No Arabic abstract
The bright X-ray binary X2127+119 in the core of the globular cluster M15 has long been thought to be in an unusual evolutionary state, in which the binary is embedded in a common envelope. Support for this idea comes from X2127+119s absorption lines, which are blue shifted at all orbital phases, indicating the existence of outflows from the system. A common-envelope scenario implies that the absorption lines should exhibit maximum blue shift near mid-eclipse (binary phase 0.0). We have re-analysed INT spectra of X2127+119 obtained in 1986, 1987 and 1988 using the latest orbital ephemeris (substantially different from that used in the original analysis), and find that the orbital phase at which the absorption lines show a maximum blue shift is not 0.0, but rather 0.25 -- 0.3. These results indicate that a common-envelope scenario for X2127+119 may not work. In addition, from spectrograms of the He II 4686 line, we report the first tentative detection of X2127+119s companion star.
The luminous low-mass X-ray binary X2127+119 in the core of the globular cluster M15 (NGC 7078), which has an orbital period of 17 hours, has long been assumed to contain a donor star evolving off the main sequence, with a mass of 0.8 solar masses (the main-sequence turn-off mass for M15). We present orbital-phase-resolved spectroscopy of X2127+119 in the H-alpha and He I 6678 spectral region, obtained with the Hubble Space Telescope. We show that these data are incompatible with the assumed masses of X2127+119s component stars. The continuum eclipse is too shallow, indicating that much of the accretion disc remains visible during eclipse, and therefore that the size of the donor star relative to the disc is much smaller in this high-inclination system than the assumed mass-ratio allows. Furthermore, the flux of X2127+119s He I 6678 emission, which has a velocity that implies an association with the stream-disc impact region, remains unchanged through eclipse, implying that material from the impact region is always visible. This should not be possible if the previously-assumed mass ratio is correct. In addition, we do not detect any spectral features from the donor star, which is unexpected for a 0.8 solar-mass sub-giant in a system with a 17-hour period.
We present time-resolved spectroscopy acquired during two epochs (spaced apart by ~15 days) of the eclipsing Low Mass X-ray Binary AC211/X2127+119 in the globular cluster M15. The spectra show variations in the HeII 4686 emission line not only modulated on the orbital period, but also on time-scales of a few days. During the first epoch of observation, the emission line shows a strong S-wave superimposed on the average double-peaked profile. The line exhibits no evidence of rotational disturbance at the orbital phases when the eclipse is observed in the optical continuum. During the second epoch, no double-peak or S-wave component is present. The HeI absorption lines detected by other authors are not present in our spectra. A Doppler image of the HeII 4686 for the first epoch supports the presence of the accretion disc. No hotspot is detected, although enhanced emission at V_X=30km/s, V_Y=160 km/s is observed. We discuss the implications of this emission in the context of an X-ray heated donor star, in which case a high mass ratio and neutron star primary are implied. Finally, we speculate on the possibility of a misaligned secondary star in AC211.
We present integral field spectroscopy of X2127+119, the luminous X-ray binary in the globular cluster M15, obtained with INTEGRAL/WYFFOS on the William Herschel Telescope. We find that tomograms of HeII 4686 line profiles appear to be incompatible with the previously-assumed view of X2127+119, in which the binary consists of a 1.4-solar-mass neutron star and a 0.8-solar-mass sub-giant companion near the main-sequence turn-off for M15. Our data imply a much smaller mass ratio M_2/M_x of 0.1. We find that models of X2127+119 with black-hole compact objects give a poor fit to our data, while a neutron-star compact object is consistent with the data, implying that X2127+119s companion may have a much lower mass (~0.1 solar masses) than previously assumed. As an 0.1-solar-mass main-sequence star would be unable to fill its Roche lobe in a binary with X2127+119s orbital period (17.1 hours), the companion is likely to be the remnant of a significantly more massive star which has had most of its envelope stripped away.
We present X-ray observations of the high-inclination low-mass X-ray binary system X2127+119 (AC211) in the globular cluster M15 (NGC 7078). The observations consist of data acquired in 1996 with the RXTE satellite and in 1995 with the ASCA satellite. Also, the MPC1 data from the 1988 GINGA observations were de-archived and re-analysed. The phase-folded 2-10 keV hardness ratios from all three missions differ significantly indicating that the system can exhibit different spectral behaviours. We find that the X-ray eclipse profiles can be described relatively well using a simple model where the secondary star passes in front of a large X-ray emitting region. For this we require a mass ratio (q=M1/M2) of about one. The radius of this X-ray emitting region is ~0.8 RL1 and its vertical extent 60 degrees above the orbital plane. We suggest that if this X-ray emitting region were an optically thick corona, it would explain various puzzling aspects of this system. We also show that the X-ray dip observed at phases around 0.65 does not conform with the idea that the dip is caused by vertically extended material associated with the stream/disc impact region, but that it could be due to structure in the inner parts of the disc.
Chromospheric model calculations of the Halpha line for selected red giant branch (RGB) and asymptotic giant branch (AGB) stars in the globular clusters M13, M15, and M92 are constructed to derive mass loss rates. The model spectra are compared to the observations obtained with the Hectochelle on the MMT telescope. These stars show strong Halpha emissions and blue-shifted Halpha cores signaling that mass outflow is present in all stars. Outflow velocities of 3-19 km/s, larger than indicated by Halpha profiles, are needed in the upper chromosphere to achieve good agreement between the model spectra and the observations. The resulting mass loss rates range from 0.6*10^{-9} to 5*10^{-9} Msun/yr, which are about an order of magnitude lower than predicted from Reimers law or inferred from the infrared excess of similar stars. The mass loss rate increases slightly with luminosity and with decreasing effective temperature. Stars in the more metal-rich M13 have higher mass loss rates by a factor of ~2 than in the metal-poor clusters M15 and M92. A fit to the mass loss rates is given by: M [Msun/yr] = 0.092 * L^{0.16} * Teff^{-2.02} * A^{0.37} where A=10^[Fe/H]. Multiple observations of stars revealed one object in M15, K757, in which the mass outflow increased by a factor of 6 between two observations separated by 18 months. Other stars showed changes in mass loss rate by a factor of 1.5 or less.