[Wien] I need theta(Gi-Gj) [the FFT of the step-function at the interstitial region] to calculate the overlap matrix which I mentioned in my previous email. M_mn interstitial(k1, k2)=<u_m, k1|u_n, k2>= sigma_ij{c*_m, k1(Gi) * theta(Gi-Gj) * c_n, k2(Gj)}

[Peter Blaha]

This will be an exception, since normally I will NOT answer to people, 
who post the same question 10 times or more to the mailing list AND my 
private email, while I'm 3 weeks abroad. (your supervisor "saved" you).

Apparently, you have looked into previous mails and yes, the required 
routines are in lapw0 and are connected with the names "reanX".

Once you know this, the rest should be fairly straight forward and 
although the code is not well documented, it has some explanations 
which should suffice to go on.

Check lapw0.F: Search eg. for "stepfunction" and it leads you to the 
right spot.

rean0.F  sets the stepfunction into a 3D array U(Gx,Gy,Gz) (in 
reciprocal space). It contains the besselfunction j_1, the sphere volume 
and some phase factors. (I'm sure this is described in D.Singh's book).

rean4.F takes an 3D array (FFT) (which should contain your 
"overlap-density"), it transforms both arrays (U and FFT) into real 
space by a FFT-procedure; multiplies the two arrays with each other and 
backtransforms to reciprocal space. The resulting FFT array (or the 1D 
array VKVXC(G) after getfft) contains the coefficients you are looking for.


fatemeh.mirjani schrieb:
> M is labeled by m,n (band index) and k1,k2 (k-points) and G is
>  the reciprocal lattice vector.
>  
> According to my search in mailing-list, I
>  think that is related to SRC_lapw0/rean files where they use FFT method.
>  but I cannot find out a variable euivalent with theta(Gi-Gj) from the
>  source in order to accomplish my interface code between Wien2k and
>  Wannier90.
> 1-Would you please guide me which variable in the source is
>  equal to theta(Gi-Gj)? and where is it?
> 2-How is clarified the dependence
>  of theta to (Gi-Gj) in the source of code?
> 
> Your help is appreciated like
>  your previous favor.
> 
> finding theta(Gi-Gj) in the source of code
> MIME-Version: 1.0
> Content-Type: text/plain; charset=utf-8
> Content-Transfer-Encoding: 7bit
> X-Originating-IP: [217.219.18.23]
> 
> Dear all;
> 
> My problem is finding theta(G) (The Fast Fourier Transform of a step-function. I call the step-function theta(r) and it is equal to zero in the spheres and is equal to 1 at the interstitial region. G is the reciprocal lattice vector.)
> 
> I need theta(Gi-Gj) [the FFT of the step-function] to calculate the overlap matrix:
> 
>                  M_mn interstitial(k1,k2)=<u_m,k1|u_n,k2>= sigma_ij{c*_m,k1(Gi) * theta(Gi-Gj) *  c_n,k2(Gj)}
> 
> M is labeled by m,n (band index) and k1,k2 (k-points).
> 
>  
> I think I should find out a variable equivalent with theta(Gi-Gj) from the source of code in order to accomplish my interface code between Wien2k and Wannier90.
> 1-Would you please guide me which variable in the source is equal to theta(Gi-Gj)? and where is it?
> 2-How is clarified the dependence of theta to (Gi-Gj) in the source of code?
> 
> I'm anxiously looking forward your reply and guidelines.
> 
> With my best regards,
> Fatemeh
> 

-- 
-----------------------------------------
Peter Blaha
Inst. Materials Chemistry, TU Vienna
Getreidemarkt 9, A-1060 Vienna, Austria
Tel: +43-1-5880115671
Fax: +43-1-5880115698
email: pblaha at theochem.tuwien.ac.at
-----------------------------------------