I strongly suggest that you research the basis for the Slater 1/2 method as
well as extensions. What Slater suggested is that an integral from the
initial state (no core hole) and the final state (complete core hole) could
be approximated by the midpoint (1/2 core hole). One can do better, albeit
it will take more calculations, for instance consider more fractional holes
(e.g. 1/4, 3/4) then integrate better.

No core hole assumes that your switch electron has passed before the system
can respond -- which is dubious because the temporal coherence (width along
the beam direction) is large, typically 100nm or so. A full core hole
assumes that the swift electron hangs around for a long time so sees more
of the final state, also not a very convincing argument for current
microscopes. (Let's ignore femtosecond EM.)

N.B., describing the swift electron correctly via mutual coherence
(effectively a density matrix) is not common. We had a go
in Ultramicroscopy 55 (1994) 165 at the spatial description, if you dig you
may find papers where the temporal part has been included. I have not fully
tracked the literature on this.

At least in standard models, the initial and final spin states should be
the same -- exchange coupling of a TEM electron and the solid is
essentially zero. (With SOC I am not sure exactly what should be done.) I
am not aware of calculations where this constraint has been enforced,
although there may be some. (I have checked this myself for some transition
metal oxides and it matters.) I very strongly suspect that in some of the
literature calculations the final spin-state differs from the original, so
the results have a buried incorrect approximation.

Thus, if you take 1/2 (or other) electrons out of the core then in my
opinion you need to ensure that the initial and final spin states (all
electrons) are the same, e.g. use runfsm. Unfortunately you might still
have a different local spin state which at least at present Wien2k cannot

On Sun, Nov 17, 2019 at 9:41 AM 丁一凡 <yfding0...@foxmail.com> wrote:

> Respected Prof. Marks,
> I remember 1/2 core hole calculations in this article "Partial core hole
> screening in the Cu L-3 edge" (DOI:10.1007/s100510170179). When calculating
> my systems, I only used a full core hole and supercell. I will follow your
> suggestion and try it. In the previous calculation, I didn't notice the
> relation between spin and core hole. Please allow me to ask a question
> here, why is runfsm the best method ?
> _______________________________________________
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
> https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien&d=DwICAg&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=IaYA9EMobH_kt7vz-PIZQe9kfpg3FT2--xUJQ2Ab52I&s=VAJEsdWNwMXjnX7a6vqaWdmGZZ0LxAyqLZiCi-twZKM&e=
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html&d=DwICAg&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=IaYA9EMobH_kt7vz-PIZQe9kfpg3FT2--xUJQ2Ab52I&s=bIndfEmmbUY8Vn29vIBEfrrBcS1ufsDySR2edvgq9uc&e=

Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
Corrosion in 4D: www.numis.northwestern.edu/MURI
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
Wien mailing list

Reply via email to