JEC COMPOSITES MAGAZINE - Issue #34 - July/August 2007 - 39

Raw material in the molecular weight of the reacting UPE resins have been proven to lead to phase separation. The macro-phase separation leads to the formation of macro-domains of polyester around MgO aggregates. These domains are surrounded by a phase enriched in PVAc. The next step, after thickening, is curing. Upon heating, the molecular weight of the polyester network is expected to decrease due to the rupture of the weak ionic bonds formed during the thickening reaction. Some re-solubilization of the macro-domains is possible before the curing reaction starts. In these experiments, curing occurs without any agitation, which does not favour re-solubilization. The low curing temperature (100°C) is not high enough to break all the ionic bonds formed during thickening reaction. This alters the micro-phase separation and leads to a coarser and more connected microgel structure when compared to non-thickened systems. A higher curing temperature (140°C) enhances the micro-phase separation process by breaking more ionic bonds. This leads to finer and more independent microgels after curing in the PVAcenriched phase, even after several days of thickening. Another positive effect is a better re-solubilization of the macro-phase separated domains, but for short thickening times only, when macro-phase separation is still limited. For longer thickening times, the elevated curing temperature does not lead to a total re-dissolution of the macro-phase separated domains of UPE resin. The use of acid-modified PVAc brings some benefits (higher polarity than with non-acid-modified PVAc and therefore better compatibility with the UPE resin) but does not prevent the macro-phase separation of thickened samples. The mechanism of the low profile effect in the absence of a thickening agent is dictated by a reaction-induced phase separation. In the presence of a thickening agent, the mechanism follows the same basic principles but is influenced by changes in compatibility. Contrary to the reaction-induced phase separation, which is limited in time because of gelation, the thickeninginduced macro-phase separation has the viscosity as a limit. Therefore, the effects of the thickening-induced phase separation are potentially more harmful, as it has no clear time limit. This is a clear indication that moulding of compounds should be done at low viscosities to limit the macro-phase separation, which should still allow good glass fibre transport during moulding. Conclusion Adding the thickening agent MgO to UPE/PVAc mixtures considerably modifies the interconnected globule morphology, which is responsible for the shrinkage control mechanism. The role of PVAc is thus reduced. The main change is the formation of macro-phase separated domains during thickening that remain during cure. The morphology of the area surrounding these domains is more continuous in the presence of MgO. The use of acid-modified PVAc improves the stability of the system, but does not prevent the macro-phase separation from occurring. MgO induces a more continuous morphology that will act more like a UPE resin (shrink), and the presence of macro-phase separated domains will reduce the amount of microvoids. A proper design of LPA should include potential material modification, by acids or other means such as surfactants, to further stabilize the compatibility of the blend before cure. Acknowledgments This work was partially supported by the Swiss National Research Fund (FNRS) under contract n° 2000-067947.02, and by the Swiss Innovation Agency (CTI) under contract 5468.2EUS in the framework of the Eureka SURFAS project E!23731. More information: http://ltc.epfl.ch No34 July-August 2007 / JEC Composites Magazine

http://ltc.epfl.ch http://www.quickstep.com.au

JEC COMPOSITES MAGAZINE - Issue #34 - July/August 2007

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #34 - July/August 2007