LMXBs form efficiently in GCs. By combining Chandra and HST ACS observations of 11 massive early-type galaxies in the Virgo Cluster, we use the most accurate identification of LMXBs and GCs to date to explore the optical properties of 270 GCs with LMXBs and 6488 GCs without detectable X-ray emission. More massive, redder, and more compact GCs are more likely to contain LMXBs. Unlike Galactic GCs, a large number of GCs with LMXBs have half-mass relaxation times >2.5 Gyr; GCs need not survive for more than five relaxation timescales to produce LMXBs. By fitting the dependence of the expected number of LMXBs per GC, k t , on the GC mass M, color ( g À z), and half-mass radius r h; cor , we find that k t / M 1:24AE0:08 ; 10 0:9 þ0:2 À0:1 ( gÀz) r À2:2AE0:2 h; cor . This rules out that the number of LMXBs per GC is linearly proportional to GC mass (99.89% confidence limit) and leads us to predict that most GCs with high X-ray luminosities contain a single LMXB. The detailed dependence of k t on GC properties appears mainly due to a dependence on the encounter rate À h and the metallicity Z, k t / À 0:82AE0:05 h Z 0:39AE0:07 . Our analysis provides strong evidence that dynamical formation and metallicity play the primary roles in determining the presence of an LMXB in extragalactic GCs. The shallower than linear encounter rate dependence requires an explanation by theories of dynamical binary formation. A metallicity-dependent variation in the number of neutron stars and black holes per unit GC mass, effects from irradiation-induced winds, or suppression of magnetic braking in metal-poor stars may all be consistent with our abundance dependence; all three scenarios require further development.

We present high-resolution spectroscopy of the oxygen K-shell interstellar absorption edge in seven X-ray binaries using the High Energy Transmission Grating Spectrometer ( HETGS) onboard the Chandra X-Ray Observatory. Using the brightest sources as templates, we found a best-fit model of two absorption edges and five Gaussian absorption lines. All of these features can be explained by the recent predictions of K-shell absorption from neutral and ionized atomic oxygen. We identify the K and K absorption lines from neutral oxygen, as well as the S ¼ 3=2 absorption edge. The expected S ¼ 1=2 edge is not detected in these data because of overlap with instrumental features. We also identify the K absorption lines from singly and doubly ionized oxygen. The O i K absorption line is used as a benchmark with which to adjust the absolute wavelength scale for theoretical predictions of the absorption cross sections. We find that shifts of 30-50 m8 are required, consistent with differences previously noticed from comparisons of the theory with laboratory measurements. Significant oxygen features from dust or molecular components, as suggested in previous studies, are not required by our HETGS spectra. With these spectra, we can begin to measure the large-scale properties of the interstellar medium (ISM ). We place a limit on the velocity dispersion of the neutral lines of P200 km s À1 , consistent with measurements at other wavelengths. We also make the first measurement of the oxygen ionization fractions in the ISM. We constrain the interstellar ratio of O ii /O i to %0.1 and the ratio of O iii /O i to P0.1. This work demonstrates the utility of X-ray spectroscopy for studies of the ISM. Future work will provide measurements of the relative abundances and ionization fractions for elements from carbon to iron.

There are three low-mass X-ray binaries (4U 0614+091, 2S 0918−549, and 4U 1543−624) for which broad line emission near 0.7 keV was previously reported. A recent high-resolution observation of 4U 0614+091 with the Chandra/LETGS found evidence for an unusually high Ne/O abundance ratio along the line of sight but failed to detect the previously reported 0.7 keV feature. We have made a search of archival ASCA spectra and identified a fourth source with the 0.7 keV feature, the 20-min ultracompact binary 4U 1850−087. In all four of these sources, the 0.7 keV residual is eliminated from the ASCA spectra by allowing excess photoelectric absorption due to a non-solar relative abundance of neon, just as in the LETGS spectrum of 4U 0614+091. The optical properties of these systems suggest that all four are ultracompact (P orb < 80 min) binaries. We propose that there is excess neon local to each of these sources, as also found in the ultracompact binary pulsar 4U 1626−67. We suggest that the mass donor in these systems is a low-mass, neon-rich degenerate dwarf and that the binaries are all ultracompact.

We present high-resolution spectroscopy of the neon K-shell and iron L-shell interstellar absorption edges in nine X-ray binaries using the High Energy Transmission Grating Spectrometer (HETGS) onboard the Chandra X-ray Observatory. We found that the iron absorption is well fit by an experimental determination of the cross-section for metallic iron, although with a slight wavelength shift of ≈20 mÅ. The neon edge region is best fit by a model that includes the neutral neon edge and three Gaussian absorption lines. We identify these lines as due to the 1s-2p transitions from Ne II, Ne III, and Ne IX. As we found in our oxygen edge study, the theoretical predictions for neutral and lowionization lines all require shifts of ≈20 mÅ to match our data. Combined with our earlier oxygen edge study, we find that a best fit O/Ne ratio of 5.4±1.6, consistent with standard interstellar abundances. Our best fit Fe/Ne ratio of 0.20±0.03 is significantly lower than the interstellar value. We attribute this difference to iron depletion into dust grains in the interstellar medium. We make the first measurement of the neon ionization fraction in the ISM. We find Ne II/Ne I≈0.3 and Ne III/Ne I≈0.07. These values are larger than is expected given the measured ionization of interstellar helium. For Ne IX, our results confirm the detection of the hot ionized interstellar medium of the Galaxy.

We present a statistical study of the low-mass X-ray binary (LMXB) populations of three nearby, old elliptical galaxies: NGC 3379, NGC 4278, and NGC 4697. With a cumulative ~1 Ms Chandra ACIS observing time, we detect 90-170 LMXBs within the D 25 ellipse of each galaxy. Cross-correlating Chandra X-ray sources and HST optical sources, we identify 75 globular cluster (GC) LMXBs and 112 field LMXBs with L X > 10 36 erg s -1 (detections of these populations are 90% complete down to luminosities in the range of 6 x 10 36 -1.5 x10 37 erg s -1 ). At the higher luminosities explored with previous studies, the statistics of this sample are consistent with the properties of GC-LMXBs reported in the literature. In the low luminosity range allowed by our deeper data (L X < 5 x 10 37 erg s -1 ), we find a significant relative lack of GC-LMXBs, when compared with field sources. Using the co-added sample from the three galaxies, we find that the incompletenesscorrected X-ray luminosity functions (XLFs) of GC and field LMXBs differ at ~4σ significance at L X < 5 x 10 37 erg s -1 . As previously reported, these XLFs are consistent at higher luminosities. The presently available theoretical models for LMXB formation and evolution in clusters are not sophisticated enough to provide a definite explanation for the shape of the observed GC-LMXB XLF. Our observations may indicate a potential predominance of GC-LMXBs with donors evolved beyond the main sequence, when compared to current models, but their efficient formation requires relatively high initial binary fractions in clusters. The field LMXB XLF can be fitted with either a single power-law model plus a localized excess at a luminosity of 5-6 x 10 37 erg s -1 , or a broken power-law with a similar low-luminosity break. This XLF may be explained with NSred-giant LMXBs, contributing to ~15% of total LMXBs population at ~5x10 37 erg s -1 . The difference in the GC and field XLFs is consistent with different origins and/or evolutionary paths between the two LMXB populations, although a fraction of the field sources are likely to have originated in GCs.