Matéria, vol 9, Nº 1 (2004) 13 - 22 http://www.materia.coppe.ufrj.br/sarra/artigos/artigo10294 A New Method to Determine the Hardness of Thin Films J. Lesage , A. Pertuz, D. Chicot LML URA CNRS 1441 University of Lille – France e-mail: jacky.lesage@univ-lille1.fr ; didier.chicot@univ-lille1.fr ABSTRACT Measuring hardness of thin films, namely films of less than 10 μm thick, using standard microhardness testers is a very complicated task for several reasons. Among them, the most important one is due to the range of indentation loads that are available with these testers. These loads are too high to allow the determination of the hardness without involving a contribution of the substrate. In order to determine the hardness of the film it is necessary to separate the two contributions by means of a mathematical model. For that purpose, it is possible to use either one or the other models available in literature. Their application, though, requires the introduction of coefficients and data which have to be deduced from other experiments or from literature. The objective of the present work is to propose a new model, likely to avoid the knowledge of other data than that obtained easily from standard experimentation. As a general rule, all the models found in literature are based on a linear additive law expressing the measured apparent hardness (composite hardness) in function of the film and substrate hardness. From the observation that any model for the composite hardness should allow the transition between a substrate hardness tendency and a film hardness tendency when applying decreasing loads, we propose to combine two types of additive laws, series and parallel, associated respectively to each of these behaviors. The ratio (t/d) n of the thickness t of the film to the diagonal d of the indent at a power n, was found to be a pertinent parameter to express the variation of the composite hardness with the indentation load. A method was proposed to determine the value of n, and finally, the procedure was applied to determine the hardness of various films Ti, TiN, TiNx, TiC, TiCN, Cr and DLC. Keywords: thin film, hardness, model 1.INTRODUCTION In order to improve the resistance to surface damage by mechanical actions, considerable research has been conducted to increase the hardness of the superficial zone of materials. This can be achieved for example by the deposition of hard thin films at the surface of materials. In the objective of designing films possessing optimum mechanical properties, it is important to determine their hardness as precisely as possible. Unfortunately, direct determination of hardness, using a conventional micro-hardness tester, is not possible for a large range of indentation loads. This is because the substrate undergoes a part of the plastic deformation when the load is such that the depth of the indent exceeds one tenth of the film thickness [1 -3 ]. As a consequence, the hardness number H C , which is calculated, is the result of contributions by both the substrate and the coating. Mathematical models are necessary in order to separate these two contributions and numerous authors have tried to construct such models by considering various hypotheses. Whatever the hypotheses, all the models available in literature assume a linear additive law to express the composite hardness H C : (1) ( ) S F S C H H a H H − + = where H F and H S are respectively the film and the substrate hardness. The differences between the models come from the more or less well-argued functions the authors have used to describe the variation of coefficient a with the applied load. One of the earliest works was that of Bückle in 1965 [4 ], who defined empirically the coefficient a as a function of a weighting factor associated to the layers of an “influence zone” affected by the indentation. A more successful model was due to Jönsson and Hogmark in 1984 [5 ], who consider the load supporting