Comments on "First Dark Matter Results from the XENON100 Experiment"

TL;DR

This work questions the reliability of XENON100's claimed limits on light WIMPs, highlighting substantial uncertainties in the LXe scintillation response at low recoil energies. It analyzes two measurement approaches for $L_{eff}$ and demonstrates a pronounced low-energy drop not captured by XENON100's constant-$L_{eff}$ assumption. A simple kinematic argument indicates that light WIMPs with $m_{WIMP} \sim 7$ GeV/$c^2$ would yield recoil energies predominantly below a few keV_r, reducing LXe's sensitivity in this mass range. The authors argue for more conservative, data-driven $L_{eff}$ modeling and explicit uncertainty accounting, which would weaken some exclusions and affect comparisons with DAMA/CoGeNT, underscoring the need for direct low-energy scintillation evidence in LXe.

Abstract

The XENON100 collaboration has recently released new dark matter limits, placing particular emphasis on their impact on searches known to be sensitive to light-mass (below 10 GeV/c^2) Weakly Interacting Massive Particles (WIMPs), such as DAMA and CoGeNT. We describe here several sources of uncertainty and bias in their analysis that make their new claimed sensitivity presently untenable. In particular, we point out additional work in this field and simple kinematic arguments that indicate that liquid xenon (LXe) may be a relatively insensitive detection medium for the recoil energies (few keV_r) expected from such low mass WIMPs.

Figures (5)

• Figure 1: Measurements of $\mathcal{L}_{\text{eff}}$ in LXe. The red vertical arrow indicates the calculated value for the kinematic cutoff in recoil energy (see text). The most recent analysis by the XENON10 collaboration peter, not considered in xenon100, follows the same trend as in Manzur et al.dan1 (dark blue points here). Light-mass WIMPs like those claimed to be excluded in xenon100 concentrate their signal beyond the left margin of this figure. A constant $\mathcal{L}_{\text{eff}}\sim$0.12 below $\sim$10 keV$_{r}$ is used in xenon100 to obtain dark matter limits.
• Figure 2: Examples of comparisons between simulation (MC) and response of LXe chambers to neutron sources theiraps (see text). The dramatic disagreement at few keV$_{r}$ can be alleviated to some extent via trigger efficiency corrections, but not completely peter.
• Figure 3: Examples of measurements of quenching factor for different scintillators, indicating in each case the calculated value for the kinematic cutoff in recoil energy (see text). An additional value of E$_{max}$=2.2 keV$_{r}$ is found for carbon recoils in organic scintillator, in good agreement with spooner.
• Figure 4: Models for $\mathcal{L}_{\text{eff}}$ used to generate the exclusions in Fig. 5. The blue line corresponds to the extrapolation to low energy used in xenon100. Black points are recent data from Manzur et al.dan1. The black line is an adiabatic fit to these (see text). The red line (logistic fit) and band correspond to ZEPLIN data zeplin (see text).
• Figure 5: XENON100 exclusions generated from the $\mathcal{L}_{\text{eff}}$ models in Fig. 4. (same color coding). The dotted blue line corresponds to the problematic $\mathcal{L}_{\text{eff}}$ contour in Fig. 1 in xenon100, ambiguously claimed by XENON100 to be both a 90% lower C.L. to their $\mathcal{L}_{\text{eff}}$ best fit and an extrapolation to Manzur et al.dan1, a crucial point we contend is misleading. Ion channeling is not included neither for CoGeNT nor DAMA regions andreasgondolo. The low-energy trend expected for the quenching factor in NaI(Tl) review, not included here, can have the effect of displacing the DAMA region away from XENON100 constraints (see text).