The ABINIT Users Mailing List ( CLOSED )

[Ari.P.Seitsonen@iki.fi]

Dear Megchiche,

  Have you tried using the s channel as the local component?
(pswatch -l 0 ...)

    Greetings from the sunny Zurich,

       apsi

> dear abinit users,
> Actually I generate the GGA pseudopotentiel for nickel using fhipp98. evry 
> time I have ghoste state wich I can't eliminate when I change the cutoff 
> raduis for the differents orbitals.
> This is a example of input file which I use to generate the pseudopotentiel 
> for nickel :
> 28.00  5  3  4 1.20  : z ncor nval iexc rnlc
>        1  0    2.00   : n l occ
>        2  0    2.00
>        2  1    6.00
>        3  0    2.00
>        3  1    6.00
>        3  2    8.00
>        4  0    2.00
>        4  1    0.00
> 2 t                  : lmax type
> 0 2.03 0.00 t
> 1 2.24 0.00 t
> 2 2.00 0.00 t
>  
> Now, when I run the pswatch I hanve this message :
> 
> #info /home/hocine/Megchiche/Work4/fhi98PP/bin/Tools/pswatch ncpp 17:02:56 
> jui 14 2003 hocine
> 
> fhi pseudopotential tool pslp - version rev 1998  
> 
>                chemical symbol  Ni
>                 nuclear charge  28.00
>    number of valence electrons  10.00
>       number of valence states   3
>     exchange-correlation model   4  GGA Perdew/Wang 1991          
>      nonlinear core-valence XC
>         parameters radial mesh   525    1.024700  0.223214E-03
>   input pseudopotentials up to   d
> 
>           === pseudo atom (Hartree a.u.) ===
> 
> <        n     l   occupation  eigenvalue(eV)  potential energy
> <  1     3     2      8.0000       -8.9542       -12.17604
> <  2     1     0      2.0000       -5.7048        -3.96404
> <  3     2     1      0.0000       -1.2784         0.00000
> 
>                   total energy     -40.49208
>                 kinetic energy      43.57303
>   ionic pseudopotential energy    -105.33637
>                 hartree energy      33.86535
>                      xc energy     -12.59409
>         local potential energy    -105.68118
>            xc potential energy      -9.01911
>                 xc energy core      -5.60830
>              xc energy valence      -6.98579
>     integrated valence density      10.00000
>           number of iterations            25   convergence  0.0E+00
>        integrated core density       5.94239
>        ... 1st derivative test       1.00000
>        ... 2nd derivative test       0.99999
> estimated equidensity radius >      58.03022
>                   y range xvgr       -23   1   6
> 
>  pslp - pseudoatom done - now testing
> 
>  --- assuming scalar-relativistic all-electron atom ---
> 
>  --- d component taken as local potential ---
>  --- input wavefunctions used for kb potentials ---
> 
>  --- kb potentials: spectrum of bound states (eV) ---
> 
>             l          e0            e1            e2
> semilocal   0       -5.7048        0.0000        0.0000
> nonlocal    0     -105.6186       -5.7048        0.0000
> semilocal   1       -1.2760        0.0000        0.0000
> nonlocal    1       -1.2760        0.0000        0.0000
> 
>  --- analysis of kb potentials: s waves  ---
> 
>  * one or more ghosts (ekb > 0, eref > eloc1)
> 
>                      kb cosine        0.1350
>                      kb energy      257.3508 eV      ekb
>    local potential groundstate     -240.2402 eV    eloc0
>         dto. 1st excited state      -21.2040 eV    eloc1
>               reference energy       -5.7048 eV     eref
> 
>  --- analysis of kb potentials: p waves  ---
> 
>  * no ghost (ekb > 0, eloc0 < eref < eloc1)
> 
>                      kb cosine        0.2525
>                      kb energy      153.7562 eV      ekb
>    local potential groundstate     -120.4497 eV    eloc0
>         dto. 1st excited state       -1.1315 eV    eloc1
>               reference energy       -1.2784 eV     eref
> 
>  --- logarithmic derivatives: at radius = 2.8857 ---
> 
>  --- nonlocal potentials ---
>  --- all-electron potential ---
>  --- semilocal potentials ---
> 
>  --- kinetic energy convergence in momentum space ---
> 
>      l  n  bracket   cutoff    norm   kinet. energy
>             (eV)      (Ry)               (Hartree)
> ck   0     1.0E+00      3   0.983775   0.125752E+00
> ck   0     1.0E-01      8   0.999354   0.158413E+00
> ck   0     1.0E-02     24   0.999976   0.161624E+00
> ck   0     1.0E-03     41   0.999999   0.161917E+00
> cx   0  1                   1.000000   0.161952E+00
> 
> ck   2     1.0E+00     52   0.998479   0.537049E+01
> ck   2     1.0E-01     66   0.999889   0.540270E+01
> ck   2     1.0E-02     80   0.999992   0.540579E+01
> ck   2     1.0E-03    136   0.999999   0.540611E+01
> cx   2  3                   1.000000   0.540614E+01
> 
> 
> thanks for your time.
> Megchiche el hocine
> 

-- 
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
 Ari Paavo Seitsonen / Ari.P.Seitsonen@iki.fi / http://www.iki.fi/~apsi/
 Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35      
 Anschrift: Physikalisch Chemisches Institut (PCI), Universität Zürich (UniZh)
 Indirizzo: Winterthurerstraße 190, CH-8057 Zürich
 Address:   Schweiz / Svizzera / Suisse / Svizra / Switzerland