Abstract for the Fourth Foresight Conference on Molecular Nanotechnology.

THE DYNAMICS OF FLUID FLOW INSIDE NANOMACHINES*

Robert E. Tuzun (a) Donald W. Noid (b)Bobby G. Sumpter

Chemical and Analytical Sciences Division
Oak Ridge National Laboratory
Oak Ridge, TN 37831-6197

(c) Ralph C. Merkle
Xerox PARC
3333 Coyote Hill Road
Palo Alto, CA 94304

This abstract is actually for two separate but closely related papers. The first paper is available in postscript at http://www.zyvex.com/nanotech/nano4/tuzun/paper1/paper1.ps.

The figures for the first paper are available in postscript in the directory http://www.zyvex.com/nanotech/nano4/tuzun/paper1/. The figures are: 1, 2a, 2b, 3a, 3b, 4, 5a, 5b, 6a, 6b.

The second paper is available in postscript at http://www.zyvex.com/nanotech/nano4/tuzun/paper2/paper2.ps.

The first figure for the second paper is available in jpeg.

The rest of the figures for the second paper are available in postscript in the directory http://www.zyvex.com/nanotech/nano4/tuzun/paper2/. The figures are: 2ap2, 2bp2, 2cp2, 3ap2, 3bp2, 3cp2, 3dp2, 3ep2, 4ap2, 4bp2, 4cp2, 5ap2, 5bp2, 5cp2, 6ap2, 6bp2, 6cp2, 7ap2, 7bp2, 7cp2, 8p2, 9ap2, 9bp2, 9cp2, 10ap, 10bp, 10cp.

The behavior of fluids in nanotechnology applications is not a well-developed area of research and is amenable to analysis by molecular dynamics (MD) techniques. The dynamics of fluid flow through nanomachines is different from other systems in that the flow is granular and that the "walls" move. We have simulated the flow of helium or argon through several sizes of carbon (graphite) nanotube. The fluid was started with some initial velocity; fluid particles were allowed to recycle through the tube via minimum image bounday conditions. Argon slowed down more quickly than helium. In addition, the behavior of the fluid strongly depended on the rigidity of the tube; a dynamic tube slowed down the fluid far more quickly than one in which the tube was held frozen. It also depended on the fluid density and tube diameter. It did not, however, depend on the tube length, because fluid flow tended to prevent the development of strong longitudinal modes, whose behavior are length-dependent.

We have also simulated the flow of C60, either as an idealized atom or a fully atomistic cage, flowing inside a carbon nanotube with helium as a carrier gas. The C60 either had zero initial velocity (to simulate the C60 being introduced into a feedstream) or moved with the initial fluid velocity. When initially motionless, the buckyball reached fluid velocity within about 5 ps, but once at fluid velocity sometimes slowed down somewhat less quickly than the fluid. The amount of fluid leakage around the C60 strongly depended on the tube diameter. In addition, simulations with rigid carbon nanotubes exhibited more fluid leakage than those with dynamic tubes.

*Research sponsored by the Division of Materials Sciences, Office of Basic Energy Sciences, U.S. Department of Energy,
under contract DE-ACO5-84OR214OO with Lockheed Martin Energy Systems, Inc.