Strain-rate and off-axis loading effects on the fibre compression strength of CFRP laminates: Experiments and constitutive modelling

https://doi.org/10.1016/j.compscitech.2020.108210Get rights and content

Abstract

A series of dynamic longitudinal compression tests have been performed on cross-ply IM7/8552 specimens cut at different off-axis angles to produce different combinations of compression and shear stresses. Together with results from previous quasi-static tests of the same kind, quasi-static and dynamic fibre kinking failure envelopes have been obtained using classical laminate theory. This new experimental data has been compared against predictions from the leading fibre kinking theories, made rate-dependent by using rate-dependent in-plane shear properties, and show that, while they can accurately predict the effects of strain rate on the uniaxial compression strength, they are unable to capture the effects of shear, neither at quasi-static nor dynamic rates. Instead, a simpler more phenomenological approach is proposed to predict the rate-dependent fibre kinking strength of FRP laminates under multi-axial loads until the micromechanics can be more accurately described.

Introduction

Longitudinal compressive failure in carbon fibre-reinforced polymer (CFRP) laminates, or fibre kinking, is arguably the most complex failure mechanism observed in these materials. There has been much work done over the years to understand the micro-scale mechanisms behind this type of failure and, while it is now widely recognised as a form of buckling instability in the fibres driven by the response of the supporting matrix, there are still specific aspects of the micromechanics that remain to be explained. Leading fibre kinking theories, such as those by Budiansky and Fleck [1,2] and Pinho [3,4], can accurately predict failure under uniaxial compression given the initial misalignment or waviness of the fibres and the nonlinear shear response of the laminate. However, under multi-axial loading, different models start to give diverging predictions based on how they address the evolution of fibre misalignment under transverse stresses [5]. In addition, even some of the aspects where all models agree have yet to be fully verified experimentally, such as the effects of strain-rate or temperature.

Matsuo and Kageyama [6] recently showed that the effects of temperature on the uniaxial compression strength of a thermoplastic CFRP composite were in good agreement with the predictions from Fleck's fibre kinking theory, where the change in strength was directly related to the change in the nonlinear shear response of the laminate. However, due to the complexity of the experimental procedures, very little work has been done to characterise the longitudinal compression behaviour of CFRPs under off-axis and dynamic loads, which is crucial for a more extensive validation of these fibre kinking models, especially for impact loading applications where the material is subjected to multi-axial stresses and high strain-rates.

Based on recent work by the authors [5], who proposed an improved testing methodology to measure the uniaxial and off-axis compressive strength of CFRP laminates using cross-ply (CP) specimens, and previous work by Ploeckl et al. [7], who showed a roughly 40% increase in the uniaxial compression strength of a quasi-isotropic (QI) laminate from quasi-static to high strain-rate testing, the work presented here aims to investigate the effects of strain-rate on the fibre kinking failure envelope by testing the same IM7/8552 CFRP from Refs. [5] under high-rate off-axis compression.

Finally, the obtained experimental data was used to evaluate the accuracy of different fibre kinking theories under quasi-static and dynamic multi-axial loading.

Section snippets

Methodology and experimental set-up

Following the quasi-static study in Refs. [5], a series of dynamic off-axis compression tests have been performed using a similar cross-ply specimen configuration and data reduction methods to extract the longitudinal compression strength from the axial measurements.

The specimens were tested at high strain-rates using a Split-Hopkinson Pressure Bar (SHPB) system, described in section 2.2, and longitudinal ply failure data was extracted from the specimen strengths using Classical Lamination

Results

With the updated dynamic specimen design, the stress concentration and boundary effect issues observed in previous iterations were successfully avoided. The results, summarized in Fig. 4 and Table 1 below, showed good repeatability and the stiffnesses were in good agreement with the quasi-static results from Ref. [5], giving confidence in the data acquisition and processing methods. Fig. 4 shows the obtained axial stress-strain curves for each of the four specimen orientations compared to the

Analysis and discussion

With these new experimental results, a comprehensive set of fibre kinking failure data for one same material system at different loading rates and biaxial stress states was compiled, which has been used to evaluate some of the leading failure theories and criteria. In particular the analytical model by Budiansky and Fleck [1,2] and the numerical 3-dimensional implementation by Pinho for the NASA LaRC criteria [3,4,19] have been compared.

Conclusions

A series of high strain-rate compression tests have been performed on uniaxial and off-axis cross-ply IM7/8552 specimens. Along with previously reported quasi-static experiments on the same material [5], a comprehensive data set on the effects of shear and strain-rate on fibre kinking strength has been compiled. This new data has been used to evaluate leading fibre kinking theories [[1], [2], [3], [4],19], showing that, when coupled with rate-dependent nonlinear in-plane shear properties, they

CRediT authorship contribution statement

Daniel Thomson: Conceptualization, Methodology, Formal analysis, Software, Validation, Visualization, Writing - original draft, Writing - review & editing. Gustavo Quino: Methodology, Investigation. Hao Cui: Conceptualization, Methodology, Software, Validation. Antonio Pellegrino: Methodology, Investigation. Borja Erice: Conceptualization, Methodology, Software, Validation, Visualization, Supervision. Nik Petrinic: Project administration, Funding acquisition, Resources, Supervision.

Declaration of competing interest

The authors would like to acknowledge Rolls-Royce plc, for their continuing support through the Solid Mechanics University Technology Centre at the University of Oxford.

Acknowledgements

The authors would like to acknowledge Rolls-Royce plc, for their continuing support through the Solid Mechanics University Technology Centre at the University of Oxford.

References (34)

Cited by (21)

  • Modelling the Non-linear fibre and matrix dominated compressive behaviour in unidirectional Fibre-reinforced composites

    2023, Composite Structures
    Citation Excerpt :

    Fibre-dominated. Quasi-static and high rate Split Hopkinson Pressure Bar (SHPB) compression tests of 8 mm gauge length dogbone specimens that were cut and ground from a 10 mm thick [0/90]20S cross-ply (CP) laminate parallel to the external plies and with small off-axis angles of 3°, 6° and 10° [23]. Matrix-dominated.

View all citing articles on Scopus
View full text