Efficient Prediction of Mole Fraction Related Vibrational Frequency Shifts

While so far it has been possible to calculate vibrational spectra of mixtures at a particular composition, we present here a novel cluster approach for a fast and robust calculation of mole fraction dependent infrared and vibrational circular dichroism spectra at the example of acetonitrile/(R)-butan-2-ol mixtures. By assigning weights to a limited number of quantum chemically calculated clusters, vibrational spectra can be obtained at any desired composition by a weighted average of the single cluster spectra. In this way, peak positions carrying information about intermolecular interactions can be predicted. We show that mole fraction dependent peak shifts can be accurately modeled and, that experimentally recorded infrared spectra can be reproduced with high accuracy over the entire mixing range. Because only a very limited number of clusters is required, the presented approach is a valuable and computationally efficient tool to access mole fraction dependent spectra of mixtures on a routine basis.