Doping dependence of the low temperature planar carrier density in overdoped YBa$_2$Cu$_3$O$_{7-\delta}$

Whether a quantum critical point (QCP) demarcates the end of the pseudogap (PG) regime in hole-doped cuprates at a singular doping level $p^* \approx 0.19$ remains an open question. A crucial part of this puzzle is how the carrier density predicted by electronic structure calculations is recovered for $p > p^*$. Here, we use magnetic fields up to 67 T to suppress superconductivity down to 50 K, allowing simultaneous measurement of the low-temperature Hall number $n_{\mathrm{H}}$ and the in-plane resistivity anisotropy $\rho_a/\rho_b$ in overdoped Y$_{1-x}$Ca$_x$Ba$_2$Cu$_3$O$_{7-\delta}$ single crystals. We confirm a previous finding [Badoux et al., Nature 531, 210 (2016)] that $n_{\mathrm{H}}$(50 K) exhibits a sharp increase below $p^*$. Using the measured resistivity anisotropy, we extract the planar carrier density $n_{\mathrm{pl}} = n_{\mathrm{H}} (\rho_a/\rho_b)^{-1}$. The doping dependence of $n_{\mathrm{pl}}$(50 K) reveals two key findings: (i) at optimal doping, $n_{\mathrm{pl}} \approx p$, and (ii) the sharp rise in $n_{\mathrm{H}}(p)$ is softened such that the full Fermi volume ($n_{\mathrm{pl}} = 1 + p$) is only partially recovered at $p^*$. This result disfavors a conventional QCP scenario in which the PG endpoint corresponds to a reconstructed Fermi surface.