MICRO-INDENTATION TESTING OF POLYMER MATRIX COMPOSITES


See the Applet related to this research work



Collaborators

This project has been carried out within the frame of a collaboration between the Laboratory of Tribology and Systems Dynamics (L.T.D.S) and the Surface Engineering Laboratory (IFoS) of Ecole Centrale de Lyon.

LTDS : L. Carpentier, Ph. Kapsa, F. Sidoroff
             M. Zidi (PhD student), N. Mokni (Graduate Student)
IFoS :   A. Chateauminois,
             M. Kharrat (PhD student)

This research work has been supported by ONERA (Chatillon, France), IFREMER (Brest, France), IFP (Rueil, France) and CEA (Le Ripault, France)



Scope

The fibre matrix interface plays an essential part in the mechanical behaviour, fatigue life and durability of fibre reinforced composite materials and many works have been devoted to its mechanical characterization. Among the micro-mechanical tests allowing a direct assessment of the interface shear strength, the micro-indentation test can provide an efficient solution. Its basic principle is very simple: starting from a polished surface perpendicular to the fibbers, a single fibre is selected and its cross-section is indented by a sharp indenter. Due to the induced stress concentration at the interface, debonding between the fibre and the matrix may occur from the fibre end. The micro-indentation test is now indeed well mastered and widely used in many situations and various domains of material science. In this particular case, it has many advantages, the greatest of which being probably the fact that it allows the characterization of the actual interface in a real composite material.
The basic result from this test is the indentation curve giving the normal load as a function of the penetration depth. Its interpretation, however, is not straightforward and a careful mechanical analysis is required to obtain intrinsic material properties from the global response of a complicated system involving the whole fibre environment (interface, matrix, neighbouring fibbers) under a highly heterogeneous stress state. Some simplifying assumptions have to be made in order to obtain tractable results. Most of the existing analysis of experimental data are based on some kind of a shear lag argument, which was initially derived by Cox with very crude approximations regarding the stress distribution. Of course, this approximation may be justified to some extent by the overall consistency of the obtained results, but some other, more precise justification would be welcome.
The purpose of the present study was to develop such a quantitative procedure for the analysis of the micro-indentation curve, but also to assess the adequacy of its underlying assumptions from some finite element computations carried out on well chosen model problems. An analytical shear lag model taking into account the fibre environment and the debonding processes has been developed. It was applied to the analysis of the micro-indentation behaviour of Glass/thermoset composites differing by the matrix and the fibre sizing.



Main Results

The data reduction scheme is based on the assumption that the measured displacement (ut) is the sum of two independent contributions:

  •  the displacement u ep due to the local elasto-plastic indentation of the fibre surface by the Vickers' indenter. It was proposed to assess this component from the indentation behaviour of a bulk homogeneous specimen having the same composition as the fibre,

  • The extraction of this latter component from the indentation curve provides the reduced indentation curve relating an equivalent stress (so =P/ pa2) to the displacement, uo, of the fibre surface . This reduced indentation curve represents the useful information regarding the interfacial behaviour.
     

      Extraction of a reduced indentation curve

    An analytical shear lag model has been developed to describe the interfacial behaviour and to assess the interfacial shear strength from the load vs uo relationship. The model was developed in order to simulate both the loading and unloading steps of the indentation process.  By means of various interface laws, it takes into account the occurrence of debonding and sliding at the fibre/matrix interface. F.E.M. simulations using appropriate model configurations have been used to validate the main assumption of this approach, namely the uniformity of the axial deformation in the fibre cross-section. This hypothesis is essential to be able to consider the problem as one-dimensional.
     

    Basic hypothesis of the micro-indentation model

    The application of the basic version of the model requires the identification of two parameters, namely the debonding load Pd and an elastic 'stiffness' parameter, n. The latter coefficient is measured from the slope of the initial elastic part of the experimental loading curve. It provides some information regarding the local environment of the tested fibre. The debonding load is identified by an identification procedure using the entire loading curve, which is more secure than the identification of an always dubious threshold point corresponding to the onset of debonding.
     

    Application of the model to an experimental indentation curve using a glass/epoxy composite



    Publications Related to the Project

    M. Kharrat, L. Carpentier, A. Chateauminois, P. Kapsa
    " Evaluation of the fibre/matrix interfacial strength of a glass fibre reinforced polymer composite using a microindentation test."
    Composites  Part A, 28, N°1 (1997) 39-46

    M. Zidi, L. Carpentier, A.Chateauminois, F.Sidoroff
    “Quantitative analysis of the micro-indentation behaviour of fibre reinforced composites:Development and validation of an analytical model”
    Composite Science and Technology 60 (2000) 429-437
    Load in PDF format - 829 Ko- (with permission from Elsevier Science)

    M. Zidi, L. Carpentier, A. Chateauminois, Ph. Kapsa, F. Sidoroff
    "Analysis of micro-indentation tests by means of an analytical model taking into account different interfacial responses."
    Composites Science and Technology 61 (2001) 369-375
    Load in PDF format - 285 Ko - (with permission from Elsvevier Science)


    | Back to top of the page | Home |
    Antoine Chateauminois - March 2001