Urea, p-benzoquinone (BQ), and ethylenediamine tetraacetic acid disodium (EDTA-2Na) were purchased from Tianjin Damao Chemical Reagent Factory. Lithium fluoride (LiF), tetracycline hydrochloride (TC), and isopropanol (IPA) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. Iron nitrate [Fe(NO₃)₃·9H₂O] was purchased from Shanghai McLean Biochemical Technology Co., Ltd. Ti₃AlC₂ MAX phase was purchased from Hangzhou Namao Technology Co., Ltd. Hydrochloric acid (37% wt.) was purchased from Guangzhou Chemical Reagent Factory. And, all the chemicals are analytically pure except for special marked, and they were directly used without further purification.

HCl solution (20 ml of 9 mol/L) was added to the Teflon beaker, and 1.0 g of LiF powder was slowly added to the beaker under the action of a magnetic agitator. After it gets fully dissolved, the solution was stirred for 10 min. To avoid overheating, 1.0 g of Ti₃AlC₂ powder was added slowly for about 20 min. After continuous stirring at 35°C for 48 h, they were centrifuged at 4,000 rpm for 20 min, washed with 3 mol/L HCl solution several times to remove the residual LiF after reaction, and then washed repeatedly with deionized water until the pH of the supernatant was close to 7. The collected Ti₃C₂ precipitate was freeze-dried thoroughly, and the obtained powder was multilayer Ti₃C₂ powder.

Fe(NO₃)₃·9H₂O (0.06 g) was dissolved and dispersed in 50 ml of deionized water, 10 g of urea was added to dissolve completely, and 0.08 g of Ti₃C₂ powder was added under an ultrasonic environment to form a uniform dispersion solution, dried thoroughly, and ground evenly. Calcination was carried out in a 50 ml crucible with a lid in a muffle furnace under an air atmosphere. The temperature rose to 550°C at a rate of 50°C/min, and the temperature was held for 1 h. The resulting brown-yellow solid was ground thoroughly to form Fe-C₃N₄/Ti₃C₂ powder (see Scheme 1).

The phase characterization of the samples was determined by an X-ray powder diffract graph (D8 Advance, Brock AXS, Germany), using Cu-Kɑ rays with a wavelength of 154 p.m. and a scanning range of 2θ = 5°∼80°; the morphology and structure of the samples were characterized by MIRA 3 LMU scanning electron microscope SEM (TESCAN Brno, S.R.O., Czech Republic) at a voltage of 15 kV; the morphology and microstructure were also observed by transmission electron microscopy TEM and high-resolution transmission electron microscopy HR-TEM (FEI Talos F200x G2,United States); the element was mapped by the energy-dispersive spectroscopy EDS (FEI Talos F200x G2, Super-X, United States); the element composition and existence form of the samples were determined by X-ray photoelectron spectroscopy XPS (Thermo Scientific K-Alpha Hangzhou Yanqu Information Technology Co., LTD) using the Al-Kα rays; the thermal stability of the samples was characterized by using a thermogravimetric analyzer TGA (TGA-4000, Platinum Elmer, United States); the specific surface area of the samples was analyzed by an automatic specific surface adsorption instrument BET (BELSORP-max, DKSH Commercial Co., LTD., Japan); the optical absorption performance of the sample was reflected by using the UV-visible diffuse reflector DRS (Shimadzu UV-2600, Japan) and calculating the bandgap of the photocatalyst. BaSO₄ was used as the reference sample; the photocatalytic activity of the samples was analyzed by fluorescence PL (Shimadzu RF-5301PC, Japan), and the excitation wavelength was 322 nm.

50 ml of 20 mg/L TC solution was added into the beaker, then 20 mg of Fe-C₃N₄/Ti₃C₂ powder was added, and shook well. The mixture was stirred away from light, the proper amount of mixture was absorbed every 10 min, and the absorbance was measured at 356 nm by using a visible spectrophotometer.

50 ml of 20 mg/L TC solution was added into the beaker, 20 mg of Fe-C₃N₄/Ti₃C₂ powder was added, and shook well. The mixture was stirred away from light until the adsorption–desorption is balanced, then placed it in a xenon lamp source (λ > 420 nm, 280W), the proper amount of mixture was absorbed every 20 min, and the absorbance was measured at 356 nm by using a visible spectrophotometer. Benzoquinone (BQ), isopropyl alcohol (IPA), and ethylenediamine tetraacetic acid disodium (EDTA-2Na) were used as the trapping agent of superoxide free radical (•O₂ ⁻), hydroxyl free radical (•OH), and hole (h⁺), respectively, to explore the degradation mechanism of pollutants in the photocatalytic reaction.

The photoelectrochemical measurements were conducted on a three-electrode electrochemical workstation (A4602, Wuhan Kesite Instrument Co., LTD.), in which the working electrode, counter electrode, and reference electrode were the obtained samples, platinum foil, and Ag/AgCl electrode, respectively. The working electrodes were prepared by the drop-coating method: typically, 2 mg of catalyst and 15 μl Nafion solution were dispersed in 1 ml of ethanol and sonicated for 30 min. Afterward, the resulted homogeneous suspension was dripped onto the FTO and dried at room temperature. Here, all photoelectrochemical measurements were carried out in 0.5 m Na₂SO₄ electrolyte, and the light source was identical to that of photocatalytic measurements. Mott–Schottky was recorded at a frequency of 1,000 Hz.

The phase of the sample was characterized by an X-ray powder diffractometer, and the results are reflected in Figure 1A which shows the XRD patterns of Ti₃AlC₂ and multilayer Ti₃C₂. A series of diffraction peaks corresponding to (002), (004), (101), (104), (105), (107), (109), and (110) crystal planes can be observed from the XRD spectra of Ti₃AlC₂. After being etched by HF, the strongest peak of Ti₃C₂ at 39.0° disappeared, and the peaks corresponding to (002) and (004) crystal planes widened and shifted to a lower angle, indicating that the Al layer in Ti₃AlC₂ had been removed, and Ti₃AlC₂ was successfully converted to Ti₃C₂. At the same time, the Ti₃C₂ layer structure becomes thinner. From Figure 1B, two evident diffraction peaks located at 12.83° and 27.05° for g-C₃N₄ are observed, corresponding to (100) and (002) crystal planes, respectively. Fe-C₃N₄ mainly had a peak at 27.05°, corresponding to the (002) crystal plane of g-C₃N₄. The characteristic peak of Fe was not observed, probably due to the small amount of doped Fe or the weak characteristic peak. The XRD pattern of g-C₃N₄/Ti₃C₂ shows two distinct peaks at 6.93° and 27.05°, respectively, corresponding to the (002) crystal plane of Ti₃C₂ and the (002) crystal plane of g-C₃N₄, indicating that g-C₃N₄ and Ti₃C₂ have achieved good recombination. The XRD pattern of Fe-C₃N₄/Ti₃C₂ is similar to that of g-C₃N₄/Ti₃C₂. The characteristic peaks of Ti₃C₂ and g-C₃N₄ are observed, and the characteristic peaks of Fe are not observed, on account of the small amount of Fe doped or the weak characteristic peaks.

Compared with multilayer Ti₃C₂, the intensity of characteristic peaks on the (002) plane of g-C₃N₄/Ti₃C₂ and Fe-C₃N₄/Ti₃C₂ at 6.93° is much weaker. It is speculated that NH₃ released during urea calcination can be inserted into the interlayer of multilayer Ti₃C₂ as a small molecule in the process of compounding with Ti₃C₂, which effectively prevents the stacking of Ti₃C₂ interlayers, plays a supporting role, promotes the expansion of interlayer spacing of multilayer Ti₃C₂, and forms a close 2D/2D interlayer contact surface with g-C₃N₄. At the same time, it is speculated that Fe doping causes different degrees of polycondensation of g-C₃N₄ and delays the phase transition process, resulting in the decrease of cell parameters and crystal plane spacing of g-C₃N₄ and the increase of specific surface area. Moreover, Ti₃C₂ and g-C₃N₄ form a 2D/2D contact surface, which directly affects the periodic stacking interlayer structure of g-C₃N₄, resulting in the decrease of the diffraction peak intensity of g-C₃N₄ at 27.05° and the broadening.

The microstructure of the samples was characterized by SEM, and the results are shown in Figure 2. It can be seen from Figure 2A that the unetched Ti₃AlC₂ is a bulk stacking structure. The morphology of Ti₃C₂ obtained by etching is shown in Figure 2B. Clear interlamellar spacing can be observed visually, showing a two-dimensional layered structure of MXene. Multilayer Ti₃C₂ stacked together changed into an accordion-like structure. It is observed from Figure 2C that g-C₃N₄ is mostly agglomerated and a small part is thin. It is observed from Figure 2D that the aggregation of g-C₃N₄ after Fe-doping is weakened, showing a thin lamellar structure. It is observed from Figure 2E that, after the combination of g-C₃N₄ and Ti₃C₂, the two form a close interlayer structure. From Figure 2F, it can be observed that there is a close interlayer structure between g-C₃N₄ and Ti₃C₂ in Fe-C₃N₄/Ti₃C₂. It is assumed that urea releases NH₃ during the calcination, and small molecule NH₃ inserts into the interlayer structure of Ti₃C₂ and acts as a gas template to strip Ti₃C₂, effectively preventing the accumulation of Ti₃C₂ layers (Yang et al., 2020).

The morphology and microstructure of the Fe-C₃N₄/Ti₃C₂ catalyst were observed by TEM (Figure 3A). It can be seen that the sample exhibits a typical layered structure. In Figure 3B, the shallowest area is g-C₃N₄, the darker overlying layer is Ti₃C₂, and the darkest black area is Fe species. More information about Fe-C₃N₄/Ti₃C₂ can be obtained from Figure 3C. The lattice spacing of Ti₃C₂ is 0.22 nm, and the lattice spacing of Fe species is 0.40 nm. In addition, the EDS element map (Figure 3D) of the sample further confirms that there is good interaction between Fe, Ti₃C₂, and g-C₃N₄. The elements C, N, O, Ti, and Fe are distributed uniformly in the samples. According to Figure 3E, the contents of Fe and Ti are 13.34 and 25.16%, respectively. These results show that g-C₃N₄ is uniformly modified by Ti₃C₂ and Fe and has close contact.

The chemical composition and valence state of the samples were characterized by X-ray photoelectron spectroscopy, and the results are shown in Figure 4. It is observed from Figure 4A that the measured spectra of Fe-C₃N₄/Ti₃C₂ show the existence of C, N, Ti, O, and Fe elements, indicating that Fe-doping and the composite of Ti₃C₂ were successfully realized in g-C₃N₄. It can be seen from the C 1s spectrum (Figure 4B) that the C 1s peak can be fitted into three peaks, namely, graphite phase carbon (C–C) at 284.38 eV, sp³-hybridized carbon (C-N=C) at 285.48 eV, and sp²-hybridized carbon (N-C=N) at 287.58 eV. The N 1s spectrum (Figure 4C) shows three distinct peaks, which are ascribed to sp²-hybridized nitrogen (C-N=C) at 398.18 eV, tertiary nitrogen group (N-(C)₃) at 399.38 eV, and free amino groups (C-N-H) at 400.10 eV, respectively. According to the Ti 2p spectrum (Figure 4D), Ti 2p XPS peaks are deconvoluted into three peaks, namely, Ti-N bond at 457.70 eV, Ti-C 2p^1/2 bond at 462.20 eV, and Ti-O 2p^1/2 bond at 463.84 eV. The existence of the Ti-O 2p^1/2 bond may be related to the slight oxidation of Ti₃C₂. According to the Fe 2p spectra (Figure 4E), there are three chemical states of Fe, namely, Fe-N bond at 710.48 eV, trivalent iron satellite peak at 718.30 eV, and trivalent iron in Fe₂O₃ at 713.08 and 724.28 eV, which indicates that there are both iron–nitrogen chemical bonds and Fe₂O₃ in the material. From the above analysis, Fe-doping and Ti₃C₂ composite were successfully realized in g-C₃N₄. In Fe-C₃N₄/Ti₃C₂, Fe mainly exists in the form of the Fe-N bond and Fe₂O₃, Ti forms the key existence in the Ti-N bond and Ti-C bond, which further confirms that Fe and Ti₃C₂ have been successfully combined with g-C₃N₄.

The thermal stability of the sample was characterized by a thermogravimetric analyzer, and the results are shown in Figure 5A. It can be observed from Figure 5A that the mass change of Ti₃C₂ is not obvious with the increase in temperature, showing excellent stability. The mass loss of g-C₃N₄, Fe-C₃N₄, g-C₃N₄/Ti₃C₂, and Fe-C₃N₄/Ti₃C₂ was very rapid within a certain temperature range, which was due to the decomposition of the C-N bond of g-C₃N₄ in the material to generate CO₂ and NO₂. At the same time, pure g-C₃N₄ was completely decomposed when the temperature exceeded 777°C, while Fe-C₃N₄, g-C₃N₄/Ti₃C₂, and Fe-C₃N₄/Ti₃C₂ were not completely decomposed, which was attributed to the stability of Fe and Ti₃C₂ residues in the sample.

The specific surface area of the sample was analyzed by an automatic surface adsorption instrument. Figure 5B shows that the specific surface areas of g-C₃N₄, Fe-C₃N₄, g-C₃N₄/Ti₃C₂, and Fe-C₃N₄/Ti₃C₂ are 38.55 m²/g, 54.13 m²/g, 49.25 m²/g, and 68.58 m²/g, respectively. Both Fe-doping and composite Ti₃C₂ are conducive to expanding the specific surface area of photocatalyst, providing more active sites and improving the adsorption performance of materials. At the same time, the larger the specific surface area, the e⁻ generated by g-C₃N₄ can migrate to the surface of the catalyst more quickly, which effectively reduces the recombination rate of electron and hole pairs and improves the photocatalytic performance of the material.

The absorption properties of the samples were recorded by UV-Vis diffuse reflectance spectroscopy. The results are shown in Figures 5C,D. It is observed from Figure 5C that the Ti₃C₂ shows broad absorption in the whole region (200∼800 nm), and there is no obvious absorption with the edge, indicating that Ti₃C₂ has metallic properties. The pristine g-C₃N₄ displays a typical absorption region at about 450 nm, which is consistent with the literature (Gebreslassie et al., 2019). The optical cutoff wavelengths of Fe-C₃N₄, g-C₃N₄/Ti₃C₂, and Fe-C₃N₄/Ti₃C₂ are 467 nm, 458, and 470 nm, respectively. It can be seen that after Fe-doping and Ti₃C₂ composite, the sample shows a redshift compared with the monomer g-C₃N₄, and the visible light absorption capacity is enhanced, indicating that Fe-doping and composite Ti₃C₂ can effectively improve the visible light absorption performance of g-C₃N₄. Due to the strong absorption of dark Ti₃C₂ in the whole wavelength range (200∼800 nm), the absorption edge of g-C₃N₄ can readily shift after recombination, so Ti₃C₂ composite can effectively enhance the light absorption of g-C₃N₄. According to the energy level theory, it is suggested that the bandgap of g-C₃N₄ forms the impurity energy level after Fe-doping. The electrons only need to absorb photons with small energy to realize the indirect transition of energy level, which can absorb photons with long-wavelength, broaden the visible light absorption range of g-C₃N₄ and improve the utilization rate of visible light. From Figure 5D the band gaps of g-C₃N₄, Fe-C₃N₄, g-C₃N₄/Ti₃C₂, and Fe-C₃N₄/Ti₃C₂ are 2.38, 2.26, 2.21, and 2.19 eV, respectively. It shows that Fe-doping and composite Ti₃C₂ can reduce the bandgap energy of g-C₃N₄, reduce the bandgap width, expand the visible light response range, and improve the visible light utilization.

The photocatalytic activity of the samples was analyzed by fluorescence spectroscopy. The results are shown in Figure 6A. It can be seen from Figure 6A that g-C₃N₄ shows a strong fluorescence emission peak at 466 nm. After Fe-doping and composite Ti₃C₂, the intensity of this peak was inhibited significantly. Confirming that Fe-doping and composite Ti₃C₂ can effectively reduce the electrons and holes recombination probability of g-C₃N₄, and the quantum efficiency is improved.

According to the relevant literature (Li et al., 2008), Fe²⁺/Fe³⁺ has a reduction potential lower than the conduction band potential of g-C₃N₄, so Fe-doping can effectively capture the photogenerated carriers of g-C₃N₄ and inhibit the recombination of electron and hole pairs. When the g-C₃N₄ layer is inserted into the Ti₃C₂ sheet, the two form many intimate interfaces, thus providing the maximum contact surface between the two phases. On such a platform, electrons can be easily transferred from g-C₃N₄ nanosheets to metal Ti₃C₂ sheets with high conductivity (Su et al., 2019), inhibiting the recombination of electron and hole pairs. At the same time, it can be observed from Figure 6A that the fluorescence intensity of Fe-C₃N₄/Ti₃C₂ is the lowest, indicating that there is a synergistic effect between Fe-doping and composite Ti₃C₂ in g-C₃N₄. The recombination rate of the excited electrons and holes in g-C₃N₄ is greatly reduced, the quantum efficiency shows excellent improvement, and the sample may show better photocatalytic activity.

As it is known to all, a higher photocurrent means a higher separation efficiency for the photogenerated charge, which is finally reflected in the better photocatalytic performance. As displayed in Figure 6B, compared with naked FTO, the photocurrent responses of g-C₃N₄, Fe-C₃N₄, and Fe-C₃N₄/Ti₃C₂ samples gradually increase, with average photocurrent densities of 17.3, 24.6, and 41.7 μA/cm². Fe-C₃N₄/Ti₃C₂ shows the highest photocurrent density, which is 2.4-fold higher than g-C₃N₄ and 1.7-fold higher than Fe-C₃N₄. The enhanced photocurrent density of Fe-C₃N₄ and Fe-C₃N₄/Ti₃C₂ electrodes, respectively, shows that Fe doping and Ti₃C₂ composite can promote g-C₃N₄ to provide more carriers to the external circuit, resulting in the greater the photocurrent and the higher the photogenerated charge. The higher the separation efficiency, the better the photocatalytic performance. The results here are mainly due to the following reasons: the reduction potential of Fe²⁺/Fe³⁺ is lower than the conduction band potential of g-C₃N₄, so Fe can effectively capture the photogenerated carriers generated by g-C₃N₄, and the e⁻–h⁺ pair recombination rate is greatly reduced (Hu et al., 2018), and Fe doping weakens the degree of polymerization of g-C₃N₄, increases its surface area, and provides more photocatalytically active sites. The Ti₃C₂ layer is embedded in thin g-C₃N₄ nanosheets, forming a heterojunction and interfacial effect between them. The photogenerated electrons tend to transfer from g-C₃N₄ to Ti₃C₂, which improves the interfacial charge transfer ability of g-C₃N₄, which can effectively inhibit the recombination of e⁻–h⁺ pairs. (Li et al., 2020). Metal Fe-doped and g-C₃N₄ composite Ti₃C₂ act simultaneously, and the photocatalytic performance is better than Fe-C₃N₄.

When the catalyst dosage was 20 mg and the initial concentration of TC is 20 mg/L, the adsorption curve of Fe-C₃N₄/Ti₃C₂ on TC is shown in Figure 7A. In the first 5 min of the dark reaction, the adsorption amount of TC on Fe-C₃N₄/Ti₃C₂ increased rapidly and reached the adsorption-desorption equilibrium after 45 min. The equilibrium adsorption amount was 10.78 mg/g, and the adsorption rate was 21.03%. The adsorption kinetics of TC on Fe-C₃N₄/Ti₃C₂ was investigated by pseudo-first-order and pseudo-second-order adsorption kinetics models. The results are shown in Figure 7A. The graph shows that the correlation coefficient of the pseudo-first-order kinetic equation R ² = 0.9784, and the correlation coefficient of the pseudo-second-order kinetic equation R ² = 0.9987, indicating that the adsorption of TC by Fe-C₃N₄/Ti₃C₂ is a second-order kinetic model, that is, chemical adsorption. The equilibrium adsorption amount q _e = 10.78 mg/g, and the adsorption rate constant k = 0.01852 g/(mgmin) can be obtained from the slope and intercept of the straight line, respectively.

When the catalyst dosage was 20 mg and the initial concentration of TC was 20 mg/L, the photocatalytic degradation curves of different samples are shown in Figure 7B. When no photocatalyst was added, TC was almost not degraded under xenon lamp irradiation, and the degradation rate was only about 1.41%, indicating that TC was relatively stable and difficult to be photodegraded. When only Ti₃C₂ was added, the photocatalytic degradation rate of TC under xenon lamp irradiation was only 4.30%, indicating that Ti₃C₂ cannot be used as a catalyst for photocatalytic degradation of TC alone. After 180 min of illumination, the photocatalytic degradation rates of TC by g-C₃N₄, g-C₃N₄/Ti₃C₂, Fe-C₃N₄, and Fe-C₃N₄/Ti₃C₂ were 44.37, 61.53, 68.01, and 86.80%, respectively. It can be seen that compared with g-C₃N₄, the degradation rate of TC by g-C₃N₄ is improved after Fe-doping and composite Ti₃C₂. The degradation rate of TC by Fe-C₃N₄/Ti₃C₂ within 180 min is 86.80%, 1.96 times that of g-C₃N₄, which greatly improves the photocatalytic degradation efficiency of TC by photocatalyst. It can be speculated that there is a synergistic effect between Fe-doping and composite Ti₃C₂ in g-C₃N₄, and the photocatalytic ability of the sample is better.

Linear fitting with ln ( C 0 / C t ) to t, and the results are shown in Figure 7C. The photocatalytic degradation of TC by Fe-C₃N₄/Ti₃C₂ conforms to the first-order reaction kinetics model, and the relevant parameters are shown in Table 1. The apparent rate constants k of Fe-C₃N₄, g-C₃N₄/Ti₃C₂, and Fe-C₃N₄/Ti₃C₂ are 1.38, 1.60, and 2.83 times of g-C₃N₄, respectively. It shows that Fe-doping and composite Ti₃C₂ are beneficial to improve the photocatalytic degradation activity of g-C₃N₄.

By adding sacrificial agents such as BQ, IPA, and EDTA-2Na to detect the common active radicals •O₂ ⁻, •OH, and h⁺ to explore the degradation mechanism of the photocatalytic reaction system. The results are shown in Figure 7D. When no radical scavengers were in the solution, the degradation rate of TC by Fe-C₃N₄/Ti₃C₂ was 86.80%. When •OH and h⁺ capture agents were added, the photocatalytic degradation of TC by Fe-C₃N₄/Ti₃C₂ was inhibited to a certain extent, and the degradation rates were reduced to 66.30 and 63.81%, respectively. When •O₂ ⁻ capture agent was added, the photocatalytic degradation of TC by Fe-C₃N₄/Ti₃C₂ was greatly inhibited, and the degradation rate decreased to 30.96%. Therefore, the influence of active species on the degradation of TC in the photocatalytic degradation system was as follows: •O₂ ⁻ > h⁺ > •OH.

Based on the above results, we speculated on the photocatalytic degradation mechanism of Fe-C₃N₄/Ti₃C₂ to TC (Figure 8). The electrons of g-C₃N₄ in Fe-C₃N₄/Ti₃C₂ were excited to CB under visible light irradiation. Thus, the photogenerated holes (h⁺) were left in the VB. In the Mott–Schottky test (Figure 6C), we made a tangent along the longest straight line in the figure, and the slope was found to be positive, indicating that g-C₃N₄ was an n-type semiconductor. According to the formula (E_g = E_VB + E_CB) and the DRS results, the theoretical estimation of the E_VB and E_CB of g-C₃N₄ were 1.35 V and −0.84 V vs. NHE. In addition, the Fermi level (E_f, vs. NHE, pH = 0) of Ti₃C₂ was −0.60 V, which was higher than the E_CB of g-C₃N₄ (Su et al., 2019). In Fe-C₃N₄/Ti₃C₂ composites, a close 2D/2D contact interface is formed between Ti₃C₂ and g-C₃N₄. Electrons could be easily transferred from g-C₃N₄ nanosheets to Ti₃C₂ surfaces, which inhibits recombination of photoinduced electrons and holes. Moreover, because the reduction potential of Fe²⁺/Fe³⁺ (Fe²⁺/Fe³⁺ = −0.77 V) was lower than the E_CB of g-C₃N₄ (Li et al., 2008), Fe could effectively capture the photogenerated electrons of g-C₃N₄ and inhibit recombination of electron and hole pairs. Then, the electrons reacted with oxygen adsorbed on the surface of the photocatalyst or dissolved in solution to form strongly oxidized •O₂ ⁻ (O₂/•O₂ ⁻ = −0.33 V). It oxidized TC to CO₂ and H₂O. At the same time, the h⁺ of g-C₃N₄ reacted with H₂O or OH⁻ in the solution to form the •OH (•OH/OH⁻ = 1.99 V) with strong oxidizability. •OH could degrade TC into CO₂ and H₂O. The holes in the valence band of g-C₃N₄ also had strong oxidation which could react with TC and degrade it. The process of Fe-C₃N₄/Ti₃C₂ photocatalytic degradation of TC is as follows.

Comparing the experimental system with other related systems, the results are shown in Table 2. It can be seen from Table 2 that the experimental system has a certain value in practicability and economy. A small amount of Fe-doping and a small amount of Ti₃C₂ composite were used to modify g-C₃N₄, and the prepared Fe-C₃N₄/Ti₃C₂ photocatalyst could be used to degrade TC efficiently with a degradation rate of 86.80%. Compared with the following table, there are certain differences in the degradation rates of different substances in different systems. The main reasons are that the types and ratios of materials, the dosage of photocatalyst, the degradation time, and the power and wavelength of the light source are different.