Modeling of real-time reliability prediction system for anisotropic conductive film (ACF) processing

This paper models a real-time reliability prediction system for application in the anisotropic conductive film (ACF) manufacturing process. The prediction system contains three major hardware/software components, namely an X-ray detection apparatus, finite element method (FEM) modeling and simulations, and post-manufacturing reliability analysis and quality control. Initially, an X-ray detection apparatus is used to scan the particle distribution in the ACF as it flows towards the roller. In the scanning operation, the coordinates of each particle are obtained and passed to a PC. Based on this scanning information, a FEM computational mesh of the ACF sample is generated in which each grid node corresponds to an individual particle. The ACF mesh is then integrated with an FEM model of the IC/substrate assembly and the relevant geometry parameters defined, i.e. the pad height, the pad length, the pad pitch, and the particle radius. Compression simulations are performed using DEFORM or ALPID software and the corresponding particle flow and redistribution calculated. A reliability analysis of the redistributed particle pattern is then performed based upon the probabilities of both opening and bridging effects. The real-time reliability prediction system modeled in this paper provides an effective reliability analysis and quality control scheme for the processing of ACF compounds.