The Research On Time-resolved Luminescence Detection And Imaging Techniques

Luminescence detection and imaging have been widely used in chemical,biological,environmental researches and clinical diagnosis,while most of them are steady-state measurements.The detected signal of the steady-state measurements always contains fluorescence,phosphorescence in different time domains and various scattering light,so it is difficult to obtain luminescence lifetime information.In contrast,time-resolved techniques can detect the luminescence change in time domain and eliminate the background signals from scattering and short-lived autofluorescence.However,the relative detection systems always require pulsed excitation sources and high-speed detectors.In addition,the excitation and detection shutter should be synchronized precisely,further leading to high-cost and high-complexity.In order to accomplish low-cost and miniaturized time-resolved luminescence detection instruments,some new time-resolved luminescence detection methods and applications are developed.Then,to improve the resolution of the luminescence microscopy,two super-resolution methods based on the time-resolved techniques and deep learning are also developed.The main research work is shown below.(1)The circuitry-free auto-phase-locked time-gated luminescence detection method for spectrally resolved luminescence lifetime detection is developed,where one mechanical chopper acts as pulse generator and detecting shutter simultaneously.The phases of each excitation and detection gate are synchronized automatically and a simple phase difference adjustment method is also developed.This method can overcome the phase-mismatching and reduce the scattering signals caused by frequency jitter.(2)A miniaturized microsecond-resolved smartphone time-gated luminescence spectroscopy is developed,where one mechanical chopper is used as the detection shutter and an optical coupler is placed at the edge of the wheel to convert the chopping signal into a transistor-transistor logic signal which is used to control the excitation source and achieve synchronization.A reflective diffraction grating disperses the light onto the image sensor of the smartphone,and the time-gated luminescence spectra can be obtained by analyzing the signal intensity at different pixels of the image captured by the smartphone.This method reduces the reliance on expensive laboratory instruments of the time-resolved technique.(3)An apparatus based on a smartphone for microsecond-resolved global luminescence lifetime imaging is developed.One mechanical chopper is used as the detection shutter and the signal converted from an optical coupler across the chopper wheel is used as the reference signal to control the excitation source through a field programmable gate arraybased digital control circuit.(4)An all-fiber time-resolved luminescence detection method is developed,where all the excitation light and luminescence are transmitted in optical fiber.An optical fiber coupled laser diode and a fiber optical switch are used as the excitation source and detection shutter,respectively.In addition,the influences of different detection parameters on the system performance and resolution are also researched.This method has characteristics of low cost,small size,ease of installation,ease of integration and low spatial coherence.(5)A super-resolution microscopy method based on the characteristics of time-resolved technique is developed.The lifetimes of luminscencet particles could be obtained with timeresolved images,and the image could be separated into several layers according to these luminescence lifetimes.After the resolution of each layer is improved,the super-resolution image could be obtained by superimposing them.It is the first time that the super-resolution microscopy was accomplished by utilizing luminescence lifetime information,further expanding the application of time-resolved technique and the super-resolution microscopy technique.(6)A dual-step microscopy resolution improvement method based on neural network is developed and its performance is improved with the help of time-resolved technique.Since the neural networks are used to achieve image translation with minor resolution difference,they could extract features more easily and accurately.