Performance Optimization And Application Research Of Radiation Detectors With Fast Time Resolution

Particle radiation detection technology with ultra-fast time resolution plays an important role in the fields of high-energy physics and medical physics.There are four key technologies that affect the time resolution of radiation detectors:fast scintillation materials,ultrafast photoelectric devices,high-speed data acquisition systems,and data analysis algorithms.Among them,the time performance of ultrafast photoelectric devices is particularly important.To improve the time resolution of radiation detectors,this dissertation uses fast timing photomultiplier tubes（FPMT）as the core photoelectric device for radiation detectors.This dissertation focuses on the structural design,optimization,and performance study of FPMT applied in radiation detectors.A design concept is proposed for using FPMT to detect Cherenkov light as the core device for radiation detectors.In order to improve the time performance of radiation detectors,the research content of this dissertation includes:This dissertation compares the scintillation and Cherenkov radiation principles,and theoretically analyzes the superiority of Cherenkov radiation as a luminescent material for radiation detectors.The detection principle of FPMT and the principle of waveform sampling-based data acquisition systems are explained.The data analysis methods for studying the performance of FPMT using waveform sampling technology are proposed.For the digitized waveform obtained by waveform sampling technology,three digital timing methods are compared,and the comparison results show that the digital constant fraction timing algorithm is more suitable for studying the time performance of radiation detectors based on FPMT,laying a foundation for the time performance research of radiation detectors in this dissertation.The structural design and optimization of FPMT has been studied in this dissertation.Starting from the drift characteristics of electrons inside FPMT,the time fluctuations generated by electrons during drift and multiplication processes are analyzed.A finite element integration technique is used to establish a model of FPMT.The design of a conical anode structure for single-anode FPMT and an 8×8 anode structure for multi-anode FPMT was proposed.And the time performance of FPMT is optimized from both structural and voltage aspects.The rise time of the optimized single anode FPMT is improved from 1.4 ns to 153 ps,and its single-photon TTS is optimized to 27.2 ps.The performance of FPMT has been studied in this dissertation.The performance of the optimized single-anode FPMT was compared and tested with advanced tubes of the same type internationally.The results showed that the single-photon detection capability and time performance of the optimized single-anode FPMT had reached the international advanced level,verifying the rationality of the structural and voltage optimization.A performance study was conducted on the ultrafast photomultiplier tube with an 8×8anodes structure,and a uniformity evaluation method for multi-anode FPMT was proposed.An 8×8 anodes FPMT with single-photon time resolution uniformity of less than 40 ps was developed.In order to further improve the coupling between luminescent materials and FPMT,a design of an 8×8 anodes Cherenkov radiator window ultrafast photomultiplier tube（CRW-FPMT）was proposed,with a single-photon time resolution of less than 40 ps,which can be used for direct particle detection.This research provides new ideas for improving the time performance of radiation detectors based on Cherenkov light detection.This dissertation focuses on the time performance study of radiation detectors.Taking TOF-PET as an example,the core unit of radiation detectors based on FPMT coupled with luminescent materials was studied for coincidence time resolution（CTR）.The experimental results show that FPMT coupled with a 3×3×5 mm³ LYSO can achieve a CTR of 92.6±1.6 ps,which is better than the results of Si PM coupled with the same size LYSO crystal in the literature.FPMT coupled with a 3×3×5 mm³ Cherenkov radiation crystal Pb F₂ can achieve a CTR of 71.9±6.6 ps,verifying the rationality and superiority of using FPMT coupled with Cherenkov material as the core unit of radiation detectors.The CTR study of radiation detectors based on CRW-FPMT was carried out,and a CTR of 137.5±11.5 ps was achieved.In addition,to address the issue of side peak events in the CTR results based on FPMT,this article proposes using a convolutional neural network（CNN）algorithm to optimize the CTR.The results show that the CNN algorithm can effectively correct the side peak events without affecting the counting rate,improving the system’s time resolution from 57.4 ps to 28.5 ps,providing a reference for the future application of machine learning algorithms in radiation detectors.