Mechanics of adhesive failure. II

In the preceding Part I of this investigation a relation was derived among the measured failure energy, $\theta $, the energy dissipated viscoelastically during joint separation, and an intrinsic failure energy $\theta _{0}$ for various rubber-to-polymer substrate joints. For some joints $\theta _{0}$ was equal to the thermodynamic work of adhesion $W_{\text{A}}$ out for others $\theta _{0}$ $\gg $ $W_{\text{A}}$. By the use of a variety of microscopical and spectroscopical techniques it is shown in the present paper that when $\theta _{0}$ $\approx $ $W_{\text{A}}$, joint failure is wholly interfacial, but that $\theta _{0}$ $\gg $ $W_{\text{A}}$ when substantial cohesive failure occurs during joint separation. It is shown that the intrinsic failure energy (which controls the total failure energy under given conditions) may be expressed as $\theta _{0}=iI+r\scr{T}_{0}+_{8}F$, where $i,r$and $s$ are the area fractions of interfacial failure, rubber cohesive failure and substrate cohesive failure respectively and $I,\scr{T}_{0}$ and $F$ are the corresponding failure energies (per area). For purely interfacial failure, $i$ = 1 and $I=W_{\text{A}}$. For strong joints, however, about 70 to 80% of the value of $\theta _{0}$ is provided by the term $r\scr{T}_{0}$. The departures from interfacial failure, which occur only with etch-treated substrate films, can be attributed to covalent bonding across the interface during cure of the elastomer. The reactive groups in the substrate surface are C=C double bonds produced by the etching treatment.