Effect of high-speed loading conditions on the fracture mode of the BGA solder joint Jong-Woong Kim a , Jin-Kyu Jang a , Sang-Ok Ha a , Sang-Su Ha a , Dae-Gon Kim b , Seung-Boo Jung a, * a School of Advanced Materials Science and Engineering, Sungkyunkwan University, 300 Cheoncheon-dong, Jangan-gu, Suwon 440-746, South Korea b Fundamental Technologies Group, Semiconductor Material R&D Center, Samsung Techwin, SAIT, San14, Nongseo-dong, Giheung-gu, Yongin 446-712, South Korea article info Article history: Received 5 November 2007 Received in revised form 18 July 2008 Available online 10 September 2008 abstract High-speed ball shear test was investigated in terms of the effects of two test parameters, shear speed and shear height, with an experimental and computational simulation utilizing 3-dimensional non-linear finite element modeling for evaluating the solder joint integrity of ball grid array (BGA) packages. A rep- resentative Pb-free solder alloy, Sn–0.7Cu, was employed in this study. Far greater shear forces were mea- sured by high-speed shear test than by low-speed shear test. The shear force further increased with shear speed mainly due to the high strain-rate sensitivity of the solder alloy. Brittle interfacial fractures can be more easily achieved by high-speed shear test, especially in higher shear speed. These were discussed with the strengthening mechanism of the solder materials by density and mobility of dislocations. Shear force decreased with shear height, and it could be found that excessively high shear heights unfavorably affected the test results leading to unexpectedly high standard deviation values or shear tip sliding from the solder ball surface. Ó 2008 Published by Elsevier Ltd. 1. Introduction Due to the environmental and health concerns, the adoption of Pb-free solders has become an inevitable trend in the electronic packaging industry. Nevertheless, compared to the Pb-containing solders, the Pb-free solders are in general stiffer and more fragile to dynamic loads, which are frequently encountered for portable electronic devices during normal or excessive operating conditions [1]. Typical way of examining the brittle Pb-free solder joint reli- ability is through board-level drop test. However, the costly and timely board-level drop test does not usually meet the time frame of the industry in developing and marketing new electronic prod- uct. Therefore, another test method with a simpler procedures and lower-cost equipment is highly expected. In order to predict solder joint reliability under drop conditions with another test method, it is important to increase the testing speed of package level test methods [2]. Currently, the most popu- lar method to evaluate the mechanical joint reliability of the solder ball joint is the ball shear test, because this test is simple and con- venient to implement. However, the ball shear test is not consid- ered suitable for predicting the joint reliability under drop loading, as the applied test speeds, usually lower than 5 mm/s, are far below the impact velocities applied to the solder joint in a drop test. Recently, a modified solder ball shear tester that is capable of performing the high-speed shearing up to several meters per second was developed, which is covering the impact velocities by the drop test [2–5]. With the capable of high-speed shearing of the solder ball joint, the Joint Electronic Device Engineering Council (JEDEC) solder ball shear test standard, JESD22-B117, was also updated with an addi- tion of ‘A’ at the end of the standard code to be JESD22-B117A on October 2006 [6]. A significant addition to the standard was of course the incorporation of the high-speed shear test conditions. However, two important shear test parameters, shear speed and shear height, were not fully standardized yet, which is only giving a range of shear speed for the high-speed loading (from 0.01 to 1 m/s) and shear height (no greater than 25% of solder ball height) for the low-speed and high-speed shear test. In our previous stud- ies considering the effects of the two test parameters in low-speed shear test, we could have known that there are significant effects of the parameters on the shear force value and failure mode [7–9]. Therefore, the effects of the parameters in high-speed shear test should also be systemically investigated for understanding of the failure behaviors of the solder joints and their mechanisms under high-speed loading. The JESD22-B117A also gives the shear failure modes for the acceptance criteria of the solder ball joint. The primary impetus for including failure mode is to provide a mean for screening the type of brittle interfacial fractures caused by too thick intermetallic compound (IMC) layers. During soldering, the solder alloy melts and then reacts with the metallization of the substrate to form IMCs at the joining interface. While forming a thin IMC layer, it is desirable to achieve a good metallurgical bond. However, exces- sively thick reaction layer is generally sensitive to stress, therefore, it might provide sites of initiation or paths of propagation for cracks, though some exceptional cases were also reported [10–12]. This is because the layer is brittle and a microstructural 0026-2714/$ - see front matter Ó 2008 Published by Elsevier Ltd. doi:10.1016/j.microrel.2008.07.066 * Corresponding author. Tel.: +82 31 290 7359; fax: +82 31 290 7371. E-mail address: sbjung@skku.ac.kr (S.-B. Jung). Microelectronics Reliability 48 (2008) 1882–1889 Contents lists available at ScienceDirect Microelectronics Reliability journal homepage: www.elsevier.com/locate/microrel