The effect of temperature on the behavior of the interphase in polymeric composites

Fibre/resin interaction in fragmentation tests: coaxial vs. simple specimens comparison

“…While all these experiments verify more or less Cox's early assertion (Cox, 1952) that the length is proportional to √ (E f /G m ) with E f fibre modulus and G m shear modulus of the matrix, Skourlis & McCullough (1993) found that the temperature dependence of the fragment length for the system they were studying was marked by an 'almost steplike increase' at temperatures lower than the glass transition temperature of the bulk resin. They assumed that an interphase of lower glass transition temperature was present with a set of properties, especially the shear modulus, different from those of the bulk matrix.…”

“…They assumed that an interphase of lower glass transition temperature was present with a set of properties, especially the shear modulus, different from those of the bulk matrix. Extending the Cox theory to a three-phase concentric model with a geometry very similar to that of our coaxial specimens, Skourlis & McCullough (1993) predicted that the fibre/ matrix stress transfer efficiency is a function of both the matrix modulus and the modulus of the interphase. They assume the final fragment length l to be given by…”

“…The demonstration was given as early as 1978 for glass fibres in epoxy or polyester resins tested up to 100-120 ЊC by Ohsawa et al (1978) who concluded that 'if the interfacial strength is greater than the matrix shear strength, the apparent strength calculated from the fragmentation test is merely the matrix shear strength'. Other authors conducted fragmentation tests on the (necessarily) same limited range of temperatures with glass (DiBenedetto & Lex, 1989;DiBenedetto, 1991) or carbon fibres (Wimolkiatisak & Bell, 1989;Skourlis & McCullough, 1993;Detassis et al, 1995) or with resin mixtures of variable stiffness associated with carbon (Rao & Drzal, 1991) or Nicalon SiC fibres (Ho et al, 1995); they all observed an increase of the fragment length with temperature or the reduction of the matrix modulus. In the last two studies, the authors point out that the extension of the fibre/matrix debonding process does not change very much from one resin system to the other and thus the comparison of the elastic reloading modes is quite relevant.…”

“…In common with the original Cox model there are a number of assumptions made by Skourlis & McCullough (1993) to derive their model that do not satisfy the actual stress state in the fragmentation test: perfect bonding, elastic behaviour of the matrix, etc. In particular, the hypothesis of perfect bonding is far from being verified for carbon/resin systems where there is a large extension of debonding, especially with some fibres.…”

Dewetting of Unreacted Epoxy/Amine Mixtures on Silica

“…In recent years, data in the literature have led us to assume that the composition of a polymer in the vicinity of a surface can differ substantially from the bulk composition (21)(22)(23). For example, concentration gradients will arise if one of the reacting components of the resin mixture has a higher affinity for the surface than the other components (24).…”