- Email: [email protected]
Calcination-temperature-dependent gas-sensing properties of mesoporous α-Fe2O3 nanowires as ethanol sensors
Accepted Manuscript Calcination-temperature-dependent gas-sensing properties of mesoporous α-Fe2O3 nanowires as ethanol sensors X.Q. Li, D.P. Li, J.C. Xu, Y.B. Han, H.X. Jin, B. Hong, H.L. Ge, X.Q. Wang PII:
S1293-2558(17)30037-7
DOI:
10.1016/j.solidstatesciences.2017.05.006
Reference:
SSSCIE 5498
To appear in:
Solid State Sciences
Received Date: 11 January 2017 Revised Date:
11 May 2017
Accepted Date: 13 May 2017
Please cite this article as: X.Q. Li, D.P. Li, J.C. Xu, Y.B. Han, H.X. Jin, B. Hong, H.L. Ge, X.Q. Wang, Calcination-temperature-dependent gas-sensing properties of mesoporous α-Fe2O3 nanowires as ethanol sensors, Solid State Sciences (2017), doi: 10.1016/j.solidstatesciences.2017.05.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Calcination-temperature-dependent gas-sensing properties of mesoporous α-Fe2O3 nanowires as ethanol sensors X.Q. Li, D.P. Li, J.C, Xu,Y.B. Han, H.X. Jin, B. Hong, H.L. Ge and X.Q. Wang*
RI PT
College of Materials Science and Engineering, China Jiliang University, Hangzhou, China, 310018
Abstract:
SC
The mesoporous α-Fe2O3 nanowires (NWs) were successfully synthesized by changing the
M AN U
calcination temperature from 550 to 750 °C (marked NWs-550, NWs-650 and NWs-750) via using SBA-15 silica as the hard templates with the nanocasting method. The characterization results indicated that the bandgap of the as-prepared samples hardly changed and the high BET surface areas changed a little with the calcination temperature from 550 to 750 °C. Mesoporous α-Fe2O3 NWs had been found
TE D
to possess the remarkable gas-sensing performance to ethanol gas. The gas-sensing behavior indicated that α-Fe2O3 NWs-650 exhibited the higher response than that of α-Fe2O3 NWs-550 and α-Fe2O3 NWs-750. The calcination-temperature-dependent gas-sensing properties were mainly attributed to the
EP
competition of surface defects and body defects by the crystallization temperature. The lower
AC C
calcination temperature could create more surface defects to improve the gas-sensing response, while the higher temperature would reduce the body defect and make the charge carriers transport easily. As the result, the suitable calcination temperature was desired to optimize the defects of nanostructures to improve the gas sensitivity.
Keywords: nanowires; nanocasting; calcination-temperature-dependent; gas-sensing properties
*Corresponding Authors. Tel: +86-571-86676157; Fax: +86-571-28911371 Email address: [email protected] (Xinqing Wang)
ACCEPTED MANUSCRIPT 1. Introduction As an environmentally friendly n-type wide bandgap semiconductor, hematite (α-Fe2O3) had been extensively investigated in catalysts, gas sensors, optical devices,
RI PT
lithium-ion batteries and electromagnetic devices [1-5]. Various nanostructures of α-Fe2O3 had been synthesized by the different methods, such as nanospheres [6-7], nanowires [8], nanorods [9], nanoribbons [10], nanotubes [11], hollow nanostructures
SC
[12] and nanoplates [13-16]. Among the above α-Fe2O3 nanostructures, mesoporous
M AN U
α-Fe2O3-based nanostructures should be one of novel candidates for gas-sensor application due to the huge specific surface area. Usually, the working mechanism of the metal oxide based gas sensor was based on the electrical conductivity, resulting from the surface reactions such as oxidation or reduction. Thus, it was very important
TE D
to control the change of the electrical conductivity after the gas adsorption. The huge surface area provided a vast number of surface defects to form the sites for the gas sorption and desorption, improving the gas-sensing sensitivity. Furthermore, the body
EP
defects could reflect the charge carriers and greatly affected the electrical conductivity
AC C
in the reducing gas [17-18].
Mesoporous nanostructures with the huge specific surface area provided a vast
number of surface defects to form the sites of gas sorption and desorption, which was suitable to act as the candidates for gas sensor. Moreover, openly porous structures would prevent the diffusion of carriers [19]. Thus, mesoporous structures could cause the enhancement of both the gas sensitivity and response rate [20]. Some efforts had been engaged in the preparation mesoporous α-Fe2O3 nanostructures for gas sensors 2
ACCEPTED MANUSCRIPT to improve the surface area and surface defects. Li et al. reported the synthesis of mesoporous α-Fe2O3 nanostructures through the solid-state chemical reaction, and exhibited the enhanced gas-sensing performance to xylene [21]. Sun et al. reported the
RI PT
synthesis of mesoporous α-Fe2O3 nanostructures through the soft template method using the triblock copolymer surfactant F127 as the template, which indicated the high gas sensitivity toward acetic acid and ethanol gas [22]. Owing to the good
SC
uniformity and high controllability for target products, nanocasting method was the
M AN U
first choice for the synthesis of the controllable nanostructures in many preparation methods [23]. SBA-15 silica with the well-ordered hexagonal straight mesoporous structures was the good candidates of hard templates for nanowires [24]. Previously, mesoporous α-Fe2O3 nanowires were synthesized using SBA-15 as hard templates
TE D
with the nanocasting method and presented the higher response to ethanol [25]. The high surface area could be obtained by the synthesis of mesoporous nanostructures, while the defects of nanostructures decreased with the increasing
EP
calcination temperature [26]. It was well known that surface defects were good for the
AC C
oxygen absorption/desorption and the body defects were harmful to the gas sensors. And it was necessary to optimize and adjust the surface defects and body defects [27]. In this study, α-Fe2O3 NWs were synthesized by using mesoporous SBA-15 silica as the hard templates with the nanocasting method, and the as-prepared samples were calcined under the temperature between 550-750 °C. The microstructure and bandgap of mesoporous α-Fe2O3 NWs was characterized with X-ray diffraction (XRD), transmission electron microscopy (TEM), nitrogen adsorption/desorption isotherm 3
ACCEPTED MANUSCRIPT and UV-vis spectrum, respectively. Furthermore, the gas-sensing properties of α-Fe2O3 NWs based gas sensors were measured and discussed in detail, and the relationship of the gas response of α-Fe2O3 NWs based gas sensors with the
RI PT
calcination temperature was concluded.
2. Experimental section
SC
All chemicals were of analytical grade and were used as purchased without any
M AN U
further purification. Ordered mesoporous silica SBA-15 was prepared by the method described previously [28]. For the synthesis of α-Fe2O3 nanowires, ferric nitrate and SBA-15 powders (the atomic ratio of Si:Fe=2:1) were dissolved and suspended in ethanol, and then hexane was added until the fine powder formed. The powder
TE D
samples were calcined at 550, 650 and 750 °C for 6 h, respectively, and was marked NWs-550, NWs-650 and NWs-750. Then 2 M NaOH aqueous solution was added to the above powder to remove the silica template and all samples were filtered and
EP
washed with deionized water and ethanol, and then dried at 80 °C for 4 h.
AC C
The morphology of all samples was examined by means of transmission electron microscopy (TEM: JME-1200EX). The crystal structure of the as-prepared products was determined by X-ray diffraction (XRD: XD-5A, Cu Kα, wavelength 0.154 nm, step 0.02°). BaSO4 powder was used as the substrate material to obtain the UV-vis spectrum with a UV3600 spectrophotometer to deduce the bandgap of both samples. Nitrogen physisorption experiments were measured at 77 K on a Micrometrics ASAP 2020 surface area and porosity analyzer. The BET (Brunauer-Emmett-Teller) surface 4
ACCEPTED MANUSCRIPT areas were calculated from the relative pressure range 0.06 to 0.2. To measure the sensing properties of the α-Fe2O3 NWs-550, NWs-650 and NWs-750, the sensors were prepared in the following way: the same amount of each
RI PT
sample was mixed with several drops of deionized water in an agate mortar to form a homogeneous paste, which was then deposited on the alumina ceramic tube assembled with platinum wire electrodes for electrical contacts. The prepared gas
SC
sensing device was aged at 300 °C for 5 days to improve the stability of the sensor.
M AN U
Then a Ni-Cr alloy wire was passed through the alumina ceramic tube and used as a heater by tuning the heating voltage. The gas sensing tests were then performed on a WS-60A gas sensing measurement system, which was a static system using atmospheric air as the interference gas. The relative humidity was about 65%. The
TE D
response (S=Ra/Rg) of a sensor was defined as the ratio of sensor resistance in air (Ra) to that in a target gas (Rg). The response and recovery times were defined as the times required for a change in the resistance to reach 90% of the equilibrium value after the
AC C
EP
detected gas was injected and removed, respectively [29].
3. Results and discussion The microstructures of SBA-15, α-Fe2O3 NWs-550, NWs-650 and NWs-750 were
characterized by TEM in Fig. 1. The TEM photo of SBA-15 in Fig. 1(a) indicated that SBA-15 presented a well-ordered array of mesopores with the pore-size of about 8 nm. After the implanting of α-Fe2O3 in mesopores and removing SBA-15 silica, α-Fe2O3 NWs in mesopores could be obtained. It could be seen in Fig. 1(b), (c) and (d) that all 5
ACCEPTED MANUSCRIPT α-Fe2O3 NWs existed in bundles with tens of nanowires. The α-Fe2O3 NWs maintained the straight mesoporous microstructures with the same diameter of about 7 nm. It should be mentioned that there was little difference between the α-Fe2O3 NWs
RI PT
for the same templates and the synthesized conditions except for the calcination temperature. In this way, the influence of the calcination temperature on the microstructure and the gas-sensing properties could be concluded.
SC
XRD was introduced to characterize the phase structure and crystallization of
M AN U
mesoporous α-Fe2O3 NWs calcined at 550, 650 and 750 °C in Fig. 2. The diffraction peaks of the α-Fe2O3 with the phase structure R-3c(167) were obviously observed. Compared three samples, the intensity of the α-Fe2O3 diffraction peaks increased a little from 140 to 180 a.u. with the temperature, indicating the improved
TE D
crystallization with the calcination temperature. Usually, the higher calcination temperature leaded the less surface and body defects of nanostructures. The decreasing body defects decreased the initial resistance, while the decreasing surface
EP
defects also decreased the resistance after the gas adsorption. Therefore, it was
AC C
necessary to optimize the surface defects and body defects by the surface area and calcination temperature. The nitrogen physisorption isotherms of the mesoporous α-Fe2O3 NWs-550,
NWs-650 and NWs-750 were shown in Fig. 3(a). The isotherms of α-Fe2O3 NWs-550, NWs-650 and NWs-750 all presented the characteristic of the type IV with the type H1 hysteresis loop, indicating the feature of mesoporous-structure according to the IUPAC. This mesoporous-structure could lead to the higher surface area for α-Fe2O3 6
ACCEPTED MANUSCRIPT NWs, which was good for the increase of surface defects. The BET (Brunauer-Emmett-Teller) surface areas of α-Fe2O3 NWs-550, NWs-650 and NWs-750 were calculated to be 95.23, 105.85 and 96.10 m2/g, respectively (Table 1).
RI PT
The α-Fe2O3 NWs-650 presented the highest surface area, which led to the more absorbed oxygen on the surface of nanowires and the larger change of surface resistance. The suitable calcination temperature led to the better mesoporous-structure,
SC
while the higher calcination temperature resulted in the collapse of micropores
M AN U
between the mesopores to some extent [28], which led to the low surface area of α-Fe2O3 NWs-750. The higher surface area should be attributed to the mesoporous-structure of α-Fe2O3 NWs. Fig. 3(b) showed the highly-centralized pore-size distribution of the α-Fe2O3 NWs-550，NWs-650 and NWs-750, indicating
TE D
the average pore diameter of α-Fe2O3 NWs-550，NWs-650 and NWs-750 were 11.38 nm, 9.86nm, 11.06nm, respectively. The corresponding pore volume of the α-Fe2O3 NWs-550，NWs-650 and NWs-750 with the value of 0.3734, 0.3715 and 0.3594
EP
cm3/g were calculated by the BJH method, and decreased with the increasing
AC C
calcination temperature. Furthermore, it could be concluded that the calcination temperature affected the BET surface area of α-Fe2O3 NWs a little. The Fig. 4 (a) showed the typical absorption spectrum of α-Fe2O3 NWs-550,
NWs-650 and NWs-750. Eg was the bandgap and could be calculated by the Tauc equation [30]: (αhv)1/n=A(hv - Eg), where α was the absorption coefficient, h was the Plank constant, v was the photon frequency and n was the electronic transition parameter of 1/2 for direct bandgap semiconductor. Since α was proportional to 7
ACCEPTED MANUSCRIPT absorbance (A) according to Lambert-Beer law, the energy intercept of the curve of (αhv)2 versus hv could give Eg when the linear region was extrapolated to the zero ordinate. The Fig. 4 (b) presented the relationship of (αhv)2 versus hv for α-Fe2O3
RI PT
NWs-550, NWs-650 and NWs-750. The Eg of α-Fe2O3 NWs-550, NWs-650 and NWs-750 (with the value of 2.01, 1.99 and 2.02 eV) were calculated by extrapolating the linear portion of the plot of (αhv)2 versus hv to α=0. The α-Fe2O3 NWs-650
SC
presented the narrowest bandgap, which was good for the electron transition from
M AN U
valence band to conduction band. As the result, the ground state resistance in the target gas decreased to some extent and the sensitivity of gas sensor was improved. From the above analysis, α-Fe2O3 NWs-550 and NWs-750 showed the similar area surface and bandgap, thus it was possible to discuss the influence of the calcination
TE D
temperature on the gas-sensing properties of α-Fe2O3 NWs.
Furthermore, the influence of the calcination temperature on the gas-sensing properties was also investigated as the following. Firstly, the response of α-Fe2O3
EP
NWs-550, NWs-650 and NWs-750 sensors to ethanol gas with 100 ppm was
AC C
measured at different operating temperatures from 240 to 370 °C in Fig. 5 (a). The response of α-Fe2O3 NWs-650 based sensor presented a rapid increase up to 37.57 at 300 °C with the operating temperature and then decreased, which is much higher than those reports with the α-Fe2O3 nanotubes, porous hematite, hematite spheres, nanocrystals, nanoparticles and nanoropes (Table 2
[21,31-38]
). And the α-Fe2O3
NWs-550 and NWs-750 sensors displayed the similar response tendency and their response reached the maximum value of 35.26 and 29.64 at the same operating 8
ACCEPTED MANUSCRIPT temperature of 300 °C, respectively. This phenomenon was commonly observed for many semiconducting metal oxide based sensors and could be explained by the balance between the speed of chemical reaction and gas diffusion [39-40]. At a low
RI PT
operating temperature, the low response could be obtained for the target gas ethanol and did not have enough thermal energy to react with the surface electron of α-Fe2O3 NWs. With the operating temperature increasing, the thermal energy obtained was
SC
high enough to overcome the activation energy of the surface reaction [41]. Moreover,
M AN U
the reduction after the maximum was due to the low gas adsorption ability of the gas molecule at high temperature. Therefore, it was concluded that the optimal operating temperature for α-Fe2O3 NWs based sensors was 300 °C and was used in the following gas-sensing tests. Mesoporous α-Fe2O3 nanowires were studied previously
TE D
and showed the different response to ethanol, which was attributed to the different aged days [25]. The response-recovery time to 100 ppm ethanol at 300 °C was given in Fig. 5 (b), (c) and (d). The response and recovery times of α-Fe2O3 NWs-550,
EP
NWs-650 and NWs-750 sensors were only 6 s, 8 s, 8 s and 29 s, 36 s, 30 s (Table 1),
AC C
resulting from the different diffusion speed of air and ethanol gas. Fig. 6 showed the responses of α-Fe2O3 NWs-550, NWs-650 and NWs-750
based sensors to different concentration of ethanol in the range of 5-1000 ppm at 300 °C. It was seen that the response of all samples increased with the ethanol concentration. The response of the mesoporous α-Fe2O3 NWs-650 sensor increased from 3.37 (5 ppm) to 77.36 (1000 ppm), while that of α-Fe2O3 NWs-550 and NWs-750 increased from 3.27 and 3.26 to 62.1 and 56.43, respectively. Due to the 9
ACCEPTED MANUSCRIPT higher surface area and narrowest bandgap, the response of the α-Fe2O3 NWs-650 based sensors was larger than those of α-Fe2O3 NWs-550 and NWs-750 sensors. With the similar area surface and bandgap, the sensitivity of α-Fe2O3 NWs-550 was a little
RI PT
higher than that of α-Fe2O3 NWs-750 for the lower calcination temperature. Owing to surface reactions such as oxidation or reduction, the working mechanism of the metal oxide gas sensor was based on the change of the electrical
SC
resistance in air and ethanol gas. In order to increase the sensitivity (S = Rair/Rgas), one
M AN U
way was to decrease the target gas resistance (ground state resistance) and another was to increase the air resistance. For n-type semiconductor, the higher crystallization degree with the less body defects resulted in the decrease of core resistance in target gas. The lower crystallization degree with the more surface defects resulted in the
TE D
increase of shell resistance in air. As the result, the suitable calcination temperature was desired to optimize the defects of nanostructures to improve the gas sensitivity. These surface reactions were directly dependent on the oxygen vacancies at the
EP
surface, which required a higher specific surface area with more active centers and
AC C
defects at the surface. Usually, the surface and body defects of α-Fe2O3 NWs decreased with the increasing calcination temperature. The decreasing surface defects would decrease the resistance in air, while the decreasing body defects decreased the initial resistance. In this way, the calcination temperature greatly affected the resistance and gas-sensing properties of α-Fe2O3 NWs. Thus, it was concluded that the resistance change between air and ethanol gas was dominated by electronic depletion layer at the surface of nanowires. And the sensitivity was improved with the lower 10
ACCEPTED MANUSCRIPT calcination temperature. Finally, the selectivity for different gases was also discussed with the concentration of 100 ppm at 300 °C in Fig. 7, and the gases included ethanol, acetone,
RI PT
methyl alcohol, normal hexane and benzene. The sensitivities of mesoporous α-Fe2O3 NWs-650 based sensors were always larger than those of α-Fe2O3 NWs-550 and NWs-750 sensor to different gases. The response of the α-Fe2O3 NWs-650 was 41.63,
SC
34.07, 7.95, 1.62 and 1.55 to ethanol, acetone, methyl alcohol, normal hexane and
M AN U
benzene, respectively. The response of α-Fe2O3 NWs-550 based sensor was 36.55, 31.36, 6.53, 1.57 and 1.31 and that of α-Fe2O3 NWs-750 sensor was 34.76, 29.82, 6.36, 1.33 and 1.28 respectively. Furthermore, the response of all samples to ethanol and acetone was much larger than those to methyl alcohol, normal hexane and
TE D
benzene. It can be seen that all samples showed the best response to ethanol gas. This should be attributed to the difference interactions between α-Fe2O3 molecules and target gas molecules. The absorbed oxygen on the surface of α-Fe2O3 NWs was
EP
preferred to react with ethanol at 300 °C. Therefore, the larger number of electrons
AC C
released back to conductive band could lead to the larger change of resistance.
4. Conclusions
Mesoporous α-Fe2O3 NWs were successfully synthesized by using SBA-15 silica as the hard templates with the nanocasting method under the calcination temperature between 550-750 °C. The BET surface areas of mesoporous α-Fe2O3 NWs-550, NWs-650 and NWs-750 exhibited the high surface areas. With increasing the 11
ACCEPTED MANUSCRIPT calcination temperature from 550 to 750 °C, the bandgap hardly changed. All results showed that all the samples exhibited high responses to ethanol gas and the lowest detection concentration at 5 ppm. Moreover, α-Fe2O3 NWs-650 exhibited the best
RI PT
gas-sensing performance for the high surface area and narrow bandgap. From the gas-sensing analysis of α-Fe2O3 NWs-550 and NWs-750 with the similar area surface and bandgap, it was concluded that the calcination temperature greatly affected the
SC
gas-sensing behavior of nanostructures. And the sensitivity could be improved at the
M AN U
low calcination temperature.
Acknowledgments
TE D
The research was funded by National Natural Science Foundation of China (Nos. 51202235) and Foundation of Science and Technology Department of Zhejiang Province (Nos.2017C33078
AC C
References
EP
and Nos. LY15B010004)
[1] Q.Y. Hao, S. Liu, X.M. Yin, Y.G. Wang, Q.H. Li, T.H. Wang, Solid State Sci. 12(2010)2125-2129
[2] J. Xu, F. Huang, Y. L. Yu, A. P. Yang, Y. S. Wang, Cryst. Eng. Commun. 13 (2011) 4873-4877.
[3] J. R. Huang, M. Yang, C. P. Gu, M. H. Zhai, Y. F. Sun, J. H. Liu, Mater. Res. Bull. 46 (2011) 1211-1218.
[4] M. Mahmoudi, A. Simchi, M. Imani J. Iran. Chem. Soc.7 (2010)21-27.
[5] T. E. García, M. Valvo, U. Lafont, C. Locati, D. Munao, E. M. Kelder J. Power Sources 196 (2011) 6425-6432
[6] L. L. Wang, Z. Lou, J. N. Deng, T. Zhang,ACS. Appl. Mater. Inter.7(2015) 13098-13104
12
ACCEPTED MANUSCRIPT [7] H. J. Wu, G. L. Wu, L. D. Wang, Powder. Technol. 269 (2015) 443-451.
[8] C. Wang, Y. Q. Wang, X. H. Liu, H. W. Yang, J. R. Sun, L. Yuan, G. W. Zhou, F. Rosei, Structure versus
properties in α-Fe2O3 nanowires and nanoblades, nanotechnology 27 (2016)035702.
RI PT
[9] D. Patil, V. Patil, P. Patil, Sensor. Actuat. B-Chem. 152 (2011) 299-306.
[10] D. Sarkar, M. Mandal, K. Mandal, ACS Appl. Mater. Inter. 5 (2013) 11995-12004.
[11] D. H. Kim, Y. S. Shim, J. M. Jeon, H. Y. Jeong, S. S. Park, Y. W. Kim, J. S. Kim, J. H. Lee, H. W. Jang, ACS
SC
Appl. Mater. Inter. 6 (2014) 14779-14784.
M AN U
[12] K. Kwon, H. Lim, Y. Sun, K. Suh, J. Phys. Chem. C 118 (2014) 2897-2907.
[13] R. Liu, Y. Jiang, Q. Lu, W. Du, F. Gao, CrystEngComm 15 (2013) 443-446.
[14] F. Lu, Q. Wu, X. Yang, L. Chen, J. Cai, C. Liang, M. Wu, P. Shen, Phys. Chem. Chem. Phys. 15 (2013)
9768-9774.
TE D
[15] J. Cai, S. Chen, M. Ji, J. Hu, Y. Ma, L. Qi, CrystEngComm 16 (2014) 1553-1559.
[16] H. Liang, W. Chen, X. Jiang, X. Xu, B. Xu, Z. Wang, J. Mater. Chem. A 2 (2014) 4340-4346.
[17] D.B. Mahadik, Y.K. Lee, C.-S. Park, H.-Y. Chung, M.-H. Hong, Hae-N.-R. Jung, W. Han, H.-H. Park, Solid
EP
State Sci. 50(2015) 1-8
AC C
[18]A. Maurya, P. Chauhan, S. K. Mishra, R. K. Srivastava, J. Alloy. Compd. 509 (2011) 8433-8440.
[19]G. X. Zhu, L. J. Guo, X. P. Shen, Z. Y. Ji, K. M. Chen, H. Zhou, Sensor. Actuat. B-Chem. 220 (2015)
977-985.
[20] G. X. Zhu, L. J. Guo, X. P. Shen,Z. Y. Ji,, K. M. Chen, H. Zhou, Sensor. Actuat. B-Chem. 220 (2015)
977-985.
[21] Y. Z. Li, Y. L. Cao, D. Z. Jia, Y. Wang, J. Xie, Sensor. Actuat. B-Chem. 198 (2014) 360-365.
[22] B. Sun, J. Horvat, H.S. Kim, W. Kim, J. Ahn, G. Wang, J. Phys. Chem. C 114 (2010) 18753-18761.
13
ACCEPTED MANUSCRIPT [23] M. Sakeye, S. Ziller, H. Amenitsch, M. Lindén and J. Smått,
J. Phys. Chem. C 120 (2016) 1854-1862.
[24] D.P. Li, Y. Zhan, J.C. Xu, H.X. Jin, D.F. Jin, B. Hong, X.L. Peng, P.F. Wang, H.L. Ge, X.Q. Wang,
Physica E 84 (2016) 395–400
Peng, H. L. Ge and X. Q. Wang,
Nanotechnology 27 (2016) 185702-185707.
RI PT
[25] D. P. Li, B. B. Zhang, J. C. Xu, Y. B. Han, H. X. Jin, D. F. Jin, X. L.
[26] L.W. Mo, M. Deng, M. S. Tang, Key Engineering Materials, 400-402 (2009) 169-174
SC
[27] S. Kavianpour, A. Yavari, Comp. Mater. Sci. 44 (2009) 1296-1306
M AN U
[28] X. Q. Wang, H. L. Ge, H. X. Jin, Y. Cui,Micropor. Mesopor. Mat.86 (2005) 335-340.
[29] L. Xu, B. Dong, Y. Wang, X. Bai, Q. Liu, H. W. Song, Sensor. Actuat. B-Chem. 147(2010) 531-538.
[30] S. K. Maji, N. Mukherjee, A. Mondal, B. Adhikary, Polyhedron 33 (2012) 145-149
[31]Y. L. Cao, H. Y. Luo, D. Z. Jia, Sensor. Actuat. B-Chem. 176 (2013) 618-624.
TE D
[32] H. J. Song, X. H. Jia, X. Q. Zhang, J. Mater. Chem. 22 (2012) 22699-22705.
[33] H. J. Song, X. H. Jia, H. Qi, X. F. Yang, H. Tang, C. Y. Min, J. Mater. Chem. 22 (2012) 3508-3516.
[34] P. Sun, W. N. Wang, Y. P. Liu, Y. F. Sun, J. Ma, G. Y. Lu, Sensor. Actuat. B-Chem. 173 (2012) 52-57.
EP
[35] P. Sun, L. You, D. W. Wang, Y. F. Sun, J. Ma, G. Y. Lu, Sensor. Actuat. B-Chem. 156 (2011) 368-374.
AC C
[36] D. W. Su,; H. S. Kim, W. S. Kim, G. X. Wang, Micropor. Mesopor. Mat. 149 (2012) 36-45.
[37] Y. R. Tao, Q. X. Gao, J. L. Di, X. C. Wu, Nanotechnol. 13 (2013) 5654-5660.
[38]S.Yan, Q. S. Wu, J. Mater. Chem. A,3 (2015) 5982-5990.
[39] C. Su, Y. C. Zou, X. F. Xu, L. Liu, Z. Liu, L. L. Liu, Colloid. Surface.A 472(2015)63-68.
[40] M. Hjiri, L. El Mir, S. G. Leonardi, A. Pistone, L. Mavilia, G. Neri, Sensor. Actuat. B-Chem 196 (2014)
413-420.
[41] Y. L. Wang, X. B. Cui, Q. Y. Yang, J. Liu, Y. Gao, P. Sun, G. Y. Lu, Sensor. Actuat. B-Chem.225 (2016)
14
ACCEPTED MANUSCRIPT 544-552.
RI PT
Caption of Table and Figures
Table 1 The structural parameters and sensitivities of α-Fe2O3 NWs-550, NWs-650 and NWs-750. Table 2 Gas responses of different hematite nanostructures to ethanol gas, as reported in the
SC
literature and the present study.
Fig.1 TEM images (scale bar 100nm) of SBA-15(a) α-Fe2O3 NWs-550 (b), NWs-650 (c) and
M AN U
NWs-750 (d).
Fig.2. XRD patterns of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
Fig.3. Nitrogen physisorption isotherms (a) pore size distribution (b) of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
The response UV-vis spectrum (a) of the α-Fe2O3 NWs-550, NWs-650 and NWs-750; the
TE D
Fig. 4
plots of (Ahv)2 versus hv (b) for the NWs-550, NWs-650 and NWs-750. The response (a) of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to ethanol vapor at
EP
Fig. 5
different operating temperatures; the response and recovery curves of α-Fe2O3 NWs-550(b), NWs-650
AC C
(c) and NWs-750 (d) sensors to 100 ppm ethanol vapor at 300 °C.
Fig. 6
response curves of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors at 300 °C with
varied ethanol.
Fig. 7 The response of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to 100 ppm of various gases at 300 °C
15
SC
RI PT
ACCEPTED MANUSCRIPT
NWs-550
95.23
NWs-650
105.85
NWs-750
96.10
Eg (eV)
response( 100 ppm ethanol 300 °C)
Response time(s)
Recovery time(s)
TE D
SBET (m2/g)
2.01
35.26
6
29
1.99
37.57
8
36
2.02
29.64
8
30
EP
Samples
M AN U
Table 1 The structural parameters and sensitivities of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
AC C
BET surface area (SBET): caculated from the N2 adsorption-desorption isotherms data. Bandgap (Eg): caculated from the UV-vis absorption spectrum.
16
SC
RI PT
ACCEPTED MANUSCRIPT
Table 2. Gas responses of different hematite nanostructures to ethanol gas, as reported in the
M AN U
literature and the present study. Sensing Materials
Method
α-Fe2O3 NWs-650
nanocasting
Mesoporous α-Fe2O3
solid-state chemical
Tsens (°C )
S
Ref.
100
300
37.57
In this paper
1000
340
70
[21]
100
270
21
[31]
Hydrothermal
500
270
42.5
[32]
Hydrolysis process
100
300
27
[33]
reaction
Solid-stats chemical
TE D
α-Fe2O3 nanoparticles
Conc. (ppm)
reaction
Hollow α-Fe2O3
spheres
EP
α-Fe2O3 hollow
Hydrothermal
100
275
12
[34]
Bundle-like
Hydrothermal
100
260
26.8
[35]
Porous α-Fe2O3
Hydrothermal
1000
250
124
[36]
α-Fe2O3 NTs
Hydrothermal
2000
270
42
[37]
α-Fe2O3 nanoropes
electrospinning and
100
240
10.2
[38]
AC C
Urchin-like α-Fe2O3
α-Fe2O3 nanorods
precursor-calcination techniques
17
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.1 TEM images (scale bar 100nm) of SBA-15(a) α-Fe2O3 NWs-550 (b), NWs-650 (c) and NWs-750 (d).
18
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.2. XRD patterns of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
19
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Fig.3. Nitrogen physisorption isotherms (a) pore size distribution (b) of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
20
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4 The response UV-vis spectrum (a) of the α-Fe2O3 NWs-550, NWs-650 and NWs-750; the
AC C
EP
plots of (Ahv)2 versus hv (b) for the NWs-550, NWs-650 and NWs-750.
21
The response (a) of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to ethanol vapor at
EP
Fig. 5
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
different operating temperatures; the response and recovery curves of α-Fe2O3 NWs-550(b), NWs-650
AC C
(c) and NWs-750 (d) sensors to 100 ppm ethanol vapor at 300 °C.
22
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
varied ethanol.
AC C
EP
Fig. 6 Response curves of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors at 300 °C with
23
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
various gases at 300 °C
AC C
EP
Fig. 7 The response of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to 100 ppm of
24
ACCEPTED MANUSCRIPT
1. Mesoporous α-Fe2O3 NWs were synthesized by changing the calcination temperature with the nanocasting method.
RI PT
2. The bandgap hardly changed and the surface areas changed a little. 3. Mesoporous α-Fe2O3 NWs-650 exhibited the highest sensitivity to ethanol gas.
4. The surface defects and body defects could affect the gas-sensing properties.
SC
5. The suitable calcination temperature should be optimized for nanostructures
AC C
EP
TE D
M AN U
to improve the sensitivity.
S1293-2558(17)30037-7
DOI:
10.1016/j.solidstatesciences.2017.05.006
Reference:
SSSCIE 5498
To appear in:
Solid State Sciences
Received Date: 11 January 2017 Revised Date:
11 May 2017
Accepted Date: 13 May 2017
Please cite this article as: X.Q. Li, D.P. Li, J.C. Xu, Y.B. Han, H.X. Jin, B. Hong, H.L. Ge, X.Q. Wang, Calcination-temperature-dependent gas-sensing properties of mesoporous α-Fe2O3 nanowires as ethanol sensors, Solid State Sciences (2017), doi: 10.1016/j.solidstatesciences.2017.05.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Calcination-temperature-dependent gas-sensing properties of mesoporous α-Fe2O3 nanowires as ethanol sensors X.Q. Li, D.P. Li, J.C, Xu,Y.B. Han, H.X. Jin, B. Hong, H.L. Ge and X.Q. Wang*
RI PT
College of Materials Science and Engineering, China Jiliang University, Hangzhou, China, 310018
Abstract:
SC
The mesoporous α-Fe2O3 nanowires (NWs) were successfully synthesized by changing the
M AN U
calcination temperature from 550 to 750 °C (marked NWs-550, NWs-650 and NWs-750) via using SBA-15 silica as the hard templates with the nanocasting method. The characterization results indicated that the bandgap of the as-prepared samples hardly changed and the high BET surface areas changed a little with the calcination temperature from 550 to 750 °C. Mesoporous α-Fe2O3 NWs had been found
TE D
to possess the remarkable gas-sensing performance to ethanol gas. The gas-sensing behavior indicated that α-Fe2O3 NWs-650 exhibited the higher response than that of α-Fe2O3 NWs-550 and α-Fe2O3 NWs-750. The calcination-temperature-dependent gas-sensing properties were mainly attributed to the
EP
competition of surface defects and body defects by the crystallization temperature. The lower
AC C
calcination temperature could create more surface defects to improve the gas-sensing response, while the higher temperature would reduce the body defect and make the charge carriers transport easily. As the result, the suitable calcination temperature was desired to optimize the defects of nanostructures to improve the gas sensitivity.
Keywords: nanowires; nanocasting; calcination-temperature-dependent; gas-sensing properties
*Corresponding Authors. Tel: +86-571-86676157; Fax: +86-571-28911371 Email address: [email protected] (Xinqing Wang)
ACCEPTED MANUSCRIPT 1. Introduction As an environmentally friendly n-type wide bandgap semiconductor, hematite (α-Fe2O3) had been extensively investigated in catalysts, gas sensors, optical devices,
RI PT
lithium-ion batteries and electromagnetic devices [1-5]. Various nanostructures of α-Fe2O3 had been synthesized by the different methods, such as nanospheres [6-7], nanowires [8], nanorods [9], nanoribbons [10], nanotubes [11], hollow nanostructures
SC
[12] and nanoplates [13-16]. Among the above α-Fe2O3 nanostructures, mesoporous
M AN U
α-Fe2O3-based nanostructures should be one of novel candidates for gas-sensor application due to the huge specific surface area. Usually, the working mechanism of the metal oxide based gas sensor was based on the electrical conductivity, resulting from the surface reactions such as oxidation or reduction. Thus, it was very important
TE D
to control the change of the electrical conductivity after the gas adsorption. The huge surface area provided a vast number of surface defects to form the sites for the gas sorption and desorption, improving the gas-sensing sensitivity. Furthermore, the body
EP
defects could reflect the charge carriers and greatly affected the electrical conductivity
AC C
in the reducing gas [17-18].
Mesoporous nanostructures with the huge specific surface area provided a vast
number of surface defects to form the sites of gas sorption and desorption, which was suitable to act as the candidates for gas sensor. Moreover, openly porous structures would prevent the diffusion of carriers [19]. Thus, mesoporous structures could cause the enhancement of both the gas sensitivity and response rate [20]. Some efforts had been engaged in the preparation mesoporous α-Fe2O3 nanostructures for gas sensors 2
ACCEPTED MANUSCRIPT to improve the surface area and surface defects. Li et al. reported the synthesis of mesoporous α-Fe2O3 nanostructures through the solid-state chemical reaction, and exhibited the enhanced gas-sensing performance to xylene [21]. Sun et al. reported the
RI PT
synthesis of mesoporous α-Fe2O3 nanostructures through the soft template method using the triblock copolymer surfactant F127 as the template, which indicated the high gas sensitivity toward acetic acid and ethanol gas [22]. Owing to the good
SC
uniformity and high controllability for target products, nanocasting method was the
M AN U
first choice for the synthesis of the controllable nanostructures in many preparation methods [23]. SBA-15 silica with the well-ordered hexagonal straight mesoporous structures was the good candidates of hard templates for nanowires [24]. Previously, mesoporous α-Fe2O3 nanowires were synthesized using SBA-15 as hard templates
TE D
with the nanocasting method and presented the higher response to ethanol [25]. The high surface area could be obtained by the synthesis of mesoporous nanostructures, while the defects of nanostructures decreased with the increasing
EP
calcination temperature [26]. It was well known that surface defects were good for the
AC C
oxygen absorption/desorption and the body defects were harmful to the gas sensors. And it was necessary to optimize and adjust the surface defects and body defects [27]. In this study, α-Fe2O3 NWs were synthesized by using mesoporous SBA-15 silica as the hard templates with the nanocasting method, and the as-prepared samples were calcined under the temperature between 550-750 °C. The microstructure and bandgap of mesoporous α-Fe2O3 NWs was characterized with X-ray diffraction (XRD), transmission electron microscopy (TEM), nitrogen adsorption/desorption isotherm 3
ACCEPTED MANUSCRIPT and UV-vis spectrum, respectively. Furthermore, the gas-sensing properties of α-Fe2O3 NWs based gas sensors were measured and discussed in detail, and the relationship of the gas response of α-Fe2O3 NWs based gas sensors with the
RI PT
calcination temperature was concluded.
2. Experimental section
SC
All chemicals were of analytical grade and were used as purchased without any
M AN U
further purification. Ordered mesoporous silica SBA-15 was prepared by the method described previously [28]. For the synthesis of α-Fe2O3 nanowires, ferric nitrate and SBA-15 powders (the atomic ratio of Si:Fe=2:1) were dissolved and suspended in ethanol, and then hexane was added until the fine powder formed. The powder
TE D
samples were calcined at 550, 650 and 750 °C for 6 h, respectively, and was marked NWs-550, NWs-650 and NWs-750. Then 2 M NaOH aqueous solution was added to the above powder to remove the silica template and all samples were filtered and
EP
washed with deionized water and ethanol, and then dried at 80 °C for 4 h.
AC C
The morphology of all samples was examined by means of transmission electron microscopy (TEM: JME-1200EX). The crystal structure of the as-prepared products was determined by X-ray diffraction (XRD: XD-5A, Cu Kα, wavelength 0.154 nm, step 0.02°). BaSO4 powder was used as the substrate material to obtain the UV-vis spectrum with a UV3600 spectrophotometer to deduce the bandgap of both samples. Nitrogen physisorption experiments were measured at 77 K on a Micrometrics ASAP 2020 surface area and porosity analyzer. The BET (Brunauer-Emmett-Teller) surface 4
ACCEPTED MANUSCRIPT areas were calculated from the relative pressure range 0.06 to 0.2. To measure the sensing properties of the α-Fe2O3 NWs-550, NWs-650 and NWs-750, the sensors were prepared in the following way: the same amount of each
RI PT
sample was mixed with several drops of deionized water in an agate mortar to form a homogeneous paste, which was then deposited on the alumina ceramic tube assembled with platinum wire electrodes for electrical contacts. The prepared gas
SC
sensing device was aged at 300 °C for 5 days to improve the stability of the sensor.
M AN U
Then a Ni-Cr alloy wire was passed through the alumina ceramic tube and used as a heater by tuning the heating voltage. The gas sensing tests were then performed on a WS-60A gas sensing measurement system, which was a static system using atmospheric air as the interference gas. The relative humidity was about 65%. The
TE D
response (S=Ra/Rg) of a sensor was defined as the ratio of sensor resistance in air (Ra) to that in a target gas (Rg). The response and recovery times were defined as the times required for a change in the resistance to reach 90% of the equilibrium value after the
AC C
EP
detected gas was injected and removed, respectively [29].
3. Results and discussion The microstructures of SBA-15, α-Fe2O3 NWs-550, NWs-650 and NWs-750 were
characterized by TEM in Fig. 1. The TEM photo of SBA-15 in Fig. 1(a) indicated that SBA-15 presented a well-ordered array of mesopores with the pore-size of about 8 nm. After the implanting of α-Fe2O3 in mesopores and removing SBA-15 silica, α-Fe2O3 NWs in mesopores could be obtained. It could be seen in Fig. 1(b), (c) and (d) that all 5
ACCEPTED MANUSCRIPT α-Fe2O3 NWs existed in bundles with tens of nanowires. The α-Fe2O3 NWs maintained the straight mesoporous microstructures with the same diameter of about 7 nm. It should be mentioned that there was little difference between the α-Fe2O3 NWs
RI PT
for the same templates and the synthesized conditions except for the calcination temperature. In this way, the influence of the calcination temperature on the microstructure and the gas-sensing properties could be concluded.
SC
XRD was introduced to characterize the phase structure and crystallization of
M AN U
mesoporous α-Fe2O3 NWs calcined at 550, 650 and 750 °C in Fig. 2. The diffraction peaks of the α-Fe2O3 with the phase structure R-3c(167) were obviously observed. Compared three samples, the intensity of the α-Fe2O3 diffraction peaks increased a little from 140 to 180 a.u. with the temperature, indicating the improved
TE D
crystallization with the calcination temperature. Usually, the higher calcination temperature leaded the less surface and body defects of nanostructures. The decreasing body defects decreased the initial resistance, while the decreasing surface
EP
defects also decreased the resistance after the gas adsorption. Therefore, it was
AC C
necessary to optimize the surface defects and body defects by the surface area and calcination temperature. The nitrogen physisorption isotherms of the mesoporous α-Fe2O3 NWs-550,
NWs-650 and NWs-750 were shown in Fig. 3(a). The isotherms of α-Fe2O3 NWs-550, NWs-650 and NWs-750 all presented the characteristic of the type IV with the type H1 hysteresis loop, indicating the feature of mesoporous-structure according to the IUPAC. This mesoporous-structure could lead to the higher surface area for α-Fe2O3 6
ACCEPTED MANUSCRIPT NWs, which was good for the increase of surface defects. The BET (Brunauer-Emmett-Teller) surface areas of α-Fe2O3 NWs-550, NWs-650 and NWs-750 were calculated to be 95.23, 105.85 and 96.10 m2/g, respectively (Table 1).
RI PT
The α-Fe2O3 NWs-650 presented the highest surface area, which led to the more absorbed oxygen on the surface of nanowires and the larger change of surface resistance. The suitable calcination temperature led to the better mesoporous-structure,
SC
while the higher calcination temperature resulted in the collapse of micropores
M AN U
between the mesopores to some extent [28], which led to the low surface area of α-Fe2O3 NWs-750. The higher surface area should be attributed to the mesoporous-structure of α-Fe2O3 NWs. Fig. 3(b) showed the highly-centralized pore-size distribution of the α-Fe2O3 NWs-550，NWs-650 and NWs-750, indicating
TE D
the average pore diameter of α-Fe2O3 NWs-550，NWs-650 and NWs-750 were 11.38 nm, 9.86nm, 11.06nm, respectively. The corresponding pore volume of the α-Fe2O3 NWs-550，NWs-650 and NWs-750 with the value of 0.3734, 0.3715 and 0.3594
EP
cm3/g were calculated by the BJH method, and decreased with the increasing
AC C
calcination temperature. Furthermore, it could be concluded that the calcination temperature affected the BET surface area of α-Fe2O3 NWs a little. The Fig. 4 (a) showed the typical absorption spectrum of α-Fe2O3 NWs-550,
NWs-650 and NWs-750. Eg was the bandgap and could be calculated by the Tauc equation [30]: (αhv)1/n=A(hv - Eg), where α was the absorption coefficient, h was the Plank constant, v was the photon frequency and n was the electronic transition parameter of 1/2 for direct bandgap semiconductor. Since α was proportional to 7
ACCEPTED MANUSCRIPT absorbance (A) according to Lambert-Beer law, the energy intercept of the curve of (αhv)2 versus hv could give Eg when the linear region was extrapolated to the zero ordinate. The Fig. 4 (b) presented the relationship of (αhv)2 versus hv for α-Fe2O3
RI PT
NWs-550, NWs-650 and NWs-750. The Eg of α-Fe2O3 NWs-550, NWs-650 and NWs-750 (with the value of 2.01, 1.99 and 2.02 eV) were calculated by extrapolating the linear portion of the plot of (αhv)2 versus hv to α=0. The α-Fe2O3 NWs-650
SC
presented the narrowest bandgap, which was good for the electron transition from
M AN U
valence band to conduction band. As the result, the ground state resistance in the target gas decreased to some extent and the sensitivity of gas sensor was improved. From the above analysis, α-Fe2O3 NWs-550 and NWs-750 showed the similar area surface and bandgap, thus it was possible to discuss the influence of the calcination
TE D
temperature on the gas-sensing properties of α-Fe2O3 NWs.
Furthermore, the influence of the calcination temperature on the gas-sensing properties was also investigated as the following. Firstly, the response of α-Fe2O3
EP
NWs-550, NWs-650 and NWs-750 sensors to ethanol gas with 100 ppm was
AC C
measured at different operating temperatures from 240 to 370 °C in Fig. 5 (a). The response of α-Fe2O3 NWs-650 based sensor presented a rapid increase up to 37.57 at 300 °C with the operating temperature and then decreased, which is much higher than those reports with the α-Fe2O3 nanotubes, porous hematite, hematite spheres, nanocrystals, nanoparticles and nanoropes (Table 2
[21,31-38]
). And the α-Fe2O3
NWs-550 and NWs-750 sensors displayed the similar response tendency and their response reached the maximum value of 35.26 and 29.64 at the same operating 8
ACCEPTED MANUSCRIPT temperature of 300 °C, respectively. This phenomenon was commonly observed for many semiconducting metal oxide based sensors and could be explained by the balance between the speed of chemical reaction and gas diffusion [39-40]. At a low
RI PT
operating temperature, the low response could be obtained for the target gas ethanol and did not have enough thermal energy to react with the surface electron of α-Fe2O3 NWs. With the operating temperature increasing, the thermal energy obtained was
SC
high enough to overcome the activation energy of the surface reaction [41]. Moreover,
M AN U
the reduction after the maximum was due to the low gas adsorption ability of the gas molecule at high temperature. Therefore, it was concluded that the optimal operating temperature for α-Fe2O3 NWs based sensors was 300 °C and was used in the following gas-sensing tests. Mesoporous α-Fe2O3 nanowires were studied previously
TE D
and showed the different response to ethanol, which was attributed to the different aged days [25]. The response-recovery time to 100 ppm ethanol at 300 °C was given in Fig. 5 (b), (c) and (d). The response and recovery times of α-Fe2O3 NWs-550,
EP
NWs-650 and NWs-750 sensors were only 6 s, 8 s, 8 s and 29 s, 36 s, 30 s (Table 1),
AC C
resulting from the different diffusion speed of air and ethanol gas. Fig. 6 showed the responses of α-Fe2O3 NWs-550, NWs-650 and NWs-750
based sensors to different concentration of ethanol in the range of 5-1000 ppm at 300 °C. It was seen that the response of all samples increased with the ethanol concentration. The response of the mesoporous α-Fe2O3 NWs-650 sensor increased from 3.37 (5 ppm) to 77.36 (1000 ppm), while that of α-Fe2O3 NWs-550 and NWs-750 increased from 3.27 and 3.26 to 62.1 and 56.43, respectively. Due to the 9
ACCEPTED MANUSCRIPT higher surface area and narrowest bandgap, the response of the α-Fe2O3 NWs-650 based sensors was larger than those of α-Fe2O3 NWs-550 and NWs-750 sensors. With the similar area surface and bandgap, the sensitivity of α-Fe2O3 NWs-550 was a little
RI PT
higher than that of α-Fe2O3 NWs-750 for the lower calcination temperature. Owing to surface reactions such as oxidation or reduction, the working mechanism of the metal oxide gas sensor was based on the change of the electrical
SC
resistance in air and ethanol gas. In order to increase the sensitivity (S = Rair/Rgas), one
M AN U
way was to decrease the target gas resistance (ground state resistance) and another was to increase the air resistance. For n-type semiconductor, the higher crystallization degree with the less body defects resulted in the decrease of core resistance in target gas. The lower crystallization degree with the more surface defects resulted in the
TE D
increase of shell resistance in air. As the result, the suitable calcination temperature was desired to optimize the defects of nanostructures to improve the gas sensitivity. These surface reactions were directly dependent on the oxygen vacancies at the
EP
surface, which required a higher specific surface area with more active centers and
AC C
defects at the surface. Usually, the surface and body defects of α-Fe2O3 NWs decreased with the increasing calcination temperature. The decreasing surface defects would decrease the resistance in air, while the decreasing body defects decreased the initial resistance. In this way, the calcination temperature greatly affected the resistance and gas-sensing properties of α-Fe2O3 NWs. Thus, it was concluded that the resistance change between air and ethanol gas was dominated by electronic depletion layer at the surface of nanowires. And the sensitivity was improved with the lower 10
ACCEPTED MANUSCRIPT calcination temperature. Finally, the selectivity for different gases was also discussed with the concentration of 100 ppm at 300 °C in Fig. 7, and the gases included ethanol, acetone,
RI PT
methyl alcohol, normal hexane and benzene. The sensitivities of mesoporous α-Fe2O3 NWs-650 based sensors were always larger than those of α-Fe2O3 NWs-550 and NWs-750 sensor to different gases. The response of the α-Fe2O3 NWs-650 was 41.63,
SC
34.07, 7.95, 1.62 and 1.55 to ethanol, acetone, methyl alcohol, normal hexane and
M AN U
benzene, respectively. The response of α-Fe2O3 NWs-550 based sensor was 36.55, 31.36, 6.53, 1.57 and 1.31 and that of α-Fe2O3 NWs-750 sensor was 34.76, 29.82, 6.36, 1.33 and 1.28 respectively. Furthermore, the response of all samples to ethanol and acetone was much larger than those to methyl alcohol, normal hexane and
TE D
benzene. It can be seen that all samples showed the best response to ethanol gas. This should be attributed to the difference interactions between α-Fe2O3 molecules and target gas molecules. The absorbed oxygen on the surface of α-Fe2O3 NWs was
EP
preferred to react with ethanol at 300 °C. Therefore, the larger number of electrons
AC C
released back to conductive band could lead to the larger change of resistance.
4. Conclusions
Mesoporous α-Fe2O3 NWs were successfully synthesized by using SBA-15 silica as the hard templates with the nanocasting method under the calcination temperature between 550-750 °C. The BET surface areas of mesoporous α-Fe2O3 NWs-550, NWs-650 and NWs-750 exhibited the high surface areas. With increasing the 11
ACCEPTED MANUSCRIPT calcination temperature from 550 to 750 °C, the bandgap hardly changed. All results showed that all the samples exhibited high responses to ethanol gas and the lowest detection concentration at 5 ppm. Moreover, α-Fe2O3 NWs-650 exhibited the best
RI PT
gas-sensing performance for the high surface area and narrow bandgap. From the gas-sensing analysis of α-Fe2O3 NWs-550 and NWs-750 with the similar area surface and bandgap, it was concluded that the calcination temperature greatly affected the
SC
gas-sensing behavior of nanostructures. And the sensitivity could be improved at the
M AN U
low calcination temperature.
Acknowledgments
TE D
The research was funded by National Natural Science Foundation of China (Nos. 51202235) and Foundation of Science and Technology Department of Zhejiang Province (Nos.2017C33078
AC C
References
EP
and Nos. LY15B010004)
[1] Q.Y. Hao, S. Liu, X.M. Yin, Y.G. Wang, Q.H. Li, T.H. Wang, Solid State Sci. 12(2010)2125-2129
[2] J. Xu, F. Huang, Y. L. Yu, A. P. Yang, Y. S. Wang, Cryst. Eng. Commun. 13 (2011) 4873-4877.
[3] J. R. Huang, M. Yang, C. P. Gu, M. H. Zhai, Y. F. Sun, J. H. Liu, Mater. Res. Bull. 46 (2011) 1211-1218.
[4] M. Mahmoudi, A. Simchi, M. Imani J. Iran. Chem. Soc.7 (2010)21-27.
[5] T. E. García, M. Valvo, U. Lafont, C. Locati, D. Munao, E. M. Kelder J. Power Sources 196 (2011) 6425-6432
[6] L. L. Wang, Z. Lou, J. N. Deng, T. Zhang,ACS. Appl. Mater. Inter.7(2015) 13098-13104
12
ACCEPTED MANUSCRIPT [7] H. J. Wu, G. L. Wu, L. D. Wang, Powder. Technol. 269 (2015) 443-451.
[8] C. Wang, Y. Q. Wang, X. H. Liu, H. W. Yang, J. R. Sun, L. Yuan, G. W. Zhou, F. Rosei, Structure versus
properties in α-Fe2O3 nanowires and nanoblades, nanotechnology 27 (2016)035702.
RI PT
[9] D. Patil, V. Patil, P. Patil, Sensor. Actuat. B-Chem. 152 (2011) 299-306.
[10] D. Sarkar, M. Mandal, K. Mandal, ACS Appl. Mater. Inter. 5 (2013) 11995-12004.
[11] D. H. Kim, Y. S. Shim, J. M. Jeon, H. Y. Jeong, S. S. Park, Y. W. Kim, J. S. Kim, J. H. Lee, H. W. Jang, ACS
SC
Appl. Mater. Inter. 6 (2014) 14779-14784.
M AN U
[12] K. Kwon, H. Lim, Y. Sun, K. Suh, J. Phys. Chem. C 118 (2014) 2897-2907.
[13] R. Liu, Y. Jiang, Q. Lu, W. Du, F. Gao, CrystEngComm 15 (2013) 443-446.
[14] F. Lu, Q. Wu, X. Yang, L. Chen, J. Cai, C. Liang, M. Wu, P. Shen, Phys. Chem. Chem. Phys. 15 (2013)
9768-9774.
TE D
[15] J. Cai, S. Chen, M. Ji, J. Hu, Y. Ma, L. Qi, CrystEngComm 16 (2014) 1553-1559.
[16] H. Liang, W. Chen, X. Jiang, X. Xu, B. Xu, Z. Wang, J. Mater. Chem. A 2 (2014) 4340-4346.
[17] D.B. Mahadik, Y.K. Lee, C.-S. Park, H.-Y. Chung, M.-H. Hong, Hae-N.-R. Jung, W. Han, H.-H. Park, Solid
EP
State Sci. 50(2015) 1-8
AC C
[18]A. Maurya, P. Chauhan, S. K. Mishra, R. K. Srivastava, J. Alloy. Compd. 509 (2011) 8433-8440.
[19]G. X. Zhu, L. J. Guo, X. P. Shen, Z. Y. Ji, K. M. Chen, H. Zhou, Sensor. Actuat. B-Chem. 220 (2015)
977-985.
[20] G. X. Zhu, L. J. Guo, X. P. Shen,Z. Y. Ji,, K. M. Chen, H. Zhou, Sensor. Actuat. B-Chem. 220 (2015)
977-985.
[21] Y. Z. Li, Y. L. Cao, D. Z. Jia, Y. Wang, J. Xie, Sensor. Actuat. B-Chem. 198 (2014) 360-365.
[22] B. Sun, J. Horvat, H.S. Kim, W. Kim, J. Ahn, G. Wang, J. Phys. Chem. C 114 (2010) 18753-18761.
13
ACCEPTED MANUSCRIPT [23] M. Sakeye, S. Ziller, H. Amenitsch, M. Lindén and J. Smått,
J. Phys. Chem. C 120 (2016) 1854-1862.
[24] D.P. Li, Y. Zhan, J.C. Xu, H.X. Jin, D.F. Jin, B. Hong, X.L. Peng, P.F. Wang, H.L. Ge, X.Q. Wang,
Physica E 84 (2016) 395–400
Peng, H. L. Ge and X. Q. Wang,
Nanotechnology 27 (2016) 185702-185707.
RI PT
[25] D. P. Li, B. B. Zhang, J. C. Xu, Y. B. Han, H. X. Jin, D. F. Jin, X. L.
[26] L.W. Mo, M. Deng, M. S. Tang, Key Engineering Materials, 400-402 (2009) 169-174
SC
[27] S. Kavianpour, A. Yavari, Comp. Mater. Sci. 44 (2009) 1296-1306
M AN U
[28] X. Q. Wang, H. L. Ge, H. X. Jin, Y. Cui,Micropor. Mesopor. Mat.86 (2005) 335-340.
[29] L. Xu, B. Dong, Y. Wang, X. Bai, Q. Liu, H. W. Song, Sensor. Actuat. B-Chem. 147(2010) 531-538.
[30] S. K. Maji, N. Mukherjee, A. Mondal, B. Adhikary, Polyhedron 33 (2012) 145-149
[31]Y. L. Cao, H. Y. Luo, D. Z. Jia, Sensor. Actuat. B-Chem. 176 (2013) 618-624.
TE D
[32] H. J. Song, X. H. Jia, X. Q. Zhang, J. Mater. Chem. 22 (2012) 22699-22705.
[33] H. J. Song, X. H. Jia, H. Qi, X. F. Yang, H. Tang, C. Y. Min, J. Mater. Chem. 22 (2012) 3508-3516.
[34] P. Sun, W. N. Wang, Y. P. Liu, Y. F. Sun, J. Ma, G. Y. Lu, Sensor. Actuat. B-Chem. 173 (2012) 52-57.
EP
[35] P. Sun, L. You, D. W. Wang, Y. F. Sun, J. Ma, G. Y. Lu, Sensor. Actuat. B-Chem. 156 (2011) 368-374.
AC C
[36] D. W. Su,; H. S. Kim, W. S. Kim, G. X. Wang, Micropor. Mesopor. Mat. 149 (2012) 36-45.
[37] Y. R. Tao, Q. X. Gao, J. L. Di, X. C. Wu, Nanotechnol. 13 (2013) 5654-5660.
[38]S.Yan, Q. S. Wu, J. Mater. Chem. A,3 (2015) 5982-5990.
[39] C. Su, Y. C. Zou, X. F. Xu, L. Liu, Z. Liu, L. L. Liu, Colloid. Surface.A 472(2015)63-68.
[40] M. Hjiri, L. El Mir, S. G. Leonardi, A. Pistone, L. Mavilia, G. Neri, Sensor. Actuat. B-Chem 196 (2014)
413-420.
[41] Y. L. Wang, X. B. Cui, Q. Y. Yang, J. Liu, Y. Gao, P. Sun, G. Y. Lu, Sensor. Actuat. B-Chem.225 (2016)
14
ACCEPTED MANUSCRIPT 544-552.
RI PT
Caption of Table and Figures
Table 1 The structural parameters and sensitivities of α-Fe2O3 NWs-550, NWs-650 and NWs-750. Table 2 Gas responses of different hematite nanostructures to ethanol gas, as reported in the
SC
literature and the present study.
Fig.1 TEM images (scale bar 100nm) of SBA-15(a) α-Fe2O3 NWs-550 (b), NWs-650 (c) and
M AN U
NWs-750 (d).
Fig.2. XRD patterns of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
Fig.3. Nitrogen physisorption isotherms (a) pore size distribution (b) of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
The response UV-vis spectrum (a) of the α-Fe2O3 NWs-550, NWs-650 and NWs-750; the
TE D
Fig. 4
plots of (Ahv)2 versus hv (b) for the NWs-550, NWs-650 and NWs-750. The response (a) of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to ethanol vapor at
EP
Fig. 5
different operating temperatures; the response and recovery curves of α-Fe2O3 NWs-550(b), NWs-650
AC C
(c) and NWs-750 (d) sensors to 100 ppm ethanol vapor at 300 °C.
Fig. 6
response curves of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors at 300 °C with
varied ethanol.
Fig. 7 The response of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to 100 ppm of various gases at 300 °C
15
SC
RI PT
ACCEPTED MANUSCRIPT
NWs-550
95.23
NWs-650
105.85
NWs-750
96.10
Eg (eV)
response( 100 ppm ethanol 300 °C)
Response time(s)
Recovery time(s)
TE D
SBET (m2/g)
2.01
35.26
6
29
1.99
37.57
8
36
2.02
29.64
8
30
EP
Samples
M AN U
Table 1 The structural parameters and sensitivities of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
AC C
BET surface area (SBET): caculated from the N2 adsorption-desorption isotherms data. Bandgap (Eg): caculated from the UV-vis absorption spectrum.
16
SC
RI PT
ACCEPTED MANUSCRIPT
Table 2. Gas responses of different hematite nanostructures to ethanol gas, as reported in the
M AN U
literature and the present study. Sensing Materials
Method
α-Fe2O3 NWs-650
nanocasting
Mesoporous α-Fe2O3
solid-state chemical
Tsens (°C )
S
Ref.
100
300
37.57
In this paper
1000
340
70
[21]
100
270
21
[31]
Hydrothermal
500
270
42.5
[32]
Hydrolysis process
100
300
27
[33]
reaction
Solid-stats chemical
TE D
α-Fe2O3 nanoparticles
Conc. (ppm)
reaction
Hollow α-Fe2O3
spheres
EP
α-Fe2O3 hollow
Hydrothermal
100
275
12
[34]
Bundle-like
Hydrothermal
100
260
26.8
[35]
Porous α-Fe2O3
Hydrothermal
1000
250
124
[36]
α-Fe2O3 NTs
Hydrothermal
2000
270
42
[37]
α-Fe2O3 nanoropes
electrospinning and
100
240
10.2
[38]
AC C
Urchin-like α-Fe2O3
α-Fe2O3 nanorods
precursor-calcination techniques
17
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.1 TEM images (scale bar 100nm) of SBA-15(a) α-Fe2O3 NWs-550 (b), NWs-650 (c) and NWs-750 (d).
18
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.2. XRD patterns of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
19
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Fig.3. Nitrogen physisorption isotherms (a) pore size distribution (b) of α-Fe2O3 NWs-550, NWs-650 and NWs-750.
20
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4 The response UV-vis spectrum (a) of the α-Fe2O3 NWs-550, NWs-650 and NWs-750; the
AC C
EP
plots of (Ahv)2 versus hv (b) for the NWs-550, NWs-650 and NWs-750.
21
The response (a) of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to ethanol vapor at
EP
Fig. 5
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
different operating temperatures; the response and recovery curves of α-Fe2O3 NWs-550(b), NWs-650
AC C
(c) and NWs-750 (d) sensors to 100 ppm ethanol vapor at 300 °C.
22
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
varied ethanol.
AC C
EP
Fig. 6 Response curves of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors at 300 °C with
23
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
various gases at 300 °C
AC C
EP
Fig. 7 The response of α-Fe2O3 NWs-550, NWs-650 and NWs-750 sensors to 100 ppm of
24
ACCEPTED MANUSCRIPT
1. Mesoporous α-Fe2O3 NWs were synthesized by changing the calcination temperature with the nanocasting method.
RI PT
2. The bandgap hardly changed and the surface areas changed a little. 3. Mesoporous α-Fe2O3 NWs-650 exhibited the highest sensitivity to ethanol gas.
4. The surface defects and body defects could affect the gas-sensing properties.
SC
5. The suitable calcination temperature should be optimized for nanostructures
AC C
EP
TE D
M AN U
to improve the sensitivity.