Rayleigh-Taylor Instability within Sediment Layers Due to Gas Retention: Preliminary Theory and Experiments

Abstract

Executive SummaryIn Hanford underground waste storage tanks, a typical waste configuration is settled beds of solid particles beneath liquid layers. The settled beds are typically also composed of layers, and these layers can have different physical and chemical properties. One postulated configuration within the settled bed is more-dense layer lying atop a less-dense layer. The different densities can be a result of different gas retention in the layers (Meacham and Kirch 2013) or different degrees of settling and compaction in the layers. This configuration can experience a Rayleigh-Taylor (RT) instability, in which the less-dense lower layer rises into the upper layer. Previous studies of gas retention and release have not considered potential buoyant motion within a settled bed of solids.The purpose of this report is to provide a review of the literature on RT instabilities with an emphasis on studies that are pertinent to RT instabilities between two solid materials of different bulk density that are formed from settled slurries. A preliminary evaluation of the potential for different gas retention in a sediment layer and gas release from the potential buoyant motion are also discussed, and the results of preliminary experiments that demonstrate RT instabilities due to gas retention are summarized.Several studies provide useful theoretical and experimental results that can be used to estimate the difference in void fraction (or in bulk density) between two layers that will result in instability within tanks, as well as the effects of yield stress and tank diameter. For soft solids (such as waste slurries), experiments have only been conducted with a slurry layer above a gas layer; experiments with two solid layers will be needed to support the stability criterion developed from the literature data. Evaluations of configurations with more than two layers that have different bulk densities were not available in the literature; however, configurations with multiple-layer systems are expected to be more stable than two layers with the same total difference in bulk density.Measurements of gas void fractions in slurries in Hanford double-shell waste tanks, when combined with density measurements of de-gassed samples, identified examples in which the gas void fraction increases with depth, and the bulk density (including gas) decreases with depth, showing that that there is a potential for buoyant motion within waste sediment layers arising from differences in bulk density.Observations of bubble retention in experiments that had gas release pathways, which were generally filled with gas, did not show a noticeable difference in gas retention above and below the depth of an open gas channel. This suggests that gas retention may not necessarily increase below the maximum channel depth. Additional experiments would be required to confirm or refute this observation.An increase in yield stress (i.e., a stronger material) would be expected to decrease the amount of gas release caused by buoyant motion within a sediment l...

Morphology of Gas Release in Physical Simulants

“…Of particular concern was the method used to generate a given volume of gas within the test material. Testing of gas retention and release has used several approaches for gas generation:  addition of peroxide (e.g., Rassat et al 2013;Gauglitz et al 2013)  addition of water-reactive metal powders such as iron or magnesium (Gauglitz et al 2012;Powell et al 2014)  radiolysis of water and organics using an external high-dose radiation source (e.g., Gauglitz et al 1996)  nucleation and expansion of soluble gas under vacuum (e.g., Rassat and Gauglitz 1995)  hydrophobic particle captive bubble expansion (e.g., Crawford et al 2013) The gas generation approach selected for cold simulant gas-release testing was nucleation and expansion of soluble gas under vacuum. In addition, several confirmatory tests support cold simulant testing employed addition of peroxide.…”

“…Such a configuration can experience a Rayleigh-Taylor (RT) instability, in which the less-dense lower layer rises into the upper layer, and this motion may cause a release of retained gas. Until the recent preliminary study of Gauglitz et al (2013), studies of gas retention and release in Hanford waste had not considered potential buoyant motion within a settled bed of sludge solids. One purpose of the Deep-Sludge Gas Release Event Project (DSGREP) at Pacific Northwest National Laboratory (PNNL) is to provide quantitative information for estimating the size of gas release events from sludge waste stored in double-shell tanks should an RT instability occur.…”