[Wien] Initial wave function

[Jorissen Kevin]

Hello, I'll try to help you out here.
* open case.corewf.  This file is written by the program lcore in each scf-cycle (so it now contains the core wave functions of the last iteration, that's what you need).  The structure of this file is as follows :
  Atom 1 (as in case.struct)   [They are not indexed in the file; for each atom, you'll see a line like : 1  core states for
                                                                                                this atom]
      core state 1                    [ there's a line like : CORE STATES = 1S ]
          large component of the wave function of core state 1             [ 781 real numbers ]
          small component of the wave function of core state 1           [idem]
      core state 2
          large
          small
      ...
       last core state
          large
          small
   Atom 2
         core state 1
          ....
END OF THE FILE.
 
So, either you write a small script that selects the correct wave function for you, or (if you just need one or two) you copy/paste the correct one yourself into a new file.
The two components are NOT to be added.  In not-relativistic quantum theory, we solve a Schrodinger equation and obtain functions as solutions.  In relativistic theory, we get more complicated solutions : four component wave functions (f1,f2,f3,f4) if we solve the Dirac equation correctly; two component wave functions (flarge,fsmall) in our approach (the scalar relativistic approximation, where spin-orbit is neglected and, if desired, added perturbatively later on).  Don't add them - they are not two parts of one scalar function, but really two separate components of a two component function !!  In the famous Singh book, read section 5.4.2
Core wave functions are described as radial functions times a spherical harmonic function (which depends on the angular variables).  The part in case.corewf is only the radial part.  This is real by the way (not complex).
For each component you will find 781 numbers.  If you use a fortran program to read them, I think you can just use something like
 read(file,*) (small(i),i=1,781)
 read(file,*) (large(i),i=1,781)
to get them (replace file by the correct logical unit number in fortran).
 
The radial coordinate for each index i is r(i) = R0 * exp[(i-1)*dx]  . 
So, the first number is the value of the large component at r= R0 (probably 0.0001 or 0.00005, see case.struct), etc.
 
 
Like I said in a previous mail, it is possible that in the output r * f(r) is written, not just f(r).  I don't remember.  It should be obvious when you plot the function.
 
I'm sorry, I do not know the abbreviation FDMNES.
 
Let me know if you require further assistance.
 
 
Kevin.
 
 

	 
	-----Original Message----- 
	From: yanming Ma [mailto:ymma66 at yahoo.com] 
	Sent: Thu 10/2/2003 9:46 PM 
	To: wien at zeus.theochem.tuwien.ac.at 
	Cc: 
	Subject: RE: [Wien] Initial wave function
	
	
	Dear kevin
	 
	I still have the problem to convert the wave function in WIEN2k to FDMNES.
	 
	>These numbers are the value of the core state wave function at every point of the logarithmic >mesh used inside of a muffin tin (i.e., they are Fcore(r(i)), and  r(i) = R0 * exp[(i-1)*dx]  ,  dx = >ln[RMT/R0] / (NPT-1) , and you find RMT, R0 and NPT in case.struct for every atom; NPT is >usually 781).  
	 
	What is the exact expression for Fcore(r(i))?
	 
	>There are TWO sets of numbers because the wave function has a large (first >set) and >small component (it's a two-component function).
	>So, select the core orbital that you want, and copy/paste the corresponding numbers; or write >a simple program that does this for you.
	 
	Can I just simply add the two component together as the final core orbital.?
	 


	Yanming Ma Ph.D
	Steacie Institute for Molecular Sciences,
	National Research Councils of Canada,
	Ottawa, Ontario
	K1A 0R6
	Canada 

	
  _____  

	Do you Yahoo!?
	The New Yahoo! Shopping <http://shopping.yahoo.com/?__yltc=s%3A150000443%2Cd%3A22708228%2Cslk%3Atext%2Csec%3Amail>  - with improved product search

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 9798 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20031003/ffc0544b/attachment.bin