Some observations from interrupted lifetest of GaInAsP/InP inverted-rib laser diodes

Conventionally, lifetests of semiconductor laser diodes usually involved operating the devices continuously at either constant power output or drive current, with periodic recording of their characteristics. In this paper, some effects arising from interrupted lifetest of 1.3 m GaInAsP-InP inverted-rib laser diodes are reported. This unconventional lifetest method involves constant power biasing at 4 mW/facet and 8 mW/facet respectively at 50°C, followed by a period during which the lifetest is interrupted and the devices left unbiased at room temperature. Subsequently, the devices were put back on constant power biasing at 50°C. Among a number of parameters, pronounced reduction in the threshold current, current for 4 mW/facet and 8 mW/facet were observed, indicating strong recovery effects commencing from the time when the life-test was interrupted. Redistribution of mobile defects in the cladding layer is postulated to be the cause of the degradation recovery, and the data supports the occurrence of an aging-current dependent defect annihilation mechanism. Such recovery effects have so far been observed to occur only in the GaInAsP-InP inverted-rib devices.     

CA型固体钽电容器的老化筛选方法

ＣＡ型固体钽电容器是重要的电子元件，为淘汰早期失效产品，确保高的可靠性水平，采用合适的筛选方法十分必要。介绍了：１．敲击漏电流闪变筛选；２．电参数性能测量筛选；３．温度循环筛选；４．高温电负荷老化筛选；５．高温浪涌电压老化筛选；６．高温漏电流老化筛选；７．密封检验筛选。筛选效果最明显的是４、６两项     

用AlN薄膜提高NiCr电阻网络的可靠性

采用低温反应射频溅射法在微晶玻璃上沉积了一层具有择优取向的氮化铝薄膜，再在其上沉积８０Ｎｉ－２０Ｃｒ电阻网络。对电阻网络进行了功率老化、高温存贮、热冲击等可靠性试验，结果表明具有ＡｌＮ薄膜的电阻网络性能得到很大提高     

冰区海洋平台疲劳可靠性评估方法的研究

给出了在随机波浪载荷和冰载荷作用下平台构件疲劳损伤的一种计算方法,并利用该方法分析了平台单个构件和平台整体的疲劳可靠性.用指数函数对波浪和海冰作用下平台构件的疲劳应力、应力循环次数和环境参数的关系进行了拟合,拟合精度很好.根据拟合的指数函数,并基于S-N曲线和Miner线性累积疲劳损伤理论,推导出构件疲劳损伤的计算公式,使构件疲劳损伤的计算过程得到很大简化.在此基础上给出了单个构件的疲劳可靠性计算公式,提出了平台整体疲劳失效的简化判断依据和平台关键失效模式的确定方法,并给出了平台整体失效概率的计算方法.计算实例表明,该方法简单可靠,可以有效地用于工程实际.     

可靠性试验评定中专家信息的描述及转化方法

专家信息是小子样条件下装备可靠性试验评定中验前信息的重要来源.将专家信息的提供过程分为专家信息的描述、收集、转化、融合等4个步骤,将专家信息的描述划分为3类(统计、模糊和其它方式)共9种描述形式,并分别就每一种描述形式给出了转化为Bayes验前分布的方法.     

微电子生产工艺可靠性评价与控制

简要介绍了可靠性评估(REM)测试结构设计,并介绍了REM试验中与时问有关的栅氧化层击穿(TDDB)、热载流子注入(HCI)效应和电迁移(EM)效应的评价试验方法及实例。REM技术与工艺过程控制(PCM)、统计工艺控制(SPC)技术结合起来就可以实现对工艺的可靠性评价与控制,某集成电路生产公司将它应用于金属化工艺中．确定了工艺输入变量与电迁移可靠性的相关性,优化了金属化工艺试验条件,提高了金属化系统的抗电迁移能力。     

半导体器件并联系统可靠性评估的经验Bayes方法

为了提高机载开关电源中半导体器件并联系统可靠性评估的准确性,运用经验 Bayes 法和经典的统计方法,研究了该系统的可靠性评估问题。分别给出了系统可靠性指标的经验 Bayes 估计,极大似然估计。利用 Monte-Carlo方法,对两种估计结果进行了比较,结果表明,经验 Bayes 估计的最大绝对误差为 0.07,它小于极大似然估计的最大绝对误差 0.368。     

Gate oxide reliabilities in MOS (metal-oxide-semiconductor) structures with Ti-polycide gates

We have investigated the thermal degradation of gate oxide in metal-oxide-semiconductor (MOS) structures with Ti-polycide gates. We found that the Ti-diffusion into the underlying polysilicon and consequently to the gate oxide occurs upon thermal cycling processes, which results in the dielectric breakdown of the gate oxide. We also found that the Ti-diffusion is suppressed by the employment of the thin (about 5 nm) titanium nitride (TiN) diffusion barrier layer, which consequently improved the reliability characterisitics of gate oxide significantly.     

PLC仍在全方位地发展(上)

以中心点法和验算点法为代表的一次二阶矩方法计算简便,但对非线性程度较高的结构功能函数,其计算结果与精确度相差过大。应用数学逼近中的拉普拉斯渐进方法将非线性功能函数在验算点处作二次展开来研究结构的可靠度问题,能较高精度的逼近精确结果。算例分析表明,当随机变量的数目较多时,由一次二阶矩方法计算的结果与精确解相差较大,而二次二阶矩方法的计算结果与精确解非常接近。     

兆瓦风力发电机系统可靠性设计理论研究

在分析兆瓦风力发电机系统的基础上,提出了兆瓦风力发电机发电系统可靠度的基本概念及计算方法,建立了可靠度最优分配的数学模型及最优兆瓦风机发电系统可靠性设计的数学模型。