Hello..
I want to scan a bond length using CAM-B3LYP and compare the result with CCSD(T) method. The basis set used for both methods are the same(6-31+G(df)). Every thing was good in the case of CAM-B3LYP method while for CCSD(T) method I receive the below error:
NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-NEF-
NUMERICAL EIGENVECTOR FOLLOWING MINIMUM SEARCH
INITIALIZATION PASS
************************************************
** ERROR IN INITNF. NUMBER OF VARIABLES ( 0) **
** INCORRECT (SHOULD BE BETWEEN 1 AND 50) **
************************************************
I used the” # opt=modredundant iop(5/48=10000) ccsd(t)/6-31+g(df) maxdisk=5GB ” keywords. the “iop” was added for solving the error but it didn’t help.
can you please help me to fic the CCSD(T) error???
any suggestion would be regarded.
Dear Maryam,
please see:
http://www.somewhereville.com/?p=1571
Best,
Martin
First , CCSD(T) takes time and it would be better you submit a CCSD calculation.
Second, if you insist on using the CCSD(T),you might change your basis set because the convergence of 6-31G* ,6-31G**and 6-31G** basis sets are too much slow and these kind of basis sets are unsuitable for the CCSD(T) effects.Upgrading the basis set to aug-cc-pVDZ reduces the average error.
look at to the reference as follows
J. Chem. Theory Comput., 2014, 10 (8), pp 3140–3150
DOI: 10.1021/ct500347q
Hi all
- Different factors may be the origin of such error. Could you upload the input file?
Unfortunately Gaussian doesn't offer analytic gradients for CCSD(T), therefore you cannot perform an opt=modredundant relaxed PES scan using this method. You have two options if you need to continue with CCSD(T) (which can often be the case when publishing).
1) You can perform the opt=modredundant scan with CCSD, and then use these optimized geometries to do your single point energy calculations with CCSD(T) (using the Geom=AllCheck option). CCSD(T) is rarely used for geometry optimization anyway.
2) You can use a different program which offers analytic gradients for CCSD(T), e.g., CFOUR (http://www.cfour.de/).
Hope this helps!
Yes, Ben is right
use
# opt freq uCSDT/aug-cc-pvtz
This is for open shell species (radicals). For closed shell, remove the u
- Badges
- Science topic