Materials Chemistry and Physics

Transcription

1 Materials Chemistry and Physics 137 (2013) 928e936 Contents lists available at SciVerse ScienceDirect Materials Chemistry and Physics journal homepage: Impact of reaction temperature, stirring and cosolvent on the solvothermal synthesis of anatase TiO 2 and TiO 2 /titanate hybrid nanostructures: Elucidating the growth mechanism Batakrushna Santara a, P.K. Giri a,b, * a Department of Physics, Indian Institute of Technology, Guwahati , India b Centre for Nanotechnology, Indian Institute of Technology, Guwahati , India highlights < Low temperature growth of 1-D anatase TiO 2 and hybrid TiO 2 /titanate nanostructures. < Impact of growth temperature, stirring and cosolvents on the evolved nanostructures. < Growth mechanism of the nanotubes and nanorods explored in details. < Nanoporous TiO 2 nanoribbons grown at 180 C have important practical applications. article info abstract Article history: Received 18 March 2012 Received in revised form 20 August 2012 Accepted 7 October 2012 Keywords: Semiconductor Nanostructures Chemical synthesis Electron microscopy Optical properties One-dimensional anatase TiO 2 and hybrid TiO 2 /titanate nanostructures are synthesized by a simple low temperature solvothermal route followed by the Na þ /H þ ion-exchange and final calcination process. We investigated the impact of reaction temperature, stirring conditions and cosolvent on the morphologies of the as-prepared nanostructures. Nanotubes and nanorods are formed in alkaline solution, while nanorods/nanowires and nanoporous nanoribbons are formed in alkaline watereethanol and alkaline watereethylene glycol mixed solvents, respectively. X-ray diffraction, Raman scattering and highresolution transmission electron microscopy studies are employed to identify the structure and phase composition. The formation of different morphologies of the as-synthesized nanostructures is investigated by field emission scanning electron microscopy and transmission electron microscopy. The growth mechanism and reaction process of the as-prepared nanostructures are explained based on the experimental observations. The photoluminescence, optical absorption and the tuning of band gap of the prepared samples are also studied. This work will be valuable for understanding the growth mechanism of various nanostructured TiO 2 and to explore the commercial applications of nanoporous nanoribbons of TiO 2. Ó 2012 Elsevier B.V. All rights reserved. 1. Introduction * Corresponding author. Department of Physics, Indian Institute of Technology, Guwahati , India. address: pravat_g@yahoo.com (P.K. Giri). The design and fabrication of one-dimensional (1D) TiO 2 nanorods, nanotubes and nanowires have attracted intensive interest due to their unique architectures, extraordinary physical and chemical properties and potential applications such as humidity sensor [1], environmental cleaning [2], lithium ion battery [3], hydrogen production [4] or storage [5] and dyesensitized solar cells [6]. To realize these promising high performance applications, the control of the highly desirable properties upon their inherent crystal structure, morphology, surface area, and porosity through their nanostructures and understanding the formation mechanism are challenging today. Considerable effort has been made to explore novel approaches from vapor-phase technique to solution-growth process to grow TiO 2 nanostructures with various properties by controlling their size and morphology. The typical vapor-phase technique requires high temperature and the control over morphology is not an easy task. Moreover, growth of anatase TiO 2 nanostructures that has more photocatalytic activity than the rutile is not possible through this technique. However, hydrothermal method is an effective /$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.

2 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e technique to synthesize TiO 2 nanostructures at relatively low temperature with optimal control over morphology, structure and phase composition without the requirement of templating. This approach requires neither expensive equipments nor specific chemicals, hence provide a more promising approach in terms of cost. In particular, it has been reported that the size, morphology and structural properties of TiO 2 nanostructures depend on the TiO 2 precursors and reaction parameters such as reaction temperature, reaction time and ph of the solution during the reaction [7,8], which can be achieved by this technique. After the first successful synthesis of titanate nanotubes by Kasuga et al. [9], extensive research has been carried out on the hydrothermal growth of titanate and TiO 2 nanorods, nanotubes and nanowires by adjusting various parameters within the hydrothermal system to determine their effects on the nanostructure formation and resultant morphology [10]. These parameters include reaction temperature, alkaline concentration, reaction duration, precursor phase, and crystallite size. Lan et al. [11] reported that nanotubes are formed after hydrothermal treatment of TiO 2 particles in 10 M NaOH solution at 150 C for 24 h, whereas nanoribbons appeared at higher temperature (180 C). Morgan et al. [12] reported that nanotubes are formed at lower temperature and at 200 C, nanotubes transformed to nanoribbons using anatase TiO 2 powder as precursor. It is accepted by several authors [13,14] that nanotubes are formed at mild hydrothermal temperatures in the range of 130e150 C, while only nanowires or nanoribbons are observed at higher temperature (>180 C). Additionally, titanate nanotubes can further transform into nanowires with prolonged reaction duration [15]. However, two additional important parameters such as cosolvent and stirring effect during the hydrothermal process have rarely been reported to achieve 1D titanate and TiO 2 nanostructures. More recently, Shen et al. [16] reported the effect of cosolvent, temperature and NaOH concentration on the morphology of titanate nanostructures. However, systematic study of solvent controlled nanostructures and the exact growth mechanism of hydrothermally synthesized titanate and TiO 2 nanotubes and nanowires is still a topic debated extensively in contemporary literature. In this study, we report on the temperature and solvent controlled hydrothermal growth of hybrid TiO 2 /titanate nanotubes, nanorods and nanowires and pure phase anatase TiO 2 nanoporous nanoribbons. The growth mechanism of the as-prepared nanostructures is elucidated from the systematic studies of field emission scanning electron microscope (FESEM) and transmission electron microscopy (TEM) imaging. We also investigate the effects of different cosolvents on the structural and optical properties of the nanostructures. One of the merits of our synthesis process is the study of the actual internal growth temperature of the solvent during the reaction. In most of the reports on the hydrothermal method, the autoclave is placed on a heated base set at an elevated temperature; this set temperature may not be the actual temperature of the solvent during the reaction. Moreover, growth temperature inside the autoclave depend on the solvent-to-solvent used in the reaction, as we have set at a particular temperature on the hot plate, but the observed temperature on the PID controller is different for different solvent of the same amount. So, the exact growth temperatures usually reported in the literature for the solvent controlled synthesis of TiO 2 nanostructures are not clear. Note that the high-pressure reactor set up (autoclave) in our laboratory has a thermal sensor with immersion tube inside the Teflon vessel, which is connected to PID temperature controller. Here we clearly studied the actual internal growth temperature of the solvent during the reaction, which is display on the PID temperature controller. 2. Experimental details All the chemicals were used without further purification. Anatase TiO 2 powders (average particle size w80 nm), sodium hydroxide pellets (NaOH), ethanol and ethylene glycol were purchased from Merck. Doubly distilled deionized (DI) water was used during all the experiments. In a typical synthesis, g of anatase TiO 2 powder was mixed with 60 ml of aqueous 10 M NaOH under stirring for 1 h, a milky solution is obtained. Then the mixed solution was transferred into a Teflon-lined autoclave (Berghof, BR- 100) of 100 ml capacity. The temperatures inside the autoclave were maintained at 130, 155, 180 C under autogenous pressure and constant magnetic stirring at 250 rpm for 16 h. The formed precipitates were obtained by centrifugation and washed several times with DI water. Then the products underwent an ultrasonic treatment with 0.1 M HCl till the ph w7 and finally the precipitates were calcined at 500 C for 5 h in air. To investigate the effects of growth mechanism, some additional experiments were carried out using different mixed cosolvents (DI water:ethanol ¼ 1:1, DI water:ethylene glycol ¼ 1:1). Characterization of as-synthesized samples was carried out by various standard tools. The crystal structures of the obtained samples are characterized by XRD patterns (Bruker D8 Advance, Cu K a radiation) and micro-raman spectroscopy (LabRam HR800, Jobin Yvon). Morphologies of the samples were studied by a FESEM (Sigma Zeiss). The high magnification morphologies and structures of as-prepared samples were observed by TEM, high-resolution TEM (HRTEM) and selected area electron diffraction (SAED) pattern (JEOL-JEM 2010 operated at 200 kv). Specimens for HRTEM investigations are prepared by dispersing powder particles in ethanol and drop casting it on a carbon coated copper grid of 400 meshes (Pacific Grid, USA). The UVeVis absorption spectroscopy measurements are recorded using a commercial spectrophotometer (PerkinEmler UV win Lab). The steady state photoluminescence (PL) spectrum is recorded at room temperature by using a 325 nm xenon lamp excitation (AB2, Thermo Spectronic). 3. Results and discussion 3.1. Structural characterization The XRD patterns of precursor anatase TiO 2 powder and solvothermal products synthesized using different solvents are shown in Fig. 1(a). Commercial anatase TiO 2 precursor was used as a reference to identify the anatase peak unambiguously. All the peaks except the peaks at 31.7 and correspond to anatase TiO 2 structure for the nanostructures synthesized in alkaline water and alkaline watereethanol mixed solvent. These additional peaks are attributed to hydrated hydrogen pentatitanate (H 2 Ti 5 O 11 $H 2 O) nanostructures (JCPDS ), indicating the presence of TiO 2 / titanate hybrid structure in the synthesized 1D nanostructures. This is also reflected in Raman spectra shown in Fig. 1(b). According to the diffraction databases, sodium titanate and hydrogen titanate have similar diffraction peak positions. It is reasonable to expect that such layered titanate structure should not be substantially affected in terms of unit-cell dimension when a small alkali cation Na þ is replaced by H þ. It is noted that no signal of Na element is detected in the energy dispersive X-ray (EDX) spectra shown later, which confirm that the Na þ is replaced by H þ during the HCl treatment with the solvothermally synthesized Na 2 Ti 5 O 11 $H 2 O product and converted into H 2 Ti 5 O 11 $H 2 O. However, no additional peak except the anatase TiO 2 is observed for the samples obtained using alkaline watereethylene glycol mixed solvent. This indicates that pure phase anatase TiO 2 is formed, when ethylene glycol was used as a cosolvent under the similar conditions of preparation.

3 930 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e936 nanostructures. However, all the Raman modes of the nanoribbons grown in ethylene glycol cosolvent show only the active Raman mode of anatase TiO 2, indicating that the phase pure anatase TiO 2 is formed. This is consistent with the XRD and HRTEM analyses. Moreover, the most intense E g (1) Raman mode of anatase at 142 cm 1 shows a considerable blue shift as compared to anatase TiO 2 precursor shown in inset of Fig. 1(b). This blue shift of the main E g (1) Raman mode was interpreted by different competing mechanism such as the non-stoichiometry due to oxygen vacancies or disorder induced defects and phonon confinement effects [17,18]. From the FESEM and TEM images, we observed that the size of the 1D nanostructures is often orders of magnitude larger than the size of the precursor TiO 2 nanoparticles. So, the phonon confinement effect is easily discarded. Therefore, we believe that the blue shift of as-synthesized nanostructures may be due to vacancies and/or defects created during the successive transformation from 2D nanosheets to 1D nanotubes, nanorods/nanowires and nanoribbons. Moreover, the presence of defects is necessary to release the stress as the TiO 6 octahedral layers, the basic unit of the crystalline structure of TiO 2 polymorphs, along with the intercalated Na þ are subsequently rolled up into nanotubes and/or splitting into nanorods. Further studies are still necessary for a clearer understanding Morphology studies Fig. 1. (a) XRD pattern of precursor anatase TiO 2 (i), the 1-D TiO 2 nanostructures grown in ethylene glycol cosolvent at 180 C (ii), alkaline water solvent (iii), and ethanol cosolvent at 130 C (iv). (b) Raman Spectra for (i) the precursor anatase TiO 2 ; the 1-D TiO 2 nanostructures synthesized using: (ii) ethylene glycol cosolvent at 180 C, (iii) alkaline water solvent and (iv) ethanol cosolvent at 130 C. The inset shows the magnified view of relative changes in the E g (1) Raman modes with various solvents. Peak positions are marked with corresponding wave numbers Raman scattering studies Raman scattering is one of the most effective tool for the study of crystallinity, phase composition, phonon confinement effect and defect structures associated with the materials. More importantly, the XRD peaks are very similar for sodium titanates, hydrogen titanates and among the polymorph of TiO 2 such as anatase, rutile, brookite and TiO 2 (B). So, XRD results do not give a conclusive idea about the structure and phase composition. The as-synthesized samples are further characterized by Raman spectroscopy to clarify about the structure and phase composition associated with the materials. All the Raman peaks of as-grown nanostructures corresponds to tetragonal anatase TiO 2 phase except the broad peak at w278 cm 1 for the sample grown by using highly alkaline water and alkaline watereethanol mixed solvents as shown in Fig. 1(b). This additional peak corresponds to hydrogen titanate, which indicates the presence of hybrid anatase TiO 2 /titanate The morphologies of the synthesized samples as obtained by FESEM are shown in Figs. 2 and 3. Fig. 2(a) shows the early growth stage of 1D nanostructures at 130 C reaction temperature using alkaline water solvent. The layered nanosheets coexisting with some partially developed nanotubes are observed at 130 C, which confirms that the 1D nanostructures are formed from the intermediate layered nanosheets. At 155 C reaction temperature, we observed both nanotubes and nanorods in the sample synthesized from alkaline water solvent as shown in Fig. 2(b) and (c), respectively. The length and the diameter of the nanotubes are observed to be w7 mm and 20e40 nm, respectively. The nanotubes are adhered among themselves forming a bunch of nanotubes and oriented in a particular direction. The bunching of adjacent parallel nanotubes into larger bundles can be explained by the oriented attachment (OA) theory. According to the OA theory [19], when structurally similar surfaces approach, atoms of opposing surfaces may combine chemically with each other. The chemical coordination results in nanotubes bundles bunching along identical crystal faces. In the present case, when the reaction temperature is increased to 180 C, comparatively longer (613e3000 nm) and thicker (35e159 nm) nanorods are formed using the same alkaline water as a solvent, which is shown in Fig. 2(d). It has been argued that the higher temperature favors the formation of nanorods by virtue of enhanced thermal stability as compared that of nanotubes [20]. However, the possible explanation for such transformation is explored further in the formation mechanism section. The morphologies of nanostructures synthesized using alkaline water and ethanol mixed solvent are shown in Fig. 3. Fig. 3(a) shows the layered nanosheets as well as splitting of the nanosheets formed at 130 C reaction temperature, which further confirmed that 1D nanostructures are actually obtained from the layered nanosheets irrespective of the solvents used. The growth of nanorods at 155 C reaction temperature using ethanol cosolvent is shown in Fig. 3(b). The nanorods have diameter ranging from 28 nm to 95 nm and length up to w3000 nm. However, we observed nanorods of diameter 38e144 nm and length up to 4500 nm coexisting with some nanowires having diameter in the range 7e20 nm as the reaction temperature increased to 180 C, shown in Fig. 3(c). The nanorods are thick and straight due to their rigidness while

4 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e Fig. 2. FESEM images of the morphology of the TiO 2 nanostructures synthesized in alkaline water solvent at reaction temperatures: (a) 130 C, (b)e(c) 155 C, and (d) 180 C. nanowires are thin and flexible and are found to be curved as shown in Fig. 3(c). In order to develop a better understanding of the growth mechanism and the structure, more detailed studies were conducted by means of TEM and HRTEM. The growth of the different TiO 2 nanostructures at different temperatures using various solvents is clearly monitored by TEM. Fig. 4(a)e(c) shows the TEM images of the as-prepared nanostructures in alkaline water solvent. A mixture of both nanotubes and nanorods is grown at 155 C reaction temperature as shown in Fig. 4(a) and (b), which is consistent with the FESEM images. The inner and outer diameters of the nanotubes are measured to be 5 nm and 9 nm, respectively. Note that the diameter of the nanotubes measured by TEM is considerably less than the diameter measured from FESEM images. This may be due to the shrinkage of the nanotubes due to the high operating electron gun voltage of the TEM. When the solvothermal reaction temperature was increased to 180 C, we observe the formation of nanorods only, as shown in Fig. 4(c). The nanotubes that are formed at 155 C are not found at 180 C, which clearly reveals that the transformation of nanotubes to nanorods occurs at higher temperature. We also noticed that the diameter and length of the nanorods increase considerably with increase in reaction temperature. The higher temperature may accelerate the solvothermal reaction and consequently the growth rate is increased. Fig. 4(d) and (f) shows the TEM images of TiO 2 nanorods grown at 155 C and 180 C, respectively, using ethanol cosolvent. Fig. 4(d) shows a single nanorod of diameter w45 nm grown at 155 C. The lattice fringes and EDX result of the corresponding nanorod are shown in inset of Fig. 4(d) and (e), respectively. The d-spacing of 3.3 Å corresponds to (101) plane of anatase TiO 2, while the d- spacing of 2.61 Å corresponds to (311) plane of H 2 Ti 5 O 11 $H 2 O. This reveals the presence of some amount of unconverted titanate with the anatase TiO 2 nanostructures, which is also reflected in XRD and Raman results (Fig. 1). The chemical composition is confirmed using the EDX spectra which detected Ti, O, Cu and C elements, as shown in Fig. 4(e). The Cu and C signals are due to the TEM grid used for the imaging. The absence of Na in the EDX spectra may suggest that Na þ ion is successfully replaced by H þ during the ion-exchange reaction when treated with HCl. The hydrogen element is unlikely to be detected in the EDX spectra due to high electron gun voltage. Thus, hybrid anatase TiO 2 /hydrogen titanate nanorods are successfully synthesized at 155 C in highly alkaline watereethanol mixed solvent. A high magnification TEM image of a single nanorod grown at 180 C is shown in Fig. 4(f). However, a unique nanostructure is obtained at 180 C when ethylene glycol was used as cosolvent. We observed nanoribbons of length up to w2.6 mm and width in the range 75e125 nm having well-resolved uniform nanopores of 8 nm diameter as shown in Fig. 4(g) and (h). This indicates that solvent also plays an important role for the formation of various nanostructures, because the solvent has a strong effect on the solubility, reactivity, and diffusion behavior of the reactants and ultimately influencing the structure and morphological features. Fig. 4(h) shows the high magnification TEM images of a single nanoribbon having nanopores of uniform shape and sizes. The SAED pattern of the corresponding nanoporous nanoribbon is shown in the inset of Fig. 4(h), which indicates the single crystalline nature of anatase TiO 2. The (101) and (105) crystal planes of anatase TiO 2 are indexed in the figure. A typical HRTEM image of the nanoporous nanoribbon is shown in Fig. 4(i), the clear lattice fringes indicating the as-synthesized products are well-crystalline. The d-spacing of 3.45 Å corresponds to (101) crystal plane of anatase TiO 2 phase, which is also the highest intensity peak

5 932 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e Optical absorption and photoluminescence studies Fig. 3. FESEM images of the morphology of the sample synthesized in alkaline watere ethanol mixed solvent at the reaction temperatures: (a) 130 C, (b) 155 C, and (c) 180 C. observed in the XRD pattern. Note that no lattice fringes corresponding to the d-spacing of any titanate phase is found. Thus the pure phase anatase TiO 2 nanostructures are obtained using ethylene glycol cosolvent which is consistent with the XRD and Raman results. The nanoporous-nanoribbons might possess very high surface area and can be suitably applicable for photocatalysis [21], environmental cleaning [22] and Li ion storage, etc. The diameters of the different nanostructures measured from the FESEM and TEM images are shown in Table 1. Light absorption characteristics of the solvothermally synthesized 1D TiO 2 nanostructures are shown in Fig. 5(a). All the samples exhibit the characteristic absorption band at around 400 nm, which is assigned to the intrinsic transition from the valence band (VB) to the conduction band (CB) of the nanostructured TiO 2. With increase in reaction temperature from 130 to 155 C, the absorption edge shows slight red shift. This may be attributed to the differences in the surface microstructures. As the size increases for growth at higher temperature, a decrease in band gap may be expected. Interestingly, an obvious rise in the intensity of absorption in the region 425e495 nm of the sample synthesized at 155 C in alkaline water solvent is observed. This may be expected due to defect created during the transformation of layered nanosheets to nanotubes. Note that the absorption edge of TiO 2 powder precursor is w380 nm (not shown) and the red shift of the absorption spectra of the as-synthesized samples is due to the presence of hybrid TiO 2 / titanate nanostructures. This is reasonable because the titanate nanostructures have lower band gap than the pure phase anatase TiO 2 [23,24]. This means that when the two phases are intermixed to form hybrid structure, their band edges overlap with each other resulting in the reduction of band gap energy. To further explain the effect of microstructures and solvent, we plotted the (ahn) 2 versus hn and estimated the direct band gap of the TiO 2 nanostructures by extrapolating the tangent of the (ahn) 2 versus hn curves to hn ¼ 0, as shown in the inset of Fig. 5(a). The band gap (E g ) decreases from 3.11 ev to 3.01 ev when growth temperature increases from 130 to 155 C. The E g increases with further increase in temperature to 180 C. Moreover, when alkaline watereethanol was used as a solvent, E g is found to be decreased to 2.97 ev. Fig. 5(b) shows the room temperature PL spectra of the assynthesized samples using various solvents at different reaction temperatures with excitation at 325 nm. The strong emission peak observed at 397 nm for all the samples is due to the near band edge emissions. Note that the intensity of the PL increases with increase in growth temperature, irrespective of the solvents used. This indicates that the crystallinity of TiO 2 increases at the higher reaction temperature. It is observed that the photoluminescence intensities of the samples grown in alkaline watereethanol mixed solvent are higher than that of alkaline water solvent. Since the growth is more in case of watereethanol mixed solvent, so larger surface area have large number of electron-hole pairs which helps the enhancement of PL intensities. As reported recently, the PL intensities depend on the surface microstructure, i.e., the intensities are enhanced in the order nanorods > nanotubes > thinfilm [25,26]. In our present work, as we observed different nanostructures at different reaction temperatures, i.e., nanosheets > nanotubes > nanorods/nanowires, so the PL enhancement is expected for higher growth temperature for a particular solvent Growth mechanism At the present time, there is an incomplete understanding on the formation mechanism of solvothermally synthesized different TiO 2 nanostructures, whether it is due to the rolling of nanosheets exfoliated from crystalline TiO 2 precursors or rolling and/or splitting of layered titanate nanosheets. To understand the growth mechanism of different 1D nanostructures, controlled experiments were carried out at different temperatures under continuous stirring during the reaction and the products using different cosolvents were also examined. A systematic study on the mechanism of formation of the as-prepared nanostructures is explored on the basis of our experimental findings.

6 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e Fig. 4. TEM images of the as-synthesized nanostructures. Using alkaline water solvents (a) and (b) at 155 C, (c) at 180 C reaction temperature; using alkaline watereethanol mixed solvent (d) at 155 C, (f) at 180 C reaction temperature; using alkaline watereethylene glycol mixed solvent (g) and (h) at 180 C reaction temperature. The inset of (h) shows the corresponding SAED pattern, and (i) shows the HRTEM lattice fringes of the nanoribbons. The inset in (d) shows the corresponding HRTEM lattice fringes of the nanorod and (e) shows the EDX spectrum of the nanorods. Wu et al. [13] argued that titanate nanosheets were first splitted and then rolling of the splitted sheets formed the nanotubes. They also claimed that the thickening of thin nanosheets is accelerated by increasing the reaction temperature and finally splitted into nanowires instead of rolling. Huang et al. [27] proposed that the transformation from nanotubes to nanowires is due to an oriented attachment (OA) [19] and Ostwald ripening (OR) [8] cooperative mechanism based on their TEM observations. With these thoughts into consideration, we carried out the experiments at different reaction temperatures under constant stirring to clarify the possible formation mechanism. It was earlier proposed that sodium titanate nanotubes were formed by rolling crystalline nanosheets exfoliated from crystalline TiO 2 materials (anatase/rutile). If this is one of the possible Table 1 The effect of solvent and reaction temperature on the morphologies, sizes and structures of the 1D TiO 2 nanostructures. Solvent Reaction temperature ( C) Morphology Diameter (nm) from FESEM Diameter (nm) from HRTEM Structure DI water 130 Nanosheets, nanotubes e e Anatase TiO 2 /titanate 155 Nanotubes 15e20 9 Nanorods 39e Nanorods 35e159 35e79 DI water:ethanol 130 Splitted nanosheets e e Anatase TiO 2 /titanate 155 Nanorods 28e95 w Nanorods 38e144 40e65 Nanowires e 20e25 DI water:ethylene glycol 180 Nanoporous nanoribbons e 75e125 Anatase TiO 2

7 934 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e936 during the calcinations at 500 C for 5 h. For simplification, the overall chemical processes of the as-prepared samples are as follows. 5TiO 2 þ 2NaOH / Na 2 Ti 5 O 11 $H 2 O Na 2 Ti 5 O 11 $H 2 O þ 2 HCl / H 2 Ti 5 O 11 $H 2 O þ 2NaCl H 2 Ti 5 O 11 $H 2 O / 5 TiO 2 þ 2H 2 O Fig. 5. (a) UVevis absorption spectra of the samples synthesized in alkaline water solvent at different reaction temperatures: (i) 180 C, (ii) 130 C, (iii) 155 C, and (iv) the sample synthesized in alkaline watereethanol mixed solvent at 180 C. The inset shows the changes in the band gap of the corresponding samples. (b) Room temperature PL spectra of the samples synthesized in alkaline water solvent at various temperatures: (i) 130 C, (ii) 155 C, (iv) 180 C; using alkaline watereethanol mixed solvent: (iii) 130 C, (v) 155 C, (vii) 180 C and (vi) the sample synthesized using alkaline watereethylene glycol mixed solvent at 180 C. formation mechanisms, then the length of the nanotubes should be restricted by the size of the raw TiO 2 materials. However, we observed that the nanotubes and nanorods are often orders of magnitude larger than the size of the raw anatase TiO 2 particles (w80 nm particle size). The similar results were reported by several authors [20,28,29], for example Huang et al. [27] reported titanate nanotubes of several hundreds of nanometer in length derived from the anatase TiO 2 powder with particle size of about 10 nm. Moreover, it is postulated that crystalline TiO 2 either anatase or rutile, does not possess any layered structure that can form the final nanotubes [30e32]. So this assumption could not be a possible growth mechanism for the 1D nanotubes and nanorods observed in this work. However, when anatase TiO 2 powder precursor reacts with concentrated alkaline NaOH solution, TieOeTi bonds are broken and the Na þ ions intercalated within the TiO 6 octrahedral units and then recrystallized into layered Na 2 Ti 5 O 11 $H 2 O nanosheets. After ultrasonic HCl treatment and washing, hydrated hydrogen pentatitanate (H 2 Ti 5 O 11 $H 2 O) is formed by the ion exchange of Na þ with H þ and the by-product NaCl is formed. The final anatase TiO 2 products are obtained by dehydration process Thus, the final anatase TiO 2 nanostructured products are obtained by the three steps reaction process: (i) the reaction of TiO 2 precursor with alkaline NaOH solution, (ii) the ion exchange between Na þ and H þ and (iii) the final calcination. The early growth stage of 1D nanostructure at 130 C in alkaline solvent shows the coexistence of nanosheets and partially developed nanotubes, indicating that nanotubes are formed from the layered titanate nanosheets. At comparatively higher temperature (155 C) and pressure, the nanosheets completely roll up to form a bunch of nanotubes well oriented in a particular directions as shown in Fig. 2(b). The driving force for the rolling may be asymmetric due to the hydrogen deficiency in surface layers [31] together with unsymmetrical surface forces due to locally high surface energy. Another reason for the rolling of nanosheets at high temperature and pressure may be due to the mechanical tension that arises during the process of dissolution/crystallization in nanosheets [14]. When structurally similar surfaces approach, atoms of opposing surfaces may combine chemically with each other. The chemical coordination results in nanotubes bundles bunching along identical crystal faces [19,33]. Interestingly, note that nanorods are also observed in the same product shown in Fig. 2(c) and 4(b). This reveals that some of the nanotubes detached from the bundles of nanotubes and transformed into nanorods. At comparatively higher temperature (180 C), we observed only nanorods instead of any intermediate nanotubes. This is expected because during the stirring, the bunch of nanotubes may split and simultaneous shrinkage and growth may occur due to high temperature and pressure to form condensed nanorods. The main advantage of the stirring is the splitting of the nanotubes and nanosheets. At the same time, the constant motion due to stirring prevents sedimentation and forces the intimate mixing of the system. After splitting of the bunch of nanotubes, the nanotubes are detached from the bundles and the pressure acts on the individual nanotube and hence the effect is more as compared to the pressure acting on the bunch of the nanotubes. That is why the nanotubes transform to condensed nanorods. The observed pressure is 8 bar showing in manometer which is in built with the high pressure reactor set up, when temperature is 180 C under stirring condition for the alkaline water solvent. On the other hand, the constant motion fields created by internal stirring are probably too effective in continuously breaking up freshly formed nanotube assemblies, so bundling is prevented again. The thickening and splitting nature of the nanosheets are clearly shown in Fig. 3(a) for the early growth stage at 130 C in alkaline watereethanol mixed solvent. However, no such nanotubes are observed at 155 C as observed in case of alkaline water solvent, when ethanol was used as a cosolvent under similar conditions. Note that the length of the nanorods is more than that of the nanorods grown in alkaline water solvent at the same reaction temperature, indicating the fast growth rate of the samples in ethanol cosolvent. The rapid growth rate in alkalinee ethanol mixed solvent accelerates the formation of thick nanosheets and further splitting of the thick nanosheets results in the formation of nanorods at higher temperature and pressure. When the reaction temperature increased to 180 C, only nanorods and nanorods/nanowires are observed in alkaline water

8 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e Fig. 6. Schematic of the growth mechanism of different 1D nanostructures formed at different temperatures using various solvents. and alkaline watereethanol mixed solvents, respectively. The possible reasons are (i) the thickening of nanosheets unable to fold and split into nanorods/nanowires. The high temperature as well as stirring may help to accelerate the reaction for the nucleation of thicker nanosheets and subsequently splitting of the thick nanosheets; (ii) the very fast growth rate and stirring promoted detachment of the nanotubes from the bunch of nanotubes which results in rapid shrinkage of the nanotubes to ultimately form nanorods at higher temperature and pressure. The observed pressures at 180 C reaction temperature under stirring condition are 8, 16 and 1 bar for the alkaline water, alkaline watereethanol and watereethylene glycol mixed solvent, respectively. Note that a quite unique morphology is observed at 180 C, when ethylene glycol was used as a cosolvent indicating that the solvent plays an important role to control the morphology of the products. Nanoporous TiO 2 nanoribbons are observed from the TEM images as shown in Fig. 4(g) and (h). A possible reason for this type of morphology may be due to the polarity and coordinating ability of a cosolvent have a strong effect on the solubility, reactivity and diffusion behavior of the reactants, thus ultimately influencing the structure and morphological features of the resulting products [34]. The parent TiO 6 octahedra of TiO 2 materials may coordinate with glycol to form chain-like structure [35], whereas the NaOH may form titanate nanosheets by sharing the vertices edges of the octahedra and resulting nanoribbons with nanopores are formed at high temperature and pressure. The growth mechanism and evolution of different 1D nanostructures are briefly elucidated in the schematic diagram shown in Fig. 6. The crossed dashed-line in the diagram indicates that the nanorods/nanotubes are not formed directly from the TiO 2 nanoparticles and it essentially involves the formation of the nanosheets followed by splitting and/or rolling to form nanorods/nanotubes of TiO Conclusions One-dimensional nanostructured TiO 2 of different morphologies, including nanotubes, nanowires, nanorods and nanoporous nanoribbons were successfully synthesized by a simple solvothermal route. The impact of reaction temperature, stirring during the reaction and different cosolvents on the morphology of the products was studied systematically. It is found that the hybrid anatase TiO 2 /titanate nanotubes and/or nanorods are formed in highly alkaline water solvent, while nanorods/nanowires are formed in highly alkaline watereethanol mixed solvent. When ethylene glycol was used as cosolvent, pure phase anatase TiO 2 nanoporous nanoribbons are obtained at 180 C. We observed that the maximum temperature limit for the transformation of nanotubes to nanorods starts at w 155 C in case of alkaline water solvent. However, the nanorods grown in alkaline watereethanol solvent is directly from the splitting of nanosheets at 155 C. The growth rate and morphologies of the 1-D nanostructures are shown to be dependent on the growth temperature and the solvent used. The growth rate is faster in alkaline water/ethanol and alkaline water/ethylene glycol cosolvent as compared to the alkaline water solvent. XRD and micro-raman scattering studies confirm the pure phase anatase and anatase/titanate hybrid structures of the assynthesized products. Based on the FESEM and TEM analyses, we have explained the mechanism of growth of different nanostructures through a schematic diagram. Optical absorption and PL studies confirm that the TiO 2 nanostrctures possess good crystallinity. This study will enable controlled growth of TiO 2 nanotubes, nanorods and nanoporous nanoribbons for practical applications such as energy conversion and storage. References [1] Y. Zhang, W. Fu, H. Yang, Q. Qi, Y. Zeng, T. Zhang, R. Ge, G. Zou, Appl. Surf. Sci. 254 (2008) [2] K. Sunada, Y. Kikuchi, K. Hashimoto, A. Fujishima, Environ. Sci. Technol. 32 (1998) 726. [3] M. Zukalova, M. Kalbac, L. Kavan, I. Exnar, M. Graetzel, Chem. Mater. 17 (2005) [4] A. Fujishima, K. Honda, Nature 238 (1972) 37. [5] S.H. Lim, J. Luo, Z. Zhong, W. Ji, J. Lin, Inorg. Chem. 44 (2005) [6] C. Xu, P.H. Shin, L. Cao, J. Wu, D. Gao, Chem. Mater. 22 (2010) 143. [7] R. Ma, K. Fukuda, T. Sasaki, M. Osada, Y. Bando, J. Phys. Chem. B 109 (2005) 6210.

9 936 B. Santara, P.K. Giri / Materials Chemistry and Physics 137 (2013) 928e936 [8] A. Elsanousi, E.M. Elssfah, J. Zhang, J. Lin, H.S. Song, C. Tang, J. Phys. Chem. C 111 (2007) [9] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Langmuir 14 (1998) [10] K. Das, S.K. Panda, S. Chaudhuri, J. Crytsal Growth 310 (2008) [11] Y. Lan, X. Gao, H. Zhu, Z. Zheng, T. Yan, F. Wu, S.P. Ringer, D. Song, Adv. Funct. Mater. 15 (2005) [12] D.L. Morgan, H.Y. Zhu, R.L. Frost, E.R. Waclawik, Chem. Mater. 20 (2008) [13] D. Wu, J. Liu, X. Zhao, A. Li, Y. Chen, N. Ming, Chem. Mater. 18 (2006) 547. [14] D.V. Bavykin, V.N. Parmon, A.A. Lapkin, F.C. Walsh, J. Mater. Chem. 14 (2004) [15] E. Horvath, A. Kukovecz, Z. Konya, I. Kiricsi, Chem. Mater. 19 (2007) 927. [16] X. Shen, J. Zhang, B. Tian, J. Mater. Sci. 47 (2012) [17] J.C. Parker, R.W. Seigel, Appl. Phys. Lett. 57 (1990) 943. [18] W.F. Zhang, Y.L. He, M.S. Zhang, Z. Yin, Q. Chen, J. Phys. D: Appl. Phys. 33 (2000) 912. [19] J. Zhang, Z. Lin, Y.Z. Lan, G.Q. Ren, D.G. Chen, F. Huang, M.C. Hong, J. Am. Chem. Soc. 128 (2006) [20] Z.Y. Yuan, B.L. Su, Colloids Surf. A 241 (2004) 173. [21] J. Du, W. Chen, C. Zhang, Y. Liu, C. Zhao, Y. Dai, Chem. Eng. J. 170 (2011) 53. [22] F. Liang, T.L. Kelly, L. Luo, H. Li, M.J. Sailor, Y.Y. Li, Appl. Mater. Interfaces (2012). [23] X. Sun, Y. Li, Chem. Eur. 9 (2003) [24] H. Yu, J. Yu, B. Cheng, M. Zhou, J. Solid State Chem. 179 (2006) 349. [25] L.M. Sikhwivhilu, S. Mpelane, B.W. Mwakikunga, S.S. Ray, Appl. Mater. Interfaces 4 (2012) [26] Y.H. Chang, C.M. Liu, C. Chen, H.E. Cheng, J. Electrochem. Soc. 159 (2012) 401. [27] J. Huang, Y. Cao, Q. Huang, H. He, Y. Liu, W. Guo, M. Hong, Cryst. Growth Des. 9 (2009) [28] C.B. Almquist, P. Biswas, J. Catal. 212 (2002) 145. [29] T. Sreethawong, Y. Suzuki, S. Yoshikawa, J. Solid State Chem. 178 (2005) 329. [30] G.H. Du, Q. Chen, R.C. Che, Z.Y. Yuan, L.M. Peng, Appl. Phys. Lett. 79 (2001) [31] S. Zhang, L.M. Peng, Q. Chen, G.H. Du, G. Dawson, W.Z. Zhou, Phys. Rev. Lett. 91 (2003) [32] A. Thorne, A. Kruth, D. Tunstall, J.T.S. Irvine, W. Zhou, J. Phys. Chem. B 109 (2005) [33] R.L. Penn, J.F. Banfield, Science 281 (1998) 969. [34] B. Wen, C. Liu, Y. Liu, New J. Chem. 29 (2005) 969. [35] X. Jiang, Y. Wang, T. Herricks, Y. Xia, J. Mater. Chem. 14 (2004) 695.