Sensors and Actuators B 114 (2006) 910–915 A selective NH 3 gas sensor based on Fe 2 O 3 –ZnO nanocomposites at room temperature Huixiang Tang, Mi Yan, Hui Zhang, Shenzhong Li, Xingfa Ma, Mang Wang, Deren Yang ∗ State Key Laboratory of Silicon Materials, Zhejiang University, Hangzhou 310027, PR China Received 16 April 2005; received in revised form 25 August 2005; accepted 26 August 2005 Available online 19 October 2005 Abstract Gas sensors based on the Fe 2 O 3 –ZnO nanocomposites with different compositions of Fe:Zn was prepared by a sol–gel and spin-coating method. Morphology of the Fe 2 O 3 –ZnO nanocomposites was characterized by transmission electron microscopy (TEM), X-ray diffraction (XRD, D/max- rA) and energy dispersive X-ray analysis (EDX). The results of electrical and sensing measurement indicated that the sensor with Fe:Zn= 2% exhibited fairly excellent sensitivity and selectivity to NH 3 at room temperature. The response and recovery time of the sensor were both less than 20 s. Finally, the mechanism for the improvement in the gas sensing properties was discussed. © 2005 Elsevier B.V. All rights reserved. Keywords: ZnO nanoparticles; Fe 2 O 3 ; Gas sensor; NH 3 1. Introduction Since many decades, the world awareness about environmen- tal problems and human safety is increasing with the techno- logical development. Therefore, sensors are required for many applications. Recently, the need to detect low ammonia concen- trations has greatly increased in many fields of technological importance, such as food technology, chemical engineering, medical diagnosis, environmental protection, monitoring of car interiors and industrial processes. Seiyama et al. proposed the gas sensors based on ZnO thin films for the first time [1]. ZnO is sensitive to many gases of inter- est, such as trimethylamine (TMA) [2–4],H 2 [5], oxygen [6–8], H 2 O [9,10], ethanol [11] and NH 3 [12], etc. It also has a rapid response with a possibility of miniaturization. However, it has some drawbacks, such as high working temperature, normally between 400 and 500 ◦ C, poor gas selectivity and relatively low gas sensitivity [13]. To overcome these disadvantages, considerable research and development are underway. There are various techniques to modify the sensing properties of the gas sensors. One critical approach is to modify the metal oxide surface by using noble ∗ Corresponding author. Tel.: +86 571 8795 1667; fax: +86 571 8795 2322. E-mail address: mseyang@zju.edu.cn (D. Yang). metals (Au, Pt or Pd) [14,15] or rare earth metals (La, Y and Ce) [16,17]. ZnO(n)/CuO(p) heterocontact configuration also showed some possibility of improving theselectivity [18]. Nanto et al. have reported that a sensor based on a ZnO thin film doped with Al, In or Ga could detect the ammonia gas whose concentra- tion was as low as 1 ppm [12]. But the working temperature was as high as 350 ◦ C. Recently, Ivanovskaya et al. suggested that a sensor based on ␣-Fe 2 O 3 /In 2 O 3 nanocomposites exhibited high sensitivity to NO 2 [19]. The present work was undertaken to investigate the gas sens- ing behavior of ZnO nanoparticle thin films doped with ␣-Fe 2 O 3 nanoparticles prepared by a sol–gel and spin-coating method. Morphological, structural and sensing properties at room tem- perature were studied. The ultimate objective of this study is to improve the gas selectivity and sensitivity of the nano-sized ZnO-based sensors at room temperature. 2. Experimental ZnO nanoparticles doped with ␣-Fe 2 O 3 nanoparticles were prepared in a similar manner to the literature procedure [20]. The ␣-Fe 2 O 3 nanoparticles were synthesized by a hydrother- mal method [21]. Some of the ␣-Fe 2 O 3 nanoparticles were ultrasonically dispersed into methanol (200 ml) at about 60 ◦ C. Subsequently, 0.01 M Zn(Ac) 2 was dissolved into the above solution. Then, a 0.03 M solution of KOH (65 ml) in methanol 0925-4005/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2005.08.010 H. Tang et al. / Sensors and Actuators B 114 (2006) 910–915 911 was added dropwise. The reaction mixture was stirred for 2 h. The resulting solution was concentrated by the evaporation of the solvent. The resulting white product was centrifugalized, washed with deionized water and ethanol to remove the ions possibly remaining in the final product. The ZnO nanoparticles doped with ␣-Fe 2 O 3 nanoparticles with different compositions (molar ratio) of Fe:Zn being 0, 1, 2, 3 and 4% were prepared. Finally, five ZnO nanoparticle solutions were obtained, labeled as sample 0, 1, 2, 3 and 4, respectively. The obtained samples were characterized by transmission electron microscopy (TEM), which was performed with a JEM 200 CX microscope operated at 160 kV. For X-ray diffraction (XRD) (D/max-rA) and energy dispersive X-ray analysis (EDX) (Phoenix) measurement, pow- der samples were used, which were prepared by annealing the final precipitates at 200 ◦ C for 3 h. Interdigitated Au electrodes were obtained by metal deposi- tion on glass substrates. The final precipitate was redispersible in ethanol. The Fe 2 O 3 –ZnO nanocomposite films were fabricated on the top of the Au electrodes by spin-coating, followed by annealing at 200 ◦ C for 3 h before electrical and sensing mea- surement. Gas sensing behavior of the ZnO nanosensors was measured at room temperature by using the Keithley 236 Source Measure Unit. The chamber was purged with N 2 until a steady baseline of the sensor resistance was reached. Then, the test vapor was injected at a fixed concentration of 0.4 ppm in N 2 . In general, the dc voltage was fixed at 10 V, and the changes of current with time were recorded. The sensor response is given here as the current ratio I g /I a , where I g and I a are the current across the sensor in the test gas and in air, respectively [22,23]. The response-recovery time of the sensor is defined as the time needed to reach 90% of the original resistance. 3. Results 3.1. Morphology of Fe 2 O 3 –ZnO nanocomposites Fig. 1 shows TEM images of the Fe 2 O 3 –ZnO nanocompos- ites with different compositions of Fe:Zn. Morphology of the pure ␣-Fe 2 O 3 prepared by a hydrothermal method was nanopar- ticles (Fig. 1a). Size of the nanoparticles was about 10 nm. When 1% Fe 2 O 3 (molar ratio) nanoparticles (sample 1) were doped into the ZnO nanoparticles, the morphology of the result- ing sample was similar to that of the pure ␣-Fe 2 O 3 (Fig. 1b). It is obvious that the prepared nanoparticles were crystalline Fig. 1. TEM images of the ZnO–Fe 2 O 3 nanocomposites with different compositions (molar ratios) of Fe:Zn: (a) pure ␣-Fe 2 O 3 nanoparticles; (b) sample 1, inset is the selected area electron diffraction (SAED) pattern of the particles; (c) sample 2; (d) sample 3; (e) sample 4, inset is the SAED pattern of the nanorods. 912 H. Tang et al. / Sensors and Actuators B 114 (2006) 910–915 Fig. 2. X-ray diffraction patternsof the ZnO–Fe 2 O 3 nanocomposites with differ- ent compositions of Fe:Zn. All the diffraction peaks could be indexed according to hexagonal structure ZnO. from the inset of Fig. 1b. The morphology of sample 2 was still nanoparticles and the size was about 10 nm (Fig. 1c). However, a few short nanorods and nanoparticles co-exited in sample 3 (Fig. 1d). Almost only nanorods exited in sample 4 (Fig. 1e). These nanorods were still crystalline (inset of Fig. 1e). The growth mechanism of the nanorods was put forward based on the above results. It may be the fact that the added Fe 2 O 3 nanopar- ticles promoted the growth of ZnO nanoparticles as the seeds. Then, those particles aggregated along one direction and formed the ZnO nanorods. 3.2. Structural and compositional characterization Fig. 2 shows the X-ray diffraction patterns of the Fe 2 O 3 –ZnO nanocomposites with different compositions of Fe:Zn. All the reflections in the XRD patterns can be indexed to the hexag- onal structure ZnO with lattice constants of a = 0.3250 nm and c = 0.5207 nm (JCPDS, 79–2205). No Fe 2 O 3 phase was detected according to the XRD patterns. It can be concluded that the crys- tallization is relatively poor when the composition of Fe is 2%. Fig. 3 shows the image of EDX and composition of sample 4. Al and Si elements came from the substrates (glass). Fe, O and Zn elements belonged to the sample. It was calculated that the composition results were almost consistent with the molar ratio of ZnO and Fe 2 O 3 from the composition table. Therefore, the nanorods in sample 4 (Fig. 1e) belonged to ZnO nanorods. 3.3. Selectivity of gas sensors based on Fe 2 O 3 –ZnO nanocomposites Fig. 4 shows the selectivity of the gas sensor based on pure ZnO nanoparticle. The sensor was exposed to TMA, ethanol, methanol, NH 3 and HCl of the same concentration level of 0.4 ppm at room temperature. From the plots, it can be deduced that the selectivity of the ZnO nanoparticles is poor. For compari- son, the current response of the gas sensors basedon Fe 2 O 3 –ZnO nanocomposites with different compositions of Fe:Zn as a function of time was measured. The same testing conditions were applied for both pure ZnO nanoparticles and Fe 2 O 3 –ZnO nanocomposites. Selectivity of the gas sensors based on 2% (Fig. 5a) and 4% Fe 2 O 3 –ZnO nanocomposites (Fig. 5b) at room Fig. 3. EDX and composition of sample 4. temperature is shown in Fig. 5. It is clear that the sensitivity of both the sensors exposed to NH 3 is fairly high (about 10,000), whereas that to the other gases is much lower. The sensitivity and selectivity are better than that of the gas sensor given in Ref. [12]. These results indicate the fairly good NH 3 selectivity of the Fe 2 O 3 –ZnO nanocomposite thin films. 3.4. Gas response and recovery characterization Fig. 6 shows the gas response of the ZnO-based gas sen- sors including a pure nano-crystalline ZnO film and the four Fe 2 O 3 –ZnO nanocomposite thin films with different composi- tions of Fe:Zn = 1, 2, 3 and 4%, when exposed to 0.4 ppm NH 3 . Fig. 4. Selectivity of the gas sensor based on pure ZnO nanoparticle. H. Tang et al. / Sensors and Actuators B 114 (2006) 910–915 913 Fig. 5. Selectivity of the gas sensors based on: (a) 2% and (b) 4% Fe 2 O 3 –ZnO nanocomposites. Table 1 summarizes the sensor signal (response magnitude), response time and recovery time of the ZnO-based sensors with different compositions of Fe:Zn. The gas sensitivity of the ZnO- based sensors increased dramatically from 100 to 10,000 as the Fe 2 O 3 nanoparticle content increasesd initially from 0 to 2%. However, the gas sensitivity decreased dramatically when the Fe 2 O 3 nanoparticle content increased to 3 and 4%. The response and recovery time of the sensor with Fe:Zn = 1% were more than 100 and 30 s, respectively. But the response and recovery time of the sensor with Fe:Zn = 2% were both less than 20 s. According to the gas sensitivity, response time and recovery time, it can be concluded that Fe:Zn = 2% is the best composition. Fig. 6. Response of gas sensors based on the Fe 2 O 3 –ZnO nanocomposites with different compositions of Fe:Zn exposed to NH 3 gas. Table 1 NH 3 sensing properties of the Fe 2 O 3 –ZnO nanocomposite gas sensors with different compositions of Fe:Zn Sample Sensor signal (S = I g /I a ) Response time (s) Recovery time (s) Pure ZnO 100 100 70 1% Fe 2 O 3 –ZnO 10 4 100 30 2% Fe 2 O 3 –ZnO 10 4 20 20 3% Fe 2 O 3 –ZnO 100 60 20 4% Fe 2 O 3 –ZnO 80 500 70 Fig. 7. Reproducibility of the gas sensor based on 2% Fe 2 O 3 –ZnO nanoparticle. 3.5. Reproducibility of gas sensor based on 2% Fe 2 O 3 –ZnO nanoparticles Fig. 7 shows the reproducibility of the sensor based on 2% Fe 2 O 3 –ZnO nanoparticles when exposed to 0.4 ppm NH 3 for three times at room temperature. It is clear that the response and recovery characteristics are almost reproducible and rather quick when exposed to NH 3 and also when exposed again to N 2 . 4. Discussion Based on the above results, the reason for the enhanced NH 3 sensitivity and selectivity of the Fe 2 O 3 –ZnO nanocomposite thin film gas sensors was put forward. Ivanovskaya et al. [19,24] have suggested that ethanol detection is a multi-step process involv- ing both reductive–oxidative and acid–base interactions with a sensor based on heterojunction oxide structures. The reactivity of oxides in acid–base reactions depends on the electronegativ- ity of the metal cation. The electronegativity is the measure of the Lewis acid site activity. The relative measure of the oxide activity in oxidation reactions can be the oxygen-oxide surface bonding energy. In fact, the less the energy of oxygen atom iso- lation from the oxide surface is, the higher the oxide oxidizing ability is. The reactivity of oxidesin acid–base reactions depends on the electronegativity of cations M n+ : χ = χ 0 (2n + 1) (1) where χ 0 is the Pauling electronegativity and n is the ion charge. The Pauling electronegativities of Fe–O and Zn–O are 914 H. Tang et al. / Sensors and Actuators B 114 (2006) 910–915 1.83 and 1.65 in Pauling units [25]. So, the adsorption of the detected gas molecules to be detected at Lewis sites would be increased when the ZnO nanoparticles was doped with Fe 2 O 3 nanoparticles. The oxides, which are characterized by the possibility of metal ion reduction without oxide phase state modification, have the greatest ability to promote oxidizing process. Fe 2 O 3 is inclined to facilitate the changing of the metal ion oxidizing state: Fe(III) ↔ Fe(II), while the oxide phase remains original [24]. That is, the complete oxidation of intermediates is going effectively at the center of Fe 2 O 3 nanoparticles. Due to the above-mentioned two reasons, the gas sensitivity of Fe 2 O 3 –ZnO nanocomposites was improved. Meanwhile, there are two reasons for the fact that the elec- tron donating ability of NH 3 is higher than those of ethanol and methanol. One is due to the higher electronegativity of the oxy- gen atom than that of the nitrogen atom. The other is that there is a lone electron pair in NH 3 . What is more important is that low- temperature ammonia oxidation was observed over iron oxide [26,27]. That is, the addition of Fe 2 O 3 promoted the interaction between the gas sensor and NH 3 . Therefore, the NH 3 selectivity was improved. However, according to the gas sensitivity, response time and recovery time, it can be concluded that the composition of Fe:Zn = 2% is the best composition. This may be mainly due to the morphology and the poor crystallization of the Fe 2 O 3 –ZnO nanocomposites. It is well known that the sensing mechanism of semiconducting oxide gas sensors is based on the surface reaction of semiconducting oxides [28]. The surface–volume ratio of nanoparticles (when the composition of Fe:Zn is 1 or 2%) is higher than that of nanorods (when the composition of Fe:Zn is 3 or 4%). From the results of XRD, the crystallization of the Fe 2 O 3 –ZnO nanoparticles was relatively poor, indicat- ing that the defects were increased. In general, there are many oxygen vacancies in the nano-sized ZnO [29,30]. So, more gas molecules are easy to be adsorbed on the active centers [22], which result in an increase of the sensitivity of Fe 2 O 3 –ZnO nanoparticles. Meanwhile, addition of a more amount of Fe 2 O 3 nanoparticles may cover the active centers of ZnO. In summary, Fe 2 O 3 nanoparticles can enhance the gas sensing properties of the ZnO nanosensors, and the composition of Fe:Zn = 2% is the best for the Fe 2 O 3 –ZnO nanocomposite thin film gas sensors. 5. Conclusion Gas sensors based on Fe 2 O 3 –ZnO nanocomposites have been prepared with different compositions of Fe:Zn. The sensor with Fe:Zn = 2% exhibited fairly excellent sensitivity and selectivity to NH 3 at room temperature. The response and recovery time of the sensor were about 20 s. The reproducibility of the ZnO gas sensor with Fe:Zn = 2% was good. So, the sensor could be used for many times. The increased sensitivity and selectivity to NH 3 may largely be attributed to the addition of Fe 2 O 3 nanoparticles, which can promote the adsorption of NH 3 molecules on the oxide surface and accelerate the oxidizing process. Acknowledgments The authors would like to appreciate the financial supports of the Natural Science Foundation of China (60225010) and 863 Project No. (2004AA513024). We also thank Prof. Youwen Wang for TEM and EDX measurement. References [1] T. Seiyama, A. Kato, K. Fjishi, M. Nagatani, A new detector for gaseous components using semiconductive thin films, Anal. Chem. 34 (1962) 1502–1503. [2] M. Egashira, Y. Shimizu, Y. Takao, Trimethylamine sensor based on semiconductive metal oxides for detection of fish freshness, Sens. Actu- ators B 1 (1990) 108–112. [3] H. Nanto, H. Sokooshi, T. Kawai, Aluminum-doped ZnO thin film gas sensor capable of detecting freshness of sea foods, Sens. Actuators B 13–14 (1993) 715–717. [4] H. Nanto, H. Solooshi, T. Kawai, T. Usuda, Zinc oxide thin-film trimethylamine sensor with high sensitivity and excellent selectivity, J. Mater. Sci. Lett. 11 (1992) 235–237. [5] S. Basu, A. Dutta, Modified heterojunction based on zinc oxide thin film for hydrogen gas-sensor application, Sens. Actuators B 22 (1994) 83–87. [6] G. Sbervelieri, P. Nelli, S. Groppelli, F. Quaranta, A. Valentini, L. Vasanelli, Oxygen gas sensing characteristics at ambient pressure of undoped and lithium-doped ZnO-sputtered thin films, Mater. Sci. Eng. B 7 (1990) 63–68. [7] U. Lampe, J. M ¨ uller, Thin-film oxygen sensors made of reactively sput- tered ZnO, Sens. Actuators B 18 (1989) 269–284. [8] F. Chaabouni, M. Abbab, B. Rezug, Metrological characteristics of ZnO oxygen sensor at room temperature, Sens. Actuators B 100 (2004) 200–204. [9] E. Traversa, A. Bearzptti, A novel humidity-detection mechanism for ZnO dense pellets, Sens. Actuators B 23 (1995) 181–187. [10] A.M. Edwin Suresh Raj, C. Maria Magdalane, K.S. nagaraja, Zinc(II) oxide–yttrium(III) oxide composite humidity sensor, Phys. Status Solidi (a) 191 (2002) 230–234. [11] Q. Wan, Q.H. Li, Y.J. Chen, T.H. Wang, Fabrication and ethanol sensing characteristics of ZnO nanowire gas sensors, Appl. Phys. Lett. 84 (2004) 3654–3656. [12] H. Nanto, T. Minami, S. Takta, Zinc-oxide thin-film ammonia gas sen- sors with high sensitivity and excellent selectivity, J. Appl. Phys. 60 (1986) 482–484. [13] J. Xu, Y. Shun, Q. Pan, J. Qin, Sensing characteristics of double layer film of ZnO, Sens. Actuators B 66 (2000) 161–163. [14] F. Boccizzi, A. Chiorino, S. Tsubota, M. Haruta, An IR study of CO-sensing mechanism on Au/ZnO, Sens. Actuators B 24–25 (1995) 540–543. [15] S.J. Gentry, T.A. Jones, The role of catalysis in solid-state gas sensors, Sens. Actuators B 10 (1986) 141–163. [16] B. Bhooloka Rao, Zinc oxide ceramic semi-conductor gas sensor for ethanol vapour, Mater. Chem. Phys. 64 (2000) 62–65. [17] X. Niu, W. Du, W. Du, K. Jiang, Preparation and gas sensing properties of Y 3+ ,La 3+ ,Ce 4+ -doped ZnO, Chin. Rare Earths 24 (6) (2003) 44–47. [18] M. Miyayama, K. Hikita, G. Uozumi, H. Yanagida, A.C. impedance analysis of gas-sensing property in CuO/ZnO heterocontacts, Sens. Actu- ators B 24–25 (1995) 383–387. [19] M. Ivanovskaya, D. Kotsikau, G. Faglia, P. Nelli, S. Irkaev, Gas-sensitive properties of thin film heterojunction structures based on Fe 2 O 3 –In 2 O 3 nanocomposites, Sens. Actuators B 93 (2003) 422–430. [20] C. Pacholski, A. Kornowski, H. Weller, Self-assembly of ZnO: from nanodots to nanorods, Angew. Chem. Int. Ed. 41 (2002) 1188–1191. [21] Z. Zhang, B.Q. Wei, P.M. Ajayan, Self-assembled patterns of iron oxide nanoparticles by hydrothermal chemical-vapor deposition, Appl. Phys. Lett. 79 (2001) 4207–4209. H. Tang et al. / Sensors and Actuators B 114 (2006) 910–915 915 [22] A. Gurlo, M. Ivanovskaya, N. B ˆ arsan, M. Schweizer-Berberich, U. Weimar, W. G ¨ opel, A. Di ´ eguez, Grain size control in nanocrystalline In 2 O 3 semiconductor gas sensors, Sens. Actuators B 44 (1997) 327–333. [23] W. Schmid, N. B ˆ arsan, U. Weimer, Sensing of hydrocarbons with tin oxide sensors: possible reaction path as revealed by consumption mea- surements, Sens. Actuators B 89 (2003) 232–236. [24] M. Ivanovskaya, D. Kotsikau, G. Faglia, P. Nelli, Influence of chemical composition and structural factors of Fe 2 O 3 /In 2 O 3 sensor on their selec- tivity and sensitivity to ethanol, Sens. Actuators B 96 (2003) 498–503. [25] L. Pauling, The Nature of the Chemical Bond, third ed., Cornell Uni- versity Press, Ithaca, NY, 1960, p. 93. [26] O.N. Sil’chenkova, V.A. Matyshak, V.N. Korchak, Intermediates deter- mined by in situ IR-spectroscopy in low-temperature oxidation of ammo- nia on Fe 2 O 3 , Kinet. Catal. 40 (1999) 526–531. [27] O.N. Sil’chenkova, V.N. Korchak, V.A. Matyshak, The mechanism of low-temperature ammonia oxidation on metal oxides according to the data of spectrokinetic measurements, Kinet. Catal. 43 (2002) 363–371. [28] J. Watson, The tin oxide gas sensor and its application, Sens. Actuators 5 (1984) 29–42. [29] T. Nagase, T. Ooie, Y. Makita, M. Nakatsuka, K. Shinozaki, N. Mizutani, A novel method for the preparation of green photoluminescent undopde zinc oxide film involving excimer laser irradiation of a sol–gel-derived precursor, Jpn. J. Appl. Phys. 39 (2000) 713–715. [30] K. Vanheusden, W.L. Warren, C.H. Seager, D.R. Tallant, J.A. Voigt, B.E. Gnade, Mechanisms behind green photoluminescence in ZnO phosphor powders, J. Appl. Phys. 79 (1996) 7983–7990. Biographies Huixiang Tang was born in 1978. Now, she is PhD candidate from State Key Lab of Silicon Materials at Zhejiang University, China. Her research project is gas sensors based on the ZnO nanoparticles. Mi Yan was born in 1965. He has been a professor of materials science at Zhejiang University since 1998. He graduated from Department of Materials Science and Engineering, Southeast University in 1980. Professor Yan has been working in the fields of functional materials and surface treatment. His current research interests are magnetic materials and related functional materials. Hui Zhang received his PhD degree from State Key Laboratory of Silicon Materials at Zhejiang University in 2004. Now, he is a teacher at Zhejiang University. His research interest is preparation and application of nano-sized compound semiconductor materials. Shenzhong Li received his master degree from State Key Laboratory of Silicon Materials at Zhejiang University in 2005. His research interest is preparation and application of nano-sized compound semiconductor materials. Xingfa Ma, Professor, executive director in chief of the Department of Adhe- sives and Coatings, vice director of the Department of Sealants and Rubber Composites of Shandong Research Institute of Non-metallic Materials, Com- mittee Member of Chinese Standard of Adhesives and Sealants. Now, he is a PhD candidate of materials physics and chemistry of Zhejiang University. His research interests include polymer coating, interface, surface modifications, polymer-based composites and organic sensitive materials for sensor. Mang Wang graduated from Department of Chemical Engineering of Zhe- jiang University in 1961. Now, he is a professor in materials physics and chemistry and polymer materials at Zhejiang University. He is currently the director of Reprographic Science and Engineering Society of China and a member of Specialist Group of Polymer Materials Division. Deren Yang was born in Yangzhou, China in 1964. He received his bache- lor degree from Department of Material Science and Engineering at Zhejiang University, and in 1991 PhD degree in the State Key Laboratory of Silicon Materials at Zhejiang University. Now, he is a Cheung Kong Professor, deputy director of the State Key Laboratory of Silicon Materials at Zhejiang Uni- versity. His current research interests are semiconductor materials, including growth, process and defect engineering of Czochralski silicon used for ultra- large scale integrated circuits (ULSI); preparation and application of silicon nano-wires, nano-tubes and other one dimensional semiconductor materials. . Sensors and Actuators B 114 (2006) 910–915 A selective NH 3 gas sensor based on Fe 2 O 3 –ZnO nanocomposites at room temperature Huixiang Tang, Mi Yan,. research interests are magnetic materials and related functional materials. Hui Zhang received his PhD degree from State Key Laboratory of Silicon Materials