Atomistic simulations of crack nucleation and intergranular fracture in bulk nc Ni

Ajing Cao's picture

We report large scale molecular dynamics simulations of dynamic cyclic uniaxial tensile deformation of pure, fully dense nanocrystalline Ni, to reveal the crack initiation, and consequently intergranular fracture is the result of coalescence of nanovoids by breaking atomic bonds at grain boundaries and triple junctions. The results indicate that the brittle fracture behavior accounts for the transition from plastic deformation governed by dislocation to one that is grain-boundary dominant when the grain size reduces to the nanoscale. The grain-boundary mediated plasticity is also manifested by the new grain formation and growth induced by stress-assisted grain-boundary diffusion observed in this work. This work illustrates that grain-boundary decohesion is one of the fundamental deformation mechanisms in nanocrystalline Ni.

The new paper is published on Phys. Rev. B, 76, 024113 (2007).

Cao_Wei_PRB_2007.pdf916.36 KB
De Xie's picture

A question?

Congrats for the pubilication!

When I look at those pictures. A quick question comes into my mind:

Did you use any Finite Element Analysis for your simulation?




Ajing Cao's picture

Hi De

Hi De,

I  didn't model nanocrystalline metals by using Finite Element Methoed since the physics governing the deformation is still not well understood. We currently focus on revealling the atomic level physics of the deformation process of nanocrystalline metals with such tiny grain size. With these fundamental understandings, we will try to use the physics-based multiscale model (FEM also) to predict the mechanical properties of these materials.

Do you have some suggestions on fracture modelling by using FEM?  Thanks!

I mentioned a reference which used the FEM to model the nanocrystalline metals combined the grain boundary seperation with grain interrior plasticity: Y. J. Wei and L. Anand, J. Mech. Phys. Solids 52, 2587 (2004).


Sandip Haldar's picture

Hi Ajing, I have started

Hi Ajing,

I have started working in Molecular simulation. In your work what  are you using for simulation i.e. any package or your own code? I am learning LAMMPS.

Sandip Haldar

Ajing Cao's picture

Hi Sandip

The code I used is modified version of Paradyn, which is originally developed by S. Plimpton at Sandia National Lab. Lampps is also a powerful package and many people use it. Wish you good luck.