ScF is a convergent trend, but rebound in Ni(111)Optimization.

167 views
Skip to first unread message

蜉蝣

unread,
Oct 31, 2024, 12:46:32 AM10/31/24
to cp...@googlegroups.com
Dear cp2k Experts,

i wan to Optimation Ni111slab,It is used for amino adsorption energy test,144 atoms.The nickel magnetic moment is 1.The optional multiplicity is 145.Open UKS、Smear.use RevPbe+D3(bj),CUTOFF 650,REL_CUTOFF 55,ALPHA 0.4,NBROYDEN 12.the Scf have a convergent trend, but rebound.I'm using the cp2k 2024.1 version.It is not known in terms of which parameters need to be modified.Hope to get some suggestions to make the calculation go smoothly.

Thanks,everyone
A new cp2k user.
23Ni111.inp
slurm-74628887.out

Marcella Iannuzzi

unread,
Oct 31, 2024, 3:30:34 AM10/31/24
to cp2k
Hi ..

Have you already tried to increase the maximum number of SCF iterations?
Regards
Marcella

冷对万夫

unread,
Oct 31, 2024, 8:54:18 AM10/31/24
to cp...@googlegroups.com
Dear Marcella Iannuzzi

Due to my email being restricted, I registered for Google Mail.I thought the email was sent, but it was blocked.
My problem is that optimizing SCF with nickel 111 does not converge. You asked me if I had increased the step size.
First,thanks for your letter. I have tried to set the SCF to 128 steps, and I have also tried to change ALPHA to 0.05 and temperature to 3000k in smear, but it is still in the current situation.


Marcella Iannuzzi

unread,
Nov 1, 2024, 9:18:19 AM11/1/24
to cp2k
Maybe 128 steps is still not enough.
Anyway, 3 Ni layers are few and I wouldn't be surprised if this model cannot reproduce the correct electronic properties. 

Regards
Marcella

冷对万夫

unread,
Nov 2, 2024, 2:04:44 AM11/2/24
to cp...@googlegroups.com
Thank you again and wish you a happy life. I will try increasing the number of atomic layers and SCF step size. We may need your help in the future.
--
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@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/13c075e8-9f85-407c-af73-d1d96ac02ec5n%40googlegroups.com.

冷对万夫

unread,
Nov 4, 2024, 12:09:14 AM11/4/24
to cp...@googlegroups.com
Dear Marcella
After taking your suggestion, I changed the crystal plane to five layers and then increased the SCF step size as well. But the situation hasn't improved yet, here are the input files and output files. Please take your free time to help see if there are any other revisions that need to be made.

Thanks.
slurm-74876436.out
x23Ni111.inp
Reply all
Reply to author
Forward
0 new messages