Bridging oxygen speciation and free oxygen (O2 −) in K-silicate glasses: Implications for spectroscopic studies and glass structure

Abstract

O 1s X-ray Photoelectron Spectroscopic (XPS) results for K-silicate glasses near the K-disilicate composition indicate the presence of at least two, and likely three, types of Bridging Oxygen (BO), which are distinguished by the number of K ions coordinated to BO (BO-K moieites). The relative abundances of these moieties cannot be accurately determined from XPS spectra, but based on crystalline K-disilicate, they are likely to be BO, BO-K1 and BO-K2 or BO-K1, BO-K2 and BO-K3 (subscript indicates the number of K+ coordinated to BO). The presence of three BO moieties is consistent with crystallographic results for crystalline K-disilicate (K2Si2SiO5) and their presence in K-silicate glass may be assessed using high quality molecular dynamics studies. The binding energy (BE) separating each BO-K peak is ~40 to ~70kJ and sufficient to affect stretching frequencies in Raman and IR spectra, and to effect the electronic properties of O and Si atoms in 17O and 29Si NMR spectra. The BO moieties may give rise to separate signals or they may coalesce to produce a broad, asymmetric band in Raman, 29Si NMR and 17O NMR spectra, just as they do in O 1s XPS spectra.The O 1s XPS and previously published 17O NMR spectra of glasses containing ~20 and ~35mol% K2O are well resolved. Both types of spectra yield the same NBO and BO abundances within experimental uncertainty. For 20mol% K2O glass, the O 1s XPS results yield ~21.5mol% NBO and ~78.5% BO whereas the 17O MAS NMR results are ~22.2% NBO and 77.8% BO. The 35mol% K2O glass yields similarly close agreement between O 1s XPS and 17O MAS NMR spectral fits. Both techniques can provide accurate oxygen species abundances provided the spectra are of high resolution, and provided complete fits are performed on the spectra.A 2D 29Si MAF NMR spectrum of K-disilicate glass indicates that two Q4 contributions are present in the spectrum, as previously observed in highly siliceous K-silicate glasses. One explanation for two peaks is that one Q4 signal represents Q4 tetrahedra attached to Q3 tetrahedra and the second peak represents Q4 attached to other Q4 species. An alternative explanation is offered based on these findings. The two Q4 signals may result from different types of BO-K moieties bonded to Si centers of Q4 species. K+, located in the second coordination sphere of BO, modifies the electronic (valence band) properties of the BO and Si atoms of the tetrahedron, thereby producing two Q4 signals. The 2D 29Si MAF NMR, 17O NMR and O 1s XPS spectra indicate ~2 (±1) mol% O2− in the K-disilicate glass. These three types of spectra yield remarkably similar results when high resolution spectra are collected and complete fitting of the spectra are performed.