ADC(2) calculation error

Questions
#1

Hello, guys.

Recently I run SP calculation using ADC(2)-s but it doesn’t work well after finishing SCF calculation.

This is the input file.

$rem
JOBTYPE = SP
METHOD = ADC(2)
BASIS = 6-31G*
DFT_D = EMPIRICAL_GRIMME
SYMMETRY = FALSE
SYM_IGNORE = TRUE
CC_SYMMETRY = FALSE
EE_SINGLETS = 3
MAX_SCF_CYCLES = 512
SCF_CONVERGENCE = 10
THRESH = 14
MEM_TOTAL = 192000
MEM_STATIC = 2000
$end

$empirical_dispersion
s6 1.00
$end

And the error message looks like it.

 Q-Chem fatal error occurred in module libmdc/newfileman.C, line 384:

 FileMan error: Error writing 425984 bytes to file FILE_HALF_TRNS1
 Path: /work/116310.dinger00/qchem202633/65.121


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

If there is anyone to know how to solve this problem, can you give me an advice?

In addition, I found the difference of EE_STATES and EE_SINGLETS and Q-chem manual indicates that EE_STATES is for the open-shell or unrestricted closed-shell calculations and EE_SINGLETS is for the restricted closed-shell calculations. I’d like to calculate S1 state energy from S0 optimized structure. S1 state is open-shell system(there is one unpaired electron), as you know, then should I choose EE_STATES option? If I turn on EE_SINGLETS instead of EE_STATES, what is happened?

Moreover, Unrestricted means the separation of the orbital of alpha electron and beta electron. I know this concept but don’t know well when Unrestricted turn on or off exactly. can you explain it through some examples?

Thank you in advance.

#2

  1. If the reference state (in your case – ground state) is closed shell, then you should use default – restricted calculation, and then you ask for EE_SINGLETS to get your S1 state. To better understand the idea of reference and target state in ADC and EOM-CC framework, you may find these papers useful:
    http://arjournals.annualreviews.org/eprint/ukNyyVKWir2agwdJ5RPx/full/10.1146/annurev.physchem.59.032607.093602
    https://iopenshell.usc.edu/pubs/pdf/revcc-30.4-151.pdf

  2. The crash looks like you may run out of disc space. Try to run a small calculation first.

  3. DFT_D option does not belong to this input. Remove it to be on the safe side.

#3

The papers you recommanded looks very useful for me! Thanks a lot. :slight_smile: