The high neutron detection efficiency, good gamma-ray discrimination and non-toxicity of 3He made of proportional counters filled with this gas the obvious choice for neutron detection, particularly in radiation portal monitors (RPM), used to control the illicit transport of nuclear material, of which neutron detectors are key components. 3He is very rare and during the last decade this gas has become increasingly difficult to acquire. With the exception of BF3, which is toxic, no other gas can be used for neutron detection in proportional counters. We present an alternative where the 3He atoms are replaced by nanoparticles made of another neutron sensitive material, 10B. The particles are dispersed in a gaseous volume, forming an aerosol with neutron sensitive properties. A proportional counter filled with such aerosol was exposed to a thermal neutron beam and the recorded response indicates that the neutrons have interacted with the particles in the aerosol. This original technique, which transforms a standard proportional gas mixture into a neutron sensitive aerosol, is a breakthrough in the field of radiation detection and has the potential to become an alternative to the use of 3He in proportional counters.
Proportional counters filled with 3He are the golden standard for thermal neutron detection. This gas has high neutron detection efficiency, good gamma-ray discrimination, is non-toxic and, until recently, was an accessible material1,2,3. These characteristics made of 3He proportional counters the obvious choice for neutron detection, particularly in radiation portal monitors (RPM), used to control the illicit transport of nuclear material, of which thermal neutron detectors are key components. 3He is, however, very rare and mostly a by-product of nuclear warhead production. With the large deployment of RPM in the last decade this gas has become increasingly difficult to acquire, with skyrocketing prices1,2. Besides domestic security, other applications where 3He is used, such as large area detectors in neutron research facilities, nuclear reactor control, cryogenic systems, well logging and medical imaging have also suffered from the reduced availability of this material4. With the exception of BF3, which is toxic and not suitable for applications outside the laboratory, no other gas exists that can be used for neutron detection in gaseous proportional counters.
Thermal neutron (n) detection in gaseous proportional counters is achieved via reactions of the type:
Due to momentum conservation, the products are emitted in opposite directions, each carrying a fraction of the total energy released (Q). The tagging of the thermal neutron is done by measuring the energy of the products released in the above reactions. These will travel in the proportional gas, ionizing its molecules and creating a track of ion-electron pairs, which number is proportional to the energy deposited in the gas. An electric field established inside the proportional counter multiplies the primary-electrons, being the resulting charge collected at an appropriate electrode. In 3He proportional counters another filling gas can be added to facilitate the collection of the reaction products energy: 3He acts as the stopping medium for the thermal neutrons via reaction (1) while the tritium and the proton are stopped by the filling gas, with higher stopping power than 3He. In large detectors both products of (1) fully deposit their energy in the proportional gas and the amplitude of the output pulses follows a Gaussian distribution with an average value proportional to Q (Fig. 1)5.
One of the alternatives to 3He is the Boron-lined proportional counter6,7,8 where the neutron sensitive material, 10B, is present in a thin layer on the inner wall of the proportional counter, making it sensitive to neutrons. The α-particle and the 7Li atom released in (2) are emitted from the boron layer in opposite directions. Due to this geometrical constrain only one of these species is, at best, emitted in the direction of the proportional gas, preventing the full collection of Q (Fig. 2).
While travelling inside the 10B layer, the α-particle and the 7Li lose a fraction of their initial energy, which is not collected in the proportional gas. As a consequence the output pulses of these detectors do not follow a Gaussian distribution as in 3He or BF3 proportional counters5,9. Instead, their pulse-height distributions are the sum of two flat distributions, each corresponding to one of the products released in (2). The result is a characteristic 2-step plateau distribution8,9,10,11 with maximal energy of 1.47 MeV and extending down to 0 (Fig. 2). This feature may represent a drawback for applications where a strong, low-energy gamma-ray background (from which the neutron must be discriminated) is present6. Another drawback of the neutron conversion in a 10B layer is that increasing the layer thickness only increases the detection efficiency up to a certain value: after some depth of interaction the α-particle and 7Li lose all their energy inside the boron layer, leaving the neutron un-accounted. In practice, the maximum detection efficiency achievable with a boron-lined proportional counter is roughly 12%10.
In this work we propose an alternative which combines the larger energy released in the 10B reaction with the full energy collection of 3He proportional counters. For that we have dispersed B4C nanoparticles in a proportional counter gas. The nanoparticles have dimensions smaller than the ranges of the α-particle and the 7Li in B4C, 3.6 and 1.8 μm, respectively12. In this situation an incoming neutron is converted inside a nanoparticle and both products of reaction (2) escape, depositing a large fraction of their energy in the proportional gas. Comparing to the case of a proportional counter filled with 3He and another gas, this solution corresponds to replacing the 3He atoms by B4C nanoparticles. The presence of the particles in the gas should not affect the properties of the proportional counter, either causing discharges or modifying its energy resolution. Under these conditions the detector response will present a full energy peak similar to the one of 3He proportional counters, but located at 2.3 MeV instead of 0.7 MeV.
Prior to the experimental developments, we have performed a Monte Carlo simulation, comprising a single B4C particle inside a proportional counter filled with Ar-CH4 (90–10%). The simulation was implemented using GEANT413 and 2 particle positions were studied: near the centre of the proportional counter (Fig. 3–Top panel) and in direct contact with the proportional counter walls (Fig. 3-Bottom panel). For each position, the particle radius was changed and the energy deposited in the proportional gas was recorded. The results show that, in the case of the particle not in contact with the detector walls (Fig. 3–Top panel) and for particle radii of 1 μm or lower, the response is indeed a single peak. For particles with a radius of 5 μm (Fig. 3–Top panel) or attached to the detector walls (Fig. 3-Bottom panel), the output is similar to the one of boron-lined proportional counters with maximal energy deposition of 1.47 MeV (for the 94% preferred branch in reaction 2). This is either due to self-absorption inside a larger particle (5 μm radius, Fig. 3–Top panel) or due to one of the secondary products being emitted onto the detector wall (particles attached to the walls, irrespective of their radius, Fig. 3-Bottom panel).
In order to evaluate the practicality of the proposed technique we have developed a test apparatus (methods section) composed of a proportional counter, filled with Ar-CH4 (90–10%), in which we have dispersed B4C nano-particles. During detector operation the gas circulates from the bottom to the top of the proportional counter, supporting the dispersion of the nanoparticles. A potential unwanted contribution to detector response could arise from nanoparticles attached to the walls of the proportional counter. In order to minimize such contribution we have equipped it with a set of electrodes (electric field gate), installed in the vicinity of its wall and surrounding the anode wire. This gate (methods section) allows establishing a reversed electric field in the region near the wall of the proportional chamber, preventing neutron interactions in nanoparticles attached to it from being detected.
Figure 4 presents the results of the irradiation of the prototype to a cold neutron beam line (Narzsiss Instrument, Paul Scherrer Institute, Switzerland), for several gate polarizations. Beside the 2-step plateau, limited by the 7Li and α-particle edges, the pulse-height distributions of Fig. 4 also present a single peak, with an energy corresponding to the sum of the α-particle and 7Li edges, representing full energy collection in the proportional gas.
For VGATE = +150 V in Fig. 4, there is no blocking effect by the gate and the α-particle and 7Li edges are clearly identifiable in the pulse-height distributions. These features and the full-energy peak have been used for energy calibration of the pulse-height distributions of Fig. 4. As VGATE is decreased to values below the ground potential, a reverse electric field is established in the region between the gate and the wall, which partially blocks interactions from material on the walls. This increases the relevance of the full-energy peak in the pulse-height distributions of Fig. 4. Nevertheless, for VGATE = −50 V, a residual 2-step plateau is present in the pulse-height distributions. This may be due to attachment of nanoparticles to the electrodes of the gate, to residual interactions on the walls or to interactions near the gate.
The response of the detector to irradiation by the neutron beam indicates that some of the nanoparticles were dispersed in the proportional gas, resulting in full collection of the energy released in the neutron capture reaction. Such feature was, until now, only possible using either 3He or BF3 proportional counters. While the former has become increasingly unavailable, the latter is toxic and not suitable for applications outside a controlled environment. In addition, the energy emitted in the neutron reaction capture in 10B is higher (2.3 MeV) than that released with 3He (0.7 MeV), improving gamma ray discrimination.
A detection efficiency of 4% was achieved for VGATE = −50 V. This figure is related with the total volume of the aerosol, the nanoparticles concentration and their composition. All these parameters can, in principle, be improved. In particular, the use of 10B enriched nanoparticles will increase the detection efficiency in the same proportion as the enrichment ratio. The electric field gate was successful in improving the pulse-height distributions; however, a technical challenge remains in decreasing the attachment of particles to the chamber wall and further increasing the full-energy peak relevance. Fluids other than gases, such as room temperature14 or cryogenic liquids15,16, are also candidates to disperse the nanoparticles.
This novel technique uses a non-toxic and abundant material and transforms a regular proportional counter gas into a neutron sensitive aerosol. The type of response recorded is an improvement over state-of-the-art-alternatives6,7,8,10,11 with events in which the total energy released in the neutron capture reaction is collected. The detection of neutrons with nanoparticles immersed in a gas constitutes a breakthrough in the field of radiation detection and has the potential to become an alternative for neutron detection.
The main component of the experimental setup is a custom-made gaseous proportional counter (Fig. 5). It is composed of a stainless steel cylinder with an inner diameter of 47.6 mm and a 50 μm diameter tungsten wire stretched along its axis. The tungsten wire (anode) is electrically insulated from the outer cylinder (cathode), which is connected to ground potential. One end of the anode wire is soldered to an SHV feedthrough while the other end is glued to a ceramic insulator on the bottom fixing ring. The bottom and top fixing rings are supported by 3 ceramic pillars, glued to the top flange. When the anode is biased with a positive voltage (1850 V were used throughout the measurements presented in this work), a radial electric field is established inside the chamber. This electric field is responsible for collecting the primary electrons converted by the ionizing radiation in the proportional gas and for drifting them towards the anode. Once the primary electrons reach a small region in the vicinity of the anode, where the electric field is higher than the ionization threshold of the gas, they undergo a proportional multiplication process by impact ionization with the gas molecules. Multiplication factors above 103 are commonly achieved in proportional counters17. In our apparatus the resulting electron charge was collected at the anode with a charge-sensitive pre-amplifier (Canberra 2006) which fed a linear amplifier (Ortec 454). A multichannel analyser digitized the output pulses from the amplifier.
During detector operation a mixture of Ar-CH4 (90–10%) was continuously flowing throughout the main chamber at a rate of 4 l/h, supplied from an adjacent pressurized canister. Gas admission into the chamber was achieved through a ¼ inch entrance located at the bottom of the chamber (gas inlet in Fig. 5). The consumed gas leaved the chamber through a ¼ inch outlet, located at the top of the chamber. This outlet was equipped with a 2-μm filter (Swagelok NI-4-VCR-2-GR-2M) and was insulated from the outer atmosphere by a gas bubbler filled with vacuum pump oil.
The stainless steel chamber was placed vertically and the particles to be dispersed were loaded into the bottom flange, immediately after the gas inlet. Prior to detector operation, the proportional gas mixture was first allowed to flow for one hour at a rate of 0.8 l/h, to ensure that the atmospheric air was removed from the inside of the stainless steel cylinder. Afterwards, with the purpose of facilitating particle dispersion, the flow was increased for a short period to values above 8 l/h. Data taking was done at a constant flow of 4 l/h.
The apparatus was installed in the Swiss Spallation Neutron Source (SINQ) at the Narziss beam line, which provides a continuous beam of cold neutrons (5 Å). The neutron beam, with vertical and horizontal dimensions of 50 mm and 10 mm, respectively, irradiated the apparatus in the centre of the stainless steel cylinder, in a direction perpendicular to its axis, at equal distances from the top and bottom flanges. With such beam collimator settings the beamline intensity was of 7 × 103 neutrons per second. A beam blocker in the neutron beam line allowed turning on and off the neutron beam.
Uncoated B4C nanoparticles (PlasmaChem GmbH, B4C Nanopowder) were used for the measurements presented in this paper. A sample was dispersed in ethanol and its particle size distribution measured by laser diffraction (Beckman Coulter LS 13 320). A mean value of 1.056 μm with d10 = 0.553 μm, d50 = 1.029 μm and d90 = 1.602 μm was measured. During data taking the detector was stable, without the occurrence of any electrical discharge or large gain fluctuations.
Eighteen copper wires (0.8 mm in diameter) were installed inside the stainless steel chamber, at a distance of 10 mm from the walls and in a direction parallel to the anode wire (Fig. 5, right). The polarization of this set of electrodes (electrostatic gate) with a negative potential established an electric field between the gate and the wall of the stainless steel chamber. Such electric field configuration should prevent the neutron conversions taking place in the material attached to (or near) the wall to be detected. The range in Ar-CH4 (90–10%) of the α-particle emitted in reaction (2) was calculated to be 8.7 mm18. Taking this information into account, the set of electrodes was placed at a distance of 10 mm from the walls. The gate blocking efficiency to neutron conversions taking place next to the walls of our chamber was determined in a former experimental measurement. For that, a thin slab of boron silicone was attached to the inner walls of the chamber. The detector was exposed to the Narziss beam line and the pulse-height distributions were recorded for several values of VGATE. For VGATE = −50V the blocking efficiency (relative to VGATE = 150 V) was measured, being 62%.
The Monte Carlo simulation (Fig. 4) was implemented with the GEANT4 software package13, version 9.5. With an extensive set of physics models and cross section database, this tool allows handling the interactions of particles with matter across a very wide energy range being used to accurately simulate the passage of particles through matter. For the simulation a standard physics list (EmStandard PhysicsList) was used and the main physical components of the experimental apparatus was reproduced, simulating its response to thermal neutrons, for different positions and diameters of a single B4C particle. A total of 107 thermal neutrons were generated for each simulation, emitted in the direction of the particle, and uniformly distributed over its cross section. For each event, the simulation records the energy deposited in the proportional counter gas.
How to cite this article: Amaro, F. D. et al. Novel concept for neutron detection: proportional counter filled with 10B nanoparticle aerosol. Sci. Rep. 7, 41699; doi: 10.1038/srep41699 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
U.S. House of Representatives Committee on Science & Technology. Caught by surprise: causes and consequences of the helium-3 supply crisis. Hearing Charter https://www.gpo.gov/fdsys/pkg/CHRG-111hhrg57170/html/CHRG-111hhrg57170.htm (Date of access: 03/12/2016) (2010).
United States Government Accountability Office. Neutron detectors alternatives to using helium-3. Report to Congressional Requesters http://gao.gov/assets/590/585514.pdf (Date of access: 03/12/2016) (2011).
Kouzes, R. T. The 3He supply problem. Pacific Northwest National Laboratory http://www.pnl.gov/main/publications/external/technical_reports/PNNL-18388.pdf (Date of access: 03/12/2016) (2009).
Sacchetti, F. et al. He-free neutron detectors and their applications. Eur. Phys. J. Plus 130, 53 (2015).
Mazed, D., Mameri, S. & Ciolini, R. Design parameters and technology optimization of 3He-filled proportional counters for thermal neutron detection and spectrometry applications. Radiation Measurements 47, 577–587 (2012).
Kouzes, R. T. et al. Neutron detection alternatives to 3He for national security applications. Nuclear Instruments and Methods in Physics Research A 623, 1035–1045 (2010).
Peerani, P. et al. Testing on novel neutron detectors as alternative to 3He for security applications. Nuclear Instruments and Methods in Physics Research A 696, 110–120 (2012).
Tsorbatzoglou, K. & McKeag, R. D. Novel and efficient 10B lined tubelet detector as a replacement for 3He neutron proportional counters. Nuclear Instruments and Methods in Physics Research A 652, 381–383 (2011).
Kouzes, R. T., Ely, J. H., Lintereur, A. T., Siciliano & E. R. & Woodring, M. L. BF3 neutron detector tests. Pacific Northwest National Laboratory http://www.pnl.gov/main/publications/external/technical_reports/PNNL-19050.pdf (Date of access: 03/12/2016) (2009).
McKinny, K. S., Anderson, T. R. & Johnson, N. H. Optimization of coating in boron-10 lined proportional counters. IEEE Transactions on Nuclear Science 60–2, 860–863 (2013).
Nelson, K. A., Edwards, N. S., Hinson, N. J., Wayant, C. D. & McGregor, D. S. A suspended boron foil multi-wire proportional counter neutron detector. Nuclear Instruments and Methods in Physics Research A 767, 14–18 (2014).
Ziegler, F. Z., Ziegler M. D. & Biersack, J. P. SRIM - the stopping and range of ions in matter (2010). Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 268, 1818–1823 (2010).
Allison, J. et al. Geant4 developments and applications. IEEE Transactions on Nuclear Science 53, 270–278 (2006).
Engler, J. Liquid ionization chambers at room temperatures. J. Phys. G: Nucl. Part. Phys. 22, 1–23 (1996).
XENON Collaboration . http://xenon.astro.columbia.edu/ (Date of access: 03/12/2016) (2011).
LUX Dark Matter . http://luxdarkmatter.org/ Date of access: 03/12/2016) (2003).
Knoll, G. F. Radiation Detection and Measurement, 4th edition, (Wiley, 2010).
Berger, M. J., Coursey, J. S., Zucker, M. A. & Chang, J. Stopping-power & range tables for electrons, protons, and helium ions. http://www.nist.gov/pml/data/star/ (Date of access: 03/12/2016) (2009).
F.D.A. and C.M.B.M. acknowledge the support of FCT, under contracts SFRH/BPD/74775/2010 and SFRH/BPD/76842/2011, respectively. We acknowledge R. Ferreira (UCQfarma), J. Catita and J. Carvalheira (Paralab), for their assistance with the particle size distribution measurement. The measurements were performed at the Swiss Spallation Neutron Source SINQ at the Paul Scherrer Institute. We acknowledge C. Schanzer of the SwissNeutronics AG and M. Horisberger for access to the neutron beam.
The authors declare no competing financial interests.
About this article
Cite this article
Amaro, F., Monteiro, C., dos Santos, J. et al. Novel concept for neutron detection: proportional counter filled with 10B nanoparticle aerosol. Sci Rep 7, 41699 (2017). https://doi.org/10.1038/srep41699