Back to Search
Start Over
A hybrid MD-DSMC coupling method to investigate flow characteristics of micro-devices
- Source :
- Journal of Computational Physics. 302:603-617
- Publication Year :
- 2015
- Publisher :
- Elsevier BV, 2015.
-
Abstract
- A new methodology is proposed to couple Molecular Dynamics (MD) and Direct Simulation Monte Carlo (DSMC) methods to simulate high Knudsen number (Kn) flows. For this purpose a two-dimensional hybrid MD-DSMC code is developed. In this method gas-surface interactions are modeled using MD, and gas-gas interactions are modeled using DSMC method. Two-way coupling between MD and DSMC is implemented by employing buffer zones for both MD and DSMC regions. Bootstrap sampling and energy minimization algorithms are employed for dynamic coupling of these two methods since MD utilizes real number of molecules during simulation whereas DSMC utilizes a lesser number of simulated molecules. The hybrid methodology combines the advantages of both methods; it has the capability of modeling the gas-surface interaction accurately considering the effect of the presence of neighboring real number of gas molecules, while in the bulk it utilizes DSMC with only the simulated number of molecules thus increasing the computational efficiency significantly compared to pure MD codes. As a result comparatively large domain sizes can be simulated with realistic behavior at the walls. The utility of the hybrid method is demonstrated by simulating high Kn flows through a micro-channel, micro-nozzle and micro-scale shock tube. The effect of partial accommodation of gas molecules with the wall is seen to be captured dynamically with this approach. (C) 2015 Elsevier Inc. All rights reserved.
- Subjects :
- Micro devices
Physics and Astronomy (miscellaneous)
Flow (psychology)
Molecular Dynamics
Energy minimization
Physics::Fluid Dynamics
Molecular dynamics
Continuum
Statistical physics
Shock tube
High Knudsen Number Flow
Coupling
Physics
Molecular-Dynamics
Numerical Analysis
Applied Mathematics
Gas-Flows
Computer Science Applications
Hybrid Methods
Computational Mathematics
Microchannel
Modeling and Simulation
Direct Simulation Monte Carlo
Fluid
Gas-Surface Interactions
Knudsen number
Direct simulation Monte Carlo
Simulation
Subjects
Details
- ISSN :
- 00219991
- Volume :
- 302
- Database :
- OpenAIRE
- Journal :
- Journal of Computational Physics
- Accession number :
- edsair.doi.dedup.....dac494a0c05c70c8e414726297f68b8c