The ball-on-ring (BOR) and 3-point bending (3PB) tests were used in this study to characterize the effect of dicing process on the fracture strength of Si dies. Dies prepared by a blade dicing process and a laser dicing process were studied. The edge-initiated fracture was distinguished from the surface-initiated fracture by the fractographic analysis. The fracture strength distributions related to surface flaws in the 3PB test as well as the BOR test were consistent regardless of dicing process. For the edge defect-induced failure mode, on the other hand, blade-sawn dies showed wider spread distribution than laser-sawn dies. It was due to the scattered nature in size and location of edge flaws induced by blade dicing. The laser-sawn dies showed a tighter distribution of the die strength although the average die strength was slightly lower than that of the blade-sawn dies. This work successfully demonstrated that the die failure caused by the edge defects can be deconvoluted from the 3PB test data by using the fractographic observation.

You do not currently have access to this content.