Quantum chemical calculations of 31P NMR chemical shifts: scopes and limitations

The aim of this work is to convince practitioners of 31P NMR methods to regard simple GIAO quantum chemical calculations as a safe tool in structural analysis of organophosphorus compounds. A comparative analysis of calculated GIAO versus experimental 31P NMR chemical shifts (CSs) for a wide range of phosphorus containing model compounds was carried out. A variety of combinations (at the HF, DFT (B3LYP and PBE1PBE), and MP2 levels using 6-31G(d), 6-31+G(d), 6-31G(2d), 6-31G(d,p), 6-31+G(d,p), 6-311G(d), 6-311G(2d,2p), 6-311++G(d,p), 6-311++G(2d,2p), and 6-311++G(3df,3pd) basis sets) were tested. On the whole, it is shown that, in contrast to what is claimed in the literature, high level of theory is not needed to obtain rather accurate predictions of 31P CSs by the GIAO method. The PBE1PBE/6-31G(d)//PBE1PBE/6-31G(d) level can be recommended for express estimation of 31P CSs. The PBE1PBE/6-31G(2d)//PBE1PBE/6-31G(d) combination can be recommended for routine applications. The PBE1PBE/6-311G(2d,2p)//PBE1PBE/6-31+G(d) level can be proposed to obtain better results at a reasonable cost. Scaling by linear regression parameters significantly improves results. The results obtained using these combinations were demonstrated in 31P CS calculations for a variety of medium (large) size organic compounds of practical interest. Care has to be taken for compounds that may be involved in exchange between different structural forms (self-associates, associates with solvent, tautomers, and conformers). For phosphorus located near the atoms of third period elements ((CH3)3PS and P(SCH3)3) the impact of relativistic effects may be notable.