Annealing effect on adhesion properties of poly(butylene terephthalate)

Injection-molded poly(butylene terephthalate) (PBT) is generally annealed at a temperature above the operating temperature to remove residual internal stress and obtain high-dimensional stability. However, this annealing treatment causes poor adhesion property and affected quality and lifetime for the products. In this study, a plausible mechanism for why such problems are caused by the annealing treatment is proposed using additive free PBT. After PBT was annealed under given conditions, surface structure and adhesion strength with an epoxy adhesive were examined. Surface roughness of the PBT film monotonically increased with increasing annealing time, and the tensile adhesive strength concurrently decreased. In addition, the height difference between hill and valley of the surface after adhesive failure became larger, meaning that the failure took place in the PBT phase beneath the interface with the epoxy region. Based on these findings, it was claimed that the poor adhesion of PBT after the annealing treatment originated from a weak boundary layer, which was formed at the outermost surface of PBT. Finally, the presence of the weak boundary layer was directly confirmed by transmission electron microscopy.     

A note on localized directional weak lower semi-continuity

In minimization problems for functionals f : M → R, M ? E a subset of some infinite dimensional Banach space E, we typically have to rely on weak (sequential) lower semi-continuity of f on the whole space E even if M is a proper subset of E. The main reason for this lack of 'localized' weak lower semi-continuity seems to be that it is not known how to get and/or to characterize weak sequential lower semi-continuity on a subset M without knowing it on the whole space. As a first step to overcome this difficulty we propose the concept of 'localized directional weak sequential lower semi-continuity' and offer a way to implement it, namely in terms of conditions on the Gateaux derivative f′ of f (weak K-monotonicity). This allows to formulate a criterium and new sufficient conditions for the existence of a minimizer.

We conclude with a discussion of applications to the variational approach to the solution of (systems of) nonlinear partial differential equations where we focus on the case of integral functionals of vector fields for which the integrand is not assumed to be quasi-convex.