KRCI help

[Peterson, Kirk]

Dear Dirac experts,

 

I guess I need another set of eyes to look at this since I can't seem to spot my mistake.  I'm attempting what should be a relatively straightforward KRCI calculation on element 125 (5 open shell electrons in 5g-, 6f-, 7d-, 8p-).   The current input generates zero determinants in every one of the possible 128 dbg irreps.  My local version of Dirac has been modified to allow calculations for Z>118, but I can't see how this would affect KRCI calculations.  Here is the relevant part of the input:

 

.CI PROGRAM

LUCIAREL

.INACTIVE

 26 29    # correlate 7s and above, 6d and below inactive

.CIROOT

 19u 1

.GAS SHELLS

 4

 7  8 /  1  3    # 7s and 7p

 8 10 /  1  0    # 8s

13 15 /  6  4    # 5g-,7d- and 8p-,6f-

15 15 / 85 121   # 50 au

 

 

The initial part of the output seems very reasonable (to me):

 

  evaluating active dbg irrep...

  # dbg irreps               :                   128

  # active MJ value (doubled):                    19

  # active fermion sym       :                     2

  # active double group irrep:                   115

===========================================================================

 *KRCICALC: General set-up for KR-CI calculation:

===========================================================================

 * Inactive orbitals     :   26  29

 * Active orbitals       :   93 128

 * Active electrons      :   15

 * GAS space setup for   4 GAS space(s) :

   - GAS space   1       :    1   3

    (constraints: min/max active electrons after space :   7/  8)

   - GAS space   2       :    1   0

    (constraints: min/max active electrons after space :   8/ 10)

   - GAS space   3       :    6   4

    (constraints: min/max active electrons after space :  13/ 15)

   - GAS space   4       :   85 121

    (constraints: min/max active electrons after space :  15/ 15)

 * CI program used       : LUCIAREL

 * Convergence of CI function(s) in the following symmetries:

    **   1 eigenstate(s) for MJ-value (doubled):  19u/2

    -- Allowed interval of 2 * MK : -15 to  15

 

 

regards,  -Kirk

[stefan]

Dear Kirk,

your input (and the output) seems reasonable to me as well. 

Although I made provisions for cases with mj > 15/2 I have never tested it to its fullest. I could imagine that you hit a bug here with your 19/2 state. 

You could try the following and change all the occurrences of "128" for KRCI to "256". This could make the CI calculation slower but should hopefully help. 

with my best regards

Stefan

src/include/dcbkrci.h

PARAMETER (MAX_NKRCI_MAX_SYM = 128)

../src/include/krciprop.h:      

PARAMETER(MXPROPKRCI_SYM = 128)

src/luciarel/mospinor_info.F90:   

integer, parameter,   public :: max_num_dbg_irreps = 128

../src/luciarel/mxpdim.inc

parameter (MXNDGIRR = 128 )

src/krmc/krmcgp.F:           

NIRR_DG     = 128

../src/krmc/krmcgp.F:           

NIRR_DG     = 128

../src/krmc/krmcinp.F:      

integer, parameter              :: max_nkrmc_max_sym = 128

../src/krmc/krmcmlf.F:           

NIRR_DG     = 128

../src/krmc/krmcmlf.F:           

NIRR_DG     = 128

[Peterson, Kirk]

Dear Stefan,

 

Miro just ran my Th atom case where mj= 2 and it fails in the same way.  So it doesn't seem to be related to my high mj value in e125.

 

best regards,

 

-Kirk

 

From: <dirac...@googlegroups.com> on behalf of stefan <stefan...@phys.chem.ethz.ch>
Reply-To: "dirac...@googlegroups.com" <dirac...@googlegroups.com>
Date: Thursday, August 3, 2023 at 3:48 AM
To: dirac-users <dirac...@googlegroups.com>
Subject: [dirac-users] Re: KRCI help

 

[EXTERNAL EMAIL]

--
You received this message because you are subscribed to the Google Groups "dirac-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dirac-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dirac-users/33d8dcde-2211-49ac-aff3-89b551696af3n%40googlegroups.com.

[Visscher, L. (Luuk)]

Dear Kirk, 

I can confirm that your Th run also fails with my standard installation of DIRAC23. I'm not a KRCI expert but given that this job ran with D17 I suspect that it has to do something with the introduction of atomic symmetry (of which I am not an expert either, this was implemented

by Ayaki Sunaga while he worked in Toulouse). 

To test this hypotheses I tricked the program into switching off atomic symmetry and run in D_infinity_h instead (what also would be the case with D17) and then the job proceeded a bit further until it ran out of memory on my machine. Can you try this work-around (see attached outpout that also has the inputs at the beginning).

best regards,

Luuk

To view this discussion on the web visit https://groups.google.com/d/msgid/dirac-users/C6BA73D1-74DA-4FDC-AB3C-C124ED8C0A69%40wsu.edu.

[Peterson, Kirk]

Dear Luuk,

 

thanks, I will certainly try this.

 

best regards,

 

-Kirk

To view this discussion on the web visit https://groups.google.com/d/msgid/dirac-users/8D3D6A4C-C021-4D22-B589-A9F52C4D6D73%40vu.nl.