sign of the transition moment in the CIS calculation
Just Got Here
2:04:19 PM PST - Mon, Dec 1st 2014
Hi there,

I am using CIS to calculate the electronic transition energy and transition dipole moment. In the output there are lines like (it is for h2o):

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

 Root   1 singlet b2             0.292135972 a.u.                7.9494 eV 

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

    Transition Moments    X  0.00000   Y -0.26644   Z -0.00000

    Transition Moments   XX  0.00000  XY  0.00000  XZ -0.00000

    Transition Moments   YY  0.00000  YZ  0.06726  ZZ  0.00000

    Dipole Oscillator Strength                         0.01383



    Occ.    5  b2  ---  Virt.    6  a1  -1.00006 X

I am wondering if I can trust the sign/direction of the transition moment in the output file (-0.26644 in this example). My impression (which might be wrong) is that the wave function has an arbitrary phase so that only the magnitude of the transition dipole moment is reliable. If my impression is correct, how can I determine the sign/direction of the transition dipole moment from the output information (presumably the transition dipole moment has a unique direction.)?

Thanks a lot!

Best regards,
Liang

Forum Regular
11:54:26 AM PST - Wed, Dec 3rd 2014
Hi Liang,

Only the magnitude, which goes into the calculation of the oscillator strength, is well defined.

Best,
-Niri

Just Got Here
10:44:39 AM PST - Fri, Dec 5th 2014
Hi Niri,

Thanks for your reply. But is there a way to infer the transition dipole moment direction from ab initio calculation? I am thinking something like transition density might be able to give a hint on this (naively I would think that the transition dipole might go from the negative parts to the positive parts in the transition density plot.), but I am really not sure about this (I am assuming that the transition density plot does not have the sign issue.). Thanks again.

Best,
Liang


Forum >> NWChem's corner >> General Topics