[ase-users] Hybrid calculator

[Olga Lopez-Acevedo]

Hi,

We have recently finished a calculator that allow the mixing of any two 
calculators with interface in ASE. The mixer is in a stable version, 
well tested and freely available in github. You can take a look at

https://github.com/csmm/multiase
http://arxiv.org/abs/1211.2075

The new calculator accepts in fact any number of calculators to be 
attached and the weights can be defined by the user. In the examples we 
have used the substractive method ONIOM. For selecting regions two 
methods are now implemented, by using atom index or by defining a box 
region. We are discussing right now which additive mixing strategies to 
add (electrostatic embedding, capping bonds ...?) Most of the 
developments should be ready by the end of the summer.

The use of lammps may be more involved, depending on the force fields 
you are interested to use. We have also there a lammps-ase interface but 
the code will still evolve this semester, also there still some tests to 
implement.

Best,

Olga


On 2013-02-07 14:35, Ali Rafieefar wrote:
> Hi All,
>
> I was wondering if anyone could give me an idea how to implement the 
> idea I have in mind using ASE.
>
> The idea is to develop a calculator (or maybe a stand-alone optimizer) 
> that can use optimizations from DFTB+ and LAMMPS calculators i.e. 
> after each optimization/lattice relaxation step in DFTB+ get the 
> energies and structure data and use it as input for LAMMPS and then 
> after one optimization by LAMMPS give out the energies and structure 
> data again to DFTB+ and continue to the point where the energies 
> converge.
>
> I don't know where to start with all of that and whether similar tasks 
> have been done by others.
>
> I appreciate any help/comments/suggestions/ideas so please don't 
> hesitate to answer even if you are not sure.
>
> Kind Regards,
> Ali Rafieefar
>
>
> _______________________________________________
> ase-users mailing list
> ase-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/ase-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20130208/23c28a27/attachment.html>