4
$\begingroup$

I am new to DFT calculation. I want to know the structure of conjugated polymers (plane angle, etc.) and their electronic and optical properties (such as HOMO-LUMO, IR, etc.).

I am using 6-311+G(d) for heavy atoms and 6-31G for hydrogen with B3LYP. My question is that setup ok for supporting my experimental data? Or I have to use different method and basis set?

Improve this question
$\endgroup$
2
  • 5
    $\begingroup$ It's okay, but I wouldn't suggest those old basis sets, and I'd definitely recommend using dispersion corrections. What program are you using? There are better newer functionals, but B3LYP isn't bad for a start at modeling conjugated polymers particularly if you want to compare optical gaps. $\endgroup$
    – Geoff Hutchison
    May 29 at 21:26
  • $\begingroup$ I am using Gaussian 09 D.01 now, I could also use Orca at some point but not so good. Are there any good functionals that don't require too much computational power? (I've use cc-pVTZ but it cost too much time). Thank you for your answers $\endgroup$
    – Duy-Minh Hoang
    Jun 1 at 7:48

1 Answer 1

Reset to default
5
$\begingroup$

Apologies I do not have enough reputation points to post this as a comment. You might have already considered other electronic structure methods, but I would like to suggest consider using density matrix renormalisation group (DMRG) based methods for the calculations. DMRG is highly suitable for modelling conjugated polymers. For example, here is an example of DMRG-CASSCF calculations on linear polyenes, and here is an example of calculations on acenes.

DMRG methods are supported by several open source quantum chemistry packages, and I found psi4, which utilises CheMPS2 particularly easy to use.

Edit :

As pointed out by @wzkchem5, DMRG does not take into account dynamical correlations. For NEVPT2/DMRG calculations on polyene systems, please see this and this.

Improve this answer
$\endgroup$
2
  • 2
    $\begingroup$ A small comment: for quantitative accuracy, dynamic correlation has to be modeled as well on top of the DMRG wavefunction, for example by NEVPT2. $\endgroup$
    – wzkchem5
    Jun 2 at 17:37
  • 1
    $\begingroup$ Thank you for the comment. Please see this paper for an example of DMRG with NEVPT2. $\endgroup$
    – squareroottwo
    Jun 3 at 18:07

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .