Kinetic-based investigation of hardening mechanisms in nanolayer composites.

摘要

Strengthening in nanolayered composite thin films has been attributed to a number of factors including the layer interface Hall-Petch mechanism, modulus mismatch, coherency strains, and layer grain size. In general in order to identify and distinguish between these different mechanisms, the effects of layer thickness on hardness are examined. It is often found (or predicted based on theory) that for comparatively large layer thickness, t, the hardness obeys the relation H = H 0 + H1t −n. It is supposed that H0 and H1t−n are independent of each other. It is often predicted and/or observed that the hardness saturates for sufficiently small t (a few tens of nm) at a level H2. In order to test the hypothesis that H0 and H 1 are independent of each other, we propose a method of obtaining the derivatives of H0 and H 1 independently through an activation analysis of nanoindentation creep, load relaxation, and rate-change experiments. Such experiments are performed on Cu/Ni, Cu/Nb and Si/Si-Ge nanolayer composites. In addition, an activation analysis is proposed in the study of the monolithic thin films of copper, nickel and niobium thin films. Activation data from these thin films indicate that bulk-like dislocation mechanisms operate. Activation data from bulk silicon suggest a kink propagation mechanism for dislocation glide. It is found that Cu/Ni and Cu/Nb multilayers obey H = H 0 + H1t −n and saturate at hardness levels H 2. However, when examined based on the activation analysis, the behaviors of Cu/Ni and Cu/Nb are systematically different. The cause of the discrepancy is discussed in the report. In Si/Si-Ge multilayers it appears that the hardnesses of the composites obey a simple volume average rule for the different layers. We discuss here the effects of incorporated carbon atoms and germanium on hardness and activation areas.