Solvent-controlled synthesis of TiO 2 1D nanostructures: Growth mechanism and characterization Kajari Das à , Subhendu K. Panda, Subhadra Chaudhuri Department of Materials Science and DST Unit of Nanoscience, Indian Association for the Cultivation of Science, Raja S. C. Mallick Road, Kolkata, WB 700 032, India article info Article history: Received 18 February 2008 Received in revised form 19 May 2008 Accepted 26 May 2008 Communicated by K. Nakajima Available online 3 July 2008 PACS: 78.67.Ch Keywords: A1. Growth models A2. Growth from solutions B1. Nanomaterials B2. Semiconducting materials abstract One-dimensional (1D) anatase TiO 2 nanostructures such as nanorods, nanowires and nanotubes with different aspect ratios were synthesized by a simple solvothermal process. The influence of the different organic solvents and the reaction time on the morphology, size and the formation of the nanostructures were investigated. The anatase TiO 2 precursor powder reacted with highly alkaline aqueous solution, yielding layered sodium titanate nanosheets. These nanosheets transformed to different 1D sodium titanates nanostructures like nanorods, nanowires and nanotubes in the different solvents i.e. highly alkaline aqueous solution, highly alkaline water–ethanol and highly alkaline water–ethylene glycol mixed solvent, respectively. Acid treatment of these 1D sodium titanates resulted hydrated titanates and finally dehydration by calcinations at 500 1C in air gave the products retaining the morphology. The synthesized samples were characterized with XRD, SEM and TEM. All the 1D nanostructures showed intense and sharp absorption spectra indicated that the products were almost defect free. Photoluminescence studies of the nanostructures showed photostable UV emission properties that arise from the band-edge free excitation. & 2008 Elsevie r B.V. All rights reserved. 1. Introduction In the past few years, the design and fabrication of the nanostructured semiconductors based on metal oxides have received considerable attention due to their interesting physical and chemical properties, and their potential applications in industry and technology [1–5]. TiO 2 is an n-type wide band-gap oxide semiconductor used for variety of applications such as dye- sensitized solar cell, environmental purification, nanodevices, gas sensors, and photocatalysts [6–13]. Many techniques such as sol–gel processing with electrophoretic deposition, spin-on process, sol–gel template method, metalorganic chemical vapor deposition, anodic oxidative hydrolysis, sonochemical synthesis, inverse microemulsion method, and molten salt-assisted pyrolysis routes have been developed to synthesize different TiO 2 nanos- tructures [14–21]. Recently, the reaction between different TiO 2 precursors and a concentrated NaOH solution under moderate hydrothermal method [22] is observed to be an effective approach to prepare 1D nanostructures of titania. However, the main attention is directed towards the control over structure and morphology of titania only by varying the reaction temperature, reaction time and pH of the solution during hydrothermal treatment, while an important experimental parameter, solvent, has rarely been deliberately controlled to achieve different 1D nanostructures of titania. Solvothermal process is the most useful technique to synthesize nanocrystalline materials with different morphologies and sizes where properties of the solvents like density, viscosity and diffusion coefficient change dramatically and the solvent behaves much differently from that expected at the normal conditions. Consequently, the solubility, diffusion process and the chemical reactivity of the reactants are greatly enhanced. The detailed studies of the effect of different organic solvents on the morphologies of the TiO 2 nanocrystals under solvothermal conditions have not been reported so far. In this paper we have synthesized phase pure anatase TiO 2 in different nanoforms such as single-crystalline nanorods, nano- wires and nanotubes using a solvothermal route and investigate the effects of the different solvents and reaction time on the shape, size, and the optical properties of the nanostructures. 2. Experimental section All the reactants and the solvents were of analytical grade and were used without further purification. In a typical procedure, 274mg of pure anatase phase TiO 2 bulk-powder was mixed with 10 N NaOH (19.2 g of NaOH in 48 ml water) aqueous solution of pH ¼ 12.77 under constant magnetic stirring for 1 h. A milky ARTICLE IN PRESS Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/jcrysgro Journal of Crystal Growth 0022-0248/$ - see front matter & 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jcrysgro.2008.05.039 à Corresponding author. Tel.: +9133 2473 4971; fax: +9133 2473 2805. E-mail address: kajari26jul@rediffmail.com (K. Das). Journal of Crystal Growth 310 (2008) 3792– 3799 white solution was appeared, which was then transferred to a Teflon-lined stainless steel autoclave with 60 ml capacity and heat treated at 150 1C for 16 h. The sample obtained was denoted by S1. Likewise, to investigate the effect of the co-solvent on the morphology and the properties of TiO 2 nanostructures, 19.2 g of NaOH was dissolved in 48 ml mixed solvent of water and an organic solvent (ethylene glycol or ethanol or polyethylene glycol- 300 (PEG-300)) with a volume ratio 1:1. 274 mg TiO 2 precursor powder was then mixed with the highly alkaline solvent mixer under constant magnetic stirring for 1 h and the solution was heat treated at 150 1C for 16 h in the autoclave. The samples obtained were named as S2, S3, and S4, respectively (Table 1). Since the solubility of NaOH decreases in the organic solvents than the water, the pH values of the mixed solvents were less than that of pure alkaline aqueous solution. For all the cases, the autoclave chambers were air-cooled to room temperature after the reac- tions. The formed white precipitates were recovered by centrifu- gation and washed several times with distilled water. An ultrasonic treatment of the products with 0.1 N HCl solution was carried out at pH$7, and the precipitates were finally calcinated at 500 1C for 4 h in air. To investigate the formation mechanism of the different nanostructures, some additional experiments were also carried out at 150 1C for different time intervals, i.e. for 30 min, 1, 8, 16 and 24 h in pure alkaline aqueous solution and for 30 min, 1, 8 and 16 h in different alkaline mixed solvents. The crystalline phases of the products were determined by X-ray powder diffraction by using a Seifert 3000P diffractometer with Cu Ka radiation (l ¼ 1.54 A ˚ ). The morphologies of the samples were studied by a scanning electron microscope (SEM; Hitachi S-2300). Microstructural properties were obtained using transmission electron microscope and high-resolution transmis- sion electron microscope (TEM and HRTEM; JEOL 2010). For the TEM observations, the powders were dispersed in 2-propanol and ultrasonicated for 15 min. A few drops of this ultrasonicated solution were taken on a carbon-coated copper grid. FTIR spectra recorded in the range 4000–400 cm À1 with a Shimadzu model FTIR spectrometer using KBr wafer. Optical absorbance of the samples was recorded by a UV–vis–NIR spectrophotometer (Hitachi, U-3410). Photoluminescence (PL) measurements were carried out at room temperature with a Fluorescence spectro- meter (F-2500) using 310 nm excitation wavelength. 3. Results and discussion The crystal structure, morphology and size of the synthesized products determined by XRD, SEM, and TEM are listed in Table 1. Fig. 1(a) shows the XRD pattern of all the samples, which revealed ARTICLE IN PRESS 20 30 40 50 60 70 80 S4 S3 S2 S1 (204) (211) (105) (200) ( 004) (101) Intensity (arb.unit) FWHM S1 = 0.465° FWHM S2 = 0.485° FWHM S3 = 0.519° FWHM S4 = 0.586° 10 20 30 40 50 60 70 80 III II I # # # # # # # # #H 2 Ti 3 O 7 # # # # # # *Na 2 Ti 3 O 7 * * * * Intensity (arb.unit) 2 θ (in degree) 2θ (in degree) Fig. 1. XRD spectra of (a) samples S1–S4 synthesized in the different solvents and (b) the TiO 2 precursor powder (I), synthesized intermediate products washing with water (II) and obtained after ultrasonic acid treatment (III). Table 1 The effect of experimental parameters on the morphologies, sizes and band gaps of the TiO 2 nanoforms Sample name Solvent Reaction temperature (1C) Reaction time (h) Morphology Size (nm) Band gap (eV) S1 Water 150 16 Nanorods D ¼ 9, L ¼ 120–170 3.69 S2 Water: ethylene glycol (1:1) 150 16 Nanotubes D ¼ 11, L ¼ 100–140 3.67 S3 Water: ethanol (1:1) 150 16 Ultralong nanowires D ¼ 4, L ¼ several micrometers 3.73 S4 Water: PEG- 300 (1:1) 150 16 Nanoparticles 16 3.63 K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–3799 3793 that irrespective of the variation of the solvents, all the peaks corresponding to the reflections from (10 1), (0 0 4), (2 0 0), (1 0 5), (211), and (20 4) planes of anatase tetragonal TiO 2 were observed, which are well matched with the standard reported values (JCPDS file No. 21-1272). The (101) peaks of all the samples were fitted to the Gaussian curves and the FWHM (full-width at half-maximum) were estimated, as mentioned in Fig. 1(a). It is noted that the FWHM were decreased in the order S1oS2o- S3oS4, which may be attributed to the improvement of crystal- linity of the samples in the order S14S24S34S4. Fig. 1(b) shows the XRD patterns of the TiO 2 precursor and the two intermediate products obtained after washing the as synthesized products with water and ultrasonicating with HCl. The spectrum II indicates the presence of the Na 2 Ti 3 O 7 and H 2 Ti 3 O 7 in the products washing with water. After treatment with 0.1 N HCl, the products were completely transformed to the hydrated titanate (H 2 Ti 3 O 7 )by substitution of the Na + by H + , which is clearly revealed from the XRD spectrum III. These two intermediate phases were appeared for all the samples. Final products were obtained after calcination of the hydrated titanates at 500 1C for 4 h. The growth process of the different 1D TiO 2 nanostructures was monitored by SEM and TEM. Fig. 2(a–c) shows the SEM images of the samples S1–S3, respectively, which reveal that the reaction of TiO 2 precursor with highly alkaline aqueous solution and highly alkaline mixed solvents gave 1D nanostructures. Although the shapes of these 1D nanostructures are not clear by the SEM images, the TEM images later clearly show nanostruc- tures’ morphologies. Fig. 2(d) shows the SEM image of the 1D nanostructures obtained in highly alkaline aqueous solution at 150 1C for 24 h. By comparing the images in Fig. 2(a) and (d), it may be concluded that the diameters of the 1D nanostructures obtained in highly alkaline aqueous solution decrease with increasing reaction time. Fig. 3(a–c) shows the low magnification TEM images of the TiO 2 samples obtained in highly alkaline aqueous solutions at 150 1C for 8, 16 and 24 h, respectively, which clearly reveal that the diameters of the nanorods decreased, whereas the lengths of the rods increased with the increasing reaction time. Ultimately, the nanorods transformed to nanowires when the reaction time was greater than 24 h. The possible explanation of this transformation is given later in the formation mechanism part. In Fig. 3(d), the low magnification TEM image of the sample S2 gives the nanotubes having inner diameter 6 nm and outer diameter 11 nm; and length 100–140 nm. Fig. 3(e) shows the HRTEM image of a nanotube. The number of the walls counted from the two sides of the tubes was not identical. Moreover, the wall thickness generally varied along the tube. The left inset in Fig. 3(e) is an enlarged picture of the tube wall as marked by the white circle. The periodicity of the fine fringes was 0.37 nm, which indicates the (1 0 1) plane of the anatase TiO 2 . The interspacing of the tube walls was 0.77 nm. Crystal structures of the TiO 2 nanorods (sample S1) were also studied through HRTEM, which is shown in Fig. 3(f). The fringes parallel to the nanorod axis correspond to an interplanar distance of about 0.35nm, which is characteristic of (10 1) plane of TiO 2 in the anatase phase. The clear lattice fringes confirm the nanorods are single-crystalline and defect free. The inset of Fig. 3(f) shows the FFT pattern of the nanorod, which also indicates the (1 0 1) plane of the anatase TiO 2 and confirms the single-crystalline nature of the nanorod. Fig. 3(g) shows the low-magnification TEM image of the ultralong nanowires of sample S3. Fig. 3(h) gives the TEM image of the nanoparticles of 16 nm obtained for sample S4. The inset of Fig. 3(h) is the HRTEM image of a nanoparticle showing lattice fringes (d ¼ 0.342 nm) corresponding to the (1 0 1) lattice plane of the anatase TiO 2 . The diameters and lengths of the different nanostructures calculated from TEM images are shown in Table 1. SEM and TEM images revealed that in all the solvents except PEG-300, 1D nanostruc- tures of different morphologies such as nanorods, nanowires and nanotubes with different diameters and lengths were formed. To understand the formation of the different 1D nanostruc- tures controlled experiments were carried out and the products at different time intervals were examined. SEM images of the ARTICLE IN PRESS Fig. 2. SEM micrographs for (a–c) the samples S1, S2 and S3, respectively and (d) the product obtained in highly alkaline aqueous solution at 150 1C for 24 h. K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–37993794 intermediate products at different time intervals are shown in Fig. 4. The reaction of the TiO 2 precursor with the highly alkaline aqueous solution at room temperature gave single layered titanate nanosheets. Fig. 4(a) shows the single layered nanosheets and some unreacted TiO 2 particles. At high temperature (150 1C) and pressure in autoclave, the splitting of the nanosheets occurred within 30 min of the reaction, which is clearly shown in Fig. 4(b). Fig. 4(c) reveals that the formation of the network of the nanorods began with the further splitted nanosheets at 150 1C for 1 h. Finally, the crystalline anatase nanorods were obtained; when the reaction time was 8 h as shown in Fig. 4(d). The bended multilayered titanate nanosheets were obtained when TiO 2 precursor reacted with the highly alkaline water–ethylene glycol mixed solvent at room temperatures, shown in Fig. 4(e). The splitting and simultaneous rolling of these bended multilayered nanosheets occurred at 150 1C for 30 min, which is clearly shown in Fig. 4(f). Fig. 4(g) shows that the product obtained for the reaction time 1 h gave nanotubes (arrow I) coexisting with splitted and bended multilayered nanosheets (arrow II), left inset of the figure shows the enlarged picture of the agglomerated nanotubes; but the product for 8h (Fig. 4(h)) gave complete formation of the nanotubes. Ultimately, well crystalline nanotubes were observed for the reaction time 16 h. In Fig. 4(i), the SEM image of the nanoparticles obtained in the highly alkaline water–PEG-300 mixed solvent at room temperature indicates that no sheet-like morphology was formed in this mixed solvent. Fig. 5 shows the FTIR spectra of the samples S1–S4. All the spectra show broad bands at 3435 and 1635 cm À1 , which correspond to the presence of the OH group and water, absorbed on the surface of the TiO 2 samples [23–25]. In addition, for samples S2–S4, two bands at 2860 and 2931 cm À1 are observed, which are attributed to the symmetric and asymmetric CH 2 stretching vibrations, respectively [26], coming from the co- solvents. The peak at 2348 cm À1 was due to CO 2 and was not related to the samples. Also a band appears around 453cm À1 for all the samples, which corresponds to a Ti–O band of anatase phase of TiO 2 [27,28]. The intensity of this band increases in the order S14S24S34S4, revealing that the crystallinity of the products improve in this order, which is in good agreement with the XRD and TEM observations of the samples. ARTICLE IN PRESS Fig. 3. Low-resolution TEM images of the (a–c) TiO 2 samples obtained in highly alkaline aqueous solution at 150 1C for 8, 16 and 24 h, respectively, and (d) TiO 2 nanotubes (sample S2); (e) HRTEM image of the sample S2 and the inset shows the enlarged picture of the tube wall; (f) HRTEM image of sample S1 and the inset shows the FFT pattern; (g) low-resolution TEM image of the sample S3; (h) low-resolution TEM image of sample S4 and the inset shows the HRTEM image of a nanoparticle. K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–3799 3795 The growth process toward the formation of 1D nanostructures can be explained from the microstructural and crystallographic evidence. Fig. 6 is the simplified schematic model showing the different stages for the growth of different TiO 2 nanostructures. TiO 6 octahedron is the basic unit of the crystalline structure of the anatase TiO 2. The three-dimensional framework of the oxide is build up by sharing the vertice edges of the octrahedra. When TiO 2 precursor reacts with highly alkaline NaOH solution, some of the Ti–O–Ti bonds of the raw material are broken and layered titanates composed of octahedral TiO 6 units with Na + metal ions are formed in the form of thin small sheets. The formation of the intermediate nanosheets of sodium titanate (Na 2 Ti 3 O 7 ) phase is confirmed by XRD spectrum and SEM image. The titanate and anatase have common structural features: both crystal lattices consist of the octahedral sharing four edges and the zigzag ribbons [29]. The strength of Na–O bonding is weaker than that of Ti–O bonding in Na 2 Ti 3 O 7 layered structure. These Na–O bonds may break at the high temperature and pressure in the autoclave and the single layered nanosheets split to form the nanorods. As the reaction time increases, further splitting of the nanosheets may increase the aspect ratio of the nanorods and hence after 24 h reaction transformation of the nanorods to nanowires was observed. After washing with water and ultrasonic acid treatment, hydrated titanates (H 2 Ti 3 O 7 ) are formed by the substitution of Na + by H + , which is revealed by the XRD. The final anatase TiO 2 products are obtained when dehydration of the hydrated titanates are occurred at the time of the calcinations of the products at 500 1C for 4 h [30]. The SEM images of the sodium ARTICLE IN PRESS Fig. 4. SEM micrographs for (a–d) the intermediate products obtained in highly alkaline aqueous solution at room temperature and at 150 1C for 30min, 1 and 8h, respectively; (e–h) the intermediate products in highly alkaline water–ethylene glycol mixed solvent at room temperature and at 150 1C for 30 min, 1 h, 8 h, respectively and inset of (g) shows the enlarged picture of the agglomerated nanotubes; (i) the intermediate product in highly alkaline water-PEG-300 mixed solvent at room temperature. 3500 3000 2500 2000 1500 1000 500 2348 453 S1 S4 S3 S2 1635 2860 2931 Transmittance (arb. unit) Wave number (cm -1 ) 3435 Fig. 5. FTIR spectra of the samples S1–S4. K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–37993796 and hydrated titanates also indicate that the morphology of the intermediate products was almost same as the final anatase TiO 2 products. It is well known that the solvent plays an important role to control the morphology of the product. In our experiments, the polarity and coordinating ability of a co-solvent have strong effect on the solubility, reactivity, and diffusion behavior of the reactants, thus ultimately influencing the structure and morpho- logical features of the resulting products. Zhang et al. [31] considered that the bending of the multilayered titanate nanosheets occurred in highly alkaline aqueous solution at 130 1C due to the imbalance of ion concentration on two different sides of the nanosheets in an asymmetrical chemical environ- ment. In our experiment, when ethylene glycol was used as the co-solvent with highly alkaline aqueous solution, multilayered sodium titanate nanosheets were formed due to the chelating property of ethylene glycol. TiO 6 octrahedra may coordinate with glycol to form chain-like structures [32], whereas the NaOH may form titanate nanosheets by sharing the vertice edges of the octrahedra. The negative charge of the titanate layers coordinated with glycol on the side underneath the surface is neutralized by Na + in the interlayer space. The Na + on the top surface of the layers may undergo frequent collision with OH À from the solution. The electronegativity of the titanate layers coordinated with glycol is greater than that of titanate layers in normal alkaline aqueous solution. The Na + deficiency on the upper surface of the multilayered titanates coordinated with glycol bends the na- nosheets to minimize the excess surface energy. The bending of the multilayered nanosheets is confirmed by the SEM image. At high temperature and pressure in the autoclave, the splitting and simultaneous rolling of the sheets give nanotube structures. Another reason for the rolling of the splitted multilayered nanosheets at high temperature may be due to the mechanical tensions that arise during the process of dissolution/crystal- lization in nanosheets [33]. Due to chelating property of the ethylene glycol, the slow nucleation rate and the very fast growth rate of the crystal may favor the rolling process of the sheets. During spontaneous crystallization and rapid growth of layers, it is possible that the width of the different layers varies, which gives rise to excess surface energy. In order to decrease the excess surface energy, the rolling of splitted nanosheets occurs. Bavykin et al. [34] described that the nanotubes were formed in highly alkaline aqueous solutions at low temperature (110–150 1C) and nanofibres were formed at the temperature 150 1C. These nanotubes were transformed to the nanorods as calcinated at 500 1C. These observations revealed that the nanotube structures are thermally unstable than the nanorods and nanofibres. In our present work, the nanotubes were obtained at 150 1C using ethylene glycol as the co-solvent and the structures remained unchanged after calcination at 500 1C for 4 h, which indicate that the nanotubes formed by the ethylene glycol assisted method are more thermally stable than the normal alkaline aqueous solution. The formation of ultralong nanowires may arise from the very fast growth rate and the rapid splitting of the sodium titanate nanosheets in highly alkaline water–ethanol mixed solvent. The rapid growth rate and the fast splitting of the nanosheets may cause various defects in the crystals, which were clearly revealed from the optical absorbance and photoluminescence studies discussed later. For the co-solvent PEG-300, the intermediate products were investigated and in this case no sheet-like morphology was observed and finally nanoparticles of 16 nm were obtained. The SEM image shown in Fig. 4(i) confirms the presence of nanoparticles instead of the nanosheets. Pre- viously, Zhu et al. [35] reported the formation of TiO 2 mesoporous structures using PEG-200 as the solvent in a hydrothermal method, which destroyed and grains were appeared when the reactions time was increased. Also, due to the presence of the inorganic polymer, the degree of the crystallization was observed to be poor. One-dimensional nanostructures were not formed, when we used only pure organic solvent without alkaline aqueous solution. Thus it may be concluded that in highly alkaline aqueous solution formation of the 1D ARTICLE IN PRESS Highly alkakline aqueous solution Nanosheets & unreacted particles High temperature & pressure High temperature & pressure Bended multilayered nanosheets & unreacted particles Nanosheets & unreacted particles Highly alkaline water-EG mixed soluiton Highly alkaline water-ethanol mixed soluiton TiO2 bulk powder Splitting Splitted nanosheets Reaction time > 8h Reaction time > 30 min Reaction time > 24h Nanorods Nanowires Reaction time > 16h Nanotubes Splitting & simultaneous rolling of the multilayred nanosheets High temperature & pressure Splitting Ultra long nanowires Fig. 6. Schematic diagram showing formation of different TiO 2 nanostructures under different synthesis conditions. K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–3799 3797 nanostructures takes place and the organic solvents used with highly alkaline aqueous solution only control the shapes of the 1D nanostructures. Fig. 7(a) shows the optical absorption spectra of the samples S1–S4 prepared in different solvents. It is interesting to note that the 1D nanostructures show intense and sharp absorption spectra indicating the formed 1D nanostructures were almost defect free and high crystalline in nature. The spectra also show that the sharpness of the absorption peaks near the band edge increased in the order S14S24S34S4, indicates that well crystalline products were obtained using water and water–ethylene glycol mixed solvent. Fig. 7(b) shows the differential absorbance spectra (dA/dl versus l) of the TiO 2 precursor powder and the sample S1. The band gaps of all the other samples were also determined from differential absorbance spectra, which are shown in Table 1. The band gaps of the TiO 2 nanostructures were varied from 3.63 to 3.73 eV, which are slightly blue shifted from the bulk value of anatase phase of the TiO 2 (3.21 eV) may be due to the nanosize effect. The room temperature PL spectra of all the samples were recorded with 310 nm excitation. Fig. 8 shows the PL spectra of all the samples. The signal at $372 nm could be attributed to emission peak from band edge free excitation [36]. The red shift takes place for the samples S3 and S4, which may be due to the oxygen vacancies [37]. In addition, another broad band between 400 and 510 nm was observed. The origin of this broad band was already reported by Daude and co-workers [38]. The lowest energy allowed phonon-assisted transitions of the anatase TiO 2 from center to the edge of the Brillouin zone are the indirect transitions, namely G 1b -X 2b (406 nm) and G 1b -X 1a (426 nm). The emissions at 455 and 504 nm are due to the transitions from intragap energy levels implicating lattice defects and oxygen vacancy. The PL results also reveal that the nanostructures showed photostable UV emission properties and also clearly indicate that the almost defect free, well crystalline TiO 2 samples were prepared at 150 1C using highly alkaline aqueous solution and highly alkaline water–ethylene glycol mixed solvent. 4. Conclusions Different 1D nanostructures such as nanorods, nanowires, and nanotubes of TiO 2 in anatase phase were synthesized by a simple solvothermal method. We have demonstrated the control over the structures, sizes and morphologies of the products by controlling the reaction temperature, time and by using different mixed solvents. In such synthesis, the reaction of the anatase TiO 2 precursor powder with highly alkaline aqueous solution produced layered sodium titanate nanosheets at room temperature. The splitting of the single layered nanosheets at high temperature and pressure in the autoclave gave the nanorods. As the reaction time increases, further splitting of the nanosheets may increase the aspect ratio of the nanorods and hence after 24 h reaction transformation of the nanorods to nanowires was observed. When ethylene glycol was used as the co-solvent with highly alkaline aqueous solution, multilayered sodium titanate na- nosheets were formed at room temperature and the splitting and simultaneous rolling of these sheets gave nanotube structures at high temperature and pressure in the autoclave. The ultralong nanowires were obtained in highly alkaline water–ethanol mixed solvent, whereas nanoparticles were formed in highly alkaline water–PEG-300 mixed solvent. A plausible mechanism for the ARTICLE IN PRESS 400 500 600 800 S1 S2 S3 S4 Absorbance (arb.unit) TiO 2 precursor 3.21eV 3.69eV S1 300 λ (nm) 700 400 500 600 800300 λ (nm) 700 dA/dλ (arb.unit) TiO 2 precursor Fig. 7. (a) Optical absorbance spectra of all the samples, (b) differential absorbance spectra (dA/dl versus l) of the TiO 2 precursor powder and the sample S1. 360 420 450 480 S4 S3 S1 S2 Intensity (arb.unit) λ (nm) 330 390 510 Fig. 8. Photoluminescence spectra of all the samples. K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–37993798 formation of different 1D nanostructures has been proposed. The final anatase TiO 2 products were obtained by dehydration of the hydrated titanates at 500 1C in air, where retention of the morphologies of the products was observed during the phase transformation. TEM and SEM observations revealed that the solvent is the most crucial factor in determining the morphologies of the products and the reaction time is the most influential factor in controlling the diameters and lengths of the products. Due to the phase purity and well crystalline nature of the TiO 2 1D nanostructures, they will be certainly applicable in the fields of photocatalysis, electrocatalysis, lithium batteries, hydrogen sto- rage, and solar-cell technologies. Acknowledgments This paper is dedicated to the memory of Prof. Subhadra Chaudhuri whose continuous support and effective guidance have made this work possible. The authors thank Mr. K. K. Das of IACS for recording the SEM micrographs. References [1] D. Li, Y. Xia, Adv. Mater. 16 (2004) 1151. [2] X. Huang, C. Pan, J. Crystal Growth 306 (2007) 117. [3] C.N.R. Rao, F.L. Deepak, G. Gundiah, A. Govindaraj, Prog. Solid State Chem. 31 (2003) 5. [4] C.S. Kim, B.K. Moon, J.H. Park, B.C. Choi, H.J. Seo, J. Crystal Growth 257 (2003) 309. [5] D.S. Seo, J.K. Lee, H. Kim, J. Crystal Growth 229 (2001) 428. [6] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Langmuir 14 (1998) 3160. [7] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Adv. Mater. 11 (1999) 1307. [8] Q. Chen, W.Z. Zhou, G.H. Du, L.M. Peng, Adv. Mater. 14 (2002) 1208. [9] G.H. Du, Q. Chen, R.C. Che, Z.Y. Yuan, L.P. Peng, Appl. Phys. Lett. 79 (2001) 3702. [10] B.D. Yao, Y.F. Chan, X.Y. Zhang, W.F. Zhang, Z.Y. Yang, N. Wang, Appl. Phys. Lett. 82 (2003) 281. [11] Z.Y. Yuan, W. Zhou, B.L. Su, Chem. Commun. (2002) 1202. [12] J.G. Yu, J.C. Yu, W.K. Ho, L. Wu, X.C. Wang, J. Am. Chem. Soc. 126 (2004) 3422. [13] D.Y. Zhang, L.M. Qi, Chem. Commun. (2005) 2735. [14] S.J. Limmer, G. Cao, Adv. Mater. 15 (2003) 427. [15] D.K. Yi, S.J. Yoo, D. -Y. Kim, Nano Lett. 2 (2002) 1101. [16] Z. Miao, D. Xu, J. Ouyang, G. Guo, X. Zhao, Y. Tang, Nano Lett. 2 (2002) 717. [17] J J. Wu, C C. Yu, J. Phys. Chem. B 108 (2004) 3377. [18] Y. Lei, L.D. Zhang, J.C. Fan, Chem. Phys. Lett. 338 (2001) 231. [19] Y. Zhu, H. Li, Y. Koltypin, Y.R. Hacohen, A. Gedanken, Chem. Commun. (2001) 2616. [20] G. Wang, G. Li, Eur. Phys. J. D 24 (2003) 355. [21] C. Xu, Y. Zhan, K. Hong, G. Wang, Solid State Commun. 126 (2003) 545. [22] Y. Lan, X.P. Gao, H.Y. Zhu, Z.F. Zheng, T.Y. Yan, F. Wu, S.P. Ringer, D.Y. Song, Adv. Funct. Mater. 15 (2005) 1310. [23] T. Nakayama, J. Electrochem. Soc. 141 (1994) 237. [24] E. Sanchez, T. Lopez, R. Gomea, A. Morales, O. Novaro, J. Solid State Chem. 122 (1996) 309. [25] Z. Ding, G.Q. Lu, P.F. Greenfield, J. Phys. Chem. B 104 (2000) 4815. [26] J. Joo, S.G. Kwon, T. Yu, M. Cho, J. Lee, J. Yoon, T. Hyeon, J. Phys. Chem. B 109 (2005) 15297. [27] S. Music, M. Gotic, S. Ivanda, S. Popovic, A. Turkovic, R. Trojko, A. Seculic, K. Furic, Mater. Sci. Eng. B 47 (1997) 33. [28] M. Ocana, V. Fornes, J.V. Serna, J. Solid State Chem. 75 (1988) 364. [29] B. Wang, Y. Shi, D. Xue, J. Solid State Chem. 180 (2007) 1038. [30] X. Jing, Y. Li, Q. Yang, Q. Yin, Mat. Sci. and Eng. B 110 (2004) 18. [31] S. Zhang, L M. Peng, Q. Chen, G.H. Du, G. Dawson, W.Z. Zhou, Phys. Rev. Lett. 91 (2003) 256103. [32] X. Jiang, Y. Wang, T. Herricks, Y. Xia, J. Mater. Chem. 14 (2004) 695. [33] D.V. Bavykin, V.N. Parmon, A.A. Lapkin, F.C. Walsh, J. Mater. Chem. 14 (2004) 3370. [34] D.V. Bavykin, J.M. Friedrich, F.C. Walsh, Adv. Mater. 18 (2006) 2807. [35] R. Tan, Y. He, Y. Zhu, B. Xu, L. Cao, J. Mater. Sci. 38 (2003) 3973. [36] L.D. Zhang, C.M. Mou, Nanostruct. Mater. 6 (1995) 831. [37] I. Justicia, P. Ordejon, G. Canto, J.L. Mozos, J. Fraxedas, G.A. Battiston, R. Gerbasi, A. Figueras, Adv. Mater. 14 (2002) 1399. [38] N. Daude, C. Gout, C. Jouanin, Phys. Rev. B 15 (1977) 3229. ARTICLE IN PRESS K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–3799 3799 . Solvent-controlled synthesis of TiO 2 1D nanostructures: Growth mechanism and characterization Kajari Das à , Subhendu K. Panda, Subhadra. HRTEM image of a nanoparticle. K. Das et al. / Journal of Crystal Growth 310 (2008) 3792–3799 3795 The growth process toward the formation of 1D nanostructures can