[Wien] charge density of a single atom

[Dr. Sharat Chandra]

Hi

I am interested in obtaining the variation in the charge density with 
respect to r, for isolated Y atom. For this I have seen the convergence of 
total energy with respect to the unit cell length, RMT*kmax and Gmax and I 
use a single k-point for all the calculations. I obtain 15 a.u. as the 
converged unit cell parameter. I have put the atom in the centre of the 
unit cell. scf shows that there are 11 electrons in the valence band, ie., 
4s2, 4p6, 4d1, 5s2. The input for lapw2 in electron density calculation is 
the range of the valence electrons energy obtained in the scf file. I 
carry out the calculation of the 3D charge density using XCrysden, using 
RHO, ANG, VAL, NODEBUG as the inputs for lapw5 program.

I calculate the the distribution of the charge density from the 
(1/2,1/2,1/2) to (1,1/2,1/2) direction along the single line to get the 
rho(r) versus r on a 300 point grid from the xsf data save by XCrysden. 
When I perform an integration of this data (by integrating the r versus 
4*pi*r^2*rho(r) data) to obtain the total charge, I get ~20 electrons. I 
should be getting either 11 electrons, or 39 electrons, which is the 
atomic number of Y, what I get is neither here nor there.

Can you please tell me what I am doing wrong? Is it that I have to use AIM 
to get the real charge density, or any other method is to be used?

Regards

Sharat Chandra