[Wien] NMR calculations with f-block elements: lapw1 crash
Peter Blaha
peter.blaha at tuwien.ac.at
Tue May 30 15:30:53 CEST 2023
Hi,
I can confirm the problems, although I get a bit different problems.
The LOMAX=3 stop in lapw1 is ok, and recompilation of both, lapw1 and
lapw2 should be ok.(in lapw2 the LOMAX setting is in modules.F).
However, the real problem is that also in SRC_nmr/modules.F the LOMAX
must be increased to 4. Without this, the case.in1_nmr file is wrong and
probably causes the second problem.
Look at your case.in1_nmr file. I guess the l=4 terms for Hf come twice
(and also my next O-atom has a wrong LO-list.
So in essence the "bug" is that x_nmr -mode in1 does not stop when
LOMAX is only 3 and you have f states.
I'll see how I can fix it for the next release.
Recompile all 3 programs (either in siteconfig or using make all
and copying the real/complex versions down.
Regards
Peter Blaha
Am 30.05.2023 um 08:52 schrieb Arkadii Pominov:
> Dear Wien2k users and developers,
>
> I am using Wien 23.2 on a two-socket 56-core/128GB Intel system
> running Ubuntu 22.04.1 LTS, gfortran-11 compiler with Openblas OMP
> libraries.
>
> I'm trying to calculate chemical shielding in a metallic compound
> containing Hf, La, Ni, and P atoms using the PBE functional.
> Initially, I used the Wien2k 21 version but then updated to the latest
> 23.2, and it still returns an error. I suspect it has something to do
> with Hf and its f-electrons. I replaced the atom with Zr, and
> everything worked (or at least it didn't crash instantly). Is there an
> inherent limitation of the NMR module that concerns/f/-elements or am
> I doing something incorrectly?
>
> I'm running these commands:
>
> cif2struct case.cif
> setrmt_lapw case
> init_lapw -b -numk 3000 -fermit 0.004 -rkmax 5.5 # I set 5.5 RKmax
> + 3000 kpoints as a starting point to not waste cpu time)
> run_lapw -p -fc 1 -cc 0.0001
> save_lapw rkm55_3k_4mry
> x_nmr_lapw -mode in1
> x_nmr_lapw -metal <- this is when it instantly crashes with Hf,
> whether with "-p" flag or not
>
>
> There are very few informative errors, here is what I found in
> nmr_q0/lapw1.error, which is caused by a STOP error in the
> SRC_lapw1/atpar.f, I assume:
>
> 'ATPAR' - LOMAX too small (LOMAX = 3, L = 4)
>
>
> And if I increase LOMAX to 4 in all modules and recompile, I get
> another error in nmr_q0/lapw1.error:
>
> 'ATPAR' - NLOAT too small (NLOAT = 11, L = 4 JATOM = 4)
>
>
> Considering NLOAT value is 3 by default, 11 is really high, so I
> assume something went wrong here, no? There is not much in the UG
> about that.
>
> Any advice and suggestions on how to make it work with Hf would be
> greatly appreciated! The bits of the input files are attached below
> for clarity.
>
> Kind regards,
> Arkadii
>
> Here is the start of my structure file after conversion from cif and
> initialization (it's quite lengthy):
> case.struct
>
> La_Hf_Ni_P
> H LATTICE,NONEQUIV.ATOMS: 10 189 P-62m
> MODE OF CALC=RELA unit=bohr
> 23.665040 23.665040 7.350468 90.000000 90.000000120.000000
> ATOM -1: X=0.81060000 Y=0.00000000 Z=0.50000000
> MULT= 3 ISPLIT= 8
> -1: X=0.00000000 Y=0.81060000 Z=0.50000000
> -1: X=0.18940000 Y=0.18940000 Z=0.50000000
> La1 NPT= 781 R0=.000010000 RMT= 2.50000 Z: 57.
> LOCAL ROT MATRIX: 0.0000000 0.5000000 0.8660254
> 0.0000000 0.8660254-0.5000000
> -1.0000000 0.0000000 0.0000000
> ATOM -2: X=0.33333333 Y=0.66666667 Z=0.00000000
> MULT= 2 ISPLIT= 4
> -2: X=0.66666667 Y=0.33333333 Z=0.00000000
> La2 NPT= 781 R0=.000010000 RMT= 2.50000 Z: 57.
> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000
> 0.0000000 1.0000000 0.0000000
> 0.0000000 0.0000000 1.0000000
> ATOM -3: X=0.42956000 Y=0.00000000 Z=0.00000000
> MULT= 3 ISPLIT= 8
> -3: X=0.00000000 Y=0.42956000 Z=0.00000000
> -3: X=0.57044000 Y=0.57044000 Z=0.00000000
> Hf1 NPT= 781 R0=.000005000 RMT= 2.50000 Z: 72.
> LOCAL ROT MATRIX: 0.0000000 0.5000000 0.8660254
> 0.0000000 0.8660254-0.5000000
> -1.0000000 0.0000000 0.0000000
> ...
>
>
> Here is the start of my case.in1c file:
>
> WFFIL EF=.72537106443116053336 Old .7253702202934816 0.7253714865
> (WFFIL, WFPRI, ENFIL, SUPWF)
> 5.50 10 6 ELPA pxq BL 64 (R-MT*K-MAX,MAX L IN WF,V-NMT,LIB)
> 0.30 5 0 (GLOBAL E-PARAMETER WITH n OTHER CHOICES,
> global APW/LAPW)
> 0 0.30 0.0000 CONT 1
> 0 -2.56 0.0010 CONT 1
> 1 0.30 0.0000 CONT 1
> 1 -1.30 0.0010 CONT 1
> 2 0.30 0.0010 CONT 1
> 0.30 5 0 (GLOBAL E-PARAMETER WITH n OTHER CHOICES,
> global APW/LAPW)
> 0 0.30 0.0000 CONT 1
> 0 -2.56 0.0010 CONT 1
> 1 0.30 0.0000 CONT 1
> 1 -1.30 0.0010 CONT 1
> 2 0.30 0.0010 CONT 1
> 0.30 7 0 (GLOBAL E-PARAMETER WITH n OTHER CHOICES,
> global APW/LAPW)
> 0 0.30 0.0000 CONT 1
> 0 -4.60 0.0001 STOP 1
> 1 0.30 0.0000 CONT 1
> 1 -2.16 0.0010 CONT 1
> 3 -0.85 0.0010 CONT 1
> 3 0.30 0.0000 CONT 1
> 2 0.30 0.0010 CONT 1
> ...
> K-VECTORS FROM UNIT:4 -9.0 1.5 451 emin / de
> (emax=Ef+de) / nband
>
>
>
> --
>
> *ARKADII POMINOV, M. Sc.
> *Ph. D. candidate in Mar & Michaelis Groups
>
> *UNIVERSITY OF ALBERTA
> *College of Natural and Applied Sciences
> Faculty of Science
>
> Department of Chemistry
> 11227 Saskatchewan Drive
> Edmonton, AB Canada T6G 2G2
>
> University of Alberta Logo <https://www.ualberta.ca/index.html>
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
--
-----------------------------------------------------------------------
Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-158801165300
Email:peter.blaha at tuwien.ac.at
WWW:http://www.imc.tuwien.ac.at WIEN2k:http://www.wien2k.at
-------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20230530/0c2746b8/attachment.htm>
More information about the Wien
mailing list