Unexpected high value for coupling in AIFDEM calculation

Questions
#1

Hello everyone.
I would like to calculate electronic coupling for PTCDI dimers. I did some calculation and got results that coupling between S1 states are very high numbers (2000-3000 cm^-1). In my system there is a double hydrogen bond, and I thought this might be a reason. And I did another calculation for pentacene dimer. Again the same situation. Then I looked at the example calculations on the manual for water cluster: in the output file attached there, there are coupling values of even several Hartrees. I would assume that this couldn’t be correct, right?.
Could anybody please help with this problem? Maybe I cannot interpret these results correctly, or something else.
Thank you in advance.

Best regards,
Tolib

#2

Please provide a complete input file for this calculation. Use the preformatted text environment, </>

#3

Here is my input:

$molecule
0 1
--PTCDI 0
   0 1
H      3.480049    0.704292    6.522990
H      3.398402   -1.249604    6.522990
H     -3.158075    3.445305    6.522990
H      3.546431    3.163756    6.522990
H     -3.446002   -3.424048    6.522990
H      3.257632   -3.706081    6.522990
H      0.330469    6.545219    6.522990
H     -0.230910   -6.805995    6.522990
O     -1.983489    5.684252    6.522990
O     -2.464167   -5.753373    6.522990
O      2.563722    5.492612    6.522990
O      2.083046   -5.945014    6.522990
N      0.288182    5.529975    6.522990
N     -0.188150   -5.791619    6.522990
C      0.110916    1.303158    6.522990
C     -0.010002   -1.564328    6.522990
C     -1.167823    0.657362    6.522990
C     -1.229136   -0.812648    6.522990
C      1.330049    0.551481    6.522990
C      1.268259   -0.917650    6.522990
C      0.170400    2.732398    6.522990
C     -0.069963   -2.992675    6.522990
C     -1.029713    3.490935    6.522990
C      1.430504    3.387439    6.522990
C     -1.329591   -3.648599    6.522990
C      1.130150   -3.751227    6.522990
C     -2.317525    1.448837    6.522990
C     -2.441493   -1.505536    6.522990
C      2.541920    1.245253    6.522990
C      2.417566   -1.710480    6.522990
C     -2.252980    2.845569    6.522990
C      2.594825    2.642503    6.522990
C     -2.494397   -2.902795    6.522990
C      2.353021   -3.107213    6.522990
C     -0.989822    4.973237    6.522990
C     -1.413839   -5.128738    6.522990
C      1.514282    4.868460    6.522990
C      1.089380   -5.234013    6.522990
H     -3.297957    0.989317    6.522990
H     -3.379604   -0.964580    6.522990
--PTCDI 1
   0 1
H      1.895809   14.932566    6.522990
H      1.814162   12.978670    6.522990
H     -4.742315   17.673579    6.522990
H      1.962191   17.392030    6.522990
H     -5.030242   10.804225    6.522990
H      1.673392   10.522192    6.522990
H     -1.253771   20.773493    6.522990
H     -1.815150    7.422278    6.522990
O     -3.567729   19.912526    6.522990
O     -4.048407    8.474900    6.522990
O      0.979482   19.720885    6.522990
O      0.498806    8.283260    6.522990
N     -1.296058   19.758248    6.522990
N     -1.772390    8.436655    6.522990
C     -1.473324   15.531431    6.522990
C     -1.594242   12.663946    6.522990
C     -2.752063   14.885635    6.522990
C     -2.813376   13.415626    6.522990
C     -0.254191   14.779754    6.522990
C     -0.315981   13.310624    6.522990
C     -1.413840   16.960671    6.522990
C     -1.654203   11.235598    6.522990
C     -2.613953   17.719209    6.522990
C     -0.153736   17.615712    6.522990
C     -2.913831   10.579675    6.522990
C     -0.454090   10.477046    6.522990
C     -3.901765   15.677110    6.522990
C     -4.025733   12.722737    6.522990
C      0.957681   15.473526    6.522990
C      0.833326   12.517793    6.522990
C     -3.837220   17.073843    6.522990
C      1.010585   16.870777    6.522990
C     -4.078636   11.325478    6.522990
C      0.768781   11.121060    6.522990
C     -2.574062   19.201510    6.522990
C     -2.998078    9.099536    6.522990
C     -0.069957   19.096733    6.522990
C     -0.494860    8.994261    6.522990
H     -4.882197   15.217591    6.522990
H     -4.963844   13.263693    6.522990
$end

$rem
   BASIS                6-31G*
   EXCHANGE             B3LYP
   IQMOL_FCHK           FALSE
   CIS_N_ROOTS          1
   CIS_TRIPLETS         FALSE
   AIFDEM_CTSTATES      TRUE
   AIFDEM               TRUE
   AIFDEM_EMBED_RANGE   FULL_QM
   AIFDEM_NTOTHRESH     99
   NTO_PAIRS            1
   MEM_TOTAL            40000
$end

Here I used FULL_QM, and pretty big system. I have another example for just water dimer, which gave the same high coupling values:

$molecule
0 1
--H2O 0
   0 1
   O      1.74078     1.59716    -1.49814
   H      2.22908     2.18316    -2.08914
   H      0.88038     2.04726    -1.32684
--H2O 1
   0 1
   O      1.388768   -1.849704   -1.917340
   H      1.568668   -0.890104   -1.890440
   H      1.759368   -2.186304   -1.077040
$end

$rem
   BASIS                6-31G*
   METHOD               wB97X-D
   CIS_N_ROOTS          3
   CIS_TRIPLETS         FALSE
   AIFDEM               TRUE
   AIFDEM_EMBED_RANGE   FULL_QM
   AIFDEM_NTOTHRESH     90
   NTO_PAIRS            1
   MEM_TOTAL		8000
$end
#4

Did anybody experience such problem?
Is this just my mistake or bug?

Here I share the part of the output for above input:

*** Ordered AIFDEM Exciton-Site Basis States Included in this Job ***

    [0] |S0>        [1] |Frg:1 S1>        [2] |Frg:2 S1>        [3] |CT D:1 A:2>
    [4] |CT D:2 A:1>


 ***AIFDEM Overlap Matrix in Exciton-Site Basis***

            1           2           3           4           5
    1   1.0000000   0.0016068  -0.0016041  -0.0000347  -0.0000342
    2   0.0016068   0.9999627   0.0000131   0.0000081  -0.0000297
    3  -0.0016041   0.0000131   0.9999627   0.0000298  -0.0000084
    4  -0.0000347   0.0000081   0.0000298   1.0000000   0.0000000
    5  -0.0000342  -0.0000297  -0.0000084   0.0000000   1.0000000

 ***AIFDEM Hamiltonian Matrix in Exciton-Site Basis***

            1           2           3           4           5
    1-2645.9669271  -4.2519130   4.2447313   0.0917381   0.0905788
    2  -4.2519130-2645.7318494  -0.0369144  -0.0215693   0.0786244
    3   4.2447313  -0.0369144-2645.7318309  -0.0789743   0.0222746
    4   0.0917381  -0.0215693  -0.0789743-2645.8029617  -0.0000023
    5   0.0905788   0.0786244   0.0222746  -0.0000023-2645.8029391
#5

It’s a bit unclear but are you referring to the diagonal elements or the off-diagonal ones when you say the couplings are large?

#6

What version of Q-Chem are you running? With v. 6.2.2, I get the following when using your input:

 ***AIFDEM Hamiltonian Matrix in Exciton-Site Basis***

            1           2           3           4           5           6
    1-152.0181799  -0.0689596   0.0019841  -0.0928897  -0.0200671   0.0235396
    2  -0.0689596-151.6170344  -0.0037664  -0.0514769   0.0057106  -0.0063614
    3   0.0019841  -0.0037664-151.5943913   0.0038836  -0.0002352   0.0003196
    4  -0.0928897  -0.0514769   0.0038836-151.1884934   0.0062192  -0.0069423
    5  -0.0200671   0.0057106  -0.0002352   0.0062192-151.6124073  -2.1748275
    6   0.0235396  -0.0063614   0.0003196  -0.0069423  -2.1748275-151.5863331
    7   0.3498633  -0.0763555   0.0029080  -0.1265199   0.0024340   0.0063971
            7
    1   0.3498633
    2  -0.0763555
    3   0.0029080
    4  -0.1265199
    5   0.0024340
    6   0.0063971
    7-151.1816553
1 Like
#7

I am referring off-diagonal elements. For example, say, matrix element [2,3] is -0.0369144, which is strangely high coupling between S1 states of this molecules.

#8

Not sure. But please note that these are nonorthogonal basis functions so the couplings may be larger than what you are used to seeing in an orthogonal basis.