ML20198H805

Determination of Radionuclide Sorption for High Level Nuclear Waste Repositories

ML20198H805
Person / Time
Issue date:	01/31/1986
From:	NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:
Shared Package
ML20198H792	List: ML20198H788 ML20198H795 ML20198H805
References
REF-WM-1 NUDOCS 8601310099
Download: ML20198H805 (18)

Text

-~

.. r Determination of Radionuclide Sorption for High-Level Nuclear Waste Repositories Technical Position January, 1986 Geochemistry Section - Geotechnical Branch Division of Waste Management U.S. Nuclear Regulatory Commission l

l l

l l

i 8601310099 060117 PDR WASTE WM-1 PDR

TABLE OF CONTENTS P_agt

1.0 INTRODUCTION

.......................................... 1 1.1 Purpose .......................................... I 1.2 Re g u l a t o ry F ra mewo r k . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.3 Definitions of Radionuclide Sorption and Related Experimental Parameters ........................... 2 -

2.0 BACKGROUND

.......................................... 2 2.1 Use of Sorption in Performance Assessment Analysis... 3 2.2 Approaches for Sorption Determination................ 4 3.0 STATEMENT OF P0SITION.................................... 5 4.0~ DISCUSSION .......................................... 6 4.1 Matrix Development............ ..................... 6 4.2 Characterization of Experimental Starting Materials and Products........................... 7 4.3 Isotherm Development for Closed-System Experimentation 9 4.4 Determination of Sorption Parameters by Multiple Experimental Approaches.............................. 10 4.5 Documentat.on of Uncertainties............ .......... 13

5.0 REFERENCES

.......................................... 14 1

I I

DETERMINATION OF RADIONUCLIOE SORPTION FOR HIGH-LEVEL NUCLEAR WASTE REPOSITORY NRC TECHNICAL POSITION

1.0 INTRODUCTION

1.1 Purpose This document presents a general approach for estimating radionuclide sorption on solids anticipated in a high-level nuclear waste repository in support of licensing findings. It is not intended to prescribe specific methods for radionuclide sorption determinations. Instead, the information is provided to the Department of Energy (00E) to be used as guidance for preparing detailed plans for radionuclice sorption determinations and submitting appropriate documentation early in the site characterization process.

1.2 Regulatory Framework Th'e Nuclear Waste Policy Act of 1982 (P. L.97-425) defines the role of three Federal agencies in the national program for disposal. of high-level radioactive wastes. The Environmental Protection Agency (EPA) has been responsible for developing " generally applicable standards for protection of the general environment from offsite releases from radioactive materials in repositories."

These standards have now been issued as final regulations (85 FR 38066). The NRC develops and issues "... technical requirements and criteria that will apply in approving or disapproving (i) applications for authorization to construct repositories; (ii) applications for licenses to receive and possess spent nuclear fuel and high-level radioactive waste in such repositories; and (iii) applications for authorization for closure and decommissioning such repositories." In this way, it will implement the EPA standards. The Department of Energy (00E) is responsible for collecting the data needed for site characterization and for constructing and operating a waste disposal facility in accordance with NRC regulations.

1

1.3 Definitions of Radionuclide Sorption and Related Experimental Parameters Sorption - one or more physicochemical processes, including ion exchange, adsorption, and chemisorption, but excluding precipitation of stoichiometric (fixed composition) solid phases, in which the radionuclide is removed from a liquid phase by interaction with a solid phase or phases.

Sorption or Desorption Ratio, R - the ratio of the concentration of 3

radionuclide on or within the solid to that in the liquid (units are L/Kg).

Retardation Factor,f R - the ratio of the velocity of the liquid to that of the radionuclide in an open (flowing) system.

Sorption Capacity - the maximum amount of radionuclide that can be sorbed on a unit mass of solid for a given set of conditions.

Starting Material - the substances added to a reaction vessel at the outset of an experiment, e.g., reactants.

2.0 BACKGROUND

A geologic repository controls the rate of radionuclide release to the accessible environment by means of two major subsystems: (1) the geologic setting; and (2) the engineered system. The geologic setting (site) is selected for its geologic, hydrologic, and geochemical attributes that enhance radionuclide isolation.

In order to ccmpensate for the uncertainty in predicting the behavior of geologic systems over long periods of time, the NRC has adopted a multi-barrier approach in its licensing criteria. In this approach the staff views the repository to be composed of three major barriers: (1) the waste package, (2) the engineered structure, and (3) the site and its environs. In general, this approach puts emphasis on: (1) engineered containment of radionuclides during the period of peak fission product decay, and (2) assurance of a controlled 2

l

release thereafter. This simplifies analysis and reduces uncertainties introduced into the analysis of the total system. During the period of engineered containment of the waste, the site geology should provide sufficient backup to account for those scenarios which may result in loss of engineered containment. Thereafter, the site geology should also have the capacity to retard the movement of the long-lived radionuclides to the accessible environment so that 10 CFR 60.112, the overall system release performance objective (EPA standard), is not exceeded. The DOE endorses the multiple barrier approach and places primary importance on the capabilities of the natural system for waste isolation (DOE, 1984).

Within the limits defined by 10 CFR 60.112 criteria, the DOE determines how much credit can be taken for sorption as a means of limiting radionuclide release. Radionuclide sorption parameters can be used in performance assessment models for characterizing the performance of a high-level nuclear waste site. These parameters are difficult to determine precisely because future geochemical conditions cannot be known with complete certainty and laboratory tests may not accurately model site behavior. However, by experimentally investigating sorption parameters using site-specific phases and conditions, it should be possible to take account of sources of uncertainty and make reasonable estimates of sorption along radionuclide release pathways in the subsurface repository.

2.1 Use of Sorption in Performance Assessment Analysis When a liquid is flowing through permeable solid media, sorption processes can

- act to retard the migration of the solute relative to the liquid flow.

Radionuclide sorption experimentation can be used to estimate this retardation

~

and, thus, quantify two aspects of repository performance. First, sorption experiments can be used to help identify which " key" radionuclides can be sorbed on repository solids. Key radionuclides are defined here as those radionuclides that might require some retardation in the host rock to meet regulatory criteria. Second, sorption studies can also be used to determine the ability of the subsurface repository (the engineered system and the geologic setting out to the accessible environment) to isolate radionuclides from the accessible environment. For example, parameters such as sorption or 3

desorption ratios, sorption capacities, and retardation factors derived from these studies can be used to help quantify the ability of the subsurface repository to retard radionuclide migration.

2.2 Approaches for Sorption Determination The NRC recognizes that there is more than one approach to determining sorption for use in performance assessment analysis. Initial experimentation used to

• characterize radionuclide sorption can be based either on thermodynamic or empirical approaches. The thermodynamic approach involves the use of simple systems (few components) and measuring fundamental thermodynamic parameters of the components participating in the dominant sorption reaction (s). These parameters could then be used in the performance assessment analysis to extrapolate to the complex systems of the repository. The empirical approach consists of determining the effect of various physical and chemical parameters on sorption under some site-specific conditions (complex systems). By combining the results of the initial experiments using the empirical approach that show the effect of varying the individual parameters on sorption, it should be possible to derive an equilibrium constant for the dominant sorption reaction if enough restrictions are placed on the systems studied. Both the thermodynamic and empirical approach yield uncertainties associated with the extrapolation to site-specific conditions not simulated in the experiments.

This technical position should not be construed as endorsing either approach.

In general, sorption experiments can be subdivided into two types: 1) closed systems; and 2) open systems (NEA, 1983; McKinley and Hadermann, 1984). Both approaches have been used to approximate one aspect of repository performance.

For characterizing sorption phenomena, closed-system experiments, such as batch sorption tests, involve contacting radionuclide-free (or deficient) solids with a radionuclide-bearing solution for the duration of the experiment followed by analytical determination of the sorption ratio, R s. (Batch desorption experiments, on the other hand, involve contacting radionuclide-free (or deficient) liquid with radionuclide-bearing solids, followed by measurement of the quantity of radionuclide leached.) Open-system experiments, such as flow-through column tests, involve the introduction of liquid solution at one end of a reaction vessel containing solid and the removal of the fluid at the 4

3

~

other end. The solid material sorbs the radionuclide(s) and, as a result, retards the migration of the contaminant (s) relative to that of the liquid.

This is expressed as a retardation factor, Rf.

There are advantages to both experimental approaches. For example, the advantages of the closed-system experiments are that they are relatively (1) simple, (2) inexpensive, and (3) better suited for mass production than the open-system experiment. On the other hand, the advantages of the open-system '

experiments are that they may better model radionuclide migration in flowing systems by revealing the presence of multiple speciation, mass action competition, colloids or particulates that might not be apparent in the closed-system experiment (batch test) (Kelmers, 1984).

3.0 STATEMENT OF POSITION It is the position of the NRC staff that sorption parameters used in performance assessment calculations shall be based on experimental data. The NRC staff considers that if the following points are included in the 00E experimental program, equilibrium constants for sorption processes can be derived. Such a strategy is vital to limiting the uncertainty involved in extrapolating experimental conditions and results to those of the repository, thus, allowing licensing decisions to be made. The 00E site programs should:

1) Develop a matrix of experiments that involves starting materials based on the anticipated range of proportions and compositions of phases under the various physicochemical conditions expected in the subsurface repository;

2) Characterize solid and liquid experimental starting materials and products;

3) For closed-system experiments, determine sorption isotherms by varying radionuclide concentrations up to an apparent concentration limit if appropriate; 5

4) Determine the applicability of sorption parameters to repository performance by using various experimental approaches involving both open and closed laboratory systems, and in situ field tests, and studies of natural analogues; and

5) Document the magnitudes of experimental and conceptual uncertainties from all anticipated sources.

Details on the individual points from the Statement of Position, along with a discussion on why the NRC staff thinks these points are important, are given below.

4.0 OISCUSSION It is the responsibility of the DOE to demonstrate that when experimentally-derived sorption parameters (plus some uncertainty factor, if necessary) are used in performance assessment calculations, there is reasonable assurance that the radionuclide migration is not underestimated.

4.1 Matrix Development A matrix of experiments should be developed as a planning tool for characterizing the sorption properties of a subsurface repository. With a matrix, the DOE should be able to demonstrate that crucial experiments that characterize sorption have not been overlooked. As a result, the DOE should be able to demonstrate with reasonable assurance that the derived sorption parameters are appropriate.

Variables such as solid composition, mineralogy, and texture, liquid composition, proportion of phases, temperature, pressure, particle size, flow rate and regime (porous and fractured media), time, and fonizing radiation should be considered in the matrix. The 00E can then effectively demonstrate its rationale for choosing some combinations of parameters for study and eliminating other combinations as inappropriate.

6

The matrix should include scoping experiments, performed early in the experimental program, which involve relatively simple systems to determine the effects of various physicochemical conditions on sorption. Following the scoping experiments, the matrix development should reflect combinations of the above parameters that simulate physicochemical conditions and phase assemblages likely to occur in the repository system. Consequently, the size of the matrix would be greatly reduced by first considering the dependence or interrelation of phases and conditions upon each other and deleting incompatible -

combinations. As experimentation progresses to a point where enough insight has been gained to establish the relationship between sorption mechanisms and experimental values, a theoretical basis may be established for reducing the number of experiments in the matrix.

Also, the priority of testing needs to be established. For example, radionuclide studies can be prioritized by comparing radionuclide inventories in the repository to release limits specified in 10 CFR 63.112 of the NRC regulations. Some radionuclides may occur in low enough quantities that, if they meet NRC release rate requirements, they will not contribute significantly to exceeding these limits. These may be assigned a lower priority than those radionuclides whose cumulative releases over 10,000 years are likely to exceed the release limits in the absence of sorption effects.

4.2 Characterization of Experimental Starting Materials ard Products In choosing appropriate solid starting materials for sorption studies, emphasis should be placed on the identification and characterization of waste form, canister, backfill, seals, packing, and host rock primary and secondary phases occurring along paths the radionuclide-bearing groundwater will take as it flows away from the waste, since these are the solids most likely to react with groundwater and thereby affect radionuclide concentrations and release rates.

Characterization of the solids should include chemical, mineralogical, textural, and particle size determinations. Thus, it is important to show with reasonable assurance that laboratory experiments involving sorption on crusned solids (for example) is relevant to sorption at the repository site. Ths surfaces of crushed material may be significantly different from the surfaces of intact material, both porous and fractured. Grinding may expose the >

7

surfaces of solid phases different from those which groundwater would contact in a repository and/or may change the reactivity of the same mineral surfaces with dissolved radionuclides.

Similarly, the range of groundwater compositions expected in a repository system should be considered in selecting liquid starting materials, Generally, in the rock-dominated environments of a high-level waste repository, groundwater compositions can be affected by reactions with solids at various temperatures and pressures. Consideratiori of the range in water compositions used in experimentation should be based on the range of compositions of analyzed groundwaters at ambient conditions, the range of compositions calculated from solid assemblages assumed to have equilibrated with the groundwater, and the range of groundwater compositions experimentally determined at elevated temperatures and pressures.

The applicability of synthetic starting materials to the conceptual model employed in developing the matrix should be addressed. Failure to do so might result in experiments that do not adequately simulate repository conditions.

For example, the preparation of radionuclide-bearing groundwater commonly involves the addition of a small amount of acidified tracer to a synthetic solution simulating the natural groundwater. The resulting solution may neither be representative of repository conditions nor be stable.

In addition to character 1:ing the starting materials, it is also imnortant to characterize the experimental products. Following the experiment, analysis of the liquid products should include the determination of major, minor, and trace element concentrations, along with pH ar.d redox conditions.

The extent of sorption of some dissolved radionuclides on engineered barrier materials and host rock can be strongly dependent on the redox potential (Eh) and acidity (pH) of the groundwater. For example, Benjamin and Leckie (1981) show that the sorpsion of Cd, Cu, Zn, and Pb on amorphous iron oxyhydroxide is strongly depero 'e. q pH. The percentage of cation sorbed varies from approximately s er to one hundred with a change in pH of two units. Likewise, Kelmers ct al. (1984) have shown that sorption ratios for neptunium and technetium are dependent on the redox condition of the system.

8

The characterization of solid products from sorption experiments is important because, for example, under the same physicochemical conditicns, different solid phases can have drastically different sorptive capacities for the same radionuclide. Characterization of the solids is important in determining which reacticns tock place and how these reactions depend on experimental technique.

In addition to determination of the composition of individual solid phases, characterization should include surface area and/or particle size measurements.

Because sorption is predominantly a surface phenomenon, the surface area of the solid may strongly affect the experimentally determined sorption parameters.

For example, neptunium sorptien ratios increased two orders of magnitude as particle diameter decreased from 200 to 2 um (Kelmers et al.,1984).

4.3 Isotherm Develcpment for Closed-System Experimentation Probable release scenarios call for radionuclide concentration gradients in the repository system. The concentrations of radionuclides in the repository can range from zero to an apparent concentration limit. The apparent concentration limit is the greatest radionuclide concentration that the liquid can maintain when the temperature, pressure, and moles of all other components in the liquid, n), are held constant. The apparent concentration limit is controlled by the solubility of some stoichiometric radionuclide-bearing sclid phase.

Figure 1, a schematic sorption isotherm, illustrates the relationship between concentration on the solid versus concentration in the liquid wnen all other parameters are held constant. Analysis of the liquid product can be used to monitor the constancy of the other parameters. Although this figure shows a linear sorption region, many sorbed species, including radionuclides, show nonlinear relationships between the quantity serbed and the solution concentration. Thus, sorption ratios can be dependent on radionuclide concentration (Serne and Relyea,1382).

i Because radionuclide concentrations are expected to vary in the repository and sorption parameters can be concentration dependent, it is reasonable to cesign experiments to determine the effect of concentration on sorption ratios.

Sorption isotherms should be determined up to an apparent concentration limit if possible, so that it can be shown that precipitation is not contributing to i the sorption ratio. Experimentally, it should be possible to determine an 9

apparent concentration limit of a radionuclide in liquid in contact with solid.

For example, at the same temperature, pressure, and nj in the liquid, two i

sorption experiments with different concentrations of the same radionuclide in the liquid starting material should yield the same radionuclide concentration in the liquid products at the apparent concentration limit (see Figure 1).

4.4 Determination of Sorption Parameters by Multiple Experimental Approaches e

If a sorption experiment could be designed that simulated all anticipated repository conditions, it would not be necessary to use multiple experimental approaches to determine sorption parameters. However, simulation of all anticipated repository conditions in sorption experimentation would be difficult and/or impractical. The fact that some parameters or conditions cannot be bounded requires the extrapolation of these conditions to those expected in the repository. This extrapolation introduces uncertainty into the medeling of sorption parameters. Therefore, multiple approaches are important because they can lend support to, and reduce the uncertainties of, experimental results from studies in which some parameters are not site specific. Some experimental parameters can be varied over a large enough range as to bound the conditions anticipated in the repository. These parameters include surface area / volume ratio (SA/V), temperature, pressure, composition, and flow rate.

Other parameters that of ten are not duplicated in the laboratory are scale, residence time, water / rock ratio, and flow characteristics, which can include saturated versus unsaturated flow and porous media versus fractured flow.

Experiments are designed so that measurable effects of physicochemical reactions :a1 be monitored in a reasonable tire, Section 60,101(a)(2) of 10 CFR Part 60 allows for tne use of accelerated tests to demonstrate compliance with performance objectives and design criteria. At the relatively low temperatures anticipatec in the repository, chemical reactions involving geologic raterials can be extremely slow. In order to accelerate these reactions so teat cnanges are measureable in experimental time, conditions other than these anticicated in the nuclear waste repository are sometimes imposec om the experimental system. For example, experiments have employed crushed solid naterial, nigh concentrations of solutes, agitation, catalysts, rapid flow rates, and elevated temperatures.

10

r-e .

In aJdition to accelerating reaction rates, laboratory experiments are designed so that the amount of material required can be handled reasonably. By scaling down systems of interest (repository size) to laboratory size, certain physical i conditions must be altered. For example, the water / rock ratio in most repository syste is is significantly less than one. However, in order to cbtain enough water for analysis in laboratory experiments, the W/R ratio is ordinarily increased significantly. This technique makes the bulk chenistry of the experimental system different from that in the repository. The proportions #

of phases in experiments has been shown to affect radionuclide scrption parameters (Palmer et al. ,1981; Raf ferty et al . ,1981; Meier et al. ,1932).

Thus, the effect of this technique on sorption parameters should be considered.

One can argue that in a fractured medium, with little porosity, most of the rock will not be in contact with the groundwater. Consequently, water / rock ratios used in experimentation should be higher than those that take into account all the rock in a repository system. If this argument is used, nowever, it follows that the solid reactants should be predominantly fracture material and not bulk rock. Sorption experiments involving crushed bulk rock might have little applicability to sorption phenomena in fractured media.

To verify the applicability of experin.entally determined sorption parameters to a repository system, the site should use multiple experimental approaches.

[This approach yas a reccmmendation of the Waste / Rock Interactions Technology (WRIT) Program (SerneandRelyea,1982).] Using this approach, sorptien parameters can be analyzed and compared. For example, the sorption ratto, R ,

g obtained from batch experiments has often been used to calculate a retardation factor, R f. The relationship between R, and Rf is taken to be Rf = 1 + pR,(1 - 4,)/#,

where p is the bulk density of the rock, and 4, is the effective porosity.

This relationship is based on ion exchange theory as applied to porous media flow. However, due to the variety of processes that contribute to sorption, the calculated Rf value may not equal the measured Rf value determined frcm a flow-through column experiment.

11

Comparison of the sorption and desorption parameters obtained from closed-system and open-system experiments is recommended. Generally, the sorption parameters (e.g., R,) derived from closed-system experiments are equal to or greater than those derived from open-system tests using the same solid material (NEA Workshop, 1983). As a result, closed-system tests may overestimate the effectiveness of a repository system to isolate radionuclides (Relyea et al., 1980). The difference in sorption ratios may be due to particle abrasion in stirred closed-system experiments or the relatively Short residence times in open system experiments (NEA Sorption Workshop,1983).

Other factors that can cause a discrepancy between the sorption parameters from open and closed systems are the presence in the liquid of multiple radionuclide species, colloids, and particulates. Changes in physicochemical parameters such as temperature, fluid velocity, radionuelide concentration, time of reaction, and fluid composition may shed some light on the causes of the discrepancy between the two types of systems. Thus, a rationale for the difference in sorption parareters cbserved using different methods contributes to the overall certainty that can be assigned to the sorption parameters.

Extrapelation of sorption parameters fron laboratory experiments to a large-scale, long-term repository system can be (11ghly uncertain. The flow characteristics of the groundwater can have a drastic effect on the applicability of laboratory-derived sorption parameters to repository perf ormar.ce. Most evporiments use crushed material as a solid medium because it is easy to handle and characterize, and accelerates solute-solid reactions, The application of crushed material to intact porous media eay be adequate but not so Men the ratur&l system is fractured ecck. Sinnock et al. (1934) and Nuttall and Ray (1981) Mye calculated that rates of radionuclide migration via fracture flow can be two orders of magnitude greator than that via porous media flow. Thus, for performance assessnent calculations, censideration of flow regime can be of the utmost importance.

If groundwater fla in a repository is predominantly via fracture flow, sorpticn tests in the 1:2bnratory may not ac'equately sinulate reposttory corditions. One method of further reducing the uncertainty caused by the inauequate simulation of various ficw characteristics could be to perfom in situ tests on site-spercific solid material (Serne and Relysa, 1982; Abelin et 12

al., 1984; Neretnieks et al., 1982). The scale of these tests can be larger than that of the experimental tests but smaller than that of the repository.

Furthermore, the in sito solid materials would certainly not have suffered the effects of handling (grinding, sieving, washing) required in laboratory tests.

Time constraints, however, would still apply in these experiments. Comparison between the laboratory and field results can illustrate the usefulness of the different approaches, but the physicochemical conditions must be carefully controlled in the in situ tests to ensure a parallelism in the approaches. ,

Therefore, in situ tests in conjunction with laboratory tests can be performed to reduce the uncertainties of extrapolation to the repository systems.

Although field tests can expand spatial scale over that which is normally handled in a laboratory, the time scale is still several orders of magnitude less than that of a repository. The study of natural analogues is a means of shedding some light on the mfgration of radionuclides in natural systems that have existed for long periods of time. Tnus, their use is a way of dealing with the uncertainty associated with extrapelating short-term laboratory and field experiments to long-term performance of a repository. Common examples of systems used as natural analogues are ore bedies such as Oklo (Brookins,1978) and the uranium deposits in the Northern Territories of Australia (Airey, ,

1983). Igneous intrusfves have also been studied because they simulate anticipated thermal histories and alteration patterns. To be useful, however, the natural analogue should include a process that is demonstrably equivalent to the same process present in the repository and have well-defined boundary conditions.

4.5 Documentation of Uncertainties There are many sources of uncertainty in sorption studfes. For example, sources of uncertainty sten from failure to duplicate anticipated repositst/

conditions, improper entrapolation of experimental results to repository '

conditions, incorrect experimental results, and analytical error. The failure to duplicate repository conditions can be caused by an incorrect understanding of the conditions, an inability to duplicate the condf tions or an inadvertent improper experimental design. Likewise, improper extrapolation can also originate from improper experimental design. Incorrect esperimental resu7ts 13 I

0 0

can be caused by imprecision or misinterpretations of the data. Ways that uncertainties can be minimized are (1) the use of multiple techniques to determine repository conditions, (2) analyses to bound ad'v erse impacts, (3) multiple experimental methods, and (4) the independent duplication of results.

The characterization of uncertainties is important for determining how much credit can be placed on sorption. By characterizing uncertainties, the DOE should be able to demonstrate with reasonable assurance that the sorption parameters are appropriate.

5.0 REFERENCES

Abelin, H., J. Gidlund, L. Moreno, and I. Neretnieks, " Migration in a Single Fracture in Granitic Rock" in Scientific Basis for Nuclear Waste Management, VII, Elsevier Science Publishing Co., p. 239-246, 1984 Airey, P. L., Radionuclide Migration Around Uranium Ore Bodies - Analogue of Radioactive Waste Repositories, NUREG/CR 3941 AAEC/C40, vol. 1, 1984.

Benjamin, M. M. and J. O. Leckte, Multiple-Site Adsorption of Cd, Cu, Zn, and Pb on Amorphous Iron Oxyhydroxide, Jou_rnal of Colloid and Interface Science vol. 79, no. 1, p. 209-221, 1981a.

Brookins, D. G. Retention of Transuranic, Other Actinide Elements and Bismuth at the Oklo Natural Reactor, Gabon, Chem. Geol ., p. 307-323, 1978.

00E/RW-0005 DRAFT, " Mission Plan for Civilian Radioactive Waste Management program", U. S. Department of Energy, April,1984.

Kelmers, A. O., Oraft Analysis of Conservatism of Radionuclide Information Measured by Batch Contact Scrption/ Apparent Concentration Limit Isotherms, Letter Report, L-290-3, 1984 Kelmers, A. D. , J. H. Kessler, W. D. Arnold, R. E. Meyer, N. H. Cutshall, G.

K. Jacobs, S. Y. Lee, Evaluation of Radionuclide Geochemical Information Developed by 00E High-Level Nuclear Waste Repository Site Projects, NUREG/CR 3730, 1984.

McKinley, I. G. and J. Hadermann, Radionuclide Sorption Database for Swiss Safety Assessment, EIR - Bericht, Nr 550, Octcber, 1934 Meier, H., E. Zimmerrac Al, G. Zeitler, and P. Menge, "The Static or Batch Method for Testing the Sorptive and Oesorptive Characteristics of Geologic Media" in Stancardization of Nothod for Measuring Migration of Radionuclides in Geomeat_a, Proceedings of the'US/FRG Bilateral Workshop, Berlin, Municn, detoter, 1982.

i 14 i

O NEA (Nuclear Energy Agency), " Sorption, Modelling and Measurement for Nuclear Waste Disposal Studies", Summary of NEA Workshop held 6-7 June 1983 in Paris, 1983.

Neretnieks, I., T. Ericksen, and P. Tahtinen, Tracer Movement in a Single Fissure in Granitic Rock: Some Experimental Results and Their Interpretation, Water Resources Res., vol. 18, no. 4, p. 849-858, 1982.

NRC, " Nuclear Regulatory Commission, 10 CFR Part 60, Disposal of High-Level Radioactive Waste in Geologic Repositories", 1984.

Nuttall, H. E. and A. K. Ray, A Combined Fracture / Porous Media Model for Contaminant Transport of Radioactive Ions" in Scientific Basis for Nuclear Waste Management, vol. 3, New York, p. 577-590, 1981.

Palmer, D. A., S. Y. Shiao, R. E. Meyer, and J. A. Wethington, Adsorption of Nuclides on Mixtures of Minerals, J. Inorganic Nucl. Chem., vol.43, p.3317-3322, 1981.

Rafferty, P. , S. Y. Shiao, C. M. Binz, and R. E. Meyer, Adsorption of Sr(II) on Clay Minerals: Effects of Salt Concentration, Loading, and pH, J. Inorganic Chem., vol. 43, p. 797-805, 1981.

Relyea, J. F., R. J. Serne, and D. Silva, Methods for Determining Radionuclide Retardation Factors: Status Report, PNL-3349, Pacific Northwest Laboratory, Richland, Washington, 1980.

Serne, R. J. and J. F. Relyea, "The Status of Radionuclide Sorption-Desorption Studies Performed by the WRIT Program" in Technology of High-Level Nuclear Waste Disposal, 1982.

Sinnock, S., Y. T. Lin, and J. P. Brannen, Preliminary Bounds on the Expected Postclosure Performance of the Yucca Mountain Repository Site, Southern Nevada, Sandia Report SAND 84-1492, 1984.

l l

15 I... . .

9 9

_a 1

o to Z APPARENT O SORPTION CAPACITY z - - - - - - - - - - - - - - - CONCENTRATION h i LIMIT E I i-Z l W I O

Z l O LINEAR ISOTHERM o i REGION (CONSTANT R,) I

/ I I

CONCENTRATION IN SOLUTION

- -_- - _ _ -- -- - - - - _ - - - - - - - - - - - _