Thanks for your attention.
I want to calculate a slab with external electric field. For small external electric field like 0.1 ev/A, I can achieve a converged total energy with the expense of more than 80 electonic steps. unfortunately, as soon as I increase the external field to 0.5 ev/A, the calculation shows no sign of convergence and its total energy is weird, like below,
RMM: 37 -0.352981988366E+06 0.80317E+03 -0.30545E+05 9483 0.148E+04 0.258E+02
RMM: 38 -0.341332444180E+06 0.11650E+05 -0.12314E+05 9272 0.148E+04 0.253E+02
RMM: 39 -0.328795963487E+06 0.12536E+05 -0.16718E+05 9301 0.147E+04 0.252E+02
RMM: 40 -0.353055503244E+06 -0.24260E+05 -0.28073E+05 9302 0.156E+04 0.256E+02
I did lots of trial calculations. as suggested by admin or others, but shows no improvement of electronic convergence. something imappropriate with my system?
Could anyone give me a suggestion on how to handle this problem? any advices are extremely welcome!
Thanks very much!
Electronic Convergence Problem!
Moderators: Global Moderator, Moderator
-
zhongchangwang
- Newbie
- Posts: 4
- Joined: Sat Jul 22, 2006 12:07 pm
- License Nr.: Unknown
- Location: Department of Materials Engineering, The University Of Tokyo
Electronic Convergence Problem!
Last edited by zhongchangwang on Sat Jan 20, 2007 9:46 am, edited 1 time in total.
-
admin
- Administrator
- Posts: 2921
- Joined: Tue Aug 03, 2004 8:18 am
- License Nr.: 458
Electronic Convergence Problem!
did you switch on the dipole corrections?
Last edited by admin on Mon Jan 22, 2007 12:40 pm, edited 1 time in total.
-
zhongchangwang
- Newbie
- Posts: 4
- Joined: Sat Jul 22, 2006 12:07 pm
- License Nr.: Unknown
- Location: Department of Materials Engineering, The University Of Tokyo
Electronic Convergence Problem!
Thanks admin for response!
I added three commands to the normal INCAR, see the following:
ENCUT=350.000 #cutoff
EDIFF=1.0e-05
ISIF=0
ISMEAR= 0 #BZ integration method (for relaxation runs)
LWAVE=.FALSE.
LCHARG =.FALSE.
LBERRY = .FALSE.
ISYM= 0
ALGO = Fast
LREAL= Auto
EFIELD= 0.5
IDIPOL= 3
LDIPOL= .TRUE.
NELM= 200
LMAXMIX = 6
Since LDIPOL was set to True, I think maybe the dipole correction was switched on.
Maybe I am wrong, could you please check the INCAR for me. Thanks very much for your attention!
I added three commands to the normal INCAR, see the following:
ENCUT=350.000 #cutoff
EDIFF=1.0e-05
ISIF=0
ISMEAR= 0 #BZ integration method (for relaxation runs)
LWAVE=.FALSE.
LCHARG =.FALSE.
LBERRY = .FALSE.
ISYM= 0
ALGO = Fast
LREAL= Auto
EFIELD= 0.5
IDIPOL= 3
LDIPOL= .TRUE.
NELM= 200
LMAXMIX = 6
Since LDIPOL was set to True, I think maybe the dipole correction was switched on.
Maybe I am wrong, could you please check the INCAR for me. Thanks very much for your attention!
Last edited by zhongchangwang on Tue Jan 23, 2007 7:00 am, edited 1 time in total.
-
theseguys
Electronic Convergence Problem!
[quote="admin"]did you switch on the dipole corrections?[/quote]
dear admin, I really want to if it is necessary to set ISYM=0 when set EFIELD=non-zero. this problem puzzle me for half a year, and I have search this problem for many times. thank you!
dear admin, I really want to if it is necessary to set ISYM=0 when set EFIELD=non-zero. this problem puzzle me for half a year, and I have search this problem for many times. thank you!
Last edited by theseguys on Mon Jul 22, 2013 4:13 pm, edited 1 time in total.