A vailable online at www.sciencedirect.com Sensors and Actuators B 129 (2008) 319–326 Influence of polymerization temperature on NH 3 response of PANI/TiO 2 thin film gas sensor Huiling Tai, Yadong Jiang ∗ , Guangzhong Xie, Junsheng Yu ∗ , Xuan Chen, Zhihua Ying State Key Laboratory of Electronic Thin Films and Integrated Devices, School of Optoelectronic Information, University of Electronic Science and Technology of China (UESTC), Chengdu 610054, PR China Received 23 May 2007; received in revised form 5 August 2007; accepted 7 August 2007 Available online 10 August 2007 Abstract Polyaniline/titanium dioxide (PANI/TiO 2 ) nanocomposite thin films were processed on a silicon substrate with gold interdigital electrodes by an in-situ self-assembly approach for NH 3 gas-sensing application, and the effect of polymerization temperature on the gas response of the PANI/TiO 2 thin film gas sensor was investigated. The results showed that the PANI/TiO 2 thin film prepared at 10 ◦ C was superior to those prepared at other temperatures in terms of response properties, which also exhibited good reproducibility, selectivity and long-term stability. UV–vis absorption and surface morphology characterization of the nanocomposite thin films were performed to explain these different gas-sensing properties. The sensing mechanism was also discussed. © 2007 Elsevier B.V. All rights reserved. Keywords: PANI/TiO 2 ; Nanocomposite thin film; NH 3 ; Polymerization temperature; Gas sensor 1. Introduction The importance of qualitative and quantitative analysis of chemical substances has been well understood for the sustainable development of human being and its habitation envi- ronment, and a large volume of research has been focused on the development and fabrication of gas sensors for detection of gaseous chemicals. Choice of suitable sensing materials along with efficient microelectronics for the detection system is the key step in such efforts [1]. At present, the nanocomposite of conducting polymer/metal oxide for gas-sensing application has attracted a lot of attention, as it has been proved that the hybridization could synergize or complement the sensitive prop- erties of pure organic or inorganic gas-sensing material [2–4]. Among various conducting polymers, polyaniline (PANI) was found to be a better choice for gas-sensing material due to its good environmental and chemical stability, ease of synthesis, ∗ Corresponding authors. Tel.: +86 28 83207157; fax: +86 28 83206123. E-mail addresses: jiangyd@uestc.edu.cn (Y. Jiang), jsyu@uestc.edu.cn (J. Yu). inexpensive monomer as well as higher sensitivity, reversible response and shorter response time compared with polypyr- role [1,5]. Therefore, recent different gas sensors based on PANI nanocomposites combined with various metal oxides have been the subject of considerable interest, i.e., PANI/SnO 2 [2,6], PANI/TiO 2 [6,7], PANI/MoO 3 [8], PANI/WO 3 [9], and PANI/In 2 O 3 [3], etc. On the other hand, as a kind of inorganic gas sensitive material, TiO 2 is a typical n-type semiconductor and has received particular attention due to its good stability and environmental-friendliness [4]. Therefore, it is hopeful to obtain novel materials with complementary behaviors between PANI and TiO 2 . Various PANI/TiO 2 nanocomposites have been prepared by using chemical polymerization or electrochemical polymerization of aniline in the presence of nano-colloidal TiO 2 . Moreover, our group also reported a PANI/TiO 2 nanocompos- ite thin film gas sensor prepared at room temperature [10].It has been revealed that the PANI/TiO 2 nanocomposite thin film gas sensor exhibited higher response values, faster response and recover rates to NH 3 than those of a pure PANI thin film sen- sor fabricated under the identical conditions. However, it was also found that the response time and long-term stability of the prepared PANI/TiO 2 gas-sensor was not good enough for prac- 0925-4005/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2007.08.013 320 H. Tai et al. / Sensors and Actuators B 129 (2008) 319–326 tical application. In order to improve the performance of the gas sensor based on the PANI/TiO 2 nanocomposite thin film and develop a usable NH 3 gas sensor, the polymerization tem- perature was systematically optimized for film processing in this work, and the nanocomposites and films were characterized by FTIR, UV–vis spectroscopy and scanning electron micro- copy (SEM). The concentration and temperature characteristics of the PANI/TiO 2 thin films as well as the reproducibility, selectivity and long-term stability of gas sensors were also investigated. 2. Experimental 2.1. Materials Polydialyldimethyldiammonium chloride (PDDA), poly- (sodium-p-styrenesulfonate) (PSS), aniline and colloidal TiO 2 (particle size < 40 nm) were all obtained from Sigma–Aldrich Co. Ammonium persulfate ((NH 4 ) 2 S 2 O 8 , APS) and hydrochlo- ric acid (HCl) were purchased from Chengdu Kelong Chemical Reagent Co. All the chemical and reagents were used as received without further purification. Fig. 1. FTIR spectra of pure PANI and PANI/TiO 2 composite. Fig. 2. UV–vis absorption spectra of a PANI thin film prepared at 20 ◦ C and PANI/TiO 2 nanocomposite thin films prepared at different polymerization tem- peratures. 2.2. Fabrication of PANI/TiO 2 nanocomposite thin films PANI/TiO 2 nanocomposite thin films were synthesized by in- situ self-assembly technique in the presence of colloidal TiO 2 with a dip-coater (KSV Co., Finland). A typical fabrication pro- cess for PANI/TiO 2 nanocomposite thin film was as follows: 0.1 g PSS was dissolved in 50 mL of deionized (DI) water, and adjusted pH to 1 using a hydrochloric acid solution. A positively charged surface was created via the deposition of a 1.0% PDDA aqueous solution for 15 min, and then the positively charged substrate was dipped into the PSS solution for 15 min to obtain a negatively charged surface. Later, the PSS-coated substrate was washed with DI water and dried by a nitrogen blow. The PSS-coated substrate was suitable to fabricate PANI/TiO 2 thin films, where simultaneous polymerization of aniline monomer and oxidation of PANI molecules occurred. The active solu- tion for PANI/TiO 2 nanocomposite material contained colloidal TiO 2 , aniline monomer and HCl followed by the addition of an oxidizing agent (APS). The optimum film of PANI/TiO 2 was achieved with a solution containing 0.1 mL of aniline, 10 mL of the sonicated colloidal TiO 2 (0.1 wt.%), 20 mL of 2.0 M HCl and 10 mL of a hydrochloride solution of APS with an equal molar ratio to aniline. The resulting mixture was kept still for 5 min after the addition of APS and then filtered. Such a fil- tered solution was used for the deposition of PANI/TiO 2 film. The PDDA/PSS substrate was removed from the reaction mix- ture after 20 min and then immersed into a 1.0 M HCl solution for 5 min and dried in air. The polymerization of PANI/TiO 2 nanocomposite was processed under various temperatures, i.e., −10, 0, 10, 20 and 30 ◦ C. A PANI thin film was also fabricated without TiO 2 at 20 ◦ C. Table 1 Maximum absorbancewavelengths of PANI/TiO 2 thin filmsprepared atdifferent polymerization temperature Polymerization temperature ( ◦ C) λ 1 (nm) λ 2 (nm) λ 3 (nm) λ 4 (nm) −10 857 425 366 399 0 874 423 363 390 10 859 415 369 378 20 865 418 360 371 30 840 423 330 359 Table 2 Response time (T 1 ) and recovery time (T 2 ) of sensors based on PANI/TiO 2 thin films prepared at 0, 10 and 20 ◦ C when exposed to NH 3 of various concentrations at room temperature Concentration of NH 3 (ppm) 0 ◦ C10 ◦ C20 ◦ C T 1 T 2 T 1 T 2 T 1 T 2 23 6 70 2 60 2 61 47 8 68 3 56 2 46 70 6 44 2 32 3 29 94 9 50 2 25 7 19 117 6 33 2 23 5 20 141 6 25 2 19 7 21 H. Tai et al. / Sensors and Actuators B 129 (2008) 319–326 321 2.3. Characterization of films FTIR spectra of pure PANI and PANI/TiO 2 nanocomposite samples palletized with KBr were performed by a NICOLET MX-1E Fourier transformed spectrometer. The UV–vis absorp- tion spectra of pure PANI and PANI/TiO 2 nanocomposite thin films were recorded on a Shimadzu UV1700 spectrometer using an uncoated glass as the reference. The surface mor- phology was assessed with a JSM-5900LV scanning electronic microscope. 3. Results and discussion 3.1. FTIR spectra of pure PANI and PANI/TiO 2 nanocomposites The resulting PANI and PANI/TiO 2 solutions described in Section 2.2 were left still at 20 ◦ C for another 6 h. The products were filtered and washed with 2.0 M HCl and DI water each for three times in turn. The products were dried at 80 ◦ C for 12 h under vacuum. The FTIR spectra of pure PANI and PANI/TiO 2 Fig. 3. Dynamic responses of the sensors based on PANI/TiO 2 nanocomposite thin films prepared at (a) −10 ◦ C, (b) 0 ◦ C, (c) 10 ◦ C, (d) 20 ◦ C, and (e) 30 ◦ CtoNH 3 at room temperature. 322 H. Tai et al. / Sensors and Actuators B 129 (2008) 319–326 Fig. 4. Responses of the nanocomposite sensors prepared at different polymer- ization temperature to NH 3 of different concentration. nanocomposites are shown in Fig. 1. The main characteristic peaks of pure PANI are assigned as follows: the bands at 1565 and 1488 cm −1 are attributed to C N and C C stretching mode of vibration for the quinonoid and benzenoid units of PANI; 1291 and 1133 cm −1 are the stretching peak of C N and C N, respectively; the peak at 797 cm −1 is assigned to C H bend- ing vibration out of the plane of the para-disubstituted benzene rings [2,11–13]. For PANI/TiO 2 nanocomposite, the IR spec- trum is almost identical to that of pure PANI, but all bands shift slightly, indicating that some interaction exists between PANI and nano-TiO 2 . In addition, the absorption band at 1408 cm −1 can be assigned to the in-plane bending vibration of O H on the surface of TiO 2 [14]. 3.2. UV–vis absorption spectra of PANI and PANI/TiO 2 nanocomposite thin films Fig. 2 depicts the UV–vis absorption spectra of PANI and PANI/TiO 2 thin films deposited on the PDDA/PSS glass sub- strate under different polymerization temperatures. It shows that Fig. 5. Schematic energy-band diagram for PANI/TiO 2 nanocomposite. three characteristic bands of the doped PANI thin film prepared at 20 ◦ C appear at about 361, 402 and 859 nm, which can be attributed to the –*, polaron–* and –polaron transitions, respectively [6,7,11]. It can be noted that the characteristic peaks of the doped PANI all appear in the PANI/TiO 2 nanocompos- ite thin films, but there are some shifts compared with pure PANI, and some new peaks are observed in the PANI/TiO 2 nanocomposite thin films shown in Table 1, which indicates that encapsulation of nano-TiO 2 particles has the effect on the doping of conducting PANI, while this effect should owe to an interaction at the interface of PANI and nano-TiO 2 particles [11]. Also it could be obviously observed that the absorption inten- sity increases with decreasing the temperature, i.e., the thickness of the in-situ self-assembly deposited PANI/TiO 2 thin films increases with decreasing the temperature, which is consistent with the previously reported results [15]. 3.3. Gas-sensing properties measurements NH 3 gas sensors were designed and fabricated to operate as a resistive element. Gold sputtered interdigital electrodes were fabricated on a 5 mm × 8 mm silicon substrate to form a trans- ducer, which were directly used to measure the resistance change of the sensitive PANI/TiO 2 nanocomposite thin film layer when exposed to NH 3 gas of different concentrations. The device was put into a test box (320 mL), and a certain amount of NH 3 gas was injected into the test chamber after the resistance reached a steady value in clean air. Gas exposure time was ca. 150 s for each pulse of NH 3 gas and the chamber was purged with clean air for ca. 200 s after each pulse to allow the surface of the sensitive film to regain atmospheric condition. A Keithley 2700 data acquisition system was used to measure the resistance variation of the sensors. The measurement was processed at room temperature. Dynamic responses of the sensors based on PANI/TiO 2 thin films fabricated under different polymerization temperatures to NH 3 are shown in Fig. 3a–e. It can be seen that the resistance of all sensors increases dramatically after exposed to NH 3 gas. It is also observed that fast resistance increases were followed by a decrease of resistance, which is most obvious for the thin films Fig. 6. Gas responses of the nanocomposite sensor prepared via polymerization at 10 ◦ C to 23 ppm NH 3 , 23 ppm CO and 500 ppm H 2 . H. Tai et al. / Sensors and Actuators B 129 (2008) 319–326 323 Fig. 7. Sensing reproducibility of the nanocomposite sensor prepared via poly- merization at 10 ◦ C to 23 ppm NH 3 at room temperature. prepared at −10 and 30 ◦ C, indicating that either more than one type of reaction sites are available or that a number of different reactions are possible [16]. It is believed that the porous structure of thin films leads to the predominance of surface phenomena over the bulk material phenomena, and therefore the resistance increases significantly with time initially when PANI nanocom- posite thin films contact with NH 3 by gas injection, which may be due to the surface adsorption effect, and the chemisorptions leads to the formation of ammonium. However, the interaction process between the thin film and the adsorbed gas is a dynami- cal process. Thus, when the thin film is exposed to NH 3 gas, the adsorption and desorption processes will simultaneously occur, and the thinner the films, the quicker the gas desorption. Then, the resistance attains a stable value when dynamic equilibrium is attained. However, further interaction mechanism study needs to be carried out to verify this speculation. Additionally, the sen- sors based on PANI/TiO 2 thin films prepared at −10 and 30 ◦ C exhibit incomplete reproducibility, which is especially serious for the thin films fabricated at −10 ◦ C, as shown in Fig. 3a, indicating a low reproducibility to NH 3 . The gas response is defined as (R gas − R air )/R air , where R air is the resistance of sensor in air and R gas is the steady resistance of sensor in the presence of a tested gas. The response values of all the samples are plotted as a function of NH 3 concentration in Fig. 4, indicating a highly linear characteristic and the highest Fig. 8. Long-term variation in resistance of the nanocomposite sensors prepared via polymerization at room temperature (25 ◦ C) and 10 ◦ C. Fig. 9. Response of the sensor prepared via polymerization at 10 ◦ CtoNH 3 of different concentrations after 0, 6, and 30 days. response value for the sensor composed of the PANI/TiO 2 thin film prepared at 10 ◦ C. We define the response and recovery time as the time required to reach 90% total resistance change. As the sensors based on PANI/TiO 2 thin films prepared at −10 and 30 ◦ C exhibited slow reaction equilibrium and low reproducibility to NH 3 , only the characteristic times of other three sensors in the case of various NH 3 concentrations are presented in Table 2. It shows that the response time (T 1 ) is almost independent on the gas concentra- tion, and a fast response time of 2 s can be observed to different NH 3 concentrations for the PANI/TiO 2 thin film prepared at 10 ◦ C; the recovery time (T 2 ), however, decreases with increas- ing the gas concentration. The reason to cause this phenomenon is that an increase in concentration leads to an increased amount of chemisorbed NH 3 , which in turn enhances the desorption rate and sensing site renewal [1]. Kukla et al. [17] proposed that the mechanism to explain the sensitivity and reversibility of PANI layers to NH 3 was a deprotonation–reprotonation process, and the resistance showed an exponential growth with an increase in NH 3 concentration. However, it can be seen from Fig. 3a–e that the resistance of the Fig. 10. Response to 23 ppm NH 3 of the sensor prepared via polymerization at 10 ◦ C as a function of temperature. 324 H. Tai et al. / Sensors and Actuators B 129 (2008) 319–326 PANI/TiO 2 nanocomposite thin films grew instantaneously and did not follow the resistance change trend observed with PANI sensors. We have postulated that PANI and TiO 2 form a p–n junction and the inter-particle electron transition from TiO 2 to PANI causes the reduction of the activation energy and enthalpy of physisorption for NH 3 gas [10]. Fig. 5 exhibits a schematic energy diagram to further illuminate the NH 3 gas sensing mech- anism of PANI/TiO 2 nanocomposite thin films [14,18], where HOMO presents the highest occupied molecular orbital, and LUMO is the lowest unoccupied molecular orbital. Li et al. [14] showed that charge separation was enhanced due to well energy bandgap matching between the conduction band of TiO 2 and the LUMO level of PANI for charge transfer. Therefore, it is also believed that such enhancement promotes NH 3 gas-sensing ability of the PANI/TiO 2 nanocomposite. Generally, a single chemical sensor has cross-sensitivity, which hinders it from practical application. Herein we measured the responses of the sensor composed of the thin film prepared at 10 ◦ C to 23 ppm CO and 500 ppm H 2 , and the results are shown in Fig. 6. It can be seen that there is a distinct difference in gas responses to the tested gases, and the sensor showed very weak responses to CO and H 2 . Based on this observation, it is believed that the sensor exhibits high selectivity to NH 3 . In addition, the response of this sensor was monitored for the repeated expo- sure and removal of 23 ppm NH 3 up to three cycles, as shown in Fig. 7, indicating high reproducibility and reversibility. Fig. 11. SEM images of PANI/TiO 2 nanocomposite thin films prepared at (a) −10 ◦ C, (b) 0 ◦ C, (c) 10 ◦ C, (d) 20 ◦ C, (e) 30 ◦ C. H. Tai et al. / Sensors and Actuators B 129 (2008) 319–326 325 Our earlier work [10] showed that the resistance of the sen- sor prepared via polymerization at room temperature (25 ◦ C) increased greatly when the sample was stored in air, whereas the response of the sensor to NH 3 decreased by a factor of 2–3 during 30 days. Therefore, the lack of long-term stability of the fabricated sensor was of concern. The resistance changes of the sensors prepared via polymerization at room tempera- ture and 10 ◦ C are shown in Fig. 8. It can be observed that the latter exhibits higher long-term stability in resistance than the former. The excellent stability of the latter was also confirmed by testing its response after a 6- and 30-day period, as shown in Fig. 9, indicating that the response values do not exhibit signifi- cant change. Meanwhile, it was found that the response/recovery time also kept stable. 3.4. Temperature characteristic It is well known that humidity and temperature have a sig- nificant effect on the operation of gas sensors. The effect of humidity on the performance of PANI/TiO 2 nanocomposite thin film sensors was investigated in our earlier work [10], and then the temperature dependence of response was studied here. The response to 23 ppm NH 3 of the sensor based on the thin film fabricated at 10 ◦ C is shown as a function of temperature in Fig. 10. It is observed that the response decreases with increas- ing temperature, which is in good agreement with the results of the reported NH 3 gas sensor based on PANI [17,19]. This indi- cates that the adsorption–desorption equilibrium shifts in the desorption direction with increasing temperature [17]. Accord- ingly, the prepared gas sensor is preferred to be operated at room temperature. 3.5. Surface morphology of PANI/TiO 2 thin films The effect of morphology of sensitive films such as grain size, structural formation, surface-to-volume ratio and film thickness on the gas sensitivity, was well recognized [3]. Therefore, the SEM images of all the PANI/TiO 2 thin films prepared under different polymerization temperatures are shown in Fig. 11.It can be seen that all the films have a very porous structure, inter- connected network of fibers and high surface area. It has been pointed out that such structure contributes to a rapid diffusion of dopants into the film [10]. However, as shown in Fig. 11, the PANI/TiO 2 thin film prepared at −10 ◦ C exhibits a two- dimensional slab surface including some disorderly fibers, and the PANI/TiO 2 thin film prepared at 0 ◦ C has a rough surface and consists of coral-like particulates. It is considered that such kind of structure affects the diffusion of gas into and out of the entire bulk of the fibrous film. In contrast, there are almost no granular particulates in the nanocomposite films prepared at 10, 20 and 30 ◦ C. The difference is that the film fabricated at 10 ◦ C has many uniform cylinders and presents a compact three-dimensional surface, whereas the films fabricated at 20 and 30 ◦ C exhibit a relative sparse network structure. All the above factors are in accordance with the observation that the sensor consisted of the thin film prepared at 10 ◦ C has a faster and higher response to NH 3 gas. Li et al. [20] proposed that during the synthesis of PANI its aggregation was triggered by heterogeneous nucleation, and the nucleation behavior of PANI is strongly dependent on the polymerization rate determined by reaction temperature. As the formation of new embryonic nuclei is much faster at high tem- perature, it is more likely that these embryonic nuclei evolve to create homogeneous nuclei before they can diffuse into heterogeneous nucleation sites to nucleate. Thus, more PANI molecules will precipitate via homogeneous nucleation at higher temperature, and the possibility of heterogeneous nucleation will be decreased. This is also in agreement with the observation that the thickness of in-situ self-assembly films increases with decreasing temperature, as shown in Fig. 2. Therefore, the films prepared at high temperatures have better uniformity than those prepared at −10 and 0 ◦ C. 4. Conclusions The NH 3 sensing capability of PANI/TiO 2 thin film sensors was investigated and the polymerization temperature was opti- mized. The PANI/TiO 2 thin film prepared at 10 ◦ C exhibited stable, reproducible and reversible resistance change in the pres- ence of NH 3 in the range of 23–141 ppm. The response time was kept 2 s, and the recovery rate was also very fast in the range of 20–60 s depending on the NH 3 concentration. The sensor also had high selectivity and long-term stability. The difference of gas-sensing property among sensors prepared under differ- ent polymerization temperatures was characterized with UV–vis spectra and SEM, and a simple schematic energy diagram was presented to illuminate the gas sensing mechanism. The in-situ self-assembly approach proposed in this work is easy-processing and feasible, which is also prone to produce PANI/TiO 2 thin film in large scale. In addition, the prepared PANI/TiO 2 gas sensor is preferred to be operated at room tem- perature, which has potential to develop the practical NH 3 gas sensor at low cost. Acknowledgements This work was supported by National Science Foundation of China via grant no. 60425101 and Program for New Century Excellent Talents in University via grant no. NCET-06-0812. References [1] G.K. Prasad, T.P. Radhakrishnan, D. Sravan Kumar, M. Ghanashyam Krishna, Ammonia sensing characteristics of thin film based on polyelec- trolyte templated polyaniline, Sens. Actuat. B 106 (2005) 626–631. [2] L. Geng, Y. Zhao, X. Huang, S. Wang, S. Zhang, S. Wu, Characterization and gas sensitivity study of polyaniline SnO 2 hybrid material prepared by hydrothermal route, Sens. Actuat. B 120 (2007) 568–572. [3] A.Z. Sadek, W. Wlodarski, K. Shin, R. Bkaner, K. Kalantar-Zadeh, A layered surface acoustic wave gas sensor based on a polyaniline/In 2 O 3 nanofibre composite, Nanotechnology 17 (2006) 4488–4492. [4] X. Zhang, G. Yan, H. Ding, Y. Shan, Fabrication and photovoltaic proper- ties of self-assembled sulfonated polyaniline/TiO 2 nanocomposite ultrathin films, Mater. Chem. Phys. 102 (2007) 249–254. [5] N. Densakulprasert, L. Wannatong, D. Chotpattananont, P. Hiamtup, A. Sirivat, J. Schwank, Electrical conductivity of polyaniline/zeolite compos- 326 H. Tai et al. / Sensors and Actuators B 129 (2008) 319–326 ites and synergetic interaction with CO, Mater. Sci. Eng. B 117 (2005) 276–282. [6] M.K. Ram, O. Yavuz, V. Lahsangah, M. Aldissi, CO gas sensing from ultrathin nano-composite conducting polymer film, Sens. Actuat. B 106 (2005) 750–757. [7] X. Ma, M. Wang, G. Li, H. Chen, R. Bai, Preparation of polyaniline-TiO 2 composite film with in situ polymerization approach and its gas-sensitivity at room temperature, Mater. Chem. Phys. 98 (2006) 241–247. [8] J. Wang, I. Matsubara, N. Murayama, S. Woosuck, N. Izu, The preparation of polyaniline intercalated MoO 3 thin film and its sensitivity to volatile organic compounds, Thin Solid Films 514 (2006) 329–333. [9] N. Parvatikar, S. Jain, S. Khasim, M. Revansiddappa, S.V. Bhoraskar, M.V.N. Ambika Prasad, Electrical and humidity sensing properties of polyaniline/WO 3 composites, Sens. Actuat. B 114 (2006) 599–603. [10] H. Tai, Y. Jiang, G. Xie, J. Yu, X. Chen, Fabrication and gas sensitivity of polyaniline-titanium dioxide nanocomposite thin film, Sens. Actuat. B 125 (2007) 644–650. [11] X. Li, G. Wang, X. Li, D. Lu, Surface properties of polyaniline/nano-TiO 2 composites, Appl. Surf. Sci. 229 (2004) 395–401. [12] A.G. Yavuz, A.E. G¨ok, Preparation of TiO 2 /PANI composites in the pres- ence of surfactants and investigation of electrical properties, Synth. Met. 157 (2007) 235–242. [13] K. Dutta, S.K. De, Optical and nonlinear electrical properties of SnO 2 - polyaniline nanocomposites, Mater. Lett. 61 (2007) 4967–4971. [14] J. Li, L. Zhu, Y. Wu, Y. Harima, A. Zhang, H. Tang, Hybrid composites of conductive polyaniline and nanocrystalline titanium oxide prepared via self-assembling and graft polymerization, Polymer 47 (2006) 7361–7367. [15] J. Stejskal, I. Sapurina, J. Prokes, J. Zemek, In situ polymerized polyaniline films, Synth. Met. 105 (1999) 195–202. [16] A.Z. Sadek, W. Wlodarski, K. Kalantar-Zadeh, C. Baker, R.B. Kaner, Doped and dedoped polyaniline nanofiber based conductometric hydrogen gas sensors, Sens. Actuat. A: Phys. 139 (2007) 53–57. [17] A.L. Kukla, Y.M. Shirshov, S.A. Piletsky, Ammonia sensors based on sensitive polyaniline films, Sens. Actuat. B 37 (1996) 135–140. [18] M.K. Ram, O. Yavuz, M. Aldissi, NO 2 gas sensing based on ordered ultra- thin films of conducting polymer and its nanocomposite, Synth. Met. 151 (2005) 77–84. [19] M. Matsuguchi, A. Okamoto, Y. Sakai, Effect of humidity on NH 3 gas sensitivity of polyaniline blend films, Sens. Actuat. B 94 (2003) 46–52. [20] D. Li, R.B. Kaner, Shape and aggregation control of nanoparticles: not shaken, not stirred, J. Am. Chem. Soc. 128 (2006) 968–975. Biographies Huiling Tai received her BS degree in the field of electronic materials and components from UESTC in 2003. Currently she is a PhD candidate of School of Optoelectronic Information at UESTC. Her major field of scientific interests is conducting polymer and its composite for gas sensor application. Yadong Jiang graduated from Department of Material Science & Engineering at UESTC with a BS degree in 1986. Then he got his MS degree and PhD degree in 1989 and 2001, respectively, from that department at UESTC. He is Professor and Dean of School of Optoelectronic Information at UESTC. His major research interests include optoelectronic material and devices, sensitive materials and sensors. Guangzhong Xie got his MS degree in Physics from Sichuan University. He is the Associate Professor of School of Optoelectronic Information at UESTC since 2001. His research interests are sensitive material and sensors. Junsheng Yu got his PhD degree from Graduate School of Bio-Applications & System Engineering at Tokyo University of Agriculture and Technology in 2001. Currently he is Professor of School of Optoelectronic Information at UESTC. His research field is organic optoelectronic materials and devices. Xuan Chen received herBS degree inchemistry fromUESTC in 2005. Currently she is a MS student of School of Optoelectronic Information at UESTC. Her major is polymer science for gas sensors. Zhihua Ying received her BS degree in 2002 and MS degree in 2005 from UESTC. She is in study for her PhD degree. Her research interests are sensitive materials and sensors. . approach for NH 3 gas- sensing application, and the effect of polymerization temperature on the gas response of the PANI/ TiO 2 thin film gas sensor was investigated A vailable online at www.sciencedirect.com Sensors and Actuators B 129 (2008) 319–326 Influence of polymerization temperature on NH 3 response of PANI/ TiO 2 thin