Heating up neutron stars with inelastic dark matter

Neutron stars can provide new insight into dark matter properties, as these dense objects capture dark matter particles very efficiently. It has recently been shown that the energy transfer in the dark matter capture process can lead to appreciable heating of neutron stars, which may be observable with forthcoming infra-red telescopes. We examine this heating in the context of inelastic dark matter, for which signals in conventional nuclear-recoil based direct detection experiments are highly suppressed when the momentum transfer is small compared to the mass splitting between dark matter states. Neutron stars permit inelastic scattering for much greater mass splittings, because dark matter particles are accelerated to velocities close to the speed of light during infall. Using an effective operator approach for fermionic DM that scatters inelastically, we show that the observation of a very cold neutron star would lead to very stringent limits on the interaction strengths that, in most cases, are much stronger than any present, or future, direct detection experiment on Earth. This holds both for elastic scattering and for inelastic scattering with mass splittings up to similar to 300 MeV.