[DFTB-Plus-User] Fwd: SpinPolarised with the xTB Hamiltonian

[Reinaldo Pis Diez]

Dear folks,

I'm sending this mail regarding how to make spin polarized calculations
within dftb+ with the xTB hamiltonian cause I believe it's lost in the
ether.

Regards,

Reinaldo

---------- Forwarded message ---------
De: Reinaldo Pis Diez <reinaldo.pisdiez at gmail.com>
Date: dom, 10 de jul. de 2022 19:08
Subject: SpinPolarised with the xTB Hamiltonian
To: <dftb-plus-user at mailman.zfn.uni-bremen.de>


Dear folks,

I'm using dftb+, rel 22.1 (from the binaries, not compiled by myself) to
see the behavior of xTB in transition metal systems with unpaired
electrons.

I'm a bit confused about how to specify the number of unpaired electrons
when the xTB method is used. I tried 'SpinPolarisation = Colinear' and the
number of unpaired electrons inside that block, but I got an error message
stating  'Missing child: SpinConstants'.

When I add a (fictitious) SpinConstants block, both with and without
ShellResolvedSpin, the same message as above is delivered by the program.
It seems that the block is being ignored but, at the same time, is needed
by the program.

In summary, can I do xTB calculations on magnetic systems using dftb+? And
most important, if yes, is it possible to set RelaxTotalSpin to Yes?

Thanks in advance for your help. Regards,

Reinaldo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20220725/9b857463/attachment.htm>