Non-analog decay of 74Rb

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Non-analog decay of 74Rb © 2001 Published by Elsevier B.V. Published version Collaboration, ISOLDE Collaboration, ISOLDE. (2001). Non-analog decay of 74Rb. Physics Letters B, 511(2-4), 145-150. https://doi.org/10.1016/S0370-2693(01)00646-3 2001 5 July 2001 Physics Letters B 511 (2001) 145–150 www.elsevier.nl/locate/npe Non-analog βdecay of 74Rb ISOLDE Collaboration M. Oinonena,b,J.Äystöa, P. Baumanni,J.Cederkälla,S.Courtini,P.Dessagnei, S. Franchooa,d, H. Fynboa,M.Górskad,J.Huikaric, A. Jokinenb,c, A. Knipperi, U. Köstera,G.LeScorneta,f,C.Miehéi,A.Nieminenc,T.Nilssona, Yu. Novikovh, K. Peräjärvic,E.Poirieri,A.Popovh,D.M.Seliverstovh,T.Siiskonena,b,H.Simona, O. Tengblade, P. Van Duppend, G. Walteri, L. Weissmana, K. Wilhelmsen-Rolandera,g aEP Division, CERN, CH-1211 Geneva 23, Switzerland bHelsinki Institute of Physics, P.O. Box 9, University of Helsinki, FIN-00014 Helsinki, Finland cDepartment of Physics, University of Jyväskylä, FIN-40351 Jyväskylä, Finland dInstituut voor Kernen Stralingsfysica, University of Leuven, Celestijnenlaan 200 D, B-3001 Leuven, Belgium eInstituto de Estructura de la Materia, CSIC, Serrano 113 bis, E-28006 Madrid, Spain fCSNSM, F-91405 Campus Orsay, France gDepartment of Physics, Stockholm University, Box 6730, S-113 85 Stockholm, Sweden hSt. Petersburg Nuclear Physics Institute, Gatchina, RU-188350 St. Petersburg, Russia iInstitut de Recherches Subatomiques, F-67037 Strasbourg cedex 2, France Received 15 February 2001; received in revised form 19 April 2001; accepted 16 May 2001 Editor: V. Metag Abstract The magnitude of the Coulomb mixing parameter δ1 IM has been experimentally deduced, for the first time, for the βdecay of 74Rb. The estimated magnitude is derived from the feeding of the non-analog first excited 0+state in 74Kr. The inferred upper limit of 0.07% is small compared to theoretical predictions. The half-life was measured to be 64.90(9) ms. 2001 Published by Elsevier Science B.V. PACS: 21.10.-k; 23.40.-s; 27.50.+e Keywords: Superallowed βdecay; Coulomb mixing; On-line mass separation 1. Introduction Accurate measurements of β-decay energies, halflives and branching ratios are necessary for studies of superallowed βdecay strength. This low-energy nuclear physics input, together with muon-decay data, E-mail address: markku.[email protected] (M. Oinonen). givespresently the most precisevalue for the up–down quark mixing matrix element Vud in the Cabibbo– Kobayashi–Maskawa (CKM) matrix [1]. Presently, the systematics of superallowed decays reach 54Co [2–4]. The largest uncertainty in Vud, as determined from β-decay results, is due to charge-dependent and other theoretical corrections [3,4]. Extending the systematics towards higher Zbrings data from a new 0370-2693/01/$ – see front matter 2001 Published by Elsevier Science B.V. PII:S0370-2693(01)00646-3 146 ISOLDE Collaboration / Physics Letters B 511 (2001) 145–150 realm into the data set and will thus eventually improve on the accepted value for Vud. In addition, it allows for further studies of the charge-dependent effects such as Coulomb mixing. The Coulomb correction δCincreases with increasing Zand may reach even a level of 1–2% for isotopes beyond 54Co [5,6]. An important contribution to the Coulomb correction is due to mixing between the involved 0+wave functions. The consequences of such an effect are observable in the βdecays as transitions feeding the non-analog 0+states. The observation of such transitions provides a test for the total Coulomb correction [7] and helps to distinguish between different theoretical values. In order to improve our knowledge of the superallowed Fermi decays, a program for performing a complete study on the β-decay of 74Rb has been launched at the on-line mass separator facility ISOLDE at CERN [8]. This program aims at highprecision measurementsof a half-life, branching ratios and a QEC value of 74Rb by using decay spectroscopy and atomic mass measurements[9,10].In addition, extensive shell-model calculations, with realistic effective interactions, are in progress [11] aiming to study the Coulomb effects involved in the decay. The eventual goal of these studies is to extract the strength of the analog transition and to significantly contribute to the superallowed decay systematics. As a first milestone, the measurement of the β-decay half-life and the first observation of non-analog βdecay of 74Rb will be reported in this Letter. Based on this observation, the Coulomb mixing between the lowest 0+ states will be discussed. 2. Experimental methods 74Rb ions were produced in spallation reactions induced by a pulsed 1-GeV proton beam incident on a Nb-foil or a ZrO2-felt target at the ISOLDE online mass separator at CERN [12]. The 60 keV massseparated ion beam was implanted in an aluminized Mylar tape with a typical collection time of 150 ms. Any residual long-lived activity was removed by transporting the implantation tape typically after every fifth proton pulse. The only isobaric contaminant was Ga. Four different types of measurements were performed for 74Rb: the determination of its half-life, a search for a high-spin isomer, a search for β-delayed proton and γ(β–γ) transitions and a search for β-delayed converted transitions (β–e−). The details of the first two experiments have been described elsewhere [13]. Essentially, the setup for these measurements consisted of a βtelescope made of a 2-mmthick plastic scintillator and a 20-mm-thick planar Ge detector, a 70% HPGe detector for γ-ray detection and a charged-particle detector telescope [14]. In the β–γexperiment a cylindrical scintillator with 70% efficiency was used as a trigger detector for βparticles. The γ-rays were detected with three 70% Ge detectors. Each Ge detector was equipped with a veto detector in front of the Ge crystal to reduce β–γsumming. Inthe β–e−experimenttransitions to excited0+states were searched for with the magnetic conversion electron spectrometer ELLI [15]. In each experiment the data were stored in list-mode and, in addition, data for half-life analysis were collected with a fast multiscaling acquisition system to minimize counting-rate induced effects. 3. Results 3.1. Half-life determination The existing experimental information on the βdecay of 74Rb prior to this work was limited to two halflife measurements [16,17]. Those measurements resulted in a half-life of 64.9(5) and 64(10) ms, respectively. In this Letter, the half-life was determined from the time dependenceof the observed events in the thin scintillator detector. Two sets of data were collected: one withoutconditions andanother one with an energy condition of Eβ>5.2 MeV. After a dead time correction, determined with an intense 26Na sample [13], the time spectra were fitted with a single exponential plus a constant background using a χ2method. Weighting of the individual data points was performed using the procedure described in Ref. [18]. The influence of pulse pile-up was negligible since the highest instantaneous count rate was below 2500ctss−1. The effect of changesin the counting rate during the measurements was estimated to be small and was included into the uncertainty of the deadtime ISOLDE Collaboration / Physics Letters B 511 (2001) 145–150 147 Fig. 1. The decay time spectrum collected at A=74 for the half-life analysis. The open squares show the experimental data and the full line is the fit. The error bars in the data are smaller than the symbols. The residual plot, discussed in the text, is shown in the lower part of the figure. correction as describedin [13]. The systematic error as contributions from contaminant activities was studied using the following procedure.First, a decay curve for 74Rb was generated assuming T1/2=64.9 ms [16]. Then, decay components taking into account the observed contaminants were added to this curve. The integrals of these components were normalized to correspond to their intensities extracted from the γspectra. The obtained time spectra were then fitted with a single-component exponential with a constant background. The effect of each contaminant on the final half-life value of 74Rb was deduced as a differencebetween the half-life value from [16] and the new value given by the fit. The effects of 74Ga and 74mGa were foundto be less than 0.01% on the half-life value. Another possible effect would be induced by a β-decaying T=0 isomeric state in 74Rb. We could set an upper limit of 0.1% for the production cross section of the isomeric state to that of the ground state of 74Rb [13] assuming T1/2(74mRb)=2 s [16]. Even if one assumes a fast Gamow–Teller transition and a half-life of only 0.1 s, the largest possible effect on the half-life of 74Rb would remain below 0.03%. Table 1 Measured half-lives for 74Rb given in milliseconds Run No conditions Eβ>5.2MeV 164.86(28)64.23(49) 264.76(22)64.41(42) 364.94(10)64.96(21) Weighted average 64.90(9)64.77(17) A typical decay-time spectrum is shown in Fig. 1. The residual, defined as a difference between the data and the fit, is plotted in the lower part of the figure and shows no deviation from a singlecomponent exponential decay. The contributions from above discussed sources of systematic uncertainty were quadratically added to the statistical uncertainty of the fit. The results are shown in Table 1. The weightedaverage of theresults are T1/2=64.90(9)ms for the non-gated and T1/2=64.77(17)ms for the energy-gateddata. Due to its better statistical accuracy we adoptthe value of 64.90(9)ms for the final half-life of 74Rb. 148 ISOLDE Collaboration / Physics Letters B 511 (2001) 145–150 3.2. Non-analog βfeeding The role of Coulomb-mixing induced corrections can be studied following the procedure described in Ref. [7] by measuring the branching ratio for nonanalog βdecay to the excited Jπ=0+states. In particular, a strong mixing could be present in the 74Rb– 74Kr case since the first excited Jπ=0+state in 74Kr lies energetically low at 508 keV [19,20]. This state decays via a 52 keV 0+ 2→2+ 1E2 transition, which subsequently decays by 456 keV E2 γemission, or a 508 keV 0+ 2→0+ 1E0 transition. In addition to the fully-convertedground-state transition, the lowenergy E2 transition also largely decays by conversion electron emission with the energy of 38 keV. The relative intensities of these transitions are not known. In the conversion electron measurement our main goal was to search for these decays of the 508 keV 0+ 2state as a signature of a non-analog Fermi transition, and subsequently, Coulomb mixing. Fig. 2 shows the conversion electron spectrum measured in coincidence with β-rays during the first 500 ms after proton pulse impact. The peak at 495 keV whose half-life was determined to be 60(17) ms, in agreement with the half-life of 74Rb, was assigned to the K conversion of the 508 keV 0+ 2→0+ 1E0 transition. The intensity of the peak compared to the total number of βdecays of 74Rb, determined by counting the short-lived decay products, is (3.7± 1.1)×10−4. In the case of the 38 keV peak, only a1σupper limit of 1.6×10−4could be set for its intensity. The above numbers lead to an upper limit of 5.3×10−4for the population of the 508 keV 0+ 2state in the βdecay of 74Rb. The 508 keV state could be also populated in γ decay following Gamow–Teller (GT) feeding of highlying Jπ=1+states in 74Kr. Consequently, a search for such γ-rays was performed. The results are shown in Fig. 3. The spectrum is a sum of β-gated, vetoed and background subtracted data collected with the three Ge detectors. The background subtraction was performed by removing the γspectrum collected during T=600–850 ms after a proton pulse impact from the one collected during T=0–250 ms. No short-lived transitions resulting from the βdecay of 74Rb were observed. In particular, the 1σupper limit for the 456 keV 2+ 1→0+ 1γtransition intensity is 8.1×10−4. The branching ratio for β-delayed proton emission, a signature of additional GT feeding, was deduced to be even lower, <5×10−5. Fig. 2. The conversion electron spectrum collected with the ELLI spectrometer. The inset shows the region around an expected peak at 38 keV. ISOLDE Collaboration / Physics Letters B 511 (2001) 145–150 149 Fig. 3. The sum of the beta-gated and background subtracted gamma spectra collected with the three Ge detectors. The spectra were vetoed with the β-rays observed at the thin scintillators in front of the Ge detectors. The inset shows the expected region of interest for the 456 keV γ-ray. 4. Discussion Different nuclear shapes have been predicted to exist in the neutron-deficient Kr nuclei [21]. These coexisting structures are usually associated with lowlying Jπ=0+states. A large monopole strength has been considered as an experimental signature of a mixing between these different shapes in a nucleus [22,23]. Assuming ΩK=2.64 ×10−8[24] and the ratio ΩK/ΩL=9.8 [25], our results on the decay of the 508 keV state combined with the experimental life-time of the state T1/2=33(7)ns [26] lead to ten times higher lower limit of |ρ(E0)|>0.27 compared to Ref. [20]. This is also in agreement with the value obtained in Ref. [19], |ρ(E0)|=0.30(4). Note that a negligible E2 strength was assumed when extracting the latter number. The result obtained in this work for the production limit of the β-decaying T=0 isomeric state, in addition to its influence on the β-decay half-life, confirms the previous result [16] with an order of magnitudelower value. Also a search for a γ-decaying isomeric state has been performed[27] with a negative result. Thus, the non-existence of a T=0isomeric state is presently well established. This measurement improves the accuracy of the half-life of 74Rb with a factor of five compared to the previous result [16]. The relative uncertainty of 1.4×10−3is, however,still five times larger thanthose measured for the lighter nuclei up to 54Co [2,18]. The largest source of uncertainty is still induced by the statistics. On the other hand, the amount of 74,74mGa and 74mRb starts to play a more significant role when the statistical uncertainty decreases. Reducing the amount of these contaminants and, especially, measuring the decay pattern of 74mGa more carefully will be important goals to further reduce the overall uncertainty. β-delayed proton emission does not contribute significantly to the branching ratio of the Fermi transition. However, our measurements does not rule out a possibility for the GT βtransitions to feed γdecaying 1+states and eventually lead to a population of the 0+state at 508 keV. Due to this and since the 38 keV converted E2 transition was not observed our number for the Coulomb mixing probability has to be considered as an upper limit. Our upper limit of 5.3×10−4for the relative population of the 0+ 2state is about an order of magnitude larger than those observed for the lower- 150 ISOLDE Collaboration / Physics Letters B 511 (2001) 145–150 mass odd–odd MT=0 nuclei from 38mKto54Co [4,7]. However, the deduced limit for the Coulomb mixing component δ1 IM <0.07% (notation from [5]) is only slightly larger than the values obtained for the lower-mass nuclei. The theoretical values for the Coulomb mixing component δ1 IM for 74Rb from [5] are 0.07–0.09% depending on the interaction used in the shell-model calculation. When correcting for the energy difference between the calculated and the experimental 0+states, the values increase to 1.4–3.1%. These values seem to be unrealistically large in the light of our experimental upper limit. Since the total Coulomb correction δCis expected to increase from about 0.5% for A<60 to about 2% for A>60 [4], the small increase in δ1 IM emphasizes the importance of the radial mismatch of the wave functions in forming the major fraction of the total Coulomb correction for 74Rb. 5. Summary Beta-decay properties of 74Rb have been studied at the ISOLDE on-line mass separator facility at CERN. A new determination of the β-decay half-life of 74Rb yielded a consistent, though five times more precise, result compared to the previous value. A refined upper limit could be set for the production of a possible β-decaying T=0 isomeric state in 74Rb. The β-delayed proton emission probability was determined to be negligible for the superallowed Fermi decay systematics. A first estimate of the magnitude of the Coulomb mixing involved in the superallowed β decay of 74Rb could be determined in this work. The upper limit for the Coulomb correction due to different degrees of configuration mixing was shown to be of the the same order of magnitude as obtained for the lighter isotopes previously studied. This emphasizes the importance of the radial mismatch part of the total Coulomb correction in the case of 74Rb. Note added in proof Recently, the half-life of 74Rb was also remeasured at the ISAC Facility at TRIUMF [28] to be 64.761(31) ms. The result has three times smaller uncertainty compared to our result and differs from our adopted result by 1.5 standard deviations. References [1] D.E. Groom et al., Eur. Phys. J. C 15 (2000) 1. [2] J.C. Hardy, I.S. Towner, V.T. Koslowsky, E. Hagberg, H. Schmeing, Nucl. Phys. A 509 (1990) 429. [3] I.S. Towner, J.C. Hardy, Proceedings of 5th International Symposium on Weak and Electromagnetic Interactions in Nuclei: WEIN ’98, June 14–21, Santa Fe, 1998. [4] J.C. Hardy, I.S. Towner, in: C. Baktash (Ed.), Proceedings of Conference Nuclear Structure, August 10–15, Gatlinburg, AIP Conference Proceedings, Vol. 481, Amer. Inst. Phys., New York, 1998. [5] W.E. Ormand, B.A. Brown, Phys. Rev. C 52 (1995) 2455. [6] H. Sagawa, N.V. Giai, T. Suzuki, Phys. Rev. C 53 (1996) 2163. [7] E. Hagberg et al., Phys. Rev. Lett. 73 (1994) 396. [8] J. Äystö et al., ISOLDE Experiment IS384, http://isolde.web. cern.ch/ISOLDE. [9] G. Bollen et al., Nucl. Instrum. Methods Phys. Res. A 368 (1996) 675. [10] D. Lunney et al., Hyperfine Interact. 99 (1996) 105. [11] M. Hjorth-Jensen, private communication, 2000. [12] D. Forkel-Wirth, G. Bollen (Eds.), ISOLDE Laboratory Portrait, Hyperfine Interact. 129 (2000). [13] M. Oinonen et al., Proceedings of 5th International Conference on Radioactive Nuclear Beams, 3–8 April, 2000, Divonne, France, Nucl. Phys. A (2001), in press. [14] A. Honkanen et al., Nucl. Instrum. Methods Phys. Res. A 395 (1997) 217. [15] J.-M. Parmonen et al., Nucl. Instrum. Methods Phys. Res. A 306 (1991) 504. [16] J.M. D’Auria et al., Phys. Lett. 66B (1977) 233. [17] C. Longour et al., Phys. Rev. Lett. 81 (1998) 3337. [18] V.T. Koslowsky et al., Nucl. Instrum. Methods Phys. Res. A 401 (1997) 289. [19] C. Chandler et al., Phys. Rev. C 56 (1997) R2924. [20] F. Becker et al., Eur. Phys. J. A 4 (1999) 103. [21] A. Petrovici, K.W. Schmid, A. Faessler, Nucl. Phys. A 665 (2000) 333, and references therein. [22] K. Heyde, R.A. Meyer, Phys. Rev. C 37 (1988) 2170. [23] J.L. Wood, E.F. Zganjar, C. De Coster, K. Heyde, Nucl. Phys. A 651 (1999) 323. [24] D.A. Bell et al., Can. J. Phys. 44 (1970) 2542. [25] E.L. Church, J. Weneser, Phys. Rev. 103 (1956) 1035; A. Passoja, T. Salonen, JYFL report 2/86, Department of Physics, University of Jyväskylä, 1986. [26] C. Chandler et al., Phys. Rev. C 61 (2000) 044309. [27] D. Rudolph et al., Phys. Rev. Lett. 76 (1996) 376. [28] G.C. Ball et al., Phys. Rev. Lett. 86 (2001) 1454.