- Email: [email protected]
Low-background gamma-ray spectrometry for the international monitoring system
Applied Radiation and Isotopes (xxxx) xxxx–xxxx
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Low-background gamma-ray spectrometry for the international monitoring system ⁎
L.R. Greenwood , M.G. Cantaloub, J.L. Burnett, A.W. Myers, C.T. Overman, J.B. Forrester, B.G. Glasgow, H.S. Miley Pacific Northwest National Laboratory, Richland, WA 99352, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: Gamma-ray spectroscopy Low background detectors IMS
PNNL has developed two low-background gamma-ray spectrometers in a new shallow underground laboratory, thereby significantly improving its ability to detect low levels of gamma-ray emitting fission or activation products in airborne particulate in samples from the IMS (International Monitoring System). The combination of cosmic veto panels, dry nitrogen gas to reduce radon and low background shielding results in a reduction of the background count rate by about a factor of 100 compared to detectors operating above ground at our laboratory.
1. Introduction The International Monitoring System (IMS) of the Comprehensive Nuclear Test Ban Treaty Organization (CTBTO) consists of a network of 80 field stations and 16 laboratories dedicated to the detection and analysis of airborne particulates to detect possible nuclear events that could indicate a treaty violation. USL16 at PNNL (Pacific Northwest National Laboratory) was certified by the IMS in 2007 to perform more detailed analyses of air particulate samples of interest collected by the field stations. Air particulate samples are solely analyzed by gamma-ray spectrometry, although the IMS also includes many other techniques for detecting treaty violations including xenon gas, seismic monitors, and infrasound monitors. The gamma-ray detectors used by the IMS network are mainly conventional, off-the-shelf HPGe (high purity germanium detectors). PNNL has developed two very low-background detectors located in a shallow underground laboratory to significantly improve the presence of weak activity concentrations of gammaemitting fission or activation products. 2. Detector design PNNL constructed a shallow underground laboratory to allow the development of low-background materials and for the deployment of very low-background radiation detectors (Aalseth et al., 2012; Forrester et al., 2013). The new laboratory is about 15 m underground and includes an overburden on top to achieve a background of about 30 m water equivalent. The ambient fast neutron flux is reduced about a factor of 100 relative to above ground locations. The section of the
⁎
underground laboratory used for low-level radiation detection operates as a class 10,000 clean room. Other parts of the laboratory operate at the class 1000 level. The gamma-ray detectors were custom ordered from Canberra, specifying very low background materials for all detector components, as supplied by Canberra, to reduce the presence of naturally occurring radioactive materials. One of the detectors is a Ptype crystal, with a relative efficiency of 112% and the second detector is n-type, with a relative efficiency of 97%. Both detectors have carbon fiber windows and copper end caps. The radiation shields were custom designed consisting of 20 cm of low background lead lined with 5 cm of oxygen-free high thermal conductivity (OFHC) copper. Whereas copper is typically used in very thin layers in above ground detectors to avoid neutron activation of the copper, that effect is greatly reduced in our underground laboratories due to the factor of 100 reduction of the neutron background. Five large cosmic veto panels surround the detectors and shields to suppress cosmic ray induced events. Some of the cosmic veto panels are from Eljen and are made of EJ-200 plastic scintillator wrapped with reflector foil in light tight boxes. The rest of the panels are from St. Gobain using BC-408 plastic scintillators. The entire assemblies are enclosed in a glove box that has a high flow of dry nitrogen gas from the boil off of a large liquid nitrogen tank outside the building. Samples are introduced and removed from the detector using a pass through that is flushed with nitrogen prior to opening to the detector enclosure. The cosmic veto nearly eliminates the 511 keV peak; the dry nitrogen flow greatly reduces the radon peaks; and the low background detector materials nearly eliminate naturally occurring peaks such as the 1460 keV peak from 40K. The sample holders for the IMS samples were designed and
Corresponding author. E-mail address: [email protected] (L.R. Greenwood).
http://dx.doi.org/10.1016/j.apradiso.2016.12.034 Received 28 July 2016; Received in revised form 29 November 2016; Accepted 20 December 2016 0969-8043/ © 2016 Elsevier Ltd. All rights reserved.
Please cite this article as: Greenwood, L.R., Applied Radiation and Isotopes (2016), http://dx.doi.org/10.1016/j.apradiso.2016.12.034
Applied Radiation and Isotopes (xxxx) xxxx–xxxx
L.R. Greenwood et al.
approximately a factor of 100 lower than normally seen at our site for commercial off-the-shelf (COTS) HPGe (high-purity germanium) detectors using a commercial shield with normal lead (20 cm thick) in an above ground location. The finished detector assemblies are shown in Fig. 3. Samples received from IMS stations are Radionuclide Aerosol Sample Analyzer (RASA) samples, compressed using our 30 t press, or Snow White and Cinderella air particulate samples which are counted as received since these samples are compressed at the field stations. The outer plastic sample holders are cleaned to remove any possible surface contamination and then bagged in air tight bags before taking them down to the underground laboratory. All personnel don protective clothing before entering the facility where the detectors are located. Samples are introduced to the sealed glove box containing the detector and shield using a pass through. The outer door is opened to introduce the sample. After closing the outer door, the pass though is flushed with dry nitrogen for several minutes. The inner door is then opened using the glove box gloves to reach the sample. The sample is then put into one of the ABS sample holders and carefully placed on the detector such that nothing touches the detector surface. After counting, the sample is removed in the reverse procedure to minimize any leakage of outside air into the inner glove box during sample insertion or removal. The background count rate for our p-type detector is typically 0.047 cps (counts per second) or 1840 counts per day per kg of Ge over the range from 40 to 2700 keV. The weight of the germanium crystal is 2.17 kg. Due to the carbon fiber window and thin Ge deadlayer, the detector has a relatively high efficiency at lower gamma-ray energies compared to a typical commercial HPGe detector. The minimum detectable activity (MDA) of Ba-140 in a 7-d count is about 3 mBq, a factor of 8 lower than the required value of 24 mBq for IMS detectors. The MDA is calculated according to the IMS requirement where the MDA is equal to the lower limit of detection at the 95% confidence level divided by the count time, gamma-ray efficiency, correction for true coincident summing, gamma-ray emission probability, and decay during counting. The LD value is calculated according to the formula:
Fig. 1. The copper-lined detector chamber is shown with the sliding top door open. Some of the cosmic veto detector panels are shown on the left side. The detector face can just be seen at the bottom of the copper section.
fabricated with a 3D printer using acrylonitrile butadiene styrene (ABS) plastic, which was found to be very low-background when screened by gamma-ray counting. The detector shields are shown in Fig. 1. The design of the low background detectors and shields was inspired by prior work at our laboratory and other low background detector systems including Baudis et al. (2011), Neder et al. (2000), and Keillor et al. (2009). These new low-background detectors were designed as a more cost-effective way of expanding our IMS laboratory compared to the PNNL-fabricated CASCADES detector (Keillor et al. (2009)). 3. Detector operation and performance The detectors are operated using Canberra Lynx digital analyzers. A pulse seen in any of the cosmic detector panels is used to veto any signals from the germanium detectors. This results in a significant reduction in cosmic events registered by the detectors, as seen in the about 98% reduction of the 511 keV background peak. The use of low background materials for the detector and shield resulted in about a 97% reduction in the 1460 keV peak from naturally occurring 40K. Typical background spectra are shown in Fig. 2. The top curve is for a COTS detector with 114% relative efficiency in an above ground laboratory. The bottom two curves show empty cave (no sample inserted) gamma-ray spectra measured on our low-background detectors located in the underground laboratory and with the addition of the cosmic veto and radon reduction systems. The net background is
LD = 2.71 + (B kg)1/2
(1)
where Bkg=background counts in the interval of ± 1.25 FWHM. The n-type detector has a high efficiency at very low gamma-ray energies such that cascade summing with x-rays is seen routinely for many gamma-ray peaks. To reduce this effect, a steel absorber with a thickness of 0.13 mm (5 mils) was inserted over the face of the detector. Partly due to the much higher efficiency at lower gamma-ray energies, the background count rate is about twice that of the p-type detector at about 0.088 cps or 4000 counts per day per kg of Ge (1.95 kg) for the energy range of 40–2900 keV.
Fig. 2. Comparison of HPGe spectra for a COTS system in an above ground laboratory (top) and the n-type and p-type low-background detectors in the underground laboratory with cosmic veto panels and radon reduction using dry nitrogen gas.
2
Applied Radiation and Isotopes (xxxx) xxxx–xxxx
L.R. Greenwood et al.
Fig. 3. Final assembly of the new detectors in our underground laboratory. Note the sample pass through, glove ports, wheel to slide open the top of the lead cave, and dewar located outside the enclosure.
Technology program (DTRA 10027-16504) of the Defense Threat Reduction Agency. Pacific Northwest National Laboratory is operated for DOE by Battelle Memorial Institute under Contract DE-AC06-76RLO 1830.
The background count rates that we have obtained with the low background detectors in our underground laboratory are comparable to what has been seen at other underground laboratories using similar systems shown by Baudis et al. (2011), Neder et al. (2000), Keillor et al., 2009, and Povinec et al. (2004). For example, the IAEA-MEL (Povinec et al. (2004)) detectors at a comparable depth underground (35 m w.e.) have count rates of 35–68 counts/hour-kg over the range of 40–2700 keV compared to our background rates of 77 counts/hour-kg for our p-type detector and 166 counts/hour-kg for our n-type detector. We are continuing to look at techniques to further reduce the background. The use of cosmic veto detectors was found to be especially effective for background reduction, somewhat offsetting the additional benefit of using underground laboratories at a greater depth. Experiments are currently in progress with the use of cosmic veto panels surrounding above ground gamma-ray detectors to determine what performance can be achieved.
References Aalseth, C.E., et al., 2012. A shallow underground laboratory for low-background radiation measurements and materials development. Rev. Sci. Inst. 83 (11). http:// dx.doi.org/10.1063/1.4761923. Baudis, L., et al., 2011. Gator: a low-background counting facility at the gran sasso underground laboratory. J. Inst. 6, P08010. http://dx.doi.org/10.1088/1748-0221/ 6/08/P08010. Forrester, J.B., et al., 2013. Construction of a shallow underground low-background detector for a CTBT radionuclide laboratory. J. Radioanal. Nucl. Chem. 296 (2), 1061–1064. http://dx.doi.org/10.1007/s10967-012-2202-3. Keillor, M.E., et al., 2009. Design and construction of an ultra-low-background 14-crystal germanium array for high efficiency and coincidence measurements. J. Radioanal. Nucl. Chem. 3 (282), 703–708. http://dx.doi.org/10.1007/s10967-009-0248-7. Neder, H., et al., 2000. Low level g-ray germanium-spectrometer to measure very low primordial radionuclide concentrations.. Appl. Radiat. Isot. 53, 191–195, (PII: S0969-8043(00)(00132-9). Povinec, P.P., et al., 2004. IAEA-MEL's underground counting laboratory in Monaco – background characteristics of HPGe detectors with anti-cosmic shielding. Appl. Radiat. Isot. 61, 85–93. http://dx.doi.org/10.1016/j.apradiso.2004.03.019.
Acknowledgements This research was sponsored by the Office of Nuclear Verification, National Nuclear Security Administration. The authors wish to acknowledge the funding support of the Nuclear Arms Control
3
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Low-background gamma-ray spectrometry for the international monitoring system ⁎
L.R. Greenwood , M.G. Cantaloub, J.L. Burnett, A.W. Myers, C.T. Overman, J.B. Forrester, B.G. Glasgow, H.S. Miley Pacific Northwest National Laboratory, Richland, WA 99352, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: Gamma-ray spectroscopy Low background detectors IMS
PNNL has developed two low-background gamma-ray spectrometers in a new shallow underground laboratory, thereby significantly improving its ability to detect low levels of gamma-ray emitting fission or activation products in airborne particulate in samples from the IMS (International Monitoring System). The combination of cosmic veto panels, dry nitrogen gas to reduce radon and low background shielding results in a reduction of the background count rate by about a factor of 100 compared to detectors operating above ground at our laboratory.
1. Introduction The International Monitoring System (IMS) of the Comprehensive Nuclear Test Ban Treaty Organization (CTBTO) consists of a network of 80 field stations and 16 laboratories dedicated to the detection and analysis of airborne particulates to detect possible nuclear events that could indicate a treaty violation. USL16 at PNNL (Pacific Northwest National Laboratory) was certified by the IMS in 2007 to perform more detailed analyses of air particulate samples of interest collected by the field stations. Air particulate samples are solely analyzed by gamma-ray spectrometry, although the IMS also includes many other techniques for detecting treaty violations including xenon gas, seismic monitors, and infrasound monitors. The gamma-ray detectors used by the IMS network are mainly conventional, off-the-shelf HPGe (high purity germanium detectors). PNNL has developed two very low-background detectors located in a shallow underground laboratory to significantly improve the presence of weak activity concentrations of gammaemitting fission or activation products. 2. Detector design PNNL constructed a shallow underground laboratory to allow the development of low-background materials and for the deployment of very low-background radiation detectors (Aalseth et al., 2012; Forrester et al., 2013). The new laboratory is about 15 m underground and includes an overburden on top to achieve a background of about 30 m water equivalent. The ambient fast neutron flux is reduced about a factor of 100 relative to above ground locations. The section of the
⁎
underground laboratory used for low-level radiation detection operates as a class 10,000 clean room. Other parts of the laboratory operate at the class 1000 level. The gamma-ray detectors were custom ordered from Canberra, specifying very low background materials for all detector components, as supplied by Canberra, to reduce the presence of naturally occurring radioactive materials. One of the detectors is a Ptype crystal, with a relative efficiency of 112% and the second detector is n-type, with a relative efficiency of 97%. Both detectors have carbon fiber windows and copper end caps. The radiation shields were custom designed consisting of 20 cm of low background lead lined with 5 cm of oxygen-free high thermal conductivity (OFHC) copper. Whereas copper is typically used in very thin layers in above ground detectors to avoid neutron activation of the copper, that effect is greatly reduced in our underground laboratories due to the factor of 100 reduction of the neutron background. Five large cosmic veto panels surround the detectors and shields to suppress cosmic ray induced events. Some of the cosmic veto panels are from Eljen and are made of EJ-200 plastic scintillator wrapped with reflector foil in light tight boxes. The rest of the panels are from St. Gobain using BC-408 plastic scintillators. The entire assemblies are enclosed in a glove box that has a high flow of dry nitrogen gas from the boil off of a large liquid nitrogen tank outside the building. Samples are introduced and removed from the detector using a pass through that is flushed with nitrogen prior to opening to the detector enclosure. The cosmic veto nearly eliminates the 511 keV peak; the dry nitrogen flow greatly reduces the radon peaks; and the low background detector materials nearly eliminate naturally occurring peaks such as the 1460 keV peak from 40K. The sample holders for the IMS samples were designed and
Corresponding author. E-mail address: [email protected] (L.R. Greenwood).
http://dx.doi.org/10.1016/j.apradiso.2016.12.034 Received 28 July 2016; Received in revised form 29 November 2016; Accepted 20 December 2016 0969-8043/ © 2016 Elsevier Ltd. All rights reserved.
Please cite this article as: Greenwood, L.R., Applied Radiation and Isotopes (2016), http://dx.doi.org/10.1016/j.apradiso.2016.12.034
Applied Radiation and Isotopes (xxxx) xxxx–xxxx
L.R. Greenwood et al.
approximately a factor of 100 lower than normally seen at our site for commercial off-the-shelf (COTS) HPGe (high-purity germanium) detectors using a commercial shield with normal lead (20 cm thick) in an above ground location. The finished detector assemblies are shown in Fig. 3. Samples received from IMS stations are Radionuclide Aerosol Sample Analyzer (RASA) samples, compressed using our 30 t press, or Snow White and Cinderella air particulate samples which are counted as received since these samples are compressed at the field stations. The outer plastic sample holders are cleaned to remove any possible surface contamination and then bagged in air tight bags before taking them down to the underground laboratory. All personnel don protective clothing before entering the facility where the detectors are located. Samples are introduced to the sealed glove box containing the detector and shield using a pass through. The outer door is opened to introduce the sample. After closing the outer door, the pass though is flushed with dry nitrogen for several minutes. The inner door is then opened using the glove box gloves to reach the sample. The sample is then put into one of the ABS sample holders and carefully placed on the detector such that nothing touches the detector surface. After counting, the sample is removed in the reverse procedure to minimize any leakage of outside air into the inner glove box during sample insertion or removal. The background count rate for our p-type detector is typically 0.047 cps (counts per second) or 1840 counts per day per kg of Ge over the range from 40 to 2700 keV. The weight of the germanium crystal is 2.17 kg. Due to the carbon fiber window and thin Ge deadlayer, the detector has a relatively high efficiency at lower gamma-ray energies compared to a typical commercial HPGe detector. The minimum detectable activity (MDA) of Ba-140 in a 7-d count is about 3 mBq, a factor of 8 lower than the required value of 24 mBq for IMS detectors. The MDA is calculated according to the IMS requirement where the MDA is equal to the lower limit of detection at the 95% confidence level divided by the count time, gamma-ray efficiency, correction for true coincident summing, gamma-ray emission probability, and decay during counting. The LD value is calculated according to the formula:
Fig. 1. The copper-lined detector chamber is shown with the sliding top door open. Some of the cosmic veto detector panels are shown on the left side. The detector face can just be seen at the bottom of the copper section.
fabricated with a 3D printer using acrylonitrile butadiene styrene (ABS) plastic, which was found to be very low-background when screened by gamma-ray counting. The detector shields are shown in Fig. 1. The design of the low background detectors and shields was inspired by prior work at our laboratory and other low background detector systems including Baudis et al. (2011), Neder et al. (2000), and Keillor et al. (2009). These new low-background detectors were designed as a more cost-effective way of expanding our IMS laboratory compared to the PNNL-fabricated CASCADES detector (Keillor et al. (2009)). 3. Detector operation and performance The detectors are operated using Canberra Lynx digital analyzers. A pulse seen in any of the cosmic detector panels is used to veto any signals from the germanium detectors. This results in a significant reduction in cosmic events registered by the detectors, as seen in the about 98% reduction of the 511 keV background peak. The use of low background materials for the detector and shield resulted in about a 97% reduction in the 1460 keV peak from naturally occurring 40K. Typical background spectra are shown in Fig. 2. The top curve is for a COTS detector with 114% relative efficiency in an above ground laboratory. The bottom two curves show empty cave (no sample inserted) gamma-ray spectra measured on our low-background detectors located in the underground laboratory and with the addition of the cosmic veto and radon reduction systems. The net background is
LD = 2.71 + (B kg)1/2
(1)
where Bkg=background counts in the interval of ± 1.25 FWHM. The n-type detector has a high efficiency at very low gamma-ray energies such that cascade summing with x-rays is seen routinely for many gamma-ray peaks. To reduce this effect, a steel absorber with a thickness of 0.13 mm (5 mils) was inserted over the face of the detector. Partly due to the much higher efficiency at lower gamma-ray energies, the background count rate is about twice that of the p-type detector at about 0.088 cps or 4000 counts per day per kg of Ge (1.95 kg) for the energy range of 40–2900 keV.
Fig. 2. Comparison of HPGe spectra for a COTS system in an above ground laboratory (top) and the n-type and p-type low-background detectors in the underground laboratory with cosmic veto panels and radon reduction using dry nitrogen gas.
2
Applied Radiation and Isotopes (xxxx) xxxx–xxxx
L.R. Greenwood et al.
Fig. 3. Final assembly of the new detectors in our underground laboratory. Note the sample pass through, glove ports, wheel to slide open the top of the lead cave, and dewar located outside the enclosure.
Technology program (DTRA 10027-16504) of the Defense Threat Reduction Agency. Pacific Northwest National Laboratory is operated for DOE by Battelle Memorial Institute under Contract DE-AC06-76RLO 1830.
The background count rates that we have obtained with the low background detectors in our underground laboratory are comparable to what has been seen at other underground laboratories using similar systems shown by Baudis et al. (2011), Neder et al. (2000), Keillor et al., 2009, and Povinec et al. (2004). For example, the IAEA-MEL (Povinec et al. (2004)) detectors at a comparable depth underground (35 m w.e.) have count rates of 35–68 counts/hour-kg over the range of 40–2700 keV compared to our background rates of 77 counts/hour-kg for our p-type detector and 166 counts/hour-kg for our n-type detector. We are continuing to look at techniques to further reduce the background. The use of cosmic veto detectors was found to be especially effective for background reduction, somewhat offsetting the additional benefit of using underground laboratories at a greater depth. Experiments are currently in progress with the use of cosmic veto panels surrounding above ground gamma-ray detectors to determine what performance can be achieved.
References Aalseth, C.E., et al., 2012. A shallow underground laboratory for low-background radiation measurements and materials development. Rev. Sci. Inst. 83 (11). http:// dx.doi.org/10.1063/1.4761923. Baudis, L., et al., 2011. Gator: a low-background counting facility at the gran sasso underground laboratory. J. Inst. 6, P08010. http://dx.doi.org/10.1088/1748-0221/ 6/08/P08010. Forrester, J.B., et al., 2013. Construction of a shallow underground low-background detector for a CTBT radionuclide laboratory. J. Radioanal. Nucl. Chem. 296 (2), 1061–1064. http://dx.doi.org/10.1007/s10967-012-2202-3. Keillor, M.E., et al., 2009. Design and construction of an ultra-low-background 14-crystal germanium array for high efficiency and coincidence measurements. J. Radioanal. Nucl. Chem. 3 (282), 703–708. http://dx.doi.org/10.1007/s10967-009-0248-7. Neder, H., et al., 2000. Low level g-ray germanium-spectrometer to measure very low primordial radionuclide concentrations.. Appl. Radiat. Isot. 53, 191–195, (PII: S0969-8043(00)(00132-9). Povinec, P.P., et al., 2004. IAEA-MEL's underground counting laboratory in Monaco – background characteristics of HPGe detectors with anti-cosmic shielding. Appl. Radiat. Isot. 61, 85–93. http://dx.doi.org/10.1016/j.apradiso.2004.03.019.
Acknowledgements This research was sponsored by the Office of Nuclear Verification, National Nuclear Security Administration. The authors wish to acknowledge the funding support of the Nuclear Arms Control
3