Commercial polysiloxane resins used in the semiconductor industry were moulded on Ag, Ni and Cu real surfaces. The samples were broken by using a pull test machine and the joint strength measured. The ‘resin’ and ‘metal’ faces obtained in the rupture process were analysed by ESCA. The chemical composition of the two faces was very different. The one from the resin side resembled that of the bulk. That from the metal side was rich in additives from the commercial resin. Some differences were found between the three metals, whilst the spiral flow of the resin in the range examined influenced the results in only a very limited way. The weak boundary layer in which the above additives accumulate seems responsible for the low adhesion level found. The Pb/Si concentration ratio at the fracture locus is correlated with the adhesive joint strength. The multifactorial nature of the adhesion in practical samples is discussed. It is suggested that the influence of the heavy elements present in the additive has to be explored in connection not only with the adhesion problem but also with the electrical reliability of the device.
No takes yet. Share an insight, caveat, or question.
Torrisi et al. (1983) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: