During the course of the synthesis of vitamin B12 in the early 1960s, Robert Woodward concluded (see Chem. Soc. Special Publications (Aromaticity), 1967, 21, 217 for the full story) that none of the above factors could rationalise the following reaction. It was thought at the time using simple molecular models (the quantitative aspect of which was pioneered in 1951 by another great organic chemist and Nobel prize winner, Derek Barton) that the product was predicted to be J and that H would be more sterically demanding and hence less likely to form. It turned out the opposite was true.
A new explanation was developed based on 'stereoelectronic' factors, i.e. recognising that the three-dimensional properties of the electrons and their phase relationship could dominate the other factors listed above. This theory of stereoelectronic control of pericyclic reactions was derived using an approach known as the conservation of orbital symmetry, together with the theoretician Roald Hoffmann. Woodward died before the Nobel prize was awarded to Hoffmann (and Fukui) for this work (published in J. Am. Chem. Soc., 1965, 87, 395. The work was reviewed here). The recognition that stereoelectronic factors (more accurately the three dimensional properties of the molecular wavefunction) can be critical to explaining reactivity has had an enormous impact on understanding organic chemistry (see Introduction to Stereoelectronics in 3rd year).
Since the original work in the 1960s, modern computer-based methods of molecular modelling allow us to investigate the original assumptions about which isomer might be expected to form. Quantitative models of both products can be readily constructed, and their relative energies calculated using a variety of methods. In this case, both so-called "Molecular Mechanics" and "Quantum Mechanics" methods suggest that both isomers are in fact fairly similar in energy (). Another Nobel prize winner, John Pople, was awarded his prize for developing the Gaussian program, one of the best known of the molecular modelling systems, and one which has been crucial in quantifying aromaticity and creating accurate models of reaction transition states and potential energy surfaces.