total energy of charged cluster

Message

jialy_25 · #1 Post by **jialy_25** » Fri Nov 10, 2006 8:38 am

I want to get the total energy of -1 charged clusters. In the INCAR, I specified:

ENCUT = 300 ! ~1.3 X ENMAX
ICHARGE = 2
ISMEAR = 0
SIGMA = 0.001
NELECT = 21 ! valence electron of neutral cluster = 20
IDIPOL = 4
POTIM - 0.1
LDIPOL = .TRUE.

Calculations were done at Gamma point only. With the LDIPOL = TRUE set, the total drift is significantly large. And the total energy seems not correct. Some are much larger, and some are much smaller. The supercell is large enough that there were no problem during simulation of neutral clusters.

What is the proper flags should I specified in INCAR to get the right calculation of negative-charged cluster?

Thank you.

#2 Post by **admin** » Wed Nov 15, 2006 1:12 pm

For a cluster, please keep at least one atom fixed in space in order to avoid drifts (total drifts in the force).
Please make sure that your supercell is cubic
I'm not sure what you mean by "the total energy seems not correct"

jialy_25 · #3 Post by **jialy_25** » Thu Nov 16, 2006 3:53 am

Thanks for your reply.

Actually i want to calculate the electron affinity(EA) of the clusters. But the total energy of the charged clusters could not give me the correct EA values (EA=Eneutral - Echarged). For example, the total energy for a neutral cluster is -21 eV. The total energy of -1 charged cluster becomes -0.6 eV after the energy correction for LDIPOL = TRUE. Obviously, this is not correct. Should I also specified DIPOL in my INCAR?