Shan Huang, William Klein, and Harvey Gould
Phys. Rev. E 103, 033305 (2021)
Machine learning has been successfully applied to a number of physics problems, but the method works less well for a system that is near a critical point. An application to nucleation in the two-dimensional Ising model shows that as the spinodal is approached and the densities of the nucleating droplet and the background become similar, the method is less accurate.
Max Okraschevski, Niklas Buerkle, Rainer Koch, and Hans-Joerg Bauer
Phys. Rev. E 103, 033304 (2021)
Methods such as weakly-compressible smoothed particle hydrodynamics are frequently used in fluid dynamics, but they are known to struggle with the direct numerical simulation of vortices and turbulence at finite resolution. By making a connection with nonequilibrium molecular dynamics, the authors identify a tensor quantity that may help understand and remedy these problems.
Mary Lou P. Bailey et al.
Phys. Rev. E 103, 032405 (2021)
Single-particle tracking makes it possible to record trajectories of individual particles, that can then be analyzed to characterize the underlying dynamics. This paper describes a tool to perform this analysis even in the case of anomalous diffusion, with a test on the in-vivo motion of a chromosomal locus in a species of yeast.
Grace H. Zhang and David R. Nelson
Phys. Rev. E 103, 022139 (2021)
Dislocations in crystals can cluster together, for example at grain boundaries, and such pileups affect the properties of the material. This work studies the behavior of one-dimensional dislocation pileups in two-dimensional crystals through mappings onto a Coulomb gas and quantum Brownian motion, and identifies two phase transitions as a function of temperature.
Sign up to receive regular email alerts from Physical Review E