[Wien] XMCD S K-edge

Mon Feb 1 18:18:14 CET 2021

Dear Krish,

Joint.f line 210

      READ(23,331) SOR,edge  <--------
      rewind (23)
      select case (edge)

 case ('K  ')
      xlo=1

      case ('L1 ')
      xlo=1

      case ('L23')
      xlo=2
...

Joint.f line 939

    if(spino.and.sor) then
       if(xlo.eq.1) then
          call dich1(xlo,DE,EC)    <--------
       else
          call dich2(xlo,EEF,DE,EC)
       endif

I don't think the subroutine "dich1" cares core energy since the core state
is not split. the subroutine "dich2" does care core energies.
Joint should know if you are calculating "K-edge"  from reading file "23" which
is generated from optics.

"23,'${scratch}$file.symmat1$updn' ,  'UNKNOWN','FORMATTED',  0
24,'${scratch}$file.symmat2$updn' ,  'UNKNOWN','FORMATTED',  0"

Check if you have these files. It might be there are some complications in file name.

Best,
Yongbin

________________________________
From: Wien <wien-bounces at zeus.theochem.tuwien.ac.at> on behalf of KRISH <krish.odc at gmail.com>
Sent: Monday, February 1, 2021 8:06 AM
To: wien at zeus.theochem.tuwien.ac.at <wien at zeus.theochem.tuwien.ac.at>
Subject: Re: [Wien] XMCD S K-edge

Prof. Peter Blaha,

I used spin polarized + spin-orbit calculation. I have got the spin-up and spin-dn S-1s core energies.
Here are the core energy values:
cas.scfcup
        1.ATOM      S                     4 CORE STATES
:1S 001: 1S                -175.485261107 Ry
:2S 001: 2S                 -14.238881461 Ry
:2PP001: 2P*                -10.241763425 Ry
:2P 001: 2P                 -10.149047898 Ry

cas.scfcdn
        1.ATOM      S                     4 CORE STATES
:1S 001: 1S                -175.486319774 Ry
:2S 001: 2S                 -14.239790578 Ry
:2PP001: 2P*                -10.243054336 Ry
:2P 001: 2P                 -10.150333457 Ry

How can I resolve the issue? Could you help me?

Thank you.
KRISH

On Sun, 31 Jan 2021 at 15:26, KRISH <krish.odc at gmail.com<mailto:krish.odc at gmail.com>> wrote:
Dear Wien2k Users,

My question is, I am not sure, how to choose the core_E2 value.

The only core values available for S-atom in the SCF file are:

   1.ATOM      S                     4 CORE STATES

:1S 001: 1S                -175.486319774 Ry

:2S 001: 2S                 -14.239790578 Ry

:2PP001: 2P*                -10.243054336 Ry

:2P 001: 2P                 -10.150333457 Ry

If I select Core_E2 as -175.486 and Core_E1 as any one of the remaining values, I get NAN in the produced case.xmcd file for RIGHT_POL, LEFT_POL, and XAS;. Has anyone ever tried K-edge XMCD? Could someone help me?

Best regards,

KRISH

On Mon, 25 Jan 2021 at 16:00, KRISH <krish.odc at gmail.com<mailto:krish.odc at gmail.com>> wrote:
Dear WIEN2k users,

The above mentioned case.injoint input file gives me syntax error when I do
x joint -up

forrtl: severe (59): list-directed I/O syntax error, unit 5, file /home/krish/WIEN2k/us-lsda-op-xm2/us-lsda-op-xm2.injoint
Image              PC                Routine            Line        Source
joint              0000000000420A0E  Unknown               Unknown  Unknown
joint              0000000000443534  Unknown               Unknown  Unknown
joint              000000000040B368  MAIN__                    196  joint.f
joint              0000000000402E5E  Unknown               Unknown  Unknown
libc-2.17.so<http://libc-2.17.so/>       00002B6258A68C05  __libc_start_main     Unknown  Unknown
joint              0000000000402D69  Unknown               Unknown  Unknown
0.002u 0.003s 0:00.00 0.0%      0+0k 0+8io 0pf+0w
error: command   /opt/ohpc/pub/wien2k/18.2/joint upjoint.def   failed

I am not sure how to fix this. Could someone help me?

Thank you.
KRISH

On Fri, 22 Jan 2021 at 18:14, KRISH <krish.odc at gmail.com<mailto:krish.odc at gmail.com>> wrote:
Dear Wien2k Users,

I am using wien2k 18.2 for the computation of XMCD spectra.
I tried to compute S K-edge XMCD spectra for my system (US). However, whilst doing that I encountered a problem for the specification K-edge values in the case.injoint file.
The injoint is given below.

1  9999 9999              : LOWER,UPPER and (optional) UPPER-VAL BANDINDEX
-0.0000    0.00100   1.0000 : EMIN DE EMAX FOR ENERGYGRID IN ryd
eV                            : output units  eV / ryd  / cm-1
XMCD                          : omitt these 4 lines for non-XMCD
-175.4856               : core energies in Ry (grep :1S case.scf) /S K-edge
1.6  0.6                      : core-hole broadening (eV) for both core states
0.1                           : spectrometer broadening (eV)
     6                        : SWITCH
     4                        : SWITCH
     2                        : NUMBER OF COLUMNS
   0.1  0.1  0.3              : BROADENING (FOR DRUDE MODEL - switch 6,7 -
ONLY)

For the M45 edge we specify 3D and 3D* edge values in Ryd. I am not sure about the K-edge. The only core values available for S-atom in the SCF file are:

   1.ATOM      S                     4 CORE STATES

:1S 001: 1S                -175.486319774 Ry

:2S 001: 2S                 -14.239790578 Ry

:2PP001: 2P*                -10.243054336 Ry

:2P 001: 2P                 -10.150333457 Ry

Can someone help, how can I select the higher core energy?

The corresponding case.inop file is given below:

9999 1       number of k-points, first k-point
-10.0 10.0 9999 Emin, Emax for matrix elements, NBvalMAX
XMCD 1 K      optional line: for XMCD of 1st atom and K-edge
6             number of choices (columns in *outmat): 2: hex or tetrag. case
1             Re xx
2             Re yy

OFF           ON/OFF   writes MME to unit 4

Am I missing anything here? Could someone help?

Thanks in advance.

Best regards,

KRISH

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20210201/e6954c74/attachment.htm>