SAMPLING AND ANALYSIS FOR RADON-222 DISSOLVED IN G R O U N D WATER AND S U R F A C E WATER L. D E W A Y N E CECIL

U.S. Geological Survey, INEL Project Office, P.O. Box 2230, Idaho Falls, ID 83402, U.S.A. T H O M A S F. G E S E L L

U.S. Department of Energy, Radiological and Environmental Sciences Laboratory, 785 DOE Place, Idaho Falls, ID 83402, U.S.A.

(Received May 1990) Abstract. Radon-222 is a naturally occurring radioactive gas in the uranium-238 decay series that has traditionally been called, simply, radon. The lung cancer risks associated with the inhalation of radon decay products have been well documented by epidemiological studies on populations of uranium miners. The realization that radon is a public health hazard has raised the need for sampling and analytical guidelines for field personnel. Several sampling and analytical methods are being used to document radon concentrations in ground water and surface water worldwide but no convenient, single set of guidelines is available. Three different sampling and analytical methods - bubbler, liquid scintillation, and field screening - are discussed in this paper. The bubbler and liquid scintillation methods have high accuracy and precision, and small analytical method detection limits of 0.2 and 10 pCi/l (picocuries per liter), respectively. The field screening method generally is used as a qualitative reconnaissance tool.

1. Introduction Radon-222 is a naturally occurring radioactive gas in the uranium-238 decay series (Figure 1) with a half-life of 3.82 days. Traditionally radon-222 is simply called radon. Radon, with an atomic number of 86, is the heaviest of the inert gas elements. The half-life is the time necessary for one half of the atoms of the radionuclide present to radioactively decay. The shorter half-lived isotopes are radon-220 (half-life of 55 seconds) and radon219 (half-life of 4 seconds) and are called thoron and actinon, respectively. With the exceptions of lead-210 (half-life of 20 years) and polonium-210 (half-life of 138 days), the half-lives of the decay products of radon are relatively short. The health risks associated with the inhalation of radon decay products have been well documented in populations of uranium miners. The decay products of radon irradiate the bronchial epithelium leading to a demonstrably increased risk of lung cancer. Risk rates developed from the uranium miner studies have been extrapolated to the radon exposures received by the general public. These exposures arise from natural concentrations of radon decay products in outdoor air as well as from elevated concentrations of radon decay products found in structures. Concentrations in structures can increase because of restricted amounts of air available for dilution and advective transport into structures. The source of radon entering structures include radon in the rocks and soil on which the structure rests, radon in the building materials, radon in water used within the structure and, in some cases, natural or liquified petroleum gas burned in non-vented gas appliances (Gesell and Prichard, 1980, Prichard and Gesell 1984). Environmental Monitoring and Assessment 20: 55-66, 1992.

56

L. D F W A Y N E C E ( ' I L A N [ ) T H O M A S F. (;E,NELL I

I

I

I

I

I

I

I

I

I

I

I

238

234

230

226

.,Q

=E c

222

t~

218

214 I

II

I

210

206

i

I

1

I

I

I

I

I

I

I

I

I

I

81

82

83

84

85

86

87

88

89

90

91

92

Atomic number, Z

0-0669

Fig. 1. The principal uranium-238 decay series. The U.S. Environmental Protection Agency has estimated that during an average life span of 70 years, between 2000 and 40000 excess lung cancer fatalities are produced in the United States by radon released from public water supplies (Cothern 1987). The average concentration of radon in public water supplies in the United States generates a lifetime risk of 1 in 10 000, a risk larger than for any contaminant currently regulated under the

RADON-222

SAMPL1NG

AND ANALYSIS

57

Safe Drinking Water Act. The realization of the potential importance of radon in water has stimulated an increase in studies of radon in water. Sampling and measurement of radon are complicated by the volatility and relatively short half-life (3.82 days) of this gaseous element. For example, radon is effectively removed by aeration or the use of an in-line granular activated charcoal filtration system. Therefore, sample techniques need to minimize aeration of the water and avoid transfer of samples through the open air. Losses of radon during transportation and storage may occur if sampling vessels are permeable to gas, or if they have leaky stopcocks and caps. Several radon sampling and analytical methods for water are used at the present time but there is no convenient, single set of guidelines. This paper documents three commonly used sampling methods and their associated analytical methods for radon in water; liquid scintillation, bubbler, and field screening.

2. Sampling Techniques Radon-222 can occur naturally in both ground water and surface water. Ground water samples for radon analyses should be collected after a well has been pumped sufficiently for water to reach chemical stability (Wood 1981). Radon sampling from surface water is less frequently conducted because natural aeration decreases radon concentrations, but it may be sampled in the same manner as ground water, using a peristaltic pump. Analytical determination of radon in surface water can provide an estimate of the location and magnitude of ground water seepage (Lee and Hollyday 1982). Water samples for radon analyses can be collected with the bubbler, liquid scintillation, and field-screening methods. These three sampling techniques are summarized and compared in Table I. 2.1. LIQUIDSCINTILLATIONMETHOD The following equipment is needed for sampling: (1) a sampling funnel and tube with standard faucet fitting (Figure 2); (2) a 20-ml glass syringe with an 18-gauge hypodermic needle; and (3) two glass scintillation vials per sampling site with 5 ml of liquidscintillation solution in each. The liquid-scintillation solution is available from the New England Nuclear Corporation. ~ For ground-water samples from domestic wells, the sampling funnel and tube are attached to a faucet at the wellhead and before the storage tank, if present. For ground-water samples from accessible wellheads, a peristaltic or submersible pump is used to supply an uncontaminated, full-column flow. (A short, flexible piece of tubing inserted into a 1000-ml beaker is an acceptable substitute for the funnel and tube.) A steady stream is allowed to flow out of the funnel for 2-3 minutes. This purges the funnel and tube and assures a fresh sample. The flow should be adjusted to minimize turbulence and allow excess water to gently spill over the edge of the funnel (U.S. Environmental Protection Agency 1978). For greater accuracy in determining radon concentrations, sampling vials containing scintillation cocktail should be weighed before and after sampling to determine the mass of the sample. Use of brand or firm names in this report is for identification purposes only and does not constitute endorsement by the U.S. GeologicalSurveyor the Departmentof Energy.

58

L. DEWAYNE CECIL AND THOMAS F. GESELL TABLE I Comparison of sampling and analytical methods for radon in water.

Method

Cost of equipment Detection Advantages (US dollars) limits

Liquid Scintillation

Included in 10 pCi/l analytical costs

Disadvantages

1. High accuracy

1. Possibility of degassing with 2. High precision syringe and a resultant loss 3. Minimal collec- of radon tion and handling time

Application 1. Elevated concentration of radon

Cost of analysis (US dollars) $10-30 per sample plus shipping

4. Relatively inexpensive 5. U.S. Environmental Protection Agency approved method Bubbler

Included in 0.2 pCi/1 analytical costs

I. High accuracy 2. High precision

1. Breaking of bubblers during transport

l.Natural ground $50 per water sample plus 2. Surface water shipping

3. U.S Environ2. Time required to mental Protecprepare bubbler tion Agency and collect approved method sample 3. Loss of vacuum during transport $5000 Field Screening

50 pCi/I

1. In situ screening with rapid results 2. Short collection and handling time

1. Low accuracy 2. Low precision

I. Screening tool

Field analysis included in equipment costs

3. Relatively inexpensive after initial equipment cost

T h e tip o f the h y p o d e r m i c needle is p l a c e d a b o u t 3 c m b e l o w the surface o f the w a t e r in the funnel. T h e syringe and h y p o d e r m i c needle are rinsed three times by d r a w i n g a few milliliters o f w a t e r f r o m the funnel a n d ejecting it. A f t e r the syringe is rinsed, 12 to 15 ml o f w a t e r are w i t h d r a w n f r o m the funnel slowly to a v o i d d e v e l o p i n g t o o great a v a c u u m . R a p i d w i t h d r a w a l o f the w a t e r s a m p l e will create a large negative pressure that m a y d r a w

RADON-222

SAMPLING

AND

ANALYSIS

59

9.1132

Fig. 2. Funnel, tubing, and syringefor collectingwater samples for analysis of radon-222.

radon and other dissolved gasses out of solution and cause some loss of radon during transfer to the sampling vial. The syringe is pointed upward and any air bubbles and excess water are ejected slowly until 10 ml of sample remain in the syringe. The sample is now ready for transfer to the sampling vial. The tip of the needle is carefully placed at the bottom of the liquid-scintillation solution. The water is ejected slowly from the syringe into the vial without causing any turbulence or air bubbles that may result in the loss of radon. The needle is removed from the vial and the cap is tightened securely to prevent leakage. The sampling vial is vigorously shaken to promote the movement of the radon gas from the water into the scintillation cocktail. A small piece of tape is placed on the cap of the sampling vial with the following: (1) site identification and sample number, and (2) collection time (date, hour, and minute). Tape must not be placed on tile wall of the collection vial and the walls should not be marked in any way because this will interfere with the liquid-scintillation counting. This procedure is repeated to obtain two separate samples from each well. The collection of two samples is recommended as a means of checking reproducibility and also as a check for leakage from a vial. The concentration reported for dissolved radon for a given sampling site is generally an average of the two vials. The two vials are placed in a small foam-packed cardboard box and sent by overnight delivery to the laboratory.

60

L. DI:.WAYNE C I ' ( ' I L AND THOMAS F. (IESELL

2.2.

BUBBLER METHOD

An all-glass deemanation bubbler (Figure 3) is used for collecting water samples for radon analysis (Yang 1987). Before sampling the bubbler is evacuated (assembled as shown in Figure 3) to less than 10 millitorr through stopcock 1, while stopcock 2 remains closed. Stopcock 1 is than closed. The bubbler is now ready for sampling ground water or surface water. For greater accuracy, bubblers should be weighed before and after sampling to determine the mass of the sample.

/1~~

~

Stopcock Screw clam P'~I~ Waterlevel

Inleport t /~-

Flexible tubing

~

0 ring

~ L~ ~iti'pcock 2 ()

Frittedglassdisc \

9.1133

Fig. 3.

All-glass deemanation bubbler used for collecting water samples for analysis of radon-222.

RADON-222

SAMPLING

AND

ANALYSIS

61

For ground-water samples, a peristaltic or submersible pump is used to supply an uncontaminated, full-column flow of sample water through a short, flexible piece of tubing to the inlet port of the evacuated bubbler. When the main chamber is about 60 percent full (as shown in Figure 3), stopcock 1 is closed. The bubbler is labeled with: (1) site identification and sample number, and (2) collection time (date, hour, and minute). Masking tape is used to secure stopcocks 1 and 2 in the closed position. The bubbler is placed inside a foam-packed cardboard box and sent by overnight delivery to the laboratory. For surface-water sampling, the inlet port end of the evacuated bubbler is gently placed into the water body until the space between stopcock 1 and the inlet port is filled with water. Stopcock l is opened to draw sample water directly into the main chamber of the bubbler until the chamber is 60 percent full. Stopcock I is then closed. Samples are labeled and transported to the labortory in the same manner described for the ground-water samples. Selection of sites for surface-water sampling should be in areas that offer complete natural mixing of the water if possible. Few data are currently available on the vertical and horizontal distribution characteristics of radon in stream and rivers. 2.3. FIELD SCREENINGMETHOD The equipment for extraction of radon consists of a standard l-liter plastic bottle and a modified cap (Reimer 1977). The cap is constructed of hard plastic and is machined and threaded to fit the top of the bottle to create a seal with the opening on the bottle and not with the shoulder or neck of the bottle (Figure 4). The cap is also machined to accommodate a fitting that contains a septum through which a hypodermic needle with syringe can be inserted to extract a sample from the closed bottle. A hypodermic needle with an opening on the wall of the needle is preferred to one with the opening at the tip because the openings at the tip can be plugged with rubber material from the septum on the cap. Needles with side-wall openings are available from the same sources that provide standard needles. The l-liter bottle should be graduated so that various quantities of water can be sampled. Typically, 750-ml water samples are collected. The exact quantity is not critical as this is a semiquantitative method and volumes within 30 ml of this amount are sufficient. Continuous addition or removal of water to achieve the desired volume should be avoided because turbulence causes degassing. Once the water sample is taken, the bottle is capped and shaken vigorously for 30 seconds. This agitation strips the dissolved gases from the water into the air space in the bottle. The bottle should then stand for 3 minutes to allow most of the remaining bubbles in the water to combine with the air in the bottle. The hypodermic needle is attached to a closed syringe and then is inserted through the septum on the cap, the bottle is squeezed to force some air into the syringe, and the syringe is removed. The quantity of gas introduced into the syringe can be varied and will depend on the type of analytical equipment used and the required accuracy of the measurement. Typically, a sample of 10 ml is sufficient. A small, tight-fitting piece of tubing is placed

62

L. D U W A Y N E

/

CECIL

AND

THOMAS

F. GESELL

Rubber septum

Quick disconnect fitting j ~

1-liter sample bottle

Fig. 4. Samplebottle for the field screeningmethod. over the tip of the hypodermic needle to prevent the escape of gases until the sample is transferred to the field analytical equipment.

3. Analytical techniques The analytical techniques for each method will only be covered briefly. Thoron (radon220) and actinon (radon-219) can interfere with the results of the analyses for radon but the contribution of these isotopes is generally considered to be insignificant because of

RADON-222

SAMPLING

AND

ANALYSIS

63

their short half-lives and the small crustal abundance of the radon-219 precursor. A more detailed discussion of the analytical methods can be found in the reports cited at the end of this paper. If questions of radon isotope specificity are important in a project, for example in studies of radioactive-waste sites or of uranium mining and milling locations, expert advice should be sought in the planning stage regarding analytical techniques for the differentiation of radon isotopes. For most uncontaminated, natural waters, the following analytical techniques should provide a good measure of the radon concentration. 3.1. LIQUID SCINTILLATIONMETHOD Analyses of water samples by this method are by direct liquid-scintillation counting (Prichard and Gesel11977). The sampling vial, with a 10-ml water sample and 5 ml of the mineral-oil based liquid-scintillation solution, is shaken in the laboratory to mix the fluids. This is followed by a wait of at least 3 hours followed by the measurement of the radioactivity in a commercial liquid-scintillation counter. The analytical method detection limit is about 10 pCi/l (1 pCi equals 37 milliBecquerel) for a 10-ml sample. 3.2. BUBBLERMETHOD Analyses of water samples using the bubbler method are by direct deemanation and alpha-scintillation counting (Yang 1987). The deemanation bubbler containing the sample is attached to a deemanation system in the laboratory as shown in Figure 5. The deemanation system consists of the following parts: (1) a bubbler; (2) a glass drying tube packed with anhydrous magnesium perchlorate to remove moisture, and sodium hydroxide on asbestos plus soda-lime to remove carbon dioxide; (3) a manometer to indicate the pressure inside the system during transfer of radon from the bubbler to the alpha-scintillation cell, and to detect leaks in the system; (4) a helium tank to purge dissolved radon into the cell; (5) an alpha-scintillation cell; and (6) a vacuum pump. The radon is purged from the water sample with helium gas into the alpha-scintillation cell. The radioactivity of the radon is determined in an alpha-scintillation counter. The analytical method detection limit for a 60-ml sample of liquid is about 0.2 pCi/1. 3.3. FIELD SCREENINGMETHOD The field-analytical instrument is an alpha-scintillometer that contains a phosphor-coated cell. These cells should be airtight in order to quantitatively accept a small gas sample from the syringe. Such a system has been described by Reimer (1977). The 10-ml gas sample is injected into the cell and the radioactivity is counted for 1 minute. Each analytical system is calibrated and the gas extraction efficiency is compared to a laboratory standard. For first-order approximations of radon concentrations, the temperature of the water can be ignored, but it is considered for more accurate calculations. The analytical method detection limit is about 50 pCi/1 for a 10-ml gas sample.

64

L. D E W A Y N E

CECIL

AND THOMAS

F. G E S E L L

Manometer (3) Alpha-scintillation cell (5)

~

J~ =

Drying

(2)

Bubbler---..~ (1)

-I I Vacuum pump (6)

9-1134

Fig. 5. Radon-222deemanationsystem.

4. Quality Assurance Because of the relatively short half-life of radon (3.82 days), standard solutions with known concentrations for quality assurance are difficult to prepare. A prototype radon222 standard for water samples is described by Inn et al. (1984). Other standard quality assurance practices that help to maintain a quality assurance program include: (1) 'blind' repicate samples submitted to evaluate within-laboratory reproducibility; (2) cross-check analyses of replicate samples among two or more laboratories to determine agreement; and (3) submission of distilled water blanks. Extraction efficiency to quantitatively determine the effectiveness of removal of radon from the water sample can be established by making multiple extractions for analyses on

RADON-222

SAMPLING

AND

ANALYSIS

65

the same sample. By comparing the radioactivity measured in each extraction, a percentof-extraction efficiency can be calculated. Generally, radon concentrations in ground water are much greater than radium-226 concentrations a n d are unrelated (Durrance 1986, p. 368). If substantial radium is present in water samples used in the determination of extraction efficiency, ingrowth o f r a d o n from the radium must be taken into account. Table I summarizes and c o m p a r e s the three methods presented here for sampling and analysis for r a d o n in water.

5. S u m m a r y and Conclusions No single sampling m e t h o d can satisfy all project requirements. If quantification o f dissolved radon concentrations is the purpose o f a study, then the bubbler or liquid scintillation m e t h o d s are preferred over the field screening method. However, if qualitative analytical results will meet the needs of a study, or if water resources to be sampled are in remote areas and rapid results are required, then the field screening method m a y be sufficient. The b u b b l e r a n d liquid scintillation methods have high accuracy and precision with relatively small analytical m e t h o d detection limits of 0.2 and 10 pCi/1 respectively. These analytical m e t h o d detection limits are adequate even for detailed, basic research project objectives. As with all analytical chemistry, accuracy, and precision are dependent on conscientious, well-trained field and laboratory personnel.

References Cothern, C.R.: 1987, 'Estimating the health risk of radon in drinking water', J. American Water Well Association, 70 (4), 153-158. Cothern, C. R., Jarvis, A.N., Whittaker, E. L., and Battist, Lewis: 1984, 'Radioactivity in environmental samples: Calibration standards, measurement methods, quality assurance, and data analysis'; Environment International, 10, 109-116. Durrance, E. M.: 1986, Radioactivity in Geology, Principles and Applications, John Wiley and Sons, Inc., New York, N.Y., 441 pp. Gesell, T.F., and Prichard, H.M.: 1980, 'The contribution of radon in tap water to indoor radon concentrations', in Gesell, T.F., and Lowder, W.M. (Eds.), Natural Radiation Environment Ill, U.S. Department of Energy, CONF-780422, vol. 2, pp. 1347-1363. Hiltebrand, D.J., Dykserl, J. E., and Raman, K.: 1987, 'Radon in water supply wells: Treatment facility requirements and costs', in Graves, Barbara (ed.), Radon, Radium and other Radioactivity in Ground Water, Chelsea, Michigan, Lewis Publishers Inc., pp. 521-534. Inn, K. G. W., Mullen, P.A., and Hutchinson, J. M.R.: 1984, 'Radioactivity standards for environmental monitoring II', Environment International, 10, 91-97. Lee. R. W., and Hollyday, E. F.: 1987, 'Radon measurements in streams to determine location and magnitude of ground-water seepage', in Graves, Barbara (ed.), Radon, Radium and other Radioactivity in Ground Water, Chelsea, Michigan, Lewis Publishers, Inc., pp. 241-249. Lowry, J, D., and Brandon, J.E.: 1981, 'Removal of radon from ground water supplies using granular activated carbon or diffused aeration', University of Maine, Department of Civil Engineering, Orono, Maine. Prichard, H.M., and Gesell, T.F.: 1977, 'Rapid measurements of 222Rn concentrations in water with a commercial liquid scintillation counter', Health Physics, 33, 577-581. Prichard, H. M., and Gesell, T. F.: 1984, 'Radon in the environment', in Lett, J. T., Ehmann, U. K., and Cox, A, B. (Eds.), Advances in Radiotion Biology, Orlando, Florida, Academic Press, Inc., pp. 391-428.

66

L. DEWAYNE CECIL AND THOMAS F. GESELL

Reid, G. W., Lassovszky, Peter, and Hathaway, Steven: 1985, 'Treatment, waste management and cost for removal of radioactivity from drinking water', Health Physics, 45, 671-694. Reimer, G. M.: 1977, 'Fixed-volume inlet system for an alpha-sensitive cell adapted for radon measurement', U.S. Geological Survey Open-File Report 77-409, 3 pp. U.S. Environmental Protection Agency: 1978, Radon in Water Sampling Program, EPA/EERF-Manual-78-1, 11 pp. Wood, W. W.: 198 ! 'Guidelines for collection and field analysis of groundwater samples for selected unstable constituents', Techniques of Water-Resources Investigations of the U.S. Geological Survey Book 1, Chapter D2, 24 pp. Yang, I.C.: 1987, 'Sampling and analysis of dissolved radon-222 in surface and ground water', in Graves, Barbara (Ed.), Radon, Radium, and Other Radioactivity in Ground Water, Chelsea, Michigan, Lewis Publishers Inc., pp. 193-203.

Sampling and analysis for radon-222 dissolved in ground water and surface water.

Radon-222 is a naturally occurring radioactive gas in the uranium-238 decay series that has traditionally been called, simply, radon. The lung cancer ...
511KB Sizes 0 Downloads 0 Views