The following is a synopsis of an article published in Health Physics of Radiation-Generating Machines; Proceedings of the 30th Midyear Topical Meeting of the Health Physics Society, San Jose, CA; January, 1997.
In 1981 The Cyclotron Corporation built a 42 MeV H- cyclotron at their facility in Berkeley, CA. Extensive factory testing of the machine included a 200-hour endurance run at full energy and maximum-rated beam current (200 µA). The resulting fast-neutron beam activated a portion of the concrete wall in the vault, especially around a 40 cm diameter "hot spot" in the forward cone of the neutron beam.
During the next eleven years the premises remained under license and were used by a second accelerator company (CTI Cyclotron Systems, Inc. now located in Knoxville, TN). However under the terms of the lease, that particular vault was not used for testing neutron-producing machines so as not to cause any further activation.
In the autumn of 1992, CTI Cyclotron Systems, Inc. undertook to vacate the premises. A careful survey and assay of the activated vault was conducted; concrete core samples at various positions and depths in the activated portion of the wall were drilled and assayed by means of an HPGe gamma ray spectrometer system comprising a 36 mm dia. by 14 mm deep, N-type, coaxial detector (EG&G Ortec "LO-AX" series) in conjunction with a personal computer-based multi-channel analyzer (Tennelec "NUCLEUS"), which was calibrated for efficiency as a function of gamma-ray energy for the particular, fixed source-detector geometry.
A least-squares fit of log-counts against log-energy in KeV, utilizing known energies and abundances in samples containing 232Th and 152Eu, over the range of energies spanning 122 KeV to 2.6 MeV, established the slope of the detector efficiency curve. The absolute efficiency intercept was established by means of NBS-certified sources containing 137Cs and 60Co.
The mean and standard deviation of the background rate measured over intervals of 100,000 seconds using a 10 gm "dummy" concrete sample from outside the vault were 0.01 Bq/gm for each of the principal activation products. Any reading below 0.03 Bq/gm (three standard deviations above zero) was considered to be at or below background.
The assay procedure entailed drilling with a deep, 2 cm dia. masonry bit, capturing the concrete dust, weighting out small (5-10 gm) samples to be sealed individually in 2.5 cm x 5 cm plastic bags which were then affixed directly to the face of the HPGe detector. The appropriate regions of interest (ROI) in the measured spectra were defined by prior calibration procedures. The lower limit of detectability (0.03 Bq/gm) was imposed by ambient background from 40K and 232Th and its daughters, which are naturally present in the earth at that location, as well as in the concerete structure of the building.
At the time of the assay, the exposure dose rate was 200 ur/hr at the surface of the wall, and 85 ur/hr measured 1 meter from the wall in front of the hot spot. The principal activation products which were still present in the concrete after eleven years were 152Eu (t1/2 = 13.4 y)and 60Co (t1/2 = 5.27 y). The peak activity concentrations were 5.0 Bq/gm 152Eu and 1.0 Bq 60Co at a depth of 15 cm at the center of the hot spot, falling by a nominal factor of 2.5 at positions 75 cm radially distant from the center.
An estimate of residual activity in the structural reinforcing rod in the concrete walls was inferred from measurements made on small steel parts--nuts, bolts, bracketry, etc. embedded in the concrete within the perimeter of the hot spot. These parts contained concentrations of 60Co of 1.4 Bq/gm at the surface and 8.0 Bq/gm at a depth of 3 cm. The variation of activity concentration in structural steel throughout the volume of concrete was assumed to be similar to that of the concrete drill samples themselves.
Demolition and Decommissioning Plan
A demolition and decommissioning plan was submitted to the Radiologic Health Branch of the State of Califirnia in December, 1992. However, the plan was not immediately implemented. The facility's radiation license was kept in force for an additional 2.5 years to accommodate the temporary installation of another radiation-producing machine at the facility in behalf of a U.S. Department of Energy National Laboratory.
In June, 1995, a final, amended demolition and decommissioning (D&D) plan was approved by the Radiologic Health Branch. The plan entailed removal and disposal of contaminated concrete from the vault wall so as to bring the exposure dose rate in the vault down to 5 ur/hr or less above ambient background. Rubble from the region of the wall immediately around the hot spot was to be sealed in 55 gal. drums and shipped to a low-level radiation waste site for disposal. The overall average activity concentration of 152Eu was 0.75 Bq/gm (20 pCi/gm) and that of 60Co was 0.22 Bq/gm (6.0 pCi/gm) within this portion of the rubble. A licensed environmental services company provided the 55-gal. drums, transported the waste, provided temporary storage, and served as a broker--handling the administrative details relating to the final disposal at an approved site in the State of Washington.
The rest of the rubble--approximately 20 cubic yards--was estimated to have less than 0.2 Bq/gm (5 pCi/gm) of activation products, and was determined to be of sufficiently low level as to qualify for disposal at an ordinary landfill. As part of the approval process for the D&D plan, an analysis of that part of the waste stream destined for the ordinary landfill was carried out utilizing a computer code entitled IMPACTS-BRC1, as recommended (and kindly provided) by the Radiologic Health Branch Staff. The software title document states: "This code represents an analysis methodology for determining the radiological impacts associated with the disposal of below-regulatory-concern (BRC) waste. It calculates the impacts under a number of scenarios, including those involved with transportation of the waste, the sorting/incinerating/recycling of the waste, and the final and post-disposal considerations".
Given certain prior assumptions about the region's climate and soil conditions, etc., and given the quantity and concentration of the various isotopes under consideration, the code analyzes the exposure effects due to transport, hazards to unauthorized intruders, handling by dump-site workers, erosion, chemical leaching, and the spreading of radioactive material through groundwater, etc., over time intervals commensurate with the isotope half-life over the active, working lifetime of the dumpsite. The code also evaluates the potential hazard associated with possible habitation on the site after closure of the landfill. In the present case, the analysis showed that the radiation exposure to workers and to the population at large was dominated by the initial transport by truck to the dumpsite, and that the maximum exposure was of the order of 0.1 person-mrem/year in the aggregate. Thus, disposal of this rubble as ordinary waste was determined to have no significant impact. It was, indeed, shown to be "below regulatory concern".
In July, 1995, the license for disposal of activated rubble was secured and the D&D plan was implemented. The process itself, while noisy, dirty, and disruptive, went reasonably smoothly and quickly. The portion of the vault walls which were removed were then replaced with fresh concrete and the premises cleaned up and surveyed.
The results of the survey were accepted by the Radiologic Health Branch as meeting the criteria as defined in the D&D plan. Accordingly, the facility radioactive materials license was finally terminated in February, 1996, and the premises released for general occupancy.
References1) Sandia National Laboratories, Safety and Reliability Analysis Division. IMPACTS-BRC, version 2.0. Developed for the U.S. Nuclear Regulatory Commission Office of Nuclear Material Safety and Safeguards Division of Waste Management; October, 1989.
Return to CRA Home Page
Return to List of Publications