[Wien] formation energy

[Stefaan Cottenier]

> I am interested in the formation energy. I've searched the mailing
> address, but I still can't get it. I found the definition of formation
> energy is like this (Ga15MnN16 for example)  :
>
> formation energy = total ENE of Ga15MnN16 -15*total ENE for Ga metal in
> standard state structure - 1* total ENE for Mn metal in standard state
> structure - 16* total ENE for N in standard structure

Correct. But more precisely: 'minus 15 * [total energy per atom] for Ga 
metal" etc. Make sure that the number of atoms of each element at the 
left and at the righ of the equation is identical. If not, the energy 
differences will not make sense.

> In our system, it's TbOCl, does that mean:
>
> formation energy = = total ENE of TbOCl - tot ENE for Tb metal in
> standard state structure - tot ENE for O in standard state - Cl in
> standard state. ?
>
> If that is correct, what is the standard state structure for O and Cl?
> Do we need to do several other calculations for Tb metal or O standard
> state to get the formation energy?

You do need to do additional calculations for all these reference 
materials, indeed. In order to avoid having to perform the entire 
structural optimization yourself, you can go here:

http://dx.doi.org/10.1080/10408436.2013.772503

Go to supplemental material. The file CIFs.tar.gz contains cif files 
with the geometry of all the crystals you need. The lattice parameters 
in these cif files are not necessarily exactly the optimized ones (with 
PBE), but close to. If you start the optimization process from there 
(with the functional that is relevant for you), you'll save some time.

Stefaan