An investigation of hardness and adhesion of sputter-deposited aluminum on silicon by utilizing a continuous indentation
- PDF / 721,193 Bytes
- 7 Pages / 593.28 x 841.68 pts Page_size
- 98 Downloads / 142 Views
P. AlexopoulosandT.-W. Wu IBMAlmaden Research Center, San Jose, California 95192
Che-Yu Li Cornell University, Ithaca, New York 14853 (Received 27 March 1987; accepted21 September 1987) The hardness of aluminum films on silicon are measured as functions of depth of the indenter. The films have thicknesses of 0.25,0.5, and 1.0/zm. The adhesion between one film and the substrate has been reduced through the prior deposition of a 10 nm layer of carbon. In each case the hardness is found to increase as the indenter approaches the film-substrate interface, but the rate of increase is greater for a film with good adhesion than for one with poor adhesion. It is suggested that this increase results from the constraint on deformation of the film by the substrate. A physical model is proposed whereby the yield stress of the film, a0, and an average effective shear strength r of the indenter-film and film-substrate interfaces, may be determined from the data.
I. INTRODUCTION This article reports the measurement of the hardness of aluminum films on silicon. The measurements were made by using a test in which the load and depth of penetration were monitored continuously as a diamond indenter was driven into the surface at a constant displacement rate, then withdrawn. For the present purposes the "hardness" will be defined as the instantaneous load on the indenter divided by the projected area of the impression formed by plastic deformation. This area is calculated based on the calibrated shape of the indenter and the measured depth of penetration corrected for the elastic rebound of the material. The hardness value calculated based on depth of indentation will, in general, differ from the conventional indentation hardness, which is based on the length of the diagonal after the indenter has been removed.1>2 The difference results from the presence of pileup (or sinkin) of material around the indenter. An important purpose of this article is to demonstrate how the interface between a soft film and a relatively hard substrate influences the results of a continuous indentation hardness test. Experimental data will be reported from films of thickness 0.25, 0.5, and 1.0 fim, including one film, the adhesion to the substrate of which has been reduced by depositing a contaminating layer of carbon between the silicon and aluminum. Other researchers have utilized indentation techniques to examine the properties of thin films and their adhesion to the substrate, where the substrate is harder than the film. Engel and Roshon3 found that an indenter J. Mater. Res. 3 (1), Jan/Feb 1988
http://journals.cambridge.org
can be used to debond a polymer coating from a copper substrate. The radius of the debonded area relates to the adhesive strength of the interface between the polymer and substrate. For a specimen of uniform hardness, the load required to drive a conical or pyramidal indenter into the surface is expected to increase at a rate proportional to the square of the indentation depth if the average pressure beneath the indenter is inde
Data Loading...
