^{210}Pb measurements at the André E. Lalonde AMS Laboratory for the radioassay of materials used in rare event search detectors
Carlos Vivo-Vilches, Benjamin Weiser, Xiaolei Zhao, Barbara B.A. Francisco, Razvan Gornea, William E. Kieser
2210
Pb measurements at the André E. Lalonde
AMS Laboratory for the radioassay of materialsused in rare event search detectors
Carlos Vivo-Vilches a,b,* , Benjamin Weiser a,b , Xiaolei Zhao c,d ,Barbara B.A. Francisco c , Razvan Gornea a , and William E. Kieser c,da Department of Physics, Carleton University, 1125 Colonel By Drive, Ottawa,ON K1S 5B6, Canada b Arthur B. McDonald Institute, 64 Bader Lane, Queen’s University, Kingston,ON K7L 3N6, Canada c A. E. Lalonde AMS Laboratory, University of Ottawa, 25 Templeton St.,Ottawa, ON K1N 6N5, Canada d Department of Physics, University of Ottawa, 25 Templeton St., Ottawa,ON K1N 6N5, Canada
Abstract
Naturally occurring radionuclide
Pb ( T / =22.2 y) is an important source ofbackground in rare event searches, such as neutrinoless double- β decay and dark matterdirect detection experiments. When a sample mass of hundreds of grams is available, γ -counting measurements can be performed. However, there are other cases where onlygrams of sample can be used. For these cases, better sensitivities are required.In this paper, in collaboration with the Astroparticle Physics group at CarletonUniversity, the capabilities of the A.E. Lalonde AMS Laboratory at the University * Corresponding author. Current address at Helmholtz-Zentrum Dresden-Rossendorf (HZDR), BautznerLandstrasse 400, Dresden, 01328, Germany
E-mail: [email protected] (C. Vivo-Vilches) a r X i v : . [ phy s i c s . i n s - d e t ] F e b f Ottawa for Pb measurements are discussed. PbF and PbO targets were used,selecting in the low energy sector, respectively, (PbF ) - or (PbO ) - ions.For fluoride targets, the blank Pb/
Pb ratio was in the 10 -14 to 10 -13 range, butcurrent output was lower and less stable. For oxide targets, current output showed bet-ter stability, despite a significant difference in current output for commercial PbO andprocessed samples, and background studies suggested a background not much higherthan that of the fluoride targets. Both target materials showed, therefore, good perfor-mance for
Pb AMS assay.Measurements of Kapton films, an ultra-thin polymer material, where masses avail-able are typically just several grams, were performed. 90% C.L. upper limits for the
Pb specific activity in the range of 0.85-2.5 Bq/kg were established for several KaptonHN films.
Keywords:
Pb contamination, Accelerator Mass Spectrometry, Rare event searches,Astroparticle physics, Radiopurity
Radioassay of materials is one of the most important tasks in the design and constructionof high sensitivity detectors for rare event searches, such as direct dark matter detection[1, 2, 3, 4], and the observation of neutrinoless double-beta decay [5, 6, 7, 8, 9, 10].Assay of naturally occurring radionuclides producing decay chains, like
U, is especiallyimportant, since each decay of the parent radionuclide causes the production of severalparticles, each of which induces background for the experiment. These include γ , β , and α particles. In secular equilibrium, all nuclides in the chain show an activity similar to the par-ent radionuclide. It is quite common, however, that one of the U daughter radionuclides,
Pb, does not fulfill this condition.In samples where all the radionuclides in the
U decay chain are in secular equilib-rium, the equal activity indicate that the
U concentration is 2 × times higher thanthat of Pb. However, the secular equilibrium can be broken by the noble gas
Rn( T / =3.8235 d). This radionuclide outgasses from any material where U is present andconsequently is present in air. The deposition and/or plating-out of
Rn daughters fromthe air to materials which only have an extremely low concentration of uranium, potentiallycould increase the concentration of
Pb above the equilibrium with
U. If this processtakes place during the production of the material, the
Pb can even be present in the bulkof the final material, and not just on its surface.2he main impact of
Pb in rare event searches comes from the α -decay of its daughternuclide, Po. The decay itself can be a direct source of background in experiments lookingfor weakly interacting massive particles (WIMPs), one of the candidates for dark matter. Inaddition, these α particles can produce neutrons through ( α ,n) reactions on low-Z elements.Because of the short range of α particles, this interaction typically takes place within thesame material where the α -decay occurs.Neutrons are also a direct source of background in WIMP searches, because they depositenergy through nuclear collisions which mimic the expected signals from these WIMPs [11].In experiments searching for neutrinoless double- β decay, α particles are not a direct source ofbackground, but neutrons produced by ( α ,n) reactions activate materials, thereby increasingthe electron recoil background.Taking into account its relatively short half-life (22.2 y), Pb is typically measured byradiometric techniques [12]:• Direct γ -counting of the 46.539 keV γ produced in 4.25% of the Pb decays. Thereare 2 challenges related to this measurement: the low intensity of this γ decay; andthe very low energy of the photon, for which the detection efficiency of germaniumdetectors is low.• β -counting of the decay of the Pb immediate daughter,
Bi ( T / = 5 .
012 d ). Thismethod requires extensive chemical preparation to separate
Bi from other radionu-clides, since most of them could cause a background in the β continuous spectrum.Because of the much shorter half-life of Bi, secular equilibrium with
Pb can beassumed.• α -counting of the Po decay ( T / = 138 .
376 d ). The main disadvantage of thistechnique is that secular equilibrium cannot be always assumed. If
Pb is depositedin the material only a few days before the measurement,
Po activity will not be inequilibrium yet, and will be much lower. In other cases, the chemical production ofthe sample can lead to
Po activities higher than
Pb ones, if the processes have ahigher chemical efficiency for polonium. Direct
Pb measurements, therefore, will bemore useful. Po α detection is useful and efficient for geological applications, whereequilibrium can be assumed. When Pb is measured by this technique, the originalamount of
Po in the sample has to be chemically removed. Then, several monthsare required for the
Pb to decay, leading to a measurable
Po activity [13].For sensitive assays, these techniques require large amounts of sample, i.e. several hun-dreds of grams. For some materials, especially low density polymers, obtaining this quantitywould be an issue. For instance, a 30 cm ×
30 cm film of 25.4 µm thick Kapton weights only3.25 g. 3CP-MS allows rapid
Pb measurements, but with limited sensitivity, due to molecu-lar background [14]. Therefore, for rapid and ultra-sensitive
Pb assay, accelerator massspectrometry (AMS) is an alternative approach worth exploring. The possibility of employ-ing AMS, using lead fluoride targets, to measure
Pb was first discussed about 20 yearsago [15, 16]. Most recently, the André E. Lalonde AMS Laboratory (AEL-AMS) has per-formed similar studies [17, 18]. These have corroborated the observation that the extractionof (PbF ) - ions from PbF targets is a good choice, offering a good ionization efficiency[19, 20].Other authors have shown that a relatively high and stable current of (PbO ) - ions canbe extracted when sputtering PbO samples mixed with silver powder [21]. Nevertheless,this type of target had not been used before for Pb AMS because of the injection of( Pb O O) - ions into the accelerator when selecting the ( Pb O ) - ion, potentiallyleading to a large, nearly isobaric, interference from Pb ions. Alternatively, fluorine hasonly one stable isotope, F, so the only lead fluoride anion injected when selecting a massof 267 u is (
PbF ) - . A post-accelerator spectrometer with high resolution, as in the caseof the AEL-AMS system, sufficiently reduces any background from molecular fragments.In this work, the use of the AEL-AMS system is evaluated for the Pb assay in mate-rials considered for the construction of low background detectors, with particular attentionto detectors for rare event searches at SNOLAB. The performance parameters of
Pb mea-surements in the 3 MV AMS system, for the two different target materials, PbF or PbO,are presented and discussed in section 2, demonstrating that AMS is one of the most appro-priate techniques for Pb assay. In section 3, the interest of
Pb AMS measurements inpolymer materials for rare event searches is addressed, as well as the chemical method andfirst results for Kapton polyimide films, showing that specific activities below 1 Bq/kg canbe measured using less than 2 g of sample.
Pb measurements at the AEL-AMSfacility
All the measurements are performed with the 3 MV system of the AEL-AMS Laboratoryat the University of Ottawa [22]. An illustration of the set-up of this system for
Pbmeasurements is presented in Figure 1.Samples are inserted as PbF or PbO, and the (PbF ) - or (PbO ) - ion, respectively, isselected by the low energy spectrometer. In both cases, the sample is mixed with silver4 igure 1: Illustration of the measurement set-up for
Pb measurements at the AEL-AMSsystem for each of the chemical species used.powder in a volume ratio of 1:1. The ions are accelerated using a terminal potential of2.5 MV. The Ar stripper gas pressure is 0.012 mbar. With the high energy analyzing magnet(radius = 2m), the charge state 3+ is selected; therefore, the final energy of the ions is9.47 MeV in the case of measurements with PbF , and 9.67 MeV with PbO.Following the high energy magnet, Pb ion current is measured with an off-axis Fara-day cup (FC). This natural isotope of lead is chosen rather than Pb as the offset Faradaycup in the
Pb position would block the
Pb beam. This allows the use of the sameterminal voltage for both the
Pb and
Pb isotopes and thus the use of the more efficientfast sequential injection method. The Pb ions continue along the beamline, throughthe electric analyzer (ESA) and the switching magnet, until they are detected in the gasionization chamber.All the pre-accelerator slits are set to ± ± Several parameters related to the
Pb measurements with the AEL-AMS system aresummarized in Table 1. When using commercial PbO targets, ( Pb O ) - currents are verystable, and typically in the range between 20 and 100 nA. When using processed PbO samplescurrents are stable as well, although with 5 times lower intensity. The lower currents in5 able 1: Parameters of
Pb measurements in the AEL-AMS system.
Pb/
Pb blankratio is presented in the next subsection.
Target material
PbF PbO
Negative stable isotope ion ( Pb F ) - ( Pb O ) - Negative ion current output (nA)
Charge state Terminal voltage (MV)
Stripper transmission (%) Accelerator to GIC optical transmission (%)
PbF ) - from PbF targets is less stable, being really high during the firstminute of sputtering, but declining exponentially during the early stage of the sputtering.After 20 min, ion current output stabilizes, but not to the same levels as in the case ofPbO targets. Even using the same mix of commercial PbF with silver, a high variability isobserved in the ion current output between targets, whether the targets are commercial leadfluoride or processed samples. In some cases ( PbF ) - current output reached several tensof nA; for others, after the first decay, it stayed at levels closer to 5 nA.Most efficiencies do not depend on the use of one or the other material. Transmissionfor the 3+ state in the Ar gas stripper is 8% and the optical transmission through the post-accelerator spectrometer to the detector is 25%. This optical transmission is estimated fromthe ratio between the directly measured Pb/
Pb ratio and the nominal
Pb/
Pb ratiofor reference samples. These reference samples are prepared by mixing:• ∼ Pb concentration of 1.71 × -16 mol/g, prepared as adilution of the NIST reference material SRM 4337 [23].• ∼ Pb-free Pb carrier solution with a Pb concentration of 37.3 mg/g. Thissolution was prepared by dissolving a 1 cm cube of ancient lead provided by PNNL(the sample number 5 in Ref. [24]) in dilute nitric acid. The final Pb concentrationwas measured by ICP-ES.Both masses are measured using a precision balance. Therefore, the final Pb/
Pbratio is approximately 1.5 × -12 , slightly changing from one target to another dependingon the exact masses of each solution. These masses are measured with a precision bal-ance. Table 2 shows the comparison between measured and nominal Pb/
Pb ratios for3 reference targets produced with this method.6 able 2:
Experimental/nominal ratio from 3 reference targets. The average experimen-tal/nominal ratio is 25.8 ± Pb ions must pass through 3 additional tightly set slits before enteringthe final detector, whereas the Pb ions enter directly into an off-axis Faraday cup with-out loss. In future development, this parameter could be improved after systematic study ofslits setting to balance background reduction and ion transmission. Nominal ratio ( × -12 ) Measured ratio ( × -12 ) Optical transmission (%) ± ± ± ± ± ± ± ± ± One of the most important sources of background in
Pb AMS measurements is theinterference of the nearly isobaric Pb ions. When using lead oxide samples, compared tolead fluoride, this interference is expected to be augmented by the injection of ( Pb O O) - ions with the ( Pb O ) - ions (242 u) in the pre-accelerator spectrometer. Nevertheless,the post-accelerator spectrometer of the AEL-AMS system is designed to have a high massresolution, which reduces the effect of this interference significantly.In order to produce the process blanks, the Pb-free Pb carrier solution described insec tion 2.2 above is used. A comparison of PbF and PbO targets produced from thissolution, with commercial ultra-pure PbF and PbO, respectively, provided an assessmentof the level of Pb contamination during the chemical processing of the samples. It also letus determine whether those commercial products could be considered for the blank material.Due to the relatively low ionization efficiencies, number of counts for blank samples aretypically very low.For PbF samples, the total number of counts from the blank targets could be as low as 1count. The blank Pb/
Pb ratio, before normalizing to the reference targets, is 1 × -14 .This has a huge statistical uncertainty. No systematic difference has been observed betweenPbF targets produced from the blank solution and commercial PbF , which shows that thismaterial is clean enough to be considered as blank material.First tests of the Pb/
Pb ratio from oxide samples using commercial PbO indicateda background one order of magnitude higher than that from the fluoride samples. In latermeasurements a systematic difference is observed between the PbO targets produced fromthe blank solution and the commercial PbO targets. As an example, the results for thetwo kinds of target during a more recent measurement are presented in Table 3. The lowercurrent output does not explain the much lower number of counts for processed targets. Even7hen
Pb/
Pb ratio for the process blanks includes a high uncertainty of a 97%, thereis quite a significant difference when compared with the
Pb/
Pb from the commercialPbO. These results suggest that the background for oxide samples is not much higher thanthat of fluoride, and the commercial PbO material should not be considered a blank.Including the correction to the reference targets, the blank
Pb/
Pb ratio for thefluoride targets is lower than 8.0 × -14 . Taking into account that 10 mg of Pb carrier persample are used (2.41 mg of Pb), this
Pb/
Pb ratio is equivalent to a
Pb activityof 0.56 mBq. In the case of PbO targets, the corrected blank
Pb/
Pb ratio is lower than2.0 × -13 , and equivalent to a Pb activity of 1.4 mBq. In Table 4, a comparison of theselevels with the backgrounds from other
Pb assay techniques is presented. AMS clearlyprovides a lower background than ICP-MS and most radiometric techniques. In comparison
Table 3:
Pb/
Pb ratio from 3 process blank PbO targets and a commercial PbO target.Total measurement time was 7570 seconds per target.
Target Number ofdetector counts Average Pb current (nA) Pb/
Pb*( × -14 ) Commercial PbO 29 4.91 14.0 ± ± *) This ratio does not include the correction of optical transmission using reference samples. Actualbackground is, therefore, 4 times higher. Table 4:
Pb specific activity derived the background measurements at the 3 MV AMSsystem compared with thaose of different radiometric techniques [13] and ICP-MS [14].
Technique
Pb background(mBq) Sample preparationtime (days) Measurementtime (min) Pb γ -counting 120 0 1000 Bi β -counting 20 10* 1000 Po α -counting 0.4 90-180* 1000ICP-MS 90 4 N/AAMS (PbF ) < < *) This sample preparation time includes the in-growth time required for the measured radionuclide toreach secular equilibrium with Pb. Po α -counting, the main advantage is that AMS requires a much shorter samplepreparation time, because of the long time (3-6 months) required for Po to reach secularequilibrium with
Pb. In addition, Po α -counting is limited to the detection of the Pbpresent at the surface of the material, while
Pb AMS gives information throughout thebulk of the material. α particles from the Pb chain
The most problematic contribution of
Pb to the background in rare event searches isdue to the α decay of its daughter, Po. This is mainly because of the ( α ,n) reactions whichmay occur when low-Z elements are present in the detector materials. Alpha particles from Po decay have a typical range of just several tens of µm in most solid materials. Therefore,( α ,n) reactions typically take place in the material where the Po decay occurred.Consequently, the impact on this background of the experiment from the
Pb concen-trations will depend on the material in which this
Pb is present. The neutron yield ofthe α particles from the Pb chain (mainly from
Po decay) in different materials wasstudied with the tool NeuCBOT [25]. In Table 5, some examples of these yields in materialscommonly used in rare event search detectors are presented. Focusing on the different kinds
Table 5:
Neutron yields of the α particles from the Pb chain in different materials,obtained with the NeuCBOT tool [25].
Material Formula Neutron yield (n/decay)
Acrylic (PMMA) [C O H ] n × -7 Aluminum Al 1.49 × -6 Kapton polyimide [C O N H ] n × -8 Polyethylene [C H ] n × -7 PVC [C H Cl] n × -8 Quartz SiO × -8 Sapphire Al O × -7 Silicon Si 1.17 × -7 Silicon carbide SiC 1.45 × -7 Teflon (PTFE/FEP) [C F ] n /[C F ] n × -6 Titanium Ti 1.17 × -8
9f polymer materials, a huge difference is observed between fluorinated ones, such as Teflon,and others. In the case of Kapton and acrylic, neutron yields are approximately 10 -7 neu-trons per decay, whereas this yield is almost 100 times higher for Teflon. Therefore, Pbconcentrations in materials like Kapton FN, consisting on Kapton polyimide coated withTeflon FEP, will have a much higher impact than equal
Pb concentrations in other kindof polyimide.
The first type of sample assayed was ultra-thin Kapton film. Kapton is an electric in-sulator typically used in circuits of low-background detectors employed to search for darkmatter and neutrinoless double-beta decay. This material is used due to its unique thermalresistance.A chemical procedure to completely digest Kapton, to extract
Pb, and to precipitatelead fluoride samples was developed. A flowchart of this procedure is shown in Figure 2.Most of this procedure can be applied to other polymers, especially those where only C, Hand O are present. A pressurized microwave digestion system (CEM MARS 230/60) wasused along with concentrated HNO , as suggested in reference [26].First, the samples were cleaned with 7 mol·L -1 HNO . After discarding the cleaningsolution, concentrated HNO (70% v/v) was added, together with the natural Pb carrier.The microwave digestion procedure for Kapton is the following:1. Initial ramp up to 150 ºC in 10 min.2. Hold at 150 ºC for 10 min.3. Ramp up up to 240 ºC in 10 min4. Hold at 240 ºC for 15 min.5. Cool down for 15 min.For each cycle of the microwave digestion system, up to 0.5 g of organic material canbe digested. Therefore, in order to be able to digest up to 2.0 g of polymer sample, 10 mgPb carrier were dissolved in 40 mL of concentrated HNO . This way, separating the totalsample of polymer in 4 aliquots with less than 0.5 g, processing these 4 replicates with 10 mLof the concentrated HNO + Pb carrier solution (one per aliquot), and finally mixing the4 replicates together, allowed us to digest such a relatively large amount of sample. The10igestion was tested by centrifuging the final solution, and checking that no precipitate wasformed. It is assumed that at least, a 99% of the sample was completely dissolved. Theresulting 1% of uncertainty is negligible in comparison with the high statistical uncertaintydue to the low number of detected counts. The Pb/
Pb ratio in this solution is the sameas in the final PbF precipitate, establishing a direct relationship between this ratio and the Pb concentration in the original Kapton sample.
Figure 2:
Flowchart of thebasic chemical steps for thepreparation of lead fluoride tar-gets for Kapton samples. Notethat, following microwave diges-tion, all processes preserve the
Pb/
Pb ratio. This solution was then fully evaporated, leaving thesolid lead nitrate, together with some residual organic ma-terial. 2% v/v HNO was added and the Pb(OH) wasprecipitated by adding NH . Samples were left to set-tle overnight, and then centrifuged, discarding the super-natant. Fluoride was formed by acid-base reaction of thePb(OH) precipitate with HF. In this last step, residualorganic material remained in the HF solution, so the finaltarget material was free of this residue. Targets producedby this method presented similar current outputs as thosefrom commercial PbF .In order to produce lead oxide, instead of fluoride sam-ples, the precipitation of lead hydroxide was replaced bylead carbonate precipitation. This was accomplished byadding a saturated sodium carbonate aqueous solution in-stead of ammonia. After centrifuging and discarding thesupernatant, the carbonate precipitate was heated in amuffle furnace for 6 h at a temperature of 650 ºC, produc-ing PbO by thermal decomposition.Initial measurements of the Kapton samples, using leadfluoride targets, are presented in Table 6. During thesemeasurements, all the samples gave Pb/
Pb ratioslower or equal to blank levels. Therefore, only 90% C.L.upper limits of the
Pb specific activity can be given.These upper limits were calculated using the Feldman andCousins method [27], using the lower limit for the back-ground to obtain the most conservative upper limit for the
Pb specific activity. The maximum
Pb specific activ-ity in the 100HN Kapton material was 0.85 Bq/kg. Takinginto account that this film has a thickness of 25.4 µm, andthat its density is 1.42 g·cm -3 , the upper limit for the Pbactivity per unit of surface is 31 mBq·m -2 . To provide an11dea of the performance of this measurement, the Pb plating out rate in high-densitypolyethylene at the underground lab of SNOLAB, due to the
Rn concentration there, is2.46 mBq·m -2 ·d -1 [11].In order to test the chemical methods for PbO, measurements of different Kapton samplesusing oxide as target material were also performed. Results, which are shown in Table 7,allow only results reported as 90% C.L. upper limits for the Pb in these samples.
Table 6:
Pb concentrations measured by AMS in different aliquots of a sample of100HN Kapton film, using lead fluoride targets.
Aliquot Samplemass (g)
Pb carrier(mg)
Pb/
Pb( × -13 ) [ Pb](Bq/kg)
Pb-Kapton-191025 1.376 ± ± < < ± ± < < ± ± < < Table 7:
Pb concentrations measured by AMS in different Kapton HN film samples,using lead oxide targets.
Sample Samplemass (g)
Pb carrier(mg)
Pb/
Pb( × -13 ) [ Pb](Bq/kg) ± ± < < ± ± < < ± ± < < ± ± < < Even though the
Pb half-life is much shorter than those of radionuclides commonlymeasured by AMS, this technique still provides a much better sensitivity than γ -counting.In the 3 MV AMS system at the University of Ottawa, Pb/
Pb background is equivalentto a
Pb activity of several hundreds of µBq. In comparison with Po α -counting, whichrequires 3-6 months after chemical treatment to achieve secular equilibrium, AMS providesmuch faster results and more comprehensive information about the entire sample, not onlyits surface activity.A study of the ion source current output from fluoride targets as a function of the tem-perature of the Cs reservoir will be performed to assess whether it is possible to improve the12on current yield without increasing its instability. Furthermore, the optical transmission formeasurements with fluoride targets could be improved using a less conservative widths forthe slits, while ensuring that the background is not severely increased.A better investigation of the background for measurements with lead oxide samples isrequired. Preliminary results suggest that this background may not be much higher than theone observed with fluoride targets. If this is confirmed, it shows the very good mass resolutionof the post-accelerator spectrometer of this AMS system for suppressing the scatter tails ofthe nearby Pb ions.In any case, with the current performance parameters, Pb measurements at the 3 MVAMS system at the University of Ottawa have already demonstrated the potential of per-forming groundbreaking assay of materials for Astroparticle Physics experiments.
Acknowledgments
The authors are deeply indebted to the André E. Lalonde Laboratory, at the Universityof Ottawa, for the access to the 3 MV AMS system to perform the measurements, to theirActinides and Fission Products Laboratory for the preparation of the samples; and to theGeochemistry Laboratory at the University of Ottawa, for the access to their microwavedigestion system and for the ICP-ES measurement of the Pb concentration of the blanksolution. We would like to thank Dr. Shawn Westerdale for his help and updates of theNeuCBOT tool, used to calculate the neutron production yield of the α particles from the Pb chain. This work has been supported by the Arthur B. McDonald Canadian Astropar-ticle Physics Research Institute, the National Research Council of Canada and the CanadaFoundation for Innovation.
References [1] NEWS-G collaboration, First results from the NEWS-G direct dark matter search ex-periment at the LSM, Astroparticle Physics 97 (2018) 54 – 62. arXiv:1706.04934 , doi:https://doi.org/10.1016/j.astropartphys.2017.10.009 .[2] PICO collaboration, Dark matter search results from the complete exposure of thePICO-60 C F bubble chamber, Phys. Rev. D 100 (2019) 022001. arXiv:1902.04031 , doi:10.1103/PhysRevD.100.022001 . 133] DEAP collaboration, Design and construction of the DEAP-3600 dark matter detec-tor, Astroparticle Physics 108 (2019) 1 – 23. arXiv:1712.01982 , doi:10.1016/j.astropartphys.2018.09.006 .[4] B. Lehnert, Backgrounds in the DEAP-3600 Dark Matter Experiment (2018). arXiv:1805.06073 .[5] EXO-200 collaboration, Search for Neutrinoless Double- β Decay with the CompleteEXO-200 Dataset, Phys. Rev. Lett. 123 (2019) 161802. arXiv:1906.02723 , doi:10.1103/PhysRevLett.123.161802 .[6] D. Leonard, D. Auty, T. Didberidze, R. Gornea, P. Grinberg, R. MacLellan, B. Methven,A. Piepke, J.-L. Vuilleumier, EXO-200 collaboration, Trace radioactive impurities infinal construction materials for EXO-200, Nuclear Instruments and Methods in PhysicsResearch Section A: Accelerators, Spectrometers, Detectors and Associated Equipment871 (2017) 169 – 179. arXiv:1703.10799 , doi:10.1016/j.nima.2017.04.049 .[7] Majorana Collaboration, Search for neutrinoless double- β decay in Ge with 26 kgyr of exposure from the Majorana Demonstrator, Phys. Rev. C 100 (2019) 025501. arXiv:1902.02299 , doi:10.1103/PhysRevC.100.025501 .[8] Majorana collaboration, The majorana demonstrator radioassay program, NuclearInstruments and Methods in Physics Research Section A: Accelerators, Spectrome-ters, Detectors and Associated Equipment 828 (2016) 22 – 36. arXiv:1601.03779 , doi:https://doi.org/10.1016/j.nima.2016.04.070 .[9] nEXO collaboration, nEXO Pre-Conceptual Design Report (2018). arXiv:1805.11142 .[10] nEXO collaboration, Sensitivity and discovery potential of the proposed nEXO exper-iment to neutrinoless double- β decay, Physical Review C 97 (2018) 065503. arXiv:1710.05075 , doi:10.1103/PhysRevC.97.065503 .[11] M. Stein, D. Bauer, R. Bunker, R. Calkins, J. Cooley, B. Loer, S. Scorza, Radondaughter plate-out measurements at SNOLAB for polyethylene and copper, NuclearInstruments and Methods in Physics Research Section A: Accelerators, Spectrome-ters, Detectors and Associated Equipment 880 (2018) 92 – 97. arXiv:1708.09476 , doi:10.1016/j.nima.2017.10.054 .[12] X. Hou, P. Roos, Critical comparison of radiometric and mass spectrometric methodsfor the determination of radionuclides in environmental, biological and nuclear wastesamples, Analytica Chimica Acta 608 (2008) 105 – 139. doi:10.1016/j.aca.2007.12.012 . 1413] Y. Ebaid, A. Khater, Determination of Pb in environmental samples, Journalof Radioanalytical and Nuclear Chemistry 270 (2006) 609 – 619. doi:10.1007/s10967-006-0470-5 .[14] D. Larivière, K. Reiber, R. Evans, R. Cornett, Determination of 210Pb at ultra-tracelevels in water by ICP-MS, Analytica Chimica Acta 549 (2005) 188 – 196. doi:10.1016/j.aca.2005.06.020 .[15] P. Steier, R. Golser, W. Kutschera, V. Liechtenstein, A. Priller, A. Valenta, C. Vocken-huber, Heavy ion AMS with a “small” accelerator", Nuclear Instruments and Methodsin Physics Research Section B: Beam Interactions with Materials and Atoms 188 (2002)283 – 287. doi:10.1016/S0168-583X(01)01114-4 .[16] C. Vockenhuber, I. Ahmad, R. Golser, W. Kutschera, V. Liechtenstein, A. Priller,P. Steier, S. Winkler, Accelerator mass spectrometry of heavy long-lived radionu-clides, International Journal of Mass Spectrometry 223-224 (2003) 713 – 732. doi:10.1016/S1387-3806(02)00944-2 .[17] A. Sookdeo, J. Cornett, W. E. Kieser, Optimizing production of Pb beams for
Pbanalysis by Accelerator Mass Spectrometry, Nuclear Instruments and Methods inPhysics Research Section B: Beam Interactions with Materials and Atoms 361 (2015)450 – 453. doi:10.1016/j.nimb.2015.02.063 .[18] A. Sookdeo, R. J. Cornett, X.-L. Zhao, C. R. J. Charles, W. E. Kieser, Measuring
Pb by accelerator mass spectrometry: a study of isobaric interferences of
Pband
Pb, Rapid Communications in Mass Spectrometry 30 (2016) 867–872. doi:10.1002/rcm.7501 .[19] G. Korschinek, J. Sellmair, A. Urban, M. Müller, A study of different ion sources for usein the
Pb experiment, Nuclear Instruments and Methods in Physics Research SectionA: Accelerators, Spectrometers, Detectors and Associated Equipment 271 (1988) 328 –331. doi:10.1016/0168-9002(88)90180-5 .[20] X.-L. Zhao, A. Litherland, J. Eliades, W. Kieser, Q. Liu, Studies of anions from sput-tering I: Survey of MF n- , Nuclear Instruments and Methods in Physics Research Sec-tion B: Beam Interactions with Materials and Atoms 268 (2010) 807 – 811. doi:10.1016/j.nimb.2009.10.036 .[21] R. Middleton, A Negative-Ion Cookbook, Department Of Physics, University of Penn-sylvania, 1990.URL doi:10.1016/j.nimb.2015.03.014 .[23] National Institute of Standards and Technology, SRM 4337 - Lead-210 RadioactivityStandard, .[24] J. L. Orrell, C. E. Aalseth, I. J. Arnquist, T. A. Eggemeyer, B. D. Glasgow, E. W. Hoppe,M. E. Keillor, S. M. Morley, A. W. Myers, C. T. Overman, S. M. Shaff, K. S. Thom-masson, Assay methods for 238U, 232Th, and 210Pb in lead and calibration of 210BiBremsstrahlung emission from lead, Journal of Radioanalytical and Nuclear Chemistry309 (2016) 1271–1281. doi:10.1007/s10967-016-4732-6 .[25] S. Westerdale, P. Meyers, Radiogenic neutron yield calculations for low-backgroundexperiments, Nuclear Instruments and Methods in Physics Research Section A: Ac-celerators, Spectrometers, Detectors and Associated Equipment 875 (2017) 57 – 64. arXiv:1702.02465 , doi:10.1016/j.nima.2017.09.007 .[26] I. J. Arnquist, C. Beck, M. L. di Vacri, K. Harouaka, R. Saldanha, Ultra-low radioactiv-ity Kapton and copper-Kapton laminates, Nuclear Instruments and Methods in PhysicsResearch Section A: Accelerators, Spectrometers, Detectors and Associated Equipment959 (2020) 163573. doi:10.1016/j.nima.2020.163573 .[27] G. J. Feldman, R. D. Cousins, Unified approach to the classical statistical analysisof small signals, Phys. Rev. D 57 (1998) 3873–3889. arXiv:physics/9711021 , doi:10.1103/PhysRevD.57.3873doi:10.1103/PhysRevD.57.3873