materials and design - kocaeli...

Materials and Design 32 (2011) 3671–3685

lable at ScienceDirect
Contents lists avai
Materials and Design

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

Review

A review: Fibre metal laminates, background, bonding types and appliedtest methods

Tamer Sinmazçelik a,b,⇑, Egemen Avcu a, Mustafa Özgür Bora a, Onur Çoban a

a Kocaeli University, Mech. Eng. Dept., Umuttepe Campus, 41380 Izmit, Kocaeli, Turkeyb TUBITAK-MRC, Materials Institute, P.K. 21, 41470 Gebze, Kocaeli, Turkey

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

Article history:Received 28 January 2011Accepted 4 March 2011Available online 12 March 2011

Keywords:B. Sandwich structuresC. Surface treatmentsE. Mechanical

0261-3069/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.matdes.2011.03.011

⇑ Corresponding author at: Kocaeli University, MCampus, 41380 Izmit, Kocaeli, Turkey. Tel.: +90 2623003.

E-mail address: [email protected] (T. Sinmazçel

During the past decades, increasing demand in aircraft industry for high-performance, lightweightstructures have stimulated a strong trend towards the development of refined models for fibre-metallaminates (FMLs). Fibre metal laminates are hybrid composite materials built up from interlacing lay-ers of thin metals and fibre reinforced adhesives. The most commercially available fibre metal lami-nates (FMLs) are ARALL (Aramid Reinforced Aluminium Laminate), based on aramid fibres, GLARE(Glass Reinforced Aluminium Laminate), based on high strength glass fibres and CARALL (Carbon Rein-forced Aluminium Laminate), based on carbon fibres. Taking advantage of the hybrid nature fromtheir two key constituents: metals (mostly aluminium) and fibre-reinforced laminate, these compos-ites offer several advantages such as better damage tolerance to fatigue crack growth and impactdamage especially for aircraft applications. Metallic layers and fibre reinforced laminate can bebonded by classical techniques, i.e. mechanically and adhesively. Adhesively bonded fibre metal lam-inates have been shown to be far more fatigue resistant than equivalent mechanically bondedstructures.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Composite materials have been subject of permanent interest ofvarious specialists during the last decades. Firstly, military applica-tions in the aircraft industry triggered off the commercial use ofcomposites after the Second World War. The innovations in thecomposite area have allowed significant weight reduction in struc-tural design. Composites offer many advantages when compared tometallic alloys, especially where high strength and stiffness toweigh ratio is concerned. Additionally, they provide excellent fati-gue properties and corrosion resistance in applications [1]. With allthese advantages, composite structures have gained widespreaduse in the aerospace industry during the last decades [2–6].

In the fifties, an important goal of aircraft materials develop-ment was to improve the crack growth properties of structuralmaterials [7]. Competing materials like advanced aluminium alloysand fibre reinforced composites have potential to increase the cost

ll rights reserved.

ech. Eng. Dept., Umuttepe303 3006; fax: +90 262 303

ik).

effectiveness of the structure. These materials still have theiradvantages and disadvantages, like the poor fatigue strength ofthe aluminium alloys and the poor impact and residual strengthproperties of carbon fibre reinforced composites. At the end ofthe seventies, the idea of using the two materials to form a hybridcomposite structural material to overcome most of the disadvan-tages of both materials was born [8]. At the Delft University ofTechnology, it was found that the fatigue crack growth rates inadhesive bonded sheet materials can be reduced, if they are builtup by laminating and adhesively bonding thin sheets of the mate-rial, instead of using on one thick monolithic sheet [7]. The advan-tage becomes highly evident if cracks start in one of the sheets ofthe laminate only, the adhesive layers behaving as crack dividers.Under these circumstances, the sheets that are still uncrackedreduce the crack growth rate in the cracked sheet. The reductionin the crack growth rates persists until a crack is initiated in theneighbouring sheet also. Based upon all the research, initial FibreMetal Laminates material, ARALL (Aramid Fibre ReinforcedAluminium Laminate) was introduced at 1978 in the Faculty ofAerospace Engineering at the Delft University of Technology(DUT) [3]. ARALL consists of alternating thin aluminium alloylayers (0.2 ± 0.4 mm) and uniaxial or biaxial aramid fibre prepreg,as shown in Fig. 1 [8].

http://dx.doi.org/10.1016/j.matdes.2011.03.011

mailto:[email protected]

http://dx.doi.org/10.1016/j.matdes.2011.03.011

http://www.sciencedirect.com/science/journal/02613069

http://www.elsevier.com/locate/matdes

Fig. 1. Schematic presentation of Fibre Metal Laminate (ARALL 2).

3672 T. Sinmazçelik et al. / Materials and Design 32 (2011) 3671–3685

Fibre metal laminates (FMLs) are hybrid composite structuresbased on thin sheets of metal alloys and plies of fibre reinforcedpolymeric materials [6]. The fibre/metal composite technologycombines the advantages of metallic materials and fibre reinforcedmatrix systems. Metals are for instance isotropic, have a high bear-ing strength and impact resistance and are easy to repair, while fullcomposites have excellent fatigue characteristics and high strengthand stiffness. The fatigue and corrosion characteristics of metalsand the low bearing strength, impact resistance and reparabilityof composites can be overcome by the combination [5,9].

These material systems are created by bonding composite lam-inate plies to metal plies [8]. The concept usually applied to alu-

Fig. 2. Classification of FML

minium with aramid and glass fibres, but also can be applied toother constituents [10]. Fig. 2 gives a classification of FML basedon metal plies. The most commercially available FMLs are ARALL1,based on aramid fibres and GLARE1 based on high strength glassfibres [8]. These two grades of FMLs are going to be explained indetail Chapter 3.

1.1. Historical development of fibre metal laminates

During the last three decades, there has been a search for light-weight materials that can replace the traditional aluminium alloysin aerospace structures [11]. For an optimal structural design, a

s based on metal plies.

T. Sinmazçelik et al. / Materials and Design 32 (2011) 3671–3685 3673

new material is needed which combines high strength, low densityand high elasticity modulus with improved toughness, corrosionresistance and fatigue properties. Fibre reinforced compositesmaterials almost cover all these demands, except for fracturetoughness.

In 1978, researches were carried out to increase the fatigue per-formance of aluminium alloys at the National Aerospace Labora-tory and at the Delft University of Technology in Netherland. Animprovement of the fatigue behaviour in laminate sheet materialswas obtained by introducing a high strength aramid fibre into theadhesive layers. As a result of all these studies, they introducedARALL, first fibre metal laminate at the Faculty of Aerospace Engi-neering at the Delft University of Technology in Netherland [3]. In1984, two international patents were accepted and a pilot produc-tion of four different types of standardized ARALL was started bythe Alcoa Company after sufficient confidence in this materialhad been gained [7]. The trades ARALL 1 and ARALL 2 were stan-dardized. Arall 1 is a variant with aluminium 7075 layers and Arall2 uses aluminium 2024 layers and it was in the as-cured condition[1].

Later, a much stiffer ARALL which consists carbon fibres insteadof aramid fibres, the CARALL Laminates, had been investigated inDUT [12]. Recent research has shown that CARALL laminates alsohave fibre failure occurred during flight-simulation fatigue testsat elevated stress levels, which resulted in poor fatigue perfor-mance. The limited failure strain of the carbon fibres (0.5–2.0%)was thought to be a disadvantage. Thus, it is sensitive to notchbehaviour comparing to monolithic aluminium alloy. Due to theproblem of galvanic corrosion between the carbon fibres and thealuminium sheet in moisture environment, more research has tobe done.

In 1990, another attempt to improve ARALL laminates, adoptinghigh strength glass fibre instead of aramid fibres, called GLARE(Glass reinforced, or ARALL with glass fibres) was developed suc-cessfully [7]. Glass reinforced aluminium (GLARE) is the successorof Aramid aluminium laminate (ARALL) [12]. A partnership be-tween AKZO and ALCOA started to operate in 1991 to produceand commercialize GLARE [1].

Finally, a new concept for fibre metal laminates (ARALL &GLARE), the Spliced Laminates, is launched by the Structural Lam-inates Company (SLC) in 1992. Spliced laminates are defined asFML in which the aluminium layers are interrupted such that thedimensions of the spliced sheets depend on the autoclave dimen-sions only [7].

1.2. Advantages and disadvantages of fibre metal laminates

Fibre metal laminates take advantages of metal and fibre-rein-forced composites, providing superior mechanical properties tothe conventional lamina consisting only of fibre-reinforced laminaor monolithic aluminium alloys [9]. Table 1 summarizes all theadvantages of fibre metal laminates depending on previousstudies.

The major disadvantage associated with epoxy based fibremetal laminates is the long processing cycle to cure the polymermatrix in the composite plies [6]. This problem increases the cycletime of whole production and decreases productivity. Thisincreases labour costs and overall cost of FMLs.

1.3. Production of FMLs

To produce FMLs, as for polymeric composite materials, themost common process involves autoclave processing. The overallgeneric scenario for production of FMLs involves about five majoractivities.

1. Preparation of tools and materials. During this step, the alumin-ium layer surfaces are pre-treated by chromic acid or phospho-ric acid, in order to improve the adhesive bonding betweenmetallic layer and fibre reinforced laminate (metallic surfacetreatment is explained in detail in Chapter 2).

2. Material deposition, including cutting, lay-up and debunking.3. Cure preparation, including the tool cleaning and the part trans-

ferring in some cases, and the vacuum bag preparation in allcases.

4. Cure, including the flow-consolidation process, the chemicalcuring reactions, as well as the bond between fibre/metallayers.

5. Post stretching, after the hot-curing (200 �C) cycle in the auto-clave, the fibre metal laminates carry a residual stress systemover the thickness of the material with a small tensile stressin the aluminium sheets and compression in the fibres, this sit-uation adversely effects fatigue resistance of FMLs. Post stretch-ing operation after curing cycle can reverses the residual stresssystem and solves this problem.

6. Inspection, usually by ultrasound, X-ray, visual techniques andmechanical tests [1,7]. Mechanical tests of FMLs are explainedwidely in Chapter 4.

1.4. Applications

Due to their advantages aforementioned above, FMLs are find-ing great use in most commonly in aerospace applications. A num-ber of companies have interest in substitute the traditionalaluminium components by FML composites [5]. Both ARALL andGLARE laminates are now being used as structural materials in air-crafts. Fibre Metal Laminates have been successfully introducedinto the Airbus A380 [4]. The Fig. 3 shows FML composite applica-tions in the Airbus A380 airplane [1,18].

ARALL has been developed for the lower wing skin panels of theformer Fokker 27 aircraft and the cargo door of the Boeing C-17[8,10]. ARALL 3 material is currently in production and flight teston the C-17 cargo doors and GLARE is selected for the Boeing777 impact resistant bulk cargo floor.

2. Adhesive bonding

Adhesive bonding process has been used in the manufacture ofaircraft structures and components for 30–40 years [19]. Bondingtechniques are used in an ever-growing number of applications:electronics (multilayer boards, bounded components), aerospaceand aeronautical industries (F-18 bonded wings), automotiveindustries and many more organizations. Bonding structural com-ponents with adhesives offers many advantages over conventionalmechanical fasteners: lower structural weight, lower fabricationcost, and improved damage tolerance [20]. Light weight sandwichconstruction and structural bonded joints form a major proportionof modern aircraft [21]. Modern metallic bonded structures usinglongitudinal lap joints in commercial aircrafts were introducedwith the advent of the Airbus A300 [19]. As stated in Ref. [19], thistechnique is currently migrating from secondary to primary struc-tural applications as one of the most interesting ways to fastenstructural parts with a high level of confidence. For example, struc-tural adhesive bonding is mainly used for attaching stringers and/or tear straps to the fuselage and wing skins, to stiffen the struc-tures against buckling. It is also applied to skin-to-core bondingin metallic honeycomb structures, such as elevators, ailerons,spoilers, and so on [22]. Bonded structures have been shown tobe far more fatigue resistant than equivalent mechanically fas-tened structures and when designed correctly, can sustain higherload levels than equivalent mechanically fastened joints [21].

Table 1Advantages of fibre metal laminates.

Materialbehaviour

1. High fatigue resistance: It is achieved by the intact bridging fibres in the wake of the crack, which restrain crack opening. FMLs have excellentfatigue characteristics [5,13]2. High strength: As mentioned before, FMLs are hybrid structures based on thin sheets of metal alloy and plies of fibre-reinforced polymericmaterials. Combination of metal alloys which have high bearing strength and fibre reinforced composites which have high strength and stiffness.FMLs are structural materials with high strength [4,5]3. High fracture toughness: Depending on the studies, FMLs have better fracture toughness than their constituent alloys. Based on this behaviour andaccounting for the lower fatigue crack growth rates presented by these materials, they can be a very attractive option for a wide set of structuralapplications [14]4. High impact resistance: Unlike composites, the damage tolerance behaviour of FML is comparable to conventional aluminium alloys. The sametypes of damage and plastic deformation are observed, only at higher impact energy levels. Impact deformation is actually a significant advantage ofFML, especially when compared to composites, because this visible damage significantly increases inspect ability and detect ability. Therefore, theuse of FML allows for repair criteria and techniques [8,15]5. High energy absorbing capacity: Based upon researches, FMLs are capable of absorbing significant energy through localized fibre fracture and shearfailure in the metal plies [6,16]

Physicalproperties

1. Low density: Owing to epoxy based polymer matrix and low density aluminium sheets, FMLs are a weight saving structural material compare toothers [7]

Durability 1. Excellent moisture resistance: The moisture absorption in FML composites is slower when compared with polymer composites, even under therelatively harsh conditions, due to the barrier of the aluminium outer layers [1]. Additionally pregreg layers are able to act as moisture barriersbetween the various aluminium layers inside of the FMLs [13]2. Excellent corrosion resistance: As mentioned above excellent moisture resistance of FMLs and high corrosion resistance of polymer based fibrelaminates ensures FMLs excellent corrosion resistance [7,8,10,13]3. Lower material degradation: FMLs have excellent moisture and corrosion resistance, FMLs degradation as result of environmental aspects issignificant lower as compared to either metallic structures, or composite structures [5]

Safety 1. Fire resistance: The high melting point of the fibres in FML (for instance, glass fibres in GLARE laminates can withstand 1100 �C) preventspenetration of the fire to the inner layers. Hence, the fire resistance of FMLs are much better than monolithic aluminium alloys depending on theirfibre melting point. FMLs are being used as fuselage materials in aircrafts. With their good fire resistance, FMLs ensures enough time to passengersfor evacuating the aircraft safely [8,17]

Cost saving 1. FML offer substantial weight savings relative to current metallic structures. Further, the number of parts required to build a component may bedramatically less than the number of parts needed to construct the same component of metal alloy. This can lead to labour savings. Sometimesoffsetting the higher price of the present materials2. Owing to fatigue insensitivity of FMLs, less repair and longer maintenance periods are sufficient for FMLs. These advantages reduce maintenancecosts of FMLs [7]

Fig. 3. Metal/fibre applications in A380 airplane from Airbus [1].


One of the most important aspect of designing aircraft struc-tures is to consider skin cracks resulting from fatigue that thecracks will be discovered within a particular inspection intervalprior to their reaching a critical crack length that can lead to cata-strophic failure. As noted in Ref. [7], at the Delft University of Tech-nology, it was found that the fatigue crack growth rates in adhesivebonded sheet materials can be reduced. If they are built up by lam-inating and adhesively bonding thin sheets of the material, insteadof using on one thick monolithic sheet. The advantage becomeshighly evident if cracks start in one of the sheets of the laminateonly, the adhesive layers behaving as crack dividers. Under thesecircumstances the sheets that are still uncracked reduce the crackgrowth rate in the cracked sheet. The reduction in the crack growthrates persists until a crack is initiated in the neighboring sheet also.The research on adhesive-bonded sheet metal laminates is then be-came the predecessor of a new class of materials; the fibre-metallaminates [7].

As reported in Ref. [19], early experiences in bonding tech-niques demonstrated that surface treatment prior to bonding isthe single most critical step which can not be disregarded, evenfor tertiary-loaded structures, since it is essential to achievelong-term service capability [21,23]. A particular surface treatmenttends to modify the substrate surface by delivering the followingfeatures: free from contamination; wettable with either primeror adhesive; highly roughened; and mechanically and hydrolyti-cally stable [23]. Surface treatment technologies need to be ex-plained from the viewpoint of adhesively bonded fibre-metallaminate research.

3. Surface treatments for adhesive bonding

As shown in Table 2, all the treatments for modification of metalsurfaces can be grouped as:

1. Mechanical.2. Chemical.3. Electrochemical.4. Coupling agent.5. Dry surface treatments.

Solvent degreasing is important, because it removes contami-nant materials which inhibit the formation of the chemical bonds.However, solvent degreasing, while providing a clean surface, doesnot promote the formation of acceptable surface conditions forlonger term bond durability [21]. The degreasing stage usuallymakes use of chlorinated solvents such as trichloroethylene,1,1,1-trichloroethane, perchloroethylene, or dichloromethane, oralternatively, non-chlorinated solvents including methyl ethylketone, methanol, isobutanol, toluene or acetone [24]. All alumin-ium alloy sheets were initially degreased prior to further surface

Table 2All treatments for modification of metal surfaces.

Treatments Nature oftreatments

Ref.

Grit-blasting Mechanical [26–29]Chromic–sulphuric acid (CAE) Acid etching [30,33]Sulfo-ferric acid (P2) Acid etching [31]Forest Product Laboratory (FPL) Acid etching [32]Alkaline Etching [25,30]Chromic acid anodizing (CAA) DC-anodizing [46]Phosphoric acid anodizing (PAA) DC-anodizing [23,34,36]Sulphiric acid anodizing (SAA) DC-anodizing [36,45]Boric-sulphiric acid anodizing (BSAA) DC-anodizing [24]Phosphoric acid anodizing (AC-PAA) AC-anodizing [36]Sulphiric acid anodizing (AC-SAA) AC-anodizing [36]Silane Coupling/oxidation [28,47,49–52]Sol–gel Coupling/oxidation [53–57]Excimer laser texturing Mechanical [54,58–62]Plasma sprayed coating Ablation/oxidation [35,63–67,69]Ion beam enchanced deposition

(IBED)Ablation/oxidation [62,68]

Table 3Chemical acid etches and their application process.

Chemicaletch type

Application process Ref.

CAE Immersion in a water solution with 330 ml/l chromic–sulphuric acid (97% v/v) and 50 g/l potassiumdichromate, at 60 oC for 15 min, and rinsed in tap water

[30]

Immersion in a water solution with 185 ml/l chromic–sulphuric acid (97% v/v) and 127 g/l ferric sulphate, at65 �C for 8 min, and rinsed in tap water (percentages byweight: 48% H2O, 37% H2SO4 and 15% FeSO4)

[30]

Immersion at 15 min in a 65 �C, followed by tap andde-ionised water rinsing

[33]

P2 Involves treatment of the adherend surface with anaqueous acidic solution containing Fe(III). (1)immersing the adherend in the P2 solution at 65 �C for8 min, (2) rinsing the specimen in DI water for 2–3 min,and (3) drying the aluminium bars in an oven at 60 �Cfor 30 min. The P2 solution was prepared by dissolving122.5 g Fe2(SO4)3 4H2O and 0.185 l of concentratedsulphuric acid in enough water to make a liter ofsolution

[31]

FPL Treat with optimized FPL for 30 min at 62 + 2 �C. Aftertreatment in the acid, the aluminium was washed for20 min in cold running tap water and finally dried in anoven at 40 �C for 30 min

[32]

Alkalineetch

Immersion in 100 g/l NaOH solution, at 60 �C for 1 min,and rinsed in tap water

[30]

Immersion in %10 wt. NaOH solution for 2 min at 60 �C [25]


pre-treatment steps. The first step in the fabrication of baselinetest specimens was methyl ethyl ketone (MEK)-wiping of alumin-ium substrates with lint-free tissues to degrease the surface [25].Each surface treatment technologies which have been used tomodify metal surfaces will be explained detail in subtitles of Sec-tion 3.

3.1. Mechanical treatment

As a preliminary preparation step in the multi-stage schedules,mechanical abrasion has been used to produce a macro-roughenedsurface, different roughness levels of the surface textures and to re-move an undesirable oxide layer, respectively [19]. This methodtypically involves abrasive scrubbing of the substrate surface withsand paper. This mechanical treatment would introduce physico-chemical changes which yield a wettable surface and modify thesurface topography, i.e., a macro-roughened surface [27].

Mechanical treatments include abrasion methods usually com-bined with degreasing. Pretreatment by blasting using alumina orsilica grit or glass beads change the topography and chemical stateof an aluminium adherend by the introduction of a ‘‘peak-and-val-ley’’ type morphology [24]. In general terms, degreasing is the min-imum pretreatment that is usually carried out prior to bonding.Grit-blasting or other mechanical abrasion methods are recognizedas providing a useful increase in initial adhesion levels [24]. Severalresearches have highlighted different aspects of grit blasting orother mechanical abrasion methods for aluminium alloys. Grit-blasting, which uses alumina grit with clean, dry, propellant air,is employed as a logical surface activation step prior to modifythe metallic surface for satisfactory bonding results [26–29].

3.2. Chemical treatment

The most commonly applied chemical treatments are based ona chromic–sulphuric acid etch [23,37]. This treatment consists ofimmersion of the substrate in a solution of sulphuric acid andpotassium dichromate. Typically, chemical treatment, i.e., acid-etching, is an intermediate production step between degreasing,alkaline cleaning, and electrochemical treatment [19]. Three classi-cal acid-etching solutions were introduced to modify the metallicsurfaces: chromic–sulphuric acid (CAE) [38], Forest Product Labo-ratory (FPL) [25], and sulfo-ferric acid (P2) etches. The most effec-tive etches incorporate mixed chromic and hydrofluoric acids.However, non-chromated acid etchants have been demonstratedto provide good adhesion results [23,39].The surfaces treated with

acid-etch were consistently reported to be out-performed by anod-ized surfaces so that the inconsistent results made them unsuitablefor stand-alone treatment in primary bonded structures [23].Chemical acid etches are given in Table 3 with concentration ofcompounds, used temperature and immersion time.

3.3. Electrochemical treatment

Good adhesion between the metal and the adhesive could beobtained when the aluminium surfaces were degreased in alka-line solutions or organic solvents and subsequently etched inaqueous chromic–sulphuric acid solutions. However, this surfacepre-treatment was found to be insufficient in certain non-bonded areas where corrosion occurred. The corrosion suscepti-bility was reduced if the aluminium surface, after chromic–sul-phuric acid etching, was anodized before bonding [34]. Inorder to obtain optimum durability of adhesive bonded alumin-ium joints, anodizing processes are extensively used in the aero-space industry. Anodizing produces a thin oxide film, again witha high degree of micro-roughness and an oxide form which ishighly resistant to hydration [21]. Anodic oxidation in solutionsof chromic acid anodizing (CAA) or phosphoric acid anodizing(PAA) is the preferred stabilizing treatment for the structuraladhesive bonding of high-strength aluminium alloys in criticalapplications such as aircraft components [35]. Chromic acidanodizing (CAA) was found to give a relatively thin and ductileoxide layer and was established as an effective pretreatmentfor adhesive bonding with good durability properties in service.The European aviation industry is still using this method [36].However, as a result of delamination in the boundary zone andcorrosion after only a few years of operation, an improvementof the usual FPL process was established. Boeing developed thephosphoric acid anodizing (PAA), which led to a clearly betterjoint durability [23,34].

As stated in Ref. [24], in almost all studies, however, the DCelectrochemical pretreatments give the best levels of initial adhe-sion and durability [23,36,40,41]. The three most widely used


electrochemical processes used in prebond applications are phos-phoric acid anodizing (PAA), chromic acid anodizing (CAA) and sul-phuric acid anodizing (SAA). Other electrolytes have been studiedbut to a much lesser extent. CAA and PAA are both extensively usedin aerospace applications [42,43]. All DC anodizing procedures arecomplex multi-stage operations incorporating degreasing, desmut-ting and deoxidizing stages [34]. The boric–sulphuric acid anodiz-ing (BSAA) process has advantages associated with both the CAAand PAA. A BSAA process has been patented by Boeing as a directreplacement to CAA to meet environmental regulations [44]. ACanodizing is a highly feasible, robust, and environmentally friendlyprocess, which is free from the hexavalent chromium found in theconventional CAA. A distinct difference from DC anodizing ishydrogen gas generation on the treated surface during the hotAC anodizing process [19]. Electrochemical treatments which areused for improving the metallic surfaces as an adhesive layer areshown in Table 4.

3.4. Coupling agent treatment

There is an ongoing need to develop environmentally friendlycoupling agent techniques, i.e., silane or sol–gel, which can be usedwith the current generation of materials, i.e., primer and film adhe-sive [19]. Silane based products are becoming an interesting mate-rial for pre-treatment deposition, because, for the environmentalcompatibility, they can be used as substitutes of traditional pre-treatments like chromates [47]. Silane coatings are environmen-tally sound, multi-metal surface pre-treatments, considered foruse on various metals such as aluminium and its alloys, copper,iron and steel, zinc and magnesium containing alloys [48]. It issufficient to mention that a large number of studies have demon-strated the usefulness of silanes such as c-glycidoxypropyltrimeth-oxysilane when used as a primer to improve the durability ofstructural aluminium bonds. This has been attributed to the forma-tion of stable, covalent bonds between the metal(oxide) and thesilane and the possibility for interphase formation giving a regionof intermediate modulus between the metal and polymer whichfacilitates stress transfer [49,50]. Silanes may be used as stand-alone processes or part of a more complex pretreatment [28,51].Bis-1,2-(triethoxysilyl)ethane (BTSE) has received much attentionas it is, after hydrolysis, highly reactive towards (covalent) metal/film bonding and cross-link formation for the creation of barrierproperties [52]. Fedel et al. [47] examined the effect of water-basedsilane pre-treatments on galvanized steel. Three different silaneswere used for the pre-treatment deposition (glycidoxypropiltri-methoxysilane, tetraethoxysilane and methyltriethoxysilane). Theelectrochemical tests showed that the silane layer acts not onlyas a coupling agent between the inorganic substrate and the organ-ic coating, but it also ensures a good barrier effect against waterand oxygen. Rider and Arnott [28] described that the silane layer

Table 4AC and DC Electrochemical Treatment.

Pre-treatment

Electrolyte(wt.%)

Voltage (V) Time(min)

Temperature(�C)

Ref.

PAA 10 (H3PO4) 10 (DC) 20 25 [36]SAA 16 (H2SO4) 1.5 (DC)

(currentdensity A/dm2)

20 20 [36]

BSAA 5.0–10.0H3BO3/30.0–50.0 H2SO4

15 ± 1 (DC) 18–22

26.7 ± 2.2 [24]

SAA 15 (DC) 22 15 [45]CAA 2.5–3.0 CrO3 40.0 ± 1.0/

50.0 ± 1.0 (DC)35–45

40.0 ± 2.0 [46]

PAA-AC 10 (H3PO4) 4 (AC) 30 s 50 [36]SAA-AC 15 (H2SO4) 10 (AC) 12 s 80 [36]

promotes the hydrolytic stability of the adhesively bonded joint,leading to increased durability in wedge tests.

Sol–gels are organic–inorganic polymers formed by hydrolysis/condensation reactions of alkoxide precursors, primarily silanes,which have found applications as electronic, optical and protectivecoatings. These coatings possess important characteristics such aschemical stability, physical strength and scratch resistance. Thesol–gel process can be used to form nanostructured inorganic films(typically 200 nm to 10 lm in overall thickness) that are moreresistant than metals to oxidation, corrosion, erosion and wearwhile also possessing good thermal and electrical properties [53].As noted in Ref. [19], sol–gel which involves the growth of me-tal–oxo polymers through both hydrolysis and a condensationreaction to form inorganic polymer networks in thicknesses rang-ing from 50 to 200 nm [54,55]. The Boegel EPII sol–gel agent con-sists of a dilute aqueous epoxy-silane and zirconium alkoxide. Thismixture provides a thin inorganic zirconium oxide film incorpo-rated with the epoxy-silane that is applied to the metallic surface[55,56]. In this process, zirconium reacts with the metallic surfaceto produce a covalent chemical bond, while the epoxy-silane offersa reactive organic group for bonding with the adhesive [55,57].

3.5. Dry surface treatments

Several dry treatments for aluminium alloy surfaces have beendeveloped to replace chemical wet treatment process: excimer la-ser texturing [54,58–62], plasma-sprayed coating [35,63–67], andion beam enhanced deposition (IBED) [62,68].

As reported in Ref. [19], laser texturing was utilized to modifyan aluminium substrate’s morphology and micro-structure, result-ing in an increased bond strength and durability [58,59,62]. IBED isa process that cleans and modifies the surface by sputtering withhigh energy argon ions under vacuum. This process requires a sur-face activation step, particularly grit-blasting, prior to IBED. Goodinitial bond strengths were obtained and the improvements inwedge durability compared with PAA were found [62]. In the aero-nautic industry, a complex and critical process is used in order toenhance both wettability and adhesive properties of aluminium al-loy surfaces.

Cold plasma treatment represents an alternative to conven-tional processes in terms of environmental impact. Plasma is anionized gas containing both charged and neutral particles, suchas electrons, ions, atoms, molecules and radicals. There are twokinds of plasma: cold and hot. The working pressure of a cold plas-ma is generally lower than atmospheric pressure which allows theproduction of chemical reactive species at low temperatures(<773 K). Since hot plasma pressure is higher than atmospheric,the temperature is about 104–105 K [35]. Cold plasma treatmentrepresents an efficient, clean and economic alternative to activatealuminium surfaces [67]. Plasma spray coatings were also usedas pretreatments prior to adhesive bonding. Davis et al. [63] re-ported that a 60Al–Si/40 polyester (by wt.) plasma spray coatingcould be engineered for specific applications and was better suitedfor the adhesive bonding process of aluminium substrates, byyielding a performance equivalent to that of the PAA treatmentwith some epoxy adhesives in the wedge test. De Iorio et al. [69]examined cold plasma treatments to modify the surface propertiesof polymeric materials (polyoxyphenylene–polyammide, polycar-bonate–ABS) and A1 6061 alloy, with the aim of increasing the sur-face adhesion in structural joints.

4. Mechanical bonding

As reported at Ref.[73], the increasing requirements for weightreduction demand more and more use of composite materials in


aerospace applications [70]. There has been a progressive increasein the number of metal parts and structures replaced by compositematerials, not only in military, but also in civil aircraft design. Forinstance, composite materials comprise 22% of the total weight ofthe Airbus A380 [71]. The centre wing box, the tail cone, the pres-sure bulkheads and the vertical and horizontal tails are only someof numerous parts of this modern aircraft [72] being made of com-posite materials. Structural coupling leads to an increase in struc-tural complexity and to additional local stress factors, but also to aconsiderable reduction in global optimizing possibilities, which, inturn, decreases the lightweight potential of composite design.Thus, the development of advanced joining technologies designedto minimize the weight penalties produced by structural couplinggains a new meaning in the present situation. This is especiallytrue for concentrated load transmissions, as it is the case formechanically fastened joints [73]. Mechanical fastening is stillone of the main methods currently used for joining compositecomponents [70], with the advantage of no special surface prepa-rations, easy disassembly and inspection.

Hybrid bolted bonded (HBB) joining technology in lap jointscombines both classical techniques, i.e. bolting and bonding. Theadvantages of HBB joints, compared to bolted joints, could be sum-merized in two main points: (i) continuous instead of discrete loadtransfer distribution along the overlap and (ii) decrease of loadtransferred by fasteners. Furthermore, in terms of security, theexistence of fasteners in HBB joints could ensure the functioningof the structure, even if the adhesive layer failure occurs. In man-ufacturing industries in which numerous bolted and riveted jointsare used, the use of HBB technology could be considered as a po-tential solution, in order to reduce the mass as well as the manu-facturing cost [74].

As stated at Ref. [73], the mechanical behaviour of compositebolted joints has been extensively studied in the past by means ofexperimental, analytical and numerical approaches [75–81], mainlyfocusing on the determination of the load capability, the load andstress distributions and failure criteria of single- and multi-rowbolted joints under the influence of varying laminate configurationsand joint geometries. One of the most effective ways to improve theload capacity of composite bolted joints entails the local reinforce-ment of the composite laminate with high-strength metal layers[82–84], thus clearly improving its bearing and shear capabilities.The special feature of this reinforcement technique consists of onlyembedding the metal layers into the bolted joining area locally,which is accomplished either by ply-addition (metallic layers are in-serted between the composite plies [82]) or ply-substitution tech-niques (composite plies are replaced by metallic layers [84]).Hence, the total load capability of this reinforcement approach de-pends not only directly on the load capability of the bolted joint,but also on the strength of the transition zone between the pure fi-bre composite material and the hybridized laminate region.

Fig. 4. Crack bridging mechanism of ARALL laminates.

5. Fibre metal laminates

5.1. Arall

The idea of putting fibre reinforcement in the adhesive bondlines between aluminium alloys has been researched at many lab-oratories over the past three decades. These researches made itpossible to develop first fibre metal laminates. Finally in 1978,ARALL has been developed at the Faculty of Aerospace Engineeringat the Delft University of Technology in Netherland. In 1982 thefirst commercial product under the trade name ARALL waslaunched by ALCOA.

ARALL laminates are made of high strength aramid fibresembedded in a structural epoxy adhesive sandwiched between

multiple layers of thin aluminium alloy sheets. A schematic pre-sentation of ARALL is shown in Fig. 1. Combination of high strengthmetals (aluminium layers) and strong fibres (aramid layers) gener-ates a new composite material with a unique set of properties.ARALL laminates offer many advantages such as high strengthand excellent fatigue properties. Moreover, they retain the advan-tages of aluminium alloys, namely lower cost, easy machining,forming, and mechanical fastening abilities, as well as substantialductility [1,3,85,86].

In the ARALL concept, unidirectional fibre prepregs are used.This concept is very tolerant of material inconsistencies. The fibreorientation is chosen into the direction of the main load. The lam-inates are designed such that the fibres do not fail, when fatiguecracks develop. That means they remain intact behind the tip ofthe propagating crack in the metallic layers. They hinder the open-ing of the crack, and consequently, the crack tip stress intensity inthe aluminium sheet is reduced. Fig. 4 shows the crack bridgingmechanism of the ARALL laminates.

The fibres are insensitive to the fatigue loading, which key ben-efit of this material regarding crack propagation. The pregregtransfers loads over the crack, decreasing the stress intensity atthe crack tip and reducing the crack growth rate. As a result ofcrack bridging mechanism, the crack growth life of FMLs is signif-icantly extended compared to monolithic aluminium. Crack bridg-ing is related to the crack opening, the loads are initially mainlytransferred through the aluminium sheets until a certain crackopening is reached [7,12,87,88]. This behaviour can lead toimprovement in the crack growth rates by a factor of a hundredand even more, as compared to monolithic aluminium alloy sheet.This improvement also allows for weight savings up to 30% in fati-gue-critical aircraft components. Comparison of ARALL laminateson a structural level with other aircraft materials show that, ARALLlaminates are very attractive materials, especially for fatiguedominated structural parts, such as the lower wing skin and the


pressurized fuselage cabin of an aircraft. In this way a new hybridmaterial ARALL has been obtained. In 1984 two international pat-ents were accepted and a pilot production of four different types ofstandardized ARALL was started by the Alcoa Company after suffi-cient confidence in this material had been gained [7]. Commercial-ized ARALL laminates are given in Table 5 [89].

All four standard ARALL products employ a thermoset adhesivesystem impregnated with unidirectional aramid fibres in a fibre toresin weight ratio of 50:50. The fibres are oriented parallel to thealuminium sheet rolling direction. ARALL laminates are producedunder very rigid quality control standards. The aluminium sheetsare anodized and primed according to existing aerospace industryspecifications. Both chromic acid anodizing and phosphoric acidanodizing treatments are used. The outer aluminium surfaces canbe left bare for later processing after an ARALL part has beenformed. Table 6 summarizes production parameters of ARALLlaminates.

ARALL panels are laid up and cured under the appropriate timesand temperatures either in mechanical press or an autoclave. Aftercuring, the aluminium layers are in a slight tensile residual stressstate and the pregreg layers in a compensating compressive resid-ual stress state. The ARALL 1 and 3 laminates are given a postcure0.4% nominal stretch to reverse these residual stress signs. TheARALL 2 and 4 laminates are not postcured stretched as a standardpractice, although they can be supplied that way [86].

Table 6Production parameters of ARALL laminates [86].

Metaltype

Cureresin

Curetemp.

Stretching Characteristics

ARALL1

7075-T6

AF-163-2

120 �C 0.4% permanentstretch

Superior fatigueresistanceHigh strength

ARALL2

2024-T3

AF-163-2

120 �C With or without0.4% stretch

Excellent fatigueresistanceIncreasedformabilityDamage tolerant

ARALL3

7475-T76

AF-163-2

120 �C 0.4% permanentstretch

Superior fatigueresistanceControlledtoughness

ARALL4

2024-T8

AF-191

175 �C With or without0.4% stretch

Excellent fatigueresistanceIncreasedelevatedTemperatureproperties

Table 5Commercially available ARALL laminates [89].

Metal type Metalthickness(mm)

Fibrelayer(mm)

Fibredirection(�)

Characteristics

ARALL 1 7075-T6 0.3 0.22 0/0 Fatigue,strength

ARALL 2 2024-T3 0.3 0.22 0/0 Fatigue,formability

ARALL 3 7475-T76 0.3 0.22 0/0 Fatigue,strength,exfoliation

ARALL 4 2024-T8 0.3 0.22 0/0 Fatigue,elevatedtemperature

As mentioned before; ARALL has been developed for the lowerwing skin panels of the former Fokker 27 aircraft and the cargodoor of the Boeing C-17. Later with new developments, ARALLhas been used in wide range of fields and applications. They wereused in fuselage and wing structures in aerospace applications andused as a ballistic material in military applications. However, theyare mostly used as a structural material in cargo doors of aircrafts[8,10,90].

5.2. Glare

GLARE laminates belong to fibre metal laminates family, theyconsist of alternating layers of unidirectional glass fibre reinforcedpregregs and high strength aluminium alloy sheets. At first, theywere developed for aeronautical applications as an improvementof ARALL with advanced glass fibre and introduced at the TechnicalUniversity of Delft in Netherlands in 1990. Later, a partnership be-tween AKZO and ALCOA started to operate in 1991 to produce andcommercialize GLARE [1,7,91,92]. Fig. 5 schematically illustrates across-ply GLARE laminate.

The biggest differentiation of GLARE compared to ARALL is thatGLARE consists of glass fibres instead of aramid fibres. This diver-sity gives superior properties to GLARE laminates. Table 7 high-lights the advantages and disadvantages of various fibres for FMLs.

The specific stiffness and strength in the fibre direction ofGLARE are enhanced over the high strength aluminium alloy usedfor the metal layers, which significantly contributes to weight sav-ings in the designs of tension-dominated structural components.Aforementioned fibre bridging mechanism impedes the growthand propagation of cracks in the aluminium-alloy layers under ten-sile fatigue loading conditions [91].

GLARE has a better adhesion between the glass fibres comparedto ARALL. Moreover, glass fibres are much more resistant to com-pression loading. As a consequence, fibre failure in the glass fibreshas rarely been observed during in fatigue load. Other advantagesof GLARE over ARALL are its higher tensile and compressivestrength, better impact behaviour and better residual strength.Better adhesion between glass fibre and resin makes GLARE lami-nates with fibres build up in two directions is possible. This beingis much more suitable for some constructions where biaxial stres-ses occur. These properties seem to make GLARE has a wider rangeof potential application [7].

Nowadays, GLARE materials are commercialized in six differentstandard grades (Table 8). They are all based on unidirectionalglass fibres embedded with epoxy adhesive resulting in prepregswith a nominal fibre volume fraction of 60%. During fabricationof composites the prepregs are laid-up in different fibre orienta-tions between aluminium alloy sheets, resulting in different stan-dard GLARE grades. For the GLARE 1, GLARE 2, GLARE 4 andGLARE 5 the composite laminate, i.e. the fibre/resin layer, arestacked symmetrically. In the case of GLARE 3 composite, the com-posite laminate have a cross-ply fibre layer stacked to the nearestouter aluminium layer of the laminate, in relation to the rollingdirection of the aluminium. For the GLARE 6 composite, the com-posite layers are stacked at +45� and �45�. Table 8 shows thesegrades, including the most important material advantages [1].GLARE is produced with a standard bonding technology. After thehot-curing (120 �C) cycle in the autoclave, the fibre-metal lami-nates carry a residual stress system over the thickness of the mate-rial, with a small tensile stress in the Al-sheets and compression inthe fibres: this is not favourable for fatigue behaviour. A post-stretching operation after the curing cycle can reverses the residualstress system and solve the problem, thus a tension stress occurs infibres and compression in the Al-sheets. But post-stretching causesextra production cost and it is practically impossible for wide pan-els [7]. Due to unique combination of properties, GLARE can be ap-

Fig. 5. Schematic illustration of a cross-ply GLARE laminates.

Table 7Identified advantages and disadvantages of various fibres for FMLs [93].

Fibre Advantage Disadvantage ConventionalFML

Aramid Outstanding toughness Weak in bending,buckling, compressionloading and transversetension

Excellent fatigueresistance in both-tensile and flexuralfatigue loading

Absorb moisture ARALL

High Young’s modulus Do not form strongbonds with othermaterials such ascomposite matrices

Low weight

Glass High tensile strength High weightHigh failure strain GLAREDo not absorb moisture Low stiffness


plied to wide range of applications. Manufacturers are giving lotsof attention on GLARE. However, it has been mostly used in aero-space applications so far. Recently, GLARE is used in the main fuse-lage skin and the leading edges of the horizontal and vertical tailplanes of new, high capacity Airbus A380. The latter applicationis a consequence of the excellent impact resistance of the FML

Table 8Commercially available GLARE grades [1,89].

Grade Sub Metal Type Metal Thickness (mm) Fibre Layer (mm)

GLARE 1 – 7475-T761 0.3–0.4 0.266

GLARE 2 GLARE 2A 2024-T3 0.2–0.5 0.266GLARE 2B 2024-T3 0.2–0.5 0.266

GLARE 3 – 2024-T3 0.2–0.5 0.266


GLARE 5 – 2024-T3 0.2–0.5 0.266


concept utilizing the high strength glass fibres with strain rate ef-fect [9,10,86,87]. With excellent impact characteristics, GLARE isbeing evaluated for use as cockpit crown, forward bulkheads, theleading edge and the flame-resistant capability of GLARE makesit suitable for flame sensitive areas such as; fire walls, cargo-liners,etc. in aerospace applications [7].

5.3. Carall

Fibre metal laminates consist of alternating layers of aluminiumalloy and adhesive impregnated fibres pregregs. Their superiorproperties depend strongly on the type of reinforcing fibres usedin pregregs. Each type has its advantages and limitations. As men-tioned before ARALL laminates are based on aramid fibres. Aramidfibres provide good specific strength and modulus, high impactresistance and outstanding toughness to ARALL laminates. How-ever, the poor compressive strength is a major limitation for thesehybrid composites. In this point of view, CARALL laminates hasdeveloped as an improvement of ARALL laminates. They containdifferent amount of carbon/epoxy pregregs instead of aramid/epoxy pregregs [93–95]. A schematic illustration of CARALL lami-nates is shown in Fig. 6.

Compared with aramid/epoxy, carbon/epoxy composites pos-sess higher specific modulus, but relatively low values of specificstrength, strain to failure and impact resistance. In terms of fatigue,it was recognized that aramid fibre composites have better low cy-cle fatigue performance but worse high cycle fatigue performance

Pregreg orientation in each fibre layer (�) Characteristics

0/0 Fatigue, strength, yield stress

0/0 Fatigue, strength90/90 Fatigue, strength

0/90 Fatigue, impact

0/90/0 Fatigue, strength, in 0� direction90/0/90 Fatigue, strength, in 90� direction

0/90/90/0 Impact, Shear, off-axis properties

+45/-45 Shear, off-axis properties-45/+45 Shear, off-axis properties

Fig. 6. A schematic illustration of CARALL laminates.


than carbon fibre composites. Moreover, the high stiffness of car-bon fibres allows for extremely efficient crack bridging and there-fore very low crack growth rates [90,96]. CARALL is producedsimilar to ARALL and GLARE laminates. Before the curing process,aluminium surfaces are treated for an optimal adhesion betweenaluminium alloy and epoxy resin. Then, they are cured in hot press.The combination of high stiffness and strength with good impactproperties gives CARALL laminates a great advantage for spaceapplications. Other applications for this laminate are impactabsorbers for helicopter struts and aircraft seats [96].

Fig. 7. Schematic view of compression loading on FML DNS specimen.
6. Test methods of fibre metal laminate composites
6.1. Bending tests

Mechanical properties of composite materials are governed bythe adhesion between fibre and matrix. Beside this, same proper-ties of FMLs are governed by the interface bond between compos-ite ply and metal ply. Determination of this adhesion is not an easytask therefore various test methods have been proposed in the lit-erature for this purpose: interlaminar shear and interfacial fracturetests. These methods give quality control information and are notsuitable for design specifications.

Interlaminar shear tests have been used to determine the inter-laminar shear strength (ILSS) of FMLs for a long time [97–105].According to the literature survey, there are three kinds of testmethods according to the interlaminar shear loading types: I. Com-pression loading [97], II. Three and five-point bending [98–101,103,105] and III. Short beam shear loading [102,104]. Thesetests will be explained individually.

Hinz et al. [97] investigated interlaminar shear properties ofFMLs by means of a double-notch shear test (DNS) according toASTM D3846 at room temperature. The interlaminar shear load be-tween the notches was primarily applied by compression on bothends of the DNS specimen as illustrated schematically in Fig. 7.According to author’s advice; for controlled deflection, a linearguided sliding carriage should be used. In this work a step motorapplied the compression load with a constant carriage speed0.06 mm/min. It is reported that the load was measured using aload cell with 500 N load capacities and recorded by means of a

Kyowa CDV-700A measuring amplifier. The compressive stress iscalculated using the cross section of the specimen. To calculatethe applied interlaminar shear stress the effective area of the shearplane was used.

The other test methods applied for ILSS investigation of FMLsare three and five-point bending tests (Fig. 8).

Lawcock et al. determined the ILSS of FMLs by three and five-point bending tests with a span of 10 mm at a crosshead speedof 1.3 mm/min [99]. The authors calculated the ILSS by Eq. (1) forthree-point bending and five-point bending:

ILSS ¼ ð33PcÞ=ð64WtÞ ð1Þ

where ‘‘Pc’’ is the initial load for interlaminar failure and ‘‘W’’ and ‘‘t’’are the width and thickness of the specimens, respectively. In an-other work; Lawcock et al. calculated ILSS of FMLs from three-pointand five-point bending tests as follows [101]:

—Three-point bending : ILSS ¼ ð3PmaxÞ=4Wt ð2Þ

—Five-point bending : ILSS ¼ ð33PmaxÞ=ð64WtÞ ð3Þ

where ‘‘Pmax’’ is the failure load and ‘‘W’’ and ‘‘t’’ are the width andthickness of the specimens, respectively.

Reyes and Cantwell also reported their investigation about ILSSof FMLs by means of three-point bending test [103]. These testswere undertaken at crosshead displacement rates between 0.1and 3 m/s and were stopped once a visible crack had propagatedfrom one of the starter defects. In all these studies; the authors

Fig. 8. Schematic view of three and five-point bending.


were not mentioned about test standards. However; Khalili et al.[105] reported that three-point bending tests were carried out onZwick 1484 by using specifications of ASTM D790 M-93.

The last test method for investigating ILSS of FMLs is short beamshear test. Park et al. performed ILSS tests to evaluate the extent ofhygrothermal degradations as a function of water absorption andthermal cycles [102]. ILSS of FMLs was evaluated by the ASTMD2344 standard specimen through a short beam shear test. Theyreported that tests were carried out with a crosshead speed of1.3 mm/min. Botelho et al. [104] also used the same test standard(ASTM D2344) for investigating the ILSS of FMLs. Authors reportedthat the tests were performed with Instron mechanical testing ma-chine using a test speed of 1.3 mm/min.

The next test method group for investigating the interfacebonding between composite and metal plies is as mentioned abovethe interfacial fracture tests. In these tests, the degree of adhesionbetween the composite and metal plies was investigated using thesingle cantilever beam (SCB) geometry [103,106,107]. This geome-try yields mixed-mode I/II loading (tension/shear) conditions at thecrack tip in the sample. It is reported in these papers that SCB spec-imens with dimensions of 20 mm � 200 mm were removed fromthe moulded plates. The SCB specimens were clamped in the testfixture shown in Fig. 9 and tested at crosshead displacement rateof 2 mm/min.

According to these studies the interfacial fracture energy wasdetermined using an experimental compliance method where theinterfacial fracture energy, ‘‘Gc’’ is given by:

Gc ¼ ð3P2ka2Þ=2B ð4Þ

where ‘‘P’’ is the applied force, ‘‘B’’ the specimen width, ‘‘C’’ thespecimen compliance and ‘‘a’’ the crack length. Also it is mentionedthat the specimen compliance was determined as a function ofcrack length and a curve fit of the following form applied:

C ¼ C0 þ ka3 ð5Þ

where ‘‘C0’’ and ‘‘k’’ are constants for a given specimen. Here, it istold that the parameter ‘‘k’’ was determined by measuring the gra-dient of the plot of the compliance ‘‘C’’ versus the cube of the cracklength ‘‘a3’’ whereas it was not necessary to determine the values of‘‘C0’’.

6.2. Fatigue tests

The fatigue properties of fibre metal laminates have been eval-uated in numerous test programs. Generally fatigue tests were

Fig. 9. Schematic illustration of SCB geometry.

conducted according to the ASTM standards. The shape and dimen-sions of specimens for fatigue testing were based on ASTM D3479;the specimen length L = 200 mm, the gauge length LG = 100 mm,the specimen width W = 25 mm [9,91,108]. Beside this Reyes andKanga were conducted ASTM E466 test standard for analyzing fati-gue behaviour of FML [109]. Except these papers there are manystudies that fatigue performance of FMLs was not investigatedaccording to these standards [8,87,89,96,106,110–119]. In thesepapers specimens were prepared as notched or cracked for the ini-tiation of crack. For fatigue tests; servo hydraulic test machineswith approximately a load capacity of 5, 10 and 25 metric tonswere used. All constant amplitude fatigue tests were conductedin tension–tension or tension–compression loading at a frequencyof 10 Hz in lab-air at room temperature [9,87,89,91,108,106,117,118]. Beside this in some papers fatigue tests were conductedwith sinusoidal waves of frequency 5 Hz [113,114]. During tests;the fatigue cracks were observed by various ways; charge coupleddevice (CCD) camera [110,115], visual observation [111], digitalcamera [112], clip gage [114], travelling optical microscope[106,116,119].

6.3. Tensile tests

The tensile behaviour of FMLs was generally investigatedaccording to ASTM D3039 [93,99,105,107,120–126]. Beside thisKawai and Arai [123] were benefit from JIS K7073 standard for con-ducting tensile testing of FMLs. In all papers dealt with tensileproperties of FMLs, the test specimen shapes and dimensions aredifferent. Some of them used rectangular tensile specimens[6,106,121] and some of them used dog bone shape tensile speci-mens [101,109,118,124]. Beside this; there are a few papers thattensile test specimen dimensions were prepared according to thestandards. In these papers; the authors were not referenced theirtensile specimen dimensions. For tensile tests; universal and servohydraulic testing machines were used: Instron tensile testing ma-chines (4469, 1195, 8502, 4204, 1125, 1251) [96,105,107,109,120–122,127], MTS tensile testing machines (810, RT/50) [99,121,123–126] and Shimadzu tensile testing machine [121]. Tensile testswere performed at room temperature at various crosshead speedsof 0,1 mm/min [124], 0,5 mm/min [96], 1 mm/min [103,93,109,120,121,123,127], 1,27 mm/min [126], 2 mm/min [6,106,125,36],5 mm/min [105]. However; in Carrillo and Cantwell’s studies ten-sile behaviour of FMLs were investigated at a constant strain rateof 0.04 min�1 on an Instron 4204 universal test machine accordingto the ASTM D3039 test standard [107,122]. It is worthy to say thatduring tensile tests the Young’s modulus of each specimen wasrecorded using a clip-on extensometer which was incapable ofmeasuring the complete stress–strain curve.

6.4. Low and High velocity impact and blast loading tests

Due to the importance of understanding the dynamic loadingeffect on the FMLs, low and high velocity and also blast loadingtests should be performed. According to the literature survey, itis well understood that there are many studies dealt with thesetest methods applied to the FMLs. Among these papers some of


them explain low velocity impact behaviour of FMLs[103,106,13,128,129], some of them dealt with high velocity im-pact performance of FMLs [13,128–131] and the others explainthe blast loading effect on the FMLs [132–137].

6.4.1. Low velocity impact testsFirst of all; the low velocity impact studies shall be mentioned.

In these studies it is seen that all the low velocity impact tests wereperformed by drop weight impact tester. Tests were carried outusing drop hammers that have various masses: 2000 g [103,106],1160 g [128] and 575 g [13,129]. In addition to this, indenter typeswere chosen as similar (hemispherical) but the diameters are dif-fering from each other as following: 5 mm [107], 10 mm [106],12.7 mm [128], 15 mm [13,129]. For the investigation of low veloc-ity impact performance; square FML samples were generally pre-pared as 75 mm � 75 mm, 100 mm � 100 mm, 125 mm �125 mm and clamped between two square or circular frames.Square frames have 75 mm � 75mm or 100 mm � 100 mm inter-nal openings [106,128]. Beside this circular frames have internaldiameters of 70 mm [107], 75 mm [103] and 80 mm [13]. The im-pact tests are instrumented as the contact force is measured with aload cell in the nose of the impactor and the strain in the specimenis measured with a strain gauge. After an impact, the carriage wascaught in order to avoid secondary impacts.

6.4.2. High velocity impact testsAfter an explanation about low velocity impact tests, in this sec-

tion high velocity impact tests applied to FMLs shall be explainedin detail. According to literature survey, it is very clear that gener-ally a gas gun consisted of a pressure vessel was used for highvelocity impact tests [13,128–131]. Generally square plates wereclamped in a steel support square aperture so that the impactorcan hit the centre of the test specimen. In this test method; afterburning through a membrane of the gas gun, the expanding gasor air accelerates a projectile or a steel ball bearing to the velocitiesranging between 25 and 100 m/s. The velocity of impactor prior toimpact can be measured as it exited the barrel by the interruptionof two laser beams a known distance apart by using light emittingdiode photovoltaic cell pairs.

The ballistic limit is defined as the lowest velocity for completepenetration or perforation. Impact testing should be conductedover a range of impact energies until complete perforation of theFML target was achieved. The resulting perforation energy wasthen could be used to calculate the specific perforation energy ofthe target by normalizing the measured perforation energy bythe areal density of the target [131]. After testing, the specimenscould be sectioned, polished and then viewed under an opticalmicroscope in order to elucidate the failure mechanisms duringimpact.

6.4.3. Blast loading testsNumerous studies [132,134–138] have investigated the local-

ized blast loading of FMLs. Blast loading was created using a plasticexplosive (PE4) shaped into a circular disc and positioned at thecentre of the plate, on top of a polystyrene foam (12–14 mm)pad to attenuate the blast. PE4 consists of 88% RDX and 12% lith-ium grease, has a detonation velocity of 8200 m/s and a TNT equiv-alency of 1.3 [132]. The detonator should be attached to the centreof the disc using 1 g of explosive, referred to as a ‘leader’. The im-pulse could be varied by varying the diameter of the explosivestrips. The FML panels 300 mm � 300 mm [136], 400 mm �400 mm [134,138] were clamped between two steel frames leav-ing an exposed area of 200mm � 200 mm [132,136], 300 mm �300mm [134,135,138]. The frames were mounted onto a ballisticpendulum. Each FML panel should be weighted and the thicknessshould be measured at four different points before testing. After

a blast; the swing of the pendulum could be used to calculatethe impulse applied to the panel by detonating the explosive.

7. Conclusions

In this paper, historical development, advantages, disadvan-tages, production and applications of FMLs were presented byexamining recent studies. In 1978 initial FMLs, ARALL laminateswhich consist alternating thin aluminium layers and aramid fibrepregreg were introduced. Later GLARE laminates based on highstrength glass fibres and CARALL laminates based on carbon fibreswere developed. GLARE laminates are the successor of ARALL lam-inates and commercially production of GLARE laminates werestarted in 1991. ARALL, GLARE and CARALL laminates are nowbeing used as structural materials in aircrafts. Yet, it can be saidthat GLARE laminates found more use than ARALL and CARALLlaminates in aircraft applications. Both laminates production in-volves five major activities; (i) surface treatment of metallic sur-face in order to improve bonding between metallic layer andfibre reinforced pregreg, (ii) material deposition, (iii) cure prepara-tion, (iv) cure process, (v) post stretching for reserving residualstress of FML which caused by curing process. FMLs provides supe-rior mechanical properties compared to fibre reinforced compos-ites and aluminium alloys. High strength, high elasticity moduluswith improved toughness and excellent fatigue properties can besaid as major advantages of FMLs. ARALL, GLARE and CARALL lam-inates explained and compared with each other in followingchapters.

In addition to the importance of reinforcement and matrix inpolymer composites, the bonding between the composite laminaeand the metallic layer is a key issue for the overall metal–fibre lam-inate performance. An adequate surface treatment of the metalliclayer is required to assure a good mechanical and adhesive bondbetween the composite laminates and metal surfaces. In this paperthe effects of surface treatments which are improved the metallicsurface morphology for a better bonding with composite laminatesare investigated. Surface treatment methods, such as mechanical,chemical, electrochemical, coupling agents and dry surface treat-ments are introduced and showed comparable performance to im-prove fibre metal laminates.

Mechanical properties of FMLs are being enhanced by the inter-face bond between composite and metal plies. This enhancementcould be controlled by various test methods. This control reportsgive quality information and are suitable for design specifications.In this review; test methods of bending, fatigue, tensile, low andhigh velocity impact and blast loading tests for determining themechanical properties of FMLs and research studies utilized fromthese test methods were explained in detail.

8. Developments for the future

FMLs consist of metallic alloy and fibre reinforced prepreg.Mostly commercially avaible GLARE, ARALL and CARALL consistvarious aluminium alloys. Many researches have been trying touse possible metallic alloys such as magnesium, titanium, etc. in-stead of aluminium alloys. It is expected that this diversity givesoptimum mechanical properties. Same efforts have been examinedfor engineering polymeric materials to replace fibre reinforcedprepreg.

Long processing cycle to cure the polymer matrix in the com-posite plies increases the cycle time of whole production and de-creases productivity. New processing methods are investigatedfor improving the productivity of curing process and decreasingthe labour costs of FMLs. These improvements will make FMLs very


attractive to various industrial applications (military, automotiveand aircraft).

Last decades many researchers have been investigating the pos-sibility of using thermoplastic materials instead of thermosetmaterials as matrix materials in FMLs. By using thermoplastic ma-trix, FMLs will find new application areas. However low compati-bility of thermoplastic matrix with metal surfaces needs to beimproved by surface modification methods explained in thisreview.

References

[1] Botelho EC, Silva RA, Pardini LC, Rezende MC. A review on the developmentand properties of continuous fiber/epoxy/aluminum hybrid composites foraircraft structures. Mater Res 2006;9(3):247–56.

[2] Bernhardt S, Ramulu M, Kobayashi AS. Low-velocity impact responsecharacterization of a hybrid titanium composite laminate. J Eng MaterTechnol 2007;129:220–6.

[3] Villanueva GR, Cantwell WJ. The high velocity impact response of compositeand FML-reinforced sandwich structures. Compos Sci Technol2004;64:35–54.

[4] Beumler T, Pellenkoft F, Tillich A, Wohlers W, Smart C. Airbus costumerbenefit from fiber metal laminates. Airbus Deutschland GmbH; 2006 May,Ref. no: L53pr0605135-Issue 1. p. 1–18.

[5] Alderliesten RC, Benedictus R. Fiber/metal composite technology for futureprimary aircraft structures. In: 48th Aiaa/Asme/Asce/Ahs/Asc structures,structural dynamics, and materials conference 15th; April 23–26, 2007;Honolulu, Hawaii; 2007. p. 1–12.

[6] Cortes P, Cantwell WJ. The prediction of tensile failure in titanium-basedthermoplastic fibre–metal laminates. Compos Sci Technol 2006;66:2306–16.

[7] Asundi A, Choi Alta YN. Fiber metal laminates: an advanced material forfuture aircraft. J Mater Process Technol 1997;63:384–94.

[8] Vogelesang LB, Vlot A. Development of fibre metal laminates for advanced. JMater Process Technol 2000;103:1–5.

[9] Chang PY, Yeh PC, Yang JM. Fatigue crack initiation in hybrid boron/glass/aluminum fiber metal laminates. Mater Sci Eng 2008;A 496:273–80.

[10] Alderliesten R. On the development of hybrid material concepts for aircraftstructures. Recent Patents Eng 2009;3:25–38.

[11] Remmers JJC, Borst RD. Delamination buckling of fibre–metal laminates.Compos Sci Technol 2001;61:2207–13.

[12] Schut J, Alderliesten RC. Delamination growth rate at low and elevatedtemperatures in glare. In: 25th International congress of the aeronauticalsciences, 2006 September 3–8; Hamburg, Germany: 2006. p. 1–7.

[13] Vlot A. Impact loading on fibre metal laminates. Int J Impact Eng1996;18(3):291–307.

[14] Castrodeza EM, Bastian FL, Perez JEI. Critical fracture toughness, Jc and D5c, ofunidirectional fibre–metal laminates. Thin-Walled Struct 2003;4:1089–101.

[15] Vogelesang B, Gunnink JW, Roebroeks GHJJ, Muè Ller RPG. Toward thesupportable and durable aircraft fuselage structure. In: Grandage JM, Jost GS,editors. Proceedings of the 18th symposium of the international committeeon aeronautical fatigue, 1995 May 3–5, Melbourne, Australia; 1995. p. 257–272.

[16] Cortes P, Cantwell WJ. Fracture properties of a fiber–metal laminates basedon magnesium alloy. J Mater Sci 2004;39:1081–3.

[17] Roebroeks GHJJ. Fibre–metal laminates recent developments andapplications. Fatigue 1994;16(1):33–42.

[18] Yehia A, Bahei-El-Din A, Amany G, Botrous B. Analysis of progressive fiberdebonding in elastic laminates. Int J Solids Struct 2003;40:7035–53.

[19] Park SY, Choi WJ, Choi HS, Kwon H, Kim SH. Recent trends in surfacetreatment technologies for airframe adhesive bonding processing: a review(1995–2008). J Adhes 2010;86:192–221.

[20] Guo YJ, Wu XR, Zhang ZL. Characterization of delamination growth behaviourof hybrid bonded laminates. Fatigue Fract Eng Mater Struct1997;20(12):1699–708.

[21] Davis M, Bond D. Principles and practices of adhesive bonded structural jointsand repairs. Int J Adhes Adhes 1999;19:91–105.

[22] di Scalea FL, Bonomo M, Tuzzeo D. Ultrasonic guided wave inspection ofbonded lap joints: noncontact method and photoelastic visualization. ResNondestr Eval 2001;13(3):153–71.

[23] Critchlow GW, Brewis DM. Review of surface pretreatments for aluminiumalloys. Int J Adhes Adhes 1996;16(4):255–75.

[24] Critchlow GW, Yendall KA, Bahrani D, Quinn A, Andrews F. Strategies for thereplacement of chromic acid anodising for the structural bonding ofaluminium alloys. Int J Adhes Adhes 2006;26:419–53.

[25] Sang Park SY, Choi WJ, Choi HS, Kwon H, Kim SH. Effects of surface pre-treatment and void content on GLARE laminate process characteristics. JMater Process Technol 2010;210:1008–16.

[26] Liu J, Chaudhury MK, Berry DH, Seeberg JE, Osborne JH, Blohowiak KY. Effectof surface morphology on crack growth at a sol–gel reinforced epoxy/aluminium interface. J Adhes 2006;82:487–516.

[27] Harris AF, Beevers A. The effects of grit-blasting on surface properties foradhesion. Int J Adhes Adhes 1999;19:445–52.

[28] Rider AN, Arnott DR. Boiling water and silane pre-treatment of aluminiumalloys for durable adhesive bonding. Int J Adhes Adhes 2000;20:209–20.

[29] Rider AN, Olsson-Jacques CL, Arnott DR. Influence of adherend surfacepreparation on bond durability. Surf Interface Anal 1999;27:1055–63.

[30] Prolongo SG, Uren A. Effect of surface pre-treatment on the adhesive strengthof epoxy–aluminium joints. Int J Adhes Adhes 2009;29:23–31.

[31] Lefebvre DR, Ahn BK, Dillard DA, Dillard JG. The effect of surface treatmentson interfacial fatigue crack initiation in aluminum/epoxy bonds. Int J Fract2002;114:191–202.

[32] Brewis DM, Critchlow GW. Locus of failure of T-peel joints formed betweenaluminium and various adhesives. Int J Adhes Adhes 1997;17:33–8.

[33] Rider AN. Factors influencing the durabilityof epoxyadhesion to silanepretreated aluminium. Int J Adhes Adhes 2006;26:67–78.

[34] Sheasby PG, Pinner R. The surface treatment and finishing of aluminium andits alloys. 6th ed. Materials Park (OH): ASM International/Stevenage, Herts(UK): Finishing Publications Ltd.; 2001.

[35] Carrino L, Napolitano G, Sorrentino L. Wettability improving of 2024aluminium alloy by oxygen cold plasma treatment. Int J Adv ManufactTechnol 2006;31:465–73.

[36] Bjorgum A, Lapique F, Walmsley J, Redford K. Anodising as pre-treatment forstructural bonding. Int J Adhes Adhes 2003;2:401–12.

[37] Bishopp A. Handbook of adhesives and sealants. Amsterdam: Elsevier; 2005.[38] Kinloch AJ, Little MSG, Watts JF. The role of the interphase in the

environmental failure of adhesive joints. Acta Mater 2000;48:4543–53.[39] Digby RP, Packham DE. Pretreatment of aluminium: topography, surface

chemistry and adhesive bond durability. Int J Adhes Adhes 1995;15(2):61–71.[40] Hadavinia H, Kinloch AJ, Little MSG, Taylor AC. The prediction of crack growth

in bonded joints under cyclic-fatigue loading I. Experimental studies. Int JAdhes Adhes 2003;23:449–61.

[41] Vine K, Cawley P, Kinloch AJ. The correlation of non-destructivemeasurements and toughness changes in adhesive joints duringenvironmental attack. J Adhes 2001;77(2):125–61.

[42] Minford JD. Handbook of aluminum bonding technology and data. NewYork: Marcel Dekker; 1993.

[43] Lunder O, Olsen B, Nisancioglu K. Pre-treatment of AA6060 aluminium alloyfor adhesive bonding. Int J Adhes Adhes 2002;22:143–50.

[44] Domingues L, Fernandes JCS, Da Cunha Belo M, Ferreira MGS, Guerra-Rosa L.Anodising of Al 2024-T3 in a modified sulphuric acid/boric acid bath foraeronautical applications. Corros Sci 2003;45:149–60.

[45] Mertens T, Kollek H. On the stability and composition of oxide layers on pre-treated titanium. Int J Adhes Adhes 2010;30:466–77.

[46] Oosting R. Toward a new durable and environmentally compliant adhesivebonding process for aluminum alloys. PhD thesis, Delft University ofTechnology, The Netherlands; 1995.

[47] Fedel M, Olivier M, Poelman M, Deflorian F, Rossi S, Druart ME. Corrosionprotection properties of silane pre-treated powder coated galvanized steel.Prog Org Coat 2009;66:118–28.

[48] Zucchi F, Trabanelli G, Grassi V, Frignani A. Proceeding of the EUROCORR.2001; 30 September–4 October; Riva del Garda, Italy; 2001.

[49] Abel ML, Digby RP, Fletcher IW, Watts JF. Evidence of specific interactionbetween c-glycidoxypropyltrimethoxysilane and oxidized aluminium usinghigh-mass resolution to F-SIMS. Surf Interface Anal 2000;29(2):115–25.

[50] Hobbs PM, Kinloch AJ. The computational molecular modelling oforganosilane primers. J Adhes 1998;66:203–28.

[51] Crook RA, Sinclair JW, Poulter LW, Schulte KJ. An environmentally-friendlyprocess for bonding aluminum using aqueous metasilicate sol–gel and silaneadhesion promoters. J Adhes 1998;68(3 and 4):315–29.

[52] DeGraevea I, Tourwé EM, Biesemans M, Willem R, Terryn H. Silane solutionstability and film morphology of water-based bis-1,2-triethoxysilyl)ethanefor thin-film deposition on aluminium. Prog Org Coat 2008;63:38–42.

[53] Varma PCR, Colreavy J, Cassidy J, Oubaha M, Duffy B, McDonagh C. Effect oforganic chelates on the performance of hybrid sol–gel coated AA 2024-T3aluminium alloys. Prog Org Coat 2009;66:406–11.

[54] Critchlow GW, Brewis DM, Emmony DC, Cottam CA. Initial investigation intothe effectiveness of CO2-laser treatment of aluminium for adhesive bonding.Int J Adhes 1995;15(4):233–6.

[55] Mazza JJ, Gaskin GB, Piero WSD, Blohowiak KY. Sol-gel technology for low-VOC, nonchromated adhesive bonding applications. Report no.: AFRL-ML-WP-TR-2004-4063; AFRL Wright-Patterson AFB; Ohio, 2004.

[56] Voast PV, Blohowiak K. Critical materials & processes bonded joint issues. In:2004, FAA Workshop on Bonded Structures 2004 June 16–18; Seattle, WA;2004.

[57] Osborne JH, Blohowiak KY, Sekits DF. Environmentally benign sol-gel surfacetreatment for aluminum bonding applications. Report no.: WL-TR-95-4060;AFRL Wright-Patterson AFB, Ohio; 1996.

[58] Critchlow GW, Cottam CA, Brewis DM, Emmony DC. Further studies into theeffectiveness of CO2-laser treatment of metals for adhesive bonding. Int JAdhes Adhes 1997;17(2):143–50.

[59] Walters CT. Laser surface preparation for adhesive bonding II. Report no.AFRL-ML-WP-TR-2006-4139; AFRL Wright-Patterson AFB, Ohio; 2004.

[60] Spadaro C, Sunseri C, Dispenza C. The influence of the nature of the surfaceoxide on the adhesive fracture energy of aluminium bonded joints asmeasured by T-peel tests. Int J Adhes Adhes 2007;76:1441–6.

[61] Wong RCP, Hoult AP, Kim JK, Yu TX. Improvement of adhesive bonding inaluminium alloys using a laser surface texturing process. J Mater ProcessTechnol 1997;63:579–84.


[62] Mazza JJ. Advanced surface preparation for metal alloys. In: 83rd Meeting ofthe AGARD Structures and Materials Panel; 1996 September 2–6; Florence,Italy; 1996. p. 10-1–10-12.

[63] Davis GD, Groff GB, Zatorski RA. Plasma spray coatings as treatments foraluminum, titanium and steel adherends. Surf Interface Anal1997;25(5):366–73.

[64] Fernandes JCS, Ferreira MGS, Haddow DB, Goruppa A, Short R, Dixon DG.Plasma-polymerised coatings used as pre-treatments for aluminium alloys.Surf Coat Technol 2002;154:8–13.

[65] Alexander MR, Zhou X, Thompson GE, Duc TM, McAlpine E, Tielsch BJ.Functionalized plasma polymer coatings for improved durability ofaluminium–epoxy adhesive joints: fractography. Surf Interface Anal2000;30:16–20.

[66] Anagreh N, Dorn L, Bilke-Krause C. Low-pressure plasma pretreatment ofpolyphenylene sulfide (PPS) surfaces for adhesive bonding. Int J Adhes Adhes2007;28(1–2):16–22.

[67] Polini W, Sorrentino L. Improving the wettability of aluminium alloy 2024 bymeans of cold plasma. Appl Surf Sci 2003;214:232–42.

[68] Koch GH, Todd GL, Deutchman A, Partyka R. Nonchemical surface treatmentfor aluminum alloys. Report no.: WL-TR-96-4084; AFRL Wright-PattersonAFB, Ohio; 1996.

[69] De Iorio I, Leone C, Nele L, Tagliaferri V. Plasma treatments of polymericmaterials and Al alloy for adhesive bonding. J Mater Process Technol1997;68:179–83.

[70] Niu MCY. Composite airframe structures, 2nd published. HongKong: Conmilit Press Ltd.; 1996.

[71] Kupke M, Kolax M. CFRP-fuselage – ensuring future competitiveness. In:Material & process technology – the driver for tomorrow’s improvedperformance. Proceeding of the 25th jubilee international SAMPE Europeconference 2004 of the society for the advancement of materials and processengineering Paris EXPO; 2004 March 30–April 1, Porte de Versailles, Paris;2004. p. 432–7.

[72] Jackson P. Airbus A380, fixed-wing civil. In: Jane’s all the world’s aircraft2005–2006, 96th ed. Coulsdon, Surrey CRS 2YH, UK; 2005 p. 241–6.

[73] Kolesnikov B, Herbeck L, Fink A. CFRP/titanium hybrid material for improvingcomposite bolted joints. Compos Struct 2008;83:368–80.

[74] Hoang-Ngoc CT, Paroissien E. Simulation of single-lap bonded and hybrid(bolted/bonded) joints with flexible adhesive. Int J Adhes Adhes2010;30:117–29.

[75] Hollmann K. Failure analysis of bolted composite joints exhibiting in-planefailure modes. J Compos Mater 1996;30(3):358–83.

[76] Hung CL, Chang FK. Strength envelope of bolted composite joints underbypass loads. J Compos Mater 1996;30(13):1402–35.

[77] Oh JH, Kim YG, Lee DG. Optimum bolted joints for hybrid compositematerials. Compos Struct 1997;38(1–4):329–41.

[78] Camanho PP, Matthews FL. Delamination onset prediction in mechanicallyfastened joints in composite laminates. J Compos Mater 1999;33(10):906–27.

[79] Dano ML, Gendron G, Picard A. Stress and failure analysis of mechanicallyfastened joints in composite laminates. Compos Struct 2000;50:287–96.

[80] Starikov R, Schön J. Quasi-static behaviour of composite joints withprotruding-head bolts. Compos Struct 2001;51:411–25.

[81] Lawlor VP, McCarthy MA, Stanley WF. An experimental study of bolt-holeclearance effects in double-lap, multi-bolt composite joints. Compos Struct2005;71:176–90.

[82] Nadler MA, Yoshino SY, Darms FJ. Boron/epoxy support strut for non-integralcryogenic tankage. In: Materials and processes, 15th SAMPE symposium, vol.15, Los Angeles; 1969. p. 179–207.

[83] Gunjaev GM, Shelesina GF, Iltshenko SI. Metal laminates with aluminium andtitanium alloys. Aviation materials and technologies. Moscow: VIAM; 2002. p.50–8 [in Russian].

[84] Kolesnikov B, Herbeck L, Fink A. Fortschrittliche Verbindungstechniken vonFaserverbunden. In: DGLR-Kongress, Dresden, Band II, 20–23 September,2004. p. 1419–28.

[85] Sun CT, Dicken A, Wut HF. Characterization of impact damage in ARALLlaminates. Compos Sci Technol 1993;49:139–44.

[86] Lee SM, editor. Handbook of composite reinforcements. Wiley VCH PublishersInc.; 1993. pp. 1–6.

[87] Shima DJ, Alderliesten RC, Spearing SM, Burianek DA. Fatigue crack growthprediction in GLARE hybrid laminates. Compos Sci Technol 2003;63:1759–67.

[88] Burianek DA, Giannakopoulos AE, Spearing SM. Modeling of facesheet crackgrowth in titanium–graphite hybrid laminates, part I. Eng Fract Mech2003;70:775–98.

[89] Khan SU, Alderliesten RC, Benedictus R. Post-stretching induced stressredistribution in fibre metal laminates for increased fatigue crack growthresistance. Compos Sci Technol 2009;69:396–405.

[90] Vlot A. Fibre metal laminates: an introduction. Kluwer Academic Publishers;2001.

[91] Kawai M, Hachinohe A. Two-stress level fatigue of unidirectional fiber–metalhybrid composite: glare 2. Int J Fatigue 2002;24:567–80.

[92] Castrodeza EM, Abdala MRWS, Bastian FL. Crack resistance curves of glarelaminates by elastic compliance. Eng Fract Mech 2006;73:2292–303.

[93] Moussavi-Torshizi SE, Dariushic S, Sadighic M, Safarpour P. A study on tensileproperties of a novel fiber/metal laminates. Mater Sci Eng A2010;527:4920–5.

[94] Lin CT, Kao PW. Fatigue delamination growth in carbon fibre-reinforcedaluminium laminates. Composites A 1996;27A:9–15.

[95] Xia Y, Wang Y, Zhou Y, Jeelani S. Effect of strain rate on tensile behavior ofcarbon fiber reinforced aluminum laminates. Mater Lett 2007;61:213–5.

[96] Lin CT, Kao PW, Yang FS. Fatigue behaviour of carbon fibre-reinforcedaluminium laminates. Composites 1991;22(2):135–41.

[97] Hinz S, Omoori T, Hojo M, Schulte K. Damage characterisation of fibre metallaminates under interlaminar shear load. Composites A 2009;40:925–31.

[98] Cepeda-Jiménez CM, Alderliesten RC, Ruano OA, Carreño F. Damage toleranceassessment by bend and shear tests of two multilayer composites: glass fibrereinforced metal laminate and aluminium roll-bonded laminate. Compos SciTechnol 2009;69:343–8.

[99] Lawcock GD, Ye L, Mai YW, Sun CT. Effects of fibre/matrix adhesion oncarbon-fibre-reinforced metal laminates I. Residual strength. Compos SciTechnol 1997;57:1609–19.

[100] Afaghi-Khatibi A, Lawcock G, Ye L, Mai YW. On the fracture mechanicalbehaviour of fibre reinforced metal laminates (FRMLs). Comput MethodsAppl Mech Eng 2000;185:173–90.

[101] Lawcock G, Ye L, Mai YW, Sun CT. The effect of adhesive bonding betweenaluminum and composite prepreg on the mechanical properties of carbon-fiber reinforced metal laminates. Compos Sci Technol 1997;1997:35–45.

[102] Park SY, Choi WJ, Choi HS. The effects of void contents on the long-termhygrothermal behaviors of glass/epoxy and GLARE laminates. Compos Struct2010;92:18–24.

[103] Reyes VG, Cantwell WJ. The mechanical properties of fibre-metal laminatesbased on glass fibre reinforced polypropylene. Compos Sci Technol2000;60:1085–94.

[104] Botelho EC, Rezende MC, Pardini LC. Hygrothermal effects evaluation usingthe Iosipescu shear test for glare laminates. J Braz Soc Mech Sci Eng2008;30(3):213.

[105] Khalili SMR, Mittal RK, Kalibar GS. A study of the mechanical properties ofsteel/aluminium/GRP laminates. Mater Sci Eng A 2005;412:137–40.

[106] Cortes P, Cantwell WJ. The fracture properties of a fibre–metal laminatebased on magnesium alloy. Composites B 2006;37:163–70.

[107] Carrillo JG, Cantwell WJ. Mechanical properties of a novel fiber–metallaminate based on a polypropylene composite. Mech Mater 2009;41:828–38.

[108] Kawai M, Kato K. Effects of R-ratio on the off-axis fatigue behavior ofunidirectional hybrid GFRP/Al laminates at room temperature. Int J Fatigue2006:1226–38.

[109] Reyes G, Kang H. Mechanical behavior of lightweight thermoplastic fiber–metal laminates. J Mater Process Technol 2007;186:284–90.

[110] Khan SU, Alderliesten RC, Benedictus R. Delamination growth in fibre metallaminates under variable amplitude loading. Compos Sci Technol2009;69:2604–15.

[111] Homan JJ. Fatigue initiation in fibre metal laminates. Int J Fatigue2006;28:366–74.

[112] Rodi R, Alderliesten R, Benedictus R. Experimental characterization of thecrack-tip-opening angle in fibre metal laminates. Eng Fract Mech2010;77:1012–24.

[113] Takamatsu T, Matsumura T, Ogura N, Shimokawa T, Kakuta Y. Fatigue crackgrowth properties of a GLARE3-5/4 fiber/metal laminate. Eng Fract Mech1999;63:253–72.

[114] Takamatsu T, Shimokawa T, Matsumura T, Miyoshi Y, Tanabe Y. Evaluation offatigue crack growth behavior of GLARE3 fiber/metal laminates using acompliance method. Eng Fract Mech 2003;70:2603–16.

[115] Alderliesten R, Rans C. The meaning of threshold fatigue in fibre metallaminates. Int J Fatigue 2009;31:213–22.

[116] Kim CW, Oh DJ. Progressive delamination with and without crackpropagation in aramid fiber reinforced metal laminates containing acircular notch. Mater Sci Eng A 2008;483–484:251–3.

[117] Austin TSP, Singh MM, Gregson PJ, Powell PM. Characterisation of fatiguecrack growth and related damage mechanisms in FRP–metal hybridlaminates. Compos Sci Technol 2008;68:1399–412.

[118] Johnson WS, Hammond MW. Crack growth behavior of internal titaniumplies of a fiber metal laminate. Composites Part A 2008;39:1705–15.

[119] Lin CT, Kao PW. Effect of fiber bridging on the fatigue crack propagation incarbon fiber-reinforced aluminum laminates. Mater Sci Eng A1995;190:65–73.

[120] Wu G, Tan Y, Yang JM. Evaluation of residual strength of notched fiber metallaminates. Mater Sci Eng A 2007;457:338–49.

[121] Ye L, Afaghi-Khatibi A, Lawcock G, Mai YW. Effect of fibre/matrix adhesion onresidual strength of notched composite laminates. Composites Part A1998;29A:1525–33.

[122] Carrillo JG, Cantwell WJ. Scaling effects in the tensile behavior of fiber-metallaminates. Compos Sci Technol 2007;67:1684–93.

[123] Kawai M, Arai Y. Off-axis notched strength of fiber–metal laminates and aformula for predicting anisotropic size effect. Composites Part A2009;40:1900–10.

[124] de Vries TJ, Vlot A, Hashagen F. Delamination behavior of spliced fiber metallaminates part 1. Experimental results. Compos Struct 1999;46:131–45.

[125] Iaccarino P, Langella A, Caprino G. A simplified model to predict the tensileand shear stress–strain behaviour of fibreglass/aluminium laminates.Compos Sci Technol 2007;67:1784–93.

[126] Botelho EC, Almeida RS, Pardini LC, Rezende MC. Elastic properties ofhygrothermally conditioned glare laminate. Int J Eng Sci 2007;45:163–72.

[127] Caprino G, Squillace A, Giorleo G, Nele L, Rossi L. Pin and bolt bearingstrength of fibreglass/aluminium laminates. Composites Part A2005;36:1307–15.


[128] Lawcock GD, Ye L, Mai YW, Sun CT. Effects of fibre/matrix adhesion oncarbon-fibre-reinforced metal laminates – II. Impact behaviour. Compos SciTechnol 1997;57:1621–8.

[129] Vlot A. Impact properties of fibre metal laminates. Compos Eng1993;3(10):911–27.

[130] Michelle S, Fatt H, Lin C, Revilock Jr DM, Hopkins DA. Ballistic impact ofGLARE fiber–metal laminates. Compos Struct 2003;61:73–88.

[131] Abdullah MR, Cantwell WJ. The impact resistance of polypropylene-basedfibre–metal laminates. Compos Sci Technol 2006;66:1682–93.

[132] Langdon GS, Nurick GN, Cantwell WJ. The response of fibre metal laminatepanels subjected to uniformly distributed blast loading. Eur J Mech A/Solids2008;27:107–15.

[133] Karagiozova D, Langdon GS, Nurick GN, Yuen KCS. Simulation of the responseof fibre–metal laminates to localised blast loading. Int J Impact Eng2010;37:766–82.

[134] Langdon GS, Lemanski SL, Nurick GN, Simmons MC, Cantwell WJ, SchleyerGK. Behaviour of fibre–metal laminates subjected to localised blast loading:part I—experimental observations. Int J Impact Eng 2007;34:1202–22.

[135] Lemanski SL, Nurick GN, Langdon GS, Simmons MS, Cantwell WJ, SchleyerGK. Understanding the behaviour of fibre metal laminates subjected tolocalised blast loading. Compos Struct 2006;76:82–7.

[136] Langdon GS, Cantwell WJ, Nurick GN. Localised blast loading of fibre–metal laminates with a polyamide matrix. Composites Part B2007;38:902–13.

[137] Langdon GS, Cantwell WJ, Nurick GN. The blast response of novelthermoplastic-based fibre–metal laminates – some preliminary results andobservations. Compos Sci Technol 2005;65:861–72.

[138] Lemanski SL, Nurick GN, Langdon GS, Simmons MC, Cantwell WJ, SchleyerGK. Behaviour of fibre metal laminates subjected to localised blast loading—part II: quantitative analysis. Int J Impact Eng 2007;34:1223–45.

materials and design - kocaeli...

Documents