Volume 536, December 2011
|Number of page(s)||25|
|Section||Interstellar and circumstellar matter|
|Published online||05 December 2011|
Star formation near an obscured AGN
Variations in the initial mass function⋆
Kapteyn Astronomical Institute, University of Groningen PO Box 800 9700 AV Groningen The Netherlands
Received: 7 June 2011
Accepted: 29 September 2011
The conditions that affect the formation of stars in radiatively and mechanically active environments are quite different from the conditions that apply to our local interstellar neighborhood. In these galactic environments, a variety of feedback processes can play a significant role in shaping the initial mass function (IMF). Here, we present a numerical study on the effects of an accreting black hole and the influence of nearby massive stars to a collapsing, 800 M⊙, molecular cloud at 10 pc distance from the black hole. Our work focusses on the star-forming ISM in the centers of (ultra-)luminous infrared galaxies ((U)LIRGS). We therefore assume that this region is enshrouded by gas and dust and that most of the UV and soft X-ray radiation from the broad line region (BLR) is attenuated along the line of sight to the model cloud. We then parametrize and study radiative feedback effects of hard X-rays emanating from the black hole BLR, increased cosmic ray rates caused by supernovae in starbursts, and strong UV radiation produced by nearby massive stars. We also investigate the importance of shear from the supermassive, 106 − 108 M⊙, black hole as the star-forming cloud orbits around it. A grid of 42 models is created and calculated with the hydrodynamical code FLASH. We find that thermal pressure from X-rays compresses the cloud, which induces a high star-formation rate early on, but reduces the overall star-formation efficiency (SFE) to about 7% through gas depletion by evaporation. We see that the turn-over mass of the IMF increases up to a factor of 2.3, Mturn = 1 − 1.5 M⊙, for the model with the highest X-ray flux (160 erg s-1 cm-2), while the high-mass slope of the IMF becomes Γ ≳ − 1 (ΓSalpeter = − 1.35). This results in more high-mass stars and a non-Salpeter IMF. Cosmic rays penetrate deeply into the cloud and increase the gas temperature to about 50 K for rates that are roughly 100 times Galactic and 200 K for 3000 times Galactic, which leads to a reduced formation efficiency of low-mass stars. While the shape of the mass function is preserved, high cosmic ray rates increase the average mass of stars, thereby shifting the turn-over mass to higher values, i.e., up to several solar masses. Owing to this process, the onset of star formation is also delayed. We find that UV radiation plays only a minor role. Because UV photons cannot penetrate a dense, n ≥ 105 cm-3, cloud deep enough, they only affect the late time accretion by heating the medium where the cloud is embedded in. When we increase the black hole mass, for a cloud that is at 10 pc distance, the turbulence caused by shearing effects reduces the SFE slightly. Furthermore, shear weakens the effect of the other parameters on the slope of the IMF as well as the turn-over mass. The run with the most massive black hole, however, causes so much shear that the hydrodynamics is completely dominated by this effect and it severely inhibits star formation. We conclude that the IMF inside active galaxies is different from the one obtained from local environments. We also find that the combined effects of X-rays, cosmic rays, UV, and shear tend to drive toward a less pronounced deviation from a Salpeter IMF.
Key words: radiative transfer / methods: numerical / stars: formation / stars: luminosity function, mass function / cosmic rays / X-rays: ISM
Figures 6–17 are available in electronic form at http://www.aanda.org
© ESO, 2011
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.