We report Hubble Space Telescope (HST) Near-infrared Camera and Multiobject Spectrometer (NICMOS) observations of the Arches and Quintuplet clusters, two extraordinary young clusters near the Galactic Center. For the first time, we have identified main sequence stars in the Galactic Center with initial masses well below 10 M ⊙ . We present the first determination of the initial mass function (IMF) for any population in the Galactic Center, finding an IMF slope which is significantly more positive (Γ ≈ −0.65) than the average for young clusters elsewhere in the Galaxy (Γ ≈ −1.4). The apparent turnoffs in the color-magnitude diagrams suggest cluster ages which are consistent with the ages implied by the mixture of spectral types in the clusters; we find τ age ∼ 2±1 Myr for the Arches cluster, and τ age ∼ 4±1 Myr for the Quintuplet. We estimate total cluster masses by adding the masses of observed stars down to the 50% completeness limit, and then extrapolating down to a lower mass cutoff of 1 M ⊙ . Using this method, we find ∼ >10 4 M ⊙ for the total mass of the Arches cluster. Such a determination for the Quintuplet cluster is complicated by the double-valued mass-magnitude relationship for clusters with ages ∼ > 3 Myr. We find a lower limit of 6300 M ⊙ for the total cluster mass, and suggest a best -2estimate of twice this value which accounts for the outlying members of the cluster. Both clusters have masses which place them as the two most massive clusters in the Galaxy.

We present the first results of a near-infrared (0.96-2.31 lm) spectroscopic survey of M, L, and T dwarfs obtained with NIRSPEC on the Keck II telescope. Our new survey has a resolving power of R ¼ =D $ 2000 and is comprised of two major data sets: 53 J-band (1.14-1.36 lm) spectra covering all spectral types from M6 to T8 with at least two members in each spectral subclass (wherever possible), and 25 flux-calibrated spectra from 1.14 to 2.31 lm for most spectral classes between M6 and T8. Sixteen of these 25 objects have additional spectral coverage from 0.96 to 1.14 lm to provide overlap with optical spectra. Spectral flux ratio indexes for prominent molecular bands are derived, and equivalent widths (EWs) for several atomic lines are measured. We find that a combination of four H 2 O and two CH 4 band strengths can be used for spectral classification of all these sources in the near-infrared and that the H 2 O indexes are almost linear with spectral type from M6 to T8. The H 2 O indexes near 1.79 and 1.96 lm should remain useful beyond T8. In the near-infrared a notable feature at the boundary between the M and L types is the disappearance of relatively weak (EW $ 1-2 Å ) atomic lines of Al i and Ca i, followed by Fe i around L2. At the boundary between L and T dwarfs it is the appearance of CH 4 in all near-infrared bands (J, H, and K) that provides a significant spectral change, although we find evidence of CH 4 as early as L7 in the K band. The FeH strength and the equivalent width of the K i lines are not monotonic, but in combination with other factors provide useful constraints on spectral type. The K i lines are sensitive to surface gravity. The CO band strength near 2.30 lm is relatively insensitive to spectral class. The peak calibrated flux (F ) in the 0.96-2.31 lm region occurs near 1.10 lm at M6 but shifts to about 1.27 lm at T8. In addition, the relative peak flux in the J, H, and K bands is always in the sense J > H > K except around L6, where the differences are small. One object, 2MASS 2244+20 (L6.5), shows normal spectral behavior in the optical but has an infrared spectrum in which the peak flux in J band is less than at H and K.

We present and use new spectra and narrow-band images, along with previously published broad-band images, of stars in the Arches cluster to extract photometry, astrometry, equivalent width, and velocity information. The data are interpreted with a wind/atmosphere code to determine stellar temperatures, luminosities, mass-loss rates, and abundances. We have doubled the number of known emission-line stars, and we have also made the first spectroscopic identification of the main sequence for any population in the Galactic Center. We conclude that the most massive stars are bona-fide Wolf-Rayet (WR) stars and are some of the most massive stars known, having M_{initial} > 100 Msun, and prodigious winds, Mdot > 10^{-5} Msun yr^{-1}, that are enriched with helium and nitrogen; with these identifications, the Arches cluster contains about 5% of all known WR stars in the Galaxy. We find an upper limit to the velocity dispersion of 22 kms^{-1}, implying an upper limit to the cluster mass of 7(10^4) Msun within a radius of 0.23 pc; we also estimate the bulk heliocentric velocity of the cluster to be v_{cluster,odot} approximately +95 kms^{-1}

No abstract

We have analyzed H and K_s-band images of the Arches cluster obtained using the NIRC2 instrument on Keck with the laser guide star adaptive optics (LGS AO) system. With the help of the LGS AO system, we were able to obtain the deepest ever photometry for this cluster and its neighborhood, and derive the background-subtracted present-day mass function (PDMF) down to 1.3 Msun for the 5 arcsec-9 arcsec annulus of the cluster. We find that the previously reported turnover at 6 Msun is simply due to a local bump in the mass function (MF), and that the MF continues to increase down to our 50 % completeness limit (1.3 Msun) with a power-law exponent of Gamma = -0.91 for the mass range of 1.3 < M/Msun < 50. Our numerical calculations for the evolution of the Arches cluster show that the Gamma values for our annulus increase by 0.1-0.2 during the lifetime of the cluster, and thus suggest that the Arches cluster initially had Gamma of -1.0 ~ -1.1, which is only slightly shallower than the Salpeter value.Comment: Accepted for publication in ApJ

A combination of high-resolution and wide-field imaging reveals two binary stars and one triple star system among the sample of the first 11 stars with planets detected by radial velocity variations. High resolution speckle or adaptive optics (AO) data probe subarcsecond scales down to the diffraction limit of the Keck 10-m or Lick 3-m, and direct images or AO images are sensitive to a wider field, extending to 10" or 38", depending upon the camera. One of the binary systems -- HD 114762 -- was not previously known to be a spatially resolved multiple system; additional data taken with the combination of Keck adaptive optics and NIRSPEC are used to characterize the new companion. The second binary system -- Tau Boo -- was a known multiple with two conflicting orbital solutions; the current data will help constrain the discrepant estimates of periastron time and separation. Another target -- 16 Cyg B -- was a known common proper motion binary, but the current data resolve a new third component, close to the wide companion 16 Cyg A. Both the HD 114762 and 16 Cyg B systems harbor planets in eccentric orbits, while the Tau Boo binary contains an extremely close planet in a tidally-circularized orbit. Although the sample is currently small, the proportion of binary systems is comparable to that measured in the field over a similar separation range. Incorporating the null result from another companion search project lowers the overall fraction of planets in binary systems, but the detections in our survey reveal that planets can form in binaries separated by less than 50 AU.Comment: 5 Tables, 16 Figures. ApJ, accepte

Numerical simulations of the dynamical friction suffered by a star cluster near the Galactic center have been performed with a parallelized tree code. Gerhard (2001) has suggested that dynamical friction, which causes a cluster to lose orbital energy and spiral in towards the galactic center, may explain the presence of a cluster of very young stars in the central parsec, where star formation might be prohibitively difficult owing to strong tidal forces. The clusters modeled in our simulations have an initial total mass of 10^5-10^6 Msun and initial galactocentric radii of 2.5-30 pc. We have identified a few simulations in which dynamical friction indeed brings a cluster to the central parsec, although this is only possible if the cluster is either very massive (~10^6 Msun), or is formed near the central parsec (<~ 5 pc). In both cases, the cluster should have an initially very dense core (> 10^6 Msun pc-3). The initial core collapse and segregation of massive stars into the cluster core, which typically happens on a much shorter time scale than that characterizing the dynamical inspiral of the cluster toward the Galactic center, can provide the requisite high density. Furthermore, because it is the cluster core which is most likely to survive the cluster disintegration during its journey inwards, this can help account for the observed distribution of presumably massive HeI stars in the central parsec.Comment: Accepted for publication in Ap

Large galaxy redshift surveys have long been used to constrain cosmological models and structure formation scenarios. In particular, the largest structures discovered observationally are thought to carry critical information on the amplitude of large-scale density fluctuations or homogeneity of the universe, and have often challenged the standard cosmological framework. The Sloan Great Wall (SGW) recently found in the Sloan Digital Sky Survey (SDSS) region casts doubt on the concordance cosmological model with a cosmological constant (i.e. the flat ΛCDM model). Here we show that the existence of the SGW is perfectly consistent with the ΛCDM model, a result that only our very large cosmological N -body simulation (the Horizon Run 2, HR2) could supply. In addition, we report on the discovery of a void complex in the SDSS much larger than the SGW, and show that such size of the largest void is also predicted in the ΛCDM paradigm. Our results demonstrate that an initially homogeneous isotropic universe with primordial Gaussian random phase density fluctuations growing in accordance with the General Relativity, can explain the richness and size of the observed large-scale structures in the SDSS. Using the HR2 simulation we predict that a future galaxy redshift survey about four times deeper or with 3 magnitude fainter limit than the SDSS should reveal a largest structure of bright galaxies about twice as big as the SGW.