Anisotropy of the Indirect Nuclear Spin-Spin Coupling Constants. III. Problems in the Structure Determination of the Molecule Dissolved in a Nematic Solvent

Anisotropy of the indirect nuclear spin-spin coupling constant was studied theoretically. All the contributions to the coupling tensors were calculated for various molecules by using the INDO molecular orbitals. As shown in part I of this series, the calculated 13C–H coupling anisotropy of 13CH3F is too small to be compared with the value obtained experimentally by Krugh and Bernheim. Thus we examined the substituent effect on the anisotropy in the 13CH3X series. It is concluded that the experimentally estimated value of the 13C–H coupling anisotropy in 13CH3F as large as 1890 Hz is erroneous and that it contains some other effects which are more important than electronic effect. We believe that the change in molecular geometry from gas state to the solute state in a nematic solvent is the most probable origin for the differences between theories and experiments. For the directly bonded C–X couplings (X is C, N or F), their anisotropies are in the same order of magnitude as their isotropic couplings. For the non-bonded C–X nuclei, they seem negligible in magnitude. For the F–F couplings, their anisotropies are exceptionally large and the orbital term is a very important source of anisotropy. Furthermore, even for the isotropic F–F couplings, the orbital and spin dipolar terms are very important and sometimes make decisive contributions exceeding the Fermi contact term.