Shear testing and failure mode analysis for evaluation of BGA ball attachment

Abstract

With the increasing focus on Ball Grid Array (BGA) technology as a high performance semiconductor package, there is an urgent need to develop industry wide standards for package quality and reliability. A significant factor in BGA package quality is the robustness of the attachment of the solder ball to the package. Currently, ball shear testing is used as a destructive test method to assess solder joint integrity. For a given shear rate, a common recognized acceptance criterion for this test is a minimum shear force of 1Kg for connections on packages with 0.75 mm solder ball diameter and 0.64 mm pads. A shear force of 1Kg alone, however, does not guarantee a quality solder joint. The solder ball failure mechanism can not be determined from shear force data alone but it is critical to understanding solder joint quality. A failure mode analysis, along with a fundamental understanding of the shear test process and results is needed to accurately assess solder joint quality.A common pad finish on BGA packages and PWB substrates is nickellgold, using either electrolytic or electroless deposition processes. There are reports of brittle fracture in BGA and PWB interconnects with both types of nickellgold surface finishes. During post assembly test and manufacturing, the solder to pad interconnection has been shown to separate under certain conditions. In this study, ball shear testing is performed in conjunction with thermal preconditioning to evaluate brittle fractures on a thermally enhanced BGA with electroless Ni/Au bond pads and on a plastic BGA with electrolytic nickellgold bond pads. Cross sectioning and fractography reveal interfacial failure in both types of packages. However, detailed metallography and scanning electron microscopy with energy dispersive x-ray analysis confirm two distinct interfacial failure mechanisms. The failure analysis is interpreted in terms of the implications on package quality and performance. Although this evaluation is performed on BGA packages, the results of the failure analysis are applicable to brittle fractures on PWBs that use a NVAu surface finish.0-7803-5502499 $10.00 01999 IEEE 1999 IEEVCPMT lnrl Electronics Manufacturing Technology Symposium 16 IntroductionAs the demand for smaller, faster and lighter electronic products with increased functionality continues to escalate, the semiconductor packaging industry has experienced a shift from traditional peripheral leaded devices to area array technology. More specifically, Ball Grid Array (BGA) technology has become one of the high performance semiconductor packages of choice for advanced applications. One of the primary factors contributing to the quality of a BGA package is the robustness of the solder bail attachment to the package. Currently, ball shear testing is used throughout the

Shear strength and aging characteristics of 63Sn−37Pb solder bumps with various solder ball and pad sizes

“…SEM observa- tions for all solder bumps with different solder ball and UBM pad sizes showed that failure during ball shear testing occurred as a form of ductile fracture in the solder regardless of reflow time. As shear crack propagated in the solder, the shear strength of the solder bumps was determined by the shear strength of the solder alloy itself, making irrelevant the shear strength on the reflow time [3,14]. The mechanical properties of the solder bump are dependent upon its microstructure and composition.…”

“…The Au layer is plated to improve the solder wettability and to protect Ni from corrosion before reflow. However, Au plating has been reported to have a detrimental effect on the reliability of the solder joints, i.e., degradation of shear strength by thermal aging and thermal fatigue tests [3,[8][9][10][11]. Although some work has been done to figure out the effect of Au plating thickness on the shear strength degradation [8,11], the relationship between the aging characteristics of the solder bumps and the bump shape, which is determined by the relative sizes of the solder ball and UBM pad, has been little reported.…”

“…With the ever increasing demand for miniaturized electronic devices with increased input/out numbers, electronic packaging technology has been shifting from peripheral leaded packages to area array packages such as BGA and flip chip [1][2][3][4]. As the size and pitch of the solder bumps decrease rapidly on the area array packages, the reliability of the electronic packages is more and more determined by solder joint reliability [1].…”

“…As the size and pitch of the solder bumps decrease rapidly on the area array packages, the reliability of the electronic packages is more and more determined by solder joint reliability [1]. In general, ball shear tests have been used to evaluate the solder bump reliability in area array packages, which makes ball shear strength of great concern in the microelectronics industry [3,5,6].…”

Effect of Surface Finishes on Ball Shear Strength in BGA Joints with Sn-3.5 mass%Ag Solder

“…12,13) This trend is continuing despite the evidence that immersion Au/electroless Ni causes or contributes to catastrophic, brittle, and interfacial solder joint fractures. [14][15][16] The electroless plating defect is manifested as brittle fracture at the interface between the Ni deposits and the intermetallic compound. The package surface finish is the important package variable, which is influencing the fracture mode.…”