The Milky Way Nuclear Star Cluster
Roberto Capuzzo-Dolcetta, Fabio Antonini, Alessandra Mastrobuono-Battisti
SStellar Clusters & Associations: A RIA Workshop on GaiaGranada, 23 rd - 27 th May 2011
The Milky Way Nuclear Star Cluster
R. Capuzzo-Dolcetta , F. Antonini , and A. Mastrobuono-Battisti Dep. of Physics, Sapienza, Univ. di Roma, Italy Dep. of Physics, Rochester Institute of Technology, Rochester, USA
Abstract
In the center of the Milky Way, as well as in many other galaxies, a compact star clusteraround a very massive black hole is observed. One of the possible explanations for theformation of such Nuclear Star Clusters is based on the ‘merging’ of globular clusters in theinner galactic potential well. By mean of sophisticated N-body simulations, we checked thevalidity of this hypothesis and found that it may actually has been the one leading to theformation of the Milky Way Nuclear Star Cluster.
The existence of very compact star clusters around the centers of many galaxies is wellascertained, nowadays. These clusters, called Nuclear Star Clusters (NSCs) are among thedensest star clusters observed, with effective radii of a few pc and central luminosities upto 10 L (cid:12) . Recently, the study of dense stellar systems has raised new interest because ofthe discovery, in an ever increasing number of galaxies, of compact nuclei in form of stellarresolved systems and/or NSCs which are now known to be present in galaxies across thewhole Hubble sequence and not only in the dE, N galaxies (Bekki & Graham, 2010). Aradial profile and velocity structure for a NSC can be determined only for the Milky WayNSC which is close enough that it can be resolved into individual stars. The Milky Way NSChas an estimated mass of 10 M (cid:12) , and it hosts a massive black hole whose mass, 4 . × M (cid:12) , is uniquely well determined.The formation mechanism of nuclear star clusters is unknown. Two competing models arepossible. In the gas model, a NSC can form from the gas that migrates to the center of thegalaxy where then forms stars. A variety of scenarios have been proposed to account for therequired fast radial inflow of gas into the galactic center, including the magneto-rotationalinstability in a differentially rotating gas disk, tidal compression in shallow density profilesor dynamical instabilities. Alternatively, in the merger model, massive clusters migrate tothe center via dynamical friction and merge to form a dense nucleus (Tremaine et al., 1975;Capuzzo-Dolcetta, 1993). Observations of NSCs in dE galaxies suggest that the majority, but a r X i v : . [ a s t r o - ph . GA ] J u l The Milky Way Nuclear Cluster not all, dE nuclei, could be the result of packing mass in form of orbitally decayed globularclusters (GCs) (Lotz et al., 2004). Numerical simulations have also shown that the basicproperties of NSCs, including their shape, mass density profile, and mass-radius relation arereproduced in the merger model under a variety of explored conditions (Capuzzo-Dolcetta etal., 2008a,b; Hartmann et al., 2011).
As we said, a NSC is present in the central region of the Milky Way. For obvious reasons, it isthe only NSC which can be resolved into individual stars in spite of its high density (Sch¨odelet al., 2007); moreover, detailed kinematic studies exist that allows a precise determinationof the mass of the central black hole, M BH = 4 . × M (cid:12) . Moreover, the relaxation timeat Sgr A ∗ influence radius is robustly estimated as 20 – 30 Gyr (Merritt, 2010) suggestingthere has not been time for a Bahcall-Wolf cusp to rise. This is consistent with the almostflat density distribution of late type stars within 0.5 pc from the center (Buchholz et al.,2009). As stated in the Introduction, at least two competing models are possible for theformation of the MW NSC. These mechanisms must account for the observed evidence of theMW NSC, i.e.: a mass of the NSC M NSC ∼ M (cid:12) , a density profile with a core of about0.5 pc and decreasing as r − . up to r (cid:39)
30 pc. Out of 30 pc there is a large nuclear stellarand molecular disk of approximately same radius.
We modeled the final evolution of a set of massive GCs which decayed in the inner regionof the Milky Way due to the dynamical friction braking exerted by the galactic stellar field.The details of the computations and the extended presentation and discussion of results aregiven in Antonini et al. (2011); here we give just a brief summary.Our Galaxy, likely triaxial in its inner part, is dense enough to make the orbits of GCs withmass around 10 M (cid:12) to decay (by dynamical friction) to the inner 50 pc in a time significantlyshorter than a Hubble time. We thus consider 12 such GCs and their final evolution toward amerger state studied via N-body simulations of their motion in the inner region of the MilkyWay, represented self-consistently as a set of N particles. The N body galactic environment issampled from a shallow ( ρ ∝ r − / ) cusp; the contribution of the central massive black holeis also considered. Simulations are done with both the high precision, parallel PhiGRape(Harfst et al., 2007) and NBSymple (Capuzzo-Dolcetta et al., 2011) codes running on theRIT Grape cluster and on the CPU+NVIDIA TESLA C2050 platform in Roma, Sapienza. The merging occurs rather quickly: after about 20 crossing times the resulting system at-tains a quasi equilibrium configuration, as it is shown by the almost steady time profile ofLagrangian radii. This corresponds to a slowly evolving (via relaxation) super stellar cluster(SSC) which oscillates around the massive black hole. Figure 1 shows the growth with timeof the spatial density profile after the various merger events. The core structure is conserved. . Capuzzo-Dolcetta et al. r ≤ . r ≥ T ∼ . As already shown by previous papers (Capuzzo-Dolcetta, 1993; Capuzzo-Dolcetta et al.,2008a,b) the dynamical friction dragging of massive globular clusters toward the galacticcenter is a viable explanation of the local high densities. The novelty of this work is theinclusion in the simulations of a massive black hole in the galactic center. Its role, althoughimportant in the vicinity of its influence radius, does not alter the general characteristics of themerger event of a set of GCs falling to the center. The merger reaches a sort of quasi-steadystate, slowly evolving due to internal relaxation which is partly affected by the presence ofthe massive black hole. With regard to the Milky Way nuclear cluster, it is relevant notinghow the original ‘core’ profile of the GCs (the NSC building blocks) is maintained for a timelong enough to justify the core actually observed in the MW NSC. Moreover, the outwardprofile radial slope, ρ ∝ r − , of the merger system is in good agreement (within 30 pc from The Milky Way Nuclear Cluster