Study Of Nerve Agents Trace Vapor Surface Acoustic Wave Sensors

Since the end of last century, the terroristic activity involving Chemical WarAgents (CWAs) has been an issue drawing attention of all over the world. To satisfy theincreasing demand, the field testing technologies detecting the CWAs have beendeveloped rapidly.The most effective strategy against the chemical terroristic attack is to raise alarmand take necessary actions before a big number of casualties occur. This would assureminimum accident magnitude and social impact. To achieve that, the detectingtechnology should have these features: high senstivity, fast response, high recognitioncapability, low cost, ease to operate and high mobility. Among various sensortechnologies, surface acoustic wave(SAW) sensor is considered as a promisingcandidate because it can meet those requirements.This dissertation focuses on a SAW sensor array detecting the CWAs. To enhancethe sensitivity, the investigation of device structure and the optimum of polymersensitive layer coating process have been carried on. Furthermore, the circuits wasdesigned and fabricated for application of SAW sensor array. Finally, several sensorswere combined into an array to improve the recognition capability. The main results areas follows:1. Love wave device was fabricated by depositing ZnO guiding layer on surfacetransverse wave(STW) device, and the characteristics of the device has beeninvestigated in detail to obtain a device with high mass sensitivity and low insetion loss.Scattering matrix theory was used for simulation of STW device, and the resultscoincided with the corresponding experiments. To achieve the highest degree of c-axisorientation of ZnO layer, the sputtering parameters were optimized, and the results wereas follows: Ar:O2=1:1, substrate temperature300℃, sputtering pressure2Pa. In thiscondition, the deposition rate was600nm/h. Insertion loss of the device decreasedoriginally with the increasing thickness of ZnO guiding layer, and reached the minimumat the thickness of300nm, then began to increase. Mass sensitivity increased originallyand reached the maximum,625cm2g1, at the thickness of600nm, then began to decrease. Quality factor (Q value) decreased monotonously. The obtained Love wavedevice was much better than conventional devices in terms of insertion loss and Qvalue.2. A design method of SAW sensor oscillation loop has been investigated based onthe principal of oscillator. Coupling inductors and phase shift filters were used tocompensate phase of the loops obtained by measuring S21parameter of thecorresponding open loops. To precisely measure the frequency signal of sensors andeliminate temperature noise, a dual channel SAW sensor oscillation circuit was designed.It contained two oscillation loops, a mixer, a low pass filter and a TTL shaping circuit.To shielding electromagnetic interference and decrease response time of sensors, ametal testing chamber was designed, of which volume was only100L. The frequencyshift of its output signal was less than30Hz during a period of2000seconds,which metthe requirement of SAW sensor applications. A circuit used for SAW sensor array wasdesigned based on experience of designing the dual channel oscillator circuit, in whichthree sensors and one reference device were contained. The testing chambers of thesesensors were arranged in line and connected, while the reference device was isolatedfrom the vapor analyte.3. The airbrush process used to deposit polymer sensitive layer on the device hasbeen investigated. In order to raise the layer thickness, thus, the sensitivity of sensor,airbrush process was optimized to improve roughness of the layer. Some conclusionswere found:(1) Size of droplets was diminished by decreasing solution flow, thereforthe layer roughness was improved.(2)The layer roughness could be improved bydecreasing concentration of solution;(3)The layer roughness was dependent on distancebetween nozzle and device.4. The investigation results mentioned above were integrated into a SAW sensorarray, and the qualitative analysis of CWAs were achieved by combination of theobtained sensor array and principal component analysis(PCA). In this case, poly{methyl[3-(2-hydroxyl,4,6-bistrifluoromethyl) phenyl]propylsiloxane}(DKAP),polyepichlorohydrin(PECH) and ethylcellulose(ECEL) were selected as sensitivematerials. Those three sensors were exposed to different organic vapors simultaneously,and each vapor had a unique response pattern. DKAP coated sensor had a sensitivity of30078Hz/(mg/m3) to DMMP vapor and the corresponding theoretical Limit of Detection(LOD) was0.00299mg/m3. Principal component analysis method was used toanalyze the response patterns of dimethyl methylphosphonate(DMMP),dichloropentane(DCP), N, N-dimethylformamide(DMF),1,2-dichloroethane, toluene,kerosene and ethanol. The result indicated that DMMP and DCP, used as simulants ofnerve agents and vesicant agents, could be distinguished from other vapors, and96.154%information was presented with two principal components. Hence, thosesimulants mentioned above can be clearly recognized by the sensor array.