• black hole physics
• cosmology
• radiative processes

In this work we want to study a possible way in which supermassive black holes can be formed: motivations for this are given by recent observations of the bright quasars at z=6 that are thought to be powered by central black holes with a mass of the order of 10^9 Msun, inferred from their observed luminosity.
We use a monte carlo code written by Shaun Cole to simulate the mass distribution and merger history of dark matter halos at various redshifts in a statistical way; we then set the initial conditions for baryons at redshift z=20 and track their evolution to z=6 coupling simple analytical prescriptions to the dark matter halo merger tree. Several other authors adopted this way of proceeding but we want to stress that we are trying to do something new: we focus our attention in the interplay between baryons, stars, light black hole seeds (10^2 Msun), heavy black hole seeds (10^5 Msun) and the radiative UV background that triggers structure formation. The problem is highly non trivial because we have to proceed in a self-consistent way: the strength of the radiative background selects different channels for structure formation inside the dark matter potential wells; the structures formed, on the other hand, determine the UV background at subsequent times. We focus on the prediction of the final central object mass, to see if it can reach the observed masses of 10^9 Msun and if it can match the empirical relation with the mass of the galaxy bulge mbulge/mBH=10^3; we also try to obtain a first indication of the evolution of the X luminosity function of quasars with redshift, that can in principle be tested with future observations. These results should serve as an indication for future work to be carried with more sophisticated tools as N-body simulations; nevertheless, our work has the advantage that it includes in the same model different physical processes that take place on a wide range of spatial scales (from 10^7 cm, that is the Schwarshild radius of a stellar black hole to 1Mpc=10^24 cm that is the typical size of the systems we examine), a feature that is very difficult to achieve with N-body simulations due to resolution limits.