"density" keyword requests Gaussian to generate relaxed density for the state specified by "root" and calculate dipole moment based on this density. The dipole moment arrow drawn by GaussView also corresponds to this. By default (i.e. without "density"), the dipole moment is calculated for reference state of TDDFT (commonly ground state).

Also note that to accurately perform electronic excitation analysis by Multiwfn, you should add IOp(9/40=4), as mentioned in Section 3.21.A of Multiwfn manual.

I have submitted 4 jobs using the parameters below for the same td-dft input file (### denotes an abbreviation):

### TD(NStates=10) IOP(2/17=4) ###
### TD(NStates=10, root=1) IOP(2/17=4) ###
### TD(NStates=10, root=2) IOP(2/17=4) ###
### TD(NStates=10, root=3) IOP(2/17=4) ###

Gaussview was then used to visualise the dipole moments, and the results were identical to the outputs generated by your program, of which the excited states all shared identical dipole moment as of ground state.

Ground state electric dipole moment in X,Y,Z: -339.511661  288.423511    2.281808 a.u.

Transition electric dipole moment between ground state (0) and excited states (a.u.)
     i     j         X             Y             Z        Diff.(eV)   Oscil.str
     0     1     0.0000000     0.0000000     0.0000000     3.72460     0.00000
     0     2     0.0000000     0.0000000     0.0000000     4.23170     0.00000
     0     3     0.0000000     0.0000000     0.0000000     5.02000     0.00000
     0     4     0.0000000     0.0000000     0.0000000     5.97110     0.00000
     0     5     0.0000000     0.0000000     0.0000000     6.44810     0.00000
     0     6     0.0000000     0.0000000     0.0000000     6.65840     0.00000
     0     7     0.0000000     0.0000000     0.0000000     6.79220     0.00000
     0     8     0.0000000     0.0000000     0.0000000     7.20070     0.00000
     0     9     0.0000000     0.0000000     0.0000000     7.31220     0.00000
     0    10     0.0000000     0.0000000     0.0000000     7.37150     0.00000

However, when I resubmitted these jobs by adding in the "density" keyword (as shown below), the dipole moments looked differently in GaussView, however, Multiwfn still gave the same results as above.

### TD(NStates=10) density IOP(2/17=4) ###
### TD(NStates=10, root=1) density IOP(2/17=4) ###
### TD(NStates=10, root=2) density IOP(2/17=4) ###
### TD(NStates=10, root=3) density IOP(2/17=4) ###

I wonder if there is a way to directly compare/subtract the dipole moment of one output file from another of identical structure (say, Gaussian output file 2 (root=1, density) - Gaussian output file 1) (root=0, density) in Multiwfn.

Thanks again

Btw, yosoro!
没想到dalao也是拉拉人hh

There is a VMD script named "drawarrow.tcl" in examples\scripts\ folder of Multiwfn package. This script defines a customized command "drawarrow". After copying content of the script into VMD console window to run it, the "drawarrow" command will be ready to use, and then you can use it to plot an arrow to graphically exhibit a vector on molecular map. Please check Section 4.15.3 of Multiwfn manual for example on how to use this command.

I would like know how to calculate the transition dipole moment (say S1 - S0) from a Gaussian td-dft output file in Multiwfn, and hence to visualise the transition dipoles (preferably an arrow on the molecule) by combi.

I tried with the example "D-pi-A" given in examples/excit/; in Multiwfn I did 18 -> 5 -> 4 and the output file dipmom looked like this:

Note: The electric dipole moments shown below include both nuclear charge and electronic contributions
Ground state electric dipole moment in X,Y,Z:   -0.000011   -0.000006   -0.000013 a.u.

Excited state electric dipole moments (a.u.):
  State         X             Y             Z        exc.(eV)    exc.(nm)
     1     -0.000011     -0.000006     -0.000013      3.9069      317.35
     2     -0.000011     -0.000006     -0.000013      4.0624      305.20
     3     -0.000011     -0.000006     -0.000013      4.4166      280.72
     4     -0.000011     -0.000006     -0.000013      4.7912      258.77
     5     -0.000011     -0.000006     -0.000013      4.8872      253.69

I then went on to proceed with my td-dft file and I got an error:

Error: This function is unavailable when singlet and triplet excited states are simultaneously occurred

This is my td-dft file input:

%chk=L1.smd.dichloromethane_td-dft.chk
# wB97XD/Def2TZVP SCF=Tight INT(grid=ultrafine) TD(50-50,NStates=10) IOP(2/17=4) SCRF=(SMD,Solvent=dichloromethane)

L1.smd.dichloromethane_td-dft

0  1
C  1.1128330848 -1.2411154091  1.2670908809
C  0.2246856582 -2.2498778555  1.5939734441
...(abbreviated)

Do I need to modify my parameters, say singlet only etc, in order to work this work? Do I also need to specify root and density as well (e.g. TD(50-50,NStates=1, root=4) density)

One step forward, say if we have "dipmom" now, what's the best way to visualise this along side with my xyz structure?

Thanks and cheers.