First observation of ²⁸O

The investigation of rare isotopes with large neutron-to-proton (N/Z) imbalances is a highly active area in nuclear physics. These nuclei provide stringent tests of nuclear structure theories, particularly sophisticated ab initio approaches where nucleon interactions are derived from first principles. Light, neutron-rich isotopes exhibit the most extreme N/Z ratios. Beyond the limits of nuclear binding – the neutron drip line – nuclei exist as very short-lived (≈10⁻²¹ s) resonances decaying via spontaneous neutron emission. Their energies and lifetimes depend critically on the underlying nuclear structure. Experimentally, these nuclei are accessible only for the lightest systems. The location of the neutron drip line has been established up to neon (Z=10), and the heaviest observed neutron-unbound nucleus for fluorine (Z=9) is ²⁸F. The observation of such systems, particularly those with extreme neutron excess like the hypothetical tetraneutron, is crucial for testing our understanding of nuclear forces and structure. This paper reports the first direct observation of ²⁸O (N/Z=2.5), unbound to four-neutron decay, and its neighboring isotope ²⁷O (three-neutron unbound). The study of these isotopes is particularly important because ²⁸O, with Z=8 and N=20, is predicted in the standard shell model to be a doubly magic nucleus, a cornerstone of our understanding of nuclear structure. The confirmation (or refutation) of its doubly magic nature will provide valuable insight into the evolution of shell structure in exotic nuclei.

The concept of nuclear shell closures, corresponding to proton and neutron numbers 2, 8, 20, 28, 50, 82, and 126, is fundamental to understanding nuclear structure. Nuclei with these 'magic' numbers exhibit enhanced stability and spherical shapes due to large energy gaps between shells. Doubly magic nuclei, possessing magic numbers for both protons and neutrons, serve as crucial benchmarks for theoretical models. The existence of doubly magic nuclei like ⁷⁸Ni (Z=28, N=50) and ¹³²Sn (Z=50, N=82) has been confirmed. However, the doubly magic nature of ⁶He (Z=2, N=4) remains less certain. The N=20 shell closure, however, is known to be significantly altered in neutron-rich nuclei around Ne, Na, and Mg isotopes (Z=10-12) – a region referred to as the 'Island of Inversion' (IoI). In this region, the energy gap between the sd and pf shells is reduced or vanishes, leading to deformed nuclei with configurations involving neutron excitations into the pf shell. Recent studies have extended the IoI to neighboring fluorine isotopes ²⁸F and ²⁹F. Conversely, the last particle-bound oxygen isotope, ²⁴O, has been found to exhibit a new shell closure at N=16, suggesting a complex interplay of shell structure and nuclear forces in this region. The structure of heavier neutron-rich oxygen isotopes, particularly ²⁸O, therefore remains an open question of significant interest.

The neutron-unbound ²⁷O and ²⁸O isotopes were produced using proton-induced nucleon knockout reactions from a 235 MeV per nucleon beam of ²⁹F at the RIKEN RI Beam Factory. A thick (151 mm) liquid-hydrogen target maximized luminosity, while the MINOS Time Projection Chamber, surrounding the target, allowed for precise determination of the reaction vertex. The SAMURAI spectrometer, including the NeuLAND and NEBULA neutron detector arrays, measured the momenta of the reaction products (charged fragments and fast neutrons). This setup allowed for the direct detection of the decay products of ²⁷O and ²⁸O: ²⁴O and three or four neutrons, respectively. The overall detection efficiency for three-neutron and four-neutron decays was approximately 2% and 0.4%, respectively, at decay energies of 0.5 MeV. Decay energies were reconstructed using the invariant-mass technique with a resolution (full width at half maximum, FWHM) of about 0.2 MeV at 0.5 MeV decay energy. Data analysis involved identifying ¹⁶O fragments using magnetic rigidity, energy loss, and time-of-flight measurements. Neutrons were identified based on their time-of-flight and energy deposited in the scintillators. Dedicated offline procedures were implemented to reject crosstalk events where a single neutron was mistakenly registered in multiple scintillators. The decay energies were reconstructed from the measured momenta of the decay products. A Monte Carlo simulation was developed to model the experimental response function, including the contribution from residual crosstalk events. The sequential decay of ²⁸O through ²⁶O was investigated by analyzing correlations in the ²⁴O+neutron subsystems. A similar analysis was performed for ²⁷O. The decay energies of ²⁷O and ²⁸O resonances were determined from a fit of the decay energy spectra, taking into account the experimental response functions and residual crosstalk. The cross-section for single-proton removal from ²⁹F populating the ²⁸O resonance was also deduced using the distorted-wave impulse approximation (DWIA).

The experiment successfully observed both ²⁷O and ²⁸O for the first time through their decay into ²⁴O and multiple neutrons. ²⁸O was found to have a decay energy of E₀₁₂₃₄ = 0.46 ± 0.04(stat) ± 0.02(syst) MeV, with an upper limit on the resonance width of 0.7 MeV (68% confidence interval). The cross-section for its production from ²⁹F via single-proton removal was determined to be 1.36 ± 0.13(syst) mb. Similarly, a ²⁷O resonance was identified with a decay energy determined through fitting and consideration of experimental response and residual crosstalk. The measured decay energies of ²⁷O and ²⁸O were compared to various theoretical predictions including large-scale shell-model calculations (using the EEdf3 interaction derived from chiral effective field theory, χEFT) and coupled-cluster calculations augmented with a statistical approach. Most theoretical models underbind both ²⁷O and ²⁸O by approximately 1 MeV. The coupled-cluster calculation results, with their associated uncertainties, showed partial consistency with the experimental results, suggesting the experimental measurements provide valuable constraints for ab initio calculations. The spectroscopic factor extracted from the ²⁹F(p,2p)²⁸O reaction cross section was C²S = 0.48 ± 0.10. This value, when compared to theoretical predictions, indicated that the N=20 shell closure is not present in ²⁸O, suggesting that the Island of Inversion extends into the oxygen isotopes. This conclusion supports the notion that the pf-shell neutron configurations significantly contribute to the structure of ²⁸O.

The successful observation of ²⁷O and ²⁸O, particularly ²⁸O's relatively low decay energy, challenges the predictions of the standard shell model which anticipated ²⁸O as a doubly magic nucleus with a significant energy gap at N=20. The experimental findings support the picture of a weakened or absent N=20 shell closure in this region, extending the influence of the Island of Inversion into the oxygen isotopes. The discrepancy between theoretical predictions and experimental decay energies, even for sophisticated ab initio models, highlights the need for further refinements in nuclear interaction models. The relatively large uncertainties in the theoretical predictions underscore the importance of precise experimental measurements like those presented here in constraining such models. The spectroscopic factor extracted from the cross-section measurement provides compelling evidence against the existence of a closed N=20 shell in ²⁸O. The agreement between the experimental data and the theoretical calculations, particularly those that incorporate continuum effects, reinforces our understanding of the breakdown of traditional shell model predictions in the context of exotic, neutron-rich nuclei.

This paper reports the first observation of the extremely neutron-rich oxygen isotopes ²⁷O and ²⁸O. Both isotopes were found as low-lying resonances, a finding made possible by a state-of-the-art experimental setup capable of detecting multiple neutrons. The experimental data, particularly the ²⁸O decay energy and the single-proton removal cross-section, provide crucial constraints for theoretical models of nuclear structure. The results confirm that the Island of Inversion extends into the oxygen isotopes and challenge the traditional shell model's predictions for ²⁸O, showing that it is not a doubly magic nucleus. Future research could focus on determining the excitation energy of the first 2⁺ state in ²⁸O to further understand its structure and the influence of the Island of Inversion.

The limited detection efficiency of the neutron detectors (around 2% for three-neutron and 0.4% for four-neutron decays) could have affected the precision of the decay energy measurements and cross-section determination. The reliance on Monte Carlo simulations to account for crosstalk events introduces some systematic uncertainty into the results. While several theoretical models were compared, a complete exploration of all possible theoretical frameworks was not possible within the scope of this study. Finally, the analysis relies on certain assumptions regarding the decay pathways and the assignment of angular momenta.