Dipole moment of electron-attached state in ORCA EA-EOM-CCSD

Question

I am trying to calculate the electric dipole moment of some (neutral) doublet molecules using ORCA 5.0.3. The molecules of interest have a nice interpretation as having 1 electron attached to a closed-shell cation, so I'd like to use the EA-EOM-CCSD method. However, I am fairly certain that the electric dipole moments in the output file are those for the cation itself, not the properties of the targeted EA state.

Here is an example input file:

! EA-EOM-CCSD cc-pVTZ
%mdci
NRoots 5
IRoot 1
end

*xyz 1 1
   Mg  0.00 0.00 0.00
   F   0.00 0.00 1.750
*

The output file predicts a dipole moment around 3.74 a.u., which is actually quite accurate for the reference state (the cation). But it is way too large for the neutral. See here:

                           *** MDCI DENSITY ***

------------------------------------------------------------------------------
                       ORCA ELECTRIC PROPERTIES CALCULATION
------------------------------------------------------------------------------

Dipole Moment Calculation                       ... on
Quadrupole Moment Calculation                   ... off
Polarizability Calculation                      ... off
GBWName                                         ... mgf.gbw
Electron density                                ... mgf.mdcip
The origin for moment calculation is the CENTER OF MASS  = ( 0.000000,  0.000000  1.450864)

-------------
DIPOLE MOMENT
-------------
                                X             Y             Z
Electronic contribution:      0.00000       0.00000      -3.04350
Nuclear contribution   :      0.00000       0.00000      -0.70497
                        -----------------------------------------
Total Dipole Moment    :      0.00000       0.00000      -3.74847
                        -----------------------------------------
Magnitude (a.u.)       :      3.74847
Magnitude (Debye)      :      9.52784

How can I force the electric property calculation to apply to the EA state, instead of the reference state?

Answer 1

Not a complete answer I'm afraid, but I don't think the current version of Orca (5.0.4) supports the calculation of any excited state properties at the EOM-CCSD level. This isn't explicitly stated in the manual that I can see, but it is hinted at in this section:

8.5.4.2 Capabilities

...

For the time being, the most useful information provided is the list of the excitation energies, the ionization potentials or the electron affinities. The ground to excited state transition moments are also available for the closed-shell implementation of EE-EOM-CCSD.

I don't know what 'EE-EOM-CCSD' is, perhaps a typo for 'EA-EOM-CCSD'? Either way, electronic dipole moments are not mentioned.

Dipole moments are supported for STEOM-CCSD however, where they are neatly printed in a section that looks like the following:

--------------------------------------------------------------------
               UNRELAXED EXCITED STATE DIPOLE MOMENTS
--------------------------------------------------------------------
               E(eV)     DX(au)      DY(au)      DZ(au)      |D|(D)
IROOT=  0:     0.000    0.817890    0.028171   -0.001944    2.080150
IROOT=  1:     4.962    0.994395    0.022802   -0.002337    2.528222
IROOT=  2:     5.051    0.252670    0.013795   -0.000642    0.643195
--------------------------------------------------------------------

Here's the input file I used, for reference (pyridine, in this case):

! STEOM-DLPNO-CCSD NOCOSX cc-pVDZ/C cc-pVDZ NormalPNO TightSCF
%Pal NProcs 8 end
%MaxCore 5000
%SCF
    AutoTRAH    True
    CNVDamp     False
    CNVZerner   False
    end
%mdci
    MaxIter     200
    NRoots      2
    DoTDM       true
    DoTriplet   false
    STEOMSOC    false
    DoDbfilter  false
    DoSTEOMNatTransOrb  True
    DoStoreSTEOM        True
    end
*xyz 0 1
N            -1.36068000    -0.04851000     0.00514000
C            -0.69236000     1.12548000     0.00316000
C             0.68982000     1.22568000     0.00009000
C             1.43473000     0.05098000    -0.00151000
C             0.77522000    -1.17369000     0.00026000
C            -0.61058000    -1.17202000     0.00377000
H            -1.31775000     2.01352000     0.00412000
H             1.17356000     2.19574000    -0.00108000
H             2.52016000     0.08962000    -0.00411000
H             1.32670000    -2.10691000    -0.00100000
H            -1.17125000    -2.10227000     0.00566000
*

But sadly there's no EA-EOM-CCSD equivalent for STEOM-CCSD (yet).

That being said, there's a new major version of Orca (6) scheduled for release in just over 2 weeks. I don't think they've announced the change-log yet, but perhaps the new version will include excited state properties for EOM-CCSD and/or EA-STEOM-CCSD. I'd recommend trying again in the new version once it's out, or see if STEOM-CCSD will work for you instead.