The frozen (inactive) disk in Sgr A: freezing the accretion of the hot gas too?

The black hole in our Galactic Center is extremely underluminous for the amount of hot gas available for accretion. Theoretical understanding of this fact rests on a likely but not entirely certain assumption that the electrons in the accreting gas are much cooler than the protons. In this case the hot gas as a whole is too hot to accrete, and is too tenous to radiate away its gravitational energy. Here we propose a drastically different picture of the accretion process in Sgr A* not based on the unchecked two-temperature assumption. Namely, we argue that there should exist a very cold inactive disk -- a remnant of a past stronger accretion activity in Sgr A*. Such a disk would be a very efficient cooling surface for the hot flow. We show that under certain conditions the cooling due to thermal conduction cannot be balanced by the viscous heating in the hot flow. Along with the heat, the hot flow loses its viscosity and thus ability to accrete. It settles (condences) onto the cold disk slighlty inside of the circularization radius. If the latter is very large, then the liberated energy, and the luminosity emitted, is orders of magnitude less than naively expected. We build a simple analytical model for this flow and calculate the expected spectra that appear to be in a very reasonble agreement with observations. Strong additional support for the presence of the inactive disk comes from the recent observations of X-ray flares in Sgr A*. The properties of these flares are very similar to those produced by stars passing through a cold disk.