A comparative study on reactive and non-reactive unbalanced magnetron sputter deposition of TiN coatings

Titanium nitride (TiN) coatings were deposited by unbalanced D.C. magnetron sputtering via the non-reactive and reactive technique using a TiN or Ti target, respectively The differences of these sputter techniques have been studied in detail. Main emphasis was laid on the characterization of the ion bombardment parameters for both techniques. The ion energy and the ion/atom flux ratio was varied in the range between 30 and 120 eV and 0.1 and 10, respectively. Coating characterization was done with respect to morphology, chemical composition, crystallographic structure, hardness, and macrostresses during thermal cycling. The use of an ion energy of 30 eV combined with an ion/atom flux ratio of 8.6 and 10 results in a microhardness of approximately 47 GPa for non-reactive and reactive TiN coatings, respectively. Their biaxial stresses and grain sizes also show comparable values for both techniques of approximately -2 GPa and 23 nm, respectively. The similar properties of TiN coatings deposited using non-reactive or reactive sputtering are, however, only valid for an intense ion bombardment. The transition from porous columnar to dense fibrous structures requires a more pronounced activation of film growth by ion bombardment in the case of reactive deposition as compared to non-reactive sputtering. Mainly, this is a result of the higher energy of the N atoms and the three times higher deposition rate in the non-reactive process compared to the reactive one. Moreover, during reactive sputtering, energy is also needed to dissociate the molecular nitrogen gas. The results obtained should serve as a fundamental basis for the understanding of the differences in growth conditions for non-reactive and reactive sputter techniques. Furthermore, an explanation of the high hardness values of the coatings is given and the influence of thermal annealing on the defect density, grain size and microhardness of the coatings is presented and discussed in detail. (C) 2002 Elsevier Science B.V. All rights reserved.