[Pw_forum] Again: unable( impossible ) to generation PP with semicore states and NLCC ?

[sonu kumar]

Dear All QE users and developers,

I am very sorry for this mail again.

I am unable  generation of PP for Cu with TM (also with rrkj scheme)
pseudization scheme , when i  include semicore states 3s and 3p and NLCC,
after making
many-many tries ( variations in  rc's, logder, V_loc, matching rad for aug
func etc....)

The most comman problem is "negative rho " problem.
I also want to use PAW , but unfortunately even unable to generate USPP.

This is leading  me to think (not 100% sure through  very limited
experience) that atomic code is unable to generate pp with both NLCC
and semicore states and also loosing  my faith in atomic code. So i need
help from PP expert or any one willing to help.


Please see below  for input file----------------------------

 &input
     title='Cu'
     zed=29,
     eminld=-10.0,
     emaxld=3.0,
     rlderiv=2.5,
     nld=3,
     iswitch=3,
     config="[Ne] 3s2 3p6 3d10.0 4s0.3 4p0.2"
     dft='LDA'
     rel=1,
  /
  &inputp
    pseudotype=3,
    nlcc=.true.,
      new_core_ps = .true.
      rcore=0.6000
    lloc=-1,
      rcloc=2.0000,
    file_pseudopw='Cu_1.UPF'
    zval=19,
    lpaw=.false.
      lnc2paw=.false.
      which_augfun ='PSQ',
      rmatch_augfun= 2.0000,
    author='sk',
    tm =.true.
    !rho0=0.01,

!nX n l   occ   nrg   rmin   rcut
  /
 8
 3S  1 0  2.00    0.00    0.650    3.0000  1
 3P  2 1  6.00    0.00    0.650    3.0000  1
 3D  3 2  10.0    0.00    0.650    2.0000   1
 3D  3 2  0.00    0.10    0.650    2.0000   1
 4S  1 0  0.30    0.00    2.000    3.2000  1
 4S  1 0  0.00    1.95    2.000    3.2000  1
 4P  2 1  0.20    0.00    2.600    3.6000  1
 4P  2 1  0.00    0.05    2.600    3.6000  1

 -------------------input
file end-------------------------------------------



With Kind Regards,
Sonu

IITD
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20101016/103b15ec/attachment.htm