| Surface-enhanced Raman scattering(SERS)has wide application prospects in some fields,such as environment detection,food safety and biomedicine with some advantages of high sensitivity,fast response,and strong specificity.The construction of SERS substrates with excellent sensitivity,uniformity and stability is a current research hotspot,and the key lies in how to efficiently and controllably construct regular and ordered micro-nano hotspot arrays.In this paper,systematic research is carried out from the three aspects of "microstructure arraySERS substrate-fast and accurate detection method",focusing on key issues which contains hot embossing process experiment and simulation of silver-based micro-pyramid structure array,rapid manufacturing of high-performance SERS substrate and fast and accurate detection of multi-material SERS based on machine learning.According to the requirements of SERS substrate for efficient and controllable manufacturing of ordered micro-nano hot spot array,silver sheets are selected as SERS functional substrate to explore the high-efficiency hot embossing process law and forming mechanism of micro pyramid structure array,so as to lay a foundation for efficient and controllable manufacturing of ordered microstructure array.Firstly,through single factor experiment,the effects of roller gap,roller temperature and roller speed on forming results are analyzed,and two forming defects are found:platform defect and no edge defect.On this basis,with the help of Design Expert software and Box-Behnken orthogonal experimental method,the coupling effects of process parameters such as roller gap,roller temperature and roller speed are analyzed,and the optimal forming process window is obtained.In order to further reveal the forming defect mechanism of micro pyramid structure,establish the simulation model of micro structure forming process,accurately simulate the two main process links of mold cavity filling and structural demolding,and analyze in detail the influence of point tracking function of software on different process parameters,structural height change,equivalent stress change and temperature change,so as to reveal the forming defect mechanism,It provides theoretical guidance for the controllable manufacturing of micro pyramid array.Based on the prepared ordered micro pyramid structure array,further research on the fabrication of high-performance SERS substrate is carried out.Firstly,silver nanoparticles are grown on the surface of microstructure by Lee-Meisel method.The SERS performance of different pyramid structures is compared to analyze the influence of morphology.Then,the electromagnetic field simulation is carried out based on FDTD solutions software to analyze the enhancement principle of SERS substrate,so as to provide theoretical guidance for the manufacture of high-performance SERS substrate.Then,the sensitivity,uniformity and stability of SERS substrate is characterized with Rhodamine 6G.The limit concentration can be obtained is 10-12 M,and the enhancement factor is calculated to be 8.8×109.Finally,the pesticide thiram is selected for rapid detection.In practical application,the sample to be tested often contained a variety of substances.In order to realize the rapid and accurate detection of different substances,this paper effectively combined SERS spectra and machine learning to realize the rapid and accurate detection of multiple substances.Selected the published SERS spectral data sets of 63 common human metabolites,preprocessed the spectra,enhanced the data by offset,noise and weighted summation,and reduced the dimension of the data by principal component analysis.Then input the data into different machine learning models,combined with the model evaluation method,selected the optimal machine learning model to realize the rapid and accurate classification of various metabolites. |
PDF Full Text Request