- Multiscale Material from Atom Modeling and Simulation to Continuum
- Guest Speaker
- Xianqiao Wang
- Guest Affiliation
- College of Engineering, University of Georgia
- Friday, November 2, 2012 4:00 pm - 5:00 pm
- Riverbend Research South Laboratory Auditorium
For several decades continuum theory has been a dominating theoretical framework for the analysis of materials and structures. This approach to predict material deformation and failure, by implicitly averaging atomic scale dynamics and defect evolution spatially and temporally is valid only for large system. It is realized that as technologies extend to the nanometer range, continuum mechanics at this new arena is questionable. Whereas atomic-scale modeling and simulation methods, e.g., molecular dynamics (MD), have provided a wealth of information for nano systems by elucidating the atomistic mechanisms that govern deformation and rupture of chemical bonds, these methods can only handle problems limited in length/time scales. Yet, ultimately we aim at the design and manufacture of synthetic and hierarchical material systems or structures in which the organization is designed and controlled on length scales ranging from nano to micro, even all the way to macro. Therefore multiscale modeling, from atom to continuum, is inevitably needed.
This talk presents an atom-based continuum (ABC) theory coupling with thermal, mechanical and electrical mechanism, aiming at a seamless transition from the atomistic to the continuum description of multi-element crystalline solids (which has more than one kind of atom in the unit cell). By accounting for the upgraded Nosé-Hoover thermostat and Lorentz force, we put forth a novel way to appreciate the full benefit of coupling the thermal, mechanical and electromagnetic fields at nano/micro scale. Contrary to many multiscale approaches, ABC theory proposed here is naturally suitable for the multi-physics analysis of multi-element crystals. Taking both efficiency and accuracy into consideration, we adopt a cluster-based summation rule for atomic force calculations in the finite element formulations. When coarse mesh is used, the majority of the degrees of freedom can be eliminated, hence, the computational cost can be reduced, accompanying the decrease of the accuracy of the simulation results. When the finest mesh is used, any lattice site is a finite element node, and the model becomes identical to a full-blown MD model, which is the standard model manifesting the discrepancies or accuracies of others by comparisons. It is possible to envision that the use of this new method in support of diverse applications, ranging from the exploitations of critical physical phenomena such as crack extension, phase transformation, and dislocation initiation at nano scale to the energy harvesting and design of bone materials at micro scale.
Bio: Dr. Xianqiao Wang is currently an Assistant Professor of College of Engineering at University of Georgia. He received his B.S. and M.S. degree in engineering mechanics from Hunan University (China) in 2004 and 2007, respectively. He obtained his Ph.D. degree in mechanical engineering from the George Washington University in 2011. After graduation, he joined the Mechanical and Aerospace Engineering Department at the George Washington University as a Research Assistant Professor. His main research areas are multiscale material modeling and simulation, computational nanomechanics, biomechanics, coupled physics analyses of nanomaterials, microcontinuum field theory, energy harvesting, and material design.