存储器空间单粒子翻转率预测

根据SSO上两卫星搭载的三个PREM测得的空间中重离子LET谱,以及利用Weibull分布模型拟合出的不同器件的σ-LET曲线,对由空间中重离子引起的单粒子翻转的翻转率进行了预测估算.将预测值与实测值对比,分析了影响翻转率的因素.对于相同器件,翻转率与设备在卫星上的位置和朝向有关.位于卫星尾部面向后退（-x）方向的翻转率高于位于底部对地（+z）方向的器件翻转率;太阳活动水平高的时间段翻转率高于太阳活动水平低的时间段.探测器接收的重离子微分LET谱的强度和硬度决定了器件的单粒子翻转率.在高于翻转LET阈值时,LET谱的强度越高,其硬度和翻转率越大.不同器件的翻转率也不相同.      

130nm体硅反相器链的单粒子瞬态脉宽特性研究

针对130 nm体硅反相器链,利用脉冲激光和重离子实验研究了目标电路单粒子瞬态（SET）的脉宽特性,并分析了电路被辐射诱发的SET脉宽特性受激光能量、重离子线性能量传递（LET）值、PMOS管栅长尺寸等因素的影响机制。重离子和脉冲激光实验结果类似,均表现为随激光能量、LET值的增加,电路被辐射诱发的SET脉宽逐步增大,且表现出明显的双（多）峰分布趋势,但辐射诱发的SET脉冲个数呈先增加再减少规律。此外,实验结果表明,在不同激光能量、LET值下,PMOS管栅长尺寸影响反相器链SET脉冲的特征不同。当激光能量、LET值较低时,PMOS管栅长尺寸大的电路产生的SET脉宽较大,而当激光能量、LET值较大时,PMOS管栅长尺寸小的电路反而产生更宽的SET脉冲。分析表明,较高激光能量、LET辐照时,寄生双极放大效应被触发可能是导致PMOS管栅长尺寸影响电路SET特征差异的主要原因。      

空间粒子辐射对卫星中集成芯片的影响

本文分析了在“风云一号(B)”气象卫星环境中各种粒子辐射在集成芯片临界体积中产生的能量沉积, 即LET(线性能量传输);得到了银河宇宙线1≤Z≤28、银河宇宙线异常成分(C、N、O、Ne、Ar、Fe)、内辐射带质子等产生的LET, 计算了其分别产生的单粒于事件(SEU)翻转率。      

反向蒙特卡罗方法在卫星辐射分析中的研究与应用

反向蒙特卡罗方法(AMC/RMC)是Geant4中一个强有力的偏置技术. 粒子产生于包含灵敏体的反向界面并在几何体中被反向追踪直至外界面或溢出能量阈值, 其计算时间只集中于对结果有贡献的粒子径迹. 与正向蒙特卡罗方法(FMC)相比, RMC更高效, 特别是当灵敏体远小于几何体其他部分和外界面时, 其优势更明显. 通过RMC与FMC的比较, 验证了RMC应用于卫星辐射剂量分析的准确性. RMC与SHIELDOSE2和SSAT的比较说明了RMC是高精度卫星辐射剂量的优选方法.      

Flash芯片电流“尖峰”现象的脉冲激光试验

利用皮秒脉冲激光单粒子效应试验装置研究了一款宇航级Flash芯片的电流“尖峰”（HCS）现象。利用激光准确定位的特点，确定电流“尖峰”是由芯片的电荷泵单元充放电引起的，不同的激光能量、入射位置会触发不同频率、相同幅值的电流“尖峰”现象，虽然电流“尖峰”发生的瞬间电流增大的现象与单粒子锁定效应表现一致，但机理完全不同。当激光能量足够高（对应于重离子LET值99.8 MeV·cm2/mg）时，在电荷泵的同一个敏感位置累积多次辐照不断触发芯片发生电流“尖峰”，芯片会因多次充放电而损坏。      

空间高能粒子与器件布线层核反应后次级粒子LET分布研究

空间高能质子和重离子是导致元器件发生单粒子效应的根本原因,为准确评估元器件在轨遭遇的单粒子效应风险,必须清楚高能质子、重离子与器件材料发生核反应的物理过程及生成的次级重离子LET（Line EnergyTransfer）分布规律。针对典型CMOS工艺器件模拟计算了不同能量质子和氦核粒子在器件灵敏单元内产生的反冲核、平均能量及线性能量转移值,并分析了半导体器件金属布线层中重金属对次级重离子LET分布的影响规律。计算结果表明：高能粒子与器件相互作用后产生大量次级重离子,且高能质子作用后产生的次级粒子的LET值主要分布为0～25MeV·cm²/mg;高能氦核粒子作用后产生的次级粒子的LET值主要分布为0～35 MeV·cm²/mg;有重金属钨（W）存在时能提高次级粒子的LET值,增加了半导体器件发生单粒子效应的概率,该研究结果可为元器件单粒子效应风险分析、航天器抗单粒子效应指标确定提供重要依据。     

0.13μm部分耗尽SOI工艺反相器链SET脉宽传播

基于0.13 μm部分耗尽绝缘体上硅（PD-SOI）工艺,设计了一款片上反相器链（DFF）单粒子瞬态（SET）脉宽测试电路并流片实现,SET脉宽测试范围为105~3 150 ps,精度为±52.5 ps。利用重离子加速器和脉冲激光模拟单粒子效应试验装置对器件进行了SET脉宽试验。采用线性能量传输（LET值）为37.6 MeV·cm2/mg的86Kr离子触发了反相器链的三级脉宽传播,利用脉冲激光正面测试器件触发了相同级数的脉宽,同时,激光能量值为5 500 pJ时触发了反相器链的双极放大效应,脉宽展宽32.4%。通过对比激光与重离子的试验结果,以及明确激光到达有源区的有效能量的影响因子,建立了激光有效能量与重离子LET值的对应关系,分析了两者对应关系偏差的原因。研究结果可为其他种类芯片单粒子效应试验建立激光有效能量与重离子LET值的对应关系提供参考。      

星用大容量静态存储器多位翻转实验研究

给出典型大容量静态存储器(SRAM)的多位翻转实验研究 结果。用HI-13串列型静电加速器和兰州重离子加速器(HIRFL)加速的重离子轰击样品,用 一套基于 网络协议的高分辨率SRAM单粒子效应检测系统检测发生的多位翻转。实验结果表明多位翻转 可以由多种机制产生：在两种Hitachi SRAM中检测到的同一字节多位翻转(SMU)是由单个离 子产生的电荷被相邻敏感节点共享所致;当IDT71256中写入测试图形“00”时,其外围电路 中产生的单粒子瞬时脉冲(SET)引起多达8位的SMU;离子大角度掠射下,IDT71256中检测到 了同一事件多位翻转(SEMU)。同时预示了两种Hitachi大容量SRAM在地球同步轨道和两条太 阳同步轨道发生SMU的频度。     

一种SEU实验数据的处理方法

对SEU数据——特别是看似不理想，但却包含了丰富的物理信息所谓“坏数据”——的处理方法进行了分析讨论，并具体针对IDT7164器件的SEU数据进行了处理，从中得到了死层厚度、灵敏体积厚度、能量阈值等进行单粒子翻转预测所需要的关键参数，并就如何根据国内加速器的实际情况开展有针对性的单粒子效应模拟试验提出了一些初步想法。     

多离子空间等离子体中快磁声波与粒子的回旋共振特性

快磁声波是空间等离子体中一种接近垂直传播的右旋极化电磁波,能够在等离子体层内外传播.快磁声波与带电粒子的回旋共振相互作用能够导致高能电子随机加速和投掷角扩散、能量质子投掷角扩散等,从而影响辐射带高能带电粒子的动态过程.分别基于完整的色散关系和高密度近似的色散关系,在不同空间等离子体条件下研究多离子空间等离子体中不同传播角的快磁声波色散曲线,并计算了快磁声波与H+,He+和O+离子的最小共振能量.结果表明,当传播角较小时,采用高密度近似与采用完整色散关系计算的离子最小共振能量没有太大差别.在中低密度中强磁场空间等离子体中,传播角≥ 88°时高密度近似色散关系会带来很大的误差,因此应利用完整色散关系计算最小共振能量.      

Microscopic track structure of equal-LET heavy ions.

The spatial distributions of ionization and energy deposition produced by high-velocity heavy ions are crucial to an understanding of their radiation quality as exhibited eg., in track segment experiments of cell survival and chromosome aberrations of mammalian cells. The stopping power (or LET) of a high velocity ion is proportional to the ratio z2/v2, apart from a slowly varying logarithmic factor. The maximum delta-ray energy that an ion can produce is proportional to v2 (non-relativistically). Therefore, two HZE ions having the same LET, but in general differing z and v will have different maximum delta-ray energies and consequently will produce different spatial patterns of energy deposition along their paths. To begin to explore the implications of this fact for the microscopic dosimetry of heavy ions, we have calculated radial distributions in energy imparted and ionization for iron and neon ions of approximately equal LET in order to make a direct comparison of their delta-ray track structure. Monte Carlo techniques are used for the charged particle radiation transport simulation.     

HZE effects on mammalian cells.

In track segment experiments cell survival and chromosome aberrations of mammalian cells have been measured for various heavy ion beams between helium and uranium in the energy range between 0.5 and 960 MeV/u, corresponding to a velocity range of 0.03 to 0.87 C, and an LET spectrum from 10 to 15 000 keV/micrometers. At low LET, the cross section (sigma) for cell killing increases with increasing LET and shows a common curve for all ions regardless of the atomic number. This indicates that in this region the track structure of the different ions is of only a minor influence, and it is rather the total energy transfer, which is important for cell killing. At higher LET values, deviations from a common sigma-LET curve can be observed which indicate a saturation effect. The saturation of the lighter ions occurs at lower LET values than for the heavier ions. These findings are also confirmed by the chromosome data, where the efficiency for the induction of chromosomal aberrations for high LET particles depends on the track structure and is nearly independent of LET. In the heavier beams (Z > or = 10) individual particles cause multiple chromosome breaks in mitotic cells.     

Biological effects of heavy ions from the standpoint of target theory.

The biological effect of heavy ions is best described through the action cross section, as a function of the end-point of interest and the charge and speed of the ion. In track theory this is called the "ion-kill" cross section, for it is the effect produced by a single heavy ion and its delta rays. As with nuclear emulsions the biological track structure passes from the grain count regime to the track width regime to the thindown region with an increase in LET. With biological cells, as with any detector capable of storing sublethal damage, with low LET irradiation the action cross section (in the ion-kill mode) is increasingly obscured by the effect of "gamma-kill", by the influence of overlapping delta rays from neighboring heavy ions. Thus at low LET response is dominated by the gamma-kill mode, so that the RBE approaches 1. The theory requires 4 radiosensitivity parameters for biological detectors, extracted from survival curves at several high LET bombardments passing through the grain count regime, and at high doses. Once these are known the systematic response of biological detectors to all high LET bombardments can be unfolded separating ion kill from gamma kill, predicting the response to a mixed radiation environment, and predicting low dose response even at the level of a single heavy ion. Cell killing parameters are now available for a variety of cell lines. Newly added is a set of parameters for cell transformation.     

RBE: mechanisms inferred from cytogenetics.

Cyclotron-accelerated heavy ion beams provide a fine degree of control over the physical parameters of radiation. Cytogenetics affords a view into the irradiated cell at the resolution of chromosomes. Combined they form a powerful means to probe the mechanisms of RBE. Cytogenetic studies with high energy heavy ion beams reveal three LET-dependent trends for 1) level of initial damage, 2) distribution of damage among cells, and 3) lesion severity. The number of initial breaks per unit dose increases from a low-LET plateau to a peak at approximately 180 keV/micrometer and declines thereafter. Overdispersion of breaks is significant above approximately 100 keV/micrometer. Lesion severity, indicated by the level of chromosomal fragments that have not restituted even after long repair times, increases with LET. Similar studies with very low energy 238Pu alpha particles (120 keV/micrometer) reveal higher levels of initial breakage per unit dose, fewer residual fragments and a higher level of misrepair when compared to high energy heavy ions at the same LET. These observations would suggest that track structure is an important factor in genetic damage in addition to LET.     

Cellular and subcellular effect of heavy ions: a comparison of the induction of strand breaks and chromosomal aberration with the incidence of inactivation and mutation.

Radiobiological effects of heavy charged particles are compared for a large variety of ions from Helium to Uranium and energies between 1 and 1000 MeV/u which correspond to LET values between 10 and 16000 keV/micrometers. The different cross section for the induction of strand breaks and chromosomal aberrations as well as for inactivation and mutation induction exhibit striking similarities when compared as function of the linear energy transfer (LET). At LET values below 100 keV/micrometers all data points of one specific effect form one single curve as a function of LET, independent of the atomic number of the ion. In this LET range, the biological effects are independ from the particle energy or track structure and depend only on the energy transfer. Therefore, LET is a good parameter in this regime. For LET values greater than 100 keV/micrometers, the curves for the different ions separate from the common curve in order of increasing atomic numbers. In this regime LET is no longer a good parameter and the physical parameters of the formation of particle tracks are important. The similarity of the sigma-LET curves for different endpoints indicates that the 'hook-structure' is produced by physical and chemical effects which occur before the biologically relevant lesions are formed. However, from the existing data of biological effects, it can be concluded that the efficiencies for cell killing are always smaller than those extrapolated from X-ray data on the basis of the energy deposition only. Therefore, cells which are directly hit by an HZE particle are not killed and undergo a finite risk of mutation and transformation.     

A comparison of two different types of geosynchronous satellite measurements during the 1989 solar proton events.

The proton telescope aboard the GOES-7 satellite continuously records the proton flux at geosynchronous orbit, and therefore provides a direct measurement of the energetic protons arriving during solar energetic particle (SEP) events. Microelectronic devices are susceptible to single event upset (SEU) caused by both energetic protons and galactic cosmic ray (GCR) ions. Some devices are so sensitive that their upsets can be used as a dosimetric indicator of a high fluence of particles. The 93L422 1K SRAM is one such device. Eight of them are on the TDRS-1 satellite in geosynchronous orbit, and collectively they had been experiencing 1-2 upset/day due to the GCR background. During the large SEP events of 1989 the upset rate increased dramatically, up to about 250 for the week of 19 Oct, due to the arrival of the SEP protons. Using the GOES proton spectra, the proton-induced SEU cross section curve for the 93L422 and the shielding distribution around the 93L422, the calculated upsets based on the GOES satellite data compared well against the log of measured upsets on TDRS-1.     

脉冲激光模拟空间载荷单粒子效应研究进展

脉冲激光作为模拟测试空间探测载荷半导体器件的单粒子效应现象的一种较新型手段,具有可以定位器件对单粒子效应敏感的具体单元以及动态测试电路系统对单粒子效应的时间响应特性的特点,能够满足工程部门、器件研发部门的不同需求。通过实验与理论研究,建立单粒子锁定与翻转效应的激光阈值能量与重离子LET值的对应关系,解决了脉冲激光模拟测试的激光结果如何定量的关键问题,据此可以定量摸底评估器件的单粒子效应敏感度,使脉冲激光测试载荷的结果更具评价以及指导意义,这对建立统一的脉冲激光单粒子效应评估试验标准以及对脉冲激光试验的推广具有重要意义。空间探测载荷发生单粒子效应后器件功能特性及电路系统的影响、防范单粒子效应电路条件影响的手段下电路系统的抗单粒子效应设计措施是的有效性,以及为空间探测专门研制的抗辐射ASIC电路评价,都需要更加精细的单粒子效应测试方法。通过建立便捷、低成本的脉冲激光定量试验的手段,解决了空间探测载荷上述单粒子效应试验的问题。     

Neoplastic transformation of hamster embryo cells by heavy ions

We have studied the induction of morphological transformation of Syrian hamster embryo cells by low doses of heavy ions with different linear energy transfer (LET), ranging from 13 to 400 keV/μm. Exponentially growing cells were irradiated with ¹²C or ²⁸Si ion beams generated by the Heavy Ion Medical Accelerator in Chiba (HIMAC), inoculated to culture dishes, and transformed colonies were identified when the cells were densely stacked and showed a crisscross pattern. Over the LET range examined, the frequency of transformation induced by the heavy ions increased sharply at very low doses no greater than 5 cGy. The relative biological effectiveness (RBE) of the heavy ions relative to 250 kVp X-rays showed an initial increase with LET, reaching a maximum value of about 7 at 100 keV/μm, and then decreased with the further increase in LET. Thus, we confirmed that high LET heavy ions are significantly more effective than X-rays for the induction of in vitro cell transformation.     

On the quantitative interpretation of cellular heavy ion action.

Current analyses of heavy ion action assume that the survival probability of a cell hit by a heavy ion depends only on the energy absorbed in its critical site. It is known, however, that the efficiency to produce a biological effect depends also on the spatial pattern of energy deposition. This has to be included in the quantitative evaluation of heavy ion action. Based on recent models of lesion formation by ionizing radiation (Goodhead and Brenner, Phys. Med. Biol. 28, 485, 1983) data with lighter ions (LET < 500 keV/micrometer) were re-analysed. It is shown that the behaviour of various cell systems can be described by a common curve which can be used to estimate the contribution of "non-linear" components (i.e. where the distribution of energy deposition plays a role) with heavy ions. It is concluded that even with Uranium ions the regions of non-linear effects does not extend beyond 50 nm from the trade core. These data will be used to assess quantitatively survival curves obtained with very heavy ion exposure.