An NMR Spectroscopy View on London Dispersion in Catalysis: Detection, Quantification, and Application in Ion Pair and Transition Metal Catalysis.

Abstract

ConspectusThe energetic contribution of London dispersion (LD) can cover a broad range from very few to hundreds of kJ mol-1 for extended interaction interfaces due to its pairwise additivity. However, for a designed and successful application of LD in chemical catalysis, there are still many obstacles and questions that remain. In principle, LD can be regarded as the attractive part of the van der Waals potential. Thus, considering the whole van der Waals potential, including the repulsive part (steric repulsion), the ideal solution to the problem in catalysis would be to design compatible interaction interfaces at exactly the correct distance. In the case of a self-assembled, flexible structure arrangement, entropic contributions and solvent interactions might be detrimental. In the case of a rigid catalyst pocket, steric hindrance might not allow for large substituents that are usually applied as dispersion energy donors (DEDs). For a working catalytic system, the following question arises: how is it possible to dissect the complex interaction interfaces in terms of energetic contributions? Usually, the energetic contribution of LD to catalysis is addressed by using calculations. However, adequately computing the correct energetic contributions can be extremely challenging for a vast conformational space with all kinds of intermolecular interactions. Thus, experimental data are essential for comparison or benchmarking.Therefore, in this Account, we describe our quest for detailed experimental data obtained via NMR spectroscopy to experimentally dissect and quantify LD in catalytic systems. In addition, we address the question of whether bulky substituents used as DEDs can be used in confined catalytic pockets. With the example of Pd phosphoramidite complexes, we show how it is possible to experimentally dissect and quantify the contribution of individual interaction areas in complicated transition metal complexes. Furthermore, a correlation between conformational rigidity and heterodimer preference clearly reveals that LD can only unfold its full potential in cases where entropic contributions are minimized. This finding can also explain the small contribution of LD in flexible and solvent-exposed molecular balances. In the field of Brønsted acid catalysis, we demonstrated that LD has a strong influence on the structures, stability, and populations of confined catalytic intermediates. LD is key for populating higher aggregates such as dimers. In addition, offsets between the experimental and computational results were observed and attributed to solvent-solute dispersion interactions. We studied the delicate interplay of attractive and repulsive interactions by adding bulky DED substituents onto a substrate, which can function as a molecular balance system. Intriguingly, the effect of LD on the free substrate was straightforwardly transferred onto the highly confined intermediates. Furthermore, this effect could even be read out in the enantioselectivities of the underlying reaction. This conceptualized a general approach regarding how LD can be used beneficially in catalysis to convert from moderate/good to excellent stereoselectivities. It showcased that bulky groups such as tert-butyl must not only be regarded as occupied volumes.