Using DIC technique to characterize the mode II interface fracture of layered system composed of multiple materials

Abstract

This article introduces a new method developed for characterizing the mode-II dominant interfacial fracture in a layered system composed of materials with significantly different mechanical properties. The proposed method consists of 3 parts. First, using digital image correlation (DIC) technique to acquire the full field strain in layered composites and then calculating the variation of elastic strain energy in multiple loading steps. Second, using DIC to acquire the relative displacement between layers and then calculating the possible strain energy release rates in multiple loading steps based on a presumptive bi-linear cohesive zone model (CZM) with unknown parameters. Third, developing an algorithm to find the optimal solutions for the presumptive bi-linear CZM by equilibrating the elastic strain energy and fracture energy with the minimum error in multiple loading steps. For verification, the proposed method was applied to characterize the mode II dominated interface fracture of a bi-material and bi-layer system under tension. The feasibility and accuracy of the proposed method are experimentally demonstrated.

Introduction

Interfacial delamination has always been a concern in civil engineering [1], [2]. Delamination may induce significant stiffness reduction in laminated composite structures, e.g. RC structural member strengthened with externally bonded fiber reinforced polymer (FRP) [3], [4]. It may also result in loss of the nonstructural protective layer, e.g. fire insulation for steel structure [5], [6]. Both of the above-mentioned conditions might lead to premature failure of the structural system.

In the past several decades, interfacial delamination has been extensively studied by researchers and engineers [7], [8]. It is widely accepted that interfacial delamination is governed by the fracture toughness of the interface [9]. Very often, delamination in composite structures is caused by a mode II (sliding mode) fracture as in the case of delamination between a beam member and externally bonded FRP strengthening layer. Theoretically, such delamination can be predicted by comparing the mode II strain energy release rate (SERR) G_II to its critical value G_IIc (governed by the property of the interface). According to linear fracture mechanics, when G_II exceeds G_IIc, mode II fracture will occur at interface. Therefore, proper characterization of the G_IIc is then critical for obtaining accurate delamination predictions.

A variety of experimental approaches have been proposed in the literature for characterization of the mode II interface fracture, including end-loaded split (ELS) test [10], center-notched flexure (CNF) test [11], end-notched flexure (ENF) test [12], three-point end-notched flexure (3ENF) [13], [14], [15], [16], [17] (Fig. 1(a)) and four-point end-notched flexure (4ENF) [18] (Fig. 1(b)), etc. However, with these traditional methods, it is difficult to track the whole development process of G_IIc with sufficient accuracy, rather, only the critical SERR or a few discrete data points on the crack-resistance curve can be obtained. For example, during a 3ENF test, the critical load is applied at the center to propagate the crack. It is very difficult to control the unstable crack propagation and therefore difficult to accurately measure the crack length. Additionally, SERR tends to drop sharply following the critical point in 3ENF test, and only the critical SERR value can be obtained. The presence of friction and shear force at crack tip also affects the accuracy of the measurement.

Moreover, the above-mentioned approaches may not be applicable to the fracture characterization of a multi-material system composed of materials that have significantly different elastic moduli and strengths. The reason is twofold. Firstly, it can be very difficult to determine the critical load in the delamination process since the load drop at crack propagation may not be notable due to the large difference between the moduli of the two layers. Secondly, almost all the existing approaches are based on linear elastic fracture mechanics which is only valid when all the layers remain elastic before and after delamination. But in reality, materials of a layered system may experience cracking or plastic deformation during loading, which will inevitably lead to additional energy dissipation and invalidate the testing result. However, it is difficult to determine whether the material is in elasticity in experiment.

In the current study, a more reliable method to characterize the mode II interface fracture for a layered composite system, of which the materials have significant different elastic moduli and strengths, is proposed. A non-contact measurement technique, digital image correlation (DIC), is engaged to acquire the strain and displacement of materials during entire loading process, which can be used to more accurately calculate the energy dissipation during the delamination process and to determine whether the materials have gone beyond elastic stage. Therefore. The proposed method is expected to overcome the aforementioned drawbacks in the conventional approaches. For demonstration purpose, characterization of the mode II fracture of a bi-material and bi-layer system composed of a structural steel substrate and a newly developed cement-based fireproofing material was conducted using the proposed approach and the result is presented in the current paper.