SUMMARY

Metal plasticity is a multiscale phenomenon that is manifested by irreversible microstructure rearrangement associated with nucleation, multiplication, interaction, and migration of dislocations. Long range elastic interactions between dislocations and other crystal defects are important to describe, along with the nonlocal, nonlinear dislocation core field. These requirements necessitate multiscale modeling techniques which (i) describe certain lattice defects and their interactions using fully resolved atomistics, (ii) preserve the net Burgers vector and associated long range stress fields of curved mixed character dislocations in a sufficiently large continuum domain in a fully 3D model, and (iii) employ the same governing equations and interatomic potentials in both atomistic and continuum domains to avoid the usage of phenomenological parameters/criteria and ad hoc procedures for passing dislocation segments between the two domains. One such approach is the concurrent atomistic-continuum (CAC) method. Unlike many other concurrent multiscale approaches, the continuum domain in CAC admits motion of dislocations and intrinsic stacking faults through a lattice without necessity of adaptive mesh refinement while employing an underlying interatomic potential as the only constitutive relation and is thus a suitable tool for dislocation-mediated metal plasticity phenomena. In this dissertation, the CAC method is advanced in multiple aspects and applied in a series of problems in plasticity of face-centered cubic (FCC) metals. First, four significant advancements in the CAC method have been made: (i) new types of finite elements are developed which yields a more accurate stacking fault energies and core structure in coarse-grained atomistic descriptions of dislocations, (ii) zero temperature, quasistatic CAC approaches are formulated to enable the constrained multiscale optimization for a sequence of non-equilibrium dislocation configurations in metals, (iii) mesh refinement schemes for both dynamic fracture and curved dislocation migration are implemented, and (iv) the code efficiency is improved using parallelized object-oriented programming. Subsequently, this enhanced CAC method is employed to study multiple plasticity problems in FCC metals, including screw dislocation cross-slip in Ni, edge dislocation bowing out from obstacles in Al, dislocation multiplication from Frank-Read sources in Cu, Ni, and Al, as well as sequential slip transfer of curved dislocation across a Sigma 3 {111} coherent twin boundary and a Sigma 11 {113} symmetric tilt grain boundary in Cu, Al, and Ni. This work makes significant contributions to the fields of mechanics of materials and multiscale modeling. It is anticipated that the new finding will improve physical understanding of dislocation-mediated plastic deformation processes in FCC metals and may assist in formulating constitutive laws and rules used in computational techniques at higher length scales.