Masters Theses

Date of Award

5-1995

Degree Type

Thesis

Degree Name

Master of Science

Major

Nuclear Engineering

Major Professor

Larry F. Miller

Committee Members

Herschel Godbee, Peter Groer, Raphael Perez

Abstract

Site-specific radiological performance assessments are required for the disposal of low-level radioactive waste (LLW) at both commercial and U.S. Department of Energy facilities. The purpose of these assessments is to provide the technical basis for demonstrating compliance with performance objectives for LLW disposal as set forth by appropriate authorities. Performance assessments are used to evaluate potential doses to individuals as a result of the release of radionuclides to the environment and intrusion into the disposal facility. Complex computer models are often used to calculate the release of radionuclides from a facility and the transport of these radionuclides through the environment. The calculated releases constitute the source term for a performance assessment. This thesis explores source term modeling of LLW disposal facilities by using two state-of-the-art computer codes, SOURCE1 and SOURCE2. An overview of the performance assessment methodology is presented, and the basic processes modeled in the SOURCE1 and SOURCE2 codes are described. A detailed derivation of contaminant transport equations for a waste disposal facility results in a first-order advective model which can be compared with the zero-order model used in the SOURCE1 and SOURCE2 codes. The derivation accounts for radioactive decay, radionuclide sorption/desorption, and unsaturated transport in a porous medium. The concept of the leach rate constant is used as a basis for the first-order advective model. Comparisons are made between the two advective models for a variety of radionuclides, transport parameters, and waste-disposal technologies. These comparisons show that, in general, the zero-order model predicts undecayed cumulative fractions leached that are slightly greater than or equal to those of the first-order model. For long-lived radionuclides, results from the two models eventually reach the same value. By contrast, for short-lived radionuclides, the zero-order model predicts a slightly higher undecayed cumulative fraction leached than does the first-order model. Variation of the distribution coefficient produces small differences between the two advective models. A new methodology, based on sensitivity and uncertainty analyses, is developed for predicting intruder scenarios. This method is demonstrated for 137Cs in a tumulus-type disposal facility. The sensitivity and uncertainty analyses incorporate input-parameter uncertainty into the evaluation of a potential time of intrusion and the remaining radionuclide inventory. Finally, conclusions from this thesis are presented, and recommendations for continuing work are made. From the work presented, it is clear that source term modeling presents a large number of areas for investigation. The work presented in this thesis enhances the understanding of source term modeling. Continued work in the recommended areas should lead to further improvement in source term codes.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS