H+ has no electrons, so the electronic energy is 0. I don't know to what extent
this is "the right zero" if you have a pseudopotential rather than a
point-charge nucleus.
Even if this was not the case (i.e. if you have a larger ion), one has to be
careful with charged unit cells. SIESTA (and any other periodic code I know)
will add a background charge, because otherwise you have an infinitely charged
system. I think the background charge changes the Hamiltonian sufficiently so
that the intuitive calculation of the adsorption energy does not work (you need
to take differences between energies of charged and uncharged cells).
Experimentally, you never have free protons. Depending on where your H+ comes
from, you could look at the reaction
G + H3O+ -> GH + H2O
If you place your H3O+ or H2O far above the surface, you can do each side of the
reaction in one calculation, and you always have the same charge.
There must also be a counter ion somewhere...
HTH,
Herbert
On 03/06/15 13:44, James Lawlor wrote:
Hi,
I'm trying to do calculate the binding energy of H+ with graphene, so this
involves finding the total energies of 3 systems - H+ isolated, graphene
isolated, and the combined system.
My current method is to use "NetCharge 1.0" in the input files of the isolated
H+ and the combined systems, which should in theory remove an electron from the
system. The problem is this returns errors for the H+ as the system is
essentially a proton and I think this is causing SIESTA to get confused.
Could anyone suggest a possible solution, or perhaps a different method?
Cheers,
James
--
James Lawlor
Theory & Modelling Group
School of Physics
Trinity College Dublin, Ireland
--
Herbert Fruchtl
Senior Scientific Computing Officer
School of Chemistry, School of Mathematics and Statistics
University of St Andrews
--
The University of St Andrews is a charity registered in Scotland:
No SC013532
