QUALITY IMPROVEMENTS IN CALIBRATION AND USE OF PORTABLE HOLDUP DETECTORS FOR URANIUM AND PLUTONIUM FISSILE DEPOSITS

Year
2015
Author(s)
Jeffrey A. Chapman - Oak Ridge National Laboratory
Sean Branney - Savannah River National Laboratory
Shaheen A. Dewji - Oak Ridge National Laboratory
Graham V. Walford - The University of Tennessee
Stephen Croft - Oak Ridge National Laboratory
Franklin H. DuBose - Georgia Institute of Technology
Andrew D. Nicholson - Oak Ridge National Laboratory
David W. Roberts - Savannah River National Laboratory
Abstract
Portable gamma spectroscopy systems used for the quantification of fissile deposits are typically based on scintillation detector systems because of the need to maintain reasonable energy resolution and detection efficiency in a compact, light-weight and simple package. Recent efforts in detector design have improved performance, both in detector efficiency and shielding effectiveness compared with older designs. Therefore this paper is focused upon their calibration, improved quantification approach and use over a broader range of applications. One driver for the rigor and discipline required in these efforts is found in quality programs such as “Quality Standards for Non Destructive Assay Characterization” (QSNDA), which defines the requirements of quality and measurement reportability. Our calibration model, termed “Comprehensive Geometry Model” (CGM) provides quantification of the detector full field of view and permits the development of specific shapes properly accommodating all physical dimensions. Reference source materials must be properly understood and their physical properties separated from any influence of the characterization of the detector field of view. This has been accomplished in the present work with the use and comparison of 133Ba and other appropriate “point” sources and point comparison made at locations where inverse square distance behavior exists for both the reference source and the “point sources”. This study has also led to the development of a simple approach to experimentally verify reference source selfattenuation and to quantify the non-uniform angular distribution of gamma rays emerging from the reference source. This approach will work for any reference source material content.Inclusion of this distribution is important as its neglect can cause significant errors (bias) in the calibration result. When the calibrations resulting from the above activities are coded into higher efficiency detectors, and combined with corresponding model integrity, the opportunity exists to make both improved fixed location and reportable dynamic scan measurements. Adoption of this approach promises to greatly improve the efficiency of process operations and additionally enable other related applications.