Careful, what you said is not completely right, there is a basis although
it is more complete than a limited set of Gaussian functions.

The angular part (potential, wave function etc) is a truncated spherical
harmonic expansion. The radial part of the wavefunction in the spheres is a
rather complex expansion - best to read carefully the user guide & lectures
for exactly what Lapw, APW+lo are.

Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
Corrosion in 4D
Co-Editor, Acta Cryst A
"Research is to see what everybody else has seen, and to think what nobody
else has thought"
Albert Szent-Gyorgi
On Dec 29, 2015 21:49, "delamora" <> wrote:

> The point about the full potential is that the radial part inside the
> sphere is numerical, you have a net of points going from r=0 to Rmt, so no
> basis is needed as in the Gaussian program where the radial part is formed
> by gaussian functions.
> ------------------------------
> Tuan,
>         What that equation means is that you have a full basis for your
> potentials, functions, etc.
>         The plane waves outside muffin tin sphere is ok in a region where
> the potential does not vary too much, but it is hopeless near the atomic
> nuclei where the potential varies as 1/r, near a nucleus the atomic like
> functions work well with this kind of potential, so the problem is to match
> both functions both in value and in slope at the muffin tin sphere.
>         It is like the step function where you have two plane waves with
> different frequency and you have to match them at the step.
>         Saludos
>                     Pablo
> ------------------------------
> *De:* <
>> en nombre de Tuan Vu <
> *Enviado:* martes, 29 de diciembre de 2015 09:22 p. m.
> *Para:*
> *Asunto:* [Wien] Formula full-potential
> Dear Pro. F. Tran
> Reason do i need details of the potential:
> Time is coming for me to protect my PhD dissertation as part of the
> University-supported project.
> One of the key question that I will answer before the dissertation council
> is that how formula 2.10 in the userguide's.pdf is formed.
> The council includes professors from many universities and academies in
> Russia, so I can't avoid that questions because the reliability
> of the calculation package depends on the full-potential, and so does my
> work. With all of my respect, I hope that you can help me
> with this question about the full-potential.
> Rest regards
> PhD. Student Tuan Vu
Wien mailing list

Reply via email to