- Email: [email protected]
Procedures for the determination of 222Rn exhalation and effective 226Ra activity in soil samples
Applied Radiation and Isotopes PERGAMON
Applied Radiation and Isotopes 50 (1999) 1039±1047
Procedures for the determination of 222Rn exhalation and eective 226 Ra activity in soil samples V. GoÂmez Escobar a, *, F. Vera Tome b, J.C. Lozano c a
Departamento de FõÂsica, Escuela PoliteÂcnica, Universidad de Extremadura, 10071 CaÂceres, Spain b Departamento de FõÂsica, Universidad de Extremadura, 06071 Badajoz, Spain c Laboratorio de Radiactividad Ambiental, Facultad de Ciencias, Universidad de Salamanca, 37008 Salamanca, Spain Received 23 March 1998; received in revised form 26 May 1998
Abstract Two methods for measuring 222 Rn exhalation and eective 226 Ra in soil samples were studied. In the ®rst determination, the method employed was based on the adsorption of radon onto activated charcoal and subsequent measurement of the activity of its daughters with an HPGe (high-purity germanium) detector. In the second, vials containing an aqueous suspension of the sample, mixed with an insoluble high eciency mineral oil scintillation cocktail, were measured using a low-level liquid scintillation counter. Studies of optimum sampling time, eciency in both procedures, variation of 226 Ra eciency with quenching, as well as the eect of sample amount and granulometry upon the quenching parameter, were carried out. The two methods were applied to the determination of 222 Rn exhalation and eective 226 Ra in environmental samples. # 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction Radon can be considered to be one of the most dangerous radioactive elements in the environment. Its character as a noble gas allows it to spread through the atmosphere. The greatest fraction of natural radiation exposure in humans results from inhalation of the decay products of radon (PorstendoÈrfer, 1991), that once ®xed in airborne particulates can be retained in human lungs, increasing lung cancer risks (Jacobi, 1988). The radon atoms (222 Rn) are formed by radioactive decay of radium (226 Ra). In the soil, radon molecules can escape from grains of soil by diusion or recoil into the soil pores (PorstendoÈrfer, 1991). This process, called emanation, increases with increasing soil moisture content, ®rst quickly and later more slowly, in a way that can be considered almost constant in the normal range of soil moisture content (between 2% and saturation). A conservative estimate of the fraction of
* Corresponding author.
radon generated that leaves solid grains (emanation rate or emanation power) is 0.25 (25%) (Malanca et al., 1994), although lower and higher values have also been reported (Wilkening, 1985). Exhalation designates the escape of radon from a material to the atmosphere. This process can occur by molecular diusion or by convection (Schery et al., 1988). Besides the moisture content, the exhalation of radon is positively correlated with temperature and wind speed and negatively with pressure, so that these factors must be considered in the determination of exhalation rates in environmental measurements. Since the topmost layer of soil has the greatest in¯uence on the exhalation rate (3±4 cm for thoron exhalation according to Megumi and Mamuro (1974)) because radon has more diculty in reaching the atmosphere from deeper layers, we have preferred an `in situ' method in the sampling procedure. There are several methods (direct or indirect) to determine radon activity (Urban and Schmitz, 1991; Harley, 1992; etc.). One of the most extensively used consists in capturing radon in activated charcoal and then measuring radon decay products (214 Pb and 214 Bi)
0969-8043/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 9 - 8 0 4 3 ( 9 8 ) 0 0 1 2 1 - 3
1040
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
Fig. 1. Diagram of the device designed for sampling.
222
Rn exhalation
in the charcoal by gamma spectrometry after equilibrium. With regard to radium, in attempting to look for possible relationships with exhalation and at least to search for tendencies, we are only interested in the fraction of radium that produces emanation of radon. This is called emanating radium (Landa and Nielson,
1987) or eective radium (Wilkening, 1985). The value of eective radium depends on the emanation rate and therefore will be in¯uenced by the moisture content (Hinton and Whicker, 1985), temperature, granulometry (Amin et al., 1995), etc. In this work, two procedures are presented for the determination of 222 Rn exhalation and eective 226 Ra in soil samples. The paper is divided into four sections. In Section 2, the two procedures are described as well as the counting equipment used in each case. The studies carried out to obtain the eciency in the two procedures are presented in Section 3, together with the experimental results. Finally, the procedures are applied to environmental samples and the results are presented in Section 4. 2. Experimental procedure 2.1.
222
Rn exhalation
The method is based on the adsorption of radon by activated charcoal and the subsequent measurement of the activity by means of its daughter products with an HPGe detector.
Fig. 2. Gamma spectrum from a 222 Rn exhalation sample with the emissions considered in the activity determination. The lower spectrum is from the same sample after treatment at 808C for 24 h.
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1041
Fig. 3. Corrected activity of 222 Rn (Bq/m2) from one sample measured several times. The line de®nes the weighting average obtained from all measurements.
The designed device (see Fig. 1) is a variation of that used by Megumi and Mamuro (1972). It consists of a metallic square case (40 cm side) in which there are two steel mesh screens that allow air to pass. In the ®rst mesh screen a layer of silica gel (3±6 mm particle diameter) was used in order to eliminate the moisture content, because the adsorption of radon by the activated charcoal can be in¯uenced by water content of the air (Nakayama et al., 1994). In the second, 480 g of activated charcoal (Mod. RL-12 of F&J Specialty Products) of 8 16 mesh were added. This amount was imposed by the volume of the container used for gamma counting (1 l Marinelli beaker). Sampling was performed by placing the sampler over the area of interest on the ground, avoiding possible escapes of air under the sides of the sampler, by inserting it 1 cm into the ground. After sampling (024 h), activated charcoal was poured into the Marinelli beaker, which was immediately closed and sealed for its later measurement in an HPGe detector (20% eciency) for 60 000 s. To calibrate the HPGe detector, a standard source Marinelli beaker was prepared with 480 g of activated charcoal to which a 226 Ra standard solution (9652 5
Bq) was added uniformly. Once the Marinelli beaker was closed and sealed, a waiting period of one month was necessary to reach radioactive equilibrium between 226 Ra and its daughters. The standard source was measured regularly several times after preparation (from 30 days until 588 days), the mean activity value obtained being (964 2 10) Bq. This result shows the stability of the HPGe detector used as well as the hermeticity of the Marinelli beaker. Moreover, another source was prepared as a blank, for which 480 g of recently treated activated charcoal was added to the Marinelli beaker. In all cases, the activated charcoal used was free of radon. For this, it had to be heated for at least 24 h at 808C in an oven before use. Fig. 2 shows the gamma spectrum from a sample of 222 Rn exhalation as well as the measurement of the same sample after being treated for 24 h at 808C. The 222 Rn activity was determined by the most important emissions of its daughters (see Fig. 2): 214 Pb (241.92 keV (7.47%), 295.22 keV (19.2%) and 352.00 keV (37.1%)) and 214 Bi (609.3 keV (46.0%), 1120.2 keV (15.0%) and 1764.5 keV (15.9%)), although the two last emissions were not considered when low
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1042
counting statistics were obtained in the measurements. Computer programs were written to carry out the decay corrections. These programs were tested, giving satisfactory results. Thus the corrected 222 Rn activity (Bq/m2) of a given sample measured at dierent times after sampling is shown in Fig. 3. In Fig. 3, the activity was calculated for all the emissions considered in the determination. It can be seen that this activity is practically constant and independent of the measurement time, thus also showing the hermeticity of the Marinelli beaker. 2.2. Eective
226
Ra in soil samples
Soil samples were taken from a layer 10 cm deep with a hollow cylindrical tube of 30 mm diameter. The samples were dried at 1108C until constant mass and then calcined at 6008C for at least 24 h in order to eliminate the organic material which can produce additional quenching. Several aliquots were suspended in 10 ml of de-ionized water inside low-diusion polyethylene (PE) vials (from Packard). Then 10 ml of insoluble high eciency mineral oil scintillation
(HEMOS) cocktail (NEF-957G, NEN Research Products) was added. The method proposed is similar to that used in our laboratory for the determination of 222 Rn and 226 Ra in aqueous samples (GoÂmez Escobar et al., 1996). The 222 Rn dissolved in water is extracted into the organic phase (cocktail) producing scintillation. The sample is measured after reaching radioactive equilibrium (one month), then the 226 Ra in the sample can be easily determined from the 222 Rn counts. Samples were measured for 500 min in a low-level Wallac 12202 Quantulus liquid scintillation spectrometer, which provides very low background count rates. It is also equipped with a pulse-shape analyzer (PSA) which separates pulses caused by alpha or beta particles into dierent spectra. An optimum PSA value of 100 was used in all cases (GoÂmez Escobar et al., 1996). To evaluate quenching, Quantulus uses a sealed 226 Ra capsule to give the external standard quench parameter (SQP(E)) as an estimator of the quench level in the sample (Kaihola, 1994). After measurement, the LSC alpha spectra were transferred to a PC through an RS 232 C serial interface and stored on hard or ¯oppy disks for subsequent analysis.
Fig. 4. Variation of 222 Rn corrected activity (Bq/m2) with sampling time. All the samples were collected with the same sampler except those marked with a square: (a) S1 standard source; (b) S2 standard source.
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1043
Fig. 5. Variation of eciency with the amount of soil sample. The values are referred to the value of the smallest mass.
3. Experimental results 3.1.
222
Rn exhalation
In order to calibrate the procedure and to obtain its eciency, the ®rst step was the preparation of adequate standard sources. For this preparation, equal aliquots of 226 Ra calibrated solution were distributed homogeneously over a volume of de-ionized water or calcined soil with a very low activity of 226 Ra, in a tray. Once well distributed, the water standard sources were evaporated to dryness and the soil standard source heated in order to eliminate excess water, followed by homogeneously distributing the soil over the Table 1 Standard sources prepared for the determination of the
222
tray. The dierent standard sources prepared are listed in Table 1. Then the samplers were attached to the standard sources (trays). Several samples were taken from each standard source, with dierent sampling times and using dierent samplers. The variation of 222 Rn corrected activities versus sampling time for the standard sources S1 and S2 is shown in Fig. 4. As can be seen, there is a linear relationship between corrected activity (Bq/m2) and sampling time, not in¯uenced by the sampler (square outlines mark the results obtained using dierent samplers). Although the standard sources were prepared using activities higher than those found in natural sites, no saturation eect seems to occur. From Fig. 4,
Rn exhalation procedure eciency
Standard source
Material of the tray
Eective area of the tray (m2) Matrix
Volume or mass
S1 S2 S3 S4
plastic plastic stainless steel stainless steel
0.0958 0.0958 0.16 0.16
400 ml 1 kg 600 ml + VYNS (3 ml) 700 ml
de-ionized water soil de-ionized water de-ionized water
1044
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
Fig. 6.
214
Po counting evolution with time after preparation for one of the planchets.
one can deduce that there is no optimum sampling time; we chose 24 h in order to assure good statistics and also because this period will include any daily cyclic changes in exhalation. As can also be observed in Fig. 4, the values obtained with the soil standard source S2 were lower than those obtained with the water standard source S1. It seems that 222 Rn has a high residence time in the soil source that hinders it reaching the activated charcoal. In order to test this assumption, another plastic tray was used, adding gradually to it amounts of a soil sample with a high concentration of 226 Ra. Fig. 5 shows the variation of relative eciency of the samples collected as a function of the soil mass added. As can be seen, there is a great in¯uence of changes in the eciency with the soil amount (thickness of the sample) when this is small, reaching a practically constant eciency for amounts greater than 200 g of sample. Taking into account these results, the best way to reproduce the environmental sampling would be to use a source of zero thickness, from which it can be assumed that all the radon escapes. For this, another two standard sources (S3 and S4; see Table 1) were prepared on stainless steel trays (0.16 m2 area) specially made to attach to our samplers. De-ionized
water was used as matrix and 226 Ra standard solution to spike the samples. For the preparation of the source S3, a small amount of a spreading agent (VYNS: vinyl chloride and vinyl acetate copolymer in cyclohexanone) was added in order to obtain a more homogeneous distribution of the 226 Ra standard solution. The eciencies calculated for the dierent measurements performed (more than 20) with these two standard sources were very reproducible, the average value obtained being (26.9 2 0.2)%. Although the standard sources S3 and S4 had thickness close to zero, the possible retention of 222 Rn was evaluated by preparing six stainless steel planchets of 5 mm diameter with the same ratio of de-ionized water, 226 Ra activity and area of the tray found in the standard sources. These planchets were measured in a gas ¯ow gridded ionization chamber from Intertechnique (Mod. Adagio 114) of 2p geometry. Fig. 6 shows the behaviour observed and the ®t to the theoretical growth of 222 Rn for one of these planchets (214 Po activity increased due to the retention of 222 Rn in the deposits). From the results obtained for the six planchets prepared, we determined that (9.8 2 0.5)% of the produced radon did not escape from the planchet.
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1045
Fig. 7. Variation of eciency with the SQP(E) parameter. The solid line represents the ®t with the function described in the text.
Taking this value into account, the global eciency in the 222 Rn exhalation procedure was (29.92 0.4)%.
3.2. Eective
226
Ra in soil samples
Recently, there has been a procedure proposed for the determination of 226 Ra in aqueous samples using a low-level liquid scintillation counter (GoÂmez Escobar et al., 1996). This uses the extraction of 222 Rn from its own aqueous phase. We propose the determination of eective 226 Ra under saturation conditions of moisture content, in which the emanation presents a maximum value although it varies slightly over a wide range (under saturation conditions, emanation is only 10% higher than for 6% relative moisture, according to Strong and Levins (1982)). Thus, the eective 226 Ra activity concentration under these conditions can be considered as a conservative value. The aqueous sample method did not consider quenching corrections for the sample. Although for some soil samples this assumption is valid, other samples show high degrees of quenching and so a correction has to be made.
For this purpose, some sources were prepared by adding dierent amounts of a standard solution of 226 Ra to 10 ml of de-ionized water to which dierent volumes of CCl4 had been added as quenching agent. Solutions thus prepared were added to PE vials containing 10 ml of HEMOS cocktail. After secular equilibrium (one month), the same sample was measured again thus determining the 222 Rn in equilibrium with 226 Ra and the 226 Ra present in the sample. The variation of the eciency, E (%), of these sources with quenching value (SQP(E) is shown in Fig. 7. The experimental results were ®tted to the function ln E a b SQP E c SQP 2 E , the values obtained for the parameters being: a = 19.1 2 3.3; b = ÿ (8.52 0.8)10 ÿ 2 and ÿ5 c = (7.9 2 0.5)10 . The eciency curve shows a strong gradient in the region of SQP(E) values between 800 and 850, which is due to the use of the PSA function. High degrees of quenching (SQP(E) below 800) produce very low eciencies and then increase alphacrosstalk in the b spectra. Because of the great in¯uence of quenching on the eciency, a study was carried out in order to determine the eect of the amount of sample in suspension
1046
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
Fig. 8. Variation of the SQP(E) parameter with the mass of soil and the grain diameter (d).
and the granulometry on quenching. The variation of the SQP(E) parameter with the amount of previously treated soil (calcined at 6008C) is presented in Fig. 8. Dierent grain diameters (d) are also distinguished in Fig. 8. It can be seen that the ¯uctuations of the SQP(E) value, even with similarly prepared samples (cf. the 1 g samples), mask the speci®c eect of the grain size on the quenching, so that it can be considered of minor importance in the range of 1±3 g of soil samples. The aliquots are thus usually taken to be of 1 g after sieving to 2 mm grain size. Although in Fig. 8 a drop in SQP(E) values is observed for amounts larger than 3 g (also observed with other samples), this behaviour is not generalizable and depends on each soil. One gram samples seem adequate for routine use. Although this amount of soil (1±3 g) is greater than the sample size usually taken for analysis in other procedures and techniques, it still requires careful sampling to assure representativity. The eective 226 Ra was therefore determined for several aliquots of the same soil sample after homogenization, in order to check the reproducibility of the procedure. A soil rich in 226 Ra was used in this test and the results were
(778 2 42), (846 2 46), (875 2 49), (8192 44), (776 2 42), (754 2 41), (778 2 42), (7402 40), (831 2 45), (7242 40), (7942 43), (725 2 39) and (745 2 40) Bq/kg. It can be seen that the results show a high reproducibility, with the weighted mean activity concentration being (7782 12) Bq/kg. The minimum detectable activity concentration (MDA) of the proposed method was evaluated using (Currie, 1968) MDA Bq=kg
LD V T E 60
with
p LD 2:71 4:65 CB T where V is the mass of sample (kg), T (min) the sample measurement time (which is the same as for the background), E the eciency and CB the background count rate (counts per minute) using a radiochemical blank. For 1 g of soil sample and 500 min of counting a MDA value of 0.05 Bq/kg was obtained. This value is notably lower than the usual activity concentrations of eective 226 Ra found in environmental soils (Wilkening, 1990).
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047 Table 2 222 Rn exhalation rate and eective
226
1047
Ra activity concentration values for three types of samples Rn exhalation rate (Bq m ÿ 2 s ÿ 1)
Place of sampling
222
Tailings of a uranium mine Surroundings of a uranium mine Extremadura University campus
1.7420.04 0.14420.003 0.03820.002
4. Application to environmental samples The methods described for the determination of Rn exhalation and eective 226 Ra in soils were applied to environmental samples. In Table 2, the results obtained from three dierent kinds of samples are shown. The Extremadura University campus was chosen as a site of low radium activity, while a uranium mine in Extremadura and its surroundings were chosen as rich and medium radium content sites, respectively. As can be seen, results of the 222 Rn exhalation rate and the eective 226 Ra activity concentration in soils show the same tendency although no quantitative correlations can be obtained. These tendencies have been further con®rmed in numerous other samples, but additional study is required before quantitative conclusions can be drawn. For this, other parameters such as moisture content of the soils and granulometry and additional corrections involving dierent layers of the soil must be considered.
222
Acknowledgements Thanks are due to the Empresa Nacional de Residuos Radiactivos (ENRESA), the Spanish national agency for radioactive waste management and to the Empresa Nacional del Uranio, S.A. (ENUSA) (Project No. 0703401) for ®nancial support.
References Amin, Y.M., Mahat, R.H., Doraisamy, S.J., Subramaniam, S.Y., 1995. The eect of grain size on the radon emanation rate. Appl. Radiat. Isot. 46, 621. Currie, L., 1968. Limits for qualitative detection and quantitative determination. Anal. Chem. 40, 586. GoÂmez Escobar, V., Vera TomeÂ, F., Lozano, J.C., MartõÂ n SaÂnchez, A., 1996. Determination of 222 Rn and 226 Ra in aqueous samples using a low-level liquid scintillation counter. Appl. Radiat. Isot. 47, 861.
Eective
226
Ra activity concentration (Bq kg ÿ 1)
66372378 4825 6.821.1 Harley, J.H., 1992. Measurements of 222 Rn: A brief history. Radiat. Prot. Dosim. 45, 13. Hinton, T.G., Whicker, F.W., 1985. A ®eld experiment on Rn ¯ux from reclaimed uranium mill tailings. Health Phys. 48, 421. Jacobi, W., 1988. Lung cancer risk from environmental exposure to radon daughters. ICRP Publication 50. Radiat. Prot. Dosim. 24, 19. Kaihola, L., 1994. Cosmic particle spectrum as a quench monitor in low level liquid scintillation spectrometry. Nucl. Instrum. Methods Phys. Res. A 339, 295. Landa, E.R., Nielson, K.K., 1987. Use of track detectors for the evaluation of emanating radium content of soil samples. Uranium 4, 97. Malanca, A., Pessina, V., Dallara, G., 1994. Radon availability from the ground of the Brazilian State of Rio Grande do Norte. Nucl. Geophys. 8, 301. Megumi, K., Mamuro, T., 1972. A method for measuring radon and thoron exhalation from the ground. J. Geophys. Res. 77, 3052. Megumi, K., Mamuro, T., 1974. Emanation and exhalation of radon and thoron gases from soil particles. J. Geophys. Res. 79, 3357. Nakayama, Y., Nagao, H., Mochida, I., Kawabuchi, Y., 1994. Adsorption of radon on active carbon. Carbon 32, 1544. PorstendoÈrfer, J., 1991. Properties and behaviour of radon and thoron and their decay products in the air. In: 5th International Symposium on the Natural Radiation Environment. ISSN 1018-5593 (Report EUR 14411 EN), p. 69. Schery, S.D., Holford, D.J., Wilson, J.L., Phillips, F.M., 1988. The ¯ow and diusion of radon isotopes in fractured porous media. Part 1. Finite slabs. Radiat. Prot. Dosim. 24, 185. Strong, K.P., Levins, D.M., 1982. Eect of moisture content on radon emanation from uranium ore and tailings. Health Phys. 42, 27. Urban, M., Schmitz, J., 1991. Radon and radon daughters metrology. In: 5th International Symposium on the Natural Radiation Environment. ISSN 1018-5593 (Report EUR 14411 EN), p. 151. Wilkening, M., 1985. Radon transport in soil and its relation to indoor radioactivity. Sci. Tot. Environ. 45, 219. Wilkening, M. (1990) Radon in the Environment. Elsevier, Amsterdam.
Applied Radiation and Isotopes 50 (1999) 1039±1047
Procedures for the determination of 222Rn exhalation and eective 226 Ra activity in soil samples V. GoÂmez Escobar a, *, F. Vera Tome b, J.C. Lozano c a
Departamento de FõÂsica, Escuela PoliteÂcnica, Universidad de Extremadura, 10071 CaÂceres, Spain b Departamento de FõÂsica, Universidad de Extremadura, 06071 Badajoz, Spain c Laboratorio de Radiactividad Ambiental, Facultad de Ciencias, Universidad de Salamanca, 37008 Salamanca, Spain Received 23 March 1998; received in revised form 26 May 1998
Abstract Two methods for measuring 222 Rn exhalation and eective 226 Ra in soil samples were studied. In the ®rst determination, the method employed was based on the adsorption of radon onto activated charcoal and subsequent measurement of the activity of its daughters with an HPGe (high-purity germanium) detector. In the second, vials containing an aqueous suspension of the sample, mixed with an insoluble high eciency mineral oil scintillation cocktail, were measured using a low-level liquid scintillation counter. Studies of optimum sampling time, eciency in both procedures, variation of 226 Ra eciency with quenching, as well as the eect of sample amount and granulometry upon the quenching parameter, were carried out. The two methods were applied to the determination of 222 Rn exhalation and eective 226 Ra in environmental samples. # 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction Radon can be considered to be one of the most dangerous radioactive elements in the environment. Its character as a noble gas allows it to spread through the atmosphere. The greatest fraction of natural radiation exposure in humans results from inhalation of the decay products of radon (PorstendoÈrfer, 1991), that once ®xed in airborne particulates can be retained in human lungs, increasing lung cancer risks (Jacobi, 1988). The radon atoms (222 Rn) are formed by radioactive decay of radium (226 Ra). In the soil, radon molecules can escape from grains of soil by diusion or recoil into the soil pores (PorstendoÈrfer, 1991). This process, called emanation, increases with increasing soil moisture content, ®rst quickly and later more slowly, in a way that can be considered almost constant in the normal range of soil moisture content (between 2% and saturation). A conservative estimate of the fraction of
* Corresponding author.
radon generated that leaves solid grains (emanation rate or emanation power) is 0.25 (25%) (Malanca et al., 1994), although lower and higher values have also been reported (Wilkening, 1985). Exhalation designates the escape of radon from a material to the atmosphere. This process can occur by molecular diusion or by convection (Schery et al., 1988). Besides the moisture content, the exhalation of radon is positively correlated with temperature and wind speed and negatively with pressure, so that these factors must be considered in the determination of exhalation rates in environmental measurements. Since the topmost layer of soil has the greatest in¯uence on the exhalation rate (3±4 cm for thoron exhalation according to Megumi and Mamuro (1974)) because radon has more diculty in reaching the atmosphere from deeper layers, we have preferred an `in situ' method in the sampling procedure. There are several methods (direct or indirect) to determine radon activity (Urban and Schmitz, 1991; Harley, 1992; etc.). One of the most extensively used consists in capturing radon in activated charcoal and then measuring radon decay products (214 Pb and 214 Bi)
0969-8043/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 9 - 8 0 4 3 ( 9 8 ) 0 0 1 2 1 - 3
1040
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
Fig. 1. Diagram of the device designed for sampling.
222
Rn exhalation
in the charcoal by gamma spectrometry after equilibrium. With regard to radium, in attempting to look for possible relationships with exhalation and at least to search for tendencies, we are only interested in the fraction of radium that produces emanation of radon. This is called emanating radium (Landa and Nielson,
1987) or eective radium (Wilkening, 1985). The value of eective radium depends on the emanation rate and therefore will be in¯uenced by the moisture content (Hinton and Whicker, 1985), temperature, granulometry (Amin et al., 1995), etc. In this work, two procedures are presented for the determination of 222 Rn exhalation and eective 226 Ra in soil samples. The paper is divided into four sections. In Section 2, the two procedures are described as well as the counting equipment used in each case. The studies carried out to obtain the eciency in the two procedures are presented in Section 3, together with the experimental results. Finally, the procedures are applied to environmental samples and the results are presented in Section 4. 2. Experimental procedure 2.1.
222
Rn exhalation
The method is based on the adsorption of radon by activated charcoal and the subsequent measurement of the activity by means of its daughter products with an HPGe detector.
Fig. 2. Gamma spectrum from a 222 Rn exhalation sample with the emissions considered in the activity determination. The lower spectrum is from the same sample after treatment at 808C for 24 h.
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1041
Fig. 3. Corrected activity of 222 Rn (Bq/m2) from one sample measured several times. The line de®nes the weighting average obtained from all measurements.
The designed device (see Fig. 1) is a variation of that used by Megumi and Mamuro (1972). It consists of a metallic square case (40 cm side) in which there are two steel mesh screens that allow air to pass. In the ®rst mesh screen a layer of silica gel (3±6 mm particle diameter) was used in order to eliminate the moisture content, because the adsorption of radon by the activated charcoal can be in¯uenced by water content of the air (Nakayama et al., 1994). In the second, 480 g of activated charcoal (Mod. RL-12 of F&J Specialty Products) of 8 16 mesh were added. This amount was imposed by the volume of the container used for gamma counting (1 l Marinelli beaker). Sampling was performed by placing the sampler over the area of interest on the ground, avoiding possible escapes of air under the sides of the sampler, by inserting it 1 cm into the ground. After sampling (024 h), activated charcoal was poured into the Marinelli beaker, which was immediately closed and sealed for its later measurement in an HPGe detector (20% eciency) for 60 000 s. To calibrate the HPGe detector, a standard source Marinelli beaker was prepared with 480 g of activated charcoal to which a 226 Ra standard solution (9652 5
Bq) was added uniformly. Once the Marinelli beaker was closed and sealed, a waiting period of one month was necessary to reach radioactive equilibrium between 226 Ra and its daughters. The standard source was measured regularly several times after preparation (from 30 days until 588 days), the mean activity value obtained being (964 2 10) Bq. This result shows the stability of the HPGe detector used as well as the hermeticity of the Marinelli beaker. Moreover, another source was prepared as a blank, for which 480 g of recently treated activated charcoal was added to the Marinelli beaker. In all cases, the activated charcoal used was free of radon. For this, it had to be heated for at least 24 h at 808C in an oven before use. Fig. 2 shows the gamma spectrum from a sample of 222 Rn exhalation as well as the measurement of the same sample after being treated for 24 h at 808C. The 222 Rn activity was determined by the most important emissions of its daughters (see Fig. 2): 214 Pb (241.92 keV (7.47%), 295.22 keV (19.2%) and 352.00 keV (37.1%)) and 214 Bi (609.3 keV (46.0%), 1120.2 keV (15.0%) and 1764.5 keV (15.9%)), although the two last emissions were not considered when low
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1042
counting statistics were obtained in the measurements. Computer programs were written to carry out the decay corrections. These programs were tested, giving satisfactory results. Thus the corrected 222 Rn activity (Bq/m2) of a given sample measured at dierent times after sampling is shown in Fig. 3. In Fig. 3, the activity was calculated for all the emissions considered in the determination. It can be seen that this activity is practically constant and independent of the measurement time, thus also showing the hermeticity of the Marinelli beaker. 2.2. Eective
226
Ra in soil samples
Soil samples were taken from a layer 10 cm deep with a hollow cylindrical tube of 30 mm diameter. The samples were dried at 1108C until constant mass and then calcined at 6008C for at least 24 h in order to eliminate the organic material which can produce additional quenching. Several aliquots were suspended in 10 ml of de-ionized water inside low-diusion polyethylene (PE) vials (from Packard). Then 10 ml of insoluble high eciency mineral oil scintillation
(HEMOS) cocktail (NEF-957G, NEN Research Products) was added. The method proposed is similar to that used in our laboratory for the determination of 222 Rn and 226 Ra in aqueous samples (GoÂmez Escobar et al., 1996). The 222 Rn dissolved in water is extracted into the organic phase (cocktail) producing scintillation. The sample is measured after reaching radioactive equilibrium (one month), then the 226 Ra in the sample can be easily determined from the 222 Rn counts. Samples were measured for 500 min in a low-level Wallac 12202 Quantulus liquid scintillation spectrometer, which provides very low background count rates. It is also equipped with a pulse-shape analyzer (PSA) which separates pulses caused by alpha or beta particles into dierent spectra. An optimum PSA value of 100 was used in all cases (GoÂmez Escobar et al., 1996). To evaluate quenching, Quantulus uses a sealed 226 Ra capsule to give the external standard quench parameter (SQP(E)) as an estimator of the quench level in the sample (Kaihola, 1994). After measurement, the LSC alpha spectra were transferred to a PC through an RS 232 C serial interface and stored on hard or ¯oppy disks for subsequent analysis.
Fig. 4. Variation of 222 Rn corrected activity (Bq/m2) with sampling time. All the samples were collected with the same sampler except those marked with a square: (a) S1 standard source; (b) S2 standard source.
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1043
Fig. 5. Variation of eciency with the amount of soil sample. The values are referred to the value of the smallest mass.
3. Experimental results 3.1.
222
Rn exhalation
In order to calibrate the procedure and to obtain its eciency, the ®rst step was the preparation of adequate standard sources. For this preparation, equal aliquots of 226 Ra calibrated solution were distributed homogeneously over a volume of de-ionized water or calcined soil with a very low activity of 226 Ra, in a tray. Once well distributed, the water standard sources were evaporated to dryness and the soil standard source heated in order to eliminate excess water, followed by homogeneously distributing the soil over the Table 1 Standard sources prepared for the determination of the
222
tray. The dierent standard sources prepared are listed in Table 1. Then the samplers were attached to the standard sources (trays). Several samples were taken from each standard source, with dierent sampling times and using dierent samplers. The variation of 222 Rn corrected activities versus sampling time for the standard sources S1 and S2 is shown in Fig. 4. As can be seen, there is a linear relationship between corrected activity (Bq/m2) and sampling time, not in¯uenced by the sampler (square outlines mark the results obtained using dierent samplers). Although the standard sources were prepared using activities higher than those found in natural sites, no saturation eect seems to occur. From Fig. 4,
Rn exhalation procedure eciency
Standard source
Material of the tray
Eective area of the tray (m2) Matrix
Volume or mass
S1 S2 S3 S4
plastic plastic stainless steel stainless steel
0.0958 0.0958 0.16 0.16
400 ml 1 kg 600 ml + VYNS (3 ml) 700 ml
de-ionized water soil de-ionized water de-ionized water
1044
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
Fig. 6.
214
Po counting evolution with time after preparation for one of the planchets.
one can deduce that there is no optimum sampling time; we chose 24 h in order to assure good statistics and also because this period will include any daily cyclic changes in exhalation. As can also be observed in Fig. 4, the values obtained with the soil standard source S2 were lower than those obtained with the water standard source S1. It seems that 222 Rn has a high residence time in the soil source that hinders it reaching the activated charcoal. In order to test this assumption, another plastic tray was used, adding gradually to it amounts of a soil sample with a high concentration of 226 Ra. Fig. 5 shows the variation of relative eciency of the samples collected as a function of the soil mass added. As can be seen, there is a great in¯uence of changes in the eciency with the soil amount (thickness of the sample) when this is small, reaching a practically constant eciency for amounts greater than 200 g of sample. Taking into account these results, the best way to reproduce the environmental sampling would be to use a source of zero thickness, from which it can be assumed that all the radon escapes. For this, another two standard sources (S3 and S4; see Table 1) were prepared on stainless steel trays (0.16 m2 area) specially made to attach to our samplers. De-ionized
water was used as matrix and 226 Ra standard solution to spike the samples. For the preparation of the source S3, a small amount of a spreading agent (VYNS: vinyl chloride and vinyl acetate copolymer in cyclohexanone) was added in order to obtain a more homogeneous distribution of the 226 Ra standard solution. The eciencies calculated for the dierent measurements performed (more than 20) with these two standard sources were very reproducible, the average value obtained being (26.9 2 0.2)%. Although the standard sources S3 and S4 had thickness close to zero, the possible retention of 222 Rn was evaluated by preparing six stainless steel planchets of 5 mm diameter with the same ratio of de-ionized water, 226 Ra activity and area of the tray found in the standard sources. These planchets were measured in a gas ¯ow gridded ionization chamber from Intertechnique (Mod. Adagio 114) of 2p geometry. Fig. 6 shows the behaviour observed and the ®t to the theoretical growth of 222 Rn for one of these planchets (214 Po activity increased due to the retention of 222 Rn in the deposits). From the results obtained for the six planchets prepared, we determined that (9.8 2 0.5)% of the produced radon did not escape from the planchet.
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
1045
Fig. 7. Variation of eciency with the SQP(E) parameter. The solid line represents the ®t with the function described in the text.
Taking this value into account, the global eciency in the 222 Rn exhalation procedure was (29.92 0.4)%.
3.2. Eective
226
Ra in soil samples
Recently, there has been a procedure proposed for the determination of 226 Ra in aqueous samples using a low-level liquid scintillation counter (GoÂmez Escobar et al., 1996). This uses the extraction of 222 Rn from its own aqueous phase. We propose the determination of eective 226 Ra under saturation conditions of moisture content, in which the emanation presents a maximum value although it varies slightly over a wide range (under saturation conditions, emanation is only 10% higher than for 6% relative moisture, according to Strong and Levins (1982)). Thus, the eective 226 Ra activity concentration under these conditions can be considered as a conservative value. The aqueous sample method did not consider quenching corrections for the sample. Although for some soil samples this assumption is valid, other samples show high degrees of quenching and so a correction has to be made.
For this purpose, some sources were prepared by adding dierent amounts of a standard solution of 226 Ra to 10 ml of de-ionized water to which dierent volumes of CCl4 had been added as quenching agent. Solutions thus prepared were added to PE vials containing 10 ml of HEMOS cocktail. After secular equilibrium (one month), the same sample was measured again thus determining the 222 Rn in equilibrium with 226 Ra and the 226 Ra present in the sample. The variation of the eciency, E (%), of these sources with quenching value (SQP(E) is shown in Fig. 7. The experimental results were ®tted to the function ln E a b SQP E c SQP 2 E , the values obtained for the parameters being: a = 19.1 2 3.3; b = ÿ (8.52 0.8)10 ÿ 2 and ÿ5 c = (7.9 2 0.5)10 . The eciency curve shows a strong gradient in the region of SQP(E) values between 800 and 850, which is due to the use of the PSA function. High degrees of quenching (SQP(E) below 800) produce very low eciencies and then increase alphacrosstalk in the b spectra. Because of the great in¯uence of quenching on the eciency, a study was carried out in order to determine the eect of the amount of sample in suspension
1046
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047
Fig. 8. Variation of the SQP(E) parameter with the mass of soil and the grain diameter (d).
and the granulometry on quenching. The variation of the SQP(E) parameter with the amount of previously treated soil (calcined at 6008C) is presented in Fig. 8. Dierent grain diameters (d) are also distinguished in Fig. 8. It can be seen that the ¯uctuations of the SQP(E) value, even with similarly prepared samples (cf. the 1 g samples), mask the speci®c eect of the grain size on the quenching, so that it can be considered of minor importance in the range of 1±3 g of soil samples. The aliquots are thus usually taken to be of 1 g after sieving to 2 mm grain size. Although in Fig. 8 a drop in SQP(E) values is observed for amounts larger than 3 g (also observed with other samples), this behaviour is not generalizable and depends on each soil. One gram samples seem adequate for routine use. Although this amount of soil (1±3 g) is greater than the sample size usually taken for analysis in other procedures and techniques, it still requires careful sampling to assure representativity. The eective 226 Ra was therefore determined for several aliquots of the same soil sample after homogenization, in order to check the reproducibility of the procedure. A soil rich in 226 Ra was used in this test and the results were
(778 2 42), (846 2 46), (875 2 49), (8192 44), (776 2 42), (754 2 41), (778 2 42), (7402 40), (831 2 45), (7242 40), (7942 43), (725 2 39) and (745 2 40) Bq/kg. It can be seen that the results show a high reproducibility, with the weighted mean activity concentration being (7782 12) Bq/kg. The minimum detectable activity concentration (MDA) of the proposed method was evaluated using (Currie, 1968) MDA Bq=kg
LD V T E 60
with
p LD 2:71 4:65 CB T where V is the mass of sample (kg), T (min) the sample measurement time (which is the same as for the background), E the eciency and CB the background count rate (counts per minute) using a radiochemical blank. For 1 g of soil sample and 500 min of counting a MDA value of 0.05 Bq/kg was obtained. This value is notably lower than the usual activity concentrations of eective 226 Ra found in environmental soils (Wilkening, 1990).
V. GoÂmez Escobar et al. / Applied Radiation and Isotopes 50 (1999) 1039±1047 Table 2 222 Rn exhalation rate and eective
226
1047
Ra activity concentration values for three types of samples Rn exhalation rate (Bq m ÿ 2 s ÿ 1)
Place of sampling
222
Tailings of a uranium mine Surroundings of a uranium mine Extremadura University campus
1.7420.04 0.14420.003 0.03820.002
4. Application to environmental samples The methods described for the determination of Rn exhalation and eective 226 Ra in soils were applied to environmental samples. In Table 2, the results obtained from three dierent kinds of samples are shown. The Extremadura University campus was chosen as a site of low radium activity, while a uranium mine in Extremadura and its surroundings were chosen as rich and medium radium content sites, respectively. As can be seen, results of the 222 Rn exhalation rate and the eective 226 Ra activity concentration in soils show the same tendency although no quantitative correlations can be obtained. These tendencies have been further con®rmed in numerous other samples, but additional study is required before quantitative conclusions can be drawn. For this, other parameters such as moisture content of the soils and granulometry and additional corrections involving dierent layers of the soil must be considered.
222
Acknowledgements Thanks are due to the Empresa Nacional de Residuos Radiactivos (ENRESA), the Spanish national agency for radioactive waste management and to the Empresa Nacional del Uranio, S.A. (ENUSA) (Project No. 0703401) for ®nancial support.
References Amin, Y.M., Mahat, R.H., Doraisamy, S.J., Subramaniam, S.Y., 1995. The eect of grain size on the radon emanation rate. Appl. Radiat. Isot. 46, 621. Currie, L., 1968. Limits for qualitative detection and quantitative determination. Anal. Chem. 40, 586. GoÂmez Escobar, V., Vera TomeÂ, F., Lozano, J.C., MartõÂ n SaÂnchez, A., 1996. Determination of 222 Rn and 226 Ra in aqueous samples using a low-level liquid scintillation counter. Appl. Radiat. Isot. 47, 861.
Eective
226
Ra activity concentration (Bq kg ÿ 1)
66372378 4825 6.821.1 Harley, J.H., 1992. Measurements of 222 Rn: A brief history. Radiat. Prot. Dosim. 45, 13. Hinton, T.G., Whicker, F.W., 1985. A ®eld experiment on Rn ¯ux from reclaimed uranium mill tailings. Health Phys. 48, 421. Jacobi, W., 1988. Lung cancer risk from environmental exposure to radon daughters. ICRP Publication 50. Radiat. Prot. Dosim. 24, 19. Kaihola, L., 1994. Cosmic particle spectrum as a quench monitor in low level liquid scintillation spectrometry. Nucl. Instrum. Methods Phys. Res. A 339, 295. Landa, E.R., Nielson, K.K., 1987. Use of track detectors for the evaluation of emanating radium content of soil samples. Uranium 4, 97. Malanca, A., Pessina, V., Dallara, G., 1994. Radon availability from the ground of the Brazilian State of Rio Grande do Norte. Nucl. Geophys. 8, 301. Megumi, K., Mamuro, T., 1972. A method for measuring radon and thoron exhalation from the ground. J. Geophys. Res. 77, 3052. Megumi, K., Mamuro, T., 1974. Emanation and exhalation of radon and thoron gases from soil particles. J. Geophys. Res. 79, 3357. Nakayama, Y., Nagao, H., Mochida, I., Kawabuchi, Y., 1994. Adsorption of radon on active carbon. Carbon 32, 1544. PorstendoÈrfer, J., 1991. Properties and behaviour of radon and thoron and their decay products in the air. In: 5th International Symposium on the Natural Radiation Environment. ISSN 1018-5593 (Report EUR 14411 EN), p. 69. Schery, S.D., Holford, D.J., Wilson, J.L., Phillips, F.M., 1988. The ¯ow and diusion of radon isotopes in fractured porous media. Part 1. Finite slabs. Radiat. Prot. Dosim. 24, 185. Strong, K.P., Levins, D.M., 1982. Eect of moisture content on radon emanation from uranium ore and tailings. Health Phys. 42, 27. Urban, M., Schmitz, J., 1991. Radon and radon daughters metrology. In: 5th International Symposium on the Natural Radiation Environment. ISSN 1018-5593 (Report EUR 14411 EN), p. 151. Wilkening, M., 1985. Radon transport in soil and its relation to indoor radioactivity. Sci. Tot. Environ. 45, 219. Wilkening, M. (1990) Radon in the Environment. Elsevier, Amsterdam.