No abstract

The thermal degradation behaviour of a polyurethane foam, synthesised from TDI and a polyether polyol, is reported. The thermal degradation behaviour of this material was evaluated by a combination of thermogravimetric analysis (TGA) and thermal volatilisation analysis (TVA). The results demonstrated that the thermal degradation is a complex process which consists of competing mechanisms which yield an array of degradation products. The TVA results revealed that the degradation occurs in two steps, with the initial step corresponding to degradation of the urethane linkages by two competing mechanisms. The first mechanism, proposed to be the predominant mechanism, involves simple depolymerisation of the urethane bond to yield TDI and polyol. A second, competing mechanism is proposed to occur which involves dissociation of the urethane linkages to yield DAT, CO2 and alkene-terminated polyol chains. The second degradation step has been shown to involve degradation of the polyol which was regenerated in the first degradation step. This is proposed to occur by random radical chain scission of the polyol to yield propene, formaldehyde, acetaldehyde, C3H6O isomers and high molar mass polyol chain fragments of various structures. Isothermal TVA studies have revealed that this occurs as low as 250 degrees C under vacuum but does not become significant until temperatures greater than 300 degrees C

SynopsisPoly(ccapro1actone) (PCL) of ca. 20,000 molecular weight is shown to be partially miscible with three aromatic-amine-cured epoxy resins. This conclusion is based on the depression of the epoxy T,, the effect on physical and mechanical properties, and the observation that a large proportion (40-55%) of the PCL ester groups are involved in hydrogen bonding. This miscibility behavior is compared to PCL blends with anhydride-cured epoxy resins, which appear to have a two-phase morphology. The different miscibilities are rationalized on the basis of the existence of functional groups (e. g., hydroxyl) in amine-cured epoxies which are capable of hydrogen bonding to the PCL ester groups. Anhydride-cured epoxy resins contain fewer potential hydrogen bonding sites. INTRODUCTIONEpoxy resins are used widely as the matrix component of high performance composite materials because of their stiffness, chemical resistance, and high-temperature stability.'v2 The fracture properties (fracture energy and tensile strength) of crosslinked epoxy resins, however, are of some concern, and a n extensive patent and scientific literature exists dealing with additives which modify these properties.s For example, the fracture toughness of an amine-cured epoxy resin can be increased by a factor of 3 by the incorporation of a carboxyl-terminated rubber: while other additives are described5s6 which can increase the tensile strength of amine-cured epoxy resins by at least 50%. As part of a larger study of the effect of additives on the properties of crosslinked epoxy resins, we describe here a study of epoxy residpoly(ecapro1actone) (PCL) blends and, in particular, the role of hydrogen bonding in influencing the miscibility of PCL with crosslinked epoxies.The majority of polymer-polymer blends are immiscible, i. e., the polymers do not mix on a molecular level.7 However, PCL has been shown to be miscible with a number of other polymers, particularly when an opportunity exists for hydrogen bonding or other polar interactions with the second p~lymer.~-l' The most extensive published study of epoxy resin/PCL blends is by Noshay and Robeson,12 who examined the miscibility of a range of anhydride-cured epoxy resins with PCL of different molecular weights and with different end groups. They concluded that above a critical molecular weight of PCL, the blends had a two-phase structure and that the * To whom correspondence should be addressed. PCL end groups reacted with the anhydride curing agent to produce a type of block copolymer. The purpose of the present work is to examine PCL blends with amine-cured epoxy resins and to compare their properties with anhydride-cured systems, because of the opportunity which exists for hydrogen bonding in amine-cured epoxy resins (reaction of an amine with an epoxide produces hydroxyl groups while reaction of an anhydride with an epoxide largely produces ester groups). EXPERIMENTALThe epoxy resins used for this study were of commercial origin and are classified here according to their major constituents: (a)...

A combination of dielectric relaxation, dynamic mechanical thermal analysis, and positron annihilation measurements is reported on melt-cast films of poly(ethylene terephthalate), poly(ethylene naphthalate), and their copolymers. The effects of change in the chemical structure of the polyesters are rationalized in terms of increased restriction of the mobility of the polar segments of the chain by the incorporation of the bulky naphthalate structure. All the quenched samples exhibit physical aging when raised to elevated temperatures, the rate depending on the degree of undercooling used in the aging experiments. The free volume surprisingly does not change significantly with temperature and leads to the suggestion that the reduction in the dielectric permittivity is a consequence of a reduction in the mobility of the local segments. The increased storage modulus is also consistent with a reduction in mobility. This proposal is further confirmed by the observation of a good correlation between the rates and extents of the physical aging as detected by dynamic mechanical and dielectric relaxation measurements. Similarities in the activation energies of the β relaxation process for all the polymers investigated indicate that the dipole relaxation processes have a common origin and can be ascribed to motion of the linking polar entity. This study implies that aging is accompanied by an increase in ordering within these polyesters.

SynopsisThe reaction product of 4-hydroxyacetanilide and 1,2-epoxy-3-phenoxypropane, when added at 19 wt % to a conventional epoxy-resin-curing agent mixture, increases the tensile strength of the cured system from 82 MPa to 123 MPa and increases the shear modulus (20°C, 1 Hz) from 970 MPa to 1560 MPa. As well as showing increased strength, the tentile-test specimens also fail in a ductile fashion, i.e., the slope of the stress-strain curve is negative a t failure, with appreciable localized deformation occurring during fracture. For notched samples (compact tension specimens), the fracture properties are strongly strain-rate-dependent. At low strain rates the additive-containing sample has a fracture energy (!?I~, critical strain energy release rate) about twice that of the additive-free control, but a t higher strain rates 91, falls to about 65% of the control value. The critical stress for crack propagation is also strain-rate-dependent and is about 50% higher than the control a t low strain rates and about 10% less than the control a t higher strain rates. Dynamic mechanical analysis and dielectric loss measurements indicate that the additive causes a decrease in the TR and a suppression of the &relaxation. Chemically, the additive accelerates the cure process but does not significantly alter the final extent of reaction of the epoxy resin. After curing, the additive is almost totally extractable by solvent indicating that it is not chemically bound to the polymer. These observations are discussed in terms of the concept of antiplasticization.