Direct Mass Measurement For Short-lived Neutron-rich Nuclides

The nuclear mass is a fundamental nuclear property which reflects the total binding energy of the nucleus and therefore all interactions among the nucleons. The correct description of nuclear binding energies has been a main challenge for nuclear models. In the last decades a variety of experimental methods and nuclear models have been developed. In general, the models have achieved a good accuracy for known masses at and near the valley of beta-stability. However, large deviations have been observed for unknown regions and especially for neutron-rich nuclei. These discrepancies are also seen in the predictions for basic nuclear properties such as the evolution of shell closures, deformations, pairing strength and the location of drip lines. Furthermore, an accurate knowledge of nuclear masses and lifetimes is required to understand the pathways of the nucleosynthesis in the stars. Presently, the theoretical predictions of different mass models can lead to quite different astrophysical conditions. These arguments are the strong motivation for new accurate experimental mass values, especially for neutron-rich nuclides.The operation of modern radioactive beam facilities has launched novel accurate methods for direct mass measurements worldwide. The combination of the in-flight separator FRS and the cooler-storage ring ESR at GSI provides unique experimental conditions with stored bare and few-electron rare isotopes for all elements up to uranium at relativistic energies. Two methods have been developed to measure the masses of stored exotic nuclei circulating in the ESR. In the first method, the Schottky Mass Spectroscopy (SMS) the large phase space of the fragments is reduced by electron cooling which enforces the stored ions to an identical mean velocity and reduces the velocity spread to roughly 3·10^-7 at low intensities. The limitation of this method for short-lived exotic nuclides is the duration of the cooling process which takes a few seconds. Therefore, a complementary method, the Isochronous Mass Spectrometry(IMS) has been pioneered for very short-lived exotic nuclei investigated with the FRS-ESR facility. For IMS the storage ring is operated in its isochronous mode where the revolution time is independent of the velocity spread of the circulating fragments. Like SMS, IMS also has the ultimate sensitivity down to a single ion.In this thesis, we report on the first large-scale IMS experiment of masses of short-lived uranium fission fragments applying this B_ρ-tagging. In the dedicated experiment, neutron-rich exotic nuclei were produced by projectile fission of 411 MeV/u ²³⁸U ions in a 1 g/cm² Be target placed at the entrance of the FRS. The intensity of the primary beam was about 2·10⁹ particles per spill, where each spill was taken every 20 seconds. At these energies, the fission fragments emerged from the target mainly as bare ions or were carrying, with lower probability, one or two bound electrons. The energy of the primary beam was chosen such that the energy of ¹³³Sn⁵⁰⁺ fragments corresponded to the relativistic Lorentz factorγ=1.41, which was exactly the isochronous energy of the ESR. All other fragments within the acceptance of the FRS-ESR facility were transmitted and stored as well. The transmission of the FRS-ESR facility has been optimized with the primary beam at the identical mean magnetic rigidity as the chosen reference fragment ¹³³Sn⁵⁰⁺. The rate of fragments stored in the ESR was on average one stored particle per measurement cycle. More than 13 000 independent measurements have been performed.The revolution times of the stored ions were about 500 ns. Time stamps were measured with a dedicated time-of-flight (TOF) detector which is equipped with a thin 17μg/cm² carbon foil. Secondary electrons were released from the foil due to each ion penetrating the foil at each revolution. These electrons were guided by crossed electric and magnetic fields to a set. of microchannel plate detectors. The signals from the detector were recorded with a commercial fast sampling oscilloscope Tektronix TDS 6154C of 40 Giga Samples per second and stored for off-line analysis. The recording lime of 1 ms was chosen to measure up to about 2000 revolutions in the ESR.The revolution times were obtained from the measured time stamps for each stored ion and were accumulated in a revolution time spectrum (TOF spectrum). In total 71 different nuclides, ranging from aluminum to barium, have been unambiguously identified in the accumulated TOF spectrum. Every hundred injections into the ESR have been accumulated into separate TOF spectra each having about 50 different stored particles. These spectra were analyzed by means of the correlation matrix approach originally developed for the SMS data analysis. In this method, only the relations between the masses in a given spectrum are analyzed. The spectra are connected together via the same nuclides in the same or different charge states present in different spectra. In this way, the absolute values of the revolution times are not essential and possible shifts of the spectra leading to broadening of the peaks, e.g. due to the slow varying electro-magnetic fields of the FRS-ESR, can be removed. The merit of the extended IMS including the matrix analysis is a strongly improved mass resolving power of m/â–³m(FWHM) = 2×10⁵. In our analysis all nuclides are included for which more than 10 particles have been recorded. This corresponds to about one stored specific isotope per day which corresponds to a production cross sections of in the order of pb. The data analysis are shown in details.The main difficulty which also limits the present mass accuracy is the lack of reliable reference masses in these neutron-rich nuclides. Most of the known masses here were obtained fromβ-endpoint measurements. However, it has been shown meanwhile that such measurements can have significant systematical errors. Large deviations to the Atomic Mass Evaluation (AME03) are also observed in recent direct measurements with penning traps. For instance, the newly measured masses for ^105-107Nb isotopic chain deviate from the tabulated values by 943(100), 905(200) and 1205(400) keV, respectively.We have used in total 36 reference masses which had to obey strict selection criteria. Their mass values had to be obtained by at least two different experimental methods which agree to each other. The uncertainty of the mass value had to be smaller than 50 keV. The corresponding values were collected from the AME'03 compilation and recent JYFLTRAP results.Masses of 35 nuclides have been determined with a typical accuracy of 120 keV. The systematic error of 119 keV was estimated. For all measured mass values this systematical error was quadratically added to the statistical error which was 2-60 keV. Masses for 8 nuclides (^85,86As,⁸⁹Se,¹²³Ag,¹³⁸Te, ^140,141I,¹⁴³Xe) are measured for the first time. The mass values obtained for ⁸⁴As and ¹²²Ag can correspond to mixtures with unresolved isomeric states with estimated excitation energies of 0(100) keV and 80(50) keV, respectively, according to NUBASE. The isomeric states in the ^134,135Te isotopes do not need to be considered due to their very short half-lives of 164 ns and 510 ns, respectively. Our directly measured value for ⁸³Ge is in very good agreement to the AME'03 extrapolation, but deviates from the recent value determined via (d,p) reaction by 447(133) keV. In addition, one isomeric state with half-live of 16μs for¹³³Sb has been identified and for the first time its excitation energy has been determined as 4564(30) keV, which is in perfect agreement, with the estimated values from independentγ-ray measurements.In order to check the obtained results we compared our mass values with the values from recent JYFLTRAP experiments. In this comparison the JYFLTRAP data were not used as reference masses. It turns out that both experimental data are in excellent agreement.The masses for ³³Al, ¹³⁴Te, ¹³⁷Te and ¹³⁹I were remeasured and show discrepancies larger than 1.5 standard deviations from the AME'03 values. We note that the tabulated values for ¹³⁴Te,¹³⁷Te and ¹³⁹I are based onβ-endpoint measurements. In general, the newly measured masses show that the AME'03 extrapolations over-bind the nuclear mass surface in the neutron rich nuclei. Based on our new measured masses, a global comparison of 9 masses not evaluated in AME'03 yieldsσ_rms deviations of 346, 481, 484, 616, 683 and 1185 keV from the FRDM, ETFSI-Q, Dufio-Zuker(28), HFB-14 and RMF models, respectively. The nuclear structure information like the separation energies, shell gaps are extracted. Due to our new mass value for ³³Al we obtain a lower gap energy at the quenched N = 20 shell closure. Our updated mass value for ¹³⁴Te yields 0.4 MeV larger shell gap towards the doubly-magic nuclide ¹³²Sn. Furthermore, our measured masses has some effect on the astrophysical r-process in the environment of low neutron density.The further development of the IMS based on the present facility focus both the accuracy of the time-of-flight detection and the ion optical of the ESR. The IMS still can contribute to the more neutron-rich nuclides as well as the N = Z nuclides near the proton drip-line with reliable and precise references. For the future, the collect ring (CR) will be built to take the place of the ESR, and works in the isochronous model thus is more suitable for the large-scale mass measurement towards the drip-lines with an improved beam quality.