whether the wave-functions in WAVECAR file are normalized?

Message

renh · #1 Post by **renh** » Thu Apr 17, 2008 12:46 pm

Hello all,

I read the Fourier components from WAVECAR file, but I found that the modulus of each wave function is not unity, though has a value around 1.

I'd like to know whether the wavefunctions are normalized? OR the deviation from unity is just some numerical error?

Following is the modulus square I calculated from the coefficients in WAVECAR file, the system is a CO molecule in 8A*8A*8A box
(
\sum_{i=1}^{nplw} | C_{i} | ^ 2
)

I appreciate any help or suggestions!

-------------------------------------------------------------------------------------

Code: Select all

ndum = 74528
nk= 1          nband= 9          enmax= 400.00000

Latt. Const.
8.00000000000000 0.000000000000000E+000 0.000000000000000E+000
0.000000000000000E+000 8.00000000000000 0.000000000000000E+000
0.000000000000000E+000 0.000000000000000E+000 8.00000000000000

Band No.      norm-square
================================
   1         1.09415319117314
   2         0.946511978865221
   3         0.840185412329243
   4         0.840185413176901
   5         0.953311515064630
   6         0.895347092733738
   7         0.895347077251360
   8         1.00013031395765
   9         1.00213347527607

<span class='smallblacktext'>[ Edited ]</span>

#2 Post by **admin** » Fri May 16, 2008 3:35 pm

please note that on WAVECAR the PW- coefficients of the wavefunction are written only. If cou compare to Kresse, Joubert,PRB59, 1758 (1999) (eqns 38-40) you will see that these contributions alone are not normalized. .