[Wien] Linearization of Low-lying s-state

Laurence Marks L-marks at northwestern.edu
Mon Mar 7 15:06:18 CET 2011


The "below -12.2" comes from what you have at the bottom of case.in1,
so this should not be an issue.

I'm not sure about your low lying states, normally a "WARNING" is just
that. However, I do wonder why you reduced the P so much and not the O
instead. You may also be going a bit too far with this, more normal is
to use a few percent with setrmt and only reduce much more if you run
into problems with overlapping spheres.

2011/3/7 David Tompsett <dat36 at cam.ac.uk>:
> Dear All,
>
> I am performing calculations on slabs of the material LiFePO4 using GGA+U in
> the latest Wien2k release.
>
> To perform min_lapw internal parameter optimisation I reduce the RMT's and
> use R=1.06 for Phosphorus. This results in a need to reduce the default for
> energy separation of core states to -9.2Ry during lstart. As a result
> Phosphorus 2p states become part of the valence and need to be treated by a
> LO. I am having trouble with large QTL-B values regarding the L=0 states
> though.
>
> The energy parameters for phosphorus from case.outputst are:
>
>                     P                                   RHFS
>
>          OCCUPANCY    ENERGY(RYD)         (R4)
> (R2)              (R)               (R-1)             (R-3)
>
>   1S      1.000    -1.5314154E+02     5.4020520E-04     1.4448157E-02
> 1.0359168E-01     1.4648276E+01
>   1S      1.000    -1.5311976E+02     5.3994095E-04     1.4444579E-02
> 1.0357948E-01     1.4649640E+01
>   2S      1.000    -1.2760822E+01     1.9082408E-01     3.2223815E-01
> 5.1970457E-01     2.8247951E+00
>   2S      1.000    -1.2738260E+01     1.8994048E-01     3.2198220E-01
> 5.1963797E-01     2.8236582E+00
>   2P*     1.000    -9.1715513E+00     1.9086915E-01     2.9430623E-01
> 4.8389989E-01     2.7216785E+00
>   2P*     1.000    -9.1408359E+00     1.8923948E-01     2.9388805E-01
> 4.8380643E-01     2.7206626E+00
>   2P      2.000    -9.1041258E+00     1.9405529E-01     2.9690219E-01
> 4.8614968E-01     2.7052889E+00     6.2902446E+01
>   2P      2.000    -9.0735022E+00     1.9234866E-01     2.9645480E-01
> 4.8603999E-01     2.7043227E+00     6.2822412E+01
>   3S      1.000    -1.0878541E+00     2.9229783E+01     4.1553776E+00
> 1.8919617E+00     7.2068651E-01
>   3S      1.000    -8.8775170E-01     3.1604015E+01     4.2438626E+00
> 1.9036101E+00     7.2402079E-01
>   3P*     1.000    -4.6442261E-01     7.9415784E+01     6.4475288E+00
> 2.3215625E+00     5.8248088E-01
>   3P*     0.000    -2.7621315E-01     1.1443191E+02     7.4136855E+00
> 2.4614687E+00     5.6064252E-01
>   3P      2.000    -4.6042138E-01     8.0649286E+01     6.4959005E+00
> 2.3301759E+00     5.8005227E-01     3.7773200E+00
>   3P      0.000    -2.7242986E-01     1.1708022E+02     7.4892061E+00
> 2.4735418E+00     5.5768379E-01     3.5655831E+00
>
> Examples of the low lying states I find in case.output1up and case.output1dn
> are:
> From up:
>      K=   0.25000   0.50000   0.50000            1
>       MATRIX SIZE 9192  WEIGHT= 2.00  PGR:
>      EIGENVALUES ARE:
>     -11.4326572  -11.4326443  -11.4304960  -11.4304836  -11.3548181
>     -11.3548178  -11.3547956  -11.3547954   -7.8222277   -7.8221979
>      -7.8198177   -7.8197860   -7.8139864   -7.8138903   -7.8117306
>      -7.8116318   -7.8093135   -7.8092692   -7.8073401   -7.8073004
>      -7.7374667   -7.7374653   -7.7373719   -7.7373716   -7.7338651
>      -7.7338636   -7.7337751   -7.7337738   -7.7332502   -7.7332486
>      -7.7329941   -7.7329924   -5.4113446   -5.4113297   -5.4091455
>      -5.4091305   -5.3250955   -5.3250950   -5.3250626   -5.3250622
>
> From dn:
>      K=   0.12500   0.25000   0.50000            1
>       MATRIX SIZE 4988  WEIGHT= 2.00  PGR:
>      EIGENVALUES ARE:
>      -8.5813282   -8.5812344   -8.5638097   -8.5637104   -8.3260488
>      -8.3257490   -8.3242101   -8.3239949   -5.6312550   -5.6309409
>      -5.6112020   -5.6109385   -5.6107774   -5.6105375   -5.6015340
>      -5.6014578   -4.8776979   -4.8773081   -4.8649380   -4.8645429
>      -4.8612661   -4.8609989   -4.8603466   -4.8597815   -4.8473038
>      -4.8470997   -4.8453215   -4.8450239   -4.6260569   -4.6254883
>      -4.6248374   -4.6245754   -4.6239155   -4.6237133   -4.6234791
>      -4.6230319   -4.6227243   -4.6219883   -4.6218599   -4.6214088
>
> One question is why does the code print?
>             0 EIGENVALUES BELOW THE ENERGY  -12.20000
>        ********************************************************
> i.e. that there are no eigenvalues more than 3Ry below the core separation
> energy?
>
> At the bottom of case. scfup I obtain:
>
>     Most likely no ghostbands, but adjust Energy-parameters or use -in1ef /
> -in1new
>
>
> :WARN : QTL-B value eq.   4.97  in Band of energy  -8.32610   ATOM=  134
> L=  0
> :WARN : You should change the E-parameter in case.in1 or use -in1ef /
> -in1new switch
>
>
>
>    QTL-B VALUE .EQ.    5.47621   in Band of energy   -8.32606   ATOM=  134
> L=  0
>
> However, from case.outputst I would have thought that the L=0 states of this
> phosphorus atom would be in the core? I have attempted adjusting the El
> parameters in case. in1 to adjust to this band energy, but the problem
> returns at a slightly difference band energy.
>
> In case.in1 for phosphorus I have used:
>   0.30    4  0      (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global
> APW/LAPW)
>  1    0.30      0.000 CONT 1
>  1   -8.80      0.001 STOP 1
>  0   -0.79      0.002 CONT 1
>  0   -10.5      0.000 CONT 1
>
> The default for the L=0 LO was 0.3.
>
> Any help as to whether these s-states are real or how to describe them would
> be very helpful.
>
> Many thanks,
> David Tompsett.
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>



-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
Web: www.numis.northwestern.edu
Chair, Commission on Electron Crystallography of IUCR
www.numis.northwestern.edu/
Electron crystallography is the branch of science that uses electron
scattering and imaging to study the structure of matter.


More information about the Wien mailing list