Glass Melter Research Articles

Viscosity of transient glass-forming melt and its relation to foaming during batch-to-glass conversion

Primary foam, which affects the heat transfer into the glass batch and the final glass quality, occurs when a sufficient quantity of transient glass-forming melt evolves with viscosity low enough to close the open porosity of the reacting glass batch. To better understand how the fraction of transient melt and its viscosity affect the primary foam temperature range, we determined, with x-ray diffraction, the fraction and composition of the transient glass-forming melt in a heated waste glass melter feed as a function of temperature. Then we prepared a set of transient melts that occurred within the foaming temperature interval and measured their viscosities with spindle and falling sphere viscometers. The results agree with the Adam-Gibbs and VFT viscosity-composition models, even for transient melt compositions outside of the models' compositional validity range. As silica and other refractory particles dissolved in the predominantly borate transient melt while temperature increased, viscosity increased from the initial value of ∼ 500 Pa s at the onset of foaming (∼650 °C) to a maximum of ∼ 770 Pa s when silica dissolution was almost complete (∼700 °C). As temperature increased further, transient melt viscosity decreased to ∼ 220 Pa s (∼850 °C) when the primary foam collapsed.

Numerical modeling of Joule heated ceramic melter

Computational fluid dynamics and heat transfer models of waste glass melters are being developed to support the vitrification campaign at the Waste Treatment and Immobilization Plant (WTP) as part of the United States nuclear waste stewardship and environmental protection programs. The WTP mission aims to achieve the stable and durable processing of vitrified waste in the safest, most efficient, and economical manner possible. Hanford’s tank waste is far more complex and varied than the waste treated by other vitrification facilities and both startup and long-term operational challenges are likely to be even greater than prior experience has shown. There are thousands of different waste compositions, which makes processing the waste a complex and challenging task. Models of test melters, pilot-scale, and full-scale melters have been developed to provide understanding and insight into the various physical and chemical phenomena occurring during vitrification. The results of these modeling activities will be used to support the WTP operations to process legacy nuclear waste into a stable form for disposal.

HEAT TRANSFER ENHANCEMENT DUE TO COLD CAP MOTION FROM BUBBLING IN A WASTE GLASS MELTER

Abstract In this study, a computational fluid dynamics (CFD) model was developed to model the motion of a solid cold cap in a waste glass melter. Forced convection bubblers at the base of the melter release air into the molten glass, which forms large bubbles that travel upwards to the cold cap and augment heat transfer from the glass to the cold cap. The CFD model employs the Navier- Stokes equations to solve for the fluctuating flowfield using a rigid body motion dynamic fluid body interaction module. This allows for movement of the floating body in response to the bubbling forces calculated at each time step. The heat flux delivered to the cold cap by the convective bubbling is studied as a function of the bubbling rate. Results for the moving cold cap are compared with the computed heat flux trends for a stationary cold cap. The heat flux delivered to the cold cap from the glass is 25% higher for the case with the moving cold cap. The heat flux was found to be proportional to v0.6 as opposed to v0.9 (where v is the normalized bubbling rate) for the stationary cold cap.

Effect of material properties on batch‐to‐glass conversion kinetics

AbstractA recently developed model of the cold cap—the reacting glass batch (melter feeds) floating on molten glass in an electric glass melter—couples heat transfer with the feed‐to‐glass conversion kinetics. The model allows for determining the distributions of temperature and various properties within the cold cap. In the present study, this model is applied to four melter feeds designed for high‐level and low‐activity nuclear wastes. Profiles of temperature, conversion degree, cold cap porosity and density, condensed matter velocity, and heating rate were determined using the material properties of the cold cap. Effects of vigorous foaming at the cold cap bottom were considered. Density, thermal conductivity, and glass production rate strongly affect the cold cap thickness and the fraction of undissolved silica entering the melt under the cold cap. The heating rate profile in the cold cap is highly nonlinear, with high heating rates observed in the foam layer.

A Shielding System Proposal for the Cabling of Electric Glass Melters

This paper analysesthe impact on workers of the electromagnetic field produced by large furnaces for glass production. The paper also propose a solution for the reduction and the containment of magnetic induction levels below the limits imposed by national and international directives. Using a three-dimensional numerical calculation code based on the hypothesis of filiform conductors in the air (Biot-savart law). The choice of the solution is based on the technique named highly coupled magnetically passive loop technique (HMCPL). The solution was then implemented and a comparison between the magnetic induction values before and after the installation of the shielding solution is presented.

Conversion degree and heat transfer in the cold cap and their effect on glass production rate in an electric melter

AbstractA predictive model of melt rate in waste glass vitrification operations is needed to inform melter operations during normal and off‐normal operations. This paper describes the development of a model of the cold cap (the reacting melter feed floating on molten glass in a glass melter) that couples heat transfer with the feed‐to‐glass conversion kinetics. The model was applied to four melter feeds designed for high‐level and low‐activity nuclear waste feeds using the material properties, either measured or estimated, to obtain temperature and conversion distribution within the cold cap. The cold cap model, when coupled with a computational fluid dynamics model of a Joule‐heated glass melter, allows the prediction of the glass production rate and power consumption. The results show reasonable agreement with the melting rates measured during pilot‐scale melter tests.

Impacts of temperature on sulfur solubility in low‐activity waste glasses

AbstractSolubility of sulfur in glass melts has been studied to estimate the maximum sulfur tolerance in nuclear waste glass melter operations. Sulfur solubility data at various temperatures are needed to investigate the possible sulfate segregation in different stages of the melting process. In this work, a crucible salt saturation method was applied to measure the sulfur solubility in three low‐activity glasses at different temperatures ranging from 950°C to 1200°C. The results show that temperature has little impact on sulfur solubility, with one exception that the sulfur solubility increases slightly from 950°C to 1100°C. The temperature effect of solubility data using SO2/O2 gas and sulfate salt as the sulfur source are compared. The decreased melter tolerance with decreasing temperature is likely caused by multiple factors including sulfur solubility.

Issue Information

Cover Photograph: Series of photos of a batch bottom during the melting of a high‐level nuclear waste glass feed simulant (modified to be transparent) on the top of the glass melt in a laboratory cuvette: a) feeding tube on glass melt level (0 s), b) primary foam is developing below batch (51 s after feed charging was initiated), c) cavities form (62 s), and d) cavities start to escape at the cold cap edge (71 s). The composition of this feed was modified by removing the iron source to make the feed transparent and thus suitable for visual observation. In the paper, the goal is to fundamentally understand the behavior of foam at the interface between batch and melt in a glass melter. [https://doi.org/10.1111/ijag.15863]. image

Through a glass darkly: In-situ x-ray computed tomography imaging of feed melting in continuously fed laboratory-scale glass melter

This study describes the first direct in-situ 3-D observation of a steady-state melting process by imaging a laboratory-scale slurry-fed glass melter in operation by x-ray computed tomography. Features of the reacting glass-feed, the foam layer underneath, and cavities in the glass melt pool are reconstructed in three-dimensional images. A slurry pool formed in a deep central caldera of dense dried feed, which penetrated into the glass melt. Slurry overflow from the caldera led to fast-dried and highly porous feed structure. A thin layer of foam separated the caldera from the melt. Bubbles ~5–15 mm in diameter were seen to grow beneath the reacting feed and move through the melt to escape at the edge. Pore morphology is benchmarked against computed tomography scans of a pellet of reacting simulated waste glass feed, and evolved gas analysis describes the gases generated as a function of temperature. Cooling artifacts are imaged and compared to previous studies of quenched cold caps. Detailed understanding of processes occurring during the conversion process in and below the reacting feed layer is necessary for the development of representative models of the melting process.

Heat transfer from glass melt to cold cap: Gas evolution and foaming

AbstractIn electric melters, the conversion heat is transferred through the foam layer at the cold‐cap bottom. Understanding cold‐cap foaming is thus important for enhancing the efficiency of both commercial and waste glass melters as well as for the development of advanced batch‐to‐glass conversion models. Observing foam behavior is still impossible “in situ,” that is, directly, in glass melters. To investigate the feed foaming behavior in laboratory conditions, we employed the feed volume expansion test, evolved gas analysis, and thermogravimetry. Combining these techniques helps assess the cold‐cap bottom temperature that directly influences the temperature gradient at the melt/cold‐cap interface, and thus the rate of melting. We also discuss the behavior of cavities formed by coalescing primary foam bubbles and ascending secondary bubbles.

Heat transfer from glass melt to cold cap: Effect of heating rate

AbstractConversion of melter feed to glass occurs in the cold cap that floats on the melt pool in a nuclear waste glass melter. The conversion rate (the melting rate or the glass production rate) is controlled by the heat flux delivered to the cold cap from the molten glass. In an attempt to analyze the intricate relationship between the rate of heating, the feed foaming response, and the rate of melting, we measured the change in feed volume at different heating rates by using several melter feeds known to exhibit a wide range of melting rates under identical melter operating conditions. As expected, the maximum foam porosity increased as the heating rate increased. However, contrary to expectation, the temperature at which the foam reached maximum volume either decreased or increased with the heating rate, depending on the feed composition. A change in maximum foam temperature from the feed volume expansion test indicates a similar change of the cold‐cap bottom temperature, which influences the heat flow to the cold cap, and thus the rate of melting.

Structural Characterization of Ternary Salt Melts for Low Activity Waste Applications

ABSTRACTReactions of alkali salts (nitrates, sulfates, carbonates, halides, borates) play a key role in the low temperature feed conversion occurring at the cold cap during processing of Hanford Low Activity Waste (LAW) glass melters. An alkali salt phase can sometimes form, and preferentially incorporate radionuclides of Cs, Cl, I, and Tc. During melting of the slurry feed, some of the feed components sequentially break down with increasing temperature to form gases (i.e., nitrates ➔ NOx, carbonates ➔ CO2, and boric acid ➔ H2O) or partially volatilize (halides). Sulfate, however, tends not to volatilize but has limited solubility in the final borosilicate glass waste form. To improve understanding of these low temperature processes and their composition dependencies, a scoping study was undertaken to synthesize salt systems that remain amorphous at room temperature, thus facilitating structural study. Melts of equimolar ratios of K2SO4-ZnSO4 (a known ionic glass-forming system) with added nitrates, halides, or carbonates, were melted and quenched. Some of the materials formed single phase glasses and some underwent crystallization upon quenching. Characterization of these quenched materials by thermal analysis, infrared absorption, and diffraction was performed. Addition of other anions to the sulfate base glass resulted in a distortion of the sulfate tetrahedron, as evidenced by infrared absorption. Carbonates strongly promoted crystallization, mostly of carbonate phases. Nitrates promoted crystallization of ZnO, and the nitrate volatilized with some incorporating into the glass. Halides tended to incorporate into the glass, but the small (F) and large (I) halogens promoted crystallization of sulfate-containing crystals, while moderate sized (Cl) halogens produced single-phase ionic glasses.

Sensitivity study of forced convection bubbling in a transparent viscous fluid as a proxy for molten borosilicate glass

Computational fluid dynamic and heat transfer models are being developed for the waste glass melters that will be used to vitrify legacy tank waste at the Hanford site. The objective of this study is to validate the ability of these models to predict the forced convection bubbling in the molten glass at prototypic conditions and scale. Bubbling is used to homogenize the melt pool and increase the melt rate of the batch. Simulations were compared with experimental data obtained with a surrogate liquid under isothermal conditions in an acrylic tank. The simulation results quantify the projected bubble area at the free surface as a function of flow rate and nozzle depth and show agreement with experimental data. Additionally, the local sensitivities of the surface tension and viscosity are evaluated for their effects, since in the melter this parameter directly affects the heat transfer from the molten glass to the reactive batch (or cold cap) layer. From the sensitivity study performed, the viscosity exhibits twice as much influence on bubble projected surface area as the surface tension. In comparison to the grid resolution used in this study, these glass properties produce higher levels of uncertainty.

Development of a Validation Approach for an Integrated Waste Glass Melter Model

Integrated models are being developed to represent the physics occurring within the high-level and low-activity waste melters that will be used to vitrify legacy tank waste at the Hanford site. These models couple the melt pool, cold cap, and plenum region within a single computational domain. Validation of the models is essential to ensure the reliability of the numerical predictions of the operational melters. Experimental data from laboratory- and pilot-scale tests are thus being used to inform and validate various aspects of the melter model. This paper presents a tiered approach to model validation consisting of a series of progressively more complex test cases designed to model the physics occurring in the full-scale system. A hierarchical methodology has been developed to segregate and simplify the physical phenomena affecting the multiphase flow and heat transfer within a waste glass melter. Four hierarchical levels are defined in a validation pyramid and built up in levels of increasing complexity from unit problems to subsystem cases, to pilot-scale systems, and then to the full-scale system.

Investigation of Flow Behavior of Joule-Heating Flow in a 2-D Model of a Reprocessing Glass Melter Cavity

Flow behavior was observed in a simplified model cavity of a multiphase High Level Liquid Waste (HLLW) reprocessing glass melter. Electrodes were set to generate Joule-heating flow in the cavity. A chaotic flow occurred because the lower part of the cavity was heated while the top surface of the cavity was cooled. Downflow and upflow occurred alternately in cavities. The shape of the cavity was a sloping bottom cavity, which was similar in shape to the real glass melter. To know the flow behavior in the cavity, 1-D flow behavior and 2-D flow behavior were measured in an experiment and simulated by an original CFD code. In the sloping bottom cavity, chaotic flow occurred in the upper part of the cavity. In the case of the sloping bottom cavity which had the same set of electrodes as the glass melter, the effect of the downflow near the electrodes decreased. The same phenomena could be predicted in the melter. The experimental results were also used to validate the CFD code, which will be helpful for developing a multiphase Joule-heating flow predicting.

Measurement of High Viscosity Liquid-liquid Two Phase Diffussion Flow in a Cylindrical Tank for Modeling a Glass Melter

Measurement of High Viscosity Liquid-liquid Two Phase Diffussion Flow in a Cylindrical Tank for Modeling a Glass Melter

Numerical comparison of bubbling in a waste glass melter

Radioactive tank waste is scheduled for vitrification at the Waste Treatment and Immobilization Plant (WTP) being constructed at the Hanford Site. Testing of the pilot-scale DuraMelter 1200 at the Vitreous State Laboratory at the Catholic University of America has demonstrated that bubbling increases the melt rate of the batch material, and as a result, melter throughput. Computational fluid dynamics (CFD) models of this pilot-scale waste glass melter are being developed to improve our understanding of the processes that occur within the melter to aid in process optimization and troubleshooting of the WTP melters. Unfortunately, model validation is complicated by the difficulty of obtaining suitable experimental data for operational melters due to the inaccessibility for direct observation and measurements of the high-temperature, opaque fluid through the water-jacketed, refractory-lined steel vessel. This study focuses on assessing the fidelity of the CFD models to accurately predict the bubbling behavior. Because of the paucity of experimental data at the resolution required for CFD validation, a code comparison was used to evaluate two common approaches for simulating flows of two immiscible Newtonian fluids on numerical grids and resolving multiphase interfaces. Here, the volume of fluid and level set methods are used to resolve the dynamically evolving interfaces between the molten glass and the air bubbles. To aid in the validation of the results of these codes, a comparison of the bubble behavior, growth, and frequency of bubble generation are presented and a grid convergence study is performed for the two approaches. The predictions from the two codes are within 6% for the average bubble radius of curvature within the bubble channels, within 2% for average terminal rise velocity of the bubbles, and within 4% for the area mean of the local maxima at the free surface. These parameters are of interest since they affect the convection within the melter and at the interface between the glass and batch layer. Ultimately, the results of this work can assist in confirming the predictive ability of waste glass melter models and provide a better understanding of the flow patterns within the WTP melters.

Predictive modeling of crystal accumulation in high-level waste glass melters processing radioactive waste

The effectiveness of high-level waste vitrification at Hanford's Waste Treatment and Immobilization Plant may be limited by precipitation/accumulation of spinel crystals [(Fe, Ni, Mn, Zn)(Fe, Cr)2O4] in the glass discharge riser of Joule-heated ceramic melters during idling. These crystals do not affect glass durability; however, if accumulated in thick layers, they can clog the melter and prevent discharge of molten glass into canisters. To address this problem, an empirical model was developed that can predict thicknesses of accumulated layers as a function of glass composition. This model predicts well the accumulation of single crystals and/or small-scale agglomerates, but excessive agglomeration observed in high-Ni-Fe glass resulted in an underprediction of accumulated layers, which gradually worsened over time as an increased number of agglomerates formed. The accumulation rate of ∼53.8 ± 3.7 μm/h determined for this glass will result in a ∼26 mm-thick layer after 20 days of melter idling.

Real‐time monitoring of crystal accumulation in the high‐level waste glass melters using an electrical conductivity method

AbstractDuring the vitrification of high‐level radioactive waste (HLW) in HLW melters in the Waste Treatment and Immobilization Plant located in Washington State, spinel crystals may precipitate from glass and accumulate in the melter riser, preventing the discharge of molten glass into canisters. Therefore, an effort is being made to develop an electrical conductivity method to monitor crystal buildup in the melter riser. A vertically configured electrical conductivity (EC) probe with an alumina shaft and Pt‐10%Rh‐electrodes was designed and tested in standard conductivity solutions and glass melts both with and without spinel crystals. The EC probe measured conductivity in conductivity solutions within 10% of their certified values and showed a linear relationship with increased spinel layer thicknesses. Testing in silicate glass containing spinel crystals allowed for the determination of spinel conductivity as a function of temperature. The conductivities of spinel crystals of 20‐24 S/m at 800°C were in excellent agreement with the conductivity of trevorite (NiFe2O4) crystals at 800°C reported in the literature. The conductivities of spinel crystals and measured changes in the conductivity across the accumulated layer allowed for a successful measurement of spinel crystal accumulation in simulated HLW glass.

X‐ray tomography of feed‐to‐glass transition of simulated borosilicate waste glasses

AbstractHigh‐level waste feed composition affects the overall melting rate by influencing the chemical, thermophysical, and morphological properties of a cold cap layer that floats on the molten glass where most feed‐to‐glass reactions occur. Data from X‐ray computed tomography imaging of melting pellets comprised of a simulated high‐aluminum feed reveal the morphology of bubbles, known as the primary foam, for various feed compositions at temperatures between 600°C and 1040°C. These feeds were formulated to make glasses with viscosities ranging from 0.5 to 9.5 Pa s at 1150°C, which was accomplished by changing the SiO2/(B2O3+Na2O+Li2O) ratio in the final glass. Pellet dimensions and profile area, average and maximum bubble areas, bubble diameter, and void fraction were evaluated. The feed viscosity strongly affects the onset of the primary foaming and the foam collapse temperature. Despite the decreasing amount of gas‐evolving components (Li2CO3, H3BO3, and Na2CO3), as the feed viscosity increases, the measured foam expansion rate does not decrease. This suggests that the primary foaming is not only affected by changes in the primary melt viscosity but also by the compositional reaction kinetic effects. The temperature‐dependent foam morphological data will be used to inform cold cap model development for a high‐level radioactive waste glass melter.