The influence of fibre orientation in aluminium–carbon laminates on low-velocity impact resistance

Abstract

The objective of this study was to assess the influence of fibre orientation in hybrid fibre metal laminates based on aluminium and carbon fibres on the impact of low-velocity impact. The analysis was conducted on the basis of fibre metal laminate impact resistance criteria, including impact force, energy absorption, bending stiffness, damage area and failure. To assess the resistance of various aluminium–carbon laminates, qualitative and quantitative evaluation criteria were employed, including the shape of the force–time curve, characteristic impact forces, energy absorption, bending stiffness, damage area and external failure analysis. Among others, authors concluded that no explicit influence of the composite layer fibre orientation on the shape and value of characteristic forces was observed. It was found that the fibre orientation and the changing number of interfaces of low durability show no explicit influence on the size and shape of delaminations.

Keywords

Hybrid fibre metal laminate carbon fibres impact damage mechanics

Introduction

Fibre metal laminates (FMLs) based on aluminium alloys and carbon fibre-reinforced composites (CFRP) are a group of composite materials which are currently being intensively researched. Due to low density and high mechanical properties including static properties and fatigue strength, it is visible that the most welcome application area for those materials would be aerostructures.^1–4 Intensive research and development work are being conducted to obtain new and alternative FMLs with the use of carbon fibres, among other materials. FMLs based on aluminium and carbon fibres exhibit higher static and fatigue strength, higher stiffness and lower density in comparison to other laminates of this type.^1,5

One of the crucial types of service loads of composite aerostructures are low-velocity impact. According to Vogelesang and Vlot² and Sohn et al.,⁶ impact damage occurs during pre-flight and taxiing operations, in the case of runway debris, hail or bird strikes, maintenance damage (e.g. dropped tools), collision of service cars or cargo and the structure, ice from propellers striking the fuselage, engine debris and tire shrapnel from tread separation and tire rupture. The authors² quoted that 13% repairs of primary structure in Boeing 747 aircraft are caused by the impact damage.

To date, studies on FML impact resistance have focused mainly on the assessment of damage mechanisms and the influence of particular types of components employed in the laminate construction on such loads.⁷ Many authors concluded that FMLs are marked by relatively high resistance to low-velocity impact, whereas the damage mechanisms which ensue are complex and strictly connected to the degradation of reinforcement fibres, composite matrix, the metal itself, as well as at the metal and composite interfaces.^8–13 There are many factors other than the possibility to use various materials in the construction of FMLs in respect of their response to impact. They include, among others, the reinforcement fibre orientation in the composite interlayers of the laminate. The influence of fibre orientation on mechanical properties has been tentatively described in the existing literature.¹⁴

In case of an out-of-plane load, the influence of the orientation of the composite materials fibres on the impact resistance among others has not been sufficiently explained yet, as only a few authors attempted to expound this issue.

In respect of conventional CFRP composites, Hitchen and Kemp¹⁵ concluded the stacking sequence affects the impact damage in carbon fibre/toughened epoxy laminate. The total delamination area was influenced by the stacking sequence and was related to the energy absorbed during impact by a novel analysis which distinguished the delamination initiation energy from the delamination extension energy and the maximum absorbed energy. The delamination initiation energy was increased by placing 45 ° plies in the surface layers and by increasing the number of dissimilar interfaces which therefore reduced the energy available for delamination extension. This reduced extension energy resulted in a reduced total delamination area. The delamination shape was influenced by splitting and fibre fracture in the ply below the delamination.

Alemi-Ardakani et al.¹⁶ have made experimental investigation and comparison of plain woven (PW), twill woven (TW) and unidirectional (UD) glass fibre composites under impact. The authors concluded that a combination of different type of reinforcement layers (e.g. the hybrid configuration [TW, (PW)₂, UD^0/90]_s provided the FRP structure with the capacity to absorb more impact energy when compared to the uniformly reinforced [(PW)₆]_s and [(TW)₃]_s laminates. They found that the damage mechanism in impacted monolithic UD laminate was notably different from that of other configurations. Instead of severe matrix crushing and fibre pull-out visible on the back side of impacted specimens with any combination of woven and unidirectional FRPs, cold cracking was observed in the uniform UD laminate configuration.

Seyed Yaghoubi et al.¹⁷ have investigated stacking sequence and geometrical effects on low-velocity impact behaviours of GLARE 5. It was concluded that laminates made of unidirectional fibres had the lowest impact resistance, followed by cross-ply and angle-ply configurations, while the quasi-isotropic lay-up showed the best resistance to impact. Moreover, the major and minor axes of the elliptical-shaped damage zone coincided with the fibre directions.

The literature provides an analysis of the influence of the 3D fabric composite reinforcement on the impact resistance in FMLs. Asaee and Taheri¹⁸ tested 3D fabric glass in FML on impact response. In their studies, however, the authors focused mainly on the influence of the 3D filling thickness in FMLs on impact. They did not correlate the results obtained with standard FMLs based on 2D fibre–polymer composite layers.

The current knowledge on the influence of stacking sequence in FMLs on impact has been reviewed and compiled by Sadighi et al.⁷ The authors concluded that it is difficult to compare relevant results published on the effect of the different lay-ups. They found that some tests were made for GLARE laminates, which have the lowest impact resistance among GLARE laminates made of unidirectional fibres in general. These are followed by cross-ply and angle-ply configurations, while the quasi-isotropic lay-up shows the best resistance to impact.

In an attempt to understand the effect of fibre orientation on the impact resistance of FMLs, an analysis of the existing literature proves largely insufficient. Although there are several scientific studies available on the influence of fibre orientation on the impact resistance of FMLs, this issue remains unresolved. It is carbon fibre aluminium laminates, characterised by fibre dominant failure mechanisms as a result of impact, that require an analysis in this respect in particular.⁷

The objective of this study was to assess the influence of fibre orientation in hybrid FMLs based on aluminium and carbon fibres on the low-velocity impact. The analysis was conducted on the basis of FML impact resistance criteria, including impact force, energy absorption, bending stiffness, damage area and failure.

Materials and methodology

Materials

The subject of examination was 2/1 FMLs composed of two thin aluminium layers and one carbon fibres-reinforced polymer layer (AlC). The 2024-T3 aluminium alloy sheets of 0.5 mm thickness were used. The composite layers consisted of unidirectional prepregs based on AS7J high-strength carbon fibres with epoxy matrix resin (Hexcel, USA). The nominal fibre content was about 60% of the volume. Before laminating, the surface of aluminium alloy sheets had been anodised in chromic acid (CAA) and next an adhesive primer was applied to improve bonding with fiber-reinforced polymers. Laminates were manufactured in the Department of Materials Engineering at the Lublin University of Technology by the means of the autoclave method (Scholz Maschinenbau, Germany) under vacuum of −0,08 MPa and 0.5 MPa pressure. The materials were cured for 2 h at 135℃.

The basic cause of damage is incompatible bending stiffness between the adjacent layers with various orientations of the reinforcing fibres. The above-mentioned lack of compatibility results mainly from elastic properties of materials, especially the ratio of Young's modules E1/E2. Assuming that stiffness is different for layers with different angular orientation, different fibre orientation in laminates (Table 1) was designed.¹⁹

Table 1.

Fibre orientation of tested aluminium–epoxy/carbon laminates.

CFRP: carbon fibre-reinforced composites.

Impact test

Samples with dimensions of 150 × 100 mm (y and x axis respectively) were subjected to low-velocity(<10 m/s) impact in room temperature by using a drop-weight impact tester (INSTRON Dynatup 9340) with a possibility to record force–time history. A hemispherical impactor tip with a diameter of 12.7 mm (0.5″) was applied. All the low-velocity impact tests conducted were based on ASTM D7136 standard.²⁰ The low-velocity impact was tested for four different energies: 5, 10, 15, and 20 J. In damage assessment, macroscopic non-impacted sides were used, whereas a C-scan and through transmission phased array methods were applied for the internal damage area (the author's position at the Department of Materials Engineering at Lublin University of Technology using OmniScan MXII, Olympus, Japan).

Results and discussion

Depending on their reinforcement fibre orientation, the impact resistance of carbon fibre aluminium laminates has been described in terms of commonly employed quantity and quality criteria, including force–time curves, energy absorption, bending stiffness, damage area, and failure of non-impacted side.

Figure 1 shows force–time curves (f–t) as well as energy absorption – time curves (E_a-t) of FMLs characterized by various carbon reinforcement fibre orientation, subjected to impact of 5–20 J.

Force–time curves

In general, the f–t curves are characterised by an ascending section of loading, a section of reaching a maximum force and, a descending part of unloading.^8,10,21 Significant force oscillation can be observed at the initial stage of the loading (during a time shorter than 0.5 ms). It is related to the stabilisation of the material-indenter system. Initial laminate loads its vibration, and consequently forces fluctuation of high frequency and low amplitude.^22,23 A growth of force which ensues at the following stage marks the beginning of the proper interaction of the impactor and the material. At this point, the material begins to degrade through the initiation and development of cracks at the carbon fibre–matrix interface, the initiation of matrix cracking and delaminations in the composite interlayers and at the interface between the metal and the composite.^24–26 Further damage in FMLs along with an increase in force leads to gradual decrease of the laminate stiffness. The force increases until it reaches the maximum force value P_m (Figure 1). At this point, the force decreases and the indenter is withdrawn. The f–t charts show significant drops in force (more than ten percent) before it reaches its maximum value. The point it reaches just before the drop is called incipient force P_i, (Figure 1), and it indicates the occurrence of a significant failure. It can be used as a measure of the material's ability to resist initial damage, referred to by certain authors as impact damage resistance of composites.^6,10,25,27 The course of f–t curves as well as the value of P_i and P_m may constitute a basis for assessment and comparison of the composite material and FMLs resistance to low-velocity impact.^{8,12,24,28–31}

On the basis of the studies conducted on the influence of fibre orientation in particular composite layers in laminates on impact resistance, no significant differences have been noted in respect of the course of f–t curves for any type of AlC laminates. The research showed that in all laminates examined, an increase in impact force leads to a growth in maximum force values. Along with a force increase, the fluctuation frequency rises as well.

A detailed analysis of the f–t curve indicates the presence of force oscillations. Many more oscillations may be observed when the force is higher, thus their exact amount is dependent on the impact energy. It is also connected with fibre orientation (number of interfaces). However, the registered oscillations can be indicative of the initiation of new cracks at the matrix/carbon interface or the growth of the existing ones, initiated at the stage of system stabilisation or immediately after that stage. According to the authors, more significant forms of damage in FMLs may be expected, i.e. delaminations in two characteristic areas. The first one is the delamination initiation at the interface of individual layers in the polymer composite with different angular orientation of carbon fibres. The second area of the delamination initiation is at the metal/composite interface. At the impact energy of 10 J and higher, the presence of P_i points in all laminates examined was noticed. In case of the energy values 15 J and 20 J, it may be asserted that laminates with multidirectional layer systems exhibit a larger amount of visible oscillations. It may suggest that those layer systems are marked by more intensive damage initiation (initiation of delamination and shear cracking) than in the case of unidirectional laminates. These differences are, however, not clear enough to formulate explicit conclusions. In the case of lower energy values, differences in force fluctuation frequency of particular multidirectional systems do not occur or are too insignificant to be recorded. No significant differences can be connected with high impact resistance of FML. In case of impact energies which do not cause rapid perforation, part of impact energy is absorbed by membrane effect.^7,24,26 Membrane effects are known to become important for deflections in the order of the plate thickness. For a membrane-dominated response, the layup sequence becomes irrelevant and is only influenced by the fraction of fibres in different directions. However, higher impact energy causes a higher response governed bending and shear when the penetration of impactor starts (plastic deformation around the impact point).

At low-velocity impacts, the large elastic response of the laminate under bending leads to the occurrence of large negative stress in the lower laminate layers. Bending cracks occur when normal stress in the layer plane exceeds the transverse tensile strength of the layer.³² The main cause of delaminations is a mismatch in elastic properties (Young's modulus) between adjacent layers of carbon fibres with different orientation.^27,32 A significant difference between transverse and longitudinal modules of material elasticity leads to greater mismatches in bending stiffness (e.g. for layers with a 0/90 fibre orientation) and the development of extensive delaminations.³³ In the case of the AlC, a mismatch in Young's modulus values applies also at the metal/composite interface. In this respect, it can be noted that unidirectional laminates (sample A) exhibit areas which are more prone to delamination development and, consequently, decreased stiffness, mainly at the metal/composite interface. Bidirectional laminates (samples B₁,B₂,B₃) and quasi-isotropic laminates (sample C) show a larger number of interfaces with Young's modulus mismatch. The observation of f–t curves, however, shows no correlation with the number of interfaces with mismatch in Young's modulus.

Figure 2 illustrates values of characteristic impact forces in laminates with various reinforcement fibre orientation.

It has been shown that an increase in impact energy leads to a growth in maximum force value (Figure 2(b)), whereas the value of initiation force remains fixed for particular laminate types (Figure 2(a)). There are visible differences in initiation force values in respect of the system. Higher P_i values (by 25%) have been noted in the case of bidirectional laminates in comparison with unidirectional and quasi-isotropic laminates. The reason is that the composites with bidirectional layer orientation are the worst in terms of damage accumulation.³⁴ Lower P_i values for laminates with fibre lay-up (0₄) and (0/±45/90), however, have no impact on the maximum force value, which is similar regardless of composite layer lay-up.

The authors of this paper believe that stacking sequence in FMLs may be of limited importance in respect of shaping the resistance to low-velocity impact expressed by the forces transferred through the laminate due to the presence of metal layers and their energy absorption capabilities through elastic–plastic deformation. This is one of the characteristic properties of FMLs in respect of conventional composite structures, for which, as it has been shown in the literature, fibre orientation in composites plays a major role in terms of impact resistance.³⁴

Energy absorption

Energy absorption curves (Figure 1), similarly to f–t curves, exhibit a stage of energy growth in time, followed by a gradual decrease. The energy drop stage, along with simultaneous determination of maximum energy value, is typical of laminates which have not undergone perforation as a result of impact. A part of energy is then returned to the system, as the elastic strain relaxes. The final energy value corresponds with the impact energy absorbed by the laminate. Upon comparing the impact energy absorption curves, it may be noticed that there are smaller differences between the aggregated energy and the absorbed energy in the case of unidirectional laminates than in the laminates with a multidirectional lay-up. This difference stems from lower values of energy aggregated in Al/(0)₄/Al laminates. At the same time, the values of final absorbed energy prove similar in all systems examined. This may point to a lower capability of impact energy accumulation in unidirectional laminates than in the laminates with a multidirectional lay-up of composite layers. The differences, however, are slight, not exceeding 10% of the value. The absorbed energy may be used to assess the laminate susceptibility to damage initiation, since damage initiation and propagation lead to impact energy absorption beside elastic deformation of the composite and elastic–plastic deformation of metal. Determining the value of absorbed energy requires the use of mathematical relations between the impact force, time of impact and indenter displacement.

It is possible to estimate the value of the energy absorbed (E_a(t)) by the laminate during the process of impact by the means of equation (1).²⁰

E_{a} (t) = \frac{m (v_{i}^{2} - v^{2})}{2} + mg δ

(1)

where

δ – indenter displacement. (m)

v_i – velocity of impacting body at the moment of contact (m/s)

v – final velocity (m/s)

m – mass of the indenter (kg)

g – gravitational acceleration, 9.80665 m/s²

The form of equation (1) must be transformed into a form which includes the variable parameters recorded on f–t curves, i.e. force in time (F(t)) and time(t).

The final velocity of the indenter v (m/s) may be described according to equation (2)²⁰

v = v_{i} + gt - \int_{0}^{t} \frac{F (t)}{m} d t

(2)

where

t – total time of indenter-material contact (s)

F – force measured at the time of impact (N)

m – mass of the indenter (kg)

g – gravitational acceleration

where as indenter displacement δ (m) must be described as equation (3)²⁰

δ = v_{i} t + \frac{g t^{2}}{2} - \int_{0}^{t} (\int_{0}^{t} \frac{F (t)}{m} d t) d t

(3)

where

t – total time of indenter-material contact (s)

v_i – velocity of impacting body at the moment of contact (m/s)

F – force measured at the time of impact (N)

m – mass of the indenter (kg)

g – gravitational acceleration

By the means of appropriate substitutions and transformations, the final expression (4) determines the energy absorbed by the laminate

\begin{matrix} E_{a} (t) = \frac{m [v_{i}^{2} - (v_{i} + gt - \int_{0}^{t} \frac{F (t)}{m} d t)^{2}]}{2} \\ + mg [v_{i} t + \frac{g t^{2}}{2} - \int_{0}^{t} (\int_{0}^{t} \frac{F (t)}{m} d t) d t \end{matrix}

(4)

Table 2 presents a compilation of energy absorbed by AlC laminates following impacts in the range of 5–20 J, depending on the fibre lay-up in the composite layers.

Table 2.

Energy absorbed by AlC laminate during impacts with the energy range of 5–20 J.

Impact energy (J)	Sample A	Sample B₁	Sample B₂	Sample B₃	Sample C
Impact energy (J)	E_a (J)	E_a (J)	E_a (J)	E_a (J)	E_a (J)
5	2.85	2.37	2.18	2.23	2.64
10	6.19	5.9	5.13	5.76	5.94
15	9.72	9.61	9.16	9.55	9.44
20	13.79	13.41	13.76	13.41	13.23

It has been shown that an increase in impact force leads to a growth in energy absorption. Table 2 provides evidence that an increase in impact energy by 5 J (in the range of 0–20 J) leads to an average growth in energy absorption by 3.7 J (standard deviation σ = 0.37), regardless of the laminate fibre configuration examined. The value of energy absorbed by the laminate amounts to ca. 60% of potential impactor energy in the case of all the laminates examined, in the range of impact energy of 5–20 J.

The influence of fibre lay-up orientation in particular layers on energy absorption is more than limited. What matters, however, are the layer interfaces. It has been noted that the Al/(0)₄/Al laminate absorbs more energy than others, regardless of the impact energy value. The differences are insignificant, amounting on average to 8%. The fact that unidirectional laminates absorb larger amounts of energy may result from a lower number of interfaces in the composite part of the laminate. The interfaces in CFRP composites are regarded as more prone to the propagation of delamination, absorbing only a part of the impact energy. It should be presumed that it is degradation mechanisms in unidirectional laminates and not necessarily delaminations on the composite interfaces that are responsible for a larger share in the energy absorption process. This may provide a basis for further analysis and correlation of energy absorption and transformation kinetics at the microstructural scale.

Bending stiffness

The impact bending stiffness can be described as the initial bend angle of a rectilinear section of force increase in the function of displacement. Bending stiffness calculations were conducted on the basis of force–displacement curves (Figure 3) according to the data provided in the literature.^35,36 The assessment employs the ascending section of the force–displacement curve following the system stabilisation stage (Figure 3).

Table 3 shows selected bending stiffness values of particular laminates.

Table 3.

Bending stiffness of aluminium/carbon laminates with various stacking sequence.

	Bending stiffness (N/mm)
	Impact energy 5 J	Impact energy 10 J	Impact energy 15 J	Impact energy 20 J
Sample A	363	413	406	421
Sample B₁	359	424	398	445
Sample B₂	350	419	409	445
Sample B₃	380	438	411	454
Sample C	349	426	415	433

The determined bending stiffness values increase along with a rise in energy impact in general. They are limited to the range of 349–454 N/mm. The differences noted for particular laminate types do not exceed 10% of the value. Bending stiffness may be employed as one of quantity criteria for assessing the laminate and composite impact resistance due to its representation of the stiffness of laminates under impact-induced bending in the beginning of impact process.³⁵ This material property is contingent mostly on its thickness, component type and other structural features of laminates.³⁰ Considering the material and not the stress, the bending stiffness plays a major role in determining the delamination. The interface delamination may be represented by the differences in bending stiffness between the laminates. The potential of delamination at the composite/composite and metal/composite interface may be hypothesized by the means of the bending stiffness value.³⁰ However, an assessment of stacking sequence influence in AlC laminates on their bending stiffness has not shown any explicit correlations (Figure 3). Regardless of composite layer lay-up, these laminates do not show any significant differences as to their bending stiffness. The only difference in terms of this property was noticed in the case of 20 J energy impact. The value of bending stiffness was ca. 10% lower in type A laminates than in type B laminates (bidirectional stacking sequence). The poorer membrane effect of UD AlC laminates is visible. The determined bending stiffness values are a sign of a greater deformation capability in those laminates than in laminates with mismatch of fibre orientation. According to the authors of this paper, the differences observed for all impact energy values which do not exceed 10% should not be a ground for conclusions regarding the influence of angular changes in the fibres of particular layers on the change in bending stiffness of the whole AlC laminate, which would assert its impact resistance. According to Liu,³⁰ it is of crucial importance to enhance the delamination resistance of a composite stacking sequence. By increasing the interface number, one would undoubtedly increase the delamination resistance of fibre-reinforced polymer composites. In the case of FML, where the composite part is thinner, the metal and metal/composite interfaces reduce the importance of stacking sequence. However, UD laminates are slightly weaker in terms of bending stiffness.

Damage

Due to the hybrid nature of laminates and significantly different properties of particular FML components, the damage mechanisms are complex, including the degradation of particular composite phases (fibres and matrix) and of metal, as well as the boundaries between particular laminate components. Fibre–metal laminates absorb the impact energy by the means of elastic deformation, elastic–plastic deformation of metal layers, as well as the initiation and propagation of internal structure damage e.g. delaminations, matrix and fibre cracking.^8,9,13,37

On the basis of research conducted,⁸ it has been stated that an increase in impact energy in FMLs, damage propagates through an increase in the area of internal damage (delamination), initiation and propagation of metal and fibre cracking, as well as perforation.^8,13 Apart from the impact energy value, however, the material aspect is also of importance in FMLs. FMLs are divided into two groups in terms of their failures, and namely ‘fibre-dominated’ and ‘aluminium-dominated’ failures. Carbon fibre aluminium and aramid fibre aluminium are both ‘fibre-dominated’, whereas aluminium-glass is ‘fibre-‘ or ‘aluminium-dominated’, depends on the behaviour of glass fibre.⁷ What should be noted is that ‘fibre-dominated’ failure mechanisms in carbon fibre aluminium laminates lead to a potential increase in the importance of fibre orientation on the impact resistance in the given laminate type.

Depending on the fibre lay-up, it is possible to observe differences as to the type and size of damage, which is one of the most significant indicators of low-velocity impact resistance in FMLs. Figure 4 illustrates a macroscopic analysis of damage to a non-impacted side (bottom) of AlC laminates, as well as ultrasound C-scan imaging illustrating the internal damage.

Macroscopic observation of a non-impacted side of AlC laminates of various reinforcement fibre orientation shows a deformation at the point of the contact with the indenter, as well as cracks in the bottom aluminium layer. Even at the lowest impact energy values, the presence of deformation points to BVID structural damage (barely visible impact damage).⁸ NVID damage (non-visible impact damage) does not occur in fibre–metal laminates, as they are typical of standard composite structures due to deformations which are visible at the macroscopic scale on the impacted side.^34,38 This may be regarded as one of the advantages FMLs have over FRP composites.

It has been noted that cracks in the bottom layer of AlC laminates occur in the case of all the laminates examined, provided that the impact energy equals 15 J at minimum. The only exception to that was observed in unidirectional laminates (sample A), where cracking of the bottom aluminium layer initiates at the 10 J energy value. Fibre orientation of carbon composite in aluminium–carbon fibre laminates has been claimed to have minor influence on the laminate degradation mechanism, which is crucial as it initiates perforation – cracking of the bottom metal layer. No differences have been noted as to the length of crack between particular stacking sequences of AlC laminates.

According to the available literature, crack initiation of the bottom aluminium layer may be consistent with its rolling direction. Some works provide extensive analyses of the cracking direction in the bottom aluminium layer as a result of impacts causing FMLs perforation.^7,10,39 Moriniere et al.³⁹ noted that the direction of metal cracking for GLARE 3/2 (0/90/90/0) is interpreted as being compatible with sheet rolling direction. In the study by Sadighi et al.,⁷ it was concluded that in the case of FMLs, where aluminium dominates in load carrying, cracking of the bottom aluminium layers propagates in the rolling direction; however, this observation was made on the basis of glass–fibre laminates. According to Vlot,⁴⁰ in terms of two types of failure of FMLs, i.e. ‘fibre’ or ‘aluminum’ critical, CARAL, due to its low strain to failure (∼2%), always shows a fibre critical behaviour irrespective of its lay-up. In the case of aluminum/carbon laminates, authors¹² proved that impact energy is absorbed by aluminium layers and further impactor penetration initiates cracking of the bottom aluminium layer in its preferred direction. Cracking of the metal layer in the direction transverse to the one of fibres from the bottom composite layer can be observed in the case of FMLs with carbon fibres, but studies concerning this subject are few and far between.^25,37 Lawcock et al.³⁷ obtained aluminium sheet cracking initiation in the transverse direction during testing of CARAL 2/1, with unidirectional fibre orientation in the composite. However, the causes of such cracking direction were not explained.

The results confirm that, the direction of crack propagation of the bottom aluminium layer is related to the fibre orientation in the composite layers, in particular to layer directly adjacent to the bottom aluminium layer. Aluminium cracking propagating perpendicularly to the adjacent layer has been observed in each laminate examined, regardless of the impact energy and the orientation of the remaining composite layers. The authors of this paper suspect that this may be related to high stiffness of carbon fibres and their lack of resistance to brittle fracture. Bending, which is critical for brittle carbon fibres, as well as shearing resulting from indenter penetration may lead to cracking of fibre strands, setting the direction of aluminium layer cracking. Moreover, it may be assumed that fibre cracks occur at this stage also in individual composite layers as a result of their microbuckling due to compressive stresses. The authors of these works conclude that fibre cracking in carbon fibre laminates may propagate even before cracking initiates in metal, which is much more deformable.^11,24,37

Although a critical failure of AlC is not visible, the earlier analysis of f–t curves indicated internal laminate failure. In order to assess the internal damage to AlC laminates upon impact, a non-destructive ultrasound examination was performed. Figure 4 presents C-scan imaging of the laminates examined from the impacted side. It shows a superposition of all the delaminations developed in the laminate structure.

A detailed analysis of the C-scan maps has indicated the presence of delaminations in all the laminates which have undergone an impact. Theoretically, the shape of the delamination results from shear stress distribution in the area surrounding the impactor, low interlayer shear resistance alongside or close to the fibre orientation direction, as well as from matrix cracking caused by bending stress.³² It was noted that the shape of delaminations in AlC laminates is always quasi-circular (Figure 4) what is impossible in conventional composites, where delaminations have a shape of peanut (quasi-izotropic laminates), cruciform (bideractional laminates) and longitudinal or no delamination but single crack through matrix along the fibre (unidirectional composites).^27,33 In the case of the AlC, a mismatch in Young's modulus values applies also at the metal/composite interface. The circular shape in the case of AlC laminate is the result of isotropic properties of aluminium and high difference in their stiffness and the stiffness of the adjacent carbon/epoxy layer. It is the reason why metal/composite interfaces are considered to be areas as prone to the initiation and development of delamination as the composite/composite interface. The influence of the metal/composite interface on the initiation and propagation of delamination limits the role of fibre lay-up in setting the direction and size of the delaminations propagating in the laminate. The difference between the Young's modulus value for the aluminium layer (ca. 70 GPa) and the CFRP composite layer (ca. 130 GPa) facilitates the propagation of vast delaminations regardless of the fibre orientation at the metal/composite interface. This has been indicated in the work Bienias et al.,⁸ where low-velocity load-induced delaminations were analysed on the basis of their cross-sections, among others in AlC laminates.

The research results have proven i.a. that an increase in impact force leads to a growth in the damage. In the case of influence of stacking sequence in FML, several minor differences in the damage area could be observed. The damage area – impact energy and absorbed energy correlations are presented in Figure 5.

Sample B₂ has the largest damage area due to a mismatch of Young modulus inside the composite and high ability to damage accumulation (delamination in the composite interlayers). Bidirectional B₂ laminate contains one interface (0/90), whereas B₁ and B₃ laminates have three (0/90) interfaces, which can lead to differences in the damage area.³⁴

Considering the damage area and the absorbed energy, it can be concluded that the correlation is quite the same as in the case of damage area vs. impact energy. Due to slight differences in the energy absorbed by different types of laminates, some additional points can be noted. Sample A is characterised by the lowest DA and the highest value of absorbed energy. Laminate type B2 always has the lowest absorbed energy and almost always the greatest DA. Samples B1, B3 and C are characterised by similar absorbed energy and similar damage area. These types of laminates are characterised by many interfaces between layers with different orientation of fibres.

Summary

The present study describes the influence of the selected fibre orientation of particular composite layers in aluminium-based FMLs on the resistance of these materials to low-velocity impact. FMLs are in general characterised by high resistance to impacts. It is known that aluminium–carbon laminates are characterized by fibre dominant failure mechanisms under impact conditions. In respect of these materials, however, it is particularly important to fully comprehend the degradation processes, as well as to assess the connection between the impact resistance and laminate structure

The objective of this study was to assess the influence of the fibre orientation of particular composite layers in carbon fibre aluminium laminates on the resistance to low-velocity impact.

The analysis comprised five laminate configurations in the 2/1 layer system, with layers differing as to fibre orientation and layer order. To assess the resistance of carbon fibre aluminium laminates, qualitative and quantitative evaluation criteria were employed, including the shape of the force–time curve, characteristic impact forces, energy absorption, bending stiffness, damage area and external failure analysis.

On the basis of the experiment conducted and further analysis performed:

No unequivocal explicit influence of the composite layer fibre orientation on the shape and value of characteristic forces was observed as a result of 5–20 J impact. Regardless of the configuration of the laminate examined, similar initiation force and maximum force values were recorded.

The energy absorption values determined did not clearly show advantage of any of the AlC fibre laminate systems designed in this respect. The energy absorbed values proved similar for all tested types of laminates. It has been noted, however, that unidirectional laminates show slightly higher maximum absorbed energy values during the impact in 5–20 J impact energy range.

It was found that the fibre orientation and the changing number of interfaces of low durability show no pronounced influence on the size and shape of delaminations in carbon fibre aluminium laminates. Despite the fibre failure dominant of AlC laminates under impact, the deformation of plastic aluminium around the impactor provides symmetrical interaction between metal and composite.

The direction of the fibre orientation at the lower composite layer has an influence on the direction of cracking propagation in the lower aluminium layer. It was found that the crack at the bottom aluminium layer propagates perpendicularly to the direction of carbon fibres of the adjacent layer, which may be relevant in the context of the further resistance of the laminate to other types of load.

Aluminium–carbon laminates and current trends in the use of thin-walled structures make it necessary to further evaluate the impact resistance of these materials, including the residual strength for which fibre orientation of the composite inside the laminate is important.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The presented research is part of a project financed by the National Science Centre allocated on the basis of decision number DEC-2012/05/N/ST8/03788.

References

Botelho

Silva

Pardini

et al.

A review on the development and properties of continuous fiber/epoxy/aluminum hybrid composites for aircraft structures. Mater Res 2006; 9: 247–256.

Vogelesang

Vlot

. Development of fibre metal laminates for advanced aerospace structures. J Mater Process Tech 2000; 103: 1–5.

Karnocka

Zglinicki

Mania

. Multi-method approach for FML mechanical properties prediction. Compos Part B Eng 2016; 91: 135–143.

Kubiak

Mania

. Hybrid versus FR laminate channel section columns – buckling and postbuckling behavior. Compos Struct 2016; 154: 142–149.

Vlot

Gunnink

. Fibre metal laminates: an introduction, Dordrecht – Boston – London: Kluwer Academic Publishers, 2001.

Sohn

Kimb

et al.

Impact damage characterization of carbon fbre/epoxy composites with multi-layer reinforcement. Compos Part B Eng 2000; 31: 681–691.

Sadighi

Alderliesten

Benedictus

. Impact resistance of fiber-metal laminates: a review. Int J Impact Eng 2012; 49: 77–90.

Bienias

Jakubczak

Surowska

et al.

Low-energy impact behaviour and damage characterization of carbon fibre reinforced polymer and aluminium hybrid laminates. Arch Civil Mech Eng 2015; 15: 925–932.

Vlot

Krull

. Impact damage resistance of various fibre metal laminates. J Phys IV France 1997; 7: 1045–1050.

10.

Caprino

Spatarob

Del Luongo

. Low-velocity impact behavior of fibre glass–aluminum laminates. Compos Part A Appl S 2004; 35: 605–616.

11.

et al.

Low velocity impact of carbon fiber aluminum laminates. Compos Struct 2015; 119: 757–766.

12.

Bienias

Surowska

Jakubczak

. The comparison of low velocity impact resistance of aluminium/carbon and glass fiber metal laminates. Polym Compos 2016; 37: 1056–1063.

13.

Bienias

Jakubczak

Dadej

. Low-velocity impact resistance of aluminium glass laminates – experimental and numerical investigation. Compos Struct 2016; 152: 339–348.

14.

Woo

Choi

Chang

. Toughness and fracture mechanisms of glass fiber/aluminum hybrid laminates under tensile loading. J Mech Sci Technol 2007; 21: 1937–1947.

15.

Hitchen

Kemp RMJ . The effect of stacking sequence on impact damage in a carbon fibre/epoxy composite. Composites 1995; 26: 207–214.

16.

Alemi-Ardakani

Milani

Yannacopoulos

et al.

A rapid approach for predisation and descrete lay-up optimization of glass fiber/polypropylene composite laminates under impact. Int J Impact Eng 2015; 84: 134–144.

17.

Seyed Yaghoubi

Liu

Liaw

. Stacking sequence and geometrical effects on low-velocity impact behaviors of GLARE 5 (3/2) fiber–metal laminates. J Thermoplast Compos 2012; 25: 223–247.

18.

Asaee

Taheri

. Experimental and numerical investigation into the influence of stacking sequence on the low-velocity impact response of new 3D FMLs. Compos Struct 2016; 140: 136–146.

19.

German

. Podstawy mechaniki kompozytów włóknistych, Cracow: Cracow University of Technology, 1996.

20.

ASTM D7136. Standard test method for measuring the damage resistance of a fiber‐reinforced-polymer matrix composites to a drop-weight impact event, book of standards, volume 15.03.

21.

Fan

Guan

Cantwell

. Numerical modeling of perforation failure in fibre metal laminates subjected to low velocity impact loading. Compos Struct 2011; 93: 2430–2436.

22.

Oya

Hamada

. Mechanical properties and failure mechanisms of carbon fibre reinforced thermoplastic laminates. Compos Part A Appl S 1997; 28: 823–832.

23.

Guan

Yang

. Low-velocity impact and damage process of composite laminates. J Compos Mater 2002; 36: 851–871.

24.

Abdullah

Cantwell

. The impact resistance of polypropylene-based fibre-metal laminates. Compos Sci Technol 2006; 66: 1682–1693.

25.

Song

Byun

et al.

Experimental and numerical investigation on impact performance of carbon reinforced aluminum laminates. J Mater Sci Technol 2010; 26: 327–332.

26.

Yang

Hahn

. The impact properties and damage tolerance and of bidirectionally reinforced fibre metal laminates. J Mater Sci 2007; 42: 948–957.

27.

Shyr

Pan

. Impact resistance and damage characteristics of composite laminates. Compos Struct 2003; 62: 193–203.

28.

Jakubczak

Bienias

. Comparison of quasi static indentation and dynamic loads of glass and carbon fibre aluminium laminates. Aircraft Eng Aerosp Tech 2016; 88: 404–410.

29.

Chang

Choi

Jeng

. Study on impact damage in laminated composites. Mech Mater 1990; 10: 83–95.

30.

Liu

. Impact-induced delamination – a view of bending stiffness mismatching. J Compos Mater 1998; 22: 674–692.

31.

Liu

Kutlu

Chang

. Matrix cracking and delamination in laminated composite beams subjected to a transverse concentrated line load. J Compos Mater 1993; 27: 436–470.

32.

González

Maimí

Camanho

et al.

Effects of ply clustering in laminated composite plates under low-velocity impact loading. Compos Sci Technol 2011; 71: 805–817.

33.

Abrate

Low-Velocity impact damage. In: Abrate

(ed). Impact on composite structures, Cambridge, United Kingdom: Cambridge University Press, 1998, pp. 135–160.

34.

Richardson

MOW

Wisheart

. Review of low-velocity impact properties of composite materials. Compos Part A Appl S 1996; 27: 1123–1131.

35.

Liu

Raju

Dang

. Impact perforation resistance of laminated and assembled composite plates. Int J Impact Eng 2000; 24: 733–746.

36.

Ghelli

D'Ubaldo

Santulli

et al.

Falling weight impact and indentation damage characterisation of sandwich panels for marine applications. OJCM 2012; 2: 8–14.

37.

Lawcock

Mai

et al.

Effects of fibre/matrix adhesion on carbon-fibre-reinforced metal laminates-II. Impact behavior. Compos Sci Technol 1997; 57: 1621–1628.

38.

Delft University of Technology. Low-velocity impact loading on fibre reinforced aluminum laminates (ARALL and GLARE) and other aircraft sheet materials. Report LR-718, 1993.

39.

Moriniere

Alderliesten

Benedictus

. Low-velocity impact energy partition in GLARE. Mech Mater 2013; 66: 59–68.

40.

Vlot

. Impact loading on fiber metal laminates. Int J Impact Eng 1996; 18: 291–307.