- Email: [email protected]
Transparent lithium loaded plastic scintillators for thermal neutron detection
Nuclear Instruments and Methods in Physics Research A 701 (2013) 58–61
Contents lists available at SciVerse ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Transparent lithium loaded plastic scintillators for thermal neutron detection R.D. Breukers a,n, C.M. Bartle b, A. Edgar c a b c
Industrial Research Limited, 69 Gracefield Road, PO Box 31310, Lower Hutt, New Zealand GNS Science Ltd, 30 Gracefield Road, PO Box 31312, Lower Hutt, New Zealand School of Chemical and Physical Sciences, Victoria University of Wellington, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 September 2012 Received in revised form 24 October 2012 Accepted 25 October 2012 Available online 1 November 2012
The fabrication of a series of novel, optically transparent, bulk plastic scintillators loaded with lithium methacrylate, and incorporating 2,5-diphenyloxazole and 5-phenyl-2-[4-(5-phenyl-1,3-oxazol-2yl)phenyl]-1,3-oxazole fluorescent centres, is described. The attenuation length, photoluminescence, and both gamma ray and thermal neutron scintillation responses were compared over a range of lithium methacrylate concentrations. The maximum concentration corresponded to a weight percentage of lithium-6 of 0.63%. The photoluminescence shows a composite 2,5-diphenyloxazole and 5-phenyl-2-[4-(5-phenyl-1,3-oxazol-2-yl)phenyl]-1,3-oxazole broad band with vibronic features in the range 350–500 nm, and lifetimes in the range 0.9–2.7 ns. An increasing luminescence in a thermal neutron beam with increasing lithium-6 content is demonstrated. & 2012 Elsevier B.V. All rights reserved.
Keywords: Plastic scintillator Lithium loaded Thermal neutron detection
1. Introduction Due to their comparatively low cost, ability to be machined into complex shapes and fast response times, plastic scintillators have found application in a wide variety of photon and particle detection applications [1]. Therefore, methods to fabricate these scintillators, especially loading with other elements that enhance their performance are of considerable interest. For example, low resolution plastic scintillators loaded with bismuth or lead complement higher cost medium resolution lanthanum halide and sodium iodide detectors [2]. Commercially available scintillators loaded with boron-10, gadolinium, lead and lithium-6 to increase the thermal neutron detection efficiency are available [3]. However, many of these commercially available scintillators are either liquids [4] or glasses [5] which may not offer all of the advantages of plastic scintillators. Plastic scintillators have very fast decay times (e.g. 5 ns) and thus exhibit high count-rate capabilities (e.g. 106 per second) [6]. The process of loading plastic scintillators with specific elements is limited by a reduced light output that often limits the elemental concentrations that can be achieved. Although lithium-6 has a high thermal neutron absorption cross-section, it has proven difficult to load into plastic scintillators. It is an objective of this paper to describe the performance of transparent polystyrene-based plastic scintillators loaded with
n
Corresponding author. Tel.: þ64 4 931 3539. E-mail addresses: [email protected], [email protected] (R.D. Breukers). 0168-9002/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.nima.2012.10.080
lithium-6. Currently lithium-6 containing scintillators are commonly glasses [5] or inorganic crystals and lithium compounds dispersed in phosphor powders [3]. The topic of transparent plastic bulk scintillators containing lithium is relatively unexplored; previous reports of the incorporation of lithium into plastic organic scintillators mainly focus on the mixing of inorganic salts and nanoparticles such as lithium fluoride [7,8], lithium chloride [9] or lithium phosphate [10] into a polymer which results in a heterogeneous material with reduced transparency. A paper relating to a transparent scintillator containing lithium by Negina et al. [11] gives no details of the scintillator’s preparation and composition and is therefore difficult to compare to this current work. Thus there have been only a few reports of promising lithiumloaded plastic scintillators for thermal neutron detection. Yet achieving loading with lithium-6 may offer advantages over boron-10 loaded scintillators, as discussed in a recent report on lithium loaded liquid scintillators by Bass et al. [12]. For example, a feature of the lithium-6 reaction is its comparatively high Q-value (4.78 MeV compared with 2.79 MeV for the boron-10 reaction). Unlike the boron-10 reaction, there is an absence of gamma emissions associated with the charged particle reaction products. The products of both reactions are otherwise short range energetic charged particles, enabling use of small detectors without significant edge effects [2]. In the case of thermal neutron absorption by lithium-6 the product triton and alpha particle are emitted with energies of 2.73 MeV and 2.05 MeV respectively, for a combined deposited energy of 4.78 MeV. In the case of the boron-10 reaction, the product lithium-7 nucleus is produced 94% of the time in an excited state emitting a 478 keV gamma ray
R.D. Breukers et al. / Nuclear Instruments and Methods in Physics Research A 701 (2013) 58–61
which often escapes the detector and is thus not detected but may contaminate the detection environment. Without detection of the gamma ray, the combined particle detected energy from the alpha particle (1.47 MeV) and the lithium ion (0.84 MeV) is 2.31 MeV and is thus considerably lower than 4.78 MeV obtained from the lithium-6 reaction. However, in practice the performance of the scintillator depends on the cross section for the neutron absorbing reaction, the percentage elemental loading that can be achieved, and the comparative light outputs of lithium loaded and boron loaded plastic scintillators. Some light absorption in lithium loading of organic scintillators at 0.15 wt% level and a relatively low neutron detection efficiency compared to boron loaded organic scintillators have been noted [2]. Historically usage has strongly favoured the boron loaded scintillators which are commercially available from Saint Gobain and Eljen Technology [3]. Improvement of the lithium loaded plastic scintillators may lead to a commercially competitive product whereby the higher reaction Q-value and the absence of gamma emission may be utilised. Here we have studied the incorporation of lithium-6 into bulk plastic polystyrene scintillators using new lithium containing compounds to give transparent scintillators. We report a method of preparing scintillators containing lithium which utilises a solution of a lithium complex which fully dissolves in the precursor monomer solution. This mixture can then be polymerised giving a transparent scintillator. Our investigation clearly shows an increase in light output with increases in the lithium-6 concentration when the scintillator is placed in a thermal neutron beam.
2. Preparation of scintillators The main goal of our research was to produce a transparent plastic scintillator loaded with lithium to act as a detector of thermal neutrons. Upon designing this scintillator we needed to incorporate the components of a conventional plastic scintillator into our material so that the products of the lithium neutron capture reaction could be converted to light, which is able to be detected by a photomultiplier tube. Styrene was chosen as the aromatic base material for our scintillator as it is cheap and has been extensively studied, as have the primary and secondary shifters 2,5-diphenyloxazole (PPO) and 5-phenyl-2-[4-(5-phenyl1,3-oxazol-2-yl)phenyl]-1,3-oxazole (POPOP) [13]. To produce our plastic scintillators we polymerised the styrene solution using the free radical initiator azobisisobutyronitrile (AIBN). A range of lithium containing compounds was trialled for solubility in the polymerisation solution. Lithium methacrylate (LiME) was investigated as the methacrylate ligand has a vinyl group which is complementary to the vinyl groups in styrene and allows the compound to be incorporated into the polymer. LiME is sparingly soluble in the styrene monomer and copolymers have been previously reported [14]. However the application of these copolymers to scintillator applications has not been investigated. To increase the lithium loading in the polymerisation solution a mixture of methacrylic acid and styrene was used. A range of scintillators with LiME concentrations varying from 1.25 to 10 wt% were fabricated using both natural lithium and enriched lithium-6. The percentage of lithium-6 by weight in 95% enriched 6Li-ME is 6.3% thus in a 4 g scintillator containing 0.4 g LiME the lithium-6 content is 0.63 wt%. The calculated lithium-6 weight percentages for both the natural lithium (7.5% lithium-6 by weight) scintillator series and the enriched scintillator series are shown in Table 1. The following is an example of the method used to fabricate the scintillators. LiME (0.4 g) was dissolved in methacrylic acid
59
Table 1 LiME content, lithium-6 percentage by weight and attenuation length of the bulk scintillator samples.
Natural (7.5 wt% Li-6)
Enriched (95 wt% Li-6)
LiME wt %
Li-6wt %
Attenuation length (mm)
1.25 2.5 5 7.5 10 1.25 2.5 5 7.5 10
0.007 0.014 0.028 0.042 0.057 0.08 0.16 0.31 0.47 0.63
41 25 32 30 27 22 31 37 41 34
Fig. 1. Photo of scintillator samples, 1.25 wt% enriched LiME (left) and 10 wt% enriched LiME (right).
(1.6 g) and this solution was then mixed with a solution of PPO (20 mg) and POPOP (2 mg) in styrene (2 g). AIBN (2 mg) was then added to this solution and the mixture sealed in a glass ampoule under argon then incubated at 50 1C to facilitate polymerisation. The scintillators fabricated by this method were 20 mm diameter, 10 mm long, transparent solid plastic bullets; however a range of sizes and shapes could be produced. One end face of each scintillator was machined flat to match the photocathode of a photomultiplier tube. The attenuation length of the flattened scintillator samples (Table 1) was measured using an Ocean Optics ST 2000 fibre-optic spectrometer. The samples were all transparent as can be seen in Fig. 1, with an average attenuation length of 32 mm for light at 429 nm the emission wavelength of POPOP (Fig. 2).
3. Luminescence measurements 3.1. Photoluminescence Photoluminescence (PL) spectra were recorded on a Fluorolog 3 fluorometer at room temperature equipped with a 450W Xe arc lamp source. Typical spectra for fluorescence and excitation are shown in Fig. 2. The fluorescence spectra show clear vibronic features which assist in attributing the spectra to two principal bands, assigned to PPO and PPO-pumped POPOP as shown. However a third weak band in the red region of the spectrum remains unattributed, and possibly arises from the polymerising agent. The excitation spectra at wavelengths characteristic of PPO and POPOP are very similar, supporting the idea of the PPO acting to pump emission from the POPOP dye in a wavelength shifting
60
R.D. Breukers et al. / Nuclear Instruments and Methods in Physics Research A 701 (2013) 58–61
Philips XP2262B photomultiplier tube located within a light-tight housing. During measurement the photomultiplier tube axis was vertical thus supporting the scintillator on the horizontal photocathode surface of the photomultiplier. A small amount of optical grease was used to couple the scintillator to the photomultiplier. The photomultiplier signals were amplified and shaped with an ORTEC 113 preamplifier and a 671 main amplifier. The pulse height signals were recorded for each source/sample combination using an ORTEC Easy-MCA. The end point of the Compton spectrum was recorded with each source and each scintillator. The average ‘enriched to natural’ ratio of responses of the correspondingly LiME loaded scintillators was within 1.07 70.05 irrespective of the gamma source used. Luminescence differences were thus small and consistent. 3.3. Thermal neutron luminescence measurements
Fig. 2. Photoluminescence spectra (solid line) of a sample containing 1.25 wt% of enriched LiME when pumped at 260 nm. The dotted, dashed, and long dashed lines show the excitation spectra for PL at 364 nm (PPO), 429 nm (POPOP) and 589 nm (unidentified) respectively.
Table 2 Measured PL lifetimes for the sample containing 1.25 wt% of enriched LiME. For the 429 nm emission, there is a minor slow component which accounts for 4% of the total light emission. Emission wavelength (nm)
s1(ns)
364 429 429 589
0.97 0.1 1.7 7 0.1 1850 7 60 2.4 7 .03
operation. The character of the spectra shows some minor changes with concentration of LiME, and for a concentration of 10% the vibronic features were not so distinctive; the relative intensity of the red band increased but it remained a minor feature. Lifetimes were measured using a 266 nm nano LED source for excitation with a stated pulse width of o1 ns. The lifetimes were extracted by reconvoluting the scattering spectrum from a solution of silica nano-spheres with a decay function comprising a sum of exponential decays, and performing a least squares fit of the reconvolution to the observed decay. The decay parameters at three emission wavelengths characteristic of POPOP, PPO, and the third weak band are summarised in Table 2. The lifetimes are consistent with the fast scintillation character of these singlet state emitters. A simple measurement on a fast oscilloscope suggested a scintillation decay time of less than 3 ns. 3.2. Gamma ray response measurements The relative responses of each scintillator for gamma rays were determined using three standard 0.37 MBq test sources, barium133, caesium-137 and cobalt-60 which cover the photon energy range from 50 keV to 1330 keV. During testing, each scintillator was mounted at the centre of the photocathode window of a
The responses of the lithium loaded scintillators were compared using an incident beam of thermal neutrons from a research reactor. The beam travelled through 5 m of air from the guide aperture terminating the beam guide system for the reactor. This beam was in turn collimated with a second aperture set 30 cm forward of the scintillator position. This collimator aperture was 10 mm wide and 2 mm high and was cut in a 2 mm thick cadmium sheet forming the body of the collimation system. The samples were positioned in the collimated thermal neutron beam in a similar manner to that described above in assessing the relative sample responses with gamma rays. The scintillators were however coupled to the photocathode via a 100 mm long, 10 mm diameter acrylic light guide. The reason for this extended coupling was to minimise the interaction between the photomultiplier, housing and stray thermal neutrons scattered from the beam. Immediately following the collimator aperture and in front of the scintillator was positioned a shutter mechanism consisting of an electrically driven movable plate made of 2 mm thick cadmium. This plate could be moved to prevent the thermal neutron beam from reaching the scintillator. To allow accurate positioning of the scintillator in the beam, the photomultiplier/scintillator detector combination unit was supported on a frame moveable in small steps by a remotely controlled mechanism. The height and horizontal position of the scintillator were thus adjustable from a location outside the experimental area with the neutron beam active. This fine control of the scintillator position in the neutron beam enabled accurate resetting after each scintillator change allowing accurate comparisons between the responses of individual scintillation samples. Because of the high flux of thermal neutrons (about 107/cm2/s) in the beam, the voltage excursions from the photomultiplier tube anode could not be resolved into individual pulses and so were integrated to provide an average signal. The voltage was recorded with a digital storage oscilloscope linked with a cable from outside the experimental area to allow remote data logging. Small corrections were made to the recorded voltage to account for detection of gamma ray emission from the collimator structure and dark current in the photomultiplier. For example, the signal level with the shutter open was compared with that when the shutter was closed and also with that when the beam was cut off by inserting a ‘beam cup’ positioned about 12 m upstream from the detector. The resulting luminescence curves for the scintillators containing enriched lithium-6 and for the scintillators containing lithium-6 at natural levels are shown in Fig. 3. 4. Discussion We have demonstrated the synthesis of a novel range of scintillators containing LiME, a compound not previously reported
R.D. Breukers et al. / Nuclear Instruments and Methods in Physics Research A 701 (2013) 58–61
61
5. Conclusions
Fig. 3. Response of scintillator samples with different levels of LiME in the thermal neutron beam. Squares represent natural lithium and diamonds enriched lithium-6. Polynomial curve fits of the data showing the average trend of scintillation response are shown as a guide to the eye.
for loading lithium into plastic scintillators. LiME contains free vinyl groups, which facilitate the solubility of this molecule into the polymerisation solution via complementary interactions with the vinyl groups in the monomer solutions. The vinyl groups also allow the lithium methacrylate to be covalently incorporated into the polymer, as opposed to being a guest into a host guest system as occurs with other molecules and nanoparticles used to load lithium into scintillators. Since the resulting scintillator block is made up of a homogeneous polymer matrix this results in it being transparent (Fig. 1). Using our novel LiME loading method we have been able to achieve lithium loadings up to 0.63 wt% (10 wt% LiME). Studies of attenuation length, photoluminescence and gamma ray scintillation response showed that increasing the loading of LiME within the scintillators did not greatly affect their optical properties. Our lithium loading of 0.63 wt% compares well with other lithium containing scintillators, for example NE 320 used in neutrino oscillation experiments by Kaifasz [4] where the lithium-6 loading was 0.15%. Increasing the LiME concentration much over 10% leads to the solution becoming saturated and produces visibly opaque scintillators. Thus if we are to endeavour to improve our system further by increasing lithium loading we will need to investigate molecules with a higher lithium content by weight or increased solubility. When we calculate the mean and the standard deviation of the ratio of the luminescence measurements for our enriched to our natural lithium loaded plastic scintillators (Fig.3), the result is 9.572.5. This result is consistent with the theoretical value of 11 expected for the ratio based on the lithium-6 wt% values in the scintillators tabulated in Table 2. The standard deviation of about 725% is attributed to the combined effect of the statistical contributions of 710% and 715% from the enriched and natural lithium loaded sample luminescence measurement respectively, and 715% due to a small observed variation, measurement to measurement, in the baseline voltage of the photomultiplier tube with the shutter and cup ‘in’ and therefore with no beam incident on the scintillator. The baseline voltage is subtracted from the total voltage measured when the beam is active. Such systematic variations are particularly relevant for the luminescence measurements with the natural lithium loaded plastic scintillators. As is evident from the ratio measurement, the total signal with the natural lithium loaded scintillator is about 10 times smaller than with the enriched scintillator and therefore the baseline variation is more significant. The errors are combined in quadrature to yield an overall standard deviation of about 725% in the ratio.
We have produced polystyrene-based bulk plastic scintillator with high transparency that contains up to 0.63 wt% lithium-6 via additions of up to 10 wt% LiME. We have demonstrated increasing luminescence in a thermal neutron beam with increasing LiME and lithium-6 content (Fig. 3). Due to the much faster decay times of organic scintillators [3,15] than for lithium glass [5] or crystal based scintillators, plastic scintillators loaded with lithium provide higher count rate and superior timing capabilities in thermal neutron detection applications. The detection of thermal neutrons without the associated gamma ray emission as occurs with boron loaded plastic scintillators also has niche applications where contamination of the radiation environment with gamma rays is a disadvantage. The ease of fabrication and ability to process the materials into a variety of shapes and sizes also means that these materials have border security applications. Recently Zaitseva et al. published a report on modification of plastic scintillators to exhibit neutron/ gamma pulse shape discrimination [16] properties previously not observed with polymer based scintillators. This is bound to spark renewed interest into plastic scintillators for many applications. We are investigating modifying our materials by increasing the loading of wavelength shifting molecules to achieve pulse shape discrimination with the polystyrene-based lithium loaded scintillators.
Acknowledgements The authors thank ANSTO for access to the research reactor, especially Ross Piltz and technical staff, and for use of the robotic detector positioning system. They also thank AINSE for financial assistance with travel and accommodation. This work was supported by the New Zealand Ministry of Business, Innovation and Employment through grant VICX0806.
References [1] P.V. Kulkarni, P.P. Antich, J.A. Anderson, J. Fernando, T.M. Aminabhavi, S.F. Harlapur, M.I. Aralaguppi, R.H. Balundgi, Polymer-Plastics Technology and Engineering 36 (1997) 1. [2] B.M. Fisher, J.N. Abdurashitov, K.J. Coakley, V.N. Gavrin, D.M. Gilliam, J.S. Nico, A.A. Shikhin, A.K. Thompson, D.F. Vecchia, V.E. Yants, Nuclear Instruments and Methods in Physics Research Section A 646 (2011) 126. [3] (a) Saint-Gobain Crystals, Scintillation Organics Brochure, 2012.(b) /http:// www.eljentechnology.com/S (accessed 21.08.2012). [4] E. Kajfasz, Proc. Xllth Morlond Workshop, Centre de Physique des Particules de Marseille, Les Arcs (France), 1992. [5] Applied Scintillation Technologies, Lithium Glass (neutron and gamma characteristics), 2012, pp. 1–2. ~ ¨ [6] I.A. Pawe"czak, J. Toke, E. Henry, M. Quinlan, H. Singh, W.U. Schroder, Nuclear Instruments and Methods in Physics Research Section A 629 (2011) 230. [7] M. Katagiri, K. Sakasai, M. Matsubayashi, T. Kojima, Nuclear Instruments and Methods in Physics Research Section A 529 (2004) 317. [8] M. Katagiri, K. Sakasai, M. Matsubayashi, T. Nakamura, Y. Kondo, Y. Chujo, H. Nanto, T. Kojima, Nuclear Instruments and Methods in Physics Research Section A 529 (2004) 274. [9] H.J. Im, S. Saengkerdsub, A.C. Stephan, M.D. Pawel, D.E. Holcomb, S. Dai, Advanced Materials 16 (2004) 1757. [10] S.A. Wallace, M. Allison, Gamma-Free Neutron Detector Based upon Lithium Phosphate Nanoparticles (2007). [11] V.R. Negina, V.N. Popov, V.V. Nazarov, Prib. Tekh. Eksp. (1980) 60. [12] C.D. Bass, E.J. Beise, H. Breuer, C.R. Heimbach, T. Langford, J.S. Nico, Physics (2012), 1–11, arXiv:1206.4036v1 [physics.ins-det]. [13] S.W. Moser, W.F. Harder, C.R. Hurlbut, M.R. Kusner, Radiation Physics and Chemistry 41 (1993) 31. [14] M. Rutkowska, A. Eisenberg, Journal of Applied Polymer Science 30 (1985) 3317. [15] M. Moszynski, B. Bengtson, Nuclear Instruments and Methods 158 (1979) 1. [16] N. Zaitseva, B.L. Rupert, I. PaweŁczak, A. Glenn, H.P. Martinez, L. Carman, M. Faust, N. Cherepy, S. Payne, Nuclear Instruments and Methods in Physics Research Section A 668 (2012) 88.
Contents lists available at SciVerse ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Transparent lithium loaded plastic scintillators for thermal neutron detection R.D. Breukers a,n, C.M. Bartle b, A. Edgar c a b c
Industrial Research Limited, 69 Gracefield Road, PO Box 31310, Lower Hutt, New Zealand GNS Science Ltd, 30 Gracefield Road, PO Box 31312, Lower Hutt, New Zealand School of Chemical and Physical Sciences, Victoria University of Wellington, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 September 2012 Received in revised form 24 October 2012 Accepted 25 October 2012 Available online 1 November 2012
The fabrication of a series of novel, optically transparent, bulk plastic scintillators loaded with lithium methacrylate, and incorporating 2,5-diphenyloxazole and 5-phenyl-2-[4-(5-phenyl-1,3-oxazol-2yl)phenyl]-1,3-oxazole fluorescent centres, is described. The attenuation length, photoluminescence, and both gamma ray and thermal neutron scintillation responses were compared over a range of lithium methacrylate concentrations. The maximum concentration corresponded to a weight percentage of lithium-6 of 0.63%. The photoluminescence shows a composite 2,5-diphenyloxazole and 5-phenyl-2-[4-(5-phenyl-1,3-oxazol-2-yl)phenyl]-1,3-oxazole broad band with vibronic features in the range 350–500 nm, and lifetimes in the range 0.9–2.7 ns. An increasing luminescence in a thermal neutron beam with increasing lithium-6 content is demonstrated. & 2012 Elsevier B.V. All rights reserved.
Keywords: Plastic scintillator Lithium loaded Thermal neutron detection
1. Introduction Due to their comparatively low cost, ability to be machined into complex shapes and fast response times, plastic scintillators have found application in a wide variety of photon and particle detection applications [1]. Therefore, methods to fabricate these scintillators, especially loading with other elements that enhance their performance are of considerable interest. For example, low resolution plastic scintillators loaded with bismuth or lead complement higher cost medium resolution lanthanum halide and sodium iodide detectors [2]. Commercially available scintillators loaded with boron-10, gadolinium, lead and lithium-6 to increase the thermal neutron detection efficiency are available [3]. However, many of these commercially available scintillators are either liquids [4] or glasses [5] which may not offer all of the advantages of plastic scintillators. Plastic scintillators have very fast decay times (e.g. 5 ns) and thus exhibit high count-rate capabilities (e.g. 106 per second) [6]. The process of loading plastic scintillators with specific elements is limited by a reduced light output that often limits the elemental concentrations that can be achieved. Although lithium-6 has a high thermal neutron absorption cross-section, it has proven difficult to load into plastic scintillators. It is an objective of this paper to describe the performance of transparent polystyrene-based plastic scintillators loaded with
n
Corresponding author. Tel.: þ64 4 931 3539. E-mail addresses: [email protected], [email protected] (R.D. Breukers). 0168-9002/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.nima.2012.10.080
lithium-6. Currently lithium-6 containing scintillators are commonly glasses [5] or inorganic crystals and lithium compounds dispersed in phosphor powders [3]. The topic of transparent plastic bulk scintillators containing lithium is relatively unexplored; previous reports of the incorporation of lithium into plastic organic scintillators mainly focus on the mixing of inorganic salts and nanoparticles such as lithium fluoride [7,8], lithium chloride [9] or lithium phosphate [10] into a polymer which results in a heterogeneous material with reduced transparency. A paper relating to a transparent scintillator containing lithium by Negina et al. [11] gives no details of the scintillator’s preparation and composition and is therefore difficult to compare to this current work. Thus there have been only a few reports of promising lithiumloaded plastic scintillators for thermal neutron detection. Yet achieving loading with lithium-6 may offer advantages over boron-10 loaded scintillators, as discussed in a recent report on lithium loaded liquid scintillators by Bass et al. [12]. For example, a feature of the lithium-6 reaction is its comparatively high Q-value (4.78 MeV compared with 2.79 MeV for the boron-10 reaction). Unlike the boron-10 reaction, there is an absence of gamma emissions associated with the charged particle reaction products. The products of both reactions are otherwise short range energetic charged particles, enabling use of small detectors without significant edge effects [2]. In the case of thermal neutron absorption by lithium-6 the product triton and alpha particle are emitted with energies of 2.73 MeV and 2.05 MeV respectively, for a combined deposited energy of 4.78 MeV. In the case of the boron-10 reaction, the product lithium-7 nucleus is produced 94% of the time in an excited state emitting a 478 keV gamma ray
R.D. Breukers et al. / Nuclear Instruments and Methods in Physics Research A 701 (2013) 58–61
which often escapes the detector and is thus not detected but may contaminate the detection environment. Without detection of the gamma ray, the combined particle detected energy from the alpha particle (1.47 MeV) and the lithium ion (0.84 MeV) is 2.31 MeV and is thus considerably lower than 4.78 MeV obtained from the lithium-6 reaction. However, in practice the performance of the scintillator depends on the cross section for the neutron absorbing reaction, the percentage elemental loading that can be achieved, and the comparative light outputs of lithium loaded and boron loaded plastic scintillators. Some light absorption in lithium loading of organic scintillators at 0.15 wt% level and a relatively low neutron detection efficiency compared to boron loaded organic scintillators have been noted [2]. Historically usage has strongly favoured the boron loaded scintillators which are commercially available from Saint Gobain and Eljen Technology [3]. Improvement of the lithium loaded plastic scintillators may lead to a commercially competitive product whereby the higher reaction Q-value and the absence of gamma emission may be utilised. Here we have studied the incorporation of lithium-6 into bulk plastic polystyrene scintillators using new lithium containing compounds to give transparent scintillators. We report a method of preparing scintillators containing lithium which utilises a solution of a lithium complex which fully dissolves in the precursor monomer solution. This mixture can then be polymerised giving a transparent scintillator. Our investigation clearly shows an increase in light output with increases in the lithium-6 concentration when the scintillator is placed in a thermal neutron beam.
2. Preparation of scintillators The main goal of our research was to produce a transparent plastic scintillator loaded with lithium to act as a detector of thermal neutrons. Upon designing this scintillator we needed to incorporate the components of a conventional plastic scintillator into our material so that the products of the lithium neutron capture reaction could be converted to light, which is able to be detected by a photomultiplier tube. Styrene was chosen as the aromatic base material for our scintillator as it is cheap and has been extensively studied, as have the primary and secondary shifters 2,5-diphenyloxazole (PPO) and 5-phenyl-2-[4-(5-phenyl1,3-oxazol-2-yl)phenyl]-1,3-oxazole (POPOP) [13]. To produce our plastic scintillators we polymerised the styrene solution using the free radical initiator azobisisobutyronitrile (AIBN). A range of lithium containing compounds was trialled for solubility in the polymerisation solution. Lithium methacrylate (LiME) was investigated as the methacrylate ligand has a vinyl group which is complementary to the vinyl groups in styrene and allows the compound to be incorporated into the polymer. LiME is sparingly soluble in the styrene monomer and copolymers have been previously reported [14]. However the application of these copolymers to scintillator applications has not been investigated. To increase the lithium loading in the polymerisation solution a mixture of methacrylic acid and styrene was used. A range of scintillators with LiME concentrations varying from 1.25 to 10 wt% were fabricated using both natural lithium and enriched lithium-6. The percentage of lithium-6 by weight in 95% enriched 6Li-ME is 6.3% thus in a 4 g scintillator containing 0.4 g LiME the lithium-6 content is 0.63 wt%. The calculated lithium-6 weight percentages for both the natural lithium (7.5% lithium-6 by weight) scintillator series and the enriched scintillator series are shown in Table 1. The following is an example of the method used to fabricate the scintillators. LiME (0.4 g) was dissolved in methacrylic acid
59
Table 1 LiME content, lithium-6 percentage by weight and attenuation length of the bulk scintillator samples.
Natural (7.5 wt% Li-6)
Enriched (95 wt% Li-6)
LiME wt %
Li-6wt %
Attenuation length (mm)
1.25 2.5 5 7.5 10 1.25 2.5 5 7.5 10
0.007 0.014 0.028 0.042 0.057 0.08 0.16 0.31 0.47 0.63
41 25 32 30 27 22 31 37 41 34
Fig. 1. Photo of scintillator samples, 1.25 wt% enriched LiME (left) and 10 wt% enriched LiME (right).
(1.6 g) and this solution was then mixed with a solution of PPO (20 mg) and POPOP (2 mg) in styrene (2 g). AIBN (2 mg) was then added to this solution and the mixture sealed in a glass ampoule under argon then incubated at 50 1C to facilitate polymerisation. The scintillators fabricated by this method were 20 mm diameter, 10 mm long, transparent solid plastic bullets; however a range of sizes and shapes could be produced. One end face of each scintillator was machined flat to match the photocathode of a photomultiplier tube. The attenuation length of the flattened scintillator samples (Table 1) was measured using an Ocean Optics ST 2000 fibre-optic spectrometer. The samples were all transparent as can be seen in Fig. 1, with an average attenuation length of 32 mm for light at 429 nm the emission wavelength of POPOP (Fig. 2).
3. Luminescence measurements 3.1. Photoluminescence Photoluminescence (PL) spectra were recorded on a Fluorolog 3 fluorometer at room temperature equipped with a 450W Xe arc lamp source. Typical spectra for fluorescence and excitation are shown in Fig. 2. The fluorescence spectra show clear vibronic features which assist in attributing the spectra to two principal bands, assigned to PPO and PPO-pumped POPOP as shown. However a third weak band in the red region of the spectrum remains unattributed, and possibly arises from the polymerising agent. The excitation spectra at wavelengths characteristic of PPO and POPOP are very similar, supporting the idea of the PPO acting to pump emission from the POPOP dye in a wavelength shifting
60
R.D. Breukers et al. / Nuclear Instruments and Methods in Physics Research A 701 (2013) 58–61
Philips XP2262B photomultiplier tube located within a light-tight housing. During measurement the photomultiplier tube axis was vertical thus supporting the scintillator on the horizontal photocathode surface of the photomultiplier. A small amount of optical grease was used to couple the scintillator to the photomultiplier. The photomultiplier signals were amplified and shaped with an ORTEC 113 preamplifier and a 671 main amplifier. The pulse height signals were recorded for each source/sample combination using an ORTEC Easy-MCA. The end point of the Compton spectrum was recorded with each source and each scintillator. The average ‘enriched to natural’ ratio of responses of the correspondingly LiME loaded scintillators was within 1.07 70.05 irrespective of the gamma source used. Luminescence differences were thus small and consistent. 3.3. Thermal neutron luminescence measurements
Fig. 2. Photoluminescence spectra (solid line) of a sample containing 1.25 wt% of enriched LiME when pumped at 260 nm. The dotted, dashed, and long dashed lines show the excitation spectra for PL at 364 nm (PPO), 429 nm (POPOP) and 589 nm (unidentified) respectively.
Table 2 Measured PL lifetimes for the sample containing 1.25 wt% of enriched LiME. For the 429 nm emission, there is a minor slow component which accounts for 4% of the total light emission. Emission wavelength (nm)
s1(ns)
364 429 429 589
0.97 0.1 1.7 7 0.1 1850 7 60 2.4 7 .03
operation. The character of the spectra shows some minor changes with concentration of LiME, and for a concentration of 10% the vibronic features were not so distinctive; the relative intensity of the red band increased but it remained a minor feature. Lifetimes were measured using a 266 nm nano LED source for excitation with a stated pulse width of o1 ns. The lifetimes were extracted by reconvoluting the scattering spectrum from a solution of silica nano-spheres with a decay function comprising a sum of exponential decays, and performing a least squares fit of the reconvolution to the observed decay. The decay parameters at three emission wavelengths characteristic of POPOP, PPO, and the third weak band are summarised in Table 2. The lifetimes are consistent with the fast scintillation character of these singlet state emitters. A simple measurement on a fast oscilloscope suggested a scintillation decay time of less than 3 ns. 3.2. Gamma ray response measurements The relative responses of each scintillator for gamma rays were determined using three standard 0.37 MBq test sources, barium133, caesium-137 and cobalt-60 which cover the photon energy range from 50 keV to 1330 keV. During testing, each scintillator was mounted at the centre of the photocathode window of a
The responses of the lithium loaded scintillators were compared using an incident beam of thermal neutrons from a research reactor. The beam travelled through 5 m of air from the guide aperture terminating the beam guide system for the reactor. This beam was in turn collimated with a second aperture set 30 cm forward of the scintillator position. This collimator aperture was 10 mm wide and 2 mm high and was cut in a 2 mm thick cadmium sheet forming the body of the collimation system. The samples were positioned in the collimated thermal neutron beam in a similar manner to that described above in assessing the relative sample responses with gamma rays. The scintillators were however coupled to the photocathode via a 100 mm long, 10 mm diameter acrylic light guide. The reason for this extended coupling was to minimise the interaction between the photomultiplier, housing and stray thermal neutrons scattered from the beam. Immediately following the collimator aperture and in front of the scintillator was positioned a shutter mechanism consisting of an electrically driven movable plate made of 2 mm thick cadmium. This plate could be moved to prevent the thermal neutron beam from reaching the scintillator. To allow accurate positioning of the scintillator in the beam, the photomultiplier/scintillator detector combination unit was supported on a frame moveable in small steps by a remotely controlled mechanism. The height and horizontal position of the scintillator were thus adjustable from a location outside the experimental area with the neutron beam active. This fine control of the scintillator position in the neutron beam enabled accurate resetting after each scintillator change allowing accurate comparisons between the responses of individual scintillation samples. Because of the high flux of thermal neutrons (about 107/cm2/s) in the beam, the voltage excursions from the photomultiplier tube anode could not be resolved into individual pulses and so were integrated to provide an average signal. The voltage was recorded with a digital storage oscilloscope linked with a cable from outside the experimental area to allow remote data logging. Small corrections were made to the recorded voltage to account for detection of gamma ray emission from the collimator structure and dark current in the photomultiplier. For example, the signal level with the shutter open was compared with that when the shutter was closed and also with that when the beam was cut off by inserting a ‘beam cup’ positioned about 12 m upstream from the detector. The resulting luminescence curves for the scintillators containing enriched lithium-6 and for the scintillators containing lithium-6 at natural levels are shown in Fig. 3. 4. Discussion We have demonstrated the synthesis of a novel range of scintillators containing LiME, a compound not previously reported
R.D. Breukers et al. / Nuclear Instruments and Methods in Physics Research A 701 (2013) 58–61
61
5. Conclusions
Fig. 3. Response of scintillator samples with different levels of LiME in the thermal neutron beam. Squares represent natural lithium and diamonds enriched lithium-6. Polynomial curve fits of the data showing the average trend of scintillation response are shown as a guide to the eye.
for loading lithium into plastic scintillators. LiME contains free vinyl groups, which facilitate the solubility of this molecule into the polymerisation solution via complementary interactions with the vinyl groups in the monomer solutions. The vinyl groups also allow the lithium methacrylate to be covalently incorporated into the polymer, as opposed to being a guest into a host guest system as occurs with other molecules and nanoparticles used to load lithium into scintillators. Since the resulting scintillator block is made up of a homogeneous polymer matrix this results in it being transparent (Fig. 1). Using our novel LiME loading method we have been able to achieve lithium loadings up to 0.63 wt% (10 wt% LiME). Studies of attenuation length, photoluminescence and gamma ray scintillation response showed that increasing the loading of LiME within the scintillators did not greatly affect their optical properties. Our lithium loading of 0.63 wt% compares well with other lithium containing scintillators, for example NE 320 used in neutrino oscillation experiments by Kaifasz [4] where the lithium-6 loading was 0.15%. Increasing the LiME concentration much over 10% leads to the solution becoming saturated and produces visibly opaque scintillators. Thus if we are to endeavour to improve our system further by increasing lithium loading we will need to investigate molecules with a higher lithium content by weight or increased solubility. When we calculate the mean and the standard deviation of the ratio of the luminescence measurements for our enriched to our natural lithium loaded plastic scintillators (Fig.3), the result is 9.572.5. This result is consistent with the theoretical value of 11 expected for the ratio based on the lithium-6 wt% values in the scintillators tabulated in Table 2. The standard deviation of about 725% is attributed to the combined effect of the statistical contributions of 710% and 715% from the enriched and natural lithium loaded sample luminescence measurement respectively, and 715% due to a small observed variation, measurement to measurement, in the baseline voltage of the photomultiplier tube with the shutter and cup ‘in’ and therefore with no beam incident on the scintillator. The baseline voltage is subtracted from the total voltage measured when the beam is active. Such systematic variations are particularly relevant for the luminescence measurements with the natural lithium loaded plastic scintillators. As is evident from the ratio measurement, the total signal with the natural lithium loaded scintillator is about 10 times smaller than with the enriched scintillator and therefore the baseline variation is more significant. The errors are combined in quadrature to yield an overall standard deviation of about 725% in the ratio.
We have produced polystyrene-based bulk plastic scintillator with high transparency that contains up to 0.63 wt% lithium-6 via additions of up to 10 wt% LiME. We have demonstrated increasing luminescence in a thermal neutron beam with increasing LiME and lithium-6 content (Fig. 3). Due to the much faster decay times of organic scintillators [3,15] than for lithium glass [5] or crystal based scintillators, plastic scintillators loaded with lithium provide higher count rate and superior timing capabilities in thermal neutron detection applications. The detection of thermal neutrons without the associated gamma ray emission as occurs with boron loaded plastic scintillators also has niche applications where contamination of the radiation environment with gamma rays is a disadvantage. The ease of fabrication and ability to process the materials into a variety of shapes and sizes also means that these materials have border security applications. Recently Zaitseva et al. published a report on modification of plastic scintillators to exhibit neutron/ gamma pulse shape discrimination [16] properties previously not observed with polymer based scintillators. This is bound to spark renewed interest into plastic scintillators for many applications. We are investigating modifying our materials by increasing the loading of wavelength shifting molecules to achieve pulse shape discrimination with the polystyrene-based lithium loaded scintillators.
Acknowledgements The authors thank ANSTO for access to the research reactor, especially Ross Piltz and technical staff, and for use of the robotic detector positioning system. They also thank AINSE for financial assistance with travel and accommodation. This work was supported by the New Zealand Ministry of Business, Innovation and Employment through grant VICX0806.
References [1] P.V. Kulkarni, P.P. Antich, J.A. Anderson, J. Fernando, T.M. Aminabhavi, S.F. Harlapur, M.I. Aralaguppi, R.H. Balundgi, Polymer-Plastics Technology and Engineering 36 (1997) 1. [2] B.M. Fisher, J.N. Abdurashitov, K.J. Coakley, V.N. Gavrin, D.M. Gilliam, J.S. Nico, A.A. Shikhin, A.K. Thompson, D.F. Vecchia, V.E. Yants, Nuclear Instruments and Methods in Physics Research Section A 646 (2011) 126. [3] (a) Saint-Gobain Crystals, Scintillation Organics Brochure, 2012.(b) /http:// www.eljentechnology.com/S (accessed 21.08.2012). [4] E. Kajfasz, Proc. Xllth Morlond Workshop, Centre de Physique des Particules de Marseille, Les Arcs (France), 1992. [5] Applied Scintillation Technologies, Lithium Glass (neutron and gamma characteristics), 2012, pp. 1–2. ~ ¨ [6] I.A. Pawe"czak, J. Toke, E. Henry, M. Quinlan, H. Singh, W.U. Schroder, Nuclear Instruments and Methods in Physics Research Section A 629 (2011) 230. [7] M. Katagiri, K. Sakasai, M. Matsubayashi, T. Kojima, Nuclear Instruments and Methods in Physics Research Section A 529 (2004) 317. [8] M. Katagiri, K. Sakasai, M. Matsubayashi, T. Nakamura, Y. Kondo, Y. Chujo, H. Nanto, T. Kojima, Nuclear Instruments and Methods in Physics Research Section A 529 (2004) 274. [9] H.J. Im, S. Saengkerdsub, A.C. Stephan, M.D. Pawel, D.E. Holcomb, S. Dai, Advanced Materials 16 (2004) 1757. [10] S.A. Wallace, M. Allison, Gamma-Free Neutron Detector Based upon Lithium Phosphate Nanoparticles (2007). [11] V.R. Negina, V.N. Popov, V.V. Nazarov, Prib. Tekh. Eksp. (1980) 60. [12] C.D. Bass, E.J. Beise, H. Breuer, C.R. Heimbach, T. Langford, J.S. Nico, Physics (2012), 1–11, arXiv:1206.4036v1 [physics.ins-det]. [13] S.W. Moser, W.F. Harder, C.R. Hurlbut, M.R. Kusner, Radiation Physics and Chemistry 41 (1993) 31. [14] M. Rutkowska, A. Eisenberg, Journal of Applied Polymer Science 30 (1985) 3317. [15] M. Moszynski, B. Bengtson, Nuclear Instruments and Methods 158 (1979) 1. [16] N. Zaitseva, B.L. Rupert, I. PaweŁczak, A. Glenn, H.P. Martinez, L. Carman, M. Faust, N. Cherepy, S. Payne, Nuclear Instruments and Methods in Physics Research Section A 668 (2012) 88.