<html>
<body>
Dear Wien2k users,<br><br>
There is a problem that made me confuse a lot. I need your help.<br><br>
I performed the LDA+U calculation for NiO as an example. The
structure file is as follow:<br><br>
<font size=2>
NiO
<br>
<b><i>F LATTICE,NONEQUIV.ATOMS:
2225_Fm-3m
<br>
MODE OF CALC=RELA unit=bohr<br>
7.927000 7.927000 7.927000 90.000000 90.000000
90.000000<br>
</font><font size=2 color="#FF0000">ATOM 1: X=0.00000000
Y=0.00000000 Z=0.00000000<br>
MULT=
1 ISPLIT= 2<br>
Ni NPT=
781 R0=0.00010000 RMT= 2.0000 Z:
28.0<br>
LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000<br>
<x-tab> </x-tab><x-tab>
</x-tab> 0.0000000
1.0000000 0.0000000<br>
<x-tab> </x-tab><x-tab>
</x-tab>
0.0000000 0.0000000 1.0000000<br>
</font><font size=2 color="#000080">ATOM 2: X=0.50000000
Y=0.00000000 Z=0.00000000<br>
MULT=
1 ISPLIT= 2<br>
O NPT= 781
R0=0.00010000 RMT= 2.0000 Z: 8.0<br>
LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000<br>
<x-tab> </x-tab><x-tab>
</x-tab> 0.0000000
1.0000000 0.0000000<br>
<x-tab> </x-tab>
0.0000000 0.0000000 1.0000000<br><br>
</font><font color="#000080">NiO.indm<br><br>
-9<br>
1<br>
1 1 2<br>
0 0<br><br>
</font><font color="#008000">NiO.inorb<br>
1 2 0<br>
PRATT 0.5<br>
1 1 2<br>
1<br>
0.52 0.00<br><br>
</i></b></font>By this set up, everything ran well. But I'm not sure that
it is right or wrong.<br><br>
But if I set up Ni as atom 2 and O as atom 1:<br><br>
<font size=2 color="#000080"><b><i>
Title
<br>
F LATTICE,NONEQUIV.ATOMS:
2225_Fm-3m
<br>
MODE OF CALC=RELA unit=bohr<br>
7.927000 7.927000 7.927000 90.000000 90.000000
90.000000<br>
ATOM 1: X=0.50000000 Y=0.00000000 Z=0.00000000<br>
MULT=
1 ISPLIT= 2<br>
O NPT=
781 R0=0.00010000 RMT= 1.6500
Z: 8.0<br>
LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000<br>
0.0000000 1.0000000 0.0000000<br>
0.0000000 0.0000000 1.0000000<br>
</font><font size=2 color="#FF0000">ATOM 2: X=0.00000000
Y=0.00000000 Z=0.00000000<br>
MULT=
1 ISPLIT= 2<br>
Ni NPT= 781
R0=0.00010000 RMT= 2.0000 Z: 28.0<br>
LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000<br>
0.0000000 1.0000000 0.0000000<br>
0.0000000 0.0000000 1.0000000<br><br>
<br>
</font><font color="#000080">NiO.indm<br><br>
-9<br>
1<br>
2 1 2<br>
0 0<br><br>
</font><font color="#008000">NiO.inorb<br>
1 1 0<br>
PRATT 0.5<br>
2 1 2<br>
1<br>
0.52 0.00<br><br>
</i></b></font>I met the error<br><br>
<br>
<pre> cycle 1
<x-tab> </x-tab>(Tue Aug 30 17:16:33 SGT
2005) <x-tab> </x-tab>(60/40 to go)
> lapw0
-p<x-tab> </x-tab>(17:16:33) starting parallel
lapw0 at Tue Aug 30 17:16:33 SGT 2005
--------
running lapw0 in single mode
2.369u 0.117s 0:02.48
99.5%<x-tab> </x-tab>0+0k 0+0io 0pf+0w
> orb -up -p<x-tab> </x-tab>(17:16:35) 0.003u
0.002s 0:00.00 0.0%<x-tab> </x-tab>0+0k 0+0io 0pf+0w
> stop error
</pre> I really do not understand why it crashed in this case. Did I
set up the input files wrongly? <br><br>
Looking at the user guide page 80 and the lecture note by Novak the
LDA+U calculation was set up for two different Ni atoms with different
index 1 and 2. The question is: should we need two different Ni atoms for
LDA+U? <br><br>
Many thanks for help.<br><br>
Regards,<br>
Khuong<br>
</body>
</html>