The use of microindentation techniques for the determination of the hardness and elastic modulus of thin films and surface layers on ceramic substrates

The need for information on the hardness and elastic modulus of thin films and surface layers has prompted investigation of microindentation techniques for determination of these properties. In this study, an ultralow load depth-sensing microindentation system was used to measure hardness and elastic modulus as a function of indenter displacement for several different thin films and surface layers on ceramic substrates. The films and layers included diamondlike carbon (DLC) films, ion implanted layers (both amorphous and crystalline), and metallic films. All of the films tested were less than 200 nm thick, and all of the indentations were made at depths of 100 nm and less. The main goal of this research was to determine whether accurate and reproducible hardness and modulus values could be obtained by this technique.

In testing the diamondlike carbon films, it was found that the ULL system is sensitive to substrate effects and can distinguish between adherent and nonadherent films. The hardness of the DLC coatings tested in this study ranged from 5-10 GPa. The elastic modulus ranged from 36-108 GPa. These values were obtained by linear extrapolation of the mechanical property data, obtained as a function of indenter displacement, to zero displacement. Values obtained at depths less than 20 nm were constant, so linear extrapolation of the data appears to be a reasonable method of obtaining the hardness of DLC films.

Tests on the unimplanted layers revealed a decrease in the relative hardness ratio (implanted to unimplanted) for both the amorphous and the damaged but crystalline layer. Because these results do not agree with published studies, several possible causes for the discrepancies were proposed. Further work is needed to determine the actual source of the differences. Plots of elastic modulus as a function of indenter displacement are not linear for either the amporphous or the damaged but crystalline layer; the values obtained decrease with indentation depth. Additional work is necessary to determine the best method for extrapolating the data to obtain an accurate modulus value for these layers.

Modulus values for nickel and gold films obtained by the ULL method agreed well with those obtained by the dynamic resonance method. However, like the modulus data for the implanted sapphire samples, the hardness versus indenter displacement data showed considerable curvature at shallow depths. Further work must be done to determine how best to treat the data in order to obtain a reliable hardness value,