Dear Prof. Artacho

With all my respect, what you say holds true if you construct a energy
surface for  the nuclei movement E_KS(R), where R are generalized atomic
positions. But that is not strictly the case I was thinking about. I was
thinking in a given E_KS calculated for a given atomic position. In this
sense the quantity E_KS, in my opinion is not a potential strictly speaking,
but the sum of the electronic kinetic and potential energies for a fixed
 external potential.

>From what Andres say - if I'm not interpreting something very wrong - it
seems to me that the value E_KS itself, for a given atomic configuration,
could be interpreted as a potential. Although that is absolutely true if the
energy surface for the atomic movements is constructed - as you well
mentioned -  in my opinion this concept cannot be easily extrapolated for a
single point energy, because E_KS is not the potential it self, but a
discrete points in the nuclear potential function ( what I called E_KS(R)
above ).

That is what I meant. Sorry if it seemed rough.

With my best regards

Ney



On Fri, Feb 18, 2011 at 6:08 PM, Emilio Artacho <[email protected]> wrote:

> Ney
>
> E_KS is the total energy of the electron system for each nuclear snapshot.
> The total energy (pot+kin) of
> the electrons represents the effective potential energy for the nuclear
> dynamics, in the adiabatic
> decomposition of both dynamics. It is thus the relevant potential energy
> surface for the nuclear
> dynamics, hence the name (I don't think it is misleading).
>
> (the springs etc of the mentioned examples are model descriptions of how
> this total energy of
> the electrons would change with nuclear positions:
> the empirically parametrised atomic *potentials* in
> empirical MD)
>
> It is Born-Oppenheimer, as Andres was saying.
>
> Emilio
>
>
>
> On Feb 18 2011, N H wrote:
>
>  In molecular dynamics all chemical bonds are represents - I mean
>> represented - by springs ... so it is obvoious that what dou you have in the
>> and is a potential energy! But that is not true if if you are comparing
>> quantum levels ... it doesn't matter if HF, CI or KS ones - because in this
>> energies what you have are components of the electronic kinetic energy plus
>> all the potential terms. And by potentiual terms I mean the electrostatic
>> interactions of the electrosn with their couter parts in the system plus the
>> interactions with the external field!
>>
>> In this sense ... calling the E_KS energy as the "potential energy" is
>> simply misleading!
>>
>> Cheers
>>
>> Ney
>>
>
>
> --
> Emilio Artacho
>
> Department of Earth Sciences, University of Cambridge
> Downing Street, Cambridge CB2 3EQ, UK
> Tel. (+44/0) 1223 333480, Fax  (+44/0) 1223 333450
> [email protected], http://www.esc.cam.ac.uk/~emilio
>
>

Responder a