In this paper, to address the problems of large blood draws, long testing times, and the inability to achieve dynamic detection of invasive testing for diabetes, stemming from the principle that type 1 diabetic patients exhale significantly higher levels of acetone than normal people, a FAIMS-MEMS gas sensor was designed to detect acetone, which utilizes the characteristics of high sensitivity, fast response, and non-invasive operation. It is prepared by MEMS processes, such as photolithography, etching, and sputtering, its specific dimensions are 4000 μm in length, 3000 μm in width and 800 μm in height and the related test system was built to detect acetone gas. The test results show that when acetone below 0.8 ppm is introduced, the voltage value detected by the sensor basically does not change, while when acetone gas exceeds 1.8 ppm, the voltage value detected by the sensor increases significantly. The detection accuracy of the sensor prepared by this method is about 0.02 ppm/mV, and the voltage change can reach 1 V with a response time of 3 s and a recovery time of 4 s when tested under 20 ppm acetone environment; this has good repeatability and stability, and has great prospects in the field of non-invasive detection of type 1 diabetes.
Keywords: FAIMS-MEMS; acetone; diabetes; reproducibility; stability.