1 Is there any reason to use guess=save? I think it is not needed.

2 Please note that Gaussian automatically reorientates your system to standard orientation during calculation. Perhaps you want to use "nosymm" keyword to disable this behavior.

Best,

Tian

I'm trying to extract data from section 4.48.11 "Plot transition dipole moment vector contributed by molecular fragments as arrows".
My molecule exhibites a donor-heteroaromatic core-acceptor(=D-pi-A) structure, and I'm approaching one molecule in two cases, but I'm having a problem.

In the first case, the TDM arrow appeared in the direction I thought.
[1. {# opt=calcfc freq b3lyp/6-31++g(d,p) guess=save} /
2. TD-DFT{# td=(singlets,nstates=6,root=1) b3lyp/6-31++g(d,p) guess=save geom=check}]

In the second case, it appeared in the opposite direction.
[1. {# b3lyp/6-31++g(d,p) guess=save geom=connectivity} /
2. TD-DFT{# td=(singlets,nstates=6,root=1) b3lyp/6-31++g(d,p) guess=save geom=check},based on actual molecular structure measured by XRC]

Looking at the isosurface hole and electron distribution, the forward arrow (D to A) seems reasonable, but when checked with VMD, it is the opposite arrow (A to D).

So I checked the log.file and was able to get the following TDM:

Ground to excited state transition electric dipole moments (Au):
       state          X           Y           Z        Dip. S.      Osc.
         1         3.2066      0.1893     -0.0835     10.3252      0.7153

-------------------------------------------------------------------------------------
Ground to excited state transition electric dipole moments (Au):
       state          X           Y           Z        Dip. S.      Osc.
         1        -3.2825     -0.1505      0.0298     10.7986      0.7658

In the second case, is there a simple Tk console command that can change the direction expressed as VMD to the forward direction? or is there a fundamental solution?

Thank you for your reply.
and Happy New Year!

Jang