Boosted or Inelastic? Discriminating Interpretations of the LZ 248 keV Event

Abstract

The LUX-ZEPLIN experiment has reported a single nuclear recoil candidate at $E_R = 248 \pm 23(\mathrm{stat}) \pm 23(\mathrm{sys})$~keV, disfavouring the background-only hypothesis at a global significance of $2.6σ$. The difficulty such an event poses is not the recoil energy itself but the absence of any accompanying excess at low energy as elastic scattering of halo dark matter yields a monotonically falling spectrum, and supplying the required momentum transfer $q \simeq 246$~MeV already demands $m_χ\gtrsim 79$~GeV. We confront the event with the two kinematically distinct mechanisms that evade this limitation, endothermic inelastic dark matter, in which a mass splitting $δ\sim \mathcal{O}(100)$~keV forbids low-energy recoils and boosted dark matter, in which a light relativistic flux supplies the momentum, treating both with the same model-independent scalar--scalar, pseudoscalar--scalar and pseudoscalar--pseudoscalar effective operators. The two scenarios prove spectrally distinguishable. The inelastic spectra sit near the observed energy for every operator, placing only $6$--$17\%$ of events below 150~keV, whereas the boosted spectra depend critically on the operator: the scalar and pseudoscalar--scalar interactions place $99\%$ and $92\%$ of their events below 150~keV, while the pseudoscalar--pseudoscalar interaction places $75\%$ above it. Momentum dependence is thus essential to the boosted interpretation, but in the inelastic case it trades against the splitting, the preferred $δ$ decreasing monotonically from $\mathcal{O}_{ss}$ to $\mathcal{O}_{ps}$ to $\mathcal{O}_{pp}$. Because the scenarios differ across the whole high-energy window, a handful of additional events would separate them, placing the question within reach of the full LZ exposure.