I wish to calculate the SOC between the ground and excited states of molecular Oxygen using SF-TDDFT. The ground state triplet would then be my reference for generating spin-flipped states and I can include this in the calculation of transition properties with WFA_REF_STATE = 0. However, SOCs are being calculated only among the excited roots. Is there a way to include the reference state in the procedure that computes SOCs? ($molecule and $rem sections below)

Kaushik, I have some doubts about this. For the case addressed by Pavel, the calculation was able to find the reference state among the spin-flipped states but I don’t think it works in mt case. My output reads:

---------------------------------------------------
SF-DFT Excitation Energies
(The first "excited" state might be the ground state)
---------------------------------------------------
Excited state 1: excitation energy (eV) = 1.7782
Total energy for state 1: -150.25536230 au
<S**2> : 2.0097
S( 1) --> S( 1) amplitude = 0.7065 alpha
S( 2) --> S( 2) amplitude = 0.7065 alpha
Excited state 2: excitation energy (eV) = 1.9766
Total energy for state 2: -150.24806990 au
<S**2> : 0.0086
S( 1) --> S( 1) amplitude = 0.7066 alpha
S( 2) --> S( 2) amplitude = -0.7066 alpha
Excited state 3: excitation energy (eV) = 1.9766
Total energy for state 3: -150.24806990 au
<S**2> : 0.0086
S( 1) --> S( 2) amplitude = 0.7066 alpha
S( 2) --> S( 1) amplitude = 0.7066 alpha
Excited state 4: excitation energy (eV) = 2.1791
Total energy for state 4: -150.24063001 au
<S**2> : 0.0086
S( 1) --> S( 2) amplitude = -0.7067 alpha
S( 2) --> S( 1) amplitude = 0.7067 alpha
Excited state 5: excitation energy (eV) = 8.9422
Total energy for state 5: -149.99209155 au
<S**2> : 1.0081
D( 7) --> S( 2) amplitude = 0.9993

All roots are far above my reference. What do you think?