Flip chip laser mark bare die strength characterization

• Published in: 2016 IEEE 18th Electronics Packaging Technology Conference (EPTC) pp. 551 - 553
• Main Authors: Muniandy, Kesvakumar, Schmadlak, Ilko, Yeung, Betty, Lauderdale, Matt, Uehling, Trent
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2016
• Subjects: Conferences; Decision support systems; Electronics packaging
• DOI: 10.1109/EPTC.2016.7861540

Bare die flip chip products have a high risk of die cracking as shown in Figure 1, during product electrical test or temperature cycling. The stresses experienced by the die during these events are understood. But the die strength impact after product laser marking on the die backside is not well understood. This area has a lot of room for improvement. Often, a lid is added to a flip chip package to protect the die from forces that could lead to cracking. A lid also acts as a protection to the die as it avoids the possibility to compromise the die strength by a laser mark, since the part would be marked on the lid and not on the bare die. A lid will add to product cost depending on body size. Without a data driven specification, millions of dollars could be wasted by putting a lid on a package that does not really need it. The objective of this project is to measure the die break strength as a function of laser mark profile, to optimize the laser mark across factories, and to update the maximum principal stress limit for die without a lid. Since unmarked die will always fail from the die edge due to defects introduced during singulation, it is the aim to achieve a laser marking that does not exceed this threshold in 4-point bend experiments. The laser depth measurement specification is not clearly defined, and in most cases of products, it is not even measured. A visual inspection criteria, which depicts a general outlook is used as a go-no-go. This can be used for the over-molded packages as it does not have the adverse effects as compared to marking on the bare silicon die. This study will further define the specification criteria for the laser depth control in regards to bare silicon die. Currently, all NPI's are evaluated and, if need be, mechanical simulations are done at cold temperature and under product test conditions. The study carried out was broken down to two different aspects, the first being testing the break strength and correlation of laser depth for the fully assembled package and the other using similar concept but done on singulated dies. The data collected from this study will be the input to the building of mechanical models to simulate and correlate the data. The initial results show that the different laser mark profiles creates different depth and correlates to the break strength of the die.