Research On Flexible Humidity Sensor For Neuromorphic Olfactory System

The human perception system has been tested by the external environment for a long time and has unparalleled perception ability.These capabilities are specifically reflected in lower power consumption,good anti-interference performance,and good biocompatibility.The flexible humidity sensor for the neuromorphic olfactory system is a biomimetic flexible functional device,mainly inspired by the functional chain of human perception,information transmission,and signal processing.This project has prepared two parts: a humidity sensor and an electrochemical neural morphology machine as perception modules and an information processing module,and achieved information transmission through MCU.The device has a small overall size,accurate respiratory pattern recognition,and high biocompatibility,which can achieve human health detection.Firstly,a resistive humidity sensor using two-dimensional material GO as the humidity sensitive material was fabricated,and the electrode structure adopts a high sensitivity and high signal-to-noise ratio finger insertion electrode structure.Among them,GO was prepared by improved Hummer method,and its composition and morphology were characterized by Raman spectrometer and atomic force microscope.The prepared GO surface contains hydroxyl functional groups that can be physically adsorbed by hydrogen bonding with water molecules in the air.As the humidity increases,the main conductive mechanism of GO films shifts from the inherent electronic conduction of GO to the ion conduction generated by adsorbed water molecules,resulting in a significant change in film conductivity.In addition,due to its large specific surface area,good biocompatibility,and unique benzene ring/oxygencontaining structure,GO provides guarantee for high-performance humidity sensing(response time<1s,and can still maintain good performance after repeated loading for1200s).Secondly,a three terminal electrochemical neuromorphic organic device was fabricated,whose channel conductivity is regulated by the gate voltage and maintains a relatively stable value after removing the gate voltage.The device materials are gate(PEDOT: PSS),proton exchange membrane layer(Nafion),and source drain channel layer(PEDOT: PSS).As the polarity of the gate voltage changes.Protons in the gate material can enter the source drain channel through the proton exchange membrane,regulating the different redox states of the channel material(PEDOT: PSS).The adjustment of this conductivity state will not disappear when the gate voltage is removed.This device can still significantly change its conductivity state under a small driving voltage,making it a good artificial synaptic device(the entire device displays>100 different non volatile conductivity states,and can still maintain good stability after 5000 s of gate regulation).Finally,through an integrated process,the GO humidity sensor,MCU,and three-terminal electrochemical neural form organic device were prepared on the same substrate,and the connection of the three modules was completed through printed circuits,reducing the redundancy of the circuits between different modules of the neural form humidity sensor and providing a guarantee for subsequent applications on wearable devices.In order to verify the possibility of using neuromorphic humidity sensors as human respiratory detection,this project systematically constructed an experimental platform.The humidity sensor signals,MCU input/output signals,and electrochemical neural morphology channel conductivity signals were collected separately to ensure the rigor of the experiment,and these signals remained consistent over time.Firstly,two sets of breaths with different frequencies and amplitudes were applied to the device,and the three terminal electrochemical nerve morphology was clearly distinguished from the channel conductivity state of the mechanical component.Then,in order to simulate neural learning,the neural morphology humidity sensor was stimulated with respiratory pulses of the same frequency and different times.After 35 stimuli,the conductivity value of the device remained basically unchanged.Finally,in order to verify respiratory detection in complex situations,a doping method(high-frequency breathing and lowfrequency breathing combined)was used to test the neural morphology humidity sensor.The experimental results indicate that even with a 5% difference,respiratory behavior can be reorganized in complex respiratory patterns.The overall structure of the neuromorphic humidity sensor is compact(15mm × 25mm),with good performance and certain application value in human health monitor.