What is the differences between ccman and ccman2?

Questions
#1

I want to run a cc2 calculation . And how better to set $rem to calculate lager molecules , as I have used ri cc2 so far.

If there are some errors in my input file.

image
image1920×2560 439 KB

#2

Is there a specific error that you are getting? If so, please paste the full input file using a preformatted text box (button that looks like </>), and similarly post the error message that you are getting.

#3 
$rem
METHOD cc2
BASIS  cc-pvdz
AUX_BASIS  rimp2-cc-pvdz
MEM_TOTAL  =  250000
cc_memory = 248000
THREADS    28
EE_SINGLETS = 1
EE_TRIPLETS = 1
IQMOL_FCHK  =  TRUE
NTO_PAIRS = 2
STATE_ANALYSIS = true
POP_MULLIKEN = true
$end

Q-Chem fatal error occurred in module /scratch/qcdevops/jenkins/workspace/build_qchem_linux_distrib/tags/qc602/qchem/ccman2/qchem/ccman2_main.C, line 26:

 Key 't' is not known.


 Please submit a crash report at q-chem.com/reporter

Many thanks.

#4

I can reproduce this with simple example of H2 molecule, and it seem to be related to STATE_ANALYSIS (because without that, the job runs to completion). Please contact Q-Chem support.

#5

Thanks John. I have one more question ,how to get NTO, i set molden_format true ,but did not get the molden.

$rem
METHOD eom-ccsd
BASIS  cc-pvdz
AUX_BASIS  rimp2-cc-pvdz
MEM_TOTAL  =  250000
cc_memory = 248000
THREADS    28
EE_SINGLETS = 1
EE_TRIPLETS = 1
IQMOL_FCHK  =  TRUE
NTO_PAIRS = 2
STATE_ANALYSIS = true
POP_MULLIKEN = true
cc_sp_t_conv = 4
cc_sp_e_conv = 6
cc_single_prec = 1
eom_single_prec = 1
cc_ref_prop = 1
CC_SP_DM = 1
cholesky_tol = 3
cc_trans_prop = 1
MOLDEN_FORMAT = TRUE
$end
#6

I think NTOs in Molden format may only work for CIS/TDDFT wave functions. The libwfa code can produce them for correlated wave functions; please check the manual.

#7

Orbitals in molden format are available for EOM-CCSD jobs but you need to remove IQMOL_FCHK = TRUE from your $rem section.
Both IQMOL_FCHK = TRUE and GUI = 2 keywords are incompatible with MODEN_FORMAT = TRUE

If I run your job like this I get the molden file for NTOs in the plots folder.

#8

Thank you very much.

#9

would be nice if there were a run-time check for this incompatibility, in the input parser.