摘要
采用铜大马士革工艺制备了用于电迁移测试的样品,对电迁移测试过程中存在的两类电阻-时间(R-t)特征曲线进行了研究。研究发现采用固定电阻变化率作为失效判定标准所得的失效时间分布曲线不能真实地反映样品的实际寿命,而采用第一次阻值跳变点对应的时间作为失效时间所得的分布曲线则更符合电迁移理论。针对两种失效判定方法所得到的不同结果进行了机理分析,结果表明,采用第一次阻值跳变点对应的时间作为失效时间分析电迁移失效更合理。
Samples used for electromigration test were prepared by copper Damascus process. Two kinds of resistance-time（ R-t） curves in electromigration test were studied. The results show that the distribution curve of time to failure from the fixed resis-tance change percentage as the failure criterion can not reflect the real lifetime of the samples,while the first jump point of resistance as failure criterion can match electromigration mechanism well. The mechanism of this discrepancy from two failure judgment methods were analyzed. And the result indecates that it is more reasonable to use the corresponding time to the first jump point of resistance as the failure time.
出处
《半导体技术》
CAS
CSCD
北大核心
2015年第10期793-797,共5页
Semiconductor Technology
