Discovery of the near-infrared counterpart to the luminous neutron-star low-mass X-ray binary GX 3+1

Using the High Resolution Camera onboard the Chandra X-ray Observatory, we have measured an accurate position for the bright persistent neutron-star X-ray binary and atoll source GX 3+1. At a location that is consistent with this new position we have discovered the near-infrared (NIR) counterpart to GX 3+1 in images taken with the PANIC and FourStar cameras on the Magellan Baade Telescope. The identification of this K_s=15.8+-0.1 mag star as the counterpart is based on the presence of a Br-gamma emission line in a NIR spectrum taken with the FIRE spectrograph on the Baade Telescope. The absolute magnitude derived from the best available distance estimate to GX 3+1 indicates that the mass donor in the system is not a late-type giant. We find that the NIR light in GX 3+1 is likely dominated by the contribution from a heated outer accretion disk. This is similar to what has been found for the NIR flux from the brighter class of Z sources, but unlike the behavior of atolls fainter (Lx ~ 1e36 to 1e37 erg/s) than GX 3+1, where optically-thin synchrotron emission from a jet probably dominates the NIR flux.

A Chandra X-ray study of the interacting binaries in the old open cluster NGC6791

We present the first X-ray study of NGC6791, one of the oldest open clusters known (8 Gyr). Our Chandra observation is aimed at uncovering the population of close interacting binaries down to Lx ~ 1e30 erg/s (0.3-7 keV). We detect 86 sources within 8 arcmin of the cluster center, including 59 inside the half-mass radius. We identify twenty sources with proper-motion cluster members, which are a mix of cataclysmic variables (CVs), active binaries (ABs), and binaries containing sub-subgiants. With follow-up optical spectroscopy we confirm the nature of one CV. We discover one new, X-ray variable candidate CV with Balmer and HeII emission lines in its optical spectrum; this is the first X-ray--selected CV confirmed in an open cluster. The number of CVs per unit mass is consistent with the field, suggesting that the 3-4 CVs observed in NGC6791 are primordial. We compare the X-ray properties of NGC6791 with those of a few old open (NGC6819, M67) and globular clusters (47Tuc, NGC6397). It is puzzling that the number of ABs brighter than 1e30 erg/s normalized by cluster mass is lower in NGC6791 than in M67 by a factor ~3 to 7. CVs, ABs, and sub-subgiants brighter than 1e30 erg/s are under-represented per unit mass in the globular clusters compared to the oldest open clusters, and this accounts for the lower total X-ray luminosity per unit mass of the former. This indicates that the net effect of dynamical encounters may be the destruction of even some of the hardest (i.e. X-ray--emitting) binaries.

X-ray sources in Galactic old open star clusters

I review the current status of studies of the X-ray sources in Galactic old open clusters. Cataclysmic variables (CVs), magnetically-active binaries (ABs), and sub-subgiants (SSGs) dominate the X-ray emission of old open clusters. Surprisingly, the number of ABs detected inside the half-mass radius with Lx >~ 1e30 erg/s (0.3-7 keV) does not appear to scale with cluster mass. Comparison of the numbers of CVs, ABs, and SSGs per unit mass in old open and globular clusters shows that each of these classes is under-abundant in globulars. This suggests that dense environments suppress the frequency of even some of the hardest binaries.

The ChaMPlane bright X-ray sources - Galactic longitudes l = 2-358 deg

The Chandra Multiwavelength Plane (ChaMPlane) Survey aims to constrain the Galactic population of mainly accretion-powered, but also coronal, low-luminosity X-ray sources (Lx <~ 1e33 erg/s). To investigate the X-ray source content in the plane at fluxes Fx >~ 3e-14 erg/s/cm^2, we study 21 of the brightest ChaMPlane sources, viz. those with >250 net counts (0.3-8 keV). By excluding the heavily obscured central part of the plane, our optical/near-infrared follow-up puts useful constraints on their nature. We have discovered two likely accreting white-dwarf binaries. CXOPS J154305.5-522709 (CBS 7) is a cataclysmic variable showing periodic X-ray flux modulations on 1.2 hr and 2.4 hr; given its hard spectrum the system is likely magnetic. We identify CXOPS J175900.8-334548 (CBS 17) with a late-type giant; if the X-rays are indeed accretion-powered, it belongs to the small but growing class of symbiotic binaries lacking strong optical nebular emission lines. CXOPS J171340.5-395213 (CBS 14) is an X-ray transient that brightened >~100 times. We tentatively classify it as a very late-type (>M7) dwarf, of which few have been detected in X-rays. The remaining sources are (candidate) active galaxies, normal stars and active binaries, and a plausible young T Tauri star. The derived cumulative number density versus flux (log N - log S) relation for the Galactic sources appears flatter than expected for an isotropic distribution, indicating that we are seeing a non-local sample of mostly coronal sources. Our findings define source templates that we can use, in part, to classify the >1e4 fainter sources in ChaMPlane.

ChaMPlane Deep Galactic Bulge Survey. I. Faint accretion-driven binaries in the Limiting Window

[Abridged] We have carried out a deep X-ray and optical survey with Chandra and HST of low-extinction regions in the Galactic bulge. Here we present the results of a search for low-luminosity (L_X <~ 1e34 ergs/s) accreting binaries in the region closest to the Galactic Center, at an angular offset of 1.4deg, that we have named the Limiting Window. Based on their blue optical colors, excess Halpha fluxes, and high X-ray--to--optical flux ratios, we identify three likely cataclysmic variables (CVs). Distance estimates put these systems farther than >~2 kpc. Based on their Halpha-excess fluxes and/or high X-ray--to--optical flux ratios, we find 22 candidate accreting binaries; however, the properties of some can also be explained if they are dMe stars or active galaxies. We investigate the CV number density towards the bulge and find that the number of observed candidate CVs is consistent with or lower than the number expected for a constant CV-to-star ratio that is fixed to the local value. The X-ray properties of two likely CVs are similar to those of the faint, hard X-ray sources in the Galactic-Center region that have been explained by (mainly) magnetic CVs. If our candidates belong to the same population, they would be the first members to be optically identified; optical or infrared identification of their Galactic-Center analogs would be impossible due to the higher obscuration. We speculate that all Galactic hard X-ray sources in our field can be explained by magnetic CVs.

The Brightest Serendipitous X-ray Sources in ChaMPlane

The Chandra Multiwavelength Plane (ChaMPlane) Survey is a comprehensive effort to constrain the population of accretion-powered and coronal low-luminosity X-ray sources (L_X < 10^33 erg s^-1) in the Galaxy. ChaMPlane incorporates X-ray, optical, and infrared observations of fields in the Galactic Plane imaged with Chandra in the past six years. We present the results of a population study of the brightest X-ray sources in ChaMPlane. We use X-ray spectral fitting, X-ray lightcurve analysis, and optical photometry of candidate counterparts to determine the properties of 21 sources. Our sample includes a previously unreported quiescent low-mass X-ray binary or cataclysmic variable (R = 20.9) and ten stellar sources (12.5 < R < 15), including one flare star (R = 17.3). We find that quantile analysis, a new technique developed for constraining the X-ray spectral properties of low-count sources, is largely consistent with spectral fitting.