Determination of Absolute Configuration Using Density Functional Theory Calculations of Optical Rotation and Electronic Circular Dichroism: Chiral Alkenes

Abstract

In principle, the absolute configuration (AC) of a chiral molecule can be deduced from its optical rotation (OR) and/or its electronic circular dichroism (ECD). In practice, this requires reliable methodologies for predicting OR and ECD. The recent application of ab initio time-dependent density functional theory (TDDFT) to the calculation of transparent spectral region OR and ECD has greatly enhanced the reliability with which these phenomena can be predicted. TDDFT calculations of OR and ECD are being increasingly utilized in determining ACs. Nevertheless, such calculations are not perfect, and as a result, ACs determined are not 100% reliable. In this paper, we examine the reliability of the TDDFT methods in the case of chiral alkenes. Sodium d line specific rotations, [alpha]D, are predicted for 26 conformationally rigid alkenes of known AC, ranging in size from 5 to 20 C atoms, and with [alpha]D values in the range of 0-500. The mean absolute deviation of predicted [alpha]D values from experimental values is 28.7. With one exception, beta-pinene, the signs of [alpha]D are correctly predicted. Errors in calculated [alpha]D values are approximately random. Our results define a "zone of indeterminacy" within which calculated [alpha]D values cannot be used to determine ACs with >95% confidence. TDDFT ECD spectra are predicted for eight of the alkenes and compared to experimental spectra. Agreement ranges from modestly good to poor, leading to the conclusion that TDDFT calculations of ECD spectra are not yet of sufficient accuracy to routinely provide highly reliable ACs. TDDFT OR calculations for two conformationally flexible alkenes, 3-tert-butylcyclohexene and trans-4-carene, are also reported. For the former, predicted rotations are incorrect in sign over the range 589-365 nm. It is possible that the AC of this molecule has been incorrectly assigned.