Adhesively-bonded Composite Joints Research Articles

Progress in adhesive-bonded composite joints: A comprehensive review

Among the myriad joining techniques, the adhesive bonding technique is widely used to join complex large-scale composite structures because of its numerous advantages compared to traditional joining techniques. This article profusely analysed the various techniques for ameliorating the performance of composite joints, such as bonding methods (secondary bonding, co-bonding, co-curing, and multi-material bonding), surface modification techniques (plasma, laser surface treatment, surface grinding, etc.), additional reinforcement techniques (Z pin, wire mesh, nanofiller, etc), and different joint geometries (stepped joints, half-stepped joints, balanced joints, and scarf joints). Also, the effect of various adhesives and fabrication techniques on the static and dynamic performance of CFRP and GFRP-based joints was studied in detail. Moreover, this review addresses the finite element modelling and optimisation techniques on adhesively bonded joints. It has been observed that the bonding methods, surface modification to enhance the roughness of the adherend, addition of nanofillers, and variations in joint geometry greatly influence the shear strength, fracture toughness, fatigue, and vibration behaviour of FRP composite joints.

Fatigue Performance Analysis of Adhesive-bonded Composite Joints Considering Size Effect

Adhesive bonding failure is one of the main failure modes threatening the safety of the wind turbine blade (WTB), which drives fatigue performance analysis of adhesive bonding becoming a vital issue. Attributed to the high cost and sample scarcity of full-scale structural testing of WTBs, fatigue analysis of adhesive-bonded composite joints becomes a critical way to investigate the bonding properties. Adhesive bonding distributes in WTB with different sizes; thus, the fatigue performance in different place have a large difference. The existing researches mainly perform fatigue properties studies on the standard specimens, while the effects of bonding size on structural fatigue properties are ignored. With these concerns, a finite element method based on Cyclic Cohesive Zone Model theory is proposed, and size effects on fatigue properties are studied for adhesive-bonded composite joints in this paper. The results can be used for fatigue performance analysis for WTB adhesive bonding.

Toughening adhesive joints through crack path engineering using integrated polyamide wires

Ensuring the progressivity of failure of adhesively-bonded composite joints is necessary to guarantee safety and to optimize maintenance operations. In our previous work, we proposed a novel surface patterning strategy to stop crack propagation by triggering bridging of adhesive ligaments. However, the brittle failure of classical bridging ligaments still releases a large amount of stored elastic energy, leading to a snap-slip crack propagation or even catastrophic sudden fracture of bonded joints. Such technology could be further improved by integrating ductile structures within the adhesive layer, but the detailed failure mechanisms require systematic investigation. In this work, we integrated thermoplastic polyamide structures within the epoxy adhesive layer of double cantilever beams to guide this transition from brittle failure to a stable softening behavior. Weak polyamide/epoxy adhesion and their embedded area fractions were critical since they affected the damage mechanisms and determined energy dissipation within bonded joints.

Material and adhesive effect in adhesively-bonded composite stepped-lap joints

Adhesive bonding is a predominant bonding technique in the aeronautical and automotive industries. Cohesive zone models, used together with the finite element method, are a viable tool to predict the strength of adhesive joints. The main objective of this study is to evaluate experimentally and numerically (by cohesive zone model) the mechanical performance of carbon-fiber reinforced polymer stepped-lap bonded joints submitted to tensile loads, for different overlap lengths ( LO) and adhesives. The failure mode analysis showed a predominant failure type for all adhesives and good correspondence with the numerical predictions. Normalized peel ( σy) and shear ( τxy) stresses in the adhesive highly increased with LO, which then reflected on different maximum load ( Pm) evolution with LO, depending on the adhesive's ductility. The damage variable SDEG (stiffness degradation) was also evaluated and emphasized on the smaller damage zone at Pmfor the brittle adhesive. A significant geometry and material effect were found on Pmof the stepped-lap joints, with benefit for large LO. In this regard, cohesive zone model revealed to be a suitable tool in determining the behavior of different joints. Comparison with joints with aluminum showed that, provided that no carbon-fiber reinforced polymer delamination occurs, stepped-lap joints between carbon-fiber reinforced polymer adherends give better results due to the higher composite stiffness.

An integrated constitutive model for tensile failure analysis and overlap design of adhesive-bonded composite joints

An integrated constitutive model is proposed for tensile failure analysis and overlap design of adhesive-bonded composite joints. In this model, two improved criteria are developed: for matrix failure and fiber failure of laminates, an improved 3D strain-based Tsai-Wu criterion is developed by introducing mode factors and an exponential progressive damage evolution; for delamination and debonding, an improved 3D coupled exponential cohesive zone model (CZM) is developed by introducing new tangential components. Both criteria are implemented in one explicit user subroutine to facilitate applications for engineering. The validity of the model is proved in comparison with the experimental results of adhesive-bonded composite single-lap joints and other numerical models. Based on the model, the influence of different overlap lengths on the tensile failure behavior of the adhesive-bonded composite single-lap joints is analyzed comprehensively. For the adhesive-bonded composite single-lap joints tested in this study, an equivalent load-weight ratio is proposed to reflect the relative load capacity, and the optimal value of the overlap length is discussed.

Quality assessment and damage detection in nanomodified adhesively-bonded composite joints using inkjet-printed interdigital sensors

In this work, the development of a planar interdigital capacitive sensor, directly onto the surface of a composite, for determining the initial quality of curing of bonded composite joints and assessing their long-term durability is presented. The sensor consisted of an interlocking comb-shaped array of silver electrodes and used to monitor the progress of cure of an adhesive resin and the subsequent damage state of the bond line in adhesively-bonded composite joints using impedance spectroscopy. The obtained results from the mechanical characterization indicated that the developed sensor did not affect the quality of the bondline while the added weight of the sensor is negligible. The curing process of the adhesive epoxy was successfully monitored while the ability of the sensor to assess the developed damage created by the mechanical loading was confirmed using transient infrared thermography.

An investigation of in-plane dynamic behavior of adhesively-bonded composite joints under dynamic compression at high strain rate

This study examines the behaviour of adhesively-bonded composite joints under dynamic compression tests. The purpose of this work is to use the split Hopkinson pressure bar (SHPB) for the dynamic characterization of adhesively bonded joints subjected to in-plane compression loading and in particular, the effect of strain rate on the mechanical behaviour and the damage kinetics. These joints are studied using glass/vinylester composite materials which are frequently used in naval applications Compression tests are performed at different strain rates using SHPB and high speed camera has been used to follow the damage progression. The experimental results have shown that the dynamic properties change with respect to the change in strain rate. Fibre buckling and delamination are the main damage criterias seen in the specimens under in-plane compressive tests. Therefore, this study not only allows us to understand the dynamic response of the adhesively bonded joints under dynamic compression but also enables us to establish damage models based on strain rate effect, for structure design purposes.

Enhancing fatigue resistance and damage characterisation in adhesively-bonded composite joints by carbon nanofibres

In the present work we report on the use of carbon nanofibres (CNFs) to simultaneously improve the cyclic fatigue resistance and the detectability of disbonding in adhesively-bonded structures made of carbon-fibre reinforced-plastic (CFRP) composites. The effects of the concentration of the CNFs (i.e. 0.4, 0.7 and 1.0 wt%) and their orientation (i.e. random versus aligned) in the epoxy-adhesive layer between two CFRP substrates are investigated. The results show that increasing the concentration of randomly-oriented CNFs (a) improves greatly the mode I fatigue resistance of the adhesive layer, including raising the crack growth threshold of the cyclic strain-energy release-rate, and (b) increases the quasi-static fracture toughness. Further improvements in the fatigue resistance occur when the CNFs are aligned perpendicular to the plane of the joint, i.e. normal to the crack plane, as opposed to being randomly-oriented in the adhesive layer. In addition, the CNFs form a conductive network that makes it possible to detect and characterise fatigue-induced disbonding using an electrical-resistance technique. A simple model is developed for the relationship between the disbond (i.e. crack size) and the electrical resistance of a bonded joint with conductive substrates. Finite element analyses are carried out to quantify the applicability of this model as a function of the conductivity of the adhesive from 10−4 S/m to 1 S/m. The results confirm that the proposed simple model is highly accurate for joints where the composite substrates have a through-thickness electrical conductivity exceeding a hundred times that of the adhesive. This research paves the way for new multi-functional adhesives with greatly enhanced fatigue resistance and disbond detection capability.

Enhancement on mechanical strength of adhesively-bonded composite lap joints at cryogenic environment using coiled carbon nanotubes

Abstract The hardness, tensile and lap joint shear behaviors of pure epoxy, straight multi-walled carbon nanotube (MWNT)/epoxy and coiled multi-walled carbon nanotube (CCNT)/epoxy adhesives conditioned at room temperature (RT) and cryogenic temperature (CT) were investigated in the present study. Experimental results showed that all adhesives had greater Vickers hardness values, Young's moduli and tensile strengths at CT. The performance of CCNT/epoxy adhesive at CT was outstanding due to the enhancement of mechanical interlocking effect between CCNTs and epoxy at low temperature. This effect led to a greater Vickers hardness value, Young's modulus and lap joint shear strength of this adhesive at CT when compared with MWNT/epoxy type. The result from finite element analysis (FEA) also proved that the contraction of matrix at CT induced additional clamping force onto the surface of nanotubes. Due to the larger surface area of CCNTs, a relatively stronger bonding strength was achieved, and thus, CCNT/epoxy adhesive had better mechanical properties at low temperature condition.

Influence of plies’ orientations on the stress distribution in adhesively bonded laminate composite joints subjected to impact loadings

Adhesive bonding of structural composite structures is increasingly used in several engineering applications. The mechanical behavior of adhesively-bonded composite joints is highly influenced by the reinforcements’ orientations. In this work, the influence of composite laminates’ orientation, on the peel and shear stresses in the adhesive layer of composite double lap joints subjected to in-plane impact loading, was investigated. To this end, three-dimensional finite elements analysis was undertaken. First, four laminate orientations/configurations were considered: 0°, ±45°, 90°, 0/90°. Second, configurations were chosen in a manner to explore the influence of plies’ positions. Starting from the 0°, one or two ply(ies) are replaced by one or two ply(ies), respectively, which are oriented in 90° or ±45° directions. It was found that the profile of the substrates and adhesive stresses variation along the axial/longitudinal direction is not affected by the substrates orientations. However, the highest adhesive stresses are achieved when fibers are all or mostly oriented along the loading/axial direction [0°] and the lowest adhesive stresses are obtained when the fibers are oriented perpendicular to the applied load [90°]. It is then concluded that the adhesive stresses increase as the substrates longitudinal stiffness increases. It is also concluded that the stress in the adhesive layer is mostly influenced by the closest laminate plies and almost not affected by the far plies.

Mixed-Mode I/II fracture behavior of asymmetric composite joints

The partitioning of the total strain energy release rate, Gtot, into the Mode I, GI, and Mode II, GII, components is challenging, especially in asymmetric cracks. Although many researchers have studied the fracture behavior of composite materials under mixed-mode loading using symmetric and/or asymmetric mixed-mode bending (MMB) specimens, few studies exist in the literature regarding the experimental investigation and analysis of the fracture behavior of asymmetric MMB specimens comprised of orthotropic layered adherends.In this study, a new approach, designated as the “extended global method”, was established the fracture mode partitioning. The “extended global method” was applied for the analysis of the experimental data. The mixed-mode fracture behavior of adhesively-bonded composite joints was experimentally investigated using asymmetric mixed-mode bending specimens. Finite element models were developed in order to validate the approach. The virtual crack closure technique was used for the calculation of the fracture components at the crack tip and an exponential traction-separation cohesive law was used to simulate the fiber bridging zone. The crack propagated along paths outside the symmetry plane and, therefore, mode partition could not be performed in commonly accepted ways followed for symmetric specimens. In addition, the experimental compliance method was used for calculating the fracture energy of the examined asymmetric mixed-mode bending specimens. Results obtained using the “extended global method” and the experimental compliance method were in good agreement with the results from FE models. The outcome of this study combined with the results obtained from pure Mode I and Mode II experiments can be used to establish failure criteria for the examined joints, that can be eventually be used for designing structural joints with the same adherends and adhesives.

Influence of Dissimilar Adherends on the Stress Distribution in Adhesively Bonded Composite Joints Subjected to Impact Loadings

The influence of nonsymmetric rotation of laminates on the shear and peel stresses in the adhesive layer of adhesively bonded double-lap composite joints (DLJ) subjected to in-plane impact compressive loadings is investigated by using a three-dimensional finite-element analysis. The compressive in-plane impact on DLJ is simulated using the direct Hopkinson bar system, and the specimen is impacted by an incident bar. It is found that the rotation of any adherend from the 0° orientation leads to a decrease in the average shear stress in the adhesive layer, but the maximum peel stress is affected only by the longitudinal stiffness of the outer adherend and decreases when this stiffness diminishes.

Wave propagation analysis in adhesively bonded composite joints using the wavelet spectral finite element method

A wavelet spectral finite element (WSFE) model is developed for studying transient dynamics and wave propagation in adhesively bonded composite joints. The adherands are formulated as shear deformable beams using the first order shear deformation theory (FSDT) to obtain accurate results for high frequency wave propagation. Equations of motion governing wave motion in the bonded beams are derived using Hamilton’s principle. The adhesive layer is modeled as a line of continuously distributed tension/compression and shear springs. Daubechies compactly supported wavelet scaling functions are used to transform the governing partial differential equations from time domain to frequency domain. The dynamic stiffness matrix is derived under the spectral finite element framework relating the nodal forces and displacements in the transformed frequency domain. Time domain results for wave propagation in a lap joint are validated with conventional finite element simulations using Abaqus®. Frequency domain spectrum and dispersion relation results are presented and discussed. The developed WSFE model yields efficient and accurate analysis of wave propagation in adhesively-bonded composite joints.

The fracture behaviour of adhesively-bonded composite joints: Effects of rate of test and mode of loading

The present paper discusses the results of an investigation into the effects of test rate and the mode of loading on the fracture energy, Gc, of adhesively-bonded fibre-composite joints. Various carbon-fibre reinforced-polymer (CFRP) matrix composite substrates have been bonded using two different types of automotive structural epoxy-adhesives. They have been tested via loading the bonded joints in mode I (tensile), mode II (in-plane shear) and mixed-mode I/II from slow rates (i.e., of about 10−5m/s) up to relatively high rates of test of about 15m/s. The high-rate tests were photographed using a high-speed digital video camera to record the deformation of the joint and the fracture behaviour. An analysis strategy has been developed for the various modes of loading (i) to account for the observed fracture behaviour, (ii) to circumvent the problems posed by oscillations in the load traces due to the presence of dynamic effects in the faster tests, and (iii) to account for the kinetic energy associated with the moving specimen arms in the faster tests. Based on the analysis strategy developed, the effect of the test rate on the fracture energy, Gc, for the different loading modes for the joints has been ascertained. Furthermore, various different fracture paths were observed in the tests. They were either cohesive, in the adhesive layer, or interlaminar in the composite substrates. The exact fracture path observed was a function of (i) the type of composite substrate, (ii) the type of adhesive, and (iii) the mode of loading employed. However, the nature of the fracture path was found to be quite insensitive to the test rate. Essentially, it was found that joints subjected to mixed-mode I/II loading were more likely to exhibit an interlaminar fracture path in the composite substrates than when loaded in either pure modes I or II. The propensity for a given joint to exhibit such a fracture path via delamination of the composite substrate has been explained by calculating the transverse tensile stresses induced in the loaded composite arms, and comparing this value to the measured transverse tensile strength of the composite. Following this approach, the underlying reasons for the observed fracture path were identified and could be predicted. Also, the proposed scheme provides a route to design against delamination failure occurring in adhesively-bonded fibre-composite test specimens.

Adhesive-Bonded Composite Joints Analysis with Delaminated Surface Ply Using Strain-Energy Release Rate

This paper presents an analytical model to determine the strain energy release rate due to an interlaminar crack of the surface ply in adhesively bonded composite joints subjected to axial tension. Single-lap shear-joint standard test specimen geometry with thick bondline is followed for model development. The field equations are formulated by using the first-order shear-deformation theory in laminated plates together with kinematics relations and force equilibrium conditions. The stress distributions for the adherends and adhesive are determined after the appropriate boundary and loading conditions are applied and the equations for the field displacements are solved. The system of second-order differential equations is solved to using the symbolic computation tool Maple 9.52 to provide displacements fields. The equivalent forces at the tip of the prescribed interlaminar crack are obtained based on interlaminar stress distributions. The strain energy release rate of the crack is then determined by using the crack closure method. Finite element analyses using the J integral as well as the crack closure method are performed to verify the developed analytical model. It has been shown that the results using the analytical method correlate well with the results from the finite element analyses. An attempt is made to predict the failure loads of the joints based on limited test data from the literature. The effectiveness of the inclusion of bondline thickness is justified when compared with the results obtained from the previous model in which a thin bondline and uniform adhesive stresses through the bondline thickness are assumed.

Fracture behaviour of adhesively bonded single lap joint with 45° ply oriented four layered symmetric and antisymmetric adherends

In aerospace and many types of civil infrastructure applications, the use of adhesive-bonded composite joints have been constantly increasing. In this study, fracture behaviour of adhesively bonded single lap joint (SLJ) with pre-existing damage at the interface of free edge of the top adherend and adhesive along the width of the joint is studied by strain energy release rate (SERR) approach using virtual crack closure technique (VCCT). Adherends are four layered laminated fiber reinforced plastic (FRP) composite plates and each lamina of adherend having either +45° or -45° fiber angle. The distributions of strain energy release rates along the delamination front are predicted by evaluating 3D local displacement fields and reaction forces required to close them, along the delamination length and width. The present investigation reveals that strain energy release rates are not constant along the delamination front, insisting on three dimensional modeling of problem. It is observed that location of damage propagation and mode of failure responsible for damage propagation varies with lay-up sequence of +45° or -45° ply oriented laminas within the top and bottom adherends. Total strain energy release rate (G) values are very high in the SLJs having (+45/-45)s FRP laminated adherends and in the SLJ with (-45/+45)s FRP laminated adherends, when compared with SLJs having (+45/-45)asand/or (-45/+45)as FRP laminated adherends. Key words: Delamination length, strain energy release rate (SERR), virtual crack closure technique (VCCT), suffix s and as, fiber angle.

Strain-Energy Release Rate Analysis of Adhesive-Bonded Composite Joints with Prescribed Interlaminar Crack

Thesis [M.S] - Wichita State University, College of Engineering, Dept. of Aerospace Engineering

The Effects of Pre-Bond Moisture on the Fracture Behaviour of Adhesively-Bonded Composite Joints

The results of an investigation into the effects of pre-bond moisture absorbed by fibre-composite substrates prior to bonding with various structural epoxy adhesives are presented. Substrates were bonded in the as-received condition (where substrates had been exposed to atmospheric moisture for periods of greater than three months) and were also bonded in the fully-dried condition (after drying under vacuum at 105°C for 28 days). Additionally, substrates were conditioned by water submersion for various durations prior to bonding. Double cantilever beam tests were performed on the resulting joints to determine the adhesive fracture energy, G IC. The effect of pre-bond moisture on the glass transition temperature of the adhesive was also determined. One adhesive was shown to exhibit an extreme sensitivity to pre-bond moisture. A severe reduction in fracture energy accompanied a change in the fracture morphology and Tg. Other adhesives were shown to be relatively insensitive to the levels of pre-bond moisture introduced.

Stress and Failure Analyses of Adhesive-Bonded Composite Joints Using ASTM D3165 Specimens

An analytical model was developed to determine the stress and strain distributions of single-lap adhesive-bonded composite joints under tension. Both the adherend and the adhesive were assumed linear elastic in the derivations. The Laminated Plate Theory was used in defining the mechanical behavior of the composite adherends, whereas the linear elasticity theory was applied to describe the material response of the adhesive. By doing so, the stresses in the adhesive can vary through the bondline thickness. After the overall system of governing equations was determined by energy method, it was solved by using a symbolic solver—Maple with appropriate boundary conditions. Results from the analytical model were verified with finite element analysis using commercial software ABAQUS. Single-lap composite joint experiments were conducted following ASTM D3165, Strength Properties of Adhesives in Shear by Tension Loading of Single-Lap-Joint Laminated Assemblies. Although all three failure modes of bonded joints, substrate, cohesive, and adhesive failure, were present as the experimental results, the substrate failure mode was the major failure mode observed. Therefore, only substrate failure mode was analyzed using the developed model in the present paper. Four failure criteria, Tsai-Hill failure criterion, von Mises failure criterion, maximum interlaminar tensile stress criterion, and maximum normal stress criterion, were used to correlate the stresses and failure load. Nonlinear regression was conducted to determine the necessary parameters in the failure criteria. Based on the experimental results, thicker bondlines result in weaker joints. The variation of failure load for joints with various bondline thicknesses was consistent with the predicted results.

Stress and Failure Analyses of Adhesive-Bonded Composite Joints Using ASTM D3165 Specimens

An analytical model was developed to determine the stress and strain distributions of single-lap adhesive-bonded composite joints under tension. Both the adherend and the adhesive were assumed linear elastic in the derivations. The Laminated Plate Theory was used in defining the mechanical behavior of the composite adherends, whereas the linear elasticity theory was applied to describe the material response of the adhesive. By doing so, the stresses in the adhesive can vary through the bondline thickness. After the overall system of governing equations was determined by energy method, it was solved by using a symbolic solver—Maple with appropriate boundary conditions. Results from the analytical model were verified with finite element analysis using commercial software ABAQUS. Single-lap composite joint experiments were conducted following ASTM D3165, Strength Properties of Adhesives in Shear by Tension Loading of Single-Lap-Joint Laminated Assemblies. Although all three failure modes of bonded joints, substrate, cohesive, and adhesive failure, were present as the experimental results, the substrate failure mode was the major failure mode observed. Therefore, only substrate failure mode was analyzed using the developed model in the present paper. Four failure criteria, Tsai-Hill failure criterion, von Mises failure criterion, maximum interlaminar tensile stress criterion, and maximum normal stress criterion, were used to correlate the stresses and failure load. Nonlinear regression was conducted to determine the necessary parameters in the failure criteria. Based on the experimental results, thicker bondlines result in weaker joints. The variation of failure load for joints with various bondline thicknesses was consistent with the predicted results.