Characterization of Remote-Handled Transuranic Waste for the Waste Isolation Pilot Plant: Interim Report (2001)

Transuranic waste is radioactive waste containing alpha-emitting radionuclides of atomic number greater than 92, half-life greater than 20 years, and activity greater than 100 nanocuries per gram of waste. Transuranic waste is composed of materials such as clothing, tools, debris, other disposable items, as well as sludge. The waste was produced during the processing of nuclear materials for defense purposes and the cleanup of various nuclear weapon sites across the nation. Alpha radiation is the primary radiation-related health hazard associated with TRU waste. Alpha radiation cannot penetrate human skin but poses a potential health hazard if alpha-emitting particles are inhaled or ingested in large enough quantities. The main alpha-emitting radionuclides in TRU waste are plutonium-238, plutonium-239, plutonium-240, and americium-241. In addition to alpha radiation, TRU waste also emits gamma radiation, which can penetrate the human skin and requires shielding during transport and handling. Gamma radiation is due mainly to fission and activation products, principally from the progenies of cesium-137 and strontium-90, and from cobalt-60.

Transuranic waste is classified as contact-handled (CH) or remote-handled (RH), according to the dose rate at the package surface. According to the legal definition, “the term ‘contact-handled transuranic waste’ means transuranic waste with a surface dose rate not greater than 200 millirem per hour. The term ‘remote-handled transuranic waste’ means transuranic waste with a surface dose rate of 200 millirem per hour or greater”¹ (Congress, 1992). Contact-handled TRU waste typically emits relatively little gamma radiation. Therefore, it can be handled directly by workers. Remote-handled TRU waste emits higher levels of gamma (penetrating) radiation. Therefore, gamma rays represent the main radiological health hazard for workers handling RH-TRU waste. Although alpha-radiation exposure has a greater health risk per unit energy deposited dose, the likelihood that workers will be exposed to alpha radiation is smaller compared to the likelihood for exposure to gamma radiation, because of the smaller penetration power for alpha radiation.² Therefore, RH-TRU waste should not be handled directly by workers and requires heavy container shielding and/or remote-handling equipment.

Transuranic waste is further classified as non-mixed or mixed. Mixed TRU waste contains both radioactive material regulated under the Atomic Energy Act and hazardous waste material regulated under RCRA. Examples of hazardous waste material are ignitable, corrosive, reactive, and toxic substances.

DOE also makes a distinction between retrievably stored waste and newly generated waste, according to the waste generation period. Retrievably stored waste is

waste produced after 1970³ but prior to the implementation of an approved RH-TRU waste characterization plan. DOE is in the process of defining this waste characterization plan, through Documents 1 and 2, to obtain EPA’s and NMED’s approval. Newly generated waste is waste that is produced after the development, approval, and implementation of the waste characterization plan and meets the characterization requirements set forth by the regulatory agencies. Newly generated waste may be waste yet-to-be generated, such as waste at the Hanford Site (see below), or may be existing waste that needs to be re-packaged in a suitable form for transportation and disposal.

For the newly generated waste total inventory (1,400 m³), the information constituting acceptable knowledge will be collected during the packaging or the re-packaging of the waste. According to DOE, AK for newly generated waste will meet the characterization requirements developed under the new RH-TRU waste characterization plan: “Waste that has to be repackaged or is being generated from a process line or decontamination/decommissioning can be generated in a way that supports disposal. Generating the data to support waste disposal to meet quality assurance requirements (i.e., two operators involved in data generation per process line; one to produce data and the other to validate) negates the need for nondestructive examination” (DOE-CBFO, 2000, section 2.3.1.6).

Figure 1 shows the geographic location of the major RH-TRU waste generator sites in DOE’s nuclear weapons complex. These are the following:

• Hanford Site, Washington,

• Idaho National Engineering and Environment Laboratory (INEEL),

• Los Alamos National Laboratory (LANL), New Mexico, and

• Oak Ridge National Laboratory (ORNL), Tennessee.

The figure also shows relative volumes of RH-TRU waste compared to CH-TRU waste for these sites. Tables 1 and 2 show RH-TRU waste inventories in terms of volume and radioactivity. These data show that there is substantial variability among generator sites concerning waste volumes and radioactivity contents.

Table 1 shows the RH-TRU waste inventory of retrievably stored waste, newly generated waste, and planned volumes of waste to be shipped to WIPP. The inventory of retrievably stored RH-TRU waste at all DOE sites is 2,197 m³. A volume of about 1,400 m³ will be generated in the future, for a total of about 3,598 m³ of RH-TRU waste. Further volume-reducing operations (see below) will decrease the waste volume to be emplaced in WIPP to 1,964 m³, as shown in the “Planned Volume” column of Table 1.

The RH-TRU waste volume inventory comprises between 1 and 4 percent in volume of the total (CH-TRU plus RH-TRU) inventory for the WIPP facility (175,564 m³). The Land Withdrawal Act, governing emplacement of TRU waste at WIPP, limits RH-TRU waste volume to 7,080 m³. According to Table 1, the volume of RH-TRU waste planned to be emplaced in WIPP is well below the statutory limit.

Tables 1 and 2 do not show the uncertainties in the inventories of RH-TRU waste at the different sites. These uncertainties arise from changes in the waste management or treatment plans. For example, DOE’s data for the Savannah River Site (SRS), in South Carolina, originally indicated a large RH-TRU waste inventory, reflecting disposal of materials stored in the separations canyons as TRU waste. However, SRS now plans to send this material to the high-level waste tanks, and process it with other high-level wastes, i.e., transform it into glass waste forms in the Defense Waste Processing Facility, for eventual disposition in a federal high-level waste repository. The uncertainties can also be significant for the sites that have yet to produce their RH-TRU waste, such as Hanford (see below).

FIGURE 1 Geographic location of the major RH-TRU waste sites and the WIPP. There are several other sites throughout the nation storing CH-TRU waste. SOURCE: DOE-CBFO, 2001c.

Table 1 shows that most of the retrievably stored RH-TRU waste is located at ORNL (1,306 m³). This volume represents nearly 60 percent of the retrievably stored volume, and 36 percent of the total volume (3,598 m³) of RH-TRU waste. About two-thirds of the retrievably stored waste at ORNL is wet sludge and one-third is debris. The sludge will be dewatered, dried, and packaged for shipment in a special facility, currently under construction.

Debris waste consists of hot cell⁴ and glovebox debris packaged in shielded containers. This waste will undergo volume-reduction operations during packaging. To meet the milestones set in the Federal Facility Compliance Act between DOE and the State of Tennessee, ORNL decided to move forward with the characterization facility to treat or re-package its RH-TRU waste and characterize it at the same time according to the same criteria set forth for CH-TRU waste characterization (EPA, 1992). Before treatment, homogeneous samples of wet sludge will be characterized by radiochemical assays performed in a laboratory on-site. Debris will be characterized by visual

examination and non-destructive examination⁵ inside a hot cell during re-packaging. This treatment will reduce ORNL RH-TRU waste volume from 1,594 to 453 m³. Therefore, most of RH-TRU waste at ORNL is considered “newly generated waste” and will be characterized at the time of its generation.

Table 1 also shows that most of the newly generated waste will be produced at the Hanford Site, where the RH-TRU waste inventory represents 37 percent of the newly generated volume of RH-TRU waste and 26 percent of the total volume. The RH-TRU waste at this site consists mostly of sludge on the bottom of the fuel pools, or equipment inside waste tanks, such as pumps, mixers, and pipelines, or RH-TRU waste buried in caissons and drums.

At INEEL, the 185 m³ of retrievably stored RH-TRU waste is composed mostly of solids generated during the destructive examination of irradiated experimental fuel pins in a hot cell.⁶ Packaging of this waste will increase its total volume, as indicated in the table. The 99 m³ of retrievably stored RH-TRU waste at LANL was characterized and packaged between 1989 and 1994. LANL characterized this waste using the same characterization plan approved for CH-TRU waste, which consists of non-destructive assay methods, radiography, and radiochemical analyses. Characterization operations were performed in a hot cell, except for small quantities of waste analyzed for radiochemical composition in a laboratory. Therefore, RH-TRU waste at LANL will probably not need to be re-characterized because the methods used are in compliance with the already approved CH-TRU characterization plan.

According to the information provided by DOE-CBFO, about 80 percent of the RH-TRU waste inventory (retrievably stored waste plus newly generated waste) will meet the characterization requirements in the new RH-TRU waste characterization plan. In fact, this waste consists of yet-to-be-generated waste (as the waste at the Hanford Site), waste that will be re-packaged (as the sludge at ORNL), or waste that was characterized according to the approved CH-TRU characterization plan (as at LANL). For the remaining 20 percent of RH-TRU waste, DOE proposes to complement the existing AK, where necessary, with additional confirmatory activities (see Chapter 4). In the National Transuranic Waste Management Program DOE writes about retrievably stored waste: “Waste that has already been generated must undergo extensive characterization to meet the requirements of the WIPP Waste Analysis Plan to meet certification requirements for disposal (for instance, nondestructive examination, RCRA constituent sampling, analysis of homogeneous waste, and visual examination)” (DOE-CBFO, 2000).

Table 2 shows that the retrievably stored RH-TRU waste activity is approximately 660,000 curies. The data show that 88.6 percent of this activity is in the retrievably stored RH-TRU waste at ORNL. Most of the radioactivity in RH-TRU waste is due to short-lived radionuclides such as cobalt-60 (half-life 5.26 years), plutonium-241 (14.4 years), strontium-90 (half-life 28 years), and cesium-137 (half-life 30 years). Therefore, in approximately 300 years, the radioactive content of RH-TRU waste due to gamma

emitters will have decayed to approximately the same level as that of CH-TRU waste. After this period, RH-TRU waste will be virtually indistinguishable from CH-TRU waste.

In the Appendix BIR (Baseline Inventory Report) of the Compliance Certification Application (CCA), DOE calculated the total activity of the RH-TRU waste inventory (from retrievably stored waste and newly generated waste) to be approximately one million curies (DOE-CAO, 1996). Therefore, the estimated amount of RH-TRU curies to be emplaced in WIPP is considerably lower than the total amount of curies from RH-TRU waste allowed by the Land Withdrawal Act (5.1 million). The total activity to be emplaced in WIPP from CH-TRU waste is approximately 6.4 million curies. DOE calculated that the percentage of total activity from the RH-TRU waste inventory represents approximately 14 percent of the total activity expected in WIPP (from CH-TRU and RH-TRU waste). However, the percentage of activity due to long-lived TRU radionuclides in RH-TRU waste is only 0.5 percent of the total activity due to long-lived TRU radionuclides in both CH- and RH-TRU waste.⁷ Only long-lived TRU radionuclides are relevant to the long-term performance of the repository because the regulatory compliance period is 10,000 years and most of the radioactivity from fission products in RH-TRU waste will decay away in approximately 300 years. Therefore, RH-TRU waste represents a small fraction of the total TRU inventory in WIPP, both from the point of view of volume and TRU activity.

Disposal of RH-TRU waste entails a number of activities at the generator sites and at WIPP, including the following:

• waste processing,

• waste characterization,

• packaging or re-packaging,

• local waste storage,

• loading of waste transportation casks,

• transportation from the generator sites to WIPP, and

• underground emplacement in WIPP.

The main issues concerning the handling of RH-TRU waste are potential radiation worker exposure and the associated costs (DOE-CBFO, 2001d). The predominant exposure pathways for workers from routine waste handling activities are direct exposure to external penetrating radiation or inhalation from surface contamination, and inhalation of airborne radioactive material generated during characterization and handling activities. Exposure pathways for workers under accident conditions include direct exposure to external penetrating radiation and inhalation of airborne particulates and gases due to accidents related to mechanical or thermal insults of waste containers (Restrepo and Millard, 2001). Because of the high surface dose rates, all of the handling operations involving RH-TRU waste are performed in hot cells, where available, and/or by remote-handling equipment. Therefore, the cost associated with the handling of RH-TRU waste is also an important issue.

Waste processing activities consist of converting the waste to a form suitable for disposal in WIPP. For instance, wet sludge must be dried first because liquids are not allowed in WIPP (if more than one percent of waste volume). Waste characterization is necessary to receive permission to ship the waste to WIPP. DOE is proposing a characterization plan, described in Chapter 4, that relies mainly on information collected about the waste, called acceptable knowledge (AK), and reduces the use of confirmatory methods. According to DOE, this would reduce the potential for worker exposure from intrusive sampling and analysis activities as well as characterization costs (Document 2, Supplemental Information).

Packaging or re-packaging will be carried out on waste that has not yet been generated (packaging) or on retrievably stored waste that has been packaged in the past but the container does not meet transportation requirements (re-packaging). Visual examination of solid waste during the re-packaging step is one of the most common characterization techniques performed in hot cells. As the waste is removed from the old container and sorted into new containers, a videocamera records on tape the items on the table while an operator describes them. This characterization method requires skilled operators to describe the waste, since it is done entirely by remote methods (see committee’s Observation 3 in Chapter 5). Volume-reducing operations, such as mechanical waste compacting, are also performed during the re-packaging process. Once the waste is packaged or re-packaged, it must be stored on-site until it receives EPA’s and NMED’s (the latter only for mixed waste) approval for shipment. Finally, RH-TRU waste must be loaded onto transportation casks,⁸ transported to WIPP, and emplaced into horizontal boreholes in the walls of the disposal rooms.

DOE is concerned about the potential exposure of workers to radiation during the characterization and handling of RH-TRU waste (DOE-CBFO, 2001d). As supplemental information for Document 2, DOE presented a report evaluating risk/cost-impacts associated with RH-TRU waste characterization options compared with characterization of CH-TRU waste. Options included full characterization, as done for CH-TRU waste, AK, and visual and/or radiographic examination. Characterization activities associated with CH-TRU waste include non-destructive assay, headspace gas sampling, visual examination/radiography, solid coring and sampling, container venting, and gas-generation testing (Appendix A). The analysis addresses risk based on the dose to workers associated with routine characterization activities and risk from accidents (Restrepo and Millard, 2001).

According to this analysis, worker doses do not increase substantially with increase in characterization of RH-TRU waste, assuming that adequate facilities and equipment will be used for characterization. However, the risk to workers from postulated accidents involving RH-TRU waste increase by factors of 2 for AK, 4 for visual examination/radiography, and 10 for full RH-TRU characterization compared to a normalized CH-TRU risk of 1.0. Using AK to characterize RH-TRU waste would minimize worker doses and risks (Restrepo and Millard, 2001, page ii).

The findings of the risk/cost analysis report are consistent with the committee’s observation that, because most operations involving the characterization of RH-TRU waste are performed remotely, workers may receive comparable radiation doses characterizing RH-TRU waste to those received in the characterization of CH-TRU waste performed outside a hot cell. Moreover, it is not clear whether workers will be exposed to radiation specifically during the characterization step, given that there are other potentially more hazardous steps involved in the complete handling of RH-TRU waste at the generators’ sites, such as packaging, handling, and transporting the waste. While data on worker exposure during characterization of CH-TRU waste is available, data on worker exposure during the characterization of RH-TRU waste are scarse.

As mentioned previously, the RH-TRU waste inventory at LANL has already been characterized and it is now stored on-site waiting for permission to be shipped to WIPP. Dose data pertaining to the characterization step per se have not been collected because of the negligible doses involved in the operations in the hot cell or on small samples in the laboratory. Staff from LANL estimated that the risk of radiation exposure is negligible during the characterization of waste, rather it is from the handling of containers outside the hot cell.⁹

The Battelle Columbus Laboratories (BCL) in Ohio is the only site currently characterizing RH-TRU waste. BCL developed its own RH-TRU waste characterization plan, which involves visual examination of the waste during re-packaging. The re-packaging step, which occurs in a hot cell, is necessary because the current RH-TRU waste containers do not meet transportation requirements. The estimated collective dose to workers handling RH-TRU waste, as determined by BCL health physics staff, is 100 person-mrem (1 person-millisievert) per container.¹⁰ This is a cumulative dose, which includes waste characterization, as well as the handling of the containers for onsite storage. Several workers are involved in the manipulation and transfer of each container. BCL also monitored for radiation exposure to individual workers during container handling and manipulation; for instance, the lid and seals of waste containers

must be attached manually outside of the hot cell. A typical dose to a worker is 10–20 mrem (0.1–0.2 millisievert) per container. According to the “Supplemental Information” attachment of Document 2, the estimate for worker dose resulting from the characterization of CH-TRU waste is approximately 15–20 mrem per container (Restrepo and Millard, 2001). This is based on DOE and generator data estimates of 15–30 contacts and handling per container, 5–10 mrem per hour contact dose rate, and assuming a reduction factor of 5 for workers conducting routine operations at some distance from the waste containers. These data show that exposure at BCL (10–20 mrem per container) is comparable to the exposure for the characterization of CH-TRU waste.

It is unclear how the exposure incurred during the characterization phase compares to that incurred in other handling operations. It is also unclear how representative BCL worker exposure data are for other RH-TRU waste generators. For instance, container handling methods and procedures (the key determinant of worker dose) may be different at other generator sites. However, data from BCL are informative because this is the only site currently characterizing RH-TRU waste.

DOE is also concerned about the costs associated with the characterization of RH-TRU waste if performed following the same characterization plan approved for CH-TRU waste. This plan is summarized in Appendix A. Briefly, the CH-TRU characterization plan requires sampling and analysis to confirm AK and the application of some characterization techniques on every container. According to DOE and the generator sites, the characterization of a RH-TRU waste container under the CH-TRU waste characterization requirements is estimated to cost in the range of $20,000 to $300,000 per container (Restrepo and Millard, 2001). In contrast, the cost for characterizing a CH-TRU waste is about $3,800 per container¹¹ (DOE-EM, 2001). The large range in characterization cost estimates is due to the variability in the waste content and the characterization processes among the sites. In addition, these cost estimates are not site-specific and take into account only a few characterization scenarios. According to the risk/cost impact analysis in Document 2, “costs incurred for full characterization of RH-TRU waste relative to CH-TRU costs increase by about an order of magnitude for full characterization to $300,000 per container. Infrastructure costs increase from $10,000 for CH-TRU to $100,000 per container for full characterization of RH-TRU waste…increased costs are due to additional storage, retrieval, handling, and characterization times necessary to compensate for increased radiation exposure conditions” (Restrepo and Millard, 2001, page ii and 19).¹²