2.1.5. N2 Adsorption-Desorption Isotherms

Figure 5 shows the N2 adsorption-desorption isotherms and pore size distribution. According to the classification of IUPAC, the isotherms of all materials are regarded as type IV, corresponding to the mesoporous structure [30]. At the initial stage, N2 adsorption capacities are greater than 0, indicating abundant micropores in the samples. Furthermore, the hysteresis loops of these samples belong to type H3, which represent that the samples are lamellar particle materials with fissure structures [31].

**Figure 5.** N2 adsorption-desorption isotherms and pore size distribution: (**a**) isotherms and (**b**) pore size.

The pore structure parameters including specific surface area (SBET), pore volume, pore size, and crystal size are summarized in Table 1. Compared with hectorite, the surface area, pore volume, and diameter of all composite samples increase, which would be beneficial to the adsorption and degradation of organic dyes. Among them, the TH-2 sample has the largest specific surface area of 491.97 m2/g, which is about twice that of hectorite and ten times that of commercial P25. The average pore size of TH-1~5 is larger than that of hectorite (3.16 nm), which proves that TiO2 is successfully introduced into the interlayer of hectorite. At the same time, when the molar ratio of Li<sup>+</sup> to Mg2+ is 1.32:5.34, most ions exchange between titanium ions and interlayer water molecules, Li+ is observed, and the interlayer spacing is largest. A higher specific surface area is beneficial to adsorption performance because it can provide more active sites, which can increase the contact area between organic pollutants and catalysts, thus increasing the photodegradation rate and improving the pollutant removal rate effectively ultimately.

Figure 5b is the pore size distribution curves calculated from Figure 5a. It can be observed that the pore size distribution curves of these samples are similar and the pore size is concentrated between 2 and 5 nm, which indicates again that hectorite and TiO2/hectorite belong to the mesoporous structure. TH-2 displays a larger pore size distribution and the largest pore size, indicating that titanium ions have more chances to intercalate in the interlayer of hectorite under the condition. The pore size of TH-5 is larger than that of TH-4, which could be because of the increase in TiO2 particles and their inhomogeneity [15].

2.1.6. Ultraviolet-Visible Diffuse Reflectance Spectra (UV-Vis DRS)

Figure 6 gives the UV-Vis DRS, and the optical absorption capacity of hectorite is weak in the wavelength range of 200–800 nm, which is probably attributed to its composition and particle size [10]. All the composite photocatalysts present an intense optical absorption under 425 nm compared with the single hectorite, indicating the electron Ti–O transformation of TiO2 after the intercalation of titanium cations [32]. The insertion of TiO2 inside the layers of hectorite may modify the electronic structure and thus the band gap energy, promoting a shift in the absorption spectra [15,32]. The band gap values were obtained with Equation (2):

$$(ahv)^{1/2} = A(hv - E\_{\mathbb{S}}),\tag{2}$$

where *α*, *hν*, *A,* and *Eg* are the optical absorption coefficient, photon energy, A constant, and band gap, respectively [33].

**Figure 6.** UV-Vis diffuse reflectance spectra and band gaps: (**a)** reflectance spectra and (**b**) band gaps.

The values of the band gap energies calculated from Equation (2) (Figure 6b) of hectorite, TH-1, TH-2, TH-3, TH-4, and TH-5, are 3.33, 3.17, 3.06, 3.11, 3.14 and 3.22 eV, respectively. Except for TH-5, the band gaps of other composite photocatalysts are smaller than that of pure TiO2 (3.2 eV). The TH-2 sample exhibits the strongest UV absorption capacity, thus providing a good opportunity to broaden the absorption band and improve the photocatalytic performance.

### 2.1.7. X-ray Photoelectron Spectroscopy (XPS)

The elemental states and surface components of the TH-2 sample were analyzed by XPS and are given in Figure 7. It mainly contains Ti, Si, Mg, F, C, and O, while the Li element in hectorite is not observed in TH-2 (Figure 7a). The phenomenon indicates that the ion exchange reaction has occurred in which Li ions are likely to be replaced by hydrogen ions or hydrate titanium ions during the reaction [10]. In addition, the leaching of Mg from the magnesia octahedron has a great influence on the skeleton structure.

**Figure 7.** XPS spectra of TH-2 composite sample: (**a**) survey spectrum, (**b**) F 1s, (**c**) O 1s, and (**d**) Ti 2p.

The presence of a peak of F 1s for TH-2 with a binding energy at 685.2 eV is shown in Figure 7b, which is associated with the physical surface adsorption of F [34]. The UV-Vis DRS of TH-2 exhibits intense absorption in the UV region without an obvious redshift, which is consistent with the previous reports of F-doped TiO2 [35,36]. In addition, the doping of the F element may contribute to the formation of oxygen holes.

According to the spectrum of O 1s for TH-2 (Figure 7c), the peak at 530.1 eV demonstrates the existence of crystal lattice oxygen O2 − and the peak at 532.1 eV is mainly related to the Si–O bond from silicon lattice [32,37]. The surface adsorption of F (≡Ti–F) and surface –OH can proceed as follows (Equation (3)) [34], while it is difficult to see the peak of –OH as F occupies more space on the surface of the TiO2/hectorite.

$$\mathbf{H} \equiv \mathbf{\overline{T}} - \mathbf{O} \mathbf{H} + \mathbf{F}^- \leftrightarrow \equiv \mathbf{\overline{T}} - \mathbf{F} + \mathbf{O} \mathbf{H}^- \mathbf{p} \mathbf{K}\_\mathbf{F} = 6.2 \tag{3}$$

Two strong peaks of Ti 2p at 458.8 eV and 464.6 eV are attributed to the Ti 2p3/2 and Ti 2p1/2 (Ti–O bond), respectively, indicating the existence of the Ti4+ chemical state in the TH-2 photocatalyst [32]. Two peaks at 457.9 eV and 463.6 eV correspond to the Ti3+ 2p3/2 and Ti3+ 2p1/2, respectively, which confirms the formation of Ti3+ [38,39]. As is well understood, Ti3+ is generally considered to be beneficial for improving photocatalytic activity.

The electron cloud density decreases with the binding energy [40,41]. Compared with pure TiO2 (458.4 and 464.1 eV), the binding energy of Ti 2p3/2 (458.8 eV) and Ti 2p1/2 (464.6 eV) of TH-2 is much higher, which should be the formation of the Ti–O–Si bond between TiO2 and hectorite [35,41,42]. The result indicates that the hectorite helps to promote the separation of e<sup>−</sup>-h+ pairs.

### *2.2. Photocatalytic Study*

The photocatalytic performance of composite samples was studied in the degradation of MB with a 125 W high-pressure mercury lamp. The blank control group and P25 were used as comparisons. At low concentration, the photocatalytic degradation process conforms to the first-order kinetic equation, which can be fitted by the Langmuir-Hinshelword (L-H) model, as shown by Equation (4) [43,44].

$$\ln\left(\frac{\mathcal{C}\_0}{\mathcal{C}}\right) = kt\_\prime \tag{4}$$

where *C*<sup>0</sup> is the adsorption and desorption equilibrium concentration of MB, *C* is the concentration of MB at time *t*, and *k* is the photocatalytic kinetic constant. *k* can be used to evaluate the photocatalytic performance, and the higher the value, the higher the catalytic efficiency.

The photocatalytic performance of TH-1, TH-2, TH-3, TH-4, TH-5, and P25 was evaluated for MB degradation under UV light irradiation in Figure 8. On the one hand, Figure 8a shows the degradation curves at 10 ppm of MB for the samples TH-1, TH-2, TH-3, TH-4, TH-5, and P25, where their corresponding removal rates are 57.5%, 97.8%, 94.5%, 80.1%, 78%, and 60.7%, respectively. Without catalysis, the removal rate of MB is only 8.2%. The result of the photolysis does not show a significant reduction in MB, indicating that the irradiation with UV light by itself is not capable of degrading the dye. The composition of TiO2 and hectorite clearly enhances the photocatalytic activity. Specifically, TH-2 and TH-3 show excellent photocatalytic activity, and photocatalysts are able to degrade 97.8% and 94.5% of MB within 60 min, while P25 and TH-1 only degrade MB by 60.7% and 57.5%, respectively. Although TH-1 has a higher surface area value than P25, its degradation is lower than that achieved by P25. The lower value from TH-1 can be attributed to the excess of defects present in the TH-1 sample, and P25 probably has a higher recombination of e−/h<sup>+</sup> pairs by affecting the photodegradation. Therefore, its degradation is lower than those of TH-2, TH-3, TH-4, and TH-5. In particular, P25 (composed of ~30% rutile and 70% anatase) presents a better performance than TH-1, attributed to the transfer of electrons from Cb of anatase to those of rutile TiO2 [32].

**Figure 8.** (**a**) Photocatalytic degradation of MB under UV light; (**b**) the first-order kinetic fitting curve of the photocatalytic MB degradation.

On the other hand, the kinetic constants (*k*), derived from the first-order kinetic fitting curve (Figure 8b), for the photodegradation of MB from highest to lowest, are shown in the order of TH-2 (0.04361 min−1), TH-3 (0.03148 min−1), TH-4 (0.0245 min−1), TH-5 (0.02121 min<sup>−</sup>1), P25 (0.01396 min−1), and TH-1 (0.00898 min−1) catalysts. The apparent rate constant of TH-2 is about 3.12 times that of P25. Based on these results, the removal rates of MB are increased by adding a quantity of TiO2 in hectorite, and the highest degradation rate is achieved when the molar ratio of lithium, magnesium, and silicon is 1.32:5.34:8 (TH-2). At the same time, the combination of TiO2 and hectorite is beneficial to the generation of oxygen vacancies. These oxygen vacancies can trap the photogenerated species and prolong their life, which improves the photoactivity. However, TH-1 could produce a defect excess and the formation of polycrystals by increasing the recombination of e−/h+ pairs, affecting the degradation of MB.

Based on the above result, the photocatalytic mechanism of TH-2 is given in Figure 9. First, TiO2 in the interlayer of hectorite can increase the specific surface area to increase the adsorption capacity and contact of MB and the photocatalyst. Secondly, TiO2 is uniformly dispersed on the hectorite to provide more active sites for the reaction. Thirdly, it is the most important that during the photocatalytic process, the electrons are stimulated from the valence band to the oxygen vacancy and Ti3+ in TiO2 under UV light irradiation, and the oxygen vacancy defect is beneficial to the adsorption of O2, while the positive charge holes are left in the conduction band. At the same time, the negatively charged interlayer surface of hectorite is conducive to improving the photocatalytic activity, and the photogenerated holes can migrate to the surface quickly under the electronic attraction on the negatively charged interlayer surface.

In addition, photocatalytic activity is closely related to the behavior of photocarriers [45,46]. Subsequently, electrons (e−) gathered on the surface of TiO2 react with dissolved oxygen molecules in water to generate the superoxide radical anion ·O2 − and other high oxidation groups. As the conduction band has a more positive potential than that of ·OH/H2O, the holes in the valence band react with water molecules (or surface hydroxyl groups) adsorbed on the surface of TiO2 to form ·OH [34]. Furthermore, due to the adsorption of F on the surface of TiO2, ·OHfree is generated in the bulk solution. Finally, MB is oxidized into decomposed products by these active substances ·O2 <sup>−</sup> and ·OH.

**Figure 9.** Schematic illustration of a photocatalytic mechanism for the degradation process of MB under UV light toward TH-2 composite.
