I am trying a very simple problem with SQS and failing. I have two atom types and they can mix on the two B2 structure sites. So I have
cat lat.in
2.6 2.6 2.6 90 90 90
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
0.000000000 0.000000000 0.000000000 A,B
0.500000000 0.500000000 0.500000000 A,B
cat str.in
2.6 2.6 2.6 90 90 90
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
0.000000000 0.000000000 0.000000000 A
0.500000000 0.500000000 0.500000000 B
Then
corrdump -s=str.in -l=lat.in -2=4 -3=4 -4=4
getclus
getclus
0 0.00000 1
1 0.00000 2
2 2.25167 8
2 2.60000 6
2 3.67696 12
3 2.60000 24
3 3.67696 24
3 3.67696 24
3 3.67696 16
4 2.60000 12
4 3.67696 12
4 3.67696 48
4 3.67696 6
4 3.67696 16
4 3.67696 16
4 3.67696 4
So now I try to do the mcsqs:
cat sqscell.out
1
4.0 0.0 0.0
0.0 4.0 0.0
0.0 0.0 4.0
mcsqs -n=128 -rc
Different occupations assigned to symmetrically equivalent site:
0.000000 0.000000 0.000000
and
0.500000 0.500000 0.500000
Aborting.
But this is crazy. The two sites in B2 are NOT equivalent.
I have tried different permutations and ways of doing this but no success.
Any help is appreciated!
Sincerely,
Ron Cohen
I left out a file:
cat rndstr.in
2.6 2.6 2.6 90 90 90
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
0.000000000 0.000000000 0.000000000 A=0.859375,B=0.140625
0.500000000 0.500000000 0.500000000 A=0.828125,B=0.171875
This problem continues. If anyone can help I would appreciate it! B2 should work! I started with a clean directory and a clean source, recompiled, and ran
orrdump -l=rndstr.in -ro -noe -nop -clus -2=3.7 -3=3.7 -4=4.2 -5=4.2 -6=4.2
Warning: unit cell is not primitive.
rcohen@tomcat:~/ATAT/SQS/FeSiB2$ getclus
2 2.25167 8
2 2.60000 6
2 3.67696 12
3 2.60000 24
3 3.67696 24
3 3.67696 24
3 3.67696 16
4 2.60000 12
4 3.67696 12
4 3.67696 48
4 3.67696 6
4 3.67696 16
4 3.67696 16
4 3.67696 4
5 3.67696 24
5 3.67696 12
5 3.67696 16
5 3.67696 4
6 3.67696 6
rcohen@tomcat:~/ATAT/SQS/FeSiB2$ less rndstr.in
rcohen@tomcat:~/ATAT/SQS/FeSiB2$ mcsqs -n=128 -rc
Different occupations assigned to symmetrically equivalent site:
0.000000 0.000000 0.000000
and
0.500000 0.500000 0.500000
Aborting.
The input is
at rndstr.in
2.6 2.6 2.6 90 90 90
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
0.000000000 0.000000000 0.000000000 Fe=0.859375,Si=0.140625
0.500000000 0.500000000 0.500000000 Fe=0.828125,Si=0.171875
OK-RTFM! I revised the rndstr.in to
cat rndstr.in
2.6 2.6 2.6 90 90 90
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
0.000000000 0.000000000 0.000000000 Fe_a=0.859375,Si_a=0.140625
0.500000000 0.500000000 0.500000000 Fe_b=0.828125,Si_b=0.171875
And now it works! 
I don’t see how to close–but this can be closed.