Pseudocontact shifts, calculation

As mentioned at the beginning of this section the size of the pseudocontact shifts in the NMR spectra could in principle be calculated for all the low spin ferric heme compounds if detailed data on the electronic g-tensors were available (Jesson (47)). Unfortunately the EPR data on the azides can not be used directly, because these complexes are not in a pure low spin state under the conditions of the NMR experiments (see section VI C). For the compounds in Figs. 10 through 20 no. successful single-crystal EPR studies were as yet reported. However only g-values determined in frozen solutions are presently available (Blumberg and Peisach (70) Salmeen and Palmer (95a)), e.g. for dicyanoferri-porphin at 1.4 °Kgi = 3.64, g 2.29, and gs 1.0 were found. 

Ligand field effects split the J manifold in a way that is not easily predicted without specific calculations. However, the overall splitting is such that many of the levels are appreciably populated at room temperature. An elegant procedure that takes such effects into account in a general way with respect to pseudocontact shifts of metal-centered origin has been provided by Bleaney [79]. 

Once the Ax values and the molecular directions are obtained, the metal centered pseudocontact shifts can be calculated of nuclei which experience also contact and ligand centered pseudocontact shifts. With this procedure the contact plus ligand centered pseudocontact shift (which is small [100]) have been calculated for several systems. In Table 2.11 the data relative to the low spin iron(III) containing heme in cytochrome b are reported [101]. 

Atom name Residue name Chemical shift Hyperfine shift (ppm) Calculated pseudocontact shift (ppm) Contact shift (ppm)... 

Marks and co-workers (12) have studied the alkyl substituted compounds 7-16. Assuming that INDO/2 molecular orbital calculations on alkyl radicals can reasonably predict experimental electron-nuclear hyperfine coupling constants, a, they have calculated the a values for each of the alkyl substituents. Taking the ratio of the contact shifts of the ortho positions in 7 and vinylic position in 16 as equal to the ratio of calculated a values and the ratio of the geometry factors as equal to the ratio of pseudocontact shifts, Marks and co-workers could solve for the contact and pseudocontact shifts in 7 and 16. Factoring the... 

Since the calculated pseudocontact shifts are smaller in magnitude than the observed isotropic shift, Edelstein, et.al., concluded that an upfield contact component contributes to the total isotropic shift, indicative of covalency in the ligand metal bonds of uranocene. 

Reducing the magnitude of the calculated pseudocontact shifts requires smaller values of the anisotropy term X -Xj / which can only result if Xj O. This result provides independent confirmation of the same result of Fischer cited above. We noted also that both the electronic structure of uranocene proposed by Warren (63), assuming a Jz= 4 ground state, and a recent model proposed by Fischer (15), assuming a Jz=3 ground state, show that XjJLs non-zero, and less than X(j, at 30°C. 

Using Fischer s value of y j2 - yj 2=8.78 BM2 the calculated pseudocontact shifts for the t-butyl groups in 1,1 -di-t-butyl-and 1,1 -dineopentyl uranocene are -12.1 ppm and 7.28 ppm, respectively, for coplanar substituents, and -14.6 ppm and 3.22 ppm, respectively, for tipped substituents. Agreement between the calculated pseudocontact shifts and the observed isotropic shifts is rather good. Calculation of the pseudocontact shifts for the eye-... 

Tipping the substituent away from the uranium center leads to better agreement between the calculated and observed shift for the t-butyl group in 1,1 -di-t-butyluranocene. With y 2- y] 2 = 12.5 BM2, a tip of 5° away from uranium affords a calculated pseudocontact shift of -13.7 ppm, in excellent agreement with the experimental isotropic shift of -13.29 ppm. 

Assuming a geometry factor of 1/6 (A + 2B + 2D + C) for the methyl group in 1,1 -dimethyluranocene, the calculated pseudocontact shifts are -16.8 ppm and -18.2 ppm, respectively, for a coplanar and a tipped substituent. By difference from the isotropic shift, the contact shifts are 6.76 ppm and 8.17 ppm, while calculation of the contact shift from B0 affords a value of 8.71 ppm. Agreement between the contact shifts calculated by both methods is excellent, particularly for the tipped substituent. 

The calculated pseudocontact shift is -26.9 ppm (coplanar), -31.8 (tipped) and by difference from the experimental isotropic shift of -23.97 ppm, the contact shifts are 2.93 ppm (coplanar), and 7.83 ppm (tipped). Calculation of the contact shift from BQ affords a value of 5.23 ppm. 

Comparison of the calculated pseudocontact shifts for the neopentyl t-butyl resonances with the isotropic shift showed that the tipped geometry affords better agreement between the two values, but still the value of calculated shift was approximately twice that of the experimental isotropic shift. However, an extremely small population of any conformation other than A will readily decrease the magnitude of the calculated pseudocontact shift for the t-butyl resonance. Assuming an extremely small population of conformations other than A, how does this affect the factored shifts of the a-protons ... 

---