Elimination of the effect of internal activity in LaCl3:Ce scintillator
aa r X i v : . [ phy s i c s . i n s - d e t ] J un Elimination of the effect of internal activity in LaCl :Cescintillator D. Chattopadhyay ∗ , Sathi Sharma, M. Saha Sarkar Saha Institute of Nuclear Physics, HBNI, 1/AF Bidhannagar, Kolkata - 700 064, India
Abstract
The Lanthanum Halide scintillator detectors have been widely used for nuclearspectroscopy experiments because of their excellent energy and time resolutions.Despite having these advantages, the intrinsic α and β contaminations in thesescintillators pose a severe limitation in their usage in rare-event detections. Inthe present work, pulse shape discrimination (PSD) with a fast digitizer has beenshown to be an efficient method to separate the effect of α contamination fromthe spectrum. The shape of the β spectrum has been generated with the helpof Monte Carlo based simulation code, and its contribution has been eliminatedfrom the spectrum. The reduction in the background events generated by bothintrinsic β and α activities has been demonstrated. The present study willencourage the application of these detectors in low cross-section measurementexperiments relevant to nuclear astrophysics.
1. Introduction
Commercially available La-halide detectors have been remarkably success-ful in the field of radiation measurement. Cerium-activated La-halide detectoroffers brilliant light output, excellent energy resolution, a fast response, excel-lent linearity [1, 2, 3, 4, 5, 6, 7] and a stable light output over a wide rangeof temperatures [8]. For Cerium-activated LaCl detector, the light output is ∼
50 photons/keV with energy resolution as low as 3.1% at 662 keV. It has fastresponse time with principal decay constant ∼
20 ns, and non-linearity below7% from 60 to 1300 keV [9, 10]. This combination of features makes La-halidedetectors useful for low energy nuclear spectroscopy experiments [11], time offlight measurements [12] and medical imaging purposes [13]. La-halide detec-tors have been used for the lifetime measurements of unstable nuclei [14, 15],and even for detecting the fusion neutrons [16]. Furthermore, due to the highZ of Lanthanum (Z=57) and high density of the crystal (for, LaCl , ρ =3.85g/cm ), large volume La-halide detectors can be useful for detection of high ∗ Corresponding author.
Email address: [email protected] (D. Chattopadhyay )
Preprint submitted to Elsevier June 19, 2020 nergy γ -rays (up to 20 MeV). However, the self-activity of the La-halide detec-tor has been observed to be a major issue that reduces the detector sensitivityand interferes with the γ rays of interest in nuclear physics experiments andcomplicates data analysis.Natural Lanthanum is composed of stable La with 99.91% abundance.The radioactive
La with half-life 1.05 × y contribute to the remaining0.09 % of the abundance [17]. Since the separation between the two isotopesis almost impossible, the contamination due to La is present in all La-halidebased scintillators.
La isotopes decay by electron capture (e − ) into Bawith 66.4 % probability and the remaining 33.6 % decay by β − - emission into Ce. In both cases,
La decays into an excited state of the correspondingdaughter nucleus resulting in the emission of subsequent γ rays of equivalentenergy. For the electron capture process, 1436 keV γ -ray along with 32-38 keVcharacteristic X-rays of Ba is emitted. Whereas, in the second case, the emissionof 789 keV γ -ray in coincidence with the β -continuum having endpoint energyof 255 keV is observed.Moreover, actinium is a chemical analogue to lanthanum, i.e. , they havevery similar chemical properties. All the actinium isotopes are radioactive. Thelongest-lived actinium isotope is Ac having a half-life of 21.772 y. La-halidebased scintillators mostly contain actinium impurities, which contribute to the α contaminations due to the decay of long-lived Ac isotopes. The inher-ent radioactivity results in an intrinsic background of about 1-2 counts/cm s[18]. As the α / β energy loss ratio is less than 1, which is typical of nearly allscintillators, the α peaks appear at lower energies than their actual energieswhen the scale is calibrated from γ -ray (or fast electron) energies [18]. The self-activity background due to α decay from Ac-series nuclei appears at a rangeof 1.5-3 MeV in the energy spectrum. The contribution of Ac componentfrom impurities has been significantly reduced with the improved techniques ofgrowing the lanthanum halides. This inevitable background can be negligible ifthe count rate of actual events is sufficiently larger than self-activity backgroundevents. However, it is difficult to estimate the real event rate for a low countrate experiment.In general, for high energy γ -ray measurements, large volume La-halide de-tectors can be useful, but the event rate of self-activity increases with crystalvolume. For smaller crystals, the presence of the 1436 keV peak indicates that asignificant fraction of the X-rays escape without detection. However, for largercrystals, these X-rays are summed with 1436 keV γ s to show a dominant sumpeak at about 1468 keV [18]. Using Cerium bromide instead of La-halide, onecan obtain higher radio-purity, but CeBr has worse energy resolution comparedto other La-halide scintillator [17]. Therefore, research on self-activity elimina-tion has significant importance for the La-halide detector to overcome the limitsof their applicability in low energy nuclear astrophysics experiments where theevent rate is low.Several studies have been done to study the internal activity of these de-tectors and find a way to eliminate them from the spectrum. Hartwell andGehrke [19] identified the presence of Ac and its daughters as contamina-2ion of α -emitting nuclides in LaCl :Ce. Milbrath et al. [20] also confirmedthe presence of Ac as the source of α contamination in LaCl :Ce throughcoincidence measurements. Szucks et al. [21] in their paper have also pointedout the demerits of the usability of LaBr detector in low energy nuclear astro-physics experiments within the desired energy range because of the mixing ofcontamination due to internal activity with the real counts. Studies by Hoel etal. [22] and Crespi et al. [23] suggested the existence of pulse shape differencesbetween α and γ pulses. Hoel et al. [22] have also observed that for LaCl :Ce,the difference in pulse shapes is modest but for LaBr the difference is too smallto be useful. Crespi et al. [23] by using the charge comparison method with afast digitizer achieved the suppression of intrinsic α background. The radioac-tive decays of La in LaBr :Ce and the shape of the internal β spectrum havebeen measured by Quarati et al. [24]. Recently, the shape of intrinsic alphapulse height spectra in lanthanum halide scintillators have been studied in detailby Wolszczak and Dorenbos [25]. Despite these efforts, the overall quantitativeunderstanding of the effect of internal activity in the spectrum of the La-halidedetector is still limited now. Lanthanum halide scintillators, namely LaCl :Ceand LaBr :Ce, show very similar trends in terms of internal activity and, inprinciple, the results obtained using one of them are directly applicable to theother. In the present work, the internal radioactivity of a LaCl :Ce detector hasbeen studied.This study aims to provide a deeper understanding of the internal α and β activities in LaCl and find a way to eliminate them from the spectrum so thatthe detector can be useful for the low energy capture cross-section measurementsrelevant to nucleosynthesis. In this paper, we present the pulse shape discrimi-nation (PSD) technique to eliminate the effect of internal α contamination fromthe spectrum in LaCl detector. The PSD parameter is optimized for betterseparation between α and γ pulses. Besides, the Monte Carlo based simulationcode using GEANT4 toolkit has been developed to eliminate the effect of β activity from the spectrum.The paper is organized as follows. In section 2, the details of the experimen-tal setup is discussed. The different techniques used for the elimination of theinternal activity are discussed in section 3. In section 4, results obtained fromthe different methods have been analyzed and discussed. Finally, the summaryand conclusions are addressed in section 5.
2. Experimental Details
In this work, an 1 ” × ” cylindrical LaCl :Ce scintillator crystal from Saint-Gobain coupled with a Hamamatsu photo-multiplier tube (PMT) was used.The PMT has been biased to -1800 V. The anode signal was processed by aCAEN digitizer DT5730 (sampling rate 500 MHz). The data were acquired inlist mode, i.e. , event by event mode, and further analyzed in off-line by LAMPS[26] data analysis software. Low activity Co,
Eu,
Ba,
Cs,
Amand
Bi sources were used for energy calibration purposes. To minimize thecontribution of room background γ -radiation in the spectrum, the detector is3
00 1000 1500 2000 2500 3000 ke V ( ec ) ke V : b k g ke V : b k g C oun t s Energy (keV)
Bare detector Detector within Pb shield contamination
789 keV + continuum c on t i nuu m Figure 1: (a) Normal room background spectrum acquired with and without lead shielding.The room background γ -ray peaks, 1460 keV from K , and 2615 keV (hidden under thealpha contamination peaks) from T h decay series are shown. The intrinsic backgroundcomponents arising from the decay of La : β -continuum, the summing bump of 789 keV( β − decay) with the β continuum, 1436 keV γ - peak (electron capture) are shown. The peaksarising from α -contamination in the detector from Ac impurity are also indicated in thefigure. kept inside a lead shielding. The spectra with and without Pb shielding areshown in Fig. 1.
3. Techniques used for elimination of the internal activity
When an energetic particle or photon is incident on a scintillator crystal,it deposits its energy in the medium. The scintillator gets excited and emitsscintillation radiation. The energy loss per unit thickness (dE/dx) inside thescintillator material varies for different incoming particles. Hence, the lumines-cence rates for different particles result in pulse shape differences. Two expo-nentials with two-time components [27] usually characterize the time evolutionof the intensity of the scintillation light. One is a fast or prompt component,and the other is a slow or delayed component. The time evolution of the num-ber of emitted scintillation photons N from a single scintillation event can bedescribed by a linear superposition of two components given by N = Aτ f e − tτf + Bτ s e − tτs (1)4here, τ f and τ s are the fast and slow decay times. The majority of thescintillation intensity is contained in the prompt component. However, thelong-lived or the slow component has an important implication as the fractionin it often depends on the nature of the incident particle. This dependence canbe utilized to differentiate between particles of different kinds that deposit thesame energy in the detector [18]. This technique is well known as pulse shapediscrimination (PSD). It is widely applied to eliminate unwanted events in amixed radiation field. The light emission intensity of a LaBr :(Ce) scintillatoris accurately modeled with a single fast component exponential decay. However,that of a LaCl :(Ce) crystal is represented well by a two-component exponentialdecay with t f ≃
26 ns and t s ≃
550 ns at 20 o C. Moreover, the ratio of B/A isapproximately 0.25 for γ -rays at 20 o C [28].
Figure 2: Typical waveforms of α and γ obtained from LaCl detector having the same totalcharge deposited. The short and long gates are shown in the figure. The inset highlights thedifference between an α and a γ pulse. In pulse shape discrimination with scintillators, the most frequently usedtechnique is the charge integration method, which determines the amount ofdelayed light output with respect to the total light output for each event toidentify the type of the corresponding ionizing radiation. DPP-PSD firmwareprovided with the CAEN 5730 digitizer [29] is based on this method. The PSDparameter is then extracted in event-by-event mode using the FPGA of thedigitizer as,
P SD = ( Q l − Q s ) Q l (2)where, Q l and Q s are the integrated charge within the long gate and short5ate, respectively, as shown in Fig. 2. The PSD is the ratio of charge deposited atthe tail part of the pulse ( Q l - Q s ) to the total charge deposited in the full pulse( Q l ). The short window (80 ns in the present case) is to be chosen in such a waythat the ratio of the tail to the total pulse would be most effective for particleidentification. For example, α and γ have different interaction mechanismswith the detector material; hence they have different responses in the detectors(shown in the inset of Fig. 2). For the same total charge deposited, α pulses arenarrower and sharper compared to γ pulses. The choice of short and long gatesis shown in Fig. 2. With this choice of gates, the PSD parameter is larger for α s than for γ s.This technique has been used to eliminate the contributions of internal ac-tivity due to α contamination from the spectrum. A two-dimensional plot ofPSD vs Channel no. ( Q l , i.e. , energy) shown in Fig. 3 distinguishes the α contamination events from the γ ’s and β ’s. Figure 3: 2D plot of PSD vs Channel No. (Energy) for normal room background for thePb-shielded detector.
LaCl :Ce detector has an internal β activity that can not be eliminated usingthe pulse shape discrimination as the shape of the pulses looks very similar forboth γ and β . For the elimination of the effect of internal β activity from6he γ spectrum of LaCl :Ce, a Monte Carlo based simulation code has beendeveloped using the GEANT4 toolkit [30]. In the present work, the GEANT4toolkit version 4.10.0 has been implemented.The three main classes in our code are detector construction, physics list,and primary generator action. In the detector construction class, the geometryand materialistic information of the detector and its encasing is defined as pro-vided by the manufacturer. The geometry consists of a small 1 ” × ” cylindricalLaCl detector surrounded by MgO reflector, and optically coupled to a Bi-alkaliPhoto-cathode through a Quartz PMT Window. In the physics list, the neces-sary physics processes, such as, G4EmStandardPhysics, G4DecayPhysics, andG4RadioactiveDecayPhysics are included. The G4EmStandardPhysics classcontains the three standard electromagnetic processes like Compton scattering,Photoelectric process, and Pair production. The G4DecayPhysics constructorhandles the decay channels for all unstable particles defined in the physics list.The same process is assigned to all unstable particles. The G4RadioactiveDecay-Physics contains the basic features of the radioactive decay of nuclei. In, primarygenerator action class, the general particle source (GPS) module has been usedas a particle generator to create different shapes with a specific position, an-gle, and energy distribution, etc. For the simulation of internal β activity, acylindrical source of same size as the detector has been used.The simulations were carried out for a large number of events (10 ) to reducethe statistical uncertainty. The response function of the internal β activities wasgenerated using the simulation code. The detailed procedure is described below.No. of La atoms (N ) inside the LaCl detector has been estimated fromthe following relation. N = N A .ρ. ( πr h ) . ( f ) A (3)where N A is the Avogadro number, ρ is the density of the detector, r isthe radius, and h is the height of the cylindrical detector, f is the fractionalabundance of La in mass A of LaCl .The number of La atoms (N) remaining after time τ can be calculated as: N = N exp ( − λ.τ ) (4)Then the number of La atoms which have decayed in this time would be: N − N = N (1 − exp ( − λ.τ )) (5)It is known that in 66.4% times La decays to
Ba via an electron capturewith the emission of ∼ γ line. Hence in the elapsed time τ no. ofemitted γ s (N ) having an energy of ∼ N = 0 . ∗ N (1 − exp ( − λ.τ )) (6)The simulated spectra (Fig. 4) have been normalized to reproduce the N counts observed in the experimental spectra to get the shape of the internal β γ - peakin N has been minimized as far as possible.The γ energy spectra of LaCl detector for chosen energies have been gen-erated by assuming a simplified expression of the peak shapes. σ ( E ) = a + b √ E (7)where, σ ( E ) is the standard deviation of the peak shape of the γ peak ofenergy E in the γ spectrum. The standard deviation is obtained from the fullwidth at half maximum (FWHM) of a γ peak. The value of the parameters a and b have been estimated by fitting the variation of FWHM of γ peaks as afunction of peak energies obtained from spectra of different radioactive sources( Co,
Cs and
Bi).
4. Results and Discussions
Fig. 1 shows the background spectrum obtained from LaCl detector ac-quired with and without lead shielding. Due to the inherent radioactivity in thedetector, the background spectrum is dominated by components of La and
Ac decay. In the electron capture decay mode of
La, a γ -ray (1436 keV)and X-ray (32 keV-K α and 38 keV-K β ) from Ba are emitted. The 1436 keV γ and the 1460 keV γ from the K present in the room background can not beresolved. For relatively larger detectors the correlated γ -ray and K X-rays(32-38keV) of Ba give rise to a sum-peak at ∼ β − decay, a continuum β − spectrum till 255 keV followed aby 789 keV γ -ray from Ce are emitted. The correlation between β particles(till 255 keV) and γ -ray (789 keV) of Ce generates a summed structure of β continuum with γ -ray. The continuum structure starts at 789 keV and spreadsto high energy at 1044 keV [3, 31]. The remaining β continuum is observedat the low energy side from 0 keV to 255 keV. In between 255 keV to 750 keVthe Compton continuum from the 789 keV and 1436 keV γ -rays are observed.Above 1700 keV, the presence of α contaminant peaks are observed. Thesepeaks originate from the α ’s emitted from the decay of Ac contamination.The α ’s produce a broad response with several peaks from roughly 1.7-2.3 MeVee(MeV electron-equivalent) [31]. Thus the background spectrum obtained fromLaCl detector is quite complicated and background components due to naturalradioactivity are mostly hidden under the intrinsic background. Bi spectra: Elimination of the effect of α contamination from thespectrum The energy spectra using
Bi source using a Pb-shielded detector acquiredwith DT5730 digitizer has been shown in Fig. 5. It is known that
Bi sourceemits three γ s at 570 keV, 1064 keV, and 1770 keV. From Fig. 5 the intrinsicactivity of LaCl :Ce is found to be dominant in the spectrum. Although 5708
500 1000 1500 C oun t s Simulated beta activity
Energy (keV)
Figure 4: Shape of the β spectrum obtained from the simulation. and 1064 keV peaks of Bi are seen, 1770 keV peak is invisible. The 1770 keVpeak is hidden under the α peaks arising from the α activity in the detector.Therefore, proper elimination of the effect of internal activity from the spectrumis needed.The effect of the contamination due to α activity is eliminated from thespectrum by using pulse shape discrimination. The PSD has been calculatedevent-by-event from the formula mentioned above. Fig. 3 shows a typical two-dimensional plot of PSD vs Channel No. (integral of Long Gate). The plotshows that one can separate the contributions for the α events from the γ events. The α subtracted spectrum has been generated by plotting only thoseevents which have PSDs corresponding to γ events. The spectra thus generatedis shown in Fig. 5. After the invoking PSD gate, the 1770 keV peak of Bisource is observed clearly in the α contamination subtracted spectrum. The detector is kept inside a 2cm lead shielding to reduce background γ radiation. In this way, the contributions from the room background have beenreduced by a factor of ∼
11 % (Fig. 1). However, the spectra are not free fromintrinsic background. Elimination of internal α and β contamination from thespectrum is required to reduce the background further.9
500 1000 1500 2000 250010 ke V : ec 1460 ke V : b k g ke V : b k g ke V ke V ke V Raw SpectrumEnergy (keV) C oun t s C oun t s Energy (keV)
Spectra gated by PSD for gammas
Bi decay - contamination X - r ay Figure 5: Raw (inset) and α -activity subtracted spectra for Bi decay.
Pulse shape discrimination method has been adopted to eliminate the effectof α activity from the spectrum. Using the two-dimensional plot of PSD vsChannel No. (integral of Long gate) for background spectra the contribution of α contamination is eliminated, as shown in Fig. 6 (b).The α subtracted spectrum still contains the effect of β contamination. Thecontribution of β activities and the shape of the spectrum due to β contamina-tion alone has been simulated using GEANT4 based simulation code. In Fig. 4,the shape of the spectrum due to the β contamination is shown. Finally, thetotal intrinsic activity, i.e. , the sum of the activities due to α and β decays, issubtracted from the raw spectrum. The reduction factor (Fig. 6) to estimatethe suppression in room background, as well as the intrinsic background, iscalculated. The time normalized background events in a Pb-shielded detectorafter elimination of intrinsic activities has been reduced by ∼
35% compared toa bare detector. 10
000 1500 2000 2500 3000
800 1000 1200 1400 1600200400600200 400 600
500 1000 1500 2000 2500 3000 (b) (d)(c) C oun t s Detector within Pb shield Intrinsic alpha activity eliminated Intrinsic alpha and beta activities eliminated1460 keV
Energy (keV)(a)
Figure 6: (a) Normal room background spectrum acquired with lead shielding compared withthat after elimination of α contamination, and also with that after elimination of both α and β activities. The room background γ -ray peaks, 1460 keV from K , and 2615 keV from T h decay series are clearly seen after the elimination of the intrinsic background. (b) The effectof removal of α activity from the spectrum, (c) and (d) results of elimination of β activityfrom an α activity eliminated spectrum at (c) low energies ( <
700 keV) and (d) in the rangefrom 700 -1700 keV. . Summary and conclusions In the present work, intrinsic activity due to α -decay events in the LaCl :Cescintillator has been eliminated by the pulse shape discrimination method usinga fast digitizer. The shape of the β continuum has been simulated using theMonte Carlo based simulation code via GEANT4. The simulated spectra afterproper normalization has been subtracted from the raw spectra for the elimi-nation of the effect of β activity. The applicability of the La-halide detectorsfor the low cross-section measurement experiments in nuclear astrophysics hasbeen improved by reduction of background events by 35%. The discriminationmethod presented in this paper would be especially useful for LaCl :Ce detec-tors, which have a slow part in the scintillation pulse useful for pulse shapediscrimination.
6. Acknowledgements
We thank Prof. C. C. Dey of Saha Institute of Nuclear Physics, Kolkatafor providing us the detector during the experiment and Mr. S. Karan of SahaInstitute of Nuclear Physics, Kolkata for the technical help.
References [1] C. M. Rozsa, Peter R. Menge, M. R. Mayhugh, Performance summary:BrilLanCe
T M scintillators LaCl and LaBr :Ce crystals: fast, efficient and high-energy-resolution scintillators,Nucl. Instr. and Meth. A , 254 (2002).[3] Peter R. Menge, G. Gautier, A. Iltis, C. Rozsa, V.Solovyev, Performance of large lanthanum bromide scintillators,Nucl. Instr. and Meth. A , 6 (2007).[4] J. Glodo, W. W. Moses, W. M. Higgins, E. V. D. van Loef, P. Wong, S. E.Derenzo, M. J. Weber, K. S. Shah, Effects of Ce concentration on scintil-lation properties of LaBr :Ce, IEEE Trans. Nucl. Sci. , 1805 (2005).[5] F. Quarati, A. J. J. Bos, S. Brandenburg, C. Dathy, P. Doren-bos, S. Kraft, R.W. Ostendorf, V. Ouspenski, Alan Owens, X-rayand gamma-ray response of a 2”x 2” LaBr :Ce scintillation detector,Nucl. Instr. and Meth. A , 115 (2007).[6] W. W. Moses, Current trends in scintillator detectors and materials,Nucl. Instr. and Meth. A , 123 (2002).127] K. S. Shah, J. Glodo, M. Klugerman, W. M. Higgins, T.Gupta, P. Wong, High energy resolution scintillation spectrometers,IEEE Trans. Nucl. Sci. , 2395 (2004).[8] M. Moszynski, A. Nassalski, A. Syntfeld-Kazuch, T. Szczesniak,W. Czarnacki, D. Wolski, G. Pausch, J. Stein, Temperaturedependences of LaBr (Ce), LaCl (Ce) and NaI(Tl) scintillators,Nucl. Instr. and Meth. A , 739 (2006).[9] E. V. D. van Loef, L. P. Dorenbos, C. W. E. van Eijk, K. Kramer, H. U.Gudel, Scintillation properties of LaCl :Ce crystals: fast, efficient, andhigh-energy resolution scintillators, IEEE Trans. Nucl. Sci. , 341 (2001).[10] K. S. Shah, J. Glodo, M. Klugerman, L. Cirignano, W. W. Moses,S. E. Derenzo, M. J. Weber, LaCl :Ce scintillator for γ -ray detection,Nucl. Instr. and Meth. A , 76 (2003).[11] E. V. D. van Loef, P. Dorenbos, C. W. E. van Eijk, High-energy-resolutionscintillator: Ce activated LaBr , Appl. Phys. Lett. , 1573 (2001).[12] Dennis R Schaart, Stefan Seifert, Ruud Vinke, Herman T van Dam, Pe-ter Dendooven, Herbert Lhner, Freek J Beekman, LaBr(3):Ce and SiPMsfor Time-Of-Flight PET: Achieving 100 Ps Coincidence Resolving Time,Phys. Med. Biol. , N179 (2010).[13] M. E. Daube-Witherspoon, S. Surti, A. Perkins, C. C. M. Kyba, R. Wiener,M. E. Werner, R. Kulp, J. S. Karp, The imaging performance of a LaBr -based PET scanner, Phys. Med. Biol. , 45 (2010).[14] E. R. White, H. Mach, L. M. Fraile, U. Koster, O. Arndt, A. Blazhev, N.Boelaert, M. J. G. Borge, R. Boutami, H. Bradley, et al. , Lifetime mea-surement of the 167.1 keV state in Ar, Phys. Rev. C , 057303 (2007).[15] N. Marginean, D. L. Balabanski, D. Bucurescu, S. Lalkovski, L. Atanasova,G. Cata-Danil, I. Cata-Danil, J. M. Daugas, D. Deleanu, P. Detis-tov, G. Deyanova, et al. , In-beam measurements of sub-nanosecond nu-clear lifetimes with a mixed array of HPGe and LaBr :Ce detectors,Eur. Phys. J. A , 329 (2010).[16] C. Cazzaniga, M. Nocente, M. Tardocchi, G. Croci, L. Giacomelli,M. Angelone, M. Pillon, S. Villari, A. Weller, L. Petrizzi, et al. ,Response of LaBr (Ce) scintillators to 2.5 MeV fusion neutrons,Rev. Sci. Instrum. , 123505-1 (2013).[17] Juris Meija, Tyler B. Coplen, Michael Berglund, Willi A. Brand,Paul De Bievre, Manfred Groning, Norman E. Holden, Johanna Ir-rgeher, Robert D. Loss, Thomas Walczyk, Thomas Prohaska, Iso-topic compositions of the elements 2013 (iupac technical report),Pure Appl. Chem.
88 (3) , 293 (2016).1318] G. F. Knoll, Radiation Detection and Measurement, 2nd Edition, JohnWiley, New York, 1989.[19] J. Hartwell, R. Gehrke, Observations on the background spectra off ourLaCl (Ce) scintillation detectors, Appl. Radiat. Isot.
63 (2) , 223 (2005).[20] B. D. Milbrath, R. C. Runkle, T. W. Hossbach, W. R. Kaye, E. A. Le-pel, B. S. McDonald, L. E. Smith, Characterization of alpha contamina-tion in lanthanum trichloride scintillators using coincidence measurements,Nucl. Instr. and Meth. A
547 (23) , 504 (2005).[21] Tamas Szucs, Daniel Bemmerer, Thomas Cowan, Detlev Degering, ZoltanElekes, Zsolt Fulop, Gyorgy Gyurky, Arnd Junghans, Matthias Kohler,Michele Marta, et al. , Eur. Phys. J. A , 8 (2012).[22] C. Hoel, L. G. Sobotka, K. S. Shah, J. Glodo, Pulse-shape discriminationofLa halide scintillators, Nucl. Instr. and Meth. A , 205 (2005).[23] F. C. L. Crespi, F. Camera, N. Blasi, A. Bracco, S. Brambilla, B.Million, R. Nicolini, L. Pellegri, S. Riboldi, M. Sassi, et al. , Alpha-gamma discrimination by pulse shape in LaBr :Ce and LaCl :Ce,Nucl. Instr. and Meth. A , 520 (2009).[24] F. G. A. Quarati, I. V. Khodyuk, C. W. E. van Eijk, P. Quarati, P.Dorenbos, Study of La radioactive decays using LaBr scintillators,Nucl. Instr. and Meth. A , 46 (2012).[25] W. Wolszczak, P. Dorenbos, Shape of intrinsic alpha pulse height spectra inlanthanide halide scintillators, Nucl. Instr. and Meth. A :Ce scintillators: Experiment and model,Phys. Rev. B , 184302 (2007); J. E. McFee, C. M. Mosquera, A. A.Faust, Comparison of model fitting and gated integration for pulse shapediscrimination and spectral estimation of digitized lanthanum halide scin-tillator pulses, Nucl. Instr. and Meth. A , 105 (2016); M. Moszynski,A. Nassalski, A. Syntfeld-Kazuch, T. Szczesniak, W. Czarnacki, D. Wolski,G. Pausch, J. Stein, Temperature dependences of LaBr (Ce), LaCl (Ce)and NaI(Tl) scintillators, Nucl. Instr. and Meth. A , 739 (2006).[28] J. E. McFee, C. M. Mosquera, A. A. Faust, Comparison of modelfitting and gated integration for pulse shape discrimination andspectral estimation of digitized lanthanum halide scintillator pulses,Nucl. Instr. and Meth. A , 105 (2016).[29] User Manual WP2081; Digital Pulse Processing in Nuclear Physics;Overview of CAEN DPP algorithms: Rev. 4 - June 12th, 2017; ApplicationNote AN3250, Pulse Shape Discrimination with different CAEN digitizersrunning DPP-PSD firmware. 1430] S. Agostinelli, et al. , Geant4a simulation toolkit,Nucl. Instr. and Meth. A506