Re: [Avogadro-Discuss] solvent box

[B.W.H. v. B. <bw...@xs...>]

Geoff,

Thanks for your comments.

I remember from a long time ago (~20years!, when I occasionally used the
Sybyl package) that you could model a solvent by providing a mol2-file
(describing a single molecule) and a so called box-file, which contained
the coordinates of the atoms involved, a reference to the corresponding
mol2-file, and the dimensions of the box. I have browsed the web, but I
dont see any reference to this file anymore, so I don't know if this
still exists. Maybe there are some current Sybyl users that can shed
some light here.

Don't go modelling of a solvent box and periodic boundary conditions
(with minimum image convention) always go together?
At the moment I'm trying to hookup my own MD code (which -optionally-)
has period boundary conditions for rectangular boxes as an extension to
Avogadro. It does work now, but it would be nice to have the PBC in
Avogadro. I don't know the code of Avogadro or OpenBabel (the advantage
of writing a plugin...) but if there is a central place where the
distance between atoms is being calculated, this is perhaps not too
difficult. After all, the concept of a unit cell is there. Perhaps I
should put some effort in understanding the inner workings of
Avogadro/OpenBabel a bit more.

Regards,
Bertwim


Geoffrey Hutchison wrote:
>> For instance, suppose I would like to have a cubic box willed with
>> decane molecules (C10H22). Note: not crystalline decane, but a liquid.
>>     
>
> This has been a question for me -- I've seen some other programs which allow you to add a solvent box (e.g., VMD), but little information on the feature. I asked a colleague who does MD simulations (including some with zeolites) and he suggested creating a library of pre-equilibrated solvent box systems for Avogadro.
>
> If there are other codes which can create solvents at appropriate densities on-the-fly, I'd appreciate some pointers. I've come up empty. Pointers to journal articles or similar would be appreciated.
>
>   
>> - Can Avogadro (or for that matter, the OB force field) handle periodic
>> boundary conditions?
>>     
>
> No, not yet. Code contributions in this line would be significantly welcome. I don't have the time to write the code.
>
>   
>> - Is there an efficient way to model many identical molecules? (e.g.
>> like in a liquid)
>>     
>
> No, there is no solvent replication, or effective fragment approaches.
>
> Hope that helps,
> -Geoff
>
>