Recent experiments at the JYFLTRAP Penning trap

The JYFLTRAP double Penning trap mass spectrometer at the Ion Guide Isotope Separator On-Line (IGISOL) facility offers excellent possibilities for high-precision mass measurements of radioactive ions. Around 400 atomic masses, including around 50 isomeric states, have been measured since JYFLTRAP became operational. JYFLTRAP has also been used as a high-resolution mass separator for decay spectroscopy experiments as well as an ion counter for fission yield studies. In this contribution, an overview of recent activities at the JYFLTRAP Penning trap is given, with a focus on nuclei discussed in the PLATAN2019 meeting.


Introduction
Atomic masses can be measured with the highest achievable precision using Penning trap mass spectrometers. Very high precision (≤ 1 keV/c 2 ) mass measurements are important for fundamental physics [1,2], such as for weak-interaction studies, testing the isospin symmetry or the unitarity of the Cabibbo-Kobayashi-Maskawa (CKM) quark mixing matrix. Mass measurements with a good precision of ≤ 10 keV/c 2 are also needed to study subtle changes in nuclear structure [2,3] or to provide accurate inputs for astrophysical reaction network calculations [4]. In this contribution, we review the recent activities at the JYFLTRAP double Penning trap mass spectrometer [5][6][7]  as the preparation trap. In the first trap, the ions of interest are selected using the buffer-gas cooling technique [12], and sent further to the second trap for precision mass measurements, or to post-trap decay spectroscopy station after the traps. Usually the Time-of-Flight Ion Cyclotron Resonance (TOF-ICR) technique [13,14] has been employed in the second trap, also known as the measurement trap, to determine the ion's cyclotron resonance frequency ν c = 1 2π q m B, where B is the magnetic field strength, and q and m are the charge and the mass of the ion, respectively. The magnetic field strength is calibrated by using reference ions with well-known mass values. A new technique to determine the ion's cyclotron resonance frequency, the Phase-Imaging Ion Cyclotron Resonance (PI-ICR) technique [15], was commissioned at JYFLTRAP in 2017, see Ref. [7]. A position-sensitive MCP ion detector with a delay-line anode (DLD40) from RoentDek GmbH [16] was installed behind the traps, and the extraction optics was modified accordingly. Since then, the new technique has been increasingly utilized for high-precision mass measurements at JYFLTRAP. The method gives consistent results with the conventional TOF-ICR technique, as demonstrated with the Q-value measurement for the neutrinoless double electron-capture on 102 Pd [17], but a higher precision can be achieved with the PI-ICR technique. Further examples of TOF-ICR and PI-ICR measurements are given in Section. 3.

Recent mass measurements with JYFLTRAP
Since the last LASER meeting held in Poznan in May 2016, several publications have come out from JYFLTRAP experiments. The Q EC value of the superallowed β + emitter 42 Sc has been measured to be 6426.350(53) keV [18], the isobaric multiplet mass equation studied for the quintet at A = 52 [19] and a general review on ion traps given in [3]. These were already discussed in Poznan. Here we present recent activities at JYFLTRAP with a focus on the PLATAN2019 contributions.

Nuclei in the vicinity of 78 Ni
Masses of nuclei close to 78 Ni are important for several reasons. Firstly, the evolution of the Z = 28 and N = 50 shell closures and the magicity of 78 Ni can be probed via mass measurements. Secondly, the masses of neutron-rich nuclei close to N = 50, i.e. the strength of the N = 50 shell closure, are relevant to understand in detail the core collapse phase of supernovae where electron captures on nuclei play a key role [20,21]. In a recent JYFLTRAP experiment, several neutron-rich Ni, Cu and Zn isotopes were measured to study these phenomena [22].
We also investigated nuclides below 68 Ni (Z = 28, N = 40), namely 69,70 Co and 67 Fe. 68 Ni has some doubly magic features, such as high excitation energy of the first 2 + state and low transition strength B(E2; 0 g.s. → 2 + 1 ) [23,24]. Previous mass measurements, however, have not shown a strong subshell closure at N = 40 in the region [25][26][27][28][29]. The new JYFLTRAP measurements support the conclusion that the N = 40 subshell closure is rather weak, and gets weaker below nickel [30,31]. The identification of the isomeric states in the cobalt isotopes is crucial for accurate empirical shell-gap energies and also for the investigation of a possible island of inversion below 68 Ni [32,33]. At JYFLTRAP, only one state was observed for 70 Co, whereas for 69 Co the data indicate that there are two states present.
The mass measurements of 69 Co and 67 Fe are also important for accurate neutron separation energies needed to calculate neutron-capture and their inverse photodisintegration rates in the astrophysical rapid neutron capture process (r process) [34,35]. Sensitivity studies [36] have shown that 67 Fe(n, γ ) 68 Fe and 68 Co(n, γ ) 69 Co have a strong impact on the calculated r-process abundances. With the new JYFLTRAP measurements, we could reduce the mass-related uncertainties in these impactful neutron-capture rates. Moreover, the photodisintegration rates are much higher than previously considered.

Neutron-rich silver isotopes
Nuclei close to the doubly magic 132 Sn are important both for nuclear structure and for the r-process. Mass measurements in this region provide information on the evolution of the Z=50 and N =82 shell closures, one-and two-neutron separation energies and pairing effects in the region. Recent studies have shown that the masses of the nuclei close to 132 Sn have the highest impact on the calculated r-process abundances for different astrophysical scenarios [37].
Recently neutron-rich silver isotopes 113−124 Ag were investigated at JYFLTRAP. Previously, 112,114−124 Ag have been studied at the ISOLTRAP Penning trap at CERN [38] and 125,126 Ag at the ESR storage ring at GSI [39]. Of these, the ESR measurements have uncertainties of 200-300 keV which are not adequate for the r-process modeling or detailed studies of nuclear structure. In addition, the 126 Ag mass obtained at ESR is 730(370) keV lower than the extrapolated value in the Atomic Mass Evaluation 2016 [40]. The ISOLTRAP measurements using the TOF-ICR technique were hampered by the existence of low-lying isomeric states and difficulties to identify the measured state. In these measurements the state for 115 Ag and 119 Ag could not be assigned as the ground state or isomer, and 121−124 Ag were assumed to be an admixture of the states which increased the uncertainty of the measurements. 120 Ag was assigned as the ground state in [38], however, a decay spectroscopy study performed at Holifield Radioactive Ion Beam Facility (HRIBF) showed that 120 Ag actually has three long-living (>10 ms) states [41]. Therefore, it is unclear which state was measured at ISOLTRAP, the ground state or the first isomeric state.
The mass measurements of silver isotopes at JYFLTRAP were performed with the PI-ICR technique [7,15]. Phase accumulation times in the measurement trap had to be selected for each case individually depending on the excitation energies of the isomers to fully separate them. Stable 133 Cs + ions were used for the calibration of the magnetic field. The measured states in the studied silver isotopes are shown in Table 1.
The excitation energies of the 119m,120m,122n,123m,124m Ag isomeric states were measured for the first time. The precision for several ground-state mass values was improved, since the isomeric states were separated in most of the cases. In 121 Ag only one state was observed, while the low-lying isomeric state is additionally known in literature [42]. In 122 Ag two states with different half-lives were observed, while three long-living states are known in literature [42]. The ground state and the first isomeric state with an unknown excitation energy in 122 Ag have similar half-lives and could not be distinguished. The excitation energies of the isomeric states in 113−118 Ag are known with a good accuracy (sub-keV) from spectroscopic measurements [42] and can be used to cross-check our mass measurements. Figure 2 shows the projection of ion cyclotron motion in the measurement trap onto the position-sensitive MCP detector for one of the PI-ICR measurements of 120 Ag.

Neutron-rich rare-earth isotopes
Neutron-rich rare-earth isotopes have been studied in two recent mass measurement campaigns at JYFLTRAP [43,44]. Altogether 22 nuclides have been measured, of which 14  [47].
In the latter experimental campaign, the masses of 162 Eu and 163 Gd were remeasured using the TOF-ICR technique with a 1600 ms quadrupolar excitation time in the second trap to resolve the ground and isomeric states from each other. In addition, the PI-ICR technique was applied for a detailed study of the ground and isomeric states in 162 Eu, see Ref. [44].
The rare-earth masses are important for understanding the formation of the rare-earth abundance peak at A = 165 in the r process. It has been proposed to form via fission cycling [48] or during the freezeout when matter is decaying toward the stability. In the latter scenario, a kink in the neutron separation energies could funnel the flow toward the midshell [49,50]. Spectroscopic data indicate that there is an onset of deformation at N = 88−90: the 2 + energies drop dramatically and the E(4 + )/E(2 + ) ratios increase suddenly. Moreover, it has been suggested that there would be a subshell closure at N = 100 based on the observed small kink in the 2 + energies at N = 100. In the JYFLTRAP campaign we wanted to investigate whether there is a kink in one-or two-neutron separation energies that could funnel the r-process flow or would support a subshell closure at N = 100.
The new JYFLTRAP data do not introduce significant changes in the trends of twoneutron separation energies (see Fig. 3). No significant kinks supporting the proposed subshell closure or a change in the nuclear structure, are observed for the studied isotopic chains. Interestingly, neutron pairing energies were found to be lower than predicted by the commonly used theoretical models when approaching the midshell at N = 104 [43,44].
The impact of the JYFLTRAP mass values on the calculated r-process abundances was studied for a representative dynamical ejecta trajectory for a 1.35 solar-mass neutron-star merger from Ref. [51], with a very low initial electron fraction Y e = 0.016 and low entropy per baryon s/k B = 8. A simple asymmetric split [52] was assumed for fission fragment distributions to ensure that the rare-earth peak forms entirely via the dynamical formation mechanism of Refs. [49,50]. Compared to a baseline study employing experimental mass values from AME16 [40] and theoretical mass values from the Finite-Range Droplet Model  [53], the new JYFLTRAP values produce a smoother abundance pattern and a better agreement with the observed r-process abundances is achieved [43,44].

Q -value measurements of rare weak decays
Decays through weak interaction offer a possibility to determine the mass of a neutrino. In ordinary β decays, the mass of a neutrino manifests itself as a distorted shape of the emitted beta spectrum close to the endpoint energy (i.e., Q-value). Such an experiment is KATRIN that uses tritium, whose β endpoint energy is 18.6 keV [54]. To detect a distortion in the beta spectrum near the end-point energy, a Q-value as small as possible is desirable. This has prompted a survey to map rare weak decays [55,56] that are β + , β − or EC decays of parent ground states to excited states in the daughter nucleus with a small Q-value (i.e., < 1 keV). The shape of the beta spectrum near the endpoint is expected to be relatively simple. These decays are also suitable for theoretical modeling of atomic effects in nuclear decay.
Several candidates for rare weak decays have been identified, all with well-known excitation energies in the daughter nucleus. However, the uncertainties in the ground-stateto-ground-state Q-values remain too high to extract a precise Q-value for a decay to an excited state. A direct parent-daughter mass ratio measurement with a relative precision of around 10 −9 would allow extraction of Q-values at the uncertainty level of a few hundred eV. It is imperative to know whether the Q-value is positive to find out the decays that could be feasible for determining the mass of a neutrino.
At JYFLTRAP, Q-values of four potential rare weak β-decay candidates were measured recently, summarized in Table 2. As the nuclei of interest are close to stability, they were produced relatively easily. Out of the studied nuclei, 135 Cs was produced using proton-induced fission of natural uranium while the others were produced using proton-or deuteron-induced fusion reactions. The analysis of the collected data is ongoing.  The first column shows the decaying parent state, the second column the excited state in the daughter nucleus, third column the excitation energy of the excited state in the daughter and the fourth column the Q-value to the excited state in the daughter. The tabulated spins and parities (J π ) of the states, as well as the excitation energies and Q-values are based on Refs. [42,57]

Nuclei close to the N = Z line in the A = 80 − 90 mass region
Heavier neutron-deficient nuclei close to the N = Z line have been recently studied using an upgraded version of the heavy-ion ion-guide, HIGISOL [58], at IGISOL. The first online experiment with the upgraded system employed 222-MeV 36 Ar 8+ ions impinging into a nat Ni target [59]. High-precision mass measurements were performed with JYFLTRAP for 82 Zr, 84 Nb, 86 Mo, 88 Tc, 88 Tc m and 89 Ru [59,60]. Two of the masses, 88 Tc m and 89 Ru, were measured for the first time and the precisions of 82 Zr, 84 Nb and 88 Tc were improved significantly. Additionally, the mass of 86 Mo given in AME16 [40] was verified. The effect of the new data on the mass surface was studied, and similar behaviour as reported in the literature [40] was observed. 88 Tc was studied at JYFLTRAP already in 2008 [61] but at the time the isomeric state 88 Tc m could not be resolved from the more abundant ground state. With the PI-ICR technique, the mass of the isomer could be determined for the first time. The isomer was measured against the ground state using the PI-ICR technique, yielding an excitation energy of E x = 70.4(31)keV [59]. The mass of the dominantly produced ground state was measured using the TOF-ICR technique, which together with the excitation energy yielded also a mass value for the isomer.
The order of the three lowest states in 88 Tc was studied based on the obtained excitation energy for the isomeric state, Weisskopf estimates for the three states, and available spectroscopic data from literature. The most likely energies and spin-parities of the first three states in 88 Tc are (E x , J π ) = (0 keV, 2 + ), (70.4 keV, 6 + ) and (95 keV, 4 + ). Shell-model calculations were also performed for comparison but those were highly sensitive to the used model space and interaction model (for details, see Ref. [59]).
In addition to the six atomic masses that were directly measured, the masses of 82 Mo and 86 Ru (T z = −1) were determined using theoretical mirror displacement energies (MDEs) and the directly measured masses of their respective mirror partners 82 Zr and 86 Mo (T z = +1) [59]. The resulting mass-excess values predict more tightly bound nuclei than literature [40], by more than 500 keV, and reduce the uncertainties of the predicted masses.
The JYFLTRAP mass measurements of 82 Zr and 84 Nb showed that mass values measured at the CSRe storage ring [62] deviate from the corresponding Penning-trap measurements, CSRe results being typically around 20 keV smaller. In order to investigate whether this is a more general feature, published results from the CSRe storage ring were gathered and compared to available Penning-trap results. A total of 17 nuclides were available for the comparison. The deviation turned out to be larger than what would be expected based on statistical fluctuations alone [59].

Other activities at JYFLTRAP
In addition to precision mass measurements, JYFLTRAP has been employed in many posttrap decay spectroscopy experiments. Either the mass-selective buffer gas cooling technique [12] in the first trap, or dipolar Ramsey cleaning [63] in the second trap, has been used to select the isotope, or even the isomer of interest for spectroscopic studies after the trap. Monoisotopic or isomeric beams are beneficial for decay experiments as the background coming from the decays of neighbouring isobars is removed. Only daughter activities will be present but those can be reduced by implanting the ions into a tape which is regularly moved using a tape station.
Since the last LASER meeting held in May 2016 in Poznan, several trap-assisted spectroscopy studies employing JYFLTRAP have been published. A beta detector and an array of broad-energy range germanium detectors have been used to study the beta decays of 88 Se [64] and 117 Rh [65]. The TASISpec silicon detector array together with a cluster and clover detectors were employed to study beta decay of 127 Cd [66]. Total absorption gamma-ray spectroscopy has been performed on the beta decays of 87 Br, 88 Br, and 94 Rb [67], 100 Tc [68] and 100gs,100m Nb and 102gs,102m Nb [69]. The study of the first multiple-beta-delayed neutron emitter above A = 100, 136 Sb [70], also involved JYFLTRAP to select only the 136 Sb ions for the studies.
JYFLTRAP has also turned out to be a great instrument to determine independent fission yields [71]. Neutron-rich isotopes at IGISOL are typically produced by proton-induced fission on nat U or 232 Th target. The yields can be determined based on the quadrupole excitation scans performed in the first trap [12]. If the isomeric excitation energy is high enough, even independent isomeric yield ratios can be obtained from these studies. The results for isomeric yield ratios for 81 Ge, 96,97 Y, 128,130 Sn, and 129 Sb in proton-induced fission on nat U and 232 Th at 25 MeV were reported in [72]. The new PI-ICR technique was recently employed for isomeric fission yield studies of 119−127 Cd and 119−127 In for 25-MeV proton-induced fission on nat U [73]. The benefit in the PI-ICR technique is that it has a superior resolving power as compared to the buffer-gas cooling technique in the first-trap as demonstrated in Fig. 1 of Ref. [73]. Hence, even low-lying isomeric states (E x ≤ 100 keV) and their fission yields can be studied. Moreover, there is no need to scan over a range of excitation frequencies. Every ion directly contributes to the yield measurement.

Summary and outlook
The scientific programme at JYFLTRAP is versatile, ranging from nuclear structure and astrophysics to experiments for neutrino physics or testing fundamental physics, such as isospin symmetry or the unitarity of the CKM quark mixing matrix. It is also used as an ion counter for fission yield studies and high-resolution mass separator for decay spectroscopy experiments. The recently commissioned PI-ICR technique has opened new possibilities to resolve and measure very low-lying isomeric states, as well as to count ions for isomeric fission yield studies. A multi-reflection time-of-flight mass spectrometer (MR-TOF) is under commission at JYFLTRAP. The MR-TOF will provide a faster method for mass measurements and help in reducing the isobaric background for the Penning-trap mass measurements. Combined with the broad range of radioactive nuclides produced at IGISOL, the prospects for high-precision mass measurements at JYFLTRAP look very promising in the future.