低功耗一氧化碳传感器研究进展

Research progress of low-power carbon monoxide sensors

  • 摘要: 低功耗一氧化碳传感器是煤炭安全开采的重要保障和分布式无线传感技术的关键基础。阐述了低功耗一氧化碳传感器的工作原理,重点介绍了电化学和半导体一氧化碳传感器的最新研究进展,分析了它们的优缺点,展望了低功耗一氧化碳传感器的发展方向和前景。① 电化学一氧化碳传感器主要由电极和电解质构成,有两电极和三电极结构,两电极电化学一氧化碳传感器没有参比电极,结构简单,易于设计和制造,成本较低,适用于低浓度一氧化碳的监测;三电极电化学一氧化碳传感器引入参比电极,具有较宽的量程和较高的精度,但成本高。电化学一氧化碳传感器响应时间较短,具有分布式无线传感应用前景,其研究重点在于铂-碳复合电极的制备,但电化学一氧化碳传感器仍然使用了水溶剂,存在漏液风险,且难以微型化;未来的发展趋势一方面要避免采用液体溶剂,研究全固态电化学一氧化碳传感器,并控制其孔隙率,缩短响应与恢复时间,另一方面要寻找廉价高效的电极材料,降低传感器成本;此外,制造微型化的电化学一氧化碳传感器也是未来重要的工作内容。② 半导体一氧化碳传感器可分为非微加热板型和微加热板型。非微加热板型一氧化碳传感器采用的是陶瓷管基底或者氧化铝平板基底,由于加热器尺寸较大,且热传导严重,功耗较高,通常在100 mW以上,难以用于分布式无线传感场合。微加热板型一氧化碳传感器热质量低,可以大大降低加热器运行功耗,还可与集成电路工艺兼容,可得到片上系统型一氧化碳传感器,降低外围电路功耗,适合物联网应用。微加热板型一氧化碳传感器体积比固体电解液电化学一氧化碳传感器更小,且具有成本低廉、灵敏度高、易于片上集成等优点,但易受湿度影响,基线易漂移;未来研究方向是对敏感材料进行修饰改进,优化封装工艺,并采用智能算法对基线进行自校准;可以使用沸石修饰敏感材料减少湿度对传感器的影响,或者在封装工艺上采用聚四氟乙烯等的憎水膜;智能算法可以采用支持向量机和人工神经网络结合算法或人工智能算法,未来希望可以实现高效率的边缘计算,以提高效率和准确率。

     

    Abstract: Low-power carbon monoxide sensor is an important guarantee for safe coal mining and a key foundation for distributed wireless sensing technology. In this paper, the working principle of low-power carbon monoxide sensor is described, and the latest research progress of electrochemical and semiconductor carbon monoxide sensor is introduced. This paper analyzes their advantages and disadvantages and proposes the development direction and prospects of low-power carbon monoxide sensor. ① The electrochemical carbon monoxide sensor is mainly composed of electrodes and electrolyte. There are two-electrode structure and three-electrode structure. The two-electrode electrochemical carbon monoxide sensor has no reference electrode. It has a simple structure and is easy to be designed and manufactured. The sensor has a low cost and is suitable for the monitoring of low-concentration carbon monoxide. The three-electrode electrochemical carbon monoxide sensor introduces a reference electrode with a larger range and higher accuracy, but the cost is high. The electrochemical carbon monoxide sensor has a short response time and has the prospect of distributed wireless sensing applications. The research focus is on the preparation of platinum-carbon composite electrodes. However, the electrochemical carbon monoxide sensor still uses liquid solvent, which has the risk of liquid leakage. Moreover, it is difficult to realize micro-miniaturization. The future development trend is to avoid the use of liquid solvents, to study all solid-state electrochemical carbon monoxide sensors, and to control the porosity to shorten the response and recovery time. On the other hand, the trend is to find cheap and efficient electrode materials to reduce sensor costs. In addition, manufacturing micro-miniaturized electrochemical carbon monoxide sensor is also an important part of future work. ② The semiconductor carbon monoxide sensor can be divided into non-micro-heating plate type and micro-heating plate type. The non-micro-heating plate type carbon monoxide sensor uses a ceramic tube substrate or an alumina plate substrate. Due to the large heater size, severe heat conduction and high power consumption with usually above 100 mW, it is difficult to be used in distributed wireless sensing occasions. The micro-heating plate type carbon monoxide sensor has low thermal quality, which can greatly reduce the operating power consumption of the heater. It is also compatible with integrated circuit process. A system-on-chip carbon monoxide sensor can be obtained, which reduces the power consumption of peripheral circuits and is suitable for IoT applications. The micro-heating plate type carbon monoxide sensor is smaller than solid electrolyte electrochemical carbon monoxide sensor, and has the advantages of low cost, high sensitivity and easy on-chip integration. However, it is easily affected by humidity and the baseline is easy to drift. The future research direction is to improve the modification of sensitive materials, optimize the packaging process, and use intelligent algorithms to self-calibrate the baseline. It is proposed to use zeolite to modify sensitive materials to reduce the impact of humidity on the sensor, or use a water-repellent membrane such as polytetrafluoroethylene in the packaging process. The intelligent algorithm can use a combination of support vector machines and artificial neural networks or artificial intelligence algorithms. In the future, it is hoped that high-efficiency edge computing can be achieved to improve the efficiency and accuracy.

     

/

返回文章
返回