SnO2/MoO3-nanostructure and alcohol detection

Sensors and Actuators B 118 (2006) 156–162 SnO2/MoO3-nanostructure and alcohol detection J. Arbiol a,b,∗ , J.R. Morante b , P. Bouvier c , T. Pagnier c , E.A. Makeeva d , M.N. Rumyantseva d , A.M. Gaskov d a TEM-MAT, SCT, Universitat de Barcelona, Lluı́s Solé i Sabarı́s 1-3, E-08028 Barcelona, CAT, Spain b EME, Departament d’Electrònica, Universitat de Barcelona, E-08028 Barcelona, CAT, Spain c LEPMI-ENSEEG, 1130, rue de la Piscine, BP 75, Saint Martin d’Hères 38402, France d Chemistry Department, Moscow State University, 1-3 Leninskie Gory, 119992 Moscow, Russia Available online 23 May 2006 Abstract SnO2 /MoO3 mixed samples from 0 to 100 mol% MoO3 have been prepared by using the sol–gel method and calcined at 500 ◦ C during 24 h. The sensor response of the samples towards alcohols, Cn H2n+1 OH (n = 1–4), and NH3 will be interpreted in terms of the acid–base properties of the nanocomposites. To support it, a detailed high-resolution transmission electron microscopy, electron energy loss spectroscopy and RAMAN analysis have been performed allowing to determine the structural, morphological and compositional evolution of this nanostructured binary system. The change on the sensing characteristics related with the Mo content on samples is explained attending to these analytical characterization results that point out the presence of Mo2 O3 in the very small grain of SnO2 , about 2.5 nm, before their segregation in determining the sensing characteristics. © 2006 Elsevier B.V. All rights reserved. Keywords: SnO2 /MoO3 ; Semiconductor gas sensor; Nanometal oxide binary systems; Surface acidity; HRTEM; EELS; Raman 1. Introduction In late 1990s, SnO2 and MoO3 mixtures were shown to have excellent catalytic properties for selective oxidation of methanol and other organic compounds [1]. It is well known that the addition of metals or metal oxides with catalytic properties can influence on the gas sensing behavior of the SnO2 material [2–10]. In this case, the addition of MoO3 to SnO2 has been proposed as an outstanding alternative to modify the sensor response to certain gas species as the presence of Mo atoms at the surface of SnO2 changes the acidity performances which varies mainly its reactivity with alcohols, ammonia or amine groups. Nowadays, these performances have strengthened the practical interest on nanocomposites of SnO2 (semiconducting oxide) and MoO3 (d-metal oxide exhibiting catalytic activity) for the development of resistive gas sensors as the catalytic characteristics in this nano-binary system are enhanced [11–15], presenting a very high active surface value. SnO2 /MoO3 composites are n-type semiconductors, just as the constituent oxides [12–14]. The introduction of molybdenum ∗ Corresponding author. E-mail address: arbiol@ub.edu (J. Arbiol). URL: http://nun97.el.ub.es/∼arbiol. 0925-4005/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2006.04.021 notably reduces the electrical conductivity of SnO2 in air, and, according to the previous literature, it may be associated either with the transfer of electrons trapped at oxygen vacancies (V0 •• and V0 • ) to Mo6+ ions [14,15], Mo6+ + V 0 • = Mo5+ + V 0 •• (1) Mo6+ + 2V 0 • = Mo4+ + 2V 0 •• (2) or with the formation of Moi 5+ interstitials in the structure of SnO2 by the quasi-chemical reaction [12,13] MoSn 6+ + e′ = Moi 5+ (3) Thus, it is assumed SnO2 could contain Mo ions in three different oxidation states [11–13]: Mo6+ , Mo5+ , and Mo4+ . Ivanovskaya et al. [12] from electron spin resonance (ESR) measurement have corroborated that some Mo atoms present Moi 5+ state in enough concentration to justify the conductivity diminution. Up to now, the gas-sensing properties of SnO2 –MoO3 materials have only been investigated at low MoO3 contents (≤35 mol%). Furthermore, only a very few works have been devoted to obtain a morphological and structural characterization of the SnO2 /MoO3 binary system in order to justify the modified sensing characteristics. All these results encouraged J. Arbiol et al. / Sensors and Actuators B 118 (2006) 156–162 157 us to prepare a series of nano-SnO2 /MoO3 mixed samples from 0 mol% MoO3 (pure SnO2 ) to 100 mol% (pure MoO3 ) in order to obtain their sensing response versus different gas species, and also perform a detailed analysis of samples morphology and nanostructure evolution able to explain the changes in the sensing characteristics. In this framework, high-resolution transmission electron microscopy (HRTEM), and electron diffraction have been used to analyze the nanostructure evolution of our samples with MoO3 /SnO2 mol% ratio. To complete our work, electron energy loss spectroscopy (EELS) has also been used to obtain the sample evolution of the local electronic state of the nanoparticles oxygen atoms (in the form of a density of states) [16–18] and finally Raman spectroscopy has been applied to support the discussion on the present MoO3 phases. 2. Experimental SnO2 /MoO3 nanocomposites were prepared by impregnation of dried ␣-stannic acid gel by (NH4 )Mo7 O24 . All of the samples were annealed at 500 ◦ C for 24 h [19]. The structural and morphological characterization of our samples was carried out by means of transmission electron microscopy (TEM) and selected area electron diffraction (SAED). In order to obtain the high-resolution TEM (HRTEM) results we used a field emission gun microscope Jeol 2010F, which works at 200 kV and has a point-to-point resolution of 0.19 nm. Electron energy loss spectroscopy (EELS) spectra were obtained in a Gatan Image Filter (GIF 2000) coupled to the Jeol 2010F microscope. Spectra obtained achieved an energy resolution of 1.2 eV. Raman spectra of SnO2 –MoO3 samples heat-treated at 500 ◦ C were obtained using a Raman spectrometer in micro Raman configuration. The excitation line used in both cases was the green (514.53 nm) line of an Ar laser (Coherent Scientific). A 50× objective with a high working distance was used. Because of the high light absorbance of MoO3 , care was taken to use low laser power (less than 2.0 mW on the sample) in order to avoid any heating. The composition of the samples was determined by electron probe X-ray microanalysis (EPXMA) on a JEOL JSM-840A/PGT IMIX (accelerating voltage: 20 kV). Fig. 1. Sensor response of the samples to consecutive injections (1 ␮l) of ethanol at 275 ◦ C. Inset: sensor response of a sample containing 3 mol% MoO3 to consecutive injections (1 ␮l) of different alcohols at 300 ◦ C (C1, methanol; C2, ethanol; C3, propanol; C4, butanol). stronger its electron-donor effect is, which facilitates hydrogen elimination from the ␣-carbon atom. The diminution in sensor response in going from propanol to butanol and from straight- to branched-chain alcohols is attributable to the steric effect (Fig. 1 inset). Then the decreasing of SnO2 response toward alcohols with the addition of MoO3 (Fig. 1) can be explained by increasing the acidity of tin dioxide surface and, hence, the contribution of dehydration to the oxidation of alcohols. The fact that introduction of molybdenum increased sensor signal to ammonium is the proof of this assumption (Fig. 3). To validate these characteristics and their dependence with the Mo content a detailed structural and morphological characterization has been done. TEM general view micrographs show the presence of small nanoparticles for those samples with low mol% MoO3 (up to 15 mol%) (Fig. 4a). However, above 15 mol% MoO3 we started to find big particles (sizes above 100 nm) cohabiting with the small nanoparticles (Fig. 4b). For samples above 71 mol% MoO3 the small nanoparticles faints and just the big particles are observed (Fig. 4c). A SAED analysis was performed in all samples in order to relate their morphological evolution with 3. Results and discussion Sensor properties of a number of SnO2 /MoO3 nanocomposites toward alcohols Cn H2n+1 OH (n = 1–4) [19] and NH3 were studied by in situ conductance measurements. In all cases, the sensor response to the alcohols decreases with increasing MoO3 content. The sensor response was found to depend on the number of carbon atoms in the alcohol molecule (Fig. 1 inset) and its structure (Fig. 2). These results are explained in terms of a model taking into account the possible mechanisms of interactions on oxide surfaces: dehydrogenation to an aldehyde (on basic surface) and dehydration to an alkene (on acid surface) [19,20]. The increase in the sensor response of pure SnO2 in going from methanol to propanol can be explained by prevalence of dehydrogenation mechanism and by role of the length of the carbon chain. Namely the larger the substituent group, the Fig. 2. Variation in the conductance of a thick nanocomposite film containing 1.6 mol% MoO3 in response to consecutive injections (1 ␮l) of C4 H9 OH isomers at 300 ◦ C; C4, n-butanol; i-C4, iso-butanol, t-C4, tert-butanol. 158 J. Arbiol et al. / Sensors and Actuators B 118 (2006) 156–162 Table 1 Phase distribution after SAED patterns analysis. Samples SnMo (mol% Mo) 0–10 15–20 27–100 SnO2 [21] ␣-MoO3 [22] ␤-MoO3 [23] XXX ----- XX --X XX → - - X → XXX --- XXX, high content; XX, moderate content; X, low content; - - -: null content. Fig. 3. Sensor response of the samples to 500 ppm NH3 in air, Tmeas = 350 ◦ C. possible changes on the materials structure. The insets shown in Fig. 4a–c correspond to the SAED patterns obtained on samples with 5, 20 and 77 mol% MoO3 , respectively. The structural phase distribution identified on SAED patterns is summarized in Table 1. For low MoO3 content (until 10 mol% MoO3 ) the unique phase able to be detected from SAED patterns was the SnO2 cassiterite phase. There exist a transition range (from 15 to 20 mol% MoO3 ) where a ␤-MoO3 phase is already detected from SAED patterns cohabiting with the SnO2 . However, above 20 mol% MoO3 the ␤-MoO3 phase rapidly disappears leading to an ␣-MoO3 phase that becomes predominant in front of the Fig. 4. (a) TEM micrograph showing a general view of the nanoparticles found in sample with 5 mol% MoO3 . (b) TEM general view of the particles found in sample with 20 mol% MoO3 . (c) TEM general view of the grains found in sample with 77 mol% MoO3 . The insets placed in figures (a)–(c) are the SAED patterns corresponding to those samples. (d) Grain size evolution of the “small” SnO2 nanoparticles. J. Arbiol et al. / Sensors and Actuators B 118 (2006) 156–162 159 SnO2 cassiterite. Likewise, this SnO2 signal also very rapidly weakens as MoO3 content increases. In fact, if we have a deeper look at the SAED patterns, we observe that the brightest spots (corresponding to big particles) can be indexed as MoO3 phases, while the diffuse or faint rings (small nanoparticles) are indexed as pure cassiterite SnO2 phases. So, the big particles (above 100 nm) correspond to MoO3 whereas small particles, 2–3 nm, are SnO2 . At this point we would like to point out that it was surprising to find the ␤-MoO3 phase on our samples, as they were treated at 500 ◦ C, and this ␤-MoO3 phase has been reported to disappear above 400 ◦ C leading to an ␣-MoO3 [24]. The use of very small nanoparticles of SnO2 as supporting material clearly changes the behavior of the obtained MoO3 allowing the presence of the ␤MoO3 phase in materials treated at such high temperatures. On the other hand, HRTEM led us to determine the nanostructure of the sample’s nanoparticles and locate the different structures. As shown in Fig. 5, and in good agreement with SAED results, we observe that the SnO2 structure is predominant for low MoO3 samples. The presence of SnO2 nanoparticles is observed until the sample with a 71 mol% MoO3 maintaining their grain size almost constant in the range of 2–3 nm. The SnO2 cassiterite structure had a growing distortion with MoO3 content (from 0 to 10 mol% MoO3 ), being observed slight changes in cell parameters. However, above 10 mol% MoO3 SnO2 the structural distortion stabilizes and it is smoothly reduced as MoO3 content increases. This distortion for low MoO3 content proves the Fig. 6. HRTEM micrograph showing the nanoparticles in sample with 53 mol% MoO3 . Fig. 5. HRTEM micrograph showing the nanoparticles in sample with 5 mol% MoO3 . presence of Mo. As observed on HRTEM micrographs (Fig. 6), above 10 mol% MoO3 we reach the Mo limit of solubility inside SnO2 structure, starting the segregation of the excess Mo and the consequently formation of MoO3 aggregates and particles. Moreover, when observing the MoO3 distribution, we notice that there is a transition region (as observed by SAED) (15–20 mol% MoO3 ) where the ␤-MoO3 phase is present. Nevertheless this phase disappear for higher MoO3 contents being replaced by the ␣-MoO3 structure. The ␣-MoO3 forms big nanoparticles (with sizes over 100 nm) that are decorated with small SnO2 nanoparticles (around 3 nm) as shown in Fig. 6. To give a more complete view of this binary system, the O Kedge energy loss near edge spectroscopy (ELNES) evolution with MoO3 content has been also analyzed (Figs. 7 and 8). Taking into account that in several samples SnO2 and MoO3 structures coexist, we have performed the EELS analysis on the SnO2 -like nanoparticles (small particles), and on the MoO3 like particles (big particles), separately. In Fig. 7a, we present the evolution with MoO3 content of the O K-edge ELNES of the SnO2 nanoparticles in those samples were they were clearly present (from 0 to 64 mol% MoO3 ). While in Fig. 7b, we present the evolution with MoO3 content of the O K-edge ELNES of the MoO3 particles in those samples above 10 mol% MoO3 . The difference in ELNES O K-edge spectra between SnO2 and MoO3 -like particles is clearly shown in Fig. 7d, where one can 160 J. Arbiol et al. / Sensors and Actuators B 118 (2006) 156–162 Fig. 7. (a) O K-ELNES evolution of the SnO2 -like nanoparticles found in samples from 0 to 64 mol% MoO3 . (b) O K-ELNES evolution of the MoO3 -like particles found in samples from 15 to 100 mol% MoO3 . (c) S1 vs. S2 intensity ratio. (d) O K-ELNES of the SnO2 -like (small nanoparticles) and MoO3 -like particles (big particles, above 100 nm) found in sample SnMo53 (for interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article). Fig. 8. Raman spectra of the SnO2 :MoO3 binary system annealed at 500 ◦ C for samples with a MoO3 content below 15%. Sample with 37% MoO3 is included as a reference of the ␣-MnO3 dominant samples. observe that the oxygen fine structure of SnO2 and MoO3 is entirely different in both materials, particularly in the energy range 530–545 eV. In this region, SnO2 shows a fine structure consisting of two main peaks (labeled as S1 and S2, at 535 and 542 eV, respectively) whereas MoO3 in the same region shows one main peak (labeled as M1, at 535 eV) with a prominent shoulder (M2, placed at 538 eV). The ELNES O K-edge spectrum referred to SnO2 nanoparticles in the first few eV is dominated by transitions to O 2p or O 2p-Sn 5p hybridized empty states which reflects a tetrahedral arrangement in the first coordination shell of oxygen (SnO2 cassiterite structure) [16,17]. On the other hand, the spectrum referred to MoO3 particles shows the transitions to the O 2p orbitals hybridized with the Mo 4d orbitals [18]. In this latest case, these features (M1 and M2) reflect the octahedral environment of the molybdenum, as the crystal field splittings lead to separations of the low part of the conduction band into t2g (M1) and eg (M2) symmetry bands. In order to analyze the contribution of Mo atoms in the SnO2 structure, we have studied the evolution of S1 versus S2 intensity ratio, (Fig. 7c), for those spectra corresponding to SnO2 -like nanoparticles shown in Fig. 7a. Attending to results shown in Fig. 7c, we observe 3 different regions. The first region, marked in a red dashed circle, include the samples synthesized with low J. Arbiol et al. / Sensors and Actuators B 118 (2006) 156–162 MoO3 content (below 15 mol% MoO3 ). In this region, the S1/S2 intensity ratio smoothly increases with respect to the pure SnO2 ratio (SnMo0 sample) when increasing the MoO3 content. This increasing is due to the relative growth of the S1 peak (placed at 535 eV) with respect to the S2 peak (542 eV), and can be directly attributed to the increasing percentage of molybdenum oxide in the SnO2 structure. These molybdenum oxide would contribute to the growth of the M1 peak (also at 535 eV) which overlaps with the S1 peak and would increase its relative intensity. This result is in good agreement with the strong and sudden decrease of the SnO2 response toward alcohols with the addition of MoO3 (Fig. 1), as the presence of molybdenum oxide at the nano-SnO2 structure, 2–3 nm, changes its acidity performances which varies its reactivity versus certain gas species. The small size of the SnO2 nanoparticles increase the surface versus volume atomic ratio, which is important to obtain better sensitivity on the gas sensor material as the active area is increased. The second main region observed in Fig. 7c is circled in a green dashed line and correspond to the samples with high MoO3 content. This second region presents a slight decrease of the S1/S2 ratio for higher MoO3 percentages. This small decrease may suggest that molybdenum oxides reaches the SnO2 solubility limit, as pointed on our SAED and HRTEM previous results, and SnO2 nanoparticles found at these high MoO3 percentages do not increase the presence of Mo in the SnO2 structure. The third region is composed by those samples placed in the transition range (from 15 to 20 mol% MoO3 ) (points are marked as red circles in Fig. 7c). In this region we observed from SAED the presence of ␤-MoO3 . As this stage is a transition region, the ELNES spectra focusing on several nanoparticles is revealing significant variations due to the transition effects. However, after these results we are able to correlate the observed saturation effect in the 20 mol% region observed in Figs. 1 and 3 with the presence of ␣-MoO3 in our samples. To complete this analysis, especially concerning the presence and origin of the ␤-MoO3, we have also performed Raman analysis, which is very sensitive to MoO3 vibration modes. Fig. 8 shows the presence of modes attributed to the ␤-MoO3 phase up to the sample with 15 mol% MoO3 , where they are overlapped by modes attributed to ␣-MoO3 modes. These spectra point out the existence of ultradispersed ␤-MoO3 at the SnO2 small grains which are coherent with the HRTEM, EELS and sensing results and, besides, these results due to the higher sensitivity and resolution than they obtained from SAED, allow to corroborates that for low MoO3 contents the vibration modes fit whit those of the ␤-MoO3 (monoclinic), pointing out that the MoO3 phase is different before starting the formation of segregated ␣-MoO3 particles (orthorhombic), when MoO3 appears to be disperse in the SnO2 structure. It is also worthy to observe that this presence is confirmed by the significant intensity reduction of the SnO2 634 cm−1 peak. So, TEM and SAED did not show, likely due to the experimental resolution, a MoO3 structure for low MoO3 content samples, since SnO2 cassiterite phase is dominant, Nevertheless, a slight distortion in atomic planes distances was observed and related with the presence of Mo. Moreover, as EELS confirms the presence of Mo–O bonds in the SnO2 structure and 161 Raman spectroscopy shows also the presence of a ␤-MoO3 phase at low MoO3 content, it is clear that the MoO3 should mainly be dispersed at the SnO2 nanoparticles, modifying its surface acidic character and, hence, the sensing mechanisms. On the other hand, when the MoO3 segregation from SnO2 grains takes place from MoO3 contents higher than 15%, the ␣-MoO3 phase appears, showing the high trend of MoO3 to be segregated and to easily aggregate forming big nanoparticles. No more surface effects appears and it explains the saturated regions of the sensing characteristics for MoO3 contents above 20%. 4. Conclusions Below the 15 mol% MoO3 , MoO3 remains dispersed as a ␤-MoO3 phase in SnO2 structure. Any evidences of the independent MoO3 cluster formation or MoO3 outer layer covering the SnO2 nanograin have been found for these low MoO3 contents. It is outstanding to point out that the molybdenum oxide is found in a monoclinic phase instead orthorhombic that is expected after the applied thermal treatment at 500 ◦ C. Likely, local stress field related to the nanosize of the supporting material has straightforward effect on the crystallographic structure of this material. The presence of this MoO3 induces effects on the SnO2 surface and/or structure that led also to a change related with the MoO3 content of the sensing characteristics against alcohol and ammonia. This sensing behaviour is well correlated with the acidity evolution of the surface of the SnO2 caused by the presence of MoO3 . Above the 20 mol% MoO3 , MoO3 appears as ␣-MoO3 phase (forming big particles). This structural feature explains the saturation on the sensor response. The structural characterization of the nano-SnO2 /MoO3 binary system justifies the sensing characteristics of this system and the combination of HRTEM, SAED, ELNES and Raman has been shown as a powerful tool for the analysis of these nano-binary metal oxide systems. Acknowledgements This work has been partially supported by CERMAE. We also kindly thank the Nanos 4 EU project. References [1] N.G. Valente, L.E. Cadus, O.F. Gorriz, L.A. Arrua, J.B. Rivarola, Synergy in the Sn–Mo–O catalysts: the selective oxidation of methanol, Appl. Catal. A 153 (1–2) (1997) 119–132. [2] A. Cabot, J. Arbiol, J.R. Morante, U. Weimar, N. Bârsan, W. Göpel, Sens. Actuators B 70 (2000) 87–100. [3] R. Dı́az, J. Arbiol, A. Cirera, F. Sanz, F. Peiró, A. Cornet, J.R. Morante, Chem. Mater. 13 (11) (2001) 4362–4366. [4] A. Ruiz, J. Arbiol, A. Cirera, A. Cornet, J.R. Morante, Mat. Sci. & Eng. C 19 (2002) 105–109. [5] J. Arbiol, A. Cirera, F. Peiró, A. Cornet, J.R. Morante, J.J. Delgado, J.J. Calvino, Appl. Phys. Letts. 80 (2) (2002) 329–331. [6] J. Arbiol, F. Peiró, A. Cornet, J.R. Morante, J.A. Pérez-Omil, J.J. Calvino, Mat. Sci. & Eng. B 91–92 (2002) 534–536. [7] J. Arbiol, J. Cerdà, G. Dezanneau, A. Cirera, F. Peiró, A. Cornet, J.R. Morante, J. Appl. Phys. 92 (2) (2002) 853–861. [8] R. Dı́az, J. Arbiol, F. Sanz, A. Cornet, J.R. Morante, Chem. Mater. 14 (8) (2002) 3277–3283. 162 J. Arbiol et al. / Sensors and Actuators B 118 (2006) 156–162 [9] A. Ruiz, G. Dezanneau, J. Arbiol, A. Cornet, J.R. Morante, Thin Solid Films 436 (1) (2003) 90–94. [10] A.M. Ruiz, G. Dezanneau, J. Arbiol, A. Cornet, J.R. Morante, Chem. Mater. 16 (5) (2004) 862–871. [11] A.M. Gaskov, M.N. Rumyantseva, Materials for solid-state gas sensors, Inorg. Mater. 36 (3) (2000) 293–301. [12] M. Ivanovskaya, P. Bogdanov, G. Faglia, P. Nelli, G. Sberveglieri, A. Taroni, On the role of catalytic additives in gas-sensitivity of SnO2 –Mo based thin film sensors, Sens. Actuators, B: Chem. 77 (2001) 268–274. [13] M. Ivanovskaya, E. Lutynskaya, P. Bogdanov, The influence of molybdenum on the properties of SnO2 ceramic sensors, Sens. Actuators, B: Chem. 48 (1998) 387–391. [14] A. Chiorino, G. Ghiotti, F. Prinetto, M.C. Carotta, D. Gnani, G. Martinelli, Preparation and characterization of SnO2 and MoOx –SnO2 nanosized powders for thick film gas sensors, Sens. Actuators, B: Chem. 58 (1999) 338–349. [15] A. Chiorino, G. Ghiotti, F. Prinetto, M.C. Carotta, M. Gallana, G. Martinelli, Characterization of materials for gas sensors: surface chemistry of SnO2 and MoOx –SnO2 nano-sized powders and electrical responses of the related thick films, Sens. Actuators, B: Chem. 59 (1999) 203–209. [16] M.S. Moreno, R.F. Egerton, P.A. Midgley, Differentiation of tin oxides using electron energy-loss spectroscopy, Phys. Rev. B 69 (2004) 233304/1–1233304/4. [17] M.S. Moreno, R.F. Egerton, J.J. Rehr, P.A. Midgley, Electronic structure of tin oxides by electron energy loss spectroscopy and real-space multiple scattering calculations, Phys. Rev. B 71 (2005) 035103/1–035103/6. [18] D. Wang, D.S. Su, R. Schlögl, Electron beam induced transformation of MoO3 to MoO2 and a new phase MoO, Z. Anorg. Allg. Chem. 630 (2004) 1007–1014. [19] E.A. Makeeva, M.N. Rumyantseva, A.M. Gaskov, Synthesis, microstructure, and gas-sensing properties of SnO2 /MoO3 nanocomposites, Inorg. Mater. 41 (4) (2005) 442–449. [20] T. Maekawa, J. Tamaki, N. Miura, N. Yamazoe, Development of SnO2 based ethanol gas sensor, Sens. Actuators, B: Chem. 9 (1992) 63. [21] SnO2 Cassiterite: Tetragonal S.G.: P42/mnm (JCPDS 77-0449). H. Seki, N. Ishizawa, N. Mizutani, M. Kato, High temperature structures of the rutiletype oxides, titanium dioxide and stannic oxide, J. Ceram. Assoc. Jpn. 92 (1984) 219–223. [22] ␣-MoO3 Ortho.: S.G.: Pbnm (JCPDS 05-0508). G. Andersson, A. Magneli, Crystal structure of molybdenum trioxide, J. Appl. Crystallogr. 21 (1950) 960–964. [23] ␤-MoO3 Monoc.: S.G.: P21/n (JCPDS 84-1360). J.B. Parise, E.M. McCarron, R. Von Dreele, J.A. Goldstone, ␤-Molybdenum trioxide produced from a novel freeze drying route, J. Solid-State Chem. 93 (1991) 193–201. [24] F. Harb, B. Gerand, G. Nowogrocki, M. Figlarz, Structural filiation between a new molybdenum oxide hydrate (MoO3 ·(1/3)H2 O) and a new monoclinic form of MoO3 obtained by dehydration, Solid State Ionics 32-33 (I) (1989) 84–90. Biographies Jordi Arbiol graduated in physics at the University of Barcelona in 1997, received his European PhD in Physics in 2001, and obtained the PhD Extraordinary Award of the Electronics Department. He joined the Electronics Department in 1997, and in 2000 he was appointed as the Assistant Professor in this department. His current research activities are centered in the structural, compositional and morphological characterization of nanosized materials and devices by means of TEM related techniques (HRTEM, EELS, EFTEM, Z-contrast, electron tomography, etc.). Joan Ramon Morante in 1980, received the PhD degree in physics from the University of Barcelona. Since 1986, he is a full professor of Electronics and director of the Electronic Materials and Engineering group, EME. He has been dean of the Physics Faculty, academic advisor of the Electronic Engineering degree and director of the Electronics Department. His activity is devoted to the electronic materials and technology, physics and chemical sensors, actuators, and Microsystems. He is co-author of more than 400 papers in international specialized journal and member of international committees and editorial boards in the field of electronic materials and technology, sensors and actuators and microsystems, and electronic systems. Pierre Bouvier works in the Raman spectroscopy group in the Laboratory of Electrochemistry and Physiochemistry of Materials and Interfaces. He is doctor INPG (2000). Works on Raman spectroscopy and imaging, and is interested in phase transitions under extreme pressure conditions and in oxide nanostructures. Thierry Pagnier leads the Raman spectroscopy group in the Laboratory of Electrochemistry and Physiochemistry of Materials and Interfaces. He is doctor in electrochemistry (1985). After 5 years at Telemecanique Electrique as head of the metallurgy research laboratory, he went back to academic research in 1990, first in solid oxide fuel cells field and from 1995 in Raman spectroscopy. Ekatherina Makeeva graduated in Chemistry at the Moscow State University in 2004 and is currently a PhD student in this University. Her work is focused on synthesis and characterization of new nanomaterials for gas sensors as well as on study of kinetics of gas-solid interaction sensing mechanism using in situ conductance measurements. Marina Rumyantseva graduated in chemistry at the Moscow State University in 1992. She received her PhD in 1996 from Moscow State University and from National Polytechnic Institute of Grenoble. From 1997 until now, she has been employed in the Chemistry Department of Moscow State University, where she has been involved in the research on synthesis and characterization of semiconductor materials for gas sensors. Alexandre Gaskov leads the Laboratory of Diagnostics of Inorganic Materials of Chemistry Department of Moscow State University since 1995. He is Doctor of Science in Inorganic Chemistry (1988) and Professor of Chemistry (1993). He has more 30 years experience of research in semiconductor material science, including crystal and thin films growth, heterostructure synthesis, surface analysis, development of semiconductor materials for IR detectors and gas sensors.