Influence of Process Parameters on Component Assembly and Drop Test Performance using a Novel Anisotropic Conductive Adhesive for Lead-free Surface Mount Assembly S. Manian Ramkumar, Ph.D. 1 and K. Srihari, Ph.D. 2 1 Center for Electronics Manufacturing and Assembly Rochester Institute of Technology Rochester, NY. Phone: 585-475-6081 Email: smrmet@rit.edu 2 Distinguished Professor and Chair Department of Systems Science and Industrial Engineering Thomas J. Watson School of Engineering and Applied Science Binghamton University, Binghamton, NY 13902-6000. Phone: 607-777-4410 Email: Srihari@binghamton.edu Abstract The novel Anisotropic Conductive Adhesive (ACA) investigated in this research uses a magnetic field to align the conductive particles in the Z-axis direction, during cure, thereby eliminating the need for pressure and the requirement to capture a monolayer of conductive particles. The formation of conductive columns within the adhesive matrix, during cure, provides a very high insulation resistance between adjacent conductors and also eliminates the need for precise printing or dispensing of adhesives onto individual fine pitch pads. The novel ACA can also be mass cured, eliminating the need for sequential component assembly. The formation of columns also alleviates the problems associated with coplanarity errors and varying lead/bump shapes in forming reliable interconnections using traditional ACAs. This research incorporated a variety of components, leaded, leadless and bumped, standard, fine and ultrafine pitch, coarse and fine particle filler formulations, different stencil thicknesses, different cure temperatures and times, and different magnetic flux densities. The print process was performed manually using metal stencils of different thicknesses and metal squeegee blades. The findings from this study, including drop tests of ‘as assembled’ and aged samples (100 hours of thermal and T&H aging) are provided. The filler particle size played a critical role in determining the continuity and the contact resistance of the adhesive joint. Standard pitch devices provided good continuity and contact resistance when assembled with larger diameter filler particles. Fine pitch devices required smaller diameter filler particles. Stencil thickness was found to be a statistically significant factor in determining the contact resistance of the adhesive joint, while magnetic flux and cure temperature were not significant statistically. A magnetic flux of 2000 Gauss and a cure schedule of 150°C for 7 minutes were found to work effectively for different material formulations. A thinner stencil (100 μm – 4 mils) was required to obtain continuity when assembling fine pitch devices whereas a thicker stencil (>200 μm – 8 mils) was required when assembling standard pitch devices. Thicker stencil prints tend to misalign the fine pitch devices considerably after placement and cure. A thick adhesive print, with fine particle material formulation, and high magnetic flux density resulted in the filler particles growing as dendrites past the surface of the adhesive. Vertical orientation for the assemblies during drop test, from a height of 900mm (36 inches), was found to be more reliable when compared to the horizontal orientation, in both the ‘as assembled’ and thermally aged conditions. The daisy chains for all of the components survived 30 drops in the vertical orientation. MicroLeadframe (MLF) devices survived 30 drops in both the horizontal and vertical orientations. These devices showed the best performance during drop testing. None of the assemblies survived the T&H aging for 100 hours, to carry out the drop tests. Introduction High density packaging technologies and the current industry mandate to eliminate ‘lead’ from assemblies have renewed the industry’s interest in new interconnection materials such as Electrically Conductive Adhesives (ECAs). ECAs have been in use for several decades in die attach applications. ECAs consist of a mixture of adhesive matrix and conductive fillers. The adhesive matrix is either a thermoplastic or a thermoset material [1]. Typically, the adhesive matrix by itself is non-conductive (has very high resistance) although there are some new intrinsically conductive adhesives. The adhesive matrix provides the adhesion, mechanical strength, and protection of the metallic contacts, while the fillers within the matrix provide the electrical conductivity between the component termination and the pad. The fillers are typically flakes or particles made of metal or metal coated polymers. ECAs can be categorized as Isotropic Conductive Adhesives (ICA) and Anisotropic Conductive Adhesives (ACA). ICAs offer electrical conductivity in all directions. The volume fraction of fillers is in the range of 20 to 30%. The particle size of the fillers in ICAs is in the range of 1 to 20 μm [2, 3]. ACAs, on the other hand, offer electrical conductivity in only one direction, i.e., in the Z-axis direction, between the component termination and the pad. A very low fraction of conductive fillers, typically in the range of 5-20% by volume, is loaded in the adhesive matrix such that conductivity is only in the Z-axis direction [4]. Particle size in ACAs is in the range of 3 to 15 μm. During the curing process, the typical ACA needs the application of pressure in order to capture a monolayer of particles between the mating 978-1-4244-2231-9/08/$25.00 ©2008 IEEE 225 2008 Electronic Components and Technology Conference Authorized licensed use limited to: Rochester Institute of Technology. Downloaded on February 17, 2009 at 10:17 from IEEE Xplore. Restrictions apply.