College of Engineering, All Events

Back to Listing

Materials Science and Engineering Colloquium

Event Type
Materials Science and Engineering
100 Materials Science and Engineering Building
Feb 5, 2018   4:00 pm  
Jaime Marian, Materials Science and Engineering, UCLA
Originating Calendar
MatSE Seminars

“Predicting the strength of W Crystals and W-Re Alloys rom parameter-free mesoscopic models”


A distinctive feature of deformation in body-centered cubic (bcc) crystals is the thermally activated motion of screw dislocations at low homologous temperatures, which manifests itself as a pronounced temperature dependence of the yield and flow stresses. As well, bcc metals are known to display a tension/compression asymmetry derived from the existence of non-Schmid resolved stresses. Bcc plasticity cannot be understood without consideration of these two phenomena, which originate in the atomic structure of dislocation cores and thus have to be studied at the appropriate scales. At the same time, the plastic behavior of metallic single-crystals is highly dependent on orientation and strain rate. However, these are macroscopic-level dependencies that cannot currently be studied with atomistic methods directly. This means that a connection must be made between calculations at the dislocation core level and at the material point level, spanning many orders of magnitude in time and space. In this work, we assemble a computational methodology grounded on an atomistic description of screw dislocation properties and non-Schmid effects, implemented into crystal plasticity and kinetic Monte Carlo models of bcc metal and alloy deformation. We find that the complete methodology is successful in predicting the experimentally measured temperature dependence of the flow stress in tungsten for several crystallographic orientations, without the need for any fitting to experimental data. As well, we simulate dislocation slip in W-Re alloys and study the softening/hardening transition using a kinetic Monte Carlo model of dislocation-solute interactions. We show results for the strength of W-Re as a function of temperature, loading orientation, solute concentration and dislocation line length, and identify conditions under which the modeling could be used to map out a high-dimensional parametric space.

link for robots only