Research on key technologies of 3D nano clustersensitive film based on SAW resonator

QI Jing, WEN Yumei, LI Ping

(1.College of Optoelectronic Engineering, Chongqing University, Chongqing, 400044,P.R.China;2.Chongqing Key Lab of Mobile Communications Technology, Chongqing University of Posts and Communications,Chongqing, 400044,P.R. China)

Research on key technologies of 3D nano clustersensitive film based on SAW resonator

QI Jing1, 2, WEN Yumei1, LI Ping1

(1.College of Optoelectronic Engineering, Chongqing University, Chongqing, 400044,P.R.China;2.Chongqing Key Lab of Mobile Communications Technology, Chongqing University of Posts and Communications,Chongqing, 400044,P.R. China)

This research attempts to design a new type of surface acoustic wave (SAW) gas sensor that integrates both the three-dimensional nanostructure technology and the SAW technology and thereby makes use of their advantages in sensitivity and specific surface area, respectively. By taking advantage of the resonant surface acoustic wave(SAWR)’s high Q-factor and low insertion loss characteristic, this paper mainly focuses on the design of the SAW chip and modification of the nanowires structure. Finally，through pumping the Sarin and HD gas into the sample pool of the 3D nano-SAW gas sensor with Back-Propagation Network recognize algorithm, As shown by the experiment results, the overall distinguishable rate is more than 90%, and the 3D nano-SAW gas sensor is superior to traditional sensing devices concerning the sensitivity, identification accuracy, and response speed.

surface acoustic wave (SAW)； nanowires； gas sensor；BP-ANN

1 Introduction

The surface acoustic wave(SAW) sensor refers to a new type of transducer that can measure frequency’s change, and it includes some environmental parameters, such as temperature and pressure. The SAW gas sensor detects gas by measuring the amplitude and frequency shifts of SAW when the target gas moves over the surface of the gas-sensitive materials[1]. One-dimensional metals or transition metal oxides, such as ZnO, SnO2, TiO2, and In2O3, are most widely applied in SAW sensor design. Besides, the nanowires or nanotubes materials, which are extensively utilized in gas sensing and detecting devices, have captured the attention of multiple researchers. Additionally, the three-dimensional (3D) gas-sensitive nanowires have a high specific surface area. When they integrate with the highly sensitive and fast responding SAW devices, they can create a new type of SAW gas sensor which will overcome the limits of traditional two-dimensional (2D) membrane sensors[2-8].

As ZnO has a faster reaction rate, and can react with gas at ambient temperature. Then, based on the method of preparing the ZnO nanowires, this paper researches the SAW technology, the aim of which is to create a new type of SAW gas sensor that can quickly detect gas at low concentration.

This new type of gas sensor takes advantage of the 3D nanostructure technology as well as the SAW technology, and it can be acutely sensitive, highly integrated, small, lightweight, and inexpensive. With micro, integrated and digitalized features, the 3D nano-SAW gas sensor can not only detect gas rapidly and reliably but also operate at room temperature.

2 Working principle

In essence, the SAW gas sensor is a SAW device coated with a film of gas-sensitive material. When the target gas moves over the surface of the SAW sensor, the gas-sensitive material will selectively absorb the molecules of the target gas, which results in the changes in its mass per unit area[9]. This can affect the properties of SAW in two ways: firstly, it will not only reduce the amplitude of SAW but also exacerbate its insertion loss. Secondly, it will cause a change in the base frequency of SAW, which can be represented as Δfs. The correlation between them is

(1)

In the function,cmrefers to the parameters of SAW substrate; Δ(m/A) refers to the increase in mass per unit area in the sensitive material;f0refers to baseband signal frequency.

When the SAW substrate is given, there is a linear relationship between the increase in mass per unit area of the sensitive material and the square of the SAW operational frequency. Therefore, to increase the Δfs, the sensitivity per unit area of the sensitive material and the operational frequency of the SAW sensor shall be increased. Sensitivity per unit area is closely related to both the sensitive material itself and the techniques, by which the SAW chip and the sensitive material can be combined[10]. Fig.1 shows the working principle of the SAW gas sensor.

Fig.1 Working principle of the SAW gas sensor

Fig.2 Two-dimensional sensitive membrane

Fig.3 Three-dimensional nanowire clusters

According to the comparison between the specific surface area of 2D sensitive membrane and 3D nanowires clusters, it can be seen that 2D sensitive membrane area is

(2)

The 3D sensitive membrane area can be expressed as follows:

(3)

(4)

Ast=2 μm,h=5 μm,d=0.5 μm, ifs=1 mm2, thenS1=1.000 mm2,S2=8.85 mm2. The specific surface area of the 3D nano structure is more than 9 times larger than that of the traditional 2D structure. Correspondingly, the sensitivity per unit area of the 3D nano structure will be more than nine times as much as that of the 2D nano structure. All threshold values are held equal, and the increase in sensitivity will be translated into a faster response time.

Fig.4 illustrates the 3D SAW gas sensor under discussion, which uses ZnO nanowires as the sensitive material[11].

Fig.4 New type of 3D nano-SAW sensor

3 Design of the SAW device

In general, there are two types of SAW gas sensors, including delay surface acoustic wave(SAWD) and resonant surface acoustic wave(SAWR). SAWR is generally characterized with highQ-factor and low insertion loss. As a device affected by frequency, SAWR is not only easy to resonate but also has good frequency stability. The SAW gas sensor studied here refers to a SAW resonator.

Through the Agilent’s Advanced design system platform simulation , selected parameters are as follows: IDT(inter digital transducer) number is 50, aperture width as 100λ, IDT period as 6.154 4 μm, other parameters keep default.

For SAWR, the location of the resonator can significantly influence its sensitivity. Traditionally, signals are processed using a two-port SAW resonator, which, however, is not the optimal choice for a SAW gas sensor because the distance between these two IDTs is too small. Fig.5 illustrates how the varying distance between the IDTs affects the properties of the resonator. When the distance is 100 wavelengths, there is the maximum harmonic suppression.

Fig.5 Impact of the distance between the IDTs on the properties of the resonator

4 Formation and modification of three-dimensional nanowires

There are different methods to grow the desired pattern of the 3D nanowires, such as hydrothermal synthesis, template synthesis, and electrochemical precipitation. The hydrothermal synthesis stands out with its low-cost and low-temperature characteristics as well as its ability to be repeated and mass produced in a non-vacuum environment. Using this method, nanowires can grow on a photolithographic substrate coated with photoresist (such as a SAW device). Its operating temperature is in the range of tens of degrees Celsius, which will not affect the IDTs on the SAW piezoelectric substrate. Furthermore, the method is easy-to-conduct and suitable for the mass production.

Fig.6 and Fig.7 show the growth process of nanowires and the crystal layers using hydrothermal synthesis and template synthesis. Firstly, the SAW device will be spin-coated or sputtered with ZnO nano particles. As s result, it can be used as the seed layer for the growth of ZnO nanowires. To grow ZnO nanowires that are spaced in a particular way, the entire SAW device will be firstly filmed with PMMA (polymethyl methacrylate). Then, it will be etched by electron beams according to the particular distance and size. The SAW device, which has been coated with PMMA, has gone through the photolithographic process. Then, it will be dipped into a mixed liquor for 18 hours for hydrothermal synthesis[12-13].

Fig.6 Growth process of ZnO nanowires on the SAW device

Fig.7 ZnO crystal layer on the surface of the SAW device.

From Fig.6 and Fig.7, it can be found that the structure and density of nanowires are not easy to control. However, these flaws can be overcome using template synthesis, by which the position and density of the nanowires will be controlled. Moreover, nanowires produced in this way usually stand more upright. The PMMA template, which has undergone the photolithographic process, will not be damaged in the low-temperature hydrothermal process. The pattern of the nanowires is of critical importance to their subsequent modification as well as the fabrication of the gas sensor.

Fig.8 Modification effect of ZnO nanowires

The formation and modification of the nanowires on the surface of the SAW device will lead to greater insertion loss as -29.30 dB , and a shift in its resonance frequencies as 300 MHz, through the Agilent vector analyzer experimental, the results are shown in Fig.9.

Fig.9 Display of SAW property after the nanowires have been formed and modified

5 Test results of 3D SAW gas sensor and the analysis

Sarin and HD gas are pumped into the sample pool of the 3D nano-SAW gas sensor whose SAW resonators are coated with SE-30 (Silicone), PTFP (a hydrogen bond acidic polysiloxane), BSP3 (a hydrogen bond acidic polysiloxane) or PECH (a poly propylene oxide). Besides, the typical response of the SAW gas sensor to the mix of Sarin and HD gas can be captured in Fig.10.

Fig.10 Responsive curves of the SAW sensor

By using Back-Propagation Network algorithm[10,14], the parameters are listed as follows:

• the neural network transfer function: tansig

• training function: trainscg

• output function: purelin

• BP network structure: 4-20-20-2

3.1.1 前列腺解剖 复习系统、局部解剖，结合实时超声图像，让学生熟悉前列腺不同区带及相邻的精囊、尿道等结构，前列腺体积测量等。

Then, 58 groups of experimental data were selected as the training sample data. The part of the data is displayed in Tab.1.

Tab.1 Training data for quantitative analysis of mixed components

The test results are shown in the Fig.11. and Fig.12. The ‘o’ indicates the value of the gas concentration in the calibration, and the ‘+’ indicates the predicted gas concentration.

Fig.11 Results of mixed qualitative judgment by test data

Fig.12 Test data errors of mixed gas qualitative judgment

As shown by the test results, the 3D nano-SAW gas sensor has high targeting capacity and disturbance resistance. Besides, comprehensive recognition rate of training sample and test sample data，the overall distinguishable rate is more than 90%.

6 Conclusions

By integrating the nanostructure technology and the SAW technology, this paper attempts to design a new type of three-dimensional nano-SAW gas sensor by integrating the nanostructure technology and the SAW technology. The research has comprehensively have been discussed in detail: the design of SAW chipas well as the growth and modification of nanowires. Finally, by using back-propagation network algorithm to recognize the mixed gas with HD and Sarin, the experiment shows that SAW gas sensor can gain a high degree of detecting capacity and disturbance resistance.

[1] FU Y Q, LUO J K, DU X Y, et al. Recent developments on ZnO films for acoustic wave based bio-sensing and microfluidic applications: a review[J]. Sensors & Actuators B Chemical, 2010, 143(2):606-619.

[2] JEONG H Y, LEE D S, HONG K C, et al. Flexible room-temperature NO2 gas sensors based on carbon nanotubes/reduced graphene hybrid films[J]. Applied Physics Letters, 2010, 96(21):213105-1-213105-3.

[3] DU J, LIANG D, TANG H, et al. InAs nanowire transistors as gas sensor and the response mechanism[J]. Nano Letters, 2009, 9(12):4348-4351.

[4] STRELCOV E, LILACH Y, KOLMAKOV A. Gas sensor based on metal-insulator transition in VO2 nanowire thermistor[J]. Nano Letters, 2009, 9(6):2322-2326.

[5] GONG J, LI Y, CHAI X, et al. UV-Light-Activated ZnO Fibers for Organic Gas Sensing at Room Temperature[J]. Journal of Physical Chemistry C, 2009, 114(2):1293-1298.

[6] KESKINEN H, TRICOLI A, MARJAMAKI M, et al. Size-selected agglomerates of SnO2 nanoparticles as gas sensors[J]. Journal of Applied Physics, 2009, 106(8):084316-1-084316-8.

[7] ARNOLD S P, PROKES S M, PERKINS F K, et al. Design and performance of a simple, room-temperature Ga2O3 nanowire gas sensor[J]. Applied Physics Letters, 2009, 95(10):103102-1-103102-3.

[8] WANG W, HU H, LIU X, Et al. Development of a Room Temperature SAW Methane Gas Sensor Incorporating a Supramolecular Cryptophane A Coating[J]. Sensors, 2016, 16(1):73.

[9] JAKUBIK W, POWRONIK P, WROTNIAK J, et al. Theoretical analysis of acoustoelectrical sensitivity in SAW gas sensors with single and bi-layer structures[J]. Sensors & Actuators B Chemical, 2016, 236:1069-1074.

[10] BEIGY H, MEYBODI M R. Adaptation of Parameters of BP Algorithm Using Learning Automata[C]// Neural Networks, 2000. Proceedings. Sixth Brazilian Symposium on. Rio de Janeiro, RJ, Brazil, Brazil: IEEE, 2000: 24-31.

[11] MA J, MAO S, WANG L, et al. Surface Acoustic Wave (SAW) Ammonia Gas Sensor Based on the ZnO Nanorod Array[J]. Sensor Letters, 2016, 14(7): 673-677.

[12] JIANG X, ZU Xihong, ZHANG Z, et al. Progress in Preparation and Application of Patterned ZnO Nanowire Arrays[J]. Materials Review, 2013(15):129-135.

[13] GAUTAM C, TIWARY C S, MACHADO L, et al. Synthesis and Porous h-BN 3D Architectures for Effective Humidity and Gas Sensors[J]. Rsc Advances, 2016, 6(91): 87888-87895.

[14] JIA L H, ZHANG X R. Analysis and Improvements of BP Algorithm[J]. Computer Technology & Development, 2006, 16(10): 101-103,107.

Biographies:

QI Jing(1983-) was born in Jiangxi Province, China. Electrical & Information Engineering from Chongqing University of Posts and Telecommunications in 2005, and the ME degree in Electrical & Information Engineering from Chongqing University in 2008. Now, he is working towards his PHD in School of Chongqing University. His research interests include sensors and instrumentation, energy harvesting circuit, short distance wireless communication, sensor search. E-mail: qijing@cqupt.edu.cn.

WEN Yumei(1964-) was born in Chongqing, China. She received the BE degree in electrical engineering from Beijing Aeronautic and Astronautic University in 1984, the ME degree in electrical engineering from China Academy of Launch Vehicle Technology in 1987, and the PhD degree in instrumentation engineering from Chongqing University in 1997. She has been a professor at College of Optoelectronic Engineering in Chongqing University since 1998. Her research interests include sensors and instrumentation, signal processing, energy harvesting devices, and LED lighting.

(编辑：魏琴芳)

2016-10-23

2017-05-09 通讯作者：漆 晶 qijing@cqupt.edu.cn

基于声表面波谐振器的三维纳米团簇敏感薄膜关键技术研究

漆 晶1,2，文玉梅1，李 平1

(1.重庆大学 光电工程学院，重庆 400044; 2.重庆邮电大学 移动通信重点实验室，重庆 400060)

设计了一种新型基于声表面波技术的气体传感器，理论分析了三维纳米线结构的比表面积大、灵敏度高等优点，采用具备高Q值和低插损的谐振型声表面波器件结构，制备了三维敏感膜结构的声表面波气体传感器。在此基础上，为提高吸附效应，对三维纳米线簇进行了修饰改进。通过将沙林气和芥子气注入放置了声表面波的气体传感器密闭腔体内，经过神经网络识别系统进行定性识别。实验结果表明，基于修饰改进后的纳米线簇敏感膜制备的声表面波气体传感器对给定毒气混合气体的整理识别率大于90%，能够满足通用的毒气定性检测要求。并且三维纳米声表面波气敏传感器的灵敏度和响应速度优于传统的传感装置，在识别系统加大样本数据量时，能够进一步提高识别精度。

声表面波；纳米线簇；气体传感器；神经网络识别算法

10.3979/j.issn.1673-825X.2017.04.011

TN65; TM93 Document code：A

1673-825X(2017)04-0494-06

The Science and Technology Project Affiliated to the Education Department of Chongqing Municipality(KJ1500433)