Integrated Nanostructures of TiO₂/g-C₃N₄/Diatomite Based on Low-Grade Diatomite as Efficient Catalyst for Photocatalytic Degradation of Methylene Blue: Performance and Mechanism

Abstract

The comprehensive utilization of low-grade diatomite resources and the effective treatment of printing and dyeing wastewater have attracted widespread attention. The combined scrubbing-magnetic separation-acid leaching-roasting process was used to increase the SiO₂ content from 59.22% to 86.93%, reduce the Al₂O₃ content from 18.32% to 6.75%, and reduce the Fe₂O₃ content from 6.85% to 1.24% in the low-grade diatomite from Heilongjiang, China. The TiO₂/g-C₃N₄/diatomite nanocomposite was prepared by a facile ultrasonic-thermal polymerization method. In this ternary structure, diatomite skeleton effectively increased the surface area with abundant adsorption sites, prevented g-C₃N₄ from restacking, and facilitated the separation of electrons and holes via the formation of TiO₂/g-C₃N₄ heterojunctions. The degradation rate was 98.77%, 90.59%, and 89.16% for the three catalytic reaction cycles of the MB solution, respectively. The composite showed a high degradation rate of the MB solution after three cycles, which indicated that the composite had good recyclability. Through the free radical capture test, it was elucidated that O₂⁻·, h⁺, and ·OH all played a role in the photocatalytic reaction of the TiO₂/g-C₃N₄/diatomite to the MB solution, in which O₂⁻· was mainly responsible for the photocatalytic oxidation mechanism, and the reaction kinetics were further investigated. This nanostructured TiO₂/g-C₃N₄/diatomite composite has fascinating visible light catalytic activity and excellent stability.

1. Introduction

Diatomite is widely used in construction [1], petroleum [2], the light industry [3,4,5], the chemical industry [6], and other fields [7,8,9], with its unique physical and chemical properties, such as multi-level pore structure [10] and nano characteristics [11], making high-grade diatomite resources increasingly scarce [12]. There has been less development and utilization of low-grade diatomite, resulting in a serious waste of resources and great pressure on the environment. The identified reserves of diatomite resources in China rank second in the world and are distributed in North, Northeast, Southwest, South Central and East, etc. Only the Linjiang-Changbai area is the deposit of high-quality diatomite resources, accounting for more than 90% of the high-quality resources in the country. Meanwhile, the annual consumption of diatomite in the world is more than 2 million tons, of which the consumption share in the United States is more than 50%, and the consumption share in China is only about 15%, indicating that the product processing of diatomite in China is low and there is still a large gap with foreign countries [13,14,15,16,17]. In addition, China’s diatomite processing technology and technical equipment is relatively backward; its processing product grade is low, and it cannot reach the performance indicators of similar foreign products. This requires us to make comprehensive use of diatomite resources while vigorously developing the purification technology of low-grade diatomite to meet the requirements of modern high-tech and new material industries, thus improving the market competitiveness of domestic diatomite products and optimizing the product structure.

Water pollution and health hazards caused by printing and dyeing wastewater have drawn considerable research interest and societal responsiveness [18,19]. Among them, methylene blue (MB), the most common organic dye, is a typical water-soluble cationic organic dye with good stability and non-degradable properties [20]. With the accumulation of MB in the ecosystem, its high toxicity can lead to genetic mutations, causing biological carcinogenesis and other health problems [21,22]. However, existing water treatment technologies for pollutant concentration control are hardly effective in limiting the degradation of environmental quality [23]. As the main living area of the people and with a large population, the water environment system must necessarily assume the productive life and development of the people. The search for green and efficient methods for the degradation of printing and dyeing wastewater has become the focus of current research [24].

The semiconductor g-C₃N₄ is popular because of its simple preparation method and its good response in the visible light region [25]. g-C₃N₄ usually suffers from low specific surface area, short photocatalytic carrier lifetime due to π-π conjugated electron system, and unsuitable photo-redox potential. The construction of heterojunctions is commonly used to suppress photogenerated electron-hole recombination and thus enhance photocatalytic activity [26,27]. TiO₂ is one of the first semiconductors to start research on the photocatalytic material, which has the advantages of being cheap, stable, and non-toxic; however, the forbidden band of anatase TiO₂ is wider at about 3.2 eV, and the optical response threshold λ_g = 387.5 nm, while the energy corresponding to ultraviolet light with a wavelength less than 380 nm only accounts for 3.39% of the total energy of the entire solar spectrum wavelength range, making its light energy utilization low [28,29]. Furthermore, the photogenerated electron-hole complexation of TiO₂ is easy, and the charge separation efficiency and photocatalytic efficiency need to be further improved [30]. Diatomite has been used to stabilize various semiconductor nanoparticles and form composites with various compositional characteristics due to its unique properties [31]. The attachment of semiconductor nanoparticles to diatomite substrates can enhance its photocatalytic performances for ultraviolet and visible light absorption and removal of organic compounds [32]. Moreover, it reduces the chance of particle aggregation and provides more active sites for photocatalytic reactions, thus improving the degradation efficiency [33].

At present, there are relatively few studies on purification techniques for low-grade diatomite; most studies are for high-grade diatomite. Meanwhile, there are relatively few studies on the preparation of the TiO₂/g-C₃N₄/diatomite ternary composite using low-grade diatomite purified concentrates. In this study, the low-grade diatomite from the diatomite planning area in Heilongjiang, China is used as the raw ore; the combined process of scrubbing-sedimentation-acid leaching-roasting is used for purification treatment; the TiO₂/g-C₃N₄/diatomite composite is prepared by the ultrasonic-thermal polymerization method; and MB is used as the simulated pollutant to investigate the feasibility of the combined process to purify low-grade diatomite and the photocatalytic degradation performance of the nanocomposite on printing and dyeing wastewater.

2. Results and Discussion

2.1. Proposed Purification Process of Low-Grade Diatomite

Based on the results of previous experiments, the suggested flowsheet for the processing of the low-grade diatomite (Figure 1) includes scrubbing, settling classification, magnetic separation, acid leaching, and roasting. The chemical compositions of the low-grade diatomite and diatomite concentrate are given in

Table 1. Chemical analysis of the diatomite.

Oxide	Weight (%)		Oxide	Weight (%)
	Raw Ore	Concentrate		Raw Ore	Concentrate
SiO2	59.22	86.93	CaO	1.75	0.40
Al2O3	18.32	6.82	MgO	0.24	0.04
Fe2O3	6.85	1.24	TiO2	0.92	0.63
K2O	1.46	1.12	P2O5	0.13	0.06
Na2O	0.53	0.56	LOI	10.50	1.54

, and the mineral composition is shown in Figure 2. The results indicate that the low-grade diatomite is composed mainly of silica, alumina, and iron, accounting for about five sixths (84.39%). Silica constitutes 59.22% of the sample, which accounts for the abundance of opal and quartz. Alumina constitutes 18.32% of the sample and is attributed to the feldspars and clay minerals. Iron constitutes 6.85% of the sample and is attributed to the magnetite and siderite. The component is followed by calcium which constitutes 1.75%, and the component is present mainly as montmorillonite clay. Magnesium and alkalis constitute about 2.23% of the sample. Phosphorous is reported as rare components. Alkalis are present as feldspars (albite and orthoclase) and clay minerals (montmorillonite, kaolinite and muscovite). Magnesium appears to be in montmorillonite, as determined by XRD. In addition, the loss on ignition (LOI) accounted for 10.50% of the sample and is attributed to the organic matter and water.

Table 1 and Figure 2 present the comparison between the raw diatomite and the concentrate. Regarding the chemical composition, Table 1 indicates that the concentrate has high silica (about 86.93%) and low iron (1.24%), but is still high in alumina (6.75%). The high alumina content is attributed to clay and feldspars particles which associate with diatoms during separation.

Table 2. Parameters of the diatomite before and after purification.

Parameter	Value
	Raw Ore	Concentrate
Specific surface area (g·m−2)	49.68	77.30
Most countable pore diameter (nm)	3.81	10.14
Average pore diameter (nm)	13.18	12.27
Pore volume (mL·g−1)	0.23	0.33

presents data for specific surface area, most countable pore diameter, average pore diameter, and pore volume. These data indicate that the specific surface area increased from 49.68 m²·g⁻¹ in the feed to 77.30 m²·g⁻¹ in the concentrate, most countable pore diameter increased from 3.81 nm to 10.14 nm, pore volume increased from 0.23 mL·g⁻¹ to 0.33 mL·g⁻¹, while average pore diameter reduced from 13.18 nm to 12.27 nm. Figure 2 shows the XRD diffraction patterns of the concentrate and raw diatomite. The data for the latter indicates the high reduction of quartz and clay in the concentrate and the increase of the broad peak of opal.

2.2. XRD Analysis

Powder XRD was performed in order to investigate the crystal structure of the phases in the nanocomposite, and the results are shown in Figure 3 and

Table 3. Phase structure and crystal of different photocatalytic materials.

Sample	Phase	Crystal Size (nm)
TiO2	Anatase	7.0
5-CN-TO	Anatase	8.3
0.1-NT-DE-550 °C-2.0 h	Anatase	8.3

. The typical diffraction peaks located at two-theta 25.31°, 36.95°, 37.79°, 38.57°, 48.05°, 53.88°, and 62.69° agree well with the primary reflections of the (101), (103), (004), (112), (200), (105), and (204) planes of the nano-TiO₂ (PDF#71-1166), respectively. The diffraction peaks of pure g-C₃N₄ appearing at two-theta 12.86° and 27.59° are indexed as reflection from (100) and (002) planes corresponding to the inter-layer structural packing and the characteristic inter-planar stacking of the aromatic systems, respectively [34,35]. The diffraction peaks corresponding to TiO₂ are clearly visible in the 5-CN-TO composite, but the diffraction peaks corresponding to g-C₃N₄ are not obvious, which is due to the sublimation of urea during the heating process, lessening the content of g-C₃N₄ in the composite. The diffraction peaks at two-theta 20.92° and 26.70° are indexed as reflection from (100) and (101) planes, corresponding to the characteristic peaks of SiO₂ (PDF#85-0865) in the diatomite. The diffraction peak at two-theta 25.31° corresponds to the (101) plane of the anatase phase TiO₂, and the diffraction peaks at 20.92° and 26.70° correspond to the characteristic peaks of SiO₂. In contrast, the absence of significant g-C₃N₄ diffraction peaks in 0.1-NT-DE-550 °C-2.0 h is attributed to the low content of g-C₃N₄.

The diffraction peak at two-theta of 25.31° (101) was selected as a representative characteristic peak of the sample, and the average crystal size of TiO₂ was calculated according to the Scherer equation. It can be seen that the average crystal size of the nano-TiO₂ anatase phase is 7.0 nm; the average crystal size of the 5-CN-TO anatase phase is 8.3 nm; the average crystal size of the 0.1-NT-DE-550 °C-2.0 h anatase phase is also 8.3 nm. This indicates that the crystal size of the anatase phase becomes larger after the nano-TiO₂ composite with g-C₃N₄ and diatomite. Among them, g-C₃N₄ promotes the growth of the TiO₂ anatase phase and avoids more lattice defects, while the addition of diatomite almost does not affect the crystal structure of TiO₂ [36].

2.3. XPS Analysis

The XPS was employed to confirm the chemical state of elements in the ternary TiO₂/g-C₃N₄/diatomite nanocomposite. Figure 4a shows that the composite is mainly composed of C, N, O, Si, and Ti. The fine spectrum of C 1s (Figure 4b) indicates that C 1s can be deconvoluted into three peaks at 283.30 eV, 284.80 eV, and 287.21 eV; the first two are carbon impurities introduced during the sample processing, and the latter is attributed to sp² hybridized C, such as (N)₂-C=N [37,38]. The fine spectrum of N 1s (Figure 4c) suggests that the core state energy peak at 397.72 eV can be split into four peaks, 395.67 eV, 397.30 eV, 398.65 eV, and 400.20 eV, corresponding to nitrogen-doped Ti-N, sp² hybridized nitrogen (C=N-C), urea polymerization resulting in N-(C)₃, and the amino functional group NH_x (x = 1, 2) [39,40,41,42,43,44]. The fine spectrum of O 1s (Figure 4d) shows that the three peaks at 528.36 eV, 531.24 eV, and 533.84 eV correspond to three different chemical states of O 1s, including O²⁻ in the TiO₂ lattice, OH⁻ groups on the surface of the composite, and O²⁻ with adsorbed H₂O. The content of OH⁻ groups on the surface of the composite is higher than that of TiO₂, while the surface OH⁻ group at 531.24 eV plays an important role in improving the photocatalytic efficiency by trapping holes (h⁺) and generating ·OH, thus inhibiting the complex between electrons (e⁻) and h⁺ [45]. The Si 2p fine spectrum (Figure 4e) indicates that the absorption peak has a binding energy of 101.90 eV, which corresponds to the Si-O of diatomite [46]. The fine spectrum of Ti 2p (Figure 4f) suggests that the binding energies 456.98 eV and 462.95 eV corresponding to Ti 2p^3/2 and Ti 2p^1/2 for the composite are shifted 2.12 eV and 1.85 eV toward lower energies compared to the binding energies 459.10 eV and 464.80 eV corresponding to Ti 2p^3/2 and Ti 2p^1/2 for TiO₂, respectively [47,48]. This shows that the presence of Ti-N electron interactions as well as the possible formation of N→Ti coordination bonds are probably due to the addition of g-C₃N₄ [29,49].

2.4. FT-IR Analysis

The FT-IR spectras for nano-TiO₂, g-C₃N₄, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h composites are shown in Figure 5. For nano-TiO₂, the bands between 400 cm⁻¹ and 700 cm⁻¹ are attributed to the Ti-O-Ti and Ti-O stretching vibration [50]. The bands at 3349 cm⁻¹ and 1634 cm⁻¹ correspond to the -OH bending vibration of adsorbed water [51]. For pure g-C₃N₄, the band around 808 cm⁻¹ is ascribed to out-of-plane bending modes of heterocyclic C-N (C-N=C and N=C-N). The peaks between 1228 cm⁻¹ and 1630 cm⁻¹ are attributed to C=N stretching vibration modes. In addition, the broad peak near 3200 cm⁻¹ is attributed to the N-H and O-H stretching vibration [52]. For 5-CN-TO, the typical characteristic peaks of TiO₂ and g-C₃N₄ can be observed, showing the vibration of Ti-O and C=N. This indicates that TiO₂ and g-C₃N₄ are successfully compounded and can avoid more lattice defects. Furthermore, the intensity of the characteristic vibration peak of g-C₃N₄ is weakened, indicating that TiO₂ and g-C₃N₄ are present in m-CN-TO at the same time, but there is less g-C₃N₄ content. For diatomite, the 1003 cm⁻¹ asymmetric strong absorption band has a shoulder absorption at 1053.41 cm⁻¹ attributed to Si-O antisymmetric stretching vibration. The band at 792 cm⁻¹ is attributed to Si-O symmetric stretching vibration. Additionally, the band at 459 cm⁻¹ is ascribed to O-Si-O antisymmetric bending vibration [53,54]. For 0.1-NT-DE-550 °C-2.0 h, the typical characteristic peaks of TiO₂, g-C₃N₄, and diatomite can be observed, showing the vibration of Ti-O, C=N and Si-O. This suggests that the TiO₂ nanoparticles, g-C₃N₄ flakes, and diatomite skeletons closely bind to each other [55].

2.5. N₂ Adsorption-Desorption Analysis

The N₂ adsorption-desorption measurements were performed to investigate the pore structure of the composites, including specific surface area, most available diameter, average pore diameter, and pore volume, as shown in Figure 6 and

Table 4. Specific surface area and pore structure parameters.

Parameter	TiO2	g-C3N4	Diatomite	5-CN-TO	0.1-NT-DE-550 °C-2.0 h
Specific surface area (m2·g−1)	301.26	87.47	148.63	179.77	145.03
Pore volume (mL·g−1)	0.39	1.63	0.34	0.43	0.33
Average pore diameter (nm)	4.73	43.55	7.96	7.78	8.44
Most available diameter (nm)	1.95	22.84	3.50	3.53	3.47

. The N₂ adsorption-desorption isotherms of nano-TiO₂, g-C₃N₄, diatomite, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h are type IV and H₃ hysteresis echoes, reflecting the pore structures such as flat slits, cracks, and wedges in the samples, where capillary coalescence occurred. The BET simulations reveal that the specific surface areas of TiO₂, g-C₃N₄, diatomite, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h are 301.26 m²·g⁻¹, 87.47 m²·g⁻¹, 148.63 m²·g⁻¹, 179.77 m²·g⁻¹, and 145.03 m²·g⁻¹, respectively. This indicates that the presence of g-C₃N₄ and diatomite reduces the specific surface areas of the composites. The BJH simulations show that the pore volumes of TiO₂, g-C₃N₄, diatomite, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h are 0.39 mL·g⁻¹, 1.63 mL·g⁻¹, 0.34 mL·g⁻¹, 0.43 mL·g⁻¹, and 0.33 mL·g⁻¹, respectively. This illustrates that the presence of g-C₃N₄ and diatomite reduces the pore volumes of the composites. The average pore diameters of TiO₂, g-C₃N₄, diatomite, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h are 4.73 nm, 43.55 nm, 7.96 nm, 7.78 nm, and 8.44 nm, respectively. This indicates that the presence of g-C₃N₄ and diatomite increases the average pore diameters of the composites. In addition, the most available diameters of TiO₂, g-C₃N₄, diatomite, 5-CN-TO, and -550 °C-2.0 h are 1.95 nm, 22.84 nm, 3.50 nm, 3.53 nm, and 3.47 nm, respectively. This shows that the presence of g-C₃N₄ and diatomite increases the most available diameters of the composites. It can be seen that the synergy of the TiO₂ nanoparticles, g-C₃N₄ flakes, and diatomite skeletons optimizes the functional structure of the composite.

2.6. SEM Analysis

The SEM images of nano-TiO₂, g-C₃N₄, diatomite, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h are shown in Figure 7, revealing the morphology characteristics of different photocatalytic materials. The nano-TiO₂ structure is spherical (Figure 7a) and the g-C₃N₄ structure is sheet-like (Figure 7b). At the same time, the agglomeration and accumulation between TiO₂ nanoparticles and between g-C₃N₄ flakes are serious. The 5-CN-TO structure is mainly spherical (Figure 7a,b), and TiO₂ nanoparticles are scattered and loaded on the g-C₃N₄ flake structure (Figure 7c). The presence of g-C₃N₄ inhibits the agglomeration of TiO₂ nanoparticles and forms heterojunctions; the electrons on the g-C₃N₄ conduction band leap to the TiO₂ conduction band through the heterojunctions; and the holes on the TiO₂ valence band are transferred to the g-C₃N₄ valence band, improving the photocatalytic performance of TiO₂ in the visible region [56,57,58,59]. Diatomite has a disc-shaped structure (Figure 7d) with a large number of nanoscale pores and a large specific surface area. With the addition of diatomite, the TiO₂/g-C₃N₄ particles in the 0.1-NT-DE-550 °C-2.0 h are more uniformly distributed and smaller in size (Figure 7e). The surface of diatomite becomes rougher after compounding, providing more surface-active sites for the adsorption and degradation of the MB solution [55]. Figure 7f shows that the 1.0-NT-DE-550 °C-2.0 h has a very serious accumulation of TiO₂, g-C₃N₄, and diatomite skeletons due to the large mass of diatomite, which reduces the photocatalytic active sites of the composite and is not conducive to the light absorption and degradation of the MB solution [60].

2.7. TEM Analysis

The TEM, HRTEM, and SAED images of TiO₂/g-C₃N₄/diatomite are shown in Figure 8. The g-C₃N₄ structure is sheet-like in the composite, TiO₂ is mainly distributed on the surface of g-C₃N₄ flakes in the form of particles, and some agglomeration phenomenon occurs but is not obvious (Figure 8a,b).

This is due to the fact that the presence of g-C₃N₄ and diatomite inhibits agglomeration and facilitates the absorption of visible light and the degradation of the MB solution. The TiO₂ nanoparticles are in close contact with the g-C₃N₄ nanosheets, and the multi-level pore structure cannot be clearly observed in the TEM images due to the small mass of diatomite. Lattice fringes from TiO₂ (101) are visible in the corresponding HRTEM image in Figure 8c with an inter-planar spacing of 0.35 nm, consistent with the card data (PDF#71-1166) and the calculation according to the corresponding ring pattern; moreover, the SADE pattern of the composite determines the lattice fringe of the anatase phase TiO₂ (101) crystal plane to be 0.35 nm (Figure 8d) [61].

2.8. UV-Vis DRS Absorption Spectra Analysis

The UV-vis DRS spectras of TiO₂, g-C₃N₄, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h are shown in Figure 9a respectively. Correspondingly, the band gap was calculated following (αhv)^1/2 = k(hv − E_g), where α, h, v, k, and E_g represent the absorption coefficient, Planck constant, light frequency, a constant, and band gap, respectively; the results can be seen in Figure 9b [62,63]. An adsorption edge in the ultraviolet range up to 400 nm appears for the nano-TiO₂, which has a band gap of 3.20 eV. The maximum absorption wavelength of pure g-C₃N₄ is about 460 nm with a band gap of 2.78 eV, and it has strong absorption in the visible region. Compared with TiO₂ and g-C₃N₄, the maximum absorption wavelength of the 5-CN-TO composite is significantly red-shifted, and the absorption intensity is significantly enhanced in the ultraviolet and visible range with a band gap of 1.84 eV.

This is due to the superposition of the ultraviolet response of TiO₂ and the visible response of g-C₃N₄ in the 5-CN-TO composite. The incorporation of g-C₃N₄ may indirectly cause a change in the electronic structure of TiO₂ through the N→Ti coordination bond, narrowing the band gap to extend the visible response [64,65,66]. Furthermore, the band gaps of nano-TiO₂ and g-C₃N₄ are 3.20 eV and 2.78 eV, respectively. Pure g-C₃N₄ can be excited by visible light and generate electrons, and the photogenerated electrons are transferred to the TiO₂ conduction band through the interface of TiO₂ and g-C₃N₄, which makes the visible light response range of the composite broaden. Compared with TiO₂/g-C₃N₄, the absorption wavelength of 0.1-NT-DE-550 °C-2.0 h is blue-shifted, but the absorption intensity is slightly larger than that of TiO₂/g-C₃N₄, around 400 nm. This indicates that the addition of diatomite reduces the absorption wavelength of TiO₂/g-C₃N₄. On the other hand, the band gaps of TiO₂/g-C₃N₄ and 0.1-NT-DE-550 °C-2.0 h are 1.84 eV and 1.91 eV, respectively, indicating that the electronic interaction between diatomite and TiO₂/g-C₃N₄ leads to a change in the band gap of TiO₂/g-C₃N₄. Moreover, the multi-level pore structure of the diatomite allows the active sites of 0.1-NT-DE-550 °C-2.0 h to increase and provides a strong adsorption capacity. The addition of diatomite facilitates the photocatalytic performance of the composites with similar band gaps [60].

2.9. Effect of Different Photocatalytic Materials

Figure 10a exhibits the influence of different photocatalytic materials on the degradation performance of the MB solution under visible light irradiation. After 360 min of reaction, the degradation rate of nano-TiO₂ is only 16.10%, indicating that TiO₂ has almost no visible light catalytic performance. g-C₃N₄ has a degradation rate of 29.41%, with a higher response to visible light than TiO₂. 5-CN-TO shows a better visible light catalytic performance with a degradation rate of 87.35%, attributed to the effective improvement of the visible light response range of TiO₂ by g-C₃N₄. Diatomite also has almost no visible light activity, but has 59.37% adsorption due to the adsorption of its multi-stage pore structure. 0.1-NT-DE-550 °C-2.0 h exhibits the best visible light catalytic performance, attributed to the synergistic effect of TiO₂, g-C₃N₄, and diatomite, with 99.51% degradation of the MB solution. The monomeric characteristic absorption peak of MB is at 664 nm, and dimer characteristic absorption peak is at 615 nm [67]. Figure 10b indicates that the intensity of the characteristic absorption peaks of the MB solution decreases sequentially after the photocatalytic reaction of TiO₂, g-C₃N₄, diatomite, 5-CN-TO, and 0.1-NT-DE-550 °C-2.0 h at 664 nm and 615 nm, respectively. Among them, the characteristic absorption peak of the MB solution almost disappears at 664 nm after 0.1-NT-DE-550 °C-2.0 h photocatalytic reaction, and only the dimer characteristic absorption peak with smaller intensity at 615 nm remains.

2.10. Cyclic Stability Analysis

The results of repeated MB conversion experiments with TiO₂/g-C₃N₄/diatomite are shown in Figure 11. The results show that the best degradation effect is achieved in the 1st cyclic photocatalytic reaction with 98.77%; the degradation effect is similar in the 2nd and 3rd reactions, with 90.59% and 89.16%, respectively. The degradation effect tends to be stable after three cycles of photocatalytic reaction, and the composite still has a high degradation rate for the MB solution, indicating that the photocatalytic performance of the TiO₂/g-C₃N₄/diatomite nanocomposite is stable, with good recyclability under visible light, and can be used for recycling in practical life and production. The enhanced photocatalytic activity and recyclability of TiO₂/g-C₃N₄/diatomite compared to unary and binary photocatalytic materials. Among them, the degradation rate of TiO₂/g-C₃N₄/diatomite prepared in this study is 1.66% higher for the MB solution in the 3rd photocatalytic reaction compared to the TiO₂/g-C₃N₄ prepared by Hoa et al. [56], which is associated with the introduction of a diatomite skeleton, improving the mechanical and thermal stability and integrity to consequently maintain steady active sites for the conversion of the MB molecule.

2.11. Reactive Radical Capture Test

In order to further verify the effect of reactive radicals in the process of photocatalytic reaction, EDTA-2Na, BQ, and TBA are added to the MB solution as h⁺, O₂⁻·, and ·OH trapping agents, and the results are shown in

Table 5. Fitting parameters of photocatalytic reaction kinetics.

Model	Capture Agent	Kinetic Equation	Parameter
			k (10−3)	R2
First-level kinetic	Blank	−ln(Ct − C0) = 12.84 × 10−3t + 0.01	12.84	0.94
	EDTA-2Na	−ln(Ct − C0) = 7.34 × 10−3t + 0.29	7.34	0.99
	BQ	−ln(Ct − C0) = 3.78 × 10−3t + 0.84	3.78	0.96
	TBA	−ln(Ct − C0) = 8.07 × 10−3t + 0.58	8.07	0.99

and Figure 12, respectively. Figure 12a shows that the MB solution is gradually degraded and discolored as the photocatalytic reaction time increases. Among them, the inhibitory effect of BQ is the most obvious, with 89.56% degradation of the MB solution by the composite, followed by EDTA-2Na and TBA, with 94.99% and 96.82% degradation rates, respectively (Figure 12b). However, the degradation rate without trapping agents is 99.51%, indicating that O₂⁻·, h⁺, and ·OH all play a role in the photocatalytic reaction process (Figure 12c). The order of degradation of the MB solution is blank > TBA > EDTA-2Na > BQ, indicating that the oxidation reaction occurring in the conduction band is the main cause of photocatalytic degradation of the MB solution. Meanwhile, diatomite loads TiO₂ and g-C₃N₄ can better promote the separation of holes and electrons, and moreover, can provide more surface-active sites, thus improving the photocatalytic performance of the composite. Table 4 and Figure 12d show that the reaction rate constant k_max = 12.84 × 10⁻³ and R² = 0.94 for the first-level kinetic equation of the catalytic reaction when no capture agent is added. In addition, the reaction rate constants k of the first-level kinetic equations for the capture agents are 7.34 × 10⁻³, 3.78 × 10⁻³, and 8.07 × 10⁻³ for EDTA-2Na, BQ, and TBA, respectively. The reaction rate constant of BQ is 1.75, 3.41, and 1.59 times higher than those of other capture agents, indicating that the O₂⁻· mainly controls the photocatalytic reaction process of the MB solution.

2.12. Photocatalytic Degradation Mechanism

The TiO₂/g-C₃N₄/diatomite composite exhibits excellent photocatalytic performance for the MB solution, and after characterization of the material structure and properties, it is found that the main reason for the improved photocatalytic performance of the composite is the significant extension of their spectral response range and the substantial increase in the number of reactive sites. Compared with the unary photocatalytic material, the photogenerated charge separation efficiency of the photocatalytic composite with diatomite loaded with TiO₂ and g-C₃N₄ is significantly improved.

The photocatalytic degradation mechanism is proposed and elucidated in light of the above experimental results. The band edges of the valence band (VB) and conduction band (CB) of a semiconductor can be obtained using the following Equations (1) and (2) [47]:

$$E_{CB}=\chi -E_e-0.5E_g$$

(1)

$$E_{VB}=E_{CB}+E_g$$

(2)

where χ is the absolute electronegativity of the semiconductor (χ is 5.81 eV and 4.73 eV for TiO₂ and g-C₃N₄, respectively [68]), E_e is the energy of the free electrons on the hydrogen scale (4.50 eV), and E_g is the band gap of the semiconductor. Based on the empirical energy band equation and the forbidden band width derived from Uv-vis DRS Tauc curves, the calculated CB and VB of g-C₃N₄ are −1.16 eV and 1.62 eV, respectively, which are more positive than those of TiO₂ (−0.29 eV and 2.91 eV, respectively). The g-C₃N₄ can be activated when the nanocomposite is irradiated under visible light leaving an equivalent number of holes with positive charge in the corresponding positions, and the electrons in full VB of g-C₃N₄ may move to the empty CB [69]. Due to its narrower band gap (2.78 eV) as compared to TiO₂ (3.20 eV), the presence of g-C₃N₄ facilitates the utilization of visible light. Moreover, well matched band structures between g-C₃N₄ and TiO₂ favor the ready transferring of photogenerated electrons at the surface of g-C₃N₄ flakes to the surface of TiO₂ particles to effectively prevent the recombination of photogenerated electrons and holes [70]. This migration effect of photogenerated carriers can effectively suppress the compounding of photogenerated electrons and holes, and improve the separation efficiency of carriers. More photogenerated electrons in the conduction band of TiO₂ and more holes in the valence band of g-C₃N₄ greatly increase the active sites on the surface of the composite, accelerating the production of O₂⁻· and ·OH in the system. The reaction of photogenerated electrons on the TiO₂ CB with dissolved O₂ forms O₂⁻·, while some of the O₂⁻· reacts with H₂O/H⁺ to produce a small amount of ·OH. The VB potential of g-C₃N₄ (1.62 eV) is higher than that of H₂O/OH (1.99 eV), so the photogenerated holes react directly with MB without producing ·OH. In addition, diatomite contains a large number of hydroxyl groups and hydrogen bonds on its surface, which can produce more Bronsted acidic sites. The acidic sites can effectively trap the positively charged photogenerated holes, providing more active adsorption sites and photocatalytic reaction centers. Si-OH can react with photogenerated holes to form ·OH., and diatomite with a large number of microporosity can adsorb MB molecules, making TiO₂/g-C₃N₄ better for photocatalytic reaction with the MB solution. Eventually, under the synergistic effect of TiO₂, g-C₃N₄, and diatomite in the composite, O₂⁻·, h⁺, ·OH, and MB molecules undergo oxidation reaction so that the strong chromophore within the MB molecule is attacked and the large conjugated system is destroyed and degraded to inorganic small molecules, H₂O and CO₂. It can be seen that the composite of TiO₂, g-C₃N₄ and diatomite increases the possibility of photocatalytic reaction of visible light on the surface of TiO₂, and increases the quantum efficiency of g-C₃N₄ and the number of active sites on the surface of the composite [60]. The reaction pathway is speculated as follows (Equations (3)–(9)):

$$\mathrm{g}\text{-}{\mathrm{C}}_3{\mathrm{N}}_4+hv\to \mathrm{g}\text{-}{\mathrm{C}}_3{\mathrm{N}}_4\left({\mathrm{e}}^{-}+{\mathrm{h}}^{+}\right)$$

(3)

$$\mathrm{Ti}{\mathrm{O}}_2+hv\to \mathrm{Ti}{\mathrm{O}}_2\left({\mathrm{e}}^{-}+{\mathrm{h}}^{+}\right)$$

(4)

$$\mathrm{g}\text{-}{\mathrm{C}}_3{\mathrm{N}}_4\left({\mathrm{e}}^{-}+{\mathrm{h}}^{+}\right)+\mathrm{Ti}{\mathrm{O}}_2\left({\mathrm{e}}^{-}+{\mathrm{h}}^{+}\right)\to \mathrm{g}\text{-}{\mathrm{C}}_3{\mathrm{N}}_4\left({\mathrm{h}}^{+}\right)+\mathrm{Ti}{\mathrm{O}}_2\left({\mathrm{e}}^{-}\right)$$

(5)

$$\mathrm{Ti}{\mathrm{O}}_2\left({\mathrm{e}}^{-}\right)+{\mathrm{O}}_2\to \mathrm{Ti}{\mathrm{O}}_2+·{\mathrm{O}}_2^{-}$$

(6)

$$·{\mathrm{O}}_2^{-}+{\mathrm{H}}_2\mathrm{O}/{\mathrm{H}}^{+}\to ·\mathrm{OH}$$

(7)

$$\mathrm{Ti}{\mathrm{O}}_2\left({\mathrm{h}}^{+}\right)+{\mathrm{H}}_2\mathrm{O}/\mathrm{O}{\mathrm{H}}^{-}\to ·\mathrm{OH}$$

(8)

$$\left(·{\mathrm{O}}_2^{-}, {\mathrm{h}}^{+}, ·\mathrm{OH}\right) +\mathrm{MB}\to \mathrm{C}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}$$

(9)

Figure 13 is a schematic diagram of the electron-hole migration and separation pathways on the surface of the TiO₂/g-C₃N₄/diatomite composite.

3. Materials and Methods

3.1. Materials

The raw diatomite was obtained from the diatomite planning area in Heilongjiang, China. Sulfuric acid (H₂SO₄) was provided by Liaoning Quanrui Reagent Co., Ltd., Jinzhou, China. Sodium hydroxide (NaOH), sodium hexametaphosphate ((NaPO₃)₆), urea (CH₄N₂O), and MB were provided by Sinopham Chemical Regent Co., Ltd., Beijing, China. Nano-TiO₂, disodium ethylenediaminetetraacetate (EDTA-2Na), benzoquinone (BQ), and tert-butanol (TBA) were provided by Shanghai Aladdin Biochemical Technology Co., Ltd., Shanghai, China. The diatomite needed to be further purified; other reagents were analytical grade and used without further purification.

3.2. Purification of Low-Grade Diatomite

200 g of low-grade diatomite was dispersed in 300 mL of water and soaked for 2 h. The soaked diatomite was scrubbed in a scrubbing machine at a rotational speed of 1300 r·min⁻¹, a material concentration of 25%, a time of 30 min, a dosage of 0.3% NaOH, a dosage of 0.3% (NaPO₃)₆, a settling time of 0.5 min, and a siphoning depth of 13 cm. The upper pulp was treated with magnetic separation in a 1.91 T wet strong magnetic separator. The non-magnetic pulp was dewatered and put into a three-necked bottle placed in an oil bath for leaching treatment at a H₂SO₄ concentration of 26.63%, a temperature of 130 °C, a solid-liquid ratio of 1:4, a rotational speed of 150 r·min⁻¹, and a time of 6 h. The leached material was dehydrated and roasted in a chamber resistance furnace at a heating rate of 10 K·min⁻¹, a temperature of 600 °C, and a time of 2.0 h to obtain the diatomite concentrate.

3.3. Synthesis of TiO₂/g-C₃N₄ and TiO₂/g-C₃N₄/Diatomite

The nano-TiO₂ and urea were ultrasonically dispersed in the 50 mL of deionized water at a mass ratio of 1:5 to obtain TiO₂/g-C₃N₄ precursor. The nano-TiO₂/urea and diatomite concentrate were ultrasonically dispersed at a mass ratio of 1:0.1 to obtain TiO₂/g-C₃N₄/diatomite precursor. The dispersed materials were dried in an electric thermostatic blast dryer at 60 °C and then fully ground in an agate bowl. The milled materials were loaded into an alumina crucible and placed in a chamber resistance furnace and heated up to 550 °C at a heating rate of 5 K·min⁻¹ and held for 2.0 h to obtain the TiO₂/g-C₃N₄ composite and the TiO₂/g-C₃N₄/diatomite composite, numbered 5-CN-TO and 0.1-NT-DE-550 °C-2.0 h, respectively. Meanwhile, g-C₃N₄ was made from urea by the above thermal polymerization process.

3.4. Characterization

The structure of the samples was measured by X-ray diffraction (XRD) with a D8 Adcance diffractometer (Bruker, Karlsruhe, Germany) using Cu Kα radiation at diffraction wavelength of 0.15406 nm, operating voltage of 40 kV, tube current of 40 mA, scan speed of 12°·min⁻¹, and scan range of 5° to 80°. The sample was prepared by selecting a copper mesh, and ultrasonically dispersed with ethanol for 5 min. The group structures were analyzed by Fourier transform infrared spectrometer (FT-IR) using a Tensor II (Bruker, Karlsruhe, Germany) equipped with an ATR accessory at wave number range of 4000~400 cm⁻¹, resolution of 4 cm⁻¹, and scan number of 16. The N₂ adsorption-desorption measurement was performed using 3H-2000PMV (Beishide Instrument, Beijing, China) at a heating temperature of 473.15 K and degassing time of 6 h. The UV-vis diffuse reflectance spectra (DRS) of the catalysts were collected on a UV-vis spectrometer (Hitachi U4150, Hitachi, Tokyo, Japan) equipped with an integrating sphere at a test range of 200~800 nm, and the reference substance was BaSO₄ white plate. The absorbance of the centrifuged MB supernatant at 664 nm and the absorption scan spectrum at a scan range of 400~780 nm, step size of 0.5 nm, and accuracy of 30 were measured using an ultraviolet visible spectrophotometer (UV-6000PC, Shanghai Metash Instruments, Shanghai, China). A scanning electron microscope (SEM) was used to characterize the morphology of the samples.

3.5. Photocatalytic Degradation of MB Mechanism

A 350 W xenon lamp (HDL-II, Anhui Aoke Test Equipment Co., Ltd., Anhui, China) was used to simulate the light source, a 10 mg·L⁻¹ MB solution was used as the contaminant, and 0.4 mg·L⁻¹ TiO₂/g-C₃N₄/diatomite composite was added to the MB solution. The reaction system was placed in a dark place before turning on the light source, and the adsorption was avoided for 60 min to make the composites reach the adsorption-desorption equilibrium. The reaction was carried out for 360 min, and the absorbance of the supernatant was measured at 664 nm after sampling at 30 min intervals and centrifugation at 3000 r·min⁻¹ to obtain C_t/C₀.

The composite after the photocatalytic reaction was separated from the MB solution by centrifugation; rinsed with deionized water and ethanol in turn for three times, respectively, to reduce the influence of other impurities; dried and collected in an electric thermostatic blast drying oven; and tested again according to the test procedure of photocatalytic performance to evaluate the stability of the photocatalytic reaction of the composite.

The role of reactive radicals during the TiO₂/g-C₃N₄/diatomite photocatalytic reaction was determined by the capture test. EDTA-2Na, BQ, and TBA were added to the MB solution as capture agents for h⁺, O₂⁻·, ·OH, respectively.

The photocatalytic reaction primary kinetic model is shown in Equation (10) [34].

$$-\ln \left(\frac{C_t}{C_0}\right)=kt$$

(10)

where C₀ is the initial MB concentration of the solution at each measurement; C_t is the MB concentration at the moment t of the reaction; k is the reaction rate constant; t is the reaction time; m is the mass of the composites.

4. Conclusions

The combined scrubbing-magnetic separation-acid leaching-roasting process was used to increase the SiO₂ content from 59.22% to 86.93%, reduce the Al₂O₃ content from 18.32% to 6.75%, and reduce the Fe₂O₃ content from 6.85% to 1.24% of the low-grade diatomite from Heilongjiang, China, under optimal process conditions. Meanwhile, an ultrasonic-thermal polymerization method applicable to the preparation of fabricating ternary TiO₂/g-C₃N₄/diatomite nanocomposite from the low-grade diatomite purified concentrate was proposed, and the degradation rate of the MB solution reached 99.51%. The interaction effects of carrier ratio, polymerization temperature, and polymerization time on the photocatalytic performance of the TiO₂/g-C₃N₄/diatomite composite was obvious; the effects of all three factors on the degradation of the MB solution reached a highly significant level, and the magnitude of the influence ability was: carrier ratio > polymerization temperature > polymerization time. O₂⁻·, h⁺, ·OH all played a role in the photocatalytic reaction of the TiO₂/g-C₃N₄/diatomite to the MB solution, where O₂⁻· mainly controlled the photocatalytic process. The TiO₂/g-C₃N₄/diatomite composite had the best degradation effect on the MB solution in the 1st cycle with 98.77%; the 2nd and 3rd cycles had similar degradation effects with 90.59% and 89.16%, respectively, and the degradation effects tended to be stable. After 3 cycle reactions, the composite still had a high degradation rate to the MB solution and could be used for recycling in practice. This nanostructured TiO₂/g-C₃N₄/diatomite composite has fascinating visible light catalytic activity and excellent stability, which opens a new path for the comprehensive development and utilization of low-grade diatomite resources.