Simon A. Hayes

Self-healing of damage in fibre-reinforced polymer-matrix composites

Self-healing resin systems have been discussed for over a decade and four different technologies had been proposed. However, little work on their application as composite matrices has been published although this was one of the stated aims of the earliest work in the field. This paper reports on the optimization of a solid-state self-healing resin system and its subsequent use as a matrix for high volume fraction glass fibre-reinforced composites. The resin system was optimized using Charpy impact testing and repeated healing, while the efficiency of healing in composites was determined by analysing the growth of delaminations following repeated impacts with or without a healing cycle. To act as a reference, a non-healing resin system was subjected to the same treatments and the results are compared with the healable system. The optimized resin system displays a healing efficiency of 65% after the first healing cycle, dropping to 35 and 30% after the second and third healing cycles, respectively. Correction for any healability due to further curing showed that approximately 50% healing efficiency could be achieved with the bisphenol A-based epoxy resin containing 7.5% of polybisphenol-A-co-epichlorohydrin. The composite, on the other hand, displays a healing efficiency of approximately 30%. It is therefore clear that the solid-state self-healing system is capable of healing transverse cracks and delaminations in a composite, but that more work is needed to optimize matrix healing within a composite and to develop a methodology for assessing recovery in performance.

Fibre/matrix stress transfer through a discrete interphase. Part 1: single-fibre model composites

Abstract This paper reports a study of the effect of an interphase on strain development in fibre-fragments. In order to form an interphase, an epoxy resin with known properties was applied to the surface of unsized reinforcing fibres and cured. These were then embedded in a matrix resin coupon, prior to fragmentation testing. The study included an examination of the effect of interphase thickness, by applying multiple coats of one of the resins, and the effect of the interphase properties, by varying the coating resins. It was found that the average fragment lengths at saturation were difficult to distinguish, as a result of the scatter introduced by the statistical distribution of fibre strengths. However, the strain interval between onset of fragmentation and saturation was found to be more sensitive to variations in the interphase properties. A finite element model was used to examine the strain development in the fragments in more detail. The mechanical properties of the fibre, interphase and matrix were accurately incorporated into the model, providing a realistic representation of the state of strain in the experimental samples. The predicted deformations around the fibre-break provided an explanation for the experimental observations.

Fibre/matrix stress transfer through a discrete interphase: 2. High volume fraction systems

The effect of plasticity on the reinforcing efficiency of a broken fibre, and the magnitude of the strain concentration experienced by the surrounding fibres, has been assessed by the use of a 3-dimensional finite-element model. It was found that the occurrence of plasticity in the matrix markedly reduced the strain concentration in fibres adjacent to a fibre fracture. The effect of increasing the fibre volume fraction on the level of strain concentration was examined and it was found that when deformation was elastic, at low applied strain, a higher fibre volume fraction led to an increase in the strain concentration. However, when plastic deformation occurred, the strain concentration factor was significantly lower and increasing the fibre volume fraction had a negligible effect. The influences of soft or stiff interphases between fibre and matrix were also studied and, during elastic deformation, these were found to be largely insignificant in determining the stress transfer processes, for the interphase thickness studied. At higher strains, the occurrence of plastic deformation in either the matrix or interphase was found to dominate the strain-transfer process and, therefore, the strain concentration. The reasons behind these dependencies are discussed and their effect on the failure of bulk composites considered.

Measurement of micro stress fields in epoxy matrix around a fibre using phase-stepping automated photoelasticity

The stress field in an epoxy resin matrix near the broken-end of a glass fibre was measured using a phase-stepping automated polariscope. Fringe order contour maps were obtained continuously in a region of 1.835×1.835 mm using this photoelastic technique. Principal stress difference, at any point in the matrix near the fibre-matrix interface and the fibre-end, could be calculated. The present technology enables continuous fringe contours and the distribution in principal stress difference around a fibre-end with partial debonding or a matrix crack to be obtained. Thus the effects of debonding and a transverse crack, emanating from a fibre-break, on the matrix stresses have been fully described.

Calibrating a nanoindenter for very shallow depth indentation using equivalent contact radius

Nanoindenter tips are usually modelled as axisymmetric cones, with calibration involving finding a fitting function that relates contact area to contact depth. For accurate calibration of shallow depth indentation, this is not ideal because it means that deeper indents tend to dominate the fitting function. For an axisymmetric object, it is always possible to define an equivalent contact radius (which, in the case, of nanoindentation is linearly related to the reduced modulus) and to obtain a fitting function that relates this equivalent contact radius to indentation depth. The equivalent contact radius approach is used here to provide shallow depth calibration of a nanoindenter tip at three separate times. The advantage of the equivalent contact radius methodology is that it provides a clearer physical interpretation of the changes in tip shape than a conventional area-based fit. We also show that the minimum depth for a reliable hardness measurement is obtainable and increases as the tip blunts with age but that consistent measurements of very near surface elastic moduli can be made if the blunting of the tip over time is fully accounted for in the tip area function calibration.

Inkjet printing of self-healing polymers for enhanced composite interlaminar properties

Inkjet printing has been used to introduce an organic system that demonstrates thermally activated self-healing in composites. The organic system is composed of monomers that, when polymerised, are capable of thermally activated self-healing through a reversible Diels–Alder mechanism. After being synthesised the monomers were formulated into inks and inkjet printed on to carbon fibre epoxy prepreg. The polymers were co-cured with the prepreg into composite laminates and the effect on the interlaminar properties of the resultant system was investigated. A single ply at the mid-plane of double cantilever beam specimens was shown to increase the initiation (by NL Point) of the interlaminar fracture toughness by 9%. The interlaminar fracture toughness with regards to crack propagation was shown to increase further by up to 27%. Increases in apparent interlaminar shear strength as measured by short beam shear of up to 11% were also observed compared to unprinted controls. After a thermal treatment the short beam shear specimens are retested and the printed specimens are shown to have significantly smaller decreases in properties compared to the control which is consistent with repair in the interlaminar region.

Direct electrical cure of carbon fiber composites

Abstract This paper presents a study of the cure of carbon fiber composites by the direct application of an electric current. The conductivity of carbon fiber enables the individual fibers to act as heating elements. The result is a large number of heating elements throughout the composite structure. These heating elements locally heat the resin surrounding them, resulting in the resin curing. This study shows that cure of composites is achieved in samples up to 60 × 25 cm. Initially, the required contact arrangement to obtain a uniformly heated panel is assessed and an optimum arrangement achieved. Following this, the change in required energy as the sample thickness and sample length are changed is measured. The degree of cure is compared to conventionally cured composites using differential scanning calorimetry (DSC). Three-point bend testing is used to determine the flexural strength and modulus of the composite samples. The results show that a comparable level of cure can be obtained using direct application of an electrical current as that obtained using autoclave cure and oven cure processes.

Photoelastic study of the stress transfer in single fibre composites

The stress transfer between a single fibre and a matrix has been studied using the technique of automated phase-stepping photoelasticity. The contours of isochromatic fringe order provide the locations and magnitude of maximum shear stress. A model composite in which a fibre-break leads to the propagation of a matrix crack has been used to compare the interfacial shear stresses around bonded and debonded interfaces in the absence and presence of a fibre-break and matrix cracks.