The Lewis electron-pair bonding model: the physical background, one century later

The shared electron-pair bonding model was suggested by Gilbert Lewis more than 100 years ago. Emerging from the chemical experience of the time, Lewis structures described contemporary aspects of chemical reality in terms of empirically adapted models without any (then unknown) quantum physical underpinnings. This Perspective details the origins and historical development of the Lewis model, which we contrast with the physical understanding of chemical bonding in terms of contemporary quantum chemistry. Some intuitively plausible classical explanations of the past, not least of which are the sharing of electrons by two atoms and the subtypes of shared electron-pair bonding and dative bonding, turned out to be well founded. Some other chemical dogmata, including the concept that bonding occurs only between two nuclei and is caused by spin coupling or that bond energy is of purely electrostatic origin, are less well founded. We now know that covalent bonding is not driven by the formation of an electron pair but rather by the lowering of the kinetic energy density of the shared electrons in the bonding region, which is provided by the interference of the atomic wavefunctions. Lewis structures remain highly useful models for describing chemical bonding in molecular structures and chemical reactions, particularly when supported by quantum chemistry. The concepts behind the three most common quantum chemical approximations - the valence bond, molecular orbital and density functional theories - are described. These methods allow us to learn that bonding is an energetic phenomenon, from which descriptors such as bond length, bond dissociation energies and force constants are derivable. The energetic origins of bonding point to bond energy decomposition analysis as a natural tool for elucidating the actions of bonding electrons.