2009ApJ...693.1554F

We analyze the available sample of double quasars, and investigate their physical properties. Our sample comprises 85 pairs, selected from the Sloan Digital Sky Survey (SDSS). We derive physical parameters for the engine and the host, and model the dynamical evolution of the pair. First, we compare different scaling relationships between massive black holes and their hosts (bulge mass, velocity dispersion, and their possible redshift dependences), and discuss their consistency. We then compute dynamical friction timescales for the double quasar systems to investigate their frequency and their agreement with the "merger driven" scenario for quasar triggering. In optical surveys, such as the SDSS, N_double,qso/N_qso~ 0.1%. Comparing typical merging timescales to expected quasar lifetimes, the fraction of double quasars should be roughly a factor of 10 larger than observed. Additionally, we find that, depending on the correlations between black holes and their hosts, the occurrence of double quasars could be redshift dependent. Comparison of our models to the SDSS quasar catalog suggests that double quasars should be more common at high redshift. We compare the typical separations at which double quasars are observed to the predictions of merger simulations. We find that the distribution of physical separations peaks at ∼30 kpc, with a tail at larger separations (∼100-200 kpc). The peak of the distribution is roughly consistent with the first episode of quasar activity found in equal mass mergers simulations. The tail of the quasar pairs distribution at large separations is instead inconsistent with any quasar activity predicted by published simulations. These large separation pairs are instead consistent with unequal mass mergers where gas is dynamically perturbed during the first pericentric passage, but the gas reaches the black hole only at the next apocenter, where the pair is observed.