E-séminaire - Size effects in metals and alloys: from dislocation plasticity to grain boundary processes - Marc Legros

Lundi 12 octobre 2020 à 10h30

Visioconférence

Marc Legros - CEMES, médaille d'argent du CNRS en 2018

Résumé

The design of functional parts, from aircraft to microelectronics, often involves the choice of a material whose properties are supposed to be known. The mechanical properties of metals and alloys are often derived from stress-strain plots performed on centimeter-size samples from which elastic modulus, elastic limit (the maximum stress before reaching plastic deformation), hardening, tenacity,… are extracted. The subsequent scaling is made through the stress, that is supposed to be independent of the sample size.

In 2004, American researchers fabricated micrometer-sized pillars in monocrystalline nickel using a focused ion beam (FIB). By compressing them, they discovered that their elastic limit increased as the dimensions of the pillars decreased, undermining the possible transposition of the mechanical properties from a scale to another. While this apparent violation of critical stress invariance did not cause turmoil among aeronautical engineers, it puzzled researchers around the world for years. This "smaller is stronger" effect had long been observed in thin films, metallic wires or nanocrystalline metals (Hall-Petch's law), and because the dislocations are the "quanta of plastic deformation", they have been suspected early. Their role has recently been rationalized in terms of initial microstructure, confinement and multiplication processes. In this field of research, electron microscopy and in particular in-situ transmission electron microscopy where one can see the movements of dislocations in real time, proved essential.

We recently extended this approach to nanocrystalline metals in which deformation by dislocations is almost shut down. Because the proportion of atoms sitting in grain boundaries (GBs) is very large in this class of material, many "GB-mechanisms" have been fantasized at the end of last century to explain their mechanical properties. We will show that GBs can indeed serve as a vector for permanent deformation, but understanding how this occurs is at the very beginning.