Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “TANK”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 343 records · Page 19

Sludge batch 9 (SB9) accepance evaluation: Radionuclide concentrations in tank 51 SB9 qualification sample prepared at SRNL

Presented in this report are radionuclide concentrations required as part of the program of qualifying Sludge Batch 9 (SB9) for processing in the Defense Waste Processing Facility (DWPF). The SB9 material is currently in Tank 51 and has been washed and prepared for transfer to Tank 40. The acceptance evaluation needs to be completed prior to the transfer of the material in Tank 51 to Tank 40. The sludge slurry in Tank 40 has already been qualified for DWPF processing and is currently being processed as Sludge Batch 8 (SB8). The radionuclide concentrations were measured or estimated in the Tank 51 SB9 Washed Qualification Sample prepared at Savannah River National Laboratory (SRNL). This sample was prepared from a three liter sample of Tank 51 sludge slurry (HTF-51-15-81) taken on July 23, 2015. The sample was delivered to SRNL where it was initially characterized in the Shielded Cells. Under the direction of Savannah River Remediation (SRR) it was then adjusted per the Tank Farm washing strategy as of October 20, 2015. This final slurry now has a compositioniv expected to be similar to that of the slurry in Tank 51 after final preparations have been made for transfer of that slurry to Tank 40.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Tank 22 Settled Sludge Core Sample Characterization and Testing

The Savannah River National Laboratory (SRNL) was requested by Savannah River Remediation (SRR) Engineering (SRR-E) to provide sample characterization and analyses of Tank 22 settled sludge core sample and the supernatant liquids (supernate) in support of Tank 22 re-baseline data, inhalation dose potential/hydrogen generation and suspension/removal of settled sludge from the bottom of Tank 22. The approximately 9.5 inches of intact Tank 22 core sample from SRR-E, after extrusion in the SRNL Shielded Cells, showed three visually different stratification layers or regions of solids in the extruded core material. Part of the initial physical characterization of these three segments of the Tank 22 core sample included density, weight percent total solids, particle size distribution, X-ray diffractions and scanning electron microscope of each segment material. Given the dissimilarity of the three Tank 22 segment materials after the physical characterization results and the fact that there was no clear line of demarcation to indicate where each segment began and ended across the entire length of the core sample, SRNL in conjunction with the SRR-E customer decided to blend all three segments of the material to form a composite Tank 22 core sample for further analysis and characterization without taking into consideration the proportional weight or mass contribution of each segment. This compositing approach assumes that the errors introduced in any characterization of the Tank 22 core sample will be comparable to the errors when the three segments are separated and combined proportionally by weight to form the composite Tank 22 sample. This compositing approach maximized the quantity of composited Tank 22 material needed for radionuclide and other physical characterizations.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Demonstration of Mixer Jet Pump Rotational Sensitivity on Mixing and Transfers of the AY-102 Tank

In support of Hanford’s waste certification and delivery of tank waste to the Waste Treatment and Immobilization Plant (WTP), Savannah River National Laboratory (SRNL) was tasked by the Washington River Protection Solutions (WRPS) to evaluate the effectiveness of mixing and transferring tank waste in the Double Shell Tank (DST) to the WTP Receipt Tank. The work discussed in this report (Phase II), is follow-on to the previous “Demonstration of Internal Structures Impacts on Double Shell Tank Mixing Effectiveness” task conducted at SRNL and the “Demonstration of Simulated Waste Transfers from Tank AY-102 to the Hanford Waste Treatment Facility”. The objective of the demonstrations performed in Phase II was to determine the impact on batch transfers when rotational parameters of the Mixer Jet Pumps (MJPs) were varied. Specifically, an evaluation was needed to ascertain the effect of the MJPs rotational characteristics on tank mixing and batch transfer consistency. Nineteen (19) different mixing/transfer demonstrations were performed by varying parameters of the MJP rotational characteristics. The demonstrations focused on visual comparisons of the results from transferring six batches of slurry from a 1/22 nd scale (geometric by diameter) Mixing Demonstration Tank (MDT) to six Receipt Tanks (RTs), where the consistency of solids in each batch could be compared. Three different simulants were used in these demonstrations, each by varying the composition of the following components: simulated Hanford Tank AY 101 supernate, gibbsite particles and silicon carbide particles/stainless steel particles.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DETERMINATION OF CORROSION INHIBITOR CRITERIA FOR TYPE III/IIIA TANKS DURING SALT DISSOLUTION OPERATIONS SUMMARY DOCUMENT

Dissolution of salt from Type III/IIIA waste tanks at the Savannah River Site may create solutions with inhibitor concentrations below those currently required (0.6M OH{sup -} and 1.1M OH{sup -} + NO{sub 2}{sup -}) per the Corrosion Control Program for high nitrate salt solutions (5.5 to 8.5M NO{sub 3}{sup -}). An experimental program was conducted to evaluate the corrosion susceptibility of grade A537 carbon steel for waste simulants containing 4.5-8.5M NaNO{sub 3} with maximum inhibitor concentrations of 0.6M NaOH and 0.2M NaNO{sub 2}. These maximum inhibitor concentrations used in this program are at a reduced level from those currently required. Current requirements were initially established for the Types I, II and IV tanks made of A285 carbon steel. The experimental program involved corrosion testing to evaluate the pitting and stress corrosion stress corrosion cracking (SCC) susceptibility of the Type III/IIIA waste tank materials. The program was conducted in two phases; the results of the first phase were reported previously (WSRC-STI-2006-00029). In this second phase, the corrosion specimens were modified to represent the 'as-fabricated' condition of the tank wall, and included specimens with mill scale, ground welds and stress-relief heat treatments. The complete description of the corrosion testing and the results are reported herein. The collective corrosion test results for A537 carbon steel in high nitrate waste simulants (4.5 - 8.5M) with the maximum inhibitor concentrations of 0.6M NaOH and 0.2M NaNO{sub 2} were as follows: (1) In long-term non-polarized U-bend testing, heat treatment, similar to the waste tank stress relief regime, reduced the incidence of cracking over the 18-month test period. Vapor space SCC was found to initiate on non-heat treated U-bend coupons. (2) In polarized U-bend testing, cracking occurred on U-bend coupons that had welds prepared similar to those in the waste tanks, i.e. ground and heat treated. (3) In electrochemical testing, pitting occurred on all coupons independent of heat treatment, inhibitor concentration, temperature, surface preparation, or welding. (4) In slow strain rate testing, cracking occurred on samples tested in solutions containing up to the maximum inhibitor concentration. The primary conclusion derived from this experimental program is that A537 carbon steel exposed to high nitrate (> 5.5M) solutions at inhibitor levels below the current specifications (0.6M OH{sup -} and 1.1M OH{sup -} + NO{sub 2}{sup -}) are susceptible to localized corrosion in the form of pitting and stress corrosion cracking. Long-term storage (e.g., greater than 100 days) of dissolved salt solutions not meeting the current inhibitor specifications for high nitrate salt chemistries in Type III waste tanks shall be avoided. Short term storage (e.g., less than 100 days) at low temperatures (e.g., less than 50 C) in these tanks is permissible for waste removal purposes. The stress relief process reduces the risk of SCC in the Type III waste tanks. On the other hand, the current inhibitor specifications should be strictly followed for the Type I, II and IV waste chemistry to prevent initiation of localized corrosion mechanisms. These tanks were not stress relieved and are more susceptible to SCC.

36 MATERIALS SCIENCE↗

CHARACTERIZATION OF THE TANK 18F SAMPLES

The Savannah River National Laboratory (SRNL) was asked by Liquid Waste Operations to characterize Tank 18F closure samples. Tank 18F slurry samples analyzed included the liquid and solid fractions derived from the 'as-received' slurry materials along with the floor scrape bottom Tank 18F wet solids. These samples were taken from Tank 18F in March 2009 and made available to SRNL in the same month. Because of limited amounts of solids observed in Tank 18F samples, the samples from the north quadrants of the tank were combined into one North Tank 18F Hemisphere sample and similarly the south quadrant samples were combined into one South Tank 18F Hemisphere sample. These samples were delivered to the SRNL shielded cell. The Tank 18F samples were analyzed for radiological, chemical and elemental components. Where analytical methods yielded additional contaminants other than those requested by the customer, these results were also reported. The target detection limits for isotopes analyzed were 1E-04 {micro}Ci/g for most radionuclides and customer desired detection values of 1E-05 {micro}Ci/g for I-129, Pa-231, Np-237, and Ra-226. While many of the minimum detection limits, as specified in the technical task request and task technical and quality assurance plans were met for the species characterized for Tank 18F, some were not met due to spectral interferences. In a number of cases, the relatively high levels of radioactive species of the same element or a chemically similar element precluded the ability to measure some isotopes to low levels. SRNL, in conjunction with the plant customer, reviewed all these cases and determined that the impacts were negligible.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

CHARACTERIZATION OF TANK 19F SAMPLES

The Savannah River National Laboratory (SRNL) was asked by Liquid Waste Operations to characterize Tank 19F closure samples. Tank 19F slurry samples analyzed included the liquid and solid fractions derived from the slurry materials along with the floor scrape bottom Tank 19F wet solids. These samples were taken from Tank 19F in April 2009 and made available to SRNL in the same month. Because of limited amounts of solids observed in Tank 19F samples, the samples from the north quadrants of the tank were combined into one Tank 19F North Hemisphere sample and similarly the south quadrant samples were combined into one Tank 19F South Hemisphere sample. These samples were delivered to the SRNL shielded cell. The Tank 19F samples were analyzed for radiological, chemical and elemental components. Where analytical methods yielded additional contaminants other than those requested by the customer, these results were also reported. The target detection limits for isotopes analyzed were based on detection values of 1E-04 {micro}Ci/g for most radionuclides and customer desired detection values of 1E-05 {micro}Ci/g for I-129, Pa-231, Np-237, and Ra-226. While many of the target detection limits, as specified in the technical task request and task technical and quality assurance plans were met for the species characterized for Tank 19F, some were not met. In a number of cases, the relatively high levels of radioactive species of the same element or a chemically similar element precluded the ability to measure some isotopes to low levels. SRNL, in conjunction with the plant customer, reviewed all these cases and determined that the impacts were negligible.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Nodal Modeling of Tank Pressurization and Draining using a Multi-Node-Ullage Approach

The purpose of the pressurization system in liquid rocket propulsion is to control the pressure in the gas space of the propellant tank (known as the ullage space) and the propellant mass flowrate to the engine. A mathematical model is required to predict the amount of pressurant necessary to ensure that pressure and temperature levels inside the tank remain within acceptable limits and that the propellant pressure leaving the tank satisfies the net positive suction pressure (NPSP) requirement of the pump feeding the engine. Nodal codes typically model tank pressurization and draining using a single node to represent the ullage and a single node to represent the propellant. As the tank drains, the ullage node grows and the propellant node shrinks. The heat transfer between ullage to wall and ullage to propellant is governed by natural convection. Designers of liquid propulsion systems often use empirical correlations to estimate the “Collapse Factor” which represents the ratio of pressurant required with heat transfer and the amount of pressurant required without heat transfer. A single node ullage model of tank pressurization was developed using GFSSP to compute the collapse factor reasonably well and later was used to model tank pressurization during test firing of the FASTRAC rocket engine. The predicted tank pressure compared well with the test data. In the early 1970’s, pressurization and drain tests with liquid methane were performed at NASA Lewis Research Center in a vacuum chamber. A 5 ft diameter spherical aluminum tank was tested to drain from 95% to 5% full using gaseous helium, hydrogen, nitrogen and methane as pressurant. Tests were conducted with different pressurant inlet temperatures and drain times. Measured data include pressurant requirement, amount of pressurant condensed, and ullage and wall temperatures at various heights in the ullage space at the end of draining. A single node GFSSP model was developed to simulate helium pressurization of the methane tank. Predicted helium consumption was 8-23% less than measured. The average error of the six test cases was 16%. A single ullage node with multiple solid node model was developed using Thermal Desktop. Predicted helium consumption compares with the test data within 2%. This paper describes the development of a GFSSP multi-node ullage model of the test configuration and compares the predicted pressurant consumption for both helium and autogenous pressurization using gaseous methane with experimental as well as TD predictions.

Nodal Model↗

Dual-Remote Raman Technology for In-Situ Identification of Tank Waste

A new Raman spectroscopic system for in-situ identification of the composition of the solid nuclear tank waste is being developed by collaborative effort between Pacific Northwest National Laboratory (PNNL) and EIC, Inc. The recent advancements in the Raman technology allow probing the chemical composition of the tank waste without sample collection. In the newly tested configuration, the Raman probe is installed on the top of the tank riser and sends the incident laser beam to the bottom of the tank, 10 – 70 feet away. The returning light containing chemical information is collected by the Raman probe and transmits it via fiber optic cable to the spectrometer located outside the tank farm area. This dual remote technology significantly expands currently limited options for the safe rapid in-situ identification of the solid tank waste needed for the retrieval decisions. The developed Raman system was extensively tested for acceptability prior tank farm deployment. This testing included calibration of the system with the respect of the distance between the Raman probe and the sample, incident laser beam angle, and presence of the optical interferences. The Raman system was successfully deployed to C-111 tank at the US DOE Hanford site. As the result of this deployment, the composition of the hardpan at the bottom of C-111 tank was identified. Further development of the dual-remote Raman technology will provide a significant safety enhancement eliminating the potential of personnel radiation exposure associated with the grab sample collection and expands options of the rapid and cost-effective in-situ chemical analysis of the tank waste.

in-situ Hanford tank, Raman measurement↗

Experimental Methods to Estimate Accumulated Solids in Nuclear Waste Tanks - 13313

The Department of Energy has a large number of nuclear waste tanks. It is important to know if fissionable materials can concentrate when waste is transferred from staging tanks prior to feeding waste treatment plants. Specifically, there is a concern that large, dense particles, e.g., plutonium containing, could accumulate in poorly mixed regions of a blend tank heel for tanks that employ mixing jet pumps. At the request of the DOE Hanford Tank Operations Contractor, Washington River Protection Solutions, the Engineering Development Laboratory of the Savannah River National Laboratory performed a scouting study in a 1/22-scale model of a waste tank to investigate this concern and to develop measurement techniques that could be applied in a more extensive study at a larger scale. Simulated waste tank solids and supernatant were charged to the test tank and rotating liquid jets were used to remove most of the solids. Then the volume and shape of the residual solids and the spatial concentration profiles for the surrogate for plutonium were measured. This paper discusses the overall test results, which indicated heavy solids only accumulate during the first few transfer cycles, along with the techniques and equipment designed and employed in the test. Those techniques include: - Magnetic particle separator to remove stainless steel solids, the plutonium surrogate from a flowing stream. - Magnetic wand used to manually remove stainless steel solids from samples and the tank heel. - Photographs were used to determine the volume and shape of the solids mounds by developing a composite of topographical areas. - Laser range finders to determine the volume and shape of the solids mounds. - Core sampler to determine the stainless steel solids distribution within the solids mounds. - Computer driven positioner that placed the laser range finders and the core sampler over solids mounds that accumulated on the bottom of a scaled staging tank in locations where jet velocities were low. These devices and techniques were very effective to estimate the movement, location, and concentrations of the solids representing plutonium and are expected to perform well at a larger scale. The operation of the techniques and their measurement accuracies will be discussed as well as the overall results of the accumulated solids test. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Dual-Remote Raman Technology for In-Situ Identification of Tank Waste - 13549

A new Raman spectroscopic system for in-situ identification of the composition of solid nuclear tank waste is being developed by collaborative effort between Pacific Northwest National Laboratory (PNNL) and EIC Laboratories, Inc. The recent advancements in Raman technology allow probing the chemical composition of the tank waste without sample collection. In the newly tested configuration, the Raman probe is installed on the top of the tank riser and sends the incident laser beam to the bottom of the tank, 10 - 70 feet away. The returning light containing chemical information is collected by the Raman probe and is transmitted via fiber optic cable to the spectrometer located outside the tank farm area. This dual remote technology significantly expands currently limited options for the safe rapid in-situ identification of the solid tank waste needed for the retrieval decisions. The developed Raman system was extensively tested for acceptability prior to tank farm deployment. This testing included calibration of the system with respect of the distance between the Raman probe and the sample, incident laser beam angle, and presence of the optical interferences. The Raman system was successfully deployed on Tank C-111 at the US DOE Hanford site. As the result of this deployment, the composition of the hardpan at the bottom of C-111 tank was identified. Further development of the dual-remote Raman technology will provide a significant safety enhancement eliminating the potential of personnel radiation exposure associated with the grab sample collection and expands options of the rapid and cost-effective in-situ chemical analysis of the tank waste. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Hanford double-shell tank extent-of-condition construction review - 15498

During routine visual inspections of Hanford double-shell waste tank 241-AY-102 (AY-102), anomalies were identified on the annulus floor which resulted in further evaluations. Following a formal leak assessment in October 2012, Washington River Protection Solutions, LLC (WRPS) determined that the primary tank of AY-102 was leaking. A formal leak assessment, documented in RPP-ASMT-53793, Tank 241-AY-102 Leak Assessment Report, identified first-of-a-kind construction difficulties and trial-and-error repairs as major contributing factors to tank failure. To determine if improvements in double-shell tank (DST) construction occurred after construction of tank AY-102, a detailed review and evaluation of historical construction records was performed for Hanford's remaining twenty-seven DSTs. Review involved research of 241 boxes of historical project documentation to better understand the condition of the Hanford DST farms, noting similarities in construction difficulties/issues to tank AY-102. Information gathered provides valuable insight regarding construction difficulties, future tank operations decisions, and guidance of the current tank inspection program. Should new waste storage tanks be constructed in the future, these reviews also provide valuable lessons-learned.(authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A summary of the Hanford single-shell tank structural analysis of record - 15526

Washington River Protection Solutions, LLC (WRPS), the US DOE Hanford Site tank contractor, developed an enhanced Single-Shell Tank Integrity Project (SSTIP) in 2009 to better understand the integrity of the single-shell tanks (SSTs). A panel of experts from industry and academia was created to recommend project activities in four main areas: structural integrity, liner degradation, leak integrity and prevention, and contamination migration. One of the panel's primary recommendations was to perform a detailed structural integrity analysis of the SSTs. WRPS and the U.S. Department of Energy, Office of River Protection commissioned the Pacific Northwest National Laboratory (PNNL) to perform this structural integrity analysis of record for the Hanford single shell tanks. This is one of seventeen panel recommendations that form the M-45-10-1 Change Package for the Hanford Federal Agreement and Compliance Order; also known as the Tri-Party Agreement. Finite element analysis was used to predict the structural response of the SSTs to the historical thermal and operating loads, plus design basis seismic loads. Bounding thermal histories were established from waste temperature records and applied to the model to include the thermal degradation of concrete modulus and strength, plus cracking due to differential thermal expansion while under in situ static loading conditions. PNNL contracted with Becht Engineering to evaluate the dynamic response of the tanks to a design basis earthquake. The degraded stiffness of the reinforced concrete was also incorporated in the seismic analysis. The combined response to static and seismic loads was then evaluated against the design requirements of the American Concrete Institute standard, ACI-349-06, for nuclear safety-related concrete structures. Specific analyses were performed for the four major tank designs.This paper presents an overview of the analysis methods used and it summarizes the major analytical findings that are common to the tanks and others that are specific to the individual tank designs. The analyses of record determined that including the bounding effects of their operating histories, each tank design currently satisfies the ACI-349-06 structural integrity design requirements. Although leak tightness is not directly addressed in the structural integrity analysis, this analysis will be a key input in the formal tank integrity assessment due in 2018. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DEMONSTRATION OF SIMULATED WASTE TRANSFERS FROM TANK AY-102 TO THE HANFORD WASTE TREATMENT FACILITY

In support of Hanford's AY-102 Tank waste certification and delivery of the waste to the Waste Treatment and Immobilization Plant (WTP), Savannah River National Laboratory (SRNL) was tasked by the Washington River Protection Solutions (WRPS) to evaluate the effectiveness of mixing and transferring the waste in the Double Shell Tank (DST) to the WTP Receipt Tank. This work is a follow-on to the previous 'Demonstration of Internal Structures Impacts on Double Shell Tank Mixing Effectiveness' task conducted at SRNL 1. The objective of these transfers was to qualitatively demonstrate how well waste can be transferred out of a mixed DST tank and to provide insights into the consistency between the batches being transferred. Twelve (12) different transfer demonstrations were performed, varying one parameter at a time, in the Batch Transfer Demonstration System. The work focused on visual comparisons of the results from transferring six batches of slurry from a 1/22nd scale (geometric by diameter) Mixing Demonstration Tank (MDT) to six Receipt Tanks, where the consistency of solids in each batch could be compared. The simulant used in this demonstration was composed of simulated Hanford Tank AZ-101 supernate, gibbsite particles, and silicon carbide particles, the same simulant/solid particles used in the previous mixing demonstration. Changing a test parameter may have had a small impact on total solids transferred from the MDT on a given test, but the data indicates that there is essentially no impact on the consistency of solids transferred batch to batch. Of the multiple parameters varied during testing, it was found that changing the nozzle velocity of the Mixer Jet Pumps (MJPs) had the biggest impact on the amount of solids transferred. When the MJPs were operating at 8.0 gpm (22.4 ft/s nozzle velocity, U{sub o}D=0.504 ft{sup 2}/s), the solid particles were more effectively suspended, thus producing a higher volume of solids transferred. When the MJP flow rate was reduced to 5 gpm (14 ft/s nozzle velocity, U{sub o}D = 0.315 ft{sup 2}/s) to each pump, dead zones formed in the tank, resulting in fewer solids being transferred in each batch to the Receipt Tanks. The larger, denser particles were displaced (preferentially to the smaller particles) to one of the two dead zones and not re-suspended for the duration of the test. As the liquid level dropped in the MDT, re-suspending the particles became less effective (6th batch). The poor consistency of the solids transferred in the 6th batch was due to low liquid level in the MDT, thus poor mixing by the MJPs. Of the twelve tests conducted the best transfer of solids occurred during Test 6 and 8 where the MJP rotation was reduced to 1.0 rpm.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Risks from Past, Current, and Potential Hanford Single Shell Tank Leaks

Due to significant delays in constructing and operating the Waste Treatment Plant, which is needed to support retrieval of waste from Hanford’s single shell tanks (SSTs), SSTs may now be required to store tank waste for two to three more decades into the future. Many SSTs were built almost 70 years ago, and all SSTs are well beyond their design lives. Recent examination of monitoring data suggests several of the tanks, which underwent interim stabilization a decade or more ago, may be leaking small amounts (perhaps 150–300 gallons per year) to the subsurface environment. A potential leak from tank T-111 is estimated to have released approximately 2,000 gallons into the subsurface. Observations of past leak events, recently published simulation results, and new simulations all suggest that recent leaks are unlikely to affect underlying groundwater above regulatory limits. However, these recent observations remind us that much larger source terms are still contained in the tanks and are also present in the vadose zone from historical intentional and unintentional releases. Recently there have been significant improvements in methods for detecting and characterizing soil moisture and contaminant releases, understanding and controlling mass-flux, and remediating deep vadose zone and groundwater plumes. To ensure extended safe storage of tank waste in SSTs, the following actions are recommended: 1) Improve capabilities for intrusion and leak detection. 2) Develop defensible conceptual models of intrusion and leak mechanisms. 3) Apply enhanced subsurface characterization methods to improve detection and quantification of moisture changes beneath tanks. 4) Maintain a flux-based assessment of past, present, and potential tank leaks to assess risks and to maintain priorities for applying mitigation actions. 5) Implement and maintain effective mitigation and remediation actions to protect groundwater resources. These actions will enable limited resources to be applied to the most beneficial actions. A systems-based approach will support extended safe storage of tank waste, reduce the risks from tank leaks, and protect human health and the environment.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Evolution of chemical conditions and estimated solubility controls on radionuclides in the residual waste layer during post-closure aging of high-level waste tanks

This document provides information specific to H-Area waste tanks that enables a flow and transport model with limited chemical capabilities to account for varying waste release from the tanks through time. The basis for varying waste release is solubilities of radionuclides that change as pore fluids passing through the waste change in composition. Pore fluid compositions in various stages were generated by simulations of tank grout degradation. The first part of the document describes simulations of the degradation of the reducing grout in post-closure tanks. These simulations assume flow is predominantly through a water saturated porous medium. The infiltrating fluid that reacts with the grout is assumed to be fluid that has passed through the closure cap and into the tank. The results are three stages of degradation referred to as Reduced Region II, Oxidized Region II, and Oxidized Region III. A reaction path model was used so that the transitions between each stage are noted by numbers of pore volumes of infiltrating fluid reacted. The number of pore volumes to each transition can then be converted to time within a flow and transport model. The bottoms of some tanks in H-Area are below the water table requiring a different conceptual model for grout degradation. For these simulations the reacting fluid was assumed to be 10% infiltrate through the closure cap and 90% groundwater. These simulations produce an additional four pore fluid compositions referred to as Conditions A through D and were intended to simulate varying degrees of groundwater influence. The most probable degradation path for the submerged tanks is Condition C to Condition D to Oxidized Region III and eventually to Condition A. Solubilities for Condition A are estimated in the text for use in sensitivity analyses if needed. However, the grout degradation simulations did not include sufficient pore volumes of infiltrating fluid for the grout to evolve to Condition A. Solubility controls for use in a flow and transport model were estimated for 27 elements in each of the chemical stages generated in the grout simulations plus local groundwater. The grout simulations were run with the initial infiltrating fluid in equilibrium with atmospheric oxygen to account for degradation of the reduction capacity of the grout. However, a lower Eh was used in pore fluids in the oxidizing conditions used to estimate solubilities to be more consistent with measured Eh values and natural systems. Solubilities of plutonium are affected by this decision, but those of other elements are not. In addition, the baseline for H-Area tanks is that they will be washed with oxalic acid prior to being filled with grout. Hence, oxalate was included in the pore fluids by assuming equilibrium with calcium oxalate. Solubility estimates were done by equilibrating a solubility controlling phase for each element with the pore fluid compositions using The Geochemist’s Workbench®. Condition B pore fluids are similar to Condition D. Therefore, solubilities for Condition B were not estimated, but assumed to be the same as in Condition D. In general solubility controlling phases were selected to bias solubilities to higher values. Several elements had no solubility controls and solubility estimates for other elements were omitted because the elements had short half-lives or were present in residual waste in very low amounts. For these it is recommended that release from the tank be instantaneous when the tank liner is breached. There is considerable uncertainty in this approach to enabling a flow and transport model to account for variable waste release. Yet, it is also flexible and requires much less computing time than a fully coupled reactive transport model. This allows some of the uncertainty to be addressed by multiple flow and transport sensitivity cases. Some of the uncertainties are addressed within this document. These include uncertainty in infiltrate composition, grout mineralogy, and disposition of certain components during the simulations. Uncertainty in the solubility estimates is addressed in part by examining sensitivity of solubilities for four key elements to uncertainty in thermodynamic data, pH, Eh, total inorganic carbon concentration, and oxalate concentration. In addition, the solubilities for the elements estimated here are compared to two other compilations of solubilities in cementitious materials.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Lifecycle Verification of Tank Liner Polymers

This report describes a method that was developed for the purpose of assessing the durability of thermoplastic liners used in a Type IV hydrogen storage tank during the tank s expected service life. In the method, a thermoplastic liner specimen is cycled between the maximum and minimum expected working temperatures while it is differentially pressurized with high-pressure hydrogen gas. The number of thermal cycling intervals corresponds to those expected within the tank s design lifetime. At prescribed intervals, hydrogen permeation measurements are done in situ to assess the ability of the liner specimen to maintain its hydrogen barrier properties and to model its permeability over the tank lifetime. Finally, the model is used to assess whether the steady-state leakage rate in the tank could potentially exceed the leakage specification for hydrogen fuel cell passenger vehicles. A durability assessment was performed on a specimen of high-density polyethylene (HDPE) that is in current use as a tank liner. Hydrogen permeation measurements were performed on several additional tank liner polymers as well as novel polymers proposed for use as storage tank liners and hydrogen barrier materials. The following technical barriers from the Fuel Cell Technologies Program MYRDD were addressed by the project: D. Durability of on-board storage systems lifetime of at least 1500 cycles G. Materials of construction vessel containment that is resistant to hydrogen permeation M. Lack of Tank Performance Data and Understanding of Failure Mechanisms And the following technical targets1 for on-board hydrogen storage systems R&D were likewise addressed: Operational cycle life (1/4 tank to full) FY 2017: 1500 cycles; Ultimate: 1500 cycles Environmental health & safety Permeation and leakage: Meets or exceeds applicable standards Loss of useable H2: FY 2017: 0.05 g/h/kg H2; Ultimate: 0.05 g/h/kg H2

hydrogen permeation polymer temperature cycling du↗

Testing of Alternative Abrasives for Water-Jet Cutting at C Tank Farm

Legacy waste from defense-related activities at the Hanford Site has predominantly been stored in underground tanks, some of which have leaked; others may be at risk to do so. The U.S. Department of Energy’s goal is to empty the tanks and transform their contents into more stable waste forms. To do so requires breaking up, and creating a slurry from, solid wastes in the bottoms of the tanks. A technology developed for this purpose is the Mobile Arm Retrieval System. This system is being used at some of the older single shell tanks at C tank farm. As originally planned, access ports for the Mobile Arm Retrieval System were to be cut using a high- pressure water-jet cutter. However, water alone was found to be insufficient to allow effective cutting of the steel-reinforced tank lids, especially when cutting the steel reinforcing bar (“rebar”). The abrasive added in cutting the hole in Tank C-107 was garnet, a complex natural aluminosilicate. The hardness of garnet (Mohs hardness ranging from H 6.5 to 7.5) exceeds that of solids currently in the tanks, and was regarded to be a threat to Hanford Waste Treatment and Immobilization Plant systems. Olivine, an iron-magnesium silicate that is nearly as hard as garnet (H 6.5 to 7), has been proposed as an alternative to garnet. Pacific Northwest National Laboratory proposed to test pyrite (FeS 2 ), whose hardness is slightly less (H 6 to 6.5) for 1) cutting effectiveness, and 2) propensity to dissolve (or disintegrate by chemical reaction) in chemical conditions similar to those of tank waste solutions. Cutting experiments were conducted using an air abrader system and a National Institute of Standards and Technology Standard Reference Material (SRM 1767 Low Alloy Steel), which was used as a surrogate for rebar. The cutting efficacy of pyrite was compared with that of garnet and olivine in identical size fractions. Garnet was found to be most effective in removing steel from the target; olivine and pyrite were less effective, but about equal to each other. The reactivity of pyrite, compared to olivine and garnet, was studied in high-pH, simulated tank waste solutions in a series of bench-top experiments. Variations in temperature, degree of agitation, grain size, exposure to air, and presence of nitrate and nitrite were also studied. Olivine and garnet showed no sign of dissolution or other reaction. Pyrite was shown to react with the fluids in even its coarsest variation (150-1000 μm). Projected times to total dissolution for most experiments range from months to ca. 12 years, and the strongest control on reaction rate is the grain size.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Investigation of thermolytic hydrogen generation rate of tank farm simulated and actual waste

To support resolution of Potential Inadequacies in the Safety Analysis for the Savannah River Site (SRS) Tank Farm, Savannah River National Laboratory conducted research to determine the thermolytic hydrogen generation rate (HGR) with simulated and actual waste. Gas chromatography methods were developed and used with air-purged flow systems to quantify hydrogen generation from heated simulated and actual waste at rates applicable to the Tank Farm Documented Safety Analysis (DSA). Initial simulant tests with a simple salt solution plus sodium glycolate demonstrated the behavior of the test apparatus by replicating known HGR kinetics. Additional simulant tests with the simple salt solution excluding organics apart from contaminants provided measurement of the detection and quantification limits for the apparatus with respect to hydrogen generation. Testing included a measurement of HGR on actual SRS tank waste from Tank 38. A final series of measurements examined HGR for a simulant with the most common SRS Tank Farm organics at temperatures up to 140 °C. The following conclusions result from this testing. HGR was measured for an actual-waste Tank 38 sample. The HGR measurements at 80, 90, 100, and 110 °C were 1.7×10 -7 , 4.7×10 -7 , 5.9×10 -7 , and 2.6×10 -6 ft 3 hr -1 gal -1 , respectively. The Limit of Detection and Limit of Quantification for the HGR measurements were 2.6×10-8 ft3 hr-1 gal-1 and 5.1×10 -8 ft 3 hr -1 gal -1 , respectively. HGR was measured in a high boiling point (HBP) simulant with major known SRS Tank Farm organic compounds. Testing used fresh organics and did not simulate the long-term exposure of these organics already experienced in the waste. Hence, testing is expected to provide high bias to the hydrogen generation rates. The final HGR measurements at 75, 100, and 140 °C were 6.2×10 -7 , 3.0×10 -6 , and 8.1×10 -5 ft 3 hr -1 gal -1 , respectively. Decreases in the HGR were noted at isothermal conditions with steady gas tracer concentration, which clouds development of a temperature dependent correlation for generation rate. While the simulant contains high concentrations of major SRS organics tested at high temperatures found in the 3H Evaporator, this limited data set may not fully bound values under all possible operating conditions. The HGR measurement apparatus was successfully employed to replicate hydrogen production rate experiments. Simulant without deliberately added organic carbon was found to contain approximately 200 mg/L of organic carbon from tramp organics in the reagents used to prepare the solution. The tests using such simulant generated hydrogen at rates consistent with, but somewhat lower than, those predicted at equivalent carbon and aluminum loadings by methods used at the Hanford Site. Prior to identifying the organic contaminants, potential sources of hydrogen in simulants without deliberately added organic carbon were examined. Testing determined that this hydrogen is not due to caustic reaction of organic materials of construction within the reaction vessel, electrolysis due to current leak, anode/cathode reactions, corrosion, or the method of heating employed in the experiment. During the HBP and Tank 38 testing, a peak was identified at a retention time greater than the retention time of the inert tracer of krypton. This peak has been identified as methane, and the data have been reexamined to obtain methane concentrations. For the HBP testing, methane concentration was constant at the highest temperature studied (140 °C) and higher than hydrogen. At the lower temperatures (100 and 75 °C), methane and hydrogen tracked closely, with methane concentration generally lower than hydrogen. For Tank 38, methane concentration was approximately 1/3 of the hydrogen concentration during testing (100 and 110 °C). At lower temperatures for Tank 38 testing (90, 80, and 27 °C), methane fell below the quantification range.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗