Probing Metal-Bridging Oxygen and Configurational Disorder in Amorphous Lead Silicates: Insights from ¹⁷O Solid-State Nuclear Magnetic Resonance

Abstract

The detailed degree of mixing between framework Si and metal cations and the nature of oxygen that only links metal cations (metal-bridging oxygen) in archetypal metal oxide glasses have been among the unsolved problems in physical chemistry. Binary lead silicate (PbO-SiO2) glass is an ideal model system for exploring the extent of cation disorder and metal-bridging oxygen because of its peculiar glass-forming ability at high PbO concentration. Here we report the first high-resolution 17O solid-state NMR spectra for binary lead silicate glasses with varying composition (near that of orthosilicate, i.e., Pb/Si = 2), where peaks due to metal-bridging oxygen (Pb-O-Pb), as well as Pb-O-Si and Si-O-Si are clearly resolved. The spectra also reveals that the metal-bridging oxygen possesses a greater topological disorder due to structural complexity in Pb coordination environments. Together with a statistical thermodynamic model of oxygen speciation proposed to describe mixing behavior in oxide glasses with any oxygen coordination number, the current 17O NMR results allow direct quantification of the degree of Pb/Si disorder and estimation of relative energy difference among the oxygen clusters [Pb-O-Pb + Si-O-Si = 2(Pb-O-Si)] of ∼−6 kJ/mol. This corresponds to a degree of mixing (Q) between Pb and network Si of ∼0.9, which implies that it deviates from random distribution (Q = 0) but shows a tendency toward Pb/Si order (Q = 1) that favors the formation of the Pb-O-Si cluster. The calculated configurational enthalpy of lead silicate glasses based on 17O NMR data showed a negative deviation and is consistent with those estimated from experimental solution calorimetry, unveiling atomistic origins of bulk macroscopic properties in the amorphous oxide. The results and methods shed light on a unique opportunity to explore the nature of intermixing and topological disorder in diverse complex amorphous oxides with varying cation types and oxygen coordination numbers by estimating the exact proportion of metal-bridging oxygen.

Structure and properties of lead silicate glasses and melts

“…Through the comparison undertaken on lead silicates glasses and lead silicates crystals by Furkawa et al (1978) and on lead oxides by Worrell and Henshal (1978) and Zahra et al (1993), the 141 cm -1 peak has been attributed to a covalent Pb-O-Pb bond in an interconnected tetragonal pyramid PbO 4 . Also, this peak at 141 cm -1 can be correlated to the NMR results of Lee and Kim (2015) who observed that the proportion of Pb-O-Pb increases with the PbO content. The charge balancing in the PbO 4 pyramids was discussed in several X-Ray diffraction analysis (Morikawa et al, 1982, Imaoka et al, 1986.…”

“…The similarity between the glass and the mineral leads us to think that in our glasses, Pb is a divalent cation in four fold coordination according to the crystalline reference and previous work (Fayon et al, 1999). Secondly, the XANES spectra are not really affected by the increase of PbO content in silicate, excepted for the first EXAFS oscillation which shows a slight increase in energy with increasing PbO content: that feature reflects that the medium-range organization around Pb is changing a little with increasing PbO content, and probably the Pb-O-Pb fraction could be increasing with the PbO content according to Lee and Kim (2015).…”

“…These glasses have already been investigated by several research groups from a thermodynamic point of view (Bockris and Mellors, 1956;Nesbitt and Fleet, 1980;Hess, 1974 andGupta 1995) or by a spectroscopic approach (Furukawa et al, 1978;Worrell and Henshall, 1978;Smets and Lommen, 1982;Wang an Zhang, 1996;Ohno et al, 1991 andFeller 2010;Lee and Kim, 2015). Despite these studies, the role of Pb is still debated.…”

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

“…5 studies have directly detected FO species both in very low silica glass compositions where stoichiometry demands its presence and in systems with somewhat higher silica contents and cations with very high field strengths or lone pair electron structures (Hung et al, 2016;Kim et al, 2015;Lee and Kim, 2014); some 29 Si NMR and Raman spectroscopic studies of very low silica glasses have also strongly suggested that FO must be present to account for observed silicate species (Fayon et al, 1999;Fayon et al, 1998;Nasikas et al, 2012;Sen et al, 2009;Sen and Tangeman, 2008). There has been significant controversy over the XPS findings and accompanying re-analyses of previously published NMR spectra (Malfait, 2015;Nesbitt et al, 2015a;Nesbitt et al, 2015b), including concerns about accuracy of alkali silicate glass compositions and the vagaries of fitting of highly overlapped Gaussian components of both XPS and NMR spectra.…”

“…silicates (Lee and Kim, 2014); estimates of about 60 to 1000 have been made from less direct spectroscopic data on a number of Mg and Ca silicates (Davis et al, 2011;Nesbitt et al, 2015a).…”