VMD-L Mailing List

From: Heather Mayes (hmayes_at_hmayes.com)
Date: Tue Jan 22 2013 - 19:27:14 CST

Josh and Peitong,

To get around the fact that dynamic bonds will connect any atoms in a
selection, I took advantage of the fact that C-H bonds are shorter
than C-C bonds. For a system with only C and H, you could set two
dynamic bond representations. For the first, select all C and use a
cut-off that will capture C-C bonds (i.e. 1.9 A). For the second,
select C and H and use a cut-off reasonable for C-H bonds (i.e. 1.3
A). Since C-C bonds will fall into both categories, there will
actually be two lines drawn between two carbons. However, since they
will exactly overlap, it is not a problem. Having the shorter cut-off
when H is included helps prevent unwanted bond representations
involving H.

My best,
-Heather

On Tue, Jan 22, 2013 at 6:18 PM, Josh Vermaas <vermaas2_at_illinois.edu> wrote:
> I believe dynamic bonds will connect ANY atoms in a selection that are
> within the distance cutoff of one another. Assuming that "type C H"
> selects all carbons and all hydrogens, it should draw all possible
> connections between C and H atoms that are within the cutoff distance
> apart. If you set the cutoff to something quite high and no bonds are
> being drawn, quite likely the selection you have made is empty (ie. VMD
> doesn't know of anything that has atomtype C or H).
>
> As to the last questions, it is possible, but you'd have to think
> carefully about how you are setting up the atomselections. C-C bonds
> happen to be longer than C-H bonds, so in principle very carefully
> chosen cutoffs will catch C-C bonds but not C-H bonds. However during
> simulation it would be a very strange system where both C-C and C-H
> bonds are being made-broken simultaneously, so I'm not sure that dynamic
> bonds would be the best way of showing what you are trying to show.
>
> -Josh Vermaas
>
> On 01/22/2013 04:44 PM, Peitong Duan wrote:
>> Hello,
>>
>> How would I set dynamic bond representations for different groups of atoms? For instance, when I am setting the C-H bond length and I enter "type C H" in "Selected Atoms", no bonds will show up. Also will a selection "type C H" apply to C-C and H-H bond lengths as well? Is it possible to set C-C bonds with a different cut-off than C-H bonds?
>>
>> Thanks.
>