Thermal Effects and Vibrational Corrections to Transition Metal NMR Chemical Shifts

Abstract

Both zero-point and classical thermal effects on the chemical shift of transition metals have been calculated at appropriate levels of density functional theory for a number of complexes of titanium, vanadium, manganese and iron. The zero-point effects were computed by applying a perturbational approach, whereas classical thermal effects were probed by Car-Parrinello molecular dynamics simulations. The systematic investigation shows that both procedures lead to a deshielding of the magnetic shielding constants evaluated at the GIAO-B3 LYP level, which in general also leads to a downfield shift in the relative chemical shifts, delta. The effect is small for the titanium and vanadium complexes, where it is typically on the order of a few dozen ppm, and is larger for the manganese and iron complexes, where it can amount to several hundred ppm. Zero-point corrections are usually smaller than the classical thermal effect. The pronounced downfield shift is due to the sensitivity of the shielding of the metal centre with regard to the metal-ligand bond length, which increase upon vibrational averaging. Both applied methods improve the accuracy of the chemical shifts in some cases, but not in general.

Density-functional computation of 93Nb NMR chemical shifts

“…Further refinement of computed transition-metal chemical shifts is possible on going from equilibrium to vibrationally and thermally averaged structures. [47] Due to the anharmonicity of the stretching potentials, averaging over the zero-point motion tends to increase metal-ligand bond distances, but rarely by more than 1 pm (see r eff values in brackets in Table 1). The concomitant effect on the 93 Nb chemical shifts is indicated to be quite small: on going from the BP86 equilibrium to the zero-point averaged structures, δ( 93 Nb) of 9 and 10 change from −2818 and −3309 ppm to −2827 and −3299 ppm, respectively (GIAO-BPW91 level), i.e.…”

Rovibrational effects on NMR shieldings in a heavy-element system: XeF2

“…[4][5][6][7][8] As the isotope shifts are seen here and elsewhere 11,12 to be quite insensitive to solvent effects, the inadequacy of DFT is a more plausible modeling deficiency than the neglect of solvent effects. 8 As it is shown here that thermal effects are notable for both the shieldings and their isotope shifts, errors also arise due to not carrying out a quantum-mechanical treatment of the finitetemperature effects.…”

“…Furthermore, the thermal contributions cannot be accurately modeled by a classical treatment (vide infra). A large part of the remaining discrepancies with experiment 4,5,7,8 are most probably due to the use of DFT for the shielding surfaces, which may produce decisive systematic errors, as shown below.…”

“…There exist semi-automated methods to treat ZPV in connection with nonrelativistic (NR) computations, but only few studies have been published on relativistic (REL) heavyelement NMR shieldings. [4][5][6][7][8] Comprehensive studies of finitetemperature rovibrational effects for such cases are completely lacking in the literature. The sensitivity of the main relativistic spin-orbit (SO) corrections to light-atom NMR shielding on rovibrational motion has been reported.…”

“…2 For example, the ZPV corrections to transition metal shieldings have been studied and combined with classical thermal effects estimated by ab initio molecular dynamics simulations. 4,5 This combination of quantum and classical treatments was also used in a study of the isotope shifts of transition metal shieldings. 7 In a very detailed study of chlorine isotope shifts of 125 Pt shieldings a quantum mechanical treatment of ZPV only was used.…”