QM/MM wth core-shell model

[Matt W]

Hi,

Unfortunately (see below) the core-shell model isn't implemented for QMMM 
in CP2K at the moment (shell charges are ignored, if I remember correctly, 
and you get what you report).

I had told Teo (Teodoro Laino) that I would implement this, and do have a 
more-or-less working core-shell QMMM implementation somewhere. But it would 
take some effort to bring back into line with a current version of CP2K and 
debug properly. It would be plausible to work on this with a bit of help if 
you were interested, but would take a bit of time to get working.

Matt

On Tuesday, March 31, 2015 at 3:28:54 PM UTC+1, Chris Lee wrote:
>
> Dear CP2K users,
>
> I'm trying to model a hydroxylated iron oxide (Fe2O3) surface using QM/MM. 
> The top half of the slab is modelled using QM, while the classical system 
> is modelled using a core-shell model for the oxygen atoms only.
>
> The issue that I'm having is that the system is supposed to be 
> electrically neutral but I'm getting an overall negative charge as the QM 
> oxygen atoms all have a charge of -2 but the QM iron and hydrogen atoms 
> have zero charge. If I disable the core-shell part of the model none of the 
> QM atoms show any charge (I think it's supposed to be this way, please 
> correct me if I'm wrong) and the classical system is overall neutral with 
> each atom showing the correct charge.
>
> Any help with this would be greatly appreciated as I'd very much like to 
> use a core-shell model for my classical system.
>
> I've attached the input files that I'm using.
>
> Chris
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20150331/b3f8744c/attachment.htm>