Pitfalls In Air Sampling For Radioactive Particulates

All uranium mills are required to perform sampling and analysis for radioactive particulates in their gaseous effluent streams and in the environment. Pacific Northwest Laboratory was requested by the U.S. Nuclear Regulatory Commission (NRC) to provide technical assistance to them for their Uranium Mill Health Physics Appraisal Program. In conducting appraisals, air sampling methods used at NRC-licensed mills were reviewed and several deficiencies noted. This paper includes only environmental and effluent particulate sampling although much of the information is applicable to both in-plant and environmental samples. First, the components of a proper sampling program are discussed: program objectives, program design, sampler design, analyses, quality assurance, and data handling. Then the specific deficiencies, or the "pitfalls" from the first 8 mill appraisals are discussed. The first consideration in establishing an air sampling program is defining the objectives of the program. What is air sampling suppose to accomplish? Many of the deficiencies we have observed have resulted because the desired objectives were not clearly established in the minds of the radiation safety staff. PROGRAM OBJECTIVES An environmental air sampling program ought to fulfill the following seven objectives. The first is to: 1) [demonstrate regulatory compliance]. Although a goal of most programs, regulatory compliance, is not well understood. One has not only to comply with the conditions of the source materials licensee, but one must also demonstrate compliance with 10CFR20 and 40CFR190. For example, 10CFR20.106 states: "A licensee shall not possess, use, or transfer licensed material so as to release to an unrestricted area radioactive material in concentrations which exceed the limits specified in Appendix B, Table II of this part .... For purposes of this section, concentrations may be averaged over a period not greater than one year." Even if a mill's license does not require sampling at the site boundary of maximum concentration, a sample may be necessary to demonstrate compliance with 10CFR20. Most mill personnel are painfully familiar with 40CFRl90.10, which states: "Operations.... shall be conducted in such a manner as to provide reasonable assurance that: (a) The annual dose equivalent does not exceed 25 millirems to the whole body.... of any member of the public as the result of exposures to planned discharges of radioactive materials, radon and its daughters excepted... from uranium fuel cycle operations..." This means a licensee's sampling program must give "reasonable assurance" that the member of the general public receiving in the most exposure gets no more than 25 millirems per year. The sampling program necessary to provide that assurance may or may not be a license requirement. However, merely meeting the license requirements and the explicit regulatory requirements does not necessariarly ensure an adequate effluent and environmental air sampling program. The second objective of the environmental air sampling program, is to 2) [identify the source(s) of contaminants]. This will include not only the routine program, but special sampling for verification of sources and nonsources. Only after sampling can a mill operator be assured that roof vents, laboratory hoods, and other localized ventilation systems are not making a significant contribution to environmental releases. An environmental sampling program should also allow the mill operator to fulfill the third objective, to 3) [estimate exposures]. Even before 40CFR190, a sampling program should have provided the mill operator with the information necessary to determine the dose to the "fence post" person, or at least to determine if doses were well below the 10CFR20 limits previously allowed. The program should 4) [detect and measure unplanned releases]. If there is a fire, a scrubber failure, or if a drum of yellowcake breaks open, measured releases will almost always be lower than conservative estimates. Whether or not a system to provide sampling during accidents is needed is almost always a cost-benefit decision. In general, uranium operations do not sample just in case an accident may occur. Yet they may decide on continuous air sampling in lieu of intermittant sampling partially because of the potential for accidents. Another objective of air sampling is 5) [to provide information on the effectiveness of control systems]. This is always a concern with new or modified equipment and may dictate sampling frequency in other situations as well. For instance, if a small leak in a bag filter cannot be detected by other means, then more frequent stack sampling may be indicated. A routine effluent and environmental monitoring program should also fulfill the sixth objective,