microlotus constructed by fe doped zno hierarchically porous nanosheets

Sensors and Actuators B 158 (2011) 9– 16 Contents lists available at ScienceDirect Sensors and Actuators B: Chemical journa l h o mepage: www.elsevier.com/locate/snb Micro-lotus constructed by Fe-doped ZnO hierarchically porous nanosheets: Preparation, characterization and gas sensing property Ang Yu a , Jieshu Qian b , Hao Pan a , Yuming Cui a , Meigui Xu a , Luo Tu a , Qingli Chai a , Xingfu Zhou a,∗ a State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemistry and Chemical Engineering, Nanjing University of Technology, Nanjing 210009, PR China b Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario, Canada M5S 3H6 a r t i c l e i n f o Article history: Received 8 November 2010 Received in revised form 21 March 2011 Accepted 23 March 2011 Available online 4 April 2011 Keywords: ZnO microsphere Nanosheet Fe doping Porous Gas sensing Oxygen vacancies a b s t r a c t Preparation of micro-lotus constructed by hierarchically porous Fe-doped ZnO nanosheets via a facile hydrothermal method is reported here. The products have been analyzed by X-ray powder diffraction (XRD), scanning electron microscopy (SEM), field emission scanning electron microscopy (FESEM), trans- mission electron microscopy (TEM), and high resolution transmission electron microscope (HRTEM). Results showed that the morphology of the sample did not change with the Fe doping amount. The photoluminescence (PL) spectra revealed the existence of oxygen vacancies in the Fe-doped ZnO porous nanosheets, which is beneficial to the adsorption of oxygen and gas response, resulting in the improved performances in the later gas sensing experiments towards several reductive gases. The effect of Fe doping percentage on the gas response has also been investigated. We found that ZnO sample with Fe doping atomic percentage of 1% showed the highest gas sensing performance, while excessive Fe doping in ZnO suppressed the gas sensing response. A possible mechanism of how Fe-doped ZnO-based sensor responses to the target gas is also proposed. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Zinc oxide, an n-type semiconductor with wide band gap of 3.3 eV, has attracted vast research interests because of its tremendous potential applications in piezoelectric nanogenera- tors, nanolasers, solar cells, gas sensors and so on [1–5]. As one of the most prominent materials for gas sensor, ZnO has already shown good response to pollutant gases such as H 2 S, NO x , and benzene [6,7], and explosive gases such as H 2 , CO, ethanol and acetone [8–10]. However, at present, the sensitivity of ZnO based gas sensor towards some chemical stable gases is still low. Thus, many efforts have been made to improve the gas sensitivity and selectivity by optimizing the morphology, surface-to-volume ratio and active center of the material. The gas sensing mechanism had been suggested by other researchers including the desorp- tion of adsorbed surface oxygen in ZnO, exchange of charges between adsorbed gas species and the ZnO surface, which is lead- ing to changes in depletion depth and changes in surface or grain boundary conduction by gas adsorption/desorption [11]. Doping of ZnO with various elements, such as noble metals, rare met- als, transition metals, or metal oxides, has been reported to be a ∗ Corresponding author. Tel.: +86 25 83587773; fax: +86 25 83587773. E-mail address: Zhouxf@njut.edu.cn (X. Zhou). useful way to improve the electrical conductivity when they are used in gas sensing devices. For examples, Navale and cowork- ers observed that undoped ZnO responses perceptibly to LPG while Ru doped ZnO sample highly senses ethanol vapors [12]. Zhu and coworkers chose to dope the ZnO nanoparticles with metal antimony from alloy with Zn/Sb = 15:1 and obtained short response–recovery time [13]. Zhang and coworkers found that the TiO 2 -doped ZnO sensor exhibited remarkably enhanced response to 100 ppm toluene even at a lower temperature of 290 ◦ C [14]. Nevertheless, there is so few reports about investigating the effect of Fe doping in ZnO nanomaterials on the gas sensing property [15], in contrast, people are more interested in using Fe-doped ZnO as high-performance ferromagnetic or photoelectric materials [16,17]. We have been interested in exploring the underlying connec- tion of the properties with the structure and morphology of the micro-nanostructured materials [18–20]. In this paper, we report a hydrothermal preparation of Fe-doped ZnO microspheres with lotus-like morphology, which are constructed by the porous ZnO nanosheets. Then both Fe-doped ZnO and undoped ZnO as gas sen- sor to ethanol and acetone were tested and the effect of Fe doping amount on the gas sensor performance was also systematically investigated, the results showed that there is a significant improve- ment of gas response of the Fe-doped ZnO compared to undoped ZnO and the optimum Fe doping atomic percentage was around 1–3%. 0925-4005/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2011.03.052 10 A. Yu et al. / Sensors and Actuators B 158 (2011) 9– 16 2. Experimental 2.1. Preparation All chemicals are analytical-grade reagents without further purification. ZnO and Fe-doped ZnO nanosheets were prepared by a hydrothermal process. 80 mmol zinc nitrate (23.808 g) and 4.8 g urea were dissolved in 400 mL deionized water under constant stir- ring and then divided into eight parts. Eight batches of different amount of ferric nitrate (0 g, 0.012 g, 0.02 g, 0.032 g, 0.04 g, 0.121 g, 0.202 g, and 0.808 g) were then added into as-prepared eight solu- tions, respectively, under constant stirring. Each solution (50 mL in volume) contained 2.98 g zinc nitrate, 0.6 g urea, and different amount of ferric nitrate, the atomic ratio of Fe to Zn was 0%, 0.3%, 0.5%, 0.8%, 1%, 3%, 5%, and 20%. The eight mixed solutions were transferred into eight 80 mL Teflon-lined autoclave and maintained at 433 K for 10 h. When the solutions were cooled to room tem- perature, the white precipitates were collected and washed with distilled water and pure ethanol several times, followed by drying at 333 K under vacuum. At last, the products were annealed at 773 K for 2 h. We refer to these samples as samples (a), (b), (c), (d), (e), (f), (g), and (h), representing the Fe doping atomic percentage of 0%, 0.3%, 0.5%, 0.8%, 1%, 3%, 5%, and 20%, respectively. 2.2. Characterization The morphology and size of the products were characterized by scanning electron microscopy (SEM, FEI Quanta-200), the fine structure of the material was obtained by a field emission scanning electron microscopy (FESEM, HITACHI S-4800). The transmission electron microscopy (TEM) and high-resolution transmission elec- tron microscopy (HRTEM) observations were performed on a JEOL JEM-2010UHR instrument at an acceleration voltage of 200 kV. The crystalline structure of samples were determined by the Bruker- D8 Advance X-ray diffractometer with graphite monochromatized high-intensity Cu K␣ radiation at 40 KV and 30 mA, the data were collected at a scanning rate of 0.2 ◦ per second for 2Â value in the range of 10–80 ◦ . Room-temperature photoluminescence (PL) spectra of the products were recorded on the Cary Eclipse EL0605- 320328 spectrofluophotometer using the 328 nm Xe laser line as the excitation source 2.3. Gas sensor fabrication and response test The ZnO and Fe-doped ZnO sample powders were mixed sepa- rately with ethanol to make sticky pastes, which were then coated on the alumina tube-like substrate, on which gold electrode was pre-assembled. After dying the coated substrates in air, they were placed in the muffle furnace at 873 K for 2 h. At last, a tiny Ni–Cr alloy coil was penetrated through the substrate tube to be a heater. The sensors were kept at working temperature for 120 h in order to improve the stability. The response tests were performed at the WS-30A gas sensing test system (Weisheng Electronics Co. Ltd., P.R. China). The gas responds value is defined as S from the calculation of Eq. (1) S = R a R g (for reducing gas) or S = R g R a (for oxidizing gas) (1) where R a and R g are the resistance in the air and test gas environ- ment, respectively. 3. Result and discussion The crystalline structures of the products were confirmed by X-ray diffraction (XRD). Fig. 1 shows the XRD patterns of the Fe- doped ZnO crystalline samples. Typical wurtzite ZnO (JCPDS Card 8070605040302010 (h) (g) (f) (e) (d) (c) (b) 201 112 103 110 102 101 002 Intensity (a.u.) 2-Th eta (deg.) 100 (a) * Fig. 1. XRD patterns of ZnO doped with different atomic percentage of Fe. No. 36-1451) diffraction patterns can be observed in all the sam- ples, and there are no obvious diffraction patterns of Fe species such as ␣-Fe, ␣-Fe 2 O 3 , ␥-Fe 2 O 3 or Fe 3 O 4 in the samples where the doping atomic percentage is lower than 5%. This indicates that all Fe ions were incorporated in the lattice of the host crystals [21]. However, when the doping atomic percentage of Fe increases to 20%, as one can see from the sample (h) in Fig. 1, a weak charac- Fig. 2. (A) SEM image with low magnification of the undoped ZnO Micro-lotus, inset is an individual lotus-like ZnO micro-nanostructure. (B) FESEM image of a ZnO Micro-lotus doped with atomic percentage of 1% Fe. Inset up right is a magnified image showing the building unit of nanosheets. Bottom left is a picture of a lotus. A. Yu et al. / Sensors and Actuators B 158 (2011) 9– 16 11 Fig. 3. SEM images of the individual ZnO micro-nanostructures doped with different percentage of Fe. The optical image at down right corner showing the colors of the samples with the different doping percentage. teristic peak at the 2Â value of 33.2 ◦ appears, which is attributed to the diffraction peak of lattice facet (1 0 4) of the crystalline Fe 2 O 3 (JCPDS Card No.33-0664). There might also be Fe 2 O 3 present in sample (b)–(g), however, the amount are too small to be detected by the XRD. Meanwhile, it appears that the intensity of the (1 0 0), (0 0 2), (1 0 1) and (1 0 2) diffraction peaks of ZnO decreased and the full width at half maximum (FWHM) of those diffraction peaks increased with the increase of Fe doping amount, revealing that the Fe doping decreased the crystallinity of the host ZnO crystals. This effect can be explained by the introduction of lattice disorder and strain induced by interstitial Fe atoms or the substitution of Fe atoms for Zn atoms [22]. Fig. 2A is the low magnification SEM image of the undoped lotus- like structured ZnO microspheres, inset of Fig. 2A shows that the micro-lotus was constructed by ZnO nanosheets. The diameter of the obtained ZnO microspheres is ca. 25 ␮m. The FESEM character- ization of ZnO sample with 1 at.% Fe-doping was performed and the image is shown in Fig. 2B. There is no notable difference of morphol- ogy and size between the undoped and Fe-doped ZnO samples in Fig. 2A and B. The structure shown in Fig. 2B resembles the shape of lotus (inset down left in Fig. 2B), and the mutilayers can be clearly observed in the Fe-doped ZnO micro-lotus. The ZnO hierarchical nanolayers, i.e. the building blocks of the lotus-like microspheres are magnified and shown as the inset up right in Fig. 2B. This 12 A. Yu et al. / Sensors and Actuators B 158 (2011) 9– 16 Fig. 4. (A and B) Typical TEM images of an individual Fe doped ZnO porous nanosheet with doping amount of 1 at.%, red dash circles highlights the nanopores in the sheet. (C) HRTEM image of the ZnO nanosheet, the red lines distinguish the boundaries between different layers of the ZnO nanosheet. (D) HRTEM image of the ZnO nanosheet, the inset shows the crystal (0 0 1) plane with distance of 0.259 nm of ZnO. structure is similar to the 3D ZnO architectures constructed by multilayered nanosheets, which had been proved to possess a high surface to volume ratio by former literatures [23–25]. When the doping atomic percentage of Fe increases to 20%, the nanosheet become sticky due to the appearance of iron oxide. The down right corner in Fig. 3 shows the optical images of all the samples, the colors of the Fe-doping ZnO samples are changing gradually with the increase of Fe doping amount. The color change is due to the increase amount of the reddish brown Fe 2 O 3 phase in ZnO, which has been confirmed by the XRD patterns. The ZnO micro-lotus doped with 1 at.% Fe was also characterized by TEM and the images are shown in Fig. 4, where we can see that the ZnO nanosheets are porous and the size of the pore varies from 5 nm to 50 nm. The irregular morphology and the size of the pore are marked by the red dash circles in both Fig. 4A and B. Fig. 4C and D are the HRTEM images of the ZnO micro-lotus sample with 1 at % Fe doping. The boundaries of the packing mutilayers of the ZnO nanosheets are marked by red dash lines in Fig. 4C, and the d-spacing value of 0.259 nm is clearly shown in Fig. 4D, which is attributed to the ZnO (0 0 1) plane. Photoluminescence (PL) spectroscopy is an important tool to characterize the intrinsic and extrinsic defects in semiconductors. Fig. 5 shows the PL spectra of all the samples using 328 nm lasers 700 600 500 400 Intensity (a.u.) Wavele n g th ( nm ) 0.3% 0.5% 0% 0.8% 1% 3% 5% 20% Fig. 5. PL spectra of the as-prepared ZnO doped with different atomic percentage of Fe. A. Yu et al. / Sensors and Actuators B 158 (2011) 9– 16 13 Space-charge Layer E f E c E v 2e - O 2 2O - E f E c E v e - O - CO 2 + H 2 O Test Gas (A) (B) Potenti al barrier Particle Particle Surface Surface Potential barrier Air Ga s Space-ch arge Layer Fig. 6. Illustration of possible mechanisms of how Fe-doped ZnO-based sensor responses to ethanol: (A) in air, and (B) in tested gas. as the excitation source. Samples (a)–(g) correspond to the sam- ples of Fe-doping atomic percentage from 0 to 5%. All the samples display strong UV emission at about 390 nm, which is attributed to the excitonic transitions with a band gap of 3.24 eV. The relatively weak blue-green light emission at around 450 nm can be ascribed to structural defects, such as oxygen vacancies or surface deep traps [26,27]. The existence of oxygen vacancies facilitates a high adsorp- tion of oxygen, which increases the chance of ZnO interacting with reductive testing gases [28]. In contrast, sample (h) with 20 at % Fe doping shows no emission in the all wavelength range, this could be explained by the destruction of defects in ZnO by the excess Fe doping atoms. However, there are two valence states of Fe dopant in ZnO, Fe 2+ and Fe 3+ , the effect of Fe doping on the ZnO lumines- cence property is complicated. The ionic radius of Fe 2+ (0.076 nm) is larger than the ionic radius of Zn 2+ (0.072 nm), so the Fe 2+ ions exist in the form of substitutional impurities. The ionic radius of Fe 3+ (0.064 nm) is smaller than the ionic radius of Zn 2+ , Fe 3+ can exist in the form of interstitial impurity besides substitutional impurity. When Fe 3+ ions replaced Zn 2+ , the oxygen ions would be attracted to the Fe 3+ ions to keep the balance of charge [29], that is why Fe 2 O 3 phase was observed in the XRD patterns. The substitutional Fe 3+ ions also affect the concentration of Zn vacancy in ZnO, which plays an important role in photoluminescence emission. In Fig. 5, the Fe-doped ZnO samples showed similar emission intensity to the undoped ZnO sample when the Fe doping amount is lower that 5 at.%. However, excess Fe doping with a percentage higher than 5 at.% decreases the exciton density, which leads to a decrease of emission intensity of 390 nm. In addition, all samples with the Fe-doping percentage from 0 to 5 at.% show the same weak blue- green light emissions at around 450 nm. The emission intensities decrease with the increase of Fe-doping percentage, indicating the loss of oxygen vacancies. The 20 at.% Fe-doping ZnO microsphere shows completely different behavior compared with other sam- ples in PL spectra. This is probably caused by the destruction of ZnO crystal lattice by the Fe 2+ and Fe 3+ , and the excessive doping lead to the formation of Fe 2 O 3 nanoparticles which cover the ZnO surface. The gas-sensing mechanism of Fe-doped ZnO-based sensors is interpreted by the resistance change originated from the chemisorbed oxygen on the depletion surface layer, as shown in Fig. 6 [30,31]. Potential barriers on the boundaries of grains reduce the mobility of the carrier, and the band bending on the concentra- tion of free charge carriers take the dominant effect on the sense response. The oxygen vacancies of Fe-doped ZnO act as electron donors to provide electrons to the conduction band. The doping effect on ZnO gas sensing is actually the catalytic effect of dopants, in the perspective of chemist, the dopants help oxygen transfer in the sensor, while in the perspective of electronic physicist, the het- erostructure of the doped samples favors the electron transfer [32]. When the surface of ZnO nanosheets is exposed to air (Fig. 6A), the oxygen molecules adsorb on ZnO surface and form O 2 − , O 2− and O − by capturing electrons from the conduction band. When the adsorption of oxygen reaches a certain level, a thick space-charge layer is formed which leads to a decrease of carrier concentration and thus results in a higher resistance. In contrast, when Fe-doped ZnO is exposed to reductive gases (Fig. 6B), for instance, ethanol or acetone, the reductive gas will react with the adsorbed oxy- gen species to produce CO 2 and H 2 O, which leads to an increase of carrier concentration and a decrease of the resistances [10]. The mechanism can be explained by several chemical reactions which are shown below: O 2 (gas) → O 2 (ads) O 2 (ads) + e − → O 2 − (ads) O 2 − (ads) + e − → 2O − (ads) CH 3 CH 2 OH(ads) + 6O − (ads) → 2CO 2 (g) + 3H 2 O(l) + 6e − CH 3 COCH 3 (ads) + 8O − (ads) → 3CO 2 (gas) + 3H 2 O(gas) + 8e − The gas sensor properties of all samples with respect to the reductive gases such as ethanol and acetone were measured at 673 K. All the samples were fabricated into three gas sensors, respectively, and all measurements have been repeated in the inter- val between 2 weeks. Fig. 7 shows the gas sensing response of all the obtained Fe-doping ZnO micro-lotus samples, the error bars represent the response variations of each sample. The value of response signal is defined as the average value of S, as expressed in Eq. (1). Fig. 7A and B shows that Fe doping is an effective way to improve ZnO sensing response to the testing gases. ZnO doped with 1 at.% Fe presents the best performance and has a remark- able enhancement of response towards ethanol (shown in Fig. 7A), the value of gas response is almost 2.5 times as large as that of undoped ZnO. A similar result was observed for sensing acetone and shown in Fig. 7B, ZnO doped with 1 at.% Fe have the highest gas sensing response towards acetone. The gas sensing response value increases gradually with the increase of Fe-doping atomic percentage from 0.3% to 1%, but decreases gradually when the dop- ing percentage goes higher than 1%. ZnO doped with 20 at.% Fe shows a low sensitivity to the testing gases. The sensing response value of 1 at.% Fe-doped ZnO is higher than that of pure ZnO and other Fe-doped ZnO samples, which is probably because of the more active adsorption center created by the appropriate per- centage of dopant. It can be concluded that Fe 2 O 3 nanoparticles 14 A. Yu et al. / Sensors and Actuators B 158 (2011) 9– 16 500450400350300250200150100 5 10 15 20 25 30 35 40 45 50 55 60 (g) (h) (f) (e) (d) (c) (b) (a) Response (Ra/Rg) Gas Concentrat ion (pp m) (a) (b) (c) (d) (e) (f) (g) (h) Sensing Toward s Ethanol 500450400350300250200150100 0 5 10 15 20 25 30 35 40 45 50 Response (Ra/Rg) Gas Concentrat ion (pp m) (a) (b ) (c) (d) (e) (f ) (g ) (h) (e) (f) (d) (c) (b) (g) (a) (h) Sensing Towards Acet one Fig. 7. The gas response of ZnO doped with different atomic percentage of Fe in (A) ethanol, and (B) acetone. adhering on the ZnO surface and the oxygen vacancies form the substituent or interstitial impurity in ZnO crystalline play great role for the improvement of its gas sensing performance. Fe 2 O 3, a widely used catalyst, can catalyze the dehydrogenation reaction and the ring-opening reaction of hydrocarbon, which is benefi- cial to gas sensing. Fig. 8 is the schematic representation of ZnO doped with different amount of iron. Appropriate Fe-doping sig- nificantly improves the gas sensing, however, excessive amount of Fe 2 O 3 covers the surface of ZnO and reduces the amount of oxygen adsorption and reactive sites, and thus, suppresses the gas-sensing properties [33]. Fig. 9A and B shows the response and recovery time of the 1 at.% Fe-doped ZnO responsing to ethanol and acetone. The response time is defined as the time needed for target gas to reach 90% of the equilibrium value, and the recovery time is the time needed for gas sensor to achieve a conductance 10% of the original value in air. The response and recovery time of the 1 at.% Fe-doped ZnO in 100 ppm ethanol is 6 s and 5 s (Fig. 9A), and the response and recovery time is 3 s and 8 s when exposed to acetone (Fig. 9B). An empirically representation of the gas response signal value of the semiconducting oxide sensor is: S = 1 + A g (P g ) ˇ (2) where P g is the partial pressure of the testing gas, which is pro- portional to the gas concentration, A g is a prefactor, and ˇ is the exponent on P g . Generally, the exponent ˇ has an ideal value of either 0.5 or 1, which is derived from the surface interaction between chemisorbed oxygen and reductive gas to the n-type semi- conductor [34,35]. The logarithm of the value calculated from Eq. (2) versus the logarithm of gas concentration was shown in Fig. 9C and D. The data can be fitted linearly very well, the slope of the fit- ting line is the correlation coefficient R of the Fe-doped ZnO sensor. The correlation coefficient of ethanol sensor is 0.9974 and ace- tone sensor is 0.9957 in the range from 100 to 500 ppm at 673 K. These results show that the sensor matches with the di-logarithm amplifying circuits for practical application in the detection range of 100–500 ppm of target gas. Fig. 8. Schematic representation of (A) undoped ZnO, (B) doped with appropriate, and (C) doped with excessive atomic percentage of Fe. Below are the SEM images of the corresponding Fe-doped ZnO samples. A. Yu et al. / Sensors and Actuators B 158 (2011) 9– 16 15 2.82.72.62.52.42.32.22.12.01.9 1.35 1.40 1.45 1.50 1.55 1.60 1.65 1.70 1.75 Experimental data Average Value Linear Fitting Y = 0.467X + 0.461 R = 0.9974 Log S Log C (C) 2.82.72.62.52.42.32.22.12.01.9 1.25 1.30 1.35 1.40 1.45 1.50 1.55 1.60 1.65 1.70 Experimental Data Average Value Linear Fitting Y = 0.4988X + 0.3138 R = 0.9957 Log S Log C (D) 1209060300 0 10 20 30 Response (Ra/Rg) Time (s) Sample (e) in 100ppm Ethanol (A) Gas In Gas Off 7s 5s 1209060300 0 10 20 Response (Ra/Rg) Time (s) Sample (e) in 100ppm Acetone (B) 8s 3s Gas In Gas Off Fig. 9. Response of ZnO doped with 1 at.% Fe in (A) ethanol, (B) acetone; (C) logarithm plots of the ethanol sensor response value versus the concentration, and (D) logarithm plots of the acetone sensor response value versus the gas concentration. 4. Conclusion In summary, Fe-doped ZnO microspheres with lotus-like mor- phology constructed by hierarchically porous nanosheets were prepared through a hydrothermal route. Systematical investigation on the effect of Fe doping on the structure and sensing property of the host ZnO crystals was presented. The results showed that the structure disturbance of ZnO crystal induced by the Fe dop- ing improves its gas response signal value to several reductive gases. The Fe doping micro-nanostructured ZnO showed signifi- cantly higher gas response signal value than the undoped ZnO, and the optimum amount of Fe doping is 1 at.%, while the excessive Fe doping reduced the gas sensing response. Acknowledgements This research was financially supported by the National Basic Research Program (2009CB623406), the National Natural Science Foundation of China (No. 20636020), the Natural Science Foun- dation of Jiangsu Province (No. 08KJB150009), and the Science & Technology Pillar Program of Jiangsu Province (No. BE2009679), and the Financial Foundation of State Key Laboratory of Materials- Oriented Chemical Engineering. Thank Dr. Jieshu Qian from University of Toronto for his significant contribution on paper writ- ing work. References [1] C. Klingshirn, ZnO: from basics towards applications, Phys. Status Solidi B 244 (2007) 3027–3073. [2] X.D. Wang, J.H. Song, J. Liu, Z.L. Wang, Direct-current nanogenerator driven by ultrasonic waves, Science 316 (2007) 102–105. [3] M.H. Huang, S. Mao, H. Feick, H.Q. Yan, Y.Y. Wu, H. Kind, E. Weber, R. Russo, P.D. Yang, Room-temperature ultraviolet nanowire nanolasers, Science 292 (2007) 1897–1899. [4] M. Law, L.E. Greene, J.C. Johnson, R. Saykally, P.D. Yang, Nanowire dye – sensi- tized solar cells, Nat. Mater. 4 (2005) 455–459. [5] Z.R. Dai, Z.W. Pan, Z.L. Wang, Novel nanostructures of functional oxides syn- thesized by thermal evaporation, Adv. Funct. Mater. 13 (2003) 9–24. [6] D. Calestani, M. Zha, R. Mosca, A. Zappettini, M.C. Carotta, V.D. Natale, L. Zanotti, Growth of ZnO tetrapods for nanostructure-based gas sensors, Sens. Actuators B 144 (2010) 472–478. [7] M.J.S. Spencer, I. Yarovsky, ZnO nanostructures for gas sensing: interaction of NO 2 , NO, O, and N with the ZnO [1 0 1 0] surface, J. Phys. Chem. C 114 (2010) 10881–10893. [8] W.T. Moon, Y.K. Jun, H.S. Kim, W.S. Kim, S.H. Hong, CO gas sensing properties in Pd-added ZnO sensors, J. Electroceram. 23 (2009) 196–199. [9] Q. Wan, Q.H. Li, Y.J. Chen, T.H. Wang, X.L. He, J.P. Li, C.L. Lin, Fabrication and ethanol sensing characteristics of ZnO nanowire gas sensors, Appl. Phys. Lett. 84 (2004) 3654–3656. [10] J. Li, H.Q. Fan, X.H. Jia, Multilayered ZnO nanosheets with 3D porous archi- tectures: synthesis and gas sensing application, J. Phys. Chem. C 114 (2010) 14684–14691. [11] Y.W. Heoa, D.P. Nortona, L.C. Tiena, Y. Kwona, B.S. Kangb, F. Renb, S.J. Peartona, J.R. LaRochec, ZnO nanowire growth and devices, Mater. Sci. Eng. R 47 (2004) 1–47. [12] S.C. Navale, V. Ravi, I.S. Mulla, Investigations on Ru doped ZnO: strain calcula- tions and gas sensing study, Sens. Actuators B 139 (2009) 466–470. [13] B.L. Zhu, C.S. Xie, D.W. Zeng, W.L. Song, A.H. Wang, Investigation of gas sensi- tivity of Sb-doped ZnO nanoparticles, Mater. Chem. Phys. 89 (2005) 148–153. [14] Y. Zeng, T. Zhang, L.J. Wang, M.H. Kang, H.T. Fan, R. Wang, Y. He, Enhanced toluene sensing characteristics of TiO 2 -doped flowerlike ZnO nanostructures, Sens. Actuators B 140 (2009) 73–78. [15] W.H. Zhang, W.D. Zhang, J.F. Zhou, Solvent thermal synthesis and gas-sensing properties of Fe-doped ZnO, J. Mater. Sci. 45 (2010) 209–215. [16] S.J. Han, J.W. Song, C.H. Yang, S.H. Park, J.H. Park, Y.H. Jeong, K.W. Rhie, A key to room-temperature ferromagnetism in Fe-doped ZnO: Cu, Appl. Phys. Lett. 81 (2002) 4212–4214. [17] S. Baek, J. Song, S. Lim, Improvement of the optical properties of ZnO nanorods by Fe doping, Physica B 399 (2007) 101–104. [18] L. Liu, B. Li, D.H. Yu, Y.M. Cui, X.F. Zhou, W.P. Ding, Temperature-induced solid- phase oriented rearrangement route to the fabrication of NaNbO 3 nanowires, Chem. Commun. 46 (2010) 427–429. [19] L. Liu, J.S. Qian, B. Li, Y.M. Cui, X.F. Zhou, X.F. Guo, W.P. Ding, Fabrication of rutile TiO 2 tapered nanotubes with rectangular cross-sections via anisotropic corrosion route, Chem. Commun. 46 (2010) 2402–2404. [20] Y.M. Cui, L. Liu, B. Li, X.F. Zhou, N.P. Xu, Fabrication of tunable core shell struc- tured TiO 2 mesoporous microspheres using linear polymer polyethylene glycol as templates, J. Phys. Chem. C 114 (2010) 2434–2439. 16 A. Yu et al. / Sensors and Actuators B 158 (2011) 9– 16 [21] A.K. Mishra, D. Das, Investigation on Fe-doped ZnO nanostructures prepared by a chemical route, Mater. Sci. Eng. B 171 (2010) 5–10. [22] C.Z. Wang, Z. Chen, Y. He, L.Y. Li, D. Zhang, Structure, morphology and properties of Fe-doped ZnO films prepared by facing-target magnetron sputtering system, Appl. Surf. Sci. 255 (2009) 6881–6887. [23] X.F. Zhou, H.L. Hu, Y.Q. Fan, S. Chen, W.P. Ding, N.P. Xu, Microspheric organi- zation of multilayered ZnO nanosheets with hierarchically porous structures, J. Phys. Chem. C 112 (2008) 11722–11728. [24] B. Li, S.Q. Liu, L. Liu, Y.M. Cui, X.F. Guo, X.F. Zhou, Frbrication of ZnO nanosheet based carambola-like microspheres and gas sensor properities, Chinese J. Inorg. Chem. 63 (2010) 591–595. [25] X.F. Zhou, X. Zhao, D.Y. Zhang, X.F. Guo, W.P. Ding, Y. Chen, Microsphere orga- nization of nanorods directed by PEG linear polymer, Langmuir 22 (2006) 1383–1387. [26] J. Zhang, L.D. Sun, C.S. Liao, C.H. Yan, A simple route towards tubular ZnO, Chem. Commun. 3 (2002) 262–263. [27] R.Q. Song, A.W. Xu, B. Deng, Q. Li, G.W. Chen, From layered basic zinc acetate nanobelts to hierarchical zinc oxide nanostructures and porous zinc oxide nanobelts, Adv. Funct. Mater. 17 (2007) 296–306. [28] Y. Zhang, J.Q. Xu, Q. Xiang, H. Li, Q.Y. Pan, P.C. Xu, Brush-like hierarchical ZnO nanostructures: synthesis, photoluminescence and gas sensor properties, J. Phys. Chem. C 113 (2009) 3430–3435. [29] Y.C. Zhang, L.H. Wu, H. Li, J.H. Xu, L.C. Han, B.C. Wang, Z.L. Tuo, E.Q. Xie, Influence of Fe doping on the optical property of ZnO films, J. Alloys Compd. 473 (2009) 319–322. [30] J.N. Zemel, Theoretical description of gas–film interaction on SnO x , Thin Solid Films 163 (1988) 189–202. [31] J. Li, H.Q. Fan, X.H. Jia, W.W. Yang, P.Y. Fang, Enhanced blue-green emission and ethanol sensing of Co-doped ZnO nanocrystals prepared by a solvothermal route, Appl. Phys. A 98 (2010) 537–542. [32] N. Han, L.Y. Chai, Q. Wang, Y.J. Tian, P.Y. Deng, Y.F. Chen, Evaluating the doping effect of Fe, Ti and Sn on gas sensing property of ZnO, Sens. Actuators B 147 (2010) 525–530. [33] C.Q. Ge, C.S. Xie, S.Z. Cai, Preparation and gas-sensing properties of Ce-doped ZnO thin-film sensors by dip-coating, Mater. Sci. Eng. B 137 (2007) 53–58. [34] I. Kim, A. Rothschild, T. Hyodo, H.L. Tuller, Microsphere templating as means of enhancing surface activity and gas sensitivity of CaCu 3 Ti 4 O 12 , Thin Films Nano Lett. 6 (2006) 193–198. [35] R.W.J. Scott, S.M. Yang, G. Chabanis, N. Coombs, D.E. Willams, G.A. Ozin, Tin dioxide opals and inverted opals: near-ideal microstructures for gas sensors, Adv. Mater. 13 (2001) 1468–1472. Biographies Ang Yu is pursuing his Master degree in the major of Chemistry in Nanjing University of Technology and currently interested in the self-assembly of nanomaterials such as ZnO, TiO 2 and their applications for gas sensor and solar cell. Jieshu Qian is currently a Ph.D. candidate in Polymer and Material Chemistry from the Department of Chemistry, University of Toronto. In 2005, he finished his under- graduate study in Chemistry from the Department for Intensive Instruction, Nanjing University, where at the same university; he got the Master degree of Physical Chemistry from the Department of Chemistry three years later. His current research interests include the self-assembly of polymer in solution and polymer as functional materials. Hao Pan is currently completing his M.A. degree in chemistry at Nanjing University of Technology. He devotes himself to the study of nano titania materials and their application in dye-sensitized solar cells. Yuming Cui is currently pursuing his Ph.D. degree in Physical Chemistry on Lab of Mesoscopic Chemistry, Department of Chemistry, Nanjing University. He got his Master degree of Chemical Technology in 2010 at Nanjing University of Technology. The research interesting of his study included self-assembly of nanoscale materials in the application for catalyst, gas sensing, and solar cell and so on. Meigui Xu began her research in 2009 at Nanjing University of Technology as a Master candidate and she major in the area of fabrication of nanomaterials and optical fiber sensor. Luo Tu is doing his study for Master degree in Nanjing University of Technology and his research interest is mainly about the synthesis and application of semiconductor. Qingli Chai continue his study for Master degree in Nanjing University of Technol- ogy and carrying out research in the field of photoelectrical treatment of the dye wastewater. Xingfu Zhou got his Ph.D. from Nanjing University in Physical Chemistry and was appointed a full professor in State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemistry and Chemical Engineering, Nanjing University of Technology in 2005, currently pursuing his research in the field of nano-scale semi- conductor materials and their applications for solar cell, gas sensors, wastewater treatment and so on. . preparation of Fe- doped ZnO microspheres with lotus-like morphology, which are constructed by the porous ZnO nanosheets. Then both Fe- doped ZnO and undoped ZnO as . explained by the destruction of defects in ZnO by the excess Fe doping atoms. However, there are two valence states of Fe dopant in ZnO, Fe 2+ and Fe 3+ , . l h o mepage: www.elsevier.com/locate/snb Micro-lotus constructed by Fe- doped ZnO hierarchically porous nanosheets: Preparation, characterization and gas sensing