Mcsqs for ternary alloys

Alloy Theoretic Automated Toolkit (ATAT)
1

Hello,
I tried to find the SQS of a ternary alloy, one constituent of which is 3 at. (say A=60 at. , B= 37 at. and C=3 at. .)

So far the smallest number of atoms in the supercell (bestsqs) is 100.
Different parameters for corrdump have been tested (2=4, 5, 6 and 7).

Is there any way to find a smaller supercell (say with 80 or 90 atoms)?

Thanks

2

Keep in mind that the smallest possible SQS is the least common denominator of all the composition on a sublattice. The 37% and 3% really make it difficult to find a small SQS.
If you compositions are of the form n/90, then a 90-atom SQS is possible, assuming you have a single sublattice.