Doctoral Dissertations

Date of Award

5-2001

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Nuclear Engineering

Major Professor

Laurence F. Miller

Committee Members

Peter G. Groer, Keith F. Eckerman, Gary T. Smith

Abstract

Quantification of the uncertainties in internal dosimetry is important because it can affect the outcome of major dose reconstruction, risk assessment or epidemiological studies. To date, the uncertainties in the internal doses have been quantified for 131I, 137Cs and 239Pu, and judgmental uncertainty factors based on expert elicitation are available for other radionuclides. The purpose of this study is to determine the uncertainties in the doses per unit intake from ingestion of 90Sr by adults and to analyze the variation of these dose factors with gender and age. To achieve these goals, the International Commission on Radiological Protection (ICRP) biokinetic model for 90Sr is employed and a parametric uncertainty analysis is performed. Necessary correlations between parameters are imposed when the calculations are performed. The uncertainties in the estimated dose factors can be quantified as a factor of 6 for the bone surface and 5 for the red bone marrow, where "a factor of is defined as the ratio of the 97.5th and the 50th percentiles of the probability distribution function obtained for each dose factor. Similarly, the uncertainties are a factor of 2.5 for bladder and stomach, 2.2 for the small intestine, 2.1 for the upper large intestine and 2.7 for the lower large intestine. For the rest of the organs the uncertainty is a factor of 3. For all organs, the main contributor to uncertainty in the estimated dose factors is the biokinetic model, followed, in most cases, by the uncertainty about the mass of the target organ. The energy deposition has the lowest contribution to uncertainty for most organs, but is important for the lower and upper large intestine. No gender differences were found in the dose factors for bone surface, red bone marrow, and the upper and lower large intestine. The doses to the soft tissues for females are larger by 20% than corresponding doses for males. When age-dependent biokinetic parameters are used, the estimated dose factors are very close in absolute value to the dose factors obtained using age-independent kinetics (i.e., a maximum difference of 40% for bone structures). The two sets of dose factors also have comparable uncertainties. A quantity called expected dose is introduced in this study, defined as the dose received over an infinite length of time, and calculated using the dose-rate weighted by the probability of surviving up to the age when the dose is delivered. The estimated expected doses are provided for different ages at exposure. For exposure at young ages the expected dose and the committed dose are similar, but the committed dose decreases to zero when exposure occurs close to age 70, while the expected dose has elevated values past age 70. The variation of the dose factors with attained age show that the maximum dose is reached about 30 years after exposure for bone surfaces, 20 years for red bone marrow, 10 years for bladder, 1-2 years for the upper and lower large intestine, and 20 years for the rest of the organs. The magnitude of the uncertainties in the dose factors is large enough to influence the outcome of epidemiological studies. The age dependency of the dose factors for adults is small compared to the uncertainties in the dose factors. Thus, the dose factors based on age-averaged parameters should suffice for most practical purposes. The biokinetics of strontium show no important gender dependency. The dose factors for soft tissues are gender-dependent because of gender differences in organ masses. The expected dose provides a more accurate description of the exposure than committed dose, and it can be used to estimate the risk of cancer and to make radiation-related decisions. The techniques developed in this study for solving first-order differential equations with uncertain correlated parameters can be applied to a large variety of models commonly used in radiation protection and nuclear engineering.

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

Share

COinS