Prediction of absorption wavelengths using a combination of semi-empirical quantum mechanics simulations and quantitative structure-property relationship modeling approaches

Light emitting or absorbing molecules have a wide range of industrial applications. To enable in-silico screening of a number of such molecules one should be able to predict their emission or absorption spectra rapidly and accurately. In the present study, we developed a combination of semi-empirical quantum mechanical (QM) simulations and a quantitative structure-property relationship (QSPR) modeling approach to compute such properties rapidly with a fair degree of accuracy. We focused our work on the prediction of the first absorption wavelength of a set of 40 fluorophores or polycyclic aromatic molecules. We also compared the results obtained from our QM-QSPR combination approach with those from different stand-alone semi empirical QM calculations and pure QSPR modeling approach. We showed that there is an advantage in the QM-QSPR combination approach for predicting and screening optical properties quickly and with a great deal of accuracy. (C) 2014 Elsevier B.V. All rights reserved.