Two metallocene ethylene‐oct‐1‐ene copolymers, differing in comonomer content and in molecular weight, were cross‐linked either by dicumyl peroxide or β‐radiation. The effect of high comonomer content on the crystalline morphology, once the materials were cross‐linked, was analyzed by differential scanning calorimetry (DSC) and dynamic mechanical analysis (DMA). The gel content was determined in boiling xylene, and the cross‐linking process was monitored by FT‐IR spectroscopy. Two endotherms were distinguished: the first one was associated to the primary crystallization, and the second one to the shorter sequences that are excluded from the primary crystallization. The successive self‐nucleation annealing (SSA) technique has revealed that as the comonomer content increases the crystal size distribution is more homogeneous, and therefore, the melting and crystallization behaviour is reversible, because of its fringed‐micellar morphology. In spite of their crystalline morphology with very low crystallites, DCP cross‐linked samples displayed a considerable decrease in crystallinity and in crystal size, whereas β‐irradiated samples showed no significant decline in crystal size. Slight changes in crystallinity were detected and attributed to the heat generation that every irradiation process involves and affects smaller crystallites preferentially. DMA analysis has confirmed DSC results on crystalline size and crystallinity variations induced by both cross‐linking processes. By means of FT‐IR spectroscopy, it was detected that a high comonomer content induces oxidation during cross‐linking. Moreover, β‐irradiation samples exhibited a lower degree of oxidation than DCP cross‐linked samples.The heating scan of DCP cross‐linked EG8411 after being submitted to successive self‐nucleation annealing.magnified imageThe heating scan of DCP cross‐linked EG8411 after being submitted to successive self‐nucleation annealing.

Summary: Two metallocene EPDMs with the same weight fraction of ethylene but differing in diene content were crosslinked, either by dicumyl peroxide (DCP) or β‐radiation. The effect of different diene and propylene content on the molecular structure and the mechanical properties once the materials were crosslinked was studied. The final gel content was very high due to the large level of unsaturations. The crosslinking process was monitored by FTIR spectroscopy by following the decay of unsaturations and the variation of the carbonyl groups that are related to the oxidation grade. It was found that β‐radiation crosslinked samples exhibited a lower oxidation grade than those crosslinked by DCP. An oscillant disc rheometer was employed to follow the evolution of the rheological properties, the scorch time, and the time corresponding to full cure during the crosslinking reaction with DCP. In addition, in order to characterize the state of cure we have studied the rheological properties in shear employing a dynamic parallel plate geometry. These results were correlated with those obtained from the molecular characterization of the soluble fraction by size exclusion chromatography. The experiments indicate that, at low irradiation doses, there is a high rate of chain scission reactions that cause an important decrease in storage modulus. Whereas, at high irradiation doses the rate of chain scission reactions diminishes, thus the storage modulus increases but it still remains at lower levels than those corresponding to the original terpolymers. The tensile properties, hardness (Shore A) and compression set tests also suggest the presence of chain scission reactions.Storage modulus (G′) versus frequency for a β‐irradiated sample.magnified imageStorage modulus (G′) versus frequency for a β‐irradiated sample.

From the point of view of the chemical reactions describing the curing process of a mixture epoxy resin~ticarboxylic acid anhydride-tertiary amine-polyol and silica filler, it is difficult to understand the network formation: several different reactions are involved, so that the interaction between the epoxy resin and the other components of the mixture is unquestionably complex. However, a linear plot is obtained whenKissinger's method is applied to dynamic DSC results (four different scanning rates), which confirms the calculation assumptions. In particular, the Borchardt and Daniels equation, where the specific rate constant is assumed to be of Arrhenius form appears to be a good mathematical model for describing the curing process under dynamic DSC conditions. The apparent activation energy determined by means of this analytical method is in good agreement with those obtained by other methods in the literature.