[gmx-users] pulling experiment

Justin A. Lemkul jalemkul at vt.edu
Mon Sep 20 04:04:29 CEST 2010

Xiaohua Zhang wrote:
> Dear gmx-users
> I want to design such kind of computer experiment:
> For a system composed of non-bonded three layers (carbon nanotubes, 
> graphene, or whatever), I want to fix layer0 (the group name) to exactly 
> (0,0,0), and pull layer2 by a constant force, for example, along 
> positive x all the time. The layer1 is free to move. I am confused to 
> define the COM pulling parameters, under the latest GROMACS 4.5.1, 
> because it seems possible to pull layer2
> pull = constant_force
> pull_geometry  =  direction_periodic
> pull_ngroups  =  2
> pull_group2 = layer2
> pull_vec2 =  1.0 0.0 0.0
> pull_k2 =  a number in unit kJ mol-1 nm-1
> but how about group1 ?
> Could you give me some suggestion?

If group1 is free to move, i.e., not under the influence of any pulling force, 
then it does not need to be considered in the pulling section.  So you will have 
one reference group (group0), one pull group (pull_ngroup = 1), and all other 
groups pertain to group2 (which are then listed as pull_group1, pull_vec1, etc).

I don't know if you can simultaneously fix layer0 to (0,0,0) in this scheme, 
since Gromacs builds all boxes relative to the origin (i.e., the box corner 
originates from the origin).  If you leave pull_group0 blank, then pulling on 
pull_group1 will be done with respect to an absolute reference of (0,0,0).




> cheers
> xiaohua
> -- 
> Xiaohua Zhang
> Suzhou Institute of Nano-Tech and Nano-Bionics
> Ruoshui Road 398, Suzhou 215123, China
> Phone: +86 512 62872552
> Mobile: +86 137 71904040
> Email: zhangxh at fudan.edu.cn <mailto:zhangxh at fudan.edu.cn>
> Email: xhzhang2009 at sinano.ac.cn <mailto:xhzhang2009 at sinano.ac.cn>


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080


More information about the gromacs.org_gmx-users mailing list