[CP2K:9331] Optimizing Geometry AND Unitary cell

[Henrique Junior]

Thank you, my friends, ASE looks like a promising solution. I'll give it a 
try.


On Wednesday, September 6, 2017 at 6:13:30 PM UTC-3, Steve Schmerler wrote:
>
> On Aug 26 14:38 -0700, Henrique Junior wrote: 
> > Dear colleagues, 
> > I’m new to CP2K and I’m going to perform a geometric optimization in my 
> > crystalline structure (to calculate vibrational frequencies, also using 
> > CP2K). In my geometric optimization, I need to, also, optimize the size 
> of 
> > my unitary cell BUT I’d like to specify a range in which the cell is 
> > allowed to expand (or contract, of course). Is this possible? Can 
> someone 
> > point me to any example? 
>
> If you need a special-purpose optimizer, you may have a look at ASE [1]. 
> They have a CP2K calculator [2] for some time. The ASE structure 
> optimizer calls CP2K (or any other backend code) in each step to kick 
> off an SCF calculation to get energy and forces. 
>
> You want to use one of their optimizers [3], together with the 
> UnitCellFilter constraint [4] to treat the cell. You would need to 
> modify the UnitCellFilter to meet you specific needs, or roll your own 
> entirely [5]. 
>
> [1] https://wiki.fysik.dtu.dk/ase/ 
> [2] https://wiki.fysik.dtu.dk/ase/ase/calculators/cp2k.html 
> [3] https://wiki.fysik.dtu.dk/ase/ase/optimize.html 
> [4] 
> https://wiki.fysik.dtu.dk/ase/ase/constraints.html#the-unitcellfilter-class 
> [5] 
> https://wiki.fysik.dtu.dk/ase/ase/constraints.html#making-your-own-constraint-class 
>
> best, 
> Steve 
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170906/45d84b66/attachment.htm>