Mn―doped mesoporous SnO2 sensor with high formaldehyde selectivity
开篇：润墨网以专业的文秘视角，为您筛选了一篇Mn―doped mesoporous SnO2 sensor with high formaldehyde selectivity范文，如需获取更多写作素材，在线客服老师一对一协助。欢迎您的阅读与分享！
mn-doped mesoporous sno2 was synthesized via a sol-gel method and then made into indirectly heated thick film gas sensor. The sensor shows far higher response value to formaldehyde than common interference gases when the Mn doping is 1mol% at 300 ℃， which is also with good response and recovery performance. Moreover， the sensor can detect formaldehyde down to 10ppm， and the response time and recovery time are less than 4s and 25s， respectively. The results of the study demonstrate that the mesoporous sensor with 1mol% Mn doping is suitable for detecting formaldehyde.
Keyword：Mesoporous SnO2， Formaldehyde， Gas sensor， Response.
INTRODUCTION
Mesoporous material is a special nanometer material， which has a lot of pores between 2 and 50 nm. It is drawing more and more attention due to its special structure and the superior performance in the chemical industry[1]， biological technology[2]， adsorption separation[3] and other fields. Increasing the specific surface area[4]and doping[5] are commonly used in improving the gas sensing properties. Compared with ordinary nanometer materials， ordered mesoporous SnO2 has a larger specific surface area， therefore it can improve the gas sensing performances to a certain extent. Doping metal or metal oxide into mesoporous SnO2 may significantly affects the physical and chemical properties， which initiates the change of the performance of gas sensor. For instance， mesoporous SnO2 sensor doped by MoO3 dramatically improved the response and selectivity to NO2[6] ； Mesoporous SnO2 sensor synthesized by sol-gel method shows the best H2 gas sensing performance when Sb2O5 doping ratio is 5wt.% and the working temperature is 400℃[7].
Manganese， widely distributed in the crust， is relatively inexpensive. However， there has been little research on the effect of gas sensing properties when adding Mn to mesoporous SnO2. In this paper， the indirectly heated thick film gas sensors were made by Mn-doped mesoporous SnO2 coated on Al2O3 tubular ceramic with a pair of Au electrodes attached with Pt wires， and then the microscopic structure and gas sensing performance affected by Mn doping were studied.
EXPERIMENT
2.1. Preparation of mesoporous SnO2
Pure mesoporous SnO2 powder was prepared from the precipitate obtained by adjusting the pH of the aqueous precursor solution to 10 using an aqueous HCl solution. The precursor solution contained n-cetylpyridinium chloride monohydrate （C16PyCl； （C5H5NC16H33）Cl・H2O） as a template， Na2SnO3・3H2O as a tin source and trimethylbenzene （mesitylene； MES）， where the concentration of C16PyCl was 2.5wt.% and the molar ratios were [C16PyCl]/[Na2SnO3・3H2O]=2.0 and [MES]/[Na2SnO3・3H2O]=2.5. After aging at 20 ℃ for 3 days in the solutions， these precipitates were treated with a 0.1M aqueous phosphoric acid （PA） solution for 2h to improve the thermal stability. The resultant powders were filtered and washed with deionized water. Then they were dried at 80℃ for 10h and then calcined at 600℃ for 5h in air. The sample was recorded as m-SnO2. The calcined powders were post-ground using an agate mortar， to reduce the size of their large agglomerates.
The preparation of Mn-doped SnO2 was similar to that of pure SnO2. MnCl2 was weighted according to the mole ratio [MnCl2]/[Na2SnO3・3H2O]=1%， 5% or 10%， configured to aqueous solution mixed with Na2SnO3・3H2O， C16PyCl and MES. The gained solids were recorded as m-SM1 （1 mol% Mn-doped mesoporous SnO2）， m-SM5 （5mol% Mn-doped mesoporous SnO2） and m-SM10 （10mol% Mn-doped mesoporous SnO2）.
Crystal phase and mesoporous structure of mesoporous SnO2 powders were characterized by X-ray diffraction analysis （XRD； Bruker， D8 Advance， Cu Kα）. The specific surface area and pore size distribution were measured by BET and BJH methods using a N2 adsorption isotherm （Micromeritics， ASAP2020）.
2.2. Fabrication of gas sensors and gas sensing performance text
Appropriate m-SnO2， m-SM1， m-SM5 or m-SM10 was weighted and homogenized by adding suitable amount deionized water， and then coated onto Al2O3 tubes. The thick films were sintered at 550℃ for 2h after being dried under air to remove water. The film thickness observed by light microscopy was about 50μm. A small Ni-Cr alloy coil with the resistance of about 33Ω was placed through the tube as a heater. The temperature was controlled through adjusting the heating power. The gas sensors were stabilized at 450℃ for 3 days in air before measurement.
The surface appearances of the thick films were characterized by a field emission scanning electron microscopy （FE-SEM， Hitachi， S-4800）.Target gases were prepared in laboratory， and the concentrations were 1000ppm. The lower concentration gases can be acquired by adjusting self-made gas sensing test system. Sensor response（S） was defined as resistance ratio， Ra/Rg， where Ra and Rg stand for the electrical resistances in dry air and the target gas， respectively. The response and recovery times were defined as the time taken to reach 90% of the original resistance.
RESULTS AND DISCUSSION
Fig. 1 shows the wide-angle and low-angle XRD patterns of pure and Mn-doped mesoporous SnO2. From the Fig.1（1） ，it can be seen that all the prominent peaks in the patterns correspond to the rutile structure of SnO2 and are indexed on the basis of standard PDF card（No.00-041-1445）. However， no characteristic peaks of the doped ions are found， which indicates that manganese ions may occupy the lattice of tin ions and the structure is almost constant after the incorporation of impurities. It is also observed that Mn doping do not significantly change the full width at half maxima of the diffraction peaks， suggesting there is little change about crystallite size before and after doping from the Scherer formula. As can be seen in the Fig.1（2），there are obvious diffraction peaks in the 2θ=2.6°-2.8°， which indicate the formation of ordered mesoporous structure was confirmed for mesoporous SnO2 before and after doping.
（1）wide-angle XRD （2）low-angle XRD
Fig.1 Wide- and low-angle XRD patterns of （a）m-SnO2，（b）m-SM1，（c）m-SM5 and（d）m-SM10
Fig. 2 shows pore size distribution and the values of specific surface area of m-SnO2，m-SM1，m-SM5 and m-SM10 powders acquired by Nitrogen adsorption isotherm. Specially， all the powders display the largest pore volumes when the pore size is 2.45nm. In addition， the specific surface area of m-SnO2 was the largest， reaching 283.5 m2・g-1. As for the m-SM1 with low concentration of Mn doping， the specific surface area remained about the same. However， when the amount of doped Mn was higher （5mol% and 10mol%） ， the specific surface areas were changed significantly， reducing to 224.2 m2・g-1 and 220.6 m2・g-1.
.
Fig.2（a）m-SnO2，（b）m-SM1，（c）m-SM5，（d）m-SM10 pore size distribution
The surface appearances of pure and Mn-doped SnO2 gas-sensing film are shown in Fig.3. The average diameters of m-SnO2，m-SM1，m-SM5 and m-SM10 are very close to each other， showing that the doping do not cause significant changes in grain sizes， which is consistent with the results of XRD test. The agglomeration phenomenon in the thick film of pure m-SnO2 is severe and the uniformity of particle distribution is poor. With the increase in the amount of Mn doping， the sizes of secondary particles in the thick film are much smaller and better distributed， which may be due to that the pinning of manganese oxide phase suppresses the growing of m-SnO2 grains.
Fig.3 FE-SEM images of （a） m-SnO2， （b） m-SM1，（c） m-SM5， （d） m-SM10.
Fig.4 shows the responses of all different types of sensors to 1000ppm formaldehyde gas at different temperatures. The response increases as the temperature rises below 300℃ and reaches the maximum at 300℃， then decreases rapidly. Compared with m-SnO2， m-SM5 and M-SM10， m-SM1 showed a much greater response value at 300℃， which indicated that 1mol% Mn was the optimum doping amount to enhance the response of mesoporous SnO2 to formaldehyde.
Fig.4 Temperature dependence of response of m-SnO2，m-SM1，m-SM5 and m-SM10 sensors.
As known to all， response and recovery performances are important factors to evaluate the quality of gas sensors. Tab.1 lists the response and recovery times of the m-SM1 sensor to 1000ppm formaldehyde at different working temperatures. The response time is 92s at 100℃. But as the temperature increases， the response time becomes shorter and reduces to only a few seconds when the temperature comes to 300～400℃. This is because desorption and catalytic reaction rate of the target gas can be accelerated at higher temperature[8]. The response and recovery times at 300～400℃ are short enough to meet the needs of the practical application， but the values of the sensor responses are too low at 350℃ and 400℃，so the operating temperature is set at 300℃.
Tab.1 Response and recovery times of m-SM1 sensor to
1000ppm formaldehyde at different temperatures.
Working temperature（℃）
100 150 200 250 300 350 400
Response time（s） 92 55 15 12 3.5 3.4 3.1
Recovery time（s） 182 169 150 108 25 22 20
In order to study the response and recovery performances of m-SM1 sensor to different concentrations of formaldehyde， sensor was successively exposed to 1000， 300， 100， 50， 30 and10ppm of formaldehyde at 300℃. As shown in Fig.5， the sensor shows a high response value and fairly short response and recovery times. When exposed to 1000ppm formaldehyde， the response value is approximately 68. With the decline of the formaldehyde concentration， the response value is also significantly reduced. For formaldehyde at standards of 300， 100， 50， 30， 10 ppm， the response values are about 37， 22， 10， 7 and 6，respectively. The response and recovery times of m-SM1 to different concentrations of formaldehyde are short and change little with the different concentrations， in which the response time does not exceed 4s and the recovery time is not more than 25s.
Fig.5 The response and recovery characteristics of SM1 sensors to different concentrations of formaldehyde at 300 ℃
The sensing properties of the all types of sensors to other gases such as CH3CH2OH，benzene，dimethylbenzene，CH2Cl3， CH3COOC2H5， DEA and CH3OH were also examined. The responses to the above gases of 1000ppm are shown in Fig.6. m-SM1 sensor has the highest response （about 68） to formaldehyde， and the responses to ethanol and methanol are about 17 and 13， respectively. The responses of m-SM1 sensor to other test gas are much lower， so the selectivity of the sensor to formaldehyde is high， which is essential for sensors to detect formaldehyde in the practical application. The responses of other types of sensors to these gases have no obvious advantage， so they cannot be used to detect a single gas.
Fig.6 The responses of all types of sensors to different gases （1000ppm） at 300 ℃.
4. INITIAL STUDY OF GAS SENSING MECHANISM
The sensing mechanism of traditional SnO2 gas sensors have been proved in previous works[9]. Oxygen with high electron affinity adsorbs on the exposed surface of the n-type semiconductor mesoporous SnO2 and ionizes to O- or O2-. As a result， the potential barriers are formed on the surface of SnO2 area， which hinder the electrons movement at grain boundaries， resulting in a increase of resistance rate. When the sensor is exposed to a reducing gas （such as formaldehyde）， the gas will react with adsorbed oxygen molecules and release the trapped electrons back to the conduction band， thereby increasing the carrier concentration and carrier mobility of SnO2.
Compared with pure mesoporous SnO2， Mn doping causes a remarkable decrease of the particle size， but the specific surface area does not rise simultaneously. That is because the external surface area is much smaller than the hole area for mesoporous material. The trivalent Mn3+ ions， which act as acceptor impurity， exist in the doped mesoporous SnO2.The Sn4+ replaced by Mn3+ can be represented by equation as
Where represents Mn substitution in Sn sites， represents lattice oxygen vacancies and the represents the original oxygen position occupied by the doping oxygen. The point defects caused by doping may increase surface active states for adsorption of detecting gas and then improve gas sensing performances of the sensor significantly.
Manganese oxide is a kind of P-type semiconductor， forming the surface hetero p-n junction when adding to the SnO2. A moderate amount of Mn doping （1mol%） easily makes all the surface hetero p-n junction of sensing film convert to schottky barrier， which gets the biggest resistance change and dramatically increases the response of the sensor. But the excess Mn doping （5mol% or 10mol%） cannot help transform hetero p-n junction to schottky barrier and more electrons are bounded by p-n junction， therefore， the resistance increases remarkably， reducing the response of the sensor elements[10].In addition，the specific surface area of 1mol% Mn doping mesoporous SnO2（280.3 m2・g -1） is only slightly lower than that of pure mesoporous SnO2， which is beneficial to improve the response of m-SM1. However， along with the increase of Mn doping （5mol % or 10mol %）， both the substantial fall of specific surface area and electron-binding effect of hetero p-n junction will cause a large likely drop in response of the sensor， although there is a increase in the surface active sites.
5. CONCLUSION
Mn-doped mesoporous SnO2 was synthesized via a sol-gel method. The formaldehyde sensing performances of mesoporous SnO2 are greatly improved when the Mn doping ratio is 1 mol%. m-SM1 sensor simultaneously shows the most excellent formaldehyde response and short response and recovery time at 300℃. This kind of sensor has good response and recovery characteristics to 10～1000ppm formaldehyde. The response time and recovery time are less than 4s and 25s， respectively. m-SM1 sensor has much higher formaldehyde response than the other interference gases， which is of great importance for detecting formaldehyde.
References and Notes
G. Michel， P. Ludovic， G. Matteo， Z. Federica. Molecular sieve catalysts as substitutes for metal chlorides in the chemical industry： Some selected examples[J]. Pure Appl. Chem. 2012，（84）：509-527.
A.E. Garcia-Bennett. Synthesis， toxicology and potential of ordered mesoporous materials in nanomedicine [J].Nanomedcine， 2011，（6）：867-877.
Z.X. Wu， D.Y. Zhao. Ordered mesoporous materials as adsorbents[J]. Chem. Commun. 2011，（47）： 3332-3338.
J.H. Pan， S.Y. Chai， C. Lee， S.E. Park， W.I. Lee. Controlled formation of highly crystallized cubic and hexagonal mesoporous SnO2 thin films[J]. J. Phys. Chem. C， 2007（111）： 55825587.
Z.H. Wen， G. Wang， W. Lu， Q. Wang， Q. Zhang， J.H. Li， Enhanced photocatalytic properties of mesoporous SnO2 induced by low concentration ZnO doping[J]. Cryst. Growth Des. 2007，（7） 17221725.
A.A. Firooz， T. Hyodo， A.R. Mahjoub， A. A. Khodadadi， Y. Shimizu. Synthesis and gas-sensing properties of nano- and meso-porous MoO3-doped SnO2[J]. Sens. Actuators B Chem. 2010（）147：554560.
L.Y. Yuan， T. Hyodo， Y. Shimizu， M. Egashira. Preparation of Mesoporous and Meso-macroporous SnO2 Powders and Application to H2 Gas Sensors[J]. Sens.Mater. 2009（21）： 241250
W. H. Zhang， W.D. Zhang. Fabrication of SnO2ZnO nano composite sensor for selective sensing of trimethylamine and the freshness of fishes[J]. Sens. Actuators B. Chem. 2008，（134）：403408.
M. Batzill， U. Diebold. The surface and materials science of tin oxide[J]. Prog. Surf. Sci. 2005，（79）：47154.
K. Jain， R.P. Pant， S.T. Lakshmikumar. Effect of Ni doping on thick film SnO2 gas sensor[J]. Sens. Actuators B Chem. 2006，（113）823829.