This thesis presents the results of an investigation into the effects of hygrothermal conditioning on the mechanical properties, and fatigue properties of epoxy based composites reinforced with carbon, glass and Kevlar 49 fibres. Cross-plied laminates (0/90) of these materials of nominal volume o fraction 60% were conditioned by drying at 60°C, by exposure to a 65% RH atmosphere at room temperature and by boiling in water. The effects of conditioning on the tensile and shear strengths and on the tensile fatigue response are discussed. The effects of exposure to an extreme diurnal cycle and to the ultra violet in isolation on the tensile properties are also discussed. The 0/90 tensile properties of the three laminates are relatively little affected by the environmental conditioning except for the case of GRP exposed to boiling water, when corrosion damage to the glass fibres significantly reduces the composite strength, and in the KFRP in which the strength is reduced by complete drying. The +/-45 strengths are more sensitive to the effects of moisture, however, it appears that the optimum strength is obtained after conditioning in the 65% RH environment. Acoustic emission monitoring of the tensile tests shows distinctive differences between KFRP and the two other types of composite and permits the identification of characteristic effects of moisture on the tensile failure mechanisms of all three materials. Tensile fatigue tests have been carried out on the laminates in the 0/90 orientation. The CFRP shows no effect of conditioning on the fatigue behaviour, and in the GRP only the boiling water conditioning affects the results. The tensile fatigue of the KFRP is affected both by boiling and by drying, the latter being the most severe. The fatigue response of the KFRP shows a dramatic down turn at lives in excess of 105 cycles. This effect appears to reflect the ease with which mechanical damage is sustained by the aromatic polyamide fibres. The residual strengths of the laminates after fatiguing is discussed and possible mechanisms for the damage accumulation in the materials during fatiguing are given.

The static and fatigue properties of advanced epoxy-based composites reinforced with carbon, glass or aromatic polyamide (Kevlar-49) fibres have been measured for a range of different loading and environmental conditions. Cross-plied laminates were tested in tension in the 0/90 and +/-45° orientations and also under flexural loading. The laminates were similar, except for the type of fibre. The effects of environmental exposure were assessed by preconditioning test specimens to equilibrium by either drying at 60°C, storage at 65%RH at ambient temperature or boiling in water. Moisture absorption was through the resin alone for CFRP and GRP and by additional absorption by the fibres for KFRP. Fatigue testing revealed that the tensile performance in the 0/90 orientation is strongly dependent on the level of cyclic strain. 0/90 CFRP has excellent fatigue and environmental resistance but GRP exhibits a steep fatigue curve and the static and low cycle fatigue strengths are both reduced by boiling. The fatigue strength of 0/90 KFRP is reduced by drying, more so than by boiling, and in all conditions the stress/log-life curves are characterised by a downward curvature or 'knee'. Tensile preloads do not significantly affect the residual fatigue properties or the equilibrium levels of moisture uptake, although extensive damage involving cracking in both longitudinal and transverse plies may lead to increased absorption rates. A tendency for Kevlar fibres to split or 'defibrillate' plays an important role in most failures of KFRP. It limits the shear strength and causes flexural failures to occur at the compression surface at low stress levels. 0/90 CFRP also fails at the compression surface in flexure but GRP fails at the tensile surface, the environmental fatigue performance resembling that under axial tensile loading. The +/-45° tensile and low cycle fatigue strengths are sensitive to the effects of conditioning, all laminates exhibiting optimum performance after conditioning at 65%RH, although generally these effects become insignificant at long lives.