While standard FPT/INDO MO calculations are widely used, and are sometimes useful in the prediction of conformational or structural behavior of spin-spin couplings, a more detailed and informative picture can be obtained if the total coupling is divided into ?r, u, TS (through space), and other specific contributions. The history of the calculations was reviewed by Kowalewski (I). Recently, Engelmann et al. (2) and Fukui et al. (3) developed the PRMO and SOS2 methods. The NNBI procedure, especially useful in the calculation of TS contributions, was described by Barlield (4). Attempts to apply PRMO to the interpretation of TS components of several H-H, H-F, and H-C couplings, and some important differences found between PRMO and NNBI results, led to the calculations described in this report. Its purpose is to point out that when the TS contribution is computed by the PRMO method, setting some atomic orbitals to be restricted, the contribution so obtained may often contain a large component due to the change in the spin polarizability near the perturbed nuclei. A few examples are discussed, PRMO and NNBI methods are compared, and the meaning of the ?r contribution is briefly discussed. The computations utilized an Amdahl 47O/V7 with a program modified from CNIND0/74 (QCPE 28 1) by the author. All the calculations mentioned in this report are routinely carried out by the program. Standard geometry was mostly used. The geometry of anisole was that previously described (5). The most serious problem with the PRMO method, in addition to those mentioned by the authors (2, 6, 7), occurs in the calculations of TS pathways. For example, if the PRMO values of the TS contributions in the HO-H*. . . HCH3 complex (where H* is the perturbed atom) are computed (setting the HO fragment restricted) we unexpectedly find that the pathways H* OH CH4 are very important (the residual couplings between H* and CH4 carbon and the nearest proton are only a small fraction of the original values). And, by setting the Fock elements between OH and CH4 zero (this is the NNBI (4) method) no significant differences between the standard INDO/FPT and these results are seen. The numerical results are shown in Table 1. The strange PRMO results can be easily explained: restricting the OH fragment makes the environment of the OH proton less polarizable and this reduces its
Read full abstract