Effect of Unfilled Underfills on Drop Impact Reliability Performance of Area Array Packages

Abstract

Board level drop test is one of the key qualification tests to ensure the solder joint reliability. It becomes critical due to leadfree solder. In this paper, we study the effect of underfills on drop test performance of a fine-pitch Ball Grid Array package (BGA), experimentally and numerically. Failure mode is also compared. There are good correlations between testing and modeling on the effect of underfill on failure modes and failure mechanisms of solder joints. Moreover, the results of testing and modeling show that important parameters affecting drop test performance are position of the package on the board, modulus of the underfill, and, circumstantially, the interfacial fracture toughness of the underfill. An unfilled underfill, as opposed to a silica-filled underfill, can provide satisfactory, or even superior, drop test performance if its interfacial fracture toughness is sufficiently high.The benefit of an unfilled underfill is better processability.

Polymeric reinforcement approaches and materials selection to improve board-level drop reliability of SnAgCu soldered area array packages

“…However, the stress in polymeric reinforcement materials increases with the modulus of materials under drop impact loading conditions and increases the likelihood of cohesive failure of materials [33]. For this reason, the modulus polymeric reinforcement materials should not be higher than their real needs, because once the materials are fractured, regardless of whether there has been adhesive failure or cohesive failure, the stress level in the solder joints will increase significantly and reduce the drop impact life of AAP assemblies.…”

Design for board trace reliability of WLCSP under drop test

“…Preferred board trace routing orientation Figures 12-13 shows the average EPEQ results at outer row of I/Os for 3 angles at which the Cu traces are routed out from the component to PCB. For pads (6,1) to (6,5), there are 3 possible angles (assuming 45° interval) for routing: 0°, 45°, and 315°, whereas for pads (1,6) to (5,6), the routing angles are 225°, 270°, and 315°. For comer pad (6,6), there are 5 possible angles.…”

“…In the literature [3][4][5][6][7][8][9][10][11][12], most researchers still focus on solder joint reliability for board level analysis ofWLCSP, mainly based on know-how gained from BGA drop test performance. However, solder joint failure may not be applicable for certain WLCSP designs, which fail at board trace interconnect, die-side Cu RDL or CuiUBM interface during drop test.…”

Bending strength of adhesive joints in microelectronic components: Comparison of edge-bonding and underfilling

“…Adhesives can be used as underfills [1][2][3][4] where a low-viscosity adhesive completely fills the gap between the component and the PCB, and edge-bonding [5][6][7] where only the perimeter of the component is bonded to the PCB using a more viscous adhesive. In underfilling, which is the more conventional approach, dots of a low-viscosity adhesive are dispensed on the printed circuit board near the periphery of the component where they flow through capillary action between the solder balls before being thermally cured [8][9][10].…”