Disposal of high-level nuclear waste above the water table in arid regions

United States Department of the Interior

Geological Survey

TABLE Hydrologists divide the subsurface into two distinct environments, a zone of ground-water saturation whose upper surface is the water table and an unsaturated zone that extends from the water table to the land surface. The United States Government's program and all foreign programs for developing permanent disposal facilities for high-level nuclear wastes have been aimed at placing such facilities in the saturated zone. However, the concept of a repository in the unsaturated zone above the water table now has begun to receive serious attention. The site being considered for the first such repository, Yucca Mountain at the Nevada Test Site, was originally chosen for study because of potentially favorable host rocks below the water table (Dixon and Glanzman, 1982; Dixon and Hoover, 1982; Scott, 1983).

INTRODUCTION

Winograd (1972, 1974) proposed placing nuclear waste in the unsaturated zone and recently summarized the advantages associated with using the thick unsaturated zones of the southwestern United States as potential environments for the disposal of solidified radioactive wastes, with particular emphasis on shallow burial (50-300 feet) of transuranic (TRU) wastes (Winograd, 1981).

A report of the Panel on Hanford Wastes, Committee on Radioactive Waste Management, National Academy of Sciences (1978, p. 3) suggested that a repository for the high-level defense wastes at Hanford, Wash., could be located in the unsaturated zone in a "tunnel or adit driven into the Rattlesnake Hills. Technical appraisal of both alternatives (that is, saturated and unsaturated zone repositories) is urgent but immediate em-

added]. A repository in the hills has the potential advantages of lesser cost, a drier environment and a longer path of travel for ground water to reach the biosphere.'' The panel also noted that:

Storage away from reactors has been proposed as a temporary measure when the onsite storage pools of electric utilities fill up with spent fuel and when repositories are not yet available. Zen (1980) recently has summarized other reasons for considering interim storage. At the present time, repositories in the United States are being designed to accept either spent fuel or reprocessed waste. Spent fuel contains both plutonium and uranium that could be extracted and further used as reactor fuel. There are other elements and isotopes present that could be useful if their extraction became economical. If we bury spent fuel, we may be burying resources that an energy-hungry and more technologically advanced world may desire to retrieve in the future. Will we have placed these where they can be exhumed only at a high cost and at risk to human life? Other benefits of interim storage are reducing the thermal output of the waste prior to disposal, reducing the radiation hazards of reprocessing, and keeping options open for future methods of disposal such as transmutation or disposal in space.

The unsaturated zone offers the possibility of fully retrievable disposal. Fully retrievable disposal, as proposed here, combines both retrievable storage and permanent disposal; the waste is at its final resting place, and removing the waste is a planned option.

A certain degree of retrievability is already required in the design of a repository. NRC-proposed rule 10 CFR Part 60 (NRC, 1981) requires that the system be designed with the option to retrieve the waste for up to 50 years, should the repository turn out to be unacceptable in its performance. Evaluations of how one might retrieve waste from typical repositories in the saturated zone (Wilems and others, 1980; Kaiser Engineers, Inc., 1980) indicate that problems could be formidable and the costs high. Thu~ it appears likely that, in the saturated zone, retrieval would be attempted only under very ~ circumstances, once a section of the repository had been filled with waste and backfilled. Even though retrieval is very unlikely to be needed, designing to allow for it increases the cost of the repository; the increase in cost rises with the duration of the desired period of retrievability (Wilems and others, 1980, p. 75).

As I shall show subsequently, the unsaturated zone probably could provide both permanent safe disposal and the option to retrieve the waste without great difficulty or danger, should reprocessing become a viable option at some time in the future or should the proposed repository prove unsound for some unforeseen reason.

In areas of moderate rain fall, the unsaturated zone can range from zero to several hundred feet in thickness. Water soaking into the land surface moves downward through the unsaturated zone to the water table. In desert areas, the unsaturated zone can extend to depths of 2,000 feet or more. In a typical desert environment, the few inches of annual precipitation is dissipated mostly as surface runoff, turning the dry gulches and washes into torrent-filled flumes for brief periods. The small amount of the precipitation that does soak into the soil is largely (and in warm or dry years, entirely) returned to the air by evapotranspiration (see, for example, National Academy of Sciences, 1978, p. 147-150).

The behavior of the water that does enter the subsurface is determined by the properties of the underlying rocks. If the underlying rocks are highly fractured, the movement of water downward through the unsaturated zone can be quite rapid. For example, Clebsch (1961, p. C124) estimated, by the tritium concentration method, that the age of ground-water samples from the U12e tunnel located 500 feet vertically below the east slope of Rainer Mesa and about 1,200 feet diagonally below the caprock (Thordarson, 1965, p. 74) was in the range of greater than 0.8 year but less than 6 years. On the other hand, in highly porous media with high bulk capillarity, the vertical movement of water becomes so slow that it is questionable whether movement occurs at all. Much of the water within the pore spaces of the rocks is simply held by capillary forces, moving neither up nor down, even though the rock may be nearly saturated; that is, the rock cannot accommodate additional water in its interstices (Thordarson, 1965, p. 26). Tunnels through such

SATURATED ZONE

absence of unfavorable features. Most exploration of the subsurface of a candidate site in the saturated zone has to wait until an exploratory shaft has been completed and the proposed host rock can be explored by horizontal drill holes or tunnels.

Winograd (oral commun., 1982) has pointed out that, in the unsaturated zone, if the host rock already has a high vertical permeability, there is no particular reason to limit the number of drill. holes from the surface. Thus the proposed repository could be explored, if necessary, like any ore body or coal bed, with drill holes every few hundred feet on a rectangular grid. Sealing of drill holes will be discussed under Repository Design,

ploration of any repository site has the potential for uncovering previously unsuspected features such as fracture zones, faults, solution features, and other possible pathways for water to the repository horizon. Depending on their nature and extent, such features might even render a potential site unlicensable.

Winograd has pointed out (oral commun., 1982) that the most unfavorable feature that might be encountered in the unsaturated zone would be a large zone of perched water, where a relatively impermeable bed or zone has collected a large amount of descending water. Such features have been encountered at the Nevada Test Site but have largely drained in a matter of months (Thordarson, 1965; Ege and others, 1980). At the worst, a number of holes might have to be drilled to drain such a zone. Once drained, such a zone would no longer constitute a hazard. Any continuing flow would be relatively small. If the site were explored with a large number of drill holes, one would expect that at least some of these would penetrate any zones of perched water that might be present and would help to drain them. In this regard, numerous exploratory drill holes would actually be desirable to demonstrate that no such zone was present.

If previously unknown faults or fracture zones are encountered in the unsaturated zone during the characterization or construction phases of the repository, the faults or fracture zones should not constitute a serious problem because of the low average flux of water (Winograd, oral commun., 1982). Below the water table, such features would be filled continuously with water from a virtually unlimited source; in the unsaturated zone, they would behave more like storm sewers (see fig. 1). At worst, such features might constitute a kind of "down spout" for surface drainage during unusual precipitation events. Consequently, they could almost certainly be handled by appropriate design of the repository drainage or, at the very worst, by sealing off tunnels that encountered them. Such a seal would need only to keep most of the water confined to the fracture zone during any precipitation events heavy enough to produce a temporary flow of water. There should be opportunity to evaluate the consequences of such precipitation events during the operational phase of the repository.

unsaturated zone. In the saturated zone, water at the level of a repository would probably be under a head of pressure roughly equivalent to the depth below the water table. The thermal load of a nuclear waste repository may induce new fractures and modify existing ones; near the waste, fractures might be closed initially by thermal stress but would tend to open on cooling. The problem of predicting the distribution, permeability, and extent of such fractures, the effect of the fractures on the rock strength, and the behavior of water in them becomes very complex because, as the water moves, it also transports heat. The ultimate objective is to avoid having thermally induced and existing natural fractures form a network that would greatly increase the access of ground water to and from the repository or that would result in a shortened flowpath to the accessible environment.

In the unsaturated zone, the interactions would be essentially thermal-mechanical. Hydrologic interactions would be minimal or absent because any water in the proximity of the repository during the thermal phase would be at pressures no greater than atmospheric. The water would simply evaporate and flow away from the waste (Geotechnical Engineers, Inc., 1979, p. 67; Evans and others, 1980). If the concentration of vapor became high enough, the vapor would recondense and accumulate in cooler rocks farther from the waste. The The direction of ground-water movement through a repository in the unsaturated zone and its environs is known, and the maximum amount that might reach the waste is readily bounded. In fuel assembly would intercept an average volume of water of about 3 liters per year, or about 40 times the volume of water it would encounter in a vertical position. If several assemblies were stacked one above another, this difference would

Of course, any descending water will not actually pass downward as an evenly distributed sheet. Local differences in vertical permeability may cause some lateral movement, with water becoming more concentrated in some areas and diminished in others. However, over the horizontal area occupied by a repository, these differences should average out. Note also that the above estimates of water that would actually contact waste make. no allowance for engineered barriers and drains aimed at reducing this contact. As mentioned previously, the heat generated by the waste also would tend to keep water away from the waste through evaporation and recondensation during the thermal pulse.

As mentioned previously, very young water has been encountered locally in the unsaturated zone; the presence of young water indicates a short downward ground-water transit time between the surface and the point of measurement (still in the unsaturated zone) for some of the water. If young water occurs locally in a few through-going fractures or fracture zones, its short transit time could give an erroneous impression of the repository's capability to contain nuclides. In such cases, calculations of ground-water transit time should take into account the relative volumes of young water and older water passing through the repository to obtain an average transit time. The effectiveness of drains in isolating any younger water from the waste may be a further consideration.

workings are available, observations can be made of the actual amount of flow into these workings; such measurements could be continued over the entire operational phase if necessary. Similarly, observations can be made to determine the effectiveness of the drainage systems and other engineered barriers to reduce or to prevent contact of water with the waste. Thus, estimates of the amount of water that might reach the waste ean ultimately be verified by actual experience and direct measurements over a period of decades.

In the saturated zone, direct measurements of the amount of water moving through and contacting waste in a sealed flooded repository can never be made. For a repository in salt, the amount of water moving through the repository should be negligible if the seals are adequate. For crystalline rocks, the amounts of water entering a repository during the operational stage can be measured, but

repository filled with water at the prevailing pressures of the surrounding ground water. Estimates of the amount of water contacting the waste will have to be based on the estimates of regional flow developed from measurements in wells surrounding the site and the physical properties of the rocks. Such estimates can never be verified by direct observations.

NRC 10 CFR Part 60 specifies that containment of nuclides within the engineered system will be substantially complete for a period of 1,000 years after permanent closure. Following this period, the release rate of any radionuclide shall not exceed 1 part in 100,000 per year of the inventory of that radionuclide calculated at 1,000 years after permanent closure. This does not apply to any radionuclide that is released at a rate of less than 0.1 percent of the calculated total annual release at 1,000 years after permanent closure.

The following calculations address the question of whether spent fuel might satisfy the 1/100,000 release rate. They assume that, after 1,000 years of containment in a canister, the average fuel assembly in a repository filled with BWR spent fuel comes in direct contact annually with the amounts of water indicated in the preceding section; that is, the canister and engineered barriers provide no further protection. We will further assume that any water that does contact the fuel assembly will become fully saturated with the radionuclides in question; that is, we will consider solubility rather than leach rate to control dissolution.

In their review of waste package development, Dayal and others (1981, p. 8) state that:

Unfortunately there is much data on leach rates but relatively little on solubilities, especially under the oxidizing conditions that would prevail in the unsaturated zone. Therefore we will proceed by detennining what concentrations of radionuclides in water would be acceptable with an annual release rate of 10- per year and comparing these "permissible" concentrations with the available data on solubilities of the actinides.

In table 1, column A shows the radiation in curies that would be emitted from the major radiation-producing elements of a 1,000-year-old BWR assembly; only those elements that individually would contribute at least 0.1 percent of the total radiation were included, and they account for 286.0 of the total of 287.3 curies, or 99.5 percent. Column B indicates the amount of an element in grams present in the assembly after 1,000 years. Column Cis 10- of that amount or what would be the ''permissible release'' annually under NRC 10 CFR Part 60. These "permissible releases" are then divided by the average annual amount of water that would encounter a vertical fuel assembly (7 5 milliliters per year) or a horizontal assembly (3 liters per year), as calculated in the preceding section (see p. 7, 8), to produce columns D and E, respectively. These represent the hypothetical concentrations that would be needed if the ''permissible release'' were to be transported in the water encountering the fuel assembly.

Column F summarizes the solubility data of Ogard and others (1981) and National Academy of Sciences (1983). Ogard and others performed leaching experiments on spent fuel and compared their results with calculated solubilities of various studies. They found that the ratios of europium, cerium, americium, and plutonium to uranium in the leachants under oxidizing conditions were very much lower than the calculated ratios in the original spent fuel. They concluded that ''These large differences in the ratios can occur if europium, cerium, americium, and plutonium are at their solubility limits and precipitate as some form of hydrous oxide as the uo2 matrix continued to dissolve with time." Their experimental contact them annually is limited and can be estimated; this may allow estimation of the rate of corrosion.

Because the canisters in the unsaturated zone

ground water rather than immersion, a simple overpack over the canister should provide adequate protection. Such an overpack might be made of titanium or even ceramic. It could even be open at the bottom, since steam or water vapor has very little transporting capacity.

In visualizing the saturated and unsaturated environments, the hydrologic conditions surrounding the canisters in crystalline rocks in the saturated zone may be considered somewhat analogous to locating the canisters in sediments at the bottom of a lake (the ground water) that is several hundred to perhaps 2,000 or 3,000 feet deep. In the case of salt, the canisters are under the lake but isolated from it by the essentially impermeable salt; they would be under lithostatic pressure and perhaps surrounded locally by brine. Hydrologic conditions in the unsaturated zone are more analogous to placing the canisters on the surface under gravel or broken rock in an area with extremely low annual rainfall but perhaps high humidity.

age. If water gained access to a repository in the saturated zone, the repository would become flooded, and the waste would be immersed in water. Consequently, it must be assumed that all nonsalt repositories in the saturated zone will eventually be filled with water.

In the unsaturated zone, flooding would be virtually impossible because of the low flux of water and high fracture permeability of the host rock; any water reaching the waste would rapidly drain off. The repository would be designed to minimize possible contact of water with waste package. For example, Winograd (oral commun., 1982) has suggested that canisters could be placed on the floor of the tunnel with flanking ditches to remove water. Alternatively, if canisters were emplaced in vertical holes, the bottom portion of each hole could be filled with coarse material to provide a

2A). The holes could be lined or grouted and collared to reduce the amount of water that might enter from fractures along the sides of the holes. The canister could rest on a pedestal in the hole (E-an Zen, oral commun., 1983) or on the bottom of the hole above a smaller diameter hole through which water could drain (D. E. White, oral commun., 1983). Alternatively, Winograd suggests that the canisters could be emplaced in subhorizontal holes in the sides of the tunnel; the holes could slope slightly upward to provide drainage into the drift so that water would not accumulate in them (fig. 28).

zone, boreholes and shafts need to be carefully sealed so that they do not become future conduits for ground water and radionuclide migration. This is especially important in repositories located in salt because of the high solubility of the host rock. The research being done on this problem has been summarized by D' Appolonia Consulting Engineers (1980, 1981). Wendall Marine (in D' Appolonia, 1980, p. 15-17) identified four issues involved in field tests of seals-the damaged zone around the hole (or shaft) caused by drilling the hole, the interface between the plug and the rock, the plug material and the permeability of the plug itseH, and quality assurance requirements of plug emplacement. He pointed out that "monitoring actual plugs in a repository is a particularly thorny issue because, in the large part, the procedure will be developed and quality assured, and then from that point on it will largely be a matter of faith that the plug is working. To attempt to put monitoring devices in a plug may itseH destroy plug integrity." Ellison and others (1981) provided some schematic designs that give a clear idea of how sealing of tunnels and shafts might be accomplished in bedded salt. At various points, the shaft or tunnel would be enlarged by removing additional material comprising the disturbed zone. The enlarged portions would be filled with concrete, bentonite, or other especially chosen material and would serve as seals or bulkheads; the rest of the tunnel or shaft would be filled with backfill. Roy

ered and pointed out that such materials should be mechanically adequate, bond well to the host rock, be able to- resist permeation of fluids particularly in the interfacial zone, be stable and chemically durable, be compatible with the surrounding rock and ground water, and be resistant to destructive expansion and contraction and to nuclide transport. The above properties and functions are also important over the long time period of repository isolation.

In the unsaturated zone, shafts and boreholes would have an entirely different relation to a repository. Shafts would increase the amount of water moving through the repository only if they diverted a significant amount of surface runoff into the subsurface. Thus they should not be located in the bottom of gullies or washes, which focus runoff; ideally they should be located on drainage divides or hilltops, although these would change with time. The focus of any sealing effort should be near the top of the shaft to prevent ingress of runoff.

FIGURE 2.-Some ways of emplacing canisters in tunnels in the unsaturated zone to allow both circulation of air to remove heat and to allow drainage of water. A, Canister is emplaced in vertical hole with dry well at bottom (Winograd, oral commun., 1982) and clearance between floor and radiation shielding to allow circulation of air. B, Emplacement of canister in subhori· zontal hole (Winograd, oral commun., 1982). C, Emplacement of canisters in holes between tunnels to allow air to move up- wards and water to drain downwards. D, Plan view of C showing angling of holes to reduce width of the tunnels. In all

The average amount of water passing through the repository would be determined by the local recharge; as long as the upper part ·of shafts or holes were sealed, they would make no additional contribution to this flux, although dipping fractures intersecting a shaft would channel flow into it. If the lower parts of the shafts (or holes) were of thermal convection; hot air, being lighter, would rise up the vertical shafts, drawing denser cooler air in through the mouth of the tunnel. The waste canisters would simply be stacked in movable storage racks by remote-handling equipment. Container performance would be monitored by sealing a readily detectable gas between an inner and outer container (a method used in the present experiment in storing waste canisters in granite at the Climax Stock, Nevada Test Site). The tunnel would slope to drain any water out its mouth. Although Hammond said nothing about the water table, he specifically cited the extensive tunnel systems in the unsaturated zone developed for weapons testing at the Nevada Test Site as an example of such a tunnel system ready for use.

Although Hammond's proposed ventilation system draws the cooler air in through the mouth of a tunnel, the same basic concept probably could be applied to a repository connected to the surface only by shafts. The only difference would be that, instead of entering at the mouth of a tunnel, cooler air would enter through a vertical shaft. As in Hammond's concept, the thermal expansion of the air heated by the waste would provide the updraft. Tunnels could be inclined as needed to both encourage the upward movement of warmer air and to promote efficient drainage.

Although Hammond's concept would be acceptable for interim storage, the stability of the tunnels probably could not be assured for the thousands of years desired for permanent disposal. Thus, canisters could not be left in the middle of tunnels where a large rockfall might rupture them. Schronhorst and others (1981) have described conceptual designs for a seH-shielded package that could be transported and emplaced with little or no additional shielding; the packages simply could be placed on the floor of the tunnel. Because such canisters would involve a considerable thickness of steel or cast iron, they might be sufficiently strong to withstand most rockfalls. If not, the tunnels might be filled later with coarse backfill when retrievability was no longer of interest. Damage to canisters also could be avoided by emplacing them in holes drilled in the floor or wall as shown in fig. 2A and B. The hole could be made sufficientlf larger than the canister to allow air circulation. Radiation shielding could be placed across th~ aecess opening in such a way as to allow air to enter and leave the hole. This would have the considerable advantage of keeping the tunnels accessible to men instead of only to remote-controlled equipment, as in Hammond's concept.

I have suggested a slightly more elaborate system in figure 2C and D. The holes containing the canisters would be drilled entirely through from one tunnel to the next and would tilt at a low angle. This would allow any water to drain down· hill into the lower tunnel; at the same time, air would enter from that tunnel, become heated, and move uphill to the upper tunnel. Thus a ventilation system could be designed with alternating cooler and warmer tunnels at slightly different levels. As shown in figure 2D, the holes could be drilled at an angle to the direction of the tunnels if one desired to keep the width of the tunnels through which the waste is transported (presumably the cooler tunnels) to a minimum. Equipment for emplacing the waste would not have to be as sturdy as that

down) a shallow incline. Canisters could have cooling fins and could rest on rails to avoid contact with the bottom of the emplacement hole where water might drain. The hole itseH could be lined with a sleeve to divert any water that might tricklein through cracks and to maintain ease of emplacement.

In addition to economy of operation, a ventila· tion system that is passive or largely passive would have the advantage of continuing to func· tion in the event of a disruption of power.

With increased ventilation, more waste could be emplaced in the same size repository. As pointed out by Hammond (1979), air cooling of a repository provides an additional benefit. An air-cooled repository does not have to be designed to bear the load of thermally induced stresses that the waste would place on a repository if the waste were buried in backfill or snugly emplaced in the walls or floor. Consequently, more waste could be emplaced in the same volume of repository since the thermal loading imposed on the host rock is the main factor that determines the spacing of the canisters. With the reduced thermal loading im· posed on the rocks, multilevel repositories, which have been considered from time to time in the past, could be used. Waste might be stored first in shallower levels with better ventilation and moved to lower levels for final storage.

In the unsaturated zone, tunnels need not be filled with backfilL Backfill apparently is required for a repository in the saturated zone. The draft of contaminated rock (Kaiser Engineers, Inc., 1980, p. 101-107). All of this might have to be carried out in an underground environment that has been heated well above the temperature that existed when the original repository was mined. Because backfilling is not necessary in repositories located in the unsaturated zone and could be deferred in those in saturated crystalline rocks, these difficulties may be avoided, and waste retrieval may be much simplier than in salt.

cial sealing required; even abrupt abandonment of

appreciably. Final decommissioning could be accomplished primarily by simply filling with coarse material all access shafts or tunnels connected to the surface. Backfilling the storage tunnels might not be necessary since we wish to promote drainage rather than slow it. The primary questions would be (1) whether ultimate collapse of the tunnels would result in subsidence at the surface and diversion of surface runoff into the repository and (2) whether backfill is needed to protect the canisters from rupture that might occur as a result of tunnel collapse. Sealing access openings to repositories in the unsaturated zone would be simple, compared to the sealing of openings to repositories in the saturated zone; failure of such seals would be of no particular consequence. Openings would be sealed primarily to keep surface runoff and people out of the repository.

Because decommissioning is relatively simple, it could be readily deferred as long as desired to provide permanently retrievable disposal. This might be desirable to assure safe performance, remove heat through ventilation, or maintain the reprocessing option. In fact, except for the need to keep people from entering the repository, a repository in the unsaturated zone abandoned during the operational period because of some overwhelming catastrophe such as nuclear war would present only a little more hazard to the natural environment than if it were completely sealed. Written warnings to future generations could be placed in the repository passageways during the operational period. A repository below the water table, however, could not be deserted so inconsequentially because abandoned shafts might provide a short cut for nuclide transport by ground water.

PRINCIPAL CONCERNS ASSOCIATED WITH REPOSITORY LOCATION IN THE UNSATURATED ZONE

The preceding discussions have shown that many of the concerns and technical problems associated with locating a repository in the saturated zone become much less important when repositories are located in the unsaturated zone. Conversely, are there any concerns or possible problems that are minor for repositories located in the saturated zone that become of major significance in the unsaturated zone? Winograd (1981) identified four major concerns or possibilities that would have to be addressed if nuclear waste were placed in the unsaturated zone: (1) exhumation by natural processes (erosion, tectonism, or meteorite impact), (2) change to a wetter climate resulting in a major rise in the water table or greatly increased flux of water in the unsaturated zone, (3) exhumation by our descendants, and (4) possible release of gaseous fission products.

The possibility of exhumation by natural processes does not appear to be a serious problem for a mined repository. Winograd demonstrated that the possibility of exhumation by erosion, tectonism, or meteorite impact should not be a serious concern for relatively shallow burial of TRU wastes at Yucca Flat; in this paper we are considering much deeper burial in a mined geologic repository. Draft NRC 10 CFR Part 60 specifies a minimum depth of 300 meters (984 feet) from the ground surface; the water table at Yucca Mountain in well G-1 was at an elevation of approximately 2,455 feet (Spengler and others, 1981), and the land surface at the proposed site is all above 4,100 feet. Thus the acceptable depth for a repository at Yucca Mountain in the unsaturated zone would lie between about 1,500 and 1,000 feet (elevations of 2,600 and 3,100 feet). By contrast, the depth of a repository in the saturated zone is limited only by factors such as rock strength, in situ stress, and temperature.

Increased precipitation will, of course, increase the flux through the repository and may cause a rise in the water table. In addition, the increased flux could produce local perched water tables and increased lateral movement above the more impermeable zones. However, the consequences of an increase in precipitation must be addressed on a site by site basis.

Winograd (1981) reviewed the evidence for past rises of the water table at Yucca Flat, NTS, during the Pleistocene when wetter climates prevailed. Winograd and Doty (1980) studied past deposits of tufa formed by springs during the Pleistocene and concluded that the water table probably did not rise more than 30 meters beneath Frenchmen Flat. Even if the water table under Yucca Mountain were to rise by several times the amount estimated for Yucca Flat, ample range of depth would still exist between the water table and the 300-meter minimum depth requirement. Winograd also reviewed the evidence for wetter climatic conditions during the Pleistocene. Conditions were clearly wetter and cooler, but semiarid conditions must have existed on the valley floors. Annual precipitation was probably no more than 50 percent higher during wetter Pleistocene climates. Because the present annual precipitation is probably on the order of 5 inches per year at Yucca Mountain, a 50-percent increase does not represent a very great amount of water. Most of this would be lost as runoff and through evapotranspiration, which would increase along with the vegetation.

A repository in the saturated zone will be much less likely to be adversely affected by changes in climate. Increased precipitation and rise of the water table could result in some increase in hydrostatic pressures, shortening of traveltimes, or changes in flow paths, but these usually would be of minor consequence.

A future civilization, seeking potable water, might inadvertently penetrate the repository with drill holes. At the Nevada Test Site, however, no water resource is present in the unsaturated zone; the target of such wells would have to be the saturated zone below the repository. Because of the relatively low flux in the unsaturated zone, such drill holes should have virtually no effect on the performance of the repository.

A future civilization might wish to reopen the repository out of curiosity or the hope of obtaining something of value. Such a civilization still would be faced with the problem of gaining access to the repository; that is, removing the fill materials from the shafts (and tunnels if backfilled or collapsed).

Even if the repository were reopened, appreciably more ground water is not likely to come in contact with the canisters because of the dry climate and the low flux of ground water in the unsaturated zone. In addition, the relatively impermeable rocks containing sorptive minerals that underlie the repository horizon and the traveltime of the ground water in the saturated zone would serve as barriers.

The waste packages could be dragged to the surface and used in some environmentally deleterious manner that we cannot foresee. This problem exists for any repository, but repositories in the unsaturated zone might present future intruders with an easier task. Intruders would not have to pump out any water from a repository in the unsaturated zone, and such a repository probably would be at a shallower depth than one in the saturated zone. Ease of intrusion is an unfortunate corollary of ease of retrieval.

The problem of warning future generations was addressed by Cameron (1981) who discussed the type of warnings that could be left on the basis of the kind of information that has been passed on versus the kind that has been lost from past civilizations. He concluded that we can assume that future generations will be able to decipher a warning message left at a repository. Weitzburg (1982) and Kaplan (1982) arrived at similar conclusions.

Spent fuel contains three major gaseous fission products: krypton-85, carbon-14, and iodine-129. These represent three quite different problems because of their differing chemical behaviors and half-lives. Krypton-85 has a half-life of 10.6 years and, being a noble gas, remains uncombined with other elements. Evans and others (1980, p. 76) note that (for krypton-85 removed and concentrated from reprocessed waste) "Engineered storage facilities would be adequate permanent storage of krypton-85 since the krypton will have sufficiently decayed in about 100 years." Thus krypton-85 presents a hazard mainly during the operational period of the repository, if canisters should begin to leak. Krypton-85 could also provide a built-in leak detection system (D. E. White, oral commun., 1983).

Carbon-14 has a half-life of about 5,700 years and is a gas if combined with oxygen (C02 and CO). Thus it is more of an intermediate term problem in geologic confinement. Much stable carbon occurs naturally in the environment to dilute the radioactive isotope, which amounts to about .004 milligrams for a BWR fuel assembly (Majumdar and others, 1982, p. 146). Carbon-14 is produced naturally from nitrogen in the upper atmosphere but is dispersed uniformly throughout the biosphere in a carbon-14/carbon-12 ratio of 10-• Thus about 4 kilograms of dead carbon (older than about 50,000 years) would suffice to dilute 0.004 milligrams of carbon-14 to the natural level, if perfect mixing or exchange could occur.

Iodine-129 has a half-life of about 16 million years; the time required for appreciable radioactive delay would be several times this and far beyond reasonable geologic prediction. However, the contribution of iodine-129 to the total radioactivity of spent fuel is small; in the 1,000-year-old

(about 0.002 percent of the total) from its 36 grams.

The primary concern over iodine-129 stems from the concentration of iodine-129 by the body in the thyroid gland. Evans and others (1980, p. 78) observed that "to maintain the dose to an individual below 75 mrem/yr [milli-roentgen equivalent per year] (EPA regulation), the isotope ratio in the thyroid must be less than 0.0013 * * *. Thus about 800 grams of iodine-127 must eventually be mixed with each gram of iodine-129 to be within the dose requirements for all future generations."

Unlike carbon, iodine usually has very low concentrations in the natural continental inorganic environment, and the requisite isotopic dilution is not readily available. Perel'man (1972) reviewed the geochemistry of iodine. He described it as a typical rare and dispersed element that migrates actively in water and in the atmosphere and concentrates in the biosphere and in living matter. Eight iodine minerals are known, but their occurrence is very rare. Most iodine ends up in the

For reprocessed waste, Evans and others (1980) and Prout and others (1982) examined alternatives for the storage and disposal of krypton-85, carbon-14, and iodine-129 upon their release during the dissolution step in the reprocessing of spent fuel. Prout and others (p. 26) concluded that burial of iodine-129 in deep ocean sediments could provide the requisite isotopic dilution with iodine-127 and that slow diffusion through the sediments along with the movement of deep ocean currents should provide the necessary mixing. Evans and others (p. 11) similarly concluded that "deep ocean sediment disposal of iodine offers the best assurance of isotopic dilution.''

Spent fuel as a waste form was reviewed by Dayal and others (1982, p. 95) who noted that:

Thus iodine-129 presents a problem in both reprocessed and unreprocessed spent fuel. For reprocessed fuel, iodine-129 would be removed from the waste and treated as a separate disposal problem. For unreprocessed fuel, most of the gases possibly could be removed by physical and chemical means short of full scale reprocessing, (for exam-. pie, perforating or chopping up of fuel elements and annealing). This solution is suggested by the following observation made by Johnson (1977, p. 18), in a study of the behavior of spent nuclear fuel in water pool storage:

Iodine-129 (as well as carbon-14) presents somewhat different problems for repositories in the saturated and unsaturated zones. In the saturated zone, the repository is effectively surrounded by ground water that sooner or later would pick up the iodine, probably as an anion with or without oxygen. Although engineered barriers could delay this process (Allard and others, 1980), naturally occurring minerals are unlikely to retard iodine's migration to any appreciable degree. Because of its long half-life, long ground-water traveltimes are of no particular help. The most effective protection would be dilution through slow release, preferably into waters with a high natural iodine content.

In the unsaturated zone, the problem is more complex (Evans and others, 1983, p. 177). Escaping iodine-129 might travel as iodine gas diffusing through rocks, with its movement controlled by the rocks' permeability to gases. Thus part of the iodine-129 might eventually diffuse from rocks directly into the atmosphere where it would mix with the iodine already there.

Reliable data on iodine in the atmosphere appear to be meager. Duce and others (1965) made a very comprehensive study of iodine, bromine, and chlorine in the atmosphere around and over Hawaii. The lowest values they encountered were about 2X 10- grams per cubic meter at several thousand feet on Mauna Loa. If comparable values occurred over Nevada and if perfect mixing were somehow achieved, about 18 cubic kilometers of air would be needed to dilute the 36 grams of iodine-129 in a single fuel assembly with 800 times that amount of iodine-127, assuming the iodine-129 was entirely lost to the atmosphere. Samples of rain from the same localities contained a minimum of about 0.1X10- grams per milliliter of iodine. Thus a square kilometer would receive about 1 gram of iodine per centimeter of rainfall. Diluting 36 grams of iodine-129 by a ratio of 800:1 would require about 3X10-• grams ofiodine-127 or about 300 centimeters (120 inches) of rain over a square 10 kilometers on a side.

These calculations are unrealistic and intended only to try to place the problem in terms that can be readily comprehended. If the iodine cannot be removed from the waste, studies should be made of rates of iodine release from the waste, and models of iodine migration, both as a gas escaping into the atmosphere and as anions moving with the ground water, need to be developed and carefully examined. The same applies to carbon-14, although the problem may be less severe because of the shorter half-life and natural abundance of nonradioactive carbon.

Leakage of radon generated by the waste should also be assessed; however, the hazard associated with a repository should be much less than the hazard presented by radon generated by uranium mine tailings because of the short half-life of radon and because the waste would be far below the surface.

With regard to site exploration and characterization, no apparent technical reasons exist to justify placing any limits on the numbers of drill holes used to explore a candidate repository in the unsaturated zone. Furthermore, it is difficult to conceive of any ''geologic surprises'' that could present serious problems in the unsaturated zone.

With regard to repository design, the problem of sealing shafts and boreholes is of little consequence in the unsaturated zone. Therefore, the number of shafts or openings into the repository need not be limited. Tunnels also do not need to be backfilled. Without backfill and with adequate ventilation, the entire repository could be accessible throughout its operational life, and the performance of both the repository and all waste packages could be readily monitored. With increased ventilation during the entire operational period, the bulk of the heat generated by the waste could be removed by allowing air to circulate around the canisters. As a result, more waste could be emplaced in the same repository volume, as the thermal loading imposed on the host rock would be greatly reduced.

Because of increased accessibility, retrieval of the waste should be simple and straightforward should the system or any components fail to perform as expected, or if retrieval of spent fuel for reprocessing were desired at some future date. Monitoring the performance of the waste package would be similarly straightforward.

Because the waste is unlikely to come in contact with any appreciable amount of water in repositories in the unsaturated zone, the performance requirements of the waste package with regard to resistance to leaching may be much less stringent than required in the saturated zone.

Where this page came from

This page was imported from U.S. Geological Survey. Published by the U.S. Geological Survey and, as a work of the United States government, in the public domain.

Nobody has written it yet — it is the source material at a new address, which is why search engines are asked to skip it and why no one earns from it. It is up for grabs: take it on, and it is yours to rewrite and to earn from.