Role of the interphase in the flow stability of reactive coextruded multilayer polymers

Polymer Engineering and Science, April, 2009 by Khalid Lamnawar, Abderrahim Maazouz

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Rheology of a Multilayer System

Two multilayer structures, prepared by a coextrusion process (T = 230[degrees]C smaller than reaction temperature T = 240[degrees]C, contact time = 70 s for all structures), were studied: one containing varying amounts of interfacial surface but a constant volume fraction of the two components, and the other containing varying volume fractions of the components, but with a constant amount of interfacial surface between them. To form multilayer films, individual films were placed on top of one another at room temperature. Compositions of 30, 50, and 70 vol% of PE were prepared while the surface between the two materials was kept constant. Multilayer films with a composition of 50 and 70 vol%, but with varying amounts of interfacial surface, were also prepared.

All film samples were fabricated under identical processing conditions to eliminate sample-to-sample errors. The obtained samples were then cut into disks with diameters of 25 mm. Subsequently, these samples were annealed at 40[degrees]C under vacuum for 1 week to allow the relaxation of chains oriented during the coextrusion process. In our previous work, we have used rheometry as a tool for probing the competition between interdiffusion and reaction at a polymer/polymer interface in a bilayer sandwich structure. The viscoelastic properties seemed to be a suitable way to characterize and understand the multilayered reactive polymers. This results in this paper deal with dynamic mechanical experiments to probe the interface effect on the rheological behavior of the reactive multilayered structure. The storage modulus, G', loss modulus G", and dynamic viscosity [eta]* were measured as functions of frequency or in dynamic time sweep measurements. Typical curves of the rheological behavior for multilayer systems with different amounts of interfacial surface were plotted and compared with their corresponding constituents.

Figure 7 portrays the viscosity vs. the angular frequency at 240[degrees]C for (a) a reactive PE-GMA/PA6 multilayer with 70 vol% of PE-GMA for a varying number of layers/interfaces and its respective neat components and (b) a nonreactive PE/PA6 bilayer system (with 50 vol% PA6) and its respective neat components.

[FIGURE 7 OMITTED]

Indeed, the flow curve of PE-GMA/PA6 (RS) showed a sharp increase in viscosity at low frequencies (yield). However, this phenomenon was not present in the case of the pure polymers. In fact, [eta]* ([omega]) for the NRS was also plotted and as can be seen no significant increase in viscosity was found, thus confirming that the phenomenon observed was caused by the reactivity at the interface at low shear rates. There was, thus, a clear distinction between the two systems. Moreover, we demonstrated in our previous works (Lamnawar and Maazouz (25), (26)) that the reaction was able to remove the interfacial slip confirming also the generation of the interface.

The multilayer viscosities showed significant increases with increasing interfacial areas. In other words, increasing the number of interfaces also increased the viscosity moduli. The results displayed that, as a consequence, the variation in viscosity of the multilayer system reflected the occurrence of both diffusion and chemical reaction. Because of the increase in interface adhesion strength due to the coupling reactions between epoxy, carboxylic, and amino groups, interdiffusion phenomena were found to increase when the kinetics of the chemical reaction at the interfaces also improved. Indeed, the interfacial reaction could only occur when PA6 and PE-GMA chain ends were able to penetrate through each layer and the copolymer interface to meet the antagonist reactive functions at the interface. Obviously, the diffusion rate of PA6 through the bulk layer was different from the one through the copolymer layer. The former was obtained through self-diffusion in the same phase, whereas the latter was dependent on the penetration rate through the interface, which should be much slower than the self-diffusion due to the denseness of the copolymer barrier. Meanwhile, a clear relationship between viscoelastic material functions of multilayer systems and compositions can be evaluated to analyze the effect of bulk and reactive functions in the polyolefin phase at the interface with PA.