COSINUS model-independent sensitivity to the DAMA/LIBRA dark matter signal

Abstract COSINUS is a dark matter direct detection experiment using NaI crystals as cryogenic scintillating calorimeters. If no signal is observed, this will constrain the dark matter scattering rate in sodium iodide. We investigate how this constraint can be used to infer that the annual modulation signal observed in the DAMA/LIBRA experiment cannot originate from dark matter nuclear recoil events, independently of the dark matter model. We achieve this by unfolding the DAMA modulation spectrum to obtain the implied unquenched nuclear recoil spectrum, which we then compare to the expected COSINUS sensitivity. We find that assuming zero background in the signal region, a 1 σ , 2 σ or 3 σ confidence limit exclusion can be obtained with 57, 130 or 250 kg day of exposure, respectively. A simple background model indicates that in the presence of background, the exposure requirements may increase by ∼ 30%.