1 Introduction

Nowadays, environmental pollution becomes a serious and urgent problem challenging the sustainable development of human society. Large quantities of wastewater was always discharged into the aquatic environment without effective treatments and then seriously harmed the environment [1], such as the dye wastewater [2]. Traditional methods of wastewater treatment, such as coagulating sedimentation, filtration and adsorption, have some disadvantages and are ineffective sometimes for removing the dyes [3]. As a new kind of technology, photocatalytic method has been used to treat wastewater and has attracted extensive attention recently [4, 5]. Photocatalyst [6, 7], which is the key of photocatalytic technology, has become a hot topic in the field of materials research. Although a wide variety of photocatalysts have been explored, most of them are not applied in large scale [8]. Up to now, TiO2 is the common used photocatalysis [9, 10] due to the low cost, chemical stability and nontoxicity. Nano-TiO2 grains exhibited excellent photocatalytic performance in degrading organic dyes in wastewater [11]. However, single nano-TiO2 particle was easily agglomerated and then weakened the photocatalytic efficiency. On the other hand, it is difficult to separate and recycle the nano-TiO2 particles from water after the photocatalytic process. Therefore, nano-TiO2 particles were usually dispersed and immobilized on the carriers [12,13,14].

A suitable carrier for TiO2 photocatalyst should possess following characteristics such as good light transmission [15], strong binding force with TiO2 particles without affecting the photocatalytic activity, large specific surface area and strong adsorption capacity for organic contaminants [16]. Due to the low cost, clay minerals were investigated as photocatalysts carrier by many researchers [11,12,13, 17,18,19,20,21,22]. The specific surface area is important for the adsorption capacity of clay minerals and the photocatalytic properties of the clay/TiO2 composites [23, 24]. For all the clay mineral carriers, montmorillonite has excellent absorption capacity as it is a natural nano material. Although kaolinite is rich in reserves, natural kaolinite always has large size, small specific area and contaminated surface. These phenomena lead to not only the low adsorption capacity of kaolinite, but also the low loading ratio of TiO2 in kaolinite/TiO2 composite, as well as the weak bonding force between TiO2 and kaolinite, which are the disadvantages for the photocatalytic performances of kaolinite/TiO2 composites. As a result, many works should be carried out in the future.

Nanoscale kaolinites [25,26,27] with a high specific surface area and better adsorption capacity, can be prepared by intercalation [28] and exfoliation method [29, 30]. On the other hand, the surface of kaolinite was cleaned and activated by acid treatment [31,32,33] in order to improve the reaction activity. However, there is little report about the influence of acid treating or intercalating/exfoliating of natural kaolinite on the photocatalytic activity of kaolinite/TiO2 composite up to now. Consequently, natural kaolin was pretreated by acid treating or intercalating/exfoliating firstly in this study. Then, kaolinite/TiO2 composites were prepared by using sol-gel method and tetrabutyl titanate as the TiO2 precursor. The degradation rates of methylene blue (MB), a commonly used dye, as well as phenol by kaolinite/TiO2 composites were measured for evaluating the effects of pretreatment on the photocatalytic performance of kaolinite/TiO2 composite.

2 Experimental

2.1 Materials

Raw kaolin (RK) used in this study was a natural kaolin collected from a mine in Zhangjiakou, China, with a particle size smaller than 45 μm. As the molar ratio of SiO2/Al2O3 (2.02) of RK was close to the theory value of kaolinite (2.0), it can be inferred that RK contained mainly kaolinite, and a small amount of impurity minerals, such as quartz.

Anhydrous ethanol (C2H5OH), potassium acetate (KAc), glacial acetic acid (HAc), concentrated hydrochloric acid (HCl) and MB (C16H18ClN3S) were analytical reagents, butyl-titanate (Ti(OC4H9)4) was chemically pure, and the deionized water was used in this study.

2.2 Pretreatment of kaolin

A certain amount of RK was added into the 4 mol L–1 HCl solution and stirred magnetically at 25 °C for 3 h, followed by washing with deionized water, centrifuging separation and drying, then obtained the acid-treated kaolin (AK).

Together with an appropriate amount of deionized water, RK and KAc (with a weight ratio of 1:1) were ground in agate mortar to get paste, then aged for 24 h to fulfill the intercalation of KAc molecules into kaolinite. Exfoliation of kaolinite was realized by washing the kaolinte/KAc intercalation compounds with anhydrous alcohol, followed by ultrasonic processing, centrifuging and drying. The as-obtained powder was carried out for other two cycles of intercalating and exfoliating process. Finally, the obtained powder was acid treated according to the method stated above to acquire the exfoliated kaolin (EK).

2.3 Preparation of kaolinite/TiO2 composites

Sol-gel method was performed to synthesize the precursor of TiO2, which composed of Ti(OC4H9)4, C2H5OH, H2O and HAc with a molar ratio of 1:32:2:0.9. In details, 2/3 of C2H5OH was magnetically stirred for 30 min with Ti(OC4H9)4 and HAc, followed by adding the mixture of deionized water and the remaining 1/3 of C2H5OH, adjusting the pH value to 3, then magnetic stirring for 60 min and ultrasonic treating for 10 min. Thus obtained the transparent yellowish sol of TiO2 precursor.

Based on our previous study, the kaolinite/TiO2 composite was designed with a ratio of TiO2 to kaolinite of 5 mmol g–1. A certain concentration of RK suspension was prepared firstly. Then, a proper amount of TiO2 precursor was dropped into the vigorously stirred RK suspension. After stirring for 3 h, the mixture was aged at room temperature for 24 h and then fully dried at 70 °C. Subsequently, the as-acquired solid mixture was calcined at 500 °C for 2 h to obtain the kaolinite/TiO2 composite (RKT).

Using the same method, kaolinite/TiO2 composites can also be prepared by using AK and EK, which were denoted as AKT and EKT, respectively.

2.4 Characterization techniques

The crystalline structure was investigated by an X-ray diffractometer (XD-3, Beijing Purkinje General Instrument Co., Ltd., China) with Cu Kα radiation (λ = 1.5406 Å) and scanning rate of 4° min−1 at 36 kV, 20 mA. The microstructure was observed by a field-emission scanning electron microscope (SEM, Quanta FEG 250, FEI Co., Ltd, Portland, USA). The chemical functional groups were determined by an infrared spectrometer (Nicolet iS10, Thermo Fisher Scientific Corp., Massachusetts, USA) with the scanning range of 4000–400 cm−1 and resolution of 4 cm−1. Ultra violet (UV)-visible spectrum was performed by the use of an UV-visible spectrometer (UV-1800PC, Shanghai United States spectrum of Instrument Co., Ltd, China) with the scanning range of 800–400 cm−1 and resolution of 1 cm−1.

2.5 Measurement of photocatalytic activity

The photocatalytic performance of the composites was evaluated by the degradation rate of MB solution. The light source was a 250 W high pressure mercury lamp whose main wavelength was 365 nm, and the height from the mercury lamp to the reaction level was 15 cm. The entire photocatalytic degradation experiment was carried out at room temperature in a self-made photocatalytic reaction vessel. A certain amount of kaolinite/TiO2 composite powder was added into a 100 mL aqueous solution of MB. Prior to irradiation, the adsorption–desorption equilibrium reaction was achieved firstly in the dark box for 1 h. Then the suspension was irradiated and sampled at regular intervals. The supernatant MB solution not containing the solid particles of the catalyst was packed in a beaker and the residual absorbance curve of the MB solution was measured by UV-1800PC UV-Vis spectrophotometer after 5 ml liquid samples were centrifuged (7000 r min–1, 4 min). The absorbance at 665 nm of the characteristic wavelength of the maximum absorption peak of MB was used to determine the concentration of aqueous MB. According to the initial absorbance of MB solution and the residual absorbance of MB solution after different time, the removal rate of composite materials on MB can be calculated. Using the similar method, phenol was used to evaluate the photocatalytic activity of kaolinite/TiO2 to the colorless organic pollutant. The absorbance at 270 nm of the maximum absorption peak of phenol was used to determine the concentration of phenol aqueous solution.

3 Results and discussion

3.1 Phase analysis

Figure 1 presents the X-ray diffraction (XRD) patterns of RK, AK and EK. It can be seen clearly that RK consisted mainly of kaolinite and a small quantity of quartz (Fig. 1a). AK had the similar XRD pattern (Fig. 1b) as that of RK, indicating that acid treatment did not damage the crystal structure of kaolinite. All the diffraction peaks of kaolinite still appeared on the XRD pattern of EK (Fig. 1c). However, they were broadened significantly and weakened a little compared with those on the patterns of RK and AK, such as (001) and (002). This phenomenon suggests that the crystallinity and grain size of kaolinite were reduced after three cycles of intercalation and exfoliation.

Fig. 1
figure 1

XRD patterns of kaolin and pretreated kaolin. a RK; b AK; c EK

Full size image

Figure 2 shows the XRD patterns of RKT, AKT and EKT. Patterns of all the three samples had typical diffraction peaks of anatase at 2θ of about 25.3, 48.2, 54.2, 55.1 and 62.9°, which are assigned to (101), (200), (105), (211) and (204), respectively. Rietveld refinement confirms the tetragonal structure of anatase in as-obtained kaolinite/TiO2 samples. For example, the unit cell parameters of anatase in RKT are a = 0.343 nm, b = 0.343 nm, c = 2.09 nm, respectively. No peaks of rutile were observed. This result indicates that the TiO2 precursor was only transformed to anatase when calcined at 500 °C. The diffraction peaks of kaolinite still appeared on the patterns of RKT and AKT, but the relative intensity of I001 of kaolinte to I100 of quartz decreased slightly compared with those of RK and AK (Figs. 1 and 2). Obviously, a small amount of kaolinite were damaged but most of them were unaffected when RKT and AKT had been calcined at 500 °C.

Fig. 2
figure 2

XRD patterns of kaolinite/TiO2 composites. a RKT; b AKT; c EKT

Full size image

However, diffraction peaks (001) and (002) of kaolinite almost disappeared on the XRD pattern of EKT, indicating that nearly all kaolinite crystals were damaged when EKT was calcined at 500 °C. This result is ascribed to the dehydroxylation of kaolinite [11]. It can be concluded that the dehydration temperature dropped as the grain size of kaolinite reduced by intercalating and exfoliating, which is coincide with the previous reports [34], and the hydroxyl was removed easily after calcination at 500 °C. Therefore, it can be assumed that hydroxyl of kaolinite was eliminated more easily after exfoliating treatment. Obviously, the dehydration of kaolinite would obtain microporous particles, which was helpful to the absorption property.

3.2 Microstructural analysis

Figure 3 exhibits SEM images of the pretreated kaolinite and the as-prepared kaolinite/TiO2 composite samples. Kaolinite particles in RK presented a typical pseudo-hexagonal shape, lamellar structure and a grain size smaller than 3 μm (Fig. 3a). Kaolinite crystals in AK had the same shape and size as those in RK (Fig. 3b). Kaolinite in EK possessed the same shape as those in RK (Fig. 3c). However, the grain size and thickness of kaolinite particles decreased significantly compared with those of RK and AK, in line with the XRD analysis. This result is also ascribed to the three cycles of intercalation and exfoliation treatment of kaolinite.

Fig. 3
figure 3

SEM images of raw kaolinite, pretreated kaolinite and kaolinite/TiO2 composites. a RK; b AK; c EK; d RKT; e AKT; f EKT

Full size image

All the three samples RKT, AKT and EKT were loaded with TiO2 particles of 10–30 nm in size (Figs. 3d-f), which was also confirmed by the energy dispersive spectrometer (EDS) and XRD analysis (Fig. 2). For RKT, TiO2 grains were mostly distributed at the corner or edge of kaolinite crystals, but seldom loaded on the surface. This phenomenon was similar to the report of Kutláková et al. [11], indicating that the smooth surface of kaolinite was not suitable for the growth of TiO2 grains.

AK was loaded with larger amount of TiO2 grains compared with RK, and TiO2 grains were uniformly distributed on the kaolinite surface. The main reason for this phenomenon may be due to the removing of impurities and the introducing of hydroxyl on the kaolinite surface by the acid treatment. Same phenomenon can also be observed in the SEM images of EKT. In addition, it should be noted that the EKT particles kept the same morphology as RKT and AKT, although the kaolinite crystals in EKT had undergone dehydroxylation.

3.3 Infrared spectroscopy analysis

The Fourier Transform Infrared (FT-IR) spectra of kaolin and kaolinite/TiO2 composite samples are given in Fig. 4. RK presented all the characteristic vibration bands of kaolinite (Fig. 4a). The stretching vibrations of inner-surface hydroxyls and inner hydroxyls were located at 3695, 3667, 3652 and 3620 cm−1, respectively. Their bending vibrations appeared at 938 and 914 cm−1, correspondingly [27]. The bands at 1115, 1030, 1010, 793 and 471 cm−1 are assigned to the vibrations of Si-O bonds, and bands at 754, 695 and 540 cm−1 are attributed to vibrations of Si-O-Al bonds [11]. The band at 3453 cm−1 is ascribed to the stretching vibration of water molecules adsorbed on the external surface of kaolinite [35, 36]. AK and EK had the similar FT-IR spectra as RK did, except for the weakening of bands at 3667 and 3652 cm−1 on the spectrum of EK.

Fig. 4
figure 4

Infrared spectra of kaolinite and kaolinite/TiO2 composites. a RK; b AK; c EK; d RKT; e AKT; f EKT

Full size image

For the FT-IR spectra of RKT, AKT and EKT, the broad band at about 1060 cm−1 and the bands in the regions of 660–550 cm−1, 470–436 cm−1 are ascribed to the vibrations of Ti-O-H, Ti-O and Ti-O-Ti bonds, respectively, indicating that these composite samples did contain TiO2 grains [11, 37]. It should be noted that the transmittance of bands at 3620–3695 cm−1 reduced significantly for RKT and AKT, and even disappeared for EKT. These phenomena coincides with the result of XRD analysis that some kaolinite in RKT and AKT were dehydrated and nearly all the kaolinite in EKT were dehydrated.

In view of the analysis of XRD, SEM and FT-IR, it can be inferred that TiO2 precursor was coated on the surface of kaolinite by the sol-gel method. Acid treating of kaolinite make the TiO2 precursor cover more uniformly and attach more tightly on the surface of kaolinite crystals. When calcined at 500 °C, the TiO2 precursor was converted to anatase and kaolinite crystals were kept undamaged for the RKT and AKT, but damaged for the EKT sample.

3.4 Photocatalytic activity of kaolinte/TiO2 composites

3.4.1 Effect of photocatalyst dosage on the photocatalytic performance

Figure 5 exhibits the degradation curves of MB aqueous solution (10 mg L–1) by the different dosage of kaolinte/TiO2 composite. The degradation rate of MB increased gradually with the extension of photocatalytic reaction time and the enhancement of photocatalyst dosage. For example, when the RKT dosage was 2 g L–1, the degradation rate of MB reached 73.97 and 89.26% after photocatalysis for 50 and 90 min, respectively. When photocatalysis for 90 min, the degradation rate increased from 82.56% through 89.26 to 96.06% with the dosage rising from 1 to 2 g L–1, 3 g L–1, correspondingly.

Fig. 5
figure 5

Degradation curves of MB solution by different dosages of kaolinite/TiO2 composites. a RKT; b AKT; c EKT

Full size image

It should be pointed out that the degradation rate of MB during the dark reaction period was indeed derived from the absorption of photocatalyst. The adsorption performance of AKT was nearly the same as that of RKT, but EKT showed the most excellent adsorption performance among the three samples. For instance, the degradation rate during dark reaction period achieved 43.61, 42.84 and 72.93% by inputting 2 g L–1 of RKT, AKT and EKT, respectively. This result is mainly ascribed to the reduced particle size of kaolinite in EK by the intercalating and exfoliating treatment, as well as the dehydration of kaolinite in EKT during the calcination process.

The photocatalytic performance of AKT and EKT were better than that of RKT, and AKT exhibited the optimum property. With the photocatalyst dosage of 2 g L–1 and photocatalysis for 60 min, the degradation rates of MB by RKT, AKT and EKT achieved 77.71%, 93.94% and 91.68%, respectively. In addition, for AKT and EKT, the degradation rate of MB by the dosage of 2 g L–1 approached to that of 3 g L–1 as the photocatalysis time rising up to 90 min. As a result, AKT was the optimum photocatalyst and 2 g L–1 would be the optimum dosage.

3.4.2 Effect of initial concentration of MB solution on the photocatalytic performance

Figure 6 shows the degradation curves of MB solutions with different initial concentration by inputting 2 g L–1 kaolinite/TiO2 composites. Under a certain photocatalyst dosage and photocatalysis time, the absorption rate and the degradation rate decreased gradually with the enhancement of initial concentration. The degradation rate of MB increased with the prolonging of photocatalysis time. No matter what the initial MB concentration was, AKT and EKT exhibited better photocatalytic activity than RKT did. It should be noted that AKT achieved the high degradation rate of 94.05% for MB solution with initial concentration of 15 mg L–1 when photocatalysis for 90 min, which was nearly similar to that for MB solution with initial concentration of 10 mg L–1. Consequently, it can be inferred that acid treatment to RK was the best and economic pretreatment method to prepare high performance kaolinite/TiO2 composite photocatalyst. Compared with other kaolinite/TiO2 composite materials [38], the AKT and EKT exhibited the better photocatalytic performance, especially the adsorption performance of the EKT sample is particularly excellent.

Fig. 6
figure 6

Degradation curves of MB solution with different initial concentration by kaolinite/TiO2 composites. a RKT; b AKT; c EKT

Full size image

In order to eliminate the effect of dye on the absorption of UV and visible light, phenol, a colorless organic pollutant, was used to evaluate photocatalytic efficiency of kaolinite/TiO2 composite. Phenol aqueous solution (10 mg L–1, 100 mL) and dosage of 200 mg photocatalyst were used in the photocatalytic experiment. Similar to the result of degradation for MB, the adsorption performance of AKT for phenol was nearly the same as that of RKT, and EKT exhibited the most excellent adsorption performance (Fig. 7). The degradation rate of phenol increased gradually with the extension of photocatalytic reaction time. When photocatalysis for 60 min, the degradation rate of phenol by RKT, AKT and EKT reached 23.5%, 37.5% and 28.5%, respectively. Obviously, these degradation rates are lower than those of MB. This result indicates that the degradation efficiency of kaolinite/TiO2 composite on phenol was lower than that on MB.

Fig. 7
figure 7

Degradation curves of phenol solution by kaolinite/TiO2 composites. a RKT; b AKT; c EKT

Full size image

4 Conclusion

The kaolinite/TiO2 composite photocatalysts were prepared successfully by sol-gel method and using RK and pretreated kaolin as the carrier and tetrabutyl titanate as TiO2 precursors, respectively. The size and thickness of EK particles decreased significantly compared with those of RK and AK. Acid treatment improved the distribution and the loading quantity of TiO2 grains. When the kaolinite/TiO2 composites were calcined at 500 °C, the tetragonal structure of anatase particles size of 30–100 nm in size were obtained, but the exfoliated kaolinite crystals were damaged. The degradation rate of MB increased with the enhancement of photocatalytic reaction time and photocatalyst dosage, but decreased with the rising of initial MB concentration. The photocatalytic performances of AKT and EKT were significantly better than RKT samples, and AKT exhibited the optimum property. Similar result was also acquired for the degradation of phenol. Both the acid treating and the exfoliating to kaolinite enhanced the photocatalytic performance of the kaolinite/TiO2 composite photocatalysts, but acid treatment may be more helpful to the preparation of high performance kaolinite/TiO2 composite photocatalyst.