The Expanded Use of MAS Solid-State NMR on Transition Metal-Rich Silicate Glasses

Upon the accessibility of high resolution Si-29 solid state NMR (Nuclear Magnetic Resonance) in the early 1980's, its application to Earth materials and their proxies, specifically to materials with a degree of disorder, such as glasses, melts, and solid solutions, allowed for petrological advancements. Unfortunately, up until recent years, the method has been largely limited to diamagnetic materials; materials with paramagnetic cations prove problematic ruling out many geologically relevant materials. The paramagnetic centers (unpaired electron spins) interact strongly with the spin of the NMR active nucleus; structural information is lost as the features of the spectra become broad and start to overlap, or the spectra themselves are so broad that they may be difficult to impossible to observe with conventional MAS NMR methods. The purview of solid state NMR has recently been expanded to paramagnetic materials following the use of P-31 and Li-7 solid state NMR to investigate battery materials (Grey et al.). Our group and others have been able to elucidate structural information using MAS and 'static' Si-29, Al-27, and P-31 NMR on crystalline silicates, oxides, and phosphates containing transition metals and REEs. In this vein, we investigated three Ni-rich silicate glasses (CaNiSi₂O₆, K₂NiSi₃O₈, K₂NiSi₂O₆) having either a nominal diopside or alkali silicate formula and one glass corresponding to Co-diopside (CaCoSi₂O₆). Stepping the frequency both up and down from the central frequency of the Si-29 nuclei, and a simple spin-echo pulse sequence, allowed for the collection of complete spectra. All spectra consisted of a very broad component (several 1000 ppm) and a narrower component (ca. 50-100 ppm). We concluded that the narrower component likely accounts for silicon atoms without a transition metal as nearest neighbors and that the intensity is related to the ratio of silicon to transition metal in the glass. The difference in shape between the broad components in each of the samples we conclude to be a result of the difference in the coordination in the transition metal, as is previously observed through x-ray and optical absorption spectroscopy. This motivates further work to investigate glasses where the coordination and structure is previously known to confirm our conclusions.