variety of tips were used as indicated in Figure 1(b). Tip correction factors were utilized for any penetrations beyond the spherical regime. In the same study [2] it was noted that V/S was constant for the first 100 nm of penetration in a given material. However, it varied from 4.7 to 11.7 µm for four single crystals ranging from tungsten to gold. Since V/S is a length scale, this was immediately considered as having a fundamental connectivity to the plasticity process. More details about this are considered in a companion paper as part of this proceedings [5]. We then hypothesized that if this µm-size length scale from nm level experiments represented a microstructural characteristic, might it not be applicable to fracture experiments as well as indentation. While at first glance this might suggest a leap of faith, it is not so strange if you consider the modified Griffith criterion scaling with the relative amount of work associated with crack exposed surface energy and crack-tip plastic volumes. Again, as detailed in the companion paper [5], this analogous behavior for nanoindentation and crack tips is demonstrated. Given that, the step of including hydrogen effects is trivial in concept but complex in detail. That is, hydrogen can be affecting both the surface energy and the plastic work contribution in ways not understood well at all. We [3] and others [6] believe that thin film decohesion, as described in the next section, can provide some of the answers. In a recent paper and a review, some 25 film/substrate multilayers were evaluated for the effects of film thickness, yield strength, test temperature and interfacial chemistry on delamination toughness [3]. It was concluded that there are brittle to ductile transitions (BDT) in metal films deposited on ceramic or semiconductor substrates. This was analogous to the BDT that occurs when either reducing thickness (increasingly plane stress) or raising test temperature (reducing yield strength) in bulk test specimens. Additional features of this are detailed in the companion paper [5]. One significant feature is that as film thickness decreases, plastic energy dissipation decreases. As shown later under results and discussion, the fracture resistance decreases to the true surface energy of the interface. This allows the local stress intensity ( [ ]1/ 2 2 i IG k E γ ≡ ) at the Griffith energy level to be established without having to deconvolve the plastic energy dissipation contribution. This may be coupled with a nanometer length scale associated with the distance of the nearest dislocation to the crack tip. Often referred to as a dislocation free zone (DFZ) this leads to the second necessary parameter, c. In a series of papers [7], it is shown that if both kIG and c are known, use of the Rice-Thomson [8] and Lin-Thomson [9] formalism leads to ( ) = 1/ 2 3/2 1/ 2 6 2 2 exp 2 exp(4/3) c k c K ys IG ys c σ π σ (2) Having the yield strength, σys, and the Griffith energy which gives kIG, the only unknown is the dislocation stand-off distance, c, which controls the local stress and hence the fracture criterion. We believe there is a connectivity between the length scale in nanoindentation, V/S, and an approach which should eventually lead to an understanding of the stand-off distance, c. While we leave that discussion for the companion paper, a more qualitative argument here is that the number of dislocations in a pile-up follows a similar relationship for cracks and nanoindentations implying the same length scale applies. Approximately six years ago [10] we suggested that observations of crack-tip and nanoindentation-tip emission of dislocations “will eventually lead to more precise mechanisms and solutions for brittle-ductile transition and contact wear phenomena.” In a series of papers [7,11,12] we suggested that the number of dislocations in an inverse pile-up might be similar for a yield instability at an indenter or at a crack tip. The number of dislocations in the
RkJQdWJsaXNoZXIy MjM0NDE=