Errors associated with depth-sensing microindentation tests
- PDF / 1,039,538 Bytes
- 11 Pages / 576 x 792 pts Page_size
- 54 Downloads / 273 Views
In the determination of mechanical properties by ultra-microindentation, various errors can appear. This paper analyzes various sources of errors in estimation of elastic modulus and hardness. These errors arise from uncertainties of the indenter geometry and properties, as well as measuring instrument limitations and errors, such as the minimum detectable load, compliance, and noise of the system. Other sources of errors are thermal drift, shape of the impression, and scatter of properties of the tested material. Characteristic features and the magnitude of individual kinds of errors are discussed, together with formulas and recommended methods for their reduction.
I. INTRODUCTION Mechanical properties of surface layers and thin films are often investigated by indentation methods, whereby a hard indenter is pressed into the specimen surface. With modern microindentation systems, depth of indenter penetration is usually measured as a function of load. This approach enables the determination not only of hardness but also of elastic modulus and other properties. Sensitivity of these devices is very high, of the order of 10" 9 m and 10~5 N, which allows the investigation of the mechanical properties at depths of penetration as small as several tens of nanometers (nm) and at loads of no more than several millinewtons (mN). For more information, see Refs. 1-9. In any kind of measurement, errors occur. They can be random or systematic, caused by the measuring device, method of evaluation, and the specimen. Microindentation methods are no exception. Here, there are several sources of errors. Besides trivial ones, connected with the function of load or displacement sensors, there are some errors that are not obvious at first sight, which can result in false conclusions. They affect not only the values of the quantity investigated, but also its distribution (e.g., elastic modulus versus depth of indenter penetration). The results can be misleading, especially if properties of thin films or specially treated surfaces are studied. In these cases, the apparent (observed) values of hardness (H) and elastic modulus (E) depend not only on the properties of the layer itself, but also on the properties of the substrate (or bulk material) and on the load or depth of penetration. In general, they change gradually from those of the layer (at very small depth of penetration) to those of the substrate at relatively large depths; i.e., they grow with the depth of penetration if the substrate is stiffer (or harder) than the film, and vice versa (Fig. 1). As a consequence, if specimens with films of various thickness or at different loads are tested, J. Mater. Res., Vol. 10, No. 6, Jun 1995 http://journals.cambridge.org
Downloaded: 30 Mar 2015
different values of H and E are obtained. We usually want to ascertain the true properties of the film, so we have to separate the component corresponding to the influence of the substrate. This is often not simple. The response of the specimen beneath the indenter depends (besides on the relative
Data Loading...
