Crash with NO_REORIENT

Questions
#1

Hello,

I have been trying to run a CVS-EOM-CCSD calculation for pyrazine with NO_REORIENT=TRUE, and I have encountered the following error:

--------------------------------------------------------------
User input:
--------------------------------------------------------------
$molecule
0 1
N        0.000100  -0.000000    1.423457
C        0.000100   1.137637    0.703525
C        0.000100   1.137637   -0.703525
N        0.000100   0.000000   -1.423457
C       -0.000100  -1.137637   -0.703525
C       -0.000100  -1.137637    0.703525
H        0.000100   2.079246    1.261889
H        0.000100   2.079246   -1.261889
H       -0.000100  -2.079246   -1.261889
H       -0.000100  -2.079246    1.261889
$end

$rem
METHOD			= eom-ccsd
BASIS			= aug-cc-pVDZ
CVS_EE_SINGLETS		= [4,4,4,4,4,4,4,4]
N_FROZEN_CORE           = 2
MEM_TOTAL 		= 64000
MEM_STATIC 		= 2000
CC_MEMORY		= 48000
CC_BACKEND		= XM
NO_REORIENT		= TRUE
$end
--------------------------------------------------------------
 ----------------------------------------------------------------
             Standard Nuclear Orientation (Angstroms)
    I     Atom           X                Y                Z
 ----------------------------------------------------------------
    1      N       0.0001000000     0.0000000000     1.4234570000
    2      C       0.0001000000     1.1376370000     0.7035250000
    3      C       0.0001000000     1.1376370000    -0.7035250000
    4      N       0.0001000000     0.0000000000    -1.4234570000
    5      C      -0.0001000000    -1.1376370000    -0.7035250000
    6      C      -0.0001000000    -1.1376370000     0.7035250000
    7      H       0.0001000000     2.0792460000     1.2618890000
    8      H       0.0001000000     2.0792460000    -1.2618890000
    9      H      -0.0001000000    -2.0792460000    -1.2618890000
   10      H      -0.0001000000    -2.0792460000     1.2618890000
 ----------------------------------------------------------------

 Q-Chem fatal error occurred in module 0, line   72:

 Bad point group code in PointGroupName

I have seen that a while ago this was a known bug. Is there any solution to it now beyond turning off the symmetry?

I need to get the symmetry of the core excited states, and it must be in the given orientation so that it is consistent with the symmetry irreps assigned by Gaussian to the normal modes. I can manually map the irreps between the different standard orientations in Gaussian and Q-Chem. Still, it would be smoother if I could get it directly from the output.

Another solution would be to get the symmetry of the normal modes from Q-Chem, but is that possible? I haven’t found it in the manual.

Thank you in advance for any advice!

#2

This was “fixed” in Q-Chem 6.2.2 (Dec. 2024) by automatically turning symmetry off when the user requests NO_REORIENT = TRUE. The underlying issue is that Q-Chem’s symmetry code needs to rotate the molecule to determine the point group but you’ve disabled that. Is there a reason to want NO_REORIENT = TRUE?

#3

Yes, I would need to get the symmetry labels for excited states in that specific orientation. Ultimately, I intend to use the symmetry of normal modes (from Gaussian) together with the symmetry of the core excited states. Thus, they need to use the same molecular orientation to be consistent. I can manually figure out the equivalences, but this is not optimum. As I said, another option would be to get the symmetry of the normal modes from Q-Chem, but I haven’t found the way to do it.

#4

Why can’t you just use the Standard Nuclear Orientation?

 ----------------------------------------------------------------
             Standard Nuclear Orientation (Angstroms)
    I     Atom           X                Y                Z
 ----------------------------------------------------------------
    1      N      -0.0000000049     0.0000666667     1.4234570000
    2      C      -1.1376370018    -0.0000171376     0.7035250000
    3      C      -1.1376370018    -0.0000171376    -0.7035250000
    4      N      -0.0000000049     0.0000666667    -1.4234570000
    5      C       1.1376370067    -0.0000495291    -0.7035250000
    6      C       1.1376370067    -0.0000495291     0.7035250000
    7      H      -2.0792459993    -0.0000865014     1.2618890000
    8      H      -2.0792459993    -0.0000865014    -1.2618890000
    9      H       2.0792460042     0.0000198347    -1.2618890000
   10      H       2.0792460042     0.0000198347     1.2618890000
 ----------------------------------------------------------------

My philosophy has always been that symmetry is a parlor trick that small molecules play to seem more interesting than they really are. Any universal explanation also needs to account for C1 geometries.