- Email: [email protected]
epoxy composites
Composites: Part B 59 (2014) 204–220
Contents lists available at ScienceDirect
Composites: Part B journal homepage: www.elsevier.com/locate/compositesb
Drop-weight impact behaviour of woven hybrid basalt–carbon/epoxy composites F. Sarasini a,⇑, J. Tirillò a, L. Ferrante a, M. Valente a, T. Valente a, L. Lampani b, P. Gaudenzi b, S. Cioffi c, S. Iannace c, L. Sorrentino c a b c
Department of Chemical Engineering Materials Environment, Sapienza-Università di Roma, Via Eudossiana 18, 00184 Rome, Italy Department of Mechanical and Aerospace Engineering, Sapienza-Università di Roma, Via Eudossiana 18, 00184 Rome, Italy Institute for Composite and Biomedical Materials, National Research Council, Piazzale Enrico Fermi 1, Località Granatello, Portici (NA) 80055, Italy
a r t i c l e
i n f o
Article history: Received 10 October 2013 Received in revised form 23 November 2013 Accepted 7 December 2013 Available online 15 December 2013 Keywords: A. Polymer-matrix composites (PMCs) Basalt fibres A. Hybrid B. Impact behaviour D. Non-destructive testing
a b s t r a c t This study addresses the effects of basalt fibre hybridization on quasi-static mechanical properties and low velocity impact behaviour of carbon/epoxy laminates. Interply hybrid specimens with two different stacking sequences (sandwich-like and intercalated) are tested at three different energies, namely 5, 12.5 and 25 J. Residual post-impact properties of the different configurations of carbon/basalt hybrid laminates are characterized by quasi static four point bending tests. Post-impact flexural tests and interlaminar shear tests are used for the mechanical characterization along with two non-destructive methods, namely acoustic emission and ultrasonic phased array, in order to get further information on both the extent of damage and failure mechanisms. Results indicate that hybrid laminates with intercalated configuration (alternating sequence of basalt and carbon fabrics) have better impact energy absorption capability and enhanced damage tolerance with respect to the all-carbon laminates, while hybrid laminates with sandwich-like configuration (seven carbon fabric layers at the centre of the laminate as core and three basalt fabric layers for each side of the composite as skins) present the most favourable flexural behaviour. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Most of the composite structures and parts used in advanced applications are fibre-reinforced composites. During their service life, composites are subjected to various loading conditions and low velocity impact is one of the most critical in particular for aerospace composite structures [1–5]. Due to their high specific stiffness and strength, carbon fibre reinforced polymer composite is the material of choice in aerospace industry. However, the toughness of carbon fibre is quite low and the resulting damage resistance is poor. In this regard, several approaches have been successfully exploited to enhance the impact damage resistance of composite laminates. One approach consists in enhancing the properties of the matrix material, as proposed by Reis et al. [6], who investigated the impact behaviour as well as the damage tolerance of Kevlar/filled epoxy matrix with two different fillers, namely cork powder and nanoclays. They found that adding fillers increased the maximum impact load, which was very dependent on the filler type especially for high impact energy, and the addition of clays increased the damaged area of around 29% while ⇑ Corresponding author. Tel.: +39 0644585408; fax: +39 066876343. E-mail address: [email protected] (F. Sarasini). 1359-8368/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.compositesb.2013.12.006
improving the residual strength of the laminates. Another approach is known as hybridization, usually with high strain to failure fibres to improve the damage resistance of composites to both low velocity and instrumented Charpy impact [7,8]. In this regard, glass fibres are the best option from the viewpoint of cost, availability and ease of processing, and hybrid carbon/glass fibre composites have consistently demonstrated better damage tolerance under impact than their carbon fibre counterparts [9–12]. In the field of fibrous reinforcement, basalt fibres have gained an increasing attention in recent years as possible replacement of the conventional glass fibres [13] due to their advantages in terms of environmental cost and chemical–physical properties. Mineral fibres obtained from basalt rocks are not new, but their suitability as reinforcement in polymer composites is a relatively new issue [14]. From the mechanical point of view, continuous basalt fibres are competitive with glass ones. The elastic modulus of basalt fibres strongly depends on the chemical composition but it is usually comparable or slightly higher than that of glass fibres, while both tensile strength and elongation at break are higher [13]. These properties suggest the potential role of basalt fibres in replacing glass ones as impact resistance improver for composite laminates with also a view to enhancing the environmental sustainability of such composites. The mechanical properties of basalt fibre
205
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
reinforced composite laminates have been thoroughly investigated for both thermoset [15–21] and thermoplastic matrices [22–25], but only limited attention has been devoted to the low-velocity impact behaviour of these class of composites [26–33]. Lopresto et al. [26] provided a thorough investigation of the mechanical properties of basalt/epoxy composites, studying also the impact resistance. The low velocity impact tests were only performed at about 100 J in order to perforate the composite specimens and reported, in comparison to glass fibre reinforced composites, higher energy absorption capability for basalt. A more detailed investigation of the effect of hybridization of basalt fibres on low velocity impact response of glass fabric reinforced epoxy composites was performed in [28,29,31], where was experimentally confirmed that basalt composites exhibit better energy absorption capability when compared to glass ones and that glass laminates’ poor damage tolerance can be enhanced by the hybridization with basalt layers. Limited research has been also performed on hybrid composites made of basalt and ductile fibres, such as nylon [27,32] and aramid [30,33]. From the review of the available literature, there is a reduced data availability on hybrid carbon/basalt composites [34–36] and, in particular, on their post-impact behaviour, even though a potential positive effect of basalt fibres in enhancing toughness (evaluated by an open hole compression test) of carbon fibre composites was pointed out [36]. The lack of data on carbon/ basalt hybrid composites is the rationale behind this experimental work, whose aim is to provide a comprehensive, though preliminary, investigation of the low-velocity impact behaviour and residual properties of basalt/carbon hybrid laminates by taking into account the effect of different stacking sequences. To attain this aim, woven basalt-carbon/epoxy hybrid composites were fabricated in interply hybrid structures and subjected to low-velocity impact using a drop weight apparatus. Specimens with different lay-up configurations were tested using three impact energies and the influence of impact energy on the flexural residual strength of composites was assessed by quasi-static four point bending tests. To determine the extent of damage, the specimens were subjected to ultrasonic phased array testing. The post-impact flexural tests were monitored by acoustic emission (AE) and the results of these analyses, along with those of the impact tests (maximum load and energy absorbed) were used to understand the role played by basalt fibre hybridization on the mechanical behaviour under impact of carbon fibre reinforced epoxy composites. In addition, interlaminar shear strength for all the hybrids is also addressed. The combined use of destructive and non-destructive characterizations provides a confirmation of the possibility of replacing glass fibres with basalt ones to improve the impact resistance of carbon-based laminates along with an indication of the most suited stacking sequence to be used to this aim.
2. Materials and methods The basalt (BAS 220.1270.P) and carbon fabrics (CC160) are plain weave fabrics supplied by Basaltex-Flocart NV (Belgium) and Saati Composites, respectively. The fibre areal densities are 220 g/m2 and 160 g/m2 for basalt and carbon fabrics, respectively. A bi-component epoxy resin (EC157+W152 MR), supplied by Elantas Camattini (Italy) is selected as polymeric matrix. The laminates have been manufactured by a laboratory Resin Transfer Moulding (RTM) system and have been cured for 12 h at room temperature and 4 h at 70 °C. All configurations are produced using thirteen fabric layers with a similar volume fraction and thickness. Two hybrid configurations have been manufactured: in the first one (BC-HS) fabrics are stacked as a sandwich-like sequence with seven carbon fibre layers (core) and three basalt fibre layers (skins) for each side of the laminate, while in the second configuration (BC-HI) seven
Table 1 Thickness and fibre volume fraction of the tested laminates. Composite type
Thickness (mm)
Fibre volume fraction
B C BC-HS BC-HI
3.40 ± 0.05 3.50 ± 0.04 3.50 ± 0.04 3.45 ± 0.05
0.315 ± 0.01 0.325 ± 0.01 0.319 ± 0.01 0.321 ± 0.01
layers of basalt fabrics and six of carbon fabrics are alternatively stacked, keeping basalt fabrics as outer layers. Not hybridised basalt (B) and carbon (C) reinforced epoxy composites have been also manufactured as reference configurations. Details of all the composite types are reported in Table 1. Specimens, with dimensions of 180 mm 60 mm 3.4 mm (length width thickness) for the structural characterisations have been cut from 300 mmwide square composite laminates. This unusual width has been chosen in order to be sure of accommodating the whole impact damaged area, thus minimizing the edge effects. Four-point bending tests have been performed on five specimens for each configuration in accordance with ASTM D 6272. A span-to-depth ratio of 23:1 and a cross-head speed of 2.5 mm/min have been used. Strain gauges have been employed to measure the effective strain and to evaluate the flexural modulus. Specimens have been tested in bending either after their production (non impacted samples) or after the low-velocity impact tests to measure their residual flexural strength. Impact conditions are selected in order to assure that the impact does not result in the full penetration of the laminate and that the damage extension into the sample does not exceed the specimen width, even under the highest energy applied. Impact tests have been performed by using a falling dart impact testing machine, model Fractovis Plus from CEAST (Pianezza – TO, Italy) over beam-like specimens with the dimensions previously described. Specimens have been tested at three impact energies (5 J, 12.5 J and 25 J) by keeping constant the indenter mass (6.929 kg) with an hemispherical impact head (diameter equal to 12.7 mm). The sample holder is circular, with the external diameter of 60 mm and the inner of 40 mm. The impacted specimens have been scanned using non-destructive ultrasonic inspection equipment OmniScan MX with standard phased array probe 3.5 MHz, linear array, 64 elements. The interlaminar shear strength (ILSS) has been evaluated in accordance with ASTM D 2344. Ten specimens have been tested for each laminate, having the following dimensions: 20 mm 6.8 mm 3.4 mm (L W t). A span-to-depth ratio of 4:1 and a cross-head speed of 1 mm/min have been used. All the mechanical tests have been performed on a Z010 from Zwick/Roell (Ulm, Germany) universal testing machine equipped with a 10 kN load cell. Post-impact flexural tests have been monitored by acoustic emission until final fracture occurred using an AMSY-5 AE system by Vallen Systeme GmbH (Icking, Germany). The AE acquisition settings used throughout this experimental work are as follows: threshold = 35 dB, Rearm Time (RT) = 0.4 ms, Duration Discrimination Time (DDT) = 0.2 ms and total gain = 34 dB. This threshold level has been defined from a 30 min track record of the background noise, with the AE setup configuration actually used, and has been set 6 dB above the maximum level of the recorded spurious signal from the electronic system. Two broad-band (100–1500 kHz, Fujicera 1045S) PZT AE sensors have been used. The sensors have been placed on the surface of the specimens at both ends to allow linear localization, with silicone grease as coupling agent. 3. Results and discussion The present investigation follows the established methodology for damage tolerance assessment which consists of four major
206
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
sequential steps, namely: impact testing, damage characterisation, determination of static residual strength and damage tolerance evaluation. The flexural and interlaminar shear mechanical properties are also evaluated, being them considered the most relevant to impact damage. 3.1. Low velocity impact response and damage evaluation Figs. 1–3 show the different damage patterns and structural responses of laminates tested at increasing impact energies. Dropweight impact test at 25 J on C composites shows penetration of the dart through its thickness with splitting on the back face, highlighting the brittle fracture of carbon laminates. In the same impact condition, B composites exhibit cross-shaped cracks and debonding on the front face with associated matrix cracking and back surface splitting but without evidence of penetration. BC-HI and BC-HS laminates show an intermediate behaviour between B and
C composites, although different patterns are detected as a consequence of the different lay-ups. In fact, the damaged area for BC-HI laminates is larger than that of BC-HS one, suggesting that the former can absorb more energy. Both hybrid laminates show bulge on the back surface as a consequence of matrix cracking and pull-out of outer fibre layers but the damage pattern is quite different. The BC-HS laminates present an impact behaviour more similar to B ones because the three outer layers, being made of basalt, prevent extensive fibre breakage unlike C composites. Conversely, the impacted BC-HI laminates exhibit back surface damages approaching the pattern of C laminates but without penetration due to the presence of alternating basalt layers which provide enhanced compliance. To assess composite impact damage, it is common to refer to the impact energy (Ei) and absorbed energy (Ea). Impact energy is the kinetic energy of the impactor right before contact with samples takes place, whereas absorbed energy is the energy dissipated
Fig. 1. Photos of damage progression on front and rear faces of B, C, BC-HI and BC-HS composite panels impacted at 5 J.
Fig. 2. Photos of damage progression on front and rear faces of B, C, BC-HI and BC-HS composite panels impacted at 12.5 J.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
207
Fig. 3. Photos of damage progression on front and rear faces of B, C, BC-HI and BC-HS composite panels impacted at 25 J.
by the system through the several mechanisms occurring after the impactor’s contact, like elastic deformation, friction, plastic deformation and, most importantly, those peculiar to the material (matrix cracking, debonding, pull-out, fibre breakage) [37]. In most of tested samples, penetration does not occur and the impactor rebounds with an energy that is the difference between Ei and Ea. The absorbed energy can be calculated from force– displacement curves (Figs. 4–6) as the area enclosed within the curve. Indeed if the displacement returns toward the axis origin during unloading, it means that some elastic energy is recovered by the laminate. Carbon laminates are the only laminates to be perforated at 25 J without showing any energy recovery (Fig. 6: displacement increases while force decreases), whereas all composites containing basalt layers exhibit a residual elastic response. Basalt and hybrid laminates absorb more energy through higher overall deformation, being more compliant, and through delaminations. Key impact parameters like peak force, impact energy (Ei), maximum displacement and absorbed energy (Ea) obtained from the transient response of each laminate are summarized in Table 2.
Fig. 4. Typical force vs. displacement response for 5 J-impacted carbon (C), basalt (B), carbon/basalt sandwich hybrid (BC-HS) and carbon/basalt intercalated hybrid (BC-HI) composites.
Fig. 5. Typical force vs. displacement response for 12.5 J-impacted carbon (C), basalt (B), carbon/basalt sandwich hybrid (BC-HS) and carbon/basalt intercalated hybrid (BC-HI) composites.
Fig. 6. Typical force vs. displacement response for 25 J-impacted carbon (C), basalt (B), carbon/basalt sandwich hybrid (BC-HS) and carbon/basalt intercalated hybrid (BC-HI) composites.
208
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Table 2 Parameters obtained from impact tests on basalt, carbon and hybrid composites.
*
Specimen
Peak force (N)
Maximum displacement (mm)
Impact energy (J)
Absorbed energy (J)
Damage degree
Damaged area (mm2)*
Energy: 5 J B C BC-HI BC-HS
3511.98 ± 64.27 3745.85 ± 212.05 3865.67 ± 83.26 3564.78 ± 173.98
2.33 ± 0.03 1.83 ± 0.12 2.20 ± 0.03 2.08 ± 0.03
4.85 ± 0.01 4.85 ± 0.07 4.97 ± 0.01 4.96 ± 0.01
1.84 ± 0.09 2.30 ± 0.09 2.14 ± 0.07 2.27 ± 0.10
0.38 ± 0.01 0.46 ± 0.01 0.43 ± 0.01 0.45 ± 0.02
310 298 690 594
Energy: 12.5 J B C BC-HI BC-HS
5654.42 ± 16.62 4209.68 ± 223.41 4792.75 ± 154.59 5090.41 ± 58.01
3.92 ± 0.09 3.71 ± 0.23 4.01 ± 0.33 3.76 ± 0.06
12.38 ± 0.01 12.38 ± 0.07 12.47 ± 0.07 12.46 ± 0.07
6.71 ± 0.19 9.24 ± 0.25 7.99 ± 0.33 6.00 ± 0.23
0.54 ± 0.01 0.74 ± 0.02 0.64 ± 0.03 0.48 ± 0.02
1048 736 966 603
Energy: 25 J B C BC-HI BC-HS
7238.07 ± 117.11 4261.90 ± 276.72 4801.88 ± 37.35 6726.30 ± 202.09
6.06 ± 0.17 – 7.25 ± 0.16 6.03 ± 0.11
24.96 ± 0.01 27.95 ± 0.12 24.95 ± 0.10
20.66 ± 0.50 – 21.84 ± 0.40 18.38 ± 1.04
0.82 ± 0.02 – 0.86 ± 0.01 0.72 ± 0.04
889 – 1027 999
Evaluated from ultrasonic C-scans.
It can be seen that the peak load increases with the increase in the incident kinetic energy, indicating greater load bearing ability of the laminates at higher energy levels. In terms of peak force, basalt laminates perform better than carbon ones at high energy levels as was the case for glass laminates with respect to carbon ones [11], while hybrid laminates show an increased peak force compared to carbon ones. As confirmed by experimental results, the laminate configuration heavily affects the transient response, with sandwich-like hybrid composites outperforming intercalated hybrids, as reported in other studies [31,33]. Carbon/epoxy laminates exhibit a lower displacement, whilst basalt and hybrid laminates, being more compliant, absorb energy through the global deformation. In order to assess the damage accumulated by the materials, a parameter called damage degree, defined as the ratio Ea/Ei, was used among the several ones available [38–40]. From Table 2, it is evident that damage degree of carbon-epoxy laminate increases dramatically moving from 5 J to 12.5 J, and reaches unity at 25 J. This is a further confirmation of the inherent brittleness of C laminates, which cannot absorb high impact energy because of their abrupt and catastrophic damage mechanism. Damage degree of basalt containing laminates increases with the impact energy, reaching a maximum value of 0.86 for the BC-HI composite, thus suggesting the positive role played by basalt fibre hybridization, which improves the impact energy absorption. The ultrasonic phased array testing method was used to evaluate damaged area that is connected with the energy absorbed by the samples. Fig. 7 shows C-scan images of different impacted samples from which it is estimated the extent of damage reported in Table 2. All C-scans are relative to an area of 59 mm 49 mm, whilst the damaged area (in mm2) for each type of specimen has to be considered through-the-thickness being the summation of damaged areas of each layer. It has been difficult to evaluate the damage caused by the impactor because of the additional damage introduced by the circular holder beneath the samples during impact tests, which was particularly important for the basalt laminates. This can explain why the data for basalt laminates in Table 2 do not exhibit a clear trend with increasing impact energy. On the whole, all the examined samples show a progression in damaged area with increasing impact energy, though the behaviour is dependent on the fabrics stacking sequence. The carbon fibre reinforced laminates exhibit a small damaged area, confirming that C composite absorbs impact load in a brittle way. On the other hand, BC-HI and BC-HS show a wider damaged area suggesting that delamination between different kinds of layers occurs with the consequence of improving the energy dissipation. To summarize,
among all the investigated composites, BC-HI composites exhibit the widest damaged areas due to the increased number of events occurring at the multiple interfaces between basalt and carbon fabric layers. This mechanism together with larger global deformation offered by the addition of basalt fibres are responsible for the higher energy absorption of hybrid laminates with respect to carbon laminates. 3.2. Interlaminar shear strength Interlaminar shear strength (ILSS) is usually a limiting design property for composites and depends on many parameters such as (a) adhesion between matrix and fibre, (b) constitutive materials, (c) fibre volume fraction and (d) stacking sequence. When the transverse shear load experienced by a laminated composite exceeds the interlaminar shear strength, a delamination failure will occur between the reinforcing layers. Fig. 8 shows that B laminates exhibit the worst interlaminar shear strength whereas C laminates provide the best one. This might result from the better interfacial bonding between carbon and epoxy matrix with respect to basalt and epoxy resin. Hybridization with basalt fibres plays a substantial role only in the BC-HS case, in accordance with the lower number of basalt/epoxy interfaces. 3.3. Residual mechanical properties and damage tolerance evaluation The basalt fibre hybridization affects also the flexural strength and modulus of laminates, both before and after impact. Carbon/ epoxy composites are stiffer than basalt/epoxy ones as the carbon fibres have higher elastic modulus than basalt ones (Table 3). Flexural tests on specimens impacted at 12.5 J (Fig. 9) highlight that hybrid composites have flexural stiffness intermediate between carbon/epoxy and basalt/epoxy ones. The different lay-up of the hybrid composites plays an important role on the laminate stiffness also considering that carbon layers are stiffer than basalt ones. Although BC-HI laminates have less carbon layers than BC-HS ones, the former possess higher flexural modulus among all basalt containing composites. This has been related to the fact that in the BC-HS sandwich structure, the outer layers, responsible for bearing the bending load, are made of basalt (less stiff than carbon). Hybridization also provides improved toughness and flexural strength. C composites show a higher flexural strength than B ones after 12.5 J impact (Fig. 9), but hybridized configurations allow to strongly improve both flexural strength and toughness with respect to carbon laminates, and the sandwich configuration seems
209
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
C–5J
C – 12.5 J
BC-HI 5 J
BC-HI 12.5 J
BC-HI 25 J
BC-HS 5 J
BC-HS 12.5 J
BC-HS 25 J
B-5J
B – 12.5 J
B - 25 J
Fig. 7. Ultrasonic C-scan images of four types of composites impacted at different impact energies.
Table 3 Summary of flexural properties for basalt, carbon and hybrid composites.
Fig. 8. Average ILSS for basalt, carbon and hybrid composites (the error bars represent the standard deviation).
Specimen
Flexural strength (MPa)
Flexural modulus (GPa)
Non-impacted B C BC-HI BC-HS
229.34 ± 3.24 593.48 ± 9.57 350.84 ± 0.30 418.68 ± 8.10
14.35 ± 0.75 38.70 ± 0.10 25.84 ± 2.78 24.20 ± 1.42
Impact Energy: 5 J B C BC-HI BC-HS
218.64 ± 3.76 349.01 ± 35.80 345.21 ± 22.14 369.58 ± 5.55
14.16 ± 0.34 33.12 ± 0.94 25.23 ± 0.07 21.99 ± 1.37
Impact Energy: 12.5 J B 207.30 ± 3.09 C 238.59 ± 0.50 BC-HI 304.45 ± 8.07 BC-HS 316.77 ± 0.79
13.55 ± 0.49 27.29 ± 0.27 23.64 ± 0.10 20.59 ± 0.06
Impact Energy: 25 J B C BC-HI BC-HS
11.48 ± 0.64 – 20.90 ± 0.22 18.58 ± 0.03
142.55 ± 3.92 – 259.72 ± 2.80 250.79 ± 13.45
210
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 9. Typical stress vs. strain curves for flexural tests on composites previously impacted at 12.5 J.
to represent the best compromise between these parameters, often resulting in competing relationship. This behaviour has been related to the higher number of carbon layers in the BC-HS configuration, characterized by a higher ultimate strength than basalt ones. Thus the flexural strength is more influenced by the fibre mechanical properties rather than by composite stacking sequence. Although the non-impacted laminates can be ranked from the higher to the lower flexural strength in the following order: C > BC-HS > BC-HI > B (Table 2), the trend is quite different for impacted samples, in particular as the impact energy increases. In fact, it should be noted that the increase of impact energy leads to increased damaged areas, which in turn reduce both flexural strength and flexural stiffness. In order to obtain a better indication of the damage tolerance (residual properties after impact) the normalized flexural strength (or stiffness) of each specimen is evaluated as the ratio of the mean strength (or stiffness) of the impacted specimen to the mean value of the flexural strength (or stiffness) of the undamaged specimen. Figs. 10 and 11 show the normalized flexural strength and flexural stiffness with respect to the increasing impact energy, respectively. After being impacted at 12.5 J, the C laminate loses almost 60% of its early flexural strength and almost 30% of its early flexural stiffness whereas the impact at 25 J leads to penetration with
Fig. 10. Normalized residual flexural strength as a function of increasing impact energy.
Fig. 11. Normalized residual flexural stiffness as a function of increasing impact energy.
catastrophic failure, confirming the worst damage resistance and tolerance capabilities among all considered laminate configurations. This behaviour is due to the characteristic energy absorption mechanism of C composites, which dissipate energy mainly by fast transverse crack propagation and fibre fracture. B composite shows a better behaviour with a more gradual degradation pattern because the impact energy was mostly absorbed by the interface failures occurring among laminate layers and at the fibre/matrix interface. It is evident the positive role played by basalt fabric on the post-impact characteristics of hybrid laminates. More specifically, BC-HI presents the most favourable degradation pattern as it shows the best damage tolerance (Figs. 10 and 11). The BC-HS laminate, despite of presenting higher quasi-static flexural properties when compared to BC-HI (Table 3), exhibits a steeper damage resistance drop (Fig. 10). This has been related to the energy dissipation mechanism of BC-HI through multiple small delaminations between dissimilar layers (Fig. 12a) whereas the BC-HS dissipates energy mainly through transverse crack propagation in carbon ‘‘core’’ of sandwich laminate and secondly through main delaminations at the basalt-skin/carbon-core interface (Fig. 12b), as confirmed by AE monitoring. 3.4. Acoustic emission monitoring The time evolution and nature of different failure modes of composites is studied through the analysis of acoustic emission signals recorded in real time during flexural tests on non-impacted and impacted composite laminates. Basalt (B) and carbon (C) laminates exhibit a different response to impact loading in terms of damage degree and damage tolerance. This behaviour suggests that the mechanisms responsible for the energy absorption are dissimilar. For non-impacted C laminates, very few AE signals are recorded until final fracture occurs, thus confirming the catastrophic and localized failure (Fig. 13). Most of AE signals have amplitudes in the range 35–50 dB, which are usually ascribed to matrix cracking, while few signals can be ascribed to fibre breakages (80–100 dB) and interfacial failures (55–65 dB) (Fig. 14) [41–43]. This behaviour is consistent with the dominant failure mode of these composites mainly due to multiple matrix cracks localized in the tensile side. The failure of carbon composites results to be therefore characterized by the nucleation of matrix cracks, which then trigger debonding phenomena and further transverse cracks leading to fibre failures. From the localization plots of AE signals (Fig. 15) it can be inferred that the extent of the damaged zone responsible for the intense acoustic emission
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 12. Close-up view of (a) BC-HI and (b) BC-HS specimens impacted at 12.5 J after flexural test.
Fig. 13. Typical AE amplitude vs. time response during flexural test on undamaged carbon laminates.
Fig. 14. Typical amplitude distribution during flexural test on undamaged carbon laminates.
211
212
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 15. Acoustic emission localization plot for undamaged carbon laminates. The inset represents a close-up view of typical failure modes for undamaged carbon laminates.
Fig. 16. Typical AE amplitude vs. time response during flexural test on carbon laminates impacted at 12.5 J.
activity is wide, as confirmed by the micrograph in the inset of Fig. 15. The impact damage causes an early onset of acoustic emission activity (signals of low amplitude and short duration) and a higher number of signals. This trend increases with increasing impact energy (Fig. 16). Impact loading causes, in brittle carbon laminates, an unstable damage in terms of matrix cracks and small debonding phenomena, which become active at lower stresses during the subsequent flexural loading. Most AE signals are still localized in the lower amplitude range (<55 dB) thus confirming the same failure modes observed in non-impacted laminates even though the presence of impact damage seems to quicken the process thanks to the splitting in the rear face. This thesis is supported by the localization of AE signals (Fig. 17), which appears to be similar to the one observed for non-impacted laminates (Fig. 15). Basalt laminates show a different behaviour, which is characterized by significant AE signals since the beginning of loading also for the undamaged material (Fig. 18). These signals can be attributed to matrix cracks and interface failures. Compared with carbon laminates, an increase in the number of signals in the range 65–75 dB (pull-out) is observed (Fig. 19). This suggests that interface failures
are the dominant failure mechanism, which is mainly localized in the compression side. This can explain the lower strength of basalt/epoxy laminates compared to carbon ones. Failure in basalt laminates takes place in compressed half section, whilst in carbon laminates the whole section is involved in the fracture process up to the tensile face. The fracture surface in basalt laminates exhibits the presence of a typical fracture pattern due to compression failure, consisting of fibre microbuckling and kink bands (inset in Fig. 19). The damaged zone so nucleated can cause longitudinal splitting leading to premature failure of the composite laminates thus preventing higher stresses in the outermost tensile fibres to be reached. Fibre microbuckling is known to occur preferentially in systems characterized by medium interface strength [44]. As for carbon laminates, the presence of impact tends to emphasize the importance of interface failures (pull-out and delaminations, Fig. 20) and to localize the damage (Fig. 21). These interface phenomena, together with higher ductility, appears to be the relevant energy dissipation mechanisms for basalt laminates [33]. Hybrid laminates with sandwich configuration (BC-HS) show, compared with carbon laminates, a higher number of AE signals
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
213
Fig. 17. Acoustic emission localization plot for carbon laminates impacted at 12.5 J.
Fig. 18. Typical AE amplitude vs. time response during flexural test on undamaged basalt laminates.
Fig. 19. Typical amplitude distribution during flexural test on undamaged basalt laminates. The inset represents a close-up view of typical failure modes for undamaged basalt laminates.
related to debonding phenomena located mainly in the compression side and due to the basalt skins. These signals are active at low stresses, as reported in Fig. 22. Moreover, a higher number of signals characterized by medium to high amplitudes (60–75 dB)
and long durations (>1 ms) related to delaminations is recorded. Basalt skins, characterized by higher deformation, hold together the carbon core that, once failed, causes the failure of the whole laminate (Fig. 12b). With increasing impact energy, it is to be noted
214
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 20. Typical amplitude distribution during flexural test on basalt laminates impacted at 25 J.
Fig. 21. Acoustic emission localization plot for (a) undamaged and (b) impacted at 25 J basalt laminates.
an increase of signals related to interface failures (Fig. 23). The presence of such delaminations, mainly localized in the compression side, quickly prevents basalt skins from hindering crack propagation in the carbon core, causing its failure. The growing damage localization with increasing impact energy supports what experimentally observed as regards the failure modes of these laminates (Fig. 24). Undamaged hybrid laminates with intercalated configuration (BC-HI) show a significant number of signals with amplitudes >60 dB related to delaminations (Fig. 25). With increasing impact energy, a considerable rise in the number of signals at the very beginning of loading along with a growing importance of multiple delaminations occur. This is highlighted in Fig. 26, which shows the amplitude distribution of AE signals for BC-HI laminates impacted at 25 J recorded during the first 110 s of
flexural loading. Such delaminations are diffused in the material due to higher number of basalt/carbon interfaces. Damage partition among internal interfaces enables BC-HI laminates, although less resistant than BC-HS hybrids, to show a lower degradation of their strength at higher impact energies, when extensive damage is present. The occurrence of splitting on the rear face, although less evident than in carbon laminates, tends to promote the failure of the outermost fibre layer localized in the tensile side, thus leading to the ultimate failure of the whole laminate. This thesis is supported by both the localization of damage that occurred during the last seconds of the test (Fig. 27) compared to the initial loading stages and the transition from a gradual to a more catastrophic failure mode at the end of flexural test, similar to what observed for carbon laminates.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
215
Fig. 22. (a) Typical AE amplitude vs. time response during flexural test on undamaged BC-HS laminates; (b) typical amplitude distribution during flexural test on undamaged BC-HS laminates; (c) amplitude vs. duration response during flexural test on undamaged BC-HS laminates.
216
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 23. (a) Typical amplitude distribution during flexural test on BC-HS laminates impacted at 25 J; (b) amplitude vs. duration response during flexural test on BC-HS laminates impacted at 25 J.
Fig. 24. Acoustic emission localization plot for BC-HS laminates impacted at 25 J. The inset represents a close-up view of typical failure modes for BC-HS laminates impacted at 25 J.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
217
Fig. 25. Typical AE amplitude vs. time response during flexural test on undamaged BC-HI laminates; (b) typical amplitude distribution during flexural test on undamaged BCHI laminates; (c) amplitude vs. duration response during flexural test on undamaged BC-HI laminates.
218
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 26. Typical amplitude distribution during flexural test on BC-HI laminates impacted at 25 J recorded during the first 110 s.
Fig. 27. Acoustic emission localization plot for undamaged BC-HI laminates (a) during the first 110 s and (b) between 110 and 130 s.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
4. Conclusions The effects of both basalt fibre hybridization and stacking sequence on low velocity impact response and damage tolerance capability of carbon fabric reinforced epoxy composites have been experimentally investigated. Two different hybrid laminate configurations have been prepared: one with a sandwich-like stacking sequence and the other one with an intercalated fabrics lay-up. From the results, the main conclusions can be summarized as follows:
Impact damage, at comparable energy levels, causes penetration of the carbon laminates while basalt ones show cross-shaped cracks on the front face and splitting on the back surface but without evidence of penetration. Basalt fibre hybridization prevents penetration in both hybrid laminates while enhancing the peak forces compared to carbon laminates. The higher ductility of basalt fibres allows larger global deformation of the composites thus contributing to a wider damaged area and energy absorption for basalt and hybrid laminates. The sandwich laminates, though substantially superior in terms of static properties (flexural and interlaminar shear strength), appear to be more sensitive to the effect of impact damage if compared to intercalated hybrids. Hybrid intercalated laminates present the best combination of post-impact flexural strength and damage tolerance among the configurations investigated. Acoustic emission analysis helped in clarifying the damage mechanisms involved in the retention of strength and stiffness properties after impact events. Carbon laminates fail by multiple matrix cracks localized in the tensile side while in basalt laminates the interface failures seem to be the relevant energy dissipation mechanisms. Hybrid laminates with sandwich configuration fail by interface failures located in the basalt skins of the compression side that prevent the skins from hindering crack propagation in the carbon core. Damage partition among multiple internal basalt/ carbon interfaces is identified as the mechanism responsible for higher energy absorption and lower strength degradation of intercalated hybrid composites.
References _ BM, Karakuzu R. An experimental investigation of the [1] Aktasß M, Atas C, Içten impact response of composite laminates. Compos Struct 2009;87:307–13. [2] Wang J, Waas AM, Wang H. Experimental and numerical study on the lowvelocity impact behavior of foam-core sandwich panels. Compos Struct 2013;96:298–311. [3] Abrate S. Impact on laminated composite materials. Appl Mech Rev 1991;44:155. [4] Abrate S. Impact on laminated composites: recent advances. Appl Mech Rev 1994;47:517. [5] Cantwell WJ, Morton J. The impact resistance of composite materials – a review. Composites 1991;22:347–62. [6] Reis PNB, Ferreira JAM, Santos P, Richardson MOW, Santos JB. Impact response of Kevlar composites with filled epoxy matrix. Compos Struct 2012;94:3520–8. [7] Gustin J, Joneson A, Mahinfalah M, Stone J. Low velocity impact of combination Kevlar/carbon fiber sandwich composites. Compos Struct 2005;69:396–406. [8] Cheon SS, Lim TS, Lee DG. Impact energy absorption characteristics of glass fiber hybrid composites. Compos Struct 1999;46:267–78. [9] Naik N, Ramasimha R, Arya H, Prabhu S, ShamaRao N. Impact response and damage tolerance characteristics of glass–carbon/epoxy hybrid composite plates. Compos B Eng 2001;32:565–74. [10] Hosur MV, Adbullah M, Jeelani S. Studies on the low-velocity impact response of woven hybrid composites. Compos Struct 2005;67:253–62. [11] Sevkat E, Liaw B, Delale F, Raju BB. Drop-weight impact of plain-woven hybrid glass–graphite/toughened epoxy composites. Compos A Appl Sci Manuf 2009;40:1090–110.
219
[12] Sayer M, Bektasß NB, Demir E, Çalliog˘lu H. The effect of temperatures on hybrid composite laminates under impact loading. Compos B Eng 2012;43:2152–60. [13] Deak T, Czigany T. Chemical composition and mechanical properties of basalt and glass fibers: a comparison. Text Res J 2009;79:645–51. [14] Ross A. Basalt fibers: alternative to glass? Compos Technol 2006;12:44–8. [15] Dorigato A, Pegoretti A. Fatigue resistance of basalt fibers-reinforced laminates. J Compos Mater 2012;46:1773–85. [16] Mingchao Wang, Zuoguang Zhang, Yubin Li, Min Li, Zhijie Sun. Chemical durability and mechanical properties of alkali-proof basalt fiber and its reinforced epoxy composites. J Reinf Plast Compos 2008;27:393–407. [17] Liu Q, Shaw MT, Parnas RS, McDonnell A-M. Investigation of basalt fiber composite mechanical properties for applications in transportation. Polym Compos 2006;27:41–8. [18] Carmisciano S, De Rosa IM, Sarasini F, Tamburrano A, Valente M. Basalt woven fiber reinforced vinylester composites: flexural and electrical properties. Mater Des 2011;32:337–42. [19] Kim MT, Rhee KY, Park SJ, Hui D. Effects of silane-modified carbon nanotubes on flexural and fracture behaviors of carbon nanotube-modified epoxy/basalt composites. Compos B Eng 2012;43:2298–302. [20] Colombo C, Vergani L, Burman M. Static and fatigue characterisation of new basalt fibre reinforced composites. Compos Struct 2012;94:1165–74. [21] Manikandan V, Winowlin Jappes JT, Suresh Kumar SM, Amuthakkannan p. Investigation of the effect of surface modifications on the mechanical properties of basalt fibre reinforced polymer composites. Compos B Eng 2012;43:812–8. [22] Szabó J, Czigány T. Static fracture and failure behavior of aligned discontinuous mineral fiber reinforced polypropylene composites. Polym Test 2003;22:711–9. [23] Karger-Kocsis J, Zhang Z, Czigány T. Special manufacturing and characteristics of basalt fiber reinforced hybrid polypropylene composites: Mechanical properties and acoustic emission study. Compos Sci Technol 2006;66:3210–20. [24] Zhang Y, Yu C, Chu PK, Lv F, Zhang C, Ji J, et al. Mechanical and thermal properties of basalt fiber reinforced poly(butylene succinate) composites. Mater Chem Phys 2012;133:845–9. [25] Akinci A, Yilmaz S, Sen U. Wear behavior of basalt filled low density polyethylene composites. Appl Compos Mater 2011;19:499–511. [26] Lopresto V, Leone C, De Iorio I. Mechanical characterisation of basalt fibre reinforced plastic. Compos B Eng 2011;42:717–23. [27] Dehkordi MT, Nosraty H, Shokrieh MM, Minak G, Ghelli D. Low velocity impact properties of intra-ply hybrid composites based on basalt and nylon woven fabrics. Mater Des 2010;31:3835–44. [28] De Rosa IM, Marra F, Pulci G, Santulli C, Sarasini F, Tirillò J, et al. Post-impact mechanical characterisation of E-glass/basalt woven fabric interply hybrid laminates. eXPRESS. Polym Lett 2011;5:449–59. [29] Rosa IM, Marra F, Pulci G, Santulli C, Sarasini F, Tirillò J, et al. Post-impact mechanical characterisation of glass and basalt woven fabric laminates. Appl Compos Mater 2011;19:475–90. [30] Wang X, Hu B, Feng Y, Liang F, Mo J, Xiong J, et al. Low velocity impact properties of 3D woven basalt/aramid hybrid composites. Compos Sci Technol 2008;68:444–50. [31] Sarasini F, Tirillò J, Valente M, Valente T, Cioffi S, Iannace S, et al. Effect of basalt fiber hybridization on the impact behavior under low impact velocity of glass/ basalt woven fabric/epoxy resin composites. Compos A Appl Sci Manuf 2013;47:109–23. [32] Tehrani Dehkordi M, Nosraty H, Shokrieh MM, Minak G, Ghelli D. The influence of hybridization on impact damage behavior and residual compression strength of intraply basalt/nylon hybrid composites. Mater Des 2013;43:283–90. [33] Sarasini F, Tirillò J, Valente M, Ferrante L, Cioffi S, Iannace S, et al. Hybrid composites based on aramid and basalt woven fabrics: impact damage modes and residual flexural properties. Mater Des 2013;49:290–302. [34] Cao Shenghu, Zhis WU, Wang X. Tensile properties of CFRP and hybrid FRP composites at elevated temperatures. J Compos Mater 2009;43:315–30. [35] Wu Z, Wang X, Iwashita K, Sasaki T, Hamaguchi Y. Tensile fatigue behaviour of FRP and hybrid FRP sheets. Compos B Eng 2010;41:396–402. [36] Zhang CH, Zhang JB, Qu MC, Zhang JN. Toughness properties of basalt/carbon fiber hybrid composites. Adv Mater Res 2011;150–151:732–5. [37] Belingardi G, Vadori R. Low velocity impact tests of laminate glass-fiber-epoxy matrix composite material plates. Int J Impact Eng 2002;27:213–29. [38] Benevolenski OI, Karger-Kocsis J, Mieck K-P, Reubmann T. Instrumented perforation impact response of polypropylene composites with hybrid reinforcement flax/glass and flax/cellulose fibers. J Thermoplast Compos Mater 2000;13:481–96. [39] Belingardi G, Cavatorta MP, Salvatore Paolino D. Repeated impact response of hand lay-up and vacuum infusion thick glass reinforced laminates. Int J Impact Eng 2008;35:609–19. [40] Bledzki AK, Mamun A, Faruk O. Abaca fibre reinforced PP composites and comparison with jute and flax fibre PP composites. eXPRESS. Polym Lett 2007;1:755–62. [41] Huguet S, Godin N, Gaertner R, Salmon L, Villard D. Use of acoustic emission to identify damage modes in glass fibre reinforced polyester. Compos Sci Technol 2002;62:1433–44. [42] Barré S, Benzeggagh ML. On the use of acoustic emission to investigate damage mechanisms in glass-fibre-reinforced polypropylene. Compos Sci Technol 1994;52:369–76.
220
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
[43] Faudree M, Baer E, Hiltner A, Collister J. Characterization of damage and fracture processes in short fiber BMC composites by acoustic emission. J Compos Mater 1988;22:1170–95.
[44] Madhukar MS, Drzal LT. Fiber-matrix adhesion and its effect on composite mechanical properties. III. Longitudinal (0 ) compressive properties of graphite/epoxy composites. J Compos Mater 1992;26:310–33.
Contents lists available at ScienceDirect
Composites: Part B journal homepage: www.elsevier.com/locate/compositesb
Drop-weight impact behaviour of woven hybrid basalt–carbon/epoxy composites F. Sarasini a,⇑, J. Tirillò a, L. Ferrante a, M. Valente a, T. Valente a, L. Lampani b, P. Gaudenzi b, S. Cioffi c, S. Iannace c, L. Sorrentino c a b c
Department of Chemical Engineering Materials Environment, Sapienza-Università di Roma, Via Eudossiana 18, 00184 Rome, Italy Department of Mechanical and Aerospace Engineering, Sapienza-Università di Roma, Via Eudossiana 18, 00184 Rome, Italy Institute for Composite and Biomedical Materials, National Research Council, Piazzale Enrico Fermi 1, Località Granatello, Portici (NA) 80055, Italy
a r t i c l e
i n f o
Article history: Received 10 October 2013 Received in revised form 23 November 2013 Accepted 7 December 2013 Available online 15 December 2013 Keywords: A. Polymer-matrix composites (PMCs) Basalt fibres A. Hybrid B. Impact behaviour D. Non-destructive testing
a b s t r a c t This study addresses the effects of basalt fibre hybridization on quasi-static mechanical properties and low velocity impact behaviour of carbon/epoxy laminates. Interply hybrid specimens with two different stacking sequences (sandwich-like and intercalated) are tested at three different energies, namely 5, 12.5 and 25 J. Residual post-impact properties of the different configurations of carbon/basalt hybrid laminates are characterized by quasi static four point bending tests. Post-impact flexural tests and interlaminar shear tests are used for the mechanical characterization along with two non-destructive methods, namely acoustic emission and ultrasonic phased array, in order to get further information on both the extent of damage and failure mechanisms. Results indicate that hybrid laminates with intercalated configuration (alternating sequence of basalt and carbon fabrics) have better impact energy absorption capability and enhanced damage tolerance with respect to the all-carbon laminates, while hybrid laminates with sandwich-like configuration (seven carbon fabric layers at the centre of the laminate as core and three basalt fabric layers for each side of the composite as skins) present the most favourable flexural behaviour. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Most of the composite structures and parts used in advanced applications are fibre-reinforced composites. During their service life, composites are subjected to various loading conditions and low velocity impact is one of the most critical in particular for aerospace composite structures [1–5]. Due to their high specific stiffness and strength, carbon fibre reinforced polymer composite is the material of choice in aerospace industry. However, the toughness of carbon fibre is quite low and the resulting damage resistance is poor. In this regard, several approaches have been successfully exploited to enhance the impact damage resistance of composite laminates. One approach consists in enhancing the properties of the matrix material, as proposed by Reis et al. [6], who investigated the impact behaviour as well as the damage tolerance of Kevlar/filled epoxy matrix with two different fillers, namely cork powder and nanoclays. They found that adding fillers increased the maximum impact load, which was very dependent on the filler type especially for high impact energy, and the addition of clays increased the damaged area of around 29% while ⇑ Corresponding author. Tel.: +39 0644585408; fax: +39 066876343. E-mail address: [email protected] (F. Sarasini). 1359-8368/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.compositesb.2013.12.006
improving the residual strength of the laminates. Another approach is known as hybridization, usually with high strain to failure fibres to improve the damage resistance of composites to both low velocity and instrumented Charpy impact [7,8]. In this regard, glass fibres are the best option from the viewpoint of cost, availability and ease of processing, and hybrid carbon/glass fibre composites have consistently demonstrated better damage tolerance under impact than their carbon fibre counterparts [9–12]. In the field of fibrous reinforcement, basalt fibres have gained an increasing attention in recent years as possible replacement of the conventional glass fibres [13] due to their advantages in terms of environmental cost and chemical–physical properties. Mineral fibres obtained from basalt rocks are not new, but their suitability as reinforcement in polymer composites is a relatively new issue [14]. From the mechanical point of view, continuous basalt fibres are competitive with glass ones. The elastic modulus of basalt fibres strongly depends on the chemical composition but it is usually comparable or slightly higher than that of glass fibres, while both tensile strength and elongation at break are higher [13]. These properties suggest the potential role of basalt fibres in replacing glass ones as impact resistance improver for composite laminates with also a view to enhancing the environmental sustainability of such composites. The mechanical properties of basalt fibre
205
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
reinforced composite laminates have been thoroughly investigated for both thermoset [15–21] and thermoplastic matrices [22–25], but only limited attention has been devoted to the low-velocity impact behaviour of these class of composites [26–33]. Lopresto et al. [26] provided a thorough investigation of the mechanical properties of basalt/epoxy composites, studying also the impact resistance. The low velocity impact tests were only performed at about 100 J in order to perforate the composite specimens and reported, in comparison to glass fibre reinforced composites, higher energy absorption capability for basalt. A more detailed investigation of the effect of hybridization of basalt fibres on low velocity impact response of glass fabric reinforced epoxy composites was performed in [28,29,31], where was experimentally confirmed that basalt composites exhibit better energy absorption capability when compared to glass ones and that glass laminates’ poor damage tolerance can be enhanced by the hybridization with basalt layers. Limited research has been also performed on hybrid composites made of basalt and ductile fibres, such as nylon [27,32] and aramid [30,33]. From the review of the available literature, there is a reduced data availability on hybrid carbon/basalt composites [34–36] and, in particular, on their post-impact behaviour, even though a potential positive effect of basalt fibres in enhancing toughness (evaluated by an open hole compression test) of carbon fibre composites was pointed out [36]. The lack of data on carbon/ basalt hybrid composites is the rationale behind this experimental work, whose aim is to provide a comprehensive, though preliminary, investigation of the low-velocity impact behaviour and residual properties of basalt/carbon hybrid laminates by taking into account the effect of different stacking sequences. To attain this aim, woven basalt-carbon/epoxy hybrid composites were fabricated in interply hybrid structures and subjected to low-velocity impact using a drop weight apparatus. Specimens with different lay-up configurations were tested using three impact energies and the influence of impact energy on the flexural residual strength of composites was assessed by quasi-static four point bending tests. To determine the extent of damage, the specimens were subjected to ultrasonic phased array testing. The post-impact flexural tests were monitored by acoustic emission (AE) and the results of these analyses, along with those of the impact tests (maximum load and energy absorbed) were used to understand the role played by basalt fibre hybridization on the mechanical behaviour under impact of carbon fibre reinforced epoxy composites. In addition, interlaminar shear strength for all the hybrids is also addressed. The combined use of destructive and non-destructive characterizations provides a confirmation of the possibility of replacing glass fibres with basalt ones to improve the impact resistance of carbon-based laminates along with an indication of the most suited stacking sequence to be used to this aim.
2. Materials and methods The basalt (BAS 220.1270.P) and carbon fabrics (CC160) are plain weave fabrics supplied by Basaltex-Flocart NV (Belgium) and Saati Composites, respectively. The fibre areal densities are 220 g/m2 and 160 g/m2 for basalt and carbon fabrics, respectively. A bi-component epoxy resin (EC157+W152 MR), supplied by Elantas Camattini (Italy) is selected as polymeric matrix. The laminates have been manufactured by a laboratory Resin Transfer Moulding (RTM) system and have been cured for 12 h at room temperature and 4 h at 70 °C. All configurations are produced using thirteen fabric layers with a similar volume fraction and thickness. Two hybrid configurations have been manufactured: in the first one (BC-HS) fabrics are stacked as a sandwich-like sequence with seven carbon fibre layers (core) and three basalt fibre layers (skins) for each side of the laminate, while in the second configuration (BC-HI) seven
Table 1 Thickness and fibre volume fraction of the tested laminates. Composite type
Thickness (mm)
Fibre volume fraction
B C BC-HS BC-HI
3.40 ± 0.05 3.50 ± 0.04 3.50 ± 0.04 3.45 ± 0.05
0.315 ± 0.01 0.325 ± 0.01 0.319 ± 0.01 0.321 ± 0.01
layers of basalt fabrics and six of carbon fabrics are alternatively stacked, keeping basalt fabrics as outer layers. Not hybridised basalt (B) and carbon (C) reinforced epoxy composites have been also manufactured as reference configurations. Details of all the composite types are reported in Table 1. Specimens, with dimensions of 180 mm 60 mm 3.4 mm (length width thickness) for the structural characterisations have been cut from 300 mmwide square composite laminates. This unusual width has been chosen in order to be sure of accommodating the whole impact damaged area, thus minimizing the edge effects. Four-point bending tests have been performed on five specimens for each configuration in accordance with ASTM D 6272. A span-to-depth ratio of 23:1 and a cross-head speed of 2.5 mm/min have been used. Strain gauges have been employed to measure the effective strain and to evaluate the flexural modulus. Specimens have been tested in bending either after their production (non impacted samples) or after the low-velocity impact tests to measure their residual flexural strength. Impact conditions are selected in order to assure that the impact does not result in the full penetration of the laminate and that the damage extension into the sample does not exceed the specimen width, even under the highest energy applied. Impact tests have been performed by using a falling dart impact testing machine, model Fractovis Plus from CEAST (Pianezza – TO, Italy) over beam-like specimens with the dimensions previously described. Specimens have been tested at three impact energies (5 J, 12.5 J and 25 J) by keeping constant the indenter mass (6.929 kg) with an hemispherical impact head (diameter equal to 12.7 mm). The sample holder is circular, with the external diameter of 60 mm and the inner of 40 mm. The impacted specimens have been scanned using non-destructive ultrasonic inspection equipment OmniScan MX with standard phased array probe 3.5 MHz, linear array, 64 elements. The interlaminar shear strength (ILSS) has been evaluated in accordance with ASTM D 2344. Ten specimens have been tested for each laminate, having the following dimensions: 20 mm 6.8 mm 3.4 mm (L W t). A span-to-depth ratio of 4:1 and a cross-head speed of 1 mm/min have been used. All the mechanical tests have been performed on a Z010 from Zwick/Roell (Ulm, Germany) universal testing machine equipped with a 10 kN load cell. Post-impact flexural tests have been monitored by acoustic emission until final fracture occurred using an AMSY-5 AE system by Vallen Systeme GmbH (Icking, Germany). The AE acquisition settings used throughout this experimental work are as follows: threshold = 35 dB, Rearm Time (RT) = 0.4 ms, Duration Discrimination Time (DDT) = 0.2 ms and total gain = 34 dB. This threshold level has been defined from a 30 min track record of the background noise, with the AE setup configuration actually used, and has been set 6 dB above the maximum level of the recorded spurious signal from the electronic system. Two broad-band (100–1500 kHz, Fujicera 1045S) PZT AE sensors have been used. The sensors have been placed on the surface of the specimens at both ends to allow linear localization, with silicone grease as coupling agent. 3. Results and discussion The present investigation follows the established methodology for damage tolerance assessment which consists of four major
206
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
sequential steps, namely: impact testing, damage characterisation, determination of static residual strength and damage tolerance evaluation. The flexural and interlaminar shear mechanical properties are also evaluated, being them considered the most relevant to impact damage. 3.1. Low velocity impact response and damage evaluation Figs. 1–3 show the different damage patterns and structural responses of laminates tested at increasing impact energies. Dropweight impact test at 25 J on C composites shows penetration of the dart through its thickness with splitting on the back face, highlighting the brittle fracture of carbon laminates. In the same impact condition, B composites exhibit cross-shaped cracks and debonding on the front face with associated matrix cracking and back surface splitting but without evidence of penetration. BC-HI and BC-HS laminates show an intermediate behaviour between B and
C composites, although different patterns are detected as a consequence of the different lay-ups. In fact, the damaged area for BC-HI laminates is larger than that of BC-HS one, suggesting that the former can absorb more energy. Both hybrid laminates show bulge on the back surface as a consequence of matrix cracking and pull-out of outer fibre layers but the damage pattern is quite different. The BC-HS laminates present an impact behaviour more similar to B ones because the three outer layers, being made of basalt, prevent extensive fibre breakage unlike C composites. Conversely, the impacted BC-HI laminates exhibit back surface damages approaching the pattern of C laminates but without penetration due to the presence of alternating basalt layers which provide enhanced compliance. To assess composite impact damage, it is common to refer to the impact energy (Ei) and absorbed energy (Ea). Impact energy is the kinetic energy of the impactor right before contact with samples takes place, whereas absorbed energy is the energy dissipated
Fig. 1. Photos of damage progression on front and rear faces of B, C, BC-HI and BC-HS composite panels impacted at 5 J.
Fig. 2. Photos of damage progression on front and rear faces of B, C, BC-HI and BC-HS composite panels impacted at 12.5 J.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
207
Fig. 3. Photos of damage progression on front and rear faces of B, C, BC-HI and BC-HS composite panels impacted at 25 J.
by the system through the several mechanisms occurring after the impactor’s contact, like elastic deformation, friction, plastic deformation and, most importantly, those peculiar to the material (matrix cracking, debonding, pull-out, fibre breakage) [37]. In most of tested samples, penetration does not occur and the impactor rebounds with an energy that is the difference between Ei and Ea. The absorbed energy can be calculated from force– displacement curves (Figs. 4–6) as the area enclosed within the curve. Indeed if the displacement returns toward the axis origin during unloading, it means that some elastic energy is recovered by the laminate. Carbon laminates are the only laminates to be perforated at 25 J without showing any energy recovery (Fig. 6: displacement increases while force decreases), whereas all composites containing basalt layers exhibit a residual elastic response. Basalt and hybrid laminates absorb more energy through higher overall deformation, being more compliant, and through delaminations. Key impact parameters like peak force, impact energy (Ei), maximum displacement and absorbed energy (Ea) obtained from the transient response of each laminate are summarized in Table 2.
Fig. 4. Typical force vs. displacement response for 5 J-impacted carbon (C), basalt (B), carbon/basalt sandwich hybrid (BC-HS) and carbon/basalt intercalated hybrid (BC-HI) composites.
Fig. 5. Typical force vs. displacement response for 12.5 J-impacted carbon (C), basalt (B), carbon/basalt sandwich hybrid (BC-HS) and carbon/basalt intercalated hybrid (BC-HI) composites.
Fig. 6. Typical force vs. displacement response for 25 J-impacted carbon (C), basalt (B), carbon/basalt sandwich hybrid (BC-HS) and carbon/basalt intercalated hybrid (BC-HI) composites.
208
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Table 2 Parameters obtained from impact tests on basalt, carbon and hybrid composites.
*
Specimen
Peak force (N)
Maximum displacement (mm)
Impact energy (J)
Absorbed energy (J)
Damage degree
Damaged area (mm2)*
Energy: 5 J B C BC-HI BC-HS
3511.98 ± 64.27 3745.85 ± 212.05 3865.67 ± 83.26 3564.78 ± 173.98
2.33 ± 0.03 1.83 ± 0.12 2.20 ± 0.03 2.08 ± 0.03
4.85 ± 0.01 4.85 ± 0.07 4.97 ± 0.01 4.96 ± 0.01
1.84 ± 0.09 2.30 ± 0.09 2.14 ± 0.07 2.27 ± 0.10
0.38 ± 0.01 0.46 ± 0.01 0.43 ± 0.01 0.45 ± 0.02
310 298 690 594
Energy: 12.5 J B C BC-HI BC-HS
5654.42 ± 16.62 4209.68 ± 223.41 4792.75 ± 154.59 5090.41 ± 58.01
3.92 ± 0.09 3.71 ± 0.23 4.01 ± 0.33 3.76 ± 0.06
12.38 ± 0.01 12.38 ± 0.07 12.47 ± 0.07 12.46 ± 0.07
6.71 ± 0.19 9.24 ± 0.25 7.99 ± 0.33 6.00 ± 0.23
0.54 ± 0.01 0.74 ± 0.02 0.64 ± 0.03 0.48 ± 0.02
1048 736 966 603
Energy: 25 J B C BC-HI BC-HS
7238.07 ± 117.11 4261.90 ± 276.72 4801.88 ± 37.35 6726.30 ± 202.09
6.06 ± 0.17 – 7.25 ± 0.16 6.03 ± 0.11
24.96 ± 0.01 27.95 ± 0.12 24.95 ± 0.10
20.66 ± 0.50 – 21.84 ± 0.40 18.38 ± 1.04
0.82 ± 0.02 – 0.86 ± 0.01 0.72 ± 0.04
889 – 1027 999
Evaluated from ultrasonic C-scans.
It can be seen that the peak load increases with the increase in the incident kinetic energy, indicating greater load bearing ability of the laminates at higher energy levels. In terms of peak force, basalt laminates perform better than carbon ones at high energy levels as was the case for glass laminates with respect to carbon ones [11], while hybrid laminates show an increased peak force compared to carbon ones. As confirmed by experimental results, the laminate configuration heavily affects the transient response, with sandwich-like hybrid composites outperforming intercalated hybrids, as reported in other studies [31,33]. Carbon/epoxy laminates exhibit a lower displacement, whilst basalt and hybrid laminates, being more compliant, absorb energy through the global deformation. In order to assess the damage accumulated by the materials, a parameter called damage degree, defined as the ratio Ea/Ei, was used among the several ones available [38–40]. From Table 2, it is evident that damage degree of carbon-epoxy laminate increases dramatically moving from 5 J to 12.5 J, and reaches unity at 25 J. This is a further confirmation of the inherent brittleness of C laminates, which cannot absorb high impact energy because of their abrupt and catastrophic damage mechanism. Damage degree of basalt containing laminates increases with the impact energy, reaching a maximum value of 0.86 for the BC-HI composite, thus suggesting the positive role played by basalt fibre hybridization, which improves the impact energy absorption. The ultrasonic phased array testing method was used to evaluate damaged area that is connected with the energy absorbed by the samples. Fig. 7 shows C-scan images of different impacted samples from which it is estimated the extent of damage reported in Table 2. All C-scans are relative to an area of 59 mm 49 mm, whilst the damaged area (in mm2) for each type of specimen has to be considered through-the-thickness being the summation of damaged areas of each layer. It has been difficult to evaluate the damage caused by the impactor because of the additional damage introduced by the circular holder beneath the samples during impact tests, which was particularly important for the basalt laminates. This can explain why the data for basalt laminates in Table 2 do not exhibit a clear trend with increasing impact energy. On the whole, all the examined samples show a progression in damaged area with increasing impact energy, though the behaviour is dependent on the fabrics stacking sequence. The carbon fibre reinforced laminates exhibit a small damaged area, confirming that C composite absorbs impact load in a brittle way. On the other hand, BC-HI and BC-HS show a wider damaged area suggesting that delamination between different kinds of layers occurs with the consequence of improving the energy dissipation. To summarize,
among all the investigated composites, BC-HI composites exhibit the widest damaged areas due to the increased number of events occurring at the multiple interfaces between basalt and carbon fabric layers. This mechanism together with larger global deformation offered by the addition of basalt fibres are responsible for the higher energy absorption of hybrid laminates with respect to carbon laminates. 3.2. Interlaminar shear strength Interlaminar shear strength (ILSS) is usually a limiting design property for composites and depends on many parameters such as (a) adhesion between matrix and fibre, (b) constitutive materials, (c) fibre volume fraction and (d) stacking sequence. When the transverse shear load experienced by a laminated composite exceeds the interlaminar shear strength, a delamination failure will occur between the reinforcing layers. Fig. 8 shows that B laminates exhibit the worst interlaminar shear strength whereas C laminates provide the best one. This might result from the better interfacial bonding between carbon and epoxy matrix with respect to basalt and epoxy resin. Hybridization with basalt fibres plays a substantial role only in the BC-HS case, in accordance with the lower number of basalt/epoxy interfaces. 3.3. Residual mechanical properties and damage tolerance evaluation The basalt fibre hybridization affects also the flexural strength and modulus of laminates, both before and after impact. Carbon/ epoxy composites are stiffer than basalt/epoxy ones as the carbon fibres have higher elastic modulus than basalt ones (Table 3). Flexural tests on specimens impacted at 12.5 J (Fig. 9) highlight that hybrid composites have flexural stiffness intermediate between carbon/epoxy and basalt/epoxy ones. The different lay-up of the hybrid composites plays an important role on the laminate stiffness also considering that carbon layers are stiffer than basalt ones. Although BC-HI laminates have less carbon layers than BC-HS ones, the former possess higher flexural modulus among all basalt containing composites. This has been related to the fact that in the BC-HS sandwich structure, the outer layers, responsible for bearing the bending load, are made of basalt (less stiff than carbon). Hybridization also provides improved toughness and flexural strength. C composites show a higher flexural strength than B ones after 12.5 J impact (Fig. 9), but hybridized configurations allow to strongly improve both flexural strength and toughness with respect to carbon laminates, and the sandwich configuration seems
209
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
C–5J
C – 12.5 J
BC-HI 5 J
BC-HI 12.5 J
BC-HI 25 J
BC-HS 5 J
BC-HS 12.5 J
BC-HS 25 J
B-5J
B – 12.5 J
B - 25 J
Fig. 7. Ultrasonic C-scan images of four types of composites impacted at different impact energies.
Table 3 Summary of flexural properties for basalt, carbon and hybrid composites.
Fig. 8. Average ILSS for basalt, carbon and hybrid composites (the error bars represent the standard deviation).
Specimen
Flexural strength (MPa)
Flexural modulus (GPa)
Non-impacted B C BC-HI BC-HS
229.34 ± 3.24 593.48 ± 9.57 350.84 ± 0.30 418.68 ± 8.10
14.35 ± 0.75 38.70 ± 0.10 25.84 ± 2.78 24.20 ± 1.42
Impact Energy: 5 J B C BC-HI BC-HS
218.64 ± 3.76 349.01 ± 35.80 345.21 ± 22.14 369.58 ± 5.55
14.16 ± 0.34 33.12 ± 0.94 25.23 ± 0.07 21.99 ± 1.37
Impact Energy: 12.5 J B 207.30 ± 3.09 C 238.59 ± 0.50 BC-HI 304.45 ± 8.07 BC-HS 316.77 ± 0.79
13.55 ± 0.49 27.29 ± 0.27 23.64 ± 0.10 20.59 ± 0.06
Impact Energy: 25 J B C BC-HI BC-HS
11.48 ± 0.64 – 20.90 ± 0.22 18.58 ± 0.03
142.55 ± 3.92 – 259.72 ± 2.80 250.79 ± 13.45
210
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 9. Typical stress vs. strain curves for flexural tests on composites previously impacted at 12.5 J.
to represent the best compromise between these parameters, often resulting in competing relationship. This behaviour has been related to the higher number of carbon layers in the BC-HS configuration, characterized by a higher ultimate strength than basalt ones. Thus the flexural strength is more influenced by the fibre mechanical properties rather than by composite stacking sequence. Although the non-impacted laminates can be ranked from the higher to the lower flexural strength in the following order: C > BC-HS > BC-HI > B (Table 2), the trend is quite different for impacted samples, in particular as the impact energy increases. In fact, it should be noted that the increase of impact energy leads to increased damaged areas, which in turn reduce both flexural strength and flexural stiffness. In order to obtain a better indication of the damage tolerance (residual properties after impact) the normalized flexural strength (or stiffness) of each specimen is evaluated as the ratio of the mean strength (or stiffness) of the impacted specimen to the mean value of the flexural strength (or stiffness) of the undamaged specimen. Figs. 10 and 11 show the normalized flexural strength and flexural stiffness with respect to the increasing impact energy, respectively. After being impacted at 12.5 J, the C laminate loses almost 60% of its early flexural strength and almost 30% of its early flexural stiffness whereas the impact at 25 J leads to penetration with
Fig. 10. Normalized residual flexural strength as a function of increasing impact energy.
Fig. 11. Normalized residual flexural stiffness as a function of increasing impact energy.
catastrophic failure, confirming the worst damage resistance and tolerance capabilities among all considered laminate configurations. This behaviour is due to the characteristic energy absorption mechanism of C composites, which dissipate energy mainly by fast transverse crack propagation and fibre fracture. B composite shows a better behaviour with a more gradual degradation pattern because the impact energy was mostly absorbed by the interface failures occurring among laminate layers and at the fibre/matrix interface. It is evident the positive role played by basalt fabric on the post-impact characteristics of hybrid laminates. More specifically, BC-HI presents the most favourable degradation pattern as it shows the best damage tolerance (Figs. 10 and 11). The BC-HS laminate, despite of presenting higher quasi-static flexural properties when compared to BC-HI (Table 3), exhibits a steeper damage resistance drop (Fig. 10). This has been related to the energy dissipation mechanism of BC-HI through multiple small delaminations between dissimilar layers (Fig. 12a) whereas the BC-HS dissipates energy mainly through transverse crack propagation in carbon ‘‘core’’ of sandwich laminate and secondly through main delaminations at the basalt-skin/carbon-core interface (Fig. 12b), as confirmed by AE monitoring. 3.4. Acoustic emission monitoring The time evolution and nature of different failure modes of composites is studied through the analysis of acoustic emission signals recorded in real time during flexural tests on non-impacted and impacted composite laminates. Basalt (B) and carbon (C) laminates exhibit a different response to impact loading in terms of damage degree and damage tolerance. This behaviour suggests that the mechanisms responsible for the energy absorption are dissimilar. For non-impacted C laminates, very few AE signals are recorded until final fracture occurs, thus confirming the catastrophic and localized failure (Fig. 13). Most of AE signals have amplitudes in the range 35–50 dB, which are usually ascribed to matrix cracking, while few signals can be ascribed to fibre breakages (80–100 dB) and interfacial failures (55–65 dB) (Fig. 14) [41–43]. This behaviour is consistent with the dominant failure mode of these composites mainly due to multiple matrix cracks localized in the tensile side. The failure of carbon composites results to be therefore characterized by the nucleation of matrix cracks, which then trigger debonding phenomena and further transverse cracks leading to fibre failures. From the localization plots of AE signals (Fig. 15) it can be inferred that the extent of the damaged zone responsible for the intense acoustic emission
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 12. Close-up view of (a) BC-HI and (b) BC-HS specimens impacted at 12.5 J after flexural test.
Fig. 13. Typical AE amplitude vs. time response during flexural test on undamaged carbon laminates.
Fig. 14. Typical amplitude distribution during flexural test on undamaged carbon laminates.
211
212
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 15. Acoustic emission localization plot for undamaged carbon laminates. The inset represents a close-up view of typical failure modes for undamaged carbon laminates.
Fig. 16. Typical AE amplitude vs. time response during flexural test on carbon laminates impacted at 12.5 J.
activity is wide, as confirmed by the micrograph in the inset of Fig. 15. The impact damage causes an early onset of acoustic emission activity (signals of low amplitude and short duration) and a higher number of signals. This trend increases with increasing impact energy (Fig. 16). Impact loading causes, in brittle carbon laminates, an unstable damage in terms of matrix cracks and small debonding phenomena, which become active at lower stresses during the subsequent flexural loading. Most AE signals are still localized in the lower amplitude range (<55 dB) thus confirming the same failure modes observed in non-impacted laminates even though the presence of impact damage seems to quicken the process thanks to the splitting in the rear face. This thesis is supported by the localization of AE signals (Fig. 17), which appears to be similar to the one observed for non-impacted laminates (Fig. 15). Basalt laminates show a different behaviour, which is characterized by significant AE signals since the beginning of loading also for the undamaged material (Fig. 18). These signals can be attributed to matrix cracks and interface failures. Compared with carbon laminates, an increase in the number of signals in the range 65–75 dB (pull-out) is observed (Fig. 19). This suggests that interface failures
are the dominant failure mechanism, which is mainly localized in the compression side. This can explain the lower strength of basalt/epoxy laminates compared to carbon ones. Failure in basalt laminates takes place in compressed half section, whilst in carbon laminates the whole section is involved in the fracture process up to the tensile face. The fracture surface in basalt laminates exhibits the presence of a typical fracture pattern due to compression failure, consisting of fibre microbuckling and kink bands (inset in Fig. 19). The damaged zone so nucleated can cause longitudinal splitting leading to premature failure of the composite laminates thus preventing higher stresses in the outermost tensile fibres to be reached. Fibre microbuckling is known to occur preferentially in systems characterized by medium interface strength [44]. As for carbon laminates, the presence of impact tends to emphasize the importance of interface failures (pull-out and delaminations, Fig. 20) and to localize the damage (Fig. 21). These interface phenomena, together with higher ductility, appears to be the relevant energy dissipation mechanisms for basalt laminates [33]. Hybrid laminates with sandwich configuration (BC-HS) show, compared with carbon laminates, a higher number of AE signals
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
213
Fig. 17. Acoustic emission localization plot for carbon laminates impacted at 12.5 J.
Fig. 18. Typical AE amplitude vs. time response during flexural test on undamaged basalt laminates.
Fig. 19. Typical amplitude distribution during flexural test on undamaged basalt laminates. The inset represents a close-up view of typical failure modes for undamaged basalt laminates.
related to debonding phenomena located mainly in the compression side and due to the basalt skins. These signals are active at low stresses, as reported in Fig. 22. Moreover, a higher number of signals characterized by medium to high amplitudes (60–75 dB)
and long durations (>1 ms) related to delaminations is recorded. Basalt skins, characterized by higher deformation, hold together the carbon core that, once failed, causes the failure of the whole laminate (Fig. 12b). With increasing impact energy, it is to be noted
214
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 20. Typical amplitude distribution during flexural test on basalt laminates impacted at 25 J.
Fig. 21. Acoustic emission localization plot for (a) undamaged and (b) impacted at 25 J basalt laminates.
an increase of signals related to interface failures (Fig. 23). The presence of such delaminations, mainly localized in the compression side, quickly prevents basalt skins from hindering crack propagation in the carbon core, causing its failure. The growing damage localization with increasing impact energy supports what experimentally observed as regards the failure modes of these laminates (Fig. 24). Undamaged hybrid laminates with intercalated configuration (BC-HI) show a significant number of signals with amplitudes >60 dB related to delaminations (Fig. 25). With increasing impact energy, a considerable rise in the number of signals at the very beginning of loading along with a growing importance of multiple delaminations occur. This is highlighted in Fig. 26, which shows the amplitude distribution of AE signals for BC-HI laminates impacted at 25 J recorded during the first 110 s of
flexural loading. Such delaminations are diffused in the material due to higher number of basalt/carbon interfaces. Damage partition among internal interfaces enables BC-HI laminates, although less resistant than BC-HS hybrids, to show a lower degradation of their strength at higher impact energies, when extensive damage is present. The occurrence of splitting on the rear face, although less evident than in carbon laminates, tends to promote the failure of the outermost fibre layer localized in the tensile side, thus leading to the ultimate failure of the whole laminate. This thesis is supported by both the localization of damage that occurred during the last seconds of the test (Fig. 27) compared to the initial loading stages and the transition from a gradual to a more catastrophic failure mode at the end of flexural test, similar to what observed for carbon laminates.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
215
Fig. 22. (a) Typical AE amplitude vs. time response during flexural test on undamaged BC-HS laminates; (b) typical amplitude distribution during flexural test on undamaged BC-HS laminates; (c) amplitude vs. duration response during flexural test on undamaged BC-HS laminates.
216
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 23. (a) Typical amplitude distribution during flexural test on BC-HS laminates impacted at 25 J; (b) amplitude vs. duration response during flexural test on BC-HS laminates impacted at 25 J.
Fig. 24. Acoustic emission localization plot for BC-HS laminates impacted at 25 J. The inset represents a close-up view of typical failure modes for BC-HS laminates impacted at 25 J.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
217
Fig. 25. Typical AE amplitude vs. time response during flexural test on undamaged BC-HI laminates; (b) typical amplitude distribution during flexural test on undamaged BCHI laminates; (c) amplitude vs. duration response during flexural test on undamaged BC-HI laminates.
218
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
Fig. 26. Typical amplitude distribution during flexural test on BC-HI laminates impacted at 25 J recorded during the first 110 s.
Fig. 27. Acoustic emission localization plot for undamaged BC-HI laminates (a) during the first 110 s and (b) between 110 and 130 s.
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
4. Conclusions The effects of both basalt fibre hybridization and stacking sequence on low velocity impact response and damage tolerance capability of carbon fabric reinforced epoxy composites have been experimentally investigated. Two different hybrid laminate configurations have been prepared: one with a sandwich-like stacking sequence and the other one with an intercalated fabrics lay-up. From the results, the main conclusions can be summarized as follows:
Impact damage, at comparable energy levels, causes penetration of the carbon laminates while basalt ones show cross-shaped cracks on the front face and splitting on the back surface but without evidence of penetration. Basalt fibre hybridization prevents penetration in both hybrid laminates while enhancing the peak forces compared to carbon laminates. The higher ductility of basalt fibres allows larger global deformation of the composites thus contributing to a wider damaged area and energy absorption for basalt and hybrid laminates. The sandwich laminates, though substantially superior in terms of static properties (flexural and interlaminar shear strength), appear to be more sensitive to the effect of impact damage if compared to intercalated hybrids. Hybrid intercalated laminates present the best combination of post-impact flexural strength and damage tolerance among the configurations investigated. Acoustic emission analysis helped in clarifying the damage mechanisms involved in the retention of strength and stiffness properties after impact events. Carbon laminates fail by multiple matrix cracks localized in the tensile side while in basalt laminates the interface failures seem to be the relevant energy dissipation mechanisms. Hybrid laminates with sandwich configuration fail by interface failures located in the basalt skins of the compression side that prevent the skins from hindering crack propagation in the carbon core. Damage partition among multiple internal basalt/ carbon interfaces is identified as the mechanism responsible for higher energy absorption and lower strength degradation of intercalated hybrid composites.
References _ BM, Karakuzu R. An experimental investigation of the [1] Aktasß M, Atas C, Içten impact response of composite laminates. Compos Struct 2009;87:307–13. [2] Wang J, Waas AM, Wang H. Experimental and numerical study on the lowvelocity impact behavior of foam-core sandwich panels. Compos Struct 2013;96:298–311. [3] Abrate S. Impact on laminated composite materials. Appl Mech Rev 1991;44:155. [4] Abrate S. Impact on laminated composites: recent advances. Appl Mech Rev 1994;47:517. [5] Cantwell WJ, Morton J. The impact resistance of composite materials – a review. Composites 1991;22:347–62. [6] Reis PNB, Ferreira JAM, Santos P, Richardson MOW, Santos JB. Impact response of Kevlar composites with filled epoxy matrix. Compos Struct 2012;94:3520–8. [7] Gustin J, Joneson A, Mahinfalah M, Stone J. Low velocity impact of combination Kevlar/carbon fiber sandwich composites. Compos Struct 2005;69:396–406. [8] Cheon SS, Lim TS, Lee DG. Impact energy absorption characteristics of glass fiber hybrid composites. Compos Struct 1999;46:267–78. [9] Naik N, Ramasimha R, Arya H, Prabhu S, ShamaRao N. Impact response and damage tolerance characteristics of glass–carbon/epoxy hybrid composite plates. Compos B Eng 2001;32:565–74. [10] Hosur MV, Adbullah M, Jeelani S. Studies on the low-velocity impact response of woven hybrid composites. Compos Struct 2005;67:253–62. [11] Sevkat E, Liaw B, Delale F, Raju BB. Drop-weight impact of plain-woven hybrid glass–graphite/toughened epoxy composites. Compos A Appl Sci Manuf 2009;40:1090–110.
219
[12] Sayer M, Bektasß NB, Demir E, Çalliog˘lu H. The effect of temperatures on hybrid composite laminates under impact loading. Compos B Eng 2012;43:2152–60. [13] Deak T, Czigany T. Chemical composition and mechanical properties of basalt and glass fibers: a comparison. Text Res J 2009;79:645–51. [14] Ross A. Basalt fibers: alternative to glass? Compos Technol 2006;12:44–8. [15] Dorigato A, Pegoretti A. Fatigue resistance of basalt fibers-reinforced laminates. J Compos Mater 2012;46:1773–85. [16] Mingchao Wang, Zuoguang Zhang, Yubin Li, Min Li, Zhijie Sun. Chemical durability and mechanical properties of alkali-proof basalt fiber and its reinforced epoxy composites. J Reinf Plast Compos 2008;27:393–407. [17] Liu Q, Shaw MT, Parnas RS, McDonnell A-M. Investigation of basalt fiber composite mechanical properties for applications in transportation. Polym Compos 2006;27:41–8. [18] Carmisciano S, De Rosa IM, Sarasini F, Tamburrano A, Valente M. Basalt woven fiber reinforced vinylester composites: flexural and electrical properties. Mater Des 2011;32:337–42. [19] Kim MT, Rhee KY, Park SJ, Hui D. Effects of silane-modified carbon nanotubes on flexural and fracture behaviors of carbon nanotube-modified epoxy/basalt composites. Compos B Eng 2012;43:2298–302. [20] Colombo C, Vergani L, Burman M. Static and fatigue characterisation of new basalt fibre reinforced composites. Compos Struct 2012;94:1165–74. [21] Manikandan V, Winowlin Jappes JT, Suresh Kumar SM, Amuthakkannan p. Investigation of the effect of surface modifications on the mechanical properties of basalt fibre reinforced polymer composites. Compos B Eng 2012;43:812–8. [22] Szabó J, Czigány T. Static fracture and failure behavior of aligned discontinuous mineral fiber reinforced polypropylene composites. Polym Test 2003;22:711–9. [23] Karger-Kocsis J, Zhang Z, Czigány T. Special manufacturing and characteristics of basalt fiber reinforced hybrid polypropylene composites: Mechanical properties and acoustic emission study. Compos Sci Technol 2006;66:3210–20. [24] Zhang Y, Yu C, Chu PK, Lv F, Zhang C, Ji J, et al. Mechanical and thermal properties of basalt fiber reinforced poly(butylene succinate) composites. Mater Chem Phys 2012;133:845–9. [25] Akinci A, Yilmaz S, Sen U. Wear behavior of basalt filled low density polyethylene composites. Appl Compos Mater 2011;19:499–511. [26] Lopresto V, Leone C, De Iorio I. Mechanical characterisation of basalt fibre reinforced plastic. Compos B Eng 2011;42:717–23. [27] Dehkordi MT, Nosraty H, Shokrieh MM, Minak G, Ghelli D. Low velocity impact properties of intra-ply hybrid composites based on basalt and nylon woven fabrics. Mater Des 2010;31:3835–44. [28] De Rosa IM, Marra F, Pulci G, Santulli C, Sarasini F, Tirillò J, et al. Post-impact mechanical characterisation of E-glass/basalt woven fabric interply hybrid laminates. eXPRESS. Polym Lett 2011;5:449–59. [29] Rosa IM, Marra F, Pulci G, Santulli C, Sarasini F, Tirillò J, et al. Post-impact mechanical characterisation of glass and basalt woven fabric laminates. Appl Compos Mater 2011;19:475–90. [30] Wang X, Hu B, Feng Y, Liang F, Mo J, Xiong J, et al. Low velocity impact properties of 3D woven basalt/aramid hybrid composites. Compos Sci Technol 2008;68:444–50. [31] Sarasini F, Tirillò J, Valente M, Valente T, Cioffi S, Iannace S, et al. Effect of basalt fiber hybridization on the impact behavior under low impact velocity of glass/ basalt woven fabric/epoxy resin composites. Compos A Appl Sci Manuf 2013;47:109–23. [32] Tehrani Dehkordi M, Nosraty H, Shokrieh MM, Minak G, Ghelli D. The influence of hybridization on impact damage behavior and residual compression strength of intraply basalt/nylon hybrid composites. Mater Des 2013;43:283–90. [33] Sarasini F, Tirillò J, Valente M, Ferrante L, Cioffi S, Iannace S, et al. Hybrid composites based on aramid and basalt woven fabrics: impact damage modes and residual flexural properties. Mater Des 2013;49:290–302. [34] Cao Shenghu, Zhis WU, Wang X. Tensile properties of CFRP and hybrid FRP composites at elevated temperatures. J Compos Mater 2009;43:315–30. [35] Wu Z, Wang X, Iwashita K, Sasaki T, Hamaguchi Y. Tensile fatigue behaviour of FRP and hybrid FRP sheets. Compos B Eng 2010;41:396–402. [36] Zhang CH, Zhang JB, Qu MC, Zhang JN. Toughness properties of basalt/carbon fiber hybrid composites. Adv Mater Res 2011;150–151:732–5. [37] Belingardi G, Vadori R. Low velocity impact tests of laminate glass-fiber-epoxy matrix composite material plates. Int J Impact Eng 2002;27:213–29. [38] Benevolenski OI, Karger-Kocsis J, Mieck K-P, Reubmann T. Instrumented perforation impact response of polypropylene composites with hybrid reinforcement flax/glass and flax/cellulose fibers. J Thermoplast Compos Mater 2000;13:481–96. [39] Belingardi G, Cavatorta MP, Salvatore Paolino D. Repeated impact response of hand lay-up and vacuum infusion thick glass reinforced laminates. Int J Impact Eng 2008;35:609–19. [40] Bledzki AK, Mamun A, Faruk O. Abaca fibre reinforced PP composites and comparison with jute and flax fibre PP composites. eXPRESS. Polym Lett 2007;1:755–62. [41] Huguet S, Godin N, Gaertner R, Salmon L, Villard D. Use of acoustic emission to identify damage modes in glass fibre reinforced polyester. Compos Sci Technol 2002;62:1433–44. [42] Barré S, Benzeggagh ML. On the use of acoustic emission to investigate damage mechanisms in glass-fibre-reinforced polypropylene. Compos Sci Technol 1994;52:369–76.
220
F. Sarasini et al. / Composites: Part B 59 (2014) 204–220
[43] Faudree M, Baer E, Hiltner A, Collister J. Characterization of damage and fracture processes in short fiber BMC composites by acoustic emission. J Compos Mater 1988;22:1170–95.
[44] Madhukar MS, Drzal LT. Fiber-matrix adhesion and its effect on composite mechanical properties. III. Longitudinal (0 ) compressive properties of graphite/epoxy composites. J Compos Mater 1992;26:310–33.