Chiral three-nucleon forces and the evolution of correlations along the oxygen isotopic chain

Background: Three-nucleon forces (3NFs) have nontrivial implications on the evolution of correlations at extreme proton-neutron asymmetries. Recent ab initio calculations show that leading-order chiral interactions are crucial to obtain the correct binding energies and neutron driplines along the O, N, and F chains [A. Cipollone, C. Barbieri, and P. Navratil, Phys. Rev. Lett. 111, 062501 (2013)]. Purpose: Here we discuss the impact of 3NFs along the oxygen chain for other quantities of interest, such has the spectral distribution for attachment and removal of a nucleon, spectroscopic factors, and radii. The objective is to better delineate the general effects of 3NFs on nuclear correlations. Methods: We employ self-consistent Green's function (SCGF) theory which allows a comprehensive calculation of the single-particle spectral function. For the closed subshell isotopes, O-14, O-16, O-22, O-24, and O-28, we perform calculations with the Dyson-ADC(3) method, which is fully nonperturbative and is the state of the art for both nuclear physics and quantum chemistry applications. The remaining open-shell isotopes are studied using the newly developed Gorkov-SCGF formalism up to second order. Results: We produce complete plots for the spectral distributions. The spectroscopic factors for the dominant quasiparticle peaks are found to depend very little on the leading-order (NNLO) chiral 3NFs. The latter have small impact on the calculated matter radii, which, however, are consistently obtained smaller than experiment. Similarly, single-particle spectra tend to be too spread with respect to the experiment. This effect might hinder, to some extent, the onset of correlations and screen the quenching of calculated spectroscopic factors. The most important effect of 3NFs is thus the fine tuning of the energies for the dominant quasiparticle states, which governs the shell evolution and the position of driplines. Conclusions: Although present chiral NNLO 3NFs interactions do reproduce the binding energies correctly in this mass region, the details of the nuclear spectral function remain at odds with the experiment showing too-small radii and a too-spread single-particle spectrum, similar to what has already been pointed out for larger masses. This suggests a lack of repulsion in the present model of NN + 3N interactions, which is mildly apparent already for masses in the A = 14-28 mass range.