Dark matter implications from the XENONnT and LZ data

Abstract

We investigate a possible dark matter origin of the high-energy nuclear-recoil-like events in XENONnT and LZ data, which cannot be explained by standard elastic spin-independent WIMP scattering. Using our unified DIAMX framework, built on openly available data and likelihood models, we perform the first combined profile-likelihood fits to multiple WIMP-search datasets with a total exposure of 7.3 tonne$\times$year. We investigate that two broad classes of dark matter nucleon interactions, with velocity-dependent cross-section or inelastic (endo- and exothermic) scattering, can reproduce the observed high-energy recoil spectrum, reaching local significances up to $4σ$. We further quantify the impact of $^{124}$Xe double electron capture (DEC) backgrounds, finding that variations in the poorly known DEC charge yields can shift the inferred significances from below $1σ$ to $4σ$. We point out that extending the same analysis to XENONnT and LZ data with recoil energies up to 300 keV, once available, will provide a powerful test of the dark matter interpretation, since the $^{124}$Xe DEC background is expected to be negligible in this high-energy range.

Figures (5)

• Figure 1: Event distributions in NR energy for XENONnT (top) and LZ (bottom). Data points are selected below or at the NR median and above the $2\sigma$ lower bound of the NR band in corrected (S1, S2); for XENONnT we additionally require cS2 $>$ 400 PE to mitigate low-energy mismodeling. Error bars show 1$\sigma$ Feldman–Cousins intervals for signal+background counts. The black solid curve and the gray band give the predicted background under the same selection criteria, while the colored dashed curves show background plus signal for four DM benchmarks: spin-independent (SI) elastic scattering with $m_{\mathrm{DM}}=200~\mathrm{GeV}$ (fixed to the 90% C.L. upper limit of the cross section reported in Ref. LZ:2024zvo), $\mathcal{Q}_3^{(7)}$ with $m_{\mathrm{DM}}=100~\mathrm{GeV}$, endothermic inelastic scattering with $\delta = 150~\mathrm{keV}$ and $m_{\mathrm{DM}} = 1~\mathrm{TeV}$, and exothermic inelastic scattering with $\delta = -200~\mathrm{keV}$ and $m_{\mathrm{DM}} = 1~\mathrm{TeV}$, the latter three evaluated at their combined XENONnT–LZ best-fit normalizations. Dotted lines show the predicted number of events outside the WIMP search region of interest (ROI), using the same normalization and extrapolated efficiencies but omitting the NR-band selections.
• Figure 2: Allowed 90% C.L. regions for the reciprocal of the cutoff scale $1/\Lambda$ (shaded bands, left $y$-axis) and local discovery significances (lines, right $y$-axis) as functions of the DM mass for the DMEFT operators $\mathcal{Q}_3^{(7)}$ and $\mathcal{Q}_4^{(7)}$. Dashed (dotted) lines denote the discovery significances if only XENONnT (LZ) datasets are considered in the inference, while the solid lines show the combined discovery significances.
• Figure 3: Contours of the local signal significance for endothermic (blue) and exothermic (green) inelastic DM in the $(m_{\mathrm{DM}}, \delta)$ space. The black dashed line marks the kinematic threshold below which endothermic scattering in xenon is kinematically forbidden. The side and top/bottom panels show the corresponding local significances marginalized over $m_{\mathrm{DM}}$ and $\delta$, respectively.
• Figure 4: Predicted total number of events inside (dashed) and outside (solid) the WIMP search ROI for the four DM benchmarks, assuming the exposure of the 7.3 tonne $\times$ year of XENONnT and LZ. The models include $\mathcal{Q}_7^{(3)}$, $\mathcal{Q}_7^{(4)}$, endothermic inelastic scattering with $\delta = 150~\mathrm{keV}$, and exothermic inelastic scattering with $\delta = -200~\mathrm{keV}$. All four are evaluated at their combined XENONnT–LZ best-fit normalizations.
• Figure 5: Corrected (S1, S2) distributions for all XENONnT and LZ datasets used in this analysis. Each event is represented by a pie chart showing the fractional contributions of the local probability density from different components of the best-fit endothermic inelastic DM model. In the 3×2 panel layout, the first five panels (the top two rows and the lower-left panel) use the best-fit benchmark ($\delta = 135~\mathrm{keV}$, $m_{\mathrm{DM}} = 70~\mathrm{GeV}$) under the baseline DEC charge-yield treatment (Case I), while the lower-right panel uses the Case II best-fit benchmark ($\delta = 130~\mathrm{keV}$, $m_{\mathrm{DM}} = 60~\mathrm{GeV}$). The pie size is proportional to the local endothermic signal fraction. Shaded regions indicate the $1\sigma$ and $2\sigma$ contours for ER (blue) and neutron (orange) backgrounds. Solid and dashed curves show the corresponding contours for the endothermic signal (red) and the $^{124}$Xe background (green), respectively. Dotted lines show the NR median curves for each dataset and illustrate the data selection criteria used in Fig. \ref{['fig:nr_below_median']}.