Global polarization of $\Lambda$ and $\bar{\Lambda}$ hyperons in Au+Au collisions at $\sqrt{s_{\rm NN}}=19.6$ and $27$ GeV

In relativistic heavy-ion collisions, a global spin polarization, $P_\mathrm{H}$, of $\Lambda$ and $\bar{\Lambda}$ hyperons along the direction of the system angular momentum was discovered and measured across a broad range of collision energies and demonstrated a trend of increasing $P_\mathrm{H}$ with decreasing $\sqrt{s_{\rm NN}}$. A splitting between $\Lambda$ and $\bar{\Lambda}$ polarization may be possible due to their different magnetic moments in a late-stage magnetic field sustained by the quark-gluon plasma which is formed in the collision. The results presented in this study find no significant splitting at the collision energies of $\sqrt{s_{\rm NN}}=19.6$ and $27$ GeV in the RHIC Beam Energy Scan Phase II using the STAR detector, with an upper limit of $P_{\bar{\Lambda}}-P_{\Lambda}<0.24$% and $P_{\bar{\Lambda}}-P_{\Lambda}<0.35$%, respectively, at a 95% confidence level. We derive an upper limit on the na\"ive extraction of the late-stage magnetic field of $B<9.4\cdot10^{12}$ T and $B<1.4\cdot10^{13}$ T at $\sqrt{s_{\rm NN}}=19.6$ and $27$ GeV, respectively, although more thorough derivations are needed. Differential measurements of $P_\mathrm{H}$ were performed with respect to collision centrality, transverse momentum, and rapidity. With our current acceptance of $|y|<1$ and uncertainties, we observe no dependence on transverse momentum and rapidity in this analysis. These results challenge multiple existing model calculations following a variety of different assumptions which have each predicted a strong dependence on rapidity in this collision-energy range.