The influence of cure conditions on the morphology and phase distribution in a rubber-modified epoxy resin using scanning electron microscopy and atomic force microscopy

In this paper, we consider the effect of cure conditions on the morphology and distribution of the rubber in a phase separated rubber-modified epoxy resin, which in effect is a two phase composite. Novel aspects of this study were measuring the elastic modulus of the dispersed rubber phase particles by atomic force microscopy (AFM) and verifying the presence of nano-dispersed rubber. The purpose of introducing dispersed rubber particles into the primary phase in these systems is to enhance their toughness. It is known that both the rubber particle size and volume fraction affect the degree to which the epoxy is toughened. It is not known, however, how the specific mechanical properties of the rubber phase itself affect the toughness. The objectives of this study were to: (1) use scanning electron microscopy (SEM) and atomic force microscopy (AFM) to determine the morphology and phase distribution of the rubber particles and (2) to measure the mechanical properties of the rubber particles using AFM. Ultimately, we would like to develop a clear understanding of how the changes in morphology and mechanical properties measured at the micro and nano-scales affect both the elastic modulus and fracture toughness of rubber-modified epoxy polymers. The epoxy system consisted of a diglycidyl ether of bisphenol-A, Epon 828, cured with piperidine and incorporating a liquid carboxyl-terminated acrlonitrile-butadiene rubber (CTBN). The carboxyl groups of the rubber are capable of reacting with the epoxy. The cure conditions considered were based on a statistically designed full factorial curing matrix, with the variables selected being cure temperature, initiator (piperidine) concentration, and rubber acrylonitrile concentration. Each of these primary variables was found to affect the phase distribution that resulted during cure. A statistical analysis of the effect of these variables on the phase morphology showed that the acrylonitrile content (%) of the rubber affected both the rubber particle size and volume fraction. The cure temperature strongly influenced the rubber particle volume fraction and modulus. Volume fractions of the rubber phase of up to 24% were obtained even though the amount of rubber added was only 12.5%. The rubber particle modulus varied from 6.20 to 7.16 MPa. Both the volume fraction and modulus of the rubber particles were found to influence the macroscopic mechanical properties of the composite. While larger volume fractions favor improved toughness, we note that that the toughness is greatest when the particle modulus values do not exceed similar to6.2 MPa. Thus, increased volume fraction by itself may not always result in increased toughness. The particles also must be sufficiently 'soft' in order to improve toughness. In the system of interest here, the processing conditions are a key factor in achieving the most appropriate material properties. By inference, this is likely to be the case as well in other rubber-modified thermosets. (C) 2004 Elsevier Ltd. All rights reserved.