No abstract

The collected data on geminal (H, H) coupling constants in substituted methanes and ethanes show that this coupling constant in methane is negative in sign, relative to vicinal coupling constants; a reassessment of the theoretical situation is required.Vicinal (H, H) coupling constants in substituted ethanes and ethylenes are qualitatively well accounted for by Karplus' theoretical treatment ; the predicted positive sign is probably correct as it leads to a predominant -, +,pattern of sign-alternation for (H, H) coupling across 2, 3 and 4 bonds. The vicinal coupling constants show variations of magnitude that are closely dependent on the electronegativity of substituent groups. Exceptions to the above pattern of sign-alternation include the positive gemha1 couplings in =CH2 and epoxide groups, and meta couplings across four bonds in benzene and heterocyclic aromatic rings.Long-range couplings across four or five bonds in unsaturated hydrocarbon groupings are mostly well accounted for by theory based on U-W configuration interaction. Some exceptional inter-nuclear couplings in saturated systems across four bonds and in unsaturated ones across five bonds that show marked stereospecifici ty, perhaps indicate residual long-range coupling through sigma bonds.* (Added in proof.)-The Dirac vector model is more likely to be relevant to long range " throughbond ' .' couplings, and Karplus' successful theory of vicinal couplings includes important terms involving H-H bonding that would not be considered on the Dirac model. However, here the predicted positive sign is the same in each case.

No abstract

No abstract

No abstract