Research On TiO₂ Nanotubes Ionization Sensor And Its Current Voltage Characteristics

The key component of gas detection is gas sensor.There are some problems in existing sensors such as large volume,complicated system,and high price,which is difficult to meet the current demand for gas detection.TiO2 nanotube sensor has become a research hotspot in the field of gas detection due to its advantages of low power consumption and easy integration.However,the current research is still in the exploration stage,and its discharge characteristics under the microstructure still need to be studied.This thesis attempts to study the micro-gap TiO2 nanotube ionization sensor and its volt-ampere characteristics.The contents are as follows:1)Anodic oxidation method is used to prepare TiO2 nanotubes and assemble them into micrometer-level pole spacing sensors.The gas discharge experiment in the air was conducted on the sensor,and it was found that the reduction of the inter-electrode distance reduced the order of magnitude of the initial voltage of the sensor.The threshold voltage measured at the 50-micron inter-electrode distance was about 25V.2)In order to analyze the discharge mechanism of the sensor,with the help of fluid mechanics simulation method,the self-sustained discharge currents of ionization sensors with different pole pitches are compared.The study shows that the small pitch structure can improve the stable discharge capability of the device.The influence of different curvature radius and different pole spacing on the volt-ampere relationship is analyzed,and it is found that both of them have a negative correlation at the same voltage.This also provides a theoretical basis for exploring the volt-ampere relationship of the sensor.3)In order to establish the volt-ampere relationship of the TiO2 nanotube ionization sensor,machine learning technology is introduced.Due to the small number of experimental samples,the processing performance of the data-driven machine learning method to establish the sensor voltammetric relationship on small samples is not efficient,so this article is based on the simulation and experimental data,the confrontation generation network expands the experimental samples,and the support vector machine is used to realize the establishment of the sensor's volt-ampere relationship.The research shows that the data augmented by the adversarial generation network can be used to build a sensor voltammetry prediction model,which improves the prediction accuracy of the sensor voltammetry relationship.Compared with BP network,support vector machine has a better prediction effect.In this paper,the voltammetry model of TiO2 nanotube ionization sensor is established by machine learning method,which provides theoretical guidance for the optimization of sensor performance,and is also of great significance to the intelligent development of sensors.