[DFTB-Plus-User] Geometry Optimization fails

[Jerry Tanoury]

Hi Sebastian,
Thank you for the quick reply.  I used GFN2-xTB to convince myself the compound was structurally sound, since I was having difficulties with a dftb+ optimization.  I'll try your revised SK files and include dftd4.

Thanks again,
Jerry

Gerald J. Tanoury, Ph.D.
Principal Scientific Fellow
Vertex Pharmaceuticals
50 Northern Avenue
Boston, MA 02210
Office: 617-341-6219
Cell: 508-740-4986
jerry_tanoury at vrtx.com
www.vrtx.com

-----Original Message-----
From: DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de> On Behalf Of Sebastian Ehlert
Sent: Thursday, March 10, 2022 3:46 PM
To: dftb-plus-user at mailman.zfn.uni-bremen.de
Subject: Re: [DFTB-Plus-User] Geometry Optimization fails

CAUTION: This email originated from outside of Vertex. Do not open links or attachments unless you know the sender and believe it is safe.



Hi Jerry,

this is known issue with the 3ob set, for some element combinations the short range repulsive is just a stub value, like found in the Mg-Cl.skf

    Spline
    24  5.22
    -2.161275291277553e-01  1.0  1.0
    4.05000   4.10000    ...

This results in an ill-defined repulsive energy at short range, which will either yield NaN (e.g. for PF4+ or SF5+) or even negative repulsion energies and nuclear fusion (like in your example), due to bond lengths shorter than the available spline values. I have a couple of revised SK files from Tomas Kubar which fix at least the stub repulsive values for those pairs [1]. However, those have not been included in the official 3ob set yet.

[1]: https://github.com/dftbparams/3ob/pull/3

Unrelated to the original question, may I ask for the reason for using GFN2-xTB for preoptimization and than reoptimize just with plain DFTB3?
I would recommend to include at least a dispersion correction like D4

    Dispersion = DFTD4 {
        s6 = 1
        s8 = 0.4727337
        s9 = 0
        a1 = 0.5467502
        a2 = 4.4955068
    }

All the best
Sebastian



On Thu, 2022-03-10 at 20:04 +0000, Jerry Tanoury wrote:
> Dear Forum,
> I am trying to optimize the geometry of the structure in the attached
> xyz file.  I have also attached my dftb_in.hsd input file.  The
> structure has taken from a pre-optimization done with GFN2 using
> Grimme’s XTB software. The problem is that the Mg and Cl atoms crash
> into each other.  The charges for Mg and Cl seem wrong.  Is there
> another set of parameters I should use for this job?
>
> Thank you for the help,
> Jerry
>
> Gerald J. Tanoury, Ph.D.
> Principal Scientific Fellow
> Vertex Pharmaceuticals
> 50 Northern Avenue
> Boston, MA 02210
> Office: 617-341-6219
> Cell: 508-740-4986
> jerry_tanoury at vrtx.com
> www.vrtx.com
>
>
>
> This email message and any attachments are confidential and intended
> for use by the addressee(s) only. If you are not the intended
> recipient, please notify me immediately by replying to this message,
> and destroy all copies of this message and any attachments. Thank you.
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at mailman.zfn.uni-bremen.de
> https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-u
> ser




This email message and any attachments are confidential and intended for use by the addressee(s) only. If you are not the intended recipient, please notify me immediately by replying to this message, and destroy all copies of this message and any attachments. Thank you.