High resolution O 1s XPS spectral, NMR, and thermodynamic evidence bearing on anionic silicate moieties (units) in PbO-SiO2 and Na2O-SiO2 glasses

Abstract

Recent technical advances in X-ray photoelectron spectroscopy (XPS) have largely overcome differential charging of insulator surfaces, with the result that O 1s linewidths for silicates can be routinely collected that are 50% to 100% narrower than had been previously obtainable. The bridging and nonbridging oxygen signals of glasses are well resolved and now can be accurately measured by XPS without resorting to ad hoc fitting procedures. All O 1s spectra of silicates collected so far are somewhat asymmetric to the high binding energy side of the peak, a characteristic due primarily to vibrational contributions in the ion state produced by photoemission (electronic excited state). This asymmetry must be considered if the proportion of bridging to nonbridging oxygen ratios are to be accurately determined from the XPS O 1s spectra of silicates. X-ray photoelectron spectroscopy (XPS) and nuclear magnetic resonance (NMR) experiments yield similar silicon-bridging oxygen (Si-O-Si) concentrations for PbO-SiO2 glasses. The data are proof that the reaction, 2PbO + SiO2 [Formula: see text] Pb2SiO4, does not go to completion (i.e., one or other reactant is not completely consumed). Instead, the data indicate that an equilibrium exists among these constituents, with the equilibrium constant for the reaction being near 15; ergo, some metal bridging oxygen (MBO or Pb-O-Pb), silicon bridging oxygen (BO or Si-O-Si), and nonbridging oxygen (NBO or M-O-Si) are present over the entire range of binary melt compositions. NMR and XPS spectroscopic techniques yield substantially different Si-O-Si concentrations in Na2O-SiO2 glasses of intermediate to high Na2O content. XPS experiments indicate higher proportions of Si-O-Si in Na2O-rich glasses than do the NMR studies. In spite of the different results, three independently collected sets of NMR experiments are similar and yield similar Si-O-Si concentrations. Five XPS experimental studies are likewise self-consistent (but different from the NMR studies). Comparison of data from the two spectroscopic techniques indicates that all experimental spectra are of good or reasonable quality, so that systematic experimental problems or the interpretation (i.e., fitting) of these spectra may be incorrect for one or both types of spectral data. Viscosity, conductivity, and the XPS data are consistent with (but not proof for) formation of an Na2Si2O5 species in the Na2O-SiO2 glasses. The most comprehensive NMR study is, by contrast, consistent with the formation of an orthosilicate species (Na4SiO4). The NMR and XPS results cannot yet be reconciled and additional experimental studies are required to resolve and to understand better Na2O-SiO2 liquids and glasses.Key words: XPS, high resolution, silicate glass, NMR.