On Tue, 2 Dec 2003, Sergey Lisenkov wrote: | DEar Prof. Soler, | | Thank you for your reply. Bu I cannot understand: Does SIESTA calculate the | real total energy? In most paper I see that authors compare total energy per | atom (for example, for carbon) with experimentasl results. How did they | obtain it? I checked it dor C-diamond, in the file carbpn.MDE the total | energy/per atom is more bigger than experimental results. | | Sergey |
Dear Sergey, let's define what we speak about. The REAL total energy (i.e., reliable absolute values, like e.g. -1.0000 Ry for hydrogen atom) is almost never available in DFT calculations, at least not with satisfactory accuracy. For one thing, the absolute numbers depend too much on implementation - limitations of the basis size, details of integrations etc. - and moreover there are basic limitations of DFT (spurious self interaction, that's why you'll never get -1 Ry for hydrogen; intraatomic correlations). One discusses absolute numbers in many-body calculations like configuration interaction, but then, for really pushing to the limits, you are technically restricted to one or few atoms. The absolute total energy in Siesta are "wrong" even by the order of magnitude, because the code uses pseudopotentials, so the (large) core contribution to the total energy is not present. However, this is not so bad, because if you are after ABSOLUTE total energy of the carbon atom SIESTA won't be the code of your choice. The RELATIVE properties like cohesion energies, defect formation energies etc. usually come about reasonably in density functional calculations, and also (with a bit of luck) with SIESTA. That's because these values are difference between two total energies, and the major part of systematic error cancels (including the missing core contribution). Nevertheless, you have to be careful to check that these difference converge depending on your basis size, energy cutoff etc. Hopefully this helps, Andrei Postnikov +-- Dr. habil. Andrei Postnikov ----- Tel. +49-541-969.2377 -- Fax .2351 ---+ | Universitaet Osnabrueck - Fachbereich Physik, D-49069 Osnabrueck, Germany | +-- [EMAIL PROTECTED] --------- http://www.home.uni-osnabrueck.de/apostnik/ --+
