[Openmd-users] Development version

[Dan Gezelter]

Brett,

   There are a couple of integrators that might get close to what you want:

   NPTf:  flexible box geometry NPT  (shearing allowed)
   NPAT:  constant normal pressure, constant lateral surface area
   NPrT:  constant normal pressure, constant lateral surface tension
   NgammaT:  constant lateral surface tension, fixed z.
	
It sounds like NPrT is closest to what you want.  The z coordinate will couple to the targetPressure,  while x&y will couple to a surfaceTension.  This won't necessarily keep x & y locked to each other. However, in most membranes, they will stay fairly close.

If that doesn't do what you want, NPTxyz is already a utility integrator (it doesn't sample an ensemble with a known free energy like the NVT, NPT, and NPTf integrators do).  It could be easily cloned to make an integrator that scales x&y together.

 --Dan

On Sep 22, 2010, at 10:03 AM, Brett Donovan wrote:

> Hi Dan
> 
> Thanks for this! I have a quick question about ensembles, I'm
> equilibrating using the NPTxyz ensemble, however I'm keen to keep the
> square dimensionality of the membrane frame, i.e. couple x and y
> pressure wise but allow z to float independently. I don't know if this
> has been implemented in the schema?
> 
> Best
> 
> Brett
> 
> 
> 
> On Wed, Sep 22, 2010 at 3:56 AM, Dan Gezelter <gezelter at nd.edu> wrote:
>> Brett,
>> 
>>  In all honesty, you are probably going to want to wait a bit until we are done messing around under the hood.  The code to do the explicit interactions between atom types is in src/nonbonded/LJ.cpp, but the InteractionManager, which will handle the branching and bookkeeping to the various interactions in src/nonbonded, is currently one of the parts of the code that doesn't compile, and I haven't had a chance to fix the static dependencies and get it checked in.  InteractionManager replaces many things that had been in the fortran code (notably the do_pair routine in src/UseTheForce/doForces.F90).
>> 
>> If you are under immense pressure for results, you could take the original version and attempt to hard-code an interaction in src/UseTheForce/DarkSide/lj.F90.
>> 
>>  --Dan
>> 
>> On Sep 21, 2010, at 9:22 AM, Brett Donovan wrote:
>> 
>>> Dear Dan
>>> 
>>> Got the dev copy. We're chomping at the bit to get the interaction
>>> matrix incorporated, I have not had the chance to go through the
>>> changes fully, do you know if its in the code as we are under quite a
>>> bit of pressure to incorporate some non LB mixing features.
>>> 
>>> Best
>>> 
>>> Brett
>>> 
>>> On Thu, Sep 16, 2010 at 12:59 PM, Dan Gezelter <gezelter at nd.edu> wrote:
>>>> Brett,
>>>> 
>>>>  We normally check out the entire development branch into an "openmd-busticated" directory for the sandbox development:
>>>> 
>>>> svn co https://openmd.net/OpenMD/svn/branches/development openmd-busticated
>>>> 
>>>> Chuck Vardeman and I are usually the only people working on this copy, and we do occasionally collide.  For example, Chuck is midway through converting to cmake as our build system and is adding exception handling, while I'm rerouting the low-level fortran force routines back to C++.  You are welcome to have a look and see what's coming.
>>>> 
>>>> A couple of sentences of warning:
>>>> 
>>>> The development branch is a bit dangerous for production runs. It isn't documented yet.  It may not do what you think it should do, and there will be times when we are messing with the build system or the low-level stuff when it won't even compile. Something that works today may be broken tomorrow. I just thought you should get some fair warning...
>>>> 
>>>>  --Dan
>>>> 
>>>> On Sep 16, 2010, at 5:14 AM, Brett Donovan wrote:
>>>> 
>>>>> Dear Dan
>>>>> 
>>>>> You mentioned a development version with the interaction matrix
>>>>> included. Where can we get this from?
>>>>> 
>>>>> Best
>>>>> 
>>>>> Brett
>>>> 
>>>> ***********************************************
>>>>  J. Daniel Gezelter
>>>>  Associate Professor and
>>>>  Director of Graduate Admissions
>>>>  Department of Chemistry and Biochemistry
>>>>  251 Nieuwland Science Hall
>>>>  University of Notre Dame
>>>>  Notre Dame, IN 46556-5670
>>>> 
>>>>  phone:  +1 (574) 631-7595
>>>>  fax:    +1 (574) 631-6652
>>>>  e-mail: gezelter at nd.edu
>>>>  web:    http://www.nd.edu/~gezelter
>>>> ************************************************
>>>> 
>>>> 
>>>> 
>> 
>> ***********************************************
>>  J. Daniel Gezelter
>>  Associate Professor and
>>  Director of Graduate Admissions
>>  Department of Chemistry and Biochemistry
>>  251 Nieuwland Science Hall
>>  University of Notre Dame
>>  Notre Dame, IN 46556-5670
>> 
>>  phone:  +1 (574) 631-7595
>>  fax:    +1 (574) 631-6652
>>  e-mail: gezelter at nd.edu
>>  web:    http://www.nd.edu/~gezelter
>> ************************************************
>> 
>> 
>> 

***********************************************
  J. Daniel Gezelter
  Associate Professor of Chemistry
  Director of Graduate Admissions
  Department of Chemistry and Biochemistry
  251 Nieuwland Science Hall
  University of Notre Dame
  Notre Dame, IN 46556-5670

  phone:  +1 (574) 631-7595
  fax:    +1 (574) 631-6652
  e-mail: gezelter at nd.edu
  web:    http://www.nd.edu/~gezelter
************************************************