Chemistry Reference
In-Depth Information
Success of the multiple time-step methodology depends on computing these
kN
N 2 interactions at each step, i.e., at intervals of h Eqs. [3]-[6] and
[8], while computing the remaining N
pair interactions (which corre-
spond to forces that vary more slowly) at longer time intervals
ð
N
k
Þ
h .
Street et al. 24 originally presented the MTS method in the context of a
distance truncated Lennard-Jones potential, so that the total number of com-
puted interactions was somewhat less that N 2 . For biological MD applications
there is evidence that cutoffs can cause undesirable artifacts. 25,26
Before discussing the implementation details, we need to state the general
issues of multiple time-step numerical methods. The central objectives are: (1)
to devise a splitting of the systematic forces into a hierarchy of two or more
force classes based on the time interval over which they vary significantly, and
(2) to incorporate these force classes into a numerical method in a way that
realizes enhanced computational efficiency and maintains stability and accu-
racy of the computed solution.
t
Splitting the Force
Streett and co-workers 24 proposed a splitting of forces based on a dis-
tance parameter r split . In the potential energy formalism we write
X
X
V
ð r Þ¼
V ij þ
V ij
½
11
j r i r j j< r split
j r i r j j r split
¼ V fast
ð r Þþ V slow
ð r Þ;
½
12
with
F fast
ð r Þ :¼r r V fast
F slow
ð r Þ :¼r r V slow
ð r Þ
ð r Þ
½
13
As a practical matter, particles will move in and out of the r split sphere for a
given particle over the course of a simulation. To avoid discontinuities that
result from a particle suddenly changing classification from the slow force
component to the fast force component, the force can be decomposed into
fast and slow components using a switching function S
ð
r
Þ
,
F i ¼ F fast
i
þ F slow
i
¼
S
ð
r
Þr r i V
ð
1
S
ð
r
ÞÞr r i V
½
14
where
8
<
1
r
<
r split l
R 2
S
ð
r
Þ¼
1
þ
ð
2 R
3
Þ
r split l
r
r split
½
15
:
0
r split <
r
Search WWH ::




Custom Search