We compare a newly developed hybrid simulation method which combines classical molecular dynamics (MD) and computational fluid dynamics (CFD) to a simulation consisting only of molecular dynamics. The hybrid code is composed of three regions: a classical MD region, a continuum domain where the dynamical equations are solved by standard CFD methods, and an overlap domain where transport information from the other two domains is exchanged. The exchange of information in the overlap region ensures that momentum, energy, and mass are conserved. The validity of the hybrid code is demonstrated by studying a single polymer tethered to a hard wall immersed in explicit solvent and undergoing shear flow. In classical molecular dynamics simulation a great deal of computational time is devoted to simulating solvent molecules, although the solvent itself is of no direct interest. By contrast, the hybrid code simulates the polymer and surrounding solvent explicitly, whereas the solvent farther away from the polymer is modeled using a continuum description. In the hybrid simulations the MD domain is an open system whose number of particles is controlled to filter the perturbative density waves produced by the polymer motion. We compare conformational properties of the polymer in both simulations for various shear rates. In all cases polymer properties compare extremely well between the two simulation scenarios, thereby demonstrating that this hybrid method is a useful way to model a system with polymers and under nonzero flow conditions. There is also good agreement between the MD and hybrid schemes and experimental data on tethered DNA in flow. The computational cost of the hybrid protocol can be reduced to less than 6% of the cost of updating the MD forces, confirming the practical value of the method.

1.
P.
Ahlrichs
and
B.
Dünweg
,
J. Chem. Phys.
111
,
8225
(
1999
).
2.
A.
Malevantes
and
J.
Yeomans
,
Europhys. Lett.
52
,
231
(
2000
).
3.
N. A.
Spenley
,
Europhys. Lett.
49
,
534
(
2000
).
4.
S.
O’Connell
and
P.
Thompson
,
Phys. Rev. E
52
,
R5792
(
1995
).
5.
G.
Ayton
,
S.
Bardenhagen
,
P.
McMurtry
,
D.
Sulsky
, and
G. A.
Voth
,
IBM J. Res. Dev.
45
,
417
(
2001
).
6.
J.
Li
,
D.
Liao
, and
S.
Yip
,
Phys. Rev. E
57
,
7259
(
1998
).
7.
J.
Li
,
D.
Liao
, and
S.
Yip
,
J. Comput.-Aided Mater. Des.
6
,
95
(
1999
).
8.
M. Allen and D. Tildesley, Computer Simulations of Liquids (Oxford University Press, Oxford, 1987).
9.
P.
Doyle
,
B.
Ladoux
, and
J.-L.
Viovy
,
Phys. Rev. Lett.
84
,
4769
(
2000
).
10.
P.
LeDuc
,
C.
Haber
,
G.
Bao
, and
D.
Wirtz
,
Nature (London)
399
,
564
(
1999
).
11.
R. Delgado-Buscalioni, P. V. Coveney, and E. Flekkøy (unpublished).
12.
B.
Dünweg
and
K.
Kremer
,
J. Chem. Phys.
99
,
6983
(
1993
).
13.
C.
Aust
,
M.
Kröger
, and
S.
Hess
,
Macromolecules
32
,
5660
(
1999
).
14.
C.
Pierleoni
and
J.-P.
Ryckaert
,
Macromolecules
28
,
5097
(
1995
).
15.
G. G. B. Dünweg and K. Kremer, in IMA Volumes in Mathematics and Its Applications, edited by S. Whittington (Springer, New York, 1998), Vol. 102.
16.
R.
Jendrejack
,
J.
de Pablo
, and
M.
Graham
,
J. Chem. Phys.
116
,
7752
(
2002
).
17.
P. A.
Thompson
and
M. O.
Robbins
,
Phys. Rev. A
41
,
6830
(
1990
).
18.
S.
Barsky
and
M. O.
Robbins
,
Phys. Rev. E
63
,
021801
(
2001
).
19.
K.
Kremer
and
G. S.
Grest
,
J. Chem. Phys.
92
,
5057
(
1990
).
20.
R.
Khare
,
J. J.
de Pablo
, and
A.
Yethiraj
,
Macromolecules
29
,
7910
(
1996
).
21.
R.
Delgado-Buscalioni
and
P. V.
Coveney
,
J. Chem. Phys.
119
,
978
(
2003
).
22.
R.
Delgado-Buscalioni
and
P. V.
Coveney
,
Phys. Rev. E
67
,
046704
(
2003
).
23.
D. J. Evans and G. P. Morris, Statistical Mechanics of Nonequilibrium Liquids (Academic, London, 1990).
24.
S.
Hess
,
M.
Kroger
, and
H.
Voigt
,
Physica A
250
,
58
(
1998
).
25.
As explained in Refs. 22 and 26, these restrictions come from the C flow (stability of the scheme and spatiotemporal resolution) and from the P system (local thermodynamic equilibrium, signal-to-noise ratio).
26.
S. V. Patankar, Numerical Heat Transfer and Fluid Flow (Hemisphere, New York, 1980).
27.
R. Delgado-Buscalioni and P. V. Coveney, Philos. Trans. R. Soc. London (to be published).
28.
M. Doi and S. Edwards, The Theory of Polymer Dynamics (Clarendon, Oxford, 1986).
29.
We checked that the monomer separation remains unchanged for the range of γ̇ considered (it varies in less than 0.005σ).
30.
For a liquid with ρ≃0.8 the USHER algorithm needs around 30 iterations to insert a LJ atom at a location where the potential energy equals the mean specific potential energy of the system (Ref. 21). If the density is decreased to 0.65, it only needs about 15 iterations (see Table I).
This content is only available via PDF.
You do not currently have access to this content.