Bruce Becker, Swen Magnuson, and Jeff Sondrup
Lockheed Martin Idaho Technologies Company
Idaho National Engineering and Environmental Laboratory

The first detailed comprehensive simulation study to evaluate fate and transport of wastes disposed in the Subsurface Disposal Area (SDA), at the Radioactive Waste Management Complex (RWMC), Idaho National Engineering and Environmental Laboratory (INEEL) has recently been conducted. One of the most crucial parts of this modeling was the source term or release model. The current study used information collected over the last five years defining contaminant specific information including: the amount disposed, the waste form (physical and chemical properties) and the type of container used for each contaminant disposed. This information was used to simulate the release of contaminants disposed in the shallow subsurface at the SDA. The DUST-MS model was used to simulate the release. Modifications were made to allow the yearly disposal information to be incorporated. The modeling includes unique container and release rate information for each of the 42 years of disposal. The results from this simulation effort are used for both a groundwater and a biotic uptake evaluation.

As part of this modeling exercise, inadequacies in the available data relating to the release of contaminants have been identified. The results from this modeling study have been used to guide additional data collection activities at the SDA for purposes of increasing confidence in the appropriateness of model predictions.

**This work was funded by the U.S. Department of Energy under DOE Idaho Operations Office Contract DE-AC07-94ID13223.

The SDA is part of the Radioactive Waste Management Complex (RWMC) located in the southwest portion of the INEEL. Low-level, mixed, and transuranic radioactive wastes were buried in shallow pits and trenches in the SDA from 1952 until 1970. Since 1970, transuranic waste has been stored in above grade facilities. Mixed waste disposal ceased in 1982. Currently, low-level waste is still being disposed in a portion of the SDA.

This work uses information generated during a year long inventory evaluation. Previous attempts to characterize the inventory had relied solely on shipping records. A review of the shipping records showed that many contaminants were not fully listed in the disposal records. A different approach to compile the inventory information was devised. The approach emphasized the use of information about the processes that generated the waste. The waste generation information was supplemented by information in other reports, shipping records, databases, and nuclear physics calculations. The results of the inventory investigations are documented in 2 reports (INEL, 1995a and INEL, 1995b)

The primary purpose of this SDA simulation study was to perform fate and transport calculations to support an Interim Risk Assessment (IRA) being conducted for the Waste Area Group (WAG) 7 Operable Unit (OU) 7-13/14 comprehensive remedial investigation/feasibility study (RI/FS) in the Federal Facility Agreement and Consent Order (FFA/CO) for the INEEL (IDHW 1991). The overall approach was to develop a simulation model representative of the releases into the shallow subsurface. This representativeness was obtained and demonstrated by calibrating the model results to observed contaminant data and is a substantial improvement over the representativeness of earlier simulation studies. A secondary purpose of the SDA simulation study was to be able to use the model to evaluate possible remediation strategies and their effects on the predicted contaminant concentrations in the OU 7-13/14 feasibility study.

The source term model is the first step in determining the concentrations of contaminants at a potential receptor location. Figure 1 shows a flow diagram of the complete modeling process. The source term model computes the release of contaminants into the environment. Subsequent models simulate the transport of contaminants to the potential receptor.

This modeling effort included several improvements over previous efforts. The improvements include:

• Incorporating yearly disposal information. Previous efforts had decayed the inventory to the present and modeled all releases from that point forward. This effort used each year as a separate input with unique waste form characteristics that effect the release.
• Incorporating site specific drum failure data. The data from previous retrieval operations was used to determine the failure rate for drums. Much of the waste was disposed in drums and the effect of including the drum failure rate is significant for the volatile contaminants.
• Incorporating site specific release rates. Beryllium and steel corrosion rates.
• Tracking full decay chains in the release and transport models.
• Partial calibration/comparison to predicted values.

The source term model computes the release of contaminants into the subsurface. The model has to account for a diverse assortment of waste types disposed over the past 4 decades. The waste comes from many facilities including other INEEL facilities such as Idaho Chemical Processing Plant, Test Reactor Area, Test Area North, and Argonne National Laboratory-West as well as off site generators such as the Rocky Flat Plant. The waste includes such diverse items as fuel end pieces, generic lab trash, anti-contamination clothing, and sludge. The containment used to transport the waste included steel drums, welded boxes, plywood boxes, cardboard boxes and other containers. All of this was accounted for in the model.

A review of the shipping records showed that many contaminants were not fully listed in the disposal records. The inventory was evaluated starting from knowledge of the process that generated the waste to determine the contaminants disposed, the amount disposed (including an upper and lower bound), the type of containment used, and the physical and chemical form of the waste. The results of the inventory investigations are documented in 2 reports (INEL, 1995a and INEL, 1995b) and are used in this modeling effort.

Contaminants are not released until the containers they were disposed in fail. Once the containers fail the contaminants are release by specific mechanisms representative of the disposed waste. Three different release mechanisms are used to simulate the release from the waste. Volatile organic compounds diffuse from the cemented sludge they were disposed in and are modeled using the analytical diffusion model. Many types of surface contamination on trash were disposed and are modeled with a surface wash-off model. This is equivalent to a first order leach model that uses a partition coefficient between the waste and the water. Many radionuclides are activation products and are created as part of activation of the base metal in a reactor vessel or the piping. These nuclides are released to the environment when the metal corrodes. This was modeled with the third release mechanism - corrosion.

The computer code DUST-MS (Disposal Unit Source Term - Multiple Species) (Sullivan, 199x) was used to model the releases to the shallow subsurface at the SDA. DUST-MS was chosen by a detailed selection process (Baumer, 1996) that listed the criteria a computer simulation needed to satisfy in order to be acceptable. After selection, the individual release models in DUST-MS were compared to analytical solutions to insure the program performed correctly.

Because of the large number of contaminants to be evaluated, a series of interface programs were developed to generate input files for the source term model and to interface between the source term and subsurface model. These interface programs format the data necessary for running the simulations while eliminating the need to manually edit files and potentially provide incorrect inputs. A flow chart showing the source term model, subsurface model, the interface programs and the data being manipulated is shown on this poster.

The information contained within the inventory evaluation reports for 1952-1983 and 1983-2003 was used to develop the waste stream (release) and inventory information databases. The databases contain the release and inventory information for each year of disposal from 1952 to 1993. Information after 1993 was estimated data and not included in the base case simulation.

Release as a function of time for each of the three waste time periods (pre 1970 waste, the 1970 to 1984 waste and the current LLW waste) was provided on a yearly basis for 10,000 years. Each of the three periods corresponds to different locations within the SDA and was input into the subsurface model in different grid blocks. Figure 2 shows the release of nitrate as a function of time.

Direct calibration of the source term model was not possible because of the lack of data immediately beneath the waste. Comparisons were made to the limited data available beneath the waste and the model was not inconsistent with the measured concentrations. The source term model was indirectly calibrated by comparing the simulated subsurface flow and transport predicted concentrations to the measured aquifer concentrations. These comparisons identified problems with four contaminants. The amount of chromium predicted was too low, the amount of Tc-99 and I-129 was too high and the amount of carbon tetrachloride was too low. Subsequent investigations revealed that the partition coefficient used for the release of chromium was for chromium metal when chromate salts were actually disposed. When the appropriate partition coefficient was used the comparison to measured data improved. Subsequent investigations into the inventory data for Tc-99 and I-129 revealed an error in the inventory data for one waste stream for a single year. For carbon tetrachloride, the only way to match measured concentrations in the subsurface and aquifer was to increase the inventory. Subsequent investigations revealed the inventory was originally underestimated.

The simulations were calibrated using the best estimate inventory information. The upper bound inventory information was used to bound the potential risks associated with the waste. The release for all of the contaminants simulated was computed as a function of time and provided as input to the subsurface model. Daughter product with a half life of greater than one year were explicitly tracked in the simulation. Daughter products with a half life of less than a year were assumed to be in secular equilibrium with the parent.

• An average infiltration rate of 8.5 cm/year, determined via monitoring and modeling, was used.
• The waste was separated into 3 groups based on changes in the type of waste accepted at the RWMC. The three time periods correspond to the pre 1970 waste, the 1970 to 1984 waste and the current low level waste. These three time periods reflect changes in the type of waste accepted at the SDA. The amount released for each group was computed based on the sum of the release of individual years within that group. Each year=s release was a function of the inventory, waste form and containment for that particular year.
• Steel boxes, concrete casks and steel drums delay the release and are accounted for in the model. Plywood boxes, cardboard boxes and other types of containment offer no barrier to release and are not accounted for. The concrete casks are modeled as diffusion release from a concrete waste form. The metal boxes are assumed to fail at the time equal to the wall thickness divided by the steel corrosion rate. Drums are assumed to fail over time in a Gaussian distribution.
• The inventory used for calibration was the best estimate contained in the buried waste inventory database. The inventory used for the BRA was the upper bound inventory from buried waste inventory database. (One possible exception to this was for VOCs where the calibration indicated potentially more contaminant disposed than was reported in the buried waste inventory database).
• Three release mechanisms are modeled - Surface wash-off, diffusion, and dissolution. Diffusion was used to model release of volatile organic compounds from sludge and release from concrete waste forms. Dissolution was used to model the release from ion exchange resins, the release from beryllium block and the release of activation products from the base metal. Surface wash-off was used to model the release of surface contamination on general trash.
• All of the mass released by the source term model was available for transport to the groundwater.
• Release of activation products from metals was by corrosion. Release of volatile organic compounds from sludge was by diffusion and release from laboratory trash and construction debris was by surface wash-off.

A simulation was developed that accounted for yearly input of waste into the disposal area and predicted the release of contaminants from that waste. Each year of disposal was characterized with its own unique release information including inventory, release mechanism, and containment type. This information was used as input to the source term model DUST-MS which provided the release into the shallow subsurface. Once released, the contaminant mass was available for transport to the aquifer. The model was part of a combined modeling effort designed to estimate the potential health risks from disposals at the SDA.

As a result of this study, additional characterization work is planned for the RWMC, to support the CERCLA RI/FS process. The results and identified uncertainties in the risk assessment have been used to direct the investigations. The complete lack of information to directly calibrate the source term model is a severe limitation in the source term modeling. A series of lysimeters are to be installed to collect leachate from directly beneath the waste. This data will be used to calibrate the source term model. In addition, site specific release rate information is unavailable for many for the contaminants that have the highest potential health risks. For activation products, the corrosion rate of the base metal controls the release. A study in INEEL soils has been started to determine an appropriate site specific steel corrosion rate. For Uranium, the chemical form of the disposed material can have significant impact on the estimated health impacts. Actinide chemistry is complex and the solubility of uranium can vary by 10 orders of magnitude or more depending on the form of the uranium disposed. An investigation into the form of the uranium disposed in the RWMC has been initiated. In addition a program to core through the waste and get contaminant profiles is being proposed. In addition to acquiring information useful in determining the release rates (by determining the mass released and the mass remaining in the waste form), information related to the contaminant mobility will be gathered as well.

1. INEL, 1995a, A Comprehensive Inventory of Radiological and Nonradiological Contaminants in Waste Buried in the Subsurface Disposal Area of the INEL RWMC During the Years 1952-1983, INEL-95/0310 (formerly EGG-WM-10903m), revision 1, August.
2. INEL, 1995b, A Comprehensive Inventory of Radiological and Nonradiological Contaminants in Waste Buried or Projected to Be Buried in the Subsurface Disposal Area of the INEL RWMC During the Years 1984-2003, INEL-95/0135, August.