Search

Nanoindentation for measuring thin film mechanical properties is probably the most popular yet ill-understood method due to its inherent complexities. As opposed to burst pressure or microtensile tests of lithographed structures, where relatively uniform stress fields may be generated, the indentation-induced stress gradients can produce unique challenges. Because of the test's simplicity and ability to mechanically probe the smallest of scales, it is becoming increasingly applied. Five possible stages of deformation are suggested from Hertzian elastic to film delamination and double buckling. In particular metal films on harder substrates are emphasized where it is shown that dislocation nucleation and arrest are only partially understood. Later stages of film delamination are illustrated with Cu/SiO 2 /Si where it is shown that the true work of adhesion is 0.6 J/m 2 . Current limitations of indentation-induced delamination measures of toughness involve large scatter associated with sensitivity of the fracture radius to the contact radius ratio.

Using a simplified version of Johnson’s core model analysis of the plastic zone, one may determine the plastic zone size around a contact or, alternatively, determine the yield strength by measuring the plastic zone. The theoretical model contains three parameters: indentation load, yield strength and zone size so that knowing any two gives the third. This is experimentally demonstrated for a series of single crystals (Fe–3wt%Si, tungsten, zinc) and polycrystals (1100-0 Al, copper and 2024-T6 aluminum). In addition, two of these are evaluated in several work-hardened states. Plastic zone sizes are estimated principally by atomic force microscopy and Zygo interferometry imaging with some verification by transmission electron microscopy. On the theoretical side, verification of the relationship is obtained by elastic–plastic numerical analysis of a bi-material system based upon an Fe–3wt%Si single crystal with a thin oxide film. It is shown that in general predictions are reasonable down to nanometer level contacts except for two cases where an indentation size effect may dominate. The proposed relationship is suggested to be an alternative measure of yield strength compared to the often cited value of H /3, particularly at light contacts.

Experiments involving the indentation of single crystals of both tungsten and an iron alloy show that the observed yield phenomena can be predicted using a superdislocation model driven by the change in shear stress between the elastically and fully plastic loading conditions. A low density of dislocation multiplication sites is required to support elastic loading which approaches applied shear stresses on the order of the theoretical shear strength of the material. Oxide film thickness and crystal orientation are examined as parameters in controlling the yield phenomena. A model based on activation of dislocation multiplication sources is suggested to explain the initiation of the yield point during indentation and the overall load–depth relationship during indentation.