390299 Protracted Colored Noise Dynamics for Polymer Systems

Tuesday, November 18, 2014: 4:30 PM
International 9 (Marriott Marquis Atlanta)
Andrew J. Peters, Richard A. Lawson, Benjamin Nation, Peter J. Ludovice and Clifford L. Henderson, School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, GA

Mean field potentials and Monte Carlo simulations have been used to study polymeric systems in a number of contexts and applications in recent years. While these methods provide great computational expedience, they include less accurate potentials and/or require intelligently selected moves that can affect the simulation outcomes. Conversely, molecular dynamics (MD) combined with realistic potentials can provide more accurate moves and states without a need for carefully calculated and selected moves, but such detail is gained in general at the expense of computational efficiency.  In polymeric systems, use of MD techniques can be challenging because of hindered chain diffusion, which can make equilibration of polymeric systems challenging and slow.  If the fluctuation distribution can be broadened intelligently in MD of polymers, then it may be possible for diffusion to be greatly increased and this leading to significantly faster equilibration while not sacrificing the basic ability to achieve proper thermodynamic sampling of equilibrium states in the system.  To achieve this intelligent fluctuation perturbation, in this work a time and directionally correlated colored noise is used instead of the white noise that is used in Langevin dynamics.  In the polymer simulations discussed in this work (which use a freely rotating mesoscale chain model), this correlated noise is applied along the polymer backbone contour over some finite time scale to encourage a reptation-like movement which drastically increases polymer diffusion.  The parameter space for this technique has been explored and will be discussed, with a focus on understanding how the length of time correlation and strength of the force noise impacts the simulation behavior. The impact of using such techniques for accelerating MD simulations in polymer systems will be demonstrated in the context of block-copolymer system simulations.  In particular, the use of these methods to explore the behavior of directed self-assembly of block copolymers as a method to extend high resolution lithography techniques will be highlighted.

Extended Abstract: File Not Uploaded