- Email: [email protected]
Mixed-Mode Cohesive Law Estimation of Composite Joints Made of Toughened Epoxy Adhesive
Available online at www.sciencedirect.com Available online at www.sciencedirect.com
ScienceDirect ScienceDirect
Structural Integrity 00 (2018) 000–000 Available online at www.sciencedirect.com Available online atProcedia www.sciencedirect.com
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Structural Integrity Procedia 00 (2018) 000–000
ScienceDirect ScienceDirect
Procedia Structural Structural IntegrityIntegrity Procedia1400(2019) (2016)354–361 000–000
www.elsevier.com/locate/procedia
2nd International Conference on Structural Integrity and Exhibition 2018 2nd International Conference on Structural Integrity and Exhibition 2018
Mixed-Mode Cohesive Law Estimation of Composite Joints Made Mixed-Mode Cohesive Law Estimation of Composite Joints Made of on Toughened Adhesive XV Portuguese Conference Fracture, PCF Epoxy 2016, 10-12 February 2016, Paço de Arcos, Portugal of Toughened Epoxy Adhesive Tauheed, Datla** Thermo-mechanicalMohd modeling ofNaresh a highV. turbine blade of an Mohd Tauheed, Naresh V.pressure Datla Department of Mechanical Engineering, IIT Delhi, Hauz Khas, New Delhi, India 110016 airplane gas turbine engine Department of Mechanical Engineering, IIT Delhi, Hauz Khas, New Delhi, India 110016 a
b
c
P. Brandão , V. Infante , A.M. Deus * Abstract Abstracta Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Joining composites using adhesive bonding is attractive because they reduce the weight of structure and allow to join complex Portugal Joining composites using adhesiveuse bonding is attractive because they reduce theand weight of structure and Pais, allow joincomposite complex b shapes. TheseDepartment benefits encourage of composite adhesive joints in aerospace automotive industries. However, IDMEC, of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco 1,to 1049-001 Lisboa, shapes. These encourage useprimary of composite adhesive joints in aerospace and automotive industries. However, composite adhesive jointsbenefits are seldom used for structures because of our limited understanding of their failure, especially under Portugal c CeFEMA, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, adhesive joints are Predicting seldom used primary structures because of our limited understanding their failure, especially under mixed-mode loads. thefor failure of composite joints is challenging because the failureof can occur cohesive in adhesive, mixed-mode loads. Predicting the failureorofwithin composite joints isPortugal challenging the these failurefailures can occur cohesive in specimen adhesive, interfacial between adhesive/adherend, the composite adherend. because Moreover, depend on the interfacial between conditions, adhesive/adherend, or within the adherend. Moreover, these failures depend oncohesive the specimen geometry, loading surface treatments, andcomposite environmental conditions. Recent studies showed that zone geometry, loading conditions, and environmental that and cohesive approach can be used to reliablysurface predict treatments, failure, but most of these studiesconditions. are limited Recent to failurestudies under showed mode I loads furtherzone for Abstract approach can be used to reliably predicttraction-separation failure, but most oflaws these(TSL) studieswere are limited to failure under mode I loads further for brittle epoxy adhesives. In this study, extracted for composite joints madeand of toughened brittle epoxy adhesives. In this study, traction-separation laws (TSL) extracted for technique. composite These joints TSLs made were of toughened epoxy adhesive through fracture tests and by applying the digital imagewere correlation (DIC) used for During their operation, modern aircraft engine components are subjected to increasingly demanding operating conditions, epoxy through fracture tests andsubjected by applying the digital image correlation (DIC)adhesive technique. These TSLs wereofused for strengthadhesive prediction of composite joints to mixed-mode loading. Composite joints were made carbon especially the high pressure turbine (HPT) blades. Such conditions cause these parts to undergo different types of time-dependent strength prediction of adherend compositeand joints subjected to mixed-mode loading. Composite adhesive joints were made of carbon fiber/epoxy composite Araldite 2015 epoxy adhesive. Mode I and mode II fracture testing were conducted using degradation, one of which is creep. A model using the finite element method (FEM) was developed, in order to be able to predict fiber/epoxy composite adherend and Araldite 2015 adhesive. Mode and IIfracture fracture testing were conducted using the double and end notch flexural specimens, respectively. Frommode these tests, TSLs were extracted by the creepcantilever behaviourbeam of HPT blades. Flight dataepoxy records (FDR) for a Ispecific aircraft, provided by a commercial aviation the double cantilever beam and end notch flexural specimens, respectively. From these fracture tests, TSLs were extracted by using a direct method based on the DIC technique. These TSLs were used in a finite element (FE) model of a lap shear joint company, were used to obtain thermal and mechanical data for three different flight cycles. In order to create the 3D model using aindirect method based on the DICstrength. technique. These TSLs were used in a finite element (FE) model of lap shear joint model ANSYS to predict the failure This FE predicted failure reasonably agreed with thea experimentally needed for the FEM analysis, a HPT blade scrap was scanned, and itsstrength chemical composition and material properties were model in ANSYS to that predict failure strength. This FE predicted strength reasonablywere agreed determined failure strength ofthe the toughened adhesive joint. obtained. The data was gathered was fed into the FEM model failure and different simulations run,with first the withexperimentally a simplified 3D determined failure of order the toughened adhesive the joint. rectangular blockstrength shape, in to better establish model, and then with the real 3D mesh obtained from the blade scrap. The © overall 2018 The Authors. Published Elsevier B.V. expected behaviour in by terms of displacement was observed, in particular at the trailing edge of the blade. Therefore such a © 2019 The Authors. Published by Elsevier B.V. B.V. © 2018 The Authors. byofElsevier model can be access usefulPublished in the goal predicting turbine blade life,(https://creativecommons.org/licenses/by-nc-nd/4.0/) given a set of FDR data. This is an open article under the CC BY-NC-ND license This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) This is an and openpeer-review access article under the CC BY-NC-ND licenseunder (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection under responsibility of Peer-review responsibility of the 2018 organizers. Selection and peer-review under responsibility of Peer-review under responsibility of the SICE 2018SICE organizers. © 2016 and Thepeer-review Authors. Published by Elsevier B.V. Selection under responsibility of Peer-review under responsibility of the SICE 2018 organizers. Keywords: Cohesive zone modelling; mixed-mode; digital Committee image correlation; toughened Peer-review under responsibility of the Scientific of PCF 2016. epoxy Keywords: Cohesive zone modelling; mixed-mode; digital image correlation; toughened epoxy Keywords: High Pressure Turbine Blade; Creep; Finite Element Method; 3D Model; Simulation. * Corresponding author. Tel.:+91-11-2659-6071 address: [email protected] *E-mail Corresponding author. Tel.:+91-11-2659-6071 E-mail address: [email protected]
2452-3216 © 2018 The Authors. Published by Elsevier B.V.
This is an © open article under theby CCElsevier BY-NC-ND 2452-3216 2018access The Authors. Published B.V. license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection under responsibility of Peer-review responsibility of the SICE 2018 organizers. This is an and openpeer-review access article under the CC BY-NC-ND licenseunder (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under * Corresponding author. Tel.: +351responsibility 218419991. of Peer-review under responsibility of the SICE 2018 organizers. E-mail address: [email protected]
2452-3216 © 2016 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the Scientific Committee of PCF 2016.
2452-3216 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of Peer-review under responsibility of the SICE 2018 organizers. 10.1016/j.prostr.2019.05.044
2
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
355
1. Introduction Manufacturing industries like aircraft, marine and automotive industries are replacing the classical joining methods with some modern techniques such as adhesive bonding this is due to their high structural integrity and reliability. However, the formation of the reliable adhesive joint in between the composite components is not an easy task and requires greater insight. To study the failure of adhesive joint and modelling of failure, cohesive zone models (CZM) has been extensively used as described in Constante and Moura (2015). Some other researcher studied to predict the fracture toughness of degraded adhesive joints by cohesive zone modeling with fracture data of accelerated aging tests. Degraded fracture toughness predictions were done by calculating the exposure index values and thereby the degraded cohesive parameters across the width of the closed joints as described in Patil et al. (2017). Some researcher successfully studies the critical energy release rate was the major parameter characterizing the fracture of adhesive. They investigated numerically and experimentally of mixed-mode fracture testing of CFRP composite joints with a film adhesive Balzani et al. (2012). Several studies have been done to evaluate mode I and mode II traction-separation laws using different methods to find the cohesive properties near the crack tip of the specimens. Few works focus on the shape of the traction-separation laws to model the adhesive layers as described detail in Campilho et al. (2013) for the best results in strength prediction. Several studies have shown that shape of the traction-separation laws are triangular in brittle adhesives, trapezoidal in ductile adhesives and exponential shapes are also fitted with different properties of adhesives as presented by Campilho et al. (2015). Some other researcher investigated the effect of adhesive layer thickness on the fracture behaviour of the adhesive joint. A recent study Carlberger et al. (2010) successfully investigates the effect of adhesive thickness on the tractionseparation law that indicates the fracture energy is more sensitive to the adhesive layer thickness than cohesive strength. When the failure is cohesive, the CZM parameters arises due to the plastic dissipation in the adhesive layers. Several studies also explore the effect of strain rate on the CZM parameters as described in Desai et al. (2015). The determination of the cohesive parameters and selection of CZM law also depends on several other parameters such as the adhesive material, moisture and temperature as deailed described in Budhe et al. (2017). Though practical joints are loaded under mixed-mode conditions, there is seldom work and this represents a research gap in the literature. In this study, traction-separation laws (TSL) were extracted for composite joints made of toughened adhesive by applying the digital image correlation (DIC) technique. These TSLs were used for strength prediction of composite joints subjected to mixed-mode loading. Composite adhesive joints were made of carbon fiber/epoxy composite adherend and Araldite 2015 epoxy adhesive. Initially, a suitable surface pre-treatment was established to be lateral sanding with 220 grit sandpaper followed by acetone cleaning. Mode I and mode II fracture testing were conducted using the double cantilever beam and end notch flexural specimens, respectively. In this study, fracture energies were taken only at the crack initiation. From these fracture tests, TSLs were extracted by using a direct method based on the digital image correlation technique. These TSLs were used in a finite element (FE) model of a single lap shear (SLS) joint model in ANSYS to predict the failure strength. This FE predicted failure strength reasonably agreed with the experimentally determined failure strength of the toughened adhesive joint. Nomenclature m B P or 𝑃𝑃��� 𝐺𝐺�� , 𝐺𝐺��� GI, GII 𝑎𝑎 , 𝑎𝑎� 𝑑𝑑𝑑𝑑 𝛥𝛥 T 𝛿𝛿𝛿 𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿 𝐺𝐺�𝛿𝛿� 𝛿𝛿�
compliance calibration coefficient specimen width maximum load critical energy release rate for mode I and mode II energy release rate depends on the current state of normal and shear stress at the crack tip initial crack length effective crack length to correct for rotation of DCB arms at crack front displacement traction separation, normal separation and shear separation curve fit polynomial equation maximum separation in the adhesive layer.
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
356
3
1.1. Material Specification and characterization The material considered throughout this study is a unidirectional carbon fiber prepreg from Hindoostan Mills Ltd. was used for the making of composite laminates. Lamination involved packing the layup in a vacuum bag and heat curing using the cycle given by the manufacturer. Considering that the loading is only in one direction, the layup used for the laminate was [0]10, i.e. 10 layers in 0° orientation. The prepreg sheets were orientated and then laid on a flat aluminium plate coated with release agent. Then, a peel fabric was laid and the layup was then sealed in a vacuum bag and vacuumed up to 0.8 bar. Vacuum assisted in the removal of trapped air and applied uniform pressure over the laminate during the curing process. The whole setup was then placed in an oven and a cuing cycle of 45 minutes at 120 ℃ was followed. The cured laminate was then released from aluminium plate. An average laminate thickness of 1.7 mm was achieved. The uniaxial tensile test were performed using Instron universal testing machine (capacity 100 kN) according to the ASTM D3039 standard in the longitudinal direction of the laminate to determine tensile modulus. E11 and ν12 were determined by applying DIC technique to the strain field. The fiber volume fraction (Vf) and E22 were determined using the mechanics of material approach given in Jones (1997). Industrial grade two-part structural toughened epoxy adhesive Araldite 2015 from Huntsman were procured. The rest of the mechanical properties of unidirectional CFRP laminate and the properties of toughened epoxy adhesive were taken from the literature as shown in Table 1. Table 1. Mechanical properties of unidirectional CFRP laminate and properties of Araldite 2015. Values in bold are for composite and adhesive were taken from Campilho et al. (2013) Modulus of elasticity Poisson’s ratio Modulus of rigidity Araldite 2015 E11 = 102 GPa
ν12 = 0.167
G12 = 4.3 GPa
E22 = 4.9 GPa
ν13 = 0.167
G13 = 4.3 GPa
E33 = 4.9 GPa
ν23 = 0.38
G23 = 3.2 GPa
2. Double cantilever beam (DCB) test
E = 1.85 � 0.21 GPa ν = 0.33
G = 0.56 � 0.21 GPa
A DCB specimen was used to determine mode I fracture parameters. Composite adhesive joints were made of carbon fiber/epoxy composite adherend and toughened epoxy adhesive (Araldite 2015). Prior to bonding the composite laminates were pre-treated by sanding with 220 grit sandpaper in the direction lateral to the length of the specimen, which was followed by acetone cleaning. Adhesive thickness, 𝑡𝑡� = 0.3 mm was maintained by providing shims or razor blade in between the adherends. ASTM D5528 standard test method was followed in DCB specimen preparation as shown in Fig.1 and determining the critical energy release rate in mode I (GIC). Here, ‘a’ is the initial crack length measured from loading point to the crack tip.
Fig. 1. The geometry of DCB specimen
Fig. 2 shows the experimental setup used for testing for DCB specimens under displacement control at 0.5 mm/min using a 10 kN universal testing machine (Shimadzu, AG-Xplus Table-top type). The 𝐺𝐺�� value was calculated by modified beam theory using Eq. 1. To obtain the cohesive parameters the displacement field about the crack tip was captured using a 5MP monochromatic CCD camera (Point Grey GS3-U3-51S5M-C 2/3” Grasshopper USB 3.0) and was analysed using the DIC technique.
4
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
DCB Fixture
357
Load cell DCB
Vertical slide
Camera & Lens Support Fixture
Fig. 2. Experimental setup for DCB test
A 35 mm fixed focal length lens along with extension tubes were used to achieve an approximate field of view of 3 mm × 3 mm, and the resolution achieved was about 1.5 µm/pixel. The camera and lens were mounted on a manual vertical slide to facilitate the vertical motion of the camera and track the crack tip during loading of the specimen. 𝐺𝐺�� � �
���
��������
(1)
Fig 3. (a) Evolution of GI with separation δ for the DCB specimen (b) Experimentally obtained and bilinear approximations for mode I traction separation law for the DCB specimens at a fixed crack length.
2.1. Determination of Mode I Traction-separation Law The traction-separation law represented 𝑇𝑇 � ��𝛿𝛿� is obtained by differentiating the following equation as described in de Moura et al. (2012), �
���������𝐺𝐺 � � �� � 𝑇𝑇�𝛿𝛿� 𝑑𝑑𝑑𝑑
This leads to, traction value at any point,
(2)
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
358
𝑇𝑇�𝛿𝛿� �
5
�����
(3)
��
A direct method was implemented for the determination of TSL. In this method, the evolution of mode I strain energy release rate, GI was plotted against crack tip opening displacement for the DCB specimen at a specific crack length as shown in the Fig. 3 (a). Then a 6th degree polynomial was fit to the evolution of GI with separation. For individual crack length, a set of images were taken at the crack tip location during loading of the DCB specimen. In each image, the normal separation was determined using DIC technique. Two reference points, one at upper and one at the lower end of the adhesive layer were used to determine the separation Mode I traction-separation law for individual crack length was obtained by differentiating the 6th degree polynomial equation of GI and separation in the DCB specimen and bilinear approximations for these experimentally obtained mode I traction-separation laws are shown in Fig. 3 (b). The criteria for approximation is the area under the curve which corresponds to the critical fracture toughness GIC and equal maximum separation for a crack length. 3. End notch flexure (ENF) test An ENF specimen is widely used to determine pure mode II fracture parameter. ASTM D7905 standard test method was followed in ENF specimen preparation as shown in Fig. 4. Figure 5 shows an experimental setup used for testing for ENF specimens and determination of GIIC are given as follows: 𝐺𝐺��� �
� �� ���� ���
(4)
��
Fig. 4. Geometry of the ENF specimen
Load cell Vertical slide
ENF specimen
ENF Camera & lens Fixture Fixture for camera Fig. 5. Experimental setup for ENF test
6
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
y = 1.13E+09δ6 - 2.33E+08δ5 + 1.80E+07δ4 6.63E+05δ3 + 1.13E+04δ2 - 1.89E+00δ + 1.99E-03
3
GII, N/mm
2 1,5 1
90 Experimental
80 70 60 50 40 30 20
0,5 0
100
Shear traction, MPa
2,5
359
10 0
0,02
0,04
Shear separation (δ), mm
0
0
0,02 0,04 Shear separation (δ), mm
Fig. 6 (a) Evolution of GII as a function of separation δ during fracture test of ENF (b) Experimentally obtained mode II traction separation law and its bilinear approximation
3.1. Determination of Mode II Traction-separation Law The procedure is identical to the one followed to extract mode I TSL. Evolution of mode II energy release rate (GII) was plotted against separation across the adhesive thickness using the direct A method as shown in Fig. 6 (a). DIC technique was used to determine the shear separation (δ). Here, the specimen undergoes bending during a test and the adhesive layer no longer remains horizontal at the crack tip location. Therefore, pure horizontal displacements given by DIC cannot be considered as shear separation. Similar to DCB specimens, a 6th degree polynomial was fit to the evolution of GII with separation, which was later differentiated toBobtain the mode II TSL and bilinear approximations to the experimentally determined mode II TSL is shown in Fig. 6 (b). 4. Strength prediction of single lap shear joints An SLS specimen is used for determining failure strength under mixed-mode loading conditions. The ASTM D3163 standard test method was followed in specimen preparation as shown in Fig. 7. Figure 8 (a) shows the test setup used to test SLS speicmens under displacement control at a constant rate of 0.5 mm/min. The failure criteria for a mixed mode cohesive zone model is given by, ��
���
�
���
����
��
(5)
A 2D finite element cohesive zone model of single lap joint was create in ANSYS APDL. The cohesive parameters extracted from the mode I and mode II fracture tests were used for the cohesive elements. Two different FE models were developed. In one model, the adhesive layer was not modelled, instead zero thickness cohesive elements were used to define the fracture path. In another model, a finite adhesive thickness of 0.3 mm was modelled and fracture path was defined within the adhesive layer by using zero thickness cohesive elements. Figure 8 (b) shows that both FE models closely predict the experimentally measured load vs. displacement response, thereby validating the ability of the FE models to predict failure loads. Moreover, since the differences between both FE models are negligible, the effect of the compliant adhesive layer on the failure response are insignificant.
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
360
7
Fig. 7. Geometry of the SLS specimen
Load cell Moving jaw SLS specimen Fixed jaw
Experimental Predicted_without_adhesive Predicted_with_adhesive
(b)6000 5000
Failure Load, N
(a)
4000 3000 2000 1000 0
0
0,5
1
1,5
2
Displacement, mm
2,5
3
Fig. 8. (a) Experimental setup for SLS test (b) Comparison between predicted and experimental load-displacement curve of single lap joint
5. Conclusions This work aimed at extracting the cohesive laws of composite adhesive joints using the direct method and to assess the use of these cohesive laws in predicting failure strength of joints under mixed-mode loading. Fracture tests with DCB and ENF specimens along with DIC technique was used to extract the mode I and mode II TSLs, respectively. It was found that the experimentally determined TSLs can be closely approximated with the bi-linear TSL. These bilinear TSLs were then used in a FE model based on cohesive zone modelling to predict the failure response of SLS joints (mixed-mode loading). The FE predicted failure response of the SLS joints closely matched with the experiments on SLS joints, thereby validating the FE model and the extracted TSLs. The FE model further showed that the compliant adhesive layer has an insignificant effect on the failure response. Acknowledgements The authors acknowledge the financial support received from the Science and Engineering Research Board (SERB), Government of India under Grant No. ECR/2015/456. References Constante, C. J., Campilho, R. D.S.G., Moura, D. C., 2015. Tensile Fracture Characterization of Adhesive Joints by Standard and Optical Techniques” Engineering Fracture Mechanics 136, 292–304. Patil, O. R., Ameli, A., Datla, N. V., 2017. Predicting Environmental Degradation of Adhesive Joints Using a Cohesive Zone Finite Element Model Based on Accelerated Fracture Tests. International Journal of Adhesion and Adhesives 76, 54–60. Balzani, C., Wagner, W., Wilckens, D., Degenhardt, R., Büsing, S., Reimerdes, H.G., 2012. Adhesive Joints in Composite Laminates—A Combined Numerical/Experimental Estimate of Critical Energy Release Rates. International Journal of Adhesion and Adhesives 32, 23–38. Carvalho, U. T.F., Campilho, R. D.S.G., 2017. Validation of Pure Tensile and Shear Cohesive Laws Obtained by the Direct Method with SingleLap Joints. International Journal of Adhesion and Adhesives 77, 41–50. Campilho, R.D.S.G., Banea, M.D., Neto, J.A.B.P., da Silva, L.F.M., 2013a. Modelling Adhesive Joints with Cohesive Zone Models: Effect of the Cohesive Law Shape of the Adhesive Layer. International Journal of Adhesion and Adhesives 44, 48–56.
8
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
361
Campilho, R.D.S.G., Banea, M.D., Moura, D. C., da Silva, L.F.M., 2015. Adhesive Thickness Effects of a Ductile Adhesive by Optical Measurement Techniques. International Journal of Adhesion and Adhesives 57, 125–32. Carlberger, T., Stigh, U., 2010. Influence of Layer Thickness on Cohesive Properties of an Epoxy-Based Adhesive-an Experimental Study. Journal of Adhesion 86, 814–33. Desai, C.K., Basu, S., Parameswaran, V., 2015. Determination of Traction Separation Law for Interfacial Failure in Adhesive Joints at Different Loading Rates. Journal of Adhesion 92, 819–39. Budhe, S., Banea, M.D., de Barros, S., da Silva, L.F.M., 2017. An Updated Review of Adhesively Bonded Joints in Composite Materials. International Journal of Adhesion and Adhesives 72, 30–42. Jones, R.M., 1975. Mechanics of composite materials. CRC press. de Moura, M.F.S.F., Gonçalves, J.P.M., Magalhães, A.G., 2012. A Straightforward Method to Obtain the Cohesive Laws of Bonded Joints under Mode I Loading. International Journal of Adhesion and Adhesives 39, 54–59. Standard Test Method for Tensile Properties of Polymer Matrix Composite Materials, ASTM D3039/D3039M-14. Standard Test Method for Determination of the Mode I Interlaminar Fracture Toughness of unidirectional Fiber-Reinforced Polymer matrix composites, ASTM D5528-13. Standard Test Method for Determination of the Mode II Interlaminar Fracture Toughness of unidirectional Fiber-Reinforced Polymer matrix composites, ASTM D7905/D7905M-14. Standard Test Method for Determining Strength of Adhesively Bonded Rigid Plastic Lap-Shear Joints by Tension Loading, ASTM D3163-01 (Reapproved 2014).
ScienceDirect ScienceDirect
Structural Integrity 00 (2018) 000–000 Available online at www.sciencedirect.com Available online atProcedia www.sciencedirect.com
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Structural Integrity Procedia 00 (2018) 000–000
ScienceDirect ScienceDirect
Procedia Structural Structural IntegrityIntegrity Procedia1400(2019) (2016)354–361 000–000
www.elsevier.com/locate/procedia
2nd International Conference on Structural Integrity and Exhibition 2018 2nd International Conference on Structural Integrity and Exhibition 2018
Mixed-Mode Cohesive Law Estimation of Composite Joints Made Mixed-Mode Cohesive Law Estimation of Composite Joints Made of on Toughened Adhesive XV Portuguese Conference Fracture, PCF Epoxy 2016, 10-12 February 2016, Paço de Arcos, Portugal of Toughened Epoxy Adhesive Tauheed, Datla** Thermo-mechanicalMohd modeling ofNaresh a highV. turbine blade of an Mohd Tauheed, Naresh V.pressure Datla Department of Mechanical Engineering, IIT Delhi, Hauz Khas, New Delhi, India 110016 airplane gas turbine engine Department of Mechanical Engineering, IIT Delhi, Hauz Khas, New Delhi, India 110016 a
b
c
P. Brandão , V. Infante , A.M. Deus * Abstract Abstracta Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Joining composites using adhesive bonding is attractive because they reduce the weight of structure and allow to join complex Portugal Joining composites using adhesiveuse bonding is attractive because they reduce theand weight of structure and Pais, allow joincomposite complex b shapes. TheseDepartment benefits encourage of composite adhesive joints in aerospace automotive industries. However, IDMEC, of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco 1,to 1049-001 Lisboa, shapes. These encourage useprimary of composite adhesive joints in aerospace and automotive industries. However, composite adhesive jointsbenefits are seldom used for structures because of our limited understanding of their failure, especially under Portugal c CeFEMA, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, adhesive joints are Predicting seldom used primary structures because of our limited understanding their failure, especially under mixed-mode loads. thefor failure of composite joints is challenging because the failureof can occur cohesive in adhesive, mixed-mode loads. Predicting the failureorofwithin composite joints isPortugal challenging the these failurefailures can occur cohesive in specimen adhesive, interfacial between adhesive/adherend, the composite adherend. because Moreover, depend on the interfacial between conditions, adhesive/adherend, or within the adherend. Moreover, these failures depend oncohesive the specimen geometry, loading surface treatments, andcomposite environmental conditions. Recent studies showed that zone geometry, loading conditions, and environmental that and cohesive approach can be used to reliablysurface predict treatments, failure, but most of these studiesconditions. are limited Recent to failurestudies under showed mode I loads furtherzone for Abstract approach can be used to reliably predicttraction-separation failure, but most oflaws these(TSL) studieswere are limited to failure under mode I loads further for brittle epoxy adhesives. In this study, extracted for composite joints madeand of toughened brittle epoxy adhesives. In this study, traction-separation laws (TSL) extracted for technique. composite These joints TSLs made were of toughened epoxy adhesive through fracture tests and by applying the digital imagewere correlation (DIC) used for During their operation, modern aircraft engine components are subjected to increasingly demanding operating conditions, epoxy through fracture tests andsubjected by applying the digital image correlation (DIC)adhesive technique. These TSLs wereofused for strengthadhesive prediction of composite joints to mixed-mode loading. Composite joints were made carbon especially the high pressure turbine (HPT) blades. Such conditions cause these parts to undergo different types of time-dependent strength prediction of adherend compositeand joints subjected to mixed-mode loading. Composite adhesive joints were made of carbon fiber/epoxy composite Araldite 2015 epoxy adhesive. Mode I and mode II fracture testing were conducted using degradation, one of which is creep. A model using the finite element method (FEM) was developed, in order to be able to predict fiber/epoxy composite adherend and Araldite 2015 adhesive. Mode and IIfracture fracture testing were conducted using the double and end notch flexural specimens, respectively. Frommode these tests, TSLs were extracted by the creepcantilever behaviourbeam of HPT blades. Flight dataepoxy records (FDR) for a Ispecific aircraft, provided by a commercial aviation the double cantilever beam and end notch flexural specimens, respectively. From these fracture tests, TSLs were extracted by using a direct method based on the DIC technique. These TSLs were used in a finite element (FE) model of a lap shear joint company, were used to obtain thermal and mechanical data for three different flight cycles. In order to create the 3D model using aindirect method based on the DICstrength. technique. These TSLs were used in a finite element (FE) model of lap shear joint model ANSYS to predict the failure This FE predicted failure reasonably agreed with thea experimentally needed for the FEM analysis, a HPT blade scrap was scanned, and itsstrength chemical composition and material properties were model in ANSYS to that predict failure strength. This FE predicted strength reasonablywere agreed determined failure strength ofthe the toughened adhesive joint. obtained. The data was gathered was fed into the FEM model failure and different simulations run,with first the withexperimentally a simplified 3D determined failure of order the toughened adhesive the joint. rectangular blockstrength shape, in to better establish model, and then with the real 3D mesh obtained from the blade scrap. The © overall 2018 The Authors. Published Elsevier B.V. expected behaviour in by terms of displacement was observed, in particular at the trailing edge of the blade. Therefore such a © 2019 The Authors. Published by Elsevier B.V. B.V. © 2018 The Authors. byofElsevier model can be access usefulPublished in the goal predicting turbine blade life,(https://creativecommons.org/licenses/by-nc-nd/4.0/) given a set of FDR data. This is an open article under the CC BY-NC-ND license This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) This is an and openpeer-review access article under the CC BY-NC-ND licenseunder (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection under responsibility of Peer-review responsibility of the 2018 organizers. Selection and peer-review under responsibility of Peer-review under responsibility of the SICE 2018SICE organizers. © 2016 and Thepeer-review Authors. Published by Elsevier B.V. Selection under responsibility of Peer-review under responsibility of the SICE 2018 organizers. Keywords: Cohesive zone modelling; mixed-mode; digital Committee image correlation; toughened Peer-review under responsibility of the Scientific of PCF 2016. epoxy Keywords: Cohesive zone modelling; mixed-mode; digital image correlation; toughened epoxy Keywords: High Pressure Turbine Blade; Creep; Finite Element Method; 3D Model; Simulation. * Corresponding author. Tel.:+91-11-2659-6071 address: [email protected] *E-mail Corresponding author. Tel.:+91-11-2659-6071 E-mail address: [email protected]
2452-3216 © 2018 The Authors. Published by Elsevier B.V.
This is an © open article under theby CCElsevier BY-NC-ND 2452-3216 2018access The Authors. Published B.V. license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection under responsibility of Peer-review responsibility of the SICE 2018 organizers. This is an and openpeer-review access article under the CC BY-NC-ND licenseunder (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under * Corresponding author. Tel.: +351responsibility 218419991. of Peer-review under responsibility of the SICE 2018 organizers. E-mail address: [email protected]
2452-3216 © 2016 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the Scientific Committee of PCF 2016.
2452-3216 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of Peer-review under responsibility of the SICE 2018 organizers. 10.1016/j.prostr.2019.05.044
2
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
355
1. Introduction Manufacturing industries like aircraft, marine and automotive industries are replacing the classical joining methods with some modern techniques such as adhesive bonding this is due to their high structural integrity and reliability. However, the formation of the reliable adhesive joint in between the composite components is not an easy task and requires greater insight. To study the failure of adhesive joint and modelling of failure, cohesive zone models (CZM) has been extensively used as described in Constante and Moura (2015). Some other researcher studied to predict the fracture toughness of degraded adhesive joints by cohesive zone modeling with fracture data of accelerated aging tests. Degraded fracture toughness predictions were done by calculating the exposure index values and thereby the degraded cohesive parameters across the width of the closed joints as described in Patil et al. (2017). Some researcher successfully studies the critical energy release rate was the major parameter characterizing the fracture of adhesive. They investigated numerically and experimentally of mixed-mode fracture testing of CFRP composite joints with a film adhesive Balzani et al. (2012). Several studies have been done to evaluate mode I and mode II traction-separation laws using different methods to find the cohesive properties near the crack tip of the specimens. Few works focus on the shape of the traction-separation laws to model the adhesive layers as described detail in Campilho et al. (2013) for the best results in strength prediction. Several studies have shown that shape of the traction-separation laws are triangular in brittle adhesives, trapezoidal in ductile adhesives and exponential shapes are also fitted with different properties of adhesives as presented by Campilho et al. (2015). Some other researcher investigated the effect of adhesive layer thickness on the fracture behaviour of the adhesive joint. A recent study Carlberger et al. (2010) successfully investigates the effect of adhesive thickness on the tractionseparation law that indicates the fracture energy is more sensitive to the adhesive layer thickness than cohesive strength. When the failure is cohesive, the CZM parameters arises due to the plastic dissipation in the adhesive layers. Several studies also explore the effect of strain rate on the CZM parameters as described in Desai et al. (2015). The determination of the cohesive parameters and selection of CZM law also depends on several other parameters such as the adhesive material, moisture and temperature as deailed described in Budhe et al. (2017). Though practical joints are loaded under mixed-mode conditions, there is seldom work and this represents a research gap in the literature. In this study, traction-separation laws (TSL) were extracted for composite joints made of toughened adhesive by applying the digital image correlation (DIC) technique. These TSLs were used for strength prediction of composite joints subjected to mixed-mode loading. Composite adhesive joints were made of carbon fiber/epoxy composite adherend and Araldite 2015 epoxy adhesive. Initially, a suitable surface pre-treatment was established to be lateral sanding with 220 grit sandpaper followed by acetone cleaning. Mode I and mode II fracture testing were conducted using the double cantilever beam and end notch flexural specimens, respectively. In this study, fracture energies were taken only at the crack initiation. From these fracture tests, TSLs were extracted by using a direct method based on the digital image correlation technique. These TSLs were used in a finite element (FE) model of a single lap shear (SLS) joint model in ANSYS to predict the failure strength. This FE predicted failure strength reasonably agreed with the experimentally determined failure strength of the toughened adhesive joint. Nomenclature m B P or 𝑃𝑃��� 𝐺𝐺�� , 𝐺𝐺��� GI, GII 𝑎𝑎 , 𝑎𝑎� 𝑑𝑑𝑑𝑑 𝛥𝛥 T 𝛿𝛿𝛿 𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿𝛿 𝐺𝐺�𝛿𝛿� 𝛿𝛿�
compliance calibration coefficient specimen width maximum load critical energy release rate for mode I and mode II energy release rate depends on the current state of normal and shear stress at the crack tip initial crack length effective crack length to correct for rotation of DCB arms at crack front displacement traction separation, normal separation and shear separation curve fit polynomial equation maximum separation in the adhesive layer.
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
356
3
1.1. Material Specification and characterization The material considered throughout this study is a unidirectional carbon fiber prepreg from Hindoostan Mills Ltd. was used for the making of composite laminates. Lamination involved packing the layup in a vacuum bag and heat curing using the cycle given by the manufacturer. Considering that the loading is only in one direction, the layup used for the laminate was [0]10, i.e. 10 layers in 0° orientation. The prepreg sheets were orientated and then laid on a flat aluminium plate coated with release agent. Then, a peel fabric was laid and the layup was then sealed in a vacuum bag and vacuumed up to 0.8 bar. Vacuum assisted in the removal of trapped air and applied uniform pressure over the laminate during the curing process. The whole setup was then placed in an oven and a cuing cycle of 45 minutes at 120 ℃ was followed. The cured laminate was then released from aluminium plate. An average laminate thickness of 1.7 mm was achieved. The uniaxial tensile test were performed using Instron universal testing machine (capacity 100 kN) according to the ASTM D3039 standard in the longitudinal direction of the laminate to determine tensile modulus. E11 and ν12 were determined by applying DIC technique to the strain field. The fiber volume fraction (Vf) and E22 were determined using the mechanics of material approach given in Jones (1997). Industrial grade two-part structural toughened epoxy adhesive Araldite 2015 from Huntsman were procured. The rest of the mechanical properties of unidirectional CFRP laminate and the properties of toughened epoxy adhesive were taken from the literature as shown in Table 1. Table 1. Mechanical properties of unidirectional CFRP laminate and properties of Araldite 2015. Values in bold are for composite and adhesive were taken from Campilho et al. (2013) Modulus of elasticity Poisson’s ratio Modulus of rigidity Araldite 2015 E11 = 102 GPa
ν12 = 0.167
G12 = 4.3 GPa
E22 = 4.9 GPa
ν13 = 0.167
G13 = 4.3 GPa
E33 = 4.9 GPa
ν23 = 0.38
G23 = 3.2 GPa
2. Double cantilever beam (DCB) test
E = 1.85 � 0.21 GPa ν = 0.33
G = 0.56 � 0.21 GPa
A DCB specimen was used to determine mode I fracture parameters. Composite adhesive joints were made of carbon fiber/epoxy composite adherend and toughened epoxy adhesive (Araldite 2015). Prior to bonding the composite laminates were pre-treated by sanding with 220 grit sandpaper in the direction lateral to the length of the specimen, which was followed by acetone cleaning. Adhesive thickness, 𝑡𝑡� = 0.3 mm was maintained by providing shims or razor blade in between the adherends. ASTM D5528 standard test method was followed in DCB specimen preparation as shown in Fig.1 and determining the critical energy release rate in mode I (GIC). Here, ‘a’ is the initial crack length measured from loading point to the crack tip.
Fig. 1. The geometry of DCB specimen
Fig. 2 shows the experimental setup used for testing for DCB specimens under displacement control at 0.5 mm/min using a 10 kN universal testing machine (Shimadzu, AG-Xplus Table-top type). The 𝐺𝐺�� value was calculated by modified beam theory using Eq. 1. To obtain the cohesive parameters the displacement field about the crack tip was captured using a 5MP monochromatic CCD camera (Point Grey GS3-U3-51S5M-C 2/3” Grasshopper USB 3.0) and was analysed using the DIC technique.
4
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
DCB Fixture
357
Load cell DCB
Vertical slide
Camera & Lens Support Fixture
Fig. 2. Experimental setup for DCB test
A 35 mm fixed focal length lens along with extension tubes were used to achieve an approximate field of view of 3 mm × 3 mm, and the resolution achieved was about 1.5 µm/pixel. The camera and lens were mounted on a manual vertical slide to facilitate the vertical motion of the camera and track the crack tip during loading of the specimen. 𝐺𝐺�� � �
���
��������
(1)
Fig 3. (a) Evolution of GI with separation δ for the DCB specimen (b) Experimentally obtained and bilinear approximations for mode I traction separation law for the DCB specimens at a fixed crack length.
2.1. Determination of Mode I Traction-separation Law The traction-separation law represented 𝑇𝑇 � ��𝛿𝛿� is obtained by differentiating the following equation as described in de Moura et al. (2012), �
���������𝐺𝐺 � � �� � 𝑇𝑇�𝛿𝛿� 𝑑𝑑𝑑𝑑
This leads to, traction value at any point,
(2)
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
358
𝑇𝑇�𝛿𝛿� �
5
�����
(3)
��
A direct method was implemented for the determination of TSL. In this method, the evolution of mode I strain energy release rate, GI was plotted against crack tip opening displacement for the DCB specimen at a specific crack length as shown in the Fig. 3 (a). Then a 6th degree polynomial was fit to the evolution of GI with separation. For individual crack length, a set of images were taken at the crack tip location during loading of the DCB specimen. In each image, the normal separation was determined using DIC technique. Two reference points, one at upper and one at the lower end of the adhesive layer were used to determine the separation Mode I traction-separation law for individual crack length was obtained by differentiating the 6th degree polynomial equation of GI and separation in the DCB specimen and bilinear approximations for these experimentally obtained mode I traction-separation laws are shown in Fig. 3 (b). The criteria for approximation is the area under the curve which corresponds to the critical fracture toughness GIC and equal maximum separation for a crack length. 3. End notch flexure (ENF) test An ENF specimen is widely used to determine pure mode II fracture parameter. ASTM D7905 standard test method was followed in ENF specimen preparation as shown in Fig. 4. Figure 5 shows an experimental setup used for testing for ENF specimens and determination of GIIC are given as follows: 𝐺𝐺��� �
� �� ���� ���
(4)
��
Fig. 4. Geometry of the ENF specimen
Load cell Vertical slide
ENF specimen
ENF Camera & lens Fixture Fixture for camera Fig. 5. Experimental setup for ENF test
6
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
y = 1.13E+09δ6 - 2.33E+08δ5 + 1.80E+07δ4 6.63E+05δ3 + 1.13E+04δ2 - 1.89E+00δ + 1.99E-03
3
GII, N/mm
2 1,5 1
90 Experimental
80 70 60 50 40 30 20
0,5 0
100
Shear traction, MPa
2,5
359
10 0
0,02
0,04
Shear separation (δ), mm
0
0
0,02 0,04 Shear separation (δ), mm
Fig. 6 (a) Evolution of GII as a function of separation δ during fracture test of ENF (b) Experimentally obtained mode II traction separation law and its bilinear approximation
3.1. Determination of Mode II Traction-separation Law The procedure is identical to the one followed to extract mode I TSL. Evolution of mode II energy release rate (GII) was plotted against separation across the adhesive thickness using the direct A method as shown in Fig. 6 (a). DIC technique was used to determine the shear separation (δ). Here, the specimen undergoes bending during a test and the adhesive layer no longer remains horizontal at the crack tip location. Therefore, pure horizontal displacements given by DIC cannot be considered as shear separation. Similar to DCB specimens, a 6th degree polynomial was fit to the evolution of GII with separation, which was later differentiated toBobtain the mode II TSL and bilinear approximations to the experimentally determined mode II TSL is shown in Fig. 6 (b). 4. Strength prediction of single lap shear joints An SLS specimen is used for determining failure strength under mixed-mode loading conditions. The ASTM D3163 standard test method was followed in specimen preparation as shown in Fig. 7. Figure 8 (a) shows the test setup used to test SLS speicmens under displacement control at a constant rate of 0.5 mm/min. The failure criteria for a mixed mode cohesive zone model is given by, ��
���
�
���
����
��
(5)
A 2D finite element cohesive zone model of single lap joint was create in ANSYS APDL. The cohesive parameters extracted from the mode I and mode II fracture tests were used for the cohesive elements. Two different FE models were developed. In one model, the adhesive layer was not modelled, instead zero thickness cohesive elements were used to define the fracture path. In another model, a finite adhesive thickness of 0.3 mm was modelled and fracture path was defined within the adhesive layer by using zero thickness cohesive elements. Figure 8 (b) shows that both FE models closely predict the experimentally measured load vs. displacement response, thereby validating the ability of the FE models to predict failure loads. Moreover, since the differences between both FE models are negligible, the effect of the compliant adhesive layer on the failure response are insignificant.
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
360
7
Fig. 7. Geometry of the SLS specimen
Load cell Moving jaw SLS specimen Fixed jaw
Experimental Predicted_without_adhesive Predicted_with_adhesive
(b)6000 5000
Failure Load, N
(a)
4000 3000 2000 1000 0
0
0,5
1
1,5
2
Displacement, mm
2,5
3
Fig. 8. (a) Experimental setup for SLS test (b) Comparison between predicted and experimental load-displacement curve of single lap joint
5. Conclusions This work aimed at extracting the cohesive laws of composite adhesive joints using the direct method and to assess the use of these cohesive laws in predicting failure strength of joints under mixed-mode loading. Fracture tests with DCB and ENF specimens along with DIC technique was used to extract the mode I and mode II TSLs, respectively. It was found that the experimentally determined TSLs can be closely approximated with the bi-linear TSL. These bilinear TSLs were then used in a FE model based on cohesive zone modelling to predict the failure response of SLS joints (mixed-mode loading). The FE predicted failure response of the SLS joints closely matched with the experiments on SLS joints, thereby validating the FE model and the extracted TSLs. The FE model further showed that the compliant adhesive layer has an insignificant effect on the failure response. Acknowledgements The authors acknowledge the financial support received from the Science and Engineering Research Board (SERB), Government of India under Grant No. ECR/2015/456. References Constante, C. J., Campilho, R. D.S.G., Moura, D. C., 2015. Tensile Fracture Characterization of Adhesive Joints by Standard and Optical Techniques” Engineering Fracture Mechanics 136, 292–304. Patil, O. R., Ameli, A., Datla, N. V., 2017. Predicting Environmental Degradation of Adhesive Joints Using a Cohesive Zone Finite Element Model Based on Accelerated Fracture Tests. International Journal of Adhesion and Adhesives 76, 54–60. Balzani, C., Wagner, W., Wilckens, D., Degenhardt, R., Büsing, S., Reimerdes, H.G., 2012. Adhesive Joints in Composite Laminates—A Combined Numerical/Experimental Estimate of Critical Energy Release Rates. International Journal of Adhesion and Adhesives 32, 23–38. Carvalho, U. T.F., Campilho, R. D.S.G., 2017. Validation of Pure Tensile and Shear Cohesive Laws Obtained by the Direct Method with SingleLap Joints. International Journal of Adhesion and Adhesives 77, 41–50. Campilho, R.D.S.G., Banea, M.D., Neto, J.A.B.P., da Silva, L.F.M., 2013a. Modelling Adhesive Joints with Cohesive Zone Models: Effect of the Cohesive Law Shape of the Adhesive Layer. International Journal of Adhesion and Adhesives 44, 48–56.
8
Mohd Tauheed et al. / Procedia Structural Integrity 14 (2019) 354–361 Mohd Tauheed/ Structural Integrity Procedia 00 (2018) 000–000
361
Campilho, R.D.S.G., Banea, M.D., Moura, D. C., da Silva, L.F.M., 2015. Adhesive Thickness Effects of a Ductile Adhesive by Optical Measurement Techniques. International Journal of Adhesion and Adhesives 57, 125–32. Carlberger, T., Stigh, U., 2010. Influence of Layer Thickness on Cohesive Properties of an Epoxy-Based Adhesive-an Experimental Study. Journal of Adhesion 86, 814–33. Desai, C.K., Basu, S., Parameswaran, V., 2015. Determination of Traction Separation Law for Interfacial Failure in Adhesive Joints at Different Loading Rates. Journal of Adhesion 92, 819–39. Budhe, S., Banea, M.D., de Barros, S., da Silva, L.F.M., 2017. An Updated Review of Adhesively Bonded Joints in Composite Materials. International Journal of Adhesion and Adhesives 72, 30–42. Jones, R.M., 1975. Mechanics of composite materials. CRC press. de Moura, M.F.S.F., Gonçalves, J.P.M., Magalhães, A.G., 2012. A Straightforward Method to Obtain the Cohesive Laws of Bonded Joints under Mode I Loading. International Journal of Adhesion and Adhesives 39, 54–59. Standard Test Method for Tensile Properties of Polymer Matrix Composite Materials, ASTM D3039/D3039M-14. Standard Test Method for Determination of the Mode I Interlaminar Fracture Toughness of unidirectional Fiber-Reinforced Polymer matrix composites, ASTM D5528-13. Standard Test Method for Determination of the Mode II Interlaminar Fracture Toughness of unidirectional Fiber-Reinforced Polymer matrix composites, ASTM D7905/D7905M-14. Standard Test Method for Determining Strength of Adhesively Bonded Rigid Plastic Lap-Shear Joints by Tension Loading, ASTM D3163-01 (Reapproved 2014).