TDNonEq keyword

Hi
I want to obtain the vertical absorption and emission energies. So, firstly, for the absorption, I put the job with input:
$molecule
0 3
coordinates
$end

$rem
BASIS = cc-pVDZ
EXCHANGE = omegaB97XD
OMEGA = 104
GUI = 2
JOB_TYPE = optimization
MAX_CIS_CYCLES = 700
SCF_CONVERGENCE = 8
MAX_SCF_CYCLES = 700
THRESH = 14
UNRESTRICTED = true
SPIN_FLIP = 1
CIS_N_ROOTS = 4
SYMMETRY_IGNORE = true
SYMMETRY = false
CIS_STATE_DERIVATIVE = 3
SOLVENT_METHOD = PCM
$end

$pcm
Tdnoneq 2
$end

$solvent
DIELECTRIC 32.613000
OPTICALDIELECTRIC 1.765709
$end

I am getting error:

A unrestricted SCF calculation will be
performed using DIIS
SCF converges when DIIS error is below 1.0e-08

Cycle Energy DIIS error

/root/qchem/bin/qchem: line 126: 10666 Segmentation fault (core dumped) ${QCPROG_S} ${inp} ${scr}
remove work dirs /root/qchem/scratch/qchem10649.0 – /root/qchem/scratch/qchem10649.0

what can I do to resolve this error. can anyone help me?

please provide a full input file (including $molecule) that exhibits the problem.

Hi
Please find the complete input below:

$molecule
0 3
O
C 1 1.341034
O 2 1.214473 1 123.691210
C 1 1.439672 2 116.137749 3 -1.390087 0
H 4 1.094090 1 105.431473 2 -179.496128 0
H 4 1.097747 1 110.510775 2 61.161311 0
H 4 1.098139 1 110.465165 2 -60.271839 0
C 2 1.499292 1 111.401707 3 -179.906358 0
C 8 1.364878 2 125.011777 1 -73.099182 0
H 9 1.094594 8 113.525968 2 178.782781 0
C 9 1.441174 8 131.851052 2 -1.548531 0
C 11 1.417793 9 125.447603 8 -9.504728 0
H 12 1.089064 11 120.611031 9 -1.894129 0
C 12 1.380544 11 121.932884 9 -179.692164 0
H 14 1.088216 12 118.466166 11 179.265234 0
C 11 1.418065 9 118.043033 8 171.632929 0
H 16 1.092480 11 118.853040 9 0.086019 0
C 16 1.381614 11 122.512799 9 -179.655967 0
H 18 1.088149 16 118.721893 11 179.672254 0
C 18 1.426076 16 120.722229 11 -0.835087 0
N 20 1.359433 18 121.623856 16 -179.711990 0
C 21 1.456134 20 120.286973 18 0.693903 0
H 22 1.094502 21 108.928102 20 179.612612 0
H 22 1.101982 21 111.629774 20 -61.206765 0
H 22 1.102548 21 111.786173 20 60.248935 0
C 21 1.456326 20 120.409372 18 178.122072 0
H 26 1.094523 21 108.900259 20 -179.402753 0
H 26 1.101901 21 111.724819 20 61.381145 0
H 26 1.102500 21 111.731552 20 -60.190882 0
C 8 1.478218 2 117.444722 1 107.049855 0
O 30 1.220310 8 125.484995 2 176.064154 0
O 30 1.346631 8 111.461093 2 -5.053075 0
C 32 1.435789 30 116.445014 8 179.916088 0
H 33 1.094256 32 105.482853 30 179.763470 0
H 33 1.098178 32 110.761673 30 60.521702 0
H 33 1.098386 32 110.563507 30 -61.069459 0
$end

$rem
BASIS = cc-pVDZ
EXCHANGE = omegaB97XD
OMEGA = 104
GUI = 2
JOB_TYPE = optimization
MAX_CIS_CYCLES = 700
SCF_CONVERGENCE = 8
MAX_SCF_CYCLES = 700
THRESH = 14
UNRESTRICTED = true
SPIN_FLIP = 1
CIS_N_ROOTS = 4
SYMMETRY_IGNORE = true
SYMMETRY = false
CIS_STATE_DERIVATIVE = 3
SOLVENT_METHOD = PCM
$end

$pcm
Tdnoneq 2
$end

$solvent
DIELECTRIC 32.613000
OPTICALDIELECTRIC 1.765709
$end

okay, I can reproduce the crash with the current Q-Chem trunk but I don’t have any guess what the problem is, other than the fact that it’s connected to TDNonEq=2 in $pcm, because when I remove that the job runs normally. This is not the non-equil model implemented by my group or my collaborators, so if you really want to run this one then please contact Q-Chem support for assistance. Alternatively, there are other non-equil models for use with CIS/TDDFT, as described in the manual here:
https://manual.q-chem.com/5.2/Ch12.S2.SS2.html#SSS3

and in the literature here:
https://doi.org/10.1002/wcms.1519

The TDNonEq=2 keyword is only utilized for calculating a single point. Since this task involves an optimized structure, it is recommended to remove the TDNonEq keyword. The vertical absorption and emission energies were calculated based on the optimized structure using TDNonEq keyword. Please refer to the Q-CHem manual. Before using TdNonEq=2 for vertical excition energies , TdNonEq=1 must be performed in the same directory.

Thanks @Haisheng and @jherbert
I read from qchem manual.
It was not that much clear to me from there
Can you please explain a little bit here
I firstly optimized my structure using the SF-TDDFT method without using any TdNonEq keyword.
After that, what I have to do?
Can you please explain the sequence, are all these 4 steps for TdNonEq keywords used for Single point calculation?

By taking the already optimized structure. I made this input
$molecule
0 3
O
C 1 1.341034
O 2 1.214473 1 123.691210
C 1 1.439672 2 116.137749 3 -1.390087 0
H 4 1.094090 1 105.431473 2 -179.496128 0
H 4 1.097747 1 110.510775 2 61.161311 0
H 4 1.098139 1 110.465165 2 -60.271839 0
C 2 1.499292 1 111.401707 3 -179.906358 0
C 8 1.364878 2 125.011777 1 -73.099182 0
H 9 1.094594 8 113.525968 2 178.782781 0
C 9 1.441174 8 131.851052 2 -1.548531 0
C 11 1.417793 9 125.447603 8 -9.504728 0
H 12 1.089064 11 120.611031 9 -1.894129 0
C 12 1.380544 11 121.932884 9 -179.692164 0
H 14 1.088216 12 118.466166 11 179.265234 0
C 11 1.418065 9 118.043033 8 171.632929 0
H 16 1.092480 11 118.853040 9 0.086019 0
C 16 1.381614 11 122.512799 9 -179.655967 0
H 18 1.088149 16 118.721893 11 179.672254 0
C 18 1.426076 16 120.722229 11 -0.835087 0
N 20 1.359433 18 121.623856 16 -179.711990 0
C 21 1.456134 20 120.286973 18 0.693903 0
H 22 1.094502 21 108.928102 20 179.612612 0
H 22 1.101982 21 111.629774 20 -61.206765 0
H 22 1.102548 21 111.786173 20 60.248935 0
C 21 1.456326 20 120.409372 18 178.122072 0
H 26 1.094523 21 108.900259 20 -179.402753 0
H 26 1.101901 21 111.724819 20 61.381145 0
H 26 1.102500 21 111.731552 20 -60.190882 0
C 8 1.478218 2 117.444722 1 107.049855 0
O 30 1.220310 8 125.484995 2 176.064154 0
O 30 1.346631 8 111.461093 2 -5.053075 0
C 32 1.435789 30 116.445014 8 179.916088 0
H 33 1.094256 32 105.482853 30 179.763470 0
H 33 1.098178 32 110.761673 30 60.521702 0
H 33 1.098386 32 110.563507 30 -61.069459 0
$end

$rem
BASIS = cc-pVDZ
EXCHANGE = omegaB97XD
OMEGA = 104
GUI = 2
JOB_TYPE = sp
MAX_CIS_CYCLES = 700
SCF_CONVERGENCE = 8
MAX_SCF_CYCLES = 700
THRESH = 14
UNRESTRICTED = true
SPIN_FLIP = 1
CIS_N_ROOTS = 10
SYMMETRY_IGNORE = true
SYMMETRY = false
CIS_STATE_DERIVATIVE = 1
SOLVENT_METHOD = PCM
$end

$pcm
TdNonEq 1
$end

$solvent
DIELECTRIC 32.613000
OPTICALDIELECTRIC 1.765709
$end

@@@
$molecule
READ
$end

$rem
BASIS = cc-pVDZ
EXCHANGE = omegaB97XD
OMEGA = 104
GUI = 2
JOB_TYPE = sp
MAX_CIS_CYCLES = 700
SCF_CONVERGENCE = 8
MAX_SCF_CYCLES = 700
THRESH = 14
UNRESTRICTED = true
SPIN_FLIP = 1
CIS_N_ROOTS = 10
SYMMETRY_IGNORE = true
SYMMETRY = false
CIS_STATE_DERIVATIVE = 1
SOLVENT_METHOD = PCM
$end

$pcm
TdNonEq 2
$end

$solvent
DIELECTRIC 32.613000
OPTICALDIELECTRIC 1.765709
$end

1st part of this job was completed, but in the 2nd part I got this error:

A unrestricted SCF calculation will be
performed using DIIS
SCF converges when DIIS error is below 1.0e-08

Cycle Energy DIIS error

/root/qchem/bin/qchem: line 126: 31313 Segmentation fault (core dumped) ${QCPROG_S} ${inp} ${scr}

I mentioned above that this is a bug, even for single-point calculations. I have submitted a ticket. Meanwhile, why not try the alternative non-equil TDDFT + PCM methods that I cited above?

Currently, SF-TDDFT methods are not compatible with the TdNonEq calculation.

TdNonEq calculation need to add “RPA = 2” and “CIS_RELAXED_DENSITY TRUE” in $rem as well as “Theory SSVPE”, “ChargeSeparation Pekar”, “StateSpecific Perturb” and “TdNonEq 1” in $pcm. Please refer Example 11.14 of Qchem manual.

The calculation results are printed in “Energy Summary Based on Constrained Equilibrium Principle”. Total energy for excited-state x (Li) is from the constrained equilibrium principle.

In our future work, we will make further improvements to this aspect specifically for SF-TDDFT methods.

I.e.,
$molecule
0 3
O
C 1 1.341034
O 2 1.214473 1 123.691210
C 1 1.439672 2 116.137749 3 -1.390087 0
H 4 1.094090 1 105.431473 2 -179.496128 0
H 4 1.097747 1 110.510775 2 61.161311 0
H 4 1.098139 1 110.465165 2 -60.271839 0
C 2 1.499292 1 111.401707 3 -179.906358 0
C 8 1.364878 2 125.011777 1 -73.099182 0
H 9 1.094594 8 113.525968 2 178.782781 0
C 9 1.441174 8 131.851052 2 -1.548531 0
C 11 1.417793 9 125.447603 8 -9.504728 0
H 12 1.089064 11 120.611031 9 -1.894129 0
C 12 1.380544 11 121.932884 9 -179.692164 0
H 14 1.088216 12 118.466166 11 179.265234 0
C 11 1.418065 9 118.043033 8 171.632929 0
H 16 1.092480 11 118.853040 9 0.086019 0
C 16 1.381614 11 122.512799 9 -179.655967 0
H 18 1.088149 16 118.721893 11 179.672254 0
C 18 1.426076 16 120.722229 11 -0.835087 0
N 20 1.359433 18 121.623856 16 -179.711990 0
C 21 1.456134 20 120.286973 18 0.693903 0
H 22 1.094502 21 108.928102 20 179.612612 0
H 22 1.101982 21 111.629774 20 -61.206765 0
H 22 1.102548 21 111.786173 20 60.248935 0
C 21 1.456326 20 120.409372 18 178.122072 0
H 26 1.094523 21 108.900259 20 -179.402753 0
H 26 1.101901 21 111.724819 20 61.381145 0
H 26 1.102500 21 111.731552 20 -60.190882 0
C 8 1.478218 2 117.444722 1 107.049855 0
O 30 1.220310 8 125.484995 2 176.064154 0
O 30 1.346631 8 111.461093 2 -5.053075 0
C 32 1.435789 30 116.445014 8 179.916088 0
H 33 1.094256 32 105.482853 30 179.763470 0
H 33 1.098178 32 110.761673 30 60.521702 0
H 33 1.098386 32 110.563507 30 -61.069459 0
$end

$rem
BASIS = cc-pVDZ
EXCHANGE = omegaB97XD
OMEGA = 104
GUI = 2
JOB_TYPE = sp
MAX_CIS_CYCLES = 700
SCF_CONVERGENCE = 8
MAX_SCF_CYCLES = 700
THRESH = 14
UNRESTRICTED = true
CIS_N_ROOTS = 10
SYMMETRY_IGNORE = true
SYMMETRY = false
CIS_STATE_DERIVATIVE = 1
SOLVENT_METHOD = PCM
pcm_print =1
RPA = 2
CIS_RELAXED_DENSITY TRUE
$end

$pcm
Theory SSVPE
ChargeSeparation Pekar
StateSpecific Perturb
TdNonEq 1
$end

$solvent
DIELECTRIC 32.613000
OPTICALDIELECTRIC 1.765709
$end

@@@

$molecule
READ
$end

$rem
BASIS = cc-pVDZ
EXCHANGE = omegaB97XD
OMEGA = 104
GUI = 2
JOB_TYPE = sp
MAX_CIS_CYCLES = 700
SCF_CONVERGENCE = 8
MAX_SCF_CYCLES = 700
THRESH = 14
UNRESTRICTED = true
CIS_N_ROOTS = 10
SYMMETRY_IGNORE = true
SYMMETRY = false
CIS_STATE_DERIVATIVE = 1
SOLVENT_METHOD = PCM
pcm_print =1
RPA = 2
CIS_RELAXED_DENSITY TRUE
$end

$pcm
Theory SSVPE
ChargeSeparation Pekar
StateSpecific Perturb
TdNonEq 2
$end

$solvent
DIELECTRIC 32.613000
OPTICALDIELECTRIC 1.765709
$end

I have added a backstop to prevent TdNonEq from being used with spin flip and to print an informative error message, starting in Q-Chem 6.1.1.

Please see the NonEquilibrium keyword for the $pcm section, which does work with spin flip.