total energy cannot converge

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
andreichibisov
Posts: 17
Joined: Thu Jan 21, 2010 2:03 am

total energy cannot converge

Post by andreichibisov » Sat Apr 09, 2011 3:06 pm

Hello all,
I want to calculate total energy of the free Ti atom in order to compute of the cluster binding energy.
However, the total energy cannot converge! Can you help me?
My input file:
ndtset 5
acell 20 20 20
prtgeo 1
enunit 1
# ionmov 3
# tolmxf 1.0d-4
diemac 1.6d0
diemix 0.5
npulayit 25
# diemac 5.0d0
iprcel 144
diecut 0.8
# nnsclo 1
# nline 20
# ortalg 4
# occopt 3
# tsmear 1.0d-3
ecut1 30
ecut2 40
ecut3 50
ecut4 60
ecut5 70
kpt 0.0 0.0 0.0
natom 1
# nband 10
nkpt 1
nsppol 2
nstep 3000
ntime 3000
nsym 1
ntypat 1
rprim 1 0 0
0 1 0
0 0 1

toldfe 1.d-5
typat 1
wtk 1
xred 0.0 0.0 0.0
znucl 22

# This line added when defaults were changed (v5.3) to keep the previous, old behaviour
iscf 7
Attachments
Ti.in
(582 Bytes) Downloaded 245 times

Locked