High-Resolution Oxygen-17 NMR of Solids

We report the observation of well-resolved solid-state oxygen-17 NMR spectra of a variety of oxides and oxyanions by means of high-field (11.7 T) “magic-angle” and “variable-angle” sample-spinning NMR spectroscopy. The results (obtained by observation of the (¹/₂, −¹/₂) spin transition) indicate that a very wide range of ¹⁷0 quadrupole coupling constants (∼0 to >5 MHz), chemical shift anisotropies (∼0 to >300 ppm), and line widths (∼0.3 to >40 ppm), in addition to the expected overall ∼ 1200 ppm isotropic chemical shift range, are to be anticipated in solid-state oxygen-17 NMR studies, giving a rich variety of spectral information. For example, in the SiO₄⁴⁻ unit of the nesosilicate forsterite (Mg₂SiO₄), each nonequivalent oxygen is partially resolved and split by a large second-order quadrupole interaction, permitting determination of the isotropic chemical shifts, quadrupole coupling constants, and electric field gradient tensor asymmetry parameters for each of the three types of oxygen present. By contrast, in the W0₄²⁻ unit of K₂WO₄, each of the three crystallographically nonequivalent oxygens is fully resolved, the isotropic chemical shift range being ∼15 ppm, there is no measurable quadrupole interaction, leading to ∼ 1 ppm line widths, and each oxygen has an ∼300 ppm chemical shift anisotropy, Δσ. In order to better predict the types of results to be obtained in future solid-state oxygen-17 NMR studies, we present empirical relationships between quadrupole coupling constant values and an average percent ionic character for several oxides and oxyanions, and between cation ionic radius and chemical shift, for a series of isoelectronic oxides and oxyanions. Overall, the results represent an initial effort in delineating the types of systems amenable to investigation by means of high-field, high-resolution solid-state oxygen-17 NMR spectroscopy and indicate that a very rich variety of oxygen-17 NMR line-width and shift parameters are to be expected for both main-group and transition-metal oxides and oxyanions in the crystalline solid state. Such studies should provide important new parameters for structural characterization of these and other, noncrystalline systems, such as some mineral phases, glasses, ceramics, and heterogeneous catalysts.