Rb83/Kr83m production and cross-section measurement with 3.4 MeV and 20 MeV proton beams
Dan Zhang, Jingkai Xia, Yifan Li, Jingtao You, Yao Li, Changbo Fu, Jianglai Liu, Ning Zhou, Jie Bao, Huan Jia, Chenzhang Yuan, Yuan He, Weixing Xiong, Mengyun Guan
883
Rb/
Kr production and cross-section measurement with 3.4 MeV and 20 MeVproton beams
Dan Zhang, Jingkai Xia, ∗ Yifan Li, Jingtao You, Yao Li, Changbo Fu, † Jianglai Liu,
1, 4, 5
NingZhou, ‡ Jie Bao, Huan Jia, Chenzhang Yuan, Yuan He, Weixing Xiong, and Mengyun Guan School of Physics and Astronomy, Shanghai Jiao Tong University, MOE Key Laboratory for Particle Astrophysicsand Cosmology, Shanghai Key Laboratory for Particle Physics and Cosmology, Shanghai 200240, China Department of Physics, University of Maryland, College Park, Maryland 20742, USA Zhiyuan College, Shanghai Jiao Tong University, Shanghai 200240, China Tsung-Dao Lee Institute, Shanghai Jiao Tong University, Shanghai 200240, China Shanghai Jiao Tong University Sichuan Research Institute, Chengdu 610213, China China Institute of Atomic Energy, Beijing 102413, China Institute of Modern Physics, the Chinese Academy of Sciences, Lanzhou 730000, China Institute of High Energy Physics, the Chinese Academy of Sciences, Beijing 100049, China
Kr with a short lifetime is an ideal calibration source for liquid xenon or liquid argon detector.The
Kr isomer can be generated through the decay of Rb isotope, and Rb is usually pro-duced by proton beams bombarding natural krypton atoms. In this paper, we report a successfulproduction of Rb / Kr with 3.4 MeV proton beam energy and measure the production rate withsuch low proton energy for the first time. Another production attempt was performed with newlyavailable 20 MeV proton beam in China, the production rate is consistent with our expectation.The produced
Kr source has been successfully injected into PandaX-II liquid xenon detector andyielded enough statistics for detector calibration.
I. INTRODUCTION
Dark matter searches in xenon detectors like PandaX-II [1, 2] and its successors [3, 4] rely on accurate eventreconstruction from scintillation (S1) and ionization (S2) signals generated by energy deposit in liquid xenon. Due tosome detector geometric deformation and electric field non-uniformity, the magnitudes of S1 and S2 signals have astrong position dependence, which degrades the event reconstruction resolution and the discrimination between nuclearrecoil and electron recoil events [5]. Therefore it is necessary to have mono-energetic signals evenly distributed in thedetector to calibrate the detector response.One of such calibration sources for this purpose is m Kr isomer, which is gaseous and can be mixed with xenonuniformly. Its half-life is only 1.83 h, so no specific removal procedure is required after a calibration campaign.In addition, the energy of the m Kr decays is small enough to calibrate noble-liquid detectors for dark mattersearch where region of interest is usually less than 100 keV [6–8]. m Kr has been used to calibrate tritium β decay experiments [9], calorimeters in the large electron-positron colliders [10] and the heavy ion detector of ALICEexperiment [11]. An experiment at Yale University reported the first use of m Kr in a liquid xenon detector forspatial and energy calibration [12].Because m Kr is a short-living source, we resort to its mother isotope Rb with a relatively long life-time ( T / =86 . Rb decays into the excited energy states of Kr through electron capture purely.The full decay scheme of Rb according to NNDC [13] shows that 74.385% of the Rb atoms decay into the m Krisomer. The simplicity of the decay mode mitigates potential side effects for detector calibration. Rb is a synthetic radioisotope which can be produced by proton beams bombarding natural krypton with peakproduction around 20 MeV proton energy. Due to limited access to such a high energy proton facility, we testedthe production with a lower energy proton beam. In this paper, we report a successful production of Rb/ m Krsource with 3.4 MeV proton beam at the China Institute of Atomic Energy, and derive an experimental yield of nat
Kr(p , xn) Rb process for proton energy below 5 MeV for the first time. The bombarding chamber design andexperiment setup are demonstrated in section II. The collection of Rb/ m Kr source and production rate measure-ment are presented in section III. Another production test performed with a recently available 20 MeV proton beam ∗ Corresponding author: [email protected]; Now at ShanghaiTech University, Shanghai, China † Now at Institute of Modern Physics, Fudan University, Shanghai 200433, China ‡ Corresponding author: [email protected] a r X i v : . [ nu c l - e x ] F e b at the Institute of Modern Physics, Chinese Academy of Sciences is shown in section IV. Finally, the injection test of Rb/ m Kr in the PandaX-II detector is shown in section V.
II. PRODUCTION OF RB/ m KR SOURCE WITH 3.4 MEV PROTONS Rb can be synthesized by the bombardment of krypton with protons or bromine with alphas. Since alphabombarding bromine yields more unexpected isotopes which might contaminate our detector, the nat
Kr(p, xn) Rbprocess is favored. The production rate depends strongly on the energy of proton beam. The optimal energy isapproximately 20 MeV, which can maximize the Rb production rate and minimize the unwanted Rb and Rbthat bring extra risk of increasing backgrounds in the gamma spectra [14].Before we had access to the 20 MeV proton beams, the China Institute of Atomic Energy provided proton beams withenergy approximately 3.4 MeV, which is slightly above the theoretical energy threshold 1.7 MeV for the Kr(p,n) Rbreaction. However, there is no precedent experimental data of this energy range ever reported before. To have a roughestimation of the production rate, we fit the experimental data on the thin target yields above 5 MeV [15–17] andmade an extrapolation to lower energy region, as shown in Fig. 1(a). Besides Rb, Rb ( T / = 18 . Kr(p,n) Rb has a threshold of 1.3 MeV (Fig. 1(b)).
E (MeV) -
10 110 M e V ) (cid:215) T h i n t a r g e t y i e l d ( M B q / C ) th Kovacs et.al.: 0.1283 (E-E ) th Mulders: 0.0092 (E-E ) th Steyn et.al.: 0.1296 (E-E ) th total: 0.0898 (E-E (a) E (MeV) M e V ) (cid:215) T h i n t a r g e t y i e l d ( M B q / C ) th Mulders et.al.: 0.0771 (E-E ) th Steyn et.al.: 1.0322 (E-E ) th total: 0.5908 (E-E (b) FIG. 1: Theoretical extension of the thin target yield of proton beam bombarding on 1 bar nat
Kr according toformer works (‘total’ means take all the previous data into account) for Rb (left) and Rb (right) [15–17].In this work, two nat
Kr target cells are prepared for bombardment so that one can be used to study the outcome indetails and the other can be supplied as calibration source. The cell design is shown in Fig. 2, details are describedas following:(I) We use 20 µ m aluminum (Al) foil as the window to separate the gas and vacuum based on the work in [18].The diameter of the window is designed as 10 mm to prevent the foil from breaking due to too large a force on theedge. This design is tested with 1.5 bar pressure dropping over the two sides.(II) In order to measure the proton beam currents, the nat Kr chamber must be insulated from the upstream part.Therefore, we use polytetrafluoroethylene (PTFE) gaskets instead of copper gasket to seal the CF35 flanges. Moreover,the bolts and the flange are wrapped by kapton tape, and paper gaskets are put between the stainless steel bolts andthe flange.(III) The energy deposited by protons on the foil may potentially result in a high temperature. According to thestopping power of proton in Al provided by PSTAR [19], 3.4 MeV proton beam deposits 0.4 MeV in 20 µ m Al.Assuming the proton beam radius of 2 mm and current of 10 µ A, the power deposit in the Al foil reaches 4 W. Thetemperature at the center of the Al foil would reach approximately 360 K if heat dissipates only through conduction,which is still far below the melting temperature of both the PTFE and Al.(IV) To determine the length of the target chamber, we calculate the effective reaction length of the proton beamin nat
Kr considering the stopping power of proton in krypton [19]. The estimated reaction length is approximately5.2 cm. Hence the length of the target chamber is chosen to be 10 ∼
15 cm.(V) The beam may hit the iron wall of the nat
Kr chamber, producing unexpected radioactive isotopes. Even
Target cellAl windowProtons
FIG. 2: Design of nat
Kr target cell.though the coulomb barrier of iron is larger than the proton energy, we put an Al dump in the nat
Kr chamber to beconservative. The Al dump is also wrapped by kapton tape to be electrically insulated.The bombardment tests were performed for 39 minutes on the first chamber and 175 minutes on the second one,with stable operation. Both chambers had 1 bar natural krypton inside. The average currents of proton beam were1.5 µ A and 1.6 µ A for the the first and second tests respectively.
III. PRODUCTION RATE MEASUREMENT WITH 3.4 MEV PROTONS
After bombardment, the target cell got contact with air before further processing to ensure that the rubidium wasin a form of chemical compound. To study the distribution of Rb production in bombardment, the first target cellwas divided into three parts: the Al window (including the foil and the flange), the Al dump and the CF35-straighttube. Each part were washed with 60-150 ml deionized water separately and 2 g zeolite beads (Merck 2 mm diameter,0.5 nm molecular sieve) were used to absorb rubidium in the solution. The solution was heated in water bath gentlyat 70 ∼ ◦ C until being dried. Then the zeolite was baked at 300 ◦ C under pumping for further degassing, and wasfinally stored in a sealed plastic bag (Fig. 3).After being degassed, the zeolite samples was measured by a germanium detector at Shanghai Jiao Tong University.The spectra of the samples are shown in Fig. 4, and compared with the MC simulation done by GEANT4 package.The radioactivity of each sample is summarized in Tab. I. The mismatch of the Compton continuum in Fig. 4(a) islikely due to the inaccurate geometry in the simulation (see Sec. IV for details). By measuring each part before andafter washing, we obtained the efficiency of the rubidium transfer as 66 ±
2% for the Al window, 83 ±
6% for the Aldump, 92 ±
1% for the CF35-straight tube (Tab. II).The relative rubidium distribution in the cell (Window : Al dump : CF35-straight tube ) is 100 : (13 ±
11) : (20 ± Rb, 100 : (23 ±
20) : (19 ±
16) for Rb and 100 : (12 ±
10) : (18 ±
15) for Rb (the decay of the isotopes hasbeen considered), respectively. It indicates that the rubidium was mainly produced near the Al window, which is asexpected because the proton energy is slightly above the threshold of the nuclear reactions and should lose all theenergy in approximately 5 cm. In addition, freshly produced rubidium isotopes are likely spreading in the chamber.As presented in Sec. IV, systematic uncertainties on geometric detection efficiencies of our germanium detector areFIG. 3: The storage of the baked zeolite after absorbing rubidium in the de-ionized water solution.
200 400 600 800 1000 1200
E (keV) - - - - - R a t e ( H z ) Rb Rb Rb Rb k e V } SimulationMeasurement (a)Sample 1 spectrum
200 400 600 800 1000 1200
E (keV) - - - - - R a t e ( H z ) Rb Rb Rb Rb k e V } SimulationMeasurement (b)Sample 2 spectrum
200 400 600 800 1000 1200
E (keV) - - - - - R a t e ( H z ) Rb Rb Rb Rb k e V } SimulationMeasurement (c)Sample 3 spectrum
FIG. 4: The comparison between the simulation and the measurement for different zeolite beads samples. Sample 1,sample 2, sample 3 absorbs rubidium from the Al window, the Al dump and the CF35-straight tube respectively.60%. Compared to the geometric effect, the possible loss due to rubidium not plating on the surface of the cell isnegligible as well as the statistical uncertainties (generally 5%).To calculate the thick target yield, the radioactivity of each zeolite sample in Tab I is summed and corrected for thehalf-lives of the decays of the transferring efficiencies. The total radioactivity generated in one target cell is 149 Bq for Rb, 18 Bq for Rb and 546 Bq for Rb, respectively. The charge of the protons used is 3 . × − C. Eventually weobtain the thick target yield via 3 MeV (effective energy after losing energy in the Al foil) proton bombardment fullystopped by nat
Kr: 0.043 ± Rb, 0.157 ± Rb. The comparison of this measurementfor thick target yield of Rb and Rb and the extrapolation from above 5 MeV experimental data can be seen inFig. 5. We also observed Rb (0.0038 ± Kr(p,n) Rb is 3 MeV [15], which is reasonable as the energy of the proton beam has some spreading.TABLE I: The radioactivity of the three zeolite samples measured from germanium detector (only statisticaluncertainties applied in this table).
Isotope Al Window (Bq) Al dump (Bq) CF35-straight tube (Bq)Time after bombarding 40 days 28 days 41 days Rb 53.4 ± ± ± Rb 3.68 ± ± ± Rb 62.6 ± ± ± TABLE II: The transfer efficiency for different parts and peaks.
Item Al Window Al dump CF35-straight tube Rb (520) 65.7 ± ± ± Rb (530) 65.7 ± ± ± Rb (553) 64.7 ± ± ± Rb (882) 66.9 ± ± ± Rb (1077) 67.8 ± ± ± ±
2% 83 ±
6% 92 ± E (MeV) - -
10 110 T h i c k t a r g e t y i e l d ( M B q / C ) Kovacs et.al.MuldersSteyn et.al.totalRb in this work (a) E (MeV) - -
10 110 T h i c k t a r g e t y i e l d ( M B q / C ) MuldersSteyn et.al.totalRb in this work (b) FIG. 5: The comparison between this work and extensions from former works (“total” means take all the previousdata into account) [15–17] for Rb (left) and Rb (right) . IV. PRODUCTION WITH 20 MEV PROTONS
Recently, a new proton facility- Chinese ADS Front End demo linac (CAFE) has been built at the Instituteof Modern Physics, Chinese Academy of Sciences, with energy up to 25 MeV [20, 21]. As one of the first users, weconducted another test with the 20 MeV proton beam bombarding on 1.1 bar natural krypton gas. The total exposureto the protons was 9.7 µ Ah (0.035 C).We preserved the previous design of the target cell and added extra water cooling for the 20 MeV proton bom-bardment. In this test, the main concern of heat loads is on the target cell instead because each proton deposits only0.11 MeV in the aluminum foil according the stopping power on PSTAR [19]. To stop the 20 MeV protons with 1 µ Aaverage currents, the heat gain on the back of target cell is 20 W. The Al beam dump is cooled by room temperaturewater with flux up to 400 cm /s, which has enough capacity for the heat removal.Multiple Rb/ m Kr sources with several kilo up to mega Becquerel were produced in the processing procedure.The window and the target chamber were washed with deionized water separately, and the chamber were washedthree times. The radioactivity ratio of the final zeolite samples is 1 st : 2 nd : 3 rd : window = 1 : 0 .
050 : 0 . . σ ) of nat Kr(p, xn) Rb, nat
Kr(p, xn) Rb and nat
Kr(p, xn) Rb at 20 MeV measured by thegermanium detector is consistent with the previous measurements as shown in Fig. 6. The initial proton energy iscalculated as 20 . ± .
03 MeV by the time of incident protons flying through a 2.47 m vacuum chamber. The protonsloss 1.4 MeV in the 25 cm target cell filled with 1.1 bar krypton gas according to the stopping power on PSTAR [19],which dominates the energy spread in Fig. 6. We validate the systematic uncertainties by measuring the detectingefficiency of a millimeter-scale source. Compared to the rubidium sources, the cylindrical shape source with a 3 mmradius and 6 mm height is small enough to be regarded as a point source. The typical size difference among the Rb/ m Kr sources is 3 cm. According to the measurements with the calibration source, a 3 cm horizontal deviationto the surface center reduces the detecting efficiency to 60% and a 3 cm vertical deviation to 40%. Therefore, thesystematic uncertainties are set to 60% in Fig. 6.
16 18 20 22 24
E (MeV) c r o ss s ec ti on ( m b ) Kovacs et.al.MuldersSteyn et.al.Rb in this work (a)
16 18 20 22 24
E (MeV) c r o ss s ec ti on ( m b ) Kovacs et.al.MuldersSteyn et.al.Rb in this work (b)
16 18 20 22 24
E (MeV) c r o ss s ec ti on ( m b ) MuldersSteyn et.al.Rb in this work (c) FIG. 6: Comparison of the nat
Kr(p, xn) Rb (a), nat
Kr(p, xn) Rb (b) and nat
Kr(p, xn) Rb (c) cross sections at20 MeV between this work and previous measurements [15–17].
V. INJECTION OF m KR IN PANDAX-II DETECTOR
We used 1 g zeolite carrying 30 Bq (60% uncertainty) Rb for detector injection test. In order to stop zeolite fromescaping the source chamber, we chose to use 0.2 µ m membrane filter (Merck, FGLP01300), which was tested to havean upper limit of the leakage of Rb isotopes 1 . × − % per hour (2.4 µ Bq/h for a 1.8 MBq Rb/ m Kr source) [7].The chamber with zeolite filled was firstly pumped individually for 60 h at 80 ◦ C and the vacuum reached 5 . × − Pa.20 L gas xenon was injected in the zeolite container to mix with m Kr. Then the mixed gas was circulated and purifiedby a getter (PS4-MT50-R-2) for 24 h. The source chamber was then connected to the PandaX-II detector throughcirculation system for 12 h.The decay from m Kr to the ground state is mainly through internal conversion electrons ( τ = 1 .
83 h). The directdecay mode from 41 . . . . . ∗ has a lifetime of 35 ns at 40 keV [22], and the electronics limits the width of thepulse on the level of 100 ns. Hence only part of the scintillation signals (S1s) of the two peaks can be separated. Forthe ionization signals (S2s), the time resolution of our detector limits to several microseconds, which makes it almostimpossible to separate the two transitions. Actually, we did not see any event with two separate S2s in our 5 × m Kr events collected. TABLE III: Decay channels of the m Kr isotope [10].
Transition energy Decay mode Branching ratio32.1 keV e(30 keV)+e(2 keV) 76%e(18 keV)+e(10 keV)+2e(2 keV) 9%e(18 keV)+X(12 keV)+2e(2 keV) 15%9.4 keV e(7.6 keV)+e(1.8 keV) 95% γ Figure 7 shows the response of m Kr in PandaX-II detector after the data process chain used in [23, 24], wherewe can see two S1 peaks from m Kr with one S2 peak. The waveforms for S1 with the two transitions separated ormixed are compared in Fig. 8.The time interval of the two S1s could be used to fit the half-life of the first excited state of m Kr as shown inFig. 9(a). From the fitting the tail with ∆ t larger than 120 ns, we obtained a half-life of 154.5 ± m Kr characteristic energy peak. The energy resolution for m Kr is 8.0%, the mean of the peak E is 40.8 keV.The fitted mean value is smaller than that provided by NNDC (41.5 keV) partly due to the baseline suppressionthreshold of the PandaX-II data-acquiring system. Meanwhile, detector response model could be further tunedaccordingly using this calibration data [1].
100 200 300 400 500 (PE) max S1 S ( P E ) FIG. 7: S2 vs S1 max by applying the data chain for the WIMP search runs and the color bar represents the counting. (a) (b)
FIG. 8: (a) The waveform with the two S1s separated. (b) The waveform with the two S1s mixed.
VI. CONCLUSIONS
We report a successful production of Rb / Kr with 3.4 MeV proton beam provided by the China Instituteof Atomic Energy. The production rate is measured as 0.0043 ± Kr(p,n) Rb process, whichis the first experimental data ever reported for such low proton energy in the world. Another production attemptwas performed with recently available 20 MeV proton beam at the Institute of Modern Physics, Chinese Academyof Sciences, which was one of the first applications on this proton facility. The produced
Kr source has beensuccessfully injected into PandaX-II liquid xenon detector and yielded enough statistics for detector calibration.
Acknowledgments
This project is supported in part by the Double First Class Plan of the Shanghai Jiao Tong University, grants fromNational Science Foundation of China (Nos. 11525522, 11775141 and 11755001), a grant from the Ministry of Scienceand Technology of China (No. 2016YFA0400301). We thank the Office of Science and Technology, Shanghai Municipal / ndf c – – D C oun t / ndf c – – S1 mixed eventsS1 seperated eventsAll events (a)
E (keV) C oun t DataFitting (b)
FIG. 9: (a) The fit of the half-life for the first excited state of m Kr. (b) The energy spectrum of m Kr withbackground subtracted.Government (No. 11DZ2260700, No. 16DZ2260200, No. 18JC1410200) and the Key Laboratory for Particle Physics,Astrophysics and Cosmology, Ministry of Education, for important support. We also thank the sponsorship from theChinese Academy of Sciences Center for Excellence in Particle Physics (CCEPP), Hongwen Foundation in Hong Kong,and Tencent Foundation in China. Finally, we thank the CJPL administration and the Yalong River HydropowerDevelopment Company Ltd. for indispensable logistical support and other help. [1] Q. Wang et al. (PandaX Collaboration), Chin. Phys. C , 125001 (2020).[2] X. Zhou et al. (PandaX Collaboration), Chin. Phys. Lett. , 011301 (2021).[3] H. Zhang et al. (PandaX Collaboration), Sci. China Phys. Mech. Astron. , 31011 (2019).[4] P. Juyal, K.-L. Giboni, X.-D. Ji, and J.-L. Liu, Nucl. Sci. Tech. , 93 (2020).[5] D. Akerib et al. (LUX Collaboration), JINST , P11022 (2017).[6] D. Akerib et al. (LUX Collaboration), Phys. Rev. D , 112009 (2017).[7] V. Hannen et al. , JINST , P10013 (2011).[8] P. Agnes et al. (DarkSide Collaboration), JINST , T12004 (2017).[9] M. Zboˇril et al. , JINST , P03009 (2006).[10] A. Chan et al. , IEEE Trans. Nucl. Sci. , 491 (1995).[11] J. Stiller (ALICE Collaboration), Nucl. Instrum. Meth. A , 20 (2013).[12] L. Kastensand, S. Cahn, A. Manzur, and D. McKinsey, Phys. Rev. C , 045809 (2009).[13] C. Dunford and T. Burrows, Report IAEA-NDS-150 (NNDC Informal Report NNDC/ONL-95/10), Rev. 95/10 Interna-tional Atomic Energy Agency, Vienna, Austria (1995).[14] D. V´enos, A. ˘Spalek, O. Lebeda, and M. Fi˘ser, Appl. Radiat. Isot. , 323 (2005).[15] J. Mulders, Int. J. Appl. Radiar. Isot. , 475 (1984).[16] G. Steyn, S. Mills, F. Nortier, and F. Haasbroek, Int. J. Appl. Radiar. Isot. , 361 (1991).[17] Z. Kov´acs, F. T´ark´anyi, S. Qaim, and G. Stocklin, Int. J. Appl. Radiar. Isot. , 329 (1991).[18] L. I. Bodine, Doctoral dissertation, University of Washington (2015).[19] M. Berger, J. Coursey, M. Zucker, and J. Chang, National Institute of Standards and Technology, Gaithersburg, MD/10.18434/T4NC7P.[20] Z. Wang et al. , Phys. Rev. Accel. Beams , 120101 (2016).[21] S. Liu, Z. Wang, H. Jia, Y. He, W. Dou, Y. Qin, W. Chen, and F. Yan, Nucl. Instrum. Meth. A , 11 (2017).[22] D. Akimov et al. , Phys. Lett. B , 245 (2002).[23] A. Tan et al. (PandaX Collaboration), Phys. Rev. Lett. , 121303 (2016).[24] X. Cui et al. (PandaX Collaboration), Phys. Rev. Lett.119