Journal of ELECTRONIC MATERIALS DOI: 10.1007/s11664-016-5270-2 Ó 2017 The Minerals, Metals & Materials Society Ethanol-Sensing Characteristics of Nanostructured ZnO: Nanorods, Nanowires, and Porous Nanoparticles CHU THI QUY,1 CHU MANH HUNG,1 NGUYEN VAN DUY,1 NGUYEN DUC HOA,1,3 MINGZHI JIAO,2 and HUGO NGUYEN2 1.—International Training Institute for Materials Science (ITIMS), Hanoi University of Science and Technology, No 1, Dai Co Viet, Hai Ba Trung, Hanoi, Viet Nam 2.—Department of Engineering Sciences, Division of Microsystem Technology, Uppsala University, Uppsala, Sweden 3.—e-mail: ndhoa@itims.edu.vn The morphology and crystalline size of metal oxide-sensing materials are believed to have a strong influence on the performance of gas sensors In this paper, we report a comparative study on the ethanol-sensing characteristics of ZnO nanorods, nanowires, and porous nanoparticles The porous ZnO nanoparticles were prepared using a simple thermal decomposition of a sheetlike hydrozincite, whereas the nanorods and nanowires were grown by hydrothermal and chemical vapor deposition methods, respectively The morphology and crystal structure of the synthesized materials were characterized by field-emission scanning electron microscopy and x-ray diffraction Ethanol gas-sensing characteristics were systematically studied at different temperatures Our findings show that for ethanol gas-sensing applications, ZnO porous nanoparticles exhibited the best sensitivity, followed by the nanowires and nanorods Gas-sensing properties were also examined with respect to the role of crystal growth orientation, crystal size, and porosity Key words: ZnO, porous nanoparticles, nanorods, nanowires, gas sensors INTRODUCTION Ease of operation, low cost, small size, and high sensitivity are very important considerations in gas sensors for monitoring ethanol in applications such as breath analysis devices for lung cancer diagnosis,1 ethanol content in alcoholic beverages,2 fermentation processes,3 and others.4 Zinc oxide (ZnO) nanostructures of various morphologies have been extensively investigated for use in gas sensing,5,6 given their exceptional physicochemical properties and ability to sense several gases,7,8 including ethanol.9,10 ZnO nanostructured thin films,11–13 nanoparticles,14,15 nanorods,16,17 nanowires,18–20 and hollow spheres21 have all been explored for gas sensing applications For example, ZnO nanorod arrays with an exposed (0001) facet were prepared by a solution pathway for a gas sensor device, which showed a response of about 67–100 ppm ethanol at (Received October 11, 2016; accepted December 28, 2016) 370°C.22 ZnO nanowires grown by a self-catalyzed vapor–liquid–solid method exhibited a response of about 1.02–100 ppm ethanol at 300°C.23 Pristine ZnO nanowires grown by chemical vapor deposition demonstrated a response of 7–5 ppm ethanol at 350°C,24 and hollow ZnO microspheres exhibited a response of about 40–250 ppm ethanol at 350°C Gas adsorption/desorption occurs mainly on the surface of materials, and their gas-sensing characteristics are strongly dependent on their morphology, porosity, and crystalline size.25 A significant increase was reported in the sensitivity of a metal oxide when its crystalline size decreased to approximately the Debye length.13 In addition, a porous material structure has been reported to provide superior sensitivity,7,20 which has prompted much interest in the synthesis of porous ZnO for gas sensor applications.26–28 For instance, Jing et al.27 reported the fabrication and gas-sensing properties of porous ZnO nanoplates, in which the sample exhibited a response (Ra/Rg) of 9–100 ppm ethanol at a working temperature of Quy, Hung, Van Duy, Hoa, Jiao, and Nguyen 380°C Hierarchically porous ZnO architectures demonstrated a response of about 24.3–100 ppm ethanol at a working temperature of 320°C.28 ZnO nanofibers showed higher sensitivity to ethanol than nanoparticles, but less sensitivity than nanoparticle–nanofibers.29 Thus it is clear from these works that the ethanol-sensing performance of ZnO depends on the synthesis method and material crystal size, porosity, and geometry, where the effective contact area between the target gas molecules and the sensing material is important The morphology of the materials also plays a crucial role in their gas-sensing performance However, none of these studies has presented a comparative study of the ethanol-sensing performance of ZnO nanorods, nanowires, and porous nanoparticles To that end, here we report our findings on the performance of three such materials The porous ZnO nanoparticles were prepared by thermal decomposition of a precipitated sheet-like hydrozincite material, whereas the nanorods and nanowires were grown using hydrothermal and chemical vapor deposition methods, respectively Ethanolsensing characteristics of the synthesized materials were measured at different temperatures ranging from 250°C to 400°C The results demonstrate that the porous ZnO nanoparticles achieved the highest response to ethanol, followed by the nanowires and nanorods EXPERIMENTAL Synthesis of ZnO Nanorods ZnO nanorods were hydrothermally grown on a thermally oxidized silicon chip with a pair of Pt electrodes similar to the one reported in Refs 30 and 31 Briefly, the chip was placed upside down in a glass bottle filled with 100 ml equimolar amounts of 0.005 M Zn(NO3)2 and hexamethylenetetramine at room temperature The bottle was kept in an oven at 85°C for 36 h The chip was then taken out of the bottle, rinsed in deionized water several times, and finally annealed at 400°C in air for 2.5 h to stabilize the synthesized material before the gas-sensing characterization Synthesis of ZnO Nanowires ZnO nanowires were also prepared by on-chip growth through a chemical vapor deposition method as reported in Ref 32 Briefly, 0.1 g of source material (50 wt.% Zn + 50 wt.% C) was loaded on an alumina boat and placed in the center zone of a quart tube furnace A silicon chip with Pt electrodes of the same design as that used for the ZnO nanoparticle sensor was placed at a distance of cm from the alumina boat During a 30-min growth period, the furnace pressure was kept at 1.5910À1 torr by flowing a mixture of argon and oxygen at a rate of 30 standard cubic centimeters per minute (sccm) and sccm, respectively, and the temperature was set at 950°C The sample was annealed at 400°C in air for 2.5 h to stabilize the synthesized material before the gas-sensing characterization Synthesis of ZnO Nanoparticles ZnO nanoparticles were prepared through the thermal decomposition of sheet-like hydrozincite into ZnO at a temperature of 400°C The hydrozincite was first prepared by a precipitation method in which 50 mL of M zinc nitrate was added to 10 mL of M sodium carbonate with vigorous stirring at room temperature for about 20 The precipitated hydrozincite was collected by centrifugation and was washed several times with ethanol prior to drying at 60°C for 12 h A small amount of hydrozincite was loaded into an alumina board and placed in the central zone of a quartz tube furnace The thermal decomposition of the hydrozincite was performed in air by heating it to 400°C for 2.5 h.33 For the sensor fabrication, the prepared material was pasted on comb-type Pt electrodes that were fabricated on a thermally oxidized silicon substrate The morphology and crystal structure of all three synthesized materials were characterized by scanning electron microscopy (SEM; JEOL 7600F) and x-ray diffraction (XRD; D8 Advance, Bruker) Brunauer–Emmett–Teller (BET) specific surface areas of the samples were determined from the N2 adsorption/desorption isotherms (Micromeritics Gemini VII) Gas-sensing characteristics were measured at different temperatures using a Keithley instrument (model 6220), as described in detail in Ref 34 RESULTS AND DISCUSSION Figure 1a and b shows SEM images of the ZnO nanorod sensor prepared by the hydrothermal method As shown in the figure, the ZnO nanorods grew from catalytic islands to form homogenous arrays The nanorods are randomly oriented and are not perpendicular to the substrate with a continuous seed layer as the one in Ref 31 The reason is that the nanorods can freely grow out to the empty space around the small seed islands, making contact with the other nanorods from neighboring islands The ZnO nanorods grown from different islands can link to one another to form an electric path between two Pt electrodes, acting as a sensing layer for resistance measurement A high-magnification SEM image, as shown in Fig 1b, reveals that the nanorods have an average diameter of approximately 250 nm The surface of the nanorods indicates that they are quite porous and were formed from many nanocrystals, possibly because of the agglomerated nanoparticles The end of the nanorod is smaller than its body, which suggests a decrease in the Zn2+ concentration in the solution with Ethanol-Sensing Characteristics of Nanostructured ZnO: Nanorods, Nanowires, and Porous Nanoparticles Fig SEM images of (a, b) ZnO nanorods, (c, d) nanowires, and (e, f) nanoporous nanoparticles growth time The length of the nanorods reaches several micrometers, as shown in Fig 1b The SEM images of the ZnO nanowire sensors prepared by chemical vapor deposition in Fig 1c and d reveal that the nanowires grew from the Au catalytic islands and obeyed the vapor–liquid–solid mechanism As the discontinuous islands of the Au catalyst were used for the growth, the nanowires nucleated on the Au islands first, and then grew longer with prolonged growth time The ZnO nanowires have an average diameter of approximately 80 nm and a length of several micrometers The density of the nanowire mat is very low, and only the long nanowires from the neighboring islands make contact with each other The SEM images of the ZnO nanoparticles obtained by thermal decomposition of hydrozincite at 400°C are shown in Fig 1e and f The ZnO nanoparticles are irregularly shaped, with an average particle size of about 30 nm, which is much smaller than the diameter of the nanorods and nanowires The decomposition of hydrozincite also leads to weight loss, thereby forming the porous structure as shown in the SEM images The formation of nanopores increases the volume-to-mass ratio of the ZnO nanoparticles, thereby creating an advantage for their gas-sensing performance.33 The crystal structures of the ZnO nanorods, nanowires, and nanoparticles investigated by XRD are shown in Fig All XRD patterns confirm that the ZnO materials have a hexagonal crystal structure, because the typical diffraction peaks are well indexed to the profile of the JCPDS standard no 361451.30 No foreign peak is observed, indicating the formation of a single phase of ZnO in those samples However, the diffraction peaks of nanowires and nanoparticles are broader than those of the nanorods, indicating smaller crystal sizes The average crystal sizes calculated based on the Scherrer equation using the (101) peaks are approximately Quy, Hung, Van Duy, Hoa, Jiao, and Nguyen Fig (a) Nitrogen adsorption/desorption isotherm of the nanoporous ZnO nanoparticles and (b) pore size distribution Fig XRD patterns of ZnO (a) nanorods (b), nanowires, and (c) nanoporous nanoparticles 70 nm, 51 nm, and 16.1 nm for nanorods, nanowires, and nanoparticles, respectively These values are smaller than the diameters estimated from the SEM images, indicating that the synthesized materials have a polycrystalline nature and inhomogeneous characteristics The specific surface area and pore size of the nanoporous ZnO nanoparticles were evaluated by nitrogen adsorption/desorption isotherm measurement, shown in Fig The data show that the BET specific surface area and pore size of the nanoparticles is 37.89 m2/g and 3.09 nm, respectively The BET surface area of the nanoporous ZnO nanoparticles is higher than that of the porous nanoplates (15.9 m2/g).27 In our study, the amount of ZnO nanorods and nanowires was insufficient for nitrogen adsorption/desorption isotherm measurement; thus their BET surface area measurements were cited from other works as approximately and 10 m2/g, respectively.35,36 It is clear that the BET surface area of the nanoporous ZnO nanoparticles is much larger than that of the nanorods and nanowires This is very important, because a large BET surface area can provide higher sensitivity The ethanol-sensing characteristics of the fabricated sensors measured at different temperatures are shown in Fig 4a, b, c, and d As indicated in Fig 4a, the transient resistance of the ZnO nanorod sensor versus time depicts a decrease in sensor base resistance upon exposure to ethanol However, the response and recovery speeds of the sensor are low Upon exposure to 50 ppm ethanol at 400°C, the sensor resistance deceased from approximately 10.5 kX to 7.3 kX within 250 s The ZnO nanowire sensor exhibited better sensing characteristics, as shown in Fig 4b Upon exposure to 50 ppm of ethanol at 400°C, the sensor resistance decreased from 72.1 kX to 42.9 kX within 26 s This sensor was also fully recovered after stopping the flow of ethanol The nanoparticle sensor, on the other hand, exhibited exceptional ethanol-sensing characteristics Upon exposure to 50 ppm of ethanol at 400°C, the sensor resistance decreased from 6.5 MX to 795 kX in only 16 s, although it did not appear to reach full saturation Comparative results of the ethanol response of the various sensors measured at Ethanol-Sensing Characteristics of Nanostructured ZnO: Nanorods, Nanowires, and Porous Nanoparticles Fig Transient resistance versus time upon exposure to different ethanol concentrations of fabricated sensors: (a) nanorods, (b) nanowires, (c) nanoparticles, and (d) sensor response as a function of ethanol concentration 400°C are shown in Fig 4d The response values for nanowires and nanorods are roughly the same, and these values are extremely low The ethanol response of the porous nanoparticle sensor to 50 ppm is 11, which is approximately 6.12- and 6.23-fold higher than that of the nanowires and nanorods, respectively Compared to the reported results of other studies, the ethanol response of nanoporous ZnO nanoparticles is comparable to that of the porous architectures,26,28 but it is higher than that of the porous nanoplates.27 The fabricated nanoporous ZnO nanoparticles also showed much higher ethanol response than did the Sb-doped ZnO nanowires23 or a-Fe2O3 hollow balls,37 which suggests that the nanoparticles are superior to the other two with regard to ethanol-sensing performance However, the selectivity and long-term stability of the device should be evaluated for practical applications One can argue that the three materials in this study were synthesized using different methods and from different precursors, and the means of bringing them into the sensor surface also differed; thus a comparison of their ethanol-sensing performance might be questionable However, the XRD analysis showed that all the materials were pure and of hexagonal polycrystalline type; thus, they are comparable as sensing materials As sensors, all the materials used substrate made from thermally oxidized silicon wafer One aspect that makes the direct comparison difficult is that the ZnO nanorods and nanowires were synthesized on-chip with discrete seed islands, and therefore the materials that were chemically bonded to the electrodes should be very stable and favorable for resistive response measurement,32 while the nanoparticle sheet was transferred and pasted between the comb-type Pt electrodes However, the ethanol-sensing performance of the nanoparticle sensor was much better than that of the other two Therefore, the different sensing characteristics of the fabricated sensors can be attributed solely to the differences in the crystalline size and volume-to-mass ratio (i.e., porosity) of the materials The Debye length of ZnO is estimated to be approximately 7.4 nm,38 which is about half the value of the nanoparticle size (16 nm) and far smaller than those of the nanowire (51 nm) and nanorod (70 nm) Thus the nanoparticles exhibit the highest ethanol response, followed by nanowires and nanorods In addition to crystal size, the preferred growth direction and the exposed facet of the materials can influence the gas-sensing performance It was reported that ZnO with an exposed (0001) facet showed a higher response than the others.22 In this study, XRD analysis reveals that the ZnO nanorods and nanoparticles have similar patterns, where the intensity of the (002) peak is comparable to that of (100) peak However, Quy, Hung, Van Duy, Hoa, Jiao, and Nguyen in the nanowires, the intensity of the (002) peak is much lower than that of the (100) peak Note that the nanowires are smaller in diameter than the nanorods but have a comparable response to ethanol; thus the lower response of the nanowires may be due to their lower (002) exposed facets The high ethanol response of the nanoporous ZnO nanoparticles is attributable mainly to the smaller crystal size, larger BET surface area, and effective exposed facets CONCLUSION ZnO nanorods, nanowires, and porous nanoparticles were successfully synthesized, and their crystal sizes were calculated The ethanol-sensing characteristics of the materials were also measured Comparisons of these materials showed that the porous nanoparticles had the best ethanol-sensing performance (i.e., highest sensing response and fastest response and recovery times), followed by nanowires and nanorods, in accordance with the increasing order of their crystal sizes and volumeto-mass ratios Thus, the high ethanol-sensing performance of the synthesized ZnO materials depends significantly on their preferred growth orientation, small crystal size, and porosity ACKNOWLEDGEMENTS This study was supported by Hanoi University of Science and Technology (Grant No T2016-LN-16) and Uppsala University REFERENCES R Capuano, M Santonico, G Pennazza, S Ghezzi, E Martinelli, C Roscioni, G Lucantoni, G Galluccio, R Paolesse, C Di Natale, and A D’Amico, Sci Rep 5, 16491 (2015) M Akamatsu, T Mori, K Okamoto, H Komatsu, K Kumagai, S Shiratori, M Yamamura, T Nabeshima, H Sakai, M Abe, J.P Hill, K Ariga, and A.C.S Appl, Mater Interfaces 7, 6189 (2015) X.-Y Zou, F Luo, R Xie, L.-P Zhang, X.-J Ju, W Wang, Z Liu, and L.-Y Chu, Anal Methods 8, 4028 (2016) Z Qin, Y Liu, W Chen, Y Wu, and S Li, Mater Sci Semicond Process 52, 75 (2016) M Akermi, N Sakly, R Ben Chaabane, and H Ben Ouada, Mater Sci Semicond Process 16, 807 (2013) A Wei, L Pan, and W Huang, Mater Sci Eng B 176, 1409 (2011) H Van Han, N.D Hoa, P Van Tong, H Nguyen, and N Van Hieu, Mater Lett 94, 41 (2013) J Liu, Z Guo, F Meng, Y Jia, T Luo, M Li, and J Liu, Cryst Growth Des 9, 1716 (2009) C Lin, S.-J Chang, W.-S Chen, and T.-J Hsueh, RSC Adv 6, 11146 (2016) 10 Z Wang, Z Tian, D Han, F Gu, and A.C.S Appl, Mater Interfaces 8, 5466 (2016) 11 J Eriksson, V Khranovskyy, F Soăderlind, P.-O Kaăll, R Yakimova, and A.L Spetz, Sens Actuators B Chem 137, 94 (2009) 12 H Gong, J.Q Hu, J.H Wang, C.H Ong, and F.R Zhu, Sens Actuators B Chem 115, 247 (2006) 13 T.T Trinh, N.H Tu, H.H Le, K.Y Ryu, K.B Le, K Pillai, and J Yi, Sens Actuators B Chem 152, 73 (2011) 14 Z.Q Zheng, J.D Yao, B Wang, and G.W Yang, Sci Rep 5, 11070 (2015) 15 A.S Kazemi, R Afzalzadeh, and M Abadyan, J Mater Sci Technol 29, 393 (2013) 16 K Mirabbaszadeh and M Mehrabian, Phys Scr 85, 35701 (2012) 17 H Ahn, J.-H Park, S.-B Kim, S.H Jee, Y.S Yoon, and D.-J Kim, Electrochem Solid State Lett 13, J125 (2010) 18 P Wang, Y Fu, B Yu, Y Zhao, L Xing, and X Xue, J Mater Chem A 3, 3529 (2015) 19 Q Wan, Q.H Li, Y.J Chen, T.H Wang, X.L He, J.P Li, and C.L Lin, Appl Phys Lett 84, 3654 (2004) 20 T.J Hsueh, C.L Hsu, S.J Chang, and I.C Chen, Sens Actuators B Chem 126, 473 (2007) 21 W Wang, Y Tian, X Wang, H He, Y Xu, C He, and X Li, J Mater Sci 48, 3232 (2013) 22 S Tian, F Yang, D Zeng, and C Xie, J Phys Chem C 116, 10586 (2012) 23 C.-L Hsu, K.-C Chen, T.-Y Tsai, and T.-J Hsueh, Sens Actuators B Chem 182, 190 (2013) 24 C.-H Kwak, H.-S Woo, F Abdel-Hady, A.A Wazzan, and J.-H Lee, Sens Actuators B Chem 223, 527 (2016) 25 N.D Hoa, N Van Duy, S.A El-Safty, and N Van Hieu, J Nanomater 2015, (2015) 26 J Li, H Fan, and X Jia, J Phys Chem C 114, 14684 (2010) 27 Z Jing and J Zhan, Adv Mater 20, 4547 (2008) 28 J Huang, Y Wu, C Gu, M Zhai, Y Sun, and J Liu, Sens Actuators B Chem 155, 126 (2011) 29 P.-P Wang, Q Qi, R.-F Xuan, J Zhao, L.-J Zhou, and G.-D Li, RSC Adv 3, 19853 (2013) 30 M Jiao, N.V Chien, N Van Duy, N.D Hoa, N Van Hieu, K Hjort, and H Nguyen, Mater Lett 169, 231 (2016) 31 N.D Khoang, H.S Hong, D.D Trung, N Van Duy, N.D Hoa, D.D Thinh, and N Van Hieu, Sens Actuators B Chem 181, 529 (2013) 32 H Nguyen, C.T Quy, N.D Hoa, N.T Lam, N Van Duy, V Van Quang, and N Van Hieu, Sens Actuators B Chem 193, 888 (2014) 33 N.T Phuong Nhung, P Van Tong, C.M Hung, N Van Duy, N.V Chien, N Van Vinh, N.T Tuyen, and N.D Hoa, RSC Adv 6, 64215 (2016) 34 P Van Tong, N.D Hoa, N Van Duy, and N Van Hieu, RSC Adv 5, 25204 (2015) 35 J Cheng and K.M Poduska, ECS J Solid State Sci Technol 3, P133 (2014) 36 W.-H Chiu, C.-H Lee, H.-M Cheng, H.-F Lin, S.-C Liao, J.-M Wu, and W.-F Hsieh, Energy Environ Sci 2, 694 (2009) 37 C.M Hung, N.D Hoa, N Van Duy, N Van Toan, D.T.T Le, and N Van Hieu, J Sci Adv Mater Devices 1, 45 (2016) 38 J Deng, Q Fu, W Luo, X Tong, J Xiong, Y Hu, and Z Zheng, Sens Actuators B Chem 224, 153 (2016) ... with Ethanol-Sensing Characteristics of Nanostructured ZnO: Nanorods, Nanowires, and Porous Nanoparticles Fig SEM images of (a, b) ZnO nanorods, (c, d) nanowires, and (e, f) nanoporous nanoparticles. .. isotherm of the nanoporous ZnO nanoparticles and (b) pore size distribution Fig XRD patterns of ZnO (a) nanorods (b), nanowires, and (c) nanoporous nanoparticles 70 nm, 51 nm, and 16.1 nm for nanorods,. .. saturation Comparative results of the ethanol response of the various sensors measured at Ethanol-Sensing Characteristics of Nanostructured ZnO: Nanorods, Nanowires, and Porous Nanoparticles Fig Transient