A popular urban legend concerns the apparent flow of stained glass windows in medieval cathedrals, where the glass windows are commonly observed to be thicker at the bottom than they are at the top. Advances in glass transition theory and experimental characterization techniques now allow for us to address this urban legend directly. In this work, we investigate the dynamics of a typical medieval glass composition used in Westminster Abbey. Depending on the thermal history of the glass, the room temperature viscosity is on the order of 10(24) to 10(25) Pa.s, about 16 orders of magnitude lower than found in a previous study of soda lime silicate glass. This measurement is in quantitative agreement with a newly derived model for the composition dependence of the nonequilibrium viscosity of glass. Despite this significantly lower value of the room temperature viscosity, the viscosity of the glass is much too high to observe measurable viscous flow on a human time scale. Using analytical expressions to describe the glass flow over a wall, we calculate a maximum flow of similar to 1 nm over a billion years.

Stretched exponential relaxation is a ubiquitous feature of homogeneous glasses. The stretched exponential decay function can be derived from the diffusion-trap model, which predicts certain critical values of the fractional stretching exponent, beta. In practical implementations of glass relaxation models, it is computationally convenient to represent the stretched exponential function as a Prony series of simple exponentials. Here, we perform a comprehensive mathematical analysis of the Prony series approximation of the stretched exponential relaxation, including optimized coefficients for certain critical values of beta. The fitting quality of the Prony series is analyzed as a function of the number of terms in the series. With a sufficient number of terms, the Prony series can accurately capture the time evolution of the stretched exponential function, including its "fat tail" at long times. However, it is unable to capture the divergence of the first-derivative of the stretched exponential function in the limit of zero time. We also present a frequency-domain analysis of the Prony series representation of the stretched exponential function and discuss its physical implications for the modeling of glass relaxation behavior. (C) 2018 Elsevier B.V. All rights reserved.

Oxide glasses often consist of multiple network formers that create the backbone of the glass network and modifiers that serve as either charge compensators or creators of non-bridging oxygens. The variety of bonding preferences results in very rich composition-property relationships. In this work, we present a statistical description of the glass structure governed by the relative enthalpic and entropic contributions to the bonding preferences in a glassy system. Using the proposed model, we derive an analytical expression to represent the bonding in mixed modifier glasses and explain the role of composition and fictive temperature on glass structure. The model provides the criteria for nonlinearity in bonding preference and reveals regions where high fluctuations in local structure are predicted.

We have developed a numerical model based on finite element analysis (FEA) with a viscoelastic material model coupling stress relaxation and structural relaxation, using the Mauro-Allan-Potuzak (MAP) non-equilibrium viscosity equation as the shift function. A modeling study of the delayed elasticity behavior in glass under different equilibrium viscosity and non-equilibrium viscosity conditions is conducted. The delayed elastic response is found to be well described by a stretched exponential function with three parameters: the maximum delayed elasticity response, the retardation time of delayed elasticity response, and the stretching exponent of delayed elasticity response. The delayed elasticity magnitude is seen to increase with lower values of the stretching exponent b(stress). At equilibrium viscosity, the retardation time shows a linear relationship with the stress relaxation time. However, when the temperature drops sharply in the non-equilibrium viscosity cases, the delayed elastic response may be frozen resulting in a lower magnitude for the delayed elasticity and the retardation time is not linear any more with the stress relaxation time. The delayed elasticity stretching exponent is seen to vary slightly at different relaxation times and normalized delayed elasticity response can roughly be collapsed into a single master curve. The impact of liquid fragility is also studied.

The stress-induced birefringence (photoelastic response) in oxide glasses has important consequences for several applications, including glass for flat panel displays, chemically strengthened cover glass, and advanced optical glasses. While the effect of composition on the photoelastic response is relatively well documented, the effect of pressure has not been systematically studied. In this work, we evaluate the effect of hot isostatic compression on the photoelastic response of ten oxide glasses within two commonly used industrial glass families: aluminosilicates and boroaluminosilicates. Hot isostatic compression generally results in decreasing modifier-oxygen bond lengths and increasing network former coordination numbers. These structural changes should lead to an increase in the stress optic coefficient (C) according to the model of Zwanziger et al., which can successfully predict the composition and structure dependence of C. However, in compressed glasses, we observe the opposite trend, viz., a decrease in the stress optic coefficient as a result of pressurization. We discuss this result based on measured changes in refractive index and elastic moduli within the context of atomic and lattice effects, building on the pioneering work of Mueller. We propose that the pressure-induced decrease in C is a result of changes in the shear modulus due to underlying topological changes in the glass network. (C) 2017 Elsevier B.V. All rights reserved.