[gmx-users] Re: water dimer is gas phase

[Erik Marklund]

Justin A. Lemkul skrev 2011-05-19 22.42:
>
>
> Vitaly Chaban wrote:
>>> For vacuum systems, "pbc = no" in conjunction with zero cutoffs is 
>>> correct
>>> to obtain a gas-phase system.
>>
>>
>> For reasonably large box, this is the same heck.
>>
>
> Setting a box size and using "pbc = xyz" precludes the use of infinite 
> cutoffs, which should be the most rigorously correct way to treat this 
> type of system. With PBC, there are long-range effects that should be 
> avoided.
>
>> Nilesh: visualize your system, using VMD or ngmx. The behavior of your
>> molecular cluster during the first ps will provide an answer. Are you
>> sure that your topology is healthy?
>>
>
> I agree with this.  The system should be going absolutely haywire :)
>
> -Justin
>
Yes, this is a good test, and the topology should be inspected. However, 
the observed result is expected for this system. A water dimer coupled 
to a thermostat will dissociate eventually (pretty rapidly) at 300 K. 
The water molecules themselves seem intact. The problem with 
visualization comes from the fact that gromacs fails to write valid pdb 
files when the coordinates are too large. I called this a bug on the 
developer list one or two years ago since this generates pdb files that 
gromacs itself can't read, but the other developers didn't catch on. You 
can postprocess the pdb and cap the coordinates so that they're confined 
to the range that the pdb format allows, but note that this will mess up 
the water moleucles as they get very far apart.

Erik

-- 
-----------------------------------------------
Erik Marklund, PhD student
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,    75124 Uppsala, Sweden
phone:    +46 18 471 4537        fax: +46 18 511 755
erikm at xray.bmc.uu.se    http://folding.bmc.uu.se/