ethanol and ozone sensing characteristics of wo3 based sensors activated by au and pd

Sensors and Actuators B 120 (2006) 338–345 Review Ethanol and ozone sensing characteristics of WO 3 based sensors activated by Au and Pd A. Labidi a,b , E. Gillet a , R. Delamare a , M. Maaref b , K. Aguir a,∗ a L2MP (CNRS UMR 6137), Service 152, FST St J´erˆome, Universit´e Paul CEZANNE Aix-Marseille III, 13397 Marseille Cedex 20, France b Unit´e de Recherche de Physique des Semiconducteurs et Capteurs, IPEST, BP 51 La Marsa 2070, Tunis, Tunisia Received 29 November 2005; received in revised form 7 February 2006; accepted 7 February 2006 Available online 29 March 2006 Abstract The sensitivity towards ethanol (C 2 H 6 O) and ozone (O 3 )ofWO 3 thin films based conductometric sensors was investigated. The performances of three sensing layers were compared: bare WO 3 , palladium (Pd) and gold (Au) activated surface WO 3 . The WO 3 thin films were deposited by thermal evaporation of oxide powders onto SiO 2 /Si transducers with platinum interdigited electrodes. All the tests were performed at the same working temperature T work = 300 ◦ C and under fixed gas concentrations: 2% ethanol and 0.8 ppm ozone using dry air as carrier gas. The morphology of the sensor surfaces were analyzed before and after working runs by atomic force microscopy (AFM) and scanning electron microscopy (SEM) in order to control the stability of the metal deposits. DC and AC electrical responses under 50 l h −1 gas flows are presented and discussed. © 2006 Elsevier B.V. All rights reserved. Keywords: DC/AC measurements; WO 3 sensor; Pd, Au activators; Ozone; Ethanol Contents 1. Introduction 338 2. Experimental 339 2.1. Sensing device preparation 339 2.2. Sensing tests procedure 340 3. Results and discussion 340 3.1. DC measurements 340 3.2. AC measurements 341 4. Conclusion 343 References 344 Biographies 344 1. Introduction Numerous metal oxide semiconductor materials were reported to be usable in conductometric gas sensors, such as ZnO, SnO 2 ,WO 3 ,TiO 2 , ␣-Fe 2 O 3 and so on. These candidates have non-stoichiometric structures, so free electrons originating ∗ Corresponding author. Tel.: +33 4 91 28 89 73; fax: +33 4 91 28 89 70. E-mail address: Khalifa.aguir@l2mp.fr (K. Aguir). from oxygen vacancies contribute to electronic conductivity when the composition of the surrounding atmosphere is altered [1–6]. Actually WO 3 is one of promising material for gas sensor application, and during the last years many works have been performed on the structural, electrical properties and sensing characteristics of WO 3 films [7–10]. It was demonstrated by different authors that WO 3 -based thin and thick films were both sensitive to a broad range of oxidizing or reducing gases such 0925-4005/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2006.02.015 A. Labidi et al. / Sensors and Actuators B 120 (2006) 338–345 339 as H 2 S [11],NO x [12,13],O 3 [14,15] and C 2 H 6 O [16,17].It is well known that WO 3 like the majority of metal oxide-based gas sensors suffer from drift and lack of selectivity, and in order to solve these problems, a number of different strategies were applied such as the sensor’s dynamic response analysis as well as mixed oxides sensors for gas detection [18]. In addition to the previous cited strategies, the modification of the metal oxide active layer by adding small amount of noble metals has been recently emerged as a promising way for the improvement of sensors selectivity. The metals that have been used frequently as surface dopants for these purposes are Au, Pt, Pd and Ag [19–23]. The main objective of our work was the understanding of the mechanisms which explain the activation of WO 3 thin film sensors by noble metals. For that, two metals have been chosen Pd and Au, and their effect on the sensor sensitivity towards ozone and ethanol was studied. The DC transient and AC responses were analyzed in order to discriminate the sensor parts which are crucial for the sensing mechanisms (grains–grain boundaries–metal/oxide interfaces ). In order to avoid bulk parameters variation the tests were carried out at the same tem- perature T work = 300 ◦ C. So the conductance changes that we measured could be attributed to surface or near surface phe- nomena. 2. Experimental 2.1. Sensing device preparation The WO 3 sensitive layer has been grown by vapor deposi- tion at room temperature on SiO 2 /Si transducer with platinum interdigitated electrodes. This structure underwent a subsequent annealing at 450 ◦ C for 1 hour in dry air. In previous stud- ies we have shown that such a fabrication process allows to obtain continuous, well crystallized films with a stoichiometry around WO 2.8 and an excellent stability under working con- ditions [24,25]. The thicknesses of the WO 3 layer and of the electrodes were 40 nm and 50 ␮m, respectively, the area of the Fig. 1. AFM (3 ␮m ×3 ␮m) micrograph of the bare WO 3 surface annealed in dry air (1 h at 450 ◦ C). active part of the sensor being 4 mm ×4 mm. On the top of the WO 3 , 0.5 nm (mean equivalent mass thickness) of the activator metal (Au or Pd) was vapor deposited in UHV at T dep = 350 ◦ C. The morphology of each sensor was controlled by AFM (Nanoscope III-Digital Instruments) and SEM (Phillips XL30 S5). Fig. 1 is an image of the bare WO 3 layer, which is formed from small grains with a mean diameter of 40 nm. The mean roughness calculated in a 1.5 ␮m ×1.5 ␮m area is 0.598 nm, it remains the same after some hours of working time. It was dif- ficult to distinguish by AFM the metals particles from the oxide grains, so we have analyzed the activated layers by SEM. Fig. 2a and b presents micrographs of Au/WO 3 and Pd/WO 3 surfaces, respectively. The mean diameter of gold particles is 5 nm, the population is homogeneous with a 6 ×10 11 cm −2 number in density. The size of palladium particles is larger (9 nm) with a4×10 11 cm −2 number in density. The three samples were observed after the sensing tests, no changes were visible in the metal layers as long as the working temperature (T work ) did not exceed the deposition temperature (T dep ). Fig. 2. SEM images of the sensors surfaces after deposition of metal layer (5 min at 350 ◦ C). (a) Au/WO 3 and (b) Pd/WO 3 . 340 A. Labidi et al. / Sensors and Actuators B 120 (2006) 338–345 Fig. 3. Experimental set-up used for C 2 H 6 O and O 3 sensing tests. 2.2. Sensing tests procedure Each sensor was tested separately using the same protocol in a small chamber where the gases are injected at a flow rate of 50 l h −1 . Dry air was used both as a reference (baseline) and as carrier gas to obtain the desired concentration of the detected gas. Fig. 3 represents schematically the experimental set-up. The dilution of C 2 H 6 O vapor in dry air was achieved using a two-arm gas-flow device. Two mass flow controllers allowed the flow rate of the dry air that act as the carrier gas to be controlled from 0 to 94lh −1 in one arm (d 1 ) in which the carrier gas passed through a balloon flask containing the vapor equilibrated with 200 cm 3 of the liquid, and from 0 to 94 l h −1 in the other arm (d 2 ). The balloon flask is put into a furnace and maintained at the fixed temperature T vap =30 ◦ C, in order to fix the partial pressure of the C 2 H 6 O vapor. In these conditions, a range of concentrations of the C 2 H 6 O in air can be calculated by applying the following equation [26]: [C] (%) = xd 1 xd 1 + d 1 + d 2 × 100 (1) where x is the molar fraction of the vapor in the balloon flask at T vap , given by: x = P vap P atm (2) with P vap the partial pressure of the vapor at a given temperature T vap , and P atm the atmospheric pressure. By varying d 1 and d 2 (d 1 + d 2 was kept constant at 50 l h −1 ), different concentration values for C 2 H 6 O in dry air can be obtained. For the test under O 3 the two mass flow controllers used for C 2 H 6 O vapor will be turned off. Once this condition is satisfied, the O 3 gas was generated by oxidizing oxygen molecules of a dry air controlled from 0 to 50 l h −1 in arm d 3 and fixed also at 50lh −1 , this flow was exposed to a pen-ray UV lamp, calibrated to give an O 3 concentration range between 0.03 and 0.8 ppm. The total flow charged by ethanol vapor or ozone was blowing on the sensor placed in the test chamber on a heating holder. The working temperature (T work ) of the sensor was controlled by a regulated power supply connected to the heating platform. The basic principles of the conductometric sensors is the varia- tion of free carriers density of the active layer exposed to a gas concentration which can be correlated to a change in conduc- tance G of the oxide. G was measured by recording the current variation at the applied constant DC potential V = 50 mV, with an HP4140B Source/Pico-ammeter. In AC regime data were acquired using a Solartron 1250 frequency response analyser in the 0.2 Hz–65 kHz frequency range. Measurements were per- formed at the working temperature T work = 300 ◦ C, for 2% of C 2 H 6 O and 0.8 ppm of O 3 . 3. Results and discussion 3.1. DC measurements In Table 1 are reported the conductances G 0 in dry air and G gas under the target gases, for the sensors WO 3 bare, Au/WO 3 and Pd/WO 3 at T work = 300 ◦ C, and resulting sensing response “S gas ” of the sensors calculated by using the relations: S gas = G gas − G 0 G 0 for reducing gases or S gas = G 0 − G gas G gas for oxidizing gases (3) The conductance in air the Au/WO 3 based sensor is two orders larger than that for the bare WO 3 , when it is smaller for Pd/WO 3 . The relative values of G 0 being as Pd/WO 3 < Au/WO 3 with G 0 (WO 3 )=5G 0 (Pd) and G 0 (Au) = 100G 0 (Pd). This indicates that the activation mode of Au and Pd are different. The transient responses of the three sensors towards pulses of 2% C 2 H 6 O and 0.8 ppm O 3 at 300 ◦ C are compared in Fig. 4a and b, respectively. The exposure time was kept constant at 15 min for each test and the time between successive pulses was also 15 min. As expected under C 2 H 6 O (reducing gas) the conduc- tance increases and it is clear that the Au/WO 3 sensor gives A. Labidi et al. / Sensors and Actuators B 120 (2006) 338–345 341 Table 1 Response “S gas ”ofWO 3 bare, Au/WO 3 and Pd/WO 3 based sensors towards 2% of C 2 H 6 O and 0.8 ppm O 3 atmospheres at 300 ◦ C Sensor G 0 (×10 −8  −1 ) G Ethanol (×10 −6  −1 ) G Ozone (×10 −9  −1 ) S Ethanol S Ozone WO 3 bare 6.72 4.3 5.5 36 7.6 Au/WO 3 104 180 6.5 177 78 Pd/WO 3 1.5 0.025 11 0.76 0.15 S was calculated by using relations (3). the largest response magnitude when bare WO 3 has the smaller response and recovery times. The kinetics of the responses to O 3 on bare WO 3 is normal for an n-type semiconductor exposed to an oxidizing gas, when on activated sensors the responses are complex (Fig. 4b): on Pd/WO 3 , there is a rapid decrease in conductance for O 3 injec- tion thereafter a slow conductance increase for the duration of the pulse and a rapid increase followed by a slow decrease during the recovery period; on Au/WO 3 , a rapid decrease in conduc- tance at the injection is followed by a slow decrease during the pulse, and a fast increase followed by a decrease characterizes the recovery period. Such a kinetics results of the superposition of two phenomena, one occurring on the bare oxide surface the other on the metallic clusters. Further experiments at different temperatures and concentrations are actually in progress in order Fig. 4. Comparison of typical transient DC responses to gas pulse at 300 ◦ C. (a) 2% of C 2 H 6 O and (b) 0.8 ppm of O 3 . to clarify these mechanisms which evidently are related to the origin of conductivity changes on activated sensors. 3.2. AC measurements The impedance sensors response versus frequency was stud- ied in the AC impedance spectrum following the analysis proce- dure that was established in our previous study [27]. The WO 3 modified sensitive layers were modeled by a serial association of three parallel RC circuits, attributed to grains (b), grains bound- aries (gb) and grains–electrodes interfaces (el). Each RC circuit rises to a semicircle in the complex plan plot of Z  (ω) versus Z  (ω) (Nyquist diagram). Under small amplitudes of sinusoidal signal, the total impedance of sensor is given by: Z Total (jω) =  i Z i (jω) (4) Z i (jω) = Z  i (ω) + jZ  i (ω) (5) where i = b, gb and el; j is the complex number j = √ −1; Z  i (ω), Z  i (ω) are the real and imaginary parts of impedance Z i (jω), respectively. In Fig. 5a and b, we report the Nyquist response of the three sensors with modeling at T work = 300 ◦ C, under 2% of C 2 H 6 O vapor as well as under 0.8 ppm of O 3 , wherein the modeled curves were made using “Equivalent Circuit” software [28]. The results of modeling for C 2 H 6 O and O 3 were reported in Tables 2 and 3, respectively; they confirm the DC analyses, the best response was obtained by the sensor doped by gold (Au/WO 3 ) which became practically conductor under ethanol. The RC modeling show the existence of two semicircles under dry air, contrary to the sensor without metals (WO 3 bare) that gives only one semicircle either under dry air or C 2 H 6 O. The first semicircle is attributed to WO 3 surface and bulk phe- nomena, the second one could be attributed probably to the carrier exchanges in the Au/WO 3 grains boundaries, because under dry air this circle appears only when Au is added to the WO 3 surface. When ethanol is introduced the second semicircle disappears, this is could be explained by the fact that the pre- adsorbed oxygen on the WO 3 and/or Au surfaces and in their grain boundaries, is consumed by the C 2 H 6 O oxidation follow- ing the reaction paths (6) and (7) [29]: C 2 H 5 OH (vap) + O − (ads) ↔ CH 3 CHO (ads) + H 2 O (vap) + e − (6) CH 3 CHO (ads) +O (lattice) ↔ CH 3 COOH (vap) +V O (7) The electrons produced by this reaction are injected into the con- duction band of WO 3 , which induces a decrease of the resistance. The response to the C 2 H 6 O is improved for sensor (Au/WO 3 ), 342 A. Labidi et al. / Sensors and Actuators B 120 (2006) 338–345 Fig. 5. Impedance measurements (symbols) and modeling (lines) at 300 ◦ C. (a) 2% of C 2 H 6 O and (b) 0.8 ppm of O 3 . A. Labidi et al. / Sensors and Actuators B 120 (2006) 338–345 343 Table 2 Modeled RC for WO 3 bare, Au/WO 3 and Pd/WO 3 based sensors towards 2% of C 2 H 6 O and dry air (baseline) at 300 ◦ C 2% of Ethanol (C 2 H 6 O) WO 3 a WO 3 +Au a WO 3 +Pd a Dry air b Ethanol b Dry air b Ethanol b Dry air b Ethanol b R s-b (×10 6 ) 15.77 0.2 1.46 0.005 59.17 37.23 C b (×10 −10 F) 1.65 1.87 1.72 1.22 1.65 1.78 n 0.993 0.986 0.991 0.971 0.986 0.975 R gb (×10 6 ) 0.018 2.37 C gb (×10 −7 F) 1.81 6.81 n 0.967 0.433 a Sensors. b Gases. Table 3 Modeled RC for WO 3 bare, Au/WO 3 and Pd/WO 3 based sensors towards 0.8 ppm of O 3 and dry air (baseline) at 300 ◦ C 0.8 ppm of Ozone (O 3 ) WO 3 a WO 3 +Au a WO 3 +Pd a Dry air b O 3 b Dry air b O 3 b Dry air b O 3 b R s-b (×10 6 ) 20.62 157.56 2.1 259.14 82.12 82.94 C b (×10 −10 F) 1.61 1.65 1.7 1.64 1.65 1.6 n 0.997 0.993 0.992 0.990 0.986 0.992 R gb (×10 6 ) 0.028 53.42 C gb (×10 −7 F) 1.98 0.1 n 0.992 0.726 a Sensors. b Gases. an opposite result was obtained with the addition of Pd to the WO 3 surface, i.e. increase of resistance and decrease of sensi- tivity. This could be caused by the oxidation of palladium on the surface or by the formation of a bimetallic component as it was found on SnO 2 and CeO 2 [30–32]. Under O 3 the sensor Au/WO 3 gives also the best response with a more pronounced grain boundaries effect. The transient responses of Fig. 4b, suggested that one of the two steps of the reaction (8) resulting of the O 3 dissociation is enhanced by the presence of the metal particles [33–35]. p 2 O 2 +∗↔O p(ads) adsorption step O p(ads) + qe ↔ O q− p(ads) electron transfer step (8) The AC analysis evidenced that it is the electron transfer step which is improved, in effect when O 3 is introduced the RC mod- eling shows a decrease of the grains boundaries capacitance due to the increases of the depletion zone and consequently increases the resistance and the electrons capture by oxygen through the interface Au/WO 3 . In this case the conductance is mainly con- trolled by the grains boundaries phenomena, i.e. by the electron transfer step in these regions, which is confirmed by the appari- tion of a second deformed semicircle. Fig. 5b shows that the model with two circles, whose results are reproduced in Table 2, represents well the experimental results at the high frequencies (first semicircle). The more important variation observed at the low frequencies between the experimental model and results can be due to the existence of athird semicircle, due to theapparitions of some diffusions phenomena in the interface grains/electrode (Pt). Unfortunately our work frequency range, does not allow us to modeling this third semicircle, that is why we have a small deviation between AC measurements and modeling in the low frequency range, as illustrated in Fig. 5b, for the sensor doped with Au (Au/WO 3 ). The other important remark is the non-sensitivity of the Pd- doped WO 3 gas sensor, which became practically insensitive to the O 3 gas. The origin of such a behavior should be the adverse effect of the PdO formation on the free carrier density. 4. Conclusion Au has been found a good sensing activator for WO 3 thin films. The sensitivities of Au/WO 3 sensors to ethanol and ozone are in the 2/1 ratio; therefore, at a working point of 300 ◦ C they can provide a stable, sensitive element for ethanol gas. On the contrary Pd/WO 3 sensors are practically insensitive in this tem- perature range to the tested gases and in these senses could be used as selective elements against ozone. The characteristics of the dynamic responses of the activated WO 3 thin films sug- gest complex phenomena which depend on the strength of the metal–substrate interaction and consequently could be induced by the formation of oxide or bimetallic species on the metal par- ticles. In the actual knowledge state in the behavior of doped 344 A. Labidi et al. / Sensors and Actuators B 120 (2006) 338–345 oxide layers one needs an understanding of activation processes at an atomic level if one want to progress in the design of predictable sensing properties. In this way we have initiated a research by XPS and synchrotron radiation photoelectron spec- troscopy (SRPES) in order to elucidate the electronic structure of the atomic species under various working conditions. References [1] W. G ¨ opel, Chemisorption and charge transfer at ionic semiconductor sur- faces: implication in designing gas sensors, Prog. Surf. Sci. 20 (1) (1985). [2] M.J. Madou, S.R. Morrison, Chemical Sensing with Solid state Devices, Academic, San Diego, 1989. [3] K.D. Schierbaum, U. Weimar, W. G ¨ opal, R. Kowalkowski, Conductance, work function and catalytic activity of SnO 2 -based gas sensors, Sens. Actu- ators B 3 (1991) 205–214. [4] K.D. Schierbaum, Engineering of oxide surfaces and metal/oxide inter- faces for chemical sensors: recent trends, Sens. Actuators B 24/25 (1995) 239–247. [5] Y.K. Chung, M.H. Kim, W.S. Um, H.S. Lee, J.K. Song, S.C. Choi, K.M. Yi, M.J. Lee, K.W. Chung, Gas sensing properties of WO 3 thick film for NO 2 gas dependent on process condition, Sens. Actuators B 60 (1999) 49–56. [6] P. Moseley, B. Tofield, Solid State Gas Sensors, Adam Hilger, Bristol, UK, 1987. [7] M. Regragui, V. Jousseaume, M. Addou, A. Outzourhit, J.C. Bern ´ ede, B. El Idrissi, Electrical and optical properties of WO 3 thin films, Thin Solid Films 397 (2001) 238–243. [8] W. Yu-De, C. Zhan-Xian, L. Yan-Feng, Z. Zhen-Lai, W. Xing-Hui, Electri- cal and gas-sensing properties of WO 3 semiconductor material, Solid-State Electron. 45 (2001) 639–644. [9] M. Stankova, X. vilanova, E. Llobet, J. Calderer, C. Bittencourt, J.J. Pireaux, X. Correig, Influence of the annealing and operating tempera- ture on the gas-sensing properties of rf sputtered WO 3 thin-film sensors, Sens. Actuators B 105 (2005) 271–277. [10] M. Gillet, K. Aguir, C. Lemire, E. Gillet, K. Schierbaum, The structure and electrical conductivity of vacuum-annealed WO 3 thin films, Thin Solid Films 467 (2004) 239–246. [11] D.J. Dwyer, Surface chemistry of gas sensors: H 2 SonWO 3 films, Sens. Actuators B 5 (1991) 155–159. [12] G. Sberveglieri, L.E. Depero, S. Groppelli, P. Nelli, WO 3 sputtered thin films for NO x monitoring, Sens. Actuators B 84 (2002) 1–5. [13] D.S. Lee, S.D. Han, J.S. Huh, D.D. Lee, Nitrogen oxides sensing charac- teristics of O 3 -based nanocrystalline thick film gas sensor, Sens. Actuators B 60 (1999) 57–63. [14] K. Aguir, C. Lemire, D.B.B. Lollman, Electrical properties of reactively sputtered WO 3 thin films as ozone gas sensor, Sens. Actuators B 84 (2002) 1–5. [15] M. Bendahan, R. Boulmani, J.L. Seguin, K. Aguir, Characterization of ozone sensors based on WO 3 reactively sputtered films: influence of O 2 concentration in the sputtering gas, and working temperature, Sens. Actu- ators B 100 (2004) 320–324. [16] R. Ionescu, A. Hoel, C.G. Granqvist, E. Llobet, P. Heszler, Ethanol and H 2 S gas detection in air and in reducing and oxidising ambience: application of pattern recognition to analyse the out p from temperature-modulated nanoparticulate WO 3 gas sensors, Sens. Actuators B 104 (2005) 124–131. [17] B.P.J. de Lacy Costello, R. Ewen, H. Gunson, N. Ratcliffe, P. Spencer, The development of sensor system for the early detection of soft rot in stored potato tubers, Meas. Sci. Technol. 11 (2000) 1685–1691. [18] K. Zakrzewska, Mixed oxides as gas sensors, Thin Solid Films 391 (2001) 229–238. [19] G. Sberveglieri, Recent developments in semiconducting thin-film gas sen- sors, Sens. Actuators B 23 (1995) 103–109. [20] P. Ivanov, E. Llobet, X. Vilanova, J. Brezmes, J. Hubalek, X. Correig, Development of high sensitivity ethanol gas sensors based on Pt-doped SnO 2 surfaces, Sens. Actuators B 99 (2004) 201–206. [21] M. Schweizer-Berberich, J.G. Zheng, U. Weimar, W. G ¨ opel, N. B ˆ arsan, E. Pentia, A. Tomescu, The effect of Pt and Pd surface doping on the response of nanocrystalline tin dioxide gas sensors to CO, Sens. Actuators B 31 (1996) 71–75. [22] I. Stambolova, K. Konstantinov, S. Vassilev, P. Peshev, Ts. Tsacheva, Lan- thanum doped SnO 2 and ZnO thin films sensitive to ethanol and humidity, Mater. Chem. Phys. 63 (2000) 104–108. [23] P.J. Shaver, Activated tungsten oxide gas detectors, Appl. Phys. Lett. 11 (1967) 7–255. [24] C. Lemire, Elaboration et caracterisation de films minces de WO 3 en vue de leur application comme capteur de gaz, Thesis-Universit ´ e Aix-Marseille III, 2001. [25] C. Bittencourt, R. Landers, E. Llobet, G. Molas, X. Crreig, M.A.P. Silva, J.E. Sueiras, J. Calderer, Effect of oxygen partial pressure and annealing temperature on the formation of sputtered tungsten oxide films, J. Elec- trochem. Soc. 149 (3) (2002) H81–H86. [26] A. Abdelghani, N. Jaffrezic-Renault, SPR fibre sensor sensitized by fluo- rosiloxane polymers, Sens. Actuators B 74 (2001) 117–123. [27] A. Labidi, C. Jacolin, M. Bendahan, A. Abdelghani, J. G ´ erin, K. Aguir, M. Maaref, Impedance spectroscopy on WO 3 gas sensor, Sens. Actuators B 106 (2005) 713–718. [28] B.A. Boukamp, Equivalent Circuit (EQUIVCRT.PAS), University of Twente, The Netherlands, 1990. [29] F. Hellegouarc’h, F. Arefi-Khonsari, R. Planade, J. Amouroux, PECVD prepared SnO 2 thin films for ethanol sensors, Sens. Actuators B 73 (2001) 27–34. [30] N. Tsud, V. Joh ´ anek, I. Star ´ a, K. Veltrusk ´ a, V. Matolin, XPS, ISS and TPD study of Pd·Sn interactions on Pd–SnO x systems, Thin Solid Films 391 (2001) 204–208. [31] L. Kepi ´ n, Carbon deposition in Pd/CeO 2 catalyst: TEM study, Catal. Today 50 (1999) 237–245. [32] T. Sk ´ ala, K. Veltrusk ´ a, M. Moroseac, I. Matolinov ´ a, A. Cirera, V. Matolin, Redox process of Pd–SnO 2 system studied by XPS, in: Proceeding of 22nd European Conference on Surface Science, Prague, Czech Republic, Poster N ◦ . 16931, 2003. [33] W. Li, S.T. Oyama, Mechanism of ozone decomposition on a maganese oxide catalyst (1), J. Am. Chem. Soc. 120 (1998) 9041–9046. [34] W. Li, G.V. Gibbs, S.T. Oyama, Mechanism of ozone decomposition on a maganese oxide catalyst (2), J. Am. Chem. Soc. 120 (1998) 9047–9052. [35] A. Gurlo, N. Bars ˆ an, M. Ivanovskaya, U. Weimar, W. G ¨ opel, In 2 O 3 and MoO 3 thin film semiconductor sensors: interaction with NO 2 and O 3 , Sens. Actuators B 47 (1998) 92–99. Biographies Ahmed Labidi, was born in La Marsa, Tunis, Tunisia in 1975. He received the DEA (post-graduate diploma) in quantum physics 2002 from the University of Tunis El Manar (Tunis, Tunisia). He is currently preparing his PhD degree in physics and Materials science in the L2MP laboratory at the Paul CEZANNE, Aix-Marseille III University (France), in cooperation with the URPSC Labora- tory at the 7 November University (Tunisia). His research interest is the electrical studies of the WO 3 gas sensors under oxidizing and reducing gases by impedance spectroscopy. Eveline Gillet, born in 1937, graduated from the University of Poitiers (France), Docteur ` es Sciences (Universit ´ e de Provence-1969). She is Professor in Physics at Paul C ´ ezanne University – Aix-Marseille (France). She worked in the area of Surface Science. In particular she studied chemisorption on transition metal nanoparticles (model catalysts). Actually she is involved in a research devoted to the electrical properties of nanostructured metal oxide semiconductor thin films and nanorods for applications to new sensing devices. Romain Delamare, was born in 1973. He is professor assistant at Paul CEZANNE, Aix Marseille III University (France). He was awarded his PhD degree in semiconductors physic from University of Orl ´ eans (France) in 2003. His principal research interests are now directed towards WO 3 gas sensors and selectivity enhancement strategies including noise spectroscopy and modelling of sensor responses. A. Labidi et al. / Sensors and Actuators B 120 (2006) 338–345 345 Mhamed Ali Maaref, was born in 1955. He is professor in Solid State Physics at the Engineering Institute of Tunis (INSAT) in 2000. He was awarded his Doctorat d’Etat in semiconductors from University of Tunis (Tunisia) in 1994. He is head of research group in Semiconductor Physics and Sensors at the Institute of Science and Technology (IPEST) University 7 Novembre (Tunisia). His principal research activities are involved on III/V semiconductors materials: Study by reflectivity and luminescence of GaAs/AlAs Super-lattices, Quantum Wells and self organized InAs/GaAs Quantum dots elaborated by MBE. Khalifa aguir, was born in 1953. He is professor at Paul CEZANNE – Aix Mar- seille III University (France). He was awarded his Doctorat d’Etat ` es Science degree from Paul Sabatier University Toulouse (France) in 1987. He is currently head of Sensors Group at Laboratoire Mat ´ eriaux & Micro ´ electronique (L2MP- CNRS) Marseille (France). His principal research interests are now directed towards WO 3 gas sensors and multisensors and selectivity enhancement strate- gies including PCA analysis and nose spectroscopy; and modelling of sensor responses. . Sensors and Actuators B 120 (2006) 338–345 Review Ethanol and ozone sensing characteristics of WO 3 based sensors activated by Au and Pd A. Labidi a,b ,. gases, for the sensors WO 3 bare, Au/ WO 3 and Pd/ WO 3 at T work = 300 ◦ C, and resulting sensing response “S gas ” of the sensors calculated by using the