A new atomistic kinetic Monte Carlo method for bridging the time gap in radiation damage studies
Haixuan Xu
Image

With a new computational tool, scientists can understand how atomic defects interact and evolve at a timescale much longer than molecular dynamics.

Although an atomic displacement cascade, which is the unit event of radiation damage production in structural materials, happens on the timescale from femto- to pico-seconds, 10-15 to 10-12 seconds, materials in commercial nuclear reactors have to serve for up to 60 years. An analogy is ultraviolet light damages skin with subsequent wrinkling happening years later. Due to the defects created by radiation, the materials undergo changes in microstructure and physical properties and become radioactive, which creates challenges for experimental investigations because of safety concerns. Therefore, theoretical methods and simulations play a critical role in predicting the performance of structural materials to avoid catastrophes.

At the Center for Defect Physics, scientists devised a new method named self-evolving atomistic kinetic Monte Carlo, or SEAKMC, to obtain accurate simulations of the atom's behavior at longer timescales. Molecular dynamics calculations are the dominant tool for studying how radiation damage occurs and evolves over very short time processes, but these calculations generally cannot examine reactions at microseconds or more because of how the calculations handle the atoms. In SEAKMC, reaction pathways and rates are calculated on the fly based on the instantaneous position of atoms in the system to determine what processes or reactions are possible. A feature called the active volume, which is associated with local regions containing defects, was introduced to decrease the computational costs and to increase the performance while maintaining the accuracy.

The researchers conducted rigorous comparisons between the new approach and the conventional molecular dynamics calculations. They found the new approach accurately describes how defects in the materials move and behave.

Roger Stoller, one of the key developers of this new method, says, "Materials scientists have been working for many years to develop a simulation method that provides atomistic fidelity to the underlying physical processes while reaching timescales long enough to make interesting comparisons with experimental results. SEAKMC is a significant advance -- able to simulate defect evolution for long times with molecular-dynamics-like accuracy."

Because SEAKMC only requires atomic positions and an interatomic potential as input, it has potential applications in physics, chemistry, and materials science. In the Center for Defect Physics, the SEAKMC has been applied to selected cases, including the long-term evolution of defect structures that are created from cascade simulations using molecular dynamics on the picosecond timescale. SEAKMC has been shown to provide a faithful reproduction of the atomic processes observed in molecular dynamics over short times, but is capable of reaching a much longer timescale.

Predicting the properties and performance of materials across multiple length and timescales is the ultimate goal, and SEAKMC is a significant step toward bridging the gap in timescales between radiation damage formation and property changes. The results obtained from SEAKMC can be used to improve and provide input to higher level mesoscale models that are used to simulate material performance in nuclear reactors. The SEAKMC can also be a general tool for studying the behavior and interaction of the broad range of defects that occur in and control the properties of all materials. As such, it can be useful in a broad range of fundamental materials investigations.

More Information

Xu HX, YN Osetsky, and RE Stoller. 2012. “Self-Evolving Atomistic Kinetic Monte Carlo: Development and Application.” Journal of Physics: Condensed Matter 24:375402. DOI: 10.1088/0953-8984/24/37/375402

Xu HX, YN Osetsky, and RE Stoller. 2011. “Simulating Complex Atomistic Processes: On-the-Fly Kinetic Monte Carlo Scheme with Selective Active Volumes.” Physical Review B 84(13):132103. DOI: 10.1103/PhysRevB.84.132103

Acknowledgements

This material is based upon work supported as part of the Center for Defect Physics, an Energy Frontier Research Center funded by the Department of Energy, Office of Science, Office of Basic Energy Sciences.

About the author(s):

Haixuan Xu is a postdoctoral fellow at the Center for Defect Physics, an Energy Frontier Research Center, at the Oak Ridge National Laboratory. His research interests are defects in metallic and ceramic materials using computational methods for energy applications and novel methods of multiscale simulation.

Newsletter Articles

Research Highlights