Research Opportunities for Managing the Department of Energy's Transuranic and Mixed Wastes (2002)

In this chapter the committee offers its views and recommendations on research opportunities for the Environmental Management Science Program (EMSP) to address challenges in managing transuranic and mixed low-level wastes. Based on its discussion of the issues that frame these challenges in Chapter 2 and its visits to Department of Energy (DOE) sites (Appendix E), the committee concluded that the most significant needs and opportunities lie in

• waste characterization and how waste characteristics may change with time,

• location and retrieval of buried wastes,

• waste treatment, and

• long-term monitoring.

The committee has been selective in its recommendations to encourage the EMSP to concentrate its limited funding in a few specific areas where the committee believes research can lead to the most significant improvements. Some technology areas, although clearly important, were excluded because in the committee’s view the science and technology base already exists to address them on a relatively short time scale—less than five years.

Each recommendation is illustrated with a brief discussion of the current baseline technologies and technology gaps,¹ challenges for next-generation technologies, and research opportunities. Some examples are included, but these should not be construed as the only opportunities that the research community might perceive. Although the selection of examples was influenced to some degree by the back-

grounds and expertise of committee members, the research recommendations were arrived at by a consensus process that considered input to the committee, the needs of the end user,² the existence of critical knowledge gaps, the potential for future cost and schedule savings, and the possibility of achieving technology breakthroughs.

The EMSP should support research to improve the efficiency of characterizing DOE’s TRU and mixed waste inventory. This should include research toward developing faster and more sensitive characterization and analysis tools to reduce costs and accelerate throughput. It should also include research to develop a fuller understanding of how waste characteristics may change with time (chemical, biological, radiological, and physical processes) to aid in decision making about disposition paths and to simplify the demonstration of regulatory compliance.

Waste characterization is defined as “the determination of the physical, chemical and radiological properties of the waste to establish the need for further adjustment, treatment, conditioning, or its suitability for further handling, processing, storage or disposal” (IAEA, 1993, p. 52). Current regulations require detailed characterization of waste for shipping and disposal, especially for transuranic (TRU) wastes destined for the Waste Isolation Pilot Plant (WIPP) as described in Sidebar 3.1. Characterization is also necessary to determine treatment options (see “Treatment” section later in this chapter). Information needed for waste characterization generally includes the identity and amount of radionuclides, liquids, volatile organic compounds (VOCs), polychlorinated biphenyls (PCBs), mercury, other metals regulated by the Environmental Protection Agency (EPA), and the rate of hydrogen generation. Such characterization increases the time and cost of preparing waste for offsite shipment as well as the potential for worker exposure to radiation.

Most transuranic and mixed wastes (TM wastes) are packaged in 55-gallon drums—hundreds of thousands of them as noted in Chapter 2. In addition, the Transuranic and Mixed Waste Focus Area (TMFA)³ esti-

mates that there are at least 12,500 large containers at five DOE sites that present special challenges for characterization because of their size. The sites cannot be closed without dispositioning these containers.⁴

One type of characterization is physical—an assembly-line style of nondestructive examination (NDE) and assay (NDA) to identify the contents of an unopened drum (see Figure 3.1). Is it sludge or debris, plastic or metal, solid or liquid? Has it been sealed, stabilized or treated properly? NDE methods also permit assessment of the heterogeneity of the drum contents and provide the means to screen for prohibited items (e.g., gas cylinders). A second type of characterization determines the chemical and radiological composition of the waste. Does the composition of the waste meet transportation and regulatory requirements for disposal? If not, how should the waste be treated? While these types of characterization provide snapshots of the waste, another consideration is that waste characteristics will change with time through radiological, chemical, and biological processes. The potential for gas generation, particularly hydrogen, is of concern. Understanding how the characteristics of containerized waste may change with time is especially important for its continued storage, shipping, and eventual disposal.

For heterogeneous debris (see Chapter 2), the baseline characterization methodology for contact-handled TRU (CH-TRU) wastes comprises a number of steps, including radiography and opening the container for visual inspection to validate process knowledge (see Sidebar 3.1). Swipes or sampling and analysis are required to obtain contaminant information. If NDA is required, waste must be repackaged into containers sized for the available instrumentation. For homogeneous solids, a statistical number of drums require coring and analysis (St. Michel and Lott, 2002).

For TRU wastes that must be handled remotely because they also contain substantial amounts of gamma-emitting isotopes (RH-TRU), the current baseline requires the same characterization steps as CH-TRU. DOE is seeking to change this requirement because of the difficulty of making such detailed characterization in remotely operated facilities and the increased risks of worker exposure (NRC, 2002b). From a cost-saving standpoint, DOE would like to characterize RH-TRU based on process knowledge only. More realistically, however, DOE believes that

FIGURE 3.1 X-ray examination of a waste drum allows operators to determine if it contains prohibited items that must be removed before shipping the waste to a disposal site. Such visual inspection of a drum, at various positions and angles, may take several hours.

Source: DOE Carlsbad Field Office.

some additional characterization will be required to validate the process knowledge (St. Michel and Lott, 2002).

DOE would like to simplify the characterization baseline for TRU wastes in order to increase the rate of shipping these wastes to WIPP. The main approach is to seek changes in current transportation and disposal requirements, for example, to reduce the many detailed characterization steps illustrated in Sidebar 3.1. In addition, the TMFA was developing improved characterization technologies. An example of state-of-the-art TMFA technology is the assay system being developed at the Idaho National Engineering and Environmental Laboratory (INEEL [see Sidebar 3.2]).

The committee believes that a gap exists in the lack of technologies available to automate sampling and characterization in a more reliable fashion. The problem becomes particularly complex for certain classes of wastes for which a single sample may not be representative (e.g., debris waste). Other technology needs include methods to nondestructively assay for radiological and nonradiological constituents (e.g., Resource Conservation and Recovery Act [RCRA] metals, low levels of TRU isotopes) and improved statistical methods to support approaches such as compositing.

Chemical and physical assays provide only a snapshot of the waste. An improved understanding of reactions that can change the waste characteristics with time is fundamental to making informed decisions on storing, shipping, and disposal. The Strategic Laboratory Council’s analysis of DOE’s environmental quality research and development portfolio found that the primary gaps in research in TRU and mixed waste disposal (DOE, 2000c, p. 28)

The current state of the art is that simple drum-by-drum methods are employed to empirically derive the thermodynamic and kinetic parameters for changes that occur in the waste as a function of broad waste categories. These methods often form the basis for assessment of the compliance of wastes with the waste acceptance criteria (e.g., gas generation rates, stability) for the pertinent disposal site.

There is a gap in basic knowledge of waste behavior, both in mechanistic understanding and in understanding other types of chemical behavior that may impact waste composition or physical integrity. These include the chemical form (speciation) of contaminants, metal-catalyzed redox reactions of waste constituents, sorption to the waste matrix, and effects of pH and ionic strength on the waste matrix. One example of

unanticipated chemical reactivity is the recent demonstration that hydrogen can be generated by reduction of water by plutonium dioxide in oxygen-lean environments (Haschke et al., 2000). This serves as a pathway for hydrogen generation in addition to radiolysis, and has potential implications for how the plutonium might migrate from the waste.

There are gaps in understanding microbial effects in waste. Little has been done regarding the microbiology of TM wastes. Conversely, much has been done toward characterizing and exploiting the microbiota of more traditional hazardous wastes, resulting in significant cost savings over traditional disposal and remediation technologies (Harkness, 2000; Steffan et al., 2000). Microbial effects may be important in organic materials stored for long periods or in mixed waste landfills. Knowledge gaps include (1) what microbes are present in mixed and TRU waste and at what abundance; (2) what the activities of the microbes are and how their activity affects the waste material; and (3) how these microbes can be exploited to improve the treatment or disposal of TM wastes (Brockman, 1995; Newman and Banfield, 2002; Reysenbach and Shock, 2002).

In recent years, significant strides have been made toward developing methods and tools for analyzing microorganisms in environmental samples. Application of these techniques to TM wastes should lead to a better understanding of biogeochemical reactions occurring in the waste materials. This information will be important for assessing treatment options or monitoring the progress of biological treatment technologies applied to the material (Brockman, 1995; Newman and Banfield, 2002). Many of these methods and tools should be directly applicable for studying and characterizing the microbiology of TM wastes. For example, some of the more popular modern molecular biology-based techniques currently being used for environmental analysis include density gradient gel electrophoresis (DGGE [Muyzer, 1999]), terminal restriction fragment length polymorphism analysis (T-RFLP [Takai et al., 2001]), and whole or partial genome sequencing, but other equally useful methods clearly exist. Other studies have demonstrated that microbial DNA can be extracted from complex environmental samples, such as soil, and cloned to assess the genetic and functional diversity of uncultured organisms (Rondon et al., 2000). Still other technologies rely on identifying chemical signatures to confirm the presence of microorganisms or evaluate their activities in environmental samples (Nichols and McMeekin, 2002; Rütter et al., 2002; Zhang, 2002). Likewise, measuring the presence and abundance of specific biomarkers (e.g., peptides) can provide an understanding of microbial activities occurring in complex matrices (Elias et al., 1999). The combination of traditional culturing methods and modern molecular or chemical ana-

lytical techniques provides powerful tools for evaluating microbial populations in complex environments.

From its fact-finding visits to DOE sites and committee members’ expertise, the committee believes that the greatest challenges for the next generation of waste characterization technologies will be to provide the following:

• more rapid, automated, NDA and NDE methods;

• more sensitive NDA and NDE technologies for larger containers and hard-to-detect chemical and radioactive contaminants; and

• improved methods, based on fundamental modeling, to derive present and future waste characteristics from a limited number of sampling parameters.

As uranium and transuranic elements such as plutonium decay, alpha particles are emitted. Associated with these alpha particles are mostly low-energy, nonpenetrating gamma rays and neutrons. Instruments for safeguarding nuclear materials rely on detecting these types of radiation. However, these measurements have drawbacks for waste assays because the radiation is subject to self-attenuation and shielding by extraneous materials, especially in larger containers. The resulting energy spectra are degraded to the extent that the resolution of currently available detectors is often insufficient to identify the specific fissioning isotope.

Multiple high-purity germanium detectors can be used to improve the sensitivity of gamma-ray spectroscopy (as can coincidence counting, see Sidebar 3.2), but this dramatically increases instrument expense and limits the number of stations that can be placed in service. Information about the spatial distribution of radioactivity within the containers is absent, which does not allow the detection of radiological “hot spots.” Calibrating the results of such measurements may become complex and require statistical methods with assumptions about waste homogeneity and source location, which can introduce substantial error.

Neutron activation analysis identifies the chemical elements and thus can be used to search for transuranics noninvasively. It is, in fact, capable of identifying several radionuclides present that do not emit high-enough-energy photons to be readily detected using gamma-ray spectroscopy. Neutron activation analysis, however, suffers from the same turnaround, expense, calibration, and inhomogeneity problems associated with gamma-ray spectroscopy. Neutron activation analysis

requires a high flux of neutrons, which is expensive, difficult, and potentially hazardous to provide. Inexpensive, bulk detection of TRU radionuclides by NDA presents a challenge for next-generation technologies.

DOE is presently developing new technologies for facility deactivation and decommissioning and subsurface contamination applications that are equally relevant for the characterization of containerized waste. Examples include portable “laboratory-on-a-chip” sensor technology for the quantitative identification of radionuclides and metals such as uranium, plutonium, cesium, strontium, mercury, and lead (Collins and Lin, 2001; Collins et al., 2002); the microcantilever sensor array technology for real-time characterization of the chemical, physical, and radiological content of ground water and mixed waste (Ji et al., 2000, 2001); and the micro-chemical sensor for in situ monitoring and characterization of volatile contaminants (Ho et al., 2001).

Similarly, the Department of Defense (DOD) has invested heavily in the development and deployment of both contact and standoff (non-contact) detectors for chemical and biological warfare agents.⁵ The ultimate program goals are to develop robust, portable, real-time sensors capable of detecting agents well below incapacitating levels. The sensors are usually connected to air sample collecting or concentrating devices, although methods to analyze water and soil samples have also been developed. Chemical sensor technologies under development are based on ion mobility, surface acoustic wave (SAW), and miniature mass spectrometry. For biomolecules and organisms, sensor development includes fiber-optic waveguide, polymerase chain reaction (PCR), and DNA chip technologies. The present sensors are either chemical or biological only, but plans are to develop combined nuclear-biological-chemical sensors.

Microorganisms can potentially affect the chemical composition and physical properties of both the contaminants and the waste matrix. Radiation-resistant bacteria were first discovered in the mid-1950s (Anderson et al., 1956). The best-studied of these organisms is Deinococcus radiodurans, which can withstand up to 5,000 grays of gamma radiation without significant loss of viability (Battista, 1997). The bacterium has been used as a host organism to create radiation-resistant recombinant organisms for degrading pollutants (Lange et al., 1998) and for treating wastes contaminated with heavy metals (Brim et al., 2000). Its entire genome sequence has been elucidated (White et al., 1999). Recent studies suggest that many TM waste materials may

contain viable and active microbial populations (Heitkamp, 2001). A previous National Research Council (NRC, 2001d) report summarized studies indicating that gas generation due to microbial degradation of cellulosic waste within WIPP will be insignificant, but it recommended monitoring.

The types of microbes present in TM wastes and the effect of microbes and microbial activity on the waste materials have largely been unstudied. These microbes could have profound effects on the ultimate fate of waste materials. The microorganisms may play a positive role by reducing the concentration of toxic organic constituents in the waste. They may play deleterious roles by increasing hydrogen or methane production, enhancing corrosion, or converting waste components into more toxic or mobile forms. Controlling or developing microorganisms to play a positive role is a challenge for future biotechnologies.

Research opportunities include noninvasive standoff imaging and image recognition methods and in-drum sensors to provide faster and more sensitive technologies for waste characterization. Research to develop predictive models of how waste characteristics may change with time, including microbial effects, can reduce the need for detailed waste analysis and provide better decision-making tools for storing, shipping, and disposing of TM wastes.

Research toward new, nondestructive and noninvasive, characterization methods may lead to significant savings in time and cost, and decreased risk of worker exposure. While these methods are currently employed in the form of real-time radiography for physical characterization and various gamma and neutron imaging techniques for radionuclide inventory determination, basic research could also yield significant improvements in these methods, as well as means of identifying other constituents of waste drums. For example, it may be feasible to identify and image RCRA metal contaminants by devising new techniques in which neutron irradiation of the waste would activate stable metals, yielding products that could subsequently be mapped for concentration and location. Other approaches might include using the radioisotopes present in the waste drums to image the contents of the drum.

New methods might also include the use of alternative forms of energy to image constituents, ranging from the use of ultrasound to identify physical forms of waste to the use of customized imaging or

local probes akin to magnetic resonance imaging (MRI) for the identification of a broad range of spin-active nuclei. Such methods, which employ detectors that are not sensitive to ionizing radiation, could avoid interference or background problems with waste that contains a significant amount of gamma-emitting isotopes, for example, RH-TRU. Promise exists for unique combinations of these methods with ionizing radiation detection methods using emissions (emitted radiation from the waste itself), induced emissions (radiation caused by external activation of waste contents), or transmissions (modification of energetic beams as they pass through waste). Similar measurement problems have been solved for medical and industrial applications. However, significant research is needed for these to meet the specific demands of the waste problem (e.g., spatial resolution, object sizes, heterogeneity, density, composition, field deployment).

Although noninvasive diagnostics are ideal, research also could improve the use of minimally invasive methods. Waste drums generally must be vented before they are shipped. This could provide a chance to emplace a variety of point detection sensors that would nondestructively convey information regarding waste constituents. When compared to conventional analytical methods that require withdrawing a sample, in situ probes could improve the speed of data acquisition and reduce associated secondary waste streams from the laboratory analyses. Examples of such probes could include fiber-optic windows for optical or spectroscopic characterization of drum contents (“optrode” approaches have been developed at Lawrence Livermore National Laboratory). Alternatively, one could envision the development of inexpensive chemical sensors operating on a variety of principles. The laboratory-on-a-chip and microcantilever sensors are examples of the type of sensor that could potentially be used for detecting changes in containerized and noncontainerized waste. EMSP projects could be coordinated with related DOD activities, especially in light of recent homeland security initiatives.⁶ Additional research is needed to develop radionuclide spectrometers suitable for use as microdetectors in combination with miniaturized chemical characterization systems.

Automation and data handling could also speed the acquisition of analytical data. Current radiography techniques rely on time-consuming visual inspection by human operators to identify prohibited items in waste drums. With increasing sophistication of image recognition algorithms, further research could yield automated systems to improve the efficiency of real-time radiography (RTR) operations. Research opportu-

nities exist in image interpretation, including self-attenuation and iterative reconstruction methodologies.

The success in application of analytical tools configured for high-throughput screening and analysis of combinatorial approaches to drug and materials discovery suggests that similar parallel approaches may be able to screen an array of large drums.⁷ Wireless technology could also be used to track and monitor the drums remotely (see “Long-Term Monitoring” section). However, remote methods to introduce samples and provide long-term power have not been developed for these sensors. It may be possible to harvest electrical power to run these sensors directly from the thermal, chemical, biological, or radiological processes associated with the waste itself.

Research is needed to evaluate the microbiology of TM wastes. The research should focus on identifying the microorganisms that exist in the waste and evaluating their function relative to the waste material. The research should determine whether these microbes affect the hazardous or radioactive components of the waste in ways that make it more or less toxic or more or less suitable for disposal in hazardous waste, low-level waste, or other landfills or repositories (e.g., WIPP). Research could focus on specific processes such as gas (e.g., H₂, CH₄, CO₂) generation and utilization, corrosion, leaching, and biological and chemical transformation of hazardous and radioactive waste components. The research should evaluate the overall effect of physical conditions (e.g., pH, temperature, radioactivity) and chemical composition (e.g., organic and inorganic components, oxygen or other electron acceptor availability) on biogeochemical processes occurring in the containers.

Research also should evaluate the effects of microbial activity on waste forms for disposal, including polymers and grouts, macroencapsulation matrices, and containers. Additional research is needed to develop new tools for rapidly diagnosing microbial activity or identifying specific microbes in TM wastes. Evaluation of chemical signatures, biochemical markers, nucleic acid sequences, or other diagnostic characteristics can lead to sensor or detector technologies for rapid and even real-time characterization or monitoring. Such basic and applied research also might lead to new technologies suitable for use in many

areas including remote sensing, long-term monitoring, and even clinical diagnostics applications. The research should complement related research efforts being conducted within DOE and at other agencies (e.g., National Aeronautics and Space Administration’s Astrobiology Institute research⁸).

One of the most beneficial cost-saving tools in the management of TM wastes would be the formulation of more reliable predictive models of how waste characteristics may change with time, well validated by experimental data. Ideally, models could predict such factors as

• gas generation rates (e.g., matrix effects on rates of radiolysis, microbial effects);

• leachability of radioactive and hazardous constituents (are methods such as the toxicity characteristic leaching procedure [TCLP] accurate predictors of susceptibility to leaching?);⁹ and

• the chemical availability of contaminants such as mercury for removal by various separation processes.

This information could simplify flowsheets, reduce the need for expensive drum-by-drum characterization, and improve the efficiency of waste packaging.

In order to construct a realistic model of hydrogen generation, for example, fundamental data would have to be compiled on rates of radiolysis for applicable organic constituents and water, rates of organic substrate diffusion under realistic conditions (clarifying matrix effects on rates of hydrogen generation), hydrogen diffusion and entrainment potential, and competing rates of chemical (e.g., recombination of hydrogen and oxygen catalyzed by metal or metal oxides) or biological reactions. These are undoubtedly complex models to derive and validate and will require new methodologies, including the means to couple interrelated parameters such as hydrogen availability (a complex function of generation and consumption, as well as physical diffusion) and metal ion oxidation states, which affect chemical reactivity.

There is a wealth of data from the actual sampling of each drum already sent to WIPP or ready to ship. A thorough analysis of these data might yield statistically valuable predictive tools. These tools might

enable relatively inexpensive sampling of a few parameters in each drum to predict the presence of problematic materials. Alternatively, within a group of drums from the same source, sampling a certain percentage of the population may adequately predict the contents of the remainder.

New experimental approaches to validate these models may also be required. If such models could be employed in justifying higher wattage or organic content limits, it could result in tremendous cost savings (avoiding additional treatments to reduce organic content) and reduced risk of worker exposure (during repackaging).

The EMSP should support research that will facilitate management of buried TRU and mixed waste in anticipation that retrieval of some waste will become necessary. This research should emphasize remote imaging and sensing technologies to locate and identify buried waste and retrieval methods that enhance worker safety.

Substantial quantities of TRU waste were disposed in near-surface excavations (shallow land disposal) prior to federal prohibition of TRU burial in 1970. Land disposal of untreated chemically contaminated wastes was not prohibited until enactment of the Resource Conservation and Recovery Act (RCRA) in the mid-1970s. Some of these wastes were buried in containers that may be retrievable;¹⁰ some were buried in bulk. In addition, a quantity of pond and lagoon sludges and associated soil remains buried (see Appendix B).

Decisions to retrieve buried waste or contaminated media generally rest on agreements among DOE, stakeholders, and regulatory agencies. Some or all of the waste buried at individual DOE sites may be left in place and monitored during long-term site stewardship programs (NRC, 2000c). However, DOE recognizes that some buried waste may require retrieval for treatment and disposition as TRU or mixed waste (DOE, 2001a).¹¹ Given the complex and changing nature of regulatory requirements and public concerns, the committee agrees that some buried wastes are likely to be retrieved in the future. Research begun now would be timely to address the additional challenges involved in locating and retrieving these materials.

DOE has no baselines for retrieving waste that has been buried at its sites throughout the country. However, initial plans to retrieve wastes at a small test plot at INEEL’s Pit 9 provide an overview of state-of-the-art technology that is commercially available to DOE (see Sidebar 3.3).

At EPA Superfund sites, two of the most commonly used approaches for the retrieval of contaminated soil and groundwater are, respectively, excavating and removing soil and solid waste and pumping and treating contaminated groundwater. Soil is excavated using backhoes, bulldozers, or front-end loaders and placed on tarps or in containers. After excavation, the soil is removed by truck and taken to a licensed hazardous waste facility for treatment. Polluted water is extracted by pumping ground water into wells and up to the surface for placement in holding tanks. EPA also allows the treatment of waste in situ. For example, an oxidant is pumped into the ground to break down chemical contaminants. In situ oxidation is often faster than pumping and treating water in contaminated aquifers (EPA, 2002b; NRC, 1995, 1997b).

Retrieval of buried waste and contaminated media generally involves excavating the entire area where the material is known or expected to be. Extending this approach to the many acres that comprise DOE burial sites is probably impractical. There is a technology gap in locating and identifying specific objects (e.g., drums, gloveboxes) and determining if they need to be retrieved. Waste characterization is done as excavation proceeds, so surprises may be encountered.¹² An earlier project at Pit 9 led to concerns that drilling to retrieve waste samples could cause an explosion or fire (NRC, 2000a). Although robotics would be ideal for increasing worker safety, the current state of the art is that robotic systems lack the versatility and reliability needed for efficient deployment in the field (Sandia, 1998).

From its fact-finding visits to DOE sites and committee members’ experience and judgment, the committee believes that the greatest challenges for the next generation of waste retrieval technologies will be to provide

• improved, noninvasive means to locate and identify buried waste whether or not it is containerized;

• remote, noninvasive assessment of the condition of buried waste containers and potential leakage from the containers; and

• remote intelligent machines (robots) for waste retrieval and repackaging or treating as necessary.

Before the waste can be retrieved it must be located and at least a preliminary characterization must be made of its condition. If the drums or other containers are intact, they can be retrieved and handled using the processes developed for stored waste. Breached containers or noncontainerized waste will be more difficult to retrieve. In either event, to minimize the number of processing steps and ensure worker safety, it would be helpful if more detailed characterization of the waste containers and their contents could be performed at the retrieval site. Hence there is opportunity to extend research for improved characterization methods, as described in the previous section, to the problem of buried waste. For example, research could lead to methods that are mobile, field deployable, and remotely operated. Next-generation technologies being developed for military purposes, such as land mine detection, might be adapted for locating buried waste (see Sidebar 3.4).

Physically retrieving wastes without exposing workers or spreading contamination will be a challenge. Next-generation robotic technology could be especially useful if the drums are not intact or if the soils surrounding the drums are highly contaminated. Robotic devices could repack the materials into a new container, preferably a smart drum capable of self-analysis and monitoring. Hanford proposed a remotely operated, multipurpose robotic vehicle with interchangeable actuators as one technology that would be capable of meeting a multitude of waste retrieval needs across the DOE complex. There is a specific need for robotic technology to retrieve RH-TRU wastes, some of which produce potentially lethal levels of radiation, from caissons located in the Hanford 618-11 waste burial grounds (Leary, 2002).

One of the best examples of next-generation robotics technology that might be further developed to retrieve buried waste drums is the HANDSS-55 system being assembled by the TMFA (see Sidebar 3.5). Drums are moved through the system automatically, and opened, and a robotic hand sorts through the contents of a drum to remove selected items as directed by video cameras and an operator using voice or touch screen commands (Frazee and Lott, 2002).

DOE has laid out an ambitious Robotics and Intelligent Machines Roadmap, which envisions threefold increases in productivity and tenfold reductions in radiation exposure to workers (Sandia, 1998). A previous NRC committee recommended robotics research for decom-

missioning nuclear facilities but cautioned that the DOE roadmap’s “envisioned leaps in technology are not likely to occur without new knowledge” (NRC, 2001b, p. 66).

Microorganisms can have a profound impact on the chemistry and fate of buried waste (Newman and Banfield, 2002). Research in the public and private sectors has led to extensive knowledge of the bio-

geochemistry and fate of traditional pollutants in the environment, for example, the DOE Natural and Accelerated Bioremediation Research (NABIR) program.¹³ The committee found no studies on the complex relationships among microbes and the organic and inorganic constituents within TM wastes themselves.

Prior to retrieval, it will be necessary to determine the condition of the waste or waste container. This need extends the challenge for imaging science research, described previously, to objects below ground. The approaches could be nonintrusive (preferred) or intrusive and could be coupled with chemical analysis. The nonintrusive approach may include ground penetrating radar, magnetometry, acoustics, chemical sensing of near-surface air samples, neutron activation, and radiological surveys. A minimally intrusive approach might use small-diameter bore-holes to emplace equipment or sensors or to collect samples.

In addition to improving image resolution, research is needed on methods to improve object identification. Is it a drum, box, or rock? Is it intact? Is the soil surrounding the object contaminated? Are the contaminants stabilized or contained? Sophisticated image analysis and identification models and software will be needed to perform these assessments. The DOD’s mine detection research might be leveraged (see Sidebar 3.4).

Research to understand biological processes that occur in buried waste can lead to better-informed decisions regarding retrieval. In simple experimental systems, the radiolytic effects of plutonium (primarily alpha-particle decay) have been shown to inhibit degradative or environmental microbes even at plutonium concentrations that do not cause chemical toxicity (Reed et al., 1999; Wildung and Garland, 1982). Understanding the relationships among waste materials and their associated microbial communities under real-world conditions (e.g., in soil, sludge, containers) could lead to improved predictability of the long-term fate and risk of the waste materials.

Significant advances in robotics will depend on research to make these devices more humanlike in their abilities to adapt to a variety of tasks, both physically and intellectually. Research toward more versatile actuators (the muscle of a robotic device), criteria-based software for independent decision making, and improved virtual reality systems for operators was recommended in a previous study of DOE facility decommissioning (NRC, 2001b). Such research would be equally relevant to developing retrieval technology for TM wastes.

The EMSP should support research for treating TRU and mixed waste to facilitate disposal. This research should include processes to simplify or stabilize waste, with emphasis on improving metal separations, eliminating incinerator emissions, and enabling alternative organic destruction methods.

Treatment is defined as “operations intended to benefit safety and/or economy [of managing wastes] by changing the characteristics of the waste” (IAEA, 1993, p. 48). Treatment may be necessary to meet regulatory requirements. The results of treatment can include volume reduction, removal of radionuclides or other contaminants, and a change in the waste’s composition.

TRU waste that meets shipping requirements can be sent to WIPP without treatment (see Chapter 2). According to DOE, after approval of Revision 19 to the Safety Analysis Report for Packaging (SARP), the volume of TRU waste that cannot be shipped due to gas generation has been reduced to approximately 3,000 cubic meters, or about 2 percent of the total inventory (Curl et al., 2002). Other shipping restrictions prohibit certain items in the waste and limit its heat production and its fissile material content.

Treatments for mixed low-level waste (MLLW) are prescribed in consent orders established between the sites and their host states in accord

with the Federal Facilities Compliance Act of 1992 (FFCA; DOE, 2000d). The availability of landfills capable of accommodating MLLW, such as the Envirocare facility in Utah, has reduced the need for treatment of hazardous and radiological constituents. However, treatment needs remain for certain wastes, particularly those containing toxic constituents, which are subject to RCRA Land Disposal Restriction treatment standards. In addition to existing wastes, new MLLW will be generated through about 2070 (see Table 2.1).

For TRU waste that does not meet shipping requirements, the baseline treatment is repackaging the waste. Repackaging may be necessary simply to remove prohibited items (see the previous section on “Characterization”). Repackaging waste to meet shipping requirements that govern heat production, fissile material content, or potential flammable gas production is extremely inefficient. According to the TMFA, repackaging these wastes so that they meet shipping regulations may result in a volume increase of ten- or perhaps twentyfold. In addition, about 98 percent of the TRU wastes that require remote handling will have to be repackaged. There is no baseline technology currently deployed for RH-TRU (Moody, 2002).

Baseline treatment technologies for MLLW developed at each DOE site as required by the FFCA were reviewed in a previous report (NRC, 1999a). Table 3.1 gives a summary of these treatment and stabilization options. Incineration is prominent among the options. However, incineration has been abandoned or is being phased out due to public concern about atmospheric emissions as well as site-specific economic considerations. As noted in Chapter 2, public opposition to a proposed incinerator at INEEL led DOE to create a Blue Ribbon Panel to recommend alternatives to incineration. A brief history of incineration and its alternatives is given in Appendix C.

The TMFA Multi-Year Program Plan states that three to five primary alternatives to incineration will be selected for comparison testing at DOE’s Western Environmental Treatment Office (WETO) in Butte, Montana, in fiscal year 2002 (DOE, 2001b). The current strategy is to select processes to represent the three general classes of alternatives: (1) thermal, (2) aqueous-based chemical oxidation, and (3) chemical separations. In addition to testing the primary alternatives at WETO, tests of other alternative methods at other locations will be conducted in a manner to make them consistent with the studies at WETO. Examples include the testing of a mediated electrochemical oxidation process at the DOD’s Aberdeen facility for chemical warfare agents, a solvent extraction method at Florida International University, and a

molten aluminum process at Sandia National Laboratories. The TMFA is currently developing guidebooks to assist DOE and permit writers in developing permit conditions for each of these alternative technologies.

Macro- and microencapsulation have become important baseline technologies for waste stabilization, i.e., treatment to prepare wastes for further handling or disposal (see Figure 3.3). Stabilization of mixed waste for disposal usually relies on its incorporation into one of several matrices—grouts or cements, glass, polymer, or ceramic—to produce a relatively homogeneous waste form, although some wastes are simply compacted.¹⁴ Macroencapsulation yields a heterogeneous waste form

by encasing the waste in a coating or block of suitable matrix, usually low- or high-density polyethylene or cement. Microencapsulation is used for the stabilization of ashes, salts, or other dry powders by mixing the waste with polymer (chiefly low density polyethylene) as feed for the extruder to produce pellets of intimately mixed waste and matrix. Versions of these technologies are used to stabilize approximately 20 percent of the waste requiring treatment for disposal at Envirocare, Utah.

Macro- and microencapsulated wastes are relatively robust mechanically when encased in a structurally rigid secondary container, and polyethylene is relatively inert to radiolysis at the levels of activity typically associated with MLLW. However, contaminants are not chemically fixed in this form, merely encased, so constituents may be more susceptible to leaching under scenarios of mechanical intrusion. For example, the EPA’s Toxicity Characteristic Leaching Procedure requires grinding

FIGURE 3.3 Macroencapsulation is used to stabilize heterogenous waste or large objects for disposal. Typically, grout or a polymer is poured over the waste so that it is encased physically. There are few data on the long-term durability of macroencapsulated wastes.

Source: DOE Richland Operations Office.

the waste form if necessary to meet size criteria for the test. This can alter the barrier provided by the encapsulation. There are few data on the long-term durability of macroencapsulated wastes.

Thermal desorption is a relatively mature technology that can remove volatile organic compounds from solid TM wastes. Commercial units are available from several vendors, (e.g., Permafix Environmental Services, Sepradyne, Envirocare). The process is being used at Oak Ridge, and the TMFA has funded process development work for treating soils and sludges at Fernald, Ohio, and organic sludges at INEEL. To drive off VOCs and moisture, waste is heated to the range of 300 to 1200ºF depending on the organics to be removed, the nature of the waste, and process details. Operating the process under reduced pressure (vacuum thermal desorption) allows lower temperatures to be used. An inert gas such as nitrogen can be used to purge the organics and prevent accidental ignition. Gases that are released are usually condensed or trapped, for example, on carbon. Thermal desorption is among the three leading options recommended as alternatives to incineration by the Blue Ribbon Panel, although the technology does not apply to all types of TM waste or reduce the waste volume (see Appendix C).

The TMFA recognized gaps in baseline technologies for treating small volumes of unique or problematic wastes. These problematic wastes include reactive materials, gas cylinders, and tritium-contaminated materials. Because their disposition requires specialized or one-of-a-kind approaches, they are often not economically attractive for private sector treatment contracts. Their limited quantities and special problems

have kept them in relatively low priority for disposition at most sites. However, because they comprise about 10 percent of DOE’s total TM waste inventory, they represent a potential roadblock to site closure. A Waste Elimination Team formed by the TMFA was involved in defining the inventory of these wastes and developing lists of technology needs (Hulet, 2002).

One example of problematic waste is mercury, and technology gaps remain for its treatment and stabilization. Mercury is present in a broad range of concentrations in several of DOE’s mixed waste streams, including large volumes of soil and debris and several types of process residues (see Table 2.3). Because it is mobile and easily vaporized, the presence of mercury creates additional effluent monitoring and control concerns in incineration and can reduce the efficiency of MLLW stabilization processes. Removing mercury before treatment simplifies downstream treatment operations.

Depending on the concentration or form of the mercury, current EPA standards require stabilizing the waste in a form that passes the TCLP test, or else treating the waste by thermal desorption or retorting, which creates a separate waste stream that cannot be recycled and will itself require stabilization. The proposed methods of stabilization are amalgamation for elemental mercury and chemical immobilization through precipitation or sorption. Some stabilizing agents are based on sulfur, whereas others have proprietary formulations. One particular matrix, which has been evaluated recently, is sulfur-containing cement. Despite the fact that these and related methods have been under investigation for several years, there still do not appear to be robust baseline methods for treatment of all mercury-containing waste streams (Morris et al., 2002; Townsend, 2001).

The lack of application of biotechnologies in TM waste treatments appears to be a technology gap.¹⁵ In the last 15 years, significant advancements have been made in biological treatment technologies for organic pollutants including chlorinated and aromatic solvents, cutting oils, PCBs, and related materials (Unterman et al., 2000). In many cases, biological treatment can result in a significant reduction in treatment costs over traditional technologies. Biological treatment approaches are sometimes coupled to other chemical or physical methods (e.g., air sparging and vapor extraction, carbon adsorption, washing) to reduce overall treatment costs. Advances in bioreactor design to allow efficient and safe treatment of contaminants have accompanied advancements in biotreatment technologies (Fortin and Desshuses, 1999; Steffan et al.,

2000). Although DOE’s NABIR program has focused on basic research related to the immobilization or removal of heavy and radioactive metals from contaminated environments, little work has been done to evaluate biological leaching or removal of these materials from TRU and mixed wastes.

Based on the committee’s fact-finding visits to DOE sites, recent approaches to developing alternatives to incineration, and committee members’ own expertise, the committee believes that the greatest challenges for next-generation treatment technologies lie in developing

• emission-free treatment processes,

• treatments for problematic or unique wastes, and

• methods to ensure the long-term durability of stabilized waste.

The committee reviewed the recommendations for next-generation alternatives to incineration by the Blue Ribbon Panel and the programs initiated by the TMFA (see Appendix C). While agreeing that the recommended technologies show promise, the committee believes that any large-scale treatment processes are likely to meet with similar public concern as incineration unless more complete knowledge can be demonstrated regarding the formation of unwanted by-products, especially after process upsets. Further, concerned citizens should be involved in selecting among technological alternatives (see Chapter 2).

Plasma arc technology is a robust technology that can treat a wide variety of wastes, although it is not emission free. This technology, first investigated by the TMFA in 1996, has continued to mature. The Naval Research Laboratory (NRL) is managing and supporting a project to establish a plasma arc hazardous waste treatment system at the Norfolk, Virginia, naval base. This system will be capable of destroying most of the 2.5 million pounds of hazardous waste generated annually at the base. A plasma arc research facility at NRL is also being used to support a Navy Advanced Technology Demonstration Project to develop a pre-prototype shipboard plasma arc system for destroying solid waste onboard Navy ships. The EMSP is supporting a plasma torch technology for decontaminating DOE facilities (NRC, 2001b).

Recently, the TMFA has chosen AEA Technology Engineering Service’s “Silver II” method for further testing.¹⁶ The process uses Ag²⁺

and concentrated nitric acid to oxidize organics, followed by electrochemical regeneration of the Ag²⁺ and recovery of the nitrogen oxides. The Silver II method produces essentially no emissions but treats a smaller spectrum of wastes than incineration or plasma processes. Silver II operates at low temperature, is easy to control, treats many organic wastes, reduces waste volume, produces no dioxins, and does not require pretreatment for small solids, slurries, or liquid wastes. However, the pretreatment of larger solid organic wastes may be required. The process is being evaluated as an option for destroying organics in Pu-238 waste at the Savannah River Site (Pierce, 2001). The U.S. Army is testing Silver II at the Aberdeen Proving Ground to destroy chemical weapons agents.

Low-emission combustion technologies are used at many of today’s petrochemical refineries. These systems (called enclosed zero flares, ground flares, population area combustors, or thermal oxidizer flares) control emission from smoke stacks, resulting in no smoke, odor, or objectionable noise.¹⁷ Such technology might provide a starting point for developing near-zero emission technologies for the more complex challenge of burning TM wastes.

The current state of the art in metal ion separation is the use of specific liquid-phase extractants or of solid-state sorbents or ion-exchange materials capable of achieving specificity for metal ion removal, particularly from aqueous waste streams. These technologies had their origin in the development of process chemistry for actinide purification and mining operations. In recent years, the emphasis in the design of such extractants shifted from increasing efficiency for producing nuclear materials to increasing separation factors for waste stream polishing (NRC, 2000b). More recently, research has begun to focus on the design of new separation systems, such as ion-specific membranes, in a desire to minimize secondary waste streams traditionally associated with liquid processing schemes.

In addition to treating legacy wastes, an important role and challenge for next-generation separation processes will be improved product separations for DOE’s continuing mission to produce nuclear materials. During its visit to Oak Ridge’s isotope production facility the committee was reminded that greater efficiency in separating highly radioactive products means a less radioactive and easier-to-manage waste stream.

Understanding factors that affect durability of matrices for mixed wastes disposed in near-surface, RCRA-compliant landfills will be especially important as DOE moves its MLLW from storage to disposal in site closures. Current testing protocols such as the TCLP are designed

for homogeneous waste forms, in which the waste is incorporated into the matrix (e.g., sludges in grout or glass). These current tests may not be good predictors of susceptibility to leaching under conditions different from those specified in the protocol (see the discussion of predictive modeling). As DOE and its subcontractors seek to reduce the costs of treating MLLW by greater use of encapsulation, developing methods to ensure the long-term durability of these heterogeneous waste forms will become increasingly important. As noted earlier, there are essentially no durability data for micro- or macroencapsulated wastes.

There are research opportunities in areas of chemical treatments (including advanced alternatives to incineration), biological treatments, stabilization, and waste form durability.

Essential to developing publicly acceptable alternatives to incineration is research to develop sensitive, reliable, and practical detection methods to track both radionuclide and hazardous chemical materials during the treatment processes. Answering the question, How much hazardous material is being released to the environment? is particularly important for public acceptance of treatment methodologies. It is also essential for controlling potential risks to workers during treatment operations.

Research opportunities for treating TM wastes are in accord with opportunities reported in a recent study of the technical needs for the Deactivation and Decommissioning Focus Area (NRC, 2001b). In particular, research in the speciation of inorganic constituents in wastes may have an impact on the selection of future treatment options. The state of the art in examining metal ion speciation is much more developed for actinide constituents in wastes than for other inorganic constituents.¹⁸ Further research is needed to assess the chemical forms of other metals (e.g., oxidation state, speciation) in the presence of complex matrices and a variety of co-contaminants. More research is needed to understand the nature of the oxide-substrate interaction. With the advent of new molecular design tools associated with nanotechnology applications, a wealth of new opportunities exists to generate novel

separation schemes based on the design of porous materials with chemical functionality specific to the separation tasks at hand.

Research can improve methods to dissolve plutonium oxide selectively in the presence of other metals and organics or lead to methods of controlling high-activity, finely divided particulate materials. For example, the Pu-238 isotope in waste from plutonium processing at the Savannah River Site exists as very finely divided oxide powder contaminating heterogeneous wastes. Due to the wattage restrictions on waste to be packaged for WIPP, it is unclear whether this waste will meet DOT requirements. The general practice for treating mixed waste would be to destroy the organic constituents through incineration or an alternative technology. The danger of dispersal of the powder precludes most thermal treatments, however, suggesting the need for a new method for removing and stabilizing the oxide powder. This raises several interesting technical challenges, such as ensuring efficient removal of finely divided powders, controlling particle dispersion and perhaps inducing agglomeration, and avoiding the generation of additional waste streams.

Hydrogen generation remains a factor in the shipment of containerized waste to WIPP. The TMFA actively sponsored work on hydrogen getters (absorbers). During a visit to the Savannah River Site, the committee heard a presentation describing significant advancements made toward the development of polymeric hydrogen getters to capture the hydrogen produced in waste drums (Duffey, 2001). These getters have proven useful for most wastes tested to date, but poisoning of the polymers—evidently by organic vapors—can occur, thereby reducing their efficiency. There are research opportunities for understanding the fundamental processes of both hydrogen production and hydrogen adsorption.

There are also opportunities to develop efficient and cost-effective biological treatment technologies for TM wastes, including recovered soils and sediments (see Table 2.1). Research should focus on the treatment of readily degradable organic material and the combined application of biological and physical or chemical treatment technologies. Although physical methods such as steam stripping, heating, or vacuum extraction may be the most common for separating the various waste components, biological treatment may be appropriate for some waste types. For example, sludges containing water and biologically degradable organic compounds may be amenable to biological treatment for removal of the hazardous components.

Research opportunities include the identification of enzyme or whole-cell treatment approaches that target specific contaminants (e.g., PCBs, mercury) or broad categories of contaminants (e.g., combustible solvents, cutting oils, chlorinated hydrocarbons). Treatment technologies must be amenable to application in the environment of the containerized waste. For example, enzyme systems that function at extreme pH levels, high salinity, or under high solvent or nonaqueous conditions would be desirable. Application of advanced molecular techniques such as directed evolution (Stemmer, 1994) might help develop appropriate biocatalysts. Research directed toward identifying ways to apply the biocatalysts, such as novel immobilization matrices, is also necessary to facilitate successful use of these catalysts.

Fundamental research may identify biological treatment processes that can facilitate the removal of RCRA wastes (e.g., Hg, Pb) and radioactive metals from contaminated media. Like hazardous organic chemicals, some toxic metals are susceptible to biological transformations. For example, mercuric ion (Hg²⁺) can undergo a range of biological transformation processes including reduction to elemental mercury, which is volatile at room temperature, or methylation, which forms the highly toxic monomethylmercury or the volatile dimethylmercury. These biotransformations have shown promise for removing mercury from wastewater (Wagner-Dobler et al., 2000).

Recent research has shown that some common iron-reducing soil bacteria can solubilize plutonium hydrous oxides that bind tightly to soils (Rusin et al., 1994). Adding a chelator enhances the solubilization process. These findings suggest that biological treatment, via either bioaugmentation or biostimulation, coupled with soil washing technologies could provide a mechanism to remediate actinide-contaminated soils. Similarly the common soil microbe Microbacterium flavescens can absorb and accumulate Pu(IV) if the siderophore desferrioxamine-B is provided (John et al., 2001).

Siderophores are chelators that are produced and released by many iron-utilizing bacteria in soil environments. The siderophores bind iron, and the iron-siderophore complex is captured by iron-utilizing microbes. The fact that these iron chelator compounds can also bind actinides suggests that they can be exploited to treat TM waste. Bioremediation approaches could involve either stimulating siderophore production by indigenous organisms or adding exogenous siderophore-producing organisms or siderophore-containing extracts to the waste or contaminated media. Once chelated, the soluble actinides could potentially be removed by soil washing or related methods.

Research is needed to develop reliable processes to transform or remove heavy and radioactive metals from mixed waste. The research should be based on the large existing volume of research on heavy-

metal biotransformation and should evaluate the effects of multiple contaminants, radioactivity, and extreme environmental conditions (e.g., low or high pH, high salts, cementing agents) on metal mobilization, immobilization, accumulation, speciation, and related transformations. The research should include multidisciplinary studies combining microbiology and genetic engineering, materials handling, reactor design, and process engineering to develop cost-effective technologies for treating TM wastes.

Biotechnology research opportunities include the identification and development of improved hydrogen and methane scavengers that can be added to waste drums. Potential scavengers could include efficient hydrogen- or methane-scavenging microorganisms or enzymes capable of binding or oxidizing hydrogen or methane. The enzymes could be improved by using genetic engineering and directed evolution of enzymes (Stemmer, 1994) to enhance their hydrogen- and methanescavenging efficiency. Additional research could include identification or development of improved methods for applying such organisms or enzymes to the drums. This could include development of immobilization techniques (e.g., sol gels, polyurethane) that improve activity and allow long-term survival of the biocatalyst while not exceeding the free liquid limits imposed by waste disposal facilities. This also could lead to studies to identify artificial electron donors that could be added to hydrogen and methane oxidation enzyme preparations to maintain their oxidative activity over long periods. Enzymes such as methane monooxygenase also may destroy other compounds such as carbon tetrachloride that poison some chemical hydrogen scavengers.

Retrieval of buried waste or contaminated media has generally required the use of very expensive engineering controls to ensure the safety of workers and the surrounding environment. A tremendous reduction in costs could be realized if waste were stabilized prior to or in the early stages of retrieval. Current containment methods involve either application of simple barriers or, in some cases, methods that partially stabilize the waste (e.g., grouting). Research should address new systems for stabilization. One could envision the development of smart materials that react with waste constituents to generate optimized coatings or combined chemical and biological processes that would stabilize the waste selectively by alteration of the matrix or generation of additional barrier layers. A recent paper describes a smart Portland cement that senses environmental conditions (Wen and Chung, 2001).

A previous report recommended research for stabilizing or contain-

ing contamination in soils or ground water (NRC, 2000a). There are important distinctions, however, between technology needs associated with the containment of subsurface contaminants in a particular geological environment and the stabilization of buried waste or contaminated media to facilitate retrieval. The need to control subsurface contaminants suggests research to devise barriers that function over the time scale envisioned for long-term site stewardship, whereas stabilization of buried waste may require only interim containment until such time as the waste is emplaced in a disposal facility.

A need exists for fundamental proof-of-concept investigations to determine the potential for microbial processes to stabilize buried TM waste by altering its composition. Promising biocatalysts may be obtained by applying traditional microbial selection and enrichment approaches to target waste materials or by “biomining” other radioactive waste materials to identify promising radiation-resistant degradative microbes. Genetic engineering could be applied to improve the metabolic capabilities of radiation-resistant organisms to develop improved biocatalysts for treating mixed waste (Brim et al., 2000; Lange et al., 1998). Organisms also can be developed to function in the extreme environmental conditions (e.g., high salt, extreme pH) found in some waste types. Research on microbiology should be coupled with research in chemistry, materials science, and process and reactor engineering to develop integrated systems to handle and treat difficult waste materials.

The NRC study of waste forms found that the matrices (e.g., grout, glass, polymers) available to stabilize MLLW for disposal are adequate (NRC, 1999a). However, the report noted that most repository performance assessments do not take credit for waste forms because quantitative tests for their long-term durability have not been developed. The report recommended that DOE’s Office of Science and Technology (OST) support work aimed at fundamental understanding of waste form durability and suggested that EMSP evaluate and fund proposed research in this area. The committee agrees that this is a valuable area for research.

Research to understand the chemical and physical processes that may leach or degrade waste forms in RCRA-compliant landfills should be combined to develop predictive models and more appropriate test procedures. As noted earlier, very little is know about the long-term durability of heterogeneous waste forms or microbiological effects on waste forms. A specific opportunity would be evaluation of the long-term stability of, and effects of biological activity on, metals immobi-

lized in sulfur cements and related materials. Bacteria are able to metabolize sulfur compounds in a variety of ways, including oxidation of reduced species (which can form sulfuric acid) and reduction of oxidized species, depending on the organisms present and the redox potential. Understanding how these reactions affect the leachability of RCRA metals from these new immobilization materials is needed.

The EMSP should support research to improve long-term monitoring of stored and disposed TRU and mixed wastes. Research should emphasize remote methods that will help verify that the storage or disposal facility works as intended over the long term, provide data for improved waste isolation systems, and inform stewardship decisions.

Long-term monitoring will play an important role in many of DOE’s site cleanup activities, especially in continuing stewardship of the sites (NRC, 2000c). In the context of TM waste, long-term monitoring will be important both during storage and after the waste has been emplaced in a disposal facility. The storage phase is likely to be long for some wastes, and it will be necessary to ascertain that the containers maintain their integrity and that internal processes in the containers are as predicted (see the section on “Characterization”). Prudence requires that wastes disposed in a RCRA-type landfill or in a repository such as WIPP be monitored until the facility is closed. Post-closure monitoring of waste in RCRA landfills will be important to ensure continuing safety and to provide data for maintaining the landfill.

In providing advice on ensuring the long-term safety of WIPP, a previous NRC committee found that “. . . the activity that would best enhance confidence in the safe and long-term performance of the repository is to monitor critical performance parameters during the long pre-closure phase of repository operations (35 to possibly 100 years)” (NRC, 2001d, p.1). For example, chambers to hold waste in the WIPP will be excavated as needed, drums will be emplaced, and lithostatic forces gradually will close the chambers—crushing the containers and sealing the waste in place. Early emplacements will be sealed during the operational life of the WIPP. Monitoring the waste during this sealing process can yield important scientific understanding of the actual closure process and also enhance safety by determining if the closure occurred properly.

In its review of DOE’s environmental quality portfolio, the Strategic Laboratory Council found a significant gap in the “fate and transport of contaminants and performance monitoring to support waste repositories and inform stewardship decisions” (DOE, 2000c, p. iii). DOE has no established plan for monitoring its MLLW disposal facilities (e.g., Hanford, the Nevada Test Site) beyond the 30-year RCRA compliance period. There are no plans for mechanical or chemical monitoring of closed rooms in WIPP during its operating period or thereafter.

Understanding and verifying the long-term behavior of these wastes and their disposition systems require more attention to monitoring than is apparent in DOE’s current planning. Research begun now to develop scientifically sound, simpler, and more reliable technologies for long-term monitoring can help ensure the safety of stored or disposed TM waste, reassure concerned citizens, provide data for new disposal facilities in the United States or abroad, and contribute generally to scientific knowledge.

Based on its fact-finding visits to DOE sites and committee members’ own expertise and judgment, the committee believes that future challenges for long-term monitoring technologies will be to extend the next-generation technologies described in the section on characterization to enhance reliability, stability, and remote operation, including

• long-lived, reliable sensors (and power supplies) that can be remotely interrogated, and

• airborne or satellite imaging.

The sensor technology discussed previously is generally applicable for long-term monitoring. State-of-the-art improvements in technology are making it possible to interrogate sensors from remote locations and to provide remote, standoff detection of both chemical and radiological hazards. Distributed sensors can be monitored by Internet and wireless technologies (Pottie and Kaiser, 2000).

In one example of a next-generation sensor technology, an electric utility prevents the overheating and shutdown of its power grid by monitoring a network of transformers and nodal sites with distributed sensors and the Internet.¹⁹ This replaces the expensive patchwork of

wired networks and human intervention required by present technologies to keep the grid in operation. The wireless sensors used in the new technology work above or below ground, and their spread-spectrum signal is impervious to electromagnetic interference. If the temperature of a transformer exceeds a safe level, the sensors trigger an alarm that alerts the utility. The utility reroutes electricity around the problem area or shuts down affected areas. By avoiding transformer failures, the utility is able to prevent costly shutdowns and blackouts. Sensors to measure other parameters can be implemented readily within the system’s architecture.

Another example includes the monitoring of mobile platforms. Transportation of materials on trucks, trains, ships, planes, and buses produces an array of potential data management system needs. These needs include environmental and safety monitoring, preventive maintenance, global positioning, antitheft measures, and real-time engine sensor data. Platform mobility and the lack of cost-effective wireless connections have been the limiting factors in developing sensor systems for mobile platform services. Again, the data can be managed through wireless sensor networks for mobile platforms. The internal temperature of a compartment can be measured to ensure that the temperature stays within range, video monitors can be utilized for theft prevention and for ensuring compliance with safety regulations, and noxious fumes can be detected to allow for timely correction. Sensor data are transmitted by a wireless wide-area network connection. Clients on mobile platforms gain cost-effective access to data about location, safety, security, engine stability, inventory, and many other parameters that can be accessed in the field or from corporate command centers.

All materials and objects (e.g., soil, water, trees, vegetation, structures, metals, paints, fabrics) create a unique spectral fingerprint. An optical sensor can determine these fingerprints by measuring reflected light, most of which registers in wavelengths, or bands, invisible to the human eye. Commercial state-of-the-art hyperspectral imaging systems operate across up to 220 wavelengths to record precise images of an otherwise hidden world. Where a standard sensor with fewer than 10 bands is capable of differentiating only between gross classes of vegetation, a hyperspectral imager can discriminate between plants and is sensitive enough to separate healthy from unhealthy growth. Hyperspectral sensors and imaging offer many attractive features for long-term, remote monitoring applications.²⁰ Current and advanced technologies include the following:

• remote sensing of earth resources,

• chemical detection and cloud tracking,

• medical photodiagnosis, and

• positional radionuclide concentrations.

A better way to monitor large, remote sites may be from airborne or satellite platforms.²¹ Overflight monitoring of nuclear power plants and other nuclear facilities for radionuclide emission is commonly practiced today. Mineral and oil prospecting is also done from the air.

There are research opportunities for developing remote, distributed sensor systems to achieve self-monitoring “smart” drums and in monitoring TM repositories to achieve more fundamental knowledge of physical, chemical, and biological processes that govern their behavior.

Smart sensors can dramatically improve the monitoring of waste storage and waste disposal in near-surface (RCRA) landfills by creating smart drums that self-analyze and report their content and location. Smart filters could monitor and control (i.e., vent or getter) the gas produced in the drum. The Internet and wireless technologies can monitor these and other distributed sensors remotely during all phases of waste disposition, including storage, transportation, and disposal. The availability of inexpensive and reliable sensors for chemical and radiological hazards in the drums would also be beneficial. To be cost-effective, the sensors must be small and mass produced like today’s MicroElectroMechanical Systems (MEMS) and tomorrow’s Nano-ElectroMechanical Systems (NEMS). To be practical, the sensors must be self-sufficient, harvesting their energy from the environment or from the waste itself.

Research on hyperspectral-imaging methods coupled with image-processing algorithms could lead to advanced remote-sensing technologies that would be well suited to monitor large, remote sites from airborne or satellite platforms.

If WIPP is used only as a geological repository for the disposal of TRU wastes, then a scientific opportunity to advance our understanding of this unique major facility will be missed. In addition to being a repository, WIPP can be an important laboratory for geoscience and sensor technology. Scientific research could be done in conjunction with measures to ensure WIPP’s long-term safety. The results will be indispensable if WIPP is enlarged or if a new salt repository is needed. A previous NRC WIPP study recommended pre-closure monitoring to gain information on the following:

• room deformation, healing of the disturbed zone around the rooms, and performance of shaft seals;

• brine migration and moisture access to the repository;

• gas generation rates and volumes; and

• effectiveness of materials placed around the waste (e.g., MgO) to modify its chemical environment (NRC, 2001d).

Sensors developed and tested at WIPP could then be used for long-term monitoring of other repositories, landfills, and burial sites. These sensors must be robust and have a lifetime of at least 10 years to monitor room closure. They should be controlled remotely and monitored by the Internet and wireless technologies, and they must be self-sufficient, harvesting their energy from the salt or the waste itself.

Microbes are likely to exist and evolve in wastes in the WIPP and in RCRA landfills. A better understanding of these microbes and their activities will help predict the long-term fate of the different waste forms and their components. Microbial activity may destroy or immobilize some waste components while increasing the motility or toxicity of others. Research should focus on specific processes, including biodegradation of the various organic components of the waste and reactions altering the geochemistry of the inorganic components. Research should evaluate biogeochemical factors that can affect the leaching or migration of toxic and radioactive materials in the environment and the effect of physical conditions (e.g., pH, temperature) and chemical composition (e.g., organic and inorganic components, oxygen or other electron acceptor availability) on biogeochemical processes occurring in the waste.

Accelerating site closure, a key feature of Office of Environmental Management (EM) planning since the 1990s, has been emphasized by

EM’s recent top-to-bottom review. Among the areas for EMSP research recommended by the committee, research in characterization that would expedite shipping wastes for off-site disposal is most likely to provide immediate payoffs. Research toward methods for treating wastes that do not meet shipping or disposal criteria might provide similar near-term payoffs.

Nevertheless, closing the larger DOE sites will require decades. Problems that are not foreseen or appreciated today are likely to be encountered in buried waste retrievals. Monitoring the WIPP during its operational period is a unique scientific opportunity. Demonstration that WIPP behaves as expected could be invaluable as DOE seeks to open other geological waste repositories. These opportunities for the long-term, breakthrough research envisioned by Congress should not be overlooked in the rush to meet short-term needs.