Effects of Mask on the Selectivity of Tin Oxide Gas Sensor

• Main Author: 謝正發
• Other Authors: 劉端祺
• Format: Others
• Language: zh-TW
• Published: 2001
• Online Access: http://ndltd.ncl.edu.tw/handle/62860425527122783313

碩士 === 國立臺灣科技大學 === 化學工程系 === 89 === To propose a way to solve the problem of poor selectivity of SnO2 semiconductor gas sensors was the goal of this research. The sensing devices utilized in this research were fabricated by screen printing. Three layers, namely the sensing layer, the insulation layer, and the filtering layer, were printed on the device. The materials of the layers were SnO2, SiO2, and SnO2 respectively. Methane and ethanol were the model gases in this study. The effects of covering the SnO2 sensing layer with an insulation SiO2 layer and with layers of SiO2 plus SnO2 were studied. Also investigated were the effects of adding Cr, Mn, Fe, Co, or Cu to the SnO2 filtering layer. A diffusion model was proposed in this research. The model was used to simulate the transport of ethanol in the filtering layer and to examine the blockage effect of the layer to ethanol. As indicated by the experimental data, masking the sensing layer with an insulation SiO2 layer could increase the in air resistance and the in 1000 ppm CH4 sensitivity of the sensor. The SiO2 layer could also lowered the temperature of maximum sensitivity from 400℃ to 350℃. Comparing to an unmasked device, a SiO2 masked sensor had about an order higher sensitivity in 1000 ppm EtOH at 300℃. In addition, the SiO2 mask could also decrease a sensor’s methane to ethanol selectivity and give the sensor a constant response time when sensing 1000 ppm CH4 at above 350℃. Comparing to a two-layer (insulation and sensing) sensor, a three-layer (filtering, insulation, and sensing) sensor had a much lower sensitivity in 1000 ppm EtOH, higher methane to ethanol selectivity, and about the same in air resistance and in 1000 ppm CH4 sensitivity. A straight line was obtained on a logarithm paper by plotting the sensitivity versus methane concentration data obtained at 500℃ from a three-layer sensor. A curve was obtained when the gas was changed to ethanol. The sensitivity of a three-layer sensor in 4000 ppm CH4 was 11.2 , lower than the 12.6 in 4000 ppm EtOH. The response time of a three-layer sensor was found inversely proportional to the concentration of methane. Adding Cr, Mn, Fe, Co, or Cu to SnO2 filtering layer could only slightly affect the sensitivity of a three-layer sensor in 1000 ppm CH4. However, the addition could remarkably change the sensitivity when the gas was switched to ethanol. The effect of the additives on the sensor’s methane to ethanol selectivity were in the order: Cu @ Mn > Fe @ Cr > Co > none. A straight line was obtained on logarithm paper by plotting the sensitivity against methane concentration data generated at 450℃from a three-layer sensor with Mn-SnO2 filtering layer. A curve was obtained when the gas was switched to ethanol. The sensitivity of the three-layer sensor with Mn-SnO2 filtration layer for 4000 ppm EtOH was 9.8. The number was about the same as the 9.7 for 200 ppm CH4.