[DFTB-Plus-User] Bulk/slab Fe optimization problems using pbc-0-3 or trans3d-0-1 parameters

[Benjamin Hourahine]

Hello Branislav,

You could constrain the cell as being hexagonal (either Isotropic or FixAngles in the geometry optimiser), also did you use hexagonal appropriate k-points (Monkhorst-Pack grid shifts are different in this case)?

Regards

Ben

       Dr. B. Hourahine, SUPA, Department of Physics,
     University of Strathclyde, John Anderson Building,
             107 Rottenrow, Glasgow G4 0NG, UK.
     +44 141 548 2325, benjamin.hourahine at strath.ac.uk

2013/4 THE Awards Entrepreneurial University of the Year
       2012/13 THE Awards UK University of the Year

    The University of Strathclyde is a charitable body,
         registered in Scotland, number SC015263
________________________________
From: DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de> on behalf of Bálint Aradi <aradi at uni-bremen.de>
Sent: 31 January 2023 08:20
To: dftb-plus-user at mailman.zfn.uni-bremen.de <dftb-plus-user at mailman.zfn.uni-bremen.de>
Subject: Re: [DFTB-Plus-User] Bulk/slab Fe optimization problems using pbc-0-3 or trans3d-0-1 parameters

CAUTION: This email originated outside the University. Check before clicking links or attachments.

Dear Branislav

> I recompiled program and now I get reasonable structures.

That's a little bit worrisome, the results should not depend on the
compilation on the program... If in doubt, try the official binaries via
Conda, they should be OK.

> I tried two different setups for unit cell and got relatively good values:
>
> ShellResolvedSCC = Yes
> ShellResolvedSpin = No
>
> Lattice parameter = 2.9338 A and nearest neighbor distance = 2.54078 A
>
> ShellResolvedSCC = No
> ShellResolvedSpin = Yes
>
> Lattice parameter = 2.9393 A and nearest neighbor distance = 2.54553 A

Actually, I'd advise to set both, ShellResolvedScc as well as
ShellResolvedSpin to "Yes".

> However, when I try to simulate 7x7x4 Fe(110) slab using LatticeOpt =
> Yes and hexagonal symmetry I get quite distorted geometry.
> Do you think this is just a parameter set inherited problem or maybe
> DFTB+ is only good when using cubic symmetry?

DFTB+ itself does not care about symmetry at all. The problem is, for
sure, parameter set related. I know, that the iron in the pbc set was
tested on small iron clusters.

https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoi.org%2F10.1016%2Fj.chemphys.2004.03.034&data=05%7C01%7Cbenjamin.hourahine%40strath.ac.uk%7Ca30df178af7043c68b1f08db03641698%7C631e0763153347eba5cd0457bee5944e%7C0%7C0%7C638107500628267885%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=uX7B2fu58%2FyHlJk5yDDfkYf0fO1js5D%2Fl4ujuIboIos%3D&reserved=0

I don't think, anybody ever checked or optimized it for slabs, so it
might be, that the set is less optimal there. One would probably adjust
the repulsive part, provided the band structure at the ab initio
geometry is reasonable.

Best regards,

Bálint


--
Dr. Bálint Aradi
Bremen Center for Computational Materials Science, University of Bremen
https://eur02.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.bccms.uni-bremen.de%2Fcms%2Fpeople%2Fb-aradi%2F&data=05%7C01%7Cbenjamin.hourahine%40strath.ac.uk%7Ca30df178af7043c68b1f08db03641698%7C631e0763153347eba5cd0457bee5944e%7C0%7C0%7C638107500628267885%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=nj4nXPnjcEP477jBB9pKA%2FZcfkEcgOmvi3leRf1QHXc%3D&reserved=0


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20230131/51332fdc/attachment.htm>