Optimizing superconductivity: from cuprates via nickelates to palladates

Motivated by cuprate and nickelate superconductors, we perform a comprehensive study of the superconducting instability in the single-band Hubbard model. We calculate the spectrum and superconducting transition temperature $T_{\rm c}$ as a function of filling and Coulomb interaction for a range of hopping parameters, using the dynamical vertex approximation. We find the sweet spot for high $T_{\rm c}$ to be at intermediate coupling, moderate Fermi surface warping, and low hole doping. Combining these results with first principles calculations, neither nickelates nor cuprates are close to this optimum within the single-band description. Instead, we identify some palladates, notably RbSr$_2$PdO$_3$ and $A^{\prime}_2$PdO$_2$Cl$_2$ ($A^{\prime}$=Ba$_{0.5}$La$_{0.5}$), to be virtually optimal, while others, such as NdPdO$_2$, are too weakly correlated.