POSCAR file for an alloy

[hpaudel]

Hello VASP users,

I am calculating the band structure for the alloy of Pb, Sn and Te. At zero doping level we have PbTe which has rock salt structure (fcc). Following is the POSCAR file for the bulk PbTe.

SYSTEM = PbTe
1.00000000000000
0.00000000 3.28500000 3.28550000
3.28500000 0.00000000 3.28500000
3.28500000 3.28500000 0.00000000
1 1
Cartesian
0.00000000 0.00000000 0.00000000 (Pb)
3.28500000 3.28500000 3.28500000 (Te)


Now I want to dope by Sn atoms (20% doping). My question is how we can include third type of atoms (Sn atoms) here? Shall I need to increase the unit cell size?

Thank you so much.

[Danny]

your unitcell contains 2 atoms...you want a doping concentration of 20%, which means 1/5, so you need at least 5 atoms of the species you want to replace...in addition to this you also need to think of the crystal structure of your alloy, will it be random substitution or nice crystal structure substitution?
In short, a unit cell with 2 atoms is to small, and the size of your supercell depends on a set of factors.

Danny

[hpaudel]

Danny,

What I understand from you now is that I need to increase the size of unit cell so that I can include more atoms. For 20% doping here I have to have 10 atoms in unit cell and replace 2 of those by new species (Sn) that gives me 1/5 doping. Is that right ?

Previous calculation shows that the alloy with less than 30% doping by Sn atoms has still fcc structure. Random substitution would be ok but probably it would take time to figure out the stable structure at given concentration.

Thank you

hpaudel

[Danny]

Yes...now the big question will be: what will you use as supercell and how do you position your substituents. is it only Pb or Te which are substituted or both? at what ratio? do the substituents prefer to cluster etc...

Danny