Hi! I'm kind of a beginner with computational chemistry and I\m trying to study the interaction between copper ion and a graphitic carbon nitride quantum dot. the calculation converges with HF but when I try using DFT the software keeps showing me these messages
EnCoef did 100 forward-backward iterations gaussian
EnCoef did100 forward-backward iterations gaussian
EnCoef did 80 forward-backward iterations gaussian
rare condition: small coef for last iteration: -0.433d-15
and repeats
I tried different functionals with different basis sets but didn't work for me
any help?
Kamaleldeen Nassar
Common options:
1- Try to increase the basis set gradually. first, preoptimize your system at lower level like PM6 semiempirical method... and later using these results as a guess in subsequent computations.
2- Try to increase the maxcycle number for SCF. For example, scf=(maxcycle=512) option.
3- The SCF=QC option is often helpful (slower, but more reliable algorithm)
Melek Hajji, very appropriate answer, thank you very much for your potential answer, also I am benefited from this answer.
Your input structure shows that you use fragments. In this case, generate a guess for wavefunction and use that guess to optimize your structure. Due to these fragments, I think your calculations are long. In the Gaussian manual, example 9.7 on page No 447 is helpful. (Exploring chemistry with electronic structure Methods, 3rd Ed.)
Dear Zaheer
by generating guess, do you mean to carry out the calculation with a low level of theory then use this guess in the DFT? or there is another way?
- Badges
- Science topic