studies on tin oxide-intercalated polyaniline nanocomposite for ammonia gas

Sensors and Actuators B 138 (2009) 76–84 Contents lists available at ScienceDirect Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb Studies on tin oxide-intercalated polyaniline nanocomposite for ammonia gas sensing applications N.G. Deshpande a,c , Y.G. Gudage a , Ramphal Sharma a,∗ , J.C. Vyas b , J.B. Kim c , Y.P. Lee c a Thin Film and Nanotechnology Laboratory, Department of Physics, Dr. B.A. Marathwada University, Auranganbad 431004 (M.S.), India b Technical Physics and Prototype Engineering Division, Bhabha Atomic Research Center, Trombay, Mumbai 400085, India c Quantum Photonic Science Research Center and BK21 Program Division of Advanced Research and Education in Physics, Hanyang University, Seoul 133-791, Republic of Korea article info Article history: Received 12 August 2008 Received in revised form 22 December 2008 Accepted 2 February 2009 Available online 20 February 2009 Keywords: Conducting polymer Solution route technique Nanocomposite PANI films Surface morphology Optical studies and gas sensor analysis abstract Thin films oftin oxide-intercalated polyaniline nanocomposite have been deposited at room temperature, through solution route technique. The as-grown films were studied for some of the useful physico- chemical properties, making use of XRD, FTIR, SEM, etc. and optical methods. XRD studies showed peak broadening andthe peak positions shift from standard values, indicating presence of tinoxide innanopar- ticles form in the polyaniline (PANI) matrix. FTIR study shows presence of the Sn–O–Sn vibrational peak and characteristic vibrational peaks of PANI. Study of SEM micrograph revealed that the composite par- ticles have irregular shape and size with micellar templates of PANI around them. AFM images show topographical features of the nanocomposite similar to SEM images but at higher resolution. Optical absorbance studies show shifting of the characteristics peaks for PANI, which may be due to presence of tin oxide in PANI matrix. On exposure to ammonia gas (100–500 ppm in air) at room temperature, it was found that the PANI film resistance increases, while that of the nanocomposite (PANI+ SnO 2 ) film decreases from the respective unexposed value. These changes on removal of ammonia gas are reversible in nature, and the composite films showed good sensitivity with relatively faster response/recovery time. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Metal oxide thin-film gas sensors are widely used for detecting gas species by measuring changes in their physical properties on exposure to specificgas, in particularmaking use ofreversibleredox reactions in presence or absence of the specific gas media. Usually, the smallchange inchemical state ofthe film material may reflect as measurable change in some physical properties, such as electrical conductivity, whichcanbe monitored by externalelectricalcircuits. Pure tin oxide, SnO 2 is a remarkable n-type semiconductor material having wide band gap (∼3.6 eV), and by making use of small quan- tity of dopant into it’s matrix, thin films of this material find use in several devices such as flat panel displays, gas sensors [1,2], etc. to name a few. However, the sensors incorporating tin oxide require an elevated temperature (≥200 ◦ C) for their optimum operation. This calls for a separate temperature controlled heater assembly to operate the device, and requiring extra power for heating. In addi- tion, the sensor operation at elevated temperature in itself causes gradual changes in the tin oxide film properties, which in turn devi- ate gas sensing properties of the device with time. Therefore, it is highly desirable to have sensors, which can operate at room tem- ∗ Corresponding author. Tel.: +91 9422793173; fax: +91 240 2403335/3115. E-mail address: ramphalsharma@yahoo.com (R. Sharma). perature, but having comparable properties with that of tin oxide for gas sensing. Conducting polymers (CPs) are in use as an alternative to metal oxide materials for gas sensing applications. Among the CPs, polyaniline (PANI) has become one of the technologically impor- tant CPs, because of it’s relatively easier synthesis, and for having excellent electronic and electro-chromic properties. It has been used in making organic solar cell, as well as gas sensor applications [3–6]. However, PANI is not as sensitive as metal oxides towards gas species, and its poor solubility in organic solvents limits its applica- tions. In spite of these problems with PANI, efforts are being made to improve itssolubilityby involving protonation withorganic acids or preparing it using emulsion polymerization in presence of sur- factants [7]. There have been several reports on improving PANI’s sensitivity and selectivity by making use of new methods, such as its synthesis in nano-structured forms [8,9], or by addition of metal catalysts [10,11], and by combination with other polymers [12]. Recently a new class of materials emerged, known as compos- ites, prepared by mixing suitably the organic and inorganic base materials in proper form. The composite materials have special properties, but as seen in some of the cases, they can also have few desirable properties from both the parent organic and inorganic class of materials. As a consequence, there are growing interests in combining both organic and inorganic materials for applications in electronics, optics, magnetism, etc. [13–15]. In literature, there are 0925-4005/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2009.02.012 N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 77 some reports concerning PANI/inorganic nanocomposite sensors [16–18]. However, very few researchers have studied the composite SnO 2 /PANI for sensor application [19,20]. We fabricated nanocomposites thin films of SnO 2 /PANI by incor- porating SnO 2 particles in the form of colloidal suspensions in PANI through solution route technique. The as-grown composite films were characterized using X-ray diffraction (XRD), Fourier transform infrared (FTIR) spectroscopy, scanning electron microscopy (SEM), atomic force microscopy (AFM),and optical absorbancestudies. The as-grown films were exposed to NH 3 gas at room temperature and the electrical response was noted. For a comparison, thin films of tin oxide, and PANI were also prepared separately, and evaluated along with the tin oxide/PANI composite films for sensing ammo- nia gas at room temperature. We report our findings in this paper and discuss a plausible mechanism for the formation and electronic behaviour of such nanocomposites. 2. Experimental 2.1. Synthesis We employed solution-route technique, to synthesize tin oxide/polyaniline nanocomposites. In this technique, formation of nanocomposites proceeds through an inorganic/organic interface reaction. Tin chloride (SnCl 4 ·5H 2 O), hydrogen peroxide (H 2 O 2 ), aniline, ammonium peroxydisulphate (APS) [(NH 4 ) 2 S 2 O 8 ] and hydrochloric acid (HCl) (all chemicals having AR grade), were pur- chased from M/s Loba Chemie, Mumbai (India). Aniline monomer was distilled under reduced pressure. Initially, SnCl 4 ·5H 2 O, was hydrolyzed, using 2 g of SnCl 4 ·5H 2 O in 50 ml of double distilled water (DDW) with constant stirring, and it’s pH was maintained at ≤4, using dilute HCl. Hydrogen peroxide was added in the above solution, which oxidizes tin ions to tin oxide, and the solution turns into a white colored suspension of SnO 2 and it serves as the starting reaction mixture for further processing. From this reaction mixture, 40 ml volume was taken and mixed with appropriate volume of ani- line, and kept below 4 ◦ C. After 30 min, the APS solution was added in the above mixture to make the reaction bath mixture. In this bath mixture pre-cleaned glass substrates were inserted vertically. It was found that after few minutes the solution color turns bluish to green, which also mark the growth of film on the substrate. The reaction mechanism was studied by monitoring thechanges in pH and temperature of the reaction bath with time [21–24],for both cases, i.e., for baths having with the tin oxide nanoparticles suspension and without it. ThepHand temperature were measured by digital ␮-pH system 361, supplied by Systronics. The as-grown films were washed with DDW and dried. Simi- larly, the precipitate was washed thoroughly using DDW, dried and casted into pellets. In order to study the response of the above films to ammonia gas, silver contacts were made on top of the film surface, by vacuum evaporation technique and making use of shadow masking. For this purpose HIND-HIGHVAC system was used. The chamber pressure during silver evaporation was kept around 0.5 × 10 −5 Torr, and during metal evaporation, film sub- strates were not heated. 2.2. Characterization The physical thickness of the as-grown nanocomposites film was measured using Fizeau fringe technique, and it was about 191nm. The XRD patterns of these films were recorded on a Bruker AXS (D8 Advanced, Germany) diffractometer in the scanning range of 20–70 ◦ (2Â) using Cu K␣ radiation having a wavelength of 1.5405 Å. The infrared spectrum of nanocomposite samples pel- letized with KBr were measured using a Fourier transformed infrared spectrometer (PerkinElmer’s Spectrum1 spectrometer). The surface morphology was studied by field emission scanning electron microscopy (FESEM JEOL-JSM 6500F). The surface mor- phology was studied using an AFM (Nanoscope IIIa produced by Vecco Digital Instruments). The root mean square (rms) sur- face roughness was determined using software provided with the microscope. Absorbance spectra were recorded in a range of 300–1000 nm by means of a PerkinElmer Lambda 25 UV–VIS spec- trophotometer. For evaluating the gas sensing properties of these films, a known concentration of ammonia gas (3N purity, sup- plied by M/s Chemtron Industries, Mumbai) was purged into a test chamber (made up of steel) kept at room temperature, by using micro-syringe. Thegas sensing behaviour oftheas-grown films was determined by measuring the current–voltage (I–V), characteristics in absence/presence ofNH 3 gas,anddata was recorded online,using a computer interfaced with the system. 3. Results and discussion 3.1. Reaction mechanism Variation in pH and temperature of reaction bath with respect to time for both cases, i.e., for bath containing tin oxide nanoparticles Fig. 1. Reaction kinetics for (a) polyaniline and (b) tin oxide/polyaniline nanocom- posite. 78 N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 suspension and without it, is shown in Fig. 1a and b. It may be seen that during the induction period (starting time of about 2–3 min), small change in bath temperature noticed, but subsequently a pro- nounced increase in bath temperature follows, indicating a faster film growth rate, and mainly due to the exothermic nature of aniline polymerization. A simple explanation to describe such a process may be follow- ing. The cation radicals, as the primary intermediates of aniline oxidation, are produced in the homogeneous aqueous medium and a fraction of them is adsorbed at available surfaces sites in random way, as allowed by thermo-dynamical state of the system. These sites initiate growth of future PANI chains, in which the monomer molecules join (with such chains) and in the process give out their own kinetic energy (of free monomer state) to the system. This transfer of energy apparently raises the temperature of the bath by an equivalent amount, and observed during the process of poly- merization (Fig. 1a and b). The heterogeneous catalytic film growth requires formation of initial seeding on the fresh substrate. In the reaction bath without a substrate into it, aniline monomers are in random thermo-dynamic state, and depending upon thermo-chemical conditions can join with each other to form small size (2 and higher monomer units) polymer molecules(such as dimer, trimer,etc.) known asoligomers. In normal conditions, the reverse of this process, i.e., the breaking of oligomers into smaller units also goes on simultaneously in the reaction bath, at about similar rate. At a fresh surface (based on its electro-phobic or electro-philic nature), the reactivity of adsorbed entities (monomer and oligomers) can be substantially enhanced compared with free species, allowing a relatively larger probability of surface attachment [25]. Once initial nucleation takes place on the fresh surface, the activation energy for the next layers (steps) in the polymerization slowly decrease as number of steps increase. Finally, after some time there is no net growth of the film when equilibrium between the joining rate and detaching rate become almost equal. The polymerization at the surface, producing a PANI film, and the polymerization in the bulk, giving rise to a PANI precipitate, proceed in succession, the former having relatively larger rates in the start, but after some finite time it equals with the latter. How- ever, when tin oxide nanoparticles are also present in the reaction bath, these SnO 2 particles impede the growth rate of the PANI film, and we see different times for the induction, and oxidative poly- merization periods. Similarly the pH, and the temperature of the bath are also different for above two cases. Therefore, tin oxide composite film formation should take somewhat larger time, and indeed we observed this difference experimentally as shown in Fig. 1a and b. It was found that for PANI reaction to occur the induction time was about 2 min, polymerization time was about 5 min, the maximum temperature evolved was 39.4 ◦ C and pH was ∼1.17. In case of tin oxide-intercalated polyaniline reaction to occur the induction time was 2 min 40 s, polymerization time was 8 min, maximum temperature was 34.9 ◦ C and pH was 1.01. 3.2. Structural analysis The XRD patterns for tin oxide, PANI and tin oxide/PANI nanocomposites, are shown in Fig. 2a–c, respectively. Fig. 2a reveals that the material deposited is SnO 2 of polycrystalline in nature. On comparing the observed XRD peaks and corresponding planes with the standard (hkl) planes a good matching was seen between the two sets, confirming that the deposited films consist of SnO 2 having primitive tetragonal structure (JCPDS DATA CARD 41-1445). The XRD pattern for tin oxide thin films showed diffraction peaks along (1 10), (1 01), (2 0 0), (2 1 1), (3 1 0) and (3 0 1), respectively. The films were preferentially oriented along (20 0) plane. The aver- age value of lattice parameters was found to be a =b = 4.755 Å and Fig. 2. XRD patterns for (a) tin oxide, (b) polyaniline and (c) tin oxide/polyaniline nanocomposite. c =3.205Å, while the standard bulk value for tin oxide crystalline structure is respectively, a =b= 4.738 Å and c =3.187 Å. This suggests that the tin oxide grains in thin film form are strained, may be due to the smaller average physical size of the grains themselves. The average crystallite size found using the standard Scherer’s formula was equal to 40 nm. Fig. 2b shows the XRD pattern for PANI films, which suggests that the film has amorphous structure. Fig. 2cis the XRD patterns for tin oxide intercalated in the PANI matrix, and one can see the presence of peaks corresponding to tin oxide nano- crystallites. However, these peaks are slightly shifted, from their respective standard positions, may be due presence of PANI matrix. In addition, we observed reduced intensity of the peaks, and rela- tively larger peak broadening, compared with XRD of pure SnO 2 film. This indicates still smaller average size of tin oxide nano- crystallites in composite film, compared that for pure SnO 2 film. The lattice constant was found to be a =b = 4.716 Å and c = 3.24 Å; while the average crystallite size was found to be nearly 23 nm. The (2 00) peak of tin oxide is seen in XRD of composite material shown in Fig. 2c, along with some other peaks. However, intensity of (3 0 1) peak is suppressed in the composite film compared to XRD of pure tin oxide. This suggests that tin oxide is present in the PANI matrix, and presence of PANI has influenced the preferred orientation of tin oxide grains in the film to some extent. 3.3. Fourier transform infrared analysis In order to find the nature of bonding in the film material we studied FTIR spectrum of tin oxide/PANI precipitate. Fig. 3 shows the FTIR spectrum for SnO 2 /PANI nanocomposites, having peaks at wave numbers 1579, 1490, 1446, 1288, 1367, 1160, and 738 cm −1 , N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 79 Fig. 3. FTIR spectrum for tin oxide/polyaniline nanocomposite. respectively. These peaks correspond to most of the characteris- tic peaks for PANI, as described in literature [26,27]. The peaks at wave numbers 1579 and 1490 cm −1 are attributed to C N and C C stretching mode for the quinoid and benzenoid rings; while the peak at wave number 1446 cm −1 is attributed to C–C aromatic ring stretching of the benzenoid diamine unit. The peaks at wave numbers 1288 and 1367 cm −1 are attributed to C–N stretching; and peak at wave number 1160 cm −1 is considered to be due to N Q N stretching. The peak at the wave number 738 cm −1 is attributed to C–H out of plane bending vibrations. However it may be noted that these peaks are slightly shifted with respect to their normal positions as seen for pure PANI films. Once again these peak shift- ings might be due to the presence of tin oxide in the PANI matrix. Furthermore, we observed a strong peak at wave number 615 cm −1 , which is dueto the antisymmetric Sn–O–Snmode in SnO 2 as shown in literature [28–30], and in a way confirms presence of tin oxide in the PANI matrix. Dutta and De [28] have observed similar results for tin oxide/PANI nanocomposites. 3.4. Surface morphological analysis The SEM micrographs of as-grown films of tin oxide, polyani- line and tin oxide/PANI nanocomposites, are shown in Fig. 4a–c, respectively. The SEM profile shown in Fig. 4a, indicates fine gran- ular surface of tin oxide, covering the entire glass substrate, with some agglomeration of finer particulates to form bigger clusters. Such agglomerations result in case of metal oxide films deposited by chemical methods [31,32]. The average grain size was ∼120 nm. In case of pure PANI, the film growth appears to be of dendritic nature, with some part of it having growth of amorphous phase (Fig. 4b). In case of tin oxide/PANI nanocomposites films (Fig. 4c), Fig. 4. FESEM micrographs for (a) tin oxide, (b) polyaniline and (c) tin oxide/polyaniline nanocomposite (inset is a high-resolution magnified image of nanocomposite). the composite particles are highly dispersed, with less amount of agglomeration. The average grain size was ∼80 nm, with dispersion of ±5 nm. The observed difference in the measurement of the grain size by XRD and SEM would be due to the fact that two or more Fig. 5. Schematic diagram of the formation of tin oxide/polyaniline nanocomposite thin films. 80 N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 crystallites may be fused together to form a particle (not resolved by SEM profile, but XRD can figure out easily) [33,34]. The grain growth of nanoparticles in such films may be understood through two basic mechanisms of aggregation. These mechanisms depend on the dispersion of colloidal particles at low- solid volume fractions (ϕ 0 → 0) having with (a) diffusion-limited cluster aggregation (DLCA), and (b) reaction-limited cluster aggregation (RLCA). In DLCA every collision between two clusters results in the for- mation of a new cluster, the aggregate of the two colliding clusters. In RLCA only a small fraction of all the collisions leads to the forma- tion of a new aggregate [35].InFig. 5, a simple schematic is shown, which provides reaction mechanism for formation of such kind of structures. In the present case, formation of polymer shell around the nano-crystalline particle/s can easily be seen in the magnified image inset of Fig. 4c, assisting the growth and further aggre- gates formation indicating a DLCA type mechanism; but it appears that for SnO 2 /PANI film formation, the RLCA mechanism preferably dominates. This is because, the reaction kinetics as reflected from Fig. 1a andb,show arelatively lower temperature of polymerization at low-pH values, and gives rise to limited aggregation. Secondly, the final thickness of the film is limited to around less than 200 nm. In contrast tothis,with theDLCA mode, amuch larger filmthickness can be achieved, but not observed by us. The concentration of surface states has correlation with the roughness and grain size via the surface-to-volume ratio, and the gas sensitivity has a proportional relationship with the film rough- ness. In order to study the surface roughness, the film samples were characterized using AFM. Fig. 6a–c show respective AFM profiles for tin oxide, PANI and SnO 2 /PANI nanocomposites. The rms sur- face roughness was found to be 46.1, 22.7 and 31.5 nm for tin oxide, PANI and SnO 2 /PANI nanocomposites, respectively. Notice that the surface roughness of nanocomposites films 31.5 nm, is in between that of the pure tin oxide and pure PANI films. 3.5. Optical analysis In case of conducting polymers, optical spectroscopy is an important technique to understand the conducting states corre- sponding to the absorption bands of inter-gap and intra-gap states [36]. Usually PANI-HCl shows three characteristic peaks of absorp- tion in wavelength bands 306–324, 402–420 and 828–835 nm, respectively. The peak in wavelength band 306–324 nm is due to the ␲–␲ * transition of benzenoid ring; the peakofwavelength band 402–420 nm, is due to the polaron–␲ * transition and the peak in wavelength band 828–835nm, is attributed to the ␲–polaron tran- sition. In addition, the peaks in wavelength bands 402–420 and 828–835 nm, arise owing to the doping level and the formation of polarons [37–39].InFig. 7, we show optical absorbancewith respect to wavelength, for pure PANI and tin oxide/PANI nanocomposite thin films. The observed absorption peak positions in present case were found at ∼324, ∼430, and ∼828 nm, for pure PANI; whereas in case of tin oxide/PANI nanocomposites these peaks were at ∼303, ∼430, and ∼800 nm, respectively. It is interesting to note that the characteristic peaks of the doped PANI appear in the SnO 2 /PANI nanocomposite thin films, but with some shift in their positions (especially for 324 and 828 nm peaks) compared with the pure film. Such shifts in the characteristic peak positions of one or both of the composite forming species are related with surface modifications, and similar shift of peak positions in CdS/PANI nanocomposites films, has been observed by Pethkar et al. [14]. In addition, there is an increase inthe absorptionat lowerwavelengths inthe SnO 2 /PANI nanocomposites case. This is characteristic property of oxides, indi- cating the presence of tin oxide. Fig. 6. AFM images for (a) tin oxide, (b) polyaniline and (c) tin oxide/polyaniline nanocomposites. 3.6. Gas sensor analysis The as-grown films of tin oxide, polyaniline, and SnO 2 /PANI composites were tested for ammonia gas at room temperature. For this films having metallic contacts were kept in the test chamber of known volume with electrical leads taken out for electrical mea- surements. A fixed amount (corresponding to 100 ppm) of NH 3 gas was injected into the test chamber, and film resistance measured with respect to time (for every 10 s interval), until it reached a steady value. This procedure was followed once again after remov- N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 81 Fig. 7. Optical absorbance versus wavelength for tin oxide/polyaniline nanocom- posites. ing NH 3 and exposing the test chamber to clean air. These steps were repeated for all three different films and for different NH 3 gas concentrations (100–500 ppm). In Fig. 8a–c, we show typical current–voltage characteristics taken for pure tin oxide, pure PANI, and the SnO 2 /PANI composite films kept at room temperature (RT), respectively. It is seen from Fig. 8a that no appreciable change noticed in the film resistance for the case of pure tin oxide film, on exposure to different concentrations of NH 3 gas, and tin oxide films remained insensitive to this gas at RT. However, in case of pure PANI films (Fig. 8b), we see large changes in the film resistance on NH 3 gas exposure. The filmresistance increases by more than an order of magnitude from its original value within a minute, indicating that the electrical resistance of PANI films is a sensitive parameter in the presence of ammonia gas, as reported earlier in literature [40,41]. The I–V characteristics of the composite films show a different but more interesting phenomenon as may be seen from Fig. 8c, that the composite SnO 2 /PANI film resistance decreases on exposure to ammonia (∼300 ppm). Furthermore, the I–V characteristics of com- posite SnO 2 /PANI films show a diode-like exponential behaviour, a characteristic of percolation in disordered systems, wherein the electrical conductance is through hopping mechanism. Kukla et al. [41] proposed that the sensitivity and reversibility of pure PANI lay- ers to NH 3 gas exposure is a deprotonation–reprotonation process, and the film resistance show an exponential rise with increase in NH 3 concentration, this mechanism seems to fit with our obser- vations. However, the decrease in resistance of composite film on exposure to ammonia gas needs further explanation. It is well known that tin oxide is an n-type semiconductor, while PANI films are normally of p-type semiconductor. This is due to the fact that duringthe polymerization process of aniline, acids(such as HCl) are used, which acts as dopant for PANI molecules, and usually bound with the central N atom of aniline (monomer) molecule, like H + N Cl − (other bonds on sides of N atom are left here for want of clarity, and more details are provided in literature, see Fig. 4 of Ref. [41]). In equilibrium at room temperature, the positive charge of bonded hydrogen shifts on N atom, making the structure looks like H N + Cl − . While the negative charge on Cl − is retained with it and remains localized, the positive charge on nitrogen becomes mobile charge in PANI matrix, via its other bonds, making the PANI as a p-type semiconductor [41]. In presence of SnO 2 crystallites, the PANI matrix gets a modified structure electronically. The PANI molecules encapsulate each SnO 2 crystallite, similarly to Fig. 5. The SnO 2 crystallites being an n-type Fig. 8. I–V curves (in the presence of ammonia gas) for (a) tin oxide, (b) polyaniline and (c) tin oxide/polyaniline nanocomposites. surrounded by p-type PANI molecules make a p–n junction like formation locally, immersed within PANI matrix of the composite film. The n-type nature of SnO 2 crystallites annihilate the holes of PANI molecules, near its boundary making a depletion layer like region, which in turn makes the overall PANI matrix electrically more insulating in nature. A tentative explanation of change in electrical resistance of com- posite film may be following. On exposing the composite film with ammonia (which can be permeated into the PANI matrix freely), some of the NH 3 molecules might reach into the depletion region, which is surrounding the SnO 2 crystallite and act as a dielectric 82 N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 Fig. 9. Sensitivity versus concentration plot for tin oxide, polyaniline and tin oxide/polyaniline nanocomposites. betweenthePANI and SnO 2 border. Thedepletion region fieldmight polarize the ammonia molecules, and in turn provide a positive charge to PANI molecules, which can become mobile on its transfer to the central N atom of PANI molecule. So in all this process cre- ates somefree holeson PANI molecules, which increasethehopping conductivity of the film, and therefore make the composite film rel- atively more conducting electrically. Once the process of polarizing the ammonia moleculesby p–n junctionlike formation issaturated, this mechanism cannot generate additional holes in the composite PANI film and therefore no additional change in the film conductiv- ity even by further additionof ammonia to it. Inthe present casethis saturation happened at around 300 ppm, as seen in Fig. 9. However, it may be noted that ammonia gas within PANI regions of the com- posite film, opens up another channel parallel to above mechanism always present in case of pure PANI films, making them more resis- tive on gas exposure. In this channel, ammonia molecules exchange the mobile hole charge with central N atom of PANI molecule and make it localized. This reduces the conductivity of the film, as found in caseof purePANIfilms. So onexposurewith ammonia, both of the channels compete with each other, and the dominating channels dictate the direction of net change in resistance of the composite film. Sensitivity (S%) is defined as the relative variation of the resistance of the sensitive film in percent per ppm of applied gas concentration, i.e., (|Rgas − Rair|/Á·Rair) × 100, whereas gas response is defined as |Rgas − Rair|/Rair, ‘Rair’ is the resistance of sensor in air, ‘Rgas’ is the steady resistance of sensor in the pres- ence of a test gas and ‘Á’ is the concentration of gas (in ppm). In Fig. 9, we show sensitivity (S%) of pure tin oxide, pure PANI, and the tin oxide/PANI nanacomposite film, on exposure to ammonia for different concentrations (100–500 ppm). For the case of pure tin oxide film, no response found (i.e., having response value 1, or no change in film resistance) within explored range. However, for purePANIfilms theresponse value increaseslinearly upto 300 ppm, and saturate thereafter or slightly decrease for larger ammonia gas concentrations. In case of SnO 2 /PANI nanocomposite film, a smooth increase of response was seen up to 300 ppm, and it remains same thereafter. It can be seen that at 300 ppm concentration of ammo- nia gas, both pure PANI, and SnO 2 /PANI composite films hadhighest response. We also studied response and recovery time of the films with respectto ammoniagas exposure. Theresponse time,and the recov- ery time are defined as the time required for a film resistance to Fig. 10. Sensing reproducibility and reversibility curves for (a) polyaniline and (b) tin oxide/polyaniline nanocomposites. reach 90% of its saturation value from the starting value on gas exposure, and on removal of the gas, respectively. In our case, the PANI films had relatively faster response times ∼8–10 s, but as usual the recovery times were relatively larger, around 160s. Notice that the larger recovery times are due to the slower out diffusion rate (concentration dependent) of the gas, which always decreases as time progress. Furthermore, these diffusion rates are small at room temperature. The SnO 2 /PANI nanocomposites films have response times of 12–15 s, and the recovery times around 80 s. It may be seen that the SnO 2 /PANI nanocomposites films showed faster recovery time (a factor of 2) as compared to the PANI films. In Fig. 10a and b, we show typical response of the film with respect to time, for repeated exposure and removal of ammonia (300 ppm) gas, and it may be seen that both PANI and SnO 2 /PANI nanocomposites films showed goodreproducible resistance changefor a numberof cycles. 4. Conclusions We synthesized tin oxide-intercalated polyaniline nanocompos- ites (SnO 2 /PANI) in thin film form, and compared the properties of the composite films with that of the thin films made from the constituent base materials. XRD studies were used to find particu- late size, while FTIR study showed presence of both SnO 2 and PANI molecules. SEM micrograph of these nanocomposite films revealed that the constituent composite particles have irregular shape and size, and encapsulated by fibrous PANI matrix. It was found that N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 83 pure SnO 2 films remain inert on NH 3 gas exposure at RT. However, presence ofSnO 2 crystallites inthe nanocomposites SnO 2 /PANI film changesthe electronicproperty ofPANImatrix in drastic way. While pure PANI films become more resistive on exposure to NH 3 gas, the composite film becomes less resistive on a similar exposure. We have provided a suitable explanation for such behaviour of these films. These SnO 2 /PANI nanocomposites films showed good sensi- tivity, reproducibility with relatively faster response for ammonia gas, at room temperature. In addition, the nanocomposites films showed faster recovery time (twice) as compared with the PANI films. However, there are still many other issues pertaining to gas sensing activity which need more attention, such as long-term sta- bility, selectivity with specific gas, etc. and need further research in this field. Acknowledgments We are thankful to BRNS-DAE Project No. 2005/34/1/BRNS/380 for financial assistance to carry out the research work. We are also thankful to Head, Department of Physics, Dr. B.A.M. Univer- sity, Aurangabad for providing the lab facilities. In addition, we highly acknowledge the help rendered by Dr. R.S. Devan and Prof. Y. Ma, Department of Physics, National Dong Hwa University, Taiwan for doing SEM characterization of our samples as well as help- ful discussions. Authors especially, N.G. Deshpande (currently), J.B. Kim and Y.P. Lee were supported by the KOSEF through Quan- tum Photonic Science Research Center, Seoul, Korea, and by MEST, Korea. References [1] M.J. Madou, S.Y. Morison, Chemical Sensing with Solid State Devices, Academic Press, San Diego, 1989. [2] N. Barsan, U. Weimar, Understanding thefundamental principles of metal oxide based gas sensors; the example of CO sensingwith SnO 2 sensors inthe presence of humidity, J. Phys. Condens. Matter 15 (2003) R813–R839. [3] M. Gratzel, Photoelectrochemical cells, Nature 414 (2001) 338–344. [4] A.G. MacDiarmid, Synthetic metals: a novel role for organic polymers (nobel lecture), Angew. Chem. Int. Ed. 40 (2001) 2581–2590. [5] A.G. MacDiarmid, A.N. Xia, J.M. Wiesinger, Electrically Active Polymers, US Pat. 5,773,568, 1998. [6] Y.S. Negi, P.V. Adhyapak, Development in polyaniline conducting polymers, J. Macromol. Sci. Polym. Rev. 42 (2002) 35–53. [7] G.G. Wallace, G.M. Spinks, A.P. Kane-Maguire, P.R. Tesdale, Conductive Elec- troactive Polymers: Intelligent Materials Systems, CRC Press, 2002. [8] J. Huang, S. Viriji, B.H. Weiller, R.B. Kaner, Nanostructured polyaniline sensors, Chem. Eur. J. 10 (2004) 1314–1319. [9] J. Huang, R.B. Kaner, The intrinsic nanofibrillar morphology of polyaniline, Chem. Commun. 4 (2006) 367–376. [10] H.S. Li, M. Josowicz, D.R. Baer, M.H. Engelhard, J. Janata, Preparation and char- acterization of polyaniline–palladium composite films, J. Electrochem. Soc. 142 (1995) 798–805. [11] A.A. Athawale, S.V. Bhagwat, P.P. Katre, Nanocomposite of Pd–polyaniline as a selective methanol sensor, Sens. Actuat. B 114 (2006) 263–267. [12] S.T. McGovern, G.M. Spinks, G.G. Wallace, Micro-humidity sensors based on a processable polyaniline blend, Sens. Actuat. B 107 (2005) 657–665. [13] S.J. Su, N. Kuramoto, Processable polyaniline–titanium dioxide nanocompos- ites: effect of titanium dioxide on the conductivity, Synth. Met. 114 (2000) 147–153. [14] S.Pethkar, R.C.Patil,J.A. Kher,K. Vijaymohanan,Deposition andcharacterization of CdS nanoparticle/polyaniline composite films, Thin Solid Films 349 (1999) 105–109. [15] K.R. Reddy, K.P. Lee, A.I. Gopalan, Self-assembly approach for the synthesis of electro-magnetic functionalized Fe 3 O 4 /polyaniline nanocomposites: effect of dopant onthe properties, Colloid Surf. A: Physicochem. Eng. Aspects 320 (2008) 49–56. [16] 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. [17] T. Taka, Humidity dependency of electrical conductivity of doped polyaniline, Synth. Met. 57 (1993) 5014–5019. [18] S.S. Joshi, T.P. Gujar, V.R. Shinde, C.D. Lokhande, Fabrication of n-CdTe/p- polyaniline heterojunction-based room temperature LPG sensor, Sens. Actuat. B 132 (2008) 349–355. [19] M.K. Ram, O. Yavuz, M. Aldissi, NO 2 gas sensingbased on ordered ultrathin films of conducting polymer and its nanocomposites, Synth. Met. 151 (2005) 77–84. [20] L. Geng, Y. Zhao, X. Huang, S. Wang, S. Zhang, S. Wu, Characterization and gas sensitivitystudy of polyaniline/SnO 2 hybridmaterial prepared by hydrothermal route, Sens. Actuat. B 120 (2007) 568–572. [21] Y. Fu, R.L. Elsenbaumer, Thermochemistry and kinetics of chemical polymer- ization of aniline determined by solution calorimetry, Chem. Mater. 6 (1994) 671–677. [22] T. Sulimenko, J. Stejskal, J. Proke ˇ s, Poly(phenylenediamine) dispersions, J. Col- loid Interf. Sci. 236 (2001) 328–334. [23] Y. Fong, J.B. Schlenoff, Polymerization ofaniline using mixed oxidizers, Polymer 36 (1995) 639–643. [24] P.M. Beadle, Y.F. Nicolau, E. Banka, P. Rannou, D. Djurado, Controlled polymer- ization of aniline at sub-zero temperatures, Synth. Met. 95 (1998) 29–45. [25] P.W. Atkins, Physical Chemistry, 3rd ed., Oxford University Press, Oxford, 1986, pp. 762–789. [26] T. Abdiryim, Z.X. Gang, R. Jamal, Comparative studies of solid-state synthesized polyaniline doped with inorganic acids, Mater. Chem. Phys. 90 (2005) 367–372. [27] S. Quillard, G. Louarn, S. Lefrant, A.G. MacDiarmid, Vibrational analysis of polyaniline: a comparative study of leucoemeraldine, emeraldine, and perni- graniline bases, Phys. Rev. B 50 (1994) 12496–12508. [28] K. Dutta,S.K. De, Optical andnonlinear electrical properties of SnO 2 –polyaniline nanocomposites, Mater. Lett. 61 (2007) 4967–4971. [29] G. Zhong, M. Liu, Preparation of nanostructured tin oxide using a sol–gel pro- cess based on tin tetrachloride and ethylene glycol, J. Mater. Sci. 34 (1999) 3213–3219. [30] S. Monredon, A. Cellot, F. Ribot, C. Sanchez, L. Armelao, L. Gueneau, L. Delattre, Synthesis and characterization of crystalline tin oxide nanoparticles, J. Mater. Chem. 12 (2002) 2396–2400. [31] T.P. Niesenand, M.R. De Guire, Review: deposition of ceramic thin films at low temperatures from aqueous solutions, J. Electroceram. 6 (2001) 169–207. [32] G. Korotcenkov, V. Tolstoy, J. Schwank, Successive ionic layer deposition (SILD) as a new sensor technology: synthesis and modification of metal oxides, Meas. Sci. Technol. 17 (2006) 1861–1869. [33] D.S. Sutrave, G.S. Shahane, V.B. Patil, L.P. Deshmukh, Micro-crystallographic and optical studies onCd 1−x Zn x Se thinfilms,Mater. Chem. Phys. 65(2000) 298–305. [34] C.M. Shen, X.G. Zhang, H.L. Li, Influence of different deposition potentials on morphology and structure of CdSe films, Appl. Surf. Sci. 240 (2005) 34–41. [35] P. Sandkuhler, J. Sefcik, M. Morbidelli, Kinetics of gel formation in dilute dis- persions with strong attractive particle interactions, Adv. Colloid Interf. Sci. 108–109 (2004) 133–143. [36] 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. [37] A.G. Macdiarmid, A.J. Epstein, The concept of secondary doping as applied to polyaniline, Synth. Met. 65 (1994) 103–116. [38] H. Jiang,Y. Geng,J.Li, F.Wang, Organic aciddoped polyaniline derivatives, Synth. Met. 84 (1997) 125–126. [39] B.J. Kim, S.G.Oh, M.G.Han, S.S. Im, Synthesis and characterization ofpolyaniline nanoparticles in SDS micellar solutions, Synth. Met. 122 (2001) 297–304. [40] D.S. Sutar, N. Padma, D.K. Aswal, S.K. Deshpande, S.K. Gupta, J.V. Yakhmi, Prepa- ration of nanofibrous polyaniline films and their application as ammonia gas sensor, Sens. Actuat. B 128 (2007) 286–292. [41] A.L. Kukla, Y.M. Shirshov, S.A. Piletsky, Ammonia sensors based on sensitive polyaniline films, Sens. Actuat. B 37 (1996) 135–140. Biographies Mr. N.G. Deshpande is currently working for his PhD degree (from 2008) in Depart- ment of Physics, Hanyang University, Seoul, South Korea under the supervision of Prof. YoungPak Lee. Current interest of research work is 1D and 2D magnetic photonic crystals and their applications. Earlier worked as Senior Research Fellow (SRF) in BRNS-DAE project related to oxides, polymers and hybrid materials for gas sensor application (2005–2008). He published nearly 18 international research papers and attended/presented (research work) at various international/national conferences. Mr Y.G. Gudage is currently working for his PhD degree (from 2006) in Depart- ment of Physics, Dr. B.A. Marathwada Univeristy, Aurangabad (M.S.), India under the supervision of Dr. Ramphal Sharma. Current research interest is photoelectrochem- ical solar cells. He worked as Senior Research Fellow (SRF) in BRNS-DAE project on gas sensor applications. He has published nearly 12 international research papers and attended various conferences. Dr. Ramphal Sharma received his PhD in 1991 from Rajasthan University, Jaipur, India. Currently, he is Associate Professor at Department of Physics, Dr. B.A.M. Uni- versity, Aurangabad (M.S.), India. Currently, he is a Brain Pool Fellow in Department of Chemistry, Hanyang University, Seoul, Korea. He has more than 15 years of expe- rience in teaching field; while 20 years of experience in research, i.e., in thin film technology. He has published more than 80 international and national papers in reputed journals. His main interest of research is gas sensor, photosensor and solar cells. He was visiting fellow of ICTP, Trieste, Italy in 1999–2001. Dr. J.C. Vyas postgraduated in Physics from University of Rajasthan, Jaipur, and received PhD from Bombay University, Mumbai. He joined BARC in 1980, and over years worked in several different fields of technical interests, such as fabrication of space quality Si solar cells, growth and characterization of non-linear optical 84 N.G. Deshpande et al. / Sensors and Actuators B 138 (2009) 76–84 single crystals, oriented thin films growth using MBE and their characterization, high-temperature superconducting thin films based weak links for device applica- tions, and thin film based gas sensors. He is a member of Indian Thermal Analysis Society, Material Research Society of India, etc. Mr. JinBae Kim received his BS and MS degrees in Department of Physics of Sunmoon University, Korea, in 2000 and 2002, respectively. He has been a PhD candidate in Department of Physics from Hanyang University from 2002. He is cur- rently focused on the physics and applications of magnetic nanostructures and magnetic photonic crystals. He has published nearly 20 papers in international journals and attended/presented his work at various reputed international/national conferences. Prof. YoungPak Lee is currently Director of Quantum Photonic Science Research Center and Distinguished Professor in Department of Physics, Hanyang University, Seoul, Korea. He received his PhD degree in Condensed-Matter Physics, Iowa State University, Ames, Iowa, U.S.A. (1987). Besides this he has worked at various reputed posts and has been awarded many honors from Ministry of Science and Technology, Korea and others. His research interest is magnetic photonic crystals, meta-materials, nanomagnetism. . of ammonia gas) for (a) tin oxide, (b) polyaniline and (c) tin oxide /polyaniline nanocomposites. surrounded by p-type PANI molecules make a p–n junction. versus concentration plot for tin oxide, polyaniline and tin oxide /polyaniline nanocomposites. betweenthePANI and SnO 2 border. Thedepletion region fieldmight polarize