1-s2.0-S0261306914001915-main

Materials and Design 59 (2014) 475–485

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier .com/locate /matdes

Investigation of the behaviour of a chopped strand mat/woven roving/foam-Klegecell composite lamination structure during Charpy testing

http://dx.doi.org/10.1016/j.matdes.2014.03.0050261-3069/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +60 3 89118411; fax: +60 3 89259659.E-mail addresses: [email protected] (A.M.T. Arifin), [email protected]

(S. Abdullah), [email protected] (Md. Rafiquzzaman), [email protected](R. Zulkifli), [email protected] (D.A. Wahab), [email protected] (A.K. Arifin).

A.M.T. Arifin a,b, S. Abdullah a,⇑, Md. Rafiquzzaman a, R. Zulkifli a, D.A. Wahab a, A.K. Arifin a

a Dept. of Mechanical and Materials Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysiab Dept. of Materials and Design Engineering, Faculty of Mechanical and Manufacturing Engineering, Universiti Tun Hussein Onn Malaysia, 86400 Batu Pahat, Johor, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history:Received 27 November 2013Accepted 3 March 2014Available online 21 March 2014

Keywords:CharpyChopped strand matFailurePolymer matrix compositeWoven roving

This paper presents the investigation of the characteristic behaviour of polymer matrix composites underCharpy impact conditions with different design configurations of the laminate structure. The aim of thisstudy was to evaluate the capability of different lamination designs for composite materials, in term ofcontact load, energy absorption, deflection and damage behaviour. In this study, laminated panels werefabricated using chopped strand mat (CSM), woven roving fabric (WR) and foam-PVC Klegecell as rein-forcement with a combination of epoxy or polyester resin, respectively. Structural panels of compositelaminates were produced using a hand lay-up technique. Each configuration design was impact testedto failure. Finite element analyses (FEA) were employed in this study to correlate the experimental valueof energy absorption with simulation results. The characteristics of different reinforcement types, matrixtype, hybrid type, architecture and orientation type were studied. These characteristics need to be con-sidered, due to their affecting the characteristic behaviour of the composite lamination structures. Basedon the results, it was found that differences in configuration design of the lamination structure of thepolymer matrix composites do influence the strength and weakness of the materials.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Polymer matrix composites are one of the three main groups incommon man-made composites, with the others being metal ma-trix composites and ceramic matrix composites. Polymer matrixcomposites, normally known as composite materials, consist oftwo or more materials that are combined to produce new materialsand to have better properties. Today, among existing materials,composites are among the best materials, due to their advantages,such as high strength and stiffness, low density (lightweight),resistance to corrosion, good electrical properties and ease of man-ufacturing. Composites have increasingly been used in variousapplications, for instance in automotive industries, maritimeindustries, aviation, equipment for sport, fishing and architecturestructures. There are two classified groups of composite materials:matrix and reinforcement [1]. As a core component, reinforcementprovides strength and is stronger than a matrix. Meanwhile, a ma-trix is a material that is used to keep the fibre (reinforcement) ori-entation and to provide protection from ambient conditions. It is

important to study impact phenomena in composite materialswhile keeping in mind that different materials are used and thatthe structures are unique and usually in a laminated form. Mostof the applications that use composite materials are exposed to im-pact conditions. Examples include marine and aerospace vehiclesand ballistic protection in the defense industry. Therefore, it isimportant to study the behaviour of composite materials that aresubjected to impact conditions. Most of the previous work on im-pact behaviour of composite materials involved evaluating the im-pact response during experimental failure testing of compositematerials, and studying the effect of the material used and ofimprovements in the composite structure.

Due to the low velocity of impact in the Charpy test, the resultsare strongly dependent on configuration, fibre type, matrix type(resin), thickness of the sample, loading velocity and type of pro-jectile. Furthermore, low velocity impact on composite materialsis very important to investigate because internal fracture can de-crease the effectiveness of materials, without any obvious damageat the impacted surface [2]. For this reason, as reported by Evci andGulgec [3], the researcher still has considerable work to under-stand the exact relationship between impact force and damagemechanism. In other words, damage initiation and propagationare dependent on both impact properties and the material’s re-sponse. The impact properties include impact force, velocity and

476 A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485

energy, and the material’s impact response depends on materialstrength, deflection of force, duration of impact and energy dissipa-tion [4]. The impactor of the Charpy test also influences the dam-age characteristics of the sample in a manner related to thedesign and mass of the impactor [5].

Ghasemnejab et al. [6] mention that the natural characteristic ofcomposite materials is to be brittle. Therefore, the energy absorp-tion under impact is linked to other mechanisms, such as fibrebreakage, matrix cracks, debonding of fibre–matrix interface anddelamination of plies. In composite material structures, if delami-nation of plies occurs between each layer, this means that a pro-gressive failure mode has occurred, and the composite shows acapability for energy absorption. According to Sohn and Hu [7],in their investigation of delamination mechanisms and energy dis-sipation of carbon fibre epoxy composites, the failure mode wasseparated between mode-I and mode-II delamination when thecomposites were tested in two extreme conditions. Shyr and Pan[8] found that the first layer in composite lamination is a keyparameter for dissipation of energy in the structure, as reportedin his study of impact behaviour and damage characteristics in dif-ferent fabric structures with various laminate thicknesses. Chanet al. [9] identified that the efficiency works to improve energyabsorption in composite lamination structures through the thick-ness direction and can indirectly control delamination of the struc-tures. Ulven and Vaidya [10] noted that the impact response inpolymer matrix composite laminates can be divided into threestages: damage initiation, penetration and total perforation. Theyalso mentioned that the contact response of the PMC and sandwichtypes under impact conditions is different because of sandwichstructures having a core material that significantly increases therigidity of the composite.

Hristov et al. [11] investigated the impact behaviour of modifiedpolypropylene/wood fibre composites, and found that the behav-iour of composite materials is different, due to differences in mate-rial used, even for a small modification of the material content. Fuet al. [12] investigated the fracture resistance of short-glass-fibrereinforced and short-carbon fibre reinforced composites, and re-ported that the impact energy of composite materials depends onthe fibre length. In optical systems, Zang and Zhang [13] identifiedthat the fibre length are strongly affect the width of the hysteresisloop and threshold switching power, whereby the polymer matrixcomposites has also been employed in the fibre grating to improvethe operation of optical switching. In the field of carbon fibre rein-forced composites, Choi and Chang [14] mentioned impact failureof the structure from matrix cracking, and this can trigger delam-ination at others ply interfaces before fracture. Kwon and Wojcik[15] found that the failure load of the composite material increaseswith a presence of small lamination cracks compared with non-delaminated structures. Pegoretti et al. [16] analysed the fracturebehaviour in an epoxy carbon laminate system using a correlationof interlaminar fracture toughness and impact energy absorption.Erkendirci [17] concluded in his work using plain weave S-2glass/HDPE thermoplastic composite that impact energy of thecomposite structure increases upon increasing the thickness andvolume fraction of the composite.

Available literature indicates that the characteristics of compos-ite materials in terms of Charpy impact behaviour have been inves-tigated by a large number of researchers [2–16]. However,numerical analysis combined with experimental investigation onCharpy impact behaviour with different design configurationshas not received much attention. Therefore, the aim of this studyis to fabricate chopped strand mat/woven roving/foam-Klegecellcomposite lamination structures with different design configura-tions, and investigate the behaviour of these composites bothexperimentally and numerically. It is believed that knowledge onthe Charpy impact behaviour of these composites would have an

essential role for many structural applications, such as marine boatstructures. Laminated panels are fabricated using CSM, WR andfoam-PVC Klegecell with a combination of epoxy or polyester resinfor reinforcement. In this study, six groups of design configurationsare used, Types-A–F. Structural panels of the composite laminatesare produced using a hand lay-up technique. Each configurationdesign is impact tested to failure. Finite element analysis (FEA) isthen employed to correlate the experimental values of energyabsorption with the simulation results.

Due to the brittle behaviour, which is one of the weaknesses ofcomposite materials, this investigation has been carried out. Thisdisadvantage causes problems for composite structures that expe-rience collision. The expected outcome from this study was to eval-uate the behaviour of different lamination structures under variousCharpy impact conditions, particularly in terms of load, energyabsorption and deflection behaviour with respect to the use of dif-ferent materials. Hence, it is important to identify resistance tofailure of composite materials with different design configurations,especially at sudden applied loads.

2. Theoretical background

2.1. Charpy testing

The Charpy impact test is commonly used to evaluate the im-pact energy and toughness of materials, and usually used in qualitycontrol processes, whereby it is one of the economical tests [18].Impact energy is defined as the energy that required to fracture astandard test specimen under impact loading. In the Charpy impacttest, the energy absorption of the specimen is determined from thechange in the height of a pendulum before and after the impact[19]. When the pendulum strikes the specimen, the specimenabsorbes the energy until it yields and it begins to undergo plasticdeformation. As reported by Nita et al. [18], at that condition, thespecimen continues to absorb the energy. Fracture occurs at thepoint when it can no longer absorb any more energy.

According to Ali et al. [19] the impact velocity of the pendulumwhen it strikes at the specimen is given by the Eq. (1), respectively.

v ¼ffiffiffiffiffiffiffiffiffiffiffi2gH1

pð1Þ

where g is the acceleration due to gravity and H is the change of ele-vation in the centre of the strike. The energy absorbed when thespecimen is fractured, which is energy lost by the pendulum is gi-ven by, as shown in Eq. (2).

U ¼ mgðh1 � h3Þ ¼ mgrðcos b� cos aÞ ð2Þ

where m is the mass of the pendulum, and h1 and h2 are elevationsof the mass centre, as shown in Fig. 1.

3. Methodology

3.1. Materials

CSM 450, woven roving WR 300 and foam-Klegecell were usedas received from the supplier as reinforcement. The matrix wasepoxy resin (ADR 246 TX) cured with Hardener ADH 160 and poly-ester resin (Polymal VE-P310P) with Methyl Ethyl Ketone Peroxide(MEPOXE) supplied by a local supplier. The hardener was used as acuring agent and to improve the interfacial adhesion and impactstrength of the composites. The materials used in this investigationare based on the same material used in a Malaysian-based marineboat [20]. To obtain the optimum matrix composition, a resin andhardener mixture of 5:1 was used. The properties of each materialemployed in the investigation are shown in Table 1, and a flow-chart of the experimental and FEA procedure is shown in Fig. 2.

Fig. 1. Illustration of Charpy impact absorbed theory [18].

Table 1Mechanical properties of reinforcement and matrix.

Material Young modulusE (GPa)

Poisson’sratio

Density(kg/m3)

Chopped strand mat 75 0.20 2540Woven roving fabric 76 0.37 2551Foam-PVC Klegecell 0.3 0.32 580Epoxy resin 2.7 0.4 1200Polyester resin 3.5 0.25 1161

Specimen Preparation namely as; Type-A, Type-B, Type-C, Type-D,

Type-E and Type-F

Preparation of Experimental

Charpy Test

Data processing

Finite Element Analysis

Modelling of materials

FEA Test (Finite Element

Analysis)

Data processing

Comparison and Validation

Result

Finish

Fig. 2. Flowchart of experimental and FEA procedure.

A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485 477

3.2. Specimen fabrication

Polymer matrix composites were fabricated using a hand lay-uptechnique. In this study, each specimen has a different configura-tion of the lamination structure, consisting of chopped strandmat (CSM 450), woven roving fabric (WR 300) and foam-Klegecell.During the fabrication process, the pattern of CSM 450, WR 300 orfoam-Klegecell was separately impregnated with epoxy or polyes-ter resin depending on the configuration of the structure. After thematerial was laid down at the mould, a controlled quantity of resinwas applied to the surface of the material. Brush or spiral rollerswere used to remove any voids in the fibre structure and to spreadthe resin evenly throughout the fibres. The process was repeateduntil the required configuration was built up. After that, the com-posite laminate specimen was cured at room temperature beforethe test specimen was cut out according to European StandardEN ISO 179-1 [21]. Fig. 3 shows the process flow for the sample fab-rication in stages, and schematic views of the laminate configura-tion for each specimen are shown in Fig. 4.

3.3. Charpy impact test

In this study, different laminate structures were fabricated andspecimens were produced to determine the impact properties ofthe composite, such as load, deflection and energy absorption.The recommendations of EN ISO 179-1 [21] are followed for theimpact test conditions in this research. A Zwick/Roell Charpy TestRig Instrument was used for investigation of impact damage resis-tance of the composite laminate structures, as shown in Fig. 5. Themass of the pendulum is 2.0 kg with a swing length of 390 mm andimpact speed of 3.85 m/s. At least five specimens were tested foreach type of specimen, with a length of 80 mm and width of10 mm. The Charpy impact test is a dynamic three point bending

test. The experimental setup consists of a specimen that is freelysupported and a pendulum that is attached to the machine usinga pinned rotating arm. The pendulum of the Charpy instrumentfalls in a circular trajectory and hits the specimen in the middleof the span length and transfers its kinetic energy to the specimen.

3.4. Numerical modelling of Charpy condition

3.4.1. Finite element analysis (FEA)Commercial finite element analysis software was employed in

this study to correlate the experimental results with simulation.For the simulation, 3D modelling was used to design impact condi-tions similar to the actual experimental conditions, as shown inFig. 6. An pendulum, supporter and specimens were simulated. Inthis study, the value of energy absorption under experimental con-ditions was studied in comparison with simulation results.

3.4.2. Material modelFinite element analysis (FEA) using commercial software was

used for investigation of the damage due to impact in the compos-ite material structures. In this process, 3D modelling was used todesign the situation of impact based on the experimental condi-tions, as shown in Fig. 7. The types of design were deformableand solid shape. The material data used in FEA (finite element anal-ysis) were obtained from monotonic experimental values, based ondifferentiation of the composite structure. The models were as-signed as homogeneous sections in dynamic conditions with expli-cit elements.

3.4.3. Simulation conditionsThe finite element model of the Charpy impact test device con-

sists of a pendulum, support and specimens. An initial velocity ofthe pendulum was assigned a value of 3.85 m/s, whereas the spec-imen was fully restrained at the beam support area. An eight-nodeslinear brick element was used in this model. Surface-to-surfacecontacts (explicit) were utilised at the contacting interfaces.Fig. 8 shows the boundary conditions at the support area for sim-ulation using an encastre type, and the pendulum used a velocity/angular velocity type as the selected step. In mesh condition, a hexshaped element was used with linear geometry, as shown in Fig. 9.

4. Results and discussion

Due to growing interest in the use of composite materials forvarious applications, it is important to investigate the response

Illustration of Hand Lay-up Technique

Finishing Process

Specimen Lamination

Specimen Grinding Process

Catalyst

Brush Roller

Spiral Roller

Platform Scale

Scissors

Resin

Specimen Preparation

Woven Roving (WR)

Chopped Strand Mat (CSM)

Foam-Klegecell

Spiral Roller

Mould

Brush Roller

Resin

CSM/WR or Foam-Klegecell

Fig. 3. Illustration of the steps in fabricating the specimen made from polymer matrix composites.

478 A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485

to impact loading of composite materials. This is because impactphenomena can occur during operations, maintenance or manufac-ture of the composite structures. In polymer matrix composites,there are several factors that affect the impact damage to the com-posite materials, such as fibre and matrix type, stacking sequenceof the structures, orientation of reinforcement, structure thicknessand velocity of impact. These are among the important factors thatshould be considered during impact studies.

Therefore, this study is divided into several parts investigatingthe effects of fibre type, matrix type, hybrid structure type andarchitecture of stacking sequence/orientation type. The followingsections discuss the results based on the type of lamination struc-tures that have been tested, as shown in Table 2. Six types of com-posite lamination structures were used to demonstrate thedifferences in characteristic behaviour between composite struc-tures, for contact load, energy absorption and deflection of thestructures. Furthermore, the entire behaviour of the structure isexamined, to understand the damage process in composite struc-tures containing a combination of different materials. However, itshould be noted that polymer matrix composites are anisotropicmaterials and it is more difficult to analyse the evolution of dam-age processes in such composites compared with other materials.Therefore, to understand the characterisations of polymer matrix

composites under impact conditions, the findings of the experi-mental work conducted are discussed and summarised below.

4.1. Effect of reinforcement type

In this experiment, two types of fibre and foam were used asreinforcement in the composite structures, which are composedof chopped strand mat (CSM), woven roving (WR) and foam-PVCKlegecell. The experimental results demonstrate that the strengthof each composite structure is different, due to differences in thereinforcement used. Fig. 10 shows the filtered load–time graphfor the two types of specimens with different lamination struc-tures. A direct comparison of energy peak loads for both structuresshows that the specimen of Type-C1 has a higher peak load com-pared with Type-F1, approaching 300 N, while Type-F1 is 262 N.One of the factors that causes lower load levels on the Type-F1

specimen is the weaknesses of resistance behaviour in foam-Klege-cell material. This behaviour is also due to delamination betweenlamination layers of the structures. The delamination is influencedby the process used to manufacture the composite materials. Thisis because the manufacturing process impacts the quality of thecomposite. As reported by Fu et al. [12], the composite impact en-ergy also depends on the fibre length, either continuous or short

Legend;

Fig. 4. Schematic views of the laminate configurations.

(a) (b)Fig. 5. Experimental setup (a) Charpy impact test instrument, (b) specimen position on the machine for the impact test.

A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485 479

and the process conditions. Additionally, the differences in damagephenomena between the two specimens are investigated.

The same graph also shows the difference in impact time, frac-ture behaviour and failure time of the structures. Differences of im-pact time and failure times occur in the structures due todifferences in thickness, which is 5 mm for Type-C1 and 23 mmfor Type-F1. Fracture of the Type-C1 specimen is faster than theG-F1 specimen, with an estimated time of around 2 s. This resultshows that the strength of the composite laminate structures isnot dependent on the thickness of the materials but rather, itsdependence is on the type of reinforcement used. As suggestedby Rashkovan and Korabel [22] the fibre (reinforcement) strength

can be increased through a surface treatment technique, to havebetter interface behaviour between matrix and fibre. Based oncomparison of G-C1 and G-F1 specimens, the selection of reinforce-ment type is very important for composite lamination structures,to ensure a strong structure and to help reduce the vulnerabilityof the material structures.

4.2. Effect of matrix type

Fig. 11 shows results from Type-E1 and Type-C2 laminationstructures for investigating the effects of matrix type. Both of thestructures have the same number of layers but with a different

Fig. 6. Illustration of Charpy test modelling with specimen in FEA.

Specimen

Impactor

Supports

Fig. 7. Finite element models for Charpy test conditions with specimen.

Boundary Condition

-Encastre Type

Velocity/Angular Velocity

Fig. 8. Boundary condition areas in the FE model.

480 A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485

matrix used, which is epoxy or polyester resin. The graph showsthat the peak load of specimen Type-E1 is higher than that ofType-C2 at 350 N and 290 N. The result shows that the epoxy resinhas a higher capability to withstand the impact compared withpolyester resin. In addition, the impact time of the epoxy compos-ite lamination is approximately 1 ms faster than the polyester

Fig. 9. FE models, before (a) and

composite lamination before the structure fails. As reported byRassmann et al. in his study of resin system effectiveness, the poly-ester laminates showed good impact properties and modulus;however, the epoxy laminates displayed good strength values[23]. The matrix effect is applies to both thermoset polymers aswell as thermoplastic polymers. Hristov et al. state that the modi-fication of polypropylene/wood fibre composites with other cou-pling agents also influences the fracture behaviour of materialsdue to differences of material properties [11].

4.3. Effect of hybrid type

The tested specimens with different lamination structures,Type-F2 and Type-D1, are shown in Fig. 12. The lamination struc-tures for specimen Type–F2 consist of foam-Klegecell at the middlewith a combination of epoxy and polyester resin matrix, whereasspecimen Type-D1 has nine layers with a combination of choppedstrand mat, woven roving and polyester resin as matrix. The resultsshow that the load values for specimen Type-D1 are higher than forspecimen Type-F2. This phenomenon indicates that the specimenof Type-D1 has a higher strength compared with Type-F2. This isdue to differences in material in the lamination structure. A mate-rial such as foam-Klegecell has advantages for certain applications,such as being lightweight, cost effective and also enabling in-creased material thickness. Normally, the hybrid structure is usedfor applications that require lightweight structure and averagestrength. Han et al. report that the capability of the hybrid systemis higher in terms of energy absorption compared with a non-hy-brid system through his research on hybrid pultruded responseto axial crushing [24]. Additionally, the positive effect of the hybridstructure is to stiffen the composite structure and to freeze the de-formed structure after the impact, as mentioned by Hufenbachet al. [25].

As shown in Fig. 13, the combination of Figs. 10–12, differenttest specimens exhibit different impact loading behaviour, espe-cially in terms of fracture behaviour. As previously discussed, thisis because different materials and lamination structures have dif-ferent capabilities to absorb energy in Charpy impact conditions.Based on the Fig. 13, it clearly shows the direction of mechanicalfailure for different of specimens before sudden fracture.

4.4. Effect of architecture and orientation type

Next, the effect of architecture and orientation type of the com-posite lamination on load values and deflection are shown in Ta-ble 3. The results indicate that the lowest value for the loadcompared with the other specimens is for Type-B. These specimensconsist of three layers with ±45� angles of woven roving. Based oncomparison with Type-A, the load of Type-B is lower even though ithas the same number of layers, but with different angles of

after (b) meshing process.

Table 2Groups and characteristics of composite lamination structures.

Type Type of resin Fibre–woven woven roving (WR) direction Type of reinforcement Fibre volume fraction Vf (%) Material combination (g/m2)

A1–5 Polyester 0�/90� CSM/WR 644.2 2 Layers 450 CSM1 Layers 200 WR

B1–5 Polyester ±45� CSM/WR 644.1 2 Layers 450 CSM1 Layers 200 WR

C1–5 Polyester 0�/90� CSM/WR 644.3 3 Layers 450 CSM2 Layers 200 WR

D1–5 Polyester 0�/90� CSM/WR 643.5 3 Layers 450 CSM6 Layers 200 WR

E1–5 Epoxy 0�/90� CSM/WR 644.4 3 Layers 450 CSM2 Layers 200 WR

F1–5 Polyester/Epoxy 0�/90� CSM/WR /foam-PVC Klegecell 644.3 2 Layers 450 CSM2 Layers 200 WR

(X) 1–5 = Type-(X), specimen 1 (S1), specimen 2 (S2), specimen 3 (S3), specimen 4 (S4) and specimen 5 (S5).

Fig. 10. Tested specimen with differentiation of lamination structures (a) specimen Type-C1 (b) specimen Type-F1.

Fig. 11. Tested specimen with different of lamination structures (a) specimen Type-E1 (b) specimen Type-C2.

A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485 481

orientation of the woven roving, which are 0�, 45� and 90� angles.As reported in [26], the strength of composite materials with dif-ferent directions of fibre shows a different trend of behaviour,especially for the loads. However, in terms of deflection the resultsdo not indicate a significant difference between the two specimens.

Comparing Type-C and Type-D, with the same material beingused but with different numbers of layers, the resulting value ofthe loads is different. Type-D contains nine layers in the lamination

structure compared with five layers in Type-C. Based on the re-sults, with increasing number of lamination layers, the load ofthe structure is increased. Contrary to this result, the deflectionof Type-D is lower than Type-C. Ku et al. report that the propertiesof composite materials improve with the addition of a fibre in thecomposite structure or by increasing the lamination layers [27].The deflection criterion between Type-E and Type-F does not showa significant difference between the other types. In this case, even

Fig. 12. Tested specimen with different lamination structures (a) specimen Type-F2 (b) specimen Type-D1.

0 1 2 3 4 5 6 7 8 9 100

100

200

300

400

500

600

Time, ms

Filt

ered

Loa

d, N

Type-D1

Type-F2

Type-E1

Type-C2

Type-C1Type-F1

Fig. 13. Direction behaviour of mechanical failure with different of specimens.

Table 3Effect of the architecture and orientation type of composite lamination based on theload and deflection values.

Type Mean load (N) Deflection (d)

A 153.13 8.87B 130.13 8.88C 291.61 7.31D 608.42 3.44E 307.55 4.75F 260.52 4.39

Type of Group

Ene

rgy

Abs

orpt

ion,

JType-A

Type-B

Type-C

Type-D

Type-E

Type-F

Fig. 14. Impact energy absorption of different types of composite structure.

482 A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485

though there is a difference in material and lamination structure,the values of deflection under experimental conditions are almostsimilar. However, in terms of the capability to provide resistance toa load, Type-F is weaker than Type-E. This is due to the soft mate-rial at the middle of Type-F, which is foam-Klegecell. Furthermore,based on the overall results of architecture and orientation, anddifferent types of composite structure, there is a significant corre-lation between load and deflection of the structure. This phenom-enon is proved by Type-D, with a high average load number andthe lowest value for deflection.

4.5. Energy absorption

In this study, an average of impact energy absorption for eachspecimen type, for different types of composite structure, was

investigated with respect to the different materials and resinsused, as shown in Fig. 14. Based on the results, Type-C exhibitedbetter energy absorbing capacity than the other composite struc-ture systems, approaching 1.3 J. In contrast, Type-B has the lowestenergy absorption. A comparison of Type-B and Type-C, shows adifference in energy absorption characteristics, with a reductionof around 51% due to differences in lamination structure. The ori-entation angles also affected the capability of the structure to ab-sorb impact energy. Thus, the total energy absorption of Type-Aand Type-B is different, due to a difference of orientation anglesbut with the same number of lamination layers. The matrix usedin processing of the composite structure also plays an importantrole and needs to be examined. By using polyester resin as a ma-trix, the capability to absorb impact energy is better than epoxy re-sin that is represented by Type-E. Due to the better adhesionbetween fibres and matrix in epoxy laminates, the absorption ofenergy is less than in polyester laminate structures. This provesthat energy propagation in the epoxy laminate structures isslightly lower than in the polyester laminate structure [23]. How-ever, the laminated structures for Type-D and Type-F show bettercombinations, because of higher energy absorption among thespecimen types, in the range of 0.8–1.2 J.

4.6. Experimental via simulation

Comparison of experimental and simulation study of energyabsorption demonstrates similar behaviour under impact

Fig. 15. Simulation condition of Charpy impact test with different step time.

A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485 483

conditions. In a simulation study, properties such as elasticity anddensity of the Type-C specimen in were considered in the model-ling process. In a modelling procedure, the parameter to evaluatethe energy absorption of the structure includes energy absorption(EA), specific energy absorption (SEA), mean impact force and peakimpact force. Energy absorption is the total strain energy absorbedduring the deformation due to the impact force. It is defined belowas:

EAðdÞ ¼Z d

0FðxÞdx ð3Þ

where d is the effective stroke length, and (x) is the length of thestructure of energy absorption. SEA is the ratio of absorbed energyto the structure mass Mt and is given by the formula:

SEA ¼ EA=Mt ð4Þ

Fig. 15 shows the simulation conditions for Charpy impact withvarying step time. The figure shows the difference of specimenbehaviour, before and after impact. It shows that in the condition(c), the stress of the specimen is high, when the impact occurs.Additionally, Fig. 16 shows the validation of energy absorption be-tween simulation and experimental results of the impact test.

Based on the simulation conducted, the Type-C design configu-ration is chosen to represent the outcome of the modelling processdue to it having the highest of energy absorption as compared toothers types. Based on the result obtained, as shown in Fig. 16,the energy absorption of the simulation is 1.7 J. On the other hand,experimental study of the specimen gave a different of energyabsorption result, approaching 1.3 J. It is essential to note that in

Fig. 16. Validation of energy absorption between simu

this simulation study, the composite plate had been assigned asan isotropic type of material in the input of FEA which has thesame elastic properties in all directions. For the numerical analysis,the composite structure was considered as homogeneous materi-als. The inhomogeneous effect of the real microstructure of com-posite structure may be the cause of the difference inexperimental and numerical results. However, the predicted re-sults based on the simulation study were found to be in reasonableagreement with experimental observations. This study is an initialwork for simulation, in order to determine the suitability of theAbaqus FEA software for verification and validation works, in thiscase for Charpy impact test conditions.

4.7. Damage behaviour in impact condition

Fig. 17 shows the damage behaviour for each of the specimensbased on the groups. Each group also has a close-up picture toshow the difference of fracture behaviour for different design con-figurations of the laminate structure. The difference in laminatestructure can affect the character of the composite materials’behaviour, as explained earlier in this paper. Composite materialscan exhibit multiple types of damage before failure, such as matrixcracking, fibre fracture, fibre pull out, fibre rupture and fibre deb-onding [28–30]. Generally, the failure of fibres can lead to failureof the whole structure. Based on observation of the damage behav-iour, the overall damage under impact is due to fibre rupture. How-ever, with different types of lamination structures, damage canalso result in fibre pull-out, delamination and fibre debonding.

lation and experimental results of impact testing.

Fig. 17. Observation of damage behaviour for each of the specimens based on the group.

484 A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485

5. Conclusions

In this study, composite specimens were made using the handlay-up technique and tested to evaluate the effects of the differ-ences in design configurations of the lamination structures. The re-sults obtained showed different impact behaviours between thespecimens, with each specimen having advantages and weak-nesses based on the respective design configurations.

Different impact behaviours were observed for the reinforce-ment types used, with the difference in strength capability betweenthe Type-C and Type-F specimens being almost 15%. This indicatesthat the selection of reinforcement type is a very important factorto take into account in order to ensure the capability of the compos-ite lamination structure. In addition, the effects of different matrixtypes showed different capabilities to withstand impact, wherebyepoxy resin had higher capability than polyester resin by 350 N. Fur-thermore, the hybrid and architecture orientation structuresshowed significantly different behaviour under impact. The hybridstructure, formed using nine layers consisting of chopped strandmat, woven roving and polyester resin as the matrix (Type-D spec-imen, 600 N), showed higher strength as compared to the architec-ture orientation structure, consisting of Foam-Klegecell in themiddle of lamination structure (Type-F specimen, 320 N). The im-pact energy absorption of the different composite lamination struc-tures showed that the Type-C specimen had better energy absorbingcapability, approaching 1.3 J, as compared to other composite struc-tures. The results on the whole show that the effects of differentreinforcement, matrix and structure types need to be consideredin the capabilities of composite materials.

In the comparative study between the experimental and FEAsimulation results, the properties of the Type-C specimen were

considered, such as elasticity and density of material. The FEA sim-ulation conducted showed similar energy absorption values withthe experimental results, demonstrating the suitability of the Aba-qus FEA software for verification and validation works on Charpyimpact test conditions.

Acknowledgements

The authors would like to express their gratitude to UniversitiKebangsaan Malaysia and Universiti Tun Hussein Onn Malaysiafor supporting these research activities.

References

[1] Srivasta VK, Pathak JP. Friction and wear properties of bushing bearing ofgraphite filled short glass fibre composite in dry sliding. J Wear 1996:145–50.

[2] Sutherland LS, Soares CG. Impact on low fibre volume, glass polyesterrectangular plates. J Compos Struct 2005:119–27.

[3] Evci C, Gulgec M. An experimental investigation on the impact response ofcomposite materials. Int J Impact Eng 2012:40–51.

[4] Sutherland LS, Soares CG. Impact characterization of low fibre volume glassreinforced polyester circular laminated plates. Int J Impact Eng 2005:1–23.

[5] Mitrevski T, Marshall IH, Thomson R. The influence of impactor shape on thedamage to composite laminates. J Compos Struct 2006:116–22.

[6] Ghasemnejab H, Furquan ASM, Mason PJ. Charpy impact damage behaviour ofsigle and multi-delaminated hybrid composite beam structures. J Mater Des2010;31:3653–60.

[7] Sohn MS, Hu XZ. Impact and high strain rate delamination characteristics ofcarbon fibre epoxy composites. Theory Appl Fract Mech 1996:17–29.

[8] Shyr TW, Pan YH. Impact resistance and damage characteristics of compositelaminates. J Compos Struct 2003:193–203.

[9] Chen H, Hong M, Liu Y. Dynamic behaviour of delaminated plates consideringprogressive failure process. J Compos Struct 2004:459–66.

[10] Ulven CA, Vaidya UK. Impact response of fire damaged polymer-basedcomposite materials. J Compos Part B 2008:97–107.

A.M.T. Arifin et al. / Materials and Design 59 (2014) 475–485 485

[11] Hristov VN, Lach R, Grellman W. Impact fracture behavior of modifiedpolypropylene/wood fiber composites. J Polym Test 2004;23:581–9.

[12] Fu SY, Lauke B, Maeder E, Hu X, Yue CY. Fracture resistance of short glass fiberreinforced and short carbon fiber reinforced polypropylene under Charpyimpact load and its dependence on processing. J Mater Process Technol1999:501–7.

[13] Zang Z, Zhang Y. Analysis of optical switching in a Yb3+-doped fiber Bragggrating by using self-phase modulation and cross-phase modulation. Appl Opt2012;51:3424–30.

[14] Choi HY, Chang KL. A model for predicting damage in graphite/epoxylaminated composites resulting from low-velocity point impact. J ComposMater 1992:2134–69.

[15] Kwon YW, Wojcik GW. Impact study of sandwich composite structures withdelamination. J Compos Mater 1998:406–30.

[16] Pegoretti A, Cristelli I, Migliaresi C. Experimental optimization of the impactenergy absorption of epoxy carbon laminates through controlleddelamination. J Sci Technol 2008:2653–61.

[17] Erkendirci OF. Charpy impact behaviour of plain weave S-2 glass/HDPEthermoplastic composites. J Compos Mater 2012:1–7.

[18] Nita A, Opran C, Murar D, Bivolaru C. Charpy impact on the molded polymericparts. Acad J Manuf Eng 2010:85–91.

[19] Ali MB, Abdullah S, Nuawi MZ, Ariffin AK. Correlation of absorbed impact withcalculated strain energy using an instrumented Charpy impact test. Indian JEng Mater Sci 2013:504–14.

[20] Davies P, Petton D. An experimental study of scale effects in marinecomposites. J Compos Part A 1998:267–75.

[21] Huber T, Bickerton S, Mussig J, Pang S, Staiger MP. Flexural and impactproperties of all-cellulose composite laminates. J Compos Sci Technol2013;88:92–8.

[22] Rashkovan IA, Korabel YG. The strength of fiber surface treatment on itsstrength and adhesion to the matrix. J Compos Sci Technol 1997:1017–22.

[23] Rassmann S, Paskaramoorthy R, Reid RG. Effect of resin system on themechanical properties and water absorption of kenaf fibre reinforcedlaminates. J Mater Des 2011;32:1399–406.

[24] Han H, Taheri F, Pegg N, Lu Y. A numerical study on the axial crushing responseof hybrid pultruded and ± 45o braided tubes. J Compos Struct2007;80:253–64.

[25] Hufenbach W, Marques Ibraim F, Langkamp A, Bohm R, Hornig A. Charpyimpact tests on the composite structures – An experimental and numericalinvestigation. Compos Sci Technol 2008;68:2391–400.

[26] Malkapuram R, Kumar V, Yuvraj SN. Recent development in natural fibrereinforced polypropylene composites. J Reinf Plast Compos 2008;28:1169–89.

[27] Ku H, Wang H, Pattarachaiyakoop N, Trada M. A review on tensile properties ofnatural fibre reinforced polymer composites. J Compos Part B 2011:856–73.

[28] Ribeiro ML, Tita V, Vandepitte D. A new damage model for compositelaminates. J Compos Struct 2012:635–42.

[29] Ude AU, Ariffin AK, Azhari CH. Impact damage characteristics in reinforcedwoven natural silk/epoxy composite face-sheet and sandwich foam, corematand honeycomb materials. Int J Impact Eng 2013;58:31–8.

[30] Ude AU, Eshkoor RA, Zulkifli R, Arifin AK, Dzuraidah AW, Azhari CH. Bombyxmori silk fibre and its composite: a review of contemporary developments. JMater Des 2014;57:298–305.