Neutron Response of the EJ-254 Boron-Loaded Plastic Scintillator
Gino Gabella, Bethany L. Goldblum, Thibault A. Laplace, Juan J. Manfredi, Joseph Gordon, Zachary W. Sweger, Edith Bourret
aa r X i v : . [ phy s i c s . i n s - d e t ] J a n Neutron Response of the EJ-254 Boron-LoadedPlastic Scintillator
Gino Gabella, Bethany L. Goldblum, Thibault A. Laplace, Juan J. Manfredi, Joseph Gordon, Zachary W. Sweger,Edith Bourret
Abstract —Organic scintillators doped with capture agentsprovide a detectable signal for neutrons over a broad energyrange. This work characterizes the fast and slow neutron responseof EJ-254, an organic plastic scintillator with 5% natural boronloading by weight. For fast neutrons, the primary mechanism forlight generation in organic scintillators is n-p elastic scattering.To study the fast neutron response, the proton light yield of EJ-254 was measured at the 88-Inch Cyclotron at Lawrence BerkeleyNational Laboratory. Using a broad-spectrum neutron source anda double time-of-flight technique, the EJ-254 proton light yieldwas obtained over the energy range of approximately 270 keV to4.5 MeV and determined to be in agreement with other plasticscintillators comprised of the same polymer base. To isolate theslow neutron response, an AmBe source with polyethylene mod-erator was made incident on the EJ-254 scintillator surroundedby an array of EJ-309 observation detectors. Events in the EJ-254target coincident with the signature 477.6 keV γ ray (resultingfrom deexcitation of the residual Li nucleus following boronneutron capture) were identified. Pulse shape discrimination wasused to evaluate the temporal differences in the response of EJ-254 scintillation signals arising from γ -ray and fast/slow neutroninteractions. Clear separation between γ -ray and fast neutronssignals was not achieved and the neutron capture feature wasobserved to overlap both the γ -ray and fast neutron bands.Taking into account the electron light nonproportionality, theneutron-capture light yield in EJ-254 was determined to be89.4 ± Index Terms —Neutron detector, organic scintillator, light yield,nuclear recoil, neutron capture
I. I
NTRODUCTION
Nuclei that exhibit a high neutron capture cross section forexample, B and Li, can be loaded in organic scintillatorsto enable detection of slow neutrons [1]–[5]. This capability is
This work was performed under the auspices of the U.S. Departmentof Energy by Lawrence Berkeley National Laboratory under Contract DE-AC02-05CH11231. The project was funded by the U.S. Department ofEnergy, National Nuclear Security Administration, Office of Defense NuclearNonproliferation Research and Development (DNN R&D). This material isbased upon work supported in part by the Department of Energy NationalNuclear Security Administration through the Nuclear Science and SecurityConsortium under Award Number DE-NA0003180.Gino Gabella, Thibault A. Laplace, Juan J. Manfredi, Joseph Gordon arewith the Department of Nuclear Engineering, University of California, Berke-ley, CA 94720 USA (email: [email protected]; [email protected]).Bethany L. Goldblum is with the Department of Nuclear Engineering,University of California, Berkeley, CA 94720 USA, and also with the NuclearScience Division, Lawrence Berkeley National Laboratory, Berkeley, CA,94720 USA (email: [email protected]).Edith Bourret is with the Materials Sciences Division, Lawrence BerkeleyNational Laboratory, Berkeley, CA, 94720 USA.Zachary W. Sweger was with the Department of Nuclear Engineering,University of California, Berkeley, CA 94720 USA. He is now with theDepartment of Physics and Astronomy, University of California, Davis, CA95616 USA. useful for basic nuclear physics studies and has a range of ap-plications in proliferation detection, radiation safety, homelandsecurity, and neutron capture therapy [6]–[10]. For example,the DarkSide-50 experiment employs a boron-loaded liquidscintillator for neutron background suppression in the searchfor dark matter [11], [12]. In border monitoring applications,organic scintillators doped with capture agents increase thedetection efficiency for shielded sources of special nuclearmaterial [13], [14]. For antineutrino-based approaches to nu-clear reactor monitoring, capture-doped organic scintillatorscan improve neutron background rejection criteria [15].This work evaluates the fast and slow neutron responseof EJ-254, a commercially available boron-loaded organicscintillator from Eljen Technology. The standard EJ-254 for-mulation is comprised of a polyvinyltoluene (PVT) polymermatrix loaded with approximately 1% B (for a 5% naturalboron loading by weight) [16]. This scintillator is useful forfast neutron spectroscopy and provides a detectable signal forthermal neutrons via scintillation light generated primarily bythe residual α particle produced in the boron neutron-capturereaction [10], [17].For fast neutrons (above approximately 1 keV), interactionsvia n-p elastic scattering give rise to recoil protons thatexcite the surrounding medium, resulting in the productionof scintillation light [18]. The proton light yield relates thelight output of the scintillator to the energy of the recoilingproton and is a unique characteristic of the scintillating ma-terial. Knowledge of the proton light yield is necessary formodeling the efficiencies of fast neutron detectors as well asfor kinematic neutron image reconstruction [12], [19], [20].For slow neutrons (below approximately 1 keV), the numberof photons produced by recoil protons becomes negligiblysmall. Neutron moderation via downscattering in the scintil-lating volume results in an increased probability of capture on B: B + n → ( Li + He Q = 2.792 MeV, 6% Li + He + γ ( ) Q = 2.310 MeV, 94%.The Li and He nuclei produced in the reaction transferenergy to the system via Coulombic interactions, and thescintillation response is dominated by α excitation due tothe ionization quenching effect. For neutrons with energies < keV, the residual Li nucleus is populated in its firstexcited state with a 94% branching ratio [21], which decays byprompt γ emission with E γ = 477 . keV. For faster neutrons,the probability of populating the first excited state in Li decreases with increasing neutron energy. This work uses thecharacteristic γ ray from Li deexcitation to aid in identifyingthe slow neutron response.Section II details the experimental setup, data analysis,and results of a measurement of the EJ-254 proton lightyield from approximately 270 keV to 4.5 MeV. Section IIIdescribes the experimental setup, data analysis, and resultsof a measurement of the slow neutron response. Concludingremarks are provided in Section IV.II. F
AST N EUTRON R ESPONSE
A. Experimental Methods
The double time-of-flight (TOF) method of Brown et al. wasused to characterize the relative proton light yield of EJ-254[22]. A high-flux, broad-spectrum neutron beam was producedby impinging 16 MeV deuterons from the 88-Inch Cyclotron atLawrence Berkeley National Laboratory onto a 3-mm-thick Betarget located in the cyclotron vault [23]. The EJ-254 organicplastic scintillator, a 5.08 cm diameter × − V. The EJ-254 scintillator was wrapped in at leastten layers of polytetrafluoroethylene tape to maximize lightcollection [24].An array of 11 EJ-309 observation scintillators was placedout of beam and at forward angles to detect n-p elasticallyscattered neutrons. The EJ-309 organic liquid scintillator fromEljen Technology has pulse shape discrimination (PSD) prop-erties, enabling differentiation of neutron and γ -ray interac-tions on an event-by-event basis. Each observation detectorconsisted of a 5.08 cm diameter × E ′ n , was determined us-ing the TOF between the target scintillator and the correspond-ing observation detector. The energy of the recoiling proton, E p , was then calculated on an event-by-event basis, using theneutron scattering angle, θ (known from the fixed positions ofthe observation detectors), and the scattered neutron energy: E p = E ′ n tan θ. (1)The light output in the target scintillator associated with agiven proton recoil energy was determined to provide theproton light yield relation. The TOF of the incoming neutron(determined using the time difference between an event inthe target scintillator and the cyclotron RF signal) was usedto reduce background contributions [22]. This kinematicallyover-constrained system provides a strong rejection criterionfor multiple scattering events in the target scintillator [25]. Fig. 1. Experimental setup for the proton light yield measurement. Theneutron beam travels along the x -axis (illustrated by the dashed line) throughthe EJ-254 target scintillator. The 11 observation cells were positioned atforward angles with respect to the incoming neutron beam. B. Data Acquisition and Pulse Processing
Data acquisition was accomplished using a CAEN V1730500 MS/s digitizer, programmed to record events when a signalwas observed in the target detector and at least one of theobservation detectors within a 400 ns window. Full waveformswere recorded in list mode with global timestamps. TheCAEN digital constant fraction discrimination algorithm (75%fraction and 4 ns delay) was applied to determine the arrivaltime of the scintillation pulses and leading edge discriminationwas applied for the cyclotron RF timing determination. Datareduction was accomplished using a custom, object-orientedC ++ library, which used elements of the ROOT v6.16 dataanalysis framework [26]. For target pulse processing, a meanwaveform was obtained by averaging pulses with energiesspanning − % of the digitizer dynamic range. Thewaveform integration length for EJ-254 was set to 300 ns toensure collection of over 95% of the scintillation light in theaverage waveform emitted within a 606 ns acquisition window.Using a pulsed LED circuit adapted from the method of Friendet al. [27], the target PMT response was shown to be linearwithin 1% over the relevant output range. C. Calibration and Data Analysis
The light output of the target detector was calibrated usingthe Compton edge (at 477 keV) of the 662 keV γ ray froma Cs source, placed > cm away from the detector. Thelight output unit was defined relative to the light yield of a477 keV electron, that is, a relative light unit of 1 correspondsto the same amount of light produced by a 477 keV recoilelectron [28]. Electron-equivalent light units were not usedbecause they are poorly defined if the light output is notproportional to the electron energy. This consideration isrelevant to the present work as EJ-254 shares the same PVTpolymer base as the EJ-200 plastic scintillator from EljenTechnology and its commercial equivalent BC-408, whichare known to have a nonproportional electron response [29],[30]. A Geant4 [31] simulation of the electron energy deposi-tion spectrum convolved with a detector resolution function[32] was used to obtain the light output associated withthe Compton edge by minimizing χ between the measuredand simulated spectra. The parameter minimizations wereperformed using the SIMPLEX and MIGRAD algorithms fromthe ROOT Minuit2 package [33]. Proton recoil energy [MeV] −
10 1 L i gh t y i e l d [ r e l a t i v e t o k e V e - ] Fig. 2. (Color online) 2-D histogram of the EJ-254 relative light yield as afunction of proton recoil energy. The binning on the abscissa was set usingthe recoil proton energy resolution. The red data points indicate the centroidsobtained from fitting the projected pulse integral spectra in each bin. The y -error bars, which are smaller than the data points in some cases, representboth the statistical and systematic uncertainty. Light yield [relative to 477 keV e-] C oun t s / . li gh t un i t s Fig. 3. (Color online) Relative light yield spectrum (blue) for the protonenergy bin centered at E p = 1645 keV fit with a piecewise power law(predominant at low amounts of light) and a normal distribution (red). Thecentroid of the distribution corresponds to the mean light output for n-p elasticscattering events within the bin. Timing calibrations were performed for both the incomingand outgoing neutron TOF measurements. In the case of theincoming neutrons, the TOF was calibrated using the photonflash produced by deuterons impinging on the Be target andthe known distance between the Be target and the EJ-254scintillator. Assuming normality, the standard deviation ofthis distribution was determined to be 6.97 ns. The outgoingneutron TOF was calibrated individually for each observationdetector using time differences between γ -ray interactions inthe target and observation detectors. The standard deviationsof the resulting distributions ranged between 0.4 and 0.5 ns.Fig. 2 provides a 2-D histogram of the measured EJ-254relative light yield as a function of proton recoil energy.A series of constraints were applied to the data, includingPSD in the observation detectors to isolate neutron events aswell as maximum and minimum outgoing neutron energiesreflective of the deuteron beam energy and intersection of then-p elastic scattering feature with the detection threshold. Thedifference between the measured and kinematically-calculatedincoming TOF (the latter obtained using the outgoing TOF and scattering angle) was also required to be less than 15.8 ns(i.e., < of the cyclotron period). The proton energyresolution, determined via a Monte Carlo simulation, was usedto define the proton-energy binning in Fig. 2. A syntheticdataset was generated taking into account the angular un-certainty due to the geometry of the detector array and theuncertainties in the neutron flight path and outgoing TOF.For each simulated neutron-scattering event, the proton recoilenergy was calculated using Eq. 1 (assuming interaction at thecenter of the detection volume) and compared to the simulatedproton energy to provide the heteroskedastic proton energyresolution function. To reduce the 2D distribution into datapoints, each bin was projected onto the light-yield axis and fitvia maximum-likelihood estimation using a Gaussian functionwith the background modeled using a continuous piecewisepower law distribution. This is illustrated for the proton energybin centered at E p = 1645 keV in Fig. 3. The red data points inFig. 2 indicate the centroids obtained from fitting the projectedpulse integral spectra in each bin.The main contributions to the light yield uncertainty werethe sensitivity to parameters in the light yield analysis and theuncertainty in the determination of the Compton edge positionduring light output calibration. The stability of the PMT gainas a function of time was investigated as a potential sourceof uncertainty and determined to be negligible. To quantifythe uncertainty from sensitivity to the light yield parameters,a Monte Carlo calculation was performed in which eachparameter was varied by sampling from a distribution definedby its associated uncertainty. The data reduction was then re-peated with these varied parameters to determine the influenceon the extracted light yield. The sampled parameters werethe incoming/outgoing neutron TOF calibration constants, themeasured detector locations, and the flight path from the Betarget to the origin of the coordinate system. The uncertaintyassociated with determination of the Compton edge positionas obtained using the error on the fit parameters was negligiblysmall. As a proxy for model error, the Compton-edge fitwas repeated while the fit region about the edge was varied,which resulted in a standard deviation of 1%. The resultinguncertainties were added in quadrature to arrive at the totallight yield uncertainty for each data point. D. Results and Discussion
Fig. 4 shows the relative proton light yield of EJ-254obtained in this work alongside the EJ-204 proton light yieldfrom Laplace et al. [28]. The error bars on the abscissarepresent a bin width, and those on the ordinate axis representthe combined statistical and systematic uncertainty. Since thetwo plastic scintillators share the same PVT base, agreementis expected between their relative proton light yield data asionization quenching is a primary process at low fluor concen-tration [34]. Indeed, over the full range of the measurement,the EJ-254 and EJ-204 proton light yield relations agree withintwo standard deviations. The EJ-254 relative proton light yielddata are summarized in the Appendix in Table II.In Fig. 5, the relative proton light yield of EJ-254 is com-pared to previous measurements by Laplace et al. [35] of the Proton recoil energy [MeV] −
10 1 L i gh t y i e l d [ r e l a t i v e t o k e V e - ] EJ-254 - This workEJ-204 - Laplace et al. [28]
Fig. 4. (Color online) EJ-254 proton light yield relation (filled black squares)compared to that of EJ-204 (open blue triangles) by Laplace et al. [28]. The x -error bars represent proton-energy bin widths. The y -error bars representthe statistical and systematic light output uncertainty. In some cases, the errorbars are smaller than the data points. EJ-309 liquid and EJ-276 plastic PSD-capable organic scintil-lators from Eljen Technology. The error bars on the abscissarepresent a bin width, and those on the ordinate axis representthe combined statistical and systematic uncertainty. At 1 MeV,the proton light yield of EJ-254 is approximately 19% lowerthan that of EJ-309 and approximately 21% higher than that ofEJ-276. The relatively higher electron number density of theplastic scintillators contributes to higher specific energy lossof the recoil protons, which in turn contributes to increasedionization quenching in the plastic media compared to theliquid EJ-309 scintillator [36]. As the solvent-to-solute transferis less efficient in plastic scintillators relative to liquids, higherfluor concentration is typically required to achieve similarscintillation efficiency [37]. For PSD-capable plastic scintilla-tors such as EJ-276, fluor concentrations must be even higherto provide conditions suitable for sufficient triplet-triplet an-nihilation [38]. Concentration quenching, which arises fromthe formation of metastable dimers of excited and unexcitedmolecules, increases with increasing fluor concentration. Thisexcimer production would be exacerbated at high excitationdensity, which may be partly responsible for the relativelylower proton light yield of EJ-276 to EJ-254 to EJ-309.III. S
LOW N EUTRON R ESPONSE
A. Experimental Methods
The identification of neutron-capture events was accom-plished using a method adapted from Sun et al. [39]. The 5.08cm dia. × . to . m. The target scintillatorwas wrapped in at least ten layers of polytetrafluoroethylenetape and optically coupled using BC-630 silicone grease to aHamamatsu H1949-51 PMT biased to − V. The obser-vation detector PMTs of Hamamatsu Type 1949-50 or 1949-51 were negatively biased using either a CAEN R1470ET orCAEN NDT1470 power supply.A diagram of the experimental setup is shown in Fig. 6. Thesource, positioned behind shielding material, and the target Proton recoil energy [MeV] −
10 1 L i gh t y i e l d [ r e l a t i v e t o k e V e - ] EJ-254 - This workEJ-276 - Laplace et al. [35]EJ-309 - Laplace et al. [35]
Fig. 5. (Color online) EJ-254 proton light yield relation (filled black squares)compared to that of EJ-276 (open red triangles) and EJ-309 (open bluecircles) by Laplace et al. [35]. The x -error bars represent proton-energy binwidths. The y -error bars represent the statistical and systematic light outputuncertainty. In some cases, the error bars are smaller than the data points.Fig. 6. (Color online) Schematic of the experimental setup used to isolateneutron-capture events. The EJ-254 scintillator is represented in blue, thepolyethylene shielding in red, and the lead shielding in gray. The AmBesource (gold) was positioned flush against the lead shielding. The targetscintillator was atop a sheet of polyethylene and all detectors rested upona large aluminum table (not pictured). scintillator, resting on a sheet of polyethylene, were placedon a large aluminum table. A minimum of 5 mm of Pband 5.1 cm of polyethylene were placed between the EJ-254detector and the source. The Pb shielding reduced contribu-tions from the 59.5 keV γ ray produced by decay of Am,and the polyethylene moderator slowed neutrons to increasethe neutron-capture rate. For B( n, α ) events in the targetscintillator populating the first excited state of Li, a 477.6 keV γ ray was emitted, which was detected in coincidence withthe signal produced in the EJ-254 detector by the recoilingreaction products. Data acquisition and pulse processing wereaccomplished as described in Section II-B. Data were recordedin two modes: a singles trigger, where events were loggedwhen a signal was observed in the target detector, and acoincident trigger, where events were logged when a signalwas observed in the target detector and at least one observationdetector within a 160 ns window. The observation detectorlocations were selected to balance detection efficiency with asufficient difference in the TOF between fast neutron and γ -rayinteractions to enable rejection of neutron scattering events. Coincidence TOF [ns] C oun t s / . n s Fig. 7. Coincident TOF plot for a single observation detector. The 2 nswindow used to select capture-gated events is represented by the verticaldashed lines.
B. Data Analysis and Calibration
Fig. 7 shows the distribution of events in TOF space for asingle observation detector. The feature centered at approxi-mately 14 ns corresponds to γ - γ scatters (i.e., γ -ray scatteringin the target detector followed by interaction in the observationdetector) and neutron-capture events in the target detectoraccompanied by the detection of a 477.6 keV γ ray in theobservation detector. A 2 ns window (indicated by the dashedvertical red lines in Fig. 7) was applied to isolate these eventsin post-processing. As described by Sun et al. [39], there isa significant contribution to the overall counts produced by γ rays scattering first off an observation detector and then backinto the target, visible in coincident TOF space as a smallerpeak earlier in time (at ∼ γ -rays scattering off the floorand surrounding materials.The light output of the EJ-254 target detector was calibratedusing the 59.5 keV γ ray from an Am source, placed > C oun t s Experimental dataSimulation
Light output [MeVee] − − − − R e s i du a l c oun t s Fig. 8. (Color online) The top panel displays the minimization between asimulated energy-deposition spectrum folded with a resolution function (red)and an experimentally-measured pulse integral spectrum (blue) produced withan
Am source incident on the EJ-254 detector. The residual plot in thebottom panel demonstrates the goodness of fit.
Due to the low energy of this γ ray and relatively highinteraction probability, the calibration was sensitive to thesurrounding materials. The amount of light per unit electron-energy deposited extracted using a full simulation of the mate-rials surrounding the detector (accounting for down-scattered γ rays) was . lower than that obtained via simulationassuming a bare scintillator. Neglecting the effect of electronlight yield nonproportionality increased the ostensible meanelectron-energy equivalent of the distribution of events in thepeak region by . . C. Results and Discussion
Fig. 9 shows the PSD parameter as a function of the totallight output in the EJ-254 scintillator. The shape of the scintil-lation pulse is expected to vary based upon the specific energyloss of the recoil particles. The ratio of prompt-to-delayed lightis greater for recoil electrons (arising from γ -ray interactions)than recoil protons (from fast neutron interactions), and thisratio is even lower for recoiling alpha particles, which domi-nate the boron neutron-capture scintillation response. Using acharge integration approach, the PSD parameter was definedas the ratio of the integral of the tail of the EJ-254 pulseto that of the total (with a total pulse integration length setto 300 ns). To optimize separation between γ -ray and fastneutron interactions, the tail start time was sampled over therange of − ns in 2 ns increments and the figure-of-merit (i.e., ratio of the peak separation to the sum of the full-width at half-maximum for each particle distribution [40]) wasmaximized at high relative light yield to obtain a tail start timeof 28 ns. Fig. 9a shows the PSD plot for the EJ-254 targetscintillator obtained without coincident constraints, whereinclear separation between the γ -ray and neutron distributionswas not observed. In Fig. 9b, coincident TOF constraintswere applied to separate neutron scattering events (black) andneutron-capture/ γ -scattering events (red). The neutron-capturefeature is centered at approximately 1.5 relative light units andoverlaps both the γ -ray and fast neutron bands.Fig. 10 provides a capture-gated relative light yield his-togram for EJ-254, with additional minimum pulse integral TABLE IC
OMPARISON OF THE RELATIVE LIGHT YIELD OF NEUTRON - CAPTURE EVENTS IN COMMERCIALLY AVAILABLE BORON - LOADED ORGANICSCINTILLATORS .Commercial name Type Neutron-capture light yield (keVee) Calibration source CitationEJ-254 Plastic 89.4 ± Am This workBC-454 Plastic 76
Cs Drake et al. [2]EJ-254 Plastic 76
Cs Eljen [2], [16]EJ-254 Plastic 100
Am Sun et al. [39]BC-454 Plastic 100
Am Normand et al. [41]BC-454 Plastic 93 Multiple Miller et al. [42]BC-454 Plastic 93 Multiple Feldman et al. [43]EJ-309B5 Liquid 100
Am Swiderski et al. [9]BC-523 Liquid 60
Am Aoyama et al. [19]BC-523A2 Liquid 60
Am Swiderski et al. [9]EJ-339A Liquid 60
Am Swiderski et al. [9]EJ-339A Liquid 50
Cs Pino et al. [10]
Light yield [relative to 59.5 keV e-] T o t a l / Q T a il R a t i o o f c ha r ge Q (a) Light yield [relative to 59.5 keV e-] T o t a l / Q T a il R a t i o o f c ha r ge Q (b)Fig. 9. (Color online) PSD plot for EJ-254 as a function of the relative lightyield. (a) Without coincident constraints. (b) With coincident TOF constraints,where the black data points correspond to neutron scattering events and thered correspond to neutron-capture and γ -ray scattering events. constraints applied to the signals in each observation detectorto reduce background contributions. The feature was fit using aGaussian distribution with a power-law background to estimateits centroid value of 1.503 ± Am photopeakas the calibration fiducial, yielding 89.4 ± Light yield [relative to 59.5 keV e-] C oun t s Fig. 10. Relative light output spectrum for capture-gated events in the EJ-254scintillator. treated properly. The keVee unit, which is often applied withan incorrect assumption of electron light proportionality, willdepend upon the calibration source used unless a correctionis applied. The nonproportionality measurement of EJ-200 byPayne et al. [30] suggests that the light yield (in number ofphotons per electron energy deposited) is approximately lower for a 59.5 keV electron compared to that for a 477 keVelectron (corresponding to the Compton edge of a 662 keV γ ray from a Cs source).To quantify the uncertainty on the neutron-capture lightyield, calibration data were recorded before and after the 4-day experiment, and the resulting calibration constants differedby 1.1%. This discrepancy is likely due to PMT instabilityduring the counting period. This difference was investigatedby isolating the neutron-capture feature from 39 chronologicalsegments of the experiment, fitting each feature, and calcu-lating the standard deviation of the resulting centroid values,which exhibited no global trend in time and were insteadscattered stochastically about the mean value. This resulted ina ± along with a 0.1% statistical uncertainty obtained on the fitparameter in the χ minimization. Contributions from theuncertainty in the thickness of aluminum, vinyl tape, andpolytetrafluoroethylene wrappings to the calibration constantwere investigated using Geant4 simulations and determinedto be negligible. A Geant4 simulation was also performed toevaluate the potential for bias in the neutron-capture light yieldcharacterization as a result of 477.6 keV γ rays scatteringin the volume, and the contribution was determined to benegligible within statistical uncertainty.Table I presents a comparison of the neutron-capture lightyield for commercially available boron-loaded organic scin-tillators. The measurement presented here, along with thosefrom Sun et al. [39] and Miller et al. [42], used a coincidenttagging setup, whereby 477.6 keV of the energy released inthe nuclear reaction is lost to the deexcitation γ ray. All otherdatasets include neutron-capture events directly populating theground state of Li (with a 6% thermal branching ratio),leading to higher energy recoiling charged particles. Whentaking into account electron light nonproportionality, the lightyield corresponding to the neutron-capture feature obtained inthis work is in rough agreement with previous measurementson EJ-254 and its commercial equivalent, BC-454. Drake etal. measured the light output of the capture feature in BC-454 using a
Cs source for calibration and obtained a lightyield of 76 keVee. Eljen Technology also reports this valuein the EJ-254 material specification sheet [16]. Correcting forelectron light nonproportionality, this number is equivalent to85.4 keVee.The measurements in Sun et al. [39] and Normand et al.[41] used an
Am source for calibration and obtained ahigher light output for the neutron capture feature. Thesediscrepancies are likely due to the lack of proper treatment ofelectron light nonproportionality in the calibration procedures.Neglecting the contribution of down-scattered γ rays frommaterials surrounding the scintillator and electron light yieldnonproportionality, the light yield of the neutron capturefeature presented in this work would have been estimated at101.7 keVee. The measurements in [42], [43] used multiple γ -ray sources to estimate the light yield on an electron-equivalentenergy scale. Their results are approximately 4% higher thanthe neutron-capture light yield measured in this work. Whilemeasurements on commercially available liquids, such as EJ-309B5, show a similar capture-feature light yield to thatobtained in [39], [41], a lower relative light yield is reportedfor BC-523, BC-523A2, and EJ-339A—possibly indicative ofhigher ionization density and/or oxygen quenching in thesematerials [9]. IV. S UMMARY
The fast neutron response of the EJ-254 boron-loaded plas-tic scintillator was characterized using a double TOF techniquefor proton recoil energies between approximately 270 keVand 4.5 MeV. The EJ-254 proton light yield agreed withthat of EJ-204 within the associated uncertainties and is thussimilar to a number of fast plastic scintillators produced byEljen Technology, including EJ-200, EJ-208, EJ-230, EJ-232, and EJ-232Q [25], [28]. This is reasonable considering allthese materials share the same PVT base and the ionizationquenching effect occurs in the host material. The neutroncapture response of EJ-254 was measured using a coincidenttagging technique. The light output of the neutron-capturefeature was determined to be 89.4 ± ± PPENDIX
The relative proton light yield data for EJ-254 are providedin Table II.
TABLE IIR
ELATIVE PROTON LIGHT YIELD DATA FOR THE
EJ-254
SCINTILLATOR .P ROTON RECOIL ENERGY BIN WIDTHS ARE PROVIDED , AS WELL AS THELIGHT OUTPUT UNCERTAINTIES . T
HE COVARIANCE MATRIX IS AVAILABLEUPON REQUEST .Proton recoil energy Light yield [ MeV ] [ relative to 477 keV electron ] +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± +0 . − . ± A CKNOWLEDGEMENTS
The authors thank the 88-Inch Cyclotron operations andfacilities staff for their help in performing these experiments.R
EFERENCES[1] J. C. Duckworth, A. W. Merrison, and A. Whittaker, “A High-EfficiencyNeutron Detector,”
Nature , vol. 165, no. 4185, p. 69, 1950. [Online].Available: https://doi.org/10.1038/165069a0[2] D. Drake, W. Feldman, and C. Hurlbut, “New electronically black neu-tron detectors,”
Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and AssociatedEquipment [3] I. A. Pawełczak and A. M. Glenn and H. P. Martinez andM. L. Carman and N. P. Zaitseva and S. A. Payne, “Boron-loaded plastic scintillator with neutron- γ pulse shape discriminationcapability,” Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and AssociatedEquipment
Scientific Reports , vol. 5, no. 1, p. 1?9,Sep 2015.[5] N. Zaitseva, A. Glenn, L. Carman, A. Mabe, and S. Payne, “New solid-state organic scintillators for wide-energy neutron detection,” LawrenceLivermore National Laboratory, Tech. Rep. LLNL-PROC-742084, 2017.[6] Y.-F. Yen, J. Bowman, R. Bolton, B. Crawford, P. Delheij, G. Hart,T. Haseyama, C. Frankle, M. Iinuma, J. Knudson, A. Masaike,Y. Masuda, Y. Matsuda, G. Mitchell, S. Penttil¨a, N. Roberson,S. Seestrom, E. Sharapov, H. Shimizu, D. Smith, S. Stephenson,J. Szymanski, S. Yoo, and V. Yuan, “A high-rate B-loadedliquid scintillation detector for parity-violation studies in neutronresonances,”
Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and AssociatedEquipment
Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and AssociatedEquipment
AppliedRadiation and Isotopes
IEEE Transactions on Nuclear Science , vol. 57,no. 1, pp. 375–380, Feb 2010.[10] F. Pino, L. Stevanato, D. Cester, G. Nebbia, L. Sajo-Bohus,and G. Viesti, “Detecting fast and thermal neutrons with aboron loaded liquid scintillator, EJ-339A,”
Applied Radiationand Isotopes
Journalof Instrumentation , vol. 11, no. 03, pp. P03 016–P03 016, mar 2016.[Online]. Available: https://doi.org/10.1088/1748-0221/11/03/P03016[12] S. Westerdale, J. Xu, E. Shields, F. Froborg, F. Calaprice, T. Alexander,A. Aprahamian, H. Back, C. Casarella, X. Fang, Y. Gupta, E. Lamere,Q. Liu, S. Lyons, M. Smith, and W. Tan, “Quenching measurementsand modeling of a boron-loaded organic liquid scintillator,”
Journalof Instrumentation , vol. 12, no. 08, pp. P08 002–P08 002, aug 2017.[Online]. Available: https://doi.org/10.1088/1748-0221/12/08/P08002[13] P. Peerani, A. Tomanin, S. Pozzi, J. Dolan, E. Miller,M. Flaska, M. Battaglieri, R. D. Vita], L. Ficini, G. Ottonello,G. Ricco, G. Dermody, and C. Giles, “Testing on novelneutron detectors as alternative to He for security applications,”
Nuclear Instruments and Methods in Physics Research Section A:Accelerators, Spectrometers, Detectors and Associated Equipment
IEEE Transactions onNuclear Science , vol. 55, no. 6, pp. 3710–3716, Dec 2008.[15] S. Dazeley, A. Bernstein, T. Classen, E. Reedy, D. Hellfeld, M. Duvall,and C. Marianno, “Antineutrino detection based on Li-doped pulseshape sensitive plastic scintillator and gadolinium-doped water,”
Inter-national Journal of Modern Physics: Conference Series , vol. 48, p.1860105, 2018.[16]
Boron Loaded Plastic Scintillator EJ-254 , El-jen Technology, July 2020. [Online]. Available:https://eljentechnology.com/images/products/data sheets/EJ-254.pdf[17] L. R. Greenwood and N. R. Chellew, “Improved b-loaded liquidscintillator with pulse-shape discrimination,” Review of Scientific Instru-ments , vol. 50, no. 4, pp. 466–471, 1979.[18] H. Klein and F. D. Brooks, “Scintillation Detectors For Fast Neutrons,”in
Proceedings of International Workshop on Fast Neutron Detectorsand Applications — PoS(FNDA2006) , vol. 025, 2007, p. 097.[19] T. Aoyama, K. Honda, C. Mori, K. Kudo, and N. Takeda,“Energy response of a full-energy-absorption neutron spectrometerusing boron-loaded liquid scintillator BC-523,”
Nuclear Instrumentsand Methods in Physics Research Section A: Accelerators,Spectrometers, Detectors and Associated Equipment
Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and AssociatedEquipment
Nuclear DataSheets , vol. 148, pp. 1 – 142, 2018, special Issue on Nuclear ReactionData.[22] J. A. Brown, B. L. Goldblum, T. A. Laplace, K. P. Harrig, L. A.Bernstein, D. L. Bleuel, W. Younes, D. Reyna, E. Brubaker, andP. Marleau, “Proton light yield in organic scintillators using a doubletime-of-flight technique,”
Journal of Applied Physics , vol. 124, no. 4,p. 045101, 2018. [Online]. Available: https://doi.org/10.1063/1.5039632[23] K. P. Harrig, B. L. Goldblum, J. A. Brown, D. L. Bleuel, L. A.Bernstein, J. Bevins, M. Harasty, T. A. Laplace, and E. F. Matthews,“Neutron spectroscopy for pulsed beams with frame overlap using adouble time-of-flight technique,”
Nuclear Instruments and Methods in
Physics Research Section A: Accelerators, Spectrometers, Detectors andAssociated Equipment , vol. 877, pp. 359 – 366, 2018.[24] M. Janecek, “Reflectivity spectra for commonly used reflectors,”
IEEETransactions on Nuclear Science , vol. 59, no. 3, pp. 490–497, June 2012.[25] J. J. Manfredi, B. L. Goldblum, T. A. Laplace, G. Gabella, J. Gordon,A. O’Brien, S. Chowdhury, J. A. Brown, and E. M. Brubaker, “Protonlight yield of fast plastic scintillators for neutron imaging,”
IEEETransactions on Nuclear Science , vol. 67, pp. 434–442, 2020.[26] R. Brun and F. Rademakers, “ROOT – An object oriented data analysisframework,”
Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and AssociatedEquipment
Nuclear Instruments and Methods in PhysicsResearch Section A: Accelerators, Spectrometers, Detectors and Asso-ciated Equipment , vol. 676, pp. 66 – 69, 2012.[28] T. A. Laplace, B. L. Goldblum, J. A. Brown, D. L. Bleuel,C. A. Brand, G. Gabella, T. Jordan, C. Moore, N. Munshi, Z. W.Sweger, A. Sweet, and E. Brubaker, “Low energy light yieldof fast plastic scintillators,”
Nuclear Instruments and Methods inPhysics Research Section A: Accelerators, Spectrometers, Detectorsand Associated Equipment
IEEE Transactions on Nuclear Science , vol. 55, no. 3, pp. 1069–1072,June 2008.[30] S. A. Payne, W. W. Moses, S. Sheets, L. Ahle, N. J. Cherepy, B. Sturm,S. Dazeley, G. Bizarri, and W. S. Choong, “Nonproportionality ofScintillator Detectors: Theory and Experiment. II,”
IEEE Transactionson Nuclear Science , vol. 58, no. 6, pp. 3392–3402, Dec 2011.[31] S. Agostinelli, J. Allison, K. Amako, J. Apostolakis, H. Araujo, P. Arce,M. Asai, D. Axen, S. Banerjee, G. Barrand, F. Behner, L. Bellagamba,J. Boudreau, L. Broglia, A. Brunengo, H. Burkhardt, S. Chauvie,J. Chuma, R. Chytracek, G. Cooperman, G. Cosmo, P. Degtyarenko,A. Dell’Acqua, G. Depaola, D. Dietrich, R. Enami, A. Feliciello,C. Ferguson, H. Fesefeldt, G. Folger, F. Foppiano, A. Forti, S. Garelli,S. Giani, R. Giannitrapani, D. Gibin, J. G. Cadenas, I. Gonzlez, G. G.Abril, G. Greeniaus, W. Greiner, V. Grichine, A. Grossheim, S. Guatelli,P. Gumplinger, R. Hamatsu, K. Hashimoto, H. Hasui, A. Heikkinen,A. Howard, V. Ivanchenko, A. Johnson, F. Jones, J. Kallenbach,N. Kanaya, M. Kawabata, Y. Kawabata, M. Kawaguti, S. Kelner,P. Kent, A. Kimura, T. Kodama, R. Kokoulin, M. Kossov, H. Kurashige,E. Lamanna, T. Lampn, V. Lara, V. Lefebure, F. Lei, M. Liendl, W. Lock-man, F. Longo, S. Magni, M. Maire, E. Medernach, K. Minamimoto,P. M. de Freitas, Y. Morita, K. Murakami, M. Nagamatu, R. Nartallo,P. Nieminen, T. Nishimura, K. Ohtsubo, M. Okamura, S. O’Neale,Y. Oohata, K. Paech, J. Perl, A. Pfeiffer, M. Pia, F. Ranjard, A. Rybin,S. Sadilov, E. D. Salvo, G. Santin, T. Sasaki, N. Savvas, Y. Sawada,S. Scherer, S. Sei, V. Sirotenko, D. Smith, N. Starkov, H. Stoecker,J. Sulkimo, M. Takahata, S. Tanaka, E. Tcherniaev, E. S. Tehrani,M. Tropeano, P. Truscott, H. Uno, L. Urban, P. Urban, M. Verderi,A. Walkden, W. Wander, H. Weber, J. Wellisch, T. Wenaus, D. Williams,D. Wright, T. Yamada, H. Yoshida, and D. Zschiesche, “Geant4 – asimulation toolkit,”
Nuclear Instruments and Methods in Physics Re-search Section A: Accelerators, Spectrometers, Detectors and AssociatedEquipment , vol. 506, no. 3, pp. 250 – 303, 2003.[32] G. Dietze and H. Klein, “Gamma-calibration of NE 213 scintillationcounters,”
Nuclear Instruments and Methods in Physics Research , vol.193, no. 3, pp. 549 – 556, 1982.[33] R. Brun and F. Rademakers, “ROOT – An object oriented data analysisframework,”
Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and Associated Equip-ment , vol. 389, no. 1, pp. 81 – 86, 1997.[34] J. Birks,
The Theory and Practice of Scintillation Counting . PergamonPress, 1964, ch. 1, pp. 447–450.[35] T. A. Laplace, B. L. Goldblum, J. E. Bevins, D. L. Bleuel, E. Bourret,J. A. Brown, E. J. Callaghan, J. S. Carlson, P. L. Feng, G. Gabella,K. P. Harrig, J. J. Manfredi, C. Moore, F. Moretti, M. Shinner,A. Sweet, and Z. W. Sweger, “Comparative scintillation performance ofEJ-309, EJ-276, and a novel organic glass,”
Journal of Instrumentation ,vol. 15, no. 11, pp. P11 020–P11 020, nov 2020. [Online]. Available:https://doi.org/10.1088%2F1748-0221%2F15%2F11%2Fp11020[36] J. B. Birks, “Scintillations from organic crystals: Specific fluorescence and relative response to different radiations,”
Proceedings of the PhysicalSociety. Section A , vol. 64, no. 10, pp. 874–877, oct 1951.[37] J. Birks,
The Theory and Practice of Scintillation Counting . PergamonPress, 1964, ch. 9, p. 321.[38] N. Zaitseva, B. L. Rupert, I. Pawełczak, A. Glenn, H. P.Martinez, L. Carman, M. Faust, N. Cherepy, and S. Payne,“Plastic scintillators with efficient neutron/gamma pulse shapediscrimination,”
Nuclear Instruments and Methods in Physics ResearchSection A: Accelerators, Spectrometers, Detectors and AssociatedEquipment
NuclearInstruments and Methods in Physics Research Section A:Accelerators, Spectrometers, Detectors and Associated Equipment
IEEE Transactions on Nuclear Science ,vol. 64, no. 7, pp. 1801–1809, 2017.[41] S. Normand, B. Mouanda, S. Haan, and M. Louvel, “Study of anew boron loaded plastic scintillator (revised),”
IEEE Transactions onNuclear Science , vol. 49, no. 4, pp. 1603–1608, Aug 2002.[42] M. C. Miller, R. S. Biddle, S. C. Bourret, R. C. Byrd, N. Ensslin,W. C. Feldman, J. J. Kuropatwinski, J. L. Longmire, M. S. Krick, D. R.Mayo et al. , “Neutron detection and applications using a BC454/BGOarray,”
Nuclear Instruments and Methods in Physics Research SectionA: Accelerators, Spectrometers, Detectors and Associated Equipment ,vol. 422, no. 1-3, pp. 89–94, 1999.[43] W. C. Feldman, G. F. Auchampaugh, and R. C. Byrd, “A novelfast-neutron detector for space applications,”