[CP2K-user] SCF not converged increasing cell parameters in non periodic calculations
Juliette Zito
juliet... at hotmail.fr
Tue Sep 8 09:00:40 UTC 2020
Hello everybody,
I'm checking the effect of increasing the cell parameters on the total
energy of a non periodic system with single point calculations. I started
from ABC [angstrom] 46.00 46.00 46.00, i.e. 10 angstroms more than the
diameter of the cubic nanocrystal I'm interested in and started increasing
the cell parameters of 2 angstroms, each time restarting from the previous
wavefunction.
However at ABC [angstrom] 54.00 54.00 54.00 I started having serious
problems with the SCF convergence.
----------------------------------- OT
---------------------------------------
Step Update method Time Convergence Total energy
Change
------------------------------------------------------------------------------
1 OT DIIS 0.80E-01 68.6 0.02074085 -61642.1809140733
-6.16E+04
2 OT SD 0.80E-01 60.7 0.02194616 -64401.2100479365
-2.76E+03
3 OT SD 0.80E-01 58.8 0.02282865 -67465.7024967550
-3.06E+03
4 OT SD 0.80E-01 58.6 0.02307889 -70736.3733263428
-3.27E+03
5 OT SD 0.80E-01 58.6 0.02244575 -74015.5498583123
-3.28E+03
6 OT SD 0.80E-01 60.3 0.02104404 -77055.6216821954
-3.04E+03
7 OT SD 0.80E-01 57.6 0.01938487 -79698.7653953436
-2.64E+03
8 OT SD 0.80E-01 61.8 0.01718284 -81913.3088630249
-2.21E+03
9 OT SD 0.80E-01 60.4 0.01352800 -83569.3974313459
-1.66E+03
10 OT DIIS 0.80E-01 61.7 0.00967401 -84543.4818645992
-9.74E+02
11 OT SD 0.80E-01 59.6 0.01690592 -83626.3970827348
9.17E+02
12 OT SD 0.80E-01 60.7 0.01064394 -85070.0655959053
-1.44E+03
13 OT SD 0.80E-01 60.7 0.00719136 -85653.2818077908
-5.83E+02
14 OT SD 0.80E-01 57.8 0.00554615 -85934.1614607915
-2.81E+02
15 OT DIIS 0.80E-01 54.5 0.00485774 -86111.4717549004
-1.77E+02
16 OT SD 0.80E-01 59.5 0.00514808 -86219.0950430464
-1.08E+02
17 OT SD 0.80E-01 58.2 0.00490076 -86376.9217575644
-1.58E+02
18 OT SD 0.80E-01 60.2 0.00507425 -86528.4409703378
-1.52E+02
19 OT SD 0.80E-01 59.8 0.00539092 -86693.3619213511
-1.65E+02
20 OT SD 0.80E-01 60.0 0.00582554 -86881.2313118624
-1.88E+02
21 OT SD 0.80E-01 60.4 0.00638478 -87102.2448740847
-2.21E+02
22 OT SD 0.80E-01 58.0 0.00708839 -87369.4705830055
-2.67E+02
23 OT SD 0.80E-01 60.6 0.00797404 -87700.9516573457
-3.31E+02
24 OT SD 0.80E-01 62.3 0.00911465 -88123.5568228465
-4.23E+02
25 OT SD 0.80E-01 61.9 0.01064681 -88681.2362322660
-5.58E+02
26 OT SD 0.80E-01 61.1 0.01279778 -89452.4383734747
-7.71E+02
27 OT SD 0.80E-01 58.3 0.01588892 -90583.8584958531
-1.13E+03
28 OT SD 0.80E-01 58.4 0.02029072 -92348.9056309448
-1.77E+03
29 OT SD 0.80E-01 59.7 0.02634650 -95231.4141703613
-2.88E+03
30 OT SD 0.80E-01 58.6 0.03432332 -100011.3988203991
-4.78E+03
31 OT SD 0.80E-01 61.0 0.04478366 -107748.6174081734
-7.74E+03
...
The same occurs for the same system with CHARGE 1 and MULTIPLICITY 2 but
also with other similar systems, always at the threshold value of 54.00
angstroms.
I attach here the input file. I'm using WAVELET as Poisson solver and the
version 6.1 of cp2k.
Is there some physical reason behind this problem? What can I do to reach
the convergence?
Best regards,
Juliette Zito
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200908/2c35f34f/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DOS.inp
Type: chemical/x-gamess-input
Size: 1200 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200908/2c35f34f/attachment.inp>
More information about the CP2K-user
mailing list