Research On Gas Detection System Based On Semiconducting Carbon Nanotube Thin Film Sensors

Rapid and continuous monitoring of toxic and explosive gases（i.e.,H₂,H₂S,NO₂）in the environment is of great importance to human life and industrial safety.Advanced gas sensing technology requires sensors with high sensitivity,miniaturization and low power consumption.Current mainstream commercial gas sensors based on resistive metal oxide semiconductors（MOS）,which are in bulk and operated at temperature higher than 200°C,are not favored in the advanced gas sensing technology.Recently,extensive investigations have been made on the use of nanomaterials,including nanowires,nanotubes and two-dimensional materials,to construct gas sensors with high sensitivity and high selectivity at room temperature.Among them,carbon nanotubes（CNTs）are considered as one of the excellent nanomaterials for the construction of highly sensitive chemical and biosensors due to their ultrathin structure,high specific surface area,high carrier mobility and high stability.Unfortunately,first,most nanomaterials have some critical defects such as poor material uniformity and reproducibility of device properties and difficulty in mass production.Second,the nanomaterials based gas sensors have the problem of long recovery time at lower operating temperature.Third,due to the high specific surface area of nanomaterials,their high affinity to gas molecules makes it difficult for the sensor to reach the equilibrium state in a short time,resulting in a long detection time.In order to detect gas continuously and rapidly in practical application,the gas sensor must overcome the above problems.For the structure of gas sensors,because the FET based sensor generally works at room temperature and has high sensitivity,and the introduction of the third gate can also provide a more flexible method to control the performance of the sensor,so it can meet the requirements of our practical application.In this work,we have batch prepared CNT thin film transistor（CNT-TFT）based gas sensor with highly uniform device performance,and realized the rapid detection of H₂,H₂S and NO₂ through a variety of modifications on the surface of CNT film.For the slow recovery of the sensors at room temperature,we propose a gate assisted recovery scheme.We only need to apply a certain voltage on the bottom gate,and through the electric field generated by the bottom gate to reduce the surface barrier of gas molecule desorption and accelerate the desorption of gas molecules,so as to realize the rapid recovery of gas,and the recovery speed is an order of magnitude higher than the natural recovery.Moreover,we can control the active and sleep state of the device by applying different values of the bottom gate voltage,so as to realize accurate detection of the sensor in the environment with the target gas all the time.In the experiment,we found that the maximum slope of the response curve of gas with different concentrations can reflect the concentration of gas the fastest.Therefore,we can quickly obtain the target gas concentration according to the response speed of the gate assisted control device in the active mode,without waiting for the equilibrium state.In order to show the feasibility of practical application of CNT-TFT based gas sensor,we built a portable detection system,combined with the gate assisted scheme,to achieve the rapid detection of H₂,detection resolution of 10 ppm,detection time of less than 10 s and rapid recovery time of 50 s at room temperature,The CNT-TFT based gas sensor has been successfully promoted to the practical application level.Compared with other reported nanomaterial based H₂ sensor detection systems,our system performs outstanding.We also consider that some problems existing in a single gas sensor（such as selectivity and humidity influence）are difficult to be solved in a short term,so we prepared the CNT-TFT based gas sensor array and test the cross sensitivity of the sensor array to a variety of gases.It is hoped that the measurement performance of the sensor can be improved by algorithm compensation.