ICMCTF2002 Session E4/F1-1: Mechanical Characterization. Micromechanical Testing and Modelling

Monday, April 22, 2002 10:50 AM in Room California

Monday Morning

Start	Invited?	Item
10:50 AM		E4/F1-1-2 A Detailed Study of SiC-on-Si Coatings. Analysis of their Mechanical Response by Nanoindentation Techniques R.E. G-Berasategui, S.J. Bull, T.F. Page (MMME, University of Newcastle, United Kingdom) As part of our aim of determining how thin a coating can we study by nanoindentation techniques and how these techniques can optimally be used to determine a wide-range of mechanical property data and mechanical response mechanisms of coated systems, we have studied a range of epitaxial 3C-SiC coatings on (100) and (111) Si. Besides calculating the variation of E and H with indenter displacement in these systems, detailed analysis has been made of load-displacement (P-δ) curves to identify the detailed effects of load support by the coating. Thus, the effect of coating thickness and deformation scale on the maximum displacement achieved (compared to the substrate alone), the amount of elastic recovery observed (compared to the substrate alone) and the critical load at which the reverse undensification transition (pop-out) in silicon occurs. In addition, analysis of P vs. δ² and P vs. S² have been made in order to look at transitions in deformation responses between the coating and substrate with increasing scale, and also to explore the detailed effects of the variation of both E and H with displacement. We have also used the new stiffness ratio technique to compare E and H data with that obtained by the usual Oliver & Pharr technique. A Triboindenter Hysitron has been used to explore responses at very low loads. High-resolution SEM techniques, together with SEM-SACP techniques have been used to explore the deformation of fracture modes observed and to give further insight into the nanoindentation observations. Overall, our observations provide a very detailed understanding of this particular coated system, including the critical spatial scales for the coating to strongly influence the mechanical responses of the system.
11:10 AM		E4/F1-1-3 Determining Strain Hardening Constitutive Models from Conical Indentation: A Sensitivity Analysis T.W. Capehart, Y.T. Cheng (GM Research And Development Center) Several procedures have previously been advanced for extracting constitutive relations from the load-displacement curves obtained from indentation experiments. This extraction required inverting the load-displacement curve to obtain the presumably unique parameters of the relevant constitutive model. This work addresses the specific problem of determining the three parameters, elastic modulus E, yield stress Y, and hardening exponent n, that define the isotropic strain hardening model, from a single load-displacement curve with a sharp conical tip. The sensitivity of the inversion process is tested through a series of finite element calculations using Abaqus. Different magnitudes of normally distributed noise are superimposed on a calculated load-displacement curve to simulate hypothetical data sets for specific values of E, Y and n. Load-displacement curves are then calculated to construct a grid of 64 points centered on these values in the 3-D parameter space. This grid is used both as a basis for constructing a cubic interpolation function and the local χ² curvature matrix. The sensitivity of the model parameter confidence regions to the magnitude of the normally distributed noise are estimated using two methods: statistical Monte Carlo simulations using a Marquardt-Levenberg algorithm to minimize χ², computed from the curvature matrix assuming a parabolic χ² surface. Both approaches reveal a strong correlation between the Y and n in the region of parameter space where Y/E~0.01 and n<0.25 that produces significant parameter bias. For realistic noise levels this parameter bias makes accurate determination of the strain hardening model parameters problematic when the determination is based on a single load-displacement curve.
11:30 AM		E4/F1-1-4 Numerical and Experimental Investigation of Sharp and Spherical Indentation of Cr-DLC Films A.W. Stewart, G.G.W. Mustoe, J.J. Moore (Colorado School of Mines) Metal-containing diamond-like carbon (Me-DLC) films are often used as tribological coatings on steels or hard metals. This paper will investigate the deformations and stresses generated during indentation of Cr-DLC films on a hard steel substrate with sharp and spherical indenter geometries. Emphasis will be on the stresses that cause failure to occur in the film/substrate system. Experimental micro-indentation data will be used to verify the onset of failure within the film/substrate system as predicted by non-linear finite element models. The finite element modeling will assess the effects of different film/substrate architectures, initial stress fields, and material strength and toughness parameters.
11:50 AM		E4/F1-1-5 Numerical and Experimental Analyses of the Indentation of Coated Systems with Substrates with Different Mechanical Properties L.A. Piana (Universidade Federal do Rio Grande do Sul, Brazil); R.M. Souza (Universidade de São Paulo, Brazil); A.O. Kunrath, T.R. Strohaecker (Universidade Federal do Rio Grande do Sul, BRAZIL) In this work, the indentation of coated systems was studied based on the effect of the mechanical properties of the substrate on the film fracture behavior. Both experimental and finite element modeling analyses were conducted to study the phenomena that result when an indenter with Rockwell C geometry applied a normal load of 1500 N on systems with TiN films deposited onto three different steel substrates: AISI 4140 with hardness 240 HV, AISI M2 with hardness 260 HV and AISI M2 with hardness 850 HV. Experimental results indicated the formation of circular cracks close to the contact edge of the indentations on specimens with 4140 substrates, in opposition to radial cracks on the other specimens, independently of the M2 substrate hardness. Both finite element and experimental (including profilometry) analyses were able to correlate these results with the amount of substrate pile-up close to the indentation edge, which affected the contact stress distribution during indentation.