanghai Macklin Biochemical Co. Ltd (Shanghai, China). All reagents were of analytical grade and used with no further purification.

Synthesis of the Photocatalysts
The bulk g-C 3 N 4 was prepared using the method reported previously [31]. In a nutshell, 20.0 g of urea was placed in a 100 mL crucible with a cover slip and heated in a muffle furnace to 550 • C for 3 h at a rate of 17.8 • C/min. After being allowed to cool to ambient temperature, the resulting yellow substance was ground into a powder in an agate mortar. The synthesized sample was denoted as CN.
The alteration to the bulk g-C 3 N 4 was treated with nitric acid vapor. Some concentrated HNO 3 was poured into a flask with three necks, and g-C 3 N 4 was added to the processing apparatus as shown in Figure 1. The self-assembling device was then placed in a 170 • C oil bath for a period of time. According to the treatment time, the g-C 3 N 4 was named CN-X (X = 2, 4, 6, 8 h).

Characterization
X-ray diffraction (XRD) patterns were obtained by a Bruker D8 ADVANCE instrument using Cu-Kα radiation (λ = 0.15405 nm, 40 KV × 60 mA) from 10 • to 80 • (2θ) with a scanning rate of 0.02 • /s. X-ray photoelectron spectroscopy (XPS) was obtained by a Thermo ESCALAB 250XI with a monochromated Al Ka X-ray source. A Bruker ALPHA-T Fourier transform infrared spectrometer was used to record the Fourier transform infrared spectra (FT-IR) of the samples, with a range from 400 to 4000 cm −1 using the KBr pellet at room temperature. Brunauer-Emmett-Teller (BET) measurement was used to determine the specific surface area of samples (Quantachrome Instruments Quadrasorb SI, United States Quantachrome Co., Ltd, Boulder, CO, USA). Using a 100 kV accelerating voltage, a field emission scanning electron microscope (SEM) (JSM-6700) was used to characterize the morphologies of the samples. A UV-vis spectrophotometer was used to generate diffuse reflectance spectra (DRS) (SHIMADZU, UV-2550). Photoluminescence (PL) spectroscopy was performed at room temperature using a Hitachi F-4500 fluorescence spectrophotometer.

Photocatalytic Experiments
Under visible light irradiation, TC and RhB were used as target pollutants to evaluate the photocatalytic activities of the as-prepared samples. For photocatalytic experiments, a photocatalytic reactor with a 200 mL beaker as a reaction vessel and a 300W-Xenon lamp with a 420 nm cutoff filter as a visible light source was used, with constant concurrent magnetic stirring. Typically, 25 mg photocatalysts were introduced into 100 mL RhB solution (10mg/L) and 50 mg photocatalysts were introduced into 100 mL TC solution (20 mg/ L). The mixture was then sonicated for 5 min and magnetically stirred in the dark for 30 min to achieve an equilibrium between adsorption and desorption. After a specific time interval of irradiation, 4 mL of the suspension was removed and filtered through a 0.22 µm filter membrane to remove the photocatalyst. A UV-vis spectrophotometer was used to measure the concentrations of TC and RhB. (SP-756PC, Shanghai Spectrum Instruments Co., Ltd, Shanghai, China).

Determination of Active Species during Photocatalytic Degradation
Radical trapping experiments were performed to ensure the presence of active species in the RhB photo-degradation process. In this study, p-Benzoquinone (p-BQ), triethanolamine (TEOA), and isopropanol (IPA) were used as scavengers of holes (h + ), hydroxyl radicals (•OH), and superoxide radicals (•O 2 − ), respectively. Prior to reaching the adsorption equilibrium, the capture agents were added to the RhB solution (10 mg/L) and agitated for 30 min without light. The photocatalytic reaction experiment was then performed.

Results and Discussion
3.1. Characterization of the As-Prepared Photocatalysts 3.1.1. Structure and Morphology Characterizations Figure 2 displays the XRD patterns of the samples to examine the impact of HNO 3 vapor on the crystalline phase of g-C 3 N 4 . The two distinct peaks at 12.97 • and 27.59 • for all samples were the in-plane (100) diffraction reflecting the interlayer stacking of the g-C 3 N 4 aromatic units and the interlayer (002) diffraction reflecting the structural packing pattern of the aromatic segments in the plane, respectively [32]. Nitric acid steam treatment improved the interlayer stacking interaction of the aromatic conjugated system by gradually increasing the (002) peak intensity of g-C 3 N 4 in comparison to untreated g-C 3 N 4 . This shows that nitric acid steam treatment can break the hydrogen bond between g-C 3 N 4 layers and cause the g-C 3 N 4 layers to peel off into nanostructures [33]. Henceforth, this confirms that the crystalline phase of g-C 3 N 4 is unaffected following treatment with nitric acid vapor. Fourier transform infrared (FT-IR) spectra were used to study the structural information of pure and processed g-C 3 N 4 . Figure 3 demonstrates that the C-N heterocycles' stretching vibrations at 1200-1650 cm −1 and the triazine units' distinctive breathing vibration at 810 cm −1 were present in all the samples, indicating that the tri-s-triazine units' chemical core structure is resistant to HNO 3 vapor erosion. However, as the bulk of g-C 3 N 4 was exfoliated, both the N-H and O-H stretching vibrations steadily increased, indicating that the increased density of the −OH group in g-C 3 N 4 was caused by the acid hydrolysis action of the bridging N groups [34].  X-ray photoelectron spectroscopy (XPS) was used to determine the elemental surface composition and chemical state of g-C 3 N 4 ( Figure 4). The XPS spectrum in Figure 4a indicates that all the samples are mainly composed of C, N, and O. Figure 4b shows the high-resolution XPS spectrum of C 1s for the bulk g-C 3 N 4 and treated g-C 3 N 4 , in which the four peaks at 284.6 eV, 286.0 eV, 288.1 eV and 289.4 eV can be attributed to C = C, C-NH 2 , N-C = N, and C-O [35]. When CN-6 was compared to bulk g-C 3 N 4 , the intensity of the peak at 286.0 eV and 289.4 eV was increased. This could be attributed to C-O groups formed during the acid hydrolysis of the tri-s-triazine unit, which bridged C-N-C [36]. The intensity of the peak at 286.0 eV and 289.4 eV increased when CN-6 was compared to bulk g-C 3 N 4 . C-O groups formed during the acid hydrolysis of the tri-s-triazine unit, which bridged C-N-C, could be responsible for this. Figure 4c displays the XPS spectra for O 1s of the bulk g-C 3 N 4 and CN-6 catalysts. The asymmetrical O 1s spectra of the two samples could be deconvoluted into these components at 533.5 eV, 532.1 eV, and 530.8 eV, attributable to the chemisorbed water, surface-adsorbed oxygen species, and surface lattice oxygen, respectively. Interestingly, the N 1s high-resolution spectrum (Figure 4d) revealed three typical peaks at 398.8 eV, 400.1 eV, and 404.4 eV corresponding to the C-N = C, −NH 2, and graphitic species, respectively [37]. The XPS results show the stability of g-C 3 N 4 preserved even after nitric acid vapor treatment.
The surface characteristics of samples from the N 2 adsorption-desorption isotherms ( Figure 5) were similar. A type-IV isotherm with a hysteresis loop was obtained in both the pure and treated g-C 3 N 4 , indicating the presence of mesoporous structures. CN-6's BET surface areas decreased by 25.7 m 2 /g when compared to bulk g-C 3 N 4 (105.6 m 2 /g), and its pore volume decreased by 0.278 ccg −1 . This is attributed to the bulk g-C 3 N 4 being exfoliated to nanosheets, then the partially completed nanosheets begin to aggregate and stack [38]. Meanwhile, the original macropore was destroyed [33]. The morphology of pure and treated g-C 3 N 4 can be illustrated via SEM images shown in Figure 6. From Figure 6a, we can see that g-C 3 N 4 samples with rough structures were composed of flakelike, aggregated, and different micron-sizes stacking layers, which are consistent with the typical characteristics of g-C 3 N 4 prepared through the method of polymerization [39,40]. In Figure 6b, bulk g-C 3 N 4 was exfoliated into nanosheets and morphologically changed to cubic and spherical shapes after nitric acid vapor treatment.

Optical Properties of the As-Prepared Photocatalysts
Diffuse reflectance UV-vis spectroscopy (DRS) was used to study the optical properties of the as-prepared samples. Figure 7a shows that a strong absorption band of bulk g-C 3 N 4 was extended up to 450 nm, indicating an apparent absorption from ultraviolet to visible light. When CN-X was delaminated into nanosheets, its absorption edge had a slightly blue shift when compared to CN. CN-X, on the other hand, showed significant absorption in visible light, indicating that these samples can absorb visible light and be used for visible light photocatalysis. This absorption could be attributed to the presence of the oxygen element in the CN-X framework following acid-exfoliated CN [41]. The bandgaps (Eg) of CN and CN-X were also estimated using a plot of (αhv) 1/2 vs. hv (n = 4 for indirect transition). Figure 7b shows that the Eg value increased from 2.69 eV for CN to 2.87eV for CN-6. This could be attributed to the ultrathin thickness's strong quantum confinement effect. It is common practice to use photoluminescence (PL) spectra as an indirect measure of charge carrier separation effectiveness. Figure 8 shows the PL spectra of g-C 3 N 4 and CN-6. In comparison to the bulk concentration of g-C 3 N 4 , CN-6 exhibited a somewhat blue shift of the intrinsic PL peak, which corresponds to the UV-vis spectrum. Furthermore, the lower PL intensity of bulk g-C 3 N 4 compared to CN-6, which may be due to surface defects of materials and stacking of nanosheets. It can be seen that a higher recombination rate is not the reason to improve the photocatalytic performance of the device in this paper [38].

Photocatalytic Activity and Stability
The photocatalytic performance of the as-prepared samples was assessed by the degradation of RhB and TC as a representative noxious pollutant under visible light irradiation conditions. The mixture of photocatalysts and impurities was magnetized and swirled in the dark for 30 min to create an adsorption-desorption equilibrium before being exposed to radiation. As shown in Figure 9a, the adsorption equilibrium was reached after 20 min of dark adsorption under these conditions, and we can obviously find that the adsorption capacity of CN-6 is stronger than that of CN. The large adsorption capability of the catalyst is advantageous to increase the photocatalytic performance because photocatalytic reactions take place on the catalyst surface. For comparison, a blank experiment without the photocatalyst was measured. The results in Figure 9a display that RhB can hardly decompose within 20 min of visible light irradiation, which can exclude the self-photolysis of RhB. The degradation performance values of CN, CN-2, CN-4, CN-6, and CN-8 were 89%, 80%, 78%, 99%, and 98%, respectively.
To gain insight into the photocatalytic activity of CN-X, the reaction kinetics of the degradation process were studied, and the result is shown in Figure 9b. The reaction kinetics were found to be appropriate for the pseudo-first-order model (Equation (1)), in which the slope of the fitted line can determine reaction rate constants (k).
−ln (C/C 0 ) = kt (1) where C and C 0 are the concentrations of RhB at irradiation times t and 0, and k is the apparent reaction rate constant. As a result, the pseudo-first-order-reaction kinetics (k) of RhB degradation relative to CN-6 was 0.16, which was twice as great as that of CN (0.08). The above result may be explained by the fact that the addition of functional groups containing oxygen causes the wells to disperse in the aqueous phase, which facilitates the catalyst's interaction with the contaminants. In addition, the absorption edge of CN-6 exhibits a somewhat blue shift after nitric acid vapor treatment, resulting in the valence band position of CN-6 becoming positively shifted and the hole potential has stronger oxidizability. In addition, to test its universality, we also employed CN-X for the photodegradation of tetracycline hydrochloride. As shown in Figure 9c, the degradation of TC by CN-X system shows the same trend as that of RhB. According to the above experiments, the catalytic performance of g-C 3 N 4 by nitric acid vapor treatment has been confirmed. The enhancement of photocatalytic activity of g-C 3 N 4 mainly occurs because of nanosheet delaminated and the introduction of oxygen-containing groups caused by nitric acid vapor treatment. Reusability and stability are vital for photocatalysts in the practical applications. In order to investigate the reusability of CN-6, five successive cycles were performed for photodegrading RhB under the same conditions. As shown in Figure 9d, the degradation stays almost unchanged in the second recycle, and the degradation efficiency of RhB was still as high as 93% after five cycles, demonstrating that CN-6 possessed superior reusability.

Discussion of Underlying Photocatalyst Mechanisms
Photoactive substances such as electrons, holes, superoxide radicals, and hydroxyl radicals are expected to photo-catalyze the degradation of RhB [42]. After using various quenchers to quench a variety of active species, we then aimed to determine the influence on the photocatalytic performance of CN-6, with the purpose of identifying the main active species in the photocatalytic process. Figure 10 shows that the degradation efficiency of RhB did not have an obvious decrease compared to the result without any scavenger after the addition of 10 mmol/L of isopropanol (IPA) as the OH quencher. This demonstrated that the predominant active species in the photocatalytic reaction were not the OH radicals. The fact that the degradation efficiency was unaffected by the addition of 10 mmol/L p-Benzoquinone (p-BQ) as an ·O 2 − scavenger further demonstrated that O 2 − was not the main active species. This is due to the conduction band of CN-6 becoming more positive, resulting in the weaker reduction of electrons, which further affects the formation of O 2 − . In contrast, with the 10 mmol/L of triethanolamine (TEOA) in the form of h + scavenger, the degradation efficiency of RhB decreased by 69%. This demonstrated that h + was the major active species of the photocatalytic process. Based on the valence band minimum (VBM) roughly evaluated from the valence band X-ray photoelectron spectroscopy (Figure 11a) and Eg result from UV-vis absorption spectra, the conduction band minimum (CBM) could be roughly calculated according to the following equation: E g = VBM-CBM As a result, Figure 11b provides a schematic depiction of the band structures of the bulk forms of g-C 3 N 4 and CN-6. The Eq. was expanded to 2.87 eV compared to the 2.69 eV of the bulk g-C 3 N 4 , which is attributed to the strong quantum confinement effect of the thickness of ultrathin. Correspondingly, the VBM of CN-6 was positively shifted to 1.96 eV from the 1.75 eV of bulk g-C 3 N 4 , indicating the oxidizability of the hole became stronger. Moreover, introducing oxygen-containing functional groups is said to promote visible light harvesting [28]. The process of RhB degradation over CN-6 is shown in Figure 12. In the presence of visible light irradiation, electrons on the VBM of CN-6 can be readily excited and transferred to the CBM of CN-6, thus leaving holes on the VBM. Photoinduced electrons will reduce O 2 adsorbed to the surface to generate O 2 − , which would be expected to oxidize RhB in an aqueous solution. The photoinduced holes will transfer to its surface and participate in the oxidation reaction [43]. More significantly, oxygen-containing functional groups such as C-OH and C = O can emerge and develop into electrophilic groups to prevent charge carriers from recombining. The high use of solar power, high surface area, and low recombination rate of photogenerated electron-hole pairs are widely recognized to be beneficial to the photo-activity of a photocatalyst. However, the BET result shows that CN-6's surface area somewhat decreased, and the DRS result shows that CN-6's absorption edge exhibited a minor blue shift after being exposed to nitric acid vapor. The high photocatalytic activity of CN-6 could thus be caused by the effective dispersal into the aqueous phase and a stronger oxidation of the hole potential. The results clearly demonstrate that nitric acid vapor treatment promoted a blue shift, positively extended its valence band position, and enhanced the oxidizability of holes.

Conclusions
In brief, CN-X was synthesized successfully by treatment with urea-pyrolysis and nitric acid vapors. CN-6 possessed excellent photocatalytic performance and stability, which is ascribed to the blue shift and the abundance of oxygen containing functional groups on surfaces after nitric acid vapor treatment. On the one hand, the introduction of oxygen-containing functional groups in the process of lamellar exfoliation can make g-C 3 N 4 disperse better in the aqueous phase, which facilitates the contact of the photocatalyst with pollutants. On the other hand, the absorption edge of CN-6 exhibits a somewhat blue shift, and the valence band position of the CN-6 is positively shifted after nitric acid vapor treatment, which leads to the stronger oxidation of hole potential. Furthermore, the results of the trap experiment demonstrated that photodegradation is dominated by h + . The results of this work provide a novel method for the self-modification of bulk g-C 3 N 4 , which can enhance the photoactivity of g-C 3 N 4 for sustainable environmental restoration.