Search for: All records

Abstract The sodium aluminosilicate (NAS) glass family is important for many different industrial applications, but glass relaxation has not yet been thoroughly studied in this system. Thermal analysis techniques such as differential scanning calorimetry (DSC) and modulated differential scanning calorimetry (MDSC) can provide insight into the enthalpy relaxation of glass by measuring the glass transition temperature (T_g), activation energy, and enthalpy of relaxation. MDSC is mostly used to study nonoxide and lowT_gglasses, and there is much debate about whether the nonreversing heat flow analysis method is accurate. To the authors’ knowledge, this is the first paper using MDSC to study these NAS compositions, and one of few papers to report MDSC on highT_goxide glasses. We report on one set of modulation conditions that obtain a linear response using Lissajous curves, as well as comparing the activation energy calculated from DSC with the enthalpy of relaxation obtained from MDSC. Our results show that the activation energy and enthalpy of relaxation do not give the same compositional minimum in relaxation, and therefore more work is needed to investigate the validity of the nonreversing heat flow approach for highT_goxide glasses. 

A phase-separated borosilicate glass, with a relative permittivity ranging from 3 to 3.5 and a loss tangent as low as 5.6 × 10−4, is presented for packaging applications for the next generation of mobile communications. Ionic polarizability for each borosilicate composition was calculated from the Clausius–Mossotti relationship for both the vitreous and crystalline structures, and the polarizability difference between the two is proportional to the dielectric loss. Small amounts of alkali modifier were added to improve the glass processability, and the loss tangent increased to the 1–7 × 10−3 range. The resulting glass is phase-separated, which has no impact in the millimeter-wave spectrum, as the wavelengths are considerably greater than the length scale of each immiscible phase. 

Abstract Machine learning (ML) is emerging as a powerful tool to predict the properties of materials, including glasses. Informing ML models with knowledge of how glass composition affects short-range atomic structure has the potential to enhance the ability of composition-property models to extrapolate accurately outside of their training sets. Here, we introduce an approach wherein statistical mechanics informs a ML model that can predict the non-linear composition-structure relations in oxide glasses. This combined model offers an improved prediction compared to models relying solely on statistical physics or machine learning individually. Specifically, we show that the combined model accurately both interpolates and extrapolates the structure of Na₂O–SiO₂glasses. Importantly, the model is able to extrapolate predictions outside its training set, which is evidenced by the fact that it is able to predict the structure of a glass series that was kept fully hidden from the model during its training. 

Abstract Previous research has shown a consistent discrepancy in the reported structure of alkaline earth aluminosilicate glasses using molecular dynamics (MD) simulations versus nuclear magnetic resonance (NMR) experiments. Past MD results have consistently shown less than 5% five‐coordinated Al units (Al^[5]) in peraluminous glass compositions, but with high fractions of triple‐bonded oxygens (TBO, i.e., triclusters). Experimental results have shown a high fraction of Al^[5]with no direct evidence for TBO. One of the main criticisms associated with high TBO content found in MD‐generated glass structures is the use of classical interatomic potentials. To investigate this issue, we analyze the formation of both TBO and Al^[5]using three independently developed potentials with varying silica content and [Al₂O₃]/[MgO] ratios for the magnesium aluminosilicate (MAS) system. We specifically choose compositions with high ratios of alumina to magnesium oxide as this region is not as commonly explored. Results indicate that Al^[5]charge compensates the Al network in metaluminous compositions (compositions with more Mg than Al) while both TBO and Al^[5]are prevalent in peraluminous ranges (high Al content compositions) to charge balance Al units. From the literature, NMR experiments report MAS glasses with varying Al^[5]fractions and show significant differences for the same reported compositions. When comparing MD results from this work, the fraction of calculated Al^[5]is within the experimental variation found in the literature. This indicates that classical potentials can accurately capture alumina environments and that both Al^[5]and TBO can coexist in relatively high fractions. From the consistency in our results, we conclude that TBOs are inherent to the aluminosilicate glass system and are not simulation artifacts. 

Abstract The impact of microstructure on hardness in phase‐separated calcium aluminosilicate glasses is investigated. Changes in hardness are governed by microstructure deformations that occur during indentation. Phase separation leads to decreased hardness due to the incongruent yielding of the droplet and matrix phases. Moreover, the deformation of microstructures possessing dilute, spherical droplets did not have a significant impact on hardness. Microstructures characterized by concentrated, acicular droplets were found to deform through a process of droplet coalescence. This process absorbs additional energy during yielding and results in glasses that deform through droplet coalescence possessing improved hardness.