Homolytic Bond Dissociation Enthalpies of C-C and C-H Bonds in Highly Crowded Alkanes

Influence of Steric and Dispersion Interactions on the Thermochemistry of Crowded (Fluoro)alkyl Compounds

ConspectusAlkanes play a pivotal role in industrial, environmental, and biological processes. They are characterized by their carbon-carbon single-bond structure, remarkable stability, and conformational diversity. Fluorination of such compounds imparts unique physicochemical properties that often enhance pharmacokinetic profiles, metabolic stability, and receptor interactions while keeping beneficial properties. However, such per- and polyfluoroalkyl substances (PFAS) show a persistent presence in the environment and potential adverse health effects, which propelled them to the forefront of global environmental and health discussions. Alkyl compounds are also prototypical for stereoelectronic (SE) effects that are widely applied in chemistry. Substituents are typically described as electron-density-donating/withdrawing and/or responsible for sterically interacting with reagents or strategic groups in the molecule. That alkane branching can result in higher stability compared to less-branched isomers has been investigated in detail also by testing quantum chemical methods, in particular density functional theory (DFT). Alkane branching results in spatially compact structures with close intramolecular contacts so that at a specific size the detailed balance of attractive London dispersion and covalent versus repulsive Pauli exchange interactions shifts to new, chemically unfragile situations. This may lead to dissociation at room temperature and opens the central question: what is the smallest crowed alkane that cannot be made synthetically? In this Account, we try to shed light on the interplay among the various (free) energy components for crowded (fluoro)alkane dissociation. In this context, homolytic cleavage of the central C-C bond in a series of model alkanes of increasing size with tert-butyl (tBu), adamantyl (Ad), and [1.1.1]propellanyl (Prop) substituents is investigated. Reference energies are calculated at the PNO-LCCSD(T)-F12b level and used to benchmark the performance of contemporary DFT functionals. In line with previous conclusions, the application of dispersion corrections to density functionals is mandatory. For crowed structures, the accurate description of the midrange correlation effects, specifically repulsive van der Waals interactions, is crucial, and we observed that the density-dependent VV10 correction is superior to D4 in this context, although the asymptotic region is better described by the latter. The best available dispersion-inclusive functionals show systematic and reasonably small residual errors and can be safely applied to large systems (>100 atoms), for which coupled cluster methods with large basis sets are not computationally feasible anymore. For qualitatively correct predictions of synthetic accessibility under equilibrium conditions (free energy), the inclusion of thermostatistical (entropy) contributions is also essential. According to our results, tetra-tert-butylmethane (C17tBu) is the largest and most crowded system with a positive dissociation free energy and should be synthesizable. The difference between hydrogenated and perfluorinated systems originates from the increase in the steric repulsion of spatially close substituents, which is not compensated to the same extent by attractive orbital and dispersion interactions. A sometimes-assumed similar steric demand for fluorine and hydrogen atoms is not corroborated by our investigations on crowded systems. Perfluorination is found to substantially decrease thermal stability, rendering perfluorinated hexamethylethane (C8tBuF) the last potentially stable representative.

C−X Bond Activation by Palladium: Steric Shielding versus Steric Attraction

The C−X bond activation (X = H, C) of a series of substituted C(n°)−H and C(n°)−C(m°) bonds with C(n°) and C(m°) = H₃C− (methyl, 0°), CH₃H₂C− (primary, 1°), (CH₃)₂HC− (secondary, 2°), (CH₃)₃C− (tertiary, 3°) by palladium were investigated using relativistic dispersion‐corrected density functional theory at ZORA‐BLYP‐D3(BJ)/TZ2P. The effect of the stepwise introduction of substituents was pinpointed at the C−X bond on the bond activation process. The C(n°)−X bonds become substantially weaker going from C(0°)−X, to C(1°)−X, to C(2°)−X, to C(3°)−X because of the increasing steric repulsion between the C(n°)‐ and X‐group. Interestingly, this often does not lead to a lower barrier for the C(n°)−X bond activation. The C−H activation barrier, for example, decreases from C(0°)−X, to C(1°)−X, to C(2°)−X and then increases again for the very crowded C(3°)−X bond. For the more congested C−C bond, in contrast, the activation barrier always increases as the degree of substitution is increased. Our activation strain and matching energy decomposition analyses reveal that these differences in C−H and C−C bond activation can be traced back to the opposing interplay between steric repulsion across the C−X bond versus that between the catalyst and substrate.