[QE-users] Hard PSP for Bismuth with ld1.x

[Vahid Askarpour]

Dear QE Users,

Following Dr. Giannozzi’s notes on PSP generation, I have successfully generated an LDA norm-conserving fully-relativistic Bi PSP (with 5d10 states) using the following input:

&input
   title='Bi',
   zed=83.,
   rel=2
   config='[Xe] 4f14  6s2 6p3 5d10',
   rlderiv=2.9, eminld=-2.0, emaxld=2.0, deld=0.01, nld=3,
   iswitch=3,
 /
 &inputp
   lpaw=.false.,
   pseudotype=2,
   file_pseudopw='Bi.UPF',
   author='Vahid',
   lloc=0
   nlcc=.true.
   tm=.true.
/
3
5D  3  2 10.00  0.00  1.60  1.60  0.0
6P  2  1   3.00  0.00  2.40  2.40  0.0
6S  1  0   2.00  0.00  2.00  2.00  0.0

I would like to make this PSP harder by adding the 5S and 5P electrons. I use the following input:

&input
   title='Bi',
   zed=83.,
   rel=2
   config='[Xe] 4f14  6s2 6p3 5d10',
   rlderiv=2.9, eminld=-2.0, emaxld=2.0, deld=0.01, nld=5,
   iswitch=3,
 /
 &inputp
   lpaw=.false.,
   pseudotype=2,
   file_pseudopw='Bi.UPF',
   author='Vahid',
   lloc=1,
   nlcc=.true.,
   tm=.true.
/
5
5D  3  2 10.00  0.00  1.20  1.20  0.0
5S  1  0   2.00  0.00  1.20  1.20  0.0
5P  2  1   6.00  0.00  1.20  1.20  0.0
6S  1  0   2.00  0.00  1.20  1.20  0.0
6P  2  1   3.00  0.00  1.20  1.20  0.0

I get 

       Warning: n=3, l=2 expected 0 nodes, found 1
       Setting wfc to zero for this iteration
       (This warning will only be printed once per wavefunction)


 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine run_pseudo (1):
     Errors in PS-KS equation
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I have tried the followings:

1. Change the rcut values
2. Use lloc=0 with 6S at the bottom of the list
3. Use lloc=-1 with various rcloc values

I always get the above error except when I set lloc=-1 and rcloc=1.2. In this case I get good matching of AE and PS energies but the bmat is NaN and so are many of the printed energies.

     n l  j  nl             e AE (Ry)       e PS (Ry)    De AE-PS (Ry) 
     3 2 1.5 5D   1( 4.00)       -2.06502       -2.06502        0.00000
     3 2 2.5 5D   1( 6.00)       -1.84371       -1.84371        0.00000
     1 0 0.5 5S   1( 2.00)     -11.51388      -11.51388        0.00000
     2 1 0.5 5P   1( 2.00)       -8.32746       -8.32746        0.00000
     2 1 1.5 5P   1( 4.00)       -6.49718       -6.49718        0.00000
     1 0 0.5 6S   1( 2.00)       -1.03807       -1.03807        0.00000
     2 1 0.5 6P   1( 2.00)       -0.42584       -0.42584        0.00000
     2 1 1.5 6P   1( 1.00)       -0.28668       -0.28668        0.00000

     eps =     NaN  iter =201

     Etot =  -43140.018305 Ry,  -21570.009153 Ha, -586949.798927 eV
     Etotps =          NaN Ry,            NaN Ha,            NaN eV

     Ekin =            NaN Ry,            NaN Ha,            NaN eV
     Encl =            NaN Ry,            NaN Ha,            NaN eV
     Ehrt =       0.000000 Ry,       0.000000 Ha,       0.000000 eV
     Ecxc =    -145.329988 Ry,     -72.664994 Ha,   -1977.315048 eV
     (Ecc =    -145.329988 Ry,     -72.664994 Ha,   -1977.315048 eV)

Does anyone have any suggestions for what I might be missing from the above procedure?

Thank you,

Vahid

Vahid Askarpour
Department of Physics and Atmospheric Science
Dalhousie University,
Halifax, NS, Canada