Nuclear structure problems solved by realistic nucleonic interaction*

I will present selected results of self-consistent mean-field (SCMF) calculations using semi-realistic interactions, aiming at disclosing roles of specific channels of the nucleonic interaction at qualitative and quantitative level. In practice, studies in this line seem to solve some puzzles in the nuclear structure physics. The semi-realistic interaction of the M3Y-type is applied, which are based on the G-matrix, and include the tensor force without any phenomenological modification. With such tensor force, the level inversion of $p0d_{3/2}$ and $p1s_{1/2}$ from $^{40}$Ca to $^{48}$Ca is nicely reproduced. The deformation of $^{80}$Zr is described well, with keeping nearly doubly magic nature of $^{90}$Zr, implying that the realistic tensor force operates with a good balance. Second, it has been pointed out that the three-nucleon LS interaction based on the chiral EFT could account for missing part of origin of the $\ell s$ splitting. Incorporating it into the SCMF calculations, long-standing puzzles with respect to charge radii of spherical nuclei can be solved. The kink in the isotope shifts in the Pb nuclei is reproduced reasonably well, without fictitious degeneracy of $n1g_{9/2}$ and $n0i_{11/2}$. Moreover, the same interaction accounts for almost equal charge radii between $^{40}$Ca and $^{48}$Ca. This seems consistent with the \textit{ab initio} result, and may anatomize it. The SCMF calculations with the same interaction also describe rapid increase of the charge radii from $^{48}$Ca to $^{52}$Ca, and predict a kink for the isotope shifts in the Sn chain at $N=82$.