Pollution Prevention Technologies for Plutonium Processing

A. U.S. Department of Energy Laboratory LALP-93-92

POLLUTION PREVENTION TECHNOLOGIES FOR PLUTONIUM PROCESSING

NUCLEAR MATERIALS TECHNOLOGY 1994 LALP-93-92 April 1994

Published by Los Alamos National Laboratory Nuclear Materials Technology Division MS E500 Los Alamos, New Mexico 87545

Coordinator K.K.S. Pillay

Editors Carolyn K. Robinson Joan Farnum

Graphic Designer Susan L. Carlson

Printing Coordination Guadalupe D. Archuleta Brenda R. Valdez

For More Information K.K.S. Pillay (505) 667-5428

Page 65 Photos: figure 3—B 1420 01 figure 4—B 1425 15

Los Alamos National Laboratory is operated by the University of California for the U.S. Department of Energy under contract W-7405-ENG-36.

An Affirmative Action/Equal Opportunity Employer.

This report was prepared as an account of work sponsored by the United States Government. Neither the Regents of the University of California, the United States Government, nor any of their employees makes any warranty, expressed or implied, or assumes legal liability or responsibility for the accuracy, com- pleteness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, mark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the Regents of the University of California, the United States Government or any agency thereof. The views and opinions expressed herein do not necessarily state or reflect those of the Regents of the University of California, the United States Government, or any agency thereof. Contents

C o n t e n t s

Preface Preventing Pollution From Plutonium Processing by K.K.S. Pillay 3

Overviews - Waste Minimization Technologies Off-Gas Scrubbing by W. Brad Smith 7

Acid Recovery and Recycle by Thomas R. Mills 12

Hydrochloric Acid Waste Stream Polishing by Louis D. Schulte, Richard R. Salazar, Benjie T. Martinez, John R. FitzPatrick, and Bradley S. Schake 16

A Novel Liquid-Liquid Extraction Technique by Daniel J.Kathios, Gordon D. Jarvinen, Stephen L. Yarbro, and Barbara F. Smith 20

Electrochemical Treatment of Mixed Hazardous Waste by Wayne H. Smith, Christine Zawodzinski, Kenneth R. Martinez, and Aaron L. Swanson 37

Application of Freeze-Drying Technology to Decontamination of Radioactive Liquids by Nicholas V. Coppa 44

Nuclear Applications for Magnetic Separations by Larry R. Avens, et al. 52

Distillation Separation of Actinides from Waste Salts by Eduardo Garcia, Vonda R. Dole, Walter J. Griego, John J. Lovato, James A. McNeese, and Keith Axler 61

Evaluating Existing Partitioning Agents for Processing Hanford High-Level Waste Solutions by S. Fredric Marsh, Zita V. Svitra, and Scott M. Bowen 66

Gas Pycnometry for Density Determination of Plutonium Parts by J. David Olivas, S. Dale Soderquist, Henry W. Randolph, Susan L. Collins, Darren Verebelyi, William J. Randall, and Gregory D. Teese 71

Contents continued on next page

1 Contents

C O N T E N T S Continued

Plutonium Dry Machining by Mark R. Miller 75

Technical Issues in Plutonium Storage by John M. Haschke and Joseph C. Martz 80

Overviews - Facility Operations Waste Management at Technical Area 55: A Historical Perspective and Status Report by Bill J. McKerley, James J. Balkey, and Ronald E.Wieneke 88

Future of Waste Management at the Los Alamos Plutonium Facility by Ronald E. Wieneke 94

Nuclear Materials Technology 1994 Report 2 Preface

P r e f a c e Preventing Pollution From Plutonium Processing

K. K. S. Pillay Introduction Nuclear Materials Technology Pollution is a dirty word. No one has to make a case anymore for the need for pollution prevention or the related concept of environmental management. Although the history of pollution control legislation in the United States dates back to the River and Harbor Act of 1899, the past four decades have been most active in developing and administering numerous environmental regulations. This process of administering environmental regulations has accelerated in recent years and has become a critical factor in the quality of life in these United States.

Since the passage of the Atomic Energy Act in 1954, the Congress of the United States has promul- gated numerous environmental statutes. Figure 1 shows the accumulation of environmental statues and their amendments that are presently applicable to nuclear defense facilities. These statutes have been codified into over 9000 regulations by federal agencies.1 The Environmental Protection Agency (EPA) alone has over 125,000 employees administering these statutes and regulations. There is an equal number of environmental regulators at the state and local government levels.

Although radioactive wastes have been accumulating in the United States for the past half century, only during the last two decades has there been a concerted effort to address the problems of long-term isolation of these wastes from the human environment. After considerable study and debate, the U.S. program for Fig.1. Accumulation of environmental statutes and their amendments radioactive waste management since the passage of the Atomic Energy Act in 1954. was codified in the Nuclear Waste Policy Act of 1982.2

3 Preface Preventing Pollution From Plutonium Processing

According to this legislation, The Nuclear Materials Technol- The following paragraphs the U.S. Department of Energy ogy (NMT) Division, in this new examine the legacy of nuclear (DOE) has the responsibility cultural climate, adopted a stra- waste, the new priority of waste to develop strategies, systems, tegic goal to systematically minimization, responsible waste and technologies for the long- eliminate the adverse environ- management, and technological term isolation of all spent mental impacts associated with challenges for the future. nuclear fuels, high-level wastes, operating the Los Alamos Pluto- and transuranic wastes in the nium Facility. This ambitious The Legacy of Nuclear Waste United States. Because of the goal of a committed group of The term “nuclear waste” is unique socio-political atmo- people in the Los Alamos generally used to represent all sphere surrounding radioactive nuclear technology community radioactive waste materials re- waste management, there are is the theme of this compilation gardless of their origin, material considerable challenges to the of technology initiatives. Our form, or radioactivity levels. scientific community for devel- strategic goal is not only part of Nuclear wastes originate from a oping technological solutions a survival strategy but also rep- multitude of activities involving that are acceptable to a resents our readiness to solve nuclear technologies in both concerned free society. national environmental prob- the civilian and defense sectors. lems associated with nuclear The steady accumulation of For nearly four decades, the U.S. processing facilities and, nuclear wastes over the past half nuclear defense facilities sought thereby, help alleviate public century has attracted consider- refuge under the Atomic Energy and governmental concerns. able public attention during the Act and claimed sovereign im- past two decades. Following the munity from compliance with Since September 1992, several reactor accidents at Three Mile most of the environmental regu- activities within NMT Division Island and Chernobyl, there lations. In response to increasing have been postponed because have been overwhelming ex- public pressure during the past of the FFCA and the long and pressions of public concerns few years, the defense sector has laborious negotiations that are over safety of all nuclear facili- undergone a cultural shift that required to reach a facility- ties and the safe management has changed past attitudes to- specific agreement with the of accumulated nuclear wastes ward complying with environ- EPA. On July 19, 1993, DOE as well as those that are being mental regulations. This shift Secretary Hazel O’Leary ap- generated. has culminated in the passage proved the Los Alamos agree- of the 1992 Federal Facility ment with Region 6 of the EPA. Studies have shown that the Compliance Act (FFCA) and According to this agreement, public’s perception of risk from other complementary regula- full compliance with the FFCA nuclear waste is similar to those tions and executive orders.3,4 is expected by the year 2000. of highly energetic, catastrophic The FFCA gives federal facilities During these negotiations, events. This perception is a three-year period to develop NMT Division began to imple- qualitatively incorrect because detailed plans for complying ment its strategic plan to meet radioactive wastes have far with all applicable environmen- all regulatory requirements of lower energy levels in compari- tal regulations. This particular waste minimization and pollu- son with reactor accidents, legislation has been a great tion prevention and to transform earthquakes, volcanic eruptions, motivator to all federal facilities, the Los Alamos plutonium pro- etc. An educational process that including the Los Alamos Pluto- cessing operations into an envi- conveys realistic risks of nuclear nium Processing Facility, to ronmentally benign activity. wastes has to be a necessary modify past practices and to In developing those technolo- part of an overall program of become conscious of the envi- gies required to meet our goals, nuclear waste management. ronmental impacts of all its we will also serve the needs of activities. the weapons complex of the next century.

Nuclear Materials Technology 1994 Report 4 Preface Preventing Pollution From Plutonium Processing

Arguments about the costs and environment are completely possibility now and can accom- benefits of nuclear technologies eliminated. The DOE has recog- plish additional waste reduction. continue to dominate public nized that if it is to successfully Waste treatment to reduce vol- discussions. Attitudes of both achieve its long-term goals, this ume, toxicity, and mobility are all governmental agencies and agency must first find a way to in various stages of development nuclear industries based on change public perceptions about and implementation. artificial cost-benefit analyses nuclear waste. Focusing on have contributed enormously waste minimization and DOE’s ambitious plan for waste to the negative attitudes sur- responsible waste management minimization can be taken a step rounding nuclear wastes in the are steps in the right direction. further using new technologies United States. Unfortunately, essential to new nuclear facilities these analyses do not factor in Waste Minimization that will have little or no impact the significant costs arising from Waste minimization is any ac- on the environment. Liquid environmental impacts consid- tion that avoids or reduces the waste-stream polishing ered unacceptable by the general generation of wastes. For nearly to eliminate radioactive and public. a decade, defense nuclear facili- hazardous components of wastes ties have been voluntarily mak- from nuclear materials process- The unmaking of nuclear wastes ing efforts to reduce nuclear ing facilities for the weapons and the abrogation of all uses of waste generation. In recent fuel cycle is feasible and can be nuclear technologies are not in years, DOE identified environ- implemented in large-scale the realm of reality. As we mental restoration and waste nuclear materials processing approach the 21st Century, we management as a major mission facilities. Approaches to mini- cannot afford to forgo the many area.6 The primary objective of mizing radioactive and hazard- peaceful uses of nuclear tech- the fourth annual Environmen- ous components in solid waste nologies. Considerable resources tal Restoration and Waste Man- streams are being developed, have been devoted during the agement Five-Year Plan for and the costs for implementing past two decades to the design 1994-98 is the reduction of such technologies are not ex- of technical solutions for long- wastes from DOE facilities.7 pected to be prohibitive. Because term management of nuclear a modern nuclear material pro- wastes and isolation of this Waste reduction can be accom- cessing facility is necessary to waste from the biosphere. There plished through source reduc- maintain a nuclear deterrent ca- is a world-wide scientific con- tion, recycling potential waste pability, a vigorous technology sensus that deep geologic dis- materials that cannot be elimi- development program for waste posal is the best option for nated or minimized, and treat- minimization is essential to disposing of existing inventories ing all wastes to reduce their maintain public confidence in of high-level and transuranic ra- volume, toxicity, or mobility nuclear programs and to change dioactive wastes.5 However, it prior to storage or disposal. present attitudes toward nuclear is now abundantly clear that the The significant reduction in waste. public and its representatives nuclear weapons requirements will not be satisfied until re- represents a major source reduc- leases of radioactive nuclides tion mechanism. Recycling of from man-made sources to the a number of waste stream components is a technical

5 Preface Preventing Pollution From Plutonium Processing

Responsible Waste Manage- materials and its impact on our Cited References ment and Future Challenges future is an international politi- 1. Philip H. Abelson, “Pathological Future nuclear processing facili- cal problem with no obvious so- Growth of Regulations,” editorial, Science 260, 1859 (June 25, 1993). ties have a major responsibility lutions. Achievements in the to properly treat and dispose of area of nuclear arms reduction are eclipsed by the increasing 2. U.S. Congress, “Nuclear Waste their wastes. These facilities Policy Act of 1982, ” Public Law 97- must be able to function without likelihood of nuclear prolifera- 425, 97th Congress (January 1983). the release of radioactive and tion. Although we can propose hazardous materials to the envi- to control proliferation through 3. U.S. Congress, “Federal Facility ronment. Dilution and disper- systematic reduction in fissile Compliance Act of 1992,” Section sion of wastes practiced in the material inventories worldwide, 6001 of Solid Waste Disposal Act, 42 past are not acceptable solutions the need for energy production USC 6961, Public Law 102-386, any longer. Immobilization and in many parts of the world using 102nd Congress (1992). isolation of wastes from the bio- fissile and fertile materials must be met. 4. “Federal Compliance with Right sphere will have only a limited to Know and Pollution Prevention role in the long-term manage- These challenges are opportuni- Requirements,” Presidential ment of nuclear wastes. Reposi- Executive Order of August 3, 1993, tory construction and geologic ties for NMT Division to de- Federal Register 58 (180), pp. 41981- disposal as a solution to nuclear velop and apply technologies 41987 (August 6, 1993). waste problems cannot be con- that can prevent environmental tinued forever; however, stabili- pollution. Technology transfer 5. Board on Radioactive Waste Man- zation and storage have to be initiatives now being actively agement, “Rethinking High-Level seriously considered as an op- promoted by DOE offer oppor- Radioactive Waste Disposal,” a posi- tion so that valuable resources tunities for NMT Division to tion statement (National Academy are not discarded for political develop university and industry Press, Washington, D.C., 1990). partnerships that can be mutu- expediency. Managed storage 6. J. D. Watkins, “Posture State- of concentrated nuclear materi- ally beneficial. Technology initiatives detailed in the following ment,” DOE/CR-0011, U.S. Depart- als may become the policy for ment of Energy, Washington, D.C. the national defense sector. pages of this report are steps (January 1993). toward achieving our overriding The scientific community faces goal of an environmentally be- 7. U.S. Department of Energy, “Envi- numerous other challenges re- nign plutonium processing facil- ronmental Restoration and Waste ✦ sulting in part from present pub- ity at Los Alamos. Management Five-Year Plan,” lic perceptions of nuclear DOE/S-00097P, Washington, D.C. technologies. These challenges (January 1993). include meeting the conflicting needs of nuclear proliferation control and nuclear energy production. Increasing concerns over the proliferation of nuclear

Nuclear Materials Technology 1994 Report 6 Waste Minimization Technologies Off-Gas Scrubbing

W. Brad Smith Introduction Nuclear Materials Processing: Effectively handling corrosive off-gases produced by aqueous- Chloride Systems based processing of Special Nuclear Materials is an enigma for most facility operations. Resolving the problem is not just limited to scrubbing the corrosive off-gases. Project research and design must also consider containment design, ventilation control, the effects of radiation and other contaminants, monitoring and control systems, maintenance, and waste minimization. The comprehensive plan for the aqueous recovery operations at the Los Alamos Plutonium Fa- cility incorporates a minimum of local corrosion-resistant material, “The comprehensive plan for the capture of all corrosive acid vapors near the source, and monitoring of the glove-box exhaust system. The first phase of the plan the aqueous recovery opera- consists of consolidation of all chloride-based processing to mini- tions at the Los Alamos mize the number of containment systems requiring special materi- Plutonium Facility incorpo- als and maintenance. The second phase involves the design and evaluation of the corrosion-resistant containment systems such as rates a minimum of local glove-box linings, dispersion-coated ventilation systems, and lined corrosion-resistant material, process piping. The third phase is comprised of various efforts to protect the local utility support systems; these efforts include the the capture of all corrosive use of an off-gas central scrubber for the chloride recovery glove- acid vapors near the source, box exhaust system and local caustic scrubbers for the chloride and and monitoring of the glove- nitrate vacuum transfer systems. This paper first examines the extent of corrosion caused by acid vapors in the Los Alamos Pluto- box exhaust system.” nium Facility, reviews the history of integrating aqueous-based chloride processing into our nuclear processing facility, and then focuses on the operation of the off-gas central scrubber, vitally important to implementing the third phase of the Los Alamos plan for aqueous recovery operations.

Acid Vapor Corrosion at the Plutonium Facility Concerns about acid vapor corrosion at the Plutonium Facility have grown with the increase in applications for acid-based aqueous processing, the buildup of local corrosion, and the trend toward shorter-lived HEPA (high efficient particulate air) filters. Other facilities are concerned about the corrosion problems caused by the increased use of hydrofluoric acid, hydrochloric acid (HCl), chlorine, and caustics.1 Acidic liquid corrosion may cause more problems than acid vapor corrosion; however, the problems caused by acid vapor corrosion can be more serious. Effects such as chloride stress cracking of stainless-steel systems, HEPA filter seal or material failures, and acid-level buildup in the utilities can be more difficult to contain, monitor, or evaluate as compared to liquid- based corrosion effects.

7 Waste Minimization Technologies Off-Gas Scrubbing

Since most standard industry The selective application of cor- History of Integrating Aqueous- techniques for determining the rosion-resistant material in the Based Chloride Processing extent and rate of corrosion are second phase of the plan has The Los Alamos graded ap- not feasible in a plutonium pro- been technically and economi- proach to implementing and cessing facility, the Los Alamos cally feasible primarily because demonstrating an integrated plan for aqueous recovery op- of the first-phase consolidation aqueous chloride flow sheet was erations is based on process of aqueous chloride operations, governed by the severity of the knowledge and the specific which routinely vent low, but effects from HCl corrosion. This system’s susceptibility to corro- significant, concentrations of graded approach proceeded as sion. Visual inspection of the HCl vapors. The Plutonium follows: early emphasis on new enclosures and the glove-box Facility was originally designed liquid-processing equipment exhaust plenums provides the for nitric acid (HNO3) opera- and transfer systems, followed only evidence of the extent of tions, so even these low concen- by secondary containment of corrosion. With the exception trations pose a long-term liquids, then ventilation of some nonstainless appurte- problem for the Facility. Corro- system upgrades, and finally nances, the glove boxes have ex- sion problems do exist with point-source off-gas scrubbing. perienced the greatest corrosion. other operations, but the impact The location is typically near from chloride corrosion presents New liquid processing equip- the glove-box bottom, where liq- the biggest potential for failure ment and transfer systems were uids collect. Similarly, the local due to stress-corrosion cracking. developed using chlorinated ducts are significantly more pit- Although the Plutonium Facility and fluoropolymer-based plas- ted near the duct bottoms than has not experienced any pri- tics, such as CPVC (chlorinated the duct tops. The scale buildup mary containment failures be- polyvinyl chloride) end plates, is highest nearer the glove-box cause of corrosion, many factors commercial-grade plastic piping exhaust, most probably due to such as general process knowl- and valves for internal glove- condensation and the subse- edge, scale buildup in the local box transfer systems, PVDF quent ferric precipitates from ducts, and localized glove-box (polyvinylideneflouride) slab dissolved iron components. corrosion indicate material in- tanks for process storage, and The rate of corrosion is also ex- compatibilities exist. Further- plastic-lined piping for all exter- tremely difficult to evaluate more, because of the variety of nal liquid transfer systems. due to the variations in operat- process operations and the size This use of plastics reduced ing times, concentrations, and of the Facility, replacing all ex- corrosion at the most susceptible chemical interactions. Specific posed systems with special cor- points, provided a level of safety upsets with processes and rosion-resistant materials would and reliability similar to typical equipment, such as the evapora- not be economically feasible metal systems, and actually im- tion process for nitrate-based and, thus, preventative mea- proved accessibility and flexibil- operations, dissolution pro- sures become necessary. Preven- ity in processing. cesses, loss of condenser cooling tative measures at Los Alamos water, or liquid carryover from consist of off-gas scrubbing at Aqueous batch processing in a vacuum transfer operations, the various sources of acid va- research and development facil- may cause more corrosion than por combined with selective ma- ity can result in spills of minor the continuous release of low- terial upgrades and real-time, amounts of process liquids be- concentration fume gas. Cur- in-line monitoring to ensure the cause of process-tubing changes rently, no system or procedure performance of these scrubbers. and routine operations or main- is in place to monitor these tenance; thus, secondary con- upset conditions. tainment of liquids is essential.

Nuclear Materials Technology 1994 Report 8 Waste Minimization Technologies Off-Gas Scrubbing

Drawing on experience gained Although the exact cause has The consolidation of the aque- with fluoropolymers for process not been determined, the de- ous chloride operations provides equipment, we used crease in filter efficiency corre- the opportunity to collect and fluoropolymer-based plastics sponds to an increase in treat all HCl fugitive emissions to protect enclosures, appurte- aqueous operations in this area. with a central scrubber and en- nances, and utilities. The glove- To trace possible causes, access hances the life cycle of the Pluto- box shell, which provides the ports for periodic sampling of nium Facility systems that don’t secondary containment, has his- the ventilation exhaust for acid use chloride-resistant material. torically been fabricated of 316 gas, for measurement of exhaust stainless steel, as are most ap- flow rates, for corrosion coupon An evaluation of commercial purtenances. The published cor- tests, and for visual examination fume gas scrubbers found one rosion rate for 316 stainless steel of scale buildup have been in- system that met the stringent rein concentrated (10 M) aqueous corporated into the local glove- quirements for integration with HCl is at least 0.05 inches per box exhaust ducts, where the the Plutonium Facility systems. year2 (versus a rate of 0.002 acid gas concentrations should Some of the features provided inches per year for aqueous be the highest. by this system include auto- HNO3). This corrosion rate matic, unattended operation; only provides for an approxi- Point-source off-gas scrubbing low maintenance; critically safe mate 10-year life cycle for was originally implemented to liquid waste storage; waste 7-gauge stainless steel. Evalua- protect the vacuum transfer minimization; and fail-safe shut- tion and testing of several differ- system. Fume gas was origi- down. The system is sized to ent glove-box linings led to a nally vented directly to the treat the exhaust from four to new specification3 for the fabri- glove-box atmosphere from fifteen glove boxes and has a to- cation and installation of dissolution, feed treatment, tal system pressure loss about fluoropolymer enclosure linings, and evaporator processes. half the available static pressure which provide both mechanical These concentrated fumes, of the glove-box exhaust system. and chemical protection.4 which condense on and etch Because this scrubbing process the glove-box windows, re- provides a utility support func- duced visibility. The humidity tion, which only enhances the Ventilation equipment has been also caused the local HEPA filter safety and life cycle of the Facil- modified for compatibility with to swell and plug prematurely. ity, it can be easily automated low concentrations of acid va- Local scrubbers now capture without the burden of a “safety pors. All mild steels used for approximately 90% of all HCl class” designation. HEPA fire screens and supports acid vapors and 50% of all have been replaced with stain- HNO3 vapors. The HCl central scrubber is less-steel components. How- designed to remove acid gas ever, recent studies5 have shown HCI Central Scrubber Design fumes from twelve glove boxes. a decrease in the efficiency of the For some processes the vacuum The process equipment will be three-stage Facility HEPA filters transfer system cannot provide installed in specially designed for the aqueous chloride pro- the necessary volume to collect enclosures to meet ALARA (as cessing ventilation system. the acid vapors. These fugitive low as reasonably achievable) Trends toward higher DOS* pen- emissions are vented through contamination control goals as- etration rates are evident for the the glove-box exhaust system sociated with maintenance op- first-stage HEPA filters but are and continuously diluted by ex- erations and instrument still an order of magnitude be- haust from other areas. Al- calibration. Fumes will be col- low the design failure point. though the local scrubbers lected in a local corrosion-resis- remove a majority of the fumes, tant exhaust duct, removed by some concern about the effects liquid absorption in a packed of chlorides on stainless-steel column scrubber, and then systems, neoprene seals, and vented to the glove-box exhaust * di (2 ethylhexyl) sebacate, which HEPA materials still exists. system. (Refer to Fig. 1 for this replaces DOP (dioctyl phthalate). and other design details that follow.)

9 Waste Minimization Technologies Off-Gas Scrubbing

The maximum static pressure Upset or maintenance condi- As HCl vapor is absorbed, the for the glove-box exhaust is a tions will also force the system pH of the scrubber solution will negative 1.3-inch water column into a bypass mode. Emergency decrease. Batch processing (WC). The maximum scrubber Shutdown (ESD) logic will be causes the influent concentration system pressure loss is a 0.7-inch developed and monitoring of HCl to vary over a wide WC. Therefore, a very sensitive installed to insure proper system spectrum. Therefore, the scrub- instrumentation system is response. Sufficient local gauges ber liquid discharge rate (see pH required to monitor and control and manual overrides will be Control Valve, AV) will be the pressure loss. Initially, the provided to insure fail-safe proportional to changes in the pressure will be maintained by operation. However, failure of HCl concentration, which is allowing the flow to decrease as any component will still not monitored by measuring the pH. the HEPA filters and scrubber affect the glove-box exhaust This sidestream of scrubber packing plug up. As the pres- system flow and pressure solution will be piped to a set of sure loss increases above a stability, minimizing the need slab tanks, and an equal amount predetermined set point, the for redundant or “safety-class” of fresh water will be added (see system will go into a bypass systems. Level Control Valve, LV) to mode (see Flow Control Valves, maintain an optimum pH of 2. designated as FV, in Fig. 1).

Fig. 1. Scrubber Piping and Instrument Diagram.

Nuclear Materials Technology 1994 Report 10 Waste Minimization Technologies Off-Gas Scrubbing

The water level in the scrubber A mass spectrometer is the most plugging of the proposed post- sump will have redundant feasible method to monitor the scrubber HEPA. The use of a monitors. One system (see effectiveness (HCl removal) of post-scrubber HEPA (which Level Control, LC) will be an the scrubbers. This type of could be installed in one of the analog signal used as an over- continuous, on-line instrumenta- utility support enclosures to ride for the pH control to main- tion is required to evaluate simplify change out and dis- tain proper pump suction with multiple acid-gas concentra- posal) between the central alarm indication for high and tions, provide wide rangeability scrubber exhaust and the main low levels. A second set of level to determine upset conditions, ducts would help insure that no switches (not shown), high-high and provide detailed glove-box contamination would spread to and low-low, will be used to exhaust air analysis on a con- the Facility exhaust ducts.¨ shut down the scrubber and tinuous basis. place the ventilation Flow Cited References Control Valves in a bypass Waste Minimization 1. U. S. Department of Energy mode. A tertiary level control, To minimize liquid waste, the Headquarters of Nuclear Safety, consisting of an open drain line, central scrubber will have a “Incident Summary,” in “Opera- shall be used to prevent any liquid discharge rate that is tions Weekly Report, No. 92-26” possibility of liquid depths proportional to changes in the (October 16-22, 1992). above 4 inches (because of system pH, as previously dis- criticality concerns). A flow- cussed. This control system will 2. Chemical Engineers Handbook, limiting device may also be reduce the projected waste from Fifth Edition, R. Perry and considered to limit the total 4000 gallons per month to C. Chilton, Eds. (McGraw Hill, Inc., amount of water used on a approximately 120 gallons per New York, 1973), p. 23-18, periodic basis. month. This waste stream will Table 23-3. be neutralized and then sent The primary contamination through the caustic waste line to 3. Los Alamos National Laboratory, control for glove boxes is the the Los Alamos Liquid Waste “Technical Specification for Enclo- negative pressure for the glove- Treatment Facility. One possible sure Linings,” Specification NMT8- box exhaust system in the alternative to treating the waste 1166-R00 (January 1990). aqueous chloride processing is to recycle or reuse the scrub- area. Therefore, the pressure ber solution and thereby elimi- 4. M. Dinehart, “Corrosion Resis- will be monitored (not shown), nate liquid waste altogether. tant Linings for Glove-Box Enclo- and any excursions beyond set sures,” in Proceedings of NACE point will override the bypass Another approach being consid- Conference, Houston, Texas, Flow Control Valves. A redun- ered for solid waste minimiza- August 1991. dant pressure switch will acti- tion is to replace local glove-box vate an alarm and also place the HEPA filters, which require ventilation Flow Control Valves frequent changing because of in a bypass mode. swelling due to the absorption of moisture from the aqueous The exhaust from the scrubber operations, with corrosion- may be saturated with water; resistant roughing filters and a the resultant humidity will affect post-scrubber HEPA. Roughing HCl detectors and air flow filters will prevent alpha con- monitors, and condensation tamination of the local ducts due will increase the possibility of to particulates, provide more corrosion in the main ducts. consistent flow and more consis- Using a local chiller to cool the tent pressure control, and reduce scrubber solution will lower the dew point as well as increase the removal efficiency of the scrubber.

11 Waste Minimization Technologies Acid Recovery and Recycle

Thomas R. Mills Introduction Actinide Materials Chemistry Plutonium processing, both for waste handling and for recovery, uses large amounts of nitric and hydrochloric acids. Typical processes involve dissolving some form of plutonium in these acids for purification. These acid-based processes effectively accomplish needed separations, but they generate large volumes of waste acid solution.

The previous approach to handling these solutions has been to neutralize them with caustic and then partition radioactive components from the bulk of the solution. The radioactive fraction is put into a concrete matrix (saltcrete), and the remaining solution is dis- “Minimizing the charged to the environment, after treatment to remove radioactivity. At Los Alamos, the bulk of nitric acid (HNO3) is separated from concentrations and nonvolatile salts (including most of the plutonium), but both frac- amounts of plutonium tions are presently disposed of rather than recovered and recycled. and neutralized salts sent This approach leads to environmental and storage problems. to waste disposal supports With both our heightened commitment to protecting the environ- the Nuclear Materials ment and future regulatory requirements, reduction of generated wastes is essential. Including handling, inspection, packaging, Technology Division's shipping, and actual storage, the cost of producing waste units can vision of zero radioactive be tremendous. Beyond cost implications, concerns over long-term and hazardous discharges.” storage of radioactive waste dictate that both the amounts be minimized and that the storage forms be highly stable. In particular, a potential for leaching of soluble salts from saltcrete exists, and such leaching could render the storage form less stable.

Recycling acids from waste solutions helps solve many of these problems because this method eliminates most of the waste generated by plutonium processing. Hazardous chemical discharges (notably nitrates) are greatly reduced, and already low levels of radioactive discharges are further reduced. This reduction in the total number of radioactive waste units requiring storage lowers disposal costs significantly; for example, acid recycle would save over $50 million in waste disposal costs in five years for a facility operating at levels similar to the previous levels of the Rocky Flats Plant. The reduced salt content in radioactive wastes, an additional benefit of acid recovery and recycle, reduces the potential for leaching of stored radioactive wastes.

Minimizing the concentrations and amounts of plutonium and neutralized salts sent to waste disposal supports the Nuclear Materials Technology (NMT) Division’s vision of zero radioactive and hazardous discharges.

Nuclear Materials Technology 1994 Report 12 Waste Minimization Technologies Acid Recovery and Recycle

Our three-year goals are 1) a 90% recycle of HNO3, 2) a 90% reduction of acid Distillation Evaporator Column concentration in waste streams, and 3) a 10-fold reduction of Water nitrate concentration in Volatile effluent liquid. Waste Acid Components Our five-year goals are Solution 1) an operating recycle system for both acids that will eliminate effluent acid Nonvolatile Recycle streams, Salts Acid 2) a reduction of nitrate wastes by a factor of 500, and 3) a reduction of chloride Fig.1. Block diagram of acid recovery and recycle system. wastes by a factor of 10. Approach A two-stage recovery and re- The acid/water vapor mixture HNO3 and Experimental Chlo- cycle process is planned. First, from the evaporator will be ride Extraction Line (EXCEL) each acid waste stream is evapo- separated in the distillation col- for HCl. ATLAS already has a rated to separate volatiles (pri- umn to yield a concentrated acid thermosyphon evaporator as an marily acid and water) from (³ 12 M HNO3, 6 M HCl) at the integral part; a distillation col- dissolved radioactive solids and column bottom and a very di- umn will be added as an up- other inorganic salts. Second, lute acid stream (“water”) at the grade. HCl recycle is at an early the evaporated acid and water column top. The separated acid stage; thus, characterization of are separated in a fractional dis- will be sufficiently concentrated HCl processes (which includes tillation column to yield a con- to be reused for processing, and flow balances) and bench-scale centrated acid and a water it will contain minimal pluto- evaporation and distillation stream. This two-stage process nium or other constituents. experiments are needed. Both is shown by the block diagram The water stream will require an evaporator and a distillation in Fig. 1. pH adjustment to neutrality column will be added to EXCEL. prior to discharge to the envi- The nonvolatile salt residue ronment. This effluent stream Following demonstrations using from evaporation, which con- will have a negligible plutonium these systems, full-scale systems tains any plutonium and inor- content and will have a salt con- will be built to handle all acid ganic salts from solution, tent much less than at present. waste streams from the Pluto- represents a small fraction of the nium Facility. The acid recovery original acid solution. This frac- Nitric and hydrochloric acid and recycle effort will have action will be neutralized and recovery and recycle will be tive participation from members dried and may receive further demonstrated at the Los Alamos of the Nitrate Systems (NMT-2), treatment (e.g., denitration or Plutonium Facility using the two the Chloride Systems (NMT-3), freeze drying1) before disposal. integrated demonstration lines: and the Actinide Materials The resulting solid will be dis- Advanced Testing Line for Ac- Chemistry (NMT-6) groups. posed of in a concrete matrix tinide Separations (ATLAS) for or an alternative ceramic form.

13 Waste Minimization Technologies Acid Recovery and Recycle

Project Status Important to retention of radio- Preliminary designs have been The Nitrate Systems Group, us- active components, a demister made for HNO3 and HCl distilla- ing the ATLAS thermosyphon bed in the evaporator minimizes tion columns for ATLAS and EX- evaporator, has already con- entrainment of fine droplets CEL. The design for a HNO3 ducted valuable initial experi- with dissolved solids. distillation system is based on ments with HNO3. Promising handling 1 L/min of 4–6 M results indicate that acid recycle The distillation column contains HNO3. A 6-inch-diameter, is feasible. A simulated process loose-poured packing to support 6-foot-tall distillation column waste solution containing 4.5 M a dispersed liquid phase trick- can concentrate at least 95% of HNO3 was evaporated with the ling downward. The packing the HNO3 in this stream to 12 M goal of separating the maximum provides maximum interfacial HNO3. The waste stream will possible fraction of contained area for vapor-liquid equilib- have less than 5% of the acid acid. At evaporation rates of rium but offers minimal resis- feed, and plutonium activity will 0.7–0.8 L/min, over 82 mol % tance to upward vapor flow. be lower than that obtained using of the HNO3 was recovered for The column has a steam boiler evaporation only. “Cold” tests further concentration. The vol- for vapor boilup and a partial are under way on the HNO3 ume of solution remaining in the condenser for liquid reflux. distillation column to determine evaporator was only 2% of the A schematic diagram of the packing performance for flow amount fed to the evaporator, evaporator and column is capacity and separating ability. and this solution was flushed shown in Fig. 2. The evaporator Following these tests, the column from the evaporator before the and column combination for will be installed in ATLAS. acid concentration process was EXCEL will have a similar begun. In a second pass of clean configuration. distillate through the evapora- Condenser tor, steady HNO3 concentrations were reached within 30 minutes. In this second pass, 43 mol % of the original nitric acid feed was Demister concentrated to 10.7 M HNO3. In a third pass, 21 mol % of the Condenser original nitric acid was collected as 12 M HNO3. Liquid Steam Level Boiler A physical design for acid re- Water cycle systems has been devel- Waste Acid oped for use in the present Solution Steam ATLAS system. The HNO3 Boiler Recycle distillation column is closely Acid coupled to the ATLAS evapora- Nonvolatile tor to take advantage of thermal Salts energy invested. Vapor exiting the evaporator becomes the Fig. 2. Schematic of thermosyphon evaporator and acid distillation feed to the column. The existing column. thermosyphon evaporator uses a steam bundle to produce vigorous boiling and liquid circulation through the boiler.

Nuclear Materials Technology 1994 Report 14 Waste Minimization Technologies Acid Recovery and Recycle

(In parallel, we are discussing Summary with a vendor possible purchase Development of acid recovery of acid recovery systems that and recycle will enable NMT Di- use technology developed at vision to reduce wastes on a fa- Pacific Northwest Laboratories.) cility-wide basis. Our research A similar system for HCl distil- will demonstrate that hazardous lation is expected to produce and radioactive chemicals can be 6 M HCl from a feed of 1 L/min reduced and eliminated from liq- of 2–4 M HCl. uid effluent streams in an existing plutonium facility. In addition, this important technol- The highest volume of the ogy will be available for the plu- HNO3 effluent stream from the tonium facility yet to come — Plutonium Facility in recent Complex 21.✦ years was about 75,000 liters per year. Although the capacity of Cited Reference ATLAS (~0.75 L/min) might 1. N. V. Coppa, “Application of conceivably handle this stream, Freeze Drying Technology to the required duty cycle of 80% Decontamination of Radioactive (1667 hours) is unrealistic for a Liquids,” this report. plutonium facility. Full-scale acid recycle systems will be designed and built using experience gained from the ATLAS and EXCEL demonstrations. These systems will address the five-year goals.

15 Waste Minimization Technologies Hydrochloric Acid Waste Stream Polishing

Louis D. Schulte, Introduction Richard R. Salazar, Aqueous processing of plutonium residues in hydrochloric acid and Benjie T. Martinez (HCl) produces waste effluent streams that require several treat- Nuclear Materials Processing: ment steps before the liquids may be discharged to the environment. Chloride Systems Effluents from HCl processing streams are currently combined and routed to controlled hydroxide precipitation as the first step in ac- John R. FitzPatrick, tinide decontamination. This step neutralizes the entire inventory Bradley S. Schake of process acid, coprecipitates many other metal salts with the ac- Inorganic and Structural tinides, and produces a solid cake that is often too high in ameri- Chemistry Group cium and plutonium for eventual WIPP (Waste Isolation Pilot Plant) Chemical Science and Technology disposal. The liquid effluent from controlled hydroxide precipita- Division tion (usually about 109 dpm per liter) is the largest contributer of alpha activity for solutions sent to the TA-50 Liquid Waste Treatment Facility. Final treatment steps at TA-50 include a ferric flocculation process, which produces additional barrels of transuranic (TRU) solid waste.

The purpose of this work is to evaluate extraction chromatography techniques and materials as an alternative method to removing actinides from aqueous Hcl effluent streams. Part of the challenge has been to develop and match extraction methods to specific waste stream requirements for best overall actinide decontamination. Effluent streams may be categorized as “high-acid streams” “The advantages of apply- (~6 M HCl) which originate from solvent extraction raffinate or ing chromatographic tech- anion exchange effluent, or “low-acid streams” (~1 M HCl) which niques to decontamination may come from the filtrate of plutonium(III) oxalate precipitation or the eluate of an anion exchange “guard column” used to remove of radioactive aqueous iron from the process stream. The advantages of applying chro- effluents include generation matographic techniques to decontamination of radioactive aqueous effluents include generation of smaller quantities of solid residues of smaller quantities of solid in forms suitable for storage, reduction of TRU waste volumes, and residues in forms suitable more efficient decontamination of aqueous effluents. for storage, reduction of High-Acid Effluent Streams TRU waste volumes, and The majority of our early research has concentrated on high-acid more efficient decontamina- streams for several reasons: 1) high americium content presents the greatest tion of aqueous effluents.” decontamination challenge, 2) increasingly low limits on activity content of grout waste forms indicate the need for isolation of the americium in a stable form for storage or vitrification, 3) removal of americium early in the processing flowsheet will reduce personnel irradiation exposure from subsequent waste handling steps, 4) decontamination of the effluent would provide feed for a simple acid recycle scheme for about 90% of the HCl used in plutonium processing, and

Nuclear Materials Technology 1994 Report 16 Waste Minimization Technologies Hydrochloric Acid Waste Stream Polishing

5) segregation of the high-acid stream from other effluent factors (DF) of up to 104. DF is streams simplifies some aspects of waste polishing design. defined for a particular solution as actinide concentration before Separation of actinides from the other elements in the waste treatment ÷ actinide concentra- stream, mostly alkali or alkaline earth metals, appears feasible tion after treatment. Elution of using present-generation technology. CMPO (n-octyl(phenyl)- plutonium(IV) from the loaded N,N-diisobutylcarbamoylmethylphosphine oxide) and related columns in flow experiments molecules have been examined for use in liquid-liquid extraction has been slower than desired schemes in hydrochloric media to remove actinides.1, 2 Extraction but is expected to be improved chromatography, using similar extractants physically sorbed to a by altering the elution condi- solid support, has been successfully used to concentrate actinides tions. Loading studies with on an analytical scale; however, testing of these techniques for plutonium(IV) indicate that the larger-scale applications has received scant attention.3–5 Our work resins should be useful for HCl has demonstrated extraction of sizable quantities of actinides from process stream decontamina- HCl media containing alkali or alkaline earth metals and shows tion. The maximum plutonium promise for development into a reversible chromatographic system. loading observed in this study was 107 mg per gram of resin, Several resins coated with CMPO diluted in either DAAP (diamyl a loading capacity approaching amylphosphonate) or TBP (tributyl phosphate) were tested. Con- one-half that of a standard an- tact studies with plutonium(IV) demonstrated a distribution coeffi- ion exchange resin (typically cient (Kd) of ~104 in 6 M HCl, and a Kd near 1 when reduced to about 1 millimole per gram of plutonium(III) in 1–4 M HCl (see Fig. 1). Kd is defined as actinide resin or ~240 mg actinide per quantity per gram of resin ÷ actinide quantity per milliliter of solu- gram of resin). tion. Flow studies of plutonium(IV) have shown decontamination Americium(III) removal has proven more difficult. Contact studies of several CMPO-bearing resins with americium(III) have shown dramatic variation in observed Kd, dependent upon the amount and type of diluent used to disperse CMPO on the bead. Contact studies to remove americium(III) on a resin loaded with pure CMPO (no diluent) in 6 M HCl demonstrated the largest Kd (~103), but this resin was kinetically too slow to be useful in a flow system. Other resins with better kinetic properties gave a Kd of about 102, high enough to re- tain significant americium(III) but with breakthrough larger than desired in flow experiments. We are continuing to design and test new extractive resin materials tailored to more effectively remove americium(III) from these Fig. 1. Changes in Pu retention on 20:20 CMPO/DAAP resin with NaNO2 high-acid effluent streams. and NH2OH additions.

17 Waste Minimization Technologies Hydrochloric Acid Waste Stream Polishing

very strongly coordinating diphosphonic acid ligands (not desirable for our process) or by flooding the column with selected stable elements in dilute acid.8 Our preliminary displacement experiments indicated that Am+3 should be effectively eluted using Al+3 or Fe+3. Further work is required to fully evaluate the promising potential of Diphonix resin for low-acid waste stream treatment.

Conclusions Major improvements in the form and quantity of actinide wastes generated from HCl plutonium processes are possible. Treatment of specific process effluents at the source, with extraction methods tailored for the individual stream, offers the best chance of successful actinide decontamination. The payoff will be concentration of actinides into smaller volumes of storable or treatable Fig. 2. Changes in Am(III) retention on Diphonix with Fe(III) addition. solid forms, decontamination of high-acid stream effluents to activity levels that allow facile HCl Low-Acid Effluent Streams recycle, reduction in alpha activity in waste solutions released The low-acid effluent streams will require a fundamentally different to TA-50, and reduction in the approach for decontamination because of the presence of additional volume of TRU solid waste chemical interferents. Oxalic acid competes with neutral ligands generated.✦ like CMPO in complexation of actinides in HCl, especially at low- acid concentrations. Iron(III) and certain other transition metal elements, present in the effluent stream from the guard column, act as interferents for CMPO-type ligands. For these reasons we are examining Diphonix,* a commercial polystyrene resin containing covalently-bound gem-diphosphonic acid groups, for these applications. Preliminary experiments with this resin have shown a very good uptake of americium from dilute HCl solutions with a measured contact Kd of about 105 (see Fig. 2). Other workers have demonstrated that the Diphonix resin is capable of selectively removing actinide(III) ions from most other elements in acidic solutions and with little interference from oxalate.6,7 Elution of actinide(III) elements from the column may be accomplished with

* Diphonix resin is commercially available from EIChroM Industries, Inc., Darien, Illinois. Nuclear Materials Technology 1994 Report 18 Waste Minimization Technologies

Hydrochloric Acid Waste Stream Polishing

Cited References 5. G. S. Barney, and R. G. Cowan, 1. E. P. Horwitz, H. Diamond, and “Separation of Actinide Ions from K. A. Martin, “The Extraction of Se- Radioactive Waste Solutions Using lected Actinides in the (III), (IV) and Extraction Chromatography,” (V) Oxidation States from Hydro- Westinghouse Hanford report, chloric Acid by OFD(iB)CMPO: WHC-SA-1520-FP (1992). The TRUEX-Chloride Process,” Solvent Extr. Ion Exch. 5 (3), 447–470 6. E. P. Horwitz et al., “Uptake of (1987). Metal Ions by a New Chelating Ion- Exchange Resin. Part 1: Acid De- 2. E. P. Horwitz, H. Diamond, pendencies of Actinide Ions,” and K. A. Martin, “Extraction of Solvent Extr. Ion Exch. 11 (5), 943- Americium(III) by Octyl(phenyl)-N,N- 966 (1993). diisobutylcarbamoylmethylphosphine oxide,” Solvent Extr. Ion Exch. 5 (3), 7. R. Chiarizia et al., “Uptake of 419–446 (1987). Metal Ions by a New Chelating Ion- Exchange Resin. Part 2: Acid De- 3. W. I. Yamada, L. L. Martella, and pendencies of Transition Metal and J. D. Navratil, “Americium Recovery Post-Transition Metal Ions,” Solvent and Purification Using a Combined Extr. Ion Exch. 11 (5), 967-985 (1993). Anion Exchange-Extraction Chroma- tography Process,” Journal of the Less 8. E. P. Horwitz, Argonne National Common Metals 86, 211–218 (1982). Laboratory, personal communication, August 1993. 4. G. J. Lumetta, D. W. Wester, J. R. Morrey, and M. J. Wagner, “Preliminary Evaluation of Chro- matographic Techniques for the Separation of Radionuclides from High-Level Radioactive Waste,” Solvent Extr. Ion Exch. 11 (4), 663–682 (1993).

19 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Daniel J. Kathios, Introduction Gordon D. Jarvinen, The cleanup of sites where nuclear materials were used and the Stephen L. Yarbro, treatment of wastes accumulated during the Cold War require the and Barbara F. Smith development of cost-effective and efficient technologies for the pro- Nuclear Materials Processing: cessing of radioactive wastes. Such technologies separate radioac- Nitrate Systems tive materials from the nonradioactive materials that make up the bulk of the wastes. The radioactive materials are sometimes further partitioned into their long-lived and short-lived radioactive components. The wastes can then be disposed of in a less expensive fashion and in a manner that significantly reduces their potential threat to the environment. Some of the radioactive materials may also be suitable for reuse.

A variety of liquid-liquid extraction (LLE) technologies are used for the processing of nuclear materials. LLE is a process in which two immiscible liquids (commonly an aqueous and an organic phase) are allowed to come in contact so that a solute (substance dissolved “LLE is a process in which in a solvent) present in one of the phases can be transferred to the two immiscible liquids other. The overall process is composed of two separate steps — (commonly an aqueous extraction and back-extraction. In the extraction step, an aqueous acidic stream, in which is dissolved a mixture of radioactive and and an organic phase) are nonradioactive species, comes in contact with an organic stream allowed to come in contact containing an extractant specifically designed to remove these radioactive metals. In the back-extraction step, this organic stream so that a solute present in comes in contact with a less acidic stream into which the extractant one of the phases can be liberates these radioactive metals. The metals from both aqueous transferred to the other.” effluent streams (leaving the extraction and back-extraction steps) are then recovered by evaporation, precipitation, electrodeposition, or other methods.

An LLE technology best suited for the applications of interest to this research must satisfy a variety of criteria. The technology must operate in a continuous mode so as to process large amounts of wastes in a reasonable amount of time. It must be resistant to the highly acidic and radioactive environments to which it will be exposed. And, because many applications will be in a contained environment, the technology must require little space, be safe and simple to use, and require little or no maintenance. LLE technologies being considered for these applications include mixer settlers, pulsed columns, spray columns, and centrifugal contactors. Each of these technologies uses its own mode of phase mixing to expedite mass transfer of the solute between the two phases. 1,2 But unfortunately, this mixing promotes aqueous/organic phase disper- sions that not only deplete the organic from the LLE process but also reduce its overall extraction and back-extraction capabilities.

Nuclear Materials Technology 1994 Report 20 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

This dispersion problem can be avoided by using a recently devel- MHF modules possess a number oped technology called a microporous hollow fiber (MHF) mem- of features that make them an brane module. This technology does not require phase mixing for attractive technology for LLE good mass transfer. Instead, the aqueous and organic phases come processes.3-5 The small size of in contact with each other through a rigid, thin, and porous mem- the hollow fibers permits a de- brane as illustrated in Fig. 1. The membrane in this diagram is sign that achieves a very high interfacial area in a given volume, which helps to minimize the size of the LLE process. Extractions and back-extractions are conducted in a dispersion-free fashion, thus avoiding phase coalescence problems commonly found in other technologies. There is no need to have a difference in aqueous and organic phase densities, allowing for a wider array of organic systems to be taken into consideration. The aqueous/ organic phase ratios can be easily varied over a wide range by merely adjusting the flow rates. The modular nature of the units makes it easy to accurately scale up experimental results to process applications and to engi- neer any multiple module configuration that would optimize the process. Finally, the modules are inexpensive (capital costs are low); they have no moving parts (maintenance Fig. 1. Diagram of a porous hydrophobic membrane. The membrane costs are low); and once run- provides a region of stable and intimate contact between the aqueous and ning, they require little supervi- organic phases where dispersion-free mass transfer can take place. sion, monitoring, or external control (operational costs are low). made of a hydrophobic material which allows the organic phase to The purpose of this research is enter its pores or “wet” the membrane. When a slight positive pres- to evaluate the potential for us- sure is induced on the aqueous phase, the organic phase is held ing MHF modules for the pro- within the pores, and a region of stable and intimate contact be- cessing of radioactive wastes. tween the two phases is established where dispersion-free mass This is to be accomplished by transfer can take place. An MHF module is illustrated in Fig. 2. analyzing the extraction and It consists of an outer casing containing a large number of hollow back-extraction capabilities of fibers fabricated from this porous membrane material. The aqueous the modules for different or- phase is passed through the inside of the hollow fibers (on the tube ganic systems over a range of side), and the organic phase is passed over the fiber bundle (on the aqueous and organic flow rates. shell side).

21 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Fig. 2. Diagram of an MHF module. The design achieves a high interfacial surface area per unit volume and allows the aqueous and organic phases to contact in a continuous-flow fashion. C fi and C fo are the respective solute concentrations in the aqueous feed and effluent, and Cei and Ceo are the respective solute concentrations in the organic feed and effluent.

and CMPO [n-octyl(phenyl)N,N- diisobutylcarbamoylmethylphosphine oxide] are the organic phase extractants used in this study. These two extractants are chosen for their ability to extract neodymium at modest acid con- The results are used to determine the number of modules required centrations (2 M nitric acid) and to achieve one theoretical extraction stage, and to provide valuable for their high selectivity for ac- insights as to how this technology can be improved and optimized tinides over fission products. for process-scale applications. Each of these extractants removes trivalent metal ions from the In this work, MHF modules are used to extract neodymium from aqueous phase by forming a com- an aqueous 2 M nitric acid stream and to concentrate the neodym- plex as shown in Fig. 3. The re- ium into a different aqueous stream by back-extracting it into verse reaction occurs during the 0.01 M nitric acid. Although neodymium is not radioactive, back-extraction process. Addi- its behavior is similar to that of trivalent americium (a long-lived tional information about these actinide) and thus serves as a surrogate for this evaluation. extractants is available in the DHDECMP (dihexyl N,N-diethylcarbamoylmethylphosphonate) literature.6-9

Fig. 3. The complexation of neodymium with CMPO. Three extractant molecules are required to complex each neodymium ion.

Nuclear Materials Technology 1994 Report 22 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Theory and Definitions

General Equations

The MHF module illustrated in Fig. 2 permits a continuous-flow countercurrent extraction (or back-extraction). The rate at which the solute is transferred from the aqueous phase to the organic phase can be determined by conducting a material balance around the aqueous stream. Doing so yields

˙ , (1) M = Qaq (C fi - C fo) where M˙ is the rate of mass transfer from the aqueous to the organic phase, Qaq is the aqueous flow rate, andC fi andC fo are the respective solute concentrations in the aqueous feed and effluent. The rate of mass transfer can also be determined by conducting a material balance around the organic stream. This approach yields ˙ M = Qor(Ceo - C ei ) , (2) whereQor is the organic flow rate, andCei andCeo are the respective solute concentrations in the organic feed and effluent. An additional way to characterize the rate of mass transfer is in terms of an overall mass transfer coefficient. For an extraction in which the aqueous phase is flowing through the inside of fibers fabricated from a hydrophobic membrane material, this rate is expressed as

˙ , (3) M = K w Ai DClm where Kw is the overall mass transfer coefficient of the module based on the aqueous phase, Ai is the total membrane surface area of the module based on the internal fiber diameter, and DClm is the logarithmic mean concentration difference given by

DC1 -DC2 DClm = . (4) ln(DC1 DC2 )

DC1 is the driving force for mass transfer at the aqueous inlet and is equal to the solute concentration in the aqueous feed minus the interfacial aqueous phase solute concentration at the aqueous inlet:

interface . (5) DC1 = C fi - C fi

interface Determining C fi is far from trivial, and its value is commonly approximated as the concentration that is in equilibrium with the solute concentration in the organic effluent. This is expressed as C C interface @ eo , (6) fi m

23 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

where m is the distribution coefficient defined as A more comprehensive development of the general equations æsolute concentrationö that characterize two-phase ç in organic phase ÷ countercurrent flow extraction m = and back-extraction processes ç ÷ . (7) ç solute concentration÷ can be found in Bennett.10 è in aqueous phase ø equilibrium Resistances to Mass Transfer

Equation 5 can thus be expressed as Mass transfer of a solute C through a porous hydrophobic eo membrane occurs in three fun- DC1 = C fi - . (8) m damental steps. First, the solute

DC2 is the driving force for mass transfer at the aqueous outlet and travels through the aqueous (in a manner consistent with the determination of DC1 ) is expressed phase to the aqueous/organic as interface where complexation takes place. Second, the solute Cei DC2 = Cfo - . (9) complex diffuses through the m membrane pore. Third, the solute complex travels through the organic medium. Expressing Equation 3 as DC M˙ = lm (10) 1 (K w Ai )

shows that the rate of mass transfer is equal to a driving force term, DClm , divided by a resistance term, 1 Kw Ai . The latter term incorporates the individual resistances to mass transfer existing in each of the three fundamental steps and i s thus referred to as the overall resistance to mass transfer. These resistances are illustrated in Fig. 4. Equations provided by Prasad3,4 and Sirkar3-5 make it possible to quantify these resistances.

Fig. 4. Resistances to mass transfer in the extraction and back- extraction processes.

Nuclear Materials Technology 1994 Report 24 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

For modules fabricated with hydrophobic membrane fibers and the fiber, Dc,or is the diffusion assembled for tube side aqueous flow and shell side organic flow, coefficient of the solute complex these resistances can be expressed as in the organic, is the membrane porosity, and is the 1 1 di ln(do / di ) di = + + . (11) tortuosity of the membrane Kw Ai kwt Ai 2 mDc,or Ai domkos Ai pores. Although Equation 11 is presented here in terms of the • • • • extraction process, its applicabil- Overall Aqueous Membrane Organic ity is equally valid for the back- Resistance (Tube Side) Resistance (Shell Side) extraction process as well. Resistance Resistance Number of Modules per Theo- retical Stage In this equation, kwt is the mass transfer coefficient of the solute in the aqueous phase inside the fibers, kos is the mass transfer coeffi- A theoretical stage in an LLE cient of the solute complex in the organic phase outside the fibers, process is one that first allows di is the internal diameter of the fiber, do is the outer diameter of mass transfer between the aqueous and the organic phases to reach equilibrium and then allows complete ( i.e., dispersion- free) separation of the two phases to occur before they exit the process. A theoretical stage with continuous aqueous and organic flow is illustrated in Fig. 5. The rate at which solute is transferred from the aqueous to the organic phase can be determined by conducting a material balance around the aqueous stream. Doing so yields

˙ TS TS M = Qaq (C fi - C fo ) , (12) TS where M˙ is the rate of mass transfer from the aqueous to the organic phase in the theoretical TS stage, and C fo is the solute concentration in the aqueous effluent leaving the theoretical stage. Conducting a material balance Fig. 5. An ideal theoretical stage in a continuous-flow LLE process. around the organic stream yields C and C are the respective solute concentrations in the aqueous and fi ei TS TS ˙ TS TS organic feed streams entering the theoretical stage, and C fo and Ceo aree M = Qor(Ceo - Cei ) , (13) the respective solute concentrations in the aqueous and organic effluent TS streams leaving the theoretical stage. where Ceo is the solute concentration in the organic effluent leaving the theoretical stage.

25 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Equations 12 and 13 can be used with the distribution coefficient TS TS TS TS equation, m = Ceo C fo , to solve for C fo and Ceo as a function of inlet concentrations and flow rates. Doing so yields

TS QaqC fi + QorCei C fo = (14) Qaq + mQor and

TS QaqC fi + QorCei Ceo = . (15) Qaq m + Qor

For a number of modules connected in series that provide the total membrane surface area required for a theoretical stage, the rate at which mass transfer occurs is

˙ TS TS TS , (16) M = K w Ai DClm TS where A is the total membrane surface area of these modules i TS based on the internal fiber diameter and DClm is the logarithmic mean concentration difference in a theoretical stage. The latter term is given by TS TS TS DC1 -DC2 DClm = TS TS , (17) ln(DC1 DC2 ) where TS TS C DC = C - eo (18) 1 fi m and C DCTS = CTS - ei . (19) 2 fo m

Dividing Equation 16 by Equation 3 yields ˙ TS TS TS M Ai DClm . (20) ˙ = M Ai DClm TS Substituting Equation 12 for M˙ and Equation 1 for M ˙ , and solving TS for Ai Ai yields ATS æC - C TS ö DC TS i ç fi fo lm N = = ÷ TS . (21) Ai è Cfi - C fo ø DClm

TS where N is the number of modules required for a theoretical stage.

Nuclear Materials Technology 1994 Report 26 Waste Minimization Technologies A Novel Liquid-Liduid Extraction Technique

Experimental Details

Preparation of Aqueous and Organic Solutions

Three different organic systems are considered in this analysis: 0.25 M CMPO in DIPB, 0.81 M DHDECMP in DIPB, and 0.25 M DHDECMP in DIPB. A total of 5 L of organic solution is prepared for each extraction (or back-extraction) experiment. Additional information about these chemicals is provided in Table I.

Table I. Extractants and Solvent used to Make the Different Organic Systems

Extractant #1: Chemical dihexyl N, N-diethylcarbamoylmethyl phosphonate Abbreviation DHDECMP Supplier Occidental Chemical Corporation

Extractant #2: Chemical n-octyl (phenyl)-N, N-diisobutylcarbamoylmethylphosphine oxide Abbreviation CMPO Supplier M&T Chemicals, Inc.

Solvent: Chemical diisopropylbenzene, 98% (approximately 50% meta, 40% para, and 10% ortho) Abbreviation DIPB Supplier Eastman Laboratory Chemicals

The aqueous solution for the extraction experiments is prepared by dissolving 2.5 g of neodymium nitrate hexahydrate in 5 L of 2 M nitric acid resulting in a neodymium concentration of 166 mg/L. The organic solution is prepared by equilibrating it against an equal volume of 2 M nitric acid. This equilibration prevents nitric acid mass transfer from affecting the neodymium extraction process.

The aqueous solution for the back-extraction experiments is prepared by making 5 L of 0.01 M nitric acid. The organic solution is first equilibrated against an equal volume of 0.01 M nitric acid and is then added to 15 g of neodymium nitrate hexahydrate, resulting in a neodymium concentration of 996 mg/L. The organic is later decanted from the trace water liberated during the dissolution process.

27 Waste Minimization Technologies

A Novel Liquid-Liquid Extraction Technique

Extraction and Back-Extraction Table II: Characteristics of Hoechst Celanese Liqui-Cel MHF Experiments Module

The experiments are conducted using the Hoechst Celanese Model No.: 5PCM-100 Liqui-Cel Laboratory LLE Sys- Description: High Efficiency Phase Contactor tem. The MHF modules subject to analysis are Hoechst Celanese Shell Characteristics: Liqui-Cel High Efficiency Phase Contactors. These modules are Material polypropylene small laboratory-scale versions Length 20.3 cm that are specifically designed for Internal Diameter 2.43 cm experimental purposes. Their Outer Diameter 3.34 cm small size makes it possible to evaluate the performance of the Fiber Characteristics: technology without having to Material polypropylene (Celgard X-10) prepare large amounts of aque- Number of Fibers 3600 ous and organic feed solutions. Effective Length 15.9 cm Additional information about Internal Diameter 240 mm the Hoechst Celanese Liqui-Cel Outer Diameter 300 mm module is provided in Table II. Effective Surface 2 A schematic of the experimental Area 4300 cm setup is illustrated in Fig. 6. In Effective Pore Size 0.05 mm the experiment, the aqueous Membrane Porosity 0.30 feed is pumped through the Membrane Tortuosity 2.6 tube side of the MHF module where it comes in contact with the organic feed and neodymium mass transfer takes place. Upon exiting the module, the aqueous stream is passed During the first half of each experiment, the organic flow rate is through a flowmeter and then held constant at 5 mL/minute, and the performance of the module through a flow cell present in a is observed over a range of aqueous flow rates. For the second half, diode array spectrophotometer the aqueous flow rate is held constant at 8 mL/minute, and the programmed for continuous, on- module’s performance is observed over a range of organic flow line neodymium concentration rates. After a particular aqueous flow rate and organic flow rate are measurement. The stream is established, a period of time must elapse for the system to reach then collected in the aqueous steady state. This point is determined when the aqueous effluent effluent reservoir. The organic neodymium concentration (measured as spectrophotometric absor- feed is pumped through the bance at 578 nm) becomes constant with respect to time. At this shell side of the module, is point, 5 mL of the aqueous effluent is removed from the sample passed through a flowmeter, and port with a syringe and is placed in a 20-mL screw-cap scintillation is ultimately collected in the or- vial for subsequent determination of its neodymium concentration. ganic effluent reservoir. Valves To ensure repeatability, three 5-mL samples are taken at each steady and pressure gauges are present state. to control the flow rates and to ensure that a positive pressure of approximately 5 psig is maintained on the aqueous side of the membrane.

Nuclear Materials Technology 1994 Report 28 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Fig. 6. Experimental setup used to evaluate the extraction and back- extraction capabilities of the MHF modules.

Neodymium Concentration Determination

Each extraction (or back-extraction) experiment generates more than 36 aqueous samples of unknown neodymium concentration. Because the absorbance peak for neodymium (exhibited at 578 nm) is very weak, an ac- ceptably precise determination of its concentration in these samples is not feasible by direct spectrophotometric measurement. However, neodymium does form a complex with arsenazo III that exhibits a strong absorbance peak at 655 nm.11 Taking advantage of this, the unknown concentrations can be accurately determined.

In this approach, 0.1 g/L of arsenazo III is added to a buffer solution of 1 M glacial acetic acid and 0.5 M sodium acetate. A 10-mL aliquot of this solution Evaluation of Distribution Coefficients is placed in a 20-mL screw-cap scintillation vial followed by The distribution coefficients are determined by using small 500 L of the aqueous sample amounts of the aqueous and organic feed solutions prepared for the of unknown neodymium con- extraction and back-extraction experiments. In this determination, centration. The resulting solu- 10 mL of the aqueous solution and 10 mL of the organic solution tion is thoroughly mixed for are placed in a 20-mL screw-cap scintillation vial and are thor- one minute using a vortexer oughly mixed for one minute using a vortexer. The mixture is then and is allowed to sit for one allowed to sit for two days to ensure that complete phase separa- day to ensure that complexation tion has taken place. Using a syringe, 5 mL of the aqueous fraction has reached equilibrium. The is removed and is placed in a new screw-cap scintillation vial for absorbance at 655 nm is then subsequent determination of its neodymium concentration. Four measured with the spectropho- distribution coefficient determinations are conducted for each aque- tometer. The neodymium con- ous/organic system. centration of each unknown is determined by comparing its measured absorbance with those obtained from a series of calibration standards.

29 Waste Minimization Technologies

A Novel Liquid-Liquid Extraction Technique

Table III. Distribution Coefficients of Neodymium for Three Different Organic Systems

Organic Distribution Distribution System Coefficient for Coefficient for Extraction Back-Extraction (2 M HNO ) (0.01 M HNO ) 3 3 -2 0.25 M CMPO 8.8 3.2 x 10 in DIPB

0.81 M DHDECMP 3.6 1.3 x 10-3 in DIPB

-5 0.25 M DHDECMP 0.25 5.8 x 10 in DIPB

Results and Discussion All extractions and back-extractions conducted with the MHF modules were clean, easy to run, and absolutely dispersion-free. No measurable loss of organic was detected during any of the experiments.

The distribution coefficients for the different organic systems are shown in Table III. A superior organic system for extraction is one in which the solute preferentially dissolves in the organic phase and thus exhibits a large distribution coefficient. From the organic systems considered, the best for this purpose is the 0.25 M CMPO in DIPB, next is the 0.81 M DHDECMP in DIPB, and last is the 0.25 M DHDECMP in DIPB. An organic system best suited for back-extraction is one in which the solute preferentially dissolves in the aqueous phase and thus exhibits a small distribution coefficient. The best system for this purpose is the 0.25 M DHDECMP in DIPB, next is the 0.81 M DHDECMP in DIPB, and last is the 0.25 M CMPO in DIPB. Under the conditions used, the organic system with the greatest affinity to remove neodymium during the extraction process is also the one least inclined to liberate the neodymium during the back-extraction process.

Nuclear Materials Technology 1994 Report 30 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Fig. 7. Extraction of neodymium from 2 M nitric acid. Neodymium concentration in aqueous effluent is shown as a function of aqueous flow rate (left) and as a function of organic flow rate (right) for the different organic systems. Aqueous feed concentration is 166 mg/L.

The results of the extraction and back-extraction experiments are illustrated in Figs. 7 and 8. As shown in the graphs, low aqueous flow rates and high organic flow rates are best for extractions, whereas low organic flow rates and high aqueous flow rates are best for back-extractions. The low flow rates of the solute-laden streams (those from which the neodymium is to be removed) provide the high residence times necessary for substantial mass transfer to take place. In addition, the high flow rates of the solute removal streams (those that enter the process with no solute) maintain low concentrations of solute in those streams and thus maximize the concentration gradients for high mass transfer rates.

Figure 7 shows that an organic system with a larger distribution coefficient provides for a better extraction. This behavior can be explained by analyzing the individual resistances to the mass transfer process. As shown in Equation 11, the membrane resistance and the shell side resistance are both smaller for an organic system having a larger value of m . In addition, as shown in Equations 4 through 9, such a system also has a greater value of DClm – the driving force for the mass transfer process. These two factors (a decreased resistance and an increased driving force) increase the mass transfer rate and thus result in a superior extraction process.

31 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Fig. 8. Back-extraction of neodymium into 0.01 M nitric acid. Neodym- ium concentration in organic effluent is shown as a function of aqueous flow rate (left) and as a function of organic flow rate (right) for the different organic systems. Organic feed concentration is 996 mg/L.

As shown in Fig. 8, superior back-extractions are achieved with an organic system exhibiting a smaller distribution coefficient. Equa- tion 11 shows that such a system does possess higher membrane and shell side resistances. However, the value of DClm (which is inherently negative for back-extraction processes) is more negative for smaller values of m and thus provides a stronger driving force for the back-extraction process. For the systems considered, the increased magnitude of this driving force is much greater than the increased resistance. As a result, the systems with smaller distribution coefficients exhibit higher mass transfer rates for the back- extraction processes.

Nuclear Materials Technology 1994 Report 32 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Fig. 9. Extraction of neodymium from 2 M nitric acid. The number of modules required for one theoretical stage is shown as a function of aqueous flow rate (left) and as a function of organic flow rate (right) for the different organic systems.

Experimental data are used with system for extraction. Not only does such a system remove more Equation 21 to determine the solute in a theoretical stage, but also fewer modules are required for number of modules required for that theoretical stage to be achieved. a theoretical stage. As shown in Figs. 9 and 10, this number in- Comparing Figs. 9 and 10 shows that for a given organic system, creases with both aqueous and the number of modules required for back-extraction is much greater organic flow rates. Thus, longer than that required for extraction. This large difference is due to the aqueous and organic residence increased membrane resistance existing in the back-extraction pro- times are conducive to fewer cess. From Equation 11, the ratio of these two resistances can be modules. expressed as

For the two stronger extraction çæ membrane resistanceö systems (the 0.25 M CMPO in è for back - extraction ø m = extraction DIPB and the 0.81 M DHDECMP æmembrane resistance . (22) in DIPB), between one and five ç ö mback-extraction modules are required for a theo- è for extraction ø retical stage. For the weaker system (the 0.25 M DHDECMP This ratio is between 200 and 4500 for the organic systems in this in DIPB), between five and analysis. Suggestions on how this membrane resistance (and there- twenty-five modules are re- fore the number of modules) can be reduced are provided in our quired. This provides another conclusions. reason for using a strong organic

33 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Fig. 10. Back-extraction of neodymium from 0.01 M nitric acid. The number of modules required for one theoretical stage is shown as a function of aqueous flow rate (left) and as a function of organic flow rate (right) for the different organic systems.

Conclusions The modules used in this analy- In addition, the modules for this MHF modules provide a cost- sis are small laboratory-scale analysis use an early version of a effective and convenient means versions specifically designed fiber bundle design that is not by which to conduct metal sepa- for experimental purposes. conducive to good shell side rations. But initial inspection of In an actual process application, flow. According to Seibert,12 the experimental data seems to pilot-scale (or larger) modules this poor flow causes a signifi- indicate that reasonable separa- can be used, each of which pro- cant portion of the fibers to be tions can be achieved only at vides as much as 10 times the bypassed by the organic and flow rates that are much lower membrane surface area of a keeps the surface area available than those found in conven- laboratory-scale version. for mass transfer from being tional, process-scale metal sepa- Additional surface area can also fully utilized. Newer modules ration applications. But by be achieved by connecting a are now being produced with an modifying the number or type number of modules in series. advanced fiber mesh that allows of modules, the organic extrac- These approaches would make the organic to flow more uni- tion systems, or the membrane it possible to conduct reasonable formly throughout the fiber material, this technology can separations at the high flow bundle. A new shell side baffle also perform well at the higher rates normally found in process technology has also been re- flow rates typically found in applications. cently introduced that may sig- process-scale applications. nificantly improve the module’s performance.

Nuclear Materials Technology 1994 Report 34 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

The technology may also be im- The performance of the modules required for a theoretical stage proved by modifying the or- may also be improved by chang- in the back-extraction process. ganic system. For extraction, ing the membrane from a hydro- As mentioned in the previous this could be achieved by using phobic to a hydrophilic material. section, this comparably large a more powerful extractant or In a hydrophilic membrane, the number of modules is due to by using a higher concentration aqueous phase fills the pores, the small distribution coefficient of the extractant in the organic and the neodymium passes (and thus large membrane resis- phase. Such an approach would through the membrane in its tance) found in the back-extrac- reduce the membrane resistance ionic form. In a hydrophobic tion process. Using a hydro- and would increase the concen- membrane (like the polypropy- philic membrane would make tration driving force for the lene membrane used in this the membrane resistance inde- mass transfer process. But there analysis), the organic phase fills pendent of the distribution coef- is a drawback to this approach. the pores, and the neodymium ficient and would significantly As mentioned in the previous passes through the membrane reduce the size of the process. section, an organic system that is in its complexed form (see Flat sheet hydrophilic mem- more inclined to capture metals Fig. 3). The diffusivity of neo- branes fabricated from during the extraction process dymium through an aqueous polybenzimidazole have dem- may be less inclined to liberate phase is significantly greater onstrated good extraction capa- those metals during the back- than that of an inherently large bilities for neodymium and cop- extraction process. Comparison and sterically hindered neodym- per13 and are reputed to have of Figs. 7 and 8 shows that this ium complex through a viscous significant acid and radiation behavior is observed in our ex- organic phase. The greater stability. MHF modules made perimental results. Because a diffusivity means less mem- from this polymer may thus be process-scale application of this brane resistance and results in potentially well-suited for the technology would undoubtedly a greater mass transfer rate. process applications of interest encompass both extraction and Changing the membrane mate- to nuclear facilities.✦ back-extraction, the perfor- rial may also reduce the large mance of the organic system in number of modules presently both processes must be taken into consideration when choos- ing an improved organic system.

35 Waste Minimization Technologies A Novel Liquid-Liquid Extraction Technique

Cited References Carbamoylmethylphosphonates - 11. S. B. Savvin, “Analytical Use of 1. Handbook of Solvent Extraction, The Effects of Size and Structure Arsenazo III: Determination of T. C. Lo, M. H. I. Baird, and of the Extractant Molecule,” Thorium, Zirconium, Uranium, and C. Hanson, Eds. (John Wiley & Radiochimica Acta 29, 143-147 Rare Earth Elements,” Talanta 8, Sons, New York, 1983), pp. 275-507. (1981). 673-685 (1961).

2. R. A. Leonard, D. G. Wygmans, 8. E. P. Horwitz, H. Diamond, 12. A. F. Seibert, X. Py, M. Mshewa, M. J. McElwee, M. O. Wasserman, and D. G. Kalina, and J. R. Fair, “Hydraulics and and G. F. Vandergift, “The Centrifu- “Carbamoylmethylphosphoryl Mass Transfer Efficiency of a gal Contactor as a Concentrator in Derivatives as Actinide Commercial-Scale Membrane Solvent Extraction Processes,” Sep. Extractants,” in Plutonium Chemis- Extractor,” Sep. Sci. Technol. 28 Sci. Technol. 28 (1-3), 177-200 (1993). try, ACS Symposium Series, (1-3), 343-359 (1993). William T. Carnall and Gregory R. 3. R. Prasad and K. K. Sirkar, Choppin, Eds. (American Chemical 13. D. Y. Takigawa, “The Effect of “Hollow Fiber Solvent Extraction: Society, Washington, D. C., 1983), Porous Support Composition and Performances and Design,” Vol. 216, pp. 433-450. Operating Parameters on the J. Membrane Sci. 50, 153-175 (1990). Performance of Supported Liquid 9. E. P. Horwitz, D. G. Kalina, L. Membranes,” Los Alamos 4. R. Prasad and K. K. Sirkar, Kaplan, G. W. Mason, and H. National Laboratory report “Dispersion-Free Solvent Extraction Diamond, “Selected LA-12027-MS UC-704 (February with Microporous Hollow Fiber Alkyl(phenyl)-N,N- 1991). Modules,” AIChE J. 34, 177-188 dailkylcarbamoylmethylphosphine (1988). Oxides as Extractants for Am(III) from Nitric Acid Media,” Sep. Sci. 5. Membrane Handbook, W. S. W. Ho Technol. 17 (10), 1261-1279 (1982). and K. K. Sirkar, Eds. (Van Nostrand Reinhold, New York, 10. C. O. Bennett and J. E. Meyers, 1992), pp. 727-763. Momentum, Heat, and Mass Transfer, 3rd ed. (McGraw-Hill, Inc, New 6. S. F. Marsh and S. L. Yarbro, York, 1982) pp. 598-623. “Comparative Evaluation of DHDECMP and CMPO as Extractants for Recovering Actinides from Nitric Acid Waste Streams,” Los Alamos National Laboratory report LA-11191 UC-10 (February 1988).

7. R. R. Shoun and W. J. McDowell, “Extraction of Trivalent Actinides and Lanthanides by

Nuclear Materials Technology 1994 Report 36 Waste Minimization Technologies Electrochemical Treatment of Mixed Hazardous Waste

Wayne H. Smith, Introduction Christine Zawodzinski, Strong oxidizing agents such as Ag(II) or Co(III) ion can be gener- Kenneth R. Martinez, ated electrochemically in nitric acid (HNO3) at the anode of a di- and Aaron L. Swanson vided cell. These species are capable of oxidizing insoluble actinide Nuclear Materials Processing: oxides to the more soluble higher oxidation state and are being Nitrate Systems investigated as a method to recover actinides from various scrap matrices.1 The electrogenerated oxidants are also capable of oxidizing various organic molecules to carbon dioxide and water and thus can serve as a potential alternative to the treatment of mixed wastes by incineration.2-5

In the treatment process, the substrate of interest is added directly to the anolyte where the catalyst is generated at a fixed rate by constant current electrolysis. The Ag(II) or Co(III) ion that is formed oxidizes the organic substrate and is itself reduced back to Ag(I) or “The most basic conclusion Co(II). These ions migrate back to the anode surface where the oxidizing agents are regenerated and the cycle is repeated. Once the of direct oxidation research organic component of the waste is destroyed, the radioactive com- is that each class of organics ponent may be recovered by standard actinide recovery techniques. reacts differently to a given Much is known about the electrochemical behavior of many organ- set of oxidation conditions; ics in broad areas such as electrosynthesis, reaction mechanisms, indeed, some types resist and kinetics. Extensive research has focused on the direct electrochemical oxidation of organics, especially methanol oxidation be- oxidation altogether.” cause of its importance in fuel-cell technology.6 The most basic conclusion of direct oxidation research is that each class of organics reacts differently to a given set of oxidation conditions; indeed, some types resist oxidation altogether. 7

These unique reactions to oxidation have led us to study the medi- ated electrochemical oxidation (MEO) of several classes of organics, beginning with isopropanol. Isopropanol is a commonly used organic solvent and has some probability of appearing in a mixed waste. It also has only three carbon atoms, which limits the number of possible intermediates that may be formed in the course of oxidation to carbon dioxide. This molecular structure should make it easier to measure the mechanistic pathway and produce some in- sight into the overall oxidation process of alcohols in general. How- ever, the complete conversion of isopropanol to carbon dioxide and water involves the transfer of 18 electrons :

- + (CH3)2CHOH - 18 e + 5 H2O —> 3 CO2 + 18 H . (1)

Thus, considerable complexity still characterizes the oxidation of this relatively “simple” molecule.

37 Waste Minimization Technologies Electrochemical Treatment of Mixed Hazardous Waste

MEO is a two-step process, the However, utilizing these tech- As seen in Figs. 1 and 2, the heterogeneous electrochemical niques usually requires a fairly efficiency for generation of both generation of the catalyst at the clean system in which the elec- Ag(II) and Co(III) decreases with electrode surface and the homo- tron transfer process of interest increasing current density. geneous reaction between the is far removed from other redox The oxidation potential of Ag(I) catalyst and the organic sub- processes in solution. Unfortu- is slightly less positive than the strate. In order to obtain a nately, both the Ag(I) to Ag(II) solvent itself, whereas the oxida- clearer understanding of the and Co(II) to Co(III) oxidations tion potential of Co(II) is slightly overall reaction, we have at- occur at the anodic solvent oxi- more positive than the solvent. tempted to study these reaction dation limit so that neither gives Therefore, the generation of components individually. a well-resolved voltammetric Co(III) is impossible without Specifically, we investigated response. simultaneous oxidation of the the effect of several critical solvent, whereas there is a small reaction parameters, such as To study the efficiency of cata- potential range in which Ag(II) current density, anolyte tem- lyst generation, we then had to may be generated without oxi- perature, acid concentration, use a combined electrochem- dation of the solvent. In either and cell separator type, on the istry-spectroscopy method in case, increasing the current efficiency of catalyst generation which a constant current density density does result in increasing in the absence of an organic is applied to the anode to carry the fraction of the current going substrate. We then repeated out the oxidation reaction, and toward the solvent oxidation those experiments with the the amount of the catalyst gener- process, decreasing the overall isopropanol present, looking at ated is monitored by measuring current efficiency of catalyst the effect of these variables on the amount of light absorbed at generation. the catalyzed destruction of the the wavelength maximum of the organic compound. Finally, we oxidized species. Measurement The effect of temperature on cur- investigated the direct electro- of the catalyst concentration was rent efficiency was determined chemical oxidation of accomplished by using a flow- by performing the electrolysis at isopropanol in the absence through cell in the spectropho- a fixed current density but at of an electron transfer catalyst. tometer and pumping the three separate temperatures: 20, solution from the electro- 35, and 50 °C. As the tempera- chemical cell. ture was increased, the genera- Catalyst Generation tion of both Ag(II) and Co(III) The first step in the catalytic All of the measurements made decreased, although this effect oxidation of organics is electro- on the two catalyst systems were was more pronounced for chemical generation of the cata- carried out in 6 M HNO3. Ag(II). Both catalysts have lim- lyst at the anode of a divided When we attempted to do the ited lifetimes in aqueous solu- electrochemical cell. Typically, electrolyses at lower solution tion because of their ability to the kinetics and mechanism of acidities, the catalysts formed oxidize water with subsequent these oxidation processes would insoluble oxide salts at the elec- reduction back to the original be studied using a small area trode surface and fell to the bot- oxidation state. As the tempera- electrode and modern electro- tom of the reaction cell. ture is increased, these parasitic chemical techniques, such as reactions proceed at a faster rate cyclic voltammetry, rotating so that the steady state concen- ring-disc voltammetry, etc. tration of the catalyst appears to decrease, even though the rate of generation remains essentially constant.

Nuclear Materials Technology 1994 Report 38 Waste Minimization Technologies Electrochemical Treatment of Mixed Hazardous Waste

Carrying out this reaction in a divided electrochemical cell is necessary to prevent the material that is oxidized at the anode from being re-reduced at the cathode. Typically, a ceramic frit, glass frit, or ion exchange membrane is used to separate the two cell compartments. We have found that a ceramic frit separator gives a higher catalyst yield than either a Nafion 117 or Nafion 324 cation exchange membrane.

The most likely explanation for this observation is charge balance within the cell. During electrolysis, positive charge builds up in the anolyte while negative charge accumulates in the catholyte. With a ceramic frit, this charge imbalance is allevi- ated by the movement of posi- Fig. 1. Effect of current density on Ag(II) generation in 6 M HNO3. tively charged ions from the anolyte to catholyte and nega- tively charged ions from catholyte to anolyte. All ions in solution will cross the barrier at rates proportional to their ionic mobility and concentration. Pro- tons that are highly mobile and in high concentration will carry the greatest fraction of the current in moving to the catholyte, while the remainder will be carried by the movement of nitrate ions to the anolyte. The movement of catalyst ions across the membrane will be minimal.

Fig. 2. Effect of current density on Co(III) generation in 6 M HNO3. 39 Waste Minimization Technologies Electrochemical Treatment of Mixed Hazardous Waste

With a cation exchange mem- Samples for GC analysis were buildup of acetone, the two- brane, alleviation of charge im- taken every half-hour from this electron oxidation product, fol- balance is restricted to the solution. The fact that there was lowed by a decrease in acetone movement of cations. Therefore, no change in isopropanol con- concentration when acetic acid, all of the current will be carried centration and no evidence of the four-electron oxidation by cation migration from other substances forming during product, began to appear. anolyte to catholyte. Whereas the time period suggested that Evidence also existed for the protons will continue to be the isopropanol is not attacked by formation of formic acid and primary charge carrier, an in- HNO3. formaldehyde in the reaction creasing fraction of the current mixture. However, the reaction will be carried by catalyst ion Under an applied constant cur- appeared to stop with the for- migration. This process is rent with no catalyst present, mation of acetic acid, which readily observed when cobalt isopropanol was oxidized very occurred when approximately is used as the catalyst. With a frit quickly and efficiently to acetic 10 equivalents of electrons per in place as a separator, no cobalt acid (Fig. 3). The GC analysis mole of isopropanol had been enters the catholyte. When a cat- clearly showed an initial removed from the solution. ion exchange membrane is used, the catholyte quickly becomes pink, the characteristic color of Co(II) in HNO3 solution. To summarize, the efficiencies of both Ag(II) and Co(III) generation are improved by use of low current density, low temperature, a ceramic frit rather than a cation exchange membrane separator, and in highly acidic media. Isopropanol Oxidation Electrochemical oxidation of isopropanol in a well-stirred solution was carried out at fixed current density at a platinum anode. Products and intermediates of the reaction were monitored by extracting a small sample of the anolyte during the run and injecting this sample directly into a gas chromatograph (GC).

Because HNO3 is itself an oxi- dizer, we first determined if solvent oxidation of the organic substrate was contributing to the 2 electrochemical oxidation pro- Fig. 3. Electrooxidation of isopropanol at 500 mA/cm . cess. A small amount of isopropanol was mixed with 6 M HNO3, and the solution was stirred for several hours.

Nuclear Materials Technology 1994 Report 40 Waste Minimization Technologies Electrochemical Treatment of Mixed Hazardous Waste

This number is consistent with the partial oxidation of isopropanol to one equivalent of acetic acid and one of carbon dioxide, as per the following equation:

- + (CH3)2CHOH - 10 e + 3 H2O —> CH3COOH + CO2 + 10 H . (2)

To further verify this result, we conducted a separate experiment starting with acetic acid. After passing 10 equivalents of electrons per mole of acetic acid, we detected no change from the original concentration.

These results are explained in the reaction mechanism that follows:

The ease of oxidation of organic compounds by class appears to be the reaction. If there were an alcohols > ketones, aldehydes > carboxylic acids. Isopropanol is appreciable concentration of very rapidly oxidized to acetone and more slowly to acetic acid. methanol, the acid-catalyzed But a product of the second oxidation step is an equivalent of esterification of acetic acid methanol. Because methanol is even more easily oxidized than would be expected. The prod- isopropanol, it is immediately converted to formaldehyde, formic uct, methyl acetate, does not acid, and ultimately carbon dioxide. Thus, only a small quantity appear in the chromatogram. of these substances is present in the reaction cell at any time during

41 Waste Minimization Technologies

Electrochemical Treatment of Mixed Hazardous Waste

Conversion of isopropanol to acetone and/or acetic acid occurs more efficiently at higher current density (Figs. 4 and 5). This higher efficiency may be due to (1) the higher acquired potential of the working electrode, which could more effectively oxidize the acetone intermediate, or (2) the generation of hydroxyl radicals by direct water oxidation, which may contribute to the overall oxidation mechanism.

Temperature has a significant influence on the oxidation rate. At room temperature some fraction of the acetone still remains in the reaction mixture even after the passage of 18 equivalents of electrons. At 50 °C the acetone is completely oxidized to acetic acid well before 18 electrons have been added. More elevated Fig. 4. Effect of current density on oxidation of isopropanol to acetone. temperatures may be necessary to achieve further oxidation of the acetic acid to carbon dioxide. These experiments are currently underway.

Finally, the oxidations were repeated using optimal conditions (high current density, high temperature, and high acidity) in the presence of a catalyst. We found that the addition of Co(III) improved the oxidation efficiency, whereas the addition of Ag(II) led to a decrease. The most likely explanation for this phenomena can be attributed to the relative redox potentials of the systems: Ag(II) < solvent < Co(III).

Fig. 5. Effect of current density on oxidation of isopropanol to acetic acid.

Nuclear Materials Technology 1994 Report 42 Waste Minimization Technolgies Electrochemical Treatment of Mixed Hazardous Waste

The results of our investigations 4. J. Farmer et al., “Electrochemical suggest that the optimum condi- Treatment of Mixed and Hazardous tions for electrochemical oxida- Wastes: Oxidation of Ethylene tion of isopropanol are high Glycol and Benzene by Silver(II),” current density, elevated tem- J. Electrochem. Soc. 139 (3), 654-661 perature, high solution acidity, (1992). and Co(III) as an electron transfer catalyst. However, even un- 5. J. C. Farmer, F. T. Wang, der optimum conditions, P. R. Lewis, and L. J. Summers, oxidation to carbon dioxide is “Electrochemical Treatment of incomplete. We are still investi- Mixed and Hazardous Waste: gating those conditions neces- Oxidation of Ethylene Glycol by sary to achieve complete Cobalt(III) and Iron(II),” in Electro- conversion.✦ chemical Engineering and the Environ- ment ‘92 (Institution of Chemical Cited References Engineers, Warwickshire, United Kingdom, 1992) ICHEME Sympo- 1. E. J. Wheelwright et al., sium Series No. 127, 203-214. “Kilogram-scale Demonstration of Plutonium Recovery from Plant 6. B. Beden, J. M. Léger, and Residues Using CEPOD II Dis- C. Lamy, “Electrocatalytic Oxida- solver, Anion Exchange, Oxalate tion of Oxygenated Aliphatic Precipitation, and Calcination to Organic Compounds at Noble Oxide,” Pacific Northwest Labora- Metal Electrodes,” in Modern tory report PNL-7653 UC-721 Aspects of Electrochemistry, J. O’M (April 1991). Bockris and L. Eberson, Eds. (Plenum Press, New York, 1992) 2. D. F. Steele, “Electrochemical Vol. 22, pp. 97-101, 234-241. Destruction of Toxic Organic Industrial Waste,” Platinum Metals 7. Organic Electrochemistry, M. M. Review 34 (1), 10-14 (1990). Baiser, Ed. (Marcel Dekker, Inc., New York, 1973) pp. 470-494. 3. P. Cox and D. Pletcher, “Electrosynthesis at Oxide Coated Electrodes Part 2. The Oxidation of Alcohols and Amines at Spinel Anodes in Aqueous Base,” J. Appl. Electrochem. 21, 11-13 (1991).

43 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids Nicholas V. Coppa Introduction and Background Actinide Materials Chemistry Freeze drying is the removal of solvent or volatile components by sublimation from a frozen solution, suspension, or multicomponent material. The freeze drying process, or lyophilization when it refers to biological materials, includes two independent steps: the freezing of the material and the removal by sublimation (drying) of the volatile component. The removal of a solvent from a material under these conditions often results in the preservation of the solid portion of the material. This preservation aspect of freeze drying technology (FDT) has led to diverse applications that have resulted in the processing of tons of materials daily throughout the world. The objective of this work is to demonstrate the applicability of FDT to the decontamination of complex radioactive materials including liquids, slurries, and sludges containing a wide variety of constituents in which the decontamination levels are in excess of 100 million. When “Our focus in applying FDT is applied to the decontamination of multicomponent radioactive materials, the product is the condensate (as opposed to the solid this technology is the portion of the material). This is the first application of FDT where elimination of radioactive primary importance is given to the condensate, and this new twist discharges from the has drawn the attention of several commercial manufacturers and developers to this work. Our focus in applying this technology is Los Alamos Plutonium the elimination of radioactive discharges from the Los Alamos Facility; however, successful Plutonium Facility; however, successful demonstration here will demonstration here will most likely lead to nuclear industry-wide applications. most likely lead to Separation of an aqueous nonradioactive solvent and a radioactive nuclear industry-wide solid (solute) during the freeze drying process occurs in the following manner (please refer to Fig. 1 throughout the following discus- applications.” sion). The solution is frozen and situated in the proximity of a condenser surface. The radioactive material in the left chamber of the freeze drying apparatus is held at a higher temperature than the condenser in the right chamber (T1 > T2). A concentration gradient develops between the two chambers due to the difference between the temperature-dependent vapor pressures (p1 and p2) of the solvent at the two locations. This concentration gradient drives the transport of the nonradioactive solvent from the sample chamber to the condenser. Heat is supplied to the frozen material to allow sublimation to occur. Because the vapor pressures can be quite low, a vacuum is applied to assist the diffusion of the sublimed solvent. Solvent removal is fully achieved by supplying enough heat to the frozen material to complete the sublimation process while removing enough heat at the condenser to maintain the concentration gradient.1

Nuclear Materials Technology 1994 Report 44 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids

Fig. 1. Schematic of a freeze drying apparatus. The sample chamber contains the frozen radioactive material (speckled) and the condenser chamber houses the cold condenser surface onto which the purified ice (gray) condenses. The temperature of the frozen material (T1) is maintained higher than that of the condenser (T2). The vapor pressure of ice at the two locations (p1 and p2) is dependent on these temperatures, and the resultant concentration gradient causes mass transport. Supplying the heat of sublimation at the sample and removing it at the condenser main- tains the transport conditions.

The drying rate can be deter- frozen solution temperature, supplying radiant energy mined through the simulta- surface area and concentration, directly to the surface of the neous solution of the Fourier and the geometry of the appara- frozen material. heat law, the Knudsen equation, tus. Nuclear waste streams have and the appropriate conduc- a relatively low solute concen- Freeze dryers are extremely tance relationship for the geom- tration in comparison to, for versatile and broad in their etry of the apparatus.2,3 (See example, foods. Consequently, applications. Laboratory freeze Reference 2 for an in-depth the greatest impediment to the dryers are usually designed to discussion of transport phe- freeze drying rate, for nuclear sublime kilograms of solvent nomena, including the Fourier, waste solutions, is the thermal per day, whereas industrial Fick, Dufour, and Sorret rela- conductivity of ice. This funda- freeze dryers are capable of tionships.) Such analysis indi- mental limitation is circum- drying thousands of kilograms cates that this rate is strongly vented through engineering per day.4 The factor that most dependent on the temperature solutions, such as maximizing greatly influences the cost of a difference between the frozen the surface area, minimizing freeze dryer is the performance solution and the condenser, the the thickness of the sample, and required by the particular application.

45 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids

Application of FDT to Nuclear FDT is versatile in that its appli- Other decontamination methods Materials cation is not limited to aqueous such as ion exchange, precipita- FDT has been applied to the de- solutions. Organic solvents can tion, or chelation have high ion contamination of waste streams be separated from other liquids specificity but require multiple containing nuclear and hazard- or solids. Liquids processed in stages, regeneration, neutraliza- ous materials with great success. this way may be reused or dis- tion, and/or filtration. They Acidic solutions containing ura- carded as nonradioactive. involve the addition of other nium, iron chromium, and components (resins, chelating nickel originating from nuclear FDT drastically reduces the agents, precipitating agents, etc.) power plant operations have volume of the waste since the that also require remediation. been decontaminated such that volume of solute or solid compo- Integration of FDT into these the concentration of radioactive nent is smaller than the solution processing lines may greatly in- materials was 1 pCi per liter or wet volume. Volume reduc- crease process efficiency and (four times lower than current tion factors greater than one decontamination effectiveness. Environmental Protection thousand have been achieved in Agency [EPA] emission levels) aqueous nitrate/nitric acid solu- Distillation and/or evaporation 14 while the decontamination fac- tions (pH < 0.5), but the exact decontamination processes in- tor was 60 million.5 (The decon- volume reduction of nuclear volve continuous, often turbu- tamination factor is equal to the waste depends on the particular lent, flow; thus, entrainment ratio of the initial and final con- moisture content. Volumes may of the radioactive component centrations.) Waste solutions also be reduced through the is an unavoidable consequence. containing sodium and potas- elimination of neutralization Decontamination factors of ten sium nitrates, cesium-137, stron- steps in transuranic (TRU) thousand are theoretically tium-90, yttrium-91, cerium-144, processing. Processing of pluto- achievable for distillation, but ruthenium-106, and iodine-131 nium- and uranium-contami- values of about one to three were reduced to dry residues nated scrap often involves the hundred are typical in practice. while the condensate was de- dissolution of the scrap using The freeze drying process differs contaminated by 10 million nitric acid. These solutions are from distillation in that the times.6 Other radioactive ele- usually neutralized using so- transport process during freeze ments, such as iodine,7 pluto- dium hydroxide for compatibil- drying involves molecular flow nium,8 and ruthenium,9 have ity with subsequent process and the thermal energy of the been concentrated in the freeze- steps. A significant percentage radioactive material is much dried solid residues to produce of the wastes stored at Hanford lower. These features are re- radiochemically pure water. was a result of chemical neutral- sponsible for the virtual elimina- Organic solvents have been ization. The application of FDT tion of solute entrainment; separated from radioactive sol- may eliminate the use of sodium consequently, decontamination ids.10 Tritium oxide has been hydroxide neutralizing agents factors obtained using FDT are separated from solids, but, in and consequently greatly reduce extremely high. that case, the condensate con- the quantity of contaminated sol- tained the radioactive fraction. ids at later process steps. During FDT can be considered safe and Zinc, cadmium, molybdenum, the freeze drying of salt/nitric compact. No high temperatures selenium, ruthenium, iron, chro- acid solutions, the nitric acid and or pressures are used. Further- mium, and nickel also have been water vapors are transported to more, because the freeze drying separated from radioactive sol- the condenser leaving behind a process occurs in a vacuum, the 14 ids.11,12 Pilot plant-scale opera- neutral salt. Furthermore, the failure of a component would tions for the concentration acidic condensate may be re- lead to an inward leak, which of radioactive materials cycled. This aspect of FDT is reduces the potential for con- (throughput of 25 L per hour) being integrated with the current tamination outside the system. 15 with decontamination factors acid-recycle effort in the of 100 thousand have also been Nuclear Materials Technology reported.13 (NMT) division.

Nuclear Materials Technology 1994 Report 46 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids

A system occupying an area of need, Los Alamos has entered In order to understand how FDT about 225 sq ft would be capable into a CRADA (cooperative re- might be integrated into the Plu- of processing 375,000 kg per search and development agree- tonium Facility’s waste stream year (based on 1500 L per day ment) with Edwards Freeze processing line, the plutonium and 250 days per year) while Drying, a commercial manufac- recovery operation from scrap maintaining a decontamination turer of plant-scale freeze dryers is used as an illustration.16 factor greater than 100 million. (1500 liters capacity per unit per (See References 17–19 for more More conventional approaches day) based in upstate New York. in-depth descriptions of the would require considerably Edwards Freeze Drying will waste stream processing line at larger space investments to work with the NMT staff in the the Plutonium Facility.) Pluto- achieve equivalent performance design, fabrication, and installa- nium scrap is defined as any characteristics. tion of the glove-box system to material that has a plutonium ensure that the operating param- component; usually other metals Plutonium Facility eters can be scaled to a plant- are present. The plutonium Demonstration sized system. scrap is dissolved in either an The FDT demonstration in the acid or base depending on the Plutonium Facility consists of In this partnership, both parties chemical composition of the two phases: have the opportunity for sub- scrap. This solution enters a pro- 1) studies using a freeze dry- stantial benefit and gain. cessing line where other metals ing apparatus in a glove Edwards Freeze Drying consid- are separated by methods such box to demonstrate perfor- ers NMT Division’s knowledge as ion exchange or precipitation. mance and to obtain design and experience with nuclear Three effluents are discharged criteria for a plant-scale materials and the Plutonium from this processing line. Each system, and Facility to be vital for the devel- varies in the concentration of the 2) integration of FDT into the opment of this new FDT appli- radioactive and nonradioactive existing Plutonium Facility cation. They see this partnership elements. The plutonium con- liquid waste processing as the foundation for the devel- centrations for the three lines line. opment of a new product line are typically 0.05, 0.08, and 0.15 g and possibly new worldwide plutonium per liter. Usually the In the first phase, data will be markets. The successful demon- radioactive component is a small acquired using mock plutonium stration of FDT at the Plutonium fraction of the total volume, but solutions for the purpose of de- Facility is crucial to meeting the even these small amounts are not termining extreme performance Laboratory’s immediate and allowed to be discharged into the characteristics. With these ex- long-range discharge goals. environment. trema known, tests will be con- Furthermore, particular nuclear ducted using aliquots of real waste problems facing other In an effort to further concentrate solutions obtained from various Department of Energy (DOE) the radioactive component and steps along the Plutonium Facil- facilities such as Rocky Flats, to pass an effluent that has a ity processing line. This work Savannah River, and Hanford lower radioactive concentration, will enable process engineers at are also being considered in the the three effluents exiting from the Facility to determine the best demonstration of this technol- the ion exchange and precipita- place(s) to employ FDT and at ogy, and solutions to those prob- tion process are sent to an evapo- what scale. lems will be integrated into the rator. The evaporator effluent Edwards Freeze Drying product has a concentration of about The integration of FDT into the development. The DOE will 7 x 10-7 g plutonium per liter Plutonium Facility waste pro- gain indispensable technology and is sent to the Liquid Waste cessing line, the second phase, for the environmental restora- Treatment Facility (TA-50) at Los will require engineering and tion of its nuclear facilities, and Alamos for further processing. manufacturing capabilities the nuclear industry in general beyond those available at will have at its disposal the tech- Los Alamos. To meet this engi- nology to meet discharge limits neering and manufacturing that are becoming increasingly more stringent.

47 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids

The residues after evaporation (called bottoms) are stabilized in cement. The cements are sent off-site for disposal and/or storage. Figure 2A illustrates the evaporator feeds, evaporator products, and storage and disposal paths.

Two of the possible application points for FDT in the processing line are (1) as a replacement for the evaporator or (2) in series after it. Either of these applications may also feature a means for stabilizing the solid residues that uses atomically mixed ceramic or cement precursors (see next section). Figures 2B and 2C illustrate these FDT application configurations. An analysis and comparison summary of evaporator performance and the performance of configurations using FDT is given in Table I and discussed in the following paragraph. Of the three evaporator feeds, the solutions with a concentration of 0.15 g plutonium per liter represent only a small fraction of the total waste processed; the largest fractions contain only 0.05 – 0.08 g plutonium per liter. Because the relative fractions are variable over time, the worst case scenario was chosen for the performance analysis; that is, we assumed that 100% of the feed has a concentration of 0.15 g plutonium per liter.

Fig. 2. Configurations for plutonium effluent processing line: (a) current processing line for plutonium scrap at the Plutonium Facility and performance characteristics, (b) proposed simple substitution of a freeze dryer and its performance characteristics, (c) proposed evaporator and freeze dryer in series and their performance characteristics.

Nuclear Materials Technology 1994 Report 48 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids

Table I. Comparison of Present Evaporator Performance to FDT and to Their Performance in Series.a

Present Evaporator FDT Performanceb Evaporator and FDTb Performance (alone) (in series) c (g Pu l-1) (dpm l-1) (g Pu l-1) (dpm l-1) (g Pu l-1) (dpm l-1) d d d Input 0.15 2.0 x 1010 0.15 2.0 x 1010 0.15 2.0 x 1010 d - Output 7.0 x 10-4 9.3 x 107 1.5 x 10-9 200 7.0 x 10 12 1.0

Decontamination 210 210 108 108 2 x 1010 2 x 1010 Factor

a bEvaporator is placed upstream relative to a freeze dryer. In light of new proprietary developments, the actual decontamination factors may be substantially greater (1000 times) c than those cited here. d Disintegrations per minute per liter, assuming a plutonium-239 half-life of 24,100 years. Maximum input values and minimum output values were chosen from Ref.16; in practice, actual values will probably be lower by a factor of 3 or greater.

Using FDT as a replacement for have to be sent to TA-50, and which would result in the simpli- the evaporator at the Plutonium this could have a significant fication or elimination of some Facility and assuming a decon- positive impact on that waste post-evaporator operations. tamination factor of 100 million, facility’s operations as well. In the liquid effluents would calendar year 1991, the pluto- Stabilization of Radioactive contain 1.5 x 10-9 g plutonium nium-239 effluents from TA-50 Solids Using Freeze-Dried per liter or 93 pCi per liter. averaged 43 pCi per liter.20 Atomically Mixed Precursors A decontamination factor of 100 Current decontamination factors A freeze drying process for the million is not at all optimistic at TA-50 are on the order of preparation of ceramic precur- because other developments about 1000 (based on the aver- sors that are atomically mixed made in conjunction with our age 1991 monthly influents and has been developed and proven industrial partner have indi- effluents for all radioactive to be very useful for overcoming cated that decontamination materials). Clearly, if FDT kinetic and thermodynamic factors far in excess of one enjoyed a successful demonstra- barriers encountered during billion should be achievable. tion at the Plutonium Facility, conventional ceramic materials (The details of these develop- this success would imply that processing. By integrating the ments were proprietary at the decontamination factors at separation aspects of FDT with time of the writing of this paper TA-50, as well as at other DOE the preparation of atomically and could not be disclosed.) facilities, would be lower (by mixed precursors (AMPs), the If a freeze dryer is placed in factors in excess of 1000) than radioactive solid residues from series and downstream from current EPA discharge limits. the separation process can be the evaporator at the Plutonium In conjunction with the prepara- transformed into materials with Facility, the effluent would tion of atomically mixed ceramic exceptional environmental stabil- contain 7 x 10-12 g plutonium precursors (see the next section), ity. These materials can be per liter, which is 0.4 pCi per FDT may prove dually advanta- synthesized faster, at lower liter or 10 times lower than the geous: first, the liquid wastes temperatures and pressures, and current EPA discharge limits. would be greatly decontami- in a fashion compatible with the In that case, discharges may not nated; second, the solid residues glove-box environment on an may be more easily stabilized, industrial scale.

49 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids

Atomic mixtures are a mixture of two or more components lacking both long- and short- range structural and composi- tional order. Atomic mixtures are prepared by rapidly freezing (dT/dt = one million °C per second) a solution containing the appropriate salts followed by freeze drying.14 (See Fig. 3 for a diagram of this freeze drying process.) The high freezing rates trap solute molecules in a disor- dered state in a matrix of ice. Then, by freeze drying, the ice Fig. 3. Schematic for the preparation of atomically mixed precursors of matrix is removed, and the dis- either ceramics or cement, which contain the solid, freeze-dried residues. ordered state of the original so- Using this approach, solid residues would be rendered more stable as well lution is preserved in the as liquid wastes decontaminated. residual solid. These atomic mixtures are extremely reactive, and the kinetics for product formation are unlike those of con- may not be considered suitable compared to that of either boro- ventional ceramic processing. for long-term storage because of silicate glasses or supercalcine A microscopic picture for the be- new performance requirements materials,24 but the high tem- havior of these mixtures is being imposed by regulating agencies peratures and pressures re- developed in conjunction with (EPA) and are not considered to quired for their synthesis have the staff from several other be stable on time scales of geo- made these materials unattrac- Laboratory divisions* using logical porportions. Other mate- tive as a waste form. Because time- and temperature-resolved rials such as glasses and mineral of the unique characteristics of neutron and x-ray scattering and analogues, for example synroc, AMPs, we are applying them to Monte-Carlo simulations of the are, or have been, candidates for the stabilization of uranium- molecular dynamics. The cal- long-term, permanent storage of and plutonium-containing solids cine temperature of AMP mate- nuclear materials, but these can- resulting from the FDT decon- rials has been shown to be lower didates also have problems. tamination process. There are (by greater than 100 °C)21 than Although the manufacturing many crystalline phases into those of conventional ceramic process for glasses is compatible which uranium, plutonium, and processes, while the reaction with the radiological environ- other radioactive elements can times are also drastically ments and procedures, serious be incorporated.25 High tem- reduced. disadvantages exist; for ex- peratures and/or pressures may ample, borosilicate glass is be eliminated. Cements with a As indicated in Fig. 2A, cements metastable with respect to a dispersion of the radioactive are currently mixed with the crystalline-phase assembly of component that is homogeneous radioactive evaporator bottoms the same composition22 and, on nanometer-length scales, for the purpose of stabilization. thus, readily devitrifies when without the use of mechanical While this composite allows for subjected to elevated tempera- mixers, may also be prepared. safe temporary storage and tures and pressures.23 Crystal- Formulations based on existing transportation, these cements line materials such as synroc mineral or cement compositions have been shown to have far su- using chemical surrogates for perior environmental stability actinide materials, such as those * Participants include Los Alamos found at the Plutonium Facility, Neutron Scattering Center are currently being developed (LANSCE) and the Materials and the products characterized. Division. Nuclear Materials Technology 1994 Report 50 Waste Minimization Technologies Application of Freeze Drying Technology to Decontamination of Radioactive Liquids

The synthesis and characteriza- 8. K. Ohtsuka, I. Kondoh, and 17. D. C. Christensen et al., tion of actinide-containing mate- T. Suzuki, “Spent Fuel Treatment “Waste from Plutonium Conversion rials are planned for the near Method,” European Patent Appli- and Scrap Recovery Operations,” future. Research using AMPs cation No. 358,431 (1990). Los Alamos National Laboratory that contain nuclear materials report, LA-11069-MS (March 1988). is focused on the development 9. S. Tachimori, J. Nucl. Sci. Technol. of several high-performance 13, 442 (1976). 18. S. L. Yarbro, “Using Solvent ceramics and cements, but this Extraction to Process Nitrate Anion research will also advance the 10. G. E. Harwood et al. ,”Process- Exchange Column Effluents,” Los frontier of multication, solid- ing of Liquid Scintillation Waste Alamos National Laboratory report, state actinide materials.¨ and Animals,” in Synthesis and LA-11007-T (October 1987). Application of Isotopically Labeled Compounds, Proceedings of Interna- 19. E. L. Christensen and W. J. Cited References tional Symposium, W. P. Duncan Maraman, “Plutonium Processing 1. J. D. Mellor, Fundamentals of and Alexander B. Susan, Eds. at the Los Alamos Scientific Labora- Freeze Drying (Academic Press, (Elsevier, Amsterdam, 1983), tory,” Los Alamos Scientific Labora- London, 1978). pp. 117-120. tory report, LA-3542 (April 1969).

2. R. B. Bird, W. E. Stewart, and 11. Y. Wang, J. Qin, Q. Ji, S. Wu, 20. R. T. Harris, “Summary of E. N. Lightfoot, Transport Phenomena and X. Wang, Fenxi Huaxue 13, 210 Operating Data,” in Annual Report, (John Wiley & Sons, New York, (1985). EM-7 Liquid Waste Management 1960), pp. 563 - 572. Group, 1991 (approved October 12. K. Ohtsuka et al., “Freeze 1992), Los Alamos National Labora- 3. T. R. Sheng and R. E. Peck, Drying in Reprocessing Nuclear tory document. “Rates for Freeze Drying,” in Water Reactor Fuels,” Jpn. Kokai Tokkyo Removal Processes: Drying and KoHo, Japan Patent No. 1,316,695 21. N. V. Coppa, et al., J. Mater. Res. Concentration of Foods and Other (1989). 7, 2017 (1992). Materials, C.J. King and S. P. Clark, Eds., AICHE Symposium Series 73 13. K. Jessnitzer et al., Nucl. Sci. 22. G. J. McCarthy, Nuclear Technol- (163), pp. 124-130 (1977). Abstr. 23(2), 2542 (1969). ogy 32, 92 (1977).

4. E. W. Flosdorf, Freeze Drying 14. N. V. Coppa,”Synthesis and 23. G. J. McCarthy et al., Nature (Reinhold, New York, 1949), Characterization of High Transition 273, 216 (1978). pp. 1-64. Temperature Superconductors,” Ph.D. thesis, Temple University 24. A. E. Ringwood et al., Nature 5. O. Takanobu and I. Kondoh, (1990), pp. 71-134. 278, 219 (1979). “Method and Apparatus for Treatment of Liquid Radioactive 15. T. R. Mills, “Acid Recovery and 25. G. J. McCarthy, “Ceramics and Waste,” Ger. Offen., European Patent Recycle,” this report. Glass Ceramics as High Level No. 3,625,602 (1987). Waste Forms,” in Ceramic & Glass 16. S. L. Yarbro, Los Alamos Radioactive Waste Forms, D. W. 6. N. V. Mikshavich et al., At. Energ. National Laboratory, private Ready and C. R. Cooley, Eds. (Dept. 27(6), 541 (1969). communication, September 1993. of Energy and Research Develop- ment, Washington, D.C., 1977), 7. K. Ohtsuka, J. Ohuchi, and T. Conf. 770102, pp. 83-99. Suzuki, “Method of Recovering Radioactive Iodine in Spent Nuclear Fuel Retreatment Process,” Euro- pean Patent Application No. 361,773 (1990).

51 Waste Minimization Technologies Nuclear Applications for Magnetic Separations

Larry R. Avens, Magnetic Separation Process Laura A. Worl, Magnetic separation is a physical separation process that segregates Ann R. Schake, materials on the basis of magnetic susceptibility.1 Because the pro- Dennis D. Padilla, cess relies on physical properties, separations can be achieved while and Karen J. deAguero producing a minimum of secondary waste. Nuclear Materials Processing: When a paramag- Chloride Systems netic particle en- Table I. Volume Magnetic Susceptibility of counters a non- Kevin B. Ramsey, Selected Compounds and Elements uniform magnetic Richard D. Maestas, field, the particle is Robert C. Martinez, * urged in the direc- Compound/Element Susceptibility x 106 and Yvette E. Valdez tion in which the Nuclear Materials Processing: field gradient Nitrate Systems FeO ...... 7178.0 increases. Diamag- Fe O ...... 1479.0 2 3 netic particles react Thomas L. Tolt UO ...... 1204.0 in the opposite Lockheed Environmental Systems and 2 sense. When the Cr O ...... 844.0 Technologies Company 2 3 field gradient is NiO ...... 740.0 of sufficiently high F. Coyne Prenger, Am ...... 707.0 intensity, paramag- Walter F. Stewart, netic particles can and Dallas D. Hill Pu ...... 636.0 be physically cap- Advanced Engineering Technology U...... 411.0 tured and separated Engineering Sciences and PuO ...... 384.0 from extraneous Application Division 2 nonmagnetic CuO ...... 242.0 material. RuO ...... 107.0 2 Th ...... 41.0 Because all actinide compounds are UO ...... 41.0 3 paramagnetic CrO ...... 14.0 3 (Table I), magnetic Si ...... -0.3 separation of actinide-containing CaO ...... -1.0 mixtures is feasible. ZrO ...... -7.8 “Magnetic separation is a 2 Magnetic separation physical separation process MgO ...... -11.0 on recycle plutonium chemical pro- that segregates materials on KCl ...... -13.0 cess residues has CaCl ...... -13.0 the basis of magnetic 2 been demonstrated NaCl ...... -14.0 on an open-gradient susceptibility.” magnetic separator.2 CaF ...... -14.0 2 The advent of reli- SiO ...... -14.0 2 able superconduct- MgF ...... -14.0 ing magnets also 2 makes magnetic Graphite ...... -14.0 separation of weak- Al O ...... -18.0 2 3 ly paramagnetic species attractive. *SI Units Nuclear Materials Technology 1994 Report 52 Nuclear Materials Technology 1993 Report Waste Minimization Technologies Nuclear Applications for Magnetic Separation

Magnetic Separation Methods In practice, a dry powder is de- The HGMS method is used to Although numerous magnetic livered onto a thin belt that separate magnetic components separation methods exist, we moves over rollers. The front from solids, liquids, or gases. A have currently selected two roller is fabricated from a neo- diagram of the method is shown methods for development: mag- dymium-iron-boron rare earth in Fig. 2. Most commonly, the netic roll (or drum-type) separa- magnet. Ferromagnetic and suf- solid is slurried with water and tion and high-gradient magnetic ficiently paramagnetic particles passed through a magnetized separation (HGMS). The useful- are attracted to the magnet and volume. Field gradients are pro- ness of the open-gradient sepa- adhere to the belt in the region duced in the magnetized vol- ration method used in the early of the magnet. As the belt moves ume by a ferromagnetic matrix work is limited by its low pro- away from the magnet, the fer- material, such as steel wool, iron cessing rate.2 romagnetic and paramagnetic shot, or nickel foam. Ferromag- particles disengage from the belt netic and paramagnetic particles The magnetic roll separation and are collected in a catch pan. are extracted from the slurry by method is used to separate Diamagnetic and nonmagnetic the ferromagnetic matrix, while magnetic particles from dry particles pass over the magnetic the diamagnetic fraction passes powders. A diagram of a roll roller relatively unaffected and through the magnetized volume. separator is shown in Fig. 1. are collected in a different catch The magnetic fraction is flushed pan. This method effectively ex- from the matrix later when the tracts particle sizes ranging from magnetic field is reduced to zero 90 to 850 µm. or the matrix is removed from the magnetized volume. The HGMS method effectively ex- tracts particle sizes ranging from ~0.3 to 90 µm and, thus, is complementary to the roll separator.

Fig. 1. Diagram of a roll magnetic separator.

53 Waste Minimization Technologies Nuclear Applications for Magnetic Separation

A second 3-inch-bore superconducting magnet, installed at our waste treatment facility, will be used to study waste stream treatment applications and will eventually be interfaced to a glove box in the Plutonium Facility for tests of chemical processing residue concentration. Finally, a 6-inch-bore magnet, installed in an open area, is being used for nonradioactive tests and for prototype development of soil decontamination applications. High-Gradient Magnetic Separation (HGMS) Model More than 10 variables indepen- dently affect the HGMS process; thus, an analytical model for the process is needed to select appropriate bench-scale experiments and to effectively analyze the resulting data. In addition, a validated analytical model sup- Fig. 2. Simplified HGMS diagram. ports prototype design and process scale-up. Much work6-12 by others has been done to model the behavior of the paramagnetic particles as they interact with the magnetized matrix. The development of the HGMS liquid waste stream treatment, Areas investigated include the method has been the central underground storage-tank waste dynamic effects of particle tra- focus of Los Alamos National treatment, chemical processing jectories in homogeneous mate- Laboratory’s magnetic separa- residue concentration, and soil rials, the influence of particle tion research.3-5 Today’s super- decontamination. buildup on the matrix elements, conducting magnet technology and the particle buildup effect produces much higher electro- Our current experimental on the flow field. These studies, magnetic fields than is possible HGMS equipment includes one though requiring simplifying with conventional electro- conventional-coil separator and assumptions, identify appropri- magnets, which are limited by three superconducting high-gra- ate independent variables the saturation of iron to a mag- dient separators. We are using that form the basis for this netic field strength of about two the 1-inch-bore conventional coil investigation. tesla. This availability of high and one 3-inch-bore supercon- fields permits a broad range of ducting magnet, installed on a HGMS applications, including vent hood in the Plutonium Fa- cility, for low-activity samples such as contaminated soils and uranium-containing surrogates.

Nuclear Materials Technology 1994 Report 54 Waste Minimization Technologies Nuclear Applications for Magnetic Separation

was performed to evaluate separator performance as a function of the independent variables in MatrixMatrix the separation process. The elementElement results were used to determine Paramagnetic the separation coefficient in the Paramagnetic Magnetic Particle proposed rate model and were particle correlated using physically meaningful, dimensionless Flow Viscous groups. StreamFlow DragViscous drag stream We assume that there exists for each matrix element a capture Gravity height, wherein each magnetic particle entering the window defined by the capture height and Fig. 3. Force balance of magnetic, viscous drag, and gravity forces on a the matrix element diameter will paramagnetic particle in the vicinity of a matrix element. be captured by the matrix element. We then define a capture cross section for a matrix element to be the ratio of the capture height to the matrix element The matrix materials currently We propose a rate model for this diameter, as shown in Fig. 4. used for HGMS are inhomogen- process that depends on a sepa- The separation coefficient is re- eous and have a complex cross ration coefficient, which we de- lated to the capture cross sec- section. In addition, the para- fine in terms of a capture cross tion, and if the capture cross magnetic particles are section and a potential function section is known, the separation nonspherical and include a defined by the force balance on efficiency can be calculated. range of particle sizes. All of the particle. A series of tests these factors preclude precise analytical treatment. If the particles are physically liberated from the host material and are Matrix Elementelement not electrically charged, the CaptureCapture Height height principal forces governing their behavior are magnetic, viscous, Flow and gravitational. Dm Dm The performance of the magnetic separator is modeled using a static force balance on an individual paramagnetic particle in the immediate vicinity of a matrix element, as shown in Fig. 3. The model assumes that if the Capture height = magnetic capture forces are Element diameter greater than the competing viscous drag and gravity forces, the particle is captured and removed from the flow stream. Fig. 4. The matrix capture cross section defined as the ratio of capture height to matrix element diamter Dm.

55 Waste Minimization Technologies Nuclear Applications for Magnetic Separation

Table II. Independent Variables Investigated

Material Characteristics Separator Parameters

Particle Size: 0.5- 50 mm Matrix Element Size: 5- 100 mm Impurity Concentration: 0.4- 400 ppm Matrix Element Spacing: 80- 1200 mm Solids Concentration: 5- 30 wt % Magnetic Field Strength: 0.5- 7.5 T Volume Magnetic Matrix Material: 43D Stainless Steel & Ni 6 Susceptibility* (x10 ): 129- 1478 Surfactant Concentration: 0.05- 0.2 wt % Residence Time: 1.0- 8.0 s Slurry pH: 7- 10 Superficial Velocity: 0.25- 4.0 cm/s

*SI units

HGMS Data Correlation The capture cross section is sig- Intuitively, one expects to have A comprehensive series of ex- nificantly less than 1.0, which to maximize both the separator periments was performed to de- implies that particles are swept length and the matrix residence termine the capture cross section around the matrix elements by time for optimum performance; as a function of the independent the fluid flow even if the par- however, only their ratio has to variables. Table II lists the vari- ticles enter the region defined by be considered, and this fact ables investigated and their the projected frontal area of the makes superficial velocity the range. For convenience, the list matrix element. In addition, the more fundamental parameter. has been divided into material capture cross section is propor- characteristics and separator pa- tional to the magnetic force and HGMS Results rameters. A total of 120 experi- the contaminant particle diam- We have completed a compre- ments were conducted; 76 eter and inversely proportional hensive series of HGMS experi- involved water/contaminant to the matrix element diameter ments with nonradioactive mixtures and 44 involved wa- and the matrix element spacing. surrogates, as discussed in the ter/soil/contaminant mixtures. These results are consistent with previous section, and have pro- our observations. The capture gressed to tests with radioactive The capture cross section for cross section is inversely pro- material. The results to date are each test was determined experi- portional to the superficial ve- very promising. mentally based on the measured locity, which is the ratio of the feed and effluent concentrations, separator length and the slurry the particle force balance, the residence time in the matrix. matrix geometry, and the contaminant particle size.

Nuclear Materials Technology 1994 Report 56 Waste Minimization Technologies Nuclear Applications for Magnetic Separation

4 1000010 To investigate chemical processing applications, we experi- mented with contaminated graphite residues. These series Illite-Beidellite : <45 m of tests revealed that HGMS per- 1000103 SiO2 : >10 m <45 m formance is strongly influenced by the graphite solids content. Copper oxide serves as the surrogate for plutonium oxide. 2 Figure 6 illustrates the effect of 10100 magnetic field and solids frac- % Extraction tion on the separation effectiveness for particles less than 90 85.6% µm. The highest separation effi- 1 ciency is observed under condi- 1010 tions of low solids fraction (<5% solids), with a fairly 99.8% minor effect from external field Plutonium Concentration (pCi/gm) strength. As discussed earlier, the effective particle size range 1 0 1 2 3 for the roll separator on dry powders is greater than 90 µm; thus, the coupling of HGMS Passes Through Separator with the roll separator would be advantageous for treatment of chemical processing residues. Fig. 5. HGMS extraction of plutonium oxide from soils (external magnetic field strength of 2.0 tesla ; 2–5 µm PuO2 particles). Our application of HGMS to the Tank Waste Remediation System (TWRS) indicates that we may be able to reduce certain types of tank waste directly to non-tran- To investigate soil decontamina- With silica, however, ~86% of suranic (non-TRU) levels in one tion applications, we worked the plutonium in a spiked step. The initial tank waste tar- closely with Lockhead Environ- sample was removed by HGMS get is Neutralized Cladding Re- mental Systems and Technolo- at two tesla. Surface adsorption moval Waste (NCRW), which is gies in a CRADA (cooperative appears to cause silica sand to formed from the chemical research and development bind plutonium oxide particles decladding of Zircaloy-clad agreement) effort. Our first tests to the surface. We also are plan- metallic uranium fuel that has with a spiked clay-type soil ning HGMS soil tests with the been made alkaline for storage. showed almost three orders of 3-inch superconducting magnet It contains zirconium hydrous- magnitude decontamination — located in the Plutonium Facil- oxide solids, TRU elements from about 1500 pCi/g to less ity. Model predictions and sur- (~1000 nCi/g in sludge), and than 4 pCi/g with an applied rogate work suggest increased mixed fission products. Our field of two tesla (see Fig. 5). decontamination levels at field surrogate components for strengths above two tesla. NCRW tank waste were generated from PUREX reagents used in NCRW production.13

57 Waste Minimization Technologies Nuclear Applications for Magnetic Separation

Magnetic Roll Separation Results A variety of hot tests have been conducted on a lab-scale roll separator in a glove box. Sev- eral lots of graphite powder, bomb reduction sand (MgO), 1 and sand, slag, and crucible (SS&C) residues have been processed with variable results. 0.5 2 The best results were obtained with graphite processing, in 0 1.5 which 85% of the plutonium Effectiveness was concentrated into 4% of the 1 bulk material. Particles were 0.1 sized greater than 125 µm. The plutonium content of the lean 0.5 Solids Fraction 0.2 fraction was sufficiently low in plutonium so that it could be Magnetic Field, B(T) 0.3 discarded directly to grout. The roll separator results were not as favorable with MgO or Fig. 6. HGMS effectiveness for contaminated graphite slurries as a SS&C residues. In the best case function of magnetic field strength and graphite solids fraction. for MgO, 68% of the plutonium The plutonium surrogate was 4- m copper oxide particles, and was concentrated in 44% of the the graphite solids were sized < 90 m for the experiments. bulk material. For SS&C, the results were inconclusive with no discrete separation. Yet, in the past, we have shown that magnetic separation can achieve higher performance with these Prior to precipitation of the HGMS is also being developed residues; for example, with the hydrous oxides, uranyl nitrate for magnetic filtration of fluid open-gradient magnetic separa- or iron nitrate was added to the waste streams generated during tor, we were able to concentrate acidic solution serving as the actinide chemical processing 81% of the plutonium in 33% of NCRW TRU surrogate. HGMS and for those generated at the the SS&C bulk material.2 tests of aged and unaged slur- waste treatment facility. The ries containing precipitated iron analytical model predicts that We feel that, if we can increase and zirconium have been con- HGMS can reduce the actinide the magnetic field intensity, a ducted at a one-tesla field, and concentration in liquid waste viable roll separation technology results indicated that 84% of streams by several orders of can be developed. The magnetic the aged ferric hydrous oxides magnitude. field intensity acting on the feed can be captured from a slurry. particles in magnetic roll separa- Because of this favorable out- tion is inversely proportional come using the iron solutions, to the thickness of the belt. HGMS tests are planned on zirconium/uranium/hydrous- oxide solutions.

Nuclear Materials Technology 1994 Report 58 Waste Minimization Technologies Nuclear Applications for Magnetic Separation

In the glove-box environment, If the separation efficiency is Cited References belt changes and roller align- high, the secondary component 1. W. J. Lyman, “High-Gradient ment are very difficult with our can be discarded, and the pri- Magnetic Separation,” in Unit Op- current roll separator. A second mary component, with reduced erations for Treatment of Hazardous roll separator of improved de- volume, can be further pro- Industrial Wastes, D. J. De Renzo, sign that uses a very thin (0.005 cessed or packaged for disposal. Ed. (Noyes Data Corporation, Park inch) stainless-steel belt has Concentration of the actinides Ridge, New Jersey, 1978), pp. 590– been obtained. This separator from extraneous materials be- 609. will be tested in a nonradioac- fore further processing yields tive environment and modified more efficient recovery or treat- 2. L. R. Avens, U. F. Gallegos, and for convenient glove-box use. ment operations; for example, J. T. McFarlan, “Magnetic Separa- concentrating the contaminants tion as a Plutonium Residue En- Benefits of Magnetic reduces the volume of chemical richment Process, Separation Science Separation reagents (acids) necessary for and Technology 25, 1967 (1990). We have shown that HGMS, subsequent operations. Further- open-gradient magnetic separa- more, increased concentration of 3. L. R. Avens et. al, “Magnetic tion, and magnetic roll separa- contaminants in dissolved feeds Separation for Soil Decontamina- tion can be used to concentrate allows a more efficient ion ex- tion,” Waste Management Confer- plutonium and uranium in change or solvent extraction unit ence ’93, Tucson, Arizona, February waste residues and contami- operation. Because less extrane- 28- March 4, 1993, Los Alamos nated soils. One of the major ous material is leached and National Laboratory report LA-UR- benefits of magnetic separation dissolved, the salt load on sub- 93-229. technology is that it only parti- sequent waste treatment operations the existing waste volume. tions is reduced. 4. L. A. Worl et. al, “Remediation A high contaminant concentra- of Hanford Tank Waste Using Mag- tion of a primary component Magnetic separation could be netic Separation,” Waste Manage- and a low contaminant concen- applied to on-site decontamina- ment Conference ’93, Tucson, tration of a secondary compo- tion of radioactive spills. How- Arizona, February 28- March 4, nent are created. ever, development of a mobile 1993, Los Alamos National Labora- magnetic separation unit would tory report LA-UR-92-3977. be required. Portable superconducting magnet systems do exist 5. F. C. Prenger et. al, “High Gradi- for mobile Magnetic Resonance ent Magnetic Separation Applied to Imaging. Development of mag- Environmental Remediation,” netic filtration may be a method Cryogenic Engineering Conference, to cut to very low levels the TRU Albuquerque, New Mexico, July effluents from Department of 12–16, 1993, Los Alamos National Energy facilities and, thereby, Laboratory report LA-UR-93-2516. greatly reduce waste treatment cost. ✦

59 Waste Minimization Technologies Nuclear Applications for Magnetic Separation

6. I. V. Akoto, “Mathematical Mod- 11. Z. J.J. Stekly and J. V. Minervini, eling of HGMS Devices,” IEEE “Shape Effect of the Matrix on the Transactions on Magnetics Mag-13, Capture Cross Section of Particles 1486–1489 (1977). in High Gradient Magnetic Separa- tion,” IEEE Transactions on Magnet- 7. F. J. Friedlander, M. Takayasu, J. ics Mag-12, 474–479 (1976). B. Retttig, and C. P. Kentzer, “Par- ticle Flow and Collection Process in 12. J. H. P. Watson, “Magnetic Fil- Single Wire HGMS Studies,” IEEE tration,” Journal of Applied Physics Transactions on Magnetics, Mag-14, 44, 4209–4213 (1973). 1158–1164, (1978). 13. “PUREX Technical Manual, 8. W. F. Lawson, W. H. Simons, and PUREX Systems and Technology R. P. Treat, “The Dynamics of a Par- Chemical Processing System Engi- ticle Attracted by a Magnetized neering,” Westinghouse Hanford Wire,” Journal of Applied Physics 48, Company report WHC-CM-5-25 3213–3224 (1977). (July 1988).

9. F. E. Luborsky and B. J. Drummond, “Buildup of Particles on Fibers in a High-Field, High- Gradient Separator,” IEEE Transac- tions on Magnetics Mag-12, 463–465 (1976).

10. J. E. Nesset and J. A. Finch, “A Static Model of High-Gradient Magnetic Separation Based on Forces within the Fluid Boundary Layer,” in Industrial Applications of Magnetic Separation, Y. A. Liu, Ed. (Institute of Electrical and Elec- tronic Engineers, New York, 1979) pp. 188- 196.

Nuclear Materials Technology 1994 Report 60 Waste Minimization Technologies Distillation Separation of Actinides from Waste Salts

Eduardo Garcia, Introduction Vonda R. Dole, Plutonium-bearing residue salts resulting from past pyrochemical Walter J. Griego, processing have accumulated in the Department of Energy (DOE) John J. Lovato, weapons complex. This problem is especially acute at Rocky Flats, James A. McNeese, where approximately 20 metric tons of salt await disposition. and Keith Axler Pyrochemical salt wastes will also be generated in the future by Nuclear Material Processing: Complex 21. In addition to containing plutonium with its attendant Chloride Systems hazards, these salts may contain reactive metals. The reactive component places the salts in the category of mixed waste and unacceptable under Waste Isolation Pilot Plant (WIPP) acceptance criteria. Because of restrictions on plutonium loading in individual waste drums, WIPP may not even have the capacity for the number of drums that would be generated, even if the waste were acceptable. The salts must therefore be treated to remove as much plutonium as possible and render inert any reactive components.

Physical separation processes, such as distillation separation, are more attractive than chemical or dissolution processes because physical processes generate much less secondary waste. A physical “It is theoretically possible separation of waste salts from plutonium is theoretically possible to remove plutonium below because of a large difference in vapor pressure. A high-temperature the 1 ppm level, which distillation process can be used to volatilize the alkali and alkaline- earth chloride salts from essentially nonvolatile actinide oxides. would then classify the salts An oxygen sparging process developed at Los Alamos pretreats as low-level waste rather the salts to convert actinide species to oxides or oxychlorides and simultaneously converts reactive metals into inert oxides so that then transuranic waste and they are no longer mixed waste. The salt is then evaporated to result in even greater cost leave a plutonium dioxide heel suitable for storage. Waste volume savings and a more benign will be essentially limited to the salts themselves. Moreover, the greatly reduced plutonium content of the salts will result in a waste form.” greatly reduced number of waste drums for WIPP disposal and a large reduction in the cost of disposal. It is theoretically possible to remove plutonium below the 1 ppm level, which would then classify the salts as low-level waste (LLW) rather than transuranic (TRU) waste and result in even greater cost savings and a more benign waste form.

61 Waste Minimization Technologies Distillation Separation of Actinides from Waste Salts

Distillation Separation Process If the temperature of the surface The distillation separation process can be modeled by the kinetic that the gas strikes is much theory of gases.1 From this theory, one can show that the rate of the colder than the temperature of mass of a gas incident on a unit area per unit time is the gas vapor, then condensation will occur with a sticking coefficient of 1. Under these condi- , (1) tions, the condensation rate is equal to the incident rate, and, thus, the amount of material where deposited can be estimated by W = gas incident rate (g cm-2 sec-1), using the calculated incident M = molecular weight (g mol-1), rate. In the case of molten salt R = gas constant (8.325 x 107 ergs K-1 mol-1), distillation, the condenser tem- T = temperature (K), perature will be much lower P = vapor pressure (torr) at temperature T, and than the temperature at which C = conversion factor (1333 dyne cm-2 torr-1). the material is evaporating, so that estimating the amount de- Substituting the values for the gas constant and the conversion posited will be possible. Most factor produces the rate equation importantly, calculations can be made for the rate of deposition of pyrochemical salts because . (2) the vapor pressure of the various components, with one exception, is known.

Table I lists vapor pressures of common pyrochemical salt components at selected tempera- Table I. Vapor Pressure of Pyrochemical Salt Components tures.2-9 The extremely low vapor pressures for plutonium Compound 850 °C 950 °C 1050 °C Ref. No. dioxide and calcium fluoride are extrapolations from higher temperature data. Table I shows -0.063 .0.64 1.2 NaCl 10 10 10 2 that the best separation can be KCI 100.23 100.90 101.5 3 achieved between plutonium dioxide and the chloride salts; 0.27 0.97 1.6 MgCl2 10 10 10 4 it also shows that a good separa- CaCl 10-2.9 10-1.9 10-1.0 4 tion between the salts and pluto- 2 nium trichloride cannot be -7.7 -6.1 -4.7 CaF2 10 10 10 5 achieved. To ensure optimal Pu 10-8 10-7 10-5 6 separation, Los Alamos pretreats the pyrochemical waste salts us- -1.8 -0.86 -0.086 PuCl3 10 10 10 7 ing an oxygen sparging process, PuO 10-15.7 10-13.5 10-11.7 8 or alternate oxidation processes, 2 to convert plutonium and ameri- -6 -5 PuOCl 10 10 9 cium species to oxides or oxychlorides.

Nuclear Materials Technology 1994 Report 62 Waste Minimization Technologies Distillation Separation of Actinides from Waste Salts

Table II. Deposition Rate (g hr-1) from 100 cm2 Surface Area The deposition rates can be used to calculate the composition of the distillate salt under ideal Compound 850 °C 950 °C 1050 °C conditions. For example, in a salt that contains only sodium NaCl 4100 20000 75000 chloride and plutonium dioxide at 850 °C with a 100 cm2 surface KCI 9100 41000 140000 area, the rate of deposition of so- MgCl 11000 55000 200000 dium chloride is 4100 g per hour 2 while that of plutonium dioxide -12 CaCl2 9 85 560 is 10 g per hour. The distillate CaF 10-3.9 10-2.3 10-1.0 salt would then contain a weight 2 concentration of 10-10 ppm -11 -9.0 -7.2 AmO2 10 10 10 plutonium. At 1050 °C, the dis- PuO 10-12 10-9.6 10-7.8 tillate salt would have a pluto- 2 nium concentration of 10-7 ppm. Pu 10-3.7 10-2.4 10-1.4 The LLW criterion for plutonium-239 is on the order of 1 ppm. Therefore, vacuum distillation of a sodium chloride, Table II lists the rate of deposi- the chloride salts, except cal- potassium chloride, or magne- tion (at selected temperatures) cium chloride, can be distilled sium chloride salt that contains of various compounds. Figure 1 at acceptable rates below only plutonium dioxide has the shows calculated deposition 900 °C. Calcium chloride can potential to produce a waste salt rates as a function of tempera- be distilled at 1100 °C. Even at with a plutonium concentration ture. These are calculated rates, 1100 °C, the vapor pressure of 10 orders of magnitude below however, and actual rates can plutonium and americium the LLW criterion. In the case of be slower by some orders of dioxide is low enough to still calcium chloride, the calculated -5 magnitude. Nevertheless, all get an excellent separation concentration is 10 ppm at from the salt. 1050 °C. Figure 2 shows a plot of plutonium concentration in various distilled pyrochemical salt components as a function of temperature. As shown in this plot, even at 1200 °C all the chloride salts will be considerably below the 1.6 ppm LLW cutoff for plutonium. The americium content must be below 0.03 ppm for LLW, and this requirement can also be met assuming all the americium is present as AmO2.

Fig. 1. Distillation rates as a function of temperature.

63 Waste Minimization Technologies Distillation Separation of Actinides from Waste Salts

A picture of the waste salt before vacuum distillation is shown in Fig. 3, and the resulting distilled salt and plutonium-dioxide heel are shown in Fig. 4.

Preliminary experiments with alternate oxidation pretreat- ments have also been very encouraging. Many different oxidation agents were considered, but sodium carbonate and calcium carbonate were selected because of desirable solubility and stability properties. They are also essentially nonhazard- Fig. 2. Calculated plutonium concentration in distilled salts ous materials. Tests with pluto- assuming only PuO present. nium metal in molten salts have 2 shown them to be effective in converting the metal to the dioxide. Experiments with waste Pretreatment of Waste Salts salts have also produced pluto- The calculations in the previous section assumed that the waste in- nium dioxide as the only pluto- cluded only chloride salt and plutonium dioxide. Actual waste salts nium species. Thermodynamic may also contain plutonium metal and plutonium trichloride and modeling predicts that the only must be pretreated to convert all actinide species to oxides. Oxygen materials that will be left in the sparging can convert plutonium trichloride to plutonium dioxide. distillation heel will be pluto- However, because neither plutonium metal nor oxygen is soluble nium dioxide and graphite. A in the chloride salts to a significant extent, using oxygen sparging final calcination step can be in- to convert plutonium metal to plutonium dioxide is more difficult. corporated into the process to Plutonium oxychloride is also generally a product of oxygen convert the graphite to carbon sparging and, based on preliminary vapor pressure measurements,9 dioxide, leaving a plutonium- may not be desirable. Alternate oxidation agents that are soluble dioxide heel suitable for storage. in the salt are being examined experimentally and modeled theoretically. Future Work Preliminary results with Results nonoptimized equipment have We are still in the early stages of distillation experiments and are us- already demonstrated that distil- ing equipment that was easily available but not necessarily designed lation can be used to purify so- for our purposes. Nevertheless, early results have been extremely dium chloride-potassium encouraging. Greater than 95% of the salt was recovered from plu- chloride salts so that they con- tonium-rich fractions of sodium chloride-potassium chloride salts tain plutonium in the range of pretreated by oxygen sparging. The plutonium concentration was hundreds of ppm. Our ultimate reduced from approximately 50% to the range of hundreds of ppm. objective is to produce salts that Although still significantly above the LLW cutoff, this concentration can be disposed of as LLW. Bet- is low enough to allow very significant savings in costs for WIPP ter equipment is being designed disposal of Rocky Flats salts. Experiments with “cold salts” in the to achieve this objective. glovebox indicate that the level of contamination in the distilled salts may be the result of handling in the glove-box environment, not the distillation process itself. The distillation heels left in the crucible consisted of loose powders that could be poured out.

Nuclear Materials Technology 1994 Report 64 Waste Minimization Technologies Distillation Separation of Actinides from Waste Salts

The encouraging preliminary results and the calculations predicting an extremely low plutonium concentration of the distillate salt lead us to conclude that the ambitious goal of converting plutonium- bearing waste salts to low-level waste is achievable.✦

Cited References Fig. 3. Plutonium-rich 1. S. Dushman, Scientific Founda- fraction of an oxygen- tions of Vacuum Technique, 2nd ed. sparged salt before vacuum (John Wiley and Sons, New York, distillation. 1962). 2. J. L. Barton and H. Bloom, J. Phys. Chem. 60, 1413 (1956).

3. E. E. Schrier and H.M. Clark, J Phys. Chem. 67, 1259 (1963).

4. O. L. Hilenbrand and N. D. Pot- ter, J. Phys. Chem. 67, 2231 (1963).

5. D. A. Schultz and A. W. Searcy, J. Phys. Chem. 67, 103 (1963).

6. R. Hultgren, P. D. Desai, D. T. Hawkins, M. Gleiser, K. K. Kelly, and D. D. Wagman, Selected Values of the Thermodynamic Properties of the Fig. 4. Plutonium-dioxide powder heel and purified condensate salt. Elements (American Society for Met- als, Metals Park, Ohio, 1973), p. 402.

7. T. E. Phipps, G. W. Sears, R. L. Calcium chloride salts constitute another major fraction of the stock- Siefert, and O. C. Simpson, piled waste salts, and equipment is being obtained to carry out the J. Phys. Chem. 18, 713 (1950). distillation process at the higher temperatures required for calcium chloride. 8. R. J. Ackerman, R. L. Faircloth, and M. H. Rand, J. Phys. Chem. 70, 3698 (1966). Experiments are also being carried out on “cold” salts to help optimize processing parameters such as time, temperature, and 9. M. A. Williamson, Los Alamos pressure. These experiments will aid condenser and other equip- National Laboratory, personal com- ment design. Surrogate metal oxides will be mixed with salts to munication, June 1993. help determine the extent to which the salts can be purified in a noncontaminated environment. Experiments with carbonate oxidation agents will also continue.

65 Waste Minimization Technologies Evaluating Existing Partitioning Agents for Processing Hanford High- Level Waste Solutions

S. Fredric Marsh Introduction Nuclear Materials Processing: The Hanford Reservation in Washington State incorporates 177 Nitrate Systems underground storage tanks that contain over 165 million curies of radioactivity in more than 65 million gallons of waste. Many differ- Zita V. Svitra and ent processes were used at Hanford during the half-century these Scott M. Bowen wastes were accumulated from nuclear materials processing for Separation and Radiochemistry national defense requirements. As a result, many different reagents Chemical Science and Technology and waste streams were neutralized and combined in underground Division tanks, often with insufficient concern about the compatibility of the components being stored. The resulting wastes are an extremely complicated and sometimes unstable mixture of sludges, saltcakes, slurries, and supernates. Adding significantly to the stored waste problem is the fact that 67 tanks are known, or are suspected, to have leaked.

The U.S. Department of Energy (DOE) is committed to the environ- “Los Alamos National mental remediation of hazardous wastes stored at the Hanford Site. Laboratory has been asked Los Alamos National Laboratory has been asked by DOE to support the Hanford Tank Waste Remediation System (TWRS) mission, by DOE to support the which is to “store, treat, and dispose of all tank waste in a safe, Hanford Tank Waste cost-effective, and environmentally sound manner.” An essential requirement for achieving the TWRS mission is identifying suitable Remediation System partitioning agents and technologies to accomplish this challenging mission, which is to 'store, goal. treat, and dispose of all tank Experimental Program waste in a safe, cost-effec- The objective of our study was to evaluate the performance of a tive, and environmentally large number of potentially useful absorber materials, including sound manner.'” many not previously studied, in realistic waste simulants of acid- dissolved sludge and acidified supernate solutions found in Hanford high-level waste (HLW) Tank 102-SY. Relatively few absorber materials have been evaluated with media approaching the complexity of Hanford HLW solutions. Nor have other inves- tigators evaluated the distribution of such a large number of elements onto so many absorbers.

We evaluated 60 commercially available or experimental absorbers for partitioning high-level radioactive waste. We use the term “absorbers” to represent solid materials that remove ionic species from aqueous solutions, whether by ion exchange, chelation, ion trapping, molecular sieving, or other mechanisms. Absorbers in our study included cation and anion exchange resins, inorganic exchangers, composite absorbers, and a series of liquid extractants sorbed on porous support beads. We measured the distributions of 14 elements onto these absorbers.

Nuclear Materials Technology 1994 Report 66 Waste Minimization Technologies Evaluating Existing Partitioning Agents for Processing Hanford High-Level Waste Solutions

The selected elements, which We also included some experi- Results represent fission products (Ce, mental absorbers to provide Only a brief summary of our Cs, Sr, Tc, and Y), actinides (U, guidance for future research results is presented here. Pu, and Am), and matrix ele- activities. A detailed description and all ments (Cr, Co, Fe, Mn, Zn, and experimental data have3 been Zr), were traced by radionu- Waste Simulants published elsewhere. clides and assayed by gamma We recognize that no waste spectrometry. Distribution simulant will exactly represent Behavior for many absorber/ coefficients for each of the 1680 the contents of an HLW tank. element combinations was sub- element/absorber/solution Nevertheless, we have used the stantially different from what combinations were measured best information available in would be expected, based on for dynamic contact periods of our effort to prepare realistic published measurements from 30 minutes, 2 hours, and 6 hours simulants. Ultimately, the most relatively clean solutions; this to provide sorption kinetics promising absorbers must be finding demonstrates the information for the specified tested with actual waste. importance of using realistic elements from these complex However, because actual waste simulants. In some cases, we media. We measured more than samples are expensive and diffi- found that inexpensive commer- 5000 distribution coefficients. cult to obtain, we suggest that cial partitioning agents outper- screening of promising candi- form specialty products whose Absorbers date absorbers precede tests costs are orders of magnitude Many organic and inorganic with actual Hanford tank waste. higher. absorbers are more resistant to chemical attack and radiation Flow Sheet Options In addition to identifying damage than are most liquid Our identification of suitable promising existing partitioning extractants. Moreover, solid partitioning agents was also agents, we determined which absorbers can readily be used intended to help designers de- partitioning agents and tech- in compact, low-cost columns velop reliable flow sheets. nologies, including some of that provide rapid multistage The selection of agents will, those that have been designated separations. A typical operation of course, depend on the par- as the Hanford baseline tech- cycle for a solid absorber col- ticular objectives of the flow nologies, are marginal or inad- umn consists of loading the ab- sheet. Although our study ini- equate. Table I summarizes sorber and elution of the sorbed tially was intended to support how well each of the 14 ele- element(s), followed by regen- the flow sheet development pro- ments is sorbed, from the two eration and reuse of the ab- gram for Hanford Tank 102-SY,1 acidic solutions, by showing the sorber. At times, the targeted many of our experimental find- number of effective absorbers in elements are sorbed so strongly ings are applicable to other flow each of the six distribution coef- that elution is difficult; in such sheet proposals, such as the so- ficient (Kd) categories. Where cases, the loaded absorber be- called “Clean Option.”2 Deci- acceptable absorbers have been comes a candidate for the final sions about whether the targeted identified, we recommend that waste form. elements should be recovered these absorbers be tested with individually, or grouped for sub- other simulated tank composi- Although most of the absorbers sequent disposal, are properly tions and with actual waste. we evaluated are cation ex- left to the flow sheet designers. Where acceptable absorbers changers, we included some an- Whichever flow sheet is have not been identified, we rec- ion exchange resins and a series adopted, we expect that our ommend an accelerated effort to of liquid extractants (sorbed on experimental data will help the identify or develop satisfactory porous carbon beads) to obtain designer anticipate the perfor- partitioning agents. more comprehensive data, as mance of existing partitioning well as to supply convenient agents. reference points to other studies.

67 Waste Minimization Technologies Evaluating Existing Partitioning Agents for Processing Hanford High-Level Waste Solutions

Table I. Number of Identified Absorbers in Various Kd Categories for Each of 14 Elements, from Two Solutions Simulating Hanford HLW Tank 102-SY

Kd Kd Kd Kd Kd Kd Kd Medium Element >1000 300-1000 100-300 20-100 10-20 5-10

Acid- Ce 0 0 0 5 1 1 Dissolved Cs 7 0 0 11 1 4 Sludge Sr 0 0 0 0 0 0 Tc 0 0 1 11 2 5 Y 0 0 0 0 0 0 Cr 0 0 0 0 0 0 Co 0 0 0 0 1 0 Fe 0 0 0 0 0 0 Mn 0 0 0 0 0 0 Zn 0 0 0 0 0 0 Zr 0 1 3 10 4 3 U 0 0 0 6 8 5 Pu 2 5 6 2 8 2 Am 0 0 0 0 0 6

Acidified Ce 1 3 3 10 3 6 Supernate Cs 8 1 5 2 4 2 Sr 0 0 0 0 3 4 Tc 0 7 7 9 3 1 Y 1 0 1 9 0 5 Cr 0 0 0 2 2 5 Co 0 0 0 1 3 4 Fe 0 0 0 0 0 3 Mn 0 0 0 0 1 5 Zn 0 0 0 5 3 2 Zr 0 0 0 1 4 3 U 0 1 4 15 8 8 Pu 0 0 0 0 0 1 Am 0 3 2 13 1 5

Elements that show good-to-excellent sorption from the simulant solution of acid-dissolved sludge are cerium, cesium, technetium, zirconium, uranium, and plutonium (Table II). Elements that show low sorption from this solution are strontium, yttrium, chromium, cobalt, iron, manganese, zinc, and americium.

Nuclear Materials Technology 1994 Report 68 Waste Minimization Technologies Evaluating Existing Partitioning Agents for Processing Hanford High-Level Waste Solutions

Table II. Absorbers* Whose Measured Distribution Coefficients Exceed 20 for the Specified Elements, from Simulated Acid-Dissolved Sludge Solution

Medium Element Absorbers

Acid- Ce JSK-1, JSK-2, JSK-3, PuroliteTM A-520-E, IonacTM SR-3 Dissolved Sludge Cs SNL/CSTs, AMP-PAN, M315-PAN, NiFC-PAN

Tc ReillexTM HPQ, IonacTM SR-3, PuroliteTM A-520-E TM Sybron (Et)3N, LANL TiO2, LANL ZrO2, LANL Nb2O5

Zr NaY-PAN, SNL/CSTs, DuracilTM 230, TiO-PAN, IonsivTM TIE-96, M315-PAN, DuoliteTM CS-100, CyanexTM 272

TM TM TM U TRU-Spec , Sybron (Et)3N, Ionac SR-3, Purolite A-520-E

Pu JSK-3, JSK-2, JSK-1, ReillexTM HPQ, TRU-SpecTM, Sybron SR-3, PuroliteTM A-520-E, CyanexTM 923

* Detailed descriptions of these absorbers, whose trade names are used here, have been published elsewhere (see Reference 3).

Elements that show good-to- simulant solutions selected for Hanford tanks, especially those excellent sorption from the this study, some of the most on the safety watch-list. The 14 simulant solution of acidified promising absorbers identified elements whose distributions supernate are cerium, cesium, in this study may have applica- have been measured should be technetium, yttrium, zinc, ura- tions at TA-55. This is especially carefully examined to determine nium, and americium (Table III). likely for those absorbers and if any should be dropped or if Elements that show poor-to- extractants that sorb americium additional elements should be marginal sorption from this so- strongly. Testing of likely ab- included in future measure- lution are strontium, chromium, sorbers with TA-55 acidic and ments. Finally, radioactive trac- cobalt, iron, manganese, zirco- alkaline waste is planned. ers, shown to provide a rapid, nium, and plutonium. reliable, and inexpensive way Future Studies to obtain large amounts of distri- Applications to Los Alamos Our screening study has met its bution data, should be used in Plutonium Facility Waste objective of identifying specific such studies. The Los Alamos Plutonium absorbers that appear capable Facility at TA-55 has assigned a of partitioning targeted elements The effects of radiation on the high priority to waste minimiza- from Hanford HLW Tank 102- most promising absorbers from tion. Because the acidic waste SY. The best absorbers from this our present and future studies streams generated at TA-55 are study and perhaps a few prom- should be determined, when ising new materials should be such information is not already in some ways similar to the 4 evaluated with simulated com- available, before partitioning positions that represent other agents are selected or before any large-scale recovery flow sheet is selected.

69 Waste Minimization Technologies Evaluating Existing Patitioning Agents for Processing Hanford High-Level Waste Solutions

Table III. Absorbers* Whose Measured Distribution Coefficients Exceed 50 for the Specified Elements, from Simulated Acidified Supernate Solution

Medium Element Absorbers

Acidified Ce CyanexTM 923, TRU-SpecTM, SNL/HTO, CMPO-DIPB, AmberlystTM 15, Supernate DHDECMP, MgO-PAN, DiphonixTM, SNL/CST111, AmberlystTM XN-1010

Cs SNL/CSTs, NiCF-PAN, AMP-PAN, M315-PAN, DurasilTM 230, SRS resorcinol, IonsivTM TIE-96

Tc Sybron (Pr) N, CyanexTM 923, AliquatTM 336, Sybron (Et) N, 3 3 ReillexTMHPQ, PuroliteTM A-520-E, IonacTM SR-6, IonacTM SR-3, TM TM TRU-Spec , LIX -26, DHDECMP, CMPO-DIPB, LANL TiO2, LANL ZrO2, LANL Nb2O5

Y CyanexTM 923, TRU-SpecTM, AmberlystTM 15, MgO-PAN, CMPO-DIPB, AmberlystTM XN-1010

Zn NiFC-PAN, DuoliteTM C-467, NaTiO-PAN

U CyanexTM 923, DiphonixTM, TRU-SpecTM, TiO-PAN, SNL/HTO, CMPO-DIPB, SNL/CST35, CyanexTM 272, DuoliteTM C-467, AmberlystTM 15

Am CyanexTM 923, TRU-SpecTM, SNL/HTO, CMPO-DIPB, AmberlystTM 15, SNL/CST111, AMP-PAN, Bone char, AmberlystTM XN-1010, LIXTM-1010 * Detailed descriptions of these absorbers, whose trade names are used here, have been published elsewhere (see Reference 3).

Conclusions Cited References M. Bowen, “Distributions of 14 Ele- Our experimental data indicate 1. S. L. Yarbro et al., “Status of Tank ments on 60 Selected Absorbers that many existing partitioning 102-SY Remediation Flowsheet De- from Two Solutions (Acidic Dis- agents may be suitable for HLW velopment at LANL,” Los Alamos solved Sludge and Alkaline National Laboratory report LA-UR- tank remediation. If reliable Supernate) Simulating Hanford 93-1725 (April 1993). partitioning agents can be iden- HLW Tank 102-SY,” Los Alamos tified, Hanford HLW tank pro- 2. J. L. Straalsund et al. “Clean Op- National Laboratory report LA- cessing might begin and reach tion: An Alternative Strategy for 12654 (October 1993). completion sooner, and at a Hanford Tank Waste Remediation: lower cost, than would other- Volume 1. Overview,” Pacific North- 4. S. F. Marsh and K. K. S. Pillay, wise be possible. We therefore west Laboratory report PNL-8388 “Effects of Ionizing Radiation on feel that the findings of our Vol. 1 (December 1992). Modern Ion Exchange Materials,” study could have a major benefi- Los Alamos National Laboratory cial impact on environmental 3. S. F. Marsh, Z. V. Svitra, and S. report LA-12655 (October 1993). remediation efforts at Hanford and elsewhere within the DOE complex.✦ Nuclear Materials Technology 1994 Report 70 Waste Minimization Technologies Gas Pycnometry for Density Determination of Plutonium Parts

J. David Olivas and Introduction S. Dale Soderquist Traditionally, determination of plutonium component density has Plutonium Metallurgy been accomplished by measuring the apparent weight loss when Henry W. Randolph, the object is immersed in a liquid of known density (Archimedes’ Susan L. Collins, principle). Liquids that are compatible with plutonium, and that Darren Verebelyi, have been successfully used in the past, include freon and William J. Randall, monobromobenzene. However, these fluids have undesirable char- and Gregory D. Teese acteristics that present health and environmental hazards. Further- Equipment Engineering Section more, the spent fluid stream is environmentally undesirable. Gas Westinghouse Savannah River pycnometry eliminates the liquid waste as well as the gaseous emis- Company sions from volatiles. The focus of current gas pycnometry research and development is to prove that this technique can reliably determine the density of plutonium subscale shapes to the desired accuracy. If this goal is met, gas pycnometry can be expanded to address all plutonium density determination needs. Background The accurate determination of part density is an important parameter in plutonium fabrication. Densities of solids are routinely determined by the hydrostatic technique of measuring the apparent “The focus of current gas weight loss when the solid is submerged in a liquid of known pycnometry research and density.1 When the sample volume, determined from Archimedes’ principle, is combined with the true weight of the sample, its den- development is to prove sity is readily calculated. However, the hydrostatic technique is that this technique can subject to error if the sample surface has cracks that are not com- reliably determine the pletely filled by the liquid. Similarly, trapped air bubbles can intro- duce errors. Furthermore, the thermal effects of heat-generating density of plutonium materials such as plutonium can induce thermal currents or raise subscale shapes to the the temperature of the fluid and thus affect the accuracy of the results. If the fluid chemically reacts with the sample, or adsorbs on desired accuracy. If this the surface of the sample, the sample will be physically altered, and goal is met, gas pycnometry results will be inaccurate. can be expanded to address all plutonium density Accuracy of the hydrostatic technique for determining plutonium determination needs.” part density thus depends, in large part, on the choice of fluid. Freon and monobromobenzene, the traditional fluids of choice for plutonium, are less attractive now because of emerging environment, safety, and health regulations. Freon production is being cur- tailed by the Montreal Protocol, whereas monobromobenzene is a suspect carcinogen. Searches for other liquids were conducted, and some were found that provided adequate results. Unfortunately, these liquids required the use of a solvent to remove them from the plutonium surface when density determination was complete. The increasing regulation of liquid waste disposal prompted efforts to identify a method of density determination that is not dependent upon the hydrostatic technique.

71 Waste Minimization Technologies Gas Pycnometry for Density Determination of Plutonium Parts

Gas pycnometry is used success- Gas fully in industries that are ill- Injector suited for hydrostatic density Actuator determination. These industries typically process powders, such as metal powders, ceramic powders, and dry cement. To use this technique, a volume of gas is allowed to expand in a cali- brated sample chamber, and the Copper pressure difference caused by Sample the sample volume is calculated. Chamber Different implementations are possible, but all rely upon the fundamental gas law. Equation 1 is the fundamental gas law at Bellows constant temperature and solving for the new volume V2 yields Equation 2:

P1V1 = P2V2 (1) and

V2=V1(P1/P2). (2) Fig. 1. Schematic of prototype gas pycnometer. The application of gas pycnometry to density determination of plutonium parts is more complicated than normal Los Alamos researchers entered into a technology development industrial applications, prima- partnership with equipment development specialists from the Sa- rily because plutonium is self- vannah River Technology Center (SRTC), a division of the heating and will increase the Westinghouse Savannah River Company. The goal of this partner- temperature of the gas in the ship was to design, build, and test a prototype system to demon- pycnometer. The fundamental strate the application of gas pycnometry on a subscale plutonium gas law at nonconstant tempera- shape. Requirements developed for the gas pycnometer system ture and its solution for volume were as follows: V2 are shown in Equations 3 • determines density within ± 0.05%, and 4: • does not leave residues or contaminants on part, • accommodates self-heating of plutonium, (P1V1)/T1=(P2V2)/T2 (3) • operates in a glove-box environment, and and • completes measurement cycle in less than two hours. Prototype Gas Pycnometer System V =V (P /P )(T /T ) . (4) 2 1 1 2 2 1 A schematic of the device developed by SRTC is shown in Fig. 1. In addition, the precision de- The sample chamber is made of a solid block of copper to provide sired for plutonium fabrication the required heat transfer characteristics for the temperature control is much higher than standard system. The designers selected a simple-to-machine, hemispherical industrial practice. Considering shell for the sample part. The chamber closely matches the shape these factors, we concluded that of the sample so that the sample will occupy a significant portion a system designed specifically of the available volume. The chamber volume should be within 50% for use with plutonium would of the volume of the part to allow a sufficiently large change in gas be required. Nuclear Materials Technology 1994 Report 72 Waste Minimization Technologies Gas Pycnometry for Density Determination of Plutonium Parts pressure and thus provide rea- The set point is selected by the heating wire was constructed to sonable accuracy. Other notable computer to ensure that heat is simulate the self-heating proper- features of the device include always being removed. In addi- ties of plutonium. A plutonium the rubber bellows in the bottom tion to maintaining the device at shell has been designed and is portion for opening and closing the set point temperature, the to be fabricated at Los Alamos. the chamber and the side- thermoelectric coolers have suf- mounted actuator for injecting ficient capacity to remove the Savannah River cold-tested the the gas charge. heat generated by a plutonium gas pycnometer with both the sample. aluminum and stainless-steel To achieve the ± 0.05% repeat- shells. The device was then ability specification for part An actuator on the side of the transported to Los Alamos density, variations from one device injects a known, repeat- where it was cold-tested again measurement cycle to the next able amount of gas into the with the aluminum shell. had to be carefully controlled. chamber. The pressure increase An analysis of measurement caused by the gas injection is Aluminum Shell Test Results at uncertainties indicated that one used to calculate the chamber Savannah River of the major design challenges volume from the fundamental The aluminum shell was used was closure of the sample cham- gas law. The difference between for most of the Savannah River ber to achieve a highly repeat- pressures in the chamber with experiments. Its true volume able volume. A rubber bellows the sample present and absent is was determined by the hydro- actuator was selected to lower used to calculate the sample vol- static technique, using water as and raise the male portion of the ume. Pressure is measured with the working fluid. The immer- chamber in order to open and a highly accurate quartz pres- sion was performed in a vacuum close the chamber. The pressure sure transducer. The computer so that air bubbles could not in the bellows is controlled to reads the transducer for twenty form and bias the results. The within 0.05 psi to assure consis- measurements before releasing mean of the results of a number tent compression of the O-ring the actuator to relieve gas injec- of trials was used as the best es- seal. Also, the bellows accom- tion pressure, and then twenty timate of the true part volume. modates angular misalignment, low pressure readings are taken. The standard deviation of the and this inherent property en- Readings are taken multiple hydrostatic measurements was sures that pressure is applied times to establish stability and 0.08%. uniformly around the O-ring give statistical significance to circumference. the values used in subsequent A number of trials were then calculations. performed using the aluminum Temperature is controlled shell in the gas pycnometer. through a feedback control A computer running custom The repeatability of the results loop. Thermistors are used to data collection and device con- was superior to the hydrostatic measure the temperature within trol software controls the entire technique (0.01% at 1 std dev). ± 0.001 K. The thermistor values system. The computer does not However, an offset of the mean are read by the control com- allow a measurement cycle to was noted. Analysis of the re- puter, which controls the heat begin if temperature stability is sults indicated that the swept removal rate by adjusting the not achieved. Once a measure- volume of the actuator, and voltage output to the thermo- ment has been initiated, no user hence the volume of the gas electric coolers. Heat must be interaction is required; thus, op- injection, was different than the constantly removed from the de- erator variability is not a factor theoretical value because of me- vice to maintain the temperature in the measurement process. chanical tolerances in mounting within ± 0.01 K of the set point. the actuator. The correct value An aluminum hemispherical for the actuator volume was shell was fabricated for initial then calculated and used in out-of-glove-box testing of the future measurements. device. In addition, a stainless- steel shell encircled by resistance

73 Waste Minimization Technologies Gas Pycnometry for Density Determination of Plutonium Parts

Stainless-Steel Shell Test Aluminum Hemisphere Data from LANL Trials Results at Savannah River 75.4 The stainless-steel shell had a 75.35 series of concentric grooves 95% Confidence machined into the outer surface. Upper Goal Resistance heating wire was 75.3 Band wound into these grooves and

connected to a power supply. 75.25 By varying the supply voltage,

the power of the heater was ad- Volume Lower Goal justed to approximate the power 75.2 generated by a plutonium shell. The power dissipation during 75.15 trials with the stainless-steel (in Cubic Centimeters) shell was 2.5 watts. 75.1 0 5 10 15 20 As anticipated, trials conducted Trial Number on the heated stainless-steel shell showed a loss in precision. The loss is attributed to uneven Fig. 2. Los Alamos aluminum shell data compared to experimental goals. heat generation, which leads to variations in the temperature of the gas in the annular space. measured with the hydrostatic Summary However, the observed precision monobromobenzene method. Los Alamos scientists, in coop- of 0.03% is still better than the No further testing is planned eration with their SRTC part- goal of 0.05%. Uniform heat with the heated stainless- ners, are evaluating gas generation from plutonium steel shell; instead, the pluto- pycnometry for determining should reduce the variability nium shell will be used. density of plutonium parts. somewhat. The plutonium shell will be This replacement for the current measured with both the gas hydrostatic immersion tech- Aluminum Shell Test Results pycnometry and hydrostatic nique will eliminate a liquid at Los Alamos monobromobenzene methods waste stream and support the After transport of the gas pyc- to determine the success of the vision of an environmentally nometer to Los Alamos, testing pycnometer system. benign plutonium facility. with the aluminum shell was Initial testing of gas pycnometry repeated to confirm that the The pycnometry method is in- with surrogate materials indi- device was not damaged in herently less flexible than the cates the technique can provide transit. Repeatability within hydrostatic technique because results that are superior to those 0.01% was achieved, confirming of the geometric constraints of from existing methods. Density proper operation of the device the pycnometer sample cham- of aluminum hemispherical and reconfirming the results ber. Near-term efforts include shells can be determined to ± from Savannah River. The fabricating a pycnometer system 0.01% of the actual density. Test- results of this testing are that will accommodate specific ing with plutonium shells will presented in Fig. 2. part geometries for Los Alamos commence in the near future.✦ missions. A longer-term effort Future Directions will examine alternative meth- Cited Reference Preparations are in progress for ods of increasing robustness, 1. F. W. Dykes and J. E. Rein, “Ordi- installing the gas pycnometer in such as replaceable chamber nary Weighing and Density Deter- a glove box in the Los Alamos inserts or premeasured gas mination,” Progress in Nuclear Plutonium Facility. After instal- volumes at higher pressures. Energy X, 11-34 (1970). lation, the aluminum shape Eventually, gas pycnometry will again be measured with will replace hydrostatic density the gas pycnometer and then determinations for plutonium. Nuclear Materials Technology 1994 Report 74 Waste Minimization Technologies Plutonium Dry Machining

Mark R. Miller Introduction Plutonium Metallurgy Reducing or eliminating hazardous wastes from plutonium machining processes at Los Alamos National Laboratory is critical to ensure continued operations under current and increasing legislative demands. In the past, Los Alamos machining processes used an argon and freon mist system to lubricate and cool the plutonium parts and to act as a fire inhibitor (plutonium is pyrophoric). Current U.S. legislation requires the elimination of all chloro-fluorocarbons, such as freon, by 1996; therefore, switching to plutonium machining methods that are environmentally friendly is crucial. Dry machining is one possibility because it does not require the use of coolants or oils, and, thus, mixed waste byproducts can be minimized or eliminated. The impact of this method is realized downstream as well, because the material recovery processes receive cleaner feed material. “Reducing or eliminating Dry machining tests at Los Alamos on various plutonium parts, hazardous wastes from which included features similar to weapon components, showed plutonium machining that plutonium can easily be dry machined, practically eliminating the need for cutting fluids. Results indicated a significant reduction processes at Los Alamos in the use of oils and solvents for research and development activi- National Laboratory is ties as well as for production processes. This reduction and eventual critical to ensure continued elimination of hazardous solvents and oils for plutonium machining processes will complement activities at the future plutonium pro- operations under current duction facility, Complex 21. and increasing legislative demands.” Background Metal machining has traditionally used cutting fluids to lubricate or cool parts at the cutting tool interface during the material removal process. These fluids have a number of benefits, such as reducing friction and wear (thus extending tool life and improving surface finish), cooling of the cutting zone, and removing metal chips from the immediate work area. The fluids are generally water-based oil, organic oils, or solvents. However, plutonium machining has been associated only with compatible organic oils or solvents, because water acts as a neutron reflector that decreases the amount of plutonium needed to sustain a nuclear chain reaction.

Cutting fluids can be used as lubricants or coolants depending on their applications. High-speed machining requires coolants to reduce heat that occurs from increased cutting speeds, whereas slower-speed machining requires lubricants to reduce material buildup on tooling and to improve surface finish. Past plutonium machining processes have used cutting fluids for both purposes, to lubricate and to cool parts.

75 Waste Minimization Technologies Plutonium Dry Machining

The plutonium shapes typically machined require that the cutting as at Rocky Flats. Freon usage at tool begin at the spindle center line and proceed outward to some Los Alamos was approximately specified diameter, or the reverse of this process. The operations 50 liters per year. Because freon are analogous to a facing operation performed from the center line is a chloro-fluorocarbon, its use outward. A direct relationship exists between the cutting speed and will soon be eliminated in the the radius of the part; as the radius (distance from the center line) United States. Clearly, alterna- increases during machining, the part surface speed increases tives to these environmentally in relation to the cutting tool.1 Therefore, close to the spindle center harmful practices at Rocky Flats line, a rotating plutonium part has very low surface speed and the and Los Alamos were desper- cutting process acts as a low-speed machining operation. As the ately needed. tool progresses outward, the surface speed of the part increases and the cutting process behaves more like a high-speed operation. This Rocky Flats investigated alterna- direct relationship between the cutting speed and the part radius is tive solvents for carbon tetra- expressed as chloride while Los Alamos searched for methods to elimi- Vc = pr ´ w , nate freon; however, Rocky Flats was shut down before an alter- where native could be found. The research performed at Los Alamos

VC = cutting speed (ft/min), revealed that by reducing the r = radius (ft), and oxygen content low enough in w = rotational speed (rpm). the glove box, plutonium could be dry machined, reducing the Plutonium is a material that must be processed in a controlled need for cutting fluids. Cur- environment in order to prevent plutonium contamination, limit rently, this process is applied radiation exposure to personnel, and reduce the amount of oxygen in approximately 85% of all present. Because plutonium is pyrophoric, heat created from pro- machining operations at cessing must be regulated, along with oxygen, to prevent igniting Los Alamos. The remaining the metal. Reducing oxygen content is also necessary to inhibit operations are carried out with oxide growth (plutonium reacts with oxygen to create plutonium a light, thin coat of oil applied oxide). Oxygen reduction within a glove box requires purging it to the surface of the material for with either nitrogen or argon. finish cuts.

Los Alamos and the Rocky Flats Plant each have used a different Methodology approach for cooling and lubricating the plutonium parts. Produc- Tests at Los Alamos were per- tion machining methods at Rocky Flats used oil as a cutting fluid formed as an engineering study that was applied by flood cooling. This method served not only to investigate the feasibility and to cool and lubricate the parts but also limited the amount of oxy- practicability of plutonium dry gen in contact with the material at the tool and part interface. machining for research and However, the oil, once in contact with plutonium, is considered a development purposes as well mixed waste and was difficult to dispose of properly. Additionally, as for production processes. the oil had to be cleaned from the surface of the parts, chips, and Experiments included upper equipment with large amounts of carbon tetrachloride, approxi- oxygen limit and plutonium mately 7800 gallons per year. This solvent, when contaminated machining tests. In addition, with plutonium, is considered a mixed hazardous waste and re- the formation of nitrogen and quires very strict disposal procedures. Research and development plutonium compounds were in- plutonium machining methods at Los Alamos used argon to propel vestigated in the current a freon mist onto the tool and part interface for the same reasons literature.

Nuclear Materials Technology 1994 Report 76 Waste Minimization Technologies Plutonium Dry Machining

Establishment of an upper oxy- At this oxygen level the burning boxes at Los Alamos to perform gen limit for dry machining was of the plutonium was concen- this inspection. However, experi- critical to the prevention of a trated in the turnings (metal enced inspectors estimated the plutonium fire. Two methods chips) as they fell from the part. surface finish to be as good as were used to test the oxygen As the turnings burned, they parts machined with cutting limit at which plutonium would quickly changed to Pu2O3 as the fluids. ignite under various machining heat caused by the friction of the conditions. The first machining cutting tool in combination with Test parts, along with production tests were performed on an en- the 4.5% oxygen created the parts, showed that, when facing gine lathe. A series of cuts were critical mix of plutonium, heat, operations are required, dry ma- taken on a 0.75-inch, delta-stabi- and oxygen needed for combus- chining plutonium can gall (or lized plutonium rod while the tion. Experience with pluto- tear) the material at the center oxygen content in the glove box nium machining and occasional line of the rotating part. Gener- was increased. Another series “fires” has shown that fires are ally, the size of this defect was of tests were performed in a fur- indeed associated with chips approximately 0.050 inches wide nace to verify the temperature as only. This concentration of heat by 0.000080 inches deep, and this well as oxygen content required energy, approximately 80%, defect appeared in approxi- for plutonium combustion. occurs because of the friction mately 30% of all parts that re- In both approaches, only small between the cutting tool rake quired this type of operation. amounts of plutonium were face and the chip, typically re- Although these plutonium parts used in order to prevent a large sulting in tool crater wear. pass inspection criteria, they do fire from occurring. Because maximum temperatures not meet the approval of the are located away from the cut- Los Alamos user. The physicists The machining investigations ting tool edge,2 the plutonium and engineers who design the were conducted on various part does not receive the fric- components for testing at the phases of plutonium, including tional heat energy that is carried Nevada Test Site require these delta-stabilized material. The away with the chips. The ex- components to surpass tests concentrated on turning perimental results, in combina- weapons-grade specifications, (the shaping process) and were tion with plutonium chip fire physically and visually. This performed on computerized nu- experience, were used to estab- anomaly may cause a part to be merical control T-base lathes. lish a maximum oxygen level rejected even though the part Various shapes were machined of 3% for dry machining of meets weapons-grade specifica- including component parts that plutonium parts. tions. However, a small amount were designed for, and tested at, of light oil swiped on the surface the Nevada Test Site. These full- Machining Tests eliminated any galling. size parts were of the type that would be manufactured for Results from the machining tests The positive test results for ma- weapons production. showed no detectable differ- chining convinced Los Alamos to ences in quality between parts adopt dry machining as the alter- Results machined without cutting fluids native to producing parts with an Upper Oxygen Limit Tests and those machined with fluids. argon/freon mist system for all The dry-machined parts were Nevada Test Site components Test results for the safe upper inspected on a AA-gage, a rotary and for related Los Alamos ma- oxygen limit during a pluto- contour inspection machine, that chining activities. About 85% of nium machining operation produced inspection data con- all plutonium components are showed that the glove-box sistent with data from parts ma- dry machined. The remaining oxygen level must increase to chined with cutting fluids. 15% require a light coat of oil to approximately 4.5% before plu- No quantitative surface finish eliminate the potential for micro- tonium ignition could occur. data could be gathered because inch tearing during the final cut. no equipment exists in glove

77 Waste Minimization Technologies Plutonium Dry Machining

The light coat of oil is brushed Additionally, no rapid formation The 20 cubic centimeters of oil on the part just prior to perform- of oxide has been observed on now introduced annually into ing the finish cut. The oil-con- dry-machined plutonium. the machining process easily taminated chips, kept separate Observations of dry-machined meets current regulations. from the dry-machined chips, parts have taken place over time are cleaned with a small amount periods of up to one year. Cer- Flood cooling techniques at of freon. The oil-free chips are tainly, if significant amounts of Rocky Flats created 2500 gallons then melt-consolidated under nitride were formed during dry of mixed waste oils each year calcium chloride to produce a machining a yellow-green oxide and required the use of approxi- plutonium button ready for fur- growth would quickly attack the mately 7800 gallons of carbon ther processing. outer surface and would be tetrachloride to clean the oil- readily visible. The empirical contaminated chips, machine Nitride Investigations and physical evidence clearly tools, and parts. Application of shows that the formation of plu- the dry machining techniques Exposing plutonium to a nitro- tonium nitride during the dry- outlined in this report (adjusted gen atmosphere at elevated tem- machining process is for the number of units ma- peratures for extended periods insignificant. chined at Rocky Flats) would of time can create a compound entirely eliminate hazardous called plutonium nitride. Pluto- (carcinogenic) carbon tetrachlo- nium nitride is extremely un- Waste Minimization and Other ride use, reduce the amount of stable in air and readily forms a Benefits mixed waste oils to approxi- hydrated oxide. Plutonium ox- Benefits from dry machining of mately 1.25 gallons per year, and ide can deteriorate the pluto- plutonium parts include im- produce approximately 50 gal- nium metal surface, reduce proved part quality, a signifi- lons of waste freon each year. density, and increase part cant decrease in chloro- Obviously, dry machining tech- weight. fluorocarbon releases, and sig- niques will complement the fu- nificant mixed waste reduction. ture plutonium production However, the temperatures and processes. time required to form even small Part quality at Los Alamos was amounts of a nitride compound enhanced by eliminating the Ongoing Research are quite high. Experimental re- freon mist that produced a local- The current trend in environ- sults have shown that 3% nitride ized cooling effect. This local- mental legislation clearly shows is formed at 625 °C after 16 that the small amount of oil and 3 ized cooling physically reduced hours of exposure to nitrogen. the part size in the immediate freon required for the dry ma- Estimates of temperatures at area of coolant application dur- chining process should be elimi- the tool part interface are signi- ing machining and thus pro- nated. Plutonium machining ficantly lower, approximately duced out-of-tolerance research and development ef- 300 – 400 °C, while cutting tool conditions. forts are addressing this issue. contact times are approximately Efforts include the investigation 2,4 microseconds. Therefore, Currently, plutonium dry ma- of diamond turning techniques even at 400 °C the creation of chining activities at Los Alamos to increase the cutting tool edge any significant amount of pluto- use approximately one liter of quality and to reduce friction on nium nitride is highly improb- freon annually, as contrasted the rake face. These performance able because the nitride does not with the 50 liters required for improvements should produce a have adequate time to form. the argon/freon mist system.

Nuclear Materials Technology 1994 Report 78 Waste Minimization Technologies Plutonium Dry Machining better surface finish as the tool approaches the center line of a rotating part. Changing spindle speeds and cutting tool feed rates should improve the surface finish in a slow-speed cutting operation. Alternative cutting fluids are also being explored, and this search has lead to a solvent called SF2I, developed by the 3-M Company. This solvent, originally tested at the Atomic Weapons Establishment in Aldermasten, England, shows promising results for plutonium machining operations. It will be further tested at Los Alamos for application as a coolant for machining operations, such as saw- ing, drilling, and milling.¨ Cited References 1. E. Oberg, F. D. Jones, and H.L. Horton, Machinery’s Handbook, 22nd ed. (Industrial Press, Inc., New York, 1987), p. 1765.

2. S. Kalpakjian, Manufacturing Processes for Engineering Materials (Addison-Wesley, Reading, Massa- chusetts, 1991), pp. 499-500.

3. F. Brown, H. M. Ockenden, and G. A. Welch, “The Preparation and Properties of Some Plutonium Compounds Part II,” J. Chem. Soc. 1955, 4196-4201 (1955).

4. F. Kreith, Principles of Heat Transfer (Harper and Row, New York, 1973), p. 81.

79 Waste Minimization Technologies Technical Issues in Plutonium Storage

John M. Haschke Introduction and Joseph C. Martz The retirement of large numbers of nuclear weapons necessitates the Plutonium Metallurgy management of an unprecedented quantity of excess plutonium. Sur- plus materials from production and reprocessing operations will add to the amount of concentrated plutonium that must be stored on an interim basis until a decision regarding ultimate disposition is made. Traditionally, the weapons complex faced a scarcity of plutonium. Thus, the future course is unfamiliar. An evaluation of technical issues is needed to develop a basis for storage decisions. Identification of acceptable storage options is a prerequisite to the definition of standards, procedures, and facilities for storage. “In this report, an assess- The selection of suitable storage options is a complex task impacted by political and economic concerns as well as technical issues. ment of issues relevant to Many factors influence decisions on plutonium storage; however, plutonium storage is pre- emphasis is often placed on a narrow set of issues, such as per- sented in order to provide a ceived proliferation resistance or some specific hazard. A comprehensive examination should include topics such as the potential for technical basis for address- remediation of hazards, the maintenance of flexibility for ultimate ing complex interfaces with disposition, and the cost of implementation. political and economic A consideration of cost must extend beyond the initial expenditures issues and for identifying for research, development, and capital investment in facilities to suitable storage options other factors that contribute to the total expense of the operation. A thorough assessment must include not only the cost of addressing for excess plutonium.” radiation exposure and contamination control but also the cost of other environmental, safety, and health (ES&H) issues. For example, disposal of radioactive waste constitutes a major and increasingly larger fraction of the total operating cost of a plutonium facility. The amount of waste that must be considered includes that produced during normal facility operation as well as that generated during facility modification and decommissioning.

In this report, an assessment of issues relevant to plutonium storage is presented in order to provide a technical basis for addressing complex interfaces with political and economic issues and for identifying suitable storage options for excess plutonium. An effort is made to summarize the merits of alternative approaches and to establish a qualitative ranking of options for an interim storage period of up to one hundred years. This work focuses on plutonium extracted from weapons. It is important to note that this source represents the mi- nority of available plutonium. Considerable quantities of reactor- grade plutonium are found throughout the world and represent a significant environmental and proliferation risk. Narrow solutions that only address material extracted from weapons may fail in the larger context of a worldwide plutonium economy.

Nuclear Materials Technology 1994 Report 80 Waste Minimization Technologies Technical Issues in Plutonium Storage

Technical Issues Credible interim storage options form and then reconfigured for Consideration of a broad range must be technically feasible and preparing material for reuse or of issues is necessary for assess- capable of near-term implemen- final disposition. Finally, a real- ing plutonium storage options. tation. Even if an option is tech- istic assessment must include A summary of the major issues nically feasible, it may be decontamination and decom- is included to provide a compre- impractical due to excessive missioning of interim processes hensive perspective. costs or limited availability of as well as installation, operation, key resources or material. Evalu- and decommissioning of any Proliferation of nuclear weapons ation of a storage option is hin- subsequent facilities. The im- capability by major nuclear dered by lack of knowledge pact of such operations on cost, states, aspiring nations, or about material properties, haz- including waste generation dur- subnational groups presents a ards, or process operations. ing the entire cycle, must be serious threat to world order Previous experience with a weighed against that of storing and stability. The selection of given option is particularly material in its existing form until storage options is complicated valuable and is given appropri- specifications for final disposi- by political pressures demand- ate consideration in this assess- tion are defined. ing that a technical remedy be ment. found to prevent proliferation. The issue of special nuclear ma- The level of interdiction re- Storage options that change the terial (SNM) control also merits quired to prevent an existing chemical form of the plutonium examination. In addition to ex- nuclear state from rapidly recon- have especially far-reaching im- isting standards for accountabil- structing a nuclear arsenal from pacts on the total cost. In all like- ity and verification of plutonium intact pits differs from that re- lihood, the interim storage form inventory, increased transpar- quired to prevent a subnational will not be identical to that cho- ency requirements are antici- group from fabricating a nuclear sen for final disposition. Addi- pated during an extended device from diverted plutonium. tional processing may ultimately storage period. Certain options The goal is simple: select a be necessary. Interim conversion are more amenable to account- physical or chemical storage from one form to another ampli- ability than others because the form that cannot be directly fies this problem, particularly if chemical form of the stored ma- used in fabricating a fission the form is significantly diluted. terial is homogeneous. Whereas weapon and that cannot be The advantage of avoiding plutonium metal and com- processed into a usable form. chemical processing is clearly pounds formed by gas-solid re- Unfortunately, there is no tech- reflected in the cost because a actions tend to be homogeneous, nical solution that achieves this unique facility is usually re- compounds prepared by con- result. quired for each process. Such densed-phase reactions or py- facilities are necessary because rolysis of precipitates are ES&H issues encompass a spec- most chemical processes are cy- frequently inhomogeneous and trum of issues that extend well clic instead of reversible. For ex- thus more difficult to inventory beyond identification of hazards ample, the technologies and and verify. Such assay and mea- and waste streams. A credible process equipment used for con- surement is particularly difficult evaluation must examine pos- verting metal to oxide are differ- for dilute plutonium forms, es- sible methods of hazard ent from those for reducing pecially those mixed with other remediation and weigh the ad- oxide to metal. The processing radioactive elements. vantages of their implementa- facility must first be configured tion against the disadvantages. for preparing the desired storage An important means for minimizing ES&H risk is the avoid- ance of all unnecessary process operations.

81 Waste Minimization Technologies Technical Issues in Plutonium Storage

A final consideration involves Storage as intact pits has distinct Fabrication of the warhead has contingency planning and pres- advantages for all technical is- little if any impact on the time ervation of flexibility for final sues. Hazards and ES&H risks required to reconstitute a large disposition. Certain physical are minimal. The pit is a well- stockpile. In fact, this very ob- and chemical forms are more certified containment vessel, and servation is reflected in the nu- suitable for a rapid, low-cost re- feasibility is certain and sup- merous arms control treaties sponse than others. Maximum ported by an extensive and ex- signed by the United States flexibility is achieved by storing cellent history of storage in the and the former Soviet Union. a form that requires minimum weapons stockpile. The desir- START I, START II, and the INF processing for interim storage, ability of intact pit storage is re- treaties all address the disposi- but this ability to respond stands flected in the comments of a tion of delivery vehicles but in conflict with a desire for in- long-time production worker have no provisions or protocols creased proliferation resistance. from the Rocky Flat’s Plant: for destruction of actual nuclear Existence of the necessary tech- “We don’t build triggers for warheads. nology base for processing and nuclear weapons. We build the verification within Complex 21 safest, most certified, best char- Storage of plutonium as intact are important factors. acterized storage containers for pits is preferred by the scientific plutonium known to man.” community over all other The options examined for stor- This option preserves maximum options under consideration. age of excess plutonium fall into flexibility and is the most cost- Nonetheless, the political cost three categories: effective option due to limited and perceptions associated with 1) storage as intact pits, facility requirements and the storage of intact pits may over- 2) storage as altered pits, and absence of radioactive waste ride the technical issues and pre- 3) storage as extracted generation. vent its ultimate implementation. material. An assessment of each category In terms of proliferation resis- Altered Pits is presented below. tance, pit storage is often consid- Pit alteration is an approach that ered highly undesirable because focuses on retaining advanta- Intact Pits of the possibility for diversion geous pit features such as Storage of plutonium as intact and use in a nuclear device. nuclear material containment pits is an approach that employs However, the potential for utiliz- while destroying the suitability the intrinsic characteristics that ing diverted plutonium from for direct reuse. Two levels of made the pit suitable for ex- other storage configurations is alteration are proposed: tended storage in a weapon con- comparable to that for an intact 1) modification of nuclear figuration. These features pit. The availability of intact geometries by mechanical include a critically safe pluto- pits for reconstruction of a large or physical methods, and nium geometry, a known arsenal by a major nuclear state 2) modification of nuclear nuclear material inventory, and may not be of concern because geometries and material a durable, hermetically sealed the production rate of weapon properties by chemical containment vessel with certi- systems is influenced by opera- methods. fied closures. A plutonium tions other than pit fabrication. Whereas mechanical and physi- facility is not needed for Indeed, experts in nonprolifera- cal methods include deformation implementation of this option tion and arms control point to and insertion of inert materials but is required for surveillance the fabrication of delivery to prevent direct reuse in weap- and disassembly of pits that are systems as the rate-limiting ons fabrication, chemical meth- identified as unsuitable for step in the reconstruction of a ods are designed to hinder or continued storage. superpower’s nuclear arsenal. deny reuse by also forming plutonium compounds that are difficult to reconstitute to a usable form. Suggested chemical methods include in situ formation of a product such as hydride, oxide, nitride, or alloy. Nuclear Materials Technology 1994 Report 82 Waste Minimization Technologies Technical Issues in Plutonium Storage

A detailed discussion of specific Verification of pit disablement The thermodynamic and kinetic methods is beyond the scope of could be accomplished by a complications cited for in situ this paper. number of nonintrusive meth- alteration are eliminated by flex- ods that would not compromise ibility in the design of process Alteration methods may be ca- classified information. equipment. As with chemical pable of denying the rapid re- alteration, the desirability of construction of an arsenal under Risk and cost are significantly storing selected material forms very specific scenarios, but the increased by use of chemical resides in the potential for pre- utility of alteration as a means methods that involve injection venting reuse and/or reprocess- of preventing proliferation by of reactive solids, liquids, or ing of plutonium. Placement of nonnuclear states and terrorist gases into the pit. Useful reac- plutonium in a dilute form such groups is doubtful. To varying tions must be complete in a rea- as a fused silica matrix or as a degrees, mechanical, physical, sonable time period to maintain mixture containing high-level and chemical alteration methods process throughput. Difficulties radioactive waste does not pre- only extend the length of time arise because of thermodynamic vent its recovery by a motivated between diversion and reconsti- and kinetic factors. In some group but would significantly tution of material into a usable cases, the reactions are highly hinder further United States form. Estimates of the time re- exothermic (for example, processing because of stringent quired for such reconstitution DH°f of PuO2 = -252 kcal/mol), ES&H concerns. A decision to have been made, but in no case and the likelihood of excessive store plutonium in such forms do they provide a significant temperatures would require is essentially equivalent to se- barrier to extraction and purifi- control of processing rate and lecting an option for final dispo- cation of nuclear material. The heat flow to prevent thermally sition and must be carefully technologies required to reform induced failure of pits as con- considered as an option for in- metal and to reprocess sug- tainment vessels. Other reac- terim storage. gested chemical forms into tions proceed at prohibitively metal are well known and can slow rates, and excessively high A qualitative assessment of be implemented at low cost, es- temperatures would be required issues for candidate storage pecially if ES&H risks are disre- to promote those processes. forms of extracted plutonium garded. None of the proposed chemical is presented in Table I. In addi- reactions has an appropriate rate tion to the plutonium com- Although pit alteration methods at room temperature, and none pounds mentioned in the may provide a small measure of of the methods is sufficiently de- discussion of chemical alter- proliferation resistance (essen- veloped to permit a thorough ation, the candidate storage tially in the political message assessment of hazards, assess- forms include plutonium metal, they might send), this gain must ment of stability during ex- plutonium dicarbide, and a be weighed against the disad- tended storage, identification of mixed plutonium-uranium ox- vantages. As noted previously, the difficulties in reprocessing ide. Results of the evaluation an important advantage of pit for final disposition, or estima- are particularly significant; alteration is retention of the pit tion of waste generation levels. none of the candidate forms as a storage vessel; however, prevent proliferation, and differ- the certified integrity of the Extracted Material ences in time periods required container is destroyed by all Extraction of plutonium de- for reprocessing are small. methods requiring access to stroys the pit configuration and the pit interior. The potential permits a broad range of candi- for release of nuclear materials date forms to be considered for during those operations will storage. The absence of classi- undoubtedly require use of a fied shapes facilitates transpar- plutonium-handling facility ency in storage, but verification and implementation of a certifi- of material transfer during ex- cation procedure for reclosure. traction remains a concern.

83 Waste Minimization Technologies Technical Issues in Plutonium Storage

Table I. Qualitative Assessment of Issues for Candidate Storage Forms of Extracted Plutonium

Candidate Proliferation Potential Remediation Anticipated Ease of SNM Storage Flexibility Forms Resistant Hazards Available Total Cost Accountability Experience

metal no yes yes low high extensive high oxide no yes yes moderate good high reduced mixed oxide no yes yes high reduced modest low dicarbide no yes (?) no (?) high very low limited reduced nitrite no yes (?) no (?) high low limited reduced hydride no yes* no low low limited high alloys no yes (?) no (?) high very low none low

(?) Insufficient data is available. *Candidate form is especially hazardous because of pyrophoricity.

Since all candidate forms have Results in Table I suggest that from literature sources shows potential ES&H hazards, the se- two storage forms, metal and ox- that massive (greater than 0.2- lection of suitable storage forms ide, are the favored candidates mm-thick) plutonium becomes hinges on the remaining issues for interim storage. The remain- pyrophoric only at temperatures in Table I. The deciding factors ing discussion of extracted pluto- near 500 °C.2 Plutonium fines are availability of methods for nium focuses on these materials. and foils ignite when heated to hazard remediation, total cost, 150–200 °C and are not suitable and storage experience. The Plutonium metal is an accepta- for storage. Oxidation of pluto- assessment of cost includes ble form for interim storage. nium is prevented by storage the loss of flexibility and the Although metal is often viewed in sealed metal containers with economic impact of first process- as a highly undesirable storage certified closures. ing for storage and then repro- form because of proliferation cessing for final disposition. concerns, the technical assess- The remaining potential hazards Cost is driven by the need to ment indicates that its prolifera- identified for storage of pluto- develop both processing and tion resistance is comparable to nium metal do not present seri- reprocessing technologies, to that of oxide. Potential hazards ous ES&H risks. Metal is not a install and replace process facili- identified for metal storage dispersible form of plutonium ties, and to dispose of associated include chemical reactivity, and must be oxidized before radioactive waste. The low as- dispersal of radioactive contami- entrainment is possible. The oxi- sessment of flexibility for mixed nation, and pressurization of the dation rate at room temperature oxide and alloys arises because storage container. The primary is such that dispersible particles reprocessing involves dissolu- hazard is the chemical reactivity are released from clean metal tion in aqueous media and sub- of metal in air and the potential only after several days in moist sequent isolation of plutonium for pyrophoric ignition. air at room temperature.3 Above by ion exchange or solvent 500 °C, air oxidation of a typical extraction. Other candidate The oxidation rate of plutonium 4-kg metal casting by the con- forms are directly recovered metal is modest and controlled stant-rate process is complete by pyrochemical methods. by a protective oxide layer at after several hours, but less than temperatures up to 450 °C. The 0.1 mass percent of the oxide oxidation rate in air is constant product exists as particles with 2 (0.2 g PuO2/cm /min) at tem- sizes in the dispersible range be- peratures above 500 °C.1 A low 10-µm geometric diameter. recent analysis of ignition data

Nuclear Materials Technology 1994 Report 84 Waste Minimization Technologies Technical Issues in Plutonium Storage

Hydrogen and other gases readily reconstituted by meth- rate of helium formation from formed by alpha radiolysis of ods such as direct oxide reduc- decay of plutonium-239 is well plastics and organic materials tion (DOR) using commercially defined.8 This decay product, in the storage environment are available chemicals and equip- which is retained as bubbles in not a hazardous pressurization ment. Plutonium dioxide is gen- massive plutonium metal and source because they are gettered erally considered to be a stable released from oxide powder, in- by reaction with plutonium. and inert material, but concerns creases the internal pressure of The products of these reactions arise because of the high surface an oxide container by about 1 frequently include plutonium area and tenacious adsorption of atm over a hundred-year period. hydride and sesquioxide that molecular species including wa- An additional pressurization present pyrophoricity hazards ter, organic compounds, and at- hazard encountered during upon exposure to air. Rupture mospheric constituents. Process short-term storage and transpor- of a storage container is possible oxides formed by pyrolysis of tation arises from desorption of if it is not hermetically sealed precipitates such as oxalate, ni- molecular species at elevated and pressure is exerted by ex- trate, or peroxide have specific temperatures created by self- pansion of the oxide product. surface areas from 20 to 60 m2 heating or fire.3 Use of vented per gram4 and chemisorb sev- storage containers provides a Assessments of the remaining eral mass percent of water.5 The method for releasing pressure issues for metal storage are removal of adsorbates including during storage, but radiolysis of favorable. Experience gained water is accomplished by heat- air produces high concentrations from storage of thousands of ing the oxide at 1000 °C.6 of gaseous nitrogen oxides that pits defines the conditions re- are highly corrosive in the pres- quired for extended storage of Potential pressurization hazards ence of moisture. metal and demonstrates that during oxide storage arise from such storage is remarkably free gases generated by radiolytic Assessment of hazards associ- of problems. SNM accountabil- and chemical reactions. Alpha- ated with plutonium storage ity is facile, maximum flexibility induced decomposition of ad- indicates that the greatest envi- is preserved, and minimal waste sorbates and organic materials ronmental risk arises from dis- is generated by storage of metal. forms gaseous products includ- persal of process oxide The cost of a processing facility ing H2, CO, and HCl. A single following a container rupture is minimal because metal stor- monolayer of water, which can or a spill.5 This conclusion is age is compatible with the be readily readsorbed on an ox- based on particle size data pyrochemical technologies of ide surface if adequate precau- showing that 100 mass percent the Complex 21 baseline. If any tions are not taken after firing, of a process oxide is in the dis- extracted form (metal, oxide, or is capable of generating 5 to 10 persible range. The dispersal other) is selected for storage, the atm of pressure in a typical risk is a thousand times greater throughput capacity of the metal storage container over time.3 for process oxide than for the extraction facility must be ap- Although the dioxide is gener- worst-case incident with metal. propriate to process the inven- ally considered to be thermody- Possible methods of remediation tory of retired pits in a timely namically stable, a recent study include sintering of oxide pow- manner that cannot exceed the shows that the dioxide reacts ders to increase the effective projected lifetime of the facility. with water to form H2 and particle size and hot-pressing of PuO2.2, a compound containing oxide into briquettes; however, Plutonium dioxide is also an Pu(VI).7 Although the kinetics the cost and waste consequences acceptable storage form. The of these radiolytic and chemical of these operations must be care- proliferation risk of oxide is con- processes are unknown and can- fully evaluated. sidered equivalent to that of plu- not be accurately predicted, the tonium because the metal is

85 Waste Minimization Technologies Technical Issues in Plutonium Storage

Remaining issues related to ox- The value of chemical alteration A comprehensive strategy must ide storage include nuclear ma- of pits and reaction of extracted also consider forms of pluto- terials accountability, storage plutonium to form proliferation- nium that are not part of the experience, and flexibility. To resistant material forms is ques- present assessment. Such mate- varying degrees each of these tionable because reconstitution rials include incinerator ash, issues impacts the others and of candidate materials to usable pyrochemical salts, anode heels, the cost. Variations from 1 to 10 forms is facile. Furthermore, etc., that are presently in storage mass percent are observed in the most candidate forms could at several DOE sites. The situa- plutonium content of process probably be used for construct- tion is extremely complex and oxides because of differences in ing a nuclear device without be- application of general guidelines the amount of unpyrolyzed an- ing reconstituted. Adequate may not be appropriate. Further ions remaining after firing.4 safeguards and security provide evaluation of these areas is Although such complications the only effective method for clearly needed. with material accountability are preventing diversion and reuse well known, storage experience of nuclear materials. The present technical assess- for oxide is limited and proce- ment provides a basis for gen- dures for outgassing and certify- The formulation of a compre- eral recommendations on ing oxides prior to storage are hensive strategy is essential for interim storage of plutonium. poorly defined. If oxide is achieving the goal of plutonium Both metal and oxide are accept- adopted as the only storage processing with minimal waste. able storage forms. Storage of form for extracted plutonium, Storage of excess plutonium metal as intact pits is preferred. the processing facility must be must be included in that strat- In the absence of a suitable plu- appropriately sized and include egy. Development and imple- tonium handling facility, intact an additional capability for effi- mentation of advanced storage is the only near-term op- ciently converting extracted technologies that reduce the tion. If interim storage of intact metal to oxide. The aqueous re- quantities of radioactive waste pits is deemed unacceptable, covery technologies in the Com- produced and those that provide storage as extracted metal is pre- plex 21 baseline produce oxide capability for decontaminating ferred. Storage as altered pits is but are designed for treating waste are essential components the least desirable option. If al- residues and pyrochemical salts, of the strategy. The maximum teration of pits is necessary, only not for oxidizing large quantities impact on waste generation may mechanical or physical methods of metal. If only oxide is stored, be realized by applying a simple that preserve the integrity of the potential impact on cost and intuitive approach: avoid all un- containment are acceptable. waste will be large. necessary processing. Conse- Metal and process oxide should quently, plutonium that is to be be stored in their existing forms Conclusions placed in interim storage should after preparation and certifica- The selection of an option for remain in its existing form; re- tion for storage. All storage interim plutonium storage on processing and interconversion vessels should be hermetically the basis of proliferation resis- of chemical forms should be sealed and should exclude or- tance cannot be justified on a avoided unless a potentially ganic and other covalently technical basis. Alteration mea- hazardous situation exists. bound materials. Regardless of sures are considered effective the storage option(s) employed, in forcing a nuclear state to adequate and continuing sur- refabricate pits before recon- veillance will be necessary to structing a large arsenal, but that identify potential problems. delay may not limit the rate at which weapons are produced.

Nuclear Materials Technology 1994 Report 86 Waste Minimization Technologies Technical Issues in Plutonium Storage

Several technology areas need Cited References 7. J. L. Stakebake, D. T. Larson, and further definition and develop- 1. J. M. Haschke, “Evaluation of J. M. Haschke, “Characterization of ment. Continuation of efforts to Source-Term Data for Plutonium the Plutonium-Water Reaction Part upgrade process technologies for Aerosolization,” Los Alamos II: Formation of a Binary Oxide efficiency and waste minimiza- National Laboratory report LA- Containing Pu(VI),” J. of Alloys and tion are essential. Additional 12315-MS (July 1992). Compounds 202, 251 (1993). work on the kinetic and equilibrium behavior of oxide adsorp- 2. J. C. Martz, J. M. Haschke, and 8. J. C. Martz, “Analysis of Pluto- tion is needed with emphasis J. L. Stakebake, “A Mechanism for nium Storage Pressure Rise,” in on the conditions required for Plutonium Pyrophoricity,” Los NMT-5:92-328, July 1992, Los preparing, certifying, and pack- Alamos National Laboratory report Alamos National Laboratory aging material for storage. LA-UR-93-2655, submitted for memo. Studies to define the chemical publication, July 1993. and radiolytic reactions between water and oxide are needed to 3. J. M. Haschke and J. C. Martz, predict long-term storage behav- “Metal-Oxide Chemistry and ior. The merits of removing am- Storage,” Los Alamos National ericium from extracted material Laboratory report LA-CP-93-159 prior to storage also requires (May 1993). evaluation. The storage of plutonium-containing residues 4. J. D. Moseley and R. O. Wing, needs to be addressed, and the “Properties of Plutonium Dioxide,” consistency of storage decisions Dow Chemical Company report and Complex 21 design must RFP-503 (August 1965). be monitored. 5. D. Y. Chung, et al., “Assessment In addition to providing a tech- of Plutonium Storage Safety Issues nical basis for selection of stor- at Department of Energy Facilities,” age options, the present study Department of Energy draft report, helps to place waste issues and Office of Defense Programs, the role of technology develop- Germantown, MD, September 1993. ment in perspective. Although development of new technolo- 6. J. L. Stakebake and L. M. Stew- gies for minimization and pro- ard, J. Colloid Interface Sci. 42, 581 cessing of radioactive waste is (1973) . necessary, development of technical input for policy decisions is also essential to ensure that the objectives are consistent and complementary.✦

87 Facility Operations Waste Management at Technical Area 55: A Historical Perspective and Status Report Bill J. McKerley, Introduction James J. Balkey, In August 1992, the transuranic (TRU) waste management opera- and Ronald E. Wieneke tions for the Nuclear Materials Technology (NMT) Division were Nuclear Materials Processing: transferred into the Nitrate Systems Group (NMT-2), one of nine Nitrate Systems operating units within the Division. In February 1993, all waste management activities at Technical Area 55 (TA-55) were consolidated. The areas of immediate concern were backlogged legacy waste items (low-level, TRU, and hazardous) stored in the basement of the Los Alamos Plutonium Facility in TA-55, cemented TRU waste drums stored at an off-site location, experimental radioactively contaminated vessels stored on the Plutonium Facility grounds, and oversized waste items stored in the Plutonium Facil- ity basement. Communication was initiated with the Laboratory’s Waste Management Group (CST-7) to discuss options for resolving these issues. This report describes the technical, regulatory, and administrative actions and accomplishments that allowed NMT Division to effectively address these important issues. Activities in the pursuance of waste minimization goals are also reviewed. “The areas of immediate concern were backlogged Background Waste management at the Plutonium Facility was one function of legacy waste items (low- the old Nuclear Materials Management Group (NMT-7). The group level, TRU, and hazardous) was responsible for such diverse activities as shipping and receiving, nuclear material storage, roasting and blending of reactor fuels, stored in the basement of the and the development of a site-wide nuclear materials model (in Los Alamos Plutonium support of Complex 21 planning) to forecast the impact of new Facility at TA-55, cemented technologies on scrap inventories and waste generation. With limited personnel resources and diverse operations, group staff TRU waste drums stored at could not keep up with the steadily increasing requirements and an off-site location, experi- demands of waste management regulations. Likewise, personnel resources were not available to assign additional technicians to mental radioactively con- waste management activities as needed to address the resulting taminated vessels stored in operational demands and problems. Thus, waste management the Plutonium Facility personnel were faced with a variety of seemingly insoluble problems without the resources to explore and implement solutions. yard, and oversized waste items stored in the Pluto- Waste Management Issues nium Facility basement.” A variety of difficulties in several operations had developed in the waste management program at the Plutonium Facility. These problems, outlined below, threatened waste certification and shipment of TRU, mixed, and low-level waste to the Laboratory’s interim storage facility and low-level waste disposal site, TA-54.

Nuclear Materials Technology 1994 Report 88 Facility Operations Waste Management at Technical Area 55: A Historical Perspective and Status Report

Water Release from Cemented Lack of Suitable Nondestructive Accumulation of Oversized TRU TRU Waste. In June of 1989, Assay Capabilities for 235U Waste. An accumulation of free liquid was found on the Waste. Another complex issue oversized waste items (bigger top of a drum of cemented TRU involved the nondestructive than would fit in 55-gallon waste in the basement of the assay of waste packages drums), requiring either size Plutonium Facility. Analysis containing uranium-235 as a reduction or packaging in stan- indicated that this liquid co-contaminant with plutonium dard waste boxes, filled storage worked its way up through the from mixed oxide fuel process- areas in the basement of the Plu- carbon filter from the inside of ing operations. At the time of tonium Facility. Shortage of per- the drum. This clear violation this waste generation, no suit- sonnel available to support the of the Waste Acceptance Criteria able nondestructive assay capa- TRU oversized waste operation for the Waste Isolation Pilot bility existed at the Plutonium also contributed to this waste Plant (WIPP), criteria that bar Facility. But the Safeguards problem. free liquids from waste pack- Assay Group (N-1), who is the ages, immediately brought into vanguard of nuclear material Accumulation of Contaminated question the integrity of the 350 detection technology develop- Experimental Vessels. A collec- waste drums that had been im- ment at the Laboratory, assayed tion of radioactively contami- mobilized using the gypsum about two dozen of these drums nated experimental vessels cement process. for uranium-235 content. This dating back to the 1960s was be- operation was outside of the ing laboriously decontaminated Prohibitive Packaging Limits domain of the Laboratory TRU by a team of NMT-2 technicians for 238Pu Waste. Transportation waste program and, conse- and was accumulating in the safety requires that materials quently, outside of the integral yard of TA-55. Methods of pre- moved by highway, in this case quality assurance program. paring the vessels for transpor- TRU wastes transported in the This technicality had to be tation and storage at TA-54 had TRUPACT II overpack (Depart- resolved before the uranium not been established with the ment of Transportation Type B Special Nuclear Material (SNM) CST-7 Waste Management shipping package), must not assays could be judged accept- Group. Additional requirements contain flammable or explosive able and waste could be certified that had to be met to transfer the mixtures of gasses. A primary for storage at TA-54. vessels out of the Plutonium source of flammable gas in TRU Facility needed to be negotiated waste is hydrogen from alpha Nonconforming 239Pu Waste as well. radiolysis of hydrogenous mate- Drums. An inventory of two rials in the waste matrix. Pluto- dozen nonconforming pluto- Accumulation of Contaminated nium-238 has the highest spe- nium-239 waste drums (due to Glove Boxes. For four years, cific activity (which is directly either SNM or weight compari- glove boxes from decontamina- proportional to the gas genera- son discrepancies) had accumu- tion and decommissioning tion rate) of the plutonium iso- lated in basement storage areas operations in the Plutonium topes handled at TA-55 and, of the Plutonium Facility. The Facility had been crated in ply- consequently, has the lowest particular discrepancies could wood crates and moved out into allowable concentration limits not be investigated and resolved the TA-55 yard for interim stor- in WIPP waste packages. Pack- due to limited staffing and the age. As in the case of the experi- aged TRU wastes, especially pressures of keeping up with mental vessels, additional those in multilayer packaging the steady stream of waste gen- documentation or preparation configurations, did not meet the erated from plutonium process- that would be required to move Waste Acceptance Criteria for ing operations. the glove boxes off-site had not WIPP and, therefore, were not been established. certifiable or suitable for storage at TA-54.

89 Facility Operations Waste Management at Technical Area 55: A Historical Perspective and Status Report

The destination of the glove hazardous waste issues) at DOE Responsibility for the Plutonium boxes would be the TA-50 Size Weapons Complex sites; there- Facility basement storage and Reduction Facility where they fore, a commitment to draft a low-level waste activities and would be cut into pieces and Federal Facility Compliance the TA-55 yard storage areas consequently repackaged in Agreement (FFCA) was made. was transferred to the NMT-8 acceptable waste containers The DOE also made an agree- Facilities Management Group, (55-gallon drums or standard ment with the EPA to cease since these functions matched waste boxes). Water had accu- generating hazardous and, con- NMT-8’s mission better. And mulated in some boxes from sequently, mixed wastes from all the shipping and receiving op- prolonged exposure to the ele- defense operations. This deci- erations were transferred to the ments and would have to be sion effectively shut down most NMT-4 Nuclear Materials Mea- analyzed and removed from radiochemical processes at the surements and Accountability the packages before shipment. Plutonium Facility as of May 8, Group. 1992. The moratorium on mixed Accumulation of Low-Level waste generation presented both Not all aspects of the reorganiza- Oversized Waste. Low-level a challenge to and an opportu- tion were positive. The dissolu- oversized waste, packaged in nity for the TA-55 mission. At a tion of the NMT-7 Nuclear 4-ft x 4-ft x 8-ft plywood boxes, time when waste management Materials Management Group had likewise accumulated in the needed manpower to resolve resulted in time-consuming at- TA-55 yard, totaling thirty-six growing waste problems, pro- tempts to coordinate interrelated boxes. These waste packages cessing operations were cur- Facility waste management op- added to the congestion of mate- tailed, resulting in idled erations that were now located rials in the yard and occupied personnel that, potentially, in separate Laboratory host valuable space. could be reassigned to address groups. The reorganization also this need. resulted in the loss of experi- Thus, waste items were stalling enced waste management per- in the “pipeline” that processed, Group Reorganization sonnel; out of thirty personnel packaged, and funneled waste Division resources were imme- originally in the NMT-7 Group, items to the waste management diately put to work on waste approximately half chose to storage and disposal site at management problems at TA-55. transfer to other types of posi- TA-54. As a result of these diffi- The TRU liquid and solid waste tions rather than move with the culties, waste operations at operations were transferred to waste management functions TA-55 were coming to a halt the NMT-2 Nitrate Systems assigned to other NMT groups. and were threatening the re- Group and placed under the Ni- New technicians and staff hired search programs that generated trate Processing Team Leaders. to replace them required site- the waste. Resources had to be This decision was based upon and operation-specific training reallocated immediately to alle- the fact that NMT-2 produced before assuming their duties. viate this situation. most of the waste being processed, and nitrate operations Waste Management Solutions New Regulatory Setting personnel had intimate knowl- The concentration of NMT-2 re- In the spring of 1992 both the edge of the waste produced. sources on waste management Department of Energy (DOE) NMT-2 group management also problems at TA-55 over the past and the Environmental Protec- had a no-nonsense reputation year and a half has yielded dra- tion Agency (EPA) realized that for defining and solving prob- matic results. Waste manage- an agreement must be reached lems; furthermore, many of the ment operations are staffing up concerning Resource Conserva- personnel idled by the morato- to realistic levels to handle the tion and Recovery Act (RCRA) rium on mixed waste generation workload, and personnel are be- compliance (which dealt with were NMT-2 employees. ing certified in their operations.

Nuclear Materials Technology 1994 Report 90 Facility Operations Waste Management at Technical Area 55: A Historical Perspective and Status Report

Dialog has been reestablished Group for a waiting period of Safe Storage Limits for 238Pu with the Laboratory waste man- 84 weeks before shipment to Waste Drums. The NMT-9 Heat agement and environmental TA-54, the approximate time for Source Technology Group has compliance organizations, pro- the cement curing process based been conducting a research effort viding a means of resolving upon earlier research by NMT to determine the safe quantity of problems in a mutually agree- personnel. Waste management plutonium-238 allowed in waste able manner. And legacy waste personnel would then visually packages. They have been re- issues are being resolved: old confirm the absence of free liq- searching existing literature, wastes are being remediated and uids before shipment to TA-54. conducting experiments under sent to TA-54 for storage, and Any packages in which moisture controlled conditions, and doing procedures are being modified was observed or that could not headspace gas sampling (with to avoid similar future prob- be inspected (such as drums the cooperation of NMT-2 waste lems. Solutions to the waste containing one-gallon cans) management personnel) on management problems outlined would require overpacking. legacy waste packages in the earlier in this paper are given The drums stored in the CMR basement of the Plutonium Facil- below. building were visually exam- ity. Work that involves CST-1 ined where possible and then Analytical Chemistry and CST-7 Repackaging of Cemented TRU overpacked in polyethylene- Waste Management and Bench- Waste Drums. Drums of ce- lined 83-gallon drums, and the mark will establish a reasonable mented TRU waste suspected annulus was filled with absor- storage limit that will ensure safe of containing free liquid were bent materials. CMR repackag- storage of plutonium-238 con- returned to TA-55 (point of gen- ing operations were conducted taining waste but that will exceed eration) from interim storage at in September and October 1993; conservative TRUPACT-II Con- TA-54. Because there was not shipment of all 218 of the CMR tent Code (TRUCON) limits for enough room for the returning drums was completed the first shipment to WIPP. Storage re- drums in the Plutonium Facility week of October. To date, 107 quirements will be based upon basement, additional space for drums in the Plutonium Facility waste matrix and containment drum storage was arranged at basement storage areas have layers. Legacy waste drums are the Chemical and Material been held the requisite time, vi- being opened to confirm packag- Research (CMR) building. The sually examined, and shipped to ing configuration, and the first 10 first challenge was to determine TA-54 for storage; the remaining drums of plutonium-238 waste the extent of the problem; the drums should be shipped out are being prepared for shipment next challenge was to find the shortly after the beginning of to TA-54 for interim storage. cause of the water release from 1994. the cement monoliths. Three Nondestructive Assay Capabili- drums were found to contain Research on the gypsum cement ties for 235U at TA-55. The WIPP significant quantities of free process and on the alternative waste program allows evaluation water attributable to variances return to portland cement is be- of certifiable wastes by instru- in the cement mixing process. ing conducted. Research into ments that can be proven to be Small quantities of water vapor the portland cement process re- equal in integrity to those used are normally released during quires addressing the control of for waste assay within the ap- the cement curing process by process-related ammonium ni- proved TRU Waste Certification the heat of hydration and some- trate emissions. Experimental Program, including its integral times condense on the inside of work is also being conducted on quality assurance program. the plastic liner. Samples of the toxic metal retention, since the The N-1 nondestructive assay liquid were found to contain evaporator bottoms (the radioac- instrumentation and methods very little radioactivity. The tive residues resulting from used in the uranium assay of plutonium is effectively immobi- evaporation) contain RCRA- the mixed plutonium/uranium lized in the cement, even though regulated concentrations of drums were rated as equivalent free liquid may appear. NMT chromium, lead, and nickel. to those used in the TRU personnel negotiated with the Waste Certification Program. CST-7 Waste Management

91 Facility Operations Waste Management at Technical Area 55: A Historical Perspective and Status Report

These assays were then used to Decontamination/Repackaging limited analysis capacity and complete the paperwork pack- of Oversized TRU Waste. Work low priority. Some of the ages, which were circulated for is progressing on the disposal of samples took a year to be ana- review and approval. In addi- oversized TRU waste items lyzed. Nondestructive assay of tion, the NMT-4 Nuclear stored in the basement of the the vessels was challenging due Materials Measurements and Plutonium Facility. In early to the thick steel walls and con- Accountability Group com- 1993, a concerted effort was sequent attenuation of the radia- pleted programming and certifi- made to identify and categorize tion used to detect and quantify cation of one of the Segmented waste items in this area. Ap- the radioisotopes in the waste. Gamma Scanning instruments, proximately a dozen items were NMT-4 ran calibration tests and capable of assaying 55-gallon decontaminated, repackaged, determined uncertainties based waste drums for uranium. This and disposed of as low-level upon a series of experiments instrument was used to assay waste. Seven items that were conducted on an uncontami- drums with low atomic weight too large to fit in standard waste nated vessel with a known waste matrices. Approximately boxes were crated and are in the source placed inside. This infor- half of the 40 mixed uranium/ process of being assayed for mation was used to devise the plutonium waste drums were SNM content by NMT-4. They counting scheme for the vessels, assayed, the paperwork pack- will be shipped to TA-54 for and the clean vessels were then ages were completed, and the interim storage pending the assayed. Calculations were waste drums were sent to TA-54 availability of the TA-50 Size made based upon the weights for storage. Reduction Facility for process- of the vessels, the results of the ing waste. RCRA constituents’ analysis, Inspection/Repackaging of and the nondestructive assay. 239Pu Waste Drums. Analysis/Decontamination of Then the vessels were classified Nonconfirming plutonium-239 Experimental Vessels. The de- for disposal. In September, drums were organized into two contamination work being con- twelve 3-ft-diameter vessels major groups: (1) those for ducted by NMT-2 personnel on were shipped to TA-54 for in- which the final gross drum the experimental vessels has terim storage until the TA-50 weights did not match the sum made steady progress, and at Size Reduction Facility is able to of the individual item weights present, twenty vessels have process them. They will eventu- and the package tare (empty been decontaminated. Two ally be placed inside standard container weight), and (2) those major determinations that had waste boxes, which are accept- for which the total drum SNM to be made prior to disposal able primary containers for assay did not agree with the were these: TRUPACT II shipment to WIPP. sum of the individual item SNM (1) whether the vessel waste assays. Paper work was first ex- residues were TRU or Removal of Contaminated amined for errors; if none could low-level waste and Glove Boxes and Oversized, be found, then the drums were (2) whether the residues Low-Level Waste Boxes. The carefully opened and the items included mixed waste. problem with the glove boxes reweighed or reassayed as ap- To make these decisions, NMT-2 and the 4-ft x 4-ft x 8-ft low-level propriate until the discrepant personnel assumed that the con- waste boxes centered upon as- item was found. The package tamination on the inside of the sembly of the proper documen- paperwork was then corrected vessels was uniform and that the tation necessary for review and and routed for reviews and ap- waste residues removed were acceptance by the CST-7 Waste provals, allowing shipment to representative of the contamina- Management personnel. After TA-54 for interim storage. tion. Samples of these residues consultation with CST-7, paper- were sent to CST-1 for analysis, work packages were prepared but analysis was delayed due to by NMT-2 waste management personnel, and shipments were scheduled. Some of the glove

Nuclear Materials Technology 1994 Report 92 Facility Operations Waste Management at Technical Area 55: A Historical Perspective and Status Report boxes that had been in storage In the past, the need for waste Epilogue and exposed to the environ- minimization was addressed in Dramatic progress has been ment for several years had water each Safe Operating Procedure made in the waste management inside the crates, which had to (SOP) as a self-contained sec- program at TA-55 over the past be drained in a manner accept- tion, but no coordinated effort year. Reorganization of the able to waste management was made to provide compre- waste management team and and health physics personnel. hensive and consistent guidance support by NMT-2 operations The thirty-six 4-ft x 4-ft x 8-ft to all operations at TA-55. personnel and management has boxes were shipped to TA-54 in restored operational viability. March 1993, freeing considerable Currently, a waste minimization Communication, vital in dealing space in the yard for other pur- plan for TA-55 is being written. with continually changing rules poses. The glove boxes, which This plan will provide guidance and regulations, has been rees- required more preparation than to all personnel (from office tablished and enhanced. Prob- the low-level waste boxes, were workers to plutonium process- lematic legacy wastes have been not shipped off-site until July ing technicians) whose opera- addressed and remediated 1993. tions generate waste on-site. where possible, and the root The redundancy of addressing causes have been identified to Waste Minimization waste minimization in each prevent recurrence. The number Waste in any form is expensive SOP will be eliminated, and of technical staff and operations to package, document, treat, and waste minimization information technicians is being increased to dispose of; costs are escalating for all personnel and all opera- handle the workload, and new as regulations become more tions will be consolidated into personnel are being trained on stringent, treatments become one source that can be revised both regulations and practical more exotic, and landfill space expeditiously as warranted. aspects of operations. Much disappears. Waste minimization work has been accomplished serves to reduce the expense of Performance indicators (PIs) are over the past year, but even waste operations by avoiding or being investigated that would more waste issues are yet to be minimizing waste in the first provide a measure of waste resolved. The waste manage- place. It is also mandated by minimization at the Plutonium ment organization must learn to DOE Order 5820.2A, “Radioac- Facility. Information on past anticipate mandated changes tive Waste Management,” and program performance has been and must plan for facility opera- the Resource Conservation and largely qualitative because only tion changes in such a way as to Recovery Act. information on waste leaving minimize the impact and burden that facility has been recorded. to the primary objective of the The philosophy of waste mini- Incorporation of PIs at TA-55 to Plutonium Facility: to provide mization at the Los Alamos Na- measure waste minimization to the DOE Weapons Complex tional Laboratory is defined by will require the commitment of first-rate research and develop- Administrative Procedure 10-8, manpower and computer time ment in actinide processing.¨ “Waste Minimization.” To date, to track the requisite variables waste minimization efforts at that comprise each PI. TA-55 have been the responsibility of the waste generators. Per- sonnel are encouraged in waste generator training to utilize waste minimization as the most desirable method of waste management in their operations. The generator has the responsibility to reduce waste not only in every day operations but also in designing future processes.

93 Facility Operations Future of Waste Management at the Los Alamos Plutonium Facility

Ronald E. Wieneke Introduction Nuclear Materials Processing: With the enactment of the Federal Facility Compliance Act of 1992 Nitrate Systems (FFCA), the national laboratories within the Department of Energy (DOE) weapons complex are no longer immune to fines for violat- ing Environmental Protection Agency (EPA) hazardous waste regulations. The Federal Facility Compliance Agreement between the laboratories and their respective states will establish a timetable to achieve full compliance with existing requirements. If the agreed- upon milestones are not met, facilities and managers will be subject to civil and possibly criminal penalties for violations. Those who oversee operations that generate hazardous waste must move quickly to implement waste management systems, but restraint “We must find innovative and good judgment will be required to avoid overburdening ways of complying with operations with excessive documentation. environmental regulations The Plutonium Facility at Los Alamos National Laboratory strives while maintaining flexibil- to meet existing environmental regulations and to anticipate future requirements. We must find innovative ways of complying with en- ity of operations and an vironmental regulations while maintaining flexibility of operations emphasis on research.” and an emphasis on research. This enormous task can only be accomplished with a spirit of cooperation among those who generate, manage, and regulate waste. This requires cooperation among (1) waste generators and waste management personnel within the Plutonium Facility, (2) waste management and environmental compliance groups at Los Alamos, and (3) external regulatory agencies. A reasonable approach to the interpretation of regulations as they apply to the complexities of radiochemical processing operations is required at all levels.

The Plutonium Facility must look ahead and anticipate changes in requirements for the management of hazardous waste. Careful planning will be needed to give us time to act in harmony with our mission, rather than being forced to react to new requirements in ways that merely minimize adverse consequences. Deadlines for Proper Storage and Treatment of Los Alamos Mixed Waste The Federal Facility Compliance Act (FFCA) waives the doctrine of sovereign immunity for federal facilities, such as DOE contractors, with respect to enforcement of environmental regulations. This waiver allows civil and possible criminal penalties to be levied by the EPA. The FFCA in essence gives the EPA the teeth that it needs to enforce compliance with the Resource Conservation and Recov- ery Act (RCRA) for hazardous and mixed waste operations. Fur- thermore, the DOE has agreed to negotiate agreements with the respective state environmental agencies on how to implement RCRA at each facility.

Nuclear Materials Technology 1994 Report 94 Facility Operations Future of Waste Management at the Los Alamos Plutonium Facility

Los Alamos must certify compli- Mixed waste, waste that has waste streams. This documenta- ance with the storage criteria for both radioactive and hazardous tion is most effective when the transuranic (TRU) mixed waste components, is a much larger process is very well defined and by the end of January 1995 and problem. DOE is severely lim- the composition of incoming must present a plan to the New ited in its ability to process process streams is well known. Mexico Environmental Depart- mixed waste to meet the EPA Documentation saves a signifi- ment (NMED) to establish treat- waste treatment standards due cant amount of money by elimi- ment priorities for low-level to a finite number of viable nating analytical tests for mixed waste by the end of May mixed waste treatment options compounds that have not clearly 1995. The treatment and dis- and a shortage of accessible been associated with the pro- posal of this waste is signifi- treatment facilities. Despite cess. On the other hand, it is par- cantly complicated by the these problems, waste must be ticularly damning when a presence of radioactivity, as has treated in accordance with the hazardous compound or condi- already been demonstrated by EPA standards before disposal. tion is discovered that was not the problems encountered with Low-level mixed waste is re- disclosed in the process knowl- stored mixed waste at other quired to be treated in accor- edge proclamation. Such a dis- DOE facilities. The ultimate dance with 40 CFR Part-268, covery throws a shadow of treatment and disposal of mixed “Land Disposal Restrictions,” doubt over all wastes so charac- waste, which must be accom- and disposed of in specially terized. For this reason, control plished in a manner that pro- designed hazardous waste land- over process knowledge docu- tects employees, the public, and fills. The landfills must incorpo- mentation must be rigorously the environment, presents both rate all specified design features exercised. a technological and financial such as double liners, leachate challenge to the entire DOE collection and treatment sys- The Plutonium Facility must weapons complex. tems, a leak detection system, document its processes carefully. and monitoring wells. All radioactive wastes must be Cost and Complexity of Waste precisely characterized to verify Disposal Disposal of TRU mixed waste that no hazardous constituents Each kind of waste generated at should be easier, provided the other than radioactivity are the Plutonium Facility has its Waste Isolation Pilot Plant present. DOE Order 5820.2A, own set of complex require- (WIPP) is successful in obtaining “Radioactive Waste Manage- ments that must be met to allow a no-migration variance for the ment,” requires that waste dis- disposal. Ever growing require- disposal phase. WIPP currently posal facilities know the ments will continue to escalate has a no-migration determina- contents of the waste that they the costs related to characterization from the EPA for the test are burying by establishing tion and disposal of waste. Even phase only. The disposal of waste acceptance criteria for low-level radioactive waste, a mixed waste (both low-level those sites that generate and relatively uncomplicated waste and TRU) is costly, and those ship the waste to them for dis- stream, will be more difficult who generate waste must avoid posal. In addition, generators and costly to dispose of in com- the production of additional must also know the hazardous ing years as the Los Alamos mixed waste whenever possible. component of the waste so that low-level waste acceptance crite- appropriate handling, storage, ria become more stringent and Documentation of Waste and treatment methods can be landfill space is depleted. All Through Process Knowledge implemented. personnel who generate waste Process knowledge is the docu- must assure that no RCRA haz- mented evidence provided by ardous components have been cognizant operating personnel inadvertently included in their of the presence or absence of low-level radioactive waste and hazardous compounds in their must demonstrate control over their waste streams.

95 Facility Operations Future of Waste Management at the Los Alamos Plutonium Facility

Mixed waste must be carefully Impact of Waste Management Problems with Conflicting analyzed for both hazardous on Research Regulations and radioactive constituents be- As the requirements of waste The complexity of the opera- cause disposal facilities must management at the Plutonium tions at the Plutonium Facility satisfy two sets of disposal site Facility grow, the program must can bring regulations from dif- criteria. One of these is the pre- become more efficient. The Plu- ferent agencies into conflict with scriptive criteria of RCRA, tonium Facility mission is to each other. For example, what which covers the hazardous perform basic and applied re- one agency may consider radio- component of the waste. The search in actinide chemistry and active waste, another may view other is the performance-based materials research in support of as valuable, recoverable scrap criteria of 40 CFR 191, “Environ- plutonium processing needs material. A significant portion mental Radiation Protection within the DOE weapons com- of the material stored at the Standards for Management and plex. Research at the Plutonium Plutonium Facility contains eco- Disposal of Spent Nuclear Fuel, Facility benefits current opera- nomically recoverable quantities High-Level and Transuranic Ra- tions involving cleanup and re- of plutonium, but it may also dioactive Wastes,” which covers covery of plutonium-bearing fall under the RCRA definition the radioactive components in scrap, safe decommissioning of waste. This material, which is the waste. To meet both of these of weapons components, and known as “scrap” with respect criteria for a waste from an un- preparations for long-term Spe- to plutonium recovery, is being known source, the generator cial Nuclear Material (SNM) stored until processing can would have to perform a full storage. Research at the Pluto- resume. This situation has been spectrum RCRA analysis on nium Facility also provides the aggravated by the changing mixed waste in accordance with basis for the design effort of the mission of the DOE complex, EPA’s SW-846, “Test Methods for next generation of DOE pluto- which has brought the concept Evaluating Solid Waste,” as well nium manufacturing facilities of economic discard limits for as perform a comprehensive (Complex 21). plutonium into question. The radioassay. Estimated cost of absence of a demand for pluto- the RCRA analysis alone for a Meanwhile, requirements for nium in weapons production mixed waste could reach tens of handling, storing, and certifying and the surplus from the decom- thousands of dollars per sample. waste are increasing the missioning of retired weapons This estimate is not unreason- workload of process operations have resulted in plutonium be- able considering the possible personnel. More time is required ing considered by some as a number of hazardous compo- to complete the necessary docu- liability. nents, employment of tech- mentation and maintain control niques to reduce radiation over waste collection and stor- When the requirements of the exposure to operators, and the age at the Plutonium Facility. Atomic Energy Act (AEA) and quality assurance requirements Every dollar that is spent for the the RCRA are both applied to for samples and sampling proto- documentation, treatment, han- mixed wastes, the result is often cols (including chain of cus- dling, packaging, or disposal of conflicting requirements. The tody). Process knowledge waste is a dollar that cannot be storage requirements for radio- documentation is, therefore, an used for research. In this age of active scrap material classified extremely important consider- shrinking research budgets and as mixed waste present particu- ation from the aspect of cost growing competition for assign- lar challenges: the AEA storage reduction. ments and funding between criteria for SNM emphasize se- sites in the DOE complex, mak- curity, personnel radiation ing optimum use of every fund- safety, and criticality safety. ing dollar is essential. For this reason, new programs are being planned and initiated that will increase the effectiveness of the waste management operations at the Plutonium Facility.

Nuclear Materials Technology 1994 Report 96 Facility Operations Future of Waste Management at the Los Alamos Plutonium Facility

Several of these safety principles The premise behind the RMMA For example, the Laboratory’s for handling fissile materials are concept is that all waste coming environmental management or- in conflict with EPA guidance from a radiation control area is ganization plans to send a repre- and RCRA requirements for potentially contaminated and sentative to work and reside at hazardous waste storage. Spe- should be disposed of as radio- the Plutonium Facility. This rep- cifically, the periodic inspection active. This premise is in direct resentative will gain an intimate of items and the labeling of ex- conflict with waste minimiza- understanding of the needs of isting stored containers as re- tion principles. Because neither those who generate the waste, quired by the EPA increase the DOE nor EPA has issued and communication will be en- personnel exposure significantly, de minimis radiation standards, hanced. A communication net- and the EPA requirement to con- nuclear facilities have had to use work is also being developed to fine water used in putting out the strict interpretation of a “no translate requirements from the fires in storage areas could radioactivity added” doctrine federal regulations to under- present a criticality hazard. for the segregation of waste standable directions at the oper- streams. This overly conserva- ating level. Monthly meetings Another example of conflicting tive stance will result in the are being held to discuss issues guidance is the TRUPACT-II generation of more suspect of mutual interest. Although all Content Code (TRUCON) limi- radioactive waste than is neces- personnel have full work sched- tations, Radioactive Material sary. Solutions that will allow ules and have difficulty commit- Management Areas (RMMAs), reasonable operating parameters ting time for additional activi- and waste minimization require- to be implemented without com- ties, the time required for these ments. Radioactive wastes are promising safety and environ- meetings is well spent. disposed of on a volumetric cost mental concerns must be found. basis; therefore, it is beneficial to Operating personnel must be minimize waste volumes to de- Improving Communication and convinced that waste manage- crease disposal costs, including Interaction at Los Alamos ment regulations are not trivial transportation and handling. Communication must be en- and must be observed consis- TRUCON limits the quantity of hanced and encouraged among tently. Waste certification and plutonium-238 to 0.049 grams personnel from the various characterization and waste per drum by specifying the al- Los Alamos organizations that stream control require addi- lowable activity in organic waste are engaged in waste manage- tional manpower and financial contained within a 55-gallon ment activities. Waste manage- expenditures. Operating person- drum with three levels of con- ment personnel in the Pluto- nel must clearly understand the tainment. This limit, which is nium Facility have been forced significance of this additional based on what many consider to to interact closely with Labora- work, even if their principal in- be an overly conservative trans- tory waste management and terest is in research. Specific portation safety analysis, may environmental management training must be developed for only amount to one contami-Matrixgroups because of the number all those who generate waste at nated tissue in a 55-gallon drum. the Plutonium Facility. The elementand complexity of regulations pertaining to radioactive, haz- training program must be well ardous, and mixed wastes. organized and must use knowl- This interaction has resulted in edgeable waste management several benefits to the Labora- personnel as instructors. tory and the Plutonium Facility.

97 Facility Operations Future of Waste Management at the Los Alamos Plutonium Facility

Need for Planning and established waste acceptance cri- manner, down to the operating Scheduling Waste teria. Waste management activi- level. The revision and review Management Activities ties are conducted in a way that process for a Plutonium Facility SOP can easily take a year. Much Scheduling waste management minimizes the impact on operations as much as possible. Waste of this time is spent on resolving activities has become essential comments and questions from because the waste handling management personnel must be careful to respect and under- reviewers. With three levels of process has become complex, program documents to update, requiring the support of dozens stand the interests of those who generate waste and try to avoid it could take three years for of personnel. Waste can no changes to filter down to the longer be collected and dis- restricting their operations un- necessarily. At times, innovative working level. This time span is posed of as it is accumulated. clearly unacceptable. Something Planning must begin before the ways of complying with regulations must be implemented. must be done to expedite the re- waste is generated, and schedul- view and approval process if oping must allow personnel to Shortcomings of the Current erations are to keep current with plan and coordinate their activi- changing requirements. ties. Special activities, such as Waste Management Program The waste management pro- glove-box removals, must be With increasing frequency, the gram must be able to react expe- scheduled to optimize use of demand for manpower to keep diently to regulatory changes. limited waste management re- up with paperwork require- For example, the TRU waste sources and to avoid conflicts ments is being met by the use program at Los Alamos consists with normal Facility activities. of contractors. If contractors are of a hierarchy of documents. Groups are experiencing staffing used, it is imperative that they EPA regulations, Department shortages, which limit the avail- 1) demonstrate regulatory of Transportation and Occupa- ability of backup personnel for expertise regarding the tional Safety and Health Admin- waste support activities. When document that is being istration regulations, AEA unplanned activities do occur, written and requirements, and DOE orders they must be handled with spe- 2) possess knowledge of the are the highest-level documents. cial care to prevent surprises operations for which the Below them are second-level that can derail normal waste dis- documents are being documents telling how the ap- posal operations. written. plicable regulations and orders The sponsor cannot simply pay will be implemented at disposal The Plutonium Facility Waste the contractor and walk away facilities (WIPP) and the Labora- Management Team from the project expecting to re- tory. The third and final level of The waste management process ceive a satisfactory product. If documents are those written at at the Plutonium Facility must anything, extra attention must the Plutonium Facility such as be performed in a cost-effective be given to the development of waste stream attachments to manner while meeting all regu- contractor-generated program the Laboratory certification latory requirements. A waste documents. Program documents plans and the Safe Operating management team has been must be written realistically. It is Procedures (SOPs), which de- formed to coordinate efforts very easy to make a commit- scribe how work is to be per- at TA-55 to attain these goals. ment that looks good on paper formed. Changes to higher-level The team provides an interface but is impossible to carry out. documents must be reflected in between regulations and opera- These unrealistic commitments lower-tier documents in a timely tions by providing guidance to are usually the first findings to those generating waste. This be identified during an audit. guidance includes how to certify and package waste according to

Nuclear Materials Technology 1994 Report 98 Facility Operations Future of Waste Management at the Los Alamos Plutonium Facility

Evolution of Waste Manage- The Plutonium Facility waste Summary ment at the Plutonium Facility management team is working to The Plutonium Facility at Los Several projects are currently improve the waste management Alamos is striving to come into underway to improve the effi- system at the Facility, not only to full compliance with all environ- ciency of the Plutonium Facility meet the needs of the present mental regulations for the stor- waste management program. but also to meet the challenges age, treatment, and disposal of The first is the computer-based of the future. Unencumbered radioactive and hazardous Nuclear Materials Technology communication is being encour- waste. The complexity of the (NMT) Waste Management aged. For example, personnel at regulations paired with the System. This system will take the Plutonium Facility have uniqueness of radiochemical the place of the current paper been issued waste management processing operations presents system, which requires stan- telephone cards to help them get challenges and frustrations to dardized forms, manual transfer in touch with the right person to those involved in the waste of data, and operator calcula- assist them. Organizations out- management process. Mutually tions. The hard-copy form used side the facility have been given agreeable solutions to waste in the current system must ac- information about how the Plu- management problems must be company the waste and func- tonium Facility waste manage- found: the EPA and DOE must tions as the data package for the ment team is organized. A new come to agreement on solutions completed container. The form facility-specific training course for the DOE complex as a whole, must then be physically circu- for those who generate waste is and the NMED and Los Alamos lated for review and approval. being planned. The course will must agree on solutions for the The new computer-based pro- be interactive and will be re- Plutonium Facility in particular. gram will either reside on one quired for all personnel who Funding shortfalls and budget- of the file servers distributed currently generate waste as well ary constraints are placing a around the network or on the as new employees. Guidance on growing emphasis on minimiz- individual personal computers process knowledge documenta- ing waste generation and maxi- connected to the network and tion will be emphasized. The mizing the efficiency of waste used for data entry in the field. Plutonium Facility is taking management practices. The A user-friendly interface will advantage of the DOE Mentor waste management team is prompt the user for data, check Program, a program by which making steady progress in im- the input, do calculations auto- DOE offers assistance from proving the quality of the waste matically, and disallow the entry experts in specialized areas, management program at the of improper information. Hard- including environmental com- Plutonium Facility. These im- copy forms for data packages pliance. These personnel, who provements will not only result will be generated outside of the are familiar with regulatory is- in enhanced operational safety processing area, eliminating pa- sues within the DOE complex, and a reduction in environmen- per transfer across the radiation are investigating compliance tal risk but will also ultimately control areas. Approvals will be problems and are discussing result in full compliance with electronic, expediting the review resolution strategies. Their rec- environmental regulations and ommendations are currently efficient utilization of taxpayer process. This system should be ✦ ready for testing in May of 1994. being evaluated by Facility dollars. In conjunction with this system, management. a bar code labeling scheme will be instituted. Waste items will be labeled with identification (ID) stickers and scanned as they are transferred between operations. Data will be entered under the item ID and transferred electronically into the NMT Waste Management System.