[Q-e-developers] problem with tin pseudopotential

[Nicola Marzari]

ciao Lorenzo - is it pslibrary 1.0?

It seemed fairly soft and very accurate, and without ghosts:

http://materialscloud.org/results/Sn_conv_patt.png
http://materialscloud.org/bands.html

			nic


On 28/01/2016 16:13, Lorenzo Paulatto wrote:
> Dear all,
> I'm working on a PAW pseudopotential for Tin, based on the one of plibrary,
> but trying to make it softer.
>
> If you run the following code you will see that the Bessel function test for
> the d channel never converges, i.e. instead of going toward the eigenvalue
> (-2.4847Ry) it goes toward a somewhat higher value: -2.4741Ry. This value also
> seems to be strongly dependent on the size of the Bessel box, which alerts me
> to the possibility of a ghost state.
>
> As this pseudo is quite similar to the one in the pslibrary, where this
> problem is actually even more evident. I would then ask you to have a look at
> the input, and tell me if you see any mistake.
>
> I've also attached a plot of the wavefunctions and of the logarithmic
> derivative (with the partial wave expansion multiplied by 1000)
>
>   &input
>     title='Sn',
>      prefix="Sn_ld1"
>     zed=50.,
>     rel=1,
>     config='[Kr] 4d10.0 5s2.0 5p1.0 5d-2 6s-2 6p-2',
>     iswitch=3,
>     dft='PBE'
>
>     nld=4,
>         eminld=-12,
>         emaxld=6,
>         deld=0.002d0,
>         rlderiv=2.40,
>
>   /
>   &inputp
>     lpaw=.true.,
>     pseudotype=3,
>     file_pseudopw='Sn.pbe-paw-4d.UPF',
>     author='ADC',
>     lloc=-2
>       rcloc=2.3,
>     which_augfun='BESSEL',
>     rmatch_augfun=2.0,
>     nlcc=.true.
>       rcore=2.0,
>       new_core_ps=.true.
>     tm=.true.
>
>     file_screen='Sn_ld1.scr'
>     file_beta  ='Sn_ld1.beta'
>     file_wfcusgen   ='Sn_ld1.usc'
>     file_wfcaegen   ='Sn_ld1.aec'
>     file_core = 'Sn_core'
>     file_screen='Sn_screen'
>
>
>   /
> 6
> 4D  3  2 10.00  0.00  1.00  2.30  0.0
> 4D  3  2  0.00 -0.30  1.00  2.30  0.0
> 5S  1  0  2.00  0.00  1.00  2.30  0.0
> 5S  1  0  0.00  2.10  1.00  2.30  0.0
> 5P  2  1  1.00  0.00  1.00  2.30  0.0
> 5P  2  1  0.00  2.30  1.00  2.30  0.0
>
> &test
>    ecutmin=10
>    ecutmax=200
>    decut=5
>    rm=50
> ! configts(1)='[Kr] 4d10.0 5s2.0 5p2.0'
> !4D  3  2  0.00  4.50  1.70  2.30  0.0
> /
> 3
> 4D  3  2 10.00  0.00  2.00  2.30  0.0
> 5S  1  0  2.00  0.00  2.00  2.30  0.0
> 5P  2  1  1.00  0.00  2.00  2.30  0.0
>
>
>
>
>
> _______________________________________________
> Q-e-developers mailing list
> Q-e-developers at qe-forge.org
> http://qe-forge.org/mailman/listinfo/q-e-developers
>

-- 
----------------------------------------------------------------------
Prof Nicola Marzari, Chair of Theory and Simulation of Materials, EPFL
Director, National Centre for Competence in Research NCCR MARVEL, EPFL
http://theossrv1.epfl.ch/Main/Contact http://nccr-marvel.ch/en/project