No abstract

SYNOPSISThe cure process and the mechanical properties of blends of diglycidyl ether of bisphenol-A-based epoxy resin and hydroxyl terminated, internally epoxidized polybutadiene rubber have been studied. Internal oxirane groups are characterized by a main absorption a t 885 cm-' in the infrared spectrum while the terminal oxirane groups of the diepoxide monomer absorb a t 913 cm-'. In the absence of prereaction of the rubber, gelation of the epoxy matrix occurs much faster than any reaction involving the internal oxirane groups or the terminal hydroxyl groups. Therefore, only weak chemical bonding between the rubber particles and the epoxy matrix exists and the fracture toughness of the blends does not show any significant improvement. Upon prereaction of the rubber with an excess diepoxide monomers, a 40% improvement in the value of the critical stress intensity factor is obtained.However, dynamic mechanical spectra of these blends acquired in the rubbery plateau region uniquely demonstrate that this improvement is due to the incorporation of the rubber into the epoxy network rather than to the presence of phase-separated rubber particles. 0 1994 John Wiley & Sons, Inc. INTRODUCTIONIt is well known that a small amount of rubber can greatly improve the fracture toughness of epoxy resins by forming discrete rubbery particles that are chemically bonded to the matrix. The toughening effect is generally accompanied by a relatively small loss of thermomechanical properties. A review of the literature indicates that both chemistry ',* and thermodynamic^^'^ play an essential role in determining the final properties of these rubber/epoxy blends. Thermodynamics include issues of miscibility and phase separation. Initially, the rubber must be able to dissolve in the monomer/curing agent/ catalyst mixture. However, miscibility must not be too great because phase separation of the rubber before gelation of the epoxy matrix is also required for efficient toughening. Too good of a miscibility leads to a diluent effect rather than a toughening effect. The balance between phase separation and polymerization depends on the solubility of the rubber in the epoxy resin, the initial fraction of rubber, and the cure conditions. This balance, in turn, determines the particle size distribution and the composition of the epoxy-rich and rubber-rich phases and, thus, directly influences the mechanical properties. Chemistry of the rubber includes issues of selectivity and reactivity. Functional groups must be present on the rubber backbone in order to selectively react with the epoxy matrix. Indeed, chemical bonding between the rubber particles and the epoxy matrix is necessary for efficient stress transfer. It has been proposed' that the rubbery particles are composed of a mixture of linear rubber-epoxy copolymers and crosslinked epoxy resin. In the absence of chemical bonding or curing, only van der Waals attractive and/or hydrogen bonding type of forces are present and not much toughening is expected. If, on the other hand, the reactivi...

The dynamic mechanical properties of blends of diglycidyl ether of bisphenol‐A‐based epoxy resin and internally epoxidized polybutadiene rubber have been studied. It is shown that the influence of the composition of the continuous phase and of the dispersed phase can be studied not only from the variations of the glass transition temperature but also from the changes in the apparent enthalpy of activation associated with this transition. As the initial rubber content increases, the composition of the dispersed phase remains practically constant while more rubber is able to dissolve in the continuous phase. Analysis of the rubbery plateau region reveals that the shear modulus of the blends is not much affected by the presence of dissolved rubber in the continuous phase but strongly depends on the volume fraction of dispersed phase. This volume fraction can be obtained from the relative drop in shear modulus after modeling the results with the Kerner equation. The results compares well with independent measurements by scanning electron microscopy. © 1994 John Wiley & Sons, Inc.

The surface induced crystallization of poly(e-caprolactone) (PCL) on an ultra-high modulus polyethylene (PE) fibre was investigated using a new approach based on the induction time t i. This approach allows estimation of the free energy difference function A~ as it appears in the theory of heterogeneous nucleation. The classical approach based on the rate of heterogeneous nucleation / is not applicable to transcrystallization because the nucleation density at the fibre surface cannot be measured. A relationship between /and t i is proposed and a theoretical justification is presented. Good agreement between the two approaches is obtained for a verification case where both / and t i can be determined. A transcrystalline growth rate study yields an estimate of the parameter 0(3-e for PCL of 680 erg 2 cm -4. The maximum growth rate g* is also obtained. The results obtained indicate the influence of certain parameters on the appearance of transcrystallinity. It is also shown how the interfacial morphology can be controlled by the knowledge of the variations of the induction time with temperature.

No abstract