Explore chapters and articles related to this topic
Multiphysics in molecular dynamics simulation
Published in Ken P. Chong, Arthur P. Boresi, Sunil Saigal, James D. Lee, Numerical Methods in Mechanics of Materials, 2017
James D. Lee, Jiaoyan Li, Zhen Zhang, Kerlin P. Robert
In MD, temperature is not an independent variable as in continuum mechanics (CM), but related to the microscopic random motions of atoms. Currently, popular techniques to control temperature include velocity rescaling, the Berendsen thermostat (Berendsen et al., 1984), the Andersen thermostat (Andersen, 1980), the Nosé–Hoover thermostat (Hoover, 1985; Nosé, 1984a, 1984b), Nosé–Hoover chains (Martyna et al., 1992), and Langevin dynamics (Lemons and Gythiel, 1997). Velocity rescaling is straightforward to implement. However, this thermostat does not allow the proper temperature fluctuations, cannot remove localized correlation, and is not time reversible. As a result, it is good for use in initialization state of a system, or say to warm up a system. The Berendsen thermostat is actually a specialized velocity rescaling thermostat, such that it has same advantages and disadvantages as velocity rescaling. Andersen thermostat maintains constant temperature condition for a material system by reassigning velocities of atoms or molecules that have collisions based on Maxwell-Boltzmann statistics. Even though the algorithm allows sampling from the canonical ensemble, the dynamics in fact is not physical. Langevin dynamics allows controlling the temperature of the canonical ensemble by the use of stochastic differential equations where a friction force term and a random force term are introduced. Compared to Langevin dynamics, Nosé–Hoover thermostat is deterministic, time reversible, and easy to implement. It has been widely used to calculate material properties. However, Nosé–Hoover thermostat is not suitable for a nanomaterial system whose temperature varies spatially and temporally during the simulation with the imposition of a temperature gradient. Li and Lee (2014a) reformulated the Nosé–Hoover thermostat to locally regulate temperatures at many distinct regions without introducing the physical linear and angular momenta and finally extend the feedback temperature force to a more general level.
Investigation of the corrosion inhibition performances of various inhibitors for carbon steel in CO2 and CO2/H2S environments
Published in Corrosion Engineering, Science and Technology, 2020
Chen Zhang, Vahdat Zahedi Asl, Yuan Lu, Jingmao Zhao
The COMPASS force field was used for Molecular dynamics simulation [24]. The simulation temperature was controlled at 333 K by Andersen thermostat. The long-range Coulomb and Vander Waals interactions were handled by the Ewald and Atom Based methods, respectively. The cut-off for these interactions was set at 1.25 nm.