SOC calculation of Bi atom by X2C

[qcb...@gmail.com]

Dear developers,

I encountered a strange problem in the spin-orbit coupling calculation of bismuth atom. SO-ECP may get reasonable results, but X2C overestimates the excitation energies significantly. For comparison, OpenMolcas was also used for the spin-orbit coupling calculation with X2C, and the results are reasonable.

          Expt   X2C-1     ECP   X2C-2
4S_3/2       0       0       0       0
2D_3/2   11419   11082   11198   16836
2D_5/2   15438   15231   15397   22029
2P_1/2   21661   21936   22103   28807
2P_3/2   33165   30437   30817   45040

X2C-1: X2C/ANO-R by OpenMolcas
ECP:   aug-cc-pVTZ-PP by Molpro
X2C-2: X2C/ANO-R by Molpro

The input of X2C is as follows.

***,Bi
memory,2000,m

geometry={
Bi;}

basis={
! ANO-RCC basis functions are omitted. One may also use ANO-RCC.
}

set,fnuc=1,dkho=101

! closed shell Bi^3-
{hf;occ,12,7,7,3,7,3,3,1;wf,86,1,0;}

! Bi 6s6p are active
{multi;start,2100.2;orbital,2140.2;
frozen,0;closed,11,6,6,3,6,3,3,1;occ,12,7,7,3,7,3,3,1;
wf,83,8,3;state,1; ! 4S
}

{multi;natorb,ci;
start,2140.2;orbital,2141.2;
frozen,0;closed,11,6,6,3,6,3,3,1;occ,12,7,7,3,7,3,3,1;
wf,83,8,3;state,1; ! 4S
wf,83,2,1;state,2; ! 2D, 2P
wf,83,3,1;state,2; ! 2D, 2P
wf,83,5,1;state,2; ! 2D, 2P
wf,83,8,1;state,2; ! 2D, 2D
}

{ci;noexc;core,11,6,6,3,6,3,3,1; save,4001.3; wf,83,8,3;state,1;}
{ci;noexc;core,11,6,6,3,6,3,3,1; save,4002.3; wf,83,2,1;state,2;}
{ci;noexc;core,11,6,6,3,6,3,3,1; save,4003.3; wf,83,3,1;state,2;}
{ci;noexc;core,11,6,6,3,6,3,3,1; save,4004.3; wf,83,5,1;state,2;}
{ci;noexc;core,11,6,6,3,6,3,3,1; save,4005.3; wf,83,8,1;state,2;}

! MR-CISD+Q energies
hlsdiag(01) =-21530.15309845  ! 4S
hlsdiag(02) =-21530.10430085  ! 2D
hlsdiag(03) =-21530.07426545  ! 2P
hlsdiag(04) =-21530.10430085  ! 2D
hlsdiag(05) =-21530.07426545  ! 2P
hlsdiag(06) =-21530.10430085  ! 2D
hlsdiag(07) =-21530.07426545  ! 2P
hlsdiag(08) =-21530.10430085  ! 2D
hlsdiag(09) =-21530.10430085  ! 2D

! Spin-Orbit calculations with the AMFI approximation.
! Other spin-orbit operators (ls/als) get similar results.
{ci;hlsmat,amfi,4001.3,4002.3,4003.3,4004.3,4005.3;}
---

Best regards,
Wenli

[Peterson, Kirk]

Dear Wenli,

 

Could you post your output file as well?  It’s odd that your final X2C excitation energies are so much higher than the diagonal elements from hlsmat.

 

Best regards,

 

-Kirk

 

From: molpr...@googlegroups.com <molpr...@googlegroups.com> on behalf of qcb...@gmail.com <qcb...@gmail.com>
Date: Wednesday, November 8, 2023 at 12:15 AM
To: molpro-user <molpr...@googlegroups.com>
Subject: [molpro-user] SOC calculation of Bi atom by X2C

[EXTERNAL EMAIL]

--
You received this message because you are subscribed to the Google Groups "molpro-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to molpro-user...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/molpro-user/2892d598-b5a8-42b7-b1a8-0f6f09196261n%40googlegroups.com.

[qcb...@gmail.com]

Dear Kirk,

Here is the output, where a smaller basis set ANO-RCC-VDZ with DKH2 was used to reproduce the problem.

PS. I also calculated the lighter pnictogen atoms As and Sb using ECP, DKH2, and X2C, and the results are correct.

Best regards,
Wenli

[Peterson, Kirk]

Dear Wenli,

 

I reproduce your results, both DK and PP, with the latest version of Molpro. It does seem that the SO matrix elements in the all-electron case are strongly overestimated.  I’m at a loss to explain it.  Perhaps Sasha M. will have some insights.

 

Best regards,

 

-Kirk

 

To view this discussion on the web, visit https://groups.google.com/d/msgid/molpro-user/ceffc42b-183c-4cfc-8296-f0799f991825n%40googlegroups.com.

[Alexander MITRUSHCHENKOV]

Dear Wenli,

I’ll look at your problem asap..

Best wishes,

Alexander

------------------------------------------------------------

Dr. Alexander Mitrushchenkov, IGR
Laboratoire de Modélisation et Simulation Multi Echelle
UMR 8208 CNRS
Univ Gustave Eiffel
5 Bd Descartes
77454 Marne la Vallée, Cedex 2, France

Phone:    +33(0)160957316
Fax:      +33(0)160957320
e-mail:   Alexander.Mitrushchenkov AT univ-eiffel.fr

[s...@isuct.ru]

Dear Alexander,

Recently, we have come across a similar problem when calculating the zero-field splitting (ZFS) in the ground state of the astatine atom in MOLPRO (see Table). Although there is no experimental ZFS available for At, we carried out more rigorous four-component calculations using the DIRAC package for comparison.

                                  ZFS, cm-1
DIRAC: 4c-DHF/dyall.aae4z         24097
DIRAC: 4c-FSCC/dyall.aae4z         23405
MOLPRO: MRCI/aug-cc-pwCVQZ-PP 21006
MOLPRO: MRCI/aug-cc-pwCVQZ-DK3 39260

It looks like this problem persists in MOLPRO for all elements as heavy as 6p (and maybe heavier).

Best regards,

Victor Solomonik

Ivanovo State University of Chemistry and Technology
Russia

понедельник, 13 ноября 2023 г. в 16:14:17 UTC+3, Alexander.Mi...@univ-eiffel.fr:

[Alexander MITRUSHCHENKOV]

Dear Victor,

I’ll try to dig out and understand the problem asap. It can be both a bug (that would be easiest) and the problem of approach (BP Hamiltonian and two-particle density approximation in MRCI) when applied to heavy elements. I will let you know.

Best wishes,

Alexander

------------------------------------------------------------

Dr. Alexander Mitrushchenkov, IGR
Laboratoire de Modélisation et Simulation Multi Echelle
UMR 8208 CNRS
Univ Gustave Eiffel
5 Bd Descartes
77454 Marne la Vallée, Cedex 2, France

Phone:    +33(0)160957316
Fax:      +33(0)160957320
e-mail:   Alexander.Mitrushchenkov AT univ-eiffel.fr

To view this discussion on the web, visit https://groups.google.com/d/msgid/molpro-user/f9b048a9-b68a-4fcf-b62a-c5cd24e363ebn%40googlegroups.com.