Structure and kinematics of shocked gas in Sgr B2: further evidence of a cloud–cloud collision from SiO emission maps

Abstract

ABSTRACT We present SiO J = 2–1 maps of the Sgr B2 molecular cloud, which show shocked gas with a turbulent substructure comprising at least three cavities at velocities of $[10,40]\, \rm km\, s^{-1}$ and an arc at velocities of $[-20,10]\, \rm km\, s^{-1}$. The spatial anticorrelation of shocked gas at low and high velocities, and the presence of bridging features in position-velocity diagrams suggest that these structures formed in a cloud–cloud collision. Some of the known compact H ii regions spatially overlap with sites of strong SiO emission at velocities of $[40,85]\, \rm km\, s^{-1}$, and are between or along the edges of SiO gas features at $[100,120]\, \rm km\, s^{-1}$, suggesting that the stars responsible for ionizing the compact H ii regions formed in compressed gas due to this collision. We find gas densities and kinetic temperatures of the order of $n_{\rm H_2}\sim 10^5\, \rm cm^{-3}$ and $\sim 30\, \rm K$, respectively, towards three positions of Sgr B2. The average values of the SiO relative abundances, integrated line intensities, and line widths are ∼10−9, $\sim 11\, \rm K\, km\, s^{-1}$, and $\sim 31\, \rm km\, s^{-1}$, respectively. These values agree with those obtained with chemical models that mimic grain sputtering by C-type shocks. A comparison of our observations with hydrodynamical simulations shows that a cloud–cloud collision that took place $\lesssim 0.5\, \rm Myr$ ago can explain the density distribution with a mean column density of $\bar{N}_{\rm H_2}\gtrsim 5\times 10^{22}\, \rm cm^{-2}$, and the morphology and kinematics of shocked gas in different velocity channels. Colliding clouds are efficient at producing internal shocks with velocities $\sim 5\!-\!50\, \rm km\, s^{-1}$. High-velocity shocks are produced during the early stages of the collision and can readily ignite star formation, while moderate- and low-velocity shocks are important over longer time-scales and can explain the widespread SiO emission in Sgr B2.