[Wien] help NaN error
Gavin Abo
gabo13279 at gmail.com
Wed Jan 4 08:33:44 CET 2023
all the compounds are having the same space group -> That's interesting
that they are all space group 1_P1. Do they all have the same number of
atoms/nonequivalent positions? If so, the compiler might not be the
cause of the issue. Steps from when I installed oneAPI if they help are
at [1].
Maybe it is not helpful but should it be, below are some comments and
historical information I have with regards to the WIEN2k sum rules from
kram.
Section "8.20 OPTIC (calculating optical properties)" and "8.12 KRAM
(Kramers-Kronig transformation)" of the WIEN2k 21.1 usersguide [2] has
the following.
/This [optic/kram] program was contributed by://
/
/Claudia Ambrosch-Draxl/
Section "D. Sumrules" in C. Ambrosch-Draxl and J. O. Sofo article [3]
contains the sentence:
/Typically, in the low energy region the contribution to the intraband
spectrum should sum up to the number of the outermost valence electrons./
If I have interpreted the above statement correctly, "1 add
intraband contributions? yes/no: 1/0" must be set in case.inkram and low
energy region (emin and emax [4] in case.injoint) must be selected were
intraband contributions are dominant.
I saw that interband contributions may be a characteristic of "metals",
slide 20 of [5]. Although as a separate note, I do recall the plasma
frequency needing to be 0 in the case of semiconductor [6].
In your case.inkram [7], I see you have set instead "0 add
intraband contributions? yes/no: 1/0" and "12.60 plasma frequencies
(from joint, opt 6)".
The original WIEN2k code by C. Ambrosch-Draxl's group seemed to
intentionally output the sum rules only when intraband contributions
were added. However, a user requested the sum rule output when
intraband contributions were not added and the kram code got modified to
print them also when there are no intraband contributions. [8]
I have always been interested to find a reference confirming and
clarifying the validity of the sum rules also the no intraband
contribution cases (and insulator & semiconductor cases), but don't
recall coming across one.
When I Googled "Chalcogenide Perovskite", the article at [9] came up having:
/Chalcogenide perovskites are an unusual type of ionic semiconductor./
Thus, it might be helpful to know if all calculations that worked and
the one that isn't if the DFT calculation (band structure/DOS) showed
them being semiconductor (Or metal or insulator).
C. Ambrosch-Draxl and S. Sharma I think redeveloped an optic code after
the original WIEN2k code contribution that has since gotten further
improved, but unfortunately it didn't go into WIEN2k but it seemed to me
that it was put to use in Elk [10].
[1]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg21942.html
[2] http://susi.theochem.tuwien.ac.at/reg_user/textbooks/usersguide.pdf
[3] https://arxiv.org/abs/cond-mat/0402523v1
[4]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg10675.html
[5]
http://susi.theochem.tuwien.ac.at/events/ws2017/notes/Laskowski_optic-xas.pdf
[6]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg22038.html
[7]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg22184.html
[8] http://zeus.theochem.tuwien.ac.at/pipermail/wien/2009-March/012256.html
[9] https://www.azom.com/article.aspx?ArticleID=21105
[10] https://elk.sourceforge.io/#contributions
Kind Regards,
Gavin
WIEN2k user
On 1/3/2023 6:42 AM, naincy Pandit wrote:
> I read your mail sir and checked the possibilities.
>
> opticplot or J is not possible because numbers are missing from the
> table absorb and others also are having NaN
>
> case.joint. the file is completely ok and having the numbers not NaN
>
> as i m working on it, from the previous work my steps are okay.
>
> as you suggest downloading one API, is it paid or free?
>
> and sir, all the compounds are having the same space group,
>
> sir, I'm working on it already and all my data is saved in itself
> (current system), if at this time I will change my compiler then all
> the data will erase and my work vanishes so can you please tell me if
> there is any alternative for this problem.
>
> On Tue, 3 Jan 2023 at 19:02, Peter Blaha <peter.blaha at tuwien.ac.at> wrote:
>
> You did not read my previous email completely, or at least did not
> answer it:
> > Use opticplot and plot "J", i.e. the unbroadened eplsilon-2 in
> > case.joint. Is this ok or already in this file some NaNs
> ....?
> Then we
> > know that the steps before kram are ok.
>
> Can you plot the unbroadened eps-2 (which comes from joint) ??
> Please look into case.joint. Are there NaNs ???
>
>
> Am 03.01.2023 um 13:35 schrieb naincy Pandit:
> > previously also i worked on it and have not faced this type
> of problem
> >
> > and yes. i have completed this optics work for more than 4
> compounds and
> > completely worked
> > previously case.inkram file contains this
> >
> > 0.1 Gamma: broadening of interband spectrum
> > 0.0 energy shift (scissors operator)
> > 0 add intraband contributions? yes/no: 1/0
> > 12.60 plasma frequencies (from joint, opt 6)
> > 0.20 Gammas for Drude terms
> >
> > and in the folder where it did not work also have the same things
> >
> > 0.1 Gamma: broadening of interband spectrum
> > 0.0 energy shift (scissors operator)
> > 0 add intraband contributions? yes/no: 1/0
> > 12.60 plasma frequencies (from joint, opt 6)
> > 0.20 Gammas for Drude terms
> >
> > both the files have same thing so how can i say my previous work or
> > steps are wrong?
> >
> > at this time this is not working there is Not a number (NaN)
> problem.
> >
> > On Tue, 3 Jan 2023 at 17:35, Peter Blaha <peter.blaha at tuwien.ac.at
> > <mailto:peter.blaha at tuwien.ac.at>> wrote:
> >
> > We (you) have first to find out if this is the first error,
> or the
> > problem is already in previous steps.
> >
> > Use opticplot and plot "J", i.e. the unbroadened eplsilon-2 in
> > case.joint. Is this ok or already in this file some NaNs
> ....? Then we
> > know that the steps before kram are ok.
> >
> > If this is ok, maybe your case.inkram is wrong ?
> >
> > PS: Have you ever done optics for another case ? Did it work
> there ?
> > Maybe your compilation of kram is wrong.
> >
> >
> > Am 03.01.2023 um 12:05 schrieb naincy Pandit:
> > > D*ear Prof. Blaha Sir and Wien2k users,*
> > >
> > > i am working with Wien2k_19.1 version with Ubuntu 22.04 i
> need to
> > get some suggestions for optical properties of chalcogenide
> > perovskites of orthorhombic phase.
> > >
> > > i have constructed the structural file with space group p_1.
> > >
> > > the purpose of my calculation is to get optical parameters.
> > >
> > > after doing the scf calculation i started working on
> optical at
> > w2web but at the end i got this type of error mentioned below
> > >
> > > Commandline: *x kram*
> > > Program input is: *""*
> > >
> > > xx
> > > yy
> > > zz
> > > Energy units: [eV]
> > > Lorentzian broadening with gamma: 0.100000000000000
> [eV]
> > > 2001 data points
> > > ENERGY INCREMENT: 1.360999999999990E-002
> > > losmo1 called to perform KK for Im to Re
> > > losmo1 called to perform KK for Im to Re
> > > losmo1 called to perform KK for Im to Re
> > > sum rule 1: Int(sigma)dw NaN
> > NaN
> > > NaN
> > > sum rule 2: Int(eloss.w)dw NaN
> > NaN
> > > NaN
> > > sum rule 3: Int(eloss/w)dw NaN
> > NaN
> > > NaN
> > > KK with broadening done
> > > 0.0u 0.0s 0:00.06 50.0% 0+0k 2120+1192io 17pf+0w
> > >
> > > Please give your valuable suggestions. How to get rid of
> this error.
> > >
> > > Thanks & Regards
> > > Naincy Pandit
> > >
> > >
> >
> > --
> >
> --------------------------------------------------------------------------
> > Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060
> Vienna
> > Phone: +43-1-58801-165300
> > Email: peter.blaha at tuwien.ac.at
> <mailto:peter.blaha at tuwien.ac.at>
> > WIEN2k: http://www.wien2k.at <http://www.wien2k.at>
> > WWW: http://www.imc.tuwien.ac.at <http://www.imc.tuwien.ac.at>
> >
> -------------------------------------------------------------------------
>
> --
> --------------------------------------------------------------------------
> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
> Phone: +43-1-58801-165300
> Email: peter.blaha at tuwien.ac.at WIEN2k: http://www.wien2k.at
> WWW: http://www.imc.tuwien.ac.at
> -------------------------------------------------------------------------
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20230104/5a83ac32/attachment-0001.htm>
More information about the Wien
mailing list