Citrate/F⁻ assisted phase control synthesis of TiO₂ nanostructures and their photocatalytic properties

Abstract

A novel and controlled sol-hydrothermal synthesis of nano-TiO₂ assisted by citrate/F⁻ has been developed. We studied factors that influence its size and morphology, such as citrate, Cl⁻, F⁻ and so on. Different ratios of citrate to TiCl₄ gave different sizes and morphologies. Cl⁻ preferentially formed a rutile phase, while citrate preferentially formed an anatase phase, and F⁻ gave a dominant {001} facet. Additionally, we found that the photocatalytic activity of the TiO₂ synthesised with the assistance of citrate was superior to the commercially available TiO₂.

---

Introduction

As one of most studied semiconducting materials, titanium dioxide (TiO₂) plays a prominent role in fundamental studies in solar energy conversion, energy storage, photocatalysis, water-spitting, and gas sensing.^1–4 The properties of nanosized TiO₂ vary with crystal size, morphology, and crystallographic structure. Therefore, controlled synthesis of nanosized TiO₂ is of great importance. Various methods have been used to prepare TiO₂ including sol–gel, reverse micelles, hydrothermal and solvothermal methods.^5–9 The sol–gel method is widely used for preparing TiO₂ nanocrystals. This method can provide excellent chemical homogeneity and the possibility of deriving unique metastable structures under mild reaction conditions, which allows compositional and microstructure tailoring through controlling the precursor chemistry and processing conditions. Under different conditions, anatase, rutile, or even brookite type TiO₂ can be constructed. Among them, the anatase type TiO₂ is most widely used because of its higher performance, larger band gap (3.2 eV) and lower recombination rate.¹⁰

Generally, sol–gel precipitates are amorphous and need a further heating treatment, of which hydrothermal treatment and the calcination process are commonly used, to induce crystallization. However, calcination frequently accelerates particle agglomeration and induces phase transformation.¹¹ Some early research has demonstrated that sol composition, pH, reaction temperature, aging time and the nature of solvent in hydrothermal treatments could influence the particle size, morphology and surface chemistry.^12,13 For example, under an acidic medium, Cl⁻ and NO₃⁻ ions favor the construction of a rutile phase, whereas SO₄²⁻ and AcO⁻ are more likely to form an anatase phase. F⁻ is of vital importance in the formation of the exposed {101} surface, which was expected to have a higher surface energy and higher chemical activities than the low-energy {110} surface.¹⁴ Moreover, under a different pH of acidic medium, mixtures of rutile and anatase phases are obtained with different proportions.¹⁵ Yin et al. and Liu et al. reported the effects of carboxylic acids, including citric acid, on the microstructure and performance of the anatase TiO₂ nanocrystals.^16,17 Herein, we report the effects to the sol–gel process, morphology and crystal phase of citrate/F⁻ as an additive to prepare TiO₂ nanostructures. Citrate ions are crucial in the formation of the anatase phase. In the sol–gel process, citrate ions have a tight combination with Ti⁴⁺ and compete with Cl⁻ in the TiO₂ formation reaction. By controlling the mole ratio of Ti⁴⁺ to citrate, we could control the mass ratio of the anatase and rutile phases of the products. More Cl⁻ and F⁻ could also affect the crystal phase and morphology of the TiO₂ nanostructures. In the different conditions, the size of TiO₂ nanostructures are not obviously changed, but the particle size distribution is narrowed when there is a higher mole ratio of Ti⁴⁺ to citrate. Subsequently, we found that the photocatalytic activities of the TiO₂ whose synthesis was assisted by citrate were superior compared with commercially available TiO₂.

Results and discussion

Effects of citrate on the sol–gel process of titania synthesis

Sodium citrate (SC) is naturally abundant and accessible in the environment. Citrate ions are also widely used to control the size and morphology in nanomaterial synthesis.¹⁷ In the synthesis of TiO₂, the α-hydroxycarboxylic and α-aminocarboxylic functional groups can interact with the Ti source or its hydrolysis species and induce the formation of anatase TiO₂ nanocrystals. In order to acquire a deeper understanding, we added various amounts of sodium citrate to a uniform sol and observed the difference among the proportions. We discovered that when a little amount of sodium citrate was added, the sol quickly deposited in a short time, and then, with the increased amounts, the precipitates were transformed into a more stable sol, which was static for several months. The effects of the Ti⁴⁺ : SC ratio (r_c) on the particle size of the sols was analyzed by DLS and is illustrated in Fig. 1. In 10.0 mL solution, a uniform amount of the titanium precursor (0.01 mmol) was added, corresponding to 0–0.01 mmol sodium citrate. When no sodium citrate is added, the average particle size in the sol is under 10 nm. If a little SC (0.5 μmol, r_c = 20 : 1 mole ratio) is added, the average particle size increases to about 30 nm. However, when r_c = 100 : 15, the average particle size abruptly changes from tens to thousands of nanometers in size. As the r_c is further increased, the average size slowly decreases again to about 40 nm (r_c = 1 : 1). Citrate ions obviously play a key role during this process.

Fig. 1 The particle sizes of different titanium precursors analyzed by DLS. The r_c increased from (a) to (g): (a) 100 : 0; (b) 100 : 5; (c) 100 : 10; (d) 100 : 15; (e) 100 : 20; (f) 100 : 30; (g) 100 : 40; (h) 100 : 100.

Typically, in the sol–gel processing of titanium alkoxides, hydrolysis and condensation reactions occur very rapidly, and precipitates emerge in a short time. Polymers and complex ligands are used to control the condensation reactions. In this process, SC did the same work equally. When there is a low mole ratio of SC : Ti⁴⁺ under 0.4, a small amount of SC may promote aggregation by decreasing the potential between the positive charges on the particles. If more SC is added into the solution, the fresh SC reacts with the precipitates again and constructs a more stable sol. Fig. 2 presents the UV-vis absorption spectra of the transparent sols before and after the SC was added. After the SC was added into the solution, the absorption of the sol had an evident blue shift, which indicated the structure of the sol had been changed by the SC.

Fig. 2 UV-vis absorption spectra of transparent sols at different reaction times: (a) 0 min; (b) 30 min; (c) 60 min; (d) 90 min (no SC added); (e) 90 min (with SC added); (f) 120 min. An equal amount of SC was added into the solution at 90 min.

Besides the mole ratio of SC to Ti⁴⁺, the pH of the solution can also influence condensation reaction. The possible species of titanium citrate anions and their transformations in the Ti citrate system under the influence of pH are shown in Scheme S1 in the ESI† according to published papers.¹⁸ As the starting material was TiCl₄, the aqueous solution was acidic at the beginning. In a low concentration solution (1 mM), precipitates appeared rapidly in several minutes, compared to several hours in a high concentration solution (0.1 M). Ultrasound and heating can accelerate this slow process. When we prepared different concentration solutions, the pH of the solution changed correspondingly. The higher concentration solutions have a lower pH value. Grassian et al.¹⁹ reported that citrate ions have different adsorption abilities to TiO₂ nanoparticles at acidic and circumneutral pHs. Under different pHs, citrate ions exist as three species (H₂Cit⁻, HCit²⁻, and Cit³⁻). When pH is below 4.0, citrate ions mostly exist as H₂Cit⁻. When the pH is between 4.0 and 6.0, citrate ions mostly exist as HCit²⁻. So as the pH increases, the concentration of COO⁻ is higher in a uniform SC solution. In uniform solutions (r_c = 5 : 1), the precipitates appeared in minutes in a pH = 6.0 solution, while it needed several hours in a pH = 2.0 solution. This illustrates that the COO⁻ of SC combined together with the titanium complexes and constructed the precipitates in a certain Ti⁴⁺ : SC mole ratio. If little COO⁻ exists, the combination of the –COOH and titanium complexes will be difficult.

According to previous research,²⁰ in pure TiCl₄ aqueous solution with a concentration of less than 0.4 mol L⁻¹, the majority of titanium ions exist as six-fold coordinated monomers of [TiO(OH₂)₅]²⁺ and their polynuclear hydrolysis species. When COO⁻ species are added into the solution, [TiO(OH₂)₅]²⁺ combines with these rapidly and loses two water molecules. The complexes of citrate and titanium have a further olation reaction and form some co-monomers. In a low concentration SC solution, one citrate ion may combine with more than one titanium ion. The formed un-stable complexes react with each other rapidly, forming a mixture of dimer, trimer and other low molecular number polymers.

However, when some fresh citrate was added, the low molecular number polymers were deconstructed by a competing reaction. The entire process was reversible regardless of whether fresh citrate or fresh titanium was added. Fig. 3 shows the FT-IR spectra of different r_c ratio samples, which indicate the formation process. In a pure Ti precursor, a single band appears at 1620 cm⁻¹ (Fig. 3a). When citrate is added, symmetric and asymmetric COO⁻ stretching vibrations appear at 1410 and 1550 cm⁻¹, the frequency separation is 140 cm⁻¹, which indicates that acetate anions act as bidentate ligands and this larger value also suggests these are bridging acetates rather than chelating acetates (Fig. 3b–d).²¹ As the r_c increases, the broad band of COO⁻ stretching vibrations becomes broader. Shoulders appear at 1635 cm⁻¹ and 1720 cm⁻¹ when r_c = 1 : 1, which are assigned to the bending vibrations of H–O–H (Fig. 3e).²² The existence of these shoulders indicate the hydrogen bonding between excess carboxylate ions. It is also evidence that the amount of citrate ions is greater than the amount of Ti ions. Two sharper bands appear after more citrate is added. Moreover, the O–H bending vibration of the hydrogen bonded C=O group at 1120 cm⁻¹ also becomes sharper, which illustrates that some citrate ions might be retained in the dry gel as ligands, or hydrogen-bonded groups to the hydroxo sites.^23,24

Fig. 3 FT-IR spectra of the titanium precursor with different titanium to citrate ratios: (a) 1 : 0; (b) 1 : 0.1; (c) 1 : 0.25; (d) 1 : 0.5; (e) 1 : 1; (f) 1 : 1.5; (g) 1 : 2.

As the compounds were dissolved in H₂O, Fig. 4 shows the ¹³C NMR spectra of 0.3 M solutions with different titanium and citrate ratios in D₂O. When the r_c is lower than 1 : 1, there were no distinct differences among the ¹³C NMR results (Fig. 4a–c). Referring to the FT-IR spectra, it was easy to define the four shifts. The shifts of methylene (CH₂) and α-hydroxyl (C–O) carbons were at 42.5 ppm and 72.1 ppm. The shifts of α-carboxyl carbons and β-carboxyl carbons both combined with the titanium were at 175.7 ppm and 172.6 ppm respectively. When citrate was superfluous in the solution, there were some free citrate ions, and the α-carboxyl carbons and β-carboxyl carbons combined with the titanium both had large downfield shifts. Meanwhile, three new shifts appeared at 86.9 ppm, 174.7 ppm and 184.6 ppm in the ¹³C NMR spectra (Fig. 4e). Under these conditions, the uncoordinated protonated carboxyl groups and hydroxyl groups were strongly hydrogen bonded to the deprotonated groups of an adjacent molecule. The formation of hydrogen bonds could be the reason for the new shifts, and the cause of the shifts of carboxyl and hydroxyl moving downfield. The shifts of α-hydroxyl (C–O), α-carboxyl carbons and β-carboxyl carbons moved to 86.9 ppm, 174.7 ppm and 184.6 ppm, respectively.²⁵ These results also coincide with the results of FT-IR spectra. We propose a possible condensation reaction pathway in the formation of the titanium complexes as shown in Fig. 5, which shows the olation and oxolation/alkoxolation reaction processes.

Fig. 4 ¹³C NMR spectra of the titanium precursor with different titanium to citrate ratios: (a) 1 : 0.3; (b) 1 : 0.6; (c) 1 : 1; (d) 1 : 2; (e) 1 : 3.

Fig. 5 Possible condensation reaction pathway for the formation of titanium complexes.

The effects of titanium : SC mole ratio and pH were discussed superficially above. Some additional factors that may have effects are Cl⁻ and OH⁻ ions. Yanagida et al. agreed that citrate ions could be combined with titanium complexes to generate (Ti(OH)_x(citrate)_yCl_z)ⁿ⁻ and related species,¹⁶ but the theory has never been precisely explained. Further research has been carried out regarding the effects of citrate ions and Cl⁻ on the morphology and crystalline phase of titania.

The effects of citrate on the morphology and crystalline phase of titania

Hydrothermal treatment is widely used in titania nanoparticle preparation. In early reports,^12,26 Cl⁻, NO₃⁻, CH₃CHOO⁻, and SO₄²⁻ were shown to favor the formation of a crystalline phase. Citrate can also favor the formation of anatase crystallites in a hydrothermal reaction. Fig. 6 shows the effect of citrate on the morphology and crystalline phase of titania. A mixture of anatase, rutile, and brookite was obtained when there was no SC used in the preparation. The titanium precursor tended to form large size rutile nanorods and small size anatase/brookite nanoparticles (Fig. 6a). When a certain amount of SC (r_c = 1 : 1) was mixed with the precursor, a uniform pure anatase product was obtained and the size of anatase particles were about 15 nm. Citrate seemingly acted as a surfactant to control the particle size and prevent the aggregation of small size particles.

Fig. 6 TEM images and XRD patterns of TiO₂ particles prepared by a hydrothermal process: (a and c) r_c = 1 : 0; (b and d) r_c = 1 : 1.

Fig. 7 shows the TEM images and XRD patterns of titanium gel and TiO₂ nanoparticles formed in different reaction conditions (additional optimization of the reaction conditions is shown in Fig. S1†). Generally, in an HCl-acidified medium, the titanium precursor is more likely to form rutile phase titania. In previous studies,²⁶ rutile crystallites began to form after the anatase crystals grew to a larger size, about 50 nm. The key for phase transformation is controlled by the volume free energy, surface energy, and strain energy. It was suggested that citrate can stop the small-size anatase nanocrystals from growing further. TiO₂ can be easily modified with carboxylates by a condensation reaction. Citrate ions combined with titanium complexes may change the surface energy of the nanocrystals. CH₃COO⁻ also shows the same effect in early research.²⁷

Fig. 7 TEM images of TiO₂ at different reaction times and conditions: (a) 80 °C for 3 h; (b) 80 °C for 72 h; (c) hydrothermal reaction for 18 h in 180 °C; (d) XRD patterns corresponding to (a–c).

Fig. 8 shows the XRD patterns of samples synthesized by hydrothermal treatment at 180 °C for 18 h. The rutile phase is the major phase (about 57%) in the pure water solution. When a little SC (r_c = 1 : 0.00625 or 1 : 0.0125) was added to the solution, the ratio of rutile phase TiO₂ had an obviously decline to about 36%, and there was no evidence of the brookite phase (Fig. 8b and c). As the amount of SC was increased (r_c = 1 : 0.025 or 1 : 0.05), the rutile phase TiO₂ completely disappeared in the product and only anatase phase was observed from the XRD patterns (Fig. 8d and e).

Fig. 8 XRD patterns of TiO₂ with different Ti⁴⁺ to citrate mole ratios: (a) 1 : 0; (b) 1 : 0.00625; (c) 1 : 0.0125; (d) 1 : 0.025; (e) 1 : 0.05.

In this series of samples, the size and shape of the TiO₂ nanostructures never changed enormously. Only the size distribution increased, as the amount of SC decreased. Therefore the shape control and the crystalline phase control are unconnected in hydrothermal processing. As mentioned above, the reaction of citrate with titanium is a competing reaction. The concentration of H⁺ or Cl⁻ can influence the species of citrate and the reaction between citrate and titanium. So we chose a critical condition to investigate the effects of H⁺ and Cl⁻ concentration. When r_c = 1 : 0.025, additional concentrated hydrochloric acid (w = 37.5%, 250 μL) was added into the solution. Fig. 9a shows the XRD pattern of the product. As additional HCl was added, the pure anatase product was transformed into a mixture of anatase and rutile, while only a little rutile phase TiO₂ (about 9%) was formed (Fig. 9b). Separately, additional NaCl (4 mmol) was added. The product contained about 29% rutile phase TiO₂. This illustrated that Cl⁻ competed with citrate during the TiO₂ nanoparticle formation process. As the concentration of Cl⁻ increased, Cl⁻ had a greater chance of combining with the titanium ions and affecting the structure of TiO₂ nanocrystals, as Cl⁻ always tends to encourage formation of the rutile phase TiO₂ crystals. Although the crystal phase of the product had an obvious change, the particle size and shape never changed (shown in Fig. 10a and b and S3†).

Fig. 9 XRD patterns of TiO₂ with different additions: (a) with nothing added; (b) with 4 mmol HCl added; (c) with 4 mmol NaCl added; (d) with 4 mmol HF aqueous solution added; (e) with 8 mmol HF aqueous solution added.

Fig. 10 TEM/SEM images of TiO₂ with different additions: (a) with nothing added; (b) with 4 mmol HCl added; (c) with 4 mmol HF aqueous solution added; (d) SEM of (c). The precursor solution contained 2 mmol Ti⁴⁺.

As Cl⁻ can affect the crystal phase of the product, F⁻ may also perform a role in the reaction process. Fig. 10 shows the TEM/SEM images before and after addition of NaCl or HF into the solution.

We chose HF as an F⁻ source to investigate the effects of F⁻ during the hydrothermal process. After the HF was added, the pattern of the TiO₂ became sharper, which indicated the TiO₂ nanoparticles grew to a larger size than before. At the same time, the diffraction peak of the {200} facet became higher compared to the {101} facet and the exposed crystal facets of the particles also changed. In recent years, the effect of F⁻ has been widely researched and discussed.^23–28 It is generally agreed that F⁻ can contribute to the preparation of TiO₂ nanosheets with dominant {001} facets. Additionally, NH₄F was also used as an F⁻ source to investigate the effects of F⁻. Fig. 11 shows the TEM images of the TiO₂ nanocrystals after NH₄F was added and Fig. 12 shows the corresponding XRD patterns. When the amounts of NH₄F were increased, the particle size of the TiO₂ nanocrystals became larger. However, when the mole ratio of Ti and F atoms was increased to 2 : 8, the product became irregular in morphology (Fig. 11d). Large sized pieces of TiO₂ were obtained, which were larger than 200 nm wide. The results illustrated that F⁻ could also take part in the competing reaction. When the concentration of F⁻ was under an appropriate level, TiO₂ nanosheets with high ratio of exposed {001} facets could be obtained. The nanosheets had 50–90% exposed {001} facets depending on the particle size. However, when F⁻ is abundant, the superfluous F⁻ etches the TiO₂ nanocrystal and some irregular TiO₂ was obtained under this condition. The amount of F and corresponding reaction times were worth further investigation in order to obtain monolithic large sized TiO₂ nanocrystals with a high ratio of exposed {001} facets.

Fig. 11 TEM/SEM images of TiO₂ with different amounts of NH₄F: (a) 1 mmol added; (b) 2 mmol added; (c) 4 mmol added; (d) 8 mmol added; (e) corresponding SEM image of (d). The precursor solution contained 2 mmol Ti⁴⁺.

Fig. 12 XRD patterns of TiO₂ with different amounts of NH₄F: (a) 1 mmol added; (b) 2 mmol added; (c) 4 mmol added; (d) 8 mmol added. The precursor solution contained 2 mmol Ti⁴⁺.

Photocatalytic measurements of the TiO₂ nanostructures

The photocatalytic activity of the nanostructures towards the degradation of rhodamine B (RhB) in aqueous solution was investigated. The potential application of TiO₂-based materials in the degradation of RhB has been reported and it is generally accepted that the photocatalytic degradation of organic compounds proceeds via two routes: direct hole oxidation or ˙OH radical oxidation.²⁹ Fig. 13 shows the decomposition of RhB dye using a series of TiO₂ nanostructured photocatalysts under UV light irradiation (365 nm). To rule out the disturbance of the catalyst, the UV-vis diffuse reflectance spectra of the as-synthesized catalyst are shown in Fig. S2.† It is obvious that there is no absorbance in the region 500–600 nm. For comparison, the photocatalytic activities of aeroxide P25 and commercial pure anatase TiO₂ (d ≤ 25 nm) were tested under the same conditions. Although the particle size and crystal phase of the series of TiO₂ were similar, the photocatalytic activities were quite distinct from each other. It seems that the photocatalytic activity is improved with the increase of r_c. It took more than 60 min for T_0.5 to degrade RhB to less than 10%, while the same result could be achieved in less than 30 min when T_0.0125 was used, which is much better than previously reported results.²⁹ The activities of T_0.2, T_0.1 and T_0.05 were nearly the same, and were between the T_0.025 and T_0.5 activities. T_0.025 and T_0.0125 both had excellent activities during the first 30 min. T_0.0125 had been characterized as a mixed crystal of the anatase phase and rutile phase, and was more active than the pure anatase phase T_0.025. In these catalysts, commercial pure anatase TiO₂ and aeroxide P25 were comparatively inefficient, and aeroxide P25 was only better than T_0.05. To investigate whether the radical species are also involved in our degradation process, we have performed a radical-trapping experiment. Isopropanol, which has been described as the best hydroxyl radical quencher, was used in our control experiment. When isopropanol was added to the RhB solution at a concentration of 10 mM, extensive inhibitions in RhB degradation were observed, which proved that photodegradation is initiated by the radical species produced by TiO₂ materials.

Fig. 13 Photodegradation of RhB under UV light with commercial anatase TiO₂, P25 and TiO₂ nanostructures.

The results suggested that citrate could affect the catalytic performance during the hydrothermal process by changing the crystal structures or surface properties of the nanostructures. As mentioned above, different amounts of citrate added into the uniform titanium tetrachloride aqueous solution could construct different titanium citrate complexes, and such reactions may occur during the hydrothermal process too. The effects of the phase changes have been studied for a long time.^30–39 The mixed phase TiO₂ had anatase phase as the major phase which shows better performance. The functional groups exposed on the particle surface also affected the photodegradation activity. The results achieved here are also better than those obtained from previous research.

Conclusions

Assisted by citrate/F⁻, a simple sol-hydrothermal synthetic method was developed to prepare TiO₂ nanostructures with controlled phase composition and morphology. We studied the factors that influence the size and morphology of the TiO₂ nanostructures, such as citrate, Cl⁻, F⁻ and so on. During the sol–gel process, citrate could combine with titanium and construct different titanium citrate complexes. Through controlling the ratio of sodium citrate to titanium tetrachloride, we could obtain different types of TiO₂. In this process, Cl ions could compete with citrate, which lead to the formation of rutile phase TiO₂ crystals, while citrate tended to form anatase phase TiO₂ crystals. Between citrate and Cl ions, citrate ions were more effective in combining with titanium precursor. By controlling the amount of citrate, the crystal phase of the nanostructures can be adjusted. F ions also could take part in the construction process, giving a dominant facet. By adding various amounts of HF or NH₄F, a series of TiO₂ nanocrystals with exposed {001} facets and the particle sizes range from 30 nm to 1 μm were obtained. The synthetic nanosheets had 50–90% {001} facets depending on the particle size. In addition, we found that the photocatalytic activities of the TiO₂ assisted by citrate ions were superior to aeroxide P25 and commercial pure anatase TiO₂.

Experimental

Materials

Aeroxide P25 was purchased from Sigma Aldrich. Commercial pure anatase TiO₂ (d ≤ 25 nm) was purchased from Acros Organic. All other chemicals were of analytical grade and were used without further purification, and were all supplied by Sinopharm Chemical Reagent Company.

Preparation of titanium sol

TiCl₄ (1 mmol) was added dropwise into 10 ml deionized water in an ice bath and this solution was stirred for 30 min. Sodium citrate (1 mmol) was then dissolved in another 10 ml deionized water. Preparation of titanium sol in different proportions of TiCl₄ and sodium citrate: 0.1 ml TiCl₄ solution was mixed with corresponding sodium citrate solution, from 0 to 0.1 ml, and was diluted to 10 ml with deionized water. To get an alkaline condition sol, the pH was adjusted using a 0.1 M NaOH solution.

Hydrothermal treatment

TiCl₄ (1 mmol) was added dropwise into 10 ml sodium citrate aqueous solution under vigorous stirring for 30 min. Then the mixture was added into a Teflon-lined stainless steel autoclave. The autoclave was heated to 180 °C and kept for 18 h in an electric oven. After static cooling down to room temperature, the TiO₂ nanocrystals were harvested by centrifugation and washed with deionized water several times.

Photocatalytic measurement

Before we measured the photocatalytic properties, the as prepared TiO₂ nanocrystals and P25 (1 mmol) were washed with 0.1 M NaOH aqueous solution and deionized water several times, and then were suspended in 20 mL deionized water. Before exposure to UV light irradiation (300 W, wavelength 365 nm), 1 mL of the treated TiO₂ aqueous suspension was added to 100 mL RhB aqueous solution (10 mg L⁻¹), and the mixture was stirred for 10 min. Then, 5.0 mL of mixture was taken out every 10 min, and TiO₂ was separated from the solution by centrifugation. The solution was reserved for fluorescence spectrum measurements. During the photodegradation process, the suspension was exposed to the air without oxygen bubbling.

Materials characterization

The crystal structures of the TiO₂ samples were characterized by powder X-ray diffraction (XRD, X’Pert-Pro MPD, Cu Kα radiation, 60 kV), XRD patterns were obtained from 20° to 80°. The phase content was estimated using the following equation.

f_A = 1/(1 + 1/K × I_R/I_A)

In this equation, f_A is the mass fraction of the anatase phase in the powder, K = 0.79 when f_A > 0.2 and K = 0.68 when f_A < 0.2, I_R and I_A are the integrated intensities of the rutile {110} and the anatase {101} peaks, respectively.

The morphologies of the TiO₂ nanostructures were investigated by a Tecnai-G220 transmission electron microscope (TEM). The diameters of the different TiO₂ sols were analyzed by dynamic light scattering measurements. The stability of the suspensions was investigated under the same conditions, and quantitative surface adsorption measurements were made. UV-vis spectra were recorded by a UV-vis spectrophotometer (UV-3150, SHIMADZU).

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 21373006, 51402203), Natural Science Foundation of Jiangsu Province for Young Scholars (BK20140326), the Natural Science Foundation of Jiangsu Higher Education Institutions (14KJB430021) and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Notes and references

1. A. I. Hochbaum and P. D. Yang, Chem. Rev., 2010, 110, 527.
2. X. Chen and S. Mao, Chem. Rev., 2007, 107, 2891.
3. S. M. Khan, M. A. Shahry and J. W. B. Ingler, Science, 2002, 297, 2243.
4. M. Paulose, O. K. Varghese, G. K. Mor, C. A. Grimes and K. G. Ong, Nanotechnology, 2006, 17, 398; J. Yu, J. Low, W. Xiao, P. Zhou and M. Jaroniec, J. Am. Chem. Soc., 2014, 136, 8839.
5. H. Zhang and J. F. Banfield, Chem. Mater., 2005, 17, 3421.
6. J. Lin, Y. Lin, P. Liu, M. J. Meziani, L. F. Allard and Y. Sun, J. Am. Chem. Soc., 2002, 124, 11514.
7. H. Yang, C. Sun, S. Qiao, J. Zou, G. Liu, S. C. Smith, H. Cheng and G. Lu, Nature, 2008, 453, 638.
8. X. Li, Q. Peng, J. Yi, X. Wang and Y. Li, Chem.–Eur. J., 2006, 12, 2383.
9. X. Wang, J. Zhuang, Q. Peng and Y. Li, Nature, 2005, 437, 121.
10. J. M. Macak, M. Zlamal, J. Krysa and P. Schmuki, Small, 2007, 3, 300.
11. C. Wang and J. Ying, Chem. Mater., 1999, 11, 3113.
12. M. Wu, G. Lin, D. Chen, G. Wang, D. He, S. Feng and R. Xu, Chem. Mater., 2002, 14, 1974.
13. R. Scotti, I. R. Bellobono, C. Canevali, C. Cannas, M. Catti, M. D. Arienzo, A. Musinu, A. Polizzi, M. Sommariva, A. Testino and F. Morazzoni, Chem. Mater., 2008, 20, 4051.
14. C. Wen, H. Jiang, S. Qiao, H. Yang and G. Lu, J. Mater. Chem., 2011, 21, 7052.
15. S. T. Aruna, S. Tirosh and A. Zaban, J. Mater. Chem., 2000, 10, 2388.
16. H. Yin, Y. Wada, T. Kitamura, T. Sumida, Y. Hasegawa and S. Yanagida, J. Mater. Chem., 2002, 12, 378.
17. Y. Liu, C. Liu and Z. Zhang, Chem. Eng. J., 2008, 138, 596.
18. (a) E. T. Kefalas, P. Panagiotidis, C. P. Raptopoulou, A. Terzis, T. Mavromoustakos and A. Salifoglou, Inorg. Chem., 2005, 8, 2596; (b) T. Kemmitt, N. I. A. Salim, G. J. Gainsford, A. Bubendorfer and M. Waterland, Inorg. Chem., 2004, 20, 6300; (c) M. Dakanali, E. T. Kefalas, C. P. Raptopoulou, A. Terzis, G. Voyiatzis, I. Kyrikou, T. Mavromoustakos and A. Salifoglou, Inorg. Chem., 2003, 15, 4632; (d) Y. F. Deng, Z. H. Zhou and H. L. Wan, Inorg. Chem., 2004, 20, 6266; (e) I. A. Mudunkotuwa and V. H. Grassian, J. Am. Chem. Soc., 2010, 132, 14986.
19. I. A. Mudunkotuwa and V. H. Grassian, J. Am. Chem. Soc., 2010, 132, 14986.
20. P. Comba and A. Merbach, Inorg. Chem., 1987, 26, 1315.
21. M. T. Tsai, J. Non-Cryst. Solids, 2002, 298, 116.
22. (a) R. Parra, M. S. Góes, M. S. Castro, E. Longo, P. R. Bueno and J. A. Varela, Chem. Mater., 2008, 20, 143; (b) M. Colomer, M. Velasco and J. Jurado, J. Sol-Gel Sci. Technol., 2006, 39, 211.
23. R. Parra, M. S. Góes, M. S. Castro, E. Longo, P. R. Bueno and J. A. Varela, Chem. Mater., 2008, 20, 143.
24. N. T. Nolan, M. K. Seery, S. J. Hinder, L. F. Healy and S. C. Pillai, J. Phys. Chem. C, 2010, 114, 13026.
25. Y. Deng, Z. Zhou and H. Wan, Inorg. Chem., 2004, 43, 6266.
26. (a) C. C. Wang and J. Y. Ying, Chem. Mater., 1999, 11, 3113–3120; (b) Y. Q. Zheng, E. W. Shi, Z. Z. Chen, W. J. Li and X. F. Hu, J. Mater. Chem., 11, 1547–1551.
27. (a) Y. Q. Dai, C. M. Cobley, J. Zeng, Y. M. Sun and Y. N. Xia, Nano Lett., 2009, 9, 2455; (b) M. M. Wu, G. Lin, D. H. Chen, G. G. Wang, D. He, S. H. Feng and R. R. Xu, Chem. Mater., 2002, 14, 1974.
28. Y. Dai, C. M. Cobley, J. Zeng, Y. Sun and Y. Xia, Nano Lett., 2009, 9, 2455.
29. (a) J. D. Zhuang, W. X. Dai, Q. F. Tian, Z. H. Li, L. Y. Xie, J. X. Wang and P. Liu, Langmuir, 2010, 26, 9686–9694; (b) R. Libanori, T. R. Giraldi, E. Longo, E. R. Leite and C. Ribeiro, J. Sol-Gel Sci. Technol., 2009, 49, 95–100.
30. H. Yu, B. Tian and J. Zhang, Chem.–Eur. J., 2011, 17, 5499.
31. J. Wu, X. Song, L. Ma and X. Wei, J. Cryst. Growth, 2011, 319, 57.
32. H. Zhang, Y. Wang, P. Liu, Y. Han, X. Yao, J. Zou, H. Cheng and H. Zhao, ACS Appl. Mater. Interfaces, 2011, 3, 2472.
33. Y. Wang, H. Zhang, Y. Han, P. Liu, X. Yao and H. Zhao, Chem. Commun., 2011, 47, 2829.
34. J. Zhang, Q. Xu, Z. Feng, M. Li and C. Li, Angew. Chem., Int. Ed., 2008, 47, 1766.
35. S. Li and U. Diebold, J. Am. Chem. Soc., 2010, 132, 64.
36. Z. Liu, X. Zhang, S. Nishimoto, M. Jin, D. A. Tryk, T. Murakami and A. Fujishima, Langmuir, 2007, 23, 10916.
37. S. Bakardjieva, J. Subrt, V. Stengl, M. J. Dianez and M. Sayagues, Appl. Catal., B, 2005, 58, 193.
38. P. Deak, B. Aradi and T. Frauenheim, J. Phys. Chem. C, 2011, 115, 3443.
39. J. Zhang, M. Li, Z. Feng, J. Chen and C. Li, J. Phys. Chem. B, 2006, 110, 927.

---

Footnotes

† Electronic supplementary information (ESI) available: Optimization of the reaction condition, UV-vis diffuse reflectance spectra of the catalyst and other additional information. See DOI: 10.1039/c5ra08816j

‡ Equal contribution to this article.

---