Abstract

Quantitative structure–activity relationship (QSAR) studies of polycyclic aromatic hydrocarbons (PAHs) often employ rapid semiempirical calculations to evaluate ionization energy (IE) and electron affinity (EA) values, assuming they are equal (but of opposite sign) to the energies of the highest occupied and lowest unoccupied molecular orbitals (HOMO and LUMO), respectively. However, regardless of the assumption of validity of Koopmans’ theorem, the reliability of this simple theoretical approach for reproducing the experimental IE and EA trends has not been tested, except for a few linear PAHs. Here the measured IEs and EAs of 17 PAHs are plotted vs. the HOMO and LUMO energies obtained with semiempirical AM1 calculations and, for comparison, HF/6-31G calculations. Good linear relationships are obtained with both methods, with correlation coefficients r > 0.98 for the IEs and r > 0.96 for the EAs. The IEs and EAs predicted by scaling the corresponding MO energies with the appropriate empirical linear equation are compared with experimental values available in the literature for PAHs (28 IEs and 22 EAs). The average (absolute) difference between evaluated and measured IEs is found to be 0.07 eV (s.d. = 0.05 eV), while for the EAs the average difference is slightly larger. The accuracy of both AM1 and HF/6-31G methods are essentially equal, the former having the significant advantage of being 60 times faster. The present study demonstrates the ability of rapid semiempirical calculations carried out on the neutral molecules to parallel the experimental IE and EA values of PAHs, and provides simple linear equations which can be routinely employed for their quantitative prediction in this class of compounds.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call