Dipole Correction in Slab Supercell
Posted: Mon Dec 14, 2009 7:04 pm
Hello all,
I am attempting to determine the formation energy for a variety of defects of varying charge states on the surface of a semiconductor. I initially began my calculations by selecting an 8-layer slab and fixing the bottom 3 layers to the bulk lattice constants. I soon realized that I had a significant dipole that arose between the relaxed and unrelaxed side of my slab. I have begun to use the dipole correction flags in VASP however I read that for charged calculations, these tags can only be used to cubic super cells. My supercell is orthorhombic. So two questions:
1. Is there a work around to use the dipole corrections in VASP to treat orthorhombic cells (I suspect not.......)
2. If I decide to do a full relaxation of all the layers, will there be a dipole arising if I put a defect on only one surface or do I have to put one on the ther surface??
Thanks in advance
I am attempting to determine the formation energy for a variety of defects of varying charge states on the surface of a semiconductor. I initially began my calculations by selecting an 8-layer slab and fixing the bottom 3 layers to the bulk lattice constants. I soon realized that I had a significant dipole that arose between the relaxed and unrelaxed side of my slab. I have begun to use the dipole correction flags in VASP however I read that for charged calculations, these tags can only be used to cubic super cells. My supercell is orthorhombic. So two questions:
1. Is there a work around to use the dipole corrections in VASP to treat orthorhombic cells (I suspect not.......)
2. If I decide to do a full relaxation of all the layers, will there be a dipole arising if I put a defect on only one surface or do I have to put one on the ther surface??
Thanks in advance