Hydrogen bond blueshifts in nitrile vibrational spectra are dictated by hydrogen bond geometry and dynamics

Vibrational Stark effect (VSE) spectroscopy has become one of the most important experimental approaches to determine the strength of noncovalent, electrostatic interactions in chemistry and biology and to quantify their influence on structure and reactivity. Nitriles (C≡N) have been widely used as VSE probes, but their application has been complicated by an anomalous hydrogen bond (HB) blueshift which is not encompassed within the VSE framework. We present an empirical model describing the anomalous HB blueshift in terms of H-bonding geometry, i.e. as a function of HB distance and angle with respect to the C≡N group. This model is obtained by comparing vibrational observables from density functional theory and electrostatics from the polarizable AMOEBA force field, and it provides a physical explanation for the HB blueshift in terms of underlying multipolar and Pauli repulsion contributions. Additionally, we compare predicted blueshifts with experimental results and find our model provides a useful, direct framework to analyze HB geometry for rigid HBs, such as within proteins or chemical frameworks. In contrast, nitriles in highly dynamic H-bonding environments like protic solvents are no longer a function solely of geometry; this is a consequence of motional narrowing, which we demonstrate by simulating IR spectra. Overall, when HB geometry and dynamics are accounted for, an excellent correlation is found between observed and predicted HB blueshifts. This correlation includes different types of nitriles and HB donors, suggesting that our model is general and can aid in understanding HB blueshifts wherever nitriles can be implemented.