[CP2K-user] [CP2K:21130] Initial wavefunction guess for MULTIPLE_UNIT_CELL run

[Konstantin Tokarev]

Dear all,

I've tried to use wfn file from single unit cell run as a starting guess in 
a multiple unit cell run, however it is rejected because of different 
number of atoms, e.g. in my attempt of using single cell wfn file for 
MULTIPLE_UNIT_CELL 2 11 CP2K printed

READ RESTART : WARNING : DIFFERENT natom, returning          160          80

So atomic guess has to be used. I wonder if there is a more efficient way 
to provide an initial guess wavefunction when "single cell" wavefunction is 
known. I suppose in this case it should be enough to translate wavefunction 
of one cell into atoms of its copy. Or, maybe even simpler, assuming that 
all stored wavefunction coefficients are given relative to coordinates of 
nuclei loaded from CELL/TOPOLOGY, just duplicate wavefunction of first 80 
atoms to make a wavefunction for 160 atoms.

Any ideas?

Another way could be to load restart wavefunction for the first 80 atoms 
and assume atomic guess for other 80, but obviously it would be less 
efficient, and would probably require modification of CP2K code.

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/55484388-445a-437f-b4a1-9c81ddcc3048n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20250209/df17c2e1/attachment-0001.htm>