Experimental assessment of Fully-Uncoupled Multi-Directional specimens for mode I delamination tests
Graphical abstract
Introduction
In the field of structural design, composite materials represent an appealing alternative to metals. Sectors like the aerospace and the automotive have adopted more and more these materials, due to their exceptional specific properties, that allow the design of efficient lightweight structures. In particular, long-fibres composite laminates are widely adopted in structural design, even for safety-critical components. Therefore, understanding the damage processes of laminates is of paramount importance.
Among others, delamination is one of the most critical damage modes, as it may significantly reduce mechanical properties of the structure, while being very difficult to detect. Hence, the design of safe structures demands a thorough knowledge of the delamination resistance of materials [1], typically in terms of critical strain Energy Release Rate (ERR). For these reasons, delamination has been long studied [2], and standards for evaluation of interlaminar properties of composite laminates have been developed, in particular for mode I [3], mode II [4] and mixed mode I/II [5]. For mode III, no standard exists yet. One of the most promising solutions proposed so far is the Edge-Crack Torsion (ECT) test [6]. However, research is still ongoing [7] and mode III toughness values obtained with this test may not be accurate and reliable [8].
Despite the fact that existing standards are recommended for non-woven UD materials only (due to the limited experience gained in round robin tests), extensive experience has been collected on their use, both for UD tape and woven materials [9]. However, this concerned mostly laminates with UD stacking sequences with the main reinforcement direction parallel to the specimen longitudinal direction (). As a consequence, only the interface between equally oriented plies at (UD interface hereafter) is usually characterised. Instead, applicability of standards to laminates with multidirectional (MD) stacking sequences is still questionable. The reason is that UD laminates have an uncoupled thermoelastic behaviour allowing optimal test conditions and accurate data reduction techniques. MD laminates, on the contrary, have a more complex behaviour that lead to a certain number of problems that prevent the attainment of consistent and reliable results [10]. Consequently, delamination testing of non-UD interfaces (MD interfaces hereafter) is a complex task. Real structures, however, are built using MD laminates, and delamination may appear at any interface.
One first issue during tests on MD laminates is the appearance of additional energy dissipation mechanisms. In standard tests involving bending of the specimen, off-axis plies may experience matrix plasticity and intralaminar damage [11]. Intralaminar damage often leads to the delamination jump (or migration) [9,[12], [13], [14], [15], [16], [17]]. Strategies to model such phenomenon are currently being investigated [18,19].
Another relevant problem with MD laminates is the presence of thermal residual stresses [20]. It was shown that such stresses may promote matrix damage and thus facilitate delamination jump [21]. Moreover, they may greatly affect the evaluation of interlaminar fracture toughness, as demonstrated by Nairn [[22], [23], [24]] and Yokozeki et al. [25]. The importance of thermal residual stresses depends on the laminate layup [23].
A third inconvenient is the difficulty in controlling the loading mode. When pure mode tests are performed, no other ERR contributions should exist. When performing mixed-mode tests, knowledge of the exact mode-mix is required, in order to meaningfully reduce and exploit experimental data. While standard delamination test procedures [[3], [4], [5]] are devised to address these issues for UD specimens, it is not guaranteed that the same results are achieved when using MD specimens, due to the presence of elastic couplings that modify the kinematics of the specimen and that may induce unwanted rotations and parasite modes contributions to the ERR. In such cases ERR modal contributions could be evaluated numerically [26], but not without difficulties [27].
Besides modal partition, another relevant issue concerning ERR is its distribution along the width of the delamination specimen. Common data reduction procedures for delamination tests are based on 2D theories: a straight delamination front and a uniform ERR distribution are assumed, even though this has long been proven to be an idealisation [28]. Nowadays, tools to predict delamination growth direction have been developed, following geometrical considerations [29], or within the framework of Cohesive Zone Model (CZM) [30].
It was shown that in MD laminates 3D effects can become important and affect the shape of the front and of ERR distribution [[31], [32], [33], [34]]. In particular, the parameter [31,32]:was found to be related to curvature of ERR distribution and delamination front. Terms in Eq. (1) are the components of the laminate stiffness matrix, as obtained by Classical Laminated Plate Theory (CLPT). On the other hand, the asymmetry of ERR distribution and of delamination front was found to be somewhat related to the parameter [33,34]:
In order to obtain reliable evaluations of critical ERR using common data reduction techniques, both and should be kept as small as possible [35]. According to [36], the value of should be much lower than 1, while in [37] it was suggested that it be lower than 0.25 (referring however to End Notched Flexure specimens).
To sum up, delamination testing of MD interfaces requires an extremely careful design of the stacking sequence for the specimens. As observed in [38,39], the ideal stacking sequence should result in specimens that avoid as much as possible additional damage mechanism, that eliminate (or at least reduce) mechanical couplings, and that are not affected by thermal residual stresses. One approach suggested is to use stacking sequences containing as much -oriented plies as possible [12,33]. This approach reduces, but does not eliminate, couplings and thermal stresses. Other authors used FE simulations to try and find suitable stacking sequences [16,21]. Eventually some authors adopted an innovative approach by using quasi-trivial (QT) solutions [40] to obtain uncoupled specimens [[41], [42], [43]]. While only specific sequences were developed (thus with limitations on possible interfaces and on the capability to investigate other aspects), this nonetheless allowed to build specimens with interesting uncoupling properties.
This same approach was pushed further in a previous work [44], where the authors presented a method based on QT solutions and special superposition rules [45,46] to conceptually design novel Fully-Uncoupled Multi-Directional (FUMD) stacking sequences. This design approach aims at generating layups for delamination specimens that are not affected by thermal residual stresses, do not have any elastic coupling, and allow to test any desired delamination interface. Such results were obtained in closed-form solution within the framework of CLPT and verified by means of Finite Element analyses in [44]. This preliminary assessment suggested that FUMD specimens could represent an extremely appealing solution for delamination testing of MD interfaces.
To confirm such approach, this paper shows the results of the very first experimental campaign carried out using FUMD sequences and taking into account different delamination interfaces and different global stiffness properties of the specimens. The main goal of such activity was to experimentally assess the capability of FUMD sequences to deliver uncoupled behaviour, to produce the expected delamination fronts, to avoid undesired rotations and to allow an easy evaluation of the critical ERR with simple procedures proposed by standards.
The paper is organised as follows. Firstly, Section 2 recalls the fundamentals of FUMD stacking sequences and explains in detail how they were used to design the delamination specimens conceived for this study. Then, the material system and all the experimental procedures adopted, from specimen fabrication to data reduction, are described. Section 4 presents and discusses the results of the study. Eventually Section 5 ends the paper with conclusions and perspectives.
Section snippets
Delamination specimens design
The theoretical principles behind the design of FUMD stacking sequences for delamination specimens have been presented for the first time in a previous work from the authors [44]. Hence, only those notions fundamental for the understanding of the present paper will be reported here. On the other hand, some more details on how to practically obtain a FUMD specimen will be given.
Material system adopted
The material system used for this study is a glass-epoxy UD-fabric, with 90% of fibre weight in the longitudinal direction (0, warp) and the remaining 10% in the transverse direction (90, weft). Its commercial reference is: HexPly© M34N/32%/430PUD/G-136x5 and it was available in the form of a prepreg. Since FUMD specimens do not address directly the problem of delamination jump, this material was used to mitigate the problem. Indeed, Ozdil et al. [[54], [55], [56]] used a similar material to
Force-displacement behaviour
The force-displacement plots of all tests are shown in Fig. 4. They include both loading phases and report all initiation points. The curves show a good consistency, especially for the first loading phase. The small scatter observed during the first unloading and the second loading phases is caused by the different delamination length, for different specimens, obtained at the end of the first loading phase, as explained in Subsection 3.4.2.The NL point is the first observed for both loading
Summary, conclusions and perspectives
In this paper, an experimental study to assess the suitability of FUMD specimens for delamination tests was presented. Five different types of FUMD specimens, having different delamination interfaces, were designed. A glass/epoxy UD-fabric material was used to reduce the likelihood of delamination jump. The specimens were tested under pure mode I delamination, according to standard procedures. For comparison purposes, also standard UD specimens of the same material were tested. Four out of five
Data availability
The raw and processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.
Author statement
T. Garulli: Conceptualization, Methodology, Software, Investigation, Writing - Original Draft, Writing - Review & Editing.
A. Catapano: Conceptualization, Resources, Writing - Review & Editing, Supervision, Project administration, Funding acquisition.
D. Fanteria: Conceptualization, Investigation, Resources, Writing - Review & Editing, Supervision, Project administration, Funding acquisition.
W. Huang: Investigation.
J. Jumel: Resources, Supervision.
E. Martin: Writing - Review & Editing,
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The first author is grateful to Nouvelle Aquitaine Région Founding for supporting this research through the AAP Recherche 2016 grant to the SMARTCOMPOSITE project. Colleagues F. Ricci, E. Monaco and N. D. Boffa from University of Naples Federico II are kindly acknowledged for performing the ultrasonic C-scan of the specimens.
References (73)
Interlaminar fracture toughness: the long and winding road to standardization
Compos. B Eng.
(1998)- et al.
Reexamination of the edge crack torsion test for determining the mode III delamination toughness of laminated composites
Eng. Fract. Mech.
(2019) - et al.
Dependence of fracture toughness of composite laminates on interface ply orientations and delamination growth direction
Compos. Sci. Technol.
(2004) - et al.
Mode I delamination of angle-ply epoxy/glass-fibre laminates exhibiting permanent deformation during fracture
Compos. Sci. Technol.
(1994) - et al.
Initiation and bifurcation mechanisms of cracks in multi-directional laminates
Compos. Sci. Technol.
(2000) - et al.
A numerical and experimental investigation of delamination behaviour in the DCB specimen
Compos. Sci. Technol.
(2000) - et al.
Mode I interlaminar fracture of carbon/epoxy cross-ply composites
Compos. Sci. Technol.
(2002) - et al.
Characterization of crack propagation in mode I delamination of multidirectional CFRP laminates
Compos. Sci. Technol.
(2012) - et al.
Influence of ply-angle on fracture in antisymmetric interfaces of CFRP laminates
Compos. Struct.
(2019) - et al.
The relationship between mixed-mode II/III delamination and delamination migration in composite laminates
Compos. Sci. Technol.
(2014)
Modelling delamination migration in angle-ply laminates
Compos. Sci. Technol.
The formation of residual stresses in laminated thermoplastic composites
Compos. Sci. Technol.
Numerical investigation to prevent crack jumping in double cantilever beam tests of multidirectional composite laminates
Compos. Sci. Technol.
Energy release rate analysis for adhesive and laminate double cantilever beam specimens emphasizing the effect of residual stresses
Int. J. Adhesion Adhes.
Correction method for evaluation of interfacial fracture toughness of DCB, ENF and MMB specimens with residual thermal stresses
Compos. Sci. Technol.
Prediction of delamination front's advancement direction in the CFRP laminates with mechanical couplings subjected to different fracture toughness tests
Compos. Struct.
Point-wise evaluation of the growth driving direction for arbitrarily shaped delamination fronts using cohesive elements
Eur. J. Mech. - A/Solids 72 (Int. J. Numer. Meth. Eng. 50 2001)
Delamination characteristics of double-cantilever beam and end-notched flexure composite specimens
Compos. Sci. Technol.
Effect of stacking sequence on energy release rate distributions in multidirectional dcb and enf specimens
Eng. Fract. Mech.
Numerical analysis of the DCB test configuration applicability to mechanically coupled Fiber Reinforced Laminated Composite beams
Compos. Struct.
A non-classical, energy release rate based approach for predicting delamination growth in graphite reinforced laminated polymeric composites
Compos. Sci. Technol.
A special class of uncoupled and quasi-homogeneous laminates
Compos. Sci. Technol.
On the determination of delamination toughness by using multidirectional DCB specimens
Polym. Test.
Effects of fiber orientation of adjacent plies on the mode I crack propagation in a carbon-epoxy laminates
Procedia Eng.
Quasi-trivial stacking sequences for the design of thick laminates
Compos. Struct.
Analysis of crack propagation in double cantilever beam tests of multidirectional laminates
Mech. Mater.
Investigations of delamination criticality and the transferability of growth criteria
Compos. Struct.
Delamination of multidirectional composite laminates at 0/θ ply interfaces
Eng. Fract. Mech.
Beam analysis of angle-ply laminate end-notched flexure specimens
Compos. Sci. Technol.
Beam analysis of angle-ply laminate mixed-mode bending specimens
Compos. Sci. Technol.
Beam analysis of angle-ply laminate DCB specimens
Compos. Sci. Technol.
Experimental characterization of the interlaminar fracture toughness of a woven and a unidirectional carbon/epoxy composite
Compos. Sci. Technol.
Effect of interfacial properties and weave structure on mode I interlaminar fracture behaviour of glass satin woven fabric composites
Compos. Appl. Sci. Manuf.
The effect of weave pattern and crack propagation direction on mode I delamination resistance of woven glass and carbon composites
Compos. B Eng.
The effect of weave pattern on the mode-I interlaminar fracture energy of e-glass/vinyl ester composites
Compos. Sci. Technol.
Characterization of interlaminar fracture behaviour of woven fabric reinforced polymeric composites
Composites
Cited by (11)
Mode decoupling in interlaminar fracture toughness tests on bimaterial specimens
2023, Engineering Fracture MechanicsDetermining cohesive parameters in an n-segment constitutive law of interfaces through DCB tests
2023, Engineering Fracture MechanicsFracture toughness resistance curves for a delamination in CFRP MD laminate composites, Part I: Nearly mode II deformation
2022, Theoretical and Applied Fracture MechanicsA consistent energy-based cohesive zone model to simulate delamination between differently oriented plies
2022, Composite StructuresCitation Excerpt :The goal of this FE model is to study the propagation of delamination in FUMD specimens. Moreover, simulated force–displacement curves and numerical delamination front geometries will be compared to their experimental counterparts taken from [20]. The CZM not only allows simulating the damage initiation, but also tracking the evolution from a straight crack front to a curved one.