Structural phase transitions using GGA + U

Question on input files/tags, interpreting output, etc.

Please check whether the answer to your question is given in the VASP online manual or has been discussed in this forum previously!

Moderators: Global Moderator, Moderator

Post Reply
Message
Author
shihongting

Structural phase transitions using GGA + U

#1 Post by shihongting » Fri Feb 15, 2008 11:15 am

I'm investigating the structural phase transitions of Nd under pressures using the GGA+U method. However, I can't get

reasonable results.
For example, experimentally, Nd is the dhcp (double hexagonal close-packed) structure at ambient conditions, and at a

pressure around 4 GPa a structural phase transition to the fcc structure is observed.
However, according to my calculations, under low pressures the energies of the fcc are less than those of the dhcp for

the same cell volumes, that means the fcc structure is energetically more favorable than the dhcp structure.
I have tested many cases of J and U setting, and can not get ideal results.
I tested the following cases:
Dudarev's scheme (LDAUTYPE = 2): U =1, 2, 3, 4, 5, 6 eV
Liechtenstein's scheme (LDAUTYPE = 1):
J=1 --> U=2,3,4,5 eV
J=2 --> U=3,4,5,6 eV
All of above cases give the results that the fcc is more stable than the dhcp under low pressures (around -1 ~ 1 GPa).

==============================================================
System = Nd: P63/mmc dhcp

GGA = 91

ISMEAR = 1
SIGMA = 0.2

ICHARG = 1
ISPIN = 2
MAGMON = 4*4

VOSKOWM = 1
AMIX = 0.2
BMIX = 0.0001
AMIX-MAG = 0.8
BMIX-MAG = 0.0001

NSW = 199
IBRION = 2
ISIF = 4

EDIFF = 0.000001
EDIFFG = -0.002
PREC = HIGH

LDAU = .TRUE.
LDAUTYPE = 1
LDAUL = 3
LDAUU = 5
LDAUJ = 1
LMAXMIX = 6

NBANDS = 60
=================================================
I attached the INCAR.
Looking forward to get helps !!!
Last edited by shihongting on Fri Feb 15, 2008 11:15 am, edited 1 time in total.

Post Reply