Getting whole set of OPDMs and Dyson orbitals

How To
#1

Dear Q-Chem developers (and users),

I’d like to calculate a certain numbers of EE singlets, EE/SF triplets, IP_BETA doublets and EA_ALPHA doublets, all from the singlet CCSD reference.

In addition to energies of the states, I’d like to get all (non-zero) alpha and beta state and transition densities of states/between pairs of states in fchk file. Here, I face two issues:

  1. The CC_TRANS_PROP keyword gives either only ground-to-excited (for value 1) or excited-to-excited (for value 2) transition densities, and I need both. Is the only way to get both to repeat the same calculation twice, once with CC_TRANS_PROP = 1 and once with CC_TRANS_PROP = 2? If that’s the case, can Q-Chem dump T and R amplitudes to some file in the first call, and then read them and only do the different analysis in the second call? I don’t see that option in the manual.

  2. I’d like to calculate triplets with MS = 1, so I assume SF_STATES is the option I need, rather than EE_TRIPLETS (which gives MS = 0). I want this because then I get alpha/beta T(MS=1) → T(MS=1) state/transition densities, which cannot be constructed by Wigner-Eckart theorem from the corresponding T(MS=0) → T(MS=0) densities (while vice versa can be done). However, if I do SF_STATES, then I don’t get S → T(MS=0) alpha/beta transition densities. Is the only way to get both T(MS=1) → T(MS=1) state/transition densities and S → T(MS=0) transition densities to specify both EE_TRIPLETS and SF_STATES?

Further I’d like to get all Dyson orbitals between (EOM-EE-)CCSD and EOM-IP/EA-CCSD states. Here, I face an issue that Q-Chem 6.2 simply freezes after SCF when I add CC_DO_DYSON to $rem, if STATE_ANALYSIS is also present. I assume this is a bug, and the only quick work-around I see is to do the calculation of all the states twice, once to get densities, and once to get Dyson orbitals? If that’s the case, dumping/reading the amplitudes would again be useful.

Thanks in advance for the help.

Best regards,
Tomislav Piteša

#2
  1. Unfortunately, you cannot compute CC-to-EOMCC and EOMCC-to-EOMCC transition properties together currently with the CC_TRANS_PROP setting, which I thought was already implemented. There is no mechanism of dumping T and R (you also need L) amplitudes to a file and read it for a subsequent calculation and it’s not planned either. Maybe, for now, consider setting the $trans_prop section described in the manual. See illustration below.
$molecule
0 1
O
H 1 1.1
H 1 1.1 2 110.
$end

$rem
method = eom-ccsd
basis = sto-3g
ee_states = [1,1,1,1]
cc_state_to_opt = [0,0]
cc_trans_prop = 1
cc_ref_prop = 1
cc_eom_prop = 1
$end

$trans_prop
state_list
ref
ee_singlets 1 1
ee_singlets 2 1
ee_singlets 3 1
ee_singlets 4 1
end_list
calc dipole
$end

Could you post an example of the input file with Dyson oribitals, so I can double check if there is a bug?

#3

Thank you for reply. Apparently, “calc dipole” in $trans_prop does not produce any densities in fchk. Moreover, with it I get zero transition densities, no matter what I put for CC_TRANS_PROP.

Here is an example where CC_DO_DYSON causes freezing:

$molecule
0 1
O 1.535338855 0.000000000 -0.438858006
C 1.535331598 -0.000007025 0.767790994
H 1.535342484 0.937663512 1.362651452
H 1.535342484 -0.937656488 1.362672535
$end

$rem
BASIS STO-3G
METHOD EOM-CCSD
CCMAN2 true
CC_SYMMETRY false
IQMOL_FCHK true
EE_SINGLETS 2
EOM_IP_BETA 3
EOM_EA_ALPHA 4
STATE_ANALYSIS true
CC_REF_PROP true
CC_EOM_PROP true
CC_TRANS_PROP 1
CC_DO_DYSON true
$end