Dresden 2020 – wissenschaftliches Programm

Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

DY: Fachverband Dynamik und Statistische Physik

DY 15: Condensed-matter simulations augmented by advanced statistical methodologies (joint session DY/CPP)

DY 15.6: Vortrag

Montag, 16. März 2020, 17:00–17:15, HÜL 186

Edgy and Parallel – Efficient Equilibration of Anisotropic Hard Particulate Systems — •Marco Klement and Michael Engel — Institute for Multiscale Simulation, FAU Erlangen, Erlangen, Germany

Particle simulations are an important method to study the phase behavior of fluids and solids. A common task is structure prediction via thermal equilibration. Examples are crystallization or melting and the aging of glasses. Near the liquid-solid phase transition of hard spheres event driven molecular dynamics (EDMD) is known as most efficient. Correlated movement of particles in EDMD aids the equilibration. For Monte Carlo methods event-chains [1] introduce a correlation of particle updates. Particles move until the next collision. One of several collision rules determines the new particle to continue the chain and the new direction of the chain. The recently published Newtonian event-chain (NEC) algorithm [2] is an advanced statistical methodology that assigns velocities to particles and performs collisions elastically. It is a close competitor in efficiency to EDMD. In this contribution we present the extension of NEC to polyhedral particles. We also discuss how parallelization strategies [3] can be adapted to this algorithm. Our implementation has been added to the software package HOOMD-blue. [4]

[1] Bernard et al., Phys. Rev. E 80, 056704 (2009)

[2] Klement et al., J. Chem. Phys. 150, 174108 (2019)

[3] Kampmann et al., J. Comput. Phys. 281, 864-875 (2015)

[4] http://glotzerlab.engin.umich.edu/hoomd-blue/