Subsolar Abundances Research Articles

Support for X‐Ray Deduced Abundances in Galaxies and Clusters: The Consistency of Iron Abundances from L and K Emission

Significantly subsolar metallicities have been reported in X-rays for a variety of sources, often in disagreement with theoretical expectations or with measurements at other wavelengths. As a test of the reliability of the X-ray abundances in systems dominated by thermal emission from hot gas in collisional ionization equilibrium, we compare Fe abundances determined independently from the Fe L and Fe K emission regions in simple one- and two-temperature thermal fits to ASCA X-ray spectra of the elliptical galaxy M87 and five galaxy clusters. We find that the Fe L and Fe K abundances are consistent within or near their 90% confidence errors, which would not necessarily be the case if there were severe problems with the atomic physics or with the instrument calibrations. Our data cover a range of plasma temperatures kT = 2-4 keV, where the Fe L emission arises primarily from Fe XXIII and XXIV and the Fe K emission arises primarily from Fe XXV. There is also a contribution from less ionized gas at lower temperatures in M87. For M87, the abundances from Fe K are systematically lower by 30%-50% than those from Fe L, and the differences are attributable to the complicated temperature structure in the central cooling flow. Allowing for the possibility of excess absorption of the emission from the cooling flow increases the systematic offset of the Fe L and Fe K abundances to 40%-80% in the simple model we consider. The cluster spectra, on the other hand, are for nearly isothermal regions outside any central cooling flow and the abundances have no systematic offset. The consistency between the Fe K and Fe L abundances, which is also seen in other clusters and in the corona of Ar Lac, indicates that whatever atomic physics uncertainties exist in the current codes are not likely to render Fe abundance measurements unreliable within the ~30%-50% agreement we find for the systems we examined. The results using various thermal emission codes differ by 15%-20%, showing the level of scatter due just to different assumptions about the atomic physics. The atomic rates for H- and He-like ions, which are relevant for the determination of abundances in elements other than Fe, are more accurately known than those for less ionized species that dominate the complicated Fe L emission. In light of the support our results give to the abundances derived from ASCA data for clusters and bright elliptical galaxies, we consider briefly the plausibility of subsolar abundances in elliptical galaxies.

The Nature of X‐Ray Emission and Mass Distributions in Two Early‐Type Galaxies

We present spectral analysis of ASCA observations of the early-type galaxies NGC 720 (E4) and NGC 1332 (E7/S0) with emphasis on constraining the relative contribution to the X-ray emission from hot gas and the integrated emission from X-ray binaries. Single-temperature spectral models yield poor fits to the spectrum (χred2~3) over the ~0.5-5 keV energy range. Two-temperature models significantly improve the spectral fits (χred2~1.5) and have soft-component temperatures and subsolar abundances consistent with previous ROSAT single-temperature models (Tsoft ~ 0.6 keV, abundances ~0.1) and hard-component temperatures (Thard 3 keV) consistent with those expected from a discrete component. The soft component dominates the emission in both galaxies, especially in the 0.4-2.4 keV band used in previous ROSAT studies: flux ratios are Fhard/Fsoft = 0.19(0.16-0.45) for NGC 720 (2 σ) and Fhard/Fsoft = 0.31(0.24-0.55) for NGC 1332 (90%). Combining these spectral results with ROSAT data we updated constraints on the mass distributions for NGC 720 and NGC 1332. For NGC 720, which yields the more precise constraints, the ellipticity of the intrinsic shape of the mass is slightly reduced (Δmass ≈ 0.05) when the discrete component is added, mass ~ 0.4-0.6 (90%). The estimates for the total mass increase with increasing discrete flux, and we find that models with Fhard/Fsoft = 0.45, the 2 σ upper limit, have masses that exceed by ~30%-50% those where Fhard/Fsoft = 0.

The Composition of the Diffuse Interstellar Medium

Recent Goddard High Resolution Spectrograph measurements of Si, S, Cr, Mn, Fe, and Zn in interstellar clouds along lines of sight in the Galactic disk and into the lower halo are discussed. The gas-phase abundance of S relative to H in the interstellar clouds appears to be indistinguishable from the solar value. For the other elements, we find well-defined upper limits in the gas-phase abundances at significantly subsolar values. For Fe, Mn, and Cr (and probably Ti), there are no convincing cases in which the relative gas-phase abundances exceed roughly -0.5 dex, i.e., these elements are not seen in interstellar gas with an abundance greater than about one-third solar. For Si, the limit is roughly -0.15 dex, and for Zn a constant abundance of -0.13 dex is found from seven clouds along one halo sight line. These subsolar maximum abundances have two possible interpretations: (1) they indicate the presence of an essentially indestructible component of interstellar dust, which contains about two-thirds of the Ti, Mn, Cr, and Fe and about one-third of the Si (based on solar composition), or (2) they indicate that the true total abundances of these elements are substantially less than in the Sun.

ROSAT Observations of Five Chromospherically Active Stars

We present X-ray spectra of three chromospherically active binaries, HR 7428, FF Aqr, and IX Per, and two active giant stars, 29 Dra and HR 9024. The spectra were obtained using the Position Sensitive Proportional Counter (PSPC) on the ROSAT satellite. For coronal plasma models with 1-temperature or 2-temperature components, we find that a range of heavy element abundances, from solar to extreme sub-solar abundances similar to those that have been inferred from analyses of ASCA spectra, provide acceptable fits to the observed PSPC spectra. The highest signal-to-noise spectrum of 29 Dra requires either a 1- or 2-temperature model with greatly sub-solar (photospheric) abundances or else a 3-component model with solar abundances. The degeneracy of these results, i.e., the fact that quite different models can produce acceptable fits to the PSPC spectra, means that it is difficult to infer the temperature structure of a corona from a PSPC spectra analysis without a priori knowledge of the prevailing coronal abundances, and vice versa. We also show that uncertainties in the intervening interstellar column densities can affect the inferred temperatures and/or abundances. Notwithstanding these difficulties, our analysis confirms that models having similar subsolar abundances to those inferred from ASCA and EUVE spectra of active coronal stars do indeed produce good fits to PSPC spectra. This general consistency between different instruments implies that explanations for the subsolar abundances that invoke systematic biases due to instrumental effects can now be ruled out.

Infrared spectroscopy of symbiotic stars: carbon abundances and 12C/13C isotopic ratios

Intermediate-resolution spectra of a sample of symbiotic stars in the K and L bands have been obtained. In many objects, Pfund and Brackett emission lines are detected, as well as the first-overtone CO absorption bands. Synthetic spectra are used to fit the CO absorption features and to obtain carbon abundances and 12 C/ 13 C ratios. Subsolar carbon abundances in the symbiotic red giants are found, with little scatter from star to star. The 12 C/ 13 C ratios are also lower than the solar value. In terms of the C abundance and isotopic ratio, the surveyed symbiotic stars are indistinguishable from normal red giants, i.e. no modifications due to binarity are detected

Spatial distribution of sodium vapor in the atmosphere of Mercury

The density and shape of the sodium exosphere of Mercury can give clues to the interactions of cosmic dust, the solar wind, and the radiation field with the planetary surface and magnetosphere. In past observations, A.E. Potter and T.H. Morgan, (1987, Icarus 71, 472–477) establish that radiation pressure affected the sodium density. In this work, we find, as before, that the disk-averaged sodium column density is correlated with the magnitude of radiation pressure in such a way that the maximum column density occurs at minimum radiation pressure and vice versa. However, superimposed on this orderly variation is a disorderly variation on a time scale less than a day, and we find anomalous distributions of sodium vapor in polar and subsolar regions. These effects may be a result of solar wind-magnetosphere interactions. Observations of north-south and east-west distributions of sodium emission from Mercury were made at three levels of radiation pressure: one corresponding to near-minimum solar radiation pressure, a second to near maximum, and a third at an intermediate radiation pressure. Corresponding north-south and east-west distributions of sodium column abundance were calculated using radiative transfer theory, and were fit to the observations by varying the assumed density and distribution of sodium vapor. The main features of the observed intensity distributions could be adequately represented by simple and reasonable models for the temperature and surface distribution of sodium vapor. For this limited data set, the calculated subsolar abundance of sodium at minimum radiation pressure was about three times larger than that at maximum radiation pressure, which is approximately consistent with previous observations. However, it was found that significant details of the observed distributions could not be fit by a simple model. The observed sodium emission distribution was concentrated in the sunward direction with polar enhancements. Several north-south distributions were anomalous in that there was clearly more sodium at one pole than the other. In addition, the sodium distribution varied substantially on a daily basis. The subsolar column abundance varied from 3.8 × 10 11 atoms/cm 2 on April 3, 1988, to 2.8 × 10 11 atoms/cm 2 on April 6, 1988. It is suggested that these anomalies may be related to the interaction of the photoionized sodium atoms with the magnestosphere. More detailed observations in the future may better define the respective roles of radiation pressure, solar wind-magnetosphere interactions, source processes, and evaporation of surface material.