Airborne Concentration cont'd

DOE-STD-1136-2004

Guide of Good Practices for Occupational Radiological Protection in Uranium Facilities

relationship between fixed air samplers and spot samples collected at various locations in typical

working areas. Discrepancies as great as two orders of magnitude are not unusual.

This deficiency of GAS monitoring for individual exposure records is caused by the high dilution

factors that tend to reduce the airborne concentrations before and after contamination reaches the filter

head. Much of the air sampled by a GAS originates in another part of the area and does not pass near

enough to pick up contamination from the source, effectively reducing the measured concentration by

dilution of the collected sample. A release of activity from a malfunctioning containment system can

produce large activity concentrations in the breathing zone of the worker. These concentrations are

effectively diluted in an unpredictable manner by one or two orders of magnitude before the

contamination reaches a monitor located only a meter away. It has been demonstrated that in some

operations (such as welding over a short period of time) differences of as much as a factor of 5 between

the right and left lapel PAS measurements can be expected.

Most of the field studies that have compared urinalysis results with air sampling in natural uranium

facilities have, in general, indicated very poor correlation between the estimated exposures and the

bioassay data. This suggests that individual exposure records of uranium workers based on GAS methods

have limited validity.

The potential for release of gaseous UF6, and subsequent generation of its soluble hydrolysis product

UO2F2, requires special air-sampling considerations in uranium conversion and gaseous diffusion plants,

relative to those plants handling less reactive compounds. In these plants, effective processing, as well as

worker safety, requires a high degree of containment. Continuous GAS operation to detect loss of

containment, coupled with spot air samples, constitute the typical sampling strategy. A study conducted at

the Oak Ridge Gaseous Diffusion Plant, concluded that shift- long air samples collected in the general

working areas were of little use in predicting worker's urinary uranium excretion. The slight correlation

observed was not statistically significant at the 95% confidence level. Thus, gaseous contaminants behave

much like particulate contaminants in that localized concentrations can be much greater than the average

concentration measured by GAS. These researchers also found that smear samples of alpha activity on work

surfaces in the area may provide a better indicator of uranium intake than the GAS records.

Although transuranic material is handled by DOE uranium facilities only as feed contamination, the

unusual characteristics of the transuranic elements make them worthy of separate consideration. The low

maximum permissible concentrations specified for these elements and their frequently low specific

activities cause extreme difficulties in detection of significant airborne activity. Operations involving

significant amounts of elements such as plutonium should be conducted in a ventilated glove-box

environment and with monitoring systems capable of detection of small releases involving a few times one

DAC. Special CAMs (GAS) and fixed BZSs are the standard air-sampling methods used in facilities of this

category in the United States.

A clear example of the wide variations in observed air-activity concentrations that can occur with

different sampling techniques is provided by data from the Three Mile Island Nuclear Generating Station,

which is typical of operations in a large open building (EGG 1988). Between June and September 1983,

over 40 multi-person entries were made into the containment building, providing 949 work-hours of PAS

data. Five stationary air monitors were operated continuously at strategic locations throughout the

building, and each entry was preceded by the collection and analysis of a high-volume grab sample. All

samples were analyzed by a gamma spectrometer, primarily to detect cesium-137, and

4-5