Extreme Environment Reliability of Components for Computing with SAC305 and Alternative High Reliability Solders
Type of DegreeDissertation
Industrial and Systems Engineering
MetadataShow full item record
The semiconductor and packaging industries have been moving away from the use of Lead (Pb) due to the increasing awareness of the health and safety concerns surrounding its use. For many applications, the industry has moved from eutectic Sn-Pb solder to the near-eutectic Sn-Ag-Cu (SAC) solders, and more applications – including those considered “extreme environment” – are likely to take place in the near future. However, the reliability of electronic assemblies with SAC solder joints has proven hard to predict based on previous experience with SnPb solders. The reliability of electronic solder joints is determined by a variety of factors including bulk solder properties and failure mechanics. Both the composition and microstructure of the solder joint will affect its bulk properties. Although an initial microstructure will be present following assembly – which will involve one or more soldering steps – this structure will continue to evolve over the lifetime of the joint. The microstructure and microstructural evolution of the Sn-Ag-Cu solders differ significantly from that of eutectic SnPb solders. Because of the risks and uncertainties involved, a new body of reliability engineering knowledge must be built of for the Sn-Ag-Cu solders based on application-specific process and service parameters. This experiment considers the thermal cycle reliability of an assortment of different electronic components and evaluates them on a 0.200” (200 mils) thick printed circuit board. Two substrate materials are tested: FR4-06 and Megtron6. Organic Solderability Preservative (OSP) surface finish is used with all test vehicles. The primary solders for package attachment in this experiment are SnPb and SAC305. Two solders designed for high-temperature reliability are also considered, including a Bi-doped SAC material and the six-element alloy Innolot (Sn3.8Ag0.7Cu3Bi1.4Sb0.15Ni). Isothermal storage at high temperature was used to accelerate the aging of the assemblies. Aging Temperatures are 25oC, 50oC, and 75oC. Aging durations are 0-Months (No Aging, baseline), 6-Months, 12-Months, and 24-Months. The test vehicles were then subjected thermal cycles of -40°C to +125°C on a 120-minute thermal profile in a single-zone environmental chamber to assess the solder joint performance. The as-reflowed failure data (No Aging Group) was found to follow specific reliability trends depending on the type and size of the component. The smaller plastic ball grid array (BGA) packages show the following pattern in Characteristic Life value, listed from best to worst: (1) Matched Innolot, (2) [S]SAC305 doped with [P]Innolot, (3) Matched SAC305, and (4) Matched SnPb. However, when considering the effects of isothermal aging on the relative reliability of various packages, the data indicate that even components that show similar initial reliability trends may display differences following aging. Following isothermal aging, several components exhibit higher reliability when paired with SnPb solder than with the SAC solder materials. Significant differences in reliability were seen between equivalent packages mounted to the two substrate materials tested (FR4-06 and Megtron6). For all of the over-molded plastic BGA components, reliability was higher on the standard glass-epoxy material (FR4-06). Performance for these packages on the high-electrical-performance polyphenylene oxide (PPO) blend material (Megtron6) was much worse. The degradation in reliability with aging were also found to be worse on Megtron6 for the Sn-Ag-Cu materials when paired with these components. However, the two Super-BGA components – SBGA 304 and SBGA 600 – dramatically reverse this substrate-based reliability trend. For the smaller plastic BGA packages, Innolot doping (micro-alloying) appears to be an effective strategy for improving characteristic life. However, as component size and pitch increase, this improvement seems to wane (and in some cases reverse itself altogether). This may be attributable to under-doping of the large-component joints. Based on assembly-level reliability, the “paste doping” strategy appears to be a promising approach to improve reliability in high-stress environments, but one that requires significant further study.