Torsional braid analysis of the aromatic amine cure of epoxy resins

Abstract

SYNOPSISThe cure behavior of diglycidyl ether of bisphenol A (DGEBA) type of epoxy resins with three aromatic diamines, 4,4'-diaminodiphenyl methane (DDM ) , 4,4'-diaminodiphenyl sulfone (44DDS), and 3,3'-diaminodiphenyl sulfone (33DDS) was studied by torsional braid analysis. For each curing agent the stoichiometry of the resin mixtures was varied from a two to one excess of amino hydrogens per epoxy group to a two to one excess of epoxy groups per amino hydrogen. Isothermal cures of the resin mixtures were carried out from 70 to 210°C (range depending on epoxy-amine mixture), followed by a temperature scan to determine the glass transition temperature ( T,) . The times to the isothermal liquidto-rubber transition were shortest for the DDM mixtures and longest for the 44DDS mixtures. The liquid-to-rubber transition times were also shortest for the amine excess mixtures when stoichiometry was varied. A relatively rapid reaction to the liquid-to-rubber transition was observed for the epoxy excess mixtures, followed by an exceedingly slow reaction process a t cure temperatures well above the T,. This slow process was only observed for epoxy excess mixtures and eventually led to significant increases in T, . Using time-temperature shifts of the glass transition temperature vs. logarithm of time, activation energies approximately 50% higher were derived for this process compared to those derived from the liquid-to-rubber transition. The rate of this reaction was virtually independent of curing agent and was attributed to etherification taking place in the epoxy excess mixtures. 0

Effect of hydroxypropyl cellulose in epoxy network formation

“…In fact, the average values of Ea decreased from 100 kJ mol Ϫ1 (0 wt % of HPC) to 88 kJ mol Ϫ1 (30 wt % of HPC). These values of Ea are comparable to those obtained through the application of the time-temperature superposition method 11 using different techniques such as DSC 1,19 and TBA, 12 where T g is the basic parameter used for the calculations of Ea. The comparison between the mentioned methods is valid, 1,19 even considering the inherent simplification of the Barrett method-which assumes order of the reaction equal to 1-and knowing that the calculation of Ea is obtained only for the first stages (initiation), where, theoretically, there are neither diffusion effects nor vitrification processes.…”

Thermochemical Response of Vinyl-Ester Resin

“…The T g was then plotted against the degree of cure for each isothermal temperature (see Figure 17). For other thermosets, such as epoxy and polyimide systems, the data points for various isothermal temperatures have been found to collapse on one master curve (Gillham and Enns 1994;Zukas 1994). However, as shown in Figure 17, this was not the case with the current system due to the competing reaction mechanism explained previously.…”

“…A mathematical expression describing the T explicitly in terms of the degree of cure is particularly convenient for implementation in cure-dependant residual stress modeling. Equation 12 was chosen because it has been successfully used by other investigators (Hong and Chung 1991;Gillham and Enns 1994;Zukas 1994). in(r,)=-Ac:…”

Polyimide-Epoxy Composites

“…However, Sanz, et al have investigated T g of epoxy systems via DMA for a myriad of epoxy compositions and compiled large amounts of reasonable data using this technique [135]. Zukas has done the same using torsional braid analysis (TBA) on many epoxy systems and produced similar conclusions to Sanz [129].…”

“…The T g of many epoxy systems has been compiled by Zukas [129]. However, as he found, there are many conflicting reports concerning the effects the stoichiometry of the reactants, typically noted as the amino hydrogen/epoxy ratio or more simply the a/e ratio, on the T g of the systems.…”