Crystallite phase-controlled preparation, characterisation and photocatalytic properties of titanium dioxide nanoparticles

Abstract

A solution-based processing method has been used to synthesise nanostructured titanium dioxide (TiO₂) with different crystalline forms ratio (anatase, rutile and their mixtures) by controlling the hydrolysis of titanium tetrachloride in aqueous solution. The contents of anatase and rutile phases in the synthesised TiO₂ powders have been successfully controlled by simply changing the proportion of in the aqueous phase. The prepared materials were characterised by X-ray diffraction, transmission electron microscopy, scanning electron microscope and Brunauer–Emmett–Teller techniques. The experimental results showed that the average size of the synthesised particles was in the range of 6–11 nm. The photocatalytic performance of the synthesised TiO₂ nanoparticles was evaluated by removal of the dye, C. I. Acid Blue 9 (AB9), under UV light illumination (30 W). According to the results, rutile appeared to be a poor photocatalyst while the mixed-phase TiO₂ (rutile–anatase) demonstrated the highest photoactivity. The efficiency parameters such as apparent quantum yield and electrical energy per order were estimated and compared for different crystalline forms of TiO₂. It was found that the photocatalysis process with synthesised mixed-phase TiO₂ nanoparticles, containing 70% anatase, had lower electrical energy consumption and higher quantum yield compared with others.

Keywords:

• nanoparticles
• nano-TiO₂
• photocatalyst
• nanocatalyst
• advanced oxidation processes

1. Introduction

Titanium dioxide (TiO₂) has attracted significant attention from researchers because of the many interesting physical and chemical properties that make it suitable for a variety of applications. For instance, TiO₂ has high corrosion resistance and chemical stability and an excellent optical transparency in the visible and near infrared regions, as well as a high refractive index that makes it useful for anti-reflection coatings in optical devices 1. It has been used mostly as a pigment in paints, sunscreens, ointments and toothpaste since its commercial production in the early twentieth century. TiO₂ pigments are inorganic chemical products used for imparting whiteness, brightness and opacity to a diverse range of applications and end-use markets, including coatings, plastics, paper and other industrial and consumer products 2,3. TiO₂ is obtained from a variety of ores that contain ilmenite, rutile, anatase and leucoxene, which are mined from deposits located throughout the world. Crystals of TiO₂ can exist in one of the three crystalline forms: rutile, anatase and brookite. Only anatase and rutile forms have good pigmentary properties, and rutile is more thermally stable than anatase. The crystallographic properties of rutile and anatase have been reported in Table 1 4. Most TiO₂ pigments have been produced from titanium mineral concentrates by the so-called chloride or sulphate process, either as the rutile or the anatase form.

Table 1. Crystallographic properties of rutile and anatase 4.

Crystal structure	Density (kg m^−3)	System	Space group	Cell parameters (nm)
	A		b		C
Rutile	4240	Tetragonal		0.4584		0.2953
Anatase	3830	Tetragonal		0.3758		0.9514

Since the discovery of photocatalytic splitting of water on TiO₂ electrodes 5, efforts have been devoted to the development of efficient water and air purification technologies, based on TiO₂ photocatalysis. Such treatments typically oxidise toxic organic compounds to non-toxic inorganic compounds, such as carbon dioxide, water, ammonium or nitrates, and chloride ions 6.

In the framework of the rapid development of nanoscience and nanotechnology, the domain of nanostructured materials, such as nanostructured TiO₂, is necessitating more research, both of an academic and of an industrial nature. Synthesis methods are a major prerequisite for achievement in this rapidly evolving field. The essential first step in the use of nanomaterials in various technologies is their production 7,8.

There are many methods of producing nanostructured TiO₂, such as the hydrothermal method 9,10, sol–gel technique 11,12, chemical vapour deposition (CVD) 13–15, physical vapour deposition 16,17, solvothermal method 18,19, electrochemical approaches (e.g. anodising of Ti) 20–22, solution combustion method 23–25, microemulsion 26,27, micelle and inverse micelle methods 28,29, ball milling 30, a flame by vapour phase 31, sonochemical reactions 32 and plasma evaporation 33,34. It should be noted that as precursors of nanocrystalline oxide powders, however, inorganic compounds are more economical than alkoxides. Since the phases and morphologies of TiO₂ are critical parameters in determining their suitability for particular applications, it is still necessary to explore new methods for synthesising nanostructured TiO₂, and especially to explore the methods fit for selectively preparing various TiO₂ phases (anatase, rutile and their mixtures).

Therefore, in this work, the solution-based processing method has been used to synthesise various crystalline phases of nanostructured TiO₂ including anatase, rutile and their mixtures, by controlling the hydrolysis of TiCl₄ in aqueous solution. The photocatalytic activity of the synthesised nanostructured TiO₂ samples has been tested in decolorisation of the dye solution containing C. I. Acid Blue 9.

C. I. Acid Blue 9, which belongs to the acidic dyes group, is soluble in cold water and methanol. It can be found in textile (as a dye for wool and silk), foodstuff and pharmaceutical wastewaters. In addition, C. I. Acid Blue 9 is the principal active ingredient in Aquashade, which can be used as an aquatic algaecide/herbicide, in natural or manmade ponds, lakes, fountains, fish farms and fish hatcheries, and may be applied by both professional applicators and homeowners. It is hazardous in case of ingestion, skin contact (irritant), eye contact (irritant), and inhalation 35,36.

2. Experimental details

2.1. Reagents

All inorganic compounds were obtained from Merck, Germany. C. I. Acid Blue 9 was obtained from Shimi Keshavarz Company, Iran. Its structure and characterisation are given in Table 2.

Table 2. Structure and characteristics of C. I. Acid Blue 9.

Other names	Brilliant Blue FCF, Erioglaucine, C.I. Food Blue 2, Brilliant Blue FCF, FD&C Blue no. 1
Colour index (C. I.) no.	42090
Molecular formula	C37H34N2Na2O9S3
Molecular mass	792.86 (g mol^−1)
Absorption maximum ()	625 (nm)
CAS No.	3844-45-9
Aqueous solubility	200 kg m^−3
Chemical class	Triphenylmethan dye
Chemical structure

2.2. Preparation of TiO₂ nanoparticles

The nanocrystalline TiO₂ reported in this study was prepared by TiCl₄ hydrolysis. Titanium tetrachloride (98% TiCl₄) was used as a main starting material without any further purification. The appreciated amount of TiCl₄ was dissolved in distilled water in an ice-water bath (Figure 1). The concentration of titanium was adjusted to 3 M. This aqueous solution was then mixed with (NH₄)₂SO₄ solution for preparation of anatase in a temperature-controlled bath. The mixture was stirred at high speed while the amount of TiCl₄ solution necessary for the desired [H₂O]: [Ti] molar ratio was added dropwise. Having maintained it at the hydrolysis temperature (70°C) for 1 h, the mixed solution was treated with 2.5 M dilute NH₄OH until the pH value was 7. For preparation of rutile crystalline, the TiCl₄ solution was diluted with 0.5 M HCl solution (70°C) and was not treated with NH₄OH solution. Then, the mixed solution was placed into a constant temperature (70°C) bath for 5 h. Subsequently, the precipitated titanium oxide (TiO₂–nH₂O) was separated from the solution by using centrifugation and repeatedly washed with distilled water to make TiO₂–nH₂O that was free of chloride ions. The hydrous titanium oxide was dried at 110°C under vacuum and ground to fine powder (TiO₂ powder 1), then was calcined at 400°C for 2 h (Figure 1). For preparation of a mixture of anatase and rutile phase TiO₂ powder, a small amount of (NH₄)₂SO₄ was added and the concentration of TiCl₄ was adjusted to 0.5 M with distilled water. The rest of the preparation steps are just like the preparation method of anatase in a temperature-controlled bath.

Figure 1. Flow chart of the steps involved in the preparation of TiO₂ nanoparticles of anatase (left) and rutile (right).

2.3. Characterisation of the nanostructured TiO₂

To determine the crystal phase composition and average crystalline size of synthesised nanostructured TiO₂ samples, X-ray diffraction (XRD) measurements were carried out at room temperature by using Siemens XRD D5000, with Cu Kα radiation. The accelerating voltage of 40 kV and emission current of 30 mA was used. The average crystalline size of the samples was calculated according to Debye–Scherrer formula 37:

where D is the average crystallite size (Å), λ is the wavelength of the X-ray radiation (Cu Kα = 1.54178 Å), β is the full width at half maximum (fwhm) intensity of the peak and θ is the diffraction angle. The weight fraction of the anatase and rutile present in the samples was estimated based on the most signal of each phase (anatase and rutile) according to the equation described by Spurr and Myers 38:

where x _A is the weight fraction of the anatase phase present in the powder sample, while I _A and I _R are the relative reflection intensities of anatase and rutile, respectively.

The samples used for transmission electron microscopy (TEM) observations were prepared by dispersing the powders in distilled water followed by ultrasonic vibration (Sonoplus Ultrasonic Homogeniser HD 2200, Germany) for 15 min, then placing a drop of the dispersion onto a copper grid coated with a layer of amorphous carbon. A LEO 906 (60 kV) TEM was used to study the morphology. The surface morphology of the catalysts was observed by scanning electron microscope (SEM). The samples for SEM imaging were coated with a thin layer of gold film to avoid charging. A LEO 440i SEM was used to study the morphology.

Nitrogen sorption analyses were obtained with a Micromeritics Tristar sorptometer using standard continuous procedures at 77.15 K on calcined samples that had been degassed at 363 K for 1 h and then at 403 K under high vacuum for at least 10 h. Surface area was calculated according to the Brunauer–Emmett–Teller (BET) model 39 over a relative pressure range of 0.05–0.30. Pore diameter distribution was calculated according to the Barret–Joyner–Halenda (BJH) method 39, modified by the Halsey thickness curve correction 40 on the desorption branch.

2.4. Photocatalytic experiments

The photocatalytic test-reaction chosen to characterise the different synthesised nanostructured TiO₂ samples was the decolorisation of the dye solution containing C. I. Acid Blue 9, selected as a model organic pollutant. In the UV/TiO₂ process, irradiation was carried out with a 30 W (UV-C) mercury lamp (Philips, Holland), which was put above a batch photocatalytic reactor of 500 mL in volume. On the surface of the solution, the light intensity was measured by Cassy Lab, Germany. The desired concentration of C. I. Acid Blue 9 (20 mg L⁻¹) and TiO₂ were fed into the pyrex reactor. The TiO₂ suspension was sonicated for 15 min, before illumination, to disperse TiO₂ uniformly in the solution by sonoplus ultrasonic homogeniser HD 2200, Germany. The value of incident photon flux by reactor volume unit at 254 nm was 2.33 × 10⁻⁶ Einstein L⁻¹ s⁻¹, which was calculated on the base of ferrioxalate actinometry measurement 41. The pH of the solution was adjusted using dilute solutions of sulphuric acid and aqueous sodium hydroxide when necessary and measured by pH meter (Metrohm, Switzerland). The UV lamp was switched on to initiate the reaction. During the irradiation, agitation was maintained to keep the solution homogenous and to increase mass transfer coefficient and, as a result, the overall dye decomposition rate. At specific time intervals, the concentration of the dye in each sample was determined using a Lightwave S2000 (England) UV/Vis spectrophotometer at λ _max = 625 nm. A linear correlation was established between the dye concentration and the absorbance at λ _max = 625 nm, in the range of 0–35 mg L⁻¹ with a coefficient of correlation r ² = 0.9993. The equation used to calculate the photocatalytic removal efficiency (X) in the treatment experiments was:

where C ₀ was the initial concentration of the dye (mg L⁻¹) and C was the concentration of the dye (mg L⁻¹) at time t.

3. Results and discussion

3.1. Preparation mechanism of different crystallite phases of TiO₂

The mechanism of forming different crystallite phases of nanostructured TiO₂ has been showed in Figure 2. From metal halide to the formation of metal oxide, there are two processes: hydrolysis and polycondensation 42. When the titanic halide reacts with water, the titanium ion first increases its coordination by using its vacant d orbitals to accept oxygen electron pairs from nucleophilic ligands (such as −OH groups) 43. Consequently, the titanium ion forms an octahedral structure of (Ti(O) _m (OH) _n (H₂O)_{6 − m − n} )^{(2m + n − 4) −}. Because the reaction is performed under an acidic medium, only −OH and −OH₂ groups will be present, and the octahedral structure of the titanium ion becomes (Ti(OH) _m (H₂O)_6 − m )^{(m − 4) −} 44. Then, the structures dehydrate each other and polycondensed into the final precipitate.

Figure 2. Proposed mechanism for preparation of TiO₂ with rutile or anatase phases: (a) the orientation of the third octahedron determines whether a rutile or an anatase nucleus is formed; (b) interaction between and octahedral hydroxyls; (c) two octahedra share edge in the presence of ; (d) formation of anatase in the presence of .

Crystals of TiO₂ can exist in one of the three crystalline forms: rutile, anatase or brookite (Table 1). In their structures, the basic building block consists of a titanium atom surrounded by six oxygen atoms in a more or less distorted octahedral configuration. In all three TiO₂ structures, the stacking of the octahedra results in three-fold coordinated oxygen atoms. The fundamental structural units in these three TiO₂ crystals forming from TiO₆ octahedron units and having different modes of arrangement and links as presented in Figure 2. In the rutile form, TiO₆ octahedra link by sharing an edge along the c-axis to form chains. These chains are then interlinked by sharing corner oxygen atoms to form a three-dimensional framework, whereas in anatase, the three-dimensional framework is formed only by edge-shared bonding among TiO₆ octahedrons. It means that octahedra in anatase share four edges and are arranged in zigzag chains along 44–48.

As can be seen in Figure 2, for forming anatase and rutile nuclei, the placement of the third octahedron is very important and determines whether a rutile or an anatase nucleus is formed. However, the presence of would influence the orientation of the third octahedron. When ions exist in the acid reaction media, these ions would interact with octahedral hydroxyls by static electricity. Because of the steric effect of , the octahedron with and another octahedron would polycondense along the converse direction in order to decrease the repulsion, and the orientation of the third octahedron is more conducive to the formation of an anatase nucleus. The more the , the more anatase nuclei can be formed. TiO₂ clusters grow further on the nucleus and then form the anatase phase. It has been reported that the presence of accelerated the growth of TiO₂ clusters to anatase 47,48. As a result, different rutile/anatase mixtures can be prepared by changing the proportion of in the aqueous phase. On the other hand, there is a weak steric effect for Cl⁻ because of its small radius. Meanwhile, the addition of Cl⁻ generally favours the formation of rutile crystallites 47–49.

As was mentioned, TiO₂ is known to possess three crystalline structures, which include rutile, anatase and brookite. In coming up with these crystal structures and to estimate the crystal grain size of anatase, rutile and brookite, the XRD experimental method is used. The XRD patterns of the synthesised anatase, rutile and their mixture have been reported in Figure 3. It can be observed from Figure 3 that anatase peaks in XRD are at 2θ = 25.3°, 37.8°, and 48.1°, while the rutile peaks are at 2θ = 27.5°, 36.2°, and 54.4°, where θ is the XRD angle. This result confirmed that the addition of sulphate was quite effective in promoting the formation of the anatase phase as the pH value rose as a result of the neutralisation of NH₄OH. The characteristics of the synthesised nanostructured TiO₂ have been reported in Table 3. The TEM images of the synthesised TiO₂ have been shown in Figure 4. These show the size distribution of TiO₂ particles. The average diameter of the particles measured from the TEM images is in good agreement with XRD results.

Figure 3. X-ray diffraction (XRD) patterns of synthesised TiO₂ samples: (a) anatase, (b) rutile (c) rutile–anatase.

Figure 4. TEM images of the synthesised TiO₂ samples calcined at 400°C for 2 h.

The prepared TiO₂ powders, either in the presence of ions or not, dried at room temperature under vacuum, were dominantly amorphous (Figure 5). The morphology of the amorphous TiO₂ powder observed by SEM is shown in Figure 5.

Figure 5. SEM image (left) and XRD pattern (right) of the amorphous TiO₂ powder dried at room temperature under vacuum.

3.2. Photocatalytic activity of the synthesised TiO₂ nanoparticles

The photocatalytic decolorisation of the solution containing C.I. Acid Blue 9 under UV illumination was used to evaluate the photoactivity of the synthesised TiO₂ nanoparticles. We found that TiO₂ and UV light had a negligible effect when they were used on their own. The reason for this observation is due to the fact that when TiO₂ is illuminated with light of energy greater than its optical band gap (λ < 390 nm), an electron is transferred to the conduction band, leaving a hole in the valance band. This is called production of an electron-hole (e − /h + ) pair. At the solid liquid interface, electron transfer can occur both from the conduction band to an electron acceptor (O₂, H⁺, etc.) in the solution and from a donor (OH⁻, H₂O, organic pollutants (P), etc.) in the solution to the valence band hole. Moreover, the hydroxyl radicals formed are strong oxidants and attack organic pollutants present at or near the surface of TiO₂. This causes photodegradation of pollutants 50–52. The mechanism is assumed up in Figure 6.

Figure 6. General mechanism of the photocatalysis on TiO₂ nanoparticle.

The characteristic absorbance at 625 nm of AB9 solution was monitored during the reaction process. Comparison of the photocatalytic activity of different synthesised TiO₂ nanoparticles has been shown in Figure 7. It is clear that the photocatalytic activity of amorphous TiO₂ is negligible. As can be seen from Figure 7, rutile appeared to be a poor photocatalyst while the mixed-phase TiO₂ (rutile–anatase) demonstrated the highest photoactivity. In a recent study by Gray et al. 53, mixed-phase TiO₂ nanocomposites were synthesised using a hydrothermal method for photooxidation and photoreduction applications. These authors found that the anatase–rutile nanocomposite exhibited the highest photoactivity among the catalysts tested. They also claim that a synergic effect exists between the anatase and rutile particles, in which spatial charge separation hinders charge recombination 53. In addition, it has long been empirically observed that mixed-phase TiO₂, i.e. TiO₂ containing both anatase and rutile, tends to exhibit higher photocatalytic activities than pure-phase TiO₂. When Bacsa and Kiwi 54 prepared TiO₂ samples with different ratios of anatase : rutile, the highest photoactivity was obtained with a 70 : 30 ratio, which is similar to that of Degussa P25. In our study, the anatase : rutile ratio of the mixed-phase TiO₂ was 70 : 30 (Table 3).

Figure 7. Comparison of the photocatalytic activity of different synthesised TiO₂ nanoparticles. [TiO₂]₀ = 150 mg L⁻¹, [AB9]₀ = 20 mg L⁻¹, pH = 6.2, V = 50 mL, I = 11.2 W m⁻²: (a) during different irradiation times, (b) at the irradiation time of 90 min.

Table 3. Physicochemical characteristics of the synthesised TiO₂ nanoparticles.

Crystalline phase of synthesised TiO2 sample	Anatase	Rutile	Anatase–Rutile
Anatase content (wt%) by XRD	100	0	70
Average crystallite size (nm)	8	11	6–10
BET surface area (m^2 g^−1)	119.96	28.31	75.29
Total pore volume (cm^3 g^−1)	0.3197	0.0747	0.2069
Pore size (nm)	9.59	10.11	12.92

A possible reason for the improved performance of such mixed-phase TiO₂ samples is that rutile acts as a sink for the electrons generated in anatase, serving to physically separate the electron-hole and thereby depress rates of recombination 55 (Figure 8(a)). This model is consistent with the fact that the band edges of rutile lie within those of anatase; i.e. the potential of the conduction band edge of anatase is more negative than that of rutile. However, it has been recently shown that just the opposite occurs: rutile undergoes band gap activation, and electrons are shuttled from rutile to anatase sites, which must be of lower energy (Figure 8(b)). This implies that one or more trap sites exist on anatase at potentials more positive than the conduction band edge of either anatase or rutile. This was recently confirmed by a photoacoustic spectroscopy study of anatase, which found trap sites on anatase at an average of 0.8 eV below the conduction band edge 55,56.

Figure 8. Models of mixed-phase TiO₂: (a) rutile islands surround anatase particles, and rutile is an electron sink; (b) a small rutile core surrounded by anatase crystallites, where electrons are shuttled from rutile to anatase.

It can be concluded that the size and morphology of rutile and anatase nanocrystals are critical to the separation and enhanced activity of mixed-phase TiO₂. As was illustrated in Figure 8(b), an emerging model of mixed-phase TiO₂ particles describes an atypically small rutile core surrounded by anatase crystallites. Catalytic ‘hot spots’ are believed to exist at the intersection of the two phases, where distorted geometry gives rise to unique surface chemistry.

The photocatalytic activity of TiO₂ materials can be influenced by many factors, including crystal structure, particle size, surface area, porosity, density of surface hydroxyl groups, surface acidity, number and nature of trap sites, and adsorption–desorption characteristics 35,53. For a catalyst, a key requirement to demonstrate high activity is to have the small primary particle size and high surface area. As can be seen in Table 3, the primary particle size of the synthesised mixed-phase TiO₂ (6–10 nm) is smaller than that others.

Figure 9 shows a typical time-dependent UV-Vis spectrum of AB9 solution during photocatalysis. The spectrum of AB9 in the visible region exhibits a main band with a maximum at 625 nm. As is clear from this figure, the absorption peaks diminished and finally disappeared under reaction, which indicated that the AB9 had been destructed. No new absorption bands appear in either the visible or ultraviolet regions. Complete decolorisation of the dye solution was observed after 90 min in the optimised conditions. Since photocatalysis of aqueous organic pollutant is an electric-energy-intensive process, and electric energy can represent a major fraction of the operating costs, simple figures-of-merit based on electric energy consumption can be very useful and informative. The photochemistry commission of the International Union of Pure and Applied Chemistry (IUPAC) proposed a figure-of-merit (or more appropriately, an Efficiency Index, as it compares electrical efficiency of different AOPs) for UV-based AOPs 57. Electrical energy per order (E _EO) compares electrical efficiency of different UV-based AOPs and is a measure of the electrical efficiency of an AOP system. It is defined (for low concentration of pollutants) as the electrical energy in kilowatt hours (kWh) required bringing about the degradation of a contaminant by one order of magnitude in 1m³ (1000 L) of contaminated water. Generally, the higher the energy efficiency of a process, the lower the E _EO. The values of E _EO can be calculated using the following Equation 57,58:

Figure 9. UV-Vis spectrum of AB9 (20 mg L⁻¹) during photocatalysis in the presence of the synthesised mixed-phase TiO₂ nanoparticles. [TiO₂]₀ = 150 mg L⁻¹, pH = 6.2, I = 11.2 W m⁻².

The electric dose (kWh m⁻³ order⁻¹) required to oxidise C. I. Acid Blue 9 (20 mg L⁻¹) in the presence of synthesised TiO₂ nanoparticles has been reported in Table 4. It is clear that a photocatalytic process with synthesised mixed-phase TiO₂ nanoparticles offered the best energy efficiency. Finally, it is useful to relate the E _EO values found in this study to the treatment costs. For instance, if the cost of the electricity in Iran and France is $0.0065 and $0.092 per kWh, the contribution to treatment cost of 20 mg L⁻¹ AB9 solution, in the presence of the mixed-phase TiO₂, from electrical energy will be $ 4.0 and $ 56.7 per m³, respectively. In addition there will be smaller cost factors for the photocatalyst used and for UV lamp replacement.

Table 4. Rate constants, half-lives, electrical energy per order and apparent quantum yield for photocatalytic decolorisation of AB9 in the presence of the synthesised TiO₂ nanoparticles. [AB9]₀ = 20 mg L⁻¹, [TiO₂]₀ = 150 mg L⁻¹, pH = 6.2.

Crystalline phase of synthesised TiO2 sample	k (min^−1)	R ^2	t 1/2 (min)	E EO (kWh m^−3 order^−1)	ϕapp × 10^−3
Rutile–Anatase	0.0374	0.98	18.53	616	6.75
Anatase	0.0159	0.96	43.59	1449	2.87
Rutile	0.0067	0.97	103.45	3439	1.21

The efficiency of the heterogeneous photocatalytic process can be quantified in term of quantum yield, which may be defined as the rate at which reactant molecules disappear or product molecules form, divided by the number of photons absorbed per unit time. In our study, the apparent quantum yield of photocatalysis process has been calculated by Equations (5) and (6) 59,60:

where k (s⁻¹) is the apparent first-order rate constant, C ₀ (mol L⁻¹) is the initial dye concentration and I ₀ (Einstein L⁻¹ s⁻¹) is the total intensity of incident photons entering the reactor. The apparent quantum yield for oxidation of C. I. Acid Blue 9 (20 mg L⁻¹) in the presence of synthesised TiO₂ nanoparticles has been reported in Table 4. As seen in Table 4, the photocatalytic process with synthesised mixed-phase TiO₂ nanoparticles has lower electrical energy consumption and higher quantum yield compared with others.

4. Conclusions

TiO₂ nanoparticles in anatase, rutile and their mixture phases have been synthesised from aqueous solution of TiCl₄. The addition of a small amount of (NH₄)₂SO₄ promoted occurrence of anatase phase and inhibited the anatase–rutile transformation such that the powder was completely anatase after calcining at 400°C for 2h. The anatase–rutile mixture (70 : 30) was prepared by controlling the proportion of in the aqueous phase. The synthesised TiO₂ samples have narrow size distribution and their average particle size was 6–11 nm. The mixed-phase TiO₂ nanoparticles have reasonable photocatalytic efficiencies, and nearly complete of C. I. Acid Blue 9 was removed in 90 min. The values of electric energy per order and apparent quantum yield calculated in the presence of different synthesised TiO₂ samples, demonstrated that the most efficient process was that using mixed-phase TiO₂ nanoparticles containing 70% anatase.

Acknowledgements

The authors thank the University of Tabriz, Iran for financial and other supports.