Segmentation Fault (mcsqs)

Alloy Theoretic Automated Toolkit (ATAT)
1

Hi Axel,

I’m encountering segmentation fault when trying to generate SQS with mcsqs for a spinel structure.
Here is my lat.in


10.124 0.000000 0.000000
0.000000 10.124 0.000000
0.000000 0.000000 10.124
0.000000 0.500000 0.500000
0.500000 0.000000 0.500000
0.500000 0.500000 0.000000
0.500000 0.000000 0.000000 Cr=0.5,V=0.5
0.000000 0.500000 0.000000 Cr=0.5,V=0.5
0.000000 0.000000 0.500000 Cr=0.5,V=0.5
0.000000 0.000000 0.000000 Cr=0.5,V=0.5
0.625000 0.625000 0.625000 Zn
0.375000 0.375000 0.375000 Zn
0.221500 0.759600 0.759600 S
0.759600 0.221500 0.759600 S
0.759600 0.759600 0.221500 S
0.759600 0.759600 0.759600 S
0.778500 0.240400 0.240400 S
0.240400 0.778500 0.240400 S
0.240400 0.240400 0.778500 S
0.240400 0.240400 0.240400 S

I generated clusters with pairs ranging upto 6A as following:

On running mcsqs on 28 (or 56 atom) supercell, it throws a segmentation fault: 11

Any insights into what might be causing the segmentation fault would be helpful (it is definitely not memory-related issue). May be I’m missing something trivial.

2

The issue is that, once you take into account the translation vectors, your 4 randomly occupied sites overlap pairwise to form only 2 distinct sites per cell. You can verify this by running

cellcvrt.exe -rr < lat.in

where -rr stands for "remove redundant".
I am sorry that the error message was not too helpful in identifying the problem!