Failure analysis of a 2.5D stacking using μinsert technology

Physical and Failure Analysis of Integrated Circuits(2013)

引用 0|浏览21
暂无评分
摘要
Failure analyses on μinsert technology used for 2.5D stacking are carried out after 1000 hours of damp heat test on nickel, copper and gold-nickel μinserts. We show that nickel μinserts are prone to delamination when the force applied on the reported die during bonding process is not homogeneous or lower than an optimal value. After the test, one out of eight tested gold-nickel μinserts is subject to a failure induced by gold-aluminum intermetallic compound growth. These intermetallic compounds engender Kirkendall voids which cause a steady increase of a few tens of percents of the electrical resistance of the gold-nickel μinserts. Failures of copper μinserts are due to residues of the titanium used to protect the copper seed layer before the electroplating step. An improper assembly seems to be the cause of failure in each case and it is easily detectable after the bonding process by measuring the electrical resistance of μinserts.
更多
查看译文
关键词
aluminium alloys,bonding processes,delamination,electroplating,failure analysis,gold alloys,μinsert technology,2.5d stacking,kirkendall voids,bonding process,copper microinsert failure,copper seed layer,damp heat test,electrical resistance measurment,electroplating step,gold-aluminum intermetallic compound growth,gold-nickel microinsert,nickel microinsert,optimal value,time 1000 hour,titanium residue,3d,edx,stem,bonding,failure mode,insert,intermetallic,pressure measurement,resistance,assembly
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要