- Email: [email protected]
Determination of activity concentration of 238U and 232Th series radionuclides in soil using a gamma-ray spectrometer in singles and coincidence modes
Applied Radiation and Isotopes 154 (2019) 108880
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Determination of activity concentration of 238U and 232Th series radionuclides in soil using a gamma-ray spectrometer in singles and coincidence modes
T
M. Bashira,b,c,∗, R.T. Newmanb, P. Jonesc a
Department of Physics, Ibrahim Badamasi Babangida University Lapai, Nigeria Department of Physics, Stellenbosch University, South Africa c Department of Subatomic Physics, iThemba Laboratory for Accelerator Based Sciences (iThemba LABS), South Africa b
H I GH L IG H T S
minimum detectable activity (MDAs) in singles mode were higher by a factor of 8–36 than for coincidence mode measurements. • The The activity concentration of U and Th series radionuclides were determined using a gamma-gamma coincidence method despite us not having any • traditional background shielding. • The activity concentration of U and Th series radionuclides in the IAEA-375 reference soil material were below the MDA for singles mode measurements. 238
238
232
232
A R T I C LE I N FO
A B S T R A C T
Keywords: Singles Gamma-gamma coincidence Activity 238 U and232Th series
Traditionally, activity concentrations of naturally occurring radioactive materials (NORMs) are measured using a gamma-ray spectrometer with a single detector shielded with lead. In this measurement, a gamma-ray spectrometer comprising an array of four LaBr3:Ce detectors without shielding was used. This spectrometer allowed for measurement in singles and coincidence (gamma-gamma) modes. In addition to using the coincidence method, a novel method of background reduction by using the photon time-of-flight was utilized. Activity concentration of 238U and 232Th series radionuclides in soil reference (IAEA-375) and beach-sand (Bs) were measured in singles and coincidence modes. In coincidence mode, the minimum detectable activities (MDAs) were lower by a factor of 8–36 than for singles mode. Activity concentration of 238U and 232Th series in soil material (IAEA-375) were determined in coincidence mode while it was below MDA in singles mode. The results correlate with the certified values to within the uncertainty although the uncertainties were high because of low statistics.
1. Introduction Naturally occurring radioactive materials (NORMs) are present in every natural substance to a certain degree (James, 2006). These NORMs are often measured using gamma-ray spectrometry with a single detector (NaI:Tl or HPGe) shielded with lead (Elisabeth et al., 2009; Gilmore, 2008; Kohler et al., 2009; Kozak et al., 2001; Lutter et al., 2009; Sykora et al., 2008). To absorb the X-rays generated from the lead shielding, the inner part of the lead shielding is lined with one or combination of the following materials; copper (Cu), cadmium (Cd), tin (Sn), perspex. Sometimes the spectrometer is situated in an underground laboratory for cosmic radiation suppression (Elisabeth et al.,
∗
2009; Lutter et al., 2009; Kohler et al., 2009; Gilmore, 2008). Moreover, a study conducted by Keillor et al. (2017) shows that commercial lead in the U.S. other than stockpiled Doe Run lead contain roughly an order of magnitude higher 210Pb levels. This could be the reason for shift from the use of 10 cm thickness of low-level radioactivity lead to 10.0–14.5 cm thick ordinary lead combined with 3.5–5.0 cm low-level lead (Sykora et al., 2008; Elisabeth et al., 2009; Lutter et al., 2009; Kohler et al., 2009; Kozak et al., 2001). Lead is also expensive and requires the use of other materials in combination with it, thereby increasing the cost of shielding and size. According to Metwally et al. (2005) and Drescher et al. (2017), the use of the gamma-gamma coincidence method has the advantage of
Corresponding author. Department of Physics, Ibrahim Badamasi Babangida University Lapai, Nigeria. E-mail addresses: [email protected], [email protected], [email protected] (M. Bashir).
https://doi.org/10.1016/j.apradiso.2019.108880 Received 12 March 2019; Received in revised form 24 August 2019; Accepted 25 August 2019 Available online 28 August 2019 0969-8043/ © 2019 Elsevier Ltd. All rights reserved.
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
minimizing spectrum background thereby reducing the minimum detectable activity (MDA). Britton et al. (2015a) used a γ − γ coincidence system comprising two large-area, planar HPGe detectors enclosed in a Pb cave to measure calibration sources. Correction for time-walk associated with low energy events and removal of accidental coincidences were accomplished, though activity concentration was not calculated. Britton et al. (2015b) developed a method for quantifying gamma coincidence signatures. A software package which uses ENSDF records as input and calculate the efficiency as well as cascade summing corrected branching ratios was created. This software was tested by measuring 54 Mn, 57,60Co, 65Zn, 88Y, 109Cd,113Sn, 137Cs, 139Ce and 241Am using gamma-gamma system composed of two planar HPGe detectors shielded with lead. The gamma-gamma system produces singles spectra and coincidence matrices using digital electronics and list-mode acquisition. The results for the investigated nuclides were found to be accurate. Antovic and Svrkota (2009a,b) measured activity of 226Ra and 232 Th daughters with a 4π gamma-ray coincidence spectrometer called PRIPYAT-2M. PRIPYAT-2M consists of six-crystal NaI:Tl each with a diameter of 15 cm and height of 10 cm. It has outer dimensions of 250 cm × 154 cm × 186 cm with iron and lead shielding. The PRIP software used enables pulse counting in integral mode and coincidence mode (double, triple, four-fold, five-fold and six-fold coincidences) using buffer memory (FIFO). However, the overlapping gamma-ray peaks centred at 1120.3 and 1238.1 keV were not resolved because of the detector's energy resolution. Markovi et al. (2017) reported on the performance of the Nutech Coincidence Low Energy Germanium Sandwich (NUCLeGeS) system for environmental analysis (210Pb and 134 Cs). The NUCLeGeS system consists of two HPGe detectors enclosed in a 10 cm thick lead shield. A digital acquisition system with 10 ns time resolution was used to collect time-stamped list mode data in anti-coincidence and coincidence modes. The results show that the spectrometer is competitive with single detector systems used in the laboratory for 210Pb analysis. Paradis et al. (2017) measured the radioactivity in environmental samples using a gamma-gamma coincidence spectrometer called Leda in singles, coincidence or anti-coincidence modes. Leda is made of 2 HPGe and 1 NaI:Tl detectors. The data acquired were time-stamped. The detection limits for all γ-emitters were improved. Below we report on an investigation into the use of a modern array of LaBr3:Ce scintillator detectors to measure activity concentrations of primordial radionuclides in soil and sand samples. Here, the measurement of gamma-gamma coincidence along time information is shown. This allows for achievement of lower minimum detectable activity relative to measurements for some radionuclides in singles mode. The setup is relatively simple requiring only the detectors and the associated electronics. Moreover, the time resolution of a LaBr3:Ce based detector is typically a few hundred picoseconds (Kumar Anil et al., 2009). It has high scintillation light output (Favalli et al., 2008), a typical energy resolution of 3% at 662 keV (Dorenbos et al., 2004) and high intrinsic gamma-ray detection efficiency (Bashir et al., 2018; Vedia et al., 2017; Regan et al., 2016). Furthermore, there is no need for operating the detectors at liquid nitrogen temperature. 238 U and 232Th are unstable and undergo a series of decays. 238U have thirteen progenies (234Th*, 234Pa*, 234U, 230Th, 226Ra*, 222Rn, 218 Po, 214Pb*, 214Bi*, 214Po, 210Pb*, 210Bi, 210Po) before stable 206Pb is reached (Gilmore, 2008; Martin, 2007). 232Th have nine progenies (228Ra, 228Ac*, 228Th, 224Ra*, 220Rn, 216Po, 212Pb*, 212Bi*, 212Po, 208 Tl*) before stable 208Pb is reached (Gilmore, 2008; Martin, 2007). According to Gilmore (2008), only radionuclides marked with * can easily be measured by gamma-ray spectrometry. An inspection of the decay scheme of 238U and 232Th (and their daughters) showed that only gamma-rays from β − decay of 214Bi and 208Tl have energies and intensities suitable for gamma-gamma coincidence measurements (with the experimental geometry used for this study). The most intense gamma-rays with their intensities in brackets from β − decay of 214Bi to 214Po are: 609.3 keV (45.5%), 1120.3 keV (14.9%),
Fig. 1. Picture of different experimental geometries investigated using an array of LaBr3:Ce gamma-ray detectors.
1238.1 keV (5.8%), 768.4 keV (4.9%), 2204.1 keV (4.9%) 1377.7 keV (4.0%), 934.1 keV (3.1%) (Wua, 2009). And the most intense gammarays from β − decay of 208Tl to 208Pb are: 2614.5 keV (99.8%), 583.2 keV (85.0%), 510.7 keV (22.6%), 860.6 keV (12.5%), 277.4 keV (6.6%) (Martin, 2007). For this study, coincidences between 609.3 keV and 1120.3 keV were chosen for the 238U series measurement, while 583.2 keV and 2614.5 keV were chosen for the 232Th series measurements because of their intensities. These chosen gamma-rays energies are also in coincidence with other cascade gamma-ray energies not of interest in this work (Martin, 2007; Wua, 2009).
2. Experimental methods At iThemba LABS, eight cylindrical LaBr3:Ce scintillator detectors are available for gamma-ray measurements. These detectors were manufactured by Saint-Gobain and have a crystal size of 5.1 cm × 5.1 cm and R2083 photomultiplier tubes (PMTs) attached to them. Different experimental geometries were investigated using eight (equidistant of 24.0 cm), four (equidistant of 15.0 cm) and two (equidistant of 10.0 cm) detectors (Fig. 1). It was found that the eight detector experimental geometry has the most scattering events (Fig. 2) because of the distance between the detectors. However, the two detector experimental geometry has the least scattering events and its detection efficiency is low. This is as a result of the number of detectors used. The four detector experimental geometry has fewer scattering events than the eight detector experimental geometry. Also, its detection efficiency is better than that for the two detector experimental geometry. Its efficiency can be improved by reducing the source-todetector distance. Furthermore, the time-of-flight of the source gamma-ray can be discerned from that of the scattered gamma-ray (Fig. 2). For the eight detector experimental geometry, the Compton scattered gamma-ray time-of-flight are ≈ 700 ps, ≈ 1400 ps, ≈ 1700 ps and ≈ 1900 ps at angle 45∘, 90∘, 135∘ and 180∘ respectively (Fig. 2). Again, the minimum timeof-flight can be estimated from Fig. 3 using the speed of light and the source-to-detector distance (d is the distance between the sample and detector front surfaces plus the beaker radius) and the distance between detectors. This enables the time-of-flight to be used and the background/scattered gamma-ray events can be significantly reduced by setting a time gate smaller than the scattered gamma-ray time-of-flight. For this study, an experimental geometry involving now only four detectors is used with samples placed 16.5 cm equidistant from the detectors. Each detector signal was linked to a XIA Pixie-16 module (XIA Pixie-16, 2018). The XIA Pixie-16 module is a 16 channel digital signal processing module, providing readout of timestamp, CFD time 2
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 2. Variation in time difference between different detector combination (at 45°, 90°, 135° and 180°) for the eight detector experimental geometry.
measure the background and container contribution. Soil reference (IAEA-375) and beach-sand (Bs) from Zanzibar were measured to obtain their activity concentration. 137Cs and 60Co point sources were used for energy calibrations. Further information about the samples and measurement time are given in Table 1. 3. Data analyses Offline, the list-mode data acquired (which gives the advantage of enabling the re-processing the raw data) were sorted into channels, time and energy. The sorted data were binned while energy and time spectra were plotted, using ROOT software (Brun and Rademakers, 1997).
Fig. 3. Picture illustrating the distance (d) between sample and detector front surfaces plus the Marinelli beaker radius; and between the detectors (L1 to L8).
and calculated energy directly from the detector output. These data consist of event ID, time stamp (48-bit), CFD time (16-bit), calculated energy (16-bit) at a 14 bit sampling resolution at 500 MHz (XIA Pixie16, 2018). In addition, a field-programmable gate array (FPGA) first captures 5 ADC samples at the rate of 100 MHz (i.e every 10 ns). Full CFD timing data are calculated from the differences between the delayed and non-delayed sums, providing the zero-crossing point (XIA Pixie-16, 2018). MIDAS software code (MIDAS) ran on a Linux workstation for data collection. These software access all the data readout from the hardware electronics through a single Peripheral Component Interconnect (PCI) bridge (PXI-8360). The time resolution of the detector is ≈ 370 ps FWHM which allow the electronics to resolve time difference at the sub-nanosecond level. Finally, each sample was placed in a Marinelli beaker (1.0 L) having dimensions of 13.0 cm diameter, 8.5 cm well-diameter, 13.5 cm height, 6.1 cm well-height and 0.2 cm thickness. Uranium ore (RGU-1) and thorium ore (RGTh-1) from IAEA were measured to facilitate the detection efficiency calibration. An empty Marinelli beaker was used to
3.1. Singles First, the energy spectra for the four detectors were gain matched and summed for each measured sample. The empty beaker spectrum (Fig. 4 upper) was normalized to the sample counting time and subtracted from the sample's spectra. The IAEA-375 soil and beach-sand samples spectra before and after the empty beaker spectrum subtraction Table 1 Reference material and sample information.
3
Sample/Source
Density (kgm−3)
Live time (seconds)
Uranium Ore (RGU-1) Thorium Ore (RGTh-1) Soil reference (IAEA-375) Beach sand (Bs) Empty beaker 137 Cs and 60Co
1409.12 1364.94 1502.63 2658.30 – –
172796 172764 172787 172983 57594 7200
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 4. Normalized summed total empty beaker spectra in singles mode (upper) and the summed total projection of coincidence events spectra for empty beaker without energy gates in coincidence mode (lower).
Fig. 5. Summed total spectra before and after empty subtraction for the IAEA-375 soil sample, and the normalized empty beaker.
Fig. 6. Summed total spectra before and after empty beaker subtraction for the beach-sand sample, and the normalized empty beaker.
4
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 7. Time difference spectra for six of the two combinations of four detectors (A) (the blue and purple colours are for detectors at 180° while red, green, black and yellow colours are for detectors at 90°), beach-sand (sample) and empty beaker time difference spectra (B), and the time gate tc (centroid peak) used to generate gamma-gamma coincidence spectra (C). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 8. Summed total gamma-gamma spectra for the beach-sand sample generated using peaks on either side of Fig. 7 time gates to show background/scattered events.
Fig. 9. Summed total gamma-gamma spectra for the IAEA-375 soil reference.
5
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 10. Summed total gamma-gamma spectra for the beach-sand sample.
Fig. 11. Summed total projection for the IAEA-375 soil reference.
Fig. 12. Summed total projection for the beach-sand sample.
are shown in Figs. 5 and 6, respectively. It can be observed from these figures that 40 K in the empty beaker is higher than in the samples. This could be due to the sample acting as a shield between the detector and
the natural background. Second, the peaks area counts (Cn) were extracted using the integral function in ROOT. Radware Gf3 was used to resolve and fit the 6
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 13. Energy gated spectra: 609.3 keV and 1120.3 keV (214Bi) for the IAEA-375 soil reference.
Fig. 14. Energy gated spectra: 583.2 keV and 2614.5 keV (208Tl) for the IAEA-375 soil reference.
Eqn. (1).
overlapping 583.2 keV and 609.3 keV peaks (Radford, 2000). Absolute full-energy peak gamma-ray detection efficiency (ε) at various peaks of interest were calculated using Eqn. (1) (Gilmore, 2008)
ε=
Cn APγ Ms Lt
3.2. Coincidence The time difference between gamma-gamma coincidence events associated with any two LaBr3:Ce detectors within the time window of 10000 ps were plotted using the time-stamped data and shown in Fig. 7 (A). In Fig. 7(C), the region tc is the true coincidence peak which contain the coincident events of interest, and the peaks on either side are random coincidence between the detectors due to background radiation from terrestrial and cosmic radiation, and/or Compton scattering as there was no shielding used (Fig. 7(B)). Two different time gates, the total time tT (−5000 ps–5000 ps) and the time gate tc (Fig. 7(C)) were used. These time gates were set on the empty beaker, IAEA-375 soil and beach-sand data to generate gammagamma spectra for any two detectors in coincidence. It is important to note that the spectra shown are all for time gate tc and the total time tT was computed to check the sensitivity of time gate tc. Although, the
(1) −1
where A is the source activity concentration in Bqkg , Pγ is the probability of emission of the gamma-ray being measured, Ms is sample mass in kg and Lt is the spectrometer data acquisition live time in seconds. The minimum detectable activity (MDA) was calculated using Eqn. (2) (Gilmore, 2008), where σB is the uncertainty in the background (empty beaker) peak count:
MDA =
2.71 + 3.29σB εPγ Ms Lt
(2) 238
232
The activity concentration of U and Th series radionuclides in IAEA-375 soil and beach-sand samples were calculated by re-arranging 7
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 15. Energy gated spectra: 609.3 keV and 1120.3 keV (214Bi) for the beach-sand sample.
Fig. 16. Energy gated spectra; Energy gated spectra: 583.2 keV and 2614.5 keV (208Tl) for beach-sand sample. Table 2 Weighted average counts in the coincidence peaks. Sample
Series
Nuclide
Counts (tc)
Counts (tT)
238
214
232
208
4.5 ± 2.1 3.0 ± 1.7
5.5 ± 2.3 3.3 ± 1.8
238
214
232
208
327.3 ± 18.1 402.5 ± 20.1
331.3 ± 18.2 424.3 ± 20.6
Table 3 Total counts in the gamma-gamma matrix generated using time gates tc and tT for all the coincidence detectors combination in IAEA-375 soil and beach-sand after empty beaker subtraction.
IAEA-375 U Th
Bi Tl
Detector Combination
Bs U Th
Bi Tl
Angle
L1-L2 L1-L3 L1-L4 L2-L3 L2-L3 L3-L4
peaks on either side of time gates (Fig. 7) were set on the beach-sand data to generate the gamma-gamma matrix in Fig. 8 to show the background/scattered events contribution, the matrix was not used in calculating activity concentration. The spectra were summed for each sample and the normalized empty beaker spectrum was subtracted from that of the IAEA-375 soil and beach-sand. Figs. 9 and 10 shows the total gamma-gamma coincident spectra. The two-dimensional gamma-
180° 90° 90° 90° 90° 180°
IAEA
Bs
Counts (tc)
Count (tT)
Counts (tc)
Counts (tT)
1338 2477 4849 3937 2826 995
30416 118671 110530 115331 108386 29635
19949 29816 23038 27561 23559 21198
56935 152482 143332 148511 141298 57045
gamma matrix corresponds to γ-rays in the decay cascade being detected in coincidence (coincidence pairs at [609.3, 1120.3] and [583.2, 2614.5]) and random coincidence. 8
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Table 4 Absolute full-energy peak detection efficiencies in singles and coincidence modes. Series
Nuclide
Energy
Efficiencies
Efficiencies
[keV]
Singles ( × 10−3)
Coincidence ( × 10−6)
609.3 1120.3 1764.5
5.0 ± 0.1 3.0 ± 0.1 2.0 ± 0.1
14.9 ± 1.0 14.9 ± 1.0 –
583.2 911.2 2614.5
5.0 ± 0.2 1.9 ± 0.1 1.3 ± 0.1
6.6 ± 0.3 – 6.6 ± 0.3
238
U 214
Bi Bi Bi
214 214 232
Th 208
Tl Ac 208 Tl 228
Fig. 18. Activity concentration of 238U series in singles and coincidence modes for the beach-sand sample measured with LaBr3:Ce and HPGe.
Table 5 Minimum Detectable Activities (MDAs) in singles and coincidence modes. Series
Nuclide
Energy
MDAs
MDAs
HPGe
[keV]
Singles
Coincidence
MDAs
Bi Bi 214 Bi
609.3 1120.3 1764.5
95.6 422.0 252.2
11.6 11.6 –
1.1 3.0 2.9
208
583.2 911.2 2614.5
127.2 385.7 166.5
15.9 – 15.9
1.9 1.6 1.0
238
U 214 214
232
Th Tl Ac Tl
228 208
Several coincidence conditions were set on the gamma-gamma coincident spectra to measure the full-energy peak in one of the two detectors. The summed total projection of coincidence events measured with any two detectors are shown in Figs. 11 and 12. These were used to get full-energy peak width of the energies of interest used for the energy gates. As shown in Figs. 13–16 when a gate is set on 609.3 keV, the 1120.3 keV 25+ → 21+ transition is seen and the 609.3 keV 21+ → 01+ transition is absent (from the decay of 214Bi to 214Po) and vice versa. Similarly, from the decay of 208Tl to 208Pb, when a gate is set on 583.2 keV, the 2614.5 keV 31−− → 01+ transition is seen and the 583.2 keV 51− → 31− transition is absent, and vice versa. As observed from Figs. 15 and 16 when a gate is set on either 583.2 keV or 609.3 keV peaks at both 1120.3 keV and 2614.5 keV were seen. This is because of the overlap of 583.2 keV and 609.3 keV full-energy peaks.
Fig. 17. Measured activity concentration of
238
U and
232
Fig. 19. Activity concentration of 232Th series in singles and coincidence modes for the beach-sand sample measured with LaBr3:Ce and HPGe.
The weighted average of the coincidence peaks count using time gate tT and tc are given in Table 2. The detection efficiency for two gamma-ray energies measured in coincidence is the product of the efficiency of the two gamma-ray energies measured in singles mode (Siegbahn, 1965) i.e, ε γc = ε γ1 ε γ2 . The true coincidence summing (TCS) for this experimental geometry is negligible (0.6% solid angle and 0.006% chance of summing for our distance) because the TCS decrease with increase in source-to-detector
Th series in coincidence mode and expected value for the IAEA-375 soil reference. 9
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
samples. Coincidence between detectors at 90° have more scattered events than those at 180° i.e, the larger the angle between detectors in coincidence the lower the number of scattering events. However, using the photon time-of-flight in addition to the coincidence method, the scattered events were significantly reduced. Also, the background was reduced by two orders of magnitude compared to the singles method. Furthermore, low-level activity concentration ≥ 11.6 Bqkg−1 and ≥ 15.9 Bqkg−1 for 238U and 232Th series radionuclides can be measured in coincidence mode. Using this spectrometer, activity concentration of radionuclides having an associated gamma-decay cascade can be measured with any sample geometry in coincidence mode.
distance (Gilmore, 2008). Moreover, both Gaussian and Poisson analysis were performed to estimate the coincidence MDA but due to the level of statistics Poisson distribution was used. The peaks counts in the summed total projection of coincidence event spectra for empty beaker in Fig. 4 (lower) extracted and Eqn. (3) (Gilmore, 2008) used to calculate coincidence MDA:
MDA =
LD ε γc Pγ1 Pγ 2 Ms Lt
(3)
where LD is the detection limit and the value taken from Gilmore (2008) for Poisson distribution (for 95% confidence) associated with background count within the counting time, and Pγ1 and Pγ2 are gamma-ray emission probabilities of gamma energy 1 and 2 respectively. The activity concentration in coincidence was calculated using Eqn. (4) suggested by Siegbahn (1965), where Nc is the coincidence peak area counts, n is the number of detectors used (in this case n = 4) and m is the number of combinations for four detectors taken two detectors at a time (in this case m = 6)
Ac =
2nNc (mε γc Pγ1 Pγ 2 Ms Lt )
Acknowledgements This authors would like to thank IBB University/TETFUND for the bursary, National Research Foundation of South Africa (99037) for providing the equipment and iThemba LABS (Laboratory for Accelerator Based Science) for the top-up funding. Also like to thank Mr. Lumkile Msebi for his support during the experiment. Appendix A. Supplementary data
(4) Supplementary data to this article can be found online at https:// doi.org/10.1016/j.apradiso.2019.108880.
4. Results and discussion References The uncertainties quoted in all the results are at the 1σ level. As observed from Table 3, coincidence between detectors at 90° have more scattered events than those at 180° when total time gate tT was used to generate the gamma-gamma matrix. But, when the time gate tc was used to generate the gamma-gamma matrix the scattered events were significantly reduced. Table 4 gives the absolute full-energy peak detection efficiencies of 238 U and 232Th series radionuclides at various gamma-ray energies of interest using singles and coincidence methods. The detection efficiencies for coincidence mode are far lower than that for singles modes. The 238U and 232Th series radionuclides MDAs in coincidence mode are all < 16 Bqkg−1 while that of singles are > 100 Bqkg−1 except for 609.3 keV which is 96 Bqkg−1 (Table 5). However, the MDAs measured with HPGe is lower than that of LaBr3:Ce in coincidence mode. In coincidence mode, our MDA for 609.3 keV was lower than that of Antovic and Svrkota (2009a,b) and higher for 583.2 and 2614.5 keV. Activity concentration of 238U and 232Th series radionuclides in IAEA-375 from the singles mode measurement was below the MDA for this experimental geometry which agrees with results from Bashir et al. (2018). Measured activity concentration of radionuclides inside the IAEA-375 soil reference in coincidence mode are consistent with certified values to within measurement uncertainties (Fig. 17). The measured activity concentrations are within the uncertainty although the uncertainties were high because of low statistics. Figs. 18 and 19 shows that the activity concentration of 238U and 232Th series in beach-sand measured in singles and coincidence modes were consistent to within measurement uncertainty. This sample was also measured using a hyper-pure germanium (HPGe) detector shielded with lead and the activity concentration agreed with our results (Figs. 18 and 19). According to UNSCEAR (2000), the world average activity concentration of 238U and 232Th are 35 Bqkg−1 and 30 Bqkg−1, respectively. Therefore, using this experimental geometry in coincidence mode low-level activity concentration of 238U ( ≥ 11.6 Bqkg−1) and 232 Th ( ≥ 15.9 Bqkg−1) series radionuclides can be measured.
Antovic, N., Svrkota, N., 2009a. Development of a method for activity measurements of 232 Th daughters with a multidetector gamma-ray coincidence spectrometer. Appl. Radiat. Isot. 67, 1133–1138. Antovic, N., Svrkota, N., 2009b. Measuring the Ra-226 activity using a multidetector γray coincidence spectrometer. J. Environ. Radioact. 100, 823–830. Bashir, M., Newman, R.T., Jones, P., 2018. Measurements of natural radioactivity in soil using an array of cerium doped lanthanum bromide scintillator detectors. In: The Proceedings of SAIP2017, the 62nd Annual Conference of South African Institute of Physics. Division B- Nuclear, Particle and Radiation Physics, vols. 131–134978-0620-82077-6, . Britton, R., Burnett, J.L., Davies, A.V., Regan, P.H., 2015a. Coincidence corrections for a multi-detector gamma spectrometer. Nucl. Instrum. Methods Phys. Res. 769, 2025. Britton, R., Jackson, M.J., Davies, A.V., 2015b. Quantifying radionuclide signatures from a γ−γ coincidence system. J. Environ. Radioact. 149, 158163. Brun, Rene, Rademakers, Fons, 1997. ROOT-an object oriented data analysis framework. In: Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, vol 389. pp. 81–86 1. Dorenbos, P., et al., 2004. Gamma-ray spectroscopy with a ⊘19 ×19 mm 3LaBr:3 0.5 % Ce 3+scinillator. IEEE Trans. Nucl. Sci. 51 (3), 1289–1296. Drescher, A., et al., 2017. Gamma-gamma coincidence performance of LaBr3:Ce scintillation detectors vs HPGe detectors in high count-rate scenarios. Appl. Radiat. Isot. 122, 116–120. Elisabeth Wieslander, J.S., Hult, Mikael, Gasparro, Joel, Marissens, Gerd, Misiaszek, Marcin, Werner, Preusse, 2009. The Sandwich spectrometer for ultra low-level γ-ray spectrometry. Appl. Radiat. Isot. 67, 731–735. Favalli, A., Mehner, H.C., Simonelli, F., 2008. Wide energy range efficiency calibration for a lanthanum bromide scintillation detector. Radiat. Meas. 43, 506–509. Gilmore, G., 2008. Practical Gamma-Ray Spectrometry, second ed. John Wiley & Sons, Ltd., West Sussex, England978-0-470-86196-7. James, Martin E., 2006. Physics for Radiation Protection: A Handbook, second ed. WILEYVCH Verlag GmbH and Co. KGaA, Weinheim. Keillor, Martin E., Aalseth, C.E., Arnquist, I.J., Eggemeyer, T.A., Fuller, E.S., Glasgow, B.D., Hoppe, E.W., Morle, S.M., Myers, A.W., Orrell, J.L., Overman, C.T., Seifert, A., Shaff, S.M., Thommasson, K.S., 2017. Recent Bremsstrahlung-based assays of Pb-210 in lead and comments on current availability of low-background lead in North America. Appl. Radiat. Isot. 126, 185–187. Kohler, M., Degering, D., Laubenstein, M., Quirin, P., Lampert, M.-O., Hult, M., Arnold, D., Neumaier, S., Reyss, J.-L., 2009. A new low-level γ-ray spectrometry system for environmental radioactivity at the underground laboratory Felsenkeller. Appl. Radiat. Isot. 67, 736–740. Kozak, Kryzysztof, Mietelski, Jerzy W., Jasifska, Miroslawa, Gaca, Pawel, 2001. Decreasing of the natural background counting - passive and active method. In: NUKLEONIKA PROCEEDINGS, vol. 46. pp. 165–169 4. Kumar Anil, G., Mazumdar, I., Gothe, D.A., 2009. Experimental measurements and GEANT4 simulations for a comparative study of efficiencies of LaBr3:Ce, NaI:Tl, and BaF2. Nucl. Instrum. Methods Phys. Res. 610, 522529. Lutter, Guillaume, Hult, Mikael, Marissens, Gerd, Andreotti, Erica, Rosengard, Ulf, Misiaszek, Marcin, Yuksel, Ayhan, Sahin, Namik, 2009. A new versatile underground gamma-ray spectrometry system. Nucl. Instrum. Methods Phys. Res. 610, 522–529. Markovi, Nikola, Roos, Per, Nielsen, Sven Poul, 2017. Digital gamma-gamma coincidence
5. Conclusion This work presents for the first time the use of coincidence techniques with LaBr3:Ce detectors without shielding to measure the activity concentration of 238U and 232Th series radionuclides in volume soil 10
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Siegbahn, Kai, 1965. Alpha-, Beta- and Gamma-Ray Spectroscopy, vol. 1 North-Holland Publishing Company Amsterdam. Sykora, I., Jeskovsky, M., Janik, R., Holy, K., Chudy, M., Provinec, P.P., 2008. Low-level single and coincidence gamma-ray spectrometry. J. Radioanal. Nucl. Chem. 276, 779–787. United Nation Scientific Committee on the Effects of Atomic Radiation, 2000. Sources and effects of ionizing radiation. In: Report to the General Assembly, with Scientific Annexes, vol. 1. Vedia, V., Carmona-Gallardo, M., Fraile, L.M., Mach, H., Udas, J.M., 2017. Performance evaluation of novel LaBr3(Ce) scintillator geometries for fast-timing applications. Nucl. Instrum. Methods Phys. Res. 857, 98105. Wua, S.C., 2009. Nuclear data sheets for A=214. Nucl. Data Sheets 110, 681. XIA Pixie-16, 2018. Pixie-16 Manual Version 3.03. Hardware Revisions: B, C, D, F, pp. 8–12. http://www.xia.com/.
HPGe system for environmental analysis. Appl. Radiat. Isot. 126, 194–196. Martin, M.J., 2007. Nuclear data sheets for A=208. Nucl. Data Sheets 108, 1583. Metwally, W.A., Mayo, C.W., Han, X., Gardner, R.P., 2005. Coincidence counting for PGNAA applications: is it the optimum method? J. Radioanal. Nucl. Chem. 265, 309–314. MIDAS Multi instance data acquisition system. https://npg.dl.ac.uk/MIDAS/. Paradis, H., de Vismes Otta, A., Cagnata, X., Piquemal, F., Gurriaran, R., 2017. A gammagamma coincidence spectrometer for the measurement of environmental samples. Appl. Radiat. Isot. 126, 179–184. Radford, D.C., 2000. Notes on the use of the program Gf3. https://radware.phy.ornl.gov/ gf3/. Regan, P.H., Shearman, R., Daniel, T., Lorusso, G., Collins, S.M., Judge, S.M., Bell, S.J., Pearce, A.K., Gurgi, L.A., Rudigier, M., Podolyk, Z., Mrginean, N., Mrginean, R., Kisyov, S., 2016. Applications of LaBr3(Ce) gamma-ray spectrometer arrays for nuclear spectroscopy and radionuclide assay. J. Phys. Conf. Ser. 763, 012004.
11
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Determination of activity concentration of 238U and 232Th series radionuclides in soil using a gamma-ray spectrometer in singles and coincidence modes
T
M. Bashira,b,c,∗, R.T. Newmanb, P. Jonesc a
Department of Physics, Ibrahim Badamasi Babangida University Lapai, Nigeria Department of Physics, Stellenbosch University, South Africa c Department of Subatomic Physics, iThemba Laboratory for Accelerator Based Sciences (iThemba LABS), South Africa b
H I GH L IG H T S
minimum detectable activity (MDAs) in singles mode were higher by a factor of 8–36 than for coincidence mode measurements. • The The activity concentration of U and Th series radionuclides were determined using a gamma-gamma coincidence method despite us not having any • traditional background shielding. • The activity concentration of U and Th series radionuclides in the IAEA-375 reference soil material were below the MDA for singles mode measurements. 238
238
232
232
A R T I C LE I N FO
A B S T R A C T
Keywords: Singles Gamma-gamma coincidence Activity 238 U and232Th series
Traditionally, activity concentrations of naturally occurring radioactive materials (NORMs) are measured using a gamma-ray spectrometer with a single detector shielded with lead. In this measurement, a gamma-ray spectrometer comprising an array of four LaBr3:Ce detectors without shielding was used. This spectrometer allowed for measurement in singles and coincidence (gamma-gamma) modes. In addition to using the coincidence method, a novel method of background reduction by using the photon time-of-flight was utilized. Activity concentration of 238U and 232Th series radionuclides in soil reference (IAEA-375) and beach-sand (Bs) were measured in singles and coincidence modes. In coincidence mode, the minimum detectable activities (MDAs) were lower by a factor of 8–36 than for singles mode. Activity concentration of 238U and 232Th series in soil material (IAEA-375) were determined in coincidence mode while it was below MDA in singles mode. The results correlate with the certified values to within the uncertainty although the uncertainties were high because of low statistics.
1. Introduction Naturally occurring radioactive materials (NORMs) are present in every natural substance to a certain degree (James, 2006). These NORMs are often measured using gamma-ray spectrometry with a single detector (NaI:Tl or HPGe) shielded with lead (Elisabeth et al., 2009; Gilmore, 2008; Kohler et al., 2009; Kozak et al., 2001; Lutter et al., 2009; Sykora et al., 2008). To absorb the X-rays generated from the lead shielding, the inner part of the lead shielding is lined with one or combination of the following materials; copper (Cu), cadmium (Cd), tin (Sn), perspex. Sometimes the spectrometer is situated in an underground laboratory for cosmic radiation suppression (Elisabeth et al.,
∗
2009; Lutter et al., 2009; Kohler et al., 2009; Gilmore, 2008). Moreover, a study conducted by Keillor et al. (2017) shows that commercial lead in the U.S. other than stockpiled Doe Run lead contain roughly an order of magnitude higher 210Pb levels. This could be the reason for shift from the use of 10 cm thickness of low-level radioactivity lead to 10.0–14.5 cm thick ordinary lead combined with 3.5–5.0 cm low-level lead (Sykora et al., 2008; Elisabeth et al., 2009; Lutter et al., 2009; Kohler et al., 2009; Kozak et al., 2001). Lead is also expensive and requires the use of other materials in combination with it, thereby increasing the cost of shielding and size. According to Metwally et al. (2005) and Drescher et al. (2017), the use of the gamma-gamma coincidence method has the advantage of
Corresponding author. Department of Physics, Ibrahim Badamasi Babangida University Lapai, Nigeria. E-mail addresses: [email protected], [email protected], [email protected] (M. Bashir).
https://doi.org/10.1016/j.apradiso.2019.108880 Received 12 March 2019; Received in revised form 24 August 2019; Accepted 25 August 2019 Available online 28 August 2019 0969-8043/ © 2019 Elsevier Ltd. All rights reserved.
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
minimizing spectrum background thereby reducing the minimum detectable activity (MDA). Britton et al. (2015a) used a γ − γ coincidence system comprising two large-area, planar HPGe detectors enclosed in a Pb cave to measure calibration sources. Correction for time-walk associated with low energy events and removal of accidental coincidences were accomplished, though activity concentration was not calculated. Britton et al. (2015b) developed a method for quantifying gamma coincidence signatures. A software package which uses ENSDF records as input and calculate the efficiency as well as cascade summing corrected branching ratios was created. This software was tested by measuring 54 Mn, 57,60Co, 65Zn, 88Y, 109Cd,113Sn, 137Cs, 139Ce and 241Am using gamma-gamma system composed of two planar HPGe detectors shielded with lead. The gamma-gamma system produces singles spectra and coincidence matrices using digital electronics and list-mode acquisition. The results for the investigated nuclides were found to be accurate. Antovic and Svrkota (2009a,b) measured activity of 226Ra and 232 Th daughters with a 4π gamma-ray coincidence spectrometer called PRIPYAT-2M. PRIPYAT-2M consists of six-crystal NaI:Tl each with a diameter of 15 cm and height of 10 cm. It has outer dimensions of 250 cm × 154 cm × 186 cm with iron and lead shielding. The PRIP software used enables pulse counting in integral mode and coincidence mode (double, triple, four-fold, five-fold and six-fold coincidences) using buffer memory (FIFO). However, the overlapping gamma-ray peaks centred at 1120.3 and 1238.1 keV were not resolved because of the detector's energy resolution. Markovi et al. (2017) reported on the performance of the Nutech Coincidence Low Energy Germanium Sandwich (NUCLeGeS) system for environmental analysis (210Pb and 134 Cs). The NUCLeGeS system consists of two HPGe detectors enclosed in a 10 cm thick lead shield. A digital acquisition system with 10 ns time resolution was used to collect time-stamped list mode data in anti-coincidence and coincidence modes. The results show that the spectrometer is competitive with single detector systems used in the laboratory for 210Pb analysis. Paradis et al. (2017) measured the radioactivity in environmental samples using a gamma-gamma coincidence spectrometer called Leda in singles, coincidence or anti-coincidence modes. Leda is made of 2 HPGe and 1 NaI:Tl detectors. The data acquired were time-stamped. The detection limits for all γ-emitters were improved. Below we report on an investigation into the use of a modern array of LaBr3:Ce scintillator detectors to measure activity concentrations of primordial radionuclides in soil and sand samples. Here, the measurement of gamma-gamma coincidence along time information is shown. This allows for achievement of lower minimum detectable activity relative to measurements for some radionuclides in singles mode. The setup is relatively simple requiring only the detectors and the associated electronics. Moreover, the time resolution of a LaBr3:Ce based detector is typically a few hundred picoseconds (Kumar Anil et al., 2009). It has high scintillation light output (Favalli et al., 2008), a typical energy resolution of 3% at 662 keV (Dorenbos et al., 2004) and high intrinsic gamma-ray detection efficiency (Bashir et al., 2018; Vedia et al., 2017; Regan et al., 2016). Furthermore, there is no need for operating the detectors at liquid nitrogen temperature. 238 U and 232Th are unstable and undergo a series of decays. 238U have thirteen progenies (234Th*, 234Pa*, 234U, 230Th, 226Ra*, 222Rn, 218 Po, 214Pb*, 214Bi*, 214Po, 210Pb*, 210Bi, 210Po) before stable 206Pb is reached (Gilmore, 2008; Martin, 2007). 232Th have nine progenies (228Ra, 228Ac*, 228Th, 224Ra*, 220Rn, 216Po, 212Pb*, 212Bi*, 212Po, 208 Tl*) before stable 208Pb is reached (Gilmore, 2008; Martin, 2007). According to Gilmore (2008), only radionuclides marked with * can easily be measured by gamma-ray spectrometry. An inspection of the decay scheme of 238U and 232Th (and their daughters) showed that only gamma-rays from β − decay of 214Bi and 208Tl have energies and intensities suitable for gamma-gamma coincidence measurements (with the experimental geometry used for this study). The most intense gamma-rays with their intensities in brackets from β − decay of 214Bi to 214Po are: 609.3 keV (45.5%), 1120.3 keV (14.9%),
Fig. 1. Picture of different experimental geometries investigated using an array of LaBr3:Ce gamma-ray detectors.
1238.1 keV (5.8%), 768.4 keV (4.9%), 2204.1 keV (4.9%) 1377.7 keV (4.0%), 934.1 keV (3.1%) (Wua, 2009). And the most intense gammarays from β − decay of 208Tl to 208Pb are: 2614.5 keV (99.8%), 583.2 keV (85.0%), 510.7 keV (22.6%), 860.6 keV (12.5%), 277.4 keV (6.6%) (Martin, 2007). For this study, coincidences between 609.3 keV and 1120.3 keV were chosen for the 238U series measurement, while 583.2 keV and 2614.5 keV were chosen for the 232Th series measurements because of their intensities. These chosen gamma-rays energies are also in coincidence with other cascade gamma-ray energies not of interest in this work (Martin, 2007; Wua, 2009).
2. Experimental methods At iThemba LABS, eight cylindrical LaBr3:Ce scintillator detectors are available for gamma-ray measurements. These detectors were manufactured by Saint-Gobain and have a crystal size of 5.1 cm × 5.1 cm and R2083 photomultiplier tubes (PMTs) attached to them. Different experimental geometries were investigated using eight (equidistant of 24.0 cm), four (equidistant of 15.0 cm) and two (equidistant of 10.0 cm) detectors (Fig. 1). It was found that the eight detector experimental geometry has the most scattering events (Fig. 2) because of the distance between the detectors. However, the two detector experimental geometry has the least scattering events and its detection efficiency is low. This is as a result of the number of detectors used. The four detector experimental geometry has fewer scattering events than the eight detector experimental geometry. Also, its detection efficiency is better than that for the two detector experimental geometry. Its efficiency can be improved by reducing the source-todetector distance. Furthermore, the time-of-flight of the source gamma-ray can be discerned from that of the scattered gamma-ray (Fig. 2). For the eight detector experimental geometry, the Compton scattered gamma-ray time-of-flight are ≈ 700 ps, ≈ 1400 ps, ≈ 1700 ps and ≈ 1900 ps at angle 45∘, 90∘, 135∘ and 180∘ respectively (Fig. 2). Again, the minimum timeof-flight can be estimated from Fig. 3 using the speed of light and the source-to-detector distance (d is the distance between the sample and detector front surfaces plus the beaker radius) and the distance between detectors. This enables the time-of-flight to be used and the background/scattered gamma-ray events can be significantly reduced by setting a time gate smaller than the scattered gamma-ray time-of-flight. For this study, an experimental geometry involving now only four detectors is used with samples placed 16.5 cm equidistant from the detectors. Each detector signal was linked to a XIA Pixie-16 module (XIA Pixie-16, 2018). The XIA Pixie-16 module is a 16 channel digital signal processing module, providing readout of timestamp, CFD time 2
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 2. Variation in time difference between different detector combination (at 45°, 90°, 135° and 180°) for the eight detector experimental geometry.
measure the background and container contribution. Soil reference (IAEA-375) and beach-sand (Bs) from Zanzibar were measured to obtain their activity concentration. 137Cs and 60Co point sources were used for energy calibrations. Further information about the samples and measurement time are given in Table 1. 3. Data analyses Offline, the list-mode data acquired (which gives the advantage of enabling the re-processing the raw data) were sorted into channels, time and energy. The sorted data were binned while energy and time spectra were plotted, using ROOT software (Brun and Rademakers, 1997).
Fig. 3. Picture illustrating the distance (d) between sample and detector front surfaces plus the Marinelli beaker radius; and between the detectors (L1 to L8).
and calculated energy directly from the detector output. These data consist of event ID, time stamp (48-bit), CFD time (16-bit), calculated energy (16-bit) at a 14 bit sampling resolution at 500 MHz (XIA Pixie16, 2018). In addition, a field-programmable gate array (FPGA) first captures 5 ADC samples at the rate of 100 MHz (i.e every 10 ns). Full CFD timing data are calculated from the differences between the delayed and non-delayed sums, providing the zero-crossing point (XIA Pixie-16, 2018). MIDAS software code (MIDAS) ran on a Linux workstation for data collection. These software access all the data readout from the hardware electronics through a single Peripheral Component Interconnect (PCI) bridge (PXI-8360). The time resolution of the detector is ≈ 370 ps FWHM which allow the electronics to resolve time difference at the sub-nanosecond level. Finally, each sample was placed in a Marinelli beaker (1.0 L) having dimensions of 13.0 cm diameter, 8.5 cm well-diameter, 13.5 cm height, 6.1 cm well-height and 0.2 cm thickness. Uranium ore (RGU-1) and thorium ore (RGTh-1) from IAEA were measured to facilitate the detection efficiency calibration. An empty Marinelli beaker was used to
3.1. Singles First, the energy spectra for the four detectors were gain matched and summed for each measured sample. The empty beaker spectrum (Fig. 4 upper) was normalized to the sample counting time and subtracted from the sample's spectra. The IAEA-375 soil and beach-sand samples spectra before and after the empty beaker spectrum subtraction Table 1 Reference material and sample information.
3
Sample/Source
Density (kgm−3)
Live time (seconds)
Uranium Ore (RGU-1) Thorium Ore (RGTh-1) Soil reference (IAEA-375) Beach sand (Bs) Empty beaker 137 Cs and 60Co
1409.12 1364.94 1502.63 2658.30 – –
172796 172764 172787 172983 57594 7200
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 4. Normalized summed total empty beaker spectra in singles mode (upper) and the summed total projection of coincidence events spectra for empty beaker without energy gates in coincidence mode (lower).
Fig. 5. Summed total spectra before and after empty subtraction for the IAEA-375 soil sample, and the normalized empty beaker.
Fig. 6. Summed total spectra before and after empty beaker subtraction for the beach-sand sample, and the normalized empty beaker.
4
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 7. Time difference spectra for six of the two combinations of four detectors (A) (the blue and purple colours are for detectors at 180° while red, green, black and yellow colours are for detectors at 90°), beach-sand (sample) and empty beaker time difference spectra (B), and the time gate tc (centroid peak) used to generate gamma-gamma coincidence spectra (C). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 8. Summed total gamma-gamma spectra for the beach-sand sample generated using peaks on either side of Fig. 7 time gates to show background/scattered events.
Fig. 9. Summed total gamma-gamma spectra for the IAEA-375 soil reference.
5
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 10. Summed total gamma-gamma spectra for the beach-sand sample.
Fig. 11. Summed total projection for the IAEA-375 soil reference.
Fig. 12. Summed total projection for the beach-sand sample.
are shown in Figs. 5 and 6, respectively. It can be observed from these figures that 40 K in the empty beaker is higher than in the samples. This could be due to the sample acting as a shield between the detector and
the natural background. Second, the peaks area counts (Cn) were extracted using the integral function in ROOT. Radware Gf3 was used to resolve and fit the 6
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 13. Energy gated spectra: 609.3 keV and 1120.3 keV (214Bi) for the IAEA-375 soil reference.
Fig. 14. Energy gated spectra: 583.2 keV and 2614.5 keV (208Tl) for the IAEA-375 soil reference.
Eqn. (1).
overlapping 583.2 keV and 609.3 keV peaks (Radford, 2000). Absolute full-energy peak gamma-ray detection efficiency (ε) at various peaks of interest were calculated using Eqn. (1) (Gilmore, 2008)
ε=
Cn APγ Ms Lt
3.2. Coincidence The time difference between gamma-gamma coincidence events associated with any two LaBr3:Ce detectors within the time window of 10000 ps were plotted using the time-stamped data and shown in Fig. 7 (A). In Fig. 7(C), the region tc is the true coincidence peak which contain the coincident events of interest, and the peaks on either side are random coincidence between the detectors due to background radiation from terrestrial and cosmic radiation, and/or Compton scattering as there was no shielding used (Fig. 7(B)). Two different time gates, the total time tT (−5000 ps–5000 ps) and the time gate tc (Fig. 7(C)) were used. These time gates were set on the empty beaker, IAEA-375 soil and beach-sand data to generate gammagamma spectra for any two detectors in coincidence. It is important to note that the spectra shown are all for time gate tc and the total time tT was computed to check the sensitivity of time gate tc. Although, the
(1) −1
where A is the source activity concentration in Bqkg , Pγ is the probability of emission of the gamma-ray being measured, Ms is sample mass in kg and Lt is the spectrometer data acquisition live time in seconds. The minimum detectable activity (MDA) was calculated using Eqn. (2) (Gilmore, 2008), where σB is the uncertainty in the background (empty beaker) peak count:
MDA =
2.71 + 3.29σB εPγ Ms Lt
(2) 238
232
The activity concentration of U and Th series radionuclides in IAEA-375 soil and beach-sand samples were calculated by re-arranging 7
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Fig. 15. Energy gated spectra: 609.3 keV and 1120.3 keV (214Bi) for the beach-sand sample.
Fig. 16. Energy gated spectra; Energy gated spectra: 583.2 keV and 2614.5 keV (208Tl) for beach-sand sample. Table 2 Weighted average counts in the coincidence peaks. Sample
Series
Nuclide
Counts (tc)
Counts (tT)
238
214
232
208
4.5 ± 2.1 3.0 ± 1.7
5.5 ± 2.3 3.3 ± 1.8
238
214
232
208
327.3 ± 18.1 402.5 ± 20.1
331.3 ± 18.2 424.3 ± 20.6
Table 3 Total counts in the gamma-gamma matrix generated using time gates tc and tT for all the coincidence detectors combination in IAEA-375 soil and beach-sand after empty beaker subtraction.
IAEA-375 U Th
Bi Tl
Detector Combination
Bs U Th
Bi Tl
Angle
L1-L2 L1-L3 L1-L4 L2-L3 L2-L3 L3-L4
peaks on either side of time gates (Fig. 7) were set on the beach-sand data to generate the gamma-gamma matrix in Fig. 8 to show the background/scattered events contribution, the matrix was not used in calculating activity concentration. The spectra were summed for each sample and the normalized empty beaker spectrum was subtracted from that of the IAEA-375 soil and beach-sand. Figs. 9 and 10 shows the total gamma-gamma coincident spectra. The two-dimensional gamma-
180° 90° 90° 90° 90° 180°
IAEA
Bs
Counts (tc)
Count (tT)
Counts (tc)
Counts (tT)
1338 2477 4849 3937 2826 995
30416 118671 110530 115331 108386 29635
19949 29816 23038 27561 23559 21198
56935 152482 143332 148511 141298 57045
gamma matrix corresponds to γ-rays in the decay cascade being detected in coincidence (coincidence pairs at [609.3, 1120.3] and [583.2, 2614.5]) and random coincidence. 8
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Table 4 Absolute full-energy peak detection efficiencies in singles and coincidence modes. Series
Nuclide
Energy
Efficiencies
Efficiencies
[keV]
Singles ( × 10−3)
Coincidence ( × 10−6)
609.3 1120.3 1764.5
5.0 ± 0.1 3.0 ± 0.1 2.0 ± 0.1
14.9 ± 1.0 14.9 ± 1.0 –
583.2 911.2 2614.5
5.0 ± 0.2 1.9 ± 0.1 1.3 ± 0.1
6.6 ± 0.3 – 6.6 ± 0.3
238
U 214
Bi Bi Bi
214 214 232
Th 208
Tl Ac 208 Tl 228
Fig. 18. Activity concentration of 238U series in singles and coincidence modes for the beach-sand sample measured with LaBr3:Ce and HPGe.
Table 5 Minimum Detectable Activities (MDAs) in singles and coincidence modes. Series
Nuclide
Energy
MDAs
MDAs
HPGe
[keV]
Singles
Coincidence
MDAs
Bi Bi 214 Bi
609.3 1120.3 1764.5
95.6 422.0 252.2
11.6 11.6 –
1.1 3.0 2.9
208
583.2 911.2 2614.5
127.2 385.7 166.5
15.9 – 15.9
1.9 1.6 1.0
238
U 214 214
232
Th Tl Ac Tl
228 208
Several coincidence conditions were set on the gamma-gamma coincident spectra to measure the full-energy peak in one of the two detectors. The summed total projection of coincidence events measured with any two detectors are shown in Figs. 11 and 12. These were used to get full-energy peak width of the energies of interest used for the energy gates. As shown in Figs. 13–16 when a gate is set on 609.3 keV, the 1120.3 keV 25+ → 21+ transition is seen and the 609.3 keV 21+ → 01+ transition is absent (from the decay of 214Bi to 214Po) and vice versa. Similarly, from the decay of 208Tl to 208Pb, when a gate is set on 583.2 keV, the 2614.5 keV 31−− → 01+ transition is seen and the 583.2 keV 51− → 31− transition is absent, and vice versa. As observed from Figs. 15 and 16 when a gate is set on either 583.2 keV or 609.3 keV peaks at both 1120.3 keV and 2614.5 keV were seen. This is because of the overlap of 583.2 keV and 609.3 keV full-energy peaks.
Fig. 17. Measured activity concentration of
238
U and
232
Fig. 19. Activity concentration of 232Th series in singles and coincidence modes for the beach-sand sample measured with LaBr3:Ce and HPGe.
The weighted average of the coincidence peaks count using time gate tT and tc are given in Table 2. The detection efficiency for two gamma-ray energies measured in coincidence is the product of the efficiency of the two gamma-ray energies measured in singles mode (Siegbahn, 1965) i.e, ε γc = ε γ1 ε γ2 . The true coincidence summing (TCS) for this experimental geometry is negligible (0.6% solid angle and 0.006% chance of summing for our distance) because the TCS decrease with increase in source-to-detector
Th series in coincidence mode and expected value for the IAEA-375 soil reference. 9
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
samples. Coincidence between detectors at 90° have more scattered events than those at 180° i.e, the larger the angle between detectors in coincidence the lower the number of scattering events. However, using the photon time-of-flight in addition to the coincidence method, the scattered events were significantly reduced. Also, the background was reduced by two orders of magnitude compared to the singles method. Furthermore, low-level activity concentration ≥ 11.6 Bqkg−1 and ≥ 15.9 Bqkg−1 for 238U and 232Th series radionuclides can be measured in coincidence mode. Using this spectrometer, activity concentration of radionuclides having an associated gamma-decay cascade can be measured with any sample geometry in coincidence mode.
distance (Gilmore, 2008). Moreover, both Gaussian and Poisson analysis were performed to estimate the coincidence MDA but due to the level of statistics Poisson distribution was used. The peaks counts in the summed total projection of coincidence event spectra for empty beaker in Fig. 4 (lower) extracted and Eqn. (3) (Gilmore, 2008) used to calculate coincidence MDA:
MDA =
LD ε γc Pγ1 Pγ 2 Ms Lt
(3)
where LD is the detection limit and the value taken from Gilmore (2008) for Poisson distribution (for 95% confidence) associated with background count within the counting time, and Pγ1 and Pγ2 are gamma-ray emission probabilities of gamma energy 1 and 2 respectively. The activity concentration in coincidence was calculated using Eqn. (4) suggested by Siegbahn (1965), where Nc is the coincidence peak area counts, n is the number of detectors used (in this case n = 4) and m is the number of combinations for four detectors taken two detectors at a time (in this case m = 6)
Ac =
2nNc (mε γc Pγ1 Pγ 2 Ms Lt )
Acknowledgements This authors would like to thank IBB University/TETFUND for the bursary, National Research Foundation of South Africa (99037) for providing the equipment and iThemba LABS (Laboratory for Accelerator Based Science) for the top-up funding. Also like to thank Mr. Lumkile Msebi for his support during the experiment. Appendix A. Supplementary data
(4) Supplementary data to this article can be found online at https:// doi.org/10.1016/j.apradiso.2019.108880.
4. Results and discussion References The uncertainties quoted in all the results are at the 1σ level. As observed from Table 3, coincidence between detectors at 90° have more scattered events than those at 180° when total time gate tT was used to generate the gamma-gamma matrix. But, when the time gate tc was used to generate the gamma-gamma matrix the scattered events were significantly reduced. Table 4 gives the absolute full-energy peak detection efficiencies of 238 U and 232Th series radionuclides at various gamma-ray energies of interest using singles and coincidence methods. The detection efficiencies for coincidence mode are far lower than that for singles modes. The 238U and 232Th series radionuclides MDAs in coincidence mode are all < 16 Bqkg−1 while that of singles are > 100 Bqkg−1 except for 609.3 keV which is 96 Bqkg−1 (Table 5). However, the MDAs measured with HPGe is lower than that of LaBr3:Ce in coincidence mode. In coincidence mode, our MDA for 609.3 keV was lower than that of Antovic and Svrkota (2009a,b) and higher for 583.2 and 2614.5 keV. Activity concentration of 238U and 232Th series radionuclides in IAEA-375 from the singles mode measurement was below the MDA for this experimental geometry which agrees with results from Bashir et al. (2018). Measured activity concentration of radionuclides inside the IAEA-375 soil reference in coincidence mode are consistent with certified values to within measurement uncertainties (Fig. 17). The measured activity concentrations are within the uncertainty although the uncertainties were high because of low statistics. Figs. 18 and 19 shows that the activity concentration of 238U and 232Th series in beach-sand measured in singles and coincidence modes were consistent to within measurement uncertainty. This sample was also measured using a hyper-pure germanium (HPGe) detector shielded with lead and the activity concentration agreed with our results (Figs. 18 and 19). According to UNSCEAR (2000), the world average activity concentration of 238U and 232Th are 35 Bqkg−1 and 30 Bqkg−1, respectively. Therefore, using this experimental geometry in coincidence mode low-level activity concentration of 238U ( ≥ 11.6 Bqkg−1) and 232 Th ( ≥ 15.9 Bqkg−1) series radionuclides can be measured.
Antovic, N., Svrkota, N., 2009a. Development of a method for activity measurements of 232 Th daughters with a multidetector gamma-ray coincidence spectrometer. Appl. Radiat. Isot. 67, 1133–1138. Antovic, N., Svrkota, N., 2009b. Measuring the Ra-226 activity using a multidetector γray coincidence spectrometer. J. Environ. Radioact. 100, 823–830. Bashir, M., Newman, R.T., Jones, P., 2018. Measurements of natural radioactivity in soil using an array of cerium doped lanthanum bromide scintillator detectors. In: The Proceedings of SAIP2017, the 62nd Annual Conference of South African Institute of Physics. Division B- Nuclear, Particle and Radiation Physics, vols. 131–134978-0620-82077-6, . Britton, R., Burnett, J.L., Davies, A.V., Regan, P.H., 2015a. Coincidence corrections for a multi-detector gamma spectrometer. Nucl. Instrum. Methods Phys. Res. 769, 2025. Britton, R., Jackson, M.J., Davies, A.V., 2015b. Quantifying radionuclide signatures from a γ−γ coincidence system. J. Environ. Radioact. 149, 158163. Brun, Rene, Rademakers, Fons, 1997. ROOT-an object oriented data analysis framework. In: Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, vol 389. pp. 81–86 1. Dorenbos, P., et al., 2004. Gamma-ray spectroscopy with a ⊘19 ×19 mm 3LaBr:3 0.5 % Ce 3+scinillator. IEEE Trans. Nucl. Sci. 51 (3), 1289–1296. Drescher, A., et al., 2017. Gamma-gamma coincidence performance of LaBr3:Ce scintillation detectors vs HPGe detectors in high count-rate scenarios. Appl. Radiat. Isot. 122, 116–120. Elisabeth Wieslander, J.S., Hult, Mikael, Gasparro, Joel, Marissens, Gerd, Misiaszek, Marcin, Werner, Preusse, 2009. The Sandwich spectrometer for ultra low-level γ-ray spectrometry. Appl. Radiat. Isot. 67, 731–735. Favalli, A., Mehner, H.C., Simonelli, F., 2008. Wide energy range efficiency calibration for a lanthanum bromide scintillation detector. Radiat. Meas. 43, 506–509. Gilmore, G., 2008. Practical Gamma-Ray Spectrometry, second ed. John Wiley & Sons, Ltd., West Sussex, England978-0-470-86196-7. James, Martin E., 2006. Physics for Radiation Protection: A Handbook, second ed. WILEYVCH Verlag GmbH and Co. KGaA, Weinheim. Keillor, Martin E., Aalseth, C.E., Arnquist, I.J., Eggemeyer, T.A., Fuller, E.S., Glasgow, B.D., Hoppe, E.W., Morle, S.M., Myers, A.W., Orrell, J.L., Overman, C.T., Seifert, A., Shaff, S.M., Thommasson, K.S., 2017. Recent Bremsstrahlung-based assays of Pb-210 in lead and comments on current availability of low-background lead in North America. Appl. Radiat. Isot. 126, 185–187. Kohler, M., Degering, D., Laubenstein, M., Quirin, P., Lampert, M.-O., Hult, M., Arnold, D., Neumaier, S., Reyss, J.-L., 2009. A new low-level γ-ray spectrometry system for environmental radioactivity at the underground laboratory Felsenkeller. Appl. Radiat. Isot. 67, 736–740. Kozak, Kryzysztof, Mietelski, Jerzy W., Jasifska, Miroslawa, Gaca, Pawel, 2001. Decreasing of the natural background counting - passive and active method. In: NUKLEONIKA PROCEEDINGS, vol. 46. pp. 165–169 4. Kumar Anil, G., Mazumdar, I., Gothe, D.A., 2009. Experimental measurements and GEANT4 simulations for a comparative study of efficiencies of LaBr3:Ce, NaI:Tl, and BaF2. Nucl. Instrum. Methods Phys. Res. 610, 522529. Lutter, Guillaume, Hult, Mikael, Marissens, Gerd, Andreotti, Erica, Rosengard, Ulf, Misiaszek, Marcin, Yuksel, Ayhan, Sahin, Namik, 2009. A new versatile underground gamma-ray spectrometry system. Nucl. Instrum. Methods Phys. Res. 610, 522–529. Markovi, Nikola, Roos, Per, Nielsen, Sven Poul, 2017. Digital gamma-gamma coincidence
5. Conclusion This work presents for the first time the use of coincidence techniques with LaBr3:Ce detectors without shielding to measure the activity concentration of 238U and 232Th series radionuclides in volume soil 10
Applied Radiation and Isotopes 154 (2019) 108880
M. Bashir, et al.
Siegbahn, Kai, 1965. Alpha-, Beta- and Gamma-Ray Spectroscopy, vol. 1 North-Holland Publishing Company Amsterdam. Sykora, I., Jeskovsky, M., Janik, R., Holy, K., Chudy, M., Provinec, P.P., 2008. Low-level single and coincidence gamma-ray spectrometry. J. Radioanal. Nucl. Chem. 276, 779–787. United Nation Scientific Committee on the Effects of Atomic Radiation, 2000. Sources and effects of ionizing radiation. In: Report to the General Assembly, with Scientific Annexes, vol. 1. Vedia, V., Carmona-Gallardo, M., Fraile, L.M., Mach, H., Udas, J.M., 2017. Performance evaluation of novel LaBr3(Ce) scintillator geometries for fast-timing applications. Nucl. Instrum. Methods Phys. Res. 857, 98105. Wua, S.C., 2009. Nuclear data sheets for A=214. Nucl. Data Sheets 110, 681. XIA Pixie-16, 2018. Pixie-16 Manual Version 3.03. Hardware Revisions: B, C, D, F, pp. 8–12. http://www.xia.com/.
HPGe system for environmental analysis. Appl. Radiat. Isot. 126, 194–196. Martin, M.J., 2007. Nuclear data sheets for A=208. Nucl. Data Sheets 108, 1583. Metwally, W.A., Mayo, C.W., Han, X., Gardner, R.P., 2005. Coincidence counting for PGNAA applications: is it the optimum method? J. Radioanal. Nucl. Chem. 265, 309–314. MIDAS Multi instance data acquisition system. https://npg.dl.ac.uk/MIDAS/. Paradis, H., de Vismes Otta, A., Cagnata, X., Piquemal, F., Gurriaran, R., 2017. A gammagamma coincidence spectrometer for the measurement of environmental samples. Appl. Radiat. Isot. 126, 179–184. Radford, D.C., 2000. Notes on the use of the program Gf3. https://radware.phy.ornl.gov/ gf3/. Regan, P.H., Shearman, R., Daniel, T., Lorusso, G., Collins, S.M., Judge, S.M., Bell, S.J., Pearce, A.K., Gurgi, L.A., Rudigier, M., Podolyk, Z., Mrginean, N., Mrginean, R., Kisyov, S., 2016. Applications of LaBr3(Ce) gamma-ray spectrometer arrays for nuclear spectroscopy and radionuclide assay. J. Phys. Conf. Ser. 763, 012004.
11