(142cj) Continuum/Molecular Hybridization Model for Small Scale Systems

Considerable effort is being paid to improve the performance and the efficiency of micro/nano electro-mechanical systems (M/NEMS), where nanoscale/microscale information transfer plays a critical role. The hierarchical nature of these systems consists of complicated nano-scale multiphysical phenomena between various sub-component interfaces with macro/mesoscale fluid flow. The continuum based theory breaks down for these systems as the confinement size is comparable to or smaller than the molecular/pseudo-particle mean free path. On the other hand, although atomistic/molecular dynamics (MD) simulation methodologies have the capabilities of calculating these phenomena, it is extremely difficult to extend simulations to real systems due to the memory requirement and computational time limitations.

We have been successful in the past in developing two separate theories to describe the multi-scale/multi-physical phenomena: (i) a novel mesoscale lattice Boltzmann method (LBM), which has the capabilities of describing transient phenomena with parallel computation facilitating the handling of complex geometries. The LBM scheme was developed for analyzing the dynamics for high Knudsen number compressible flows and heat transfer [1-3], and it was found to accurately capture the characteristic velocity and temperature slip at the interface in the nano-scale by introducing mixed boundary conditions. (ii) In the molecular scale, we developed a coarse-grained molecular dynamics model (CGMD) (with the molecular parameters obtained from ab-initio calculations) which can accurately describe the nanoscale systems with particle and solid surface coupling [4, 5]. In this study, we will hybridize the above two parallel approaches on continuum and molecular levels, and will thoroughly examine the critical fluid–fluid and fluid–solid interfaces within the multiscale framework [6].

This will be achieved by generating a buffer zone at the component interface where both CGMD and LBM solutions are valid. Here, the mean particle velocities obtained from CGMD provide boundary conditions for the LBM solution at one side of the overlap region and the constrained dynamics algorithm forces the instantaneous mean particle velocity to equate the continuum solution at the other boundary. The average continuum velocity in each cell is obtained by averaging the velocities of all CGMD particles within a differential volume that is centered on the point of interest. The equations of motion in the CGMD are also modified to match the external force obtained from the LBM via proper scaling.

To validate our hybrid model, we examine our results of the test case for simple fluid flow in a confined geometry and will extend our analysis towards the complex multi-physical phenomena in N/MEMS applications.

[1] W. T. Kim, M. S. Jhon, Y. Zhou, I. Staroselsky, and H. Chen, “Nanoscale air bearing modeling via lattice Boltzmann method,” J. Appl. Phys. 97, 10P304 (2005)

[2] H.M Kim, D. Kim, W. T. Kim, P.S. Chung, and M. S. Jhon, “Langmuir slip model for air bearing simulation using the lattice boltzmann method,” IEEE Trans. Magn., 43, 2244 (2007)

[3] S.S. Ghai, W.T. Kim, R.A. Escobar, C.H. Amon, and M.S. Jhon, “A novel heat transfer model and its application to information storage systems,” J. Appl. Phys., 97, 10P703 (2005)

[4] S. H. Vemuri, P.S. Chung, R.L. Smith, L.T. Biegler, and M.S. Jhon, “Perfluoropolyether Lubricant Interactions with Novel Graphene Overcoat via Coarse-grained Molecular Dynamics,” IEEE Trans. Magn., (accepted)

[5] P.S. Chung and M.S. Jhon, “Molecular spreading of pure and binary mixture PFPE nano films on carbon-overcoated disks,” IEEE Trans. Magn., 46, 2405 (2010)

[6] P.S. Chung, M.S. Jhon, and L.T. Biegler, “The holistic strategy in multiscale modeling,” Adv. Chem. Eng., 40, 59 (2011)