Analysis of the reliability of package-on-package devices manufactured using various underfill methods

Abstract

As next-generation electronic packages continue to dictate smaller devices and more functionality, package-on-package (POP) configurations have started to gain popularity in the SMT industry. These stacked package devices enable board space savings, simplified system design, enhanced performance, and lower pin count. Although POPs are experiencing rapid growth for certain applications such as mobile handsets, digital cameras, PDAs, and MP3 players, concerns over POP drop test and thermal cycling performance reliability issues have been raised. Recently, the electronics industry has gathered a great deal of POP reliability data to help optimize the POP manufacturing and application process. A number of studies and tests have been conducted to investigate the board-level reliability of POPs in relation to drop test and thermal cycling performance. The test conditions have examined packages manufactured with and without underfill and have also analyzed the impact of different underfill dispensing patterns (i.e. full underfill, cornerbond and edgebond) However, few papers discuss the effects of the underfilling strategy — such as undefilling the bottom component only or underfilling both top and bottom components, or the effects of solder alloy choice on the reliability of POP packaging. In this paper, the effects of underfill dispensing type and POP ball alloy type on the reliability of POP devices during drop testing and thermal cycle testing were evaluated. It was found that both underfill dispensing type and alloy type have a profound effect on POP reliability. The study results revealed that underfilling only the bottom component seems to have no significant contribution to POP drop test reliability. Underfilling both the top and bottom components yields better drop test performance than underfilling only the bottom component. In addition, the SAC105 (98.5%Sn + 1.0%Ag + 0.5%Cu) bump alloy shows better drop test performance than the SAC305 (96.5%Sn + 3.0%Ag + 0.5%Cu) alloy.