JEC Group have brought together the international community of composites leaders and executives in our Composites Circle as an unique networking opportunity to meet with both peers and future partners.
Curing of unsaturated polyester (UPE) resins is a cross-linking chain growth copolymerization of unsaturated polyester molecules with predominantly styrene monomer. This reaction is associated with significant polymerization shrinkage. As a consequence, composite parts moulded with these resins suffer from lower dimensional stability and poor surface quality. Surface quality is significantly improved through the addition of Low Profile Additives (LPA). Many different types of LPA are available; all of them are thermoplastics, with different degrees of compatibility with the UPE resin.
1- M. MONTAGNE, PRANGINS, SWITZERLAND
2- X. BULLIARD, EPFL – LTC, LAUSANNE, SWITZERLAND
3- V. MICHAUD, EPFL – LTC, LAUSANNE, SWITZERLAND
4- J-A. E. MÅNSON, EPFL – LTC, LAUSANNE, SWITZERLAND
(Published on July-August 2007 - JEC Magazine #34)
The shrinkage control mechanism in UPE resins in the presence of an LPA is a reaction-induced phase separation (spinodal decomposition) that occurs during the initial stage of cure. Phase separation is driven by changes in the molecular weight and the polarity of the UPE resin upon polymerization. This leads to the formation of two phases, one phase enriched in polyester and the other, in LPA. The morphology related to improved shrinkage control is described as co-continuous, although it corresponds to interconnected polyester globules in a matrix enriched in LPA. Micro-voiding, occurring at the end of the reaction or during cooling down, contributes to shrinkage control and dimensional stability.
Most of the research on the shrinkage control mechanism was carried out in the absence of thickening agents. During the manufacture of Sheet Moulding Compounds (SMC), thickening agents are added in order to significantly increase the viscosity, ensuring easy handling before the moulding stage and homogeneous flow of glass fibres, during. The main thickening agents employed in industry are MgO and Mg(OH)2. Thickening is an acid-base reaction between the acid groups of UPE and LPA molecules and alkaline magnesium oxide (or hydroxide). The bonds created during the thickening reaction are ionic by nature and will rupture when heated. As thickening takes place before curing, it can affect the compatibility of a UPE/ LPA blend. Industrial practice has shown that the surface quality of moulded parts decreases in the presence of thickening agents.
In this study, a maleic anhydride/propylene glycol resin was combined with both an acid-modified (PVAc-A) and a homopolymer PVAc (PVAc-B) with a molecular weight of approximately 90,000 g/mol.
The effect of thickener addition on phase stability
The morphology of a reference sample containing 14 wt. % of PVAc-A (acid modified) in the absence of MgO is featureless and remained this way for at least 5 days at 30°C.
The morphology changes of the same blend during thickening at 30ºC in the presence of MgO are shown in Figure 1. Figure 1.a corresponds to the early stage of thickening and reveals some MgO aggregates. The presence of MgO creates instability, which is characterized by the formation of round-shaped domains. These domains coalesce to a size of 500-800 μm around the thickener aggregates (Figure 1.b). A real SMC consists of approximately 43% organic phase and 57% filler and glass fibre by volume. This has a significant limiting effect on the size achievable by these macro-domains.
To identify the phases shown in Figure 1, samples thickened for two days were cured at 140°C on a hot plate that was mounted under a microscope. Upon heating, the macro-domains exhibited small changes in shape but did not re-dissolve. Heating activated the copolymerization between styrene and the UPE resin, which induced a micro-phase separation in the zone surrounding the macro-domains (Figure 2.a). A magnification in the upper corner of this figure shows the structure that developed during micro-phase separation. Upon additional curing, microvoid formation created the black zones around the macro-domains (Figure 2.b). It appears likely that the macrodomains are based on UPE resin with little or no PVAc and that the continuous region surrounding the macro-domains corresponds to a phase enriched in PVAc, still containing UPE resin. No differences could be detected when replacing PVAc-A with PVAc-B (homopolymer).
The following terminology will be used to avoid confusion: macro-phase separation refers to phase separation induced by the thickening reaction before curing and micro-phase separation refers to phase separation induced during curing by the cross-linking reaction between UPE chains and styrene.
UPE/PVAc-A mixtures were cured at 100ºC and 140ºC. Figure 3.a shows the reference morphology for a blend containing 14 wt. % of PVAc-A in the absence of MgO. The micro-structure of the samples cured at 100ºC is, as expected, coarser than that of the samples cured at 140ºC. At elevated temperatures, gelation occurs earlier and the time between micro-phase separation and gelation available for the development of the microstructure is shorter. Growth and coarsening of the microgels are suppressed earlier, leading to finer microgels.
The addition of MgO did not lead to a severe change in the microstructure, even after prolonged thickening time. Figure 3.b shows an example for a sample thickened during two days. The microgels remained rather independent and are smaller than for the sample without MgO. A lower magnification examination of the fracture surface showed the presence of macro-phase separated domains of UPE resin at the bottom of the samples (Figure 4). This macro-phase separated phase (being mostly or only UPE resin) effectively increases the PVAc concentration in the other phase. Previous work has shown that, with increasing PVAc concentrations, the diameter of the microgels gets smaller. These macro-phase separated domains in the samples cured at 140ºC appeared after a slightly longer thickening time, because the higher curing temperature could have induced some resolubilization of the macro-domains.
The reference morphology of PVAc-B in the absence of MgO and cured at 140°C is shown in Figure 5.a. As for PVAc-A, with short to moderate thickening time, smaller and more independent microgels are evident. However, a significant change occurred for longer thickening time (4 days). The polyester microgels were no longer independent and fused to form a continuous phase (Figure 5.b). This was not observed in the case of PVAc-A. A macro-phase separation, as with PVAc-A, was observed at the bottom of the samples. Macro-phase separated domains were visible after a slightly shorter thickening time, when compared to PVAc-A. For both types of PVAc, a decrease in the size of microgels can be observed for a thickening time of 2 days, which is most likely a PVAc concentration effect.