Towards the accurate and efficient calculation of optical rotatory dispersion using augmented minimal basis sets

It has been recognized that quantum-chemical predictions of dispersive (nonresonant) chiroptical phenomena are exquisitely sensitive to the periphery of the electronic wavefunction. To further elaborate and potentially exploit this assertion, linear-response calculations of specific optical rotation were performed within the framework of density-functional theory (DFT) by augmenting small basis sets (e.g., STO - 3G and 3 - 21G) for the core and valence electrons with diffuse functions taken from substantially larger bases (e.g., aug-cc-pVXZ where X = D, T, or Q). Of particular interest was the ability of such computationally efficient (augmented small-basis) model chemistries to reproduce results derived from more expensive (canonical large-basis) schemes. The results appear to be quite promising, with the augmented minimal-basis ansatz often yielding wavelength-resolved rotatory powers close to those deduced from standard DFT(B3LYP)/aug-cc-pVXZ treatments. Analogous linear-response analyses were performed by means of coupled-cluster singles and doubles (CCSD) theory, once again leading to augmented small-basis estimates of specific rotation in reasonable accord with their large-basis counterparts. Although CCSD predictions were deemed to be slightly worse than those obtained from DFT, they still were of sufficient quality for such reduced-basis calculations to be considered viable for exploratory work.