The preparation of TiO₂ composite materials modified with Ce and tourmaline and the study of their photocatalytic activity

Abstract

Rare earth cerium (Ce), borosilicate mineral tourmaline and titanium dioxide (TiO₂) composite photocatalysts were prepared via a sol–gel method. Through detecting the microstructure and the methyl orange (MO) photodegradation of composite photocatalysts with different preparation processes and component contents, this paper discusses the effects of tourmaline and Ce on the preparation and photocatalytic activity of the composite catalysts. Among the different composite photocatalysts, the sample with the following addition sequence and proportions possessed the highest MO degradation ratio in 3 h (94.6%): 0.10 wt% of Ce was added into the primary alkoxide solution and 0.40 wt% of tourmaline added to the titanium sol, the synergistic effects on the microstructure of the photocatalyst resulted in nanoparticles wrapped around fine tourmaline particles during preparation and a high activity to photodegrade organic pollutants in water.

---

1 Introduction

Due to the advantages of inexpensive preparation, innocuity, high efficiency and strong oxidizing power, titanium dioxide (TiO₂) photocatalysts have shown broad application prospects in the field of photodegradation of organics for pollutant treatment or water decontamination with antibacterial or environmental benefits, and TiO₂ has attracted much attention in the last decade as it is considered a green and most promising environmental protection material.^1,2 In terms of TiO₂ preparation, various chemical and physical treatment techniques have been investigated, such as sol–gel methods, hydrothermal methods, chemical vapor deposition methods and so on.^3–5 Despite its fascinating properties, TiO₂ has a wide band gap of 3.2 eV and a high recombination rate of the photogenerated electron–hole pairs. Thus, the low solar energy utilization rate (the ultraviolet portion with a wavelength of less than 387.5 nm) and quantum efficiency has resulted in the limiting of the industrial applications of TiO₂ when prepared by traditional methods.³ Therefore, TiO₂ must be modified to narrow down the band gap and hinder the recombination of photogenerated electron–hole pairs, to enhance the light availability and photocatalytic activity.^6,7 Modifying TiO₂ with metal nanoparticles is effective to obtain a more efficient photocatalytic activity. Zhang et al.⁸ have modified TiO₂ with four kinds of noble metal (Pt, Rh, Pd and Au) synthesis composite materials. By observing the effect on the photocatalytic degradation of formaldehyde, the author found the best loading was 1% Pt/TiO₂. However, noble metals are too expensive to apply practically. Moreover, some workers have employed effective non-precious transition metals to modify TiO₂, which could make the band gap narrow and produce more photogenerated electron–hole pairs, to enhance the redox ability of TiO₂.^9–12 Recently, it has been reported that the incorporation of lanthanide ions or oxides into TiO₂ results in highly efficient photocatalytic activity due to their special f and d electron orbital structure.^13–15 Among these ions, cerium (Ce) produced a much more effective photocatalytic activity when being used to dope TiO₂.^16–20

In addition, researchers have utilized the electrostatic poles of tourmaline crystals to improve TiO₂ photocatalysis in recent years.^21–25 Tourmaline is a complex borosilicate mineral, belonging to the trigonal space group, and has spontaneous and permanent polarity, which can produce an electric dipole. The general chemical formula of tourmaline can be written as XY₃Z₆Si₆O₁₈(BO₃)₃W₄ (where X is Na⁺, Ca²⁺, K⁺ or vacancies; Y is Mg²⁺, Mn²⁺, Fe²⁺, Al³⁺, Fe³⁺, Mn³⁺ or Li⁺; Z is Al³⁺, Fe³⁺, Cr³⁺ or Mg²⁺; and W is OH⁻, F⁻, or O²⁻).^26,27 As a heteropolar mineral, the three fold symmetry axis of tourmaline is the C axis, and there is neither an axis nor center of symmetry perpendicular to the C axis. The unique elements and structure mean that tourmaline features many important properties, such as piezoelectricity, pyroelectricity, irradiating in the far-infrared, and a strong electric field that exists on the surface of a tourmaline granule, similar to an electric dipole, especially in a small granule with a diameter of microns or less. Liang et al.²¹ prepared tourmaline/TiO₂ composite films using a sol–gel method, modifying TiO₂ by forming different microstructures through adding tourmaline and changing the preparation method, and it was found that the photocatalytic activity of TiO₂ was effectively enhanced.

In view of the strong electric field and strong far-infrared radiation of natural tourmaline, and the unique characteristics of the rare earth elements, we propose the application of sol–gel technology to prepare nanometer TiO₂ composite materials with traces of rare earth Ce and tourmaline. Though evaluating the photocatalytic degradation of methyl orange (MO), the aim of our study is to explore the synergistic-effect of Ce and tourmaline on the TiO₂ composite materials, including the impact of the addition sequence and the content of Ce and tourmaline on the microstructure of the TiO₂ composite photocatalysts, and the ultimate enhancement of their photocatalytic activity. It is expected to provide a green and promising material to improve organic pollutant degradation for protecting the environment.

2 Experimental

2.1 Photocatalysts

TiO₂ and the different TiO₂ composite photocatalysts with Ce or tourmaline were prepared as follows (the preparation process and component proportions for samples A–F are listed in Table 1).

Table 1 The process parameters and component proportion of the different photocatalyst samples^a

Sample	Amount of components			Preparation process	Component proportion of the ultimate composite (wt%)
	Ti(OC4H9)4 (g)	Ce(NO3)3·6H2O (g)	Tourmaline (g)		TiO2	Ce	Tourmaline
a (a) The primary alkoxide solution (Ti(OC4H9)4 : C2H5OH = 1 : 5, mass ratio). (b) HCl and CH3COOH solution. (c) The cerous nitrate solution with ethanol. (t) The tourmaline suspension with ethanol.
A	11.000	—	—	(a) + (b)	100	—	—
B	10.623	0.008	—	(a) + (c) + (b)	99.9	0.1	—
C	10.594	—	0.010	(a) + (b) + (t)	99.6	—	0.4
D	10.583	0.008	0.010	(a) + (c) + (t) + (b)	99.5	0.1	0.4
E	10.583	0.008	0.010	(a) + (b) + (c) + (t)	99.5	0.1	0.4
F	10.583	0.008	0.010	(a) + (c) + (b) + (t)	99.5	0.1	0.4

---

As a control sample, the primary alkoxide solution TiO₂ sol was prepared from Ti(OC₄H₉)₄ dissolved in C₂H₅OH (1 : 5 mass ratio) and marked as (a). HCl, CH₃COOH and deionized water (16.180 g, 7.4 mL and 1.3 mL, respectively, marked as (b)) were mixed with this solution (pH = 2.0). After stirring for 3 h, the resultant alkoxide solution was kept at room temperature for 2 h to allow hydrolysis reactions.

For the different TiO₂ composite photocatalysts, the blending aqueous solution, composed of cerous nitrate (0.008 g of Ce(NO₃)₃·6H₂O) and ethanol (65 mL), was marked as (c); the tourmaline suspension in ethanol (0.010 g of tourmaline, 65 mL of ethanol), was marked as (t). Either the (c) solution, (t) solution, or both solutions were added dropwise following the different sequences (preparation process in Table 1), and stirred for 3 h. The tourmaline particles used here were black tourmaline from the Hebei Province, China. The particle size of the black tourmaline (D₅₀) was 0.40 μm. The main chemical composition of the tourmaline was as follows (wt%, dry mass): Al₂O₃ 35.98; B₂O₃ 10.94; K₂O 0.04; Na₂O 0.91; MgO 0.20; SiO₂ 34.60; Fe₂O₃ 15.80; CaO trace. In the experiments, Ti(OC₄H₉)₄, CH₃COOH, HCl and other chemicals used were analytical reagents.

Stainless steel meshes were used as substrates for the thin films. Before preparation, the stainless steel meshes (100 mm × 100 mm) were subjected to pre-treatment, which was as follows: the stainless steel meshes were washed using acid and rinsed with deionized water and ethanol, respectively. Finally, the meshes were well prepared after drying. The photocatalyst films were prepared with the above sol solutions by dipping–withdrawing on the stainless steel meshes in ambient atmosphere. The meshes were firstly dipped into the sol for 10 min. The withdrawal speed was 50 mm min⁻¹. The substrates coated with gel films were dried naturally for 1 h, and subsequently heat-treated at 600 °C for 2 h. All the photocatalyst film samples were thus prepared and the ultimate component proportion of different photocatalyst samples was calculated, as shown in Table 1.

2.2 Characterization

200 mL deionized water was injected into a plastic bottle, which was wrapped with a 4 mm-thick circle layer of tourmaline particles outside. The electrical conductivity and surface tension of the deionized water were observed three times at intervals of 30 min using a digital conductivity meter (DDS-11AT, China) and an automatic tensiometer (JK99B, China), respectively.

The different sample A–F films on stainless steel meshes were scraped, and these sample powders were collected. For the crystal structures of the photocatalyst samples, the powders of samples A–F were characterized via X-ray diffraction (XRD, Philips X’Pert MPD, Holland) with Cu Kα radiation (λ = 0.1540 nm).

The composition of the different photocatalyst powders was determined using energy dispersive spectroscopy (EDS) analysis, and the microstructures of the different sample films were observed by scanning electron microscopy (SEM, Hitachi, S-4800, Japan) and transmission electron microscopy (TEM, JEOL JEM-100CXII, Japan). The Brunauer–Emmett–Teller (BET) surface area was determined by nitrogen adsorption measurements at 77 K (Quantachrome, Autosorb-iQ, USA).

The absorption edges of the different samples were measured by ultraviolet (UV)-vis diffuse reflectance spectra (UV-vis Spectrophotometer, Hitachi, U-3900H, Japan) and BaSO₄ was used as a reflectance standard.

The photocatalytic activity of the different film samples was evaluated using MO degradation. An UV lamp (25 W, 253.7 nm) was used as a light source, with a 40 cm distance to the bottom of the reactor, and the average intensity of the UV irradiance was 49 mW cm⁻². For MO photodegradation, sample films (100 mm × 100 mm) were put into the bottom of the reactor, with a pure quartz glass seal on the top. 100 mL of an aqueous solution of MO (20 mg L⁻¹) was dropped into the reactor, the effective depth of the reaction liquid was about 9.6 mm. The changes of MO concentration were detected using an UV-vis spectrophotometer (UV-vis, TU-1810PC, China) at intervals of 20 min. During MO photodegradation, the pH values of every sample reaction were measured with a pH meter (Leici, PHS-3E, China) at 0 h, 1 h, 2 h and 3 h. The optical images of MO photodegradation by samples A–F were also taken at different times (0 h and 3 h).

For investigating the generation of OH radicals by photocatalysts, a stock terephthalic acid solution (with 5 × 10⁻⁴ M terephthalic acid and 2 × 10⁻³ M NaOH) was prepared as described in the literature.^28–30 Samples A–F were each added into the stock solution, respectively. The suspensions were irradiated by the UV lamp for 30 min, collected and centrifugated for PL measurements. Fluorescence spectra were measured on a fluorescence spectrophotometer (Hitachi, F-4500, Japan) with an excitation wavelength of 320 nm.

For the discussion of the effect on the photocatalytic activity of the TiO₂ composite film, the different component proportions of tourmaline and Ce in sample F were studied.

3 Results and discussion

3.1 The electrical conductivity and surface tension of deionized water with tourmaline

Table 2 shows the differences of the electrical conductivity and surface tension of deionized water wrapped with or without tourmaline. The electrical conductivity of deionized water stayed around 1.58 μs cm⁻¹ (without tourmaline), and increased to 2.07 μs cm⁻¹ after being wrapped in tourmaline for 120 min. The surface tension of deionized water was about 72 mN m⁻¹ (without tourmaline), and was reduced to 65 mN m⁻¹, approximately (with tourmaline).

Table 2 The electrical conductivity and surface tension of deionized water as effected by tourmaline

Time (min)	Electrical conductivity of deionized water (μs cm^−1)		Surface tension of deionized water (mN m^−1)
	With tourmaline	Without tourmaline	With tourmaline	Without tourmaline
0	1.56 ± 0.01	1.56 ± 0.01	71.69 ± 0.06	71.69 ± 0.06
30	1.81 ± 0.01	1.58 ± 0.02	64.53 ± 0.06	72.35 ± 0.04
60	1.94 ± 0.02	1.59 ± 0.03	64.87 ± 0.03	71.77 ± 0.05
90	2.02 ± 0.03	1.57 ± 0.02	65.25 ± 0.02	72.14 ± 0.03
120	2.07 ± 0.02	1.58 ± 0.02	64.54 ± 0.04	71.84 ± 0.02

---

The tourmaline particles were wrapped around the outside of the plastic bottle of deionized water, therefore the electric field of tourmaline would not affect the electrical conductivity and surface tension of deionized water. However, the strong far-infrared radiation emitted by the tourmaline particles could penetrate through the plastic bottle and provide energy to enhance molecular vibration and diminish clusters of water molecules. The strong far-infrared radiation of tourmaline is due to the high amount of vibration in its crystal structure, including multiple infrared active vibrating bonds, such as the stretching and bending vibrations of the Si–O–Si bond in the [SiO₄] tetrahedron, the O–H bond in hydroxyl and the other bonds of the metal ions with oxygen. The unique elements in the structure of tourmaline cause its high emissivity in the far-infrared, reaching to 0.90, in the wavelength range of 4–14 μm.²¹ Tourmaline can emit far-infrared radiation, which activates the intramolecular and intermolecular vibrations in water to diminish the clusters of molecules and promote the ionization of H₂O, therefore the intermolecular forces in water is reduced.^26,27 Thus the higher electrical conductivity and lower surface tension were detected.

3.2 XRD characterization

XRD characterization was used to investigate the phase structure and phase composition of the A–F photocatalyst samples (Fig. 1). The data indicated that the pure TiO₂ particles of the control sample (sample A) exhibited well crystallized phases of anatase (JCPDS no. 21-1272) and rutile (JCPDS no. 21-1276). The sample B photocatalyst also had both anatase and rutile phases, wherein the rutile phase was the major phase. The sample D photocatalyst had a larger proportion of the anatase phase than sample A and B. The sample C, E and F photocatalysts each had anatase as the main phase. However, no tourmaline structure (2θ values = 22.28, 30.28, and 34.78, JCPDS. no. 43-1464) was found because the amount was below the detectable limit. From Fig. 1, it was also noteworthy that no peaks corresponding to CeO₂ were observed in any of the doped TiO₂ samples (Sample B, D, E and F). According to the literature,^18,19,31 one reason was that the concentration of the dopant Ce was too low to be detected via XRD. The other was that as the ionic radius of Ce⁴⁺ (0.092 nm) was much larger than that of Ti⁴⁺ (0.065 nm), it was hard to achieve the substitution of Ce⁴⁺ for Ti⁴⁺ in the composite photocatalysts. By comparison, there was no peak shift phenomenon or change of the lattice constants between the B–F samples and A, indicating that the Ce (0.102 nm) elements were not incorporated into the structure of TiO₂. Therefore, Ce existed in the form of uniformly dispersed small cluster CeO₂ on the surface or in the interstitial sites of TiO₂.

Fig. 1 XRD patterns of photocatalyst samples A–F (the details of every sample are listed in Table 1).

The 2θ values of 25.38, 37.88, and 48.18 corresponded to the (101), (004), and (200) crystal planes of the anatase TiO₂ phase, respectively. With regard to samples C–F, the peak widths were broadened and indicated smaller crystallite sizes compared with sample A. Because the component tourmaline has spontaneous polarization properties, it could be seen as a pair of electric dipoles: the two ends of the tourmaline particles equate to the positive and negative poles of the electric field (as shown in Scheme 1). During the preparation of the C, E and F photocatalysts, the monomers formed sol particles through hydrolysis and condensation reactions. The positively charged ion particles [Ti(OR)₄(H₃O)_nⁿ⁺] in the sol were attracted by the negative electrode of the tourmaline particles, and Cl⁻ was attracted via the positive electrode. Furthermore, the electric field and strong far-infrared radiation²¹ of the tourmaline particles could activate vibrations within the water molecules and diminish the clusters of the water molecules, thus providing the high conductivity and low surface tension of water (Table 2) as in the above discussion. The sol particles aggregated and grew via the electric field attraction of tourmaline, thus more and more crystal nuclei were formed around tourmaline. Therefore, the ions moved more rapidly and the sol particles grew faster than in the control TiO₂ sample, and spatial effects made smaller crystallite particle sizes. It was assumed that the small size of the single particles would reduce the backlog inside the lattice, and that the far-infrared emission of tourmaline could strengthen bond vibration, therefore the bond distance of Ti–Ti tends to form the larger anatase phase (0.304–0.379 nm) instead of the rutile phase (0.195–0.198 nm). Moreover, CH₃COOH acted on the nonhydrolytic processing of the [001]-oriented lattice plane and the form of the anatase crystal, which could be promoted by the electric field and the strong far-infrared radiation of tourmaline particles for the stability of anatase crystalline-phase.^32,33 In the sample C, E and F photocatalysts, including tourmaline facilitated anatase as the main phase. During the preparation of sample B, tourmaline particles were absent and the cerous nitrate solution was dropped in to the solution before HCl and CH₃COOH were added. Ce³⁺ as foreign ions might disturb the formation of normal sol, therefore the form of the sol particles was not perfect and uniform, and the rutile phase appeared. During the preparation of sample D, tourmaline particles and cerous nitrate solution were dropped in to the solution before HCl and CH₃COOH were added. The electric field and strong far-infrared radiation of tourmaline, apart from the foreign ion Ce³⁺, might also disturb the formation of normal sol, therefore the form of sol particles was not perfect and uniform and more of the rutile phase appeared than in C, E and F. It was exposed that the addition sequence of Ce and tourmaline could affect the phases and particle size of the composite photocatalysts.

Scheme 1 Synthetic route of the sample F photocatalyst nanoparticles.

3.3 Microstructure characterization

Fig. 2 shows the SEM and TEM micrographs of the sample A–F photocatalytic films. In the SEM morphologies it is shown that the D₅₀ of tourmaline was 0.40 μm, and the diameters of the pure TiO₂ particles in sample A were around 150 nm. The nanoparticles of the sample B–F films were smaller than the pure TiO₂ particles of sample A and had a larger surface area, which was consistent with the results of the XRD characterization and TEM images of samples B and F. In the sample C, E and F films, the tourmaline particles were wrapped up by the nanoparticles, which formed nanoparticle clusters and increased the surface area (Fig. 2C, E, F and I). In the sample B film, compared with sample A, smaller nanoparticles (Fig. 2B and H) formed the cascade distribution and rough surface, which also made the surface area higher. The SEM micrographs showed that both Ce and tourmaline made the diameters of TiO₂ particles smaller than that of the pure TiO₂ sample, therefore the sample B–F photocatalytic films each had a higher surface area than that of the control sample TiO₂ film (they rose from 23.82 m² g⁻¹ to about 40 m² g⁻¹ Table 3). It was shown that the addition of Ce and tourmaline affects the microstructure of TiO₂ nanoparticles.

Fig. 2 SEM (A: sample A photocatalyst; B: sample B photocatalyst; C: sample C photocatalyst; D: sample D photocatalyst; E: sample E photocatalyst; F: sample F photocatalyst; G: tourmaline particles) and TEM (H: sample B photocatalyst; I: sample F photocatalyst) micrographs of the tourmaline and different photocatalyst samples.

Table 3 The specific surface areas and pH values during the MO degradation reactions with samples A–F

Sample	SBET (m^2 g^−1)	pH values for MO degradation reaction at interval time
		0 h	1 h	2 h	3 h
A	23.82	5.23	5.29	5.18	5.17
B	38.26	5.26	5.25	5.19	5.16
C	39.88	5.23	5.15	5.21	5.19
D	39.75	5.28	5.18	5.17	5.21
E	40.65	5.30	5.21	5.18	5.18
F	41.23	5.18	5.13	5.15	5.18

---

Through EDS analysis it was observed that in the sample C film the nanoparticles around the clusters contained only the elements Ti and O, from the TiO₂, and Cl, from the HCl of the titanium sol. The nanoparticles of the clusters contained not only Ti, O, and Cl, but also the elements Fe, Al, Si, Ca, K, and Mg, which belonged to the tourmaline. It illustrated that the tourmaline particles were wrapped up by the TiO₂ nanoparticles. The sample E and F films showed a similar conclusion: the tourmaline particles were wrapped up by the Ce-doped TiO₂ nanoparticles. Combining the results of the XRD characterization with the SEM and TEM micrographs confirmed the microstructure of sample F and its synthetic route (as shown in Scheme 1). In the sample D film, the tourmaline particles were added before the titanium sol, which might have disturbed the primary formation of the sol particle nucleus, so the photocatalyst nanoparticles were not wrapped around the tourmaline perfectly, this was in accordance with the discussion of the XRD results as above.

3.4 UV-vis diffuse reflectance spectra of different composite photocatalyst

The absorption edges of the sample A–F composites are shown in Fig. 3. Compared with sample A and sample C (without the doping Ce), it is revealed that samples B, D, E and F absorb light in the range of 400–500 nm (the visible light region), because of the doping with Ce to narrow the band gap of the TiO₂ composite photocatalysts. However, the low doping content of Ce induced a low absorption of visible light. In consideration of the higher absorption of UV light, UV light (253.7 nm) was still used as a light source for the photodegradation of MO.

Fig. 3 UV-vis diffuse reflectance spectra of the sample A–F photocatalysts.

3.5 Photocatalytic activity towards methyl orange of the different composite photocatalysts

Fig. 4 shows the photocatalytic degradation of MO via a stainless steel net surface plated with the sample A–F films. With regard to the TiO₂ composite films, the content of tourmaline was 0.40 wt% and the content of Ce was 0.10 wt%. The degradation time was 180 min, and the MO decolorization ratio increased significantly with the increase of the reaction time. It was known from Fig. 3 that the TiO₂ film had the lowest photocatalytic degradation ratio of MO at 59.0%, the sample B film was 68.3%, the sample C–E films were all around 83% and the sample F film was at 94.6%. It was shown that the photocatalytic efficiency of the TiO₂ composite photocatalysts could be improved with tourmaline or rare earth metals. And the efficiency of the MO photocatalytic degradation via the sample F composite film was higher than those of both sample B and C. This conclusion was confirmed by the optical images of MO photodegradation in Fig. 5.

Fig. 4 The MO photocatalytic degradation ratio of the sample A–F photocatalysts.

Fig. 5 The optical images of the MO photodegradation by samples A–F under UV light after different times (0 h and 3 h).

Among these samples with different composites and preparation processes, sample F had the highest photocatalytic efficiency. Tourmaline is a unique polar mineral and the tourmaline particles can be seen as many individual micro-electrical fields. During the preparation of sample F, as shown in Scheme 1, the anodic electric field of the tourmaline particles weakened the adsorption of the titanium ions, which stopped the crystal growing and formed titanium crystals with smaller sizes. This morphology provides the larger specific surface with many more photocatalytic active sites, so that the composite photocatalyst system can more fully make contact with the reactant and increase the utilization rate of the reaction at the activity point. During the photocatalytic reaction, as shown in Scheme 2, TiO₂ was excited by light with energy greater than the band gap, then electrons of the valence band jumped across the band gap into the conduction band and left a hole in the valence band. The photogenerated electrons reacted with O₂ dissolved in the water to form a superoxide radical (O₂⁻); hydroxyl ions (OH⁻) trapped photogenerated holes to produce a hydroxyl radical (˙OH), which can then perform a hydrogen abstraction reaction with organic pollutants. The strong oxidation ability of O₂⁻ and the ˙OH free radicals could degrade MO. However, during electron transitions from the valence band to the conduction band, there will be a considerable proportion of the photogenerated electrons returned to the valence band and recombined with holes. The cerium doped in the experiment exists in the tetravalent cerium state because of the easy oxidization at high temperature, equivalent to narrowing the band gap in titanium dioxide, which can expand the spectral response range to improve the photocatalytic efficiency. In addition, the tourmaline particle with its own permanent electrode, can be seen as a pair of electric dipoles, according to the following formula: E_r = (2/3)E₀(a/r)³ (where a is a tourmaline particle radius and r is the distance from the center). According to our previous study,²¹ the electric field is about 10⁴ to 10⁷ (the highest value) V m⁻¹ on a tourmaline surface in the ten micron range, which promoted the generation of OH⁻ from H₂O and the MO adsorption onto the surface of catalysts. The photocatalyst nanoparticles were excited by irradiation under UV light to generate electrons and holes, thus the photogenerated electrons were also adsorbed on the anode of the tourmaline particles to form O₂⁻ and avoided the recombination of photogenerated electron–hole pairs, which also improved the utilization rate of photogenerated holes. At the same time, the far-infrared radiation from the tourmaline particles can enhance the molecular motions and promote the decrease of water molecule clusters (H₂O)_n to improve the solubility of O₂, which is conducive to the organic pollutant degradation reaction. The natural electric field and the far-infrared radiation from the tourmaline particles, and the reaction of the rare earth element Ce, jointly affect the TiO₂ photocatalytic degradation of organic pollutants.

Scheme 2 Synergistic effects of tourmaline and Ce on MO photocatalytic degradation by the TiO₂ composite photocatalyst (sample F).

The pH value for the MO degradation reaction with every sample was detected as shown in Table 3. It is known that the pH value for every reaction was kept around 5.2; no obvious changes were seen. As acetic acid was used for the preparation of each sample (as seen in Table 1), the reaction solutions were found to be acidic (pH = 5.2–5.3) at the beginning (at 0 h), which is beneficial for MO degradation by all of the samples. During MO degradation, Ce could promote the photogeneration of electron–hole pairs and the tourmaline could promote the ionization of H₂O to form H⁺ and OH⁻. Thus the holes promoted by Ce and tourmaline would rapidly form radicals with high reactivity such as OH radicals (as shown in Fig. 6), to be utilized to degrade MO. Therefore, the pH value is kept around 5.2, and the MO degradation is accelerated by radicals during the reaction.

Fig. 6 The fluorescence spectra for the OH radicals of samples A–F in the terephthalic acid solution (after 30 min of UV light illumination).

According to work from other researchers,^23,34 the rutile phase of TiO₂ has a lower photocatalytic activity than the anatase phase, but the anatase and rutile phases of mixed-phase TiO₂ have a higher photocatalytic activity than pure phase TiO₂. However, the synergistic effect of tourmaline and the rare earth metal greatly enhanced the TiO₂ photocatalytic activity compared to the control TiO₂ sample with the anatase and rutile mixed-phase. Therefore, the best photocatalytic activity was found following the preparation process of sample F.

In the relevant literature,^17,18 cerium doping was introduced into the sample B film as Ce³⁺ (cerium nitrate). Because of the high temperature, the Ce³⁺ is easily oxidized and exists in the form of Ce⁴⁺, which formed the CeO₂–TiO₂ system in the film, and could narrow the band gap of the photocatalyst to make the excitation of photogenerated-electrons easier.³⁵ Furthermore, the smaller sized photocatalyst nanoparticles had a higher surface area and more active sites, which improved the photocatalytic efficiency of sample B (68.3%) compared with TiO₂ (59.0%). In the sample C–E films, the photocatalytic degradation ratio of MO was around 83% for each of them after 3 h. The sample C composite film was similar to a photocatalyst studied previously,²¹ but tourmaline is a unique polar mineral and formed titanium crystals with smaller sizes to increase the photocatalytic surface area. Furthermore, tourmaline could hinder the recombination of the photogenerated electron–holes in TiO₂ and enhance the photocatalytic activity. During the preparation of samples D and E, the cerous nitrate solution and tourmaline were added together before or after the formation of the titanium sol, and tourmaline might disturb the form of uniform sol particles with Ce. Therefore, the sample C–E films had improved the photocatalytic activity, but did not get the highest photocatalytic activity because of the disturbance between Ce and tourmaline during the preparation. Therefore, the addition sequence during the preparation process can effect the photocatalytic activity of the TiO₂ composite photocatalyst.

3.6 Effect of the tourmaline content on the photocatalytic activity

In consideration of sample F possessing the highest photocatalytic activity as discussed in Section 3.5, the optimal contents of tourmaline for this composite photocatalyst sample were determined. As is seen in Fig. 7, in the same light conditions and photocatalytic time, the composite photocatalyst samples (following the preparation procedure of sample F) comprising different contents of tourmaline (0 wt%, 0.20 wt%, 0.40 wt% and 0.60 wt%, respectively) demonstrated different photocatalytic activities. The activities of MO photocatalytic degradation via samples with tourmaline were stronger than the sample without tourmaline (sample B). Moreover, the mass fraction with 0.40 wt% had the strongest catalytic degradation activity of MO (within 3 h, the photocatalytic degradation ratio of MO reached 94.6%). With a further increase in the mass fraction of tourmaline, the MO photocatalytic degradation of sample F declined.

Fig. 7 The degradation ratio of MO using the composite photocatalyst samples with different tourmaline contents (0 wt%; 0.20 wt%; 0.40 wt%; 0.60 wt%).

The reason of this trend is that when the tourmaline content was too small, it was difficult to form an activity center; upon reaching a moderate amount, the tourmaline could effectively control the TiO₂ grain growth in crystals and reduce the recombination rate of electron and hole, so that the TiO₂ composite catalyst produced more ˙OH to improve the activity of the catalyst. However, the high content of tourmaline would adsorb excessive photogenerated electrons and reduce O²⁻ content, thus affecting the photocatalytic activity. If the tourmaline mass fraction continues to increase, tourmaline will package on the TiO₂ surface, which would lead to a decrease in the amount of photocatalytic activity centers and reduce the photocatalytic degradation activity. Moreover, the immoderate addition of tourmaline would disturb the preparation of the film on the stainless steel meshes. Therefore, 0.40 wt% was the optimal amount of tourmaline added to sample F.

3.7 Effect of the Ce content on the photocatalytic activity

As is seen in Fig. 8, in the same light conditions and photocatalytic time, the composite photocatalyst samples (following the preparation procedure for sample F) comprising different contents of Ce (0 wt%, 0.05 wt%, 0.10 wt% and 0.15 wt%, respectively) demonstrated different photocatalytic activities. In 3 h, the MO photocatalytic degradation ratios of samples with different Ce content were 68.3%, 90.0%, 94.6% and 88.9%, respectively. Among these samples, the film with 0.10% of Ce content (the same as sample F that Section 3.5 discussed) had the highest ratio of the photocatalytic degradation. When the Ce content was increased to 0.15%, the photocatalytic activity began to slow down. It means that too much or too little Ce content was not conducive to improve the photocatalytic activity of the composite photocatalyst sample. Under UV light, Ce⁴⁺ also could generate electron–holes, therefore the doped rare earth compound could increase the catalytic activity. But during the heating part of the preparation, Ce⁴⁺ will gradually migrate from the anatase phase to the TiO₂ powder surface, and excess dopant would result in the rare earth ion being deposited on the TiO₂ surface to form the composite with a center of charge carriers, thus lowering the photocatalytic activity. Therefore, 0.10 wt% was the appropriate amount of Ce to add to sample F.

Fig. 8 The degradation ratio of MO by the composite photocatalyst samples with different Ce content (0 wt%; 0.05 wt%; 0.10 wt%; 0.15 wt%).

Through the XRD and SEM characterization, our study inferred that the addition sequence of Ce and tourmaline affected the microstructure of TiO₂ composite photocatalysts. In sample F, Ce was added before the primary titanium sol and could disperse uniformly with primary titanium solution and not disturb the formation of the sol nuclei around tourmaline. Moreover, the tourmaline particles were added after the primary titanium sol, and resulted in smaller size TiO₂ nanoparticles, a higher surface area and more activity sites. By means of the MO photocatalytic degradation ratio, it was revealed that the microstructure could enhance the photocatalytic activity; meanwhile, doping with Ce narrowed the band gap and tourmaline hindered the recombination of the photogenerated electron–hole pairs and promoted the oxidative degradation of MO. Because of the synergy-effect of Ce and tourmaline on the microstructure and photocatalytic activity of TiO₂ composite photocatalyst, sample F with 0.10 wt% Ce and 0.40 wt% tourmaline had the highest MO photocatalytic degradation ratio.

4 Conclusions

During the preparation of the TiO₂ composite photocatalyst, the tourmaline particles act as many individual micro-electrical fields, which could restrict the crystal growth and result in smaller titanium nanoparticles. This concave convex morphology was significant for increasing the surface areas and active sites to improve the photocatalytic efficiency. Besides, the electrostatic adherence and physical adsorption of tourmaline were also conducive to the contact between organic pollutants and active sites. In the electric field of tourmaline, the photogenerated electrons were adsorbed on the anode of tourmaline, further avoiding the recombination of photogenerated electrons and holes. Under the comprehensive effect of the far-infrared radiation and the electric field of tourmaline, the ionization process of water was accelerated, which is conducive to the formation of hydroxyl radicals, and thus improved the photocatalytic activity. Furthermore, Ce narrowed the band gap of the TiO₂ composite photocatalyst, which also enhanced the degradation reaction of organic pollutants.

This paper prepared tourmaline, Ce and TiO₂ composite photocatalysts via a sol–gel method. By detecting the photocatalytic activity of the composite photocatalysts with different preparation methods and component contents, and analyzing the effects of tourmaline and Ce on the composite catalyst, we took a preliminary study on the synergistic mechanism of tourmaline and Ce on the TiO₂ composite photocatalyst and summarized the appropriate preparation process and component proportion (sample F with 0.10 wt% Ce and 0.40 wt% tourmaline). This study has provided the preparation of a TiO₂ photocatalyst improved with Ce and tourmaline, which is a green and promising technique in organic pollutant degradation and in the area of environmental protection.

Acknowledgements

This project was supported by Leading Talent Cultivation Plan for Hebei Province Higher Education Innovation Team of China (Grant no. LJRC020).

Notes and references

1. M. A. Ahmed, J. Photochem. Photobiol., A, 2010, 238, 63.
2. P. Gao, J. Liu and T. Zhang, J. Hazard. Mater., 2012, 229–230, 209.
3. Y. Xie, Q. N. Zhao and X. J. Zhao, Catal. Lett., 2007, 118, 231.
4. X. Feng, K. Shankar and O. K. Varghese, Nano Lett., 2008, 8, 3781.
5. Z. Lai, F. Peng and Y. Wang, J. Mater. Chem., 2012, 22, 2390.
6. Z. Hu and J. C. Yu, J. Mater. Chem. A, 2013, 1, 12221.
7. Z. Hu, J. C. Yu and T. Ming, Appl. Catal., B, 2015, 168–169, 483.
8. C. B. Zhang, H. He and K. Tanaka, Appl. Catal., B, 2006, 65, 37.
9. S. Sun, J. J. Ding and J. Bao, Appl. Surf. Sci., 2012, 258, 5031.
10. T. Nogawa, T. Isobe and S. Matsushita, Mater. Lett., 2012, 82, 174.
11. Q. R. Deng, X. H. Xia and M. L. Guo, Mater. Lett., 2011, 65, 2051.
12. Y. H. Liu, Z. L. Wang and W. B. Fan, Ceram. Int., 2014, 40, 3887.
13. H. Y. Wei, Y. S. Wu and N. Lun, J. Mater. Sci., 2004, 39, 1305.
14. F. B. Li, X. Z. Li and C. H. Ao, Chemosphere, 2005, 59, 787.
15. T. Z. Tong, J. L. Zhang and B. Z. Tian, J. Colloid Interface Sci., 2007, 315, 382.
16. S. Kityakarn, Y. Pooarporn and P. Songsiriritthigul, Electrochim. Acta, 2012, 83, 113.
17. S. N. R. Inturi, T. Boningari and M. Suidan, Appl. Catal., B, 2014, 144, 333.
18. A. Charanpahari, S. S. Umare and R. Sasikala, Appl. Surf. Sci., 2013, 282, 408.
19. Z. H. Meng, L. H. Wan and L. J. Zhang, J. Ind. Eng. Chem., 2014, 20, 4102.
20. X. G. Sun, C. M. Li and L. Y. Ruan, J. Alloys Compd., 2014, 585, 800.
21. J. S. Liang, J. P. Meng and G. C. Liang, Trans. Nonferrous Met. Soc., 2006, 16, s542.
22. S. H. Song and M. Kang, J. Ind. Eng. Chem., 2008, 14, 785.
23. G. K. Zhang and X. Qin, Mater. Res. Bull., 2013, 48, 3743.
24. Z. Wu, H. Wang and K. Zheng, J. Phys. Chem. C, 2012, 116, 12814.
25. R. R. Yeredla and H. Xu, J. Phys. Chem. C, 2008, 112, 532.
26. G. Donnay, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1977, 33, 927.
27. T. Nakamura and T. Kubo, Ferroelectrics, 1992, 137, 13.
28. Z. Shen, S. Sun and W. Wang, J. Mater. Chem. A, 2015, 3, 3285.
29. F. Wang, W. Kwok Hung Ng and J. C. Yu, Appl. Catal., B, 2012, 111–112, 409.
30. K. Ishibashi, A. Fujishima and T. Watanabe, Electrochem. Commun., 2000, 2, 207.
31. X. Cao, D. Li and W. Jing, J. Mater. Chem., 2012, 22, 15309.
32. J. F. Ye, W. Liu and J. G. Cai, J. Am. Chem. Soc., 2011, 133, 933.
33. P. Gao, A. R. Li and D. D. Sun, J. Hazard. Mater., 2014, 279, 96.
34. Y. L. Zhang, H. H. Gan and G. K. Zhang, Chem. Eng. J., 2011, 172, 936.
35. J. H. Park, W. S. Kim and D. H. Jo, J. Ind. Eng. Chem., 2014, 20, 1965.

---