电荷耦合器件在质子辐照下的粒子输运仿真与效应分析

应用蒙特卡洛方法计算了质子在科学级电荷耦合器件(charge-coupled device, CCD) 结构中的能量沉积, 并结合该CCD的质子辐照试验及退火试验数据, 分析了器件的辐射损伤机理. 仿真计算体硅内沉积的位移损伤剂量和栅氧化层的电离损伤剂量, 辐照与退火试验过程中主要考察暗信号、电荷转移效率两个参数的变化规律. 研究结果显示, 暗信号和电荷转移效率的变化规律与位移、电离损伤剂量一致; 退火后暗信号大幅度降低, 辐照导致的表面暗信号增加占总暗信号增加的比例至少为80%; 退火后电荷转移效率恢复较小, 电荷转移效率降低的原因主要为体缺陷. 通过总结试验规律, 推导出了电荷转移效率退化程度的预估公式及其损伤因子k_damage.

Particle transport simulation and effect analysis of CCD irradiated by protons

Monte Carlo method is used to calculate the energy deposition of proton-irradiated scientific CCD (charge coupled device) structure, and the radiation damage mechanism of the device is analyzed by combining the proton irradiation with the annealing experiments. The ionizing dose in gate oxide layer and the displacement damage dose in silicon deposition are simulated. During irradiation and annealing experiments two main parameters, dark signal and charge transfer efficiency, are investigated. Results show that variations of dark signal and charge transfer efficiency are the same as those with ionizing dose and displacement damage dose. During irradiation, dark signal rises obviously as the fluence of 10 MeV proton increases. Defects and their annealing temperature:the divacancy levels show little annealing effect below 300℃, while the oxygen-vacancy complex is stable up to 350℃, and the phosphorous-vacancy has a characteristic annealing temperature of 150℃. Interface states are annealed totally at 175℃. So the annealing only affects oxide-trapped-charges. Dark signal is greatly reduced after annealing, this phenomenon means that the dark signal is mainly affected by ionization. The surface dark signal proportion of the total dark signal can be calculated by the reduction of dark signal during annealing and this is at least 80% or more. As the fluence of 10 MeV proton increases, the charge transfer efficiency reduces obviously. After annealing, the recovery of charge transfer efficiency changes very little, so the charge transfer efficiency is unaffected by oxide-trapped-charges, since it is reduced due mainly to bulk defects. The final device damage will always be proportional to the amount of initial damage and also to the electrical effect on the device. Hence NIEL scaling implies a universal relation:device damage=k_damage×displacement damage dose, where k_damage is a damage constant depending on the device and the parameter affected, and the displacement damage dose (DD) is the product of the NIEL and the particle fluence. MULASSIS is used to calculate the displacement damage dose in depletion area of P-area and deduce k_damage by combining with the experimental value of charge transfer efficiency; k_damage is calculated to be about 3.50×10^-14. The formula for degradation degree of charge transfer efficiency is CTE_{after irradiated} = 1-Dd×k_damage, this formula is used to estimated CTE and the result is compared with the value from experiment. It is shown that the simulated data is in agreement with the experimental data.