Dynamic Performance of Particles Loaded Cross-Plied GFRP Composite

Fiber reinforced polymer composites (FRPs) are being increasingly used for a wide range of engineering applications owing to their high specific strength and stiffness. However, there through-the-thickness performance lacks some of the most demanding physical and mechanical property requirements for structural applications, such as aerospace vehicles and military components. Charpy impact test was done on end notched specimens of cross-plied glass fiber reinforced epoxy resin (GFRP) composites. Alumina and flyash particles were added in the epoxy resin with the weight ratio of 2.5%. The effect of notched length and fillers on dynamic performance of GFRP composite was obtained. Finally, SEM images were used to exp lain the changes in mechanical properties.


Introduction
Fiber reinforced po ly mer co mposites possess high specific moduli and specific strengths and are widely used in many structural applications including aerospace, sporting goods, automobile, civil and marine structures. While the inplane and fiber dominant properties make these composites useful in these applications, their through-the-thickness properties are limited by the relat ively poor properties of the matrix resin and the weak fiber-matrix interfacial bond. In order for FRPs to offer better choice for aerospace and military components over monolithic metallic structures possessing no delamination problems, significant imp rovements in the through-the-thickness properties are necessary.
It is well known that co mposite structures in the form of laminates are extremely susceptible to crack initiat ion and propagation along the laminar interfaces in various failu re modes. In fact, delamination is one of the most prevalent life-limiting crack growth modes in laminate composites as delamination may cause severe reductions in in-plane strength and stiffness, leading to catastrophic failure of the whole structure [1]. Delamination may be introduced by external loading as in static bending, compression or tension, in cyclic fatigue or by impacts of low-to-high energies, during manufacturing or in service. Impact loading during service is a common phenomenon for aerospace composite structures. There are situations like tool drops, runway debris, bird strikes, hailstorms and ballistic strikes, which induce considerable damage to the composite. In order to produce an impact resistant structure, it is important to understand the dynamics of the impact event and thus to predict the extent of the induced damage and estimate the residual properties so that the composites are designed with improved structural integrity and mechanical performance after impact [2][3]. Many useful techniques have been successfully devised to imp rove the delamination resistance in the past three decades [4][5][6][7]. In addition to the resistance to interlaminar fracture and impact damage, FRPs for advanced aerospace structures often require the most demanding mu lti-functional properties. The introduction of nanotechnology in the field of co mposite materials with nanoscale fillers, such as carbon nanotubes (CNTs) and carbon nanofibers (CNFs), has offered new opportunities to improve these mechan ical and mu ltifunctio nal properties of FRPs [8]. In light of their excellent Young's modulus and strength, extremely h igh aspect ratio, large surface area, and excellent thermal and electrical properties, these nanofillers can be incorporated into the FRPs to modify the properties of polymer matrix [9][10][11]. This paper aims to provide an overview of the enhancement of charpy impact energy and dynamic charpy impact toughness properties of GFRPs, especially the impact and delamination resistance, arising fro m the incorporation of alu mina and flyash fillers in the co mposites.

Materials and Sample Preparation
The composites were made fro m co mmercially available 90/ 0° Cross-ply E-Glass Fibre. The matrix material was Epo xy resin, which is procured fro m Resinova Chemie Ltd., Kanpur, India. The epo xy resin was contained Araldite (LY-554) and hardener (HY-951). They were mixed properly with the weight rat io 10:1.
The alu mina powder was of Sasol, Germany. The part icle size was in the range of 15-60 µm. The mixture has an alu mina percentage of 72-77%. The surface area is 170-250 m 2 /g. The crystalline size of alu mina powder is 4.5-10 n m.
Hand lay-up method was used for the fabrication of composite specimens. Fourteen layers of cross-plied E-glass-fiber sheets were composed with the epoxy resin to make a single large co mposite plate. The resin and hardener were taken in the rat io of 10:1 by weight. A constant load of 20 kg was p laced on the top of layer to control the weight ratio of fiber and epo xy resin, because excess amount of epoxy resin was squeezed out from the inside the fiber layers. The specimen was left for curing at amb ient conditions under the influence off applied weights for 24 hours. Later on post curing was done in furnace at 50-60℃ for 1 hr. Similarly, specimen 2 and 3 was fabricated with the introduction fillers in epo xy resin before the addition of hardener.
Three different composites were made. In itially, plain GFRP without filler material was prepared. Later, it was blended with 2.5% Fly-ash by weight of epo xy and with 2.5% A lu mina by weight of epo xy respectively (which is referred as specimen 1, 2 and 3 respectively).
The weight fraction of the glass in the composite was controlled through the application of weights during curing. However, burn-off method was used to measure the exact weight rat io of fiber and epo xy resin. First, co mposite specimen (10x10 mm) was weighted and placed in crucible. The crucible with co mposite template was kept in furnace at 250 O C for 5 hr. The epo xy resin was burn out and then the weight of fiber was measured. Finally, weight fraction of glass fiber and epoxy resin was obtained as 0.541 in specimen-1, 0.5327 in s p ec imen -2 and 0.4663 in specimen-3 respectively.
The composite plates obtained above were then cut into ENF specimen, wh ich dimension is shown in Figure 1.
For each specimen (i.e. 1, 2, and 3) three d ifferent notch lengths were made as shown in the Table 1. The in itial notch length was introduced by placing thick plastic sheet between glass ply 7 and 8 during the Hand-Layup procedure. The notch was later opened through a sharp knife-edge and mallet.

Charpy Impact Test
The impact test was performed with instrumented charpy equipment (Model; Resil Impactor-50, CEAST, S.p.A., Italy), as shown in Fig. 2. The impact-hamme r and vice lever with specimen adapter used were according to Charpy impact test. The impact length and impact velocity were 0.327m and 3.46m/s. Charpy test was performed on ten ENF samp les of each class of composites (type-1, 2, and 3). The average results were repo rted to obtain the impact energy and toughness of each sample. The impact energy required fo r each specimen was indicated by the mach ine and was reco rded. The fractured specimens are p resented in Fig . 3.

Scanning Electron Microscopy Test
The fractography study was used with the help of scanning electron microscopy (SEM), Philips-XL-20, wh ich is availab le in the Dept. of Physics, BHU. Fo r SEM observatio n on the fractured specimens, the black soot-like material was directly mounted to the sample holder with silver glue which is electrically conductive.  Fig. 4 shows the variation of impact energy with change in the initial crack length of different specimen made by third phase in the addition of to cross-plied E-glass fiber reinforc ed epoxy resin co mposites. The results clearly indicate that the variation in impact energy with the initial crack lengths 25.4, 43.1 and 50.8 mm respectively. Since, the specimen goes under delamination under the dynamic impact loading condition (Charpy imp act test) therefore the specimen with lowest init ial crack length would absorbed more energy for delamination, which is exp lained by the graph (as delamination occurred till the point of impact), i.e. specimen with lowest initial crack length 25.4 mm has absorbed more energy than any other samp le in both cases, while, the specimen with higher in itial crack length has absorbed lower energy than other samples [4].

Variation of Impact Energy
For the specimen-1 the crack didn't propagated for 25.4 mm and 17.4 mm in itial crack length. This shows that the min imu m init ial crack length of 25.4, 43.1 mm specimen is required. However, for specimen-2 and 3 the crack propagated for 25.4 mm in itial crack length. This shows an increase in delamination tendency of GFRP with fillers but with increase in absorbed impact energy. The results of sample of same co mposition and same in itial crack length showed that impact energy increases with change in crack length for every case. Hence, it can be concluded fro m the results that the impact energy absorbed by the specimen goes mostly in increasing the crack length [6]. Hence, energy main ly goes in delamination.

Variation of Dynamic Fracture Toughness
The dynamic fracture toughness can be obtained fro m the following relat ion, wh ich is given below K where, I, h and w are Impact Energy , breadth and width of the specimen. In all cases the value o f average dynamic fractu re toug hness of alumina samp les are mo re than the flyash/GFRP and GFRP samples. Th is indicates that alumina part icles increase the interface bond strength between epoxy resins, which increase the toughness of GFRP co mposites [7].

Microscopy Analysis
SEM images for GFRP sample shows matrix failure and de-bonding as visible in Figure 6. There is no fibre b reakage in the de-laminat ion area. The fib res are shown in micrograph by region A. The matrix left after de-bonding is shown by region B. There is considerable brittle failure of matrix as shown by region C. In Figure 7, minute crack flo w lines are seen around the de-bonded reg ion D because the energy is trans mitted through the matrix to the fibre resulting in de-bonding. De-bonding shows weak fibre-matrix interface [7,8]. SEM analysis of the Fly-ash specimen shows distribution of fly ash particles in the matrix, resin fly ash interface, g lass fibre distribution in the matrix, glass fibre matrix interface, deformation behaviour etc [9]. This appeared that the maximu m cracks are v isible in the flayash/epoxy resin rich area, as shown in Fig. 8.
It was observed that the fly-ash particles basically contain the metal o xide, as the particles are not spherical (Fig. 7-A). Also the mixing is rather heterogeneous, containing the lu mps of the fly-ash (Fig. 7-A). The fly-ash sites where the large clusters were there acted as the site of stress-concentra tion, resulting in the init iation of cracks (Fig. 7 -B) On the other hand, where the particles are well distributed and are small, and the area resisted the dislocation's movement.  . SEM images for fractured specimen of alumina/GFRP composite SEM images for alu mina/ GFRP co mposite samp le indic ate that there is no fibre failure in the de-lamination area [10]. There is significant de-bonding between fibre and matrix as shown by region A and B. The matrix underwent britt le failure as shown by C in Figure 9. The fracture toughness of sample improved considerably because the alu mina particles converted into nano particles lead ing to large increase in surface area. Th is increase in surface area resu lts in absorption of large amount of impact energy [11]. However, as visible from the SEM there was improper mixing of alu mina leading to clusters. These clusters acted as stressconcentrati on point shown by D. As cluster size decreases crack flow lines are vis ible as shown by E. Small voids are also visible.

Conclusions
The experimental results confirmed that the add ition of filler improved the toughness of GFRP co mposite under dynamic load ing. In the case of alu mina, dynamic fracture toughness, K dC, was mo re than the flyash filled GFRP co mposites, because, the shape and size o f f ille rs play a key role in changing the toughness of the matrix. A lso, micro graphs showed that cracks were arrested by the particles and most of the fibres fractured due to increase of interface bond strength.