Phillip A. Gauglitz

Foam Generation in Homogeneous Porous Media

Abstract In steady gas–liquid flow in homogeneous porous media with surfactant present, there is often observed a critical injection velocity or pressure gradient ∇ p min at which “weak” or “coarse” foam is abruptly converted into “strong foam”, with a reduction of one to two orders of magnitude in total mobility: i.e., “foam generation”. Earlier research on foam generation is extended here with extensive data for a variety of porous media, permeabilities, gases (N 2 and CO 2 ), and surfactants. For bead and sandpacks, ∇ p min scales like (1/ k ), where k is permeability, over 2 1 2 orders of magnitude in k ; for consolidated media, the relation is more complex. For dense-CO 2 foam, ∇ p min exists but can be less than 23 KPa/m (1 psi/ft). If pressure drop, rather than flow rates, is fixed, one observes an unstable regime between stable “strong” and “coarse” foam regimes; in the unstable regime ∇ p is nonuniform in space or variable in time. Results are interpreted in terms of the theory of foam mobilization at a critical pressure gradient.

An Approach to Understanding Cohesive Slurry Settling, Mobilization, and Hydrogen Gas Retention in Pulsed Jet Mixed Vessels

The Hanford Waste Treatment and Immobilization Plant (WTP) is being designed and built to pretreat and vitrify a large portion of the waste in Hanford’s 177 underground waste storage tanks. Numerous process vessels will hold waste at various stages in the WTP. Some of these vessels have mixing-system requirements to maintain conditions where the accumulation of hydrogen gas stays below acceptable limits, and the mixing within the vessels is sufficient to release hydrogen gas under normal conditions and during off-normal events. Some of the WTP process streams are slurries of solid particles suspended in Newtonian fluids that behave as non-Newtonian slurries, such as Bingham yield-stress fluids. When these slurries are contained in the process vessels, the particles can settle and become progressively more concentrated toward the bottom of the vessels, depending on the effectiveness of the mixing system. One limiting behavior is a settled layer beneath a particle-free liquid layer. The settled layer, or any region with sufficiently high solids concentration, will exhibit non-Newtonian rheology where it is possible for the settled slurry to behave as a soft solid with a yield stress. In this report, these slurries are described as settling cohesive slurries.

Shear Strength Correlations for Kaolin/Water Slurries: A Comparison of Recent Measurements with Historical Data

This report documents testing funded by CH2M Hill Plateau Remediation and performed by Pacific Northwest National Laboratory (PNNL) in collaboration with Fauske and Associates, LLC (FAI) to determine the behavior of vessel spanning bubbles. The shear strengths of four samples of kaolin/water mixtures obtained by PNNL from FAI were measured and are reported here. The measured shear strengths of these samples were then used to determine how the Rassat correlation fit these new measurements or if a new correlation was needed. These results were then compared with previously reported data.

The Disruption of Vessel-Spanning Bubbles with Sloped Fins in Flat-Bottom and 2:1 Elliptical-Bottom Vessels

Radioactive sludge was generated in the K-East Basin and K-West Basin fuel storage pools at the Hanford Site while irradiated uranium metal fuel elements from the N Reactor were being stored and packaged. The fuel has been removed from the K Basins, and currently, the sludge resides in the KW Basin in large underwater Engineered Containers. The first phase to the Sludge Treatment Project being led by CH2MHILL Plateau Remediation Company (CHPRC) is to retrieve and load the sludge into sludge transport and storage containers (STSCs) and transport the sludge to T Plant for interim storage. The STSCs will be stored inside T Plant cells that are equipped with secondary containment and leak-detection systems. The sludge is composed of a variety of particulate materials and water, including a fraction of reactive uranium metal particles that are a source of hydrogen gas. If a situation occurs where the reactive uranium metal particles settle out at the bottom of a container, previous studies have shown that a vessel-spanning gas layer above the uranium metal particles can develop and can push the overlying layer of sludge upward. The major concern, in addition to the general concern associated with the retention and release of a flammable gas such as hydrogen, is that if a vessel-spanning bubble (VSB) forms in an STSC, it may drive the overlying sludge material to the vents at the top of the container. Then it may be released from the container into the cell’s secondary containment system at T Plant. A previous study demonstrated that sloped walls on vessels, both cylindrical coned-shaped vessels and rectangular vessels with rounded ends, provided an effective approach for disrupting a VSB by creating a release path for gas as a VSB began to rise. Based on the success of sloped-wall vessels, a similar concept is investigated here where a sloped fin is placed inside the vessel to create a release path for gas. A key potential advantage of using a sloped fin compared to a vessel with a sloped wall is that a small fin decreases the volume of a vessel available for sludge storage by a very small fraction compared to a cone-shaped vessel. The purpose of this study is to quantify the capability of sloped fins to disrupt VSBs and to conduct sufficient tests to estimate the performance of fins in full-scale STSCs. Experiments were conducted with a range of fin shapes to determine what slope and width were sufficient to disrupt VSBs. Additional tests were conducted to demonstrate how the fin performance scales with the sludge layer thickness and the sludge strength, density, and vessel diameter based on the gravity yield parameter, which is a dimensionless ratio of the force necessary to yield the sludge to its weight.( ) Further experiments evaluated the difference between vessels with flat and 2:1 elliptical bottoms and a number of different simulants, including the KW container sludge simulant (complete), which was developed to match actual K-Basin sludge. Testing was conducted in 5-in., 10-in., and 23-in.-diameter vessels to quantify how fin performance is impacted by the size of the test vessel. The most significant results for these scale-up tests are the trend in how behavior changes with vessel size and the results from the 23-in. vessel. The key objective in evaluating fin performance is to determine the conditions that minimize the volume of a VSB when disruption occurs because this reduces the potential for material inside the STSC from being released through vents.

The yield stress of foamy sands

The yield stress of a mixture of foam and solids, or foamy sand, was investigated theoretically using a two-dimensional (2-D) periodic model. The range of solid fractions considered ranged from about 40 to 68%. The yield stress of a foamy sand increases with gas fraction at a given solid fraction and increases with solid fraction at a given gas fraction. At a fixed fraction of solid plus gas, yield stress is relatively insensitive to gas or solid fraction alone. There exists a maximum liquid fraction above which the yield stress disappears. These trends agree with those reported for foamy sands encountered in tunneling through soft sediments and proppant-laden fracturing fluids used in the petroleum industry.

A Discussion of SY-101 Crust Gas Retention and Release Mechanisms

The flammable gas hazard in Hanford waste tanks was made an issue by the behavior of double-shell Tank (DST) 241-SY-101 (SY-101). Shortly after SY-101 was filled in 1980, the waste level began rising periodically, due to the generation and retention of gases within the slurry, and then suddenly dropping as the gases were released. An intensive study of the tanks behavior revealed that these episodic releases posed a safety hazard because the released gas was flammable, and, in some cases, the volume of gas released was sufficient to exceed the lower flammability limit (LFL) in the tank headspace (Allemann et al. 1993). A mixer pump was installed in SY-101 in late 1993 to prevent gases from building up in the settled solids layer, and the large episodic gas releases have since ceased (Allemann et al. 1994; Stewart et al. 1994; Brewster et al. 1995). However, the surface level of SY-101 has been increasing since at least 1995, and in recent months the level growth has shown significant and unexpected acceleration. Based on a number of observations and measurements, including data from the void fraction instrument (VFI), we have concluded that the level growth is caused largely by increased gas retention in the floating crust. In September 1998, the crust contained between about 21 and 43% void based on VFI measurements (Stewart et al. 1998). Accordingly, it is important to understand the dominant mechanisms of gas retention, why the gas retention is increasing, and whether the accelerating level increase will continue, diminish or even reverse. It is expected that the retained gas in the crust is flammable, with hydrogen as a major constituent. This gas inventory would pose a flammable gas hazard if it were to release suddenly. In May 1997, the mechanisms of bubble retention and release from crust material were the subject of a workshop. The evaluation of the crust and potential hazards assumed a more typical void of roughly 15% gas. It could be similar to percolati on in single-shell tank (SST) waste forms. The much higher void being currently observed in SY-101 represents essentially a new crust configuration, and the mechanisms for sudden gas release need to be evaluated. The purpose of this study is to evaluate the situation of gas bubbles in crust based on the previous work on gas bubble retention, migration, and release in simulants and actual waste. We have also conducted some visual observations of bubble migration through simulated crusts to help understand the interaction of the various mechanisms.

Estimated Maximum Gas Retention from Uniformly Dispersed Bubbles in K Basin Sludge Stored in Large-Diameter Containers

This letter report addresses the KE Basin sludge that will be retrieved and stored in large-diameter containers (LDCs.) A fraction of the hydrogen gas bubbles generated from the corrosion of uranium metal and oxides may be retained within the sludge matrix. Those entrapped bubbles will expand the sludge bed volume and, therefore, will affect how much sludge can be loaded into a container. The entrapped gas bubbles will also impact the overall thermal conductivity and heat capacity of the sludge bed. The evaluation summarized here was performed to estimate the maximum gas holdup (volume fraction gas) that could occur sludge stored in large-diameter containers, assuming uniform gas generation (i.e., uniform distribution of metallic uranium particles). This report represents an evaluation of the retention of uniformly distributed bubbles and an estimate of the maximum gas fraction that might be retained in K Basin LDCs based on existing literature data on bubble retention and Basin sludge characterization data. Existing data show that the maximum gas fraction varies, depending on physical properties and the configuration of the material or waste.

Strong-Sludge Gas Retention and Release Mechanisms in Clay Simulants

The Hanford Site has 28 double-shell tanks (DSTs) and 149 single-shell tanks (SSTs) containing radioactive wastes that are complex mixes of radioactive and chemical products. The mission of the Department of Energys River Protection Project is to retrieve and treat the Hanford tank waste for disposal and close the tank farms. A key aspect of the mission is to retrieve and transfer waste from the SSTs, which are at greater risk for leaking, into DSTs for interim storage until the waste is transferred to and treated in the Waste Treatment and Immobilization Plant. There is, however, limited space in the existing DSTs to accept waste transfers from the SSTs, and approaches to overcoming the limited DST space will benefit the overall mission. The purpose of this study is to summarize and analyze the key previous experiment that forms the basis for the relaxed controls and to summarize progress and results on new experiments focused on understanding the conditions that result in low gas retention. The previous large-scale test used about 50 m3 of sediment, which would be unwieldy for doing multiple parametric experiments. Accordingly, experiments began with smaller-scale tests to determine whether the desired mechanisms can be studied without the difficulty of conducting very large experiments. The most significant results from the current experiments are that progressively lower gas retention occurs in tests with progressively deeper sediment layers and that the method of gas generation also affects the maximum retention. Based on the results of this study, it is plausible that relatively low gas retention could occur in sufficiently deep tank waste in DSTs. The current studies and previous work, however, have not explored how gas retention and release will behave when two or more layers with different properties are present.

Vessel-Spanning Bubble Formation in K-Basin Sludge Stored in Large-Diameter Containers

The K Basin sludge to be retrieved and stored in the large diameter containers (LDCs) contains some fraction of uranium metal that generates hydrogen gas, which introduces potential upset conditions. One postulated upset condition is a rising plug of sludge supported by a hydrogen bubble that is driven into the vent filters at the top of the container. In laboratory testing with actual K Basin sludge, vessel-spanning bubbles that lifted plugs of sludge were observed in 3-inch-diameter graduated cylinders. This report presents a series of analytical assessments performed by the Pacific Northwest National Laboratory to address the potential for the generation of a vessel spanning bubble in the LDCs. The assessments included the development and evaluation of static and dynamic bubble formation models over the projected range of K Basin sludge physical properties. Additionally, the theory of circular plates was extrapolated to examine conditions under which a plug of sludge would collapse and release a spanning bubble.

Surface Tension Estimates for Droplet Formation in Slurries with Low Concentrations of Hydrophobic Particles, Polymer Flocculants or Surface-Active Contaminants

In support of the K-Basin project, Pacific Northwest National Laboratory (PNNL) was requested to evaluate the appropriate surface tension value to use in models predicting the formation of droplets from spray leaks of K-Basin slurries. The specific issue was whether it was more appropriate to use the surface tension of pure water in model predictions for all plausible spray leaks or to use a lower value. The surface tension of K-Basin slurries is potentially affected not only by particles but by low concentrations of nonionic polyacrylamide flocculant and perhaps by contaminants with surfactant properties, which could decrease the surface tension below that of water. A lower surface tension value typically results in smaller droplets being formed with a larger fraction of droplets in the respirable size range, so using the higher surface tension value of pure water is not conservative and thus needs a strong technical basis.