> Dear all, > > I want to do a test calculation on the local magnetic moment of iron on > the > interatomic distance. The cutoff radii for my pseudopotential generation > is > 2.0 Bohr. When the Fe-Fe distance is set to be 3.0 Bohr (this means that > one > atom is in the core region of the other atom), is the calculation > reliable?
Dear Leila, I remember to have posed such question long ago to one of Siesta fathers and the answer was positive, in a sense that there is no constraint technically forbidding overlap of pseudopotential cutoff radii at adjacent atoms, like say muffin-tins in FLAPW codes. (More specific comments from pseudopotential gurus will be mostly welcome). However, you should know what you are doing, because the structure of wave atomic function inside the cutoff radius indeed starts to differ, and the error increases as you come closer. Discussing specifically Fe, the overlap of 3d will probably not pose big problem, because their nodal structure is correct. For 4s, it is not, but they are delocalized anyway, and I don't think the details of their pseudopotential count too much. However, you should take care of 3p, including them in the valence as you come for closer distances. Moreover I think this is basically difficult to cover in Siesta, using the same fixed basis, a large range of varying interatomic distances with the same accuracy. That is, you should check against all-electron calculation whether the error is acceptable for your purposes. > If not, shall I reduce the cutoff radii of my input file for the > pseudopotential generation? This is good for transferability but results in harder pseudopotential and introduce other kind of problems. > Or, can I do all-electronic calculations with > SIESTA? No Best regards Andrei > > Thank you in advance! > > Best wishes! > > Yours sincerely, > > Leila >
