A study of NMR chemical shielding in 5-coordinate phosphorus compounds (phosphoranes)

• Published in: Tetrahedron 2005-12, Vol.61 (52), p.12343-12349
• Main Authors: Chesnut, D.B., Quin, L.D.
• Format: Article
• Language: English
• Subjects: 31P NMR; Density functional theory; Møller–Plesset theory; Phosphoranes
• ISSN: 0040-4020; 1464-5416
• DOI: 10.1016/j.tet.2005.09.130

Calculations of the phosphorus NMR chemical shielding in 5-coordinate phosphorus compounds have been carried out using the gauge-including-atomic-orbital (GIAO) 6-311+G(nd,p) basis set at both scaled density functional theory (sDFT) and estimated infinite order Møller–Plesset (EMPI) approaches. Results are generally in accord with previous studies on 3-coordinate phosphorus compounds but fail badly for compounds containing multiple chlorine atoms and indicate a need for a relativistic treatment of these species. We observe that some compounds with reported experimental 31P NMR chemical shifts far downfield of the calculated values are in fact in the range known from experiment and calculation to be in that expected for phosphonium ions; the reported structures need to be reconsidered. The first ab initio calculations, using scaled DFT and EMPI methods, of the 31P NMR chemical shift of phosphoranes have given satisfactory values with a root-mean-square-error of 15–20 ppm from experimental shifts. The calculations are useful in predicting shifts for new compounds. Of special interest are calculations for the cyclic compounds illustrated above. The shifts are δ −97.7 and δ −89.8, respectively, and are quite unlike shifts around δ +19 reported for some P-phenyl derivatives purported to have the five-membered ring system. These shifts suggest instead 4-coordinate (phosphonium) phosphorus.