Solder alloys of 80Sn-20Pb were electrodeposited on the Cu-based leadframe alloy(PMC-102) from an organic sulfonate bath with or without an additive using direct or pulse currents, and influences of plating parameters on the microstructure of the electrodeposits and behavior of the intermetallic compound formed at 80Sn-20Pb/PMC-102 interfaces were investigated.
Alloy deposits of 80Sn-20Pb were obtained either from an organic sulfonate bath containing 28% $Pb^{2+}$ at 10 $A/dm^2$ or from a solution with additive at 8 $A/dm^2$. At a cathodic current density of 10 $A/dm^2$, the cathodic polarization for 80Sn-20Pb plating in solution with grain refiner agent increased 500 mV more than that in solution without the agent, thereby producing a very fine electrodeposit. The grain size of pulse electrodeposit was extremely fine at low duty cycle and low pulse frequency but getting coaser with increasing the duty cycle and/or the frequency, approaching that of deposit formed by direct current. X-ray diffraction studies showed that the electrodeposit of 80Sn-20Pb is a mixture of Pb and β-Sn crystals. Whereas the Pb crystal was found to be randomly oriented, the β-Sn crystal exhibited a transformation of {200} preferred orientation to {220} one with decreasing the duty cycle and/or the frequency.
On aging at 150℃ in vacuum(Burn-In), intermetallic compound layers of $\eta'$-phase and ε-phase were formed between PMC-102 and 80Sn-20Pb alloy deposit, and the thickness of total intermetallic compound layer($\eta'$+ε) increased with increasing aging time. Growth rate of intermetallic compound depended on the microstructure of electrodeposit, that is, samples with coaser grain, in particular, those electroplated from bath not containing additive by dc had lower growth rate.
The density of surface crack occurring in 90˚ bending test increased with increasing the thickness of intermetallic compound layer for samples with the same microstructure of electrodeposit. Comparing the surface crack density of samples with different microstructure of electrodeposit however, relation between the surface crack density and the thickness of intermetallic compound layer was not clear. For the same aged samples, surface crack density of samples electrodeposited from a bath containing an additive was the least. The improved resistance to cracking of solder layer electrodeposited from a bath containing an additive seems to be resulted from the diffusion barrier effect of the additive adsorbed during electrolysis.
