Terminate called after throwing an instance of 'std::runtime_error'

Questions
#1

Hello, does anyone encounter this problem in frequency calculation?
terminate called after throwing an instance of ‘std::runtime_error’
what(): hamiltonian: no capability for term scrf_hcore:fock_deriv
Aborted
Error: in the serial run

Input file:
$rem
!QSYS wt=6h
jobtype freq
method CAM-B3LYP
basis def2-SVP
mem_total 200000
threads 16
SOLVENT_METHOD = PCM
$end

$pcm
theory iefpcm
$end

$solvent
Dielectric 1.889
$end

$molecule
0 1
C -2.9896453878 3.6762459123 -0.1356192182
C -3.7343983450 2.8438072706 -0.9685760015
C -2.1702826718 3.1078697204 0.8374484395
C -3.6559311490 1.4610183380 -0.8399185865
C -2.1002456054 1.7264981648 0.9841370912
C -2.8398098168 0.8860935457 0.1424186749
N -2.7620777357 -0.5203989398 0.2828371535
C -3.9348071905 -1.3076167274 0.1940242643
C -1.5076919764 -1.1396754390 0.5007758651
C -5.1253879143 -0.8834229187 0.7979238684
C -6.2780876717 -1.6558209008 0.7005496197
C -6.2627695739 -2.8699378786 0.0172514904
C -5.0781100718 -3.2997285669 -0.5769953509
C -3.9256821490 -2.5247463722 -0.4991036695
C -0.3740200353 -0.7199318928 -0.2067554493
C 0.8583347398 -1.3254685873 0.0154726361
C 0.9802342067 -2.3690086119 0.9305495440
C -1.3790277169 -2.1823978462 1.4269335559
C -0.1468910016 -2.7949616911 1.6302466381
H -3.0477188020 4.7611819444 -0.2439401438
H -4.3773021811 3.2743088958 -1.7398675021
H -1.5850560555 3.7476348117 1.5021934576
H -4.2315986114 0.8141919553 -1.5038283013
H -1.4666203312 1.2874382413 1.7562358747
H -5.1424586271 0.0603196406 1.3451929200
H -7.1973548337 -1.3075067271 1.1772005807
H -7.1676976296 -3.4769485231 -0.0514336720
H -5.0504179744 -4.2466093958 -1.1212016662
H -3.0049365213 -2.8617607896 -0.9776062843
H -0.4656197999 0.0886698241 -0.9336277148
H 1.7316663624 -0.9840796948 -0.5450831881
H 1.9474466735 -2.8473890457 1.0973197243
H -2.2549243514 -2.5121424555 1.9877344388
H -0.0662365510 -3.6078846605 2.3556716111
$end

#2

Note that IEF-PCM solvation model does not have analytic Hessians. Does the problem persist if you replace this with C-PCM (which is the default)?

#3

Thanks. I made it after removing IEFPCM part in the input file. But the geometry was optimised through IEFPCM solvent mode according to literature. Should I keep the same parameters for consistency? Or just using the default C-PCM is also OK?

#4

If you are looking to compute frequencies then you should reoptimize using C-PCM because otherwise your molecule is not at a local minimum. Physically speaking, the difference between the tw structures is likely to be very small (but you will be able to verify), as discussed in this review: https://doi.org/10.1002/wcms.1519

I will take a look at why there wasn’t a better error message for this job.

#5

Update: This warning message at top of Q-Chem output file

!!********************************************!!
!!******************Warning*******************!!
!!Desired analytical derivatives not available!!
!!Finite difference job might take a long time!!
!!********************************************!!
!!********************************************!!

 2-order derivative to be evaluated numerically with 1-order analytical derivatives

tells you that analytic Hessians are not available and therefore frequencies will be computed as finite-difference of analytic gradients. I ran your input on a smaller molecule and it ran to completion without the error that you noticed, so I would take a look at the last SCF calculation before the error and see if it looks like there were any issues with that. However, the better solution is to switch to C-PCM and use analytic Hessians.