2005MNRAS.364L..23N

A few dozen young high-mass stars orbit Sgr A* at distances as short as ∼0.1 pc, where star formation should be quenched by the strong tidal shear from Sgr A*. The puzzling young stellar population is believed to come into existence in one of two ways: (i) `normal' star formation at several tens of parsecs in a very massive star cluster that then spiralled in; or (ii) star formation in situ in a massive self-gravitating disc. We propose to constrain these two scenarios via the expected X-ray emission from young low-mass stars that should have formed alongside the massive stars. To this end we compare the recent Chandra observations of X-ray emission from young stars in the Orion nebula, and the Chandra observations of the Sgr A* field. We show that the cluster spiral-in model is ruled out irrespectively of the initial mass function (IMF) of the young stars. In addition, for the in situ model, we find that no more than a few thousand low-mass stars could have formed alongside the massive stars. This is more than a factor of 10 fewer than expected if these stars were formed with the standard IMF as elsewhere in the Galaxy. The young stars in the Galactic Centre are thus the first solid observational evidence for star formation in active galactic nucleus (AGN) discs and also require the IMF of these stars to be top-heavy. We briefly consider the implications of these results for AGN in general.