Stitched shape memory alloy wires enhance damage recovery in self-healing fibre-reinforced polymer composites

A major issue in composite technology is matrix micro-cracking due to low-velocity impact damage, which may severely limit service lifetimes of composite parts. In a novel approach, remarkable levels of healing of impact damage are obtained using shape memory alloy (SMA) wires to close longitudinal cracks in woven glass fibre-reinforced polymer plates with an epoxy-polycaprolactone (EP-PCL) matrix that shows dual-phase continuity. Thermal actuation of SMA wires stitched through the thickness of the stacked glass fibre plies introduces compressive loads to the cracks thanks to anchoring of the SMA loops at the fabric surfaces and debonding of the intervening threads, which prevents local deformation of the SMA, so that crack closure by about 200 μm is achievable. Concomitant expansion of the vascular network formed by the molten PCL fills the compressed cracks, resulting in highly effective healing on cooling, as demonstrated by C-scan images. Specimens stitched with SMA wires hence show almost complete healing, i.e. damage area recovery of 85%, after low-velocity impact at up to 17 J followed by heat treatment at 150 °C. This represents a 55% improvement over previous results for unstitched EP-PCL composites, and hence significantly greater degrees of healing than so far reported for this range of impact energies and this type of system.

Damage recovery after impact in E-glass reinforced poly(epsilon-caprolactone)/epoxy blends

Amaël Maximilien Cohades, Véronique Michaud

Damage recovery after low-velocity impact has been assessed in woven E-glass fibre-reinforced polymer composites with an epoxy matrix and a blend of epoxy and 25 vol% of poly(e-caprolactone) (PCL). Impact was carried out at three energy levels (8.5, 17, 34 J) and composites with epoxy-PCL blends demonstrated similar energy absorption capacity as compared to pure epoxy composites even though the extent of damage (quantified by C-scans and optical microscopy) was higher. Ultimate compressive residual strength of the modified composites was, for the different impact energy levels, 23-33% lower as compared to unmodified composites. Healing efficiency after a thermal mending cycle at 150 degrees C for 30 min has been quantified using three complementary characterization methods; impact damage could be recovered from 20% to 100% depending on the impact energy level. These modified matrix composites are thus able to fully recover low-velocity impact damage at energy levels often met in real structures. (C) 2017 Elsevier Ltd. All rights reserved.

Elsevier2017

Functional composites with damage control and repair

Amaël Maximilien Cohades

Fibre Reinforced Polymer (FRP) composite structures are subjected during service to low energy impacts or unexpected loads, leading to damage. Microcracks are generally first formed in the matrix and can reach up to several hundred microns in thickness. In many living systems, an initial self-sealing phase followed by self-healing of the original tissue leads to highly effective repair of minor damage events. A bio-inspired strategy for the efficient healing of microcracks in FRPs is hence to: (i) autonomously close cracks whose thickness is above a certain threshold to ensure better crack faces registry; (ii) use a healing matrix with structural properties close to that of a conventional FRP matrix, that is able to expand and fill the cracked regions repeatedly after multiple damage events. This strategy was investigated in the present work through (i) the use of Shape Memory Alloy wires (SMAs) as stitches in FRPs and (ii) a healing matrix based on thermoset-thermoplastic phase-separated blends. Blends composed of epoxy resin and different thermoplastics (PCL, PLA and PMMA) were evaluated for their potential as healing matrix, based on their room temperature toughness, stiffness and their capacity to heal when subjected to a moderate heating cycle. Three types of blend morphology resulted from polymerisation-induced phase separation during cure, depending on the thermoplastic content, including an interconnected particulate epoxy phase and a co-continuous thermoplastic phase at high thermoplastic contents. An optimal composition was found for epoxy-25vol%PCL blends. The PCL phase expands by 14% in volume upon melting at 150 °C, therefore enabling filling of small cracks. When further integrated to FRPs (with an adapted vacuum infusion moulding process), this healing matrix led to composites with similar stiffness and strength to that of pure epoxy composites, but also to full recovery of compression after impact strength for low damage extent (impacts of 8.5 J). To provide healing for larger damage extent, NiTiCu SMA wires, with a diameter of 150µm, were stitched in the dry fabric before processing. After damage and upon heating at 150 °C, SMA stitches, that have been stretched and partly debonded upon crack propagation, efficiently closed the cracks. This procedure demonstrated the capacity of the wires to close 200 µm thick cracks in the FRPs and led to fully heal impact damage events up to 17 J, corresponding to the main concern of maintenance activities in the composite industry (e.g. tools dropped from 1-2 m height). Finally, as an alternative to blends, PCL electrospun nanofibrous veils were interleaved into fabrics, infused with epoxy and cured to reach a microstructure combining both phase-separated domains as well as intact nanofibre regions. Interlaminar crack propagation demonstrated up to 48% toughness increase as compared to reference specimens. However, healing was prevented due to reduced flow of PCL in the fine channels resulting from phase separation, showing the limitation of this approach as compared to the use of blends and stitches. Phase-separated epoxy-PCL composites (with or without stitched SMA wires), considering their manufacturing feasibility through conventional industrial processes, their acceptable mechanical properties and their ability to fully heal low-velocity impact damage, demonstrated their relevance for composite structures that are subjected to moderate loads and not easily accessible to repair.

EPFL2018

Tailored Capsule Based Self-Healing for Epoxy Matrix Composites

Sam Neuser

Self-healing materials are a novel class of materials able to repair structural or, more generally, functional damage to a part. They hence have the potential to improve reliability and cost effectiveness, because failure or loss of functionality may be postponed or completely avoided. This concept has been applied to all classes of materials, but is particularly suited to polymers and their composites. Indeed, it was first demonstrated for epoxies containing capsules of dicyclopentadiene and dispersed Grubbs’ catalyst. However, the high cost of the catalyst precludes commercial use of this self-healing system. In the present thesis, two alternative approaches to self-healing of an epoxy matrix were investigated: (1) a solvent based self-healing epoxy; (2) a two-component epoxy/amine self-healing system using a novel encapsulation technique. The solvent-based system showed efficient healing in an under-cured epoxy matrix. The solvent in question was ethyl phenylacetate (EPA) encapsulated with urea-formaldehyde (UF) or UF-polyurethane. It was demonstrated that during a crack event, the solvent diffused into the epoxy matrix following a non-Fickian case II profile, with a geometrical swelling of the surface layer of the crack. For a healing time of 24h, a critical crack gap of 30μm was identified, beyond which the swelling was insufficient to bridge the crack faces and the healing efficiency was limited to around 25%. For smaller crack gaps, solvent swelling led to intimate contact of the crack faces and reaction of residual functional groups through a reduction of the glass transition temperature Tg of the epoxy, resulting in improved healing efficiency. The use of shape memory alloys (SMA) that are strained during the fracture event and subsequently heated to reduce the crack opening was also investigated. High healing efficiencies were observed and it was found that longer SMA wire activation times further improved the healing efficiency up to almost 80% through improved crack closure and an increased degree of polymerization. The behavior of this material during aging was then investigated in partially cracked samples, guaranteeing small crack gaps. It was found that the premature diffusion of the solvent through the microcapsule membrane into the epoxy matrix led to a decrease in healing efficiency from around 75% in fresh samples, to about 15% in samples aged for 75 days. Increasing the thickness of the polymeric shell of the capsules did not significantly improve performance. Moreover, moisture uptake by the epoxy also contributed to the reduced healing efficiency. Because the healing agent is hydrophobic, a higher water content in the epoxy matrix limited solvent diffusion into the epoxy and hence the extent of healing. A potential solution to the problem of premature diffusion of solvent into the epoxy is suggested to be the use of other shell materials or additional coatings on the capsules. The response of the solvent based self-healing system to dynamic fatigue was also studied. While in static testing, matrix swelling governs crack healing, a different mechanism was observed in the case of dynamic loading. In contrast to the unmodified epoxy, which showed rapid crack propagation under all the conditions investigated, in the presence of the solvent, crack arrest was observed for stress intensities of up to 0.48 MPam1/2. This phenomenon was attributed to crack pinning owing to diffusion of the solvent into the matrix surrounding the crack tip. The elongation at break of the solvated epoxy increased threefold, and the Young’s modulus decreased by more than 50%. Using finite element analysis (FEM) to model the stress distribution around the solvent-infused crack tip, it was shown that the local crack tip stress was reduced by 43%, accounting for this novel toughening effect. While solvent healing performed well in an under-cured matrix and interesting phenomena were observed, a fully cured matrix with a two-component self-healing system may be preferable for real applications because it avoids issues pertaining to premature solvent diffusion. The possibility of encapsulating an amine curing agent in a core-shell structure was therefore investigated as an alternative. A microfluidic glass chip was used to encapsulate a hydrophilic amine/water based curing agent in an acrylate resin shell using ex-situ exposure to UV light. Low viscosity epoxy was encapsulated with UF using a modified agitated emulsion encapsulation technique. Successful self-healing was demonstrated for epoxy specimens containing these microcapsules, along with embedded SMA wires, with healing efficiencies of up to 100% and 40% for low and high temperature curing epoxy, respectively. Overall, this work represents a significant step towards affordable epoxy self-healing materials, the feasibility of solvent and epoxy/amine healing having been successfully demonstrated for an epoxy matrix. The epoxy/amine self-healing system is thought to be particularly promising, considering the early stage of development. Furthermore, the integration of SMA wires proved to improve healing efficiencies dramatically, especially with the epoxy/amine microcapsules. With further optimization and scale-up of the microfluidic encapsulation method, a new low cost self-healing system, and hence potentially compatible with a wide range of polymeric materials, including composites, is therefore within reach because it is based on a commonly used epoxy/amine adhesive.