Effect of olefin pyramidalization on the proton affinity of tricyclo[3.3.3.0(3,7)]undec-3(7)-ene as determined by ab initio calculations and kinetic method measurements

The effects of alkene pyramidalization on proton affinity (PA) were investigated, using the pyramidalized olefin tricyclo[3.3.3.0(3,7)]undec-3(7)-ene (1), with bicyclo[3.3.0]oct-1 (5)-ene (2) as a reference compound. The expectation that the relief of olefin strain energy associated with protonation of 1 will result in a substantially greater proton affinity for 1 compared with 2 was confirmed by ab initio calculations. This was also evident from measurements made by the kinetic method in which competitive dissociations of proton-bound cluster ions of the olefin of interest and a reference base were examined in an ion trap mass spectrometer. However, the ab initio calculations show a smaller PA difference (11.7 kcal/mol) between 1 and 2 than the experiments which yield a difference of 23 +/- 2 kcal/mol. This discrepancy is reconciled by proposing that, in the experiments involving 1, protonation leads not to carbocation 1-H+ but to a rearranged carbocation. This type of isomerization was demonstrated experimentally by measurements of the PAs of 3,3-dimethyl-1-butene (7) and 2,3-dimethyl-2-butene (10), and ab initio calculations indicate that the tertiary cyclopropylcarbinyl carbocation, 5, is considerably lower in energy than 1-H+. Cation 5 could be formed either directly from 1-H+ or by protonation of vinylcyclopropane 4, an isomer of 1 which could be formed under the conditions of the experiment. The most stable alkene that can be formed by deprotonation of 5 is the diene 6, which is calculated to have PA = 219.6 kcal/mol at the MP4SDQ/6-31G* level of theory. The difference of 20.7 kcal/mol between the calculated PAs of 2 and 6 is in good agreement with the measured difference Delta PA = 23 +/- 2 kcal/mol and thus supports the hypothesis that the experiments involving 1 have measured the PA corresponding to formation of carbocation 5.