I use SIESTA regularly so I'll tell you the way I do it. You can use sisl
as @nickpapior proposed, so I will give you a different alternative. Using the program gdis
, with it can open multiple molecular coordinate formats, including xyz, mol, pdb, etc. Then, in the top left corner, you have the option of saving the molecule in SIESTA format. This is nice when you have a molecule with many atoms. Then what I do is to delete all the presets gdis puts in the fdf file except for SystemLabel
, NumberOfAtoms
, NumberOfSpecies
, block ChemicalSpeciesLabel
, block AtomicCoordinatesAndAtomicSpecies
. Then at the end I add the line include defaults.fdf
. defaults.fdf
is a file with all the parameters I use. Usually, within the same project that file doesn't change much from simulation to simulation. That's not a 100 % automatic solution but is pretty close. If you are in Ubuntu or Debian-based system you can install it easily from the command line with apt
.
One little warning, if you are creating a .fdf
file from a .pdb
in that format molecules are enumerated, for example, C1, C2, C3, ... Then, before creating the fdf
you have to select all atoms of the same species and relabel them. That's easy to do inside the program in editor mode.
Inside the program, you can select and configure some settings for the fdf
but I still recommend you have your own defaults.