AGHCF gaseous-effluent tritium sampling system: design considerations and performance testing results.

Performance testing results involving the operation of a gaseous-effluent tritium sampling system are provided. A system description, including design improvements made over the course of several years of operation, is also presented. The sampling technique is based on using ethylene glycol bubblers for collecting tritiated water entrained as vapor in a nitrogen purge gas from a hot cell facility. Tritiated gas is converted to water vapor using a high temperature copper oxide bed; newly-formed water molecules are readily collected in a second set of ethylene glycol bubblers. For a single bubbler containing 20 mL of ethylene glycol, the tritium collection efficiency was determined to be 98.3%. To optimize performance and to minimize the volume of mixed waste generated, tests were performed to evaluate variations in the collection efficiency as a function of sorbent volume. For purposes of comparison the performance of water-filled bubblers was also evaluated. With two water bubblers connected in series the overall collection efficiency was quite satisfactory at 95.4%, although appreciable water losses were evident. The bed oxidation efficiency for tritiated gas was evaluated over a range of temperatures; at a catalyst bed temperature of 350 degrees C nearly all the gas was converted to water vapor. System memory effects were assessed by measuring the amount of residual contamination present in the bubblers after purging the sampling system for a week offline. Contamination problems arising from previous sampling periods were inconsequential. The overall uncertainty associated with the total quantity of tritium discharged to the environment based on the results of the performance testing was estimated to be +/-24% at the 95% confidence level.

Summary of stationary and personal air sampling measurements made during a plutonium glovebox decommissioning project.

Workplace air sampling was performed during the decommissioning of a previously active plutonium glovebox facility located at Argonne National Laboratory. Personal air samplers (PAS) were used to measure breathing zone activity concentrations of workers engaged in size-reducing contaminated gloveboxes. Stationary air samplers (SAS) were used to measure the work area activity concentrations and test their application in providing representative sampling of breathing zone activity concentrations. The relative response of these samplers (PAS:SAS) was tracked during the course of the decommissioning work, with results yielding favorable agreement to within a factor of +/-5. A cascade impactor was used to determine the particle size distribution of workplace aerosols. The average activity median aerodynamic diameter (AMAD) was estimated to be 3.0 microm, with a corresponding geometric standard deviation of 2.4. A gas-flow proportional counter was utilized to measure the gross alpha activity collected on both the SAS glass fiber and the PAS cellulose fiber filters. A subset of this filter group was subsequently analyzed using an alpha spectrometer post radiochemical processing and isotopic separation. The quantity of alpha activity measured on the SAS filters was generally within +/-30% of the alpha spectrometry measurements. However, measurements made of the activity present on the PAS cellulose fiber filters were consistently underestimated using a gas-flow proportional counter, suggesting a small correction factor of 15-20% to account for the absorption of alpha particle emissions.

Design, calibration, and operation of 220Rn stack effluent monitoring systems at Argonne National Laboratory.

A group of stack effluent monitoring systems have been developed to monitor discharges of 220Rn from a hot cell facility at Argonne National Laboratory. The stack monitors use flow-through scintillation cells and are completely microprocessor-based systems. A method for calibrating the stack monitors in the laboratory and in the field is described. A nominal calibration factor for the stack monitoring systems in use is 15.0 cts min-1 per kBq m-3 (0.56 cts min-1 per pCi L-1) +/- 26% at the 95% confidence level. The plate-out fraction of decay products in the stack monitor scintillation cells, without any pre-filtering, was found to be nominally 25% under normal operating conditions. When the sample was pre-filtered upstream of the scintillation cell, the observed cell plate-out fraction ranged from 16-22%, depending on the specific sampling conditions. The instantaneous 220Rn stack concentration can be underestimated or overestimated when the steady state condition established between 220Rn and its decay products in the scintillation cell is disrupted by sudden changes in the monitored 220Rn concentration. For long-term measurements, however, the time-averaged response of the monitor represents the steady state condition and leads to a reasonable estimate of the average 220Rn concentration during the monitoring period.

Measured absorbed dose rates from semi-infinite hemispherical volumes of 133Xe.

Surface absorbed dose rates from different hemispheric volumes of 133Xe have been measured directly with an extrapolation chamber. The results indicate that a linear relationship exists between the radius of the cloud volume and the surface absorbed dose rate for radii between 0 and 23 cm. If cloud volumes with radii larger than 23 cm are taken to be infinite with respect to the range of the charged particles emitted, the absorbed dose rate calculated based on that assumption will be within the uncertainty of any measurement of absorbed dose rate that might be made. For hemispheric volumes having radii less than or equal to 23 cm, the surface absorbed dose rate in tissue-equivalent material, in mGy h-1, is approximated (+/- 20%) by the product of [1.30 mGy h-1 cm-1 kBq-1 cm3] X [cloud radius, cm] X [cloud activity concentration, kBq cm-3].