Facing the exhaustion of fossil fuel resources and serious environmental pollution due to industrial and domestic wastes, scientists are actively looking for efficient photocatalysts which can convert solar energy into chemical energy and be used in the photodegradation of environmental pollutants, hydrogen production via water splitting, air treatment, and CO₂ conversion to hydrocarbon fuel.^[1–5] In the last few years, graphitic carbon nitride (g-C₃N₄), as a new material, has come into being in this area, and it is a non-metallic semiconductor with a band gap energy of about 2.7 eV (λ ∼ 460 nm), durable, non-toxic, highly recyclable and exhibits superior photocatalytic ability.^[2,6–10] Therefore, g-C₃N₄ has become an attractive research object in the field of photocatalyst.

To obtain g-C₃N₄, it is necessary to use carbon and nitrogen-rich precursors such as urea,^[11,12] thiourea,^[11] melamine,^[13,14] cyanamide,^[15] dicyandiamide,^[16] etc. Common methods for preparing g-C₃N₄ include high-pressure pyrolysis,^[17] thermal polymerization,^[14] hydrothermal,^[18,19] etc. One of the C and N-rich precursors, not expensive and easy to find, is urea. Our previous study showed that with a heating time of 2 h, g-C₃N₄ nanosheets prepared by thermal polymerization in Ar atmosphere at 550 °C exhibited the highest photocatalytic efficiency.^[20] In the present work, the heating temperature is fixed at 550 °C while the heating time is changed from 0.5 to 2.5 h to study the effect of the heating time on the formation of crystal structure, optical property, and photocatalytic performance of g-C₃N₄ nanosheets.

The XRD patterns of the g-C₃N₄ samples prepared at the heating temperature of 550 °C for different time are shown in Fig. 1(a). The samples all exhibit three diffraction peaks located at 12.7°,24.8°, and 27.5° which are corresponding to the (001), (101), and (002) diffraction planes, respectively. The peak intensity increases as the heating time increases. A comparison between these peaks and the (001), (101), and (002) peaks of standard g-C₃N₄ materials demonstrates that the synthesized g-C₃N₄ is crystallized in hexagonal phase (JCPDS file No. 87-1526). The strongest peak at 27.5° corresponds to the interlayer stacking of the conjugated aromatic system^[21] with an interlayer distance of 3.42 Å. It is obvious that as the heating times increase, the (002) peak shifts slightly towards the larger diffraction angle, indicating the narrowing of the interlayer distance and the decrease of crystal parameter c. The calculations show that parameter c decreases from 6.85 Å for CN-0.5 sample to 6.77 Å for CN-2.5 sample (Table 1). The lattice parameter a can be calculated via peak (101) located at 24.8°. The calculations show that parameter a decreases from 4.59 Å to 4.55 Å when the heating time increases from 0.5 h to 2.5 h (Table 1). Parameters a and c varying with the heating time are plotted in Fig. 1(b), which shows that th eg-C₃N₄ crystal is packed more tightly as the processing time increases, similar to what happens as the heating temperature increases.^[20] In addition, figure 1(a) shows that the diffraction intensity increases with the peak width narrowing when the heating time increases (Table 1), indicating that the polymerization of g-C₃N₄ takes place more strongly and crystallinity of g-C₃N₄ is higher as the heating time increases. The XRD peak position and peak width change noticeably as the heating time increases from 1.0 h to 1.5 h, which shows that a fairly ordered lattice can be achieved after heating urea at 550 °C for 1.5 h.

Fig. 1.

	Figure Option ViewDownloadNew Window
	Fig. 1. (a) XRD patterns of g-C3N4 samples heated at 550 °C for different heating time and (b) lattice parameters varying with heating time.

Table 1.

Table 1.

Table 1. Peak positions, heights, and full widths at half maximum (FWHM) of (002) XRD peaks, and lattice parameters a and c of as-prepared samples. . Samples Peak position/(°) Height FWHM/(°) a/Å c/Å CN-0.5 27.20 212.06 2.28 4.586 6.846 CN-1.0 27.17 556.71 2.76 4.573 6.842 CN-1.5 27.49 513.75 1.26 4.568 6.805 CN-2.0 27.55 782.80 1.02 4.553 6.779 CN-2.5 27.52 1215.16 1.22 4.545 6.768

Table 1. Peak positions, heights, and full widths at half maximum (FWHM) of (002) XRD peaks, and lattice parameters a and c of as-prepared samples. .

Figure 2 shows the infrared absorption spectra of the as-synthesized g-C₃N₄ samples with different heating time. According to the previous studies, the breathing modes of CN heterocycles of triazine units are responsible for the absorption peak at 810 nm while the N–H group substituted into aromatic primary amine usually shows the stretching absorption in the region of 3000–3600 nm. The stretching mode of the C = N bonds in aromatic rings is responsible for the absorption peak at 1640 cm⁻¹ while the stretching modes of the C–N bonds cause some absorption bands in the range of 1200–1500 cm⁻¹. The FTIR spectra of the g-C₃N₄ samples change slightly as the preparation time increases as indicated by the leftward shifting of the peaks 1242 cm⁻¹ and 1413 cm⁻¹ and the rightward shifting of the peaks 1321 cm⁻¹ and 1457 cm⁻¹ in Fig. 2(b). In the absorption region of 3000–3600 cm⁻¹, some stretching frequencies are observed at 3084 cm⁻¹, 3162 cm⁻¹, 3269 cm⁻¹, and 3413 cm⁻¹, among which the 3413 cm⁻¹ and 3162 cm⁻¹ bands can be attributed to asymmetric stretching and symmetric stretching of the NH₂ group while the 3084 cm⁻¹ and 3269 cm⁻¹ bands are due to the O–H stretching vibrations.^[22] It is obviously seen that the 3413 cm⁻¹ band shifts gradually toward higher frequency with heating time increasing, consistent with the XRD analysis in which the crystal becomes more ordered and the crystal parameters decrease. In addition, the intensity ratio of 3413 cm⁻¹ to 3162 cm⁻¹ first increases, and then decreases. This also strengthens the change in crystal structure of g-C₃N₄ as heating time increases.

Fig. 2.

	Figure Option ViewDownloadNew Window
	Fig. 2. (a) FTIR spectra of g-C3N4 samples prepared at 550 °C with different preparation time and (b) FTIR peak positions varying with wavelength.

The SEM images of g-C₃N₄ synthesized with different heating time are shown in Fig. 3. The g-C₃N₄ sample obtained at 550 °C for 0.5 h in Fig. 3(a) shows heterogeneous particle morphology. Some particles are granular while the others are in the form of nanosheets with a thickness of around 10 nm. When the heating time increases to 1.0 h, the granular particles no longer exist, leaving nanosheets with curved edges. For 1.0 h, the sheets are quite large with very few torn holes on the surface. However, when the heating time increases to 2.0 h and 2.5 h, the samples are still in the form of nanosheets, however, the surfaces of the sheets are punctured or torn into smaller sheets. The g-C₃N₄ obtained at 550 °C for 2.5 h consists of nanosheets that are small in size and more porous than the other samples, which is beneficial for photocatalytic performance due to its large specific surface area.

Fig. 3.

	Figure Option ViewDownloadNew Window
	Fig. 3. FE-SEM images of g-C3N4 nanosheets synthesized at 550 °C with different preparation time: (a) 0.5 h, (b) 1.0 h, (c) 2.0 h, and (d) 2.5 h.

The BET technique is used to measure the specific surface area and pore size distributions of the as-prepared g-C₃N₄ samples with heating time of 1.0 h, 2.0 h, and 2.5 h as shown in Fig. 4. All samples show type IV isotherms with H3 hysteresis loops, indicating a mesoporous structure that forms slit-shaped pores with a non-uniform size and/or shape. The calculated surface area, average pore volume, and average pore size are listed in Table 2. The surface area of the samples increases from 60 m²/g to 71 m²/g and 88 m²/g for CN-1.0, CN-2.0, and CN-2.5, respectively, which are larger than the specific surface area of pure g-C₃N₄ reported previously.^[23,24] This difference is due to the nanosheets being punctured and torn as the heating time increases as seen in the SEM images. All samples exhibit a wide pore size distribution (Fig. 4(b)) and a large average pore width in a range of 10–30 nm.

Fig. 4.

	Figure Option ViewDownloadNew Window
	Fig. 4. (a) N2 adsorption–desorption isotherms of g-C3N4 heated for 1.0 h, 2.0 h, and 2.5 h and (b) plots of BJH pore volume versus pore diameter.

Table 2.

Table 2.

Table 2. Surface areas, pore volumes, and pore sizes of the as-prepared samples. . Samples Surface area/m2⋅g−1 Pore volume/cm3⋅g−1 Pore diameter/Å CN-1.0 60.0004 0.218945 145.9624 CN-2.0 70.8108 0.426483 240.9142 CN-2.5 87.8561 0.218932 99.6773

Table 2. Surface areas, pore volumes, and pore sizes of the as-prepared samples. .

UV–vis diffuse reflectance spectra (DRS) of g-C₃N₄ obtained with different preparation time in Fig. 5(a) indicate an absorption edge at around 450 nm for all samples. It is obvious that the absorption edge first shifts slightly to the right as the heating time increases from 0.5 h to 1.0 h and then inversely shifts to the left with the increasing preparation time. For indirect band gap semiconductor such as g-C₃N₄, ΔE_g can be determined by using Tauc plot of (α hν)^1/2 versus photon energy (hν) as shown in Fig. 5(b). The band gap is 2.73 eV and 2.66 eV for CN-1.0 and CN-2.0, respectively, revealing that the visible light can be used for exciting photocatalysis.

Fig. 5.

	Figure Option ViewDownloadNew Window
	Fig. 5. (a) Absorption spectra of g-C3N4 nanosheets obtained at 550 °C with different preparation time and (b) Wood–Tauc method determining band gap energy of indirect semiconductor.

Figure 6 shows the photoluminescence spectra PL (with excitation wavelength of 350 nm) of the as-synthesized g-C₃N₄ samples with different processing time, indicating a broad blue–green emission band for all samples. The PL band concentrates at 450 nm, in accordance with the UV–vis absorption edge. This PL band can be decomposed into four component peaks as shown in the inset of Fig. 6(a) (for CN-1.5 sample), including P1 (430.4 nm, 2.88 eV), P2 (453.6 nm, 2.73 eV), P3 (491.6 nm, 2.52 eV), and P4 (545.8 nm, 2.27 eV) which accord well with previous reports.^[20,25] The PL peaks shift slightly towards the longer wavelength as the processing time increases (see Fig. 6(b)), consistent with the decrease of the energy bandgap obtained in the absorption spectra. In addition, the PL intensity first gradually increases to the highest value at CN-2.0 and then decreases.

Fig. 6.

	Figure Option ViewDownloadNew Window
	Fig. 6. (a) Photoluminescence spectra of g-C3N4 nanosheets obtained at 550 °C with different preparation time, with inset showing the normalized PL spectra and dot lines being Gaussian fitting for CN-2.0 sample, and (b) PL peak positions varying with heating time.

The photocatalytic activity of the g-C₃N₄ sample is assessed through the degradation of Rhodamine B (RhB) solution under the irradiation of xenon lamp in which the UV–vis absorption technique was used to indirectly measure the concentration of remaining RhB. Figure 7 shows the C/C₀ ratio as a function of time, indicating that the RhB degradation highly depends on the heating time. Samples CN-0.5 and CN-1.0 exhibit low photocatalytic efficiency, only about 60% of RhB is degraded after 2 h of exposing. The photocatalytic activity increases sharply as the preparing time increases to 2 h. The CN-2.0 and CN-2.5 degrade almost all RhB after 2 h of exposing. This improvement of photocatalytic performance can be explained by the mesoporous structure of the as-prepared g-C₃N₄ with long heating time. Although CN-2.0 has lower specific surface area than CN-2.5, its pore volume and pore size are both larger, so their photocatalytic performances are almost the same. In addition, the narrower energy band gap of the sample with long processing time also contributes to the improvement of the photocatalytic activity.

Fig. 7.

	Figure Option ViewDownloadNew Window
	Fig. 7. Adsorption and photocatalytic activities of the g-C3N4 prepared at 550 °C with different preparation time in decomposition of RhB solution under xenon lamp irradiation.

The structural and physical properties of g-C₃N₄ nanosheets are greatly influenced by the heating time. The crystal lattice is formed more orderly and packed more densely as the heating time increases. The increase in fabrication time also causes the g-C₃N₄ nanosheets to be slightly torn, making them more porous and the specific surface area increase. The energy band gap changes negligibly as the processing time increases. However, the photoluminescence peaks shift slightly toward longer wavelength. The changes in structure, morphology, and physical properties determine the change in the photocatalytic performance of the g-C₃N₄ nanosheets. The g-C₃N₄ nanosheets obtained in long heating time, greater than 2.0 h, possess high photocatalytic activities: the RhB decomposed after 2.0 h of xenon lamp irradiating is accounted for 95%.