Estimating accuracy of 17O NMR measurements in oxide glasses: Constraints and evidence from crystalline and glassy calcium and barium silicates

27Al NMR Studies of Aluminosilicate Glasses

“…Convincing (albeit not unambiguous) evidence for the occurrence of O [3] coordinations in glasses is hitherto only provided for an CaAl 2 O 4 glass composition by using heteronuclear 27 Fig. 18) also predict another controversial entity in glass structures-"free O 2À ions," i.e., O [0] species that solely coordinate modifier cations [258][259][260]. This implies the formation of M-O [0] -M motifs that are intimately related to tendencies of modifier clustering.…”

“Free” oxide ions in silicate melts: Thermodynamic considerations and probable effects of temperature

“…However, the common necessity of fitting partially overlapping peak shapes, some of which are clearly non-Gaussian in shape because of quadrupolar effects, introduces significant uncertainty and limits the precision of this approach in constraining FO concentrations and values of K. For example, in the study mentioned above as having a clear negative result for the direct detection of FO at a level of about 0.5% in a low-silica Ca silicate glass (Thompson et al, 2012), the indirect approach of measuring BO and NBO from spectra and calculating FO from composition would allow, within uncertainties, a few % FO, as the BO content was slightly higher than predicted from stoichiometry. In contrast, similar data for a Ba silicate glass, for which the 17 O NMR spectrum is better resolved, showed a slightly lower BO to NBO ratio than expected from composition, which can"t readily be explained by this type of oxygen speciation and is probably the result of uncertainties in fitting or other aspects of the data analysis.…”

“…(Chase, 1998) b. (Barin et al, 1977) c. (Hess, 1995) d. (Zr/Hf)O 2 -SiO 2 amorphous thin films, 17 O NMR (Kim et al, 2015) e. PbO-SiO 2 glasses, 17 O NMR (Lee and Kim, 2014) f. PbO-SiO 2 glasses, O 1s XPS (Nesbitt et al, 2015a;Nesbitt and Dalby, 2007) g. glasses near to Mg 2 SiO 4 , 29 Si NMR (Davis et al, 2011) h. MgSiO 3 glass, 29 Si NMR, 1% FO as upper limit (Davis et al, 2011) i. based on re-analysis of published NMR data (Nesbitt et al, 2015a) j. glasses near to CaMgSiO 4 , 17 O NMR and 29 Si NMR (Nasikas et al, 2012) k. 44% SiO 2 glass, FO not detected in 17 O NMR (Thompson et al, 2012) l. CaSiO 3 glass, 29 Si NMR, 1% FO as upper limit (Zhang et al, 1997) m. O 1s XPS (Nesbitt et al, 2011) n. 60% SiO 2 glass, FO not detected in 17 O NMR (Stebbins and Sen, 2013) o. O 1s XPS (Sawyer et al, 2015) p. from model of high temperature oxide-melt equilibria (Toop and Samis, 1962) formed by ion beam sputtering (modified from (Kim et al, 2015)). Oxygen species are labeled:…”

“…Spinning sidebands are marked by "*". (Thompson et al, 2012). The crystal contains "free" oxide ions, which give the complex peak at 260-280 ppm.…”

“…Future CP-HETCOR NMR experiments on Mg silicate binary compositions could, however, be quite revealing.Well-defined negative results for the direct detection of FO ions in more silica-rich glasses are also important in constraining possible ranges of K ox values for reaction (2). In a CaO-SiO 2 glass with a silica content of 44 mol% (near the low end of the range obtainable by normal quenching methods), the region of the spectrum expected for FO was carefully explored, extending well above and below the known frequency range for this species in crystalline oxides and silicates(Thompson et al, 2012) (Fig. 2).…”