Optimization Designs And Applications Of High Performance Photomultiplier Tubes

Photomultiplier tubes(PMTs)are extraordinary photodetector devices that can convert weak light signals into electric signals and amplify them.The sensitivity to weak light signal makes PMTs become attractive devices in medical imaging and high-energy physics detection.Currently,there are three main types of PMT that are widely needed and applied,namely PMTs with Dynodes,Microchannel Plate(MCP)PMTs and Si-PMTs.Based on the strict requirements of high detection efficiency,high gain,fast response and low noise for large high-precision detection devices,this thesis optimizes the design of a kind of ultra-compact Dynode-PMT for neutrino detection,simulates the gain and timing characteristics of the MCP-PMT multiplication system and builds a Si-PMT-based TOF-PET detection system module.The main contents of this thesis are as follows:1)In response to the requirements of the high performance 3-3.5 inch PMT for the JUNO detector under construction,the upgraded Mediterranean KM3 NET and the Antarctic Ice Cube detector,a kind of ultra-compact PMT with a length of only 103 mm has been optimized in the CST Particle Studio,which is the shortest 3 inch PMT structure.The effect of the distance between the PMT multiplication system and the photocathode on its collection efficiency and time characteristics was analyzed,and the optimal placement of the PMT multiplication system was given.The effects of different voltage ratio schemes on the potential distribution and electron trajectory of the PMT were compared,and a more reasonable parameter setting of the voltage ratio scheme was given for the optimized PMT model.The timing performance of the PMT is effectively improved.The quantum efficiencies,single photoelectron spectrums and gains of the optimized 3-inch and 3.5-inch PMT prototypes were tested.2)The gain and timing performance of ALD based MCP-PMT multiplier system were simulated.The dependences of the gain and timing performance on the SEE yield properties were assessed by using the Monte Carlo and particle-in-cell methods.Three important secondary electron emissions,the backscattered,rediffused and true SEEs,were discussed numerically.The secondary electron cascade processes in the MCP channel were calculated and mimiced.The simulation result indicates that the opportunities for improving the gain of the MCP by improving SEE yield corresponding to the incident energy of 0 ~ 100 e V.The backscattered and rediffused electrons are found to have strong effects on the gain and timing performance of the MCP.Although the higher the SEE yield the higher the MCP gain,the drawback is the extremely high SEE yield will make the MCP saturated prematurely and degrade the time resolution.The simulation results will be used to guide the design and selection of emissive material for ALD-coated MCP development.3)The influence factors limiting the high linearity of MCP-PMT were studied by simulation.The dependences of MCP gain nonlinearity on the number of incident electrons,operating voltage,and SEE yield properties were studied.The simulation results show that the gain of the MCP single channel decreases as the number of incident electrons increases due to the space charge effects.The higher operating voltage and SEE yield of the MCP,the faster the gain deteriorates.To mitigate the gain saturation effect of MCP single channel,a novel structural design of MCP-PMT has been proposed by adjusting the design of the MCP chevron pair.A significant improvement in the output pulse peak can be obtained.4)Double-ended readout TOF-PET detector based on coupling a long scintillation crystal to Si-PMTs at both ends for correcting the depth-dependent effects to improve the coincidence time resolution(CTR)was stuied.In particular,we focused our attention to analyze timing performance using 4 different correction methods.For a 3 mm × 3 mm × 25 mm unpolished lutetium fine silicate(LFS)crystal with double-ended readout and practical head-on irradiation,a CTR of 246 ps FWHM can be achieved using depth-dependent timing-correction and weighted average time method compared to 393 ps using the conventional single-ended readout,which corresponds to a 37% improvement.