Quenching-Induced Structural Distortion of Graphitic Carbon Nitride Nanostructures: Enhanced Photocatalytic Activity and Electrochemical Hydrogen Production.

Engineered nanomaterials are emerging in the field of environmental chemistry. This study involves the analysis of the structural, electronic, crystallinity, and morphological changes in graphitic carbon nitride (g-C3N4), an engineered nanomaterial, under rapid cooling conditions. X-ray diffraction, scanning electron microscopy, high-resolution transmission electron microscopy, Brunauer-Emmett-Teller, Fourier transform infrared, Raman, band gap, and Mott-Schottky analyses strongly proved that the liquid N2-quenched sample of g-C3N4 has structural distortion. The photocatalytic efficiency of engineered g-C3N4 nanostructures was analyzed through the degradation of reactive red 120 (RR120), methylene blue (MB), rhodamine B, and bromophenol as a representative dye. The photocatalytic dye degradation efficiency was analyzed by UV-vis spectroscopy and total organic carbon (TOC) analysis. The photocatalytic efficiency of g-C3N4 under different quenching conditions included quenching at room temperature in ice and liquid N2. The degradation efficiencies are found to be 4.2, 14.7, and 82.33% for room-temperature, ice, and liquid N2 conditions, respectively. The pseudo-first-order reaction rate of N2-quenched g-C3N4 is 9 times greater than the ice-quenched g-C3N4. Further, the TOC analysis showed that 55% (MB) and 59% (RR120) of photocatalytic mineralization were achieved within a time duration of 120 min by the liquid N2-quenched g-C3N4 nanostructure. In addition, the quenched g-C3N4 electrocatalytic behavior was examined via the hydrogen (H2) evolution reaction in acidic medium. The liquid N2-quenched g-C3N4 catalyst showed a lower overpotential with high H2 evolution when compared with the other two g-C3N4-quenched samples. The results obtained provide an insight and extend the scope for the application of engineered g-C3N4 nanostructures in the degradation of organic pollutants as well as for H2 evolution.

Introduction

Rapid industrial development and demand for cheap energy leads to various environmental hazards as it affects the living organism because of their cryogenic effect.[1−4] Fossil fuels are the main resources and play a crucial role to supply world energy demand; however, they will produce environmental problems mainly in water bodies.[5−7] Material research studies have devoted their efforts to tackle the problems of these hazardous materials.[8,9] In materials science, the discovery of graphene ushered in the study of two-dimensional (2D) materials[10,11] because these materials offer high surface to volume ratio, high carrier mobility, and enhanced light absorption.[12,13] Unique properties of low-dimensional materials are attracting, which trigged the development and discovery of other low-dimensional materials such as graphene oxide, graphdiyne, graphitic carbon nitride (g-C3N4), metal chalcogenide, and Mxene.[14−17] Among the 2D materials, g-C3N4 is a metal-free polymer semiconductor, which is constituted of carbon and nitrogen.[18,19] g-C3N4 has varied photocatalytic properties in the visible region and received tremendous interest among the researchers.[20,21] Previous studies have focused on the synthesis of highly efficient photocatalysts through the modification of electronic structures and morphology engineering.[22] Attempts have been made to improve the photocatalytic efficiency of g-C3N4 by making hybrids.[23,24] These compounds and their hybrids have potential industrial applications such as H2 production, CO2 reduction, supercapacitor and sensor fabrication, and organic pollutant degradation.[25−30]

The photocatalytic properties of g-C3N4 predominantly suffered from poor surface area, low electrical conductivity, and rapid charge carrier recombination. These advantages can be overcome by turning its nanostructure.[31,32] For instance, engineered TiO2 displays superior photocatalytic activity, compared to its conventional form because of the minimization of charge carrier.[33,34] The studies on TiO2 reveal that changes made in morphology and electronic structure enhance the photocatalytic efficiency. Earlier studies have used various methods to tune the nanostructure of g-C3N4, such as alkaline modification, acid treatment, metal doping (N, P, F), treatment of inorganic semiconductors/g-C3N4 heterojunctions, and formation of carbon/g-C3N4 hybrids.[35−39] Similarly, various approaches have been applied for the creation of other nanostructures. Doping and the hybrid formation method have some serious drawbacks such as dopant loss and the use of strong acid. In addition, in self-doped TiO2, the obtained efficiency is not enough to degrade the hazardous pollutant.[40] Therefore, we need to explore new methodologies to improve the photocatalytic efficiency of the g-C3N4 nanostructure.

In this study, for the first time, we demonstrate the enhancement of photocatalytic properties of engineered g-C3N4 using simple quenching techniques with various quenchers. The methodology involves the synthesis of g-C3N4 using pyrolysis at 450 °C, followed by quenching under varied conditions such as room temperature and ice and liquid nitrogen cooling. The photocatalytic degradation properties of the obtained catalyst were analyzed, using the degradation of organic dyes such as reactive red 120 (RR120), methylene blue (MB), bromophenol B (BP), and rhodamine B (RhB). Dye discoloration and its degradation were analyzed by the UV–vis spectroscopy and total organic carbon (TOC) analysis measurements.

Experiments

Materials

For g-C3N4 preparation, urea was chosen as a raw material, which was purchased from Sigma-Aldrich, India. Organic pollutants including anionic, cationic, and phenol compounds, namely, reactive red 120, MB, RhB, and BP, were procured from Alfa Aesar, India. The radical scavenger benzoquinone, ethanol, and electrolyte were obtained from Sigma-Aldrich, India. All reactions were carried out with distilled water.

g-C3N4 Preparation Methods

The preparation of engineered g-C3N4 involves simple pyrolysis. In this study, 2 g of urea was ground and placed into a furnace heated to 450 °C with a heating rate of 2 °C/min for a time duration of 2 h. During pyrolysis, the crucible was sealed to avoid the compound loss. After pyrolysis, the crucible was cooled to room temperature by various quenching processes such as room-temperature quenching, ice quenching, and liquid nitrogen quenching. In the quenching process, the direct contact of the quencher and g-C3N4 was avoided. The room-temperature and ice- and liquid nitrogen-quenched samples were labeled as G1, G2, and G3, respectively.

Physical Characterization

The physical properties such as crystalline morphologies, surface areas, band gaps, structural defects, and presence of functional groups were studied by X-ray diffraction (XRD), scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HR-TEM), Brunauer–Emmett–Teller (BET), ultraviolet diffuse reflectance spectroscopy (UV–DRS), Fourier transform infrared (FTIR), and Raman spectroscopy. The instruments employed for the analysis were XRD (XRD Lab, X-Shimadzu), SEM (JEOL, JSE-6390 SEM), high-resolution SEM (JEOL JEM 2100 HR-TEM), the BET surface area (JASCO V-660), FTIR spectroscopy (Thermo Scientific Nicolet 6700 FTIR), UV–vis (JASCO, V-40), DRS UV (Model: Shimadzu model), and Raman spectroscopy (HORIBA LabRam HR Evolution). Mott–Schottky analysis was carried out using a CHI 660D electrochemical workstation.

Photocatalyst Study

The photocatalytic properties of the engineered g-C3N4 for the degradation of RR120, MB, RhB, and BP were studied under visible light (150 W Xe lamp, >420 nm) in the home-made reactor. Initially, 1 × 10–5 M probe molecule and 5 mg of the catalyst were thoroughly mixed using a magnetic stirrer. The pollutant and engineered g-C3N4 catalyst were stirred for 30 min to reach the adsorption equilibrium. At periodic time intervals, 3 mL of the probe solution was taken out to measure the absorption. The photocatalytic degradation of RR120, MB, and RhB was calibrated at 520, 664, and 550 nm absorbance maxima, respectively. The initial and time-dependent dye concentration are denoted as C0 and C, where t refers to the reaction time and “k’” to the pseudo-first-order rate constant. The expression used for calculating the dye efficiency is given below.[41]

The dye degradation reaction obeys the pseudo-first-order reaction. The first-order reaction rate constant (k) was calculated by the expression 2

Results and Discussion

Physical Characterization

The crystalline and structural changes of the G1, G2, and G3 samples were studied using XRD and are shown in Figure a. In the figure, G1 has a d-spacing of 0.319 nm, which is consistent with the values of the previously reported g-C3N4 nanostructures.[42,43] The d-spacing for G2 and G3 was about 0.32 nm, which is slightly higher than the naturally cooled g-C3N4. Furthermore, the diffraction angle was shifted to a lower angle region from 27.9°, 27.6°, and 27.1° for samples G1, G2, and G3, respectively. The peak shift maybe associated with the lattice dislocation or a defect, which may be induced during the quenching. Further, the grain sizes of G1, G2, and G3 were found to be 3.1, 7.15, and 11.75 nm, respectively. Generally, g-C3N4 showed two distinguishable XRD peaks, which can be assigned to the (001) and (002) crystal lattice.[44] However, the disappearance of the structural peak (001) observed in G2 and G3 at a lower angle confirms the structural alteration in the quenched samples.[45] The XRD analysis proves the introduction of defects in the samples G2 and G3. SEM analysis was carried out to examine the morphology at the nanoscopic level. All the engineered g-C3N4 (G1, G2, and G3) samples exhibited a flakelike structure with different sizes. To better understand the morphology of g-C3N4, HR-TEM analysis was carried out. Figure shows the HR-TEM images of the various quenched g-C3N4 samples. Figure a–d represents the HR-TEM images of G1 where the smooth and uniform sheetlike structure can be seen at lower and higher magnification. In the case of G2, similar sheetlike structures are clearly seen with perfect edges (Figure e–h). On the other hand, the HR-TEM images of G3 represented in Figure i–l show a sheetlike structure with several pores uniformly distributed on the surface. In the high magnification, the image depicts that the pores are in the size of a few nanometers. The creation of pores on the 2D structure is due to the quenching process. Such types of pores on the surface have been previously reported on graphene sheets because of KOH activation and metal doping.[46,47] Morphology changes in G3 samples can be explained using the principles of thermodynamics. The temperature of liquid N2 is −196 °C, which is much lower than the room temperature and the temperature of ice. Once a hot g-C3N4 is surrounded by liquid N2, it will cause rapid cooling. This uneven thermal profile along the g-C3N4 surface cools faster than ice and room-temperature conditions. This ultrafast cooling induces the pores on the surface, leading to a different morphology as observed in G3. To further understand the surface area and pore nature of the g-C3N4-quenched samples, the BET surface area measurement was carried out. Figure depicts the N2 adsorption–desorption isotherms of the quenched g-C3N4 samples. All the g-C3N4 samples exhibit type IV adsorption–desorption isotherm (Figure a–c), which revealed that mesoporous structures could be present in g-C3N4.[48]Figure d–f represents the pore size distribution curve of G1, G2, and G3, respectively. The obtained average pore diameters were 5.4, 5.2, and 5.2 nm for samples G1, G2, and G3, respectively. Naturally quenched g-C3N4 has a low surface area of 12.67 m2/g and a low pore volume of 0.017 cm3/g. For G2, the surface area was 99.96 m2/g and the pore volume was 0.13 cm3/g. The G3 sample has the highest surface area of 807.12 m2/g and the largest pore volume of 1.06 cm3/g. X-ray photoelectron spectroscopy (XPS) spectra of G1, G2, and G3 samples were measured to understand the chemical composition. Figure a represents the full survey spectra of G1, G2, and G3, in which two prominent peaks were observed at 286.9 and 397.8 eV corresponding to C 1s and N 1s, respectively. The high-resolution C 1s spectra consist of two peaks including 283.6 and 287.2 eV, which are shown in Figure b.[49,50] Further, the peak binding energy was slightly changed for the samples G2 and G3. Likewise, the N 1s spectra peak position and peak integral area were changed in G2 and G3 samples, respectively. The deconvoluted N 1s spectra are shown in Figure d–f. All spectra consist of three peaks located at 397.6, 398.3, and 399.8 eV assigned to C–N–C, N–(C)3, and N–H bonds, respectively.[51] It is important to note that we calculated the peak area ratio between two coordinated N2C atoms and three coordinated N3C atoms. The peak ratio (N2C/N3C) was decreased in G3 when compared with that in G1 and G2, and also the binding energy was shifted. This observation indicates that actual N atoms are vacant in quenched g-C3N4. On the other hand, the N vacancy is mainly ascribed to the N2C lattice site because of the unsaturated coordination and also higher population. These results were well consistent with previously reported results.[52] The overall XPS spectra emphasize that the quenched g-C3N4 has structural defects compared with pure g-C3N4.

Figure 1

(a) XRD pattern and (b–d) SEM images of G1, G2, and G3.

Figure 2

(a) HR-TEM images of (a–d) G1, (e–h) G2, and (i–l) G3.

Figure 3

Adsorption–desorption isotherms of (a) G1, (b) G2, and (c) G3 and the corresponding pore size distribution curves of (d) G1, (e) G2, and (f) G3.

Figure 4

(a) Full survey spectra of G1, G2, and G3. (b) C 1s XPS spectra of G1, G2, and G3. (c) N 1s XPS spectra of G1, G2, and G3. Deconvolution spectra of N 1s (d) G1, (e) G2, and (f) G3.

Figure a represents the UV–DRS spectra of engineered g-C3N4 samples, and the variation in band gap across the samples is observed. From the figure, no obvious regular change can be found, that is, the band gap has decreased for G2 and then increased for G3 when compared to that of G1, which is mainly due to the change of the electronic structure in the quenched samples.[53] The FTIR spectra of G1, G2, and G3 samples were also studied and are shown in Figure b. The observed vibration band of the engineered samples (N–H: 3186, C=N: 1634, and C–N: 1326 cm–1) agrees with the functional groups of g-C3N4. The absorption band at 810 cm–1 can be assigned to the tri-s-triazine-based structure.[54,55] Besides, C=N and C–N appeared at 1260–1480 and 1572–1633 cm–1, respectively. These C=N and C–N bands have a slight difference in the G3 sample. On the other hand, the band intensity also decreased, which confirms the defect developed in G3. Raman spectroscopy is an important tool to analyze the 2D material structures. Raman spectra of pure and N2-quenched g-C3N4 are shown in Figure c. It is observed that G1 and G3 samples exhibit different Raman patterns, which suggests that the g-C3N4 crystallinity was affected in the N2-quenching process. Broad peaks were observed at 1200–1700 cm–1, which are due to the C–N asymmetric structural vibrations. This peak is similar to the G and D peaks observed for the carbon-based material. Apart from the broad peaks, a few sharp peaks were observed at a lower angle, which is due to the different types of s-triazine.[56] Further, the structural disorder developed in g-C3N4 can be effectively analyzed by the intensity ratio between the D and G bands (ID/IG). The calculated relative ID/IG intensities were 0.86 and 1.01 for G1 and G3, respectively. The change in relative intensity suggested that the N2-cooled g-C3N4 has crystal distraction in its structure.[57]

Figure 5

(a) UV–DRS, (b) FTIR spectroscopy, and (c) Raman spectra of G1, G2, and G3.

To investigate the band potential analysis of G1, G2, and G3, Mott–Schottky analysis was carried out. The Mott–Schottky measurement was conducted in 0.5 M Na2SO4 aqueous solution at pH 7 under dark condition. The flat band potentials of prepared samples were determined using the Mott–Schottky expression at 298 K. The estimated flat band potential was found from the average value of the x-intercepts of the linear portion of the Mott–Schottky plot. The measured potentials versus Ag/AgCl were converted to the reversible hydrogen electrode (RHE) scale via the Nernst equationHere, VRHE is the converted potential versus RHE, VAg/AgCl is the experimental potential measured against the Ag/AgCl reference electrode, and the value of VAg/AgClvsNHE0 is 0.209 V at 25 °C. Figure a represents the flat band potential of the engineered g-C3N4 samples. The observed flat band potentials of G1, G2, and G3 are −0.628, −0.778, and −0.998 eV, respectively. Using the Mott–Schottky analysis, the band potential of the various engineered g-C3N4 was examined.[58]Figure b depicts the valence band (VB) and conduction band (CB) position of the engineered g-C3N4 sample. The observed upshift in CB position in G1, G2, and G3 indicates the reduction capability of the photoexcited electron, whose activity can be enhanced under visible light. The overall physical characterization mentioned above indicates that the liquid N2-quenched samples have defects and a disordered structure. This structurally modified g-C3N4 sample shows an increase in photocatalytic efficiency that was studied using various pollutants.

Figure 6

(a) Mott–Schottky plot and (b) band positions of G1, G2, and G3.

Photocatalytic Activity and Analysis

The photocatalytic degradation efficiencies of engineered G1, G2, and G3 samples are monitored by the dye degradation method. Figure a–c shows the UV absorbance spectra of RR120 measured at different time intervals of the photocatalytic reaction for the samples G1, G2, and G3, respectively. The corresponding residual concentration against the time plot is shown in Figure d. The dye degradation efficiency was calculated using the empirical eq . After 120 min of irradiation, the photocatalytic degradation efficiency was found to be 28, 49, and 82% for G1, G2, and G3, respectively. Obviously, the efficiency of G3 is much higher than that of G1 and G2. To gain understanding regarding the difference in the reaction rate between the catalysts, the rate constant analysis was performed using expression 1. The calculated quasi-first-order reaction rate constant of G3 was nearly 9-fold greater than G2. This experimental observation convincingly proved that the sample G3 as a photocatalyst outperformed samples G1 and G2. In addition to validate the photocatalytic efficiency of G3, it was tested against other dye molecules. Figure a–c shows the photocatalytic degradation of MB, RhB, and BP in the presence of G3. The photocatalytic degradation efficiencies of MB, RhB, and BP are 95, 76, and 92%, respectively. From this observation, it can be said that the sample G3 acts as a good photocatalyst for degrading anionic–cationic dyes and phenol derivatives.[59] In order to understand the photocatalytic mechanism of G3, radical scavenging analysis was performed.[60]Figure a depicts the C/C0 against the time plot of G3 samples after the addition of EtOH and benzoquinone radicals as a scavenger. From the figure, it is clear that benzoquinone is more effective than alpha hydrogen alcohol in inhibiting the reaction. On the basis of the findings, it can be deduced that the superoxide radical plays a dynamic role in the photocatalytic degradation of the organic pollutants.[61] On the other hand, EtOH also suppressed the photocatalytic reaction; however, the effect was less. From this study, it can be concluded that superoxide and hydroxyl radicals play an important role in the degradation process. The enhanced degradation can be explained by the electronic structure of g-C3N4.

Figure 7

UV–vis absorbance spectra of photocatalytic reduction of RR120 aqueous solution under visible-light irradiation for (a) G1, (b) G2, and (c) G3. (d) Degradation rate curves of RR120 and MB by G1, G2, and G3 under visible-light irradiation (RR120 = 1 × 10–5 M, g-C3N4 = 5 mg L–1, and reaction time = 120 min).

Figure 8

Photocatalytic degradation of (a) MB, (b) RhB, and (c) BP by G3. (d) Fluorescence spectral changes with irradiation time on the G3 sample (MB = 1 × 10–5 M, RhB = 1 × 10–5 M, BP = 1 × 10–5 M, g-C3N4 = 5 mg L–1, and reaction time = 120 min).

Figure 9

(a) Effects of various scavengers on the photocatalytic degradation of RR120 in the presence of G3. (b) Recycling studies of the RR120 degradation in the presence of G3. TOC removal of (c) MB and (d) RR120 by G3 (RR120 = 1 × 10–5 M, g-C3N4 = 5 mg L–1, and reaction time = 120 min).

As seen in Figure b, the CB potentials are −0.628, −0.728, and −0.998 eV for G1, G2, and G3, respectively, and the CB potential of NHE is 1.58 eV. The CB potential is more negative than O2/O2• (−0.33 eV vs NHE) standard redox potential. Also, the band gap change in the G3 sample is the reason for the concentration of more electrons in the CB. These electrons play a critical role in the formation of the superoxide radical. When g-C3N4 is irradiated by light, the charge carriers are formed at the VB and CB edges. The charge carriers migrate at the surface and react with oxygen and OH to produce a free radical. Further, these radicals oxidize and reduce the target molecule which is adsorbed on the surface.[62] It is widely accepted that the large surface area and pores on the surface serve as a good adsorption site for organic pollutants. From BET results, it is observed that the G3 sample has a large surface area as well as a high porosity.[63,64] Plenty of molecules are adsorbed by G3, resulting in high degradation efficiency. Further, we have performed the catalytic durability test for the sample G3. As seen in Figure b, there was no degradation efficiency loss observed even after the fourth cycle. These results suggest that the catalyst has high stability.[65] In addition, TOC analysis was carried out to understand the organic carbon removal efficiency of the photocatalyst from the dye.[66]Figure c,d shows the time-dependent TOC amount for RR120 and MB dyes during photocatalysis. From the figure, it is observed that during the photocatalysis of RR120 and MB, a mineralization of 55 and 59% is achieved, respectively. The TOC results clearly indicate that the catalyst strongly mineralizes the organic dye (RR120 and MB) during photocatalysis. The overall results highlight that the rapid cooling process is one of the cost-effective ways to modify the morphology and the electronic structure of the catalyst, leading to the enhancement of the photocatalytic properties.

Electrochemical Hydrogen Evolution

Electrochemical water splitting is a promising pathway to produce renewable and green H2. The presently available electrocatalytic material such as Pt is costly and strongly hinders the scalable application.[67] In this work, the HER activity of G1, G2, and G3 was studied in 0.5 M H2SO4. The polarization curve is shown in Figure , and the sample G3 provides enhanced hydrogen production with a lower onset potential of −0.350 mV. On the other hand, G1 and G2 showed the onset potentials of −0.469 and −0.469 mV, respectively. The polarization curves of the current density for the corresponding materials are found to be −48.5, −75.7, and −305.02 μA/cm2 for G1, G2, and G3, respectively. In addition, we also measured the electrochemical impedance spectroscopy (EIS) of the prepared g-C3N4 sample. Figure b depicts the Nyquist plot of the EIS for G1, G2, and G3. The sample G3 showed lower charge resistance compared to the other samples, and hence, G3 shows higher hydrogen production.

Figure 10

(a) Hydrogen evolution reaction activity and (b) impedance analysis of the G1, G2, and G3 samples.

Conclusions

The photocatalytic properties of the engineered g-C3N4 prepared by quenching at different temperatures are better than the samples prepared by pyrolysis alone. The unique morphology of quenched g-C3N4 enhances absorbance, band gap, and flat band potential and reduces the particle size. The pseudo-first-order reaction rate of the nitrogen-quenched g-C3N4 is 9 times greater than the ice-quenched g-C3N4. Superoxide radicals play a significant role in the dye degradation, which was justified by the scavenger studies. The enhanced degradation of RR120 and MB is due to the increase of band potential of the CB. This increase may be due to the migration of the photoinduced electron. The results obtained in this study provide insights into the application of g-C3N4 as an effective catalyst for the degradation of organic pollutants. There is scope for further research in the understanding and consolidation of the catalytic properties of g-C3N4.