One-Step Nonaqueous Synthesis of Pure Phase TiO 2 Nanocrystals from TiCl 4 in Butanol and Their Photocatalytic Properties

Pure phase TiO2 nanomaterials were synthesized by an autoclaving treatment of TiCl4 with butanol as a single alcohol source. It was found that the control of molar ratio of TiCl4 to butanol played an important role in determining the TiO2 crystal phase and morphology. A high molar ratio of TiCl4 to butanol favored the formation of anatase nanoparticles, whereas rutile nanorods were selectively obtained at a low molar ratio of TiCl4 to butanol. Evaluation of the photocatalytic activity of the synthesized TiO2 was performed in terms of decomposition of organic dye rhodamine B under ultraviolet irradiation. It turned out that the as-synthesized TiO2 crystallites possessed higher photocatalytic activities toward bleaching rhodamine B than Degussa P25, benefiting from theirhigh surface area, small crystal size as well as high crystallinity.


Introduction
The surface photochemistry as well as physical and chemical stability of the semiconductor, TiO 2 , makes it an attractive material for multiple applications such as photocatalysis, photovoltaic cells, photochromics, photonic crystals, smart surface coatings, and sensors.Photocatalytic properties of TiO 2 depend significantly on the availability of active sites, and thus, depend on the percentage of exposed surface, surface area, phase composition, and degree of crystallinity [1][2][3][4][5][6][7][8][9][10].TiO 2 exists as three different crystalline structures: rutile, anatase, and brookite.Rutile is the most thermodynamically stable phase, whereas the latter two are metastable phase.Research so far has been centered on the synthesis of anatase nanoparticles.However, recently the preparation of nanometer-sized rutile has received increasing attention due to its promising potential as a photocatalyst [11][12][13][14].For this, well-crystallized anatase or rutile TiO 2 nanocrystals with small crystallite size and high surface area are highly desirable.
Besides, nonaqueous and surfactant-free sol-gel routes in organic solvents offer the possibility of enabling the synthesis of pure phase TiO 2 with high crystallinity and well-defined and uniform morphologies [15][16][17][18][19]. Therein, the simplest route to TiO 2 nanocrystals involves the reaction of TiCl 4 with alcohol and usually takes place at lower temperature.The alcohol not only acts as the oxygen-supplying agent for TiO 2 but also strongly influences particle size, shape, and even composition and crystal structure.The slow reaction rates in combination with the stabilizing effect of the organic species lead to the formation of highly crystalline products that are often characterized by uniform morphologies and crystallite sizes in the range of just a few nanometers.By reacting TiCl 4 with various alcohol homologs, different crystal phases (rutile and anatase) and morphologies (nanoparticles, nanorods) have been obtained.However, to the best of our knowledge, to manipulate the crystal phase and morphologies of TiO 2 nanocrystals using a single alcohol source has not previously been documented in the literature yet.Motivated by the above results, in this work, by reacting TiCl 4 with a single alcohol, butanol, we further extend nonaqueous surfactant-free route to synthesize pure phase of anatase or rutile nanocrystals.The results disclose that the reaction of TiCl 4 with butanol at a fairly low temperature leads to anatase TiO 2 nanocrystals with sizes between 5 and 10 nm.Moreover, the reaction system can be easily manipulated to produce rutile nanorods with diameters between 10 and 20 nm by simply decreasing the molar ratio of TiCl 4 to butanol.Last, the photocatalytic activities of the pure phase TiO 2 nanostructures are evaluated by decomposing the organic dye rhodamine B (RB).

Experimental
The procedure for the synthesis of TiO 2 nanomaterials is as follows.A certain molar of TiCl 4 (0.005 mol) was quickly dripped into butanol (0.02 or 0.10 mol) under vigorous stirring at room temperature.The pipets and autoclaves were dried in vacuum before use.The obtained solution was transferred to and sealed in teflonlined autoclave (15 mL), and the autoclave was then put in an oven for solvothermal treatment at 150 • C for 12 h.The products were collected by filtration, repeatedly washed by ethanol, and dried under vacuum.XRD patterns of the samples were recorded on a Rigaku, D/max-2500 X-ray diffractometer.Raman spectra were recorded on a Jobin-Yvon HR800 instrument with an Ar + laser source of 488 nm wavelength in a macroscopic configuration.High-resolution transmission electron microscopy (HRTEM) is performed with a JEM-3010 electron microscope.Nitrogen porosimetry was performed on a Micromeritics ASAP 2010 instrument.Surface areas were calculated using the Brunauer-Emmett-Teller (BET) equation.Pore size distributions were calculated using the Barret-Joyner-Halenda (BJH) model based on nitrogen desorption isotherm.
The photoreactor was designed with an internal light source (50 W high pressure mercury lamp with main emission wavelength 313 nm and an average light intensity of 2.85 mW cm −2 ) surrounded by a water-cooling quartz jacket to cool the lamp, where a 100 mL aliquot of the RB solution with an initial concentration of 20 mg L −1 and TiO 2 solid catalyst (0.1 g) were filled.The solution was stirred in the dark for 30 min to obtain a good dispersion and establish adsorption-desorption equilibrium between the organic molecules and the catalyst surface.Decreases in the concentrations of dyes were analyzed by a Cary 500 UV-VIS-NIR spectrophotometer at λ = 554 nm.At given intervals of illumination, the samples of the reaction solution were taken out and analyzed.

Results and Discussion
XRD and Raman techniques were used to characterize the structure and size of as-obtained nanomaterials.The XRD patterns (Figure 1(a)) clearly demonstrate that the nanomaterials obtained with high molar ratio of TiCl 4 to butanol are of anatase phase (JCPDS 21-1272).In contrast, a rutile sample (JCPDS 21-1276) is produced in the synthesis with a low molar ratio of TiCl 4 to butanol.The crystallite sizes, using the well-known Scherrer equation based on the half widths of (101) diffraction peak for anatase and (110) diffraction peak for rutile are 8.7 and 15.6 nm for the titania prepared at high and low molar ratio of TiCl 4 to butanol, respectively.Moreover, the XRD pure phase anatase and  rutile TiO 2 in the Raman spectra further confirms the single phase feature of the TiO 2 products in our present approach.That is, the spectrum of anatase TiO 2 , obtained at low molar ratio of TiCl 4 to butanol, reveals three Raman bands at 606, 437, and 243 cm −1 ranging between 200 and 800 cm −1 .The Raman bands at 393, 514, and 614 cm −1 are attributed to the anatase sample obtained at high molar ratio of TiCl 4 to butanol.No Raman active peaks for any impurities attributed to organic species further indicates that the rutile and anatase phase of TiO 2 thus obtained are of good quality.
The sample powders were further characterized with TEM and HRTEM to gain information on the morphology and size distribution of the crystallites.In the TEM images (Figure 2(a)), the anatase crystallites appeared as uniform irregular particles with sizes close to their XRD crystallite sizes.The diameter of the anatase nanoparticles ranges from 5 to 10 nm.  , respectively, using the tetragonal rutile unitcell, the angle 130 • in this case is attributed to the contact between (110) and (111) planes based on the interplanar angles.More information about the rutile crystals can be derived form the HRTEM image.The lattice spacings being about 0.32 and 0.22 nm between adjacent lattice planes of the TiO 2 , nanorods are corresponding to the distance between (110) and (111) crystal planes of the rutile phase, respectively.Therefore, the growth directions of nanorods are concluded to be perpendicular to (110) crystal planes.The preferential growth along this direction is arising from the unique crystal structure of rutile.It is known that rutile has 4 2 screw axes along [001] direction [20].The screw structure will promote the crystal growth along this direction, leading to a crystal morphology dominated by the (110) faces, and thereof the size calculated from the broadening of the 110 peak in the XRD pattern corresponds to the diameter of the nanorods.As expected, the results of XRD analysis are consistent with the TEM observation.It is therefore very clear that the change of the molar ratio of TiCl 4 to butanol is feasible for control of the crystal phase, and morphology of the TiO 2 materials.
The porous structures of the resulting samples were studied by nitrogen sorption.Figure 3 presents the nitrogen adsorption-desorption isotherms and BJH pore size distribution curves of the samples.The anatase nanoparticles exhibit a type IV adsorption isotherm with an H3 hysteresis loop according to BDDT classification, which is a typical characteristic of mesoporous materials.Herein, formation of such mesoporous structure is attributed to the aggregation of the primary nanoparticles.The surface area and median pore diameters derived from isotherm are 132 m 2 g −1 and 20 nm, respectively.As for the rutile nanorods, the hysteresis loop appears at 0.78 < P/P 0 < 1.0 in the isotherms, corresponding to the filling of textural macropores produced by rutile nanorods aggregates.The surface area of the as-obtained rutile sample is 83 m 2 g −1 .
The data presented above demonstrate that the as-adopted nonaqueous solventhermal process enable highly flexible manipulation of the crystal phase and morphology of TiO 2 nanomaterials by reacting of TiCl 4 with single alcohol.More importantly, the synthesis route is one-step, template-free, surfactant-free, as well as performed at low temperature.The reaction can be described as follows [18,19]: At the initial stage of mixing, TiCl 4 reacts with butanol producing TiCl 4−x (CH 3 CH 2 CH 2 CH 2 O) x , CH 3 CH 2 CH 2 CH 2 Cl, as well as HCl.Owing to the steric hindrance of butanol, the reaction is not vigorous and so is the evolution of HCl.Therefore, a certain amount of HCl can be dissolved in butanol through forming oxonium salts between HCl and butanol.Then the dissolved HCl can act as catalyst for the etherification of butanol in the system, and water is produced as a by-product.The hydrolysis of TiCl 4 that follows can produce different Ti(OH) 4−x Cl x species because the ligand field strength of OH − is higher than that of Clions.In the autoclaving stage at elevated temperatures, the polycondensation of Ti(OH) 4−x Cl x species will easily lead to extensive formation of the Ti-O-Ti network and then crystallization.Moreover, [OC 4 H 9 − ] can serve as a barrier against the hydrolysis of TiCl 4 .And this slow reaction rate is also the main reason for the growth of TiO 2 nanostructures with good crystallinity.The results that pure phase of anatase nanoparticles and rutile nanorods are selectively produced at high and low molar ratio of TiCl 4 to butanol, respectively, seems to not agree with the known chemistry in the synthesis of TiO 2 materials in aqueous phase.It has been reported that higher acidity in solution (high molar ratio of TiCl 4 to H 2 O) could result in the formation of the rutile phase rather than anatase phase.Herein, the difference may be caused by following factors: firstly, when the amount of TiCl 4 reactant is fixed, the retention amount of HCl in the reaction system increases as the amount of butanol reactant added increases.The observation that a white acid fog is released from the teflonline tube at high molar ratio of TiCl 4 to butanol rather than low molar ratio after opening the autoclave when the autoclave is cooled naturally to room temperature indirectly confirms this.As a result, selective formation of rutile phase is favored in the case of larger amount of HCl retained in the system.Secondly, the steric hindrance of butanol not only serves as a barrier against the hydrolysis of TiCl 4 in water but also has the ability to tune the growth behavior of TiO 2 crystals.This has been confirmed by other experimental results.That is, when butanol is replaced by ethanol while other conditions are kept unchanged, only anatase TiO 2 is obtained at both high and low molar ratio of TiCl 4 to ethanol.Also, one experiment employing the final solution volume as the condition for rutile synthesis but keeping the molar ratio of TiCl 4 to butanol as the condition for anatase synthesis is performed.The result shows that pure anatase phase of TiO 2 is obtained again, suggesting that HCl retained in the system plays important role in the synthesis and the effect of the initial volume of the reactants solution in the autoclave can be neglected.Therefore, the difference of TiO 2 phase and morphology may be attributed to the contribution of the equilibrium concentration of H + , Cl − , and steric hindrance effect of butanol.The photocatalytic degradation of RB has been chosen as a model reaction to evaluate the photocatalytic activities of the present samples, and it is compared with that of commercial Degussa P25. Figure 4 shows the decolorization curves of RB on different catalysts.The photocatalytic control experiments indicate that there is no appreciable decolorization of RB over TiO 2 either in the absence of UV irradiation or in the absence of the catalyst.However, the absorption peaks corresponding to RB at 554 nm diminishes gradually as the exposure time is extended in the presence of different TiO 2 samples under UV light.The activities of TiO 2 samples are even higher than that of P25.99, 80, and 60% decolorization of RB are observed after 30 min UV light irradiation for anatase nanoparticles, rutile nanorods, and P25, respectively.Factors including high surface area, small crystal size, and high crystallinity are able to account for the high photocatalytic activities of as-synthesized TiO 2 samples.High surface area can provide more active sites and adsorb more reactive species.Moreover, small crystal size can shorten the route for an electron migrates from the conduction band of TiO 2 to its surface.Meanwhile, the high crystallinity means few defects in photocatalysis.It is well known that lattice defects may act as recombination centers for photoinduced electrons and holes, thus significantly reducing the net photocatalytic activity.Therefore, the as-synthesized highly crystallized anatase nanoparticles and rutile nanorods having diameter of only 5-10 and 10-20 nm, respectively, as well as high surface area of 132 and 83 m 2 g −1 , respectively, possess high photocatalytic activity than that of P25 (a mean particle size of 30 nm and specific surface area of 50 m 2 g −1 ).

Conclusion
Summarizing, this work demonstrates that the nonaqueous technique of autoclaving treatment of TiCl 4 /butanol mixtures can be developed as a new approach for the synthesis of TiO 2 nanocrystallites with controlled crystal size, phase and morphology.The key factor determining the crystal phase and morphology is the molar ratio of TiCl 4 to butanol.
HCl plays an important role in determining the possible formation mechanism and the crystal structures.The asadopted approach is one-step, template-free, surfactant-free, as well as performed at low temperature, enlightening a way to synthesize other metal oxides.

Figure 1 :
Figure 1: XRD patterns and Raman spectra of TiO 2 samples prepared at high molar ratio of TiCl 4 to butanol (A) and low molar ratio of TiCl 4 to butanol (B).
Figure 2(b) shows the HRTEM image of individual particles.The fringes observed are corresponding to the interplanar distances, which agree well with the lattice spacing of the (101) low-index facets of the anatase TiO 2 .Moreover, the well-developed lattice fringes clearly indicate the high crystallinity.Figure 2(c) shows the TEM image of rutile crystallites.It can be seen that the crystallites show rod-like feature and the nanorods have a diameter of 10-20 nm.Moreover, most of the nanorods have sharp tips, and the measured angle between the tip edge and mainbody side of the nanorod is about 130 • , indicating the high crystallinity of rutile nanorods.Since contact angles between (110) and (101), (110) and (111), and (110) and (221) planes are calculated to be 112.5 • , 132.3 • , and 151.2

Figure 4 :
Figure 4: The time course of the photodegradation of RB on UV light-irradiated TiO 2 catalysts; inset shows the adsorption curve in dark over different catalysts.