higher precision on EIGENVAL values and spin-orbit splitting
Posted: Mon Jul 27, 2009 5:08 am
Dear all,
Have any one of you modified the source code to change the precision of EIGENVAL values (currently they have only 4 digits after decimal point)? I looked up the source code and found quite a few places where EIGENVAL was defined. Do I need to change all of them if I intend to increase the precision of the eigen values?
The reason I want to increase the precision is that I saw some literature that the spin-orbit splitting of graphene was calculated and I wanted to reproduce those calculations. But the spin-orbit splitting of graphene is very small (about micron eV level). I suppose the spin-orbit splitting is read from the bandstructure (EIGENVAL values). So if my assumption is correct, the EIGENVAL values must have higher precision.
Any suggestion/clarification will be highly appreciated.
Cosmos
Have any one of you modified the source code to change the precision of EIGENVAL values (currently they have only 4 digits after decimal point)? I looked up the source code and found quite a few places where EIGENVAL was defined. Do I need to change all of them if I intend to increase the precision of the eigen values?
The reason I want to increase the precision is that I saw some literature that the spin-orbit splitting of graphene was calculated and I wanted to reproduce those calculations. But the spin-orbit splitting of graphene is very small (about micron eV level). I suppose the spin-orbit splitting is read from the bandstructure (EIGENVAL values). So if my assumption is correct, the EIGENVAL values must have higher precision.
Any suggestion/clarification will be highly appreciated.
Cosmos