VMD-L Mailing List
From: Ashar Malik (asharjm_at_gmail.com)
Date: Mon Feb 01 2016 - 03:51:35 CST
Please reply to the same message if you have more questions regarding it.
As to your latest post:
if it is a good idea create a molecule (nanoparticle) that big ...
One can model particles as big as a planet or larger.
. I was hoping to create a hard sphere/soft sphere simulation of
> nanoparticle with a monolayer and give it a repulsive potential in the
> solvent I am using to reduce the required computation, since the size of
> the system has already become too big.
What you are doing is essentially coarse - graining your system. Yes this
practice is carried out. It can be done as long as the calculations you do
and the results you infer are not dependent on the level of detail of the
system. Just like gravity doesn't effect (significantly) the atoms - hence
you can remove atomic details from your calculations when you are studying
planetary motions etc.
Can that be done in NAMD or LAMMPS?
That is not the right question to ask. One would go about doing what you
are doing by first arriving at a force field. This is basically what you
said about repulsive potentials etc. Depending on what plays a role in your
system and what doesn't you would have to arrive at potentials which would
govern the dynamics of your system. NAMD is an engine which solves your
system according to the potentials you provide it. You have to determine
> I have already created my solvent box in NAMD. I just need to create a
> capped nanoparticle and solvate it.
I am not sure why you would do this? In essence the most intensive part of
any simulation is the handling of the solvent (e.g. water). If you coarse
grain everything else accept water - you will hardly see much improvement
in your computational time. You need to use a coarse grained model for
water as well to go along with this entire exercise.
You said that you only want a sphere - which is hollow. Hollow doesn't and
shouldn't mean vacuum -- you would need to define some way in which you
model the contents of this sphere with a continuum --- JMol is also a
visualizer. Systems are defined by atomic coordinates in 3-D (x,y and z).
You need to create those - for everything that constitutes the boundary. If
you want just one GIANT sphere - then there is no question of making it
hollow because you would have to give it a VDW radii and perhaps some other
potential so that it reproduces some properties of the system you are
trying to model.
Perhaps it would be good if you discussed it with your supervisor.
Hope some of what I said will help.
On Mon, Feb 1, 2016 at 8:55 PM, Ana Celia Vila Verde <acavilaverde_at_gmail.com
> Dear Shubham,
> The sphere command is for drawing, not for setting up the initial
> configuration for a simulation.
> If you need to simulate a nanoparticle covered with polymers, then I
> recommend you use a nanoparticle builder. VMD is not the best tool for
> this job. I've never used the nanoparticle builder tools myself but it's
> possible that some of them even allow you to create a nanoparticle covered
> with polymers. I guess you can also make the particle hollow so you don't
> waste time simulating the inner atoms of the nanoparticle, if it makes a
> big difference in the total number of atoms. If you do this, you'll need to
> think of a potential that enforces the nanoparticle shape.
> I hope it helps,
> On 01/02/16 08:29, Shubham Mishra wrote:
>> I am trying to create a sphere of diameter in nanometers range
>> representing a nanoparticle in a simulation. I tried graphic tools of VMD
>> and I am obtaining a sphere as required using sphere command but I am not
>> sure how to give properties of a particle viz. mass and degrees of freedom.
>> I need to simulate a nanoparticle covered with a monolayer of polymeric
>> capping agent and I don't require effect of material properties of
>> nanoparticle at the current stage. So I don't want to use nanoparticle
>> builders as mentioned in DISCUS or OpenMD and simulate the nanoparticle as
>> a macromloecule.
>> If there is any possible way to achieve this on VMD, if so some insight
>> would be much appreciated.
>> Shubham Mishra
-- Best, /A