Search

Package structures are continually becoming more complicated to achieve high-performance devices. Thus, the 2D structure of package needs to be modified to a 2.5D or 3D structure. Therefore, standoff properties at the solder joint are required to prevent Si chip damage and electrical shorts when the solder joint is located between the substrate and interposer. Cu-core solder balls (CCSB) are the most popular interconnection choice for a 2.5D package because they have better standoff properties than the general SAC solder. The mechanical properties of CCSBs were evaluated by ball shear tests and measuring bump height as compared with Sn–3.0 wt.%Ag–0.5 wt.% Cu solder. The bump heights of CCSBs and Sn–3.0 wt.%Ag–0.5 wt.% Cu were measured under several compressive pressure conditions at room temperature, 175 °C and 235 °C. The stand-off height of the CCSBs dramatically better than that of Sn–3.0 wt.%Ag–0.5 wt.% Cu solder under high pressure and temperature. The shear strength of the CCSBs was stronger than that of SAC solder at all multiple reflows due to the finer microstructure and higher stacking of dislocations at the interface of the Cu core ball surface.

The microstructures and mechanical reliability of Sn–58Bi solder and epoxy Sn–58Bi composite solder joint were investigated with organic solderability preservative surface finishes. The mechanical reliabilities of Sn58Bi and epoxy Sn58Bi solder were evaluated by the board-level drop test and the 3-point bend test after temperature-humidity storage testing. The microstructure and chemical composition of the solder joints were characterized by scanning electron microscopy and energy dispersive X-ray spectroscopy, respectively. The addition of epoxy in solder paste did not affect the morphology of the intermetallic compound. The thickness of the scalloped-shaped Cu6Sn5 intermetallic compound of solder/OSP joint increased with aging time. The drop number until fail for the epoxy Sn58Bi/OSP joint was higher than that for the Sn–58Bi/OSP joint; the average numbers of drops withstood by the Sn–58Bi/OSP joint and epoxy Sn–58Bi/OSP joint following the reflow process were fewer than 10 drops and 180 drops, respectively. The drop number of solder/OSP joints decreased with increasing aging time. The result of the 3-point bend tests shows that the number of bend cycles for the epoxy Sn–58Bi/OSP joint was 30 times higher than that for the Sn–58Bi/OSP joint. The number of bend cycles for solder/OSP joints was decreased with increasing aging time.

As demanding for high performance and miniaturization of electronic devices, interconnection materials required higher reliability in mechanical, thermal and electrical. The importance of electromigration issue has increased because of these trends. We evaluated the electromigration behavior of Sn58Bi solder and Sn58Bi epoxy solder under high temperature and constant current flow. The electromigration test-kit was a designed and fabricated flip chip-type module and the diameter of the solder bump was 250 μm. A current was passed through the two solder joints, producing a current density of 3.0 × 103 A/cm2 at 100 °C. The microstructure of solder joint after electromigration test were investigated with field-emission scanning electron microscopy during electromigration, a Bi-rich layer was observed at the anode side of the solder joint and the formation of Kirkendall voids was observed at the cathode side of the solder joints. Different inorganic materials affect electromigration in the eutectic Sn58Bi solder joints.

An epoxy Sn–58wt.%Bi solder joint was evaluated by a three-point bending test with electroless nickel immersion gold (ENIG) and electroless nickel electroless palladium immersion gold (ENEPIG) surface finishes aged at 85 °C and 85% relative humidity. Scanning electron microscopy and electron probe microanalysis were carried out to study intermetallic compound variation. The morphology, total thickness, and chemical composition of intermetallic compound in epoxy Sn58Bi solder joints were the same as those of Sn–58wt.%Bi solder joints with each surface finish. The average number of bending-to-failure cycles for the epoxy Sn–58wt.%Bi solder/ENIG joints and epoxy Sn–58wt.%Bi solder/ENEPIG was more than 4000 and 5000, respectively. The average number of bending-to-failure cycles of the epoxy Sn–58wt.%Bi solder joint decreased with increasing age. Three-point bending reliability of epoxy Sn–58wt.%Bi solder joints was higher than that of Sn–58wt.%Bi solder with both surface finishes. Cracking of all solder joints subjected to as-reflowed was propagated through the solder matrix. However, after aging for 1000 h, cracking occurred primarily between intermetallic compound layers.