Ductile fracture of solder-Cu interface and inverse identification of its interfacial model parameters. (November 2017)

• Record Type: Journal Article
• Authors: Xu, Yangjian; Zhao, Shuai; Jin, Guohui; Wang, Xiaogui; Liang, Lihua
• Is Part Of: Mechanics of materials. Volume 114(2017)
• Page Start: 279
• Page End: 292
• Publication Date: 2017-11
• Subjects: Parameter identification -- Ductile interfacial failure -- Cohesive zone model -- Solder -- IMC; Strength of materials -- Periodicals; Mechanics, Applied -- Periodicals; Résistance des matériaux -- Périodiques; Mécanique appliquée -- Périodiques; Mechanics, Applied; Strength of materials; Periodicals; Electronic journals; 620.11
• DOI: 10.1016/j.mechmat.2017.08.013
• Languages: English
• ISSNs: 0167-6636

Abstract

Abstract: Interfacial failure is a crucial issue in the study of reliability in electronic packaging. The failure mechanism is complicated since ductile failure of the solders is generally accompanied with brittle fracture of the intermetallic compounds(IMCs). Besides, it is also greatly influenced by loading conditions. In this paper, fractures of SAC305-Cu interfaces under mode I and mode II loading conditions have been experimentally studied. Their interfacial micro-structures before and after failures have been analyzed using SEM and EDS. At the same time, numerical simulations of interfacial failure have also been performed using an improved cohesive zone model. Finally, the failure mechanisms for both mode I and mode II fractures were disclosed. In addition, though the cohesive zone models(CZM) were widely used to investigate the interfacial failure, their parameter identification is still an important concern, especially when considering both the accuracy and efficiency in the analysis. In our work, a parameter inverse analysis method has been developed, based on the Kalman filter algorithm and the response surface interpolation technique. This interpolation method can hugely improve the computational efficiency, but it will also introduce a large numerical error in the present cases, in which the interfacial failure is accompanied by large plastic deformation of solder, or the parameter searching ranges are set over-large. Aiming at this issue, a range-reduced inverse analysis scheme was further developed in our work. The developed method and scheme were validated robust and accurate through verifications of both pseudo and real experimental data. The cohesive zone model parameters acquired can be effective to simulate the interfacial failure of Cu and Pb-free solder. The whole inverse analysis is automatically achieved and can be easily applied to other complicated interfacial failure problems.