Any particle dark matter ( DM ) scenario featuring a suppressed power spectrum of astrophysical relevance results in a delay of galaxy formation .
As a consequence , such scenarios can be constrained using the global 21-cm absorption signal initiated by the UV radiation of the first stars .
The Experiment to Detect the Global Epoch of Reionization Signature ( EDGES ) recently reported the first detection of such an absorption signal at redshift \sim 17 .
While its amplitude might indicate the need for new physics , we solely focus on the timing of the signal to test non-cold DM models .
Assuming conservative limits for the stellar-to-baryon fraction ( f _ { * } < 0.03 ) and for the minimum cooling temperature ( T _ { vir } > 10 ^ { 3 } Kelvin ) motivated by radiation-hydrodynamic simulations , we are able to derive unprecedented constraints on a variety of non-cold DM models .
For example , the mass of thermal warm DM is limited to m _ { TH } > 6.1 keV , while mixed DM scenarios ( featuring a cold and a hot component ) are constrained to a hot DM fraction below 17 percent .
The ultra-light axion DM model is limited to masses m _ { a } > 8 \times 10 ^ { -21 } eV , a regime where its wave-like nature is pushed far below the kiloparsec scale .
Finally , sterile neutrinos from resonant production can be fully disfavoured as a dominant DM candidate .
The results of this paper show that the 21-cm absorption signal is a powerful discriminant of non-cold dark matter , allowing for significant improvements over to the strongest current limits .
Confirming the result from EDGES is paramount in this context .