| Metal oxide semiconductor nanomaterials have been widely used in gas sensor re-search due to their strong conductivity,high sensitivity,stable performance,low power con-sumption,and ease of integration.However,current metal oxide semiconductor gas sensors often suffer from drawbacks such as high operating temperatures and long response/recov-ery times.In this paper,SnO2nanoparticles were prepared using a bio-template method,and sheet-like SnO2,flower-shaped Cu/SnO2,ZnO nanorods,and Cu/ZnO nanorod gas-sensitive materials were prepared using a hydrothermal method.Their morphological fea-tures and composition were characterized and analyzed,their gas sensitivity performance was tested,and the gas sensitivity response mechanism was studied.A sensor array was self-assembled,and a BP neural network prediction model optimized by the sparrow algo-rithm was used to achieve quantitative prediction of three types of toxic gas mixtures.The research work in this paper is mainly divided into the following aspects:(1)SnO2gas-sensitive materials were synthesized using the bio-template method,and the SnO2materials consisted of irregular,dispersed nanoparticles.The flake SnO2gas-sensitive materials was synthesized using the hydrothermal method,showing a uniformly distributed nanosheet structure,and the flower-like Cu/SnO2gas-sensitive materials was a three-dimensional nanoflower structure assembled from nanosheets.The SnO2nanoparti-cles at 180℃ for 50 ppm ammonia with a response value of 37.5%and a response/recovery time of 71/28.8 s,respectively.The flower-like Cu/SnO2had a response value of 35.73%for50 ppm formaldehyde at 200℃ with a response/recovery time of 59.4/32.4 s,respectively,and had good reproducibility and long-term stability.(2)The ZnO and Cu/ZnO nanorods gas-sensitive materials were prepared by hydrother-mal method.The ZnO nanorods showed a three-dimensional radial nanorod structure,and the Cu/ZnO nanorods did not change the main structure of ZnO nanorods due to copper doping.The response value of Cu/ZnO nanorods to 50 ppm formaldehyde at 220℃ was51.06%,and the response/recovery time was 58.5/21.6 s,respectively,and had good re-producibility and long-term stability.The gas-sensitive response mechanisms of SnO2and ZnO nanomaterials were analyzed and studied.(3)The gas sensors made of SnO2and ZnO gas-sensitive materials were assembled into a sensing array to test the ternary gas mixture.The four features of maximum response value S,response time Tres,recovery time Trecand response area square were extracted and combined with the BP neural network prediction model optimized by the sparrow algorithm to achieve the prediction of the gas mixture concentration.In brief,this paper synthesizes SnO2and ZnO gas-sensitive materials,characterizes their microscopic morphology and composition,prepares gas sensors,tests their gas-sensitive performance,analyses the gas-sensitive mechanism of the gas-sensitive materials.The self-assembles a sensor array combined with a BP neural network model after the sparrow opti-mization algorithm in achieving quantitative prediction of mixed gas concentrations,which has good application prospects. |