Mixed-mode fracture of sandwich composites: Performance improvement with multiwalled carbon nanotube sonicated resin

Abstract

An experimental investigation on sandwich composite materials composed of glass-fiber face sheet and polyvinyl-chloride foam core has been carried out. The research demonstrates improvement in mixed-mode delamination fracture toughness values of samples under mixed-mode bending condition. The improvement is recorded with addition of a certain percentage by weight of multiwalled carbon nanotubes in comparison to conventional samples. An easy and cost-effective methodology of multiwalled carbon nanotube insertion through sonication of epoxy resin followed by mixing with hardener and vacuum resin infusion technology for manufacturing of sandwich composites has been utilized in this study. The study also identifies the optimum weight percentage of multiwalled carbon nanotube addition in the resin system for maximum performance gain in mixed-mode fracture toughness. The results of observations in this study have been supported by field emission scanning electron microscope studies as well as high-resolution transmission electron microscope analysis.

Keywords

Foam glass-epoxy sandwich composite carbon nanotube mixed-mode fracture

Introduction

In spite of high strength to weight ratio and many other advantages of sandwich composite structures, interfacial face-core delamination remains as one of the major challenges for these types of structures. Sandwich composites are subjected to mixed-mode delamination failures in almost all practical loading situations due to significant differences in material properties and thickness of the materials used for the face sheet and the core. Interfacial face-core debonding is considered to be one of the major causes for loss of structural integrity and/or even catastrophic collapse of these structures. Definition of “interface” for a sandwich composite structure is an important point to be noted here. Since the resin infiltrates into the micropores of foam core and also in between the glass fibers, there is no distinct interface region as observed in a metal–metal bond. Typically, the interface region between glass fiber-reinforced plastics (GFRP) face sheet and the polyvinyl chloride (PVC) foam core is formed as a result of bonding between resin system of the face sheet and PVC foam core. Thus, the interface in this type of sandwich is a smeared layer of glass fiber strands, the resin system (used for GFRP face sheet) and some regions of PVC foam core. As a result, there is no distinct layer of resin system bonding and separating out the face sheet and core which may be separately termed as an interface region.

Research literature has been focused on improving the interfacial delamination resistance of these structures. Examples include use of shear key insertions [1,2], stitching technology [3 –6], z-pin fiber insertions [7 –12], peel stoppers [13–14] and also use of multiwalled carbon nanotube (MWCNT) in epoxy resin mix to promote better bonding [15 –37]. In this research, the effect of MWCNT in the epoxy resin mix has been investigated to determine performance improvement of the mixed-mode response of glass fiber face sheet and PVC foam core sandwich composites.

The use of MWCNTs for toughening resin systems has become a popular practice after the seminal paper by Iijima [17]. The use of several types of CNTs through different methodologies for improvement in fracture characteristics is typically limited to laminated composites in literature. In case of composite laminates, improvements of 60% and 17% in interlaminar fracture toughness have been achieved with aligned CNTs and sprayed MWCNT, respectively [22,23]. Another work on CFRP laminates recorded an improvement of 32% in interlaminar toughness by spraying MWCNTs on Teflon coated peel cloth and transferring them to woven prepregs [31]. In laminated composites, improvements of fracture toughness by 23% and 20% are reported with the use of 0.5 wt% and 0.1 wt% of functionalized CNTs and MWCNTs, respectively [30,29] in comparison to conventional cases. In hybrid nanocomposite, Karapappas et al. [35] recorded an increase of 60% in mode I fracture toughness with 1 wt% MWCNT and 75% increase in mode II fracture toughness (G_IIC) with 0.5 wt% MWCNT of resin in comparison to the neat epoxy reinforced with carbon fibers only. Garcia et al. [36] used vertically aligned CNTs (VACNTs) at the interface between the plies of carbon composite laminate and reported an improvement of 50% in mode I fracture toughness and as high as 200% in mode II fracture toughness values. In nanocomposites, an enhancement in mixed-mode fracture toughness approximately by 31% for mode ratio 1 with 0.5 wt% of MWCNT has also been reported [37]. The above set of literature shows different percentage increase in performance with addition of MWCNT in nanocomposites or laminated composites only. It should be pointed out here that the percentage gain in fracture performance of the samples depends largely on the type of resin system used, type of functional group of the MWCNT and also on the material characteristics of the constituent materials which are to be bonded with the resin system. In fact, MWCNTs not only toughens the resin system but also infiltrates within the adjacent constituent materials that are to be bonded and thereby improves mechanical performance.

Literature report use of several types of CNTs for improvement in fracture characteristics of nano, hybrid or laminated composites only. But to the best of the authors’ knowledge, there appears to be no work done on influence of MWCNTs on mixed-mode fracture toughness of glass-epoxy (G/E) and PVC core sandwich composites. This manuscript strives to investigate the influence of MWCNTs on mixed-mode fracture of G/E–PVC core sandwich composite along with determination of optimum percentage of MWCNTs. A sandwich composite differs significantly from laminated composites, since it consists of at least two different materials (having significant differences in material and geometrical properties). Thereby the characteristics of a bi-material interface (present in sandwich composites) cannot be directly extrapolated from results obtained from single-material interfaces (as present in laminated composites).

This manuscript demonstrates improvement in fracture performance of sandwich composite with different wt% of MWCNT addition on different mode mixities (measured by the ratio of mode II to mode I fracture toughness values) of mixed-mode bending (MMB) samples. In addition, the methodology used in this research work for MWCNTs addition in epoxy resin is much simpler than that used by other researchers for addition of MWCNT in resin mix for composite laminate manufacture.

Details about the process of manufacturing are presented in Process of manufacture section along with HRTEM investigations. Experimental investigation section describes the experimental investigations carried out with the manufactured samples; Results and discussions section discusses the results as obtained in experimental investigation (MMB test) and micrograph studies (FESEM) for samples with and without MWCNT; Conclusions section provides discussion and conclusion to the manuscript including possibilities of future work.

Process of manufacture

Epoxy resin (Araldite CY 230-1 IN of Huntsman Advanced Materials with 1300–1800 mPa s viscosity and 1.13–1.16 g/cc specific gravity at 25℃) and MWCNTs (Cheap Tubes Inc.) with outer diameter less than 8 nm, inside diameter 2–5 nm, individual length of 10–30 µm, specific surface area 500 m²/g, bulk density 0.27 g/cc) are used. The above mentioned resin and MWCNTs are thoroughly mixed with an ultrasonic liquid processor (OSCAR make SONAPROS PR- 500 MP processor with frequency (20 ± 3) kHz, ultrasonic power 500 W and 18 mm Ø Titanium horn) for 2.5 h intermittently (15 s on and 5 s off for the first hour, 10 s on and 5 s off in second hour and 5 s on 5 s off for the last half an hour) at 60% amplitude maintaining room temperature (25℃). The process of ultrasonication of the same epoxy resin has been repeated with 0.1 wt%, 0.2 wt%, 0.3 wt%, 0.4 wt% and 0.5 wt%, respectively. Hardener (Aradur HY 951 IN of Huntsman Advanced Materials with 10–20 mPa s viscosity and 0.97–0.99 g/cc specific gravity at 25℃) in 10 wt% of epoxy resin is mixed with the mixture after sonication by stirring at 150 for 15 min.

Dispersion of MWCNTs at different weight percentages of epoxy resin is studied through HRTEM analyses. Hardened mixtures of sonicated epoxy resin and hardener are trimmed with LEICA EM trim and cut to 100 nm thick samples using LEICA EMFCS Ultra-Microtome machine. These samples are then placed on copper grids and examined through JEOL JEM2100 HRTEM with 80 k magnification at an operating voltage of 200 kV. HRTEM images in Figure 1 for resin samples with MWCNTs (0.2–0.4 wt% of epoxy resin) can be observed to show relatively good dispersion; however, the optimum wt% of MWCNT for the best value of mixed-mode fracture toughness cannot be stated clearly from visual observations. Resin samples with MWCNTs for 0.1 wt% and 0.5 wt% of resin demonstrate low dispersion and agglomeration effects, respectively. An efficient dispersion or agglomeration of MWCNTs in resin is related to relative van-der Waals forces, curvature, and relative surface energy between MWCNTs and resin. In this research, ultrasonication is done to overcome the attractive forces, but MWCNTs may reagglomerate on discontinuation of external forces after sonication [21]. It should be noted in here that efficient load transfer between the constituent materials of the face sheet and the core is determined by the better nanotube–matrix interface properties and primarily influenced by the optimum dispersion of MWCNTs in the nanocomposites [22].

Figure 1.

HRTEM images of samples with different weight percentage of multiwalled CNT: (a) 0.1 wt% CNT, (b) 0.2 wt% CNT, (c) 0.3 wt% CNT, and (d) 0.4 wt% CNT, (e) 0.5 wt%.

Manufactured sandwich composite samples having MWCNTs in different percentages by weight of epoxy resin are tested in MMB setup for determination of optimum wt% of MWCNT in epoxy resin for enhancement of mixed-mode fracture toughness of the samples. Semi-rigid closed cell PVC foam with a thickness of 30 mm, cell size of approximately 400 µm, density 100 kg/m³ and trade name of Divinycell H100 manufactured by DIAB Inc. is used as the core material in this research work. The PVC foam core is covered with one layer of stitched combination mat followed by a layer of woven roving mat made from glass fibers on both sides. The assembly is placed inside an air tight vacuum bagging system (face sheet and core covered by flow media, porous Teflon sheet and highly permeable breather cloth). The resin system prepared through sonication of resin with MWCNTs and mixing with hardener is then allowed to pass through the preform to cast the sandwich composite panel by vacuum resin infusion (VRI) technology.

The input and output infusion lines are blocked after complete part wetting of the system which is then cured under vacuum for more than 48 h at room temperature (25℃). The process is repeated for manufacturing sandwich panels with different wt% of MWCNTs. Initial delamination in the sample in the form of pre-crack required for the delaminated MMB test is introduced in the samples with the help of 80 µm thick non-stick impermeable Teflon sheet between the face sheet and the PVC core during resin infusion process for sample preparation by VRI method.

Experimental investigation

Figure 2 shows a schematic diagram of the test setup along with sample dimensions. Samples of dimensions 190 mm × 35 mm × 35 mm (where length = 190 mm and width = 35 mm and thickness = 35 mm as per Figure 2) are cut from cured sandwich panels. The initial delamination length is kept as 50 mm. Samples are pasted with aluminum hinge tabs on its both top and bottom face sheets at initially delaminated part using DP 460 adhesive. Distance between the end point of glued part of hinge and that of delamination (a), as shown in Figure 2, is kept at 25 ± 1.5 mm. It should be noted that there exists previous literature [38 –42] on sandwich composites without MWCNTs to determine mixed-mode fracture toughness values with similar experimental setup (MMB apparatus).

Figure 2.

Schematic diagram of MMB test setup with sample.

Mixed-mode fracture toughness G_MMB of the MMB specimens is calculated using the formula provided by Quispitupa et al. [41]. MMB specimens of sandwich composite panels can be considered as superposition of cracked sandwich beam (CSB) and DCB specimens. Analytic expressions for the MMB compliance and energy release rate for symmetric sandwich specimens (h_f₁ = h_f₂ = h_f and E_f₁ = E_f₂ = E_f) have been derived based on load partitioning as

C = [\frac{c}{L} C 1 + \frac{c - L}{2 L} C 2] (\frac{c}{L} - α 1 \frac{c + L}{2 L}) + (\frac{c + L}{L})^{2} C CSB

(1)

where C is the MMB compliance, c is the lever arm distance, 2 L is the span length, α₁ is a parameter defined in Quispitupa et al. [41], a is the crack length, η is the elastic foundation modulus [41]. Whereas A, B, D are the extensional, coupling, and bending stiffnesses of DCB sample as defined in Quispitupa et al. [41]. C_CSB is the compliance of the cracked sandwich beam specimen [41], P is the applied load, b the beam width, D_d and D_i are the flexural stiffnesses of the debonded and intact region of the beam defined in Quispitupa et al. [41]. C₁ and C₂ are the compliances of the upper and lower legs of the DCB specimen defined in Quispitupa et al. [41]. The mode ratio G_II/G_I is given by

\frac{G II}{G I} = (\frac{P II a}{2 P I})^{2} \frac{[\frac{1}{D d} - \frac{1}{D i}]}{\frac{1}{h c G xz} + \frac{a^{2}}{(D - \frac{B^{2}}{A})} + \frac{12}{E f h_{f}^{3}} [a^{2} + 2 a η^{\frac{1}{4}} + η^{\frac{1}{2}}]}

(3)

where P_I and P_II are the loads on DCB and CSB specimens as defined in Quispitupa et al. [41].

A minimum of lever arm distance is required to avoid contact between crack faces is given by

c > \frac{\propto 1 L}{2 - \propto 1}

(4)

The condition represented in equation (4) is maintained in all the specimens under MMB tests to maintain the validity of the methodology applied in these experiments.

The aluminum hinge tabs are fitted with the attachments of the lever at top and that of the base at bottom of the steel base of the MMB setup as shown in Figure 3. The lever transfers compressive load to the sample at the upper surface of the top face sheet through a roller attached to the lever. The sandwich beam sample is supported on a clamp and a roller support at bottom which is attached to the base. The whole setup is placed on the loading base of Tinius Olsen make UTM. A C-clamp like attachment and a 10 kN load cell at a loading rate of 0.5 mm/min are then utilized to provide compressive load under different mode ratios on the lever through the wheels attached to the lever as shown in Figure 3. Samples are continuously loaded for propagation of crack by at least 20 mm from the end point of delamination or the point of crack initiation.

Figure 3.

Initially loaded MMB sample in UTM: (a) initial and (b) zoomed in region.

The results of the MMB tests are recorded as per recommendation of ASTM standard D6671/D6671M-13 [40] supported by visual observations. However, it should be noted in here that the case for this experiment is different from determination of mixed-mode fracture toughness in ASTM D6671/D6671M-13 [40]. In this experiment, the initial delamination present in the sample is oriented towards one end of the cross-section in between two dissimilar layers of face sheet and core, whereas ASTM D6671/D6671M-13 prescribes a non-adhesive insert serving as delamination initiator at the mid plane between two laminated layers of similar material properties and geometrical sections. The mean of the results of investigations on at least 10 samples (yielding the results between mean ±two times the standard deviation) is considered for any conclusion derived from the experiment. The MMB samples are subjected to an upward directed load at the left end of the debonded face sheet and a downward directed load at the center, thereby subjecting the specimen to both bending and tensile forces through lever action under different mode ratios. The initiation of crack after reaching the end point of delamination and crack propagation at the face/core interlayer occurs slowly with increase in load. The mean of load readings taken from either side of the sample is taken at the point of initiation of the crack and at other discrete points at each 2.5 mm distance from the end point of delamination with the help of camera attachments. Deflected shape of the sample after the experiment is shown in Figure 4. Except for the changes in stiffness, load capacity and ductility, between the samples with and without MWCNTs, no other significant difference in behavior is observed between them. The crack kinked into the core at a later stage of crack propagation in a large number of samples which agrees with the findings of a previous study [42] where it is stated that the direction of crack propagation depends upon the fracture toughness values of the associated components, i.e. the fracture toughness of the face sheet, interface and the core. The mixed-mode fracture toughness values calculated from the MMB tests of samples with initial delamination incorporate contribution from both mode I and mode II types of fractures.

Figure 4.

Final deflection of the MMB sample.

Results and discussions

Experimental investigations have been carried out for specimens with different percentages of MWCNT sonicated resins. But typical plots for load vs. displacements of sandwich specimens with MWCNTs (0.3 wt%) and without CNT are shown in Figure 5. The plot with 0.3 wt% MWCNT has been selected for representation since this percentage provided the best results with regard to performance improvements for mixed-mode fracture.

Figure 5.

Load–displacement plot of MMB Samples with different mode ratio: (a) mode ratio 1.0, (b) mode ratio 0.8, (c) mode ratio 0.5.

Plots for all the mode ratios show enhancements in peak load-carrying capacities of the specimens with 0.3 wt% MWCNT of resin over the samples without CNT (neat sample). The quantitative performance improvements are shown in Table 1. Even though there is slight increase in initial stiffness of samples with MWCNT (CNT samples) for all mode ratios indicating higher initial resistance to fracture, the increase is not that significant, and the quantitative estimate has not been evaluated for this portion. Typically, the pre-peak non-linear portions of the plots demonstrate formation of micro-cracks. Attainment of higher peak strengths and displacement at the peak strength for the MWCNT doped samples in comparison to the conventional samples demonstrate improved resistance offered by CNT samples. The improved resistance is induced through prevention of coalescence of the micro-cracks into a macro-crack. This resistance to coalescence of micro-cracks is typically initiated by the random dispersion of MWCNTs in the resin mix. As the mode ratio is decreased, the mode I component becomes more significant (for mode ratio 0.5 in comparison to that of 1), the drop in load immediately after attainment of peak-strength of the sample becomes sharper. However, for individual mode ratios, while comparing between samples with and without CNT, it can be observed that sharp fall in load after peak is reduced for CNT samples in comparison to the neat samples. The curve becomes much more flatter indicating that the random dispersion of MWCNT in the CNT samples effectively acts to prevent crack propagation in the sample. The displacement ductility of the sample (observed as displacement distance between attainments of peak strength to a drop of 20% from the peak strength) is also observed to increase for samples with MWCNT for all mode ratios in comparison to samples without CNT.

Table 1.

Increase in peak load-carrying capacity of samples with 0.3 wt% of resin with respect to the sample without CNT.

Mode ratio (G_II/G_I)	Peak load (P_ult) of samples without CNT (N)	Peak load (P_ult) of samples with 0.3 wt% CNT (N)	Increase in peak load-carrying capacity (%)
1	1282	1725	34.55
0.8	1269	1612	27.03
0.5	1201	1479	23.15

Even without addition of CNT, typically the shear strength is slightly higher than the normal strength. This fact is corroborated by increase in peak strength with increase in mode ratio in Table 1 (mentioned by Quispitupa et al. [42]) for sandwich composites. Since shear resistance offered by the resin mix system in between the glass fiber face sheet and the PVC foam is higher than the mode I resistance to crack, a better regain in strength along with increase in displacement ductility is observed with increase in mode ratio. The random dispersion of MWCNT in the samples is also observed to improve the performance of the samples with increase in mode ratios.

Random dispersion of CNTs in the sample and their modes of failure or pull-out could be observed from destructive tests on tested samples. The samples in which experimentations have been completed are cut into small pieces and then coated with gold palladium alloy using coater machine of Quorum series Q150 RES. These coated samples are observed under a Carl Zeiss MERLIN machine at extra high tension voltage of 5 kV for FESEM studies. Figure 6(a) shows random dispersion of MWCNT within the resin mix at the interface observed at 25 k magnification. Figure 6(b) shows small white circular patches which are practically torn fiber ends, observed at 150 k magnification. Figure 6(c) shows MWCNT bridging the resin system and glass fiber at 200 k magnification. Figure 6(d) shows bridging of MWCNT between two resin blocks individually attached to GFRP and foam respectively at 115 k magnification. Better mechanical interlocking mechanism developed by the MWCNT between the glass fibers, matrix and PVC core results in increase in resistance to mixed-mode fracture in MMB samples in comparison to the samples without MWCNTs.

Figure 6.

Micrographs of CNT dispersed in the resin system at the interface of sandwich composite: (a) CNT dispersed in resin system at interface region at 25 k magnification (b) numerous CNT pull-out zoomed in 150 k magnification, (c) Zoomed in CNT bridging and at 200 k magnification and (d) CNT bridging crack at 115 k magnification.

Figure 7 shows that for all % by weight additions of MWCNT (to the sample), the improvement in performance (in comparison to neat samples) increases till 0.3% after which it decreases. This trend of the plots demonstrates that the optimum percentage of MWCNT addition to G/E–PVC sandwich containing the specific type of resin system is 0.3 wt%. Figure 7 also shows error bars in experimental observations. It can also be observed from the figures that values of the maximum fracture toughness at 0.3 wt% MWCNT decreases with reduction in mode II components (or decrease in mixed-mode ratio values) which has also been previously demonstrated by Quispitupa et al. [42]. Primarily the reason for decrease in performance with increased wt% of CNT (> 0.3 wt%) is due to agglomeration effect which subsequently leads to localized stress concentration and zones of crack initiation. The figure also shows that over a wide range of mode ratios selected for this work, 0.3% gives the best percentage gain uniformly. However, it should be kept in mind that this 0.3% which gives the best gain in performance typically depends on the constituents of the sandwich construction as well as the epoxy resin system used in this process. It should be highlighted at this point that authors have reported an improvement of 34.34% in interface fracture toughness value of sandwich composite sample using MWCNT for predominantly mode I loading [24] with 1.5 wt.% addition of MWCNT. This significant difference in percentages of MWCNT (with that reported in this manuscript) is primarily because of the resin system used for the two works and also on type of loading applied. Moreover, the current article focuses on MMB showing different mode ratios of fracture, whereas the previous document focused on double cantilever beam experiments demonstrating primarily mode I behavior. This behavioral trend in results as observed in this manuscript (as well as the previous one [24]) with the addition of MWCNTs has also been previously demonstrated by Fiedler et al. [27] and Zhou el al. [28] for mode I stress intensity factor (K_IC) of carbon/epoxy composites mixed with MWCNT. Wichmann et al. [32] also reported 0.3 wt% as the upper limit of CNTs mixed in the resin for maximum percentage gain in nano-reinforced laminated composites with VARTM.

Figure 7.

Variation of mixed-mode fracture toughness of MMB sample with percentage of CNT sonication: (a) mode ratio 1.0, (b) mode ratio 0.8, (c) mode ratio 0.5 and (d) superposed plots for different mode ratios.

Figure 8 shows increment in mixed-mode fracture energy (G_MMB) with different mode ratios for different wt% of MWCNT specimen. Figure demonstrates that the increase in G_MMB values is highest for 0.3 wt% MWCNT, and least for 0.5 wt% of MWCNT. The justification of above fact can be sought from HRTEM and FESEM analyses. The larger volume of cylindrical resin system around MWCNTs provides reinforcement effect against pull-out mechanism [43]. Obviously, better dispersion of MWCNT within the resin system leads to better mechanical performance against interfacial delamination. It should also be noted that there is an optimum quantity and addition of MWCNT above the optimum quantity results in agglomeration and thereby do not lead to better performance.

Figure 8.

Increment in mixed-mode fracture energy (GMMB) with mode ratio for different wt% of MWCNT specimen.

Conclusions

The manuscript highlights the differences in experimentally observed response of interface fracture toughness of samples with and without MWCNT. Out of the different process of addition of MWCNT in composites (as can be obtained from literature), a simple method of sonicating the MWCNT in epoxy resin has been chosen in this research for the PVC-core/glass-fiber/epoxy-resin sandwich to demonstrate significant improvement in mechanical performance of samples against interfacial delamination subjected to mixed-mode loading condition. The research demonstrates an improvement of 34.55%, 27.03% and 23.15% in peak load-carrying capacity for samples with 0.3 wt% MWCNT (weight % being taken with respect to amount of epoxy resin used for sample fabrication) in comparison to samples without MWCNT for mode ratios 1, 0.8 and 0.5, respectively. Good dispersion of MWCNT is also observed with 0.3 wt% MWCNT addition in comparison to other percentages of addition of MWCNT by weight. Samples with 0.3 wt% MWCNT recorded an increase in fracture toughness values of 54.79%, 49.27% and 46.76% in comparison to samples without for mode ratio 1, 0.8 and 0.5, respectively. HRTEM images and FESEM analyses demonstrate good dispersion and fiber bridging of MWCNT in the epoxy resin system, respectively.

Footnotes

Acknowledgements

The authors acknowledge the contribution of Mr. Subhajit Das in manufacturing of the mixed-mode bending apparatus used in the experiment. The first author would like to thank many students and technicians who had helped him for carrying out the experiments. Any opinions, findings and conclusions or recommendations expressed in this manuscript are those of the writers and do not necessarily reflect those of the Space Research Organization, India.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by Indian Space Research Organization, India under Award No. IIT/KCSTC/Chair./New:Appr./13-14/70.

References

Mitra

. A methodology for improving shear performance of marine grade sandwich composites: sandwich composite panel with shear-key. Compos Struct 2010; 92: 1065–1072.

Mitra

Raja

. Improving delamination resistance capacity of sandwich composite columns with initial face/core debond. Compos Part B Eng 2012; 43: 1604–1612.

Mouritz

Leong

Herzberg

. A review of the effect of stitching on the inplane mechanical properties of fibre-reinforced polymer composites. Compos Part A Appl Sci Manuf 1997; 28: 971–991.

Shah-Khan

Mouritz

. Fatigue behavior of stitched GRP laminates. Compos Sci Technol 1998; 56: 695–701.

Kim

Lee

Park

et al.

Evaluation of durability and strength of stitched foam-cored sandwich structures. Compos Struct 1999; 47: 543–550.

Mouritz

. Ballistic impact and explosive blast resistance of stitched laminates. Compos Part B Eng 2001; 32: 431–439.

Freitas

Magee

Dardzinski

et al.

Fiber insertion process for improved damage tolerance in aircraft laminates. J Adv Mater 1994; 25: 36–43.

Palazotto

Gummadi

LNB

Vaidya

et al.

Low velocity impact damage characteristics of z-fiber reinforced sandwich panels – an experimental study. Compos Struct 1999; 43: 275–288.

Vaidya

Nelson

Sinn

et al.

Processing and high strain impact response of multifunctional sandwich composites. Compos Struct 2001; 52: 429–440.

10.

Cartie

Fleck

. The effect of pin reinforcement upon the through thickness compressive strength of foam-cored sandwich panels. Compos Sci Technol 2003; 63: 2401–2409.

11.

Mouritz

. Compression properties of z-pinned sandwich composites. J Mater Sci 2006; 41: 5771–5774.

12.

Mouritz

. Review of z-pinned composite laminates. Compos Part A Appl Sci Manuf 2007; 38: 2383–2397.

13.

Grenestedt

. Development of a new peel-stopper for sandwich structures. Compos Sci Technol 2001; 61: 1555–1559.

14.

Wonderly

Grenestedt

. Dynamic performance of a peel stopper for composite sandwich ship structures. J Compos Mater 2004; 38: 805–831.

15.

Siddiqui

Marom

et al.

Dispersion and functionalization of carbon nanotubes for polymer-based nanocomposites: a review. Compos Part A Appl Sci Manuf 2010; 41: 1345–1367.

16.

Qian

Hreenhalgh

Shaffer

MSP

et al.

Carbon nanotube-based hierarchical composites: a review. J Mater Chem 2010; 20: 4751–4762.

17.

Iijima

. Helical microtubules of graphitic carbon. Nature 1991; 354: 56–58.

18.

Ajayan

Tour

. Nanotube composites. Nature 2007; 447: 1066–1068.

19.

Thostenson

Chou

. Nanocomposites in context. Compos Sci Technol 2005; 65: 491–516.

20.

Lubineau

Rahaman

. A review of strategies for improving the degradation properties of laminated continuous-fiber/epoxy composites with carbon based nano reinforcements. Carbon 2012; 50: 2377–2395.

21.

Tang

Zhang

et al.

Interlaminar fracture toughness and CAI strength of fibre-reinforced composites with nanoparticles – a review. Compos Sci Technol 2013; 86: 26–37.

22.

Wicks

de Villoria

Wardle

. Interlaminar and intralaminar reinforcement of composite laminates with aligned carbon nanotubes. Compos Sci Technol 2010; 70: 20–28.

23.

Almuhammadi

Alfano

Yang

et al.

Analysis of interlaminar fracture toughness and damage mechanisms in composite laminates reinforced with sprayed multi-walled carbon nanotubes. Mater Des 2014; 53: 921–927.

24.

Patra

Mitra

. Interface fracture of sandwich composites: influence of MWCNT sonicated epoxy resin. Compos Sci Technol 2014; 101: 94–101.

25.

Godara

Mezzo

Luizia

et al.

Influence of carbon nanotube reinforcement on the processing and the mechanical behaviour of carbon fiber/epoxy composites. Carbon 2009; 47: 2914–2923.

26.

Fan

Santare

Advani

. Interlaminar shear strength of glass fiber reinforced epoxy composites enhanced with multi-walled carbon nanotubes. Compos Part A Appl Sci Manuf 2010; 39: 540–554.

27.

Fiedler

Gojny

Wichmann

MHG

et al.

Fundamental aspects of nano-reinforced composites. Compos Sci Technol 2006; 66: 3115–3125.

28.

Zhou

Pervin

Lewis

et al.

Fabrication and characterization of carbon/epoxy composites mixed with multi-walled carbon nanotubes. Mater Sci Eng A 2008; 475: 157–165.

29.

Mujika

Vargas

Ibarretxe

et al.

Influence of the modification with MWCNT on the interlaminar properties of long carbon fiber composites. Compos Part B 2012; 43: 1336–1340.

30.

Davis

Whelan

. An experimental study of interlaminar shear fracture toughness of a nanotube reinforced composite. Compos Part B 2011; 42: 105–116.

31.

Joshi

Dikshit

. Enhancing interlaminar fracture characteristics of woven CFRP prepreg composites through CNT dispersion. J Compos Mater 2012; 46: 665–675.

32.

Wichmann

Sumfleth

Gojny

et al.

Glass-fibre-reinforced composites with enhanced mechanical and electrical properties – benefits and limitations of a nanoparticle matrix. Eng Fract Mech 2006; 73: 2346–2359.

33.

Hamer

Leibovich

Green

et al.

Mode I and Mode II fracture energy of MWCNT reinforced nanofibrilmats interleaved carbon/epoxy laminates. Compos Sci Technol 2014; 90: 48–56.

34.

Seyhan

Tanoglu

Schulte

. Mode I and mode II fracture toughness of E-glass non-crimp fabric/carbon nanotube (CNT) modified polymer based composites. Eng Fract Mech 2008; 75: 5151–5162.

35.

Karapappas

Vavouliotis

Tsotra

et al.

Enhanced fracture properties of carbon reinforced composites by the addition of multi-wall carbon nanotubes. J Compos Mater 2009; 43: 977–985.

36.

Garcia

Wardle

Hart

. Joining prepreg composite interfaces with aligned carbon nanotubes. Compos Part A Appl Sci Manuf 2008; 39: 1065–1070.

37.

Ayatollahi

Shadlou

Shokrieh

. Mixed mode brittle fracture in epoxy/multi-walled carbon nanotube nanocomposites. Eng Fract Mech 2011; 78: 2620–2632.

38.

Reeder

Crews

Jr . Mixed-mode bending method for delamination testing. AIAA J 1990; 28: 1270–1276.

39.

Kinloch

Wang

Williams

et al.

The mixed-mode delamination of fibre-composite materials. Compos Sci Technol 1992; 47: 225–237.

40.

ASTM Standard D6671/D6671M − 13^ɛ1. Standard test method for mixed mode I-mode II interlaminar fracture toughness of unidirectional fiber reinforced polymer matrix composites¹, 2013.

41.

Quispitupa

Berggreen

Carlsson

. On the analysis of the mixed mode bending sandwich specimen for debond fracture characterization. Eng Fract Mech 2009; 76: 594–613.

42.

Quispitupa

Berggreen

Carlsson

. Design analysis of the mixed mode bending sandwich specimen. J Sand Str Mater 2010; 12: 253–272.

43.

Lachman

Wagner

. Correlation between interfacial molecular structure and mechanics in CNT/epoxy nano-composites. Compos Part A Appl Sci Manuf 2010; 41: 1093–1098.