Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Geologic Problems at Low-Level Radioactive Waste-Disposal Sites INTRODUCTION JOHN B. ROBERTSON U.S. Geological Survey AB STRACT Less-than-desirable geohydrologic containment has occurred at three commercially operated and three De- partment of Energy-operated low-level radioactive waste-disposal sites in the United States. Studies of these sites indicate that the problems fall into eight general categories: "bathtub eject" (water accumulation in filled trenches), trench-cap integrity, erosion, high water table, hydrogeologic complexity, flooding, complex leachate chemistry, and rapid radionuclide migration in groundwater. Problems have been encountered in both high-permeability and low-permeability burial media. All of these problen~s appear avoidable by applying the following more practical, comprehensive, and commonsense earth-science guidelines for site selection and design: · A very arid environment eliminates most problems. · The bathtub effect can be avoided by using physically stable waste forms and by improving the design of the cap. · Acceptable humid-zone sites can be constructed in permeable media if the water table is sufficiently deep and capillary forces (the "wick effect") are used to divert percolating water from the waste. · An important factor is to select sites in relatively simple geohydrologic environments to facilitate the prediction of their containment properties. The United States has been generating low-level radioactive waste since the "atomic age" dawned in the 1940s. Most of this waste has been disposed of by crude, shallow-land-burial tech- niques, although prior to 1970, significant quantities were also dumped at sea. The term "low-level waste" is a catchall clas- sification lacking specific definition; it includes a variety of radioactive materials that do not fall into one of three other more specifically defined categories of waste: high-level wastes, transuranic wastes, and uranium mill tailings. Some "low-level" wastes are extremely radioactive and may contain relatively large quantities of fission products, with half-lives longer than 25 yr; strontium-90 is an example. Before 1970, low-level wastes were also allowed to contain significant quantities of long-lived 104 transuranic isotopes such as plutonium-239 (half-life of 24,000 years>. Until 1962, all low-level waste was disposed of by the federal government at federally operated facilities such as Oak Ridge National Laboratory, Tennessee. With the commercialization of nuclear power and expanded use of nuclear medicine and other waste-generation activities, the private sector was given the responsibility for low-level waste disposal, with state and federal regulation. Between 1962 and 1967, five commercially operated shallow- land-burial sites for low-level waste opened for business at Beatty, Nevada; Maxey Flats, Kentucky; West Valley, New York; Richland, Washington; and Sheffield, Illinois (Figure 7.1~. Three of those five sites are now closed because of various technical and legal problems. A sixth commercial site was opened
Low-Level Radioactive Waste-Disposal Sites / jeering Laboratory,~Argonne/7 West Valley \Beatty/ I Sheffield ~ ~J': ~ _ ~a~¢ 3~ , | Los Alamos l_> ~ ~ Oak Ridge~Savannah River ~ ~ ~ il,~l ~ K~7 ~ · DO E Storage and Disposal Site · Commercial Disposal Site 0 200 400 600 800 1 000 I 1 1 1 1 1 Miles in 1971 at Barnwell, South Carolina, and currently remains open. That site plus the Beatty and Richland sites now handle all of the nation's commercially generated low-level wastes, which amount to some 75,000 m3/yr. In addition, federal government nuclear research and de- fense activities generate approximately an equal volume of low- level wastes per year, which is buried at five major U. S. De- partment of Energy (DOE) facilities and several minor sites (Figure 7. 1~. Because the locations of the three currently operating com- mercial sites do not represent a politically acceptable or geo- graphically equitable distribution of the wastes, and because their limited capacity is not adequate for anticipated waste- generation rates, the need for additional sites has been rec- ognized over the past few years. Congress passed the Low- Level Waste Policy Act in 1980, which mandates that states establish additional sites on a regional basis before 1986. DOE will also require additional burial sites within the next several years. It would seem prudent, therefore, to apply the best earth-science criteria to the screening, selection, and design of new sites. The source of some of our best geohydrologic information for that purpose is the performance record of the older sites. The principal concern, of course, is to protect groundwater and surface-water supplies from contamination. SITE-SELECTION CRITERIA FOR EXISTING SITES During the period when the six commercial disposal sites were chosen, there were no uniform regulations providing compre- hensive site-specific geohydrologic criteria to be applied to the selection and operation of disposal sites. The Atomic Energy 105 FIGURE 7.1 Location of principal low-level radioactive-waste-disposal sites in the United States. Commission had some general guidelines and performance standards for low-level sites but allowed states to set their own standards if they assumed responsibility for regulation sites. It is not clear what specific geohydrologic criteria (if any) were applied to each of the six sites. It is apparent that the criteria were simplistic and that the dominant criterion for the humid zone sites was that they be placed in low-permeability, clay-rich sediments or in shale. Another important criterion was that the site be underlain by easily excavatable material. The West Valley, New York, site is in fairly uniform, clay-rich glacial till; the Maxey Flats, Kentucky, site is in a low-perme- ability (but fractured) shale. The Sheffield, Illinois, site was apparently intended to be in clay-rich glacial till, but it was discovered later that the till contained some permeable, grav- elly sand lenses. The Barnwell, South Carolina, site was placed in sandy, clayey coastal plain sediments, which are somewhat more permeable than the tills and shales of the other eastern sites. For the two arid western sites (Richland, Washington, and Beatty, Nevada) the low rainfall rate appears to have been the dominant geohydrologic criterion applied. Those sites have an average annual rainfall of 165 and 101 mm (6.5 and 4 inches), respectively, and are both situated in mixed coarse-grained unconsolidated sediments (Robertson, 1980~. The DOE sites were apparently selected with even less de- finitive and documented earth-science criteria. The dominant approach appears to have been to locate the sites within the bounds of the facility reservation. Geohydrologic factors appear to have been a secondary consideration. Among the geologic media and hydrologic settings selected for the DOE sites are the following: mixed glacial tills, coastal plain sediments, thin floodplain sediments on permeable ba- salt, coarse-grained glacio-fluvial sediments, fractured perme-
106 able shales, and fractured luff. Permeability of these materials ranges from about 10 - ~ to 10 cm/sec. Annual precipitation rates range from about 101 to 1370 mm (4 to 54 inches). Water-table depths range from less than 2 m to a few hundred meters (Robertson, 1980~. PROBLEMS ENCOUNTERED AT EXISTING SITES Several geohydrologic problems have been encountered at ex- isting sites, which can be partially attributed, in retrospect, to inadequate attention to earth-science criteria in site selection and to inadequate site characterization and design. Nearly all of these problems or shortcomings are related to eight factors: "bathtub effect,'' trench-cap integrity, erosion, high water table, hydrogeologic complexity, flooding, complex leachate chem- istry, and rapid radionuclide migration. Nearly all of these factors are interrelated and interdependent. Bathtub Effect This effect occurs in wet-climate, low-permeability sites such as West Valley, New York, and Maxey Flats, Kentucky. (A1- though the principal medium at Maxey Flats is low-permea- bility shale, the shale contains a few thin beds of fractured sandstone with higher permeability.) Because of differential subsidence, desiccation cracks, and other reasons, the trench- capping material on waste-filled trenches becomes increasingly permeable, thus enhancing the infiltration of precipitation. Water then accumulates in the trenches, sometimes seeping out on the ground surface, carrying leached radionuclides. This prob- lem has often been blamed more on the low permeability of the natural media than on the high permeability of trench caps and backfill material. Trench-Cap Integrity This problem is closely related to the bathtub effect. As wastes and backfill material decompose and compact, settlement cracks, depressions, and holes develop in the capping material, pro- viding ready avenues of water infiltration. Desiccation cracks can also develop in clay-rich caps during extended dry periods, with the same hydrologic effect. E. roslon Erosion has been a problem or is considered to be a potential problem at some sites. At the Sheffield, Illinois, site, for in- stance, unanticipated rapid runoff from large snow accumula- tion caused undesirable erosion and piping problems in 1979 (Tames B. Foster, U.S. Geological Survey, personal commu- nication, 1979~. Questions on potential long-term erosion prob- JOHN B. ROBERTSON lems have been raised at sites such as West Valley, Beatty, and DOE's Idaho National Engineering Laboratory (INEL). High Water Table In a few cases (Oak Ridge National Laboratory and West Valley, for instance), burial trenches were excavated below the water table, thus providing constant submergence and leaching of some wastes. Although that condition might be undesirable, it is not necessarily detrimental if groundwater flow rates are sufficiently slow. In some places at the Oak Ridge site, raising the ground surface by fill material actually induced the upward movement of the water table into the waste (Webster, 1979~. Hydrologic Complexity Hydrologic complexity has been a problem of varying magni- tude at many sites. It simply means the sites proved to be physically more complex than originally anticipated, so that long-term (or even short-term) performance predictions were found to be in error. An example is the complex of glacial- fluvial stratigraphy of the Sheffield, Illinois, site (Foster and Erickson, 1979~. The original array of test wells used to char- acterize the stratigraphy was not adequate to define the dis- tribution of permeable sand units. In another example at the Maxey Flats, Kentucky, site, groundwater flow is controlled by fractures (Zehner, 1979~; although the average flow rate may be very low, it cannot easily be characterized quantitatively. Flooding Flooding has been a problem on the INEL disposal site. On two occasions, in 1962 and 1969, the site has been inundated by local runoff (Barraclough et al., 1976~. Remedial engineering measures have since been taken to reduce the likelihood and magnitude of further flooding. Although INEL is an arid site with 203 mm (8 inches) of annual rainfall, the problem resulted in open burial trenches' being filled with water and consid- erable infiltration of water over the entire site. Flooding was caused by a combination of unusual meteorologic conditions and the location of the site within a topographic basin. Complex Leachate Chemistry Because of the variety and complexity of low-level waste, the chemical characteristics of leachates from buried wastes are comparably variable and complex. Many nonradioactive or- ganic and inorganic compounds are buried with the waste or result from the biological and chemical decomposition of trash materials. This results in unpredictable oxidation states of some nuclides and chemical complexation with chelating agents and other ligands. Such complexes can be more mobile in ground- water than can uncomplexed cations. At the Maxey Flats site,
Low-Level Radioactive Waste-Disposal Sites for example, plutonium has been observed in trench leachates and in groundwater in chemically complexed forms (Cleveland and Bees, 19814. Rapid Radionuclide Migration All the above-mentioned problems can contribute and have contributed to the migration of waste radionuclides away from burial sites at faster rates or in different directions than ex- pected. Contributing to this problem at some sites is the rel- atively high permeability of the media in which the wastes are buried. This has been a concern at Oak Ridge, Maxey Flats, Sheffield, Barnwell, and INEL, among others. Plutonium and other isotopes have migrated laterally through a thin permeable fractured sandstone bed at Maxey Flats. Tritium has migrated vertically and laterally through permeable sand layers at Shef- field and Barnwell, and several isotopes have migrated verti- cally through permeable basalt at INEL. Thus, an apparent dilemma arises: if both low-permeability and high-permeability sites can have problems, are both con- ditions unacceptable or is one preferable to another? POSSIBLE ANSWERS TO THE QUESTIONS None of the problems observed at existing sites has been di- sastrous in terms of harm to human life there is no evidence of public drinking-water contamination nor harmful radiation exposure to humans due to groundwater contamination from these sites. However, the problems are nonetheless undesir- able. Essentially, all of these potential problems at future sites appear amenable to practical solutions by applying more sen- sible and appropriate earth-science criteria to site selection, characterization, design, and operation. Bathtub Effect and Trench-Cap Integrity There are at least three potential solutions to the bathtub effect: 1. Require stable noncompactable waste forms and backfill, combined with more stable, low-permeability trench capping. 2. Place trenches in permeable media above the water table with low-permeability trench cap. 3. Locate the site in a very arid environment. These options are all specified in the proposed Nuclear Reg- ulatory Commission's low-level waste management regulations, 10 CFR Part 61 fFederal Register 46~142>, July 24, 19814. It is generally agreed that low-permeability clay-rich sediments can be good burial media when used conjunctively with improve- ments in waste form and capping technology. Geohydrologic Complexity Problems related to conditions of geohydrologic complexity can, of course, be reduced by avoiding media dominated by secondary permeability features or complex stratigraphy. Com- 107 plexity is relative; consequently, this is a subjective criterion, requiring judgment. High Water Table In addition to specifying minimum depth to the saturated zone, the potential occurrence of a high water table can be avoided by reducing the permeability of trench caps, providing good land-surface drainage, and avoiding large increases in land- surface elevation from backfilling. If the hydraulic conductivity of the burial medium is below 10-6 or 1O-7 cm/see, ground- water flow rates will be slow enough that radionuclide migration is dominated by molecular diffusion. In such circumstances, it is not really important to exclude groundwater from the waste. In Canada, for instance, burial below the water table is per- mitted in glacial clays with low hydraulic conductivity. Complex Leachate Chemistry The problems resulting from complex leachate chemistry can also be reduced by simplifying waste forms; requiring more stable, less leachable wastes; and excluding potential com- plexing agents from the waste. Rapid Radionuclide Migration in Iligh-Permeability Media This problem can be avoided by applying the following guide- lines to sites in high-permeability materials: 1. Waste must be placed well above the water table. 2. Contact of waste with infiltrating water must be mini- mized by stable, low-permeability trench covers or by effective use of the "wick effect." The wick effect results from natural capillary suction of certain types of unsaturated sediments, which draws water away from wastes. This principle has been effectively demonstrated in the field by French researchers (Rancor, 1980~. The Barnwell, South Carolina, site appears to have benefited by this effect, but conclusive evidence is not yet available. CONCLUSIONS The key to effective isolation of low-level radioactive waste (or any hazardous waste) is minimizing contact of water with the waste and minimizing migration rates in groundwater. Perhaps the most fundamental lesson learned from examining the his- tory of earth-science aspects of existing sites is that no single dominant geohydrologic criterion is more critical than others for every site. Criteria for all facets of the disposal system must be considered together to obtain good performance. Reliance on a single criterion, such as burial in low-permeability clay, while ignoring other criteria such as waste form and trench caps, can lead to failure. It is apparent that both the Nuclear Regulatory Commission and the Department of Energy have recognized these problems and the potential solutions. Con- sequently, both organizations are incorporating, in one form
108 or another, the solutions recommended here into their re- spective regulatory criteria and technical guidelines. Because of lessons learned from earlier and current disposal practices, the next generation of low-level radioactive-waste-disposal sites promises to be much more reliable from an earth-science point of view. REFERENCES Barraclough, J. T., J. B. Robertson, and V. J. Janzer (1976). Hydrology of the solid waste burial ground, as related to the potential migration of radionuclides, Idaho National Engineering Laboratory, U.S . Geol. Surv. Open-File Rep. 76471, 183 pp. Cleveland, J. M., and T. F. Bees (1981). Characterization of plutonium in Maxey Flats radioactive trench leachates, Science 212, 1506-1509. Foster, J. B., and J. R. Erickson (1979). Preliminary report on the JOHN B. ROBERTSON hydrogeology of a low-level radioactive-waste disposal site near Shef- field, Illinois, U.S. Geol. Surv. Open-File Rep. 79-1545, 87 pp. Rancon, D. (1980). Application de la technique des barrieres capillaires aux stockades entranchees, in Proceedings of an IAEA-NEA Sym- posium on Underground Disposal of Radioactive Wastes, Otaniem`, Finland, July 2-6, 1979, I, pp. 241-265. Robertson, J. B. (1980). Shallow land burial of low-level radioactive wastes in the USA, in Proceedings of an IAEA-NEA Symposium on Underground Disposal of Radioactive Wastes, Otaniemi, Finland, July 2-6, 1979, 11, pp. 253-269. Webster, D. A. (1979). Land burial of solid waste at Oak Ridge National Laboratory, Tennessee: A case history, in Management of Low-Level Radioactive Waste, M. A. Carter, A. A. Moghissi, and B. Kahn, eds., Pergamon Press, New York, pp. 731-745. Zehner, H. H. (1979). Preliminary hydrogeologic investigation of the Maxey Flats radioactive waste burial site, Fleming County, Ken- tucky, U.S. Geol. Surv. Open-File Rep. 79-1329, 66 pp.