Sensors and Actuators B 137 (2009) 513–520 Contents lists available at ScienceDirect Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb Gas sensing characteristics and porosity control of nanostructured films composed of TiO 2 nanotubes ଝ Min-Hyun Seo a , Masayoshi Yuasa b , Tetsuya Kida b , Jeung-Soo Huh c , Kengo Shimanoe b,∗ , Noboru Yamazoe b a Department of Molecular and Material Sciences, Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Kasuga, Fukuoka 816-8580, Japan b Department of Energy and Material Sciences, Faculty of Engineering Sciences, Kyushu University, Kasuga, Fukuoka 816-8580, Japan c Department of Materials Science and Metallurgy, Kyungpook National University, Daegu 702-701, South Korea article info Article history: Received 15 August 2008 Received in revised form 19 December 2008 Accepted 30 January 2009 Available online 10 February 2009 Keywords: TiO 2 nanotube VOC Porosity control Porous film Hydrothermal treatment abstract Preparation and morphology control of TiO 2 nanostructured films for gas sensor applications were inves- tigated. To examine the effect of the morphology of sensing films on the sensing characteristics, TiO 2 with different morphologies, nanoparticles and nanotubes, were used for the film preparation. TiO 2 nanotubes were prepared by a hydrothermal treatment of TiO 2 nanoparticles in a NaOH solution at 160, 200, and 230 ◦ C for 24 h and subsequent washing with an HCl solution. Uniform sized TiO 2 nanotubes of 1 ␮min length and 50 nm in diameter were formed at 230 ◦ C. The sensing films composed of nanotubes prepared at 230 ◦ C showed a high sensor response to toluene at 500 ◦ C as compared with those composed of TiO 2 nanoparticles. Scanning electron microscope (SEM) analysis and pore size distribution measurements indicated that the sensing films composed of the TiO 2 nanotubes had a high porous morphology with a peak pore size of around 200 nm, which can promote the diffusion of toluene deep inside the films and improve the sensor response. The obtained results demonstrated the importance of microstructure con- trol of sensing layers for improving the sensitivity to large size molecules like volatile organic compounds (VOCs). © 2009 Elsevier B.V. All rights reserved. 1. Introduction Semiconductor gas sensors based on metal oxides have aroused considerable interest owing to their high sensitivity to pollutant gases, low cost, and small size [1–3]. Various oxide materials such as SnO 2 [4],WO 3 [5],TiO 2 [6–9], and ZnO [10] have so far been used for gas sensors. Among them, TiO 2 is a well-known important functional material used for a variety of applications such as pho- tocatalysts [11], dye-sensitized solar cells [12], batteries [13], and pigments [14]. For gas sensor applications, it has been reported that TiO 2 with a large surface area shows good sensing properties to CO [7],H 2 [8], and NO x [9]. The important feature of TiO 2 -based gas sensors is that they can be operated at high temperature because of the good chemical stability of TiO 2 . Recently, utilization of nanostructured TiO 2 for various devices has attracted much attention due to prospects for upgrading the device performance through the nanostructure control of devices [15,16]. Various nanostructured TiO 2 such as nanorods [17], ଝ Paper presented at the International Meeting of Chemical Sensors 2008 (IMCS- 12), July 13–16, 2008, Columbus, OH, USA. ∗ Corresponding author. Tel.: +81 92 583 7876; fax: +81 92 583 7538. E-mail address: simanoe@mm.kyushu-u.ac.jp (K. Shimanoe). nanowires [18], nanosheets [19], and nanotubes [20–23] have been prepared by wet-chemical methods. TiO 2 nanotubes were firstly reported by Kasuga et al. in 1998 [20]. Since then, the chemical and physical properties of TiO 2 nanotubes have been studied intensively due to the ease of the preparation using a simple hydrothermal method. The nanotubular architecture can achieve high specific surface area and thus TiO 2 nanotubes prepared by a hydrother- mal method have been successfully utilized for photocatalysis [24], dye-sensitized solar cells [25], and lithium-ion batteries [26]. TiO 2 nanotubes and nanofibers prepared by anodization and electrospinning methods have been used for the detection of H 2 [27,28],NO 2 [29], and water vapor [30,31]. Varghese et al. reported a large change in the electric resistance of arrays of TiO 2 nan- otubes in response to H 2 [27,28]. However, the sensing mechanism of nanostructured TiO 2 films composed of nanotubes or nanofib ers has not yet been understood well. It is thus of considerable interest and importance to examine the detailed gas sensing properties of nanostructured films base d on TiO 2 nanotubes. For semiconductor gas sensors, the porosity of sensing films is an important parameter [32–34]; porous sensing films can facilitate gas diffusion deep inside the films and reach high gas sensitivity. In particular, the microstructure control is important to detect large size molecules such as volatile organic compounds (VOCs). Note that it is possible to prepare porous gas sensing films using TiO 2 0925-4005/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2009.01.057 514 M H. Seo et al. / Sensors and Actuators B 137 (2009) 513–520 Fig. 1. SEM images of the sensing films composed of (a) commercial TiO 2 nanoparticles (P-25) and those hydrothermally treated at (b) 160, (c) 200 and (d) 230 ◦ C. (a’)–(d’) show the corresponding images after calcination at 600 ◦ C. nanotubes because of their high anisotropic shape; packing tubular particles would produce loosely-packed particulate films with ran- domly distributed pores while hindering intimate contacts among the particles. Such a porous film is expected to show sensitivities to large sized gas molecules like VOCs. In this study, we fabricated porous gas sensing filmscomposed of TiO 2 nanotubes prepared by a hydrothermal treatment and studied the gas sensing properties of the porous TiO 2 nanotubular films. The investigation was carried out with a particular emphasis on the promotion of the gas sensitivity through the porosity control of gas sensing films. 2. Experimental TiO 2 nanotubes were prepared by a hydrothermal method as reported in the literature [18].0.5gofaTiO 2 commercial powder (Degussa P-25 (mean particle size: ca. 20 nm)) was hydrothermally treated with 30 mL of a NaOH solution (10 mol/L) at 160, 200 and Fig. 2. TEM images of TiO 2 aggregates and nanotubes prepared by the hydrothermal treatment at (a) 160, (b) 200 and (c) 230 ◦ C, followed by calcination at 600 ◦ C. M H. Seo et al. / Sensors and Actuators B 137 (2009) 513–520 515 Fig. 3. Pore size distribution of the sensing films composed of (a) commercial TiO 2 nanoparticles (P-25) and those hydrothermally treated at (b) 160, (c) 200 and (d) 230 ◦ C. They were calcined at 600 ◦ C. 230 ◦ C for 24 h in a Teflon-lined autoclave. After the treatment, the TiO 2 powder was washed with 50 mL of a HCl solution (0.2 mol/L) under ultrasonic irradiation for 1 h. Then, the obtained products were filtered and dried to recover TiO 2 nanotubes. The resulting products were characterized by X-ray diffraction (XRD) with Cu-K␣ radiation, scanning electron microscopy (SEM), and transmission electron microscopy (TEM). For electrical and sensing characterizations, TiO 2 thick films were fabricated by a screen-printing method. By using a binary dispersant of ␣-terpineol (95 mass%) mixed with ethyl cellulose (5 mass%), the sensor material was converted into a paste, which was screen-printed on an alumina substrate attached with a pair of Au electrodes (line width: 180 ␮m, distance between lines: 90 ␮m, sensing layer area: 64 mm 2 ). After screen-printing, the fabricated sensor devices were calcined at 600 ◦ C for 1 h. The porosity of the films was measured by mercury porosimetry. The electrical and gas sensing properties of the films were tested using CO, H 2 , ethanol, and toluene as target gases at 450–550 ◦ C. Measurements were performed using a conventionalgas flow appa- ratus equipped with an electric furnace at a gas flow rate of Fig. 4. XRD patterns of (a) commercial TiO 2 nanoparticles (P-25) and those hydrothermally treated at (b) 160, (c) 200 and (d) 230 ◦ C. Fig. 5. XRD patterns of (a) commercial TiO 2 nanoparticles (P-25) and those hydrothermally treated at (b) 160, (c) 200 and (d) 230 ◦ C. They were calcined at 600 ◦ C. 100 cm 3 /min. The sensor response was defined as R air /R gas , where R air and R gas are the electric resistances in air and that in a test gas, respectively. 3. Results and discussion 3.1. Characterization of nanotubular TiO 2 films Fig. 1 shows SEM images of the surface of TiO 2 thick films com- posed of TiO 2 nanotubes prepared by the hydrothermal treatment at 160, 200, and 230 ◦ C, together with that of a film composed of commercial TiO 2 nanoparticles (P-25). The surfaces of the films were observed before and after calcination at 600 ◦ C. It is obvious that the morphology of the films was significantly changed after the hydrothermal treatment, depending on the treatment temperature. The treatment at 160 ◦ C produced heavily aggregated TiO 2 particles, as shown in Fig. 1(b). On the other hand, the treatment at 200 and 230 ◦ C resulted in the formation of tubular TiO 2 of 1 ␮m and 50 nm in length and diameter, respectively, as shown in Fig. 1(c) and (d). With increasing the temperature of the hydrothermal treatment, the length of the tubes increased and the distribution of the diam- eter became narrower. Nanotubes with a more uniform size were formed by the hydrothermal treatment at 230 ◦ C. The SEM images confirm that the films composed of the TiO 2 nanotubes are more porous than those composed of TiO 2 nanoparticles, as expected. It is also noted that the calcination induced no drastic change in the morphology of TiO 2 , but resulted in slight sintering of TiO 2 aggre- gates and nanotubes obtained at 160 and 200 ◦ C, respectively, as shown in Fig. 1(b’) and (c’). The nanostructures of the TiO 2 nanotubes calcined at 600 ◦ C were observed by TEM, as shown in Fig. 2. The TEM observation Table 1 Specific surface area of commercial TiO 2 nanoparticles (P-25) and those hydrother- mally treated at 160, 200 and 230 ◦ C, together with the peak pore size of the sensing films composed of the particles. They were calcined at 600 ◦ C. Sample Specific surface area (m 2 /g −1 ) Peak pore size (nm) Commercial TiO 2 (P-25) 46.3 36.0 160 ◦ C 76.4 20.4 200 ◦ C 22.7 138.6 230 ◦ C 23.1 201.3 516 M H. Seo et al. / Sensors and Actuators B 137 (2009) 513–520 Fig. 6. Electric resistance in air as a function of operating temperature for the thick films composed of (a) commercial TiO 2 nanoparticles (P-25) and those hydrother- mally treated at (b) 160, (c) 200 and (d) 230 ◦ C. revealed that the tubular structure is stable even after calcination at 600 ◦ C and TiO 2 aggregates obtained at160 ◦ C have no tubular struc- ture. The wall thickness of nanotubes prepared at 200 and 230 ◦ Cis estimated to be ca. 5 nm from the TEM images. In addition, a uni- form size distribution was confirmed for the nanotubes treated at 230 ◦ C, suggesting that the optimum temperature of the hydrother- mal treatment is 230 ◦ C. The mechanism of the nanotube formation by a hydrothermal treatment has been reported as follows [23]; first, TiO 2 transformed into a layered compound of Na 2 Ti 2 O 5 ·H 2 O by NaOH. The washing of the product with HCl results in the ion exchange of Na + with H + . The HCl treatment removes Na + from the layered compound to form exfoliated sheets. The sheets roll up to form tube-like particles. The TEM image shown in Fig. 2(c) also con- firms the rolling up of nanosheets into nanotubes. On the basis of the above mechanism, it is considered that the treatment at higher temperature of 230 ◦ C promoted the transformation of TiO 2 parti- cles into nanosheets and resulted in the formation of more uniform nanotubes, as observed in SEM and TEM images. Fig. 3 shows the pore size distribution of the films composed of TiO 2 nanoparticles, aggregates, and nanotubes after calcination at 600 ◦ C. The distribution of pores peaks at approximately 36 and 138 nm for the films composed of commercial TiO 2 particles and TiO 2 nanotubes obtained at 200 ◦ C, respectively, indicating that the porosity of the film was increased by using the nanotubular particles. Furthermore, the peak pore size of the TiO 2 nanotubes obtained at 230 ◦ C increased, while the porosity of the TiO 2 aggre- gates obtained at 160 ◦ C decreased. The observed trend is in good agreement with the SEM results. The pore size at the maximum pore volume for the films composed of TiO 2 nanoparticles, aggre- gates, and nanotubes is summarized in Table 1, together with the specific surface area of TiO 2 measured by the B.E.T. method. The surface area of calcinated nanotubes was lower than that of TiO 2 nanoparticles. A peculiar feature was observed in the TiO 2 aggre- gates obtained at 160 ◦ C that show a high surface area but a small peak pore size. The crystal structure of the obtained TiO 2 nanotubes was ana- lyzed by XRD. Fig. 4 shows the XRD patterns of the products hydrothermally treated at different temperatures. The patterns show that the crystal phase of TiO 2 changed from a mixed phase of rutile and anatase to that of H 2 Ti 3 O 7 and anatase after the hydrothermal treatment, irrespective of reaction temperature. The Fig. 7. Sensor responses to (a) CO (500 ppm), (b) H 2 (500 ppm), (c) ethanol (47 ppm), and (d) toluene (50 ppm) gases in the temperature range of 450–550 ◦ C for the devices using commercial TiO 2 nanoparticles (P-25) and those hydrothermally treated at 160, 200 and 230 ◦ C. M H. Seo et al. / Sensors and Actuators B 137 (2009) 513–520 517 Fig. 8. Sensor responses to CO (500 ppm), H 2 (500 ppm), ethanol (47 ppm), and toluene (50 ppm) gases at 500 ◦ C for the devices using (a) commercial TiO 2 nanoparticles (P-25) and those hydrothermally treated at (b) 160, (c) 200 and (d) 230 ◦ C. main peak for the TiO 2 nanotubes was assigned to H 2 Ti 3 O 7 accord- ing to recent structural characterizations [35,36], although different assignments of nanotubes to H 2 Ti 2 O 5 ·H 2 O and H 2 Ti 4 O 9 ·H 2 Ohave also been reported [23,37]. Fig. 5 shows the XRD patterns of the products after calcination at 600 ◦ C. The crystallization of H 2 Ti 3 O 7 to anatase occurred in the TiO 2 nanotubes obtained at 200 and 230 ◦ C, although the H 2 Ti 3 O 7 phase remained. On the other hand, for TiO 2 aggregates obtained at 160 ◦ C, the H 2 Ti 3 O 7 phase com- pletely transformed into the anatase phase after calcination. It has been reported that the thermal stability of the protonated titanate phase is improved by the presence of sodium ions and the removal of sodium ions promotes the thermal conversion to anatase [35].It is suggested that the reaction of TiO 2 with NaOH is incomplete at 160 ◦ C to produce partly Na-intercalated layered compounds on the basis of the proposed mechanism discussed above and the SEM and TEM results, and that the subsequent washing with HCl effectively removed sodium ions from the layered compounds. Consequently, the low sodium content in aggregates obtained at 160 ◦ C may assist in the formation of the anatase phase. 3.2. Gas sensing properties of nanotubular TiO 2 films Fig. 6 shows the electrical resistances in air as a function of temperature ranging from 450 to 550 ◦ C for the fabricated films composed of TiO 2 nanoparticles, aggregates, and nanotubes. The film using TiO 2 nanotubes obtained at 230 ◦ C gave the highest resis- tance, reflecting their higher porosity than the other films. However, the electric resistance of the other films was not correlated well with their porosity. This is because the electric resistance is depen- dent on various parameters such as grain size, tube length, film thickness, crystal structure, and physical parameters such as car- rier density and effective mobility. In addition, the observed thin wall thickness (ca. 5 nm) of the nanotubes may also be responsible for the high electric resistance. The electric resistance decreased with increasing the operating temperature, following the typical behavior of oxide semiconduc- tor. However, the electric resistance at lower than 500 ◦ C exceeded 10 9 , which is too high to be measured using a conventional elec- tric circuit. Thus, the optimal operating temperature is judged to be 500 ◦ C from a practical point of view. The sensor response of the fabricated films composed of TiO 2 nanoparticles, aggregates, and nanotubes was examined against traces of CO, H 2 , ethanol, and toluene gases at 450, 500, and 550 ◦ C, as shown in Fig. 7. The fabricated sensors using TiO 2 (n-type semi- conductor) responded to the target gases by a decrease in the electric resistance. Thus, the probable sensing mechanism is that Fig. 9. Schematic model of diffusion of toluene gas inside a porous nanotubular film. 518 M H. Seo et al. / Sensors and Actuators B 137 (2009) 513–520 the target gas molecules diffuse into the sensing layer through pores and react with adsorbed oxygen on TiO 2 , as reported [1]. This reduces the thickness of the surface depletion layer in TiO 2 , thereby decreasing the electric resistance. Fig. 8 summarizes the sensor response of the devices to the test gases at 500 ◦ C. The film using commercial TiO 2 nanoparticles exhibited the highest sensor responses to all gases. We have recently developed a new model on the roles of shape and size of component crystals in semiconductor gas sensors on the basis of electron-depleted conditions in com- ponent crystals [38]. The model confirms that the gas sensitivity increases with decreasing the crystal size as experimentally proved for many cases, and predicts that crystals in spherical shape would show higher sensitivity than those in columnar shape. Thus, the observed higher sensor response of commercial TiO 2 nanoprticles than nanotubes is consistent with the developed model. On the other hand, the film using TiO 2 nanotubes prepared at 230 ◦ C showed comparably high sensitivity to toluene among the gases tested, despite their lower surface area than that of com- mercial TiO 2 nanoparticles. As noted above, for semiconductor gas sensors, target gases diffuse in a sensing film through pores and react with surface oxygen adsorbed on component particles to induce the resistance change. The concentration of the target gases decreases inside the sensing film as they diffuse. The component particles located deep inside the film may remain intact or inacces- sible for the target gases provided that the film is not sufficiently porous. This would lead to a decrease in the sensor response due to a decrease in the utility factor of the sensing film or a decrease in the accessibility of the target gas. Thus, the effect of the porosity of sensing films on the sensor response is more pronounced for target gases with large molecular sizes. As revealed by the pore size dis- tribution measurements, the film using TiO 2 nanotubes prepared at 230 ◦ C has the peak pore size of around 200 nm, which is much larger than those for the film using TiO 2 aggregates and nanotubes prepared at 160 and 230 ◦ C, respectively. Such macropores can pro- vide high-diffusivity paths for large toluene molecules and improve the utility factor of the sensing film. Thus, the observed particular Fig. 10. Response transients to toluene (50 ppm) at 500 ◦ C for the devices using (a) commercial TiO 2 nanoparticles (P-25) and (b) those hydrothermally treated at 230 ◦ C. increase in the sensor response to toluene for the film using TiO 2 nanotubes prepared at 230 ◦ C can be ascribed to the high porosity of the nanotubular film, as is schematically shown in Fig. 9. In addi- tion, almost constant sensor responses to CO and H 2 were observed for the films using TiO 2 aggregates and nanotubes. This is probably because more combustible gases than toluene are difficult to dif- fuse deep inside the films and tend to burn at the film surfaces even if the porosity of the films is high. Importantly, the selective detec- tion of toluene was achieved by the porosity improvement of the sensing film and the high temperature operation. There have been some papers reporting the response charac- teristics of TiO 2 -based sensors to VOCs. Teleki et al. reported the sensor response of S = 6 to ethanol gas (30–300 ppm) in dry air at 400 ◦ C for TiO 2 doped with Nb [39]. On the other hand, Gessner et Fig. 11. SEM images of the TiO 2 nanotubular films calcined at (a) 600 and (b) 700 ◦ C. (a’) and (b’) show the corresponding high magnification images. M H. Seo et al. / Sensors and Actuators B 137 (2009) 513–520 519 Fig. 12. Sensor responses to toluene (50 ppm) for the TiO 2 nanotubular films cal- cined at (a) 600 and (b) 700 ◦ C. al. reported the sensor response of S = 4 to toluene gas (100 ppm) in dry air at 400–50 0 ◦ C [40]. Although the preparation methods of TiO 2 and the experimental conditions are different from those in the present study, the reported sensor responses were lower than our results. Therefore, our approach of controlling the film pore size is quite effective in improving the sensor performance to large size molecules like ethanol and toluene. Fig. 10 shows the response transients to toluene (50 ppm) at 500 ◦ C of the sensors using TiO 2 nanotubes prepared at 230 ◦ C and commercial TiO 2 nanoparticles (P-25). Even if a difference in the response transients is seen by both sensors, it seems not to be the important difference because the speed of response and recovery depends on dead volume in the equipment, as reported by Kida et al. [41]. According to a basic viewpoint of gas diffusion [42],itis know that the diffusion through the microstructure affects sensor response (R air /R gas ). From the above, it is thought that the sensor response of nanotubes observed in Fig. 10 is as high as that of nanoparticles. To provide a further experimental evidence of supporting the above discussions, the porosity of the nanotubular film was con- trolled by calcination and then the sensor response was tested again. Fig. 11 shows the SEM images of the film composed of TiO 2 nanotubes prepared at 230 ◦ C after calcination at 600 and 700 ◦ C. From the SEM images, it appears that the porosity obviously decreased due to sintering of nanotubes after calcination at higher temperature. As a result, the sensitivity to toluene gas significantly decreased, as shown in Fig. 12. Thus, the results obtained indicate again the importance of themicrostructure control ofsensing layers for detecting large sized gas molecules. 4. Conclusion TiO 2 nanotubes of 1 ␮m in length and 50 nm in diameter were formed by the hydrothermal treatment of TiO 2 nanoparticles with NaOH at 200 and 230 ◦ C. Uniform sized nanotubes were obtained at 230 ◦ C. The tubular structure of the TiO 2 nanotubes was stable even after calcination at 600 ◦ C for 1 h. The calcination of the nanotubes resulted in the formation of the anatase phase. A porous film with a peak pore size of around 200 nm was successfully formed using the TiO 2 nanotubes obtained at 230 ◦ C. The sensing film using the nan- otubes exhibited higher sensor response to toluene at 500 ◦ C than CO and H 2 . The porous morphology of the nanotubular films facili- tated diffusion of large sized molecules like toluene through pores in the film, leading to the improved sensor response and selectivity to toluene. Acknowledgement This study was partly supported by NISSAN Science Foundation, a Grant-in-Aid for Exploratory Research (No. 19659402) from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and by Environmental Research and Technology Develop- ment Fund from the Ministry of Environment of Japan. References [1] N. Yamazoe, G. Sakai, K. Shimanoe, Oxide semiconductor gas sensors, Catal. Surv. Asia 1 (2003) 63–75. [2] D.E. Williams, Semiconducting oxides as gas-sensitive resistors, Sens. Actuators B: Chem. 57 (1999) 1–16. [3] M. Fleischer, H. Meixner, Fastgas sensors based on metal oxides which are stable at high temperatures, Sens. Actuators B: Chem. 43 (1997) 1–10. [4] C. Xu, J. Tamaki, N. Miura, N. Yamazoe, Grain size effects on gas sensitivity of porous SnO 2 -based elements, Sens. Actuators B: Chem. 3 (1991) 147–155. [5] Y.G. Choi, G. Sakai, K. Shimanoe, N. Miura, N. Yamazoe, Wet process-prepared thick filmsof WO 3 for NO 2 sensing, Sens. ActuatorsB:Chem.95 (2003) 258–265. [6] A.M. Ruiz, A. Corneta, K. Shimanoe, J.R. Morante, N. Yamazoe, Effects of various metal additives on the gas sensing performances of TiO 2 nanocrystals obtained from hydrothermal treatments, Sens. Actuators B: Chem. 108 (2005) 34–40. [7] G.S. Devi, T. Hyodo, Y. Shimizu, M. Egashira, Synthesis of mesoporous TiO 2 - based powders and their gas-sensing properties, Sens. Actuators B: Chem. 87 (2002) 122–129. [8] C.M. Carney, S.H. Yoo, S.A. Akbar, TiO 2 –SnO 2 nanostructures and their H 2 sens- ing behavior, Sens. Actuators B: Chem. 108 (2005) 29–33. [9] A.M. Ruiz, G. Sakai, A. Cornet, K. Shimanoe, J.R. Morante, N. Yamazoe, Cr-doped TiO 2 gas sensor for exhaust NO 2 monitoring, Sens. Actuators B: Chem. 93 (2003) 509–518. [10] H.W. Ryu, B.S. Park, S.A. Akbar, W.S. Lee, K.J. Hong, Y.J. Seo, D.C. Shin, J.S. Park, G.P. Choi, ZnO sol–gel derived porous film for CO gas sensing, Sens. Actuators B: Chem. 96 (2003) 717–722. [11] A. Fujishima, T.N. Rao, D.A. Tryk, Titanium dioxide photocatalysis, J. Photochem. Photobiol. C 1 (2000) 1–21. [12] B. O’Regan, M. Grätzel, A low-cost, high-efficiency solar cell based on dye- sensitized colloidal TiO 2 films, Nature 353 (1991) 737–740. [13] S.Y. Huang, L. Kavan, I. Exnar, M. Grätzel, Rocking chair lithium battery based on nanocrystalline TiO 2 (Anatase), J. Electrochem. Soc. 142 (1995) L142–L144. [14] J.H. Braun, A. Baidins, R.E. Marganski, TiO 2 pigment technology: a review, Prog. Org. Coat. 20 (1992) 105–138. [15] Z.L. Wang, Nanobelts, nanowires and nanodiskettes of semiconducting oxides- from materials to nanodevices, Adv. Mater. 15 (2003) 432–436. [16] D.V. Bavykin, J.M. Friedrich, F.C. Walsh, Protonated titanates and TiO 2 nanos- tructured materials: synthesis, properties, and applications, Adv. Mater. 18 (2006) 2807–2824. [17] X. Gao, H. Zhu, G. Pan, S. Ye, Y. Lan, F. Wu, D. Song, Preparation and electro- chemical characterization of anatase nanorods for lithium-inserting electrode material, J. Phys. Chem. B 108 (2004) 2868–2872. [18] Z. Miao, D. Xu, J. Ouyang, G. Guo, X. Zhao, Y. Tang, Electrochemically induced sol–gel preparation of single-crystalline TiO 2 nanowires, Nano Lett. 2 (2002) 717–720. [19] T. Sasaki, M. Watanabe, Semiconductor nanosheet crystallites of quasi-TiO 2 and their optical properties, J. Phys. Chem. B 101 (1997) 10159–10161. [20] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Formation of titanium oxide nanotube, Langmuir 14 (1998) 3160–3163. [21] T. Kasuga, Formation of titanium oxide nanotubes using chemical treatments and their characteristic properties, Thin Solid Films 496 (2006) 141–145. [22] Y. Yang, X. Wang, L. Li, Crystallization and phase transition of titanium oxide nanotube arrays, J. Am. Ceram. Soc. 91 (2008) 632–635. [23] C.C. Tsai,H.Teng,Structural featuresofnanotubes synthesized from NaOH treat- ment on TiO 2 with different post-treatments, Chem. Mater. 18 (2006) 367–373. [24] H. Tokudome, M. Miyauchi, N-doped TiO nanotube with visible light activity, Chem. Lett. 33 (2004) 1108–1109. [25] S. Uchida, R. Chiba, M. Tomiha, N. Masaki, M. Shirai, Application of titania nan- otubes to a dye-sensitized solar cell, Electrochemistry 70 (2002) 418–420. [26] J. Xu, C. Jia, B. Cao, W.F. Zhang, Electrochemical properties of anatase TiO 2 nan- otubes as an anode material for lithium-ion batteries, Electrochim. Acta 52 (2007) 8044–8047. [27] O.K. Varghese, D. Gong, M. Paulose, K.G. Ong, E.C. Dickey, C.A. Grimes, Extreme changes in the electrical resistance of titania nanotubes with hydrogen expo- sure, Adv. Mater. 15 (2003) 624–627. [28] O.K. Varghese, D. Gong, M. Paulose, K.G. Ong, C.A. Grimes, Hydrogen sens- ing using titania nanotubes, Sens. Actuators B: Chem. 93 (2003) 338– 344. [29] I D. Kim, A. Rothschild, B.H. Lee, D.Y. Kim, S.M. Jo, H.L. Tuller, Ultrasensi- tive chemiresistors based on electrospun TiO 2 nanofibers, Nano Lett. 6 (2006) 2009–2013. 520 M H. Seo et al. / Sensors and Actuators B 137 (2009) 513–520 [30] Z. Li, H. Zhang, W. Zheng, W. Wang, H. Huang, C. Wang, A.G. MacDiarmid, Yen Wei, Highly sensitive and stable humidity nanosensors based on LiCl doped TiO 2 electrospun nanofibers, J. Am. Chem. Soc. 130 (2008) 5036– 5037. [31] Y. Zhang, W. Fu, H. Yang, Q. Qi, Y. Zeng, T. Zhang, R. Ge, G. Zou, Synthesis and characterization of TiO 2 nanotubes for humidity sensing, Appl. Surf. Sci. 254 (2008) 5545–5547. [32] G. Sakai, N. Matsunaga, K. Shimanoe, N. Yamazoe, Theory of gas-diffusion con- trolled sensitivity for thin film semiconductor gas sensor, Sens. Actuators B: Chem. 80 (2001) 125–131. [33] N.S. Baik, G. Sakai, K. Shimanoe, N. Miura, N. Yamazoe, Hydrothermal treat- ment of tin oxide sol solution for preparation of thin-film sensor with enhanced thermal stabilityandgas sensitivity, Sens. Actuators B:Chem.65 (2000) 97–100. [34] Y. Shimizu, T. Maekawa, Y. Nakamura, M. Egashira, Effects of gas diffusivity and reactivity on sensing properties of thick film SnO 2 -based sensors, Sens. Actuators B: Chem. 46 (1998) 163–168. [35] S. Xiaoming, L. Yadong, Synthesis and characterization of ion-exchangeable titanate nanotubes, Chem. Eur. J. 9 (2003) 2229–2238. [36] Q. Chen, G.H. Du, S. Zhang, L.M. Peng, The structure of trititanate nanotubes, Acta Crystallogr. B 58 (2002) 587–593. [37] A. Nakahira, W. Kato, M. Tamai, T. Isshili, K. Nishio, H. Aritani, Synthesis of nan- otube from a layered H 2 Ti 4 O 9 ·H 2 O in a hydrothermal treatment using various titania sources, J. Mater. Sci. 39 (2004) 4239–4245. [38] N. Yamazoe, K. Shimanoe, Roles of shape andsizeofcomponent crystals in semi- conductor gas Sensors: I. Response to oxygen, J. Electrochem. Soc. 155 (2008) J85–J92. [39] A. Teleki, N. Bjelobrk, S.E.Pratsinis, Flame-made Nb-andCu-doped TiO 2 sensors for CO and ethanol, Sens. Actuators B: Chem. 130 (2008) 449–457. [40] T. Gessner, K. Gottfried, R. Hoffmann, C. Kaufmann, U. Weiss, E. Charetdinov, P. Hauptmann, R. Lucklum, B. Zimmermann, U. Dietel, G. Springer, M. Vogel, Metal oxide gas sensor for high temperature application, Microsystem Technol. 6 (2000) 169–174. [41] T. Kida, T. Kuroiwa, M. Yuasa, K. Shimanoe, N. Yamazoe, Study on the response and recovery properties of semiconductor gas sensors using a high-speed gas- switching system, Sens. Actuators B: Chem. 134 (2008) 928–933. [42] N. Matsunaga, G. Sakai, K. Shimanoe, N. Yamazoe, Formulation of gas diffusion dynamics for thin film semiconductor gas sensor based on simple reaction- diffusion equation, Sens. Actuators B: Chem. 96 (2003) 226–233. Biographies Min-Hyun Seo received his M. Eng. Degree in Dept. of sensor and display in 2007 from Kyungpook National University. He is currently a Ph.D. course student at the Department of Molecular and Material Sciences in Kyushu University. His current research interests include the development of gas sensors and nanoparticle process- ing. Masayoshi Yuasa has been an Assistant Professor at Kyushu University since 2005. He received his M. Eng. Degree in materials science in 2003. His current research interests include the development of gas sensors and active electrocatalysts for oxygen reduction. Tetsuya Kida has been an Associate Professor at Kyushu University since 2006. He received his M. Eng. Degree in materials science in 1996 and his Dr. Eng. Degree in 2001 from Kyushu University. His current research interests include the develop- ment of gas sensor, nanoparticle processing, and self-assembled inorganic–organic hybrid materials. Jeung-Soo Huh has been a Professor of Kyungpook National university, Korea since 1995. He received B.S. and M.S. from Dept. of Materials Science & Engineering, Seoul National University in 1983 and 1985. He obtained his Ph.D. in Electronic Materials from M.I.T (Massachusetts Institute of Technology) in 1993. His current research interests include gas sensor, odor sensing and medical application of these sensors. Kengo Shimanoe has been a Professor at Kyushu University since 2005. He received the B.E. Degree in Applied Chemistry in 1983 and the M. Eng. Degree in 1985 from Kagoshima University and Kyushu University, respectively. He joined Nippon Steel Corp. in 1985, and received his Dr. Eng. Degree in 1993 from Kyushu University. His current research interests include the development of gas sensors and other functional devices. Noboru Yamazoe had been a Professor at Kyushu University since 1981 until he retired in 2004. He received his M. Eng. Degree in Applied Chemistry in 1963 and his Dr. Eng. Degree in 1969 from Kyushu University. His research interests were directed mostly to the development and application of functional inorganic materials. . homepage: www.elsevier.com/locate/snb Gas sensing characteristics and porosity control of nanostructured films composed of TiO 2 nanotubes ଝ Min-Hyun Seo a ,. 2009 Keywords: TiO 2 nanotube VOC Porosity control Porous film Hydrothermal treatment abstract Preparation and morphology control of TiO 2 nanostructured films for gas sensor applications