Abstract
Self-healing is receiving increasing interest worldwide as a technology to autonomously address the effects of damage in composite materials. This paper describes the results of four point bend flexural testing (ASTM-D6272-02) of T300/914 carbon fibre reinforced epoxy with resin filled embedded hollow glass fibres (HGF) which provided a self-healing functionality. The study investigated the effect of the embedded HGF on the host CFRP mechanical properties and also the healing efficiency of the laminates after they were subjected to quasi-static impact. Specimens were tested in the undamaged, damaged and healed conditions using a commercial two-part epoxy healing agent (Cytec Cycom 823). Microscopic characterisation of the embedded HGF was also undertaken to characterise the effect on the host laminate fibre architecture.
1. Introduction
One focus in the development of advanced composite materials has been to address their vulnerability to impact damage. The orthotropic nature of composite materials results in a relatively low through-thickness strength which is often mitigated by creating damage tolerant designs for structural components. For example, CFRP used in aerospace applications is typically assigned an allowable compressive strain level of <0.4% [1,2] whereas commercially available carbon fibres typically have a compressive strain to failure of around 1%. This results in conservative design and over-weight structures. Furthermore, the sensitivity of advanced composite materials to impact damage results in high maintenance and inspection costs in order to ensure that potentially significant damage events do not go undetected.
Self-healing is a novel alternative to damage tolerant design and removes the need to perform temporary repairs to damaged structures. It derives inspiration from biological responses to damage, such as the human process of haemostasis, in order to impart an autonomous healing functionality into a composite material. This technology has the potential to mitigate damage resulting from an impact event, thereby providing an opportunity to improve the design allowables for CFRP or offer other benefits such as reduced maintenance and inspection schedules.
To date, research into self-healing of polymeric materials has considered a number of approaches [3–7]. These include the use of either glass tubes or microcapsules, containing a healing agent, which is released into a damage site upon fracture. Polymerisation of the healing agent then restores a degree of mechanical performance. There has also been significant research into the use of such systems in fibre reinforced polymers (FRP) [8–15]. Research at the University of Bristol is seeking to impart self-healing functionality to a composite by using embedded resin filled hollow glass fibres (HGF) [13–15]. A bespoke fibre making facility [16–19] has been used to produce HGF of between 30–100 𝞵m diameter and a hollowness of around 50%. These were then embedded within glass fibre reinforced plastic (GFRP) and infused with uncured resin to provide a healing functionality to the laminate. This paper focuses on the progression of this work by considering its application to carbon fibre reinforced plastic (CFRP), a material more widely used in aerospace applications.
2. Experimental methodology
To date, the self-healing work at Bristol has incorporated HGF within GFRP laminates as discrete plies [13–15]. It was decided that this approach would not be suitable for CFRP laminates as it would effectively produce a hybrid glass–carbon laminate and result in a significant reduction in mechanical properties. A less detrimental method was derived whereby a small number of discrete HGF’s are dis- tributed within a CFRP ply where they act solely as distributed storage vessels for the healing agent. It was imperative that the embedded HGF did not detrimentally affect the mechanical performance of the CFRP laminate but could provide a sufficient volume of healing resin to address any damage. Therefore, the distribution of HGF within a laminate poses a problem of optimisation. Four point bend flexural testing (according to ASTM-D6272-02 [20]) was selected to assess the self-healing efficiency of the resulting CFRP as it has already been successfully used to demonstrate self-healing in GFRP [13–15].
2.1. Specimen manufacture
Carbon fibre/epoxy (Hexcel T300/914) pre-preg was selected as the host laminate as it is widely used for aerospace applications. Quasi-isotropic (QI) plates (230 mm x 160 mm x 2.6 mm) were manufactured using hand lay-up [16 ply ( 45°/90°/45°/0°)2S]. Cure was undertaken according to manufacturer’s recommendations. Two different HGF distributions, of fibre pitch spacing of 70 𝞵m and 200 𝞵m, were wound directly onto uncured CFRP plies prior to lamination. These resulted in 3% and 1% volume fraction of HGF, respectively, for a typical CFRP sample. This was to investigate the effect of HGF on the host laminate properties and the healing effectiveness of different healing agent volumes. The 70 𝞵m pitch ensured that the fibres were effectively nestled side by side, and the 200 𝞵m spacing ensured roughly three fibre diameters of space between HGF so that optimal embedment was achieved within the laminate. HGF was located at two 0°/ 45° interfaces within the lay-up as follows and in the same orientation as the 0° ply: ( -45°/90°/45°/0°/HGF/ 45°/90°/45°/0°/0°/45°/90°/ 45°/ HGF/0°/45°/90°/ -45°)
This interface was selected as the location for HGF embedment to fulfil two requirements:
- The 0° direction (i.e., the longest dimension of the test samples) ensured that a maximum length of HGF was contained within the sample and consequently a maximum amount of healing resin was available.
- Previous work [21–23] regarding the embedment of optical fibre within a composite, states that an optimum is achieved when the embedded fibre possesses the same orientation of at least one of the adjacent plies.
2.2. Laminate microstructure
Specimen sections from two different test panels were potted in epoxy resin, and polished (P120–P2500, diamond 6 𝞵m and 3 𝞵m) and the microstructure examined optically. This provided a visual assessment of the disruption caused by the different HGF configurations on the host CFRP plies and identified any inconsistencies. Furthermore, it provided images to accurately determine the outer diameter and hollowness of the HGF, and the mode of damage propagation through the material.
A fibre spacing of 70 𝞵m highlighted several issues about the resulting quality of embedding HGF within a laminate. This small pitch spacing was selected to ensure the HGF were in close proximity (Fig. 1) and thereby facilitate a high degree of healing efficiency. However, during manufacture HGF could be displaced resulting in severe disruption to the host ply (Fig. 1b, c). This was attributed to fluctuations in the fibre pitch control at low HGF spacing combined with the low ‘tack’ of the 914 epoxy at ambient temperature. This was seen to result in lengths of HGF detaching from and reattaching to the surface producing fibre clumping and resin rich regions which were generated within the laminate after completion of the cure cycle.
Fig. 1. (a) HGF spaced at 70 𝞵m showing, (b) good embedding and (c) HGF clumping within host laminate.
Fig. 2. (a) HGF spaced at 200 𝞵m showing, (b) consistent spacing and (c) excellent embedding within host laminate.
A fibre spacing of 200 𝞵m resulted in a much higher quality laminate (Fig. 2). The resin rich regions surrounding the HGF were minimised and there was no evidence of HGF clumping. The large spacing between fibres allowed the low ‘tack’ of the 914 resin system to hold them in place, and any that did detach were able to reattach without contacting any neighbouring HGF. The large spacing between HGF permitted improved consolidation of carbon fibres and matrix resin during laminate cure further improving the overall HGF embedment and reducing the disruption to the laminate.
2.3. Mechanical testing
A support span to depth ratio of 32:1 and a support span to load span ratio of 3:1 were selected according to ASTM D6272-02. This resulted in specimen dimensions of 100 mm x 20 mm x 2.6 mm, each sample was measured and average values were used for the strength calculations in accordance with the ASTM standard. The variation in thickness between samples was less than 1.5%. Ten samples were cut from a plate with the use of a water-cooled diamond grit saw. The sample edges were polished with SiC paper (P2500) to avoid any unwanted edge effects. Samples were then dried, sealed in sample bags and stored in a temperature and humidity controlled environment prior to testing.
Fig. 3. (a) Typical load–displacement curves for quasi-static indentation of CFRP (3 damaged and 3 healed samples), cross-sectional damage for impact force (b) 1700 N (c) 2000 N.
Immediately prior to mechanical testing, the HGF within each specimen were infiltrated using a vacuum assist technique with pre-mixed, two-part epoxy healing resin (Cytec Cycom 823) with a mix ratio 4:1 by weight.
Quasi-static impact damage was imparted to each specimen using a 5 mm diameter spherical tup mounted on a Hounsfield H20K-W (20 kN load cell) electromechanical test machine. Under load control, the sample was supported by a steel ring of 27 mm outer diameter and 14 mm inner diameter. The indentations were stopped at a peak load of either 1700 N or 2000 N. A typical load–displacement curve (Fig. 3) shows two yield points are evident, one at 1400 N, and another at 1700 N with a maximum load achieved at 2000 N upon which the load can no longer be sustained and the tup begins to penetrate the laminate causing significant back face damage. Up to this point the damage is contained within the laminate and can be likened to BVID: The impact surface suffers a minor indent ( 0.3 mm), the back face experiences minimal distortion due to back face delamination and minimal back face fibre break out. The shear crack/delamination distribution within the laminate forms the characteristic ‘pine tree’ distribution as would be expected from a drop weight impact event.
After indentation, the specimens were left at 70 °C for 45 min to reduce healing resin viscosity (25 cps) and facilitate infiltration into damage sites. This was followed by a cure schedule of 125 °C for 75 min. Whilst this process diverges from the original aim of achieving an autonomic healing ability, the use of a premixed healing resin and elevated temperature after a damage event was an attempt to make the Cycom 823 more suitable as a healing resin and attempt to demonstrate the highest level of healing efficiency possible with this system. Cycom 823 is not designated as a healing agent and was chosen as being the best available at this time. In fact, no resin system currently exists which is specifically designed for this type of application (i.e., low viscosity, insensitivity to mix ratio, rapid cure under ambient conditions and unlimited shelf-life). However, one practical advantage of using such a resin system for this demonstration phase of the study is that the temperature activation provides excellent control of cure initiation, eliminating time constraints on the testing/manufacturing process.
After healing, test specimens were mounted on a Roell Amsler HCT25 electro-mechanical test machine with roller spacing determined by the specimen dimensions and ASTM D6272-02. An Instron 8800 controller/data-logger was used to control the test machine and record data. Specimens were loaded to catastrophic failure at which point the cross-head was stopped and the load removed. Specimens were monitored to ensure a consistent failure mode and optical microscopy used to record detailed observations. Results were obtained from ten undamaged, five damaged and five healed specimens.
3. Damage analysis
Microscopic analysis of samples after quasi-static indentation highlighted the mode of damage development and the interaction with the HGF. The cross-sectional damage distribution was typical of an impact damage event [24,25]. A localised ‘crushing’ zone was evident on the upper surface followed by shear cracks and delaminations of increasing length through the thickness of the laminate culminating in the largest delamination at the back face between the final two plies of the stack.
The process of self-healing relies upon the occurrence of two phenomena:
- HGF fibre fracture initiated by a damage event.
- Connectivity between HGF and damage network within material.
The region of HGF immediately beneath the impactor was subject to ‘crushing’ forces, the consequences of which were difficult to discern clearly using optical microscopy. However, the majority of HGF fracture within the laminate is evident from Fig. 4a where shear cracks and delaminations intercepted regions of HGF. Furthermore, Fig. 4b shows similar damage infiltrated with healing resin mixed with Ardrox 983 UV fluorescent dye. It can be seen from this image that the larger delaminations contain only a limited volume of healing resin. This could be due to the inability of capillary action alone to draw significant volumes of resin into the relatively thick cracks, combined with limited length of HGF available for evacuation of resin. This was due to the small laminates used for this investigation which only provided around 20 mm of HGF either side of the impact damage site. It can be seen from Fig. 4b that delaminations thicker than 30 𝞵m may be too thick to allow healing resin infiltration. However, the absence of healing resin may also be a result of the specimen polishing process removing cured resin from the interface.
Delaminations were observed to propagate along interfaces between plies of dissimilar fibre direction. It can be seen that this initiates HGF fracture via two mechanisms:
- In Fig. 4c, it can be seen that clusters of HGF can cause a delamination to deviate from its path, presumably due to fibre clusters and the resulting resin rich regions causing a weakness in the laminate. However, the propagating crack does not pass directly through a HGF/matrix interface, but instead deviates around them causing fibre rupture and release of healing resin.
- HGF are ruptured as a delamination propagates along a ply interface (Fig. 4d). This suggests that the HGF’s susceptibility to fracture is similar to the matrix as no obvious crack deviation is evident. However, there is no apparent evidence of reinforcing fibre failure under these loading conditions.
Fig. 4. (a) Damage distribution within laminate; (b) damage infiltration with healing resin + fluorescent dye; (c) delaminations deviating from interface; and (d) propagating along interface causing HGF fracture.
Intra-ply shear cracks link multiple delaminations. The presence of these shear cracks is essential for self-healing as they provide connectivity between different damage sites in the laminate and facilitate healing at multiple interfaces. Shear cracks can also be seen to have initiated HGF fracture in preference to reinforcing fibre fracture (Fig. 5). However, the thickness of shear cracks (10 𝞵m) is generally smaller than delaminations (30 𝞵m) and so shear cracks would be expected to generate a larger capillary action to transport healing resin as the capillary force generated is inversely proportional to the radius of the opening between two surfaces.
4. Mechanical properties
The results of the four point bend flexural testing are shown in Table 1. Comparisons are made between the performance of undamaged, damaged and healed specimens for the two HGF pitch spacings alongside a baseline CFRP laminate with no HGF.
Fig. 6 shows typical load–displacement curves arising from the four point bend flexural testing. The relative placement of the undamaged, damaged and healed traces for 70 𝞵m HGF spacing gives a clear representation of the effects of damage and healing events on the flexural performance for CFRP with embedded HGF.
Analysis of the flexural test data (Table 1) shows that the fibres spaced at 70 𝞵m caused the largest reduction in undamaged strength (8%). This effect can be attributed to the significant disruption to the fibre architecture observed in Fig. 1. However, after a quasi-static impact at 2000 N peak load, this configuration exhibited a large amount of damage tolerance when compared to the 200 𝞵m fibre spacing and plain baseline laminates. This apparent damage tolerance can be attributed to energy absorbed by the crushing of HGF [26]. It appears that the initial reduction in strength due to the presence of a significant volume of HGF is offset by the increased damage tolerance. The large volume fraction of HGF in this configuration also provides a large reservoir of healing agent as confirmed by a 97% recovery of undamaged strength (equivalent to 89% of the undamaged strength of the baseline laminate).
Fig. 5. Shear cracks intercepting HGF within a self-healing CFRP laminate.
Table 1 Results of flexural four point bend testing of CFRP with embedded HGF
Fig. 6. Comparative load–displacement curves under 4 point bend for undamaged, damaged and healed for 70 𝞵m spaced HGF.
Fig. 6 shows that a damaged specimen containing HGF under load experienced a number of intermittent decreases in load as failure was approached. These are probably attributable to the propagation of matrix cracks and delaminations up to a critical level at which they effect the load bearing performance of the laminate. Conversely, the healed specimen appears to suppress these damage sites, inhibiting crack propagation and thereby delaying the failure of the laminate to higher load levels. Fig. 6 also highlights the effectiveness of self-healing as the load–displacement curve for the healed laminate lies close to the undamaged curve.
The 200 𝞵m HGF spacing specimens exhibit little reduction in undamaged strength (2%) attributable to the reduced disruption to the host laminate (Fig. 2). These specimens behaved similarly to the baseline laminate when damaged, presumably due to the limited amount of HGF available for crushing. However, healed samples achieved 82% of their undamaged strength (equivalent to 80% of the undamaged strength of the baseline laminate), despite the significantly lower volume of available healing resin.
Both HGF spacings investigated (70 𝞵m and 200 𝞵m) showed similar trends, experiencing an initial reduction in flexural strength in the undamaged state compared to an unmodified baseline. This can be attributed to three effects:
- Distortion of the reinforcing fibre architecture.
- Generation of resin rich regions (crack nucleation/propagation sites).
- Displacement of reinforcing fibres with non-structural HGF (reduction in carbon fibre volume fraction).
Secondly, the damage event caused a reduction in strength due to the generation of shear cracks and delaminations which then propagate under load leading to premature failure. Finally, a strength recovery was experienced after fracture of HGF where healing agent had infiltrated damage sites and mitigated some of the effects of damage.
5. Conclusions
The incorporation of HGF within a CFRP laminate has been shown to produce minimal degradation in flexural strength and ply disruption. At fibre spacings of over three HGF diameters (200 𝞵m), very good embedment of the HGF can be achieved, even on the most demanding 0°/45° interface.
The embedding of HGF by winding directly onto CFRP pre-preg is aided by the tension in the fibre produced during the fibre manufacturing process and further improved during cure as the initial viscosity drop before gelation facilitates resin and fibre migration around the larger HGF.
The presence of uniformly distributed HGF at a ply interface does not appear to cause obvious crack path deviation, suggesting that they do not create sites of weakness under these loading conditions. However, small clusters of HGF combined with resin rich regions can cause significant disruption resulting in crack path deviation.
Laminates with higher volume fractions of HGF exhibited a larger degree of damage tolerance under quasi-static impact events which is attributed to energy absorption from HGF crushing.
The concept of self-healing CFRP has been developed and demonstrated for four point bend flexural testing. Laminates with resin filled HGF at two different fibre spacings (70 𝞵m and 200 𝞵m) were shown to provide recovery in flexural strength due to self-healing. It is noted, however, that in this demonstration the healing process used an un-optimised resin system employed in a most favourable manner, i.e. premixed and allowed to infiltrate damage at raised temperature, to eliminate the problems of stoichiometric mix sensitivity and resin viscosity on damage infiltration.
Intra-ply shear cracks and delaminations during an impact event cause HGF fracture and the release of healing resin into the interconnected damage sites. There is evidence to suggest that if crack faces separate by more than 30 𝞵m the capillary forces may not be sufficient, or there may be insufficient resin volume released, to fully infiltrate the damage site.
The optimum flexural strength recovery was for self- healing CFRP with HGF at 70 𝞵m spacing. This configuration achieved 97% of the undamaged state and 89% of the baseline laminate performance. However, this fibre spacing resulted in significant disruption to the host laminate fibre architecture and consequently a large reduction in initial mechanical properties.
A reduction in scatter is evident for the flexural strength of the healed samples observed in Table 1. Although there is no evidence that the healed samples have experienced a different failure mode, the action of healing has produced a more stable and predictable failure.
Future work will use an instrumented drop-weight impact tester to create a more realistic damage condition within the material and apply compression after impact (CAI) testing in order to achieve a more rigorous assessment of the effects of HGF on the host laminate and the subsequent healing performance.
The 10 largest coal producers and exporters in Indonesia:
- Bumi Resouces
- Adaro Energy
- Indo Tambangraya Megah
- Bukit Asam
- Baramulti Sukses Sarana
- Harum Energy
- Mitrabara Adiperdana
- Samindo Resources
- United Tractors
- Berau Coal
