ARTICLE IN PRESS
Applied Radiation and Isotopes 61 (2004) 129–132
Substrate characterization for underwater gamma spectrometry: tank measurement results utilizing efficiencies calculated via Monte-Carlo code R. Oconea,*, A. Kostezhb, V. Kurinenkob, A. Tyshchenkob, G. Derkachb, P. Leonea a
A.P.A.T. Agenzia per la Protezione dell’Ambiente e per i Servizi Tecnici, Via Vitaliano Brancati, 48, Rome 00144, Italy b UHMI Ukrainian Hydro Meteorological Institute, Nauki av, 37, Kiev 03028, Ukraine
Abstract In order to study the sediment contamination, underwater gamma-ray spectrometry measurements performed by the sub-marine detector Canberra HpGe with a relative efficiency of 80% were carried out in an equipped tank at the Ukrainian Hydrometeorological Institute of Kiev. Different substrates, certified sources and experiment geometrical set-up were arranged. Efficiencies were calculated by in situ object counting system (ISOCS) software. ISOCS performance tests using certified sources were carried out by comparing laboratory measurements, and measurement results for mineralogy and density performances are reported. r 2004 Elsevier Ltd. All rights reserved. Keywords: In situ gamma-ray spectrometry; Efficiency calculation code; Sediment contamination
1. Introduction In sediment contamination studies use of a submarine gamma-ray detector avoids cores sampling, treatment, and subsequent laboratory measurements. In this geometry, the calibration of the efficiency is the main difficulty to overcome. Actually, unlike the traditional laboratory procedures used, for the in situ measurements, it is not possible to use calibration sources for the efficiency versus energy equation construction. However, it is possible to calculate them by using numerical codes. In the present work, the code in situ object counting system (ISOCS) (Canberra, 2002) was used in the calculations. ISOCS is based on the MCNP code which is a general Monte-Carlo radiation transport code. The source-detector-universe geometry is specified via mathematical descriptions of the surfaces and volumes that make up the objects in the ‘‘universe’’. Each emitted gamma ray is tracked as it interacts with *Corresponding author. Fax: +39-06-50072059. E-mail address:
[email protected] (R. Ocone).
the atoms in the materials it traverses, taking into account the double-differential cross sections for photoatomic reactions.
2. Materials, methods, and results Underwater gamma-ray spectrometry measurements, with a submersible detector, were carried out in an equipped tank at the Ukrainian Hydrometeorological Institute of Kiev during July 2002. This facility was composed of: main tank, 1.2 m width and 0.75 m high, containing substrate and water; second tank, 0.45 m width and 1 m high, containing water and the detector; mobile crane to support and move the second tank and allow the immersion operations of the detector in the main tank; the source positioning system composed of a plastic graduated cylinder housed in the main tank. Both the tanks were made of transparent plastic material. Sources containing 137Cs, 226Ra, 152Eu, produced and certified by the Laboratory of Radioecology (2000), of spherical shape with a diameter less than 2 mm and
0969-8043/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2004.03.033
ARTICLE IN PRESS R. Ocone et al. / Applied Radiation and Isotopes 61 (2004) 129–132
130
located at centre of a stainless-steel cylinder were utilized. Dimensions of the cylinders were 3 mm diameter, 10 mm length, and 0.2 mm wall thickness. The main tank and the positioning system were both filled with the same homogenized substrate and covered by water. Three different substrates were used: sand, silt and clay, gravel. Major element concentrations were determined by X-ray fluorescence with a multi-channel spectrometer SRM-25. These results, with the grain size distribution and the dry and wet densities are reported in Table 1. Natural and artificial radionuclides contents, reported in Table 2, were determined by low-background gamma-ray spectrometry system ORTEC with uncertainties in the range of 15%–20%. A Canberra HpGe detector, with an efficiency of 77.2% relative to that of a 3 30 NaI, a Peak/Compton ratio of 75.8:1 and with a resolution of 1.84 keV at 1.33 MeV was used for these measurements. The detector and the circular collimator were equipped with an external protective housing for underwater measurement applications and for avoiding external contamination. Canberra Genie2000 was the gamma-ray spectrometry software used to analyse the measured spectra. The detector properties at APAT before shipping the detector to UHMI and after arrival were checked by measuring 241Am and 137Cs counting efficiencies. The efficiencies measured in APAT were (3.5270.18) 103 and (4.2970.22) 103 for 241Am and 137Cs, respectively. The same efficiencies measured at UHMI were (3.2270.16) 103 and (3.9570.20) 103. It can be
observed that the efficiencies changed only for few per cent after shipping the detector to UHMI laboratory. ISOCS functioning performance was verified by measuring a sample of known activity both using an ORTEC GWL 100210 gamma-ray spectrometer calibrated by using certified reference materials and by using the sub-marine detector with ISOCS procedure. For this purpose, a sample of dry sand with a known activity of 137Cs was prepared in a cylindrical geometry with 140 mm diameter, 50 mm height and with a density of 1.66 g/cm3. Activity values of 380719 and 382719 Bq were measured with the ORTEC system and with the sub-marine detector, respectively. The sources produced by the Laboratory of Radioecology were measured by the sub-marine detector with ISOCS software. Three measurement sets were performed to check each source that was positioned on the detector axis at 2071 cm distance for each lower activity source and at 5071 cm distance for each higher activity one. Mean values of these measurements are reported in Table 3. Preliminary measurements at different distance from the axis and at different depths had been performed to measure the significant attenuation of gamma rays in the sediment. Finally, several measurement sets for each particular substrate covered with water were carried out, using the tested certified sources, with the purpose of verify ISOCS performance with respect to substrate mineralogy and density. The detector was perpendicularly dipped in contact with the substrate and the sources
Table 1 Grain size, density, and major elemental compounds content (%) in substrates Substrate
Gravel Sand Silt and clay
Grain size (mm)
2–10 o2 0.002C0.05
SiO2
2.91 81.30 56.70
TiO2
Al2O3
0.07 0.03 0.29
0.62 0.08 3.58
Fe2O3
0.67 0.00 0.79
MnO
0.03 0.01 0.02
MgO
0.38 0.01 0.51
CaO
76.9 0.00 1.92
Na2O
0.21 0.00 0.42
K2O
H2O
0.23 0.00 0.77
17.80 18.00 35.00
Density (g/cm3) Dry
Wet
1.87 1.75 1.33
2.27 2.13 1.78
Table 2 Natural and artificial radionuclides content in the substrates Substrate
Radionuclide concentrations (Bq/kg) 235
40
K
137
214
226
234m
0.6a
153.0 39.6 542.0
0.9 0.3
11.9 8.2 29.7
12.1 8.1 29.7
16.6a
U
Gravel Sand Silt and clay a b
Estimated. Not detected.
b
0.5a
Cs
b
Pb
Ra
Pa
b
33.0a
228
212
6.9 1.0 28.5
7.6 1.1 33.2
Ac
Pb
212
Bi
7.5 b
31.6
ARTICLE IN PRESS R. Ocone et al. / Applied Radiation and Isotopes 61 (2004) 129–132
were placed on the detector axis at 571 cm distance for each lower activity source and at 1571 cm distance for each higher activity one. Measurement results are reported in Tables 4–6.
Table 3 Activity values measured at distances of 20 and 50 cm without substrate Certified activity (kBq) 73%
Measured Uncertainty activity (kBq) (kBq)
226
34 145
38.2 181.7
2.2 10.5
137
38 164
40.8 169.6
1.6 5.9
36 181
34.8 188.9
1.3 6.6
Ra
Cs
152
Eu
Table 4 Activity values measured at distances of 5 and 15 cm with silt and clay as substrate Certified Measured Uncertainty Measured/ activity activity (kBq) (kBq) certified (kBq) 73% 226
Ra
137
Cs
152
Eu
34 145
33.8 153.1
1.7 7.7
1.00 1.06
38 164
36.5 154.5
1.8 7.8
0.96 0.94
36 181
34.8 182.4
1.7 9.1
0.97 1.01
Table 5 Activity values measured at distances of 5 and 15 cm with gravel as substrate Certified Measured Uncertainty Measured/ activity activity (kBq) (kBq) certified (kBq) 73% 226
34 145
27.5 185.9
1.4 9.3
0.81 1.28
137
38 164
34.0 193.2
1.7 9.7
0.89 1.18
152
36 181
30.5 193.3
1.5 9.7
0.85 1.07
Ra
Cs
Eu
131
Table 6 Activity values measured at distances of 5 and 15 cm with sand as substrate Certified Measured Uncertainty Measured/ activity activity (kBq) (kBq) certified (kBq) 73% 226
34 145
33.6 159.1
1.7 8.0
0.99 1.10
137
38 164
37.6 164.3
1.9 8.2
0.99 1.00
152
36 181
31.1 173.9
1.6 8.7
0.86 0.96
Ra
Cs
Eu
After background subtraction 226Ra activity was calculated at energies of 295.21, 351.92, 609.32, 1120.28, and 1764.51 keV, and the measurement results reported in Tables 3–6 are given in the form of combined activity. In measurements with 152Eu coincidence summing corrections were not necessary to perform because of the separation between the source and the detector. No significant discrepancy was observed among the activity values calculated from the emission lines from 121 to 1408 keV. Hence, the 152Eu measurement results reported in Tables 3–6 are given in the form of combined activity of the most abundant photon emissions.
3. Discussion Table 3 shows that the agreement between certified and measured activities is in general good. A major disagreement for 226Ra activities can be associated with the high background variability present in the laboratory. The ISOCS functioning performance measurements show a really good agreement allowing to state that the whole ‘‘in situ object counting system’’ can be successfully used when samples cannot be moved and analysed with fixed detectors in the laboratory and/or where empirical calibrations are not available. In measurements with the attenuation medium (Tables 4–6), since the detector was dipped in the water, measurements were not affected from laboratory background variability, so agreement in 226Ra case is improved. It can be observed from Tables 4–6 that the discrepancies between measured and certified activities increase with the density and grain size. It is concluded therefore that they originate in the influence of nonhomogeneity of the absorber on the measured activities. The efficiency calculations using the ISOCS software
ARTICLE IN PRESS 132
R. Ocone et al. / Applied Radiation and Isotopes 61 (2004) 129–132
were namely performed assuming homogeneous absorbing medium, using the macroscopic absorber density.
Acknowledgements The authors thank Dr. O. Voitsekhovitch, Dr. V. Kanivets, Mr. M. Blasi, Mr. A. Marchetti.
References Canberra, 2002. ISOCS Calibration Software for Genie-2000 Model S573, 9231013D. Laboratory of Radioecology of The Ministry of an Ecology and Natural Resources of Ukraina, Ukranian Scientific Centre of the Ecology of Sea, Odessa (Ukraina), 2000. Package of sources SGR-1/99. Passport No. 12-123 (manufacturing date: 08.02.2000).
