[Wien] DOS in a single k-pointitrary

[Oleg Artamonov]

Dear Gerhard,

Thank you for detailed reply. I'll try to explain my position.
I would like to service my angle resolved e,2e experiment with
the inelastic scattering of polarized electrons from W(110).
We assume that the experimental spin-asymmetry is related to
the spin-asymmetry of the valence states. For this reason
I need to get the spin-asymmetry distribution of the valence
electrons in the energy interval  about 1 eV near the Fermi level
along the specified direction in the surface BZ.
I am running now the spin-resolved calculations with the orbital
potential as advised by P. Novak to resolve spin-states.
I can compose a script to describe the line in the surface BZ,
but I need to get an idea how to calculate the DOS for the arbitrary
k-point.

Thank you in advance,
                                        Oleg Artamonov.

----- Original Message ----- 
From: "Gerhard Fecher" <fecher at uni-mainz.de>
To: "A Mailing list for WIEN2k users" <wien at zeus.theochem.tuwien.ac.at>
Sent: Friday, December 12, 2008 11:42 AM
Subject: Re: [Wien] DOS in a single k-pointitrary


> What you are looking for is most probably the Bloch spectral function.
> Usually this is calculated easily from the Greens function, but Wien2k is 
> not a Green's function method as for example KKR, LMTO, or ASW are
> therfore I have no idea how to calculate the Bloch spectral function in 
> Wien2k.
>
> In an ordered lattice, however, the Bloch spectral function is simply a 
> sum of delta-functions (equation in my last mail) and just
> reveals the bandstructure. That means, the spaghetti plot of Wien2k tells 
> you already where the Bloch spectral function is non-zero.
> If you like to have any other point or line, you simply need to change the 
> input file case.klist_band. This you may do either by hand or by using
> XCrysden to be sure where you are in the Brillouin zone.
> Anyway, I do not see a real reason what the knowledge about states in 
> points or on lines with arbitrary symmetry should tell,
> at least if you do not do an angular resolved photoemission experiment, 
> and even there the knowledge about the states itself does not tell you
> about the matrix elements between initial and final states. As physics 
> does not jump, the states on a line close to a high symmetry line
> should be close, and inspecting the band structure in the high symmetry 
> directions gives you the idea how the spaghettis should look like
> if you are slightly off a certain direction. If you find singular points 
> then something is completely wrong, anyway.
>
> The Bloch spectral function is much more interesting for statistically 
> disordered systems.
> In that case you do not have a band structure anymore as the (infinite) 
> periodicity of youre lattice is lost.
> Roughly spoken, in such systems the spaghettis will be smeared about each 
> point in k and E.
> For Wien2k, there is no need to think further about that situation, as it 
> does not allow to calculate for statistically disordered systems.
>
> Indeed, one should remark that the use of supercells to model 
> statistically disordered systems is very dangerous,
> as the artificially brought in symmetry may result in band gaps that are 
> not existent in real live as they become completely
> smeared over by the statistical disorder.
>
> Ciao
> Gerhard
>
> ====================================
> Dr. Gerhard H. Fecher
> Institut of Inorganic and Analytical Chemistry
> Johannes Gutenberg - University
> 55099 Mainz
> ________________________________________
> Von: wien-bounces at zeus.theochem.tuwien.ac.at 
> [wien-bounces at zeus.theochem.tuwien.ac.at] im Auftrag von Oleg Artamonov 
> [arto at mail.nnz.ru]
> Gesendet: Donnerstag, 11. Dezember 2008 19:27
> An: A Mailing list for WIEN2k users
> Betreff: Re: [Wien] DOS in a single k-pointitrary
>
> Dear  Gerhard,
>
> Yes, You are right and a lot of states should be empty.
> I consider the DOS in the arbitrary k-point as a number of occupied
> states in a selected energy range and in the chosen volume of the
> k-space around the k-point. This value depends on the dispersion of
> states. Please, give me a hint how it is possible to calculate this
> value in Wien2k.
>                Thanks,
>                        Oleg Artamonov.
>
>
> On Thu, 2008-12-11 at 18:04 +0100, Gerhard Fecher wrote:
>> rho(E) = 1/N_k sum delta( epsilon_i,k - E)
>> delta( epsilon_i,k - E) is 1 if you hit the state on the spaghetti and 0 
>> if you do not hit it with youre k-point
>> that means for most of the arbitrary k-points at arbitrary energies the 
>> density of states is zero.
>>
>> Ciao
>> Gerhard
>>
>> ====================================
>> Dr. Gerhard H. Fecher
>> Institut of Inorganic and Analytical Chemistry
>> Johannes Gutenberg - University
>> 55099 Mainz
>> ________________________________________
>> Von: wien-bounces at zeus.theochem.tuwien.ac.at 
>> [wien-bounces at zeus.theochem.tuwien.ac.at] im Auftrag von Oleg Artamonov 
>> [arto at mail.nnz.ru]
>> Gesendet: Donnerstag, 11. Dezember 2008 17:34
>> An: A Mailing list for WIEN2k users
>> Betreff: [Wien] DOS in a single k-pointitrary
>>
>> Dear Wien2k users,
>>
>> Perhaps, somebody has calculated the DOS in an arbitrary single
>> k-point or the DOS distribution along a line in the BZ.
>> How is it possible to do that in Wien2k?
>>
>> Thanks in advance,
>>                 Oleg Artamononov.
>>
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>