No Arabic abstract
PG1159 stars are hot, hydrogen-deficient (pre-) white dwarfs with atmospheres mainly composed of helium, carbon, and oxygen. The unusual surface chemistry is the result of a late helium-shell flash. Observed element abundances enable us to test stellar evolution models quantitatively with respect to their nucleosynthesis products formed near the helium-burning shell of the progenitor asymptotic giant branch stars. Because of the high effective temperatures (Teff), abundance determinations require ultraviolet spectroscopy and non-local thermodynamic equilibrium model atmosphere analyses. Up to now, we have presented results for the prototype of this spectral class and two cooler members (Teff in the range 85,000-140,000 K). Here we report on the results for two even hotter stars (PG1520+525 and PG1144+005, both with Teff = 150,000 K) which are the only two objects in this temperature-gravity region for which useful far-ultraviolet spectra are available, and revisit the prototype star. Previous results on the abundances of some species are confirmed, while results on others (Si, P, S) are revised. In particular, a solar abundance of sulphur is measured in contrast to earlier claims of a strong S deficiency that contradicted stellar evolution models. For the first time, we assess the abundances of Na, Al, and Cl with newly constructed non-LTE model atoms. Besides the main constituents (He, C, O), we determine the abundances (or upper limits) of N, F, Ne, Na, Al, Si, P, S, Cl, Ar, and Fe. Generally, good agreement with stellar models is found.
Metal-poor massive stars dominate the light we observe from star-forming dwarf galaxies and may have produced the bulk of energetic photons that reionized the universe at high redshift. Yet, the rarity of observations of individual O stars below the $20%$ solar metallicity ($Z_odot$) of the Small Magellanic Cloud (SMC) hampers our ability to model the ionizing fluxes of metal-poor stellar populations. We present new Hubble Space Telescope far-ultraviolet (FUV) spectra of three O-dwarf stars in the galaxies Leo P ($3%,Z_odot$), Sextans A ($6%,Z_odot$), and WLM ($14%,Z_odot$). We quantify equivalent widths of photospheric metal lines and strengths of wind-sensitive features, confirming that both correlate with metallicity. We infer the stars fundamental properties by modeling their FUV through near-infrared spectral energy distributions and identify stars in the SMC with similar properties to each of our targets. Comparing to the FUV spectra of the SMC analogs suggests that (1) the star in WLM has an SMC-like metallicity, and (2) the most metal-poor star in Leo P is driving a much weaker stellar wind than its SMC counterparts. We measure projected rotation speeds and find that the two most metal-poor stars have high $v ,mathrm{sin}(i),geq,290,mathrm{km},mathrm{s}^{-1}$, and estimate just a $3-6%$ probability of finding two fast rotators if the metal-poor stars are drawn from the same $v ,mathrm{sin}(i)$ distribution observed for O dwarfs in the SMC. These observations suggest that models should include the impact of rotation and weak winds on ionizing flux to accurately interpret observations of metal-poor galaxies in both the near and distant universe.
Spectra of B stars in the wavelength range of 911-1100 A have been obtained with the EURD spectrograph onboard the Spanish satellite MINISAT-01 with ~5 A spectral resolution. IUE spectra of the same stars have been used to normalize Kurucz models to the distance, reddening and spectral type of the corresponding star. The comparison of 8 main-sequence stars studied in detail (alpha Vir, epsilon Tau, lambda Tau, tau Tau, alpha Leo, zeta Lib, theta Oph, and sigma Sgr) shows agreement with Kurucz models, but observed fluxes are 10-40% higher than the models in most cases. The difference in flux between observations and models is higher in the wavelength range between Lyman alpha and Lyman beta. We suggest that Kurucz models underestimate the FUV flux of main-sequence B stars between these two Lyman lines. Computation of flux distributions of line-blanketed model atmospheres including non-LTE effects suggests that this flux underestimate could be due to departures from LTE, although other causes cannot be ruled out. We found the common assumption of solar metallicity for young disk stars should be made with care, since small deviations can have a significant impact on FUV model fluxes. Two peculiar stars (rho Leo and epsilon Aqr), and two emission line stars (epsilon Cap and pi Aqr) were also studied. Of these, only epsilon Aqr has a flux in agreement with the models. The rest have strong variability in the IUE range and/or uncertain reddening, which makes the comparison with models difficult.
We computed a comprehensive set of theoretical ultraviolet spectra of hot, massive stars with the radiation-hydrodynamics code WM-Basic. This model atmosphere and spectral synthesis code is optimized for computing the strong P Cygni-type lines originating in the winds of hot stars, which are the strongest features in the ultraviolet spectral region. The computed set is suitable as a spectral library for inclusion in evolutionary synthesis models of star clusters and star-forming galaxies. The chosen stellar parameters cover the upper left Hertzsprung-Russell diagram at L >~ 10^2.75 Lsun and T_eff >~ 20,000 K. The adopted elemental abundances are 0.05 Zsun, 0.2 Zsun, 0.4 Zsun, Zsun, and 2 Zsun. The spectra cover the wavelength range from 900 to 3000 {AA} and have a resolution of 0.4 {AA}. We compared the theoretical spectra to data of individual hot stars in the Galaxy and the Magellanic Clouds obtained with the International Ultraviolet Explorer (IUE) and Far Ultraviolet Spectroscopic Explorer (FUSE) satellites and found very good agreement. We built a library with the set of spectra and implemented it into the evolutionary synthesis code Starburst99 where it complements and extends the existing empirical library towards lower chemical abundances. Comparison of population synthesis models at solar and near-solar composition demonstrates consistency between synthetic spectra generated with either library. We discuss the potential of the new library for the interpretation of the rest-frame ultraviolet spectra of star-forming galaxies. Properties that can be addressed with the models include ages, initial mass function, and heavy-element abundance. The library can be obtained both individually or as part of the Starburst99 package.
We present an online catalog containing spectra and supporting information for cataclysmic variables that have been observed with the Far Ultraviolet Spectroscopic Explorer (FUSE). For each object in the catalog we list some of the basic system parameters such as (RA,Dec), period, inclination, white dwarf mass, as well as information on the available FUSE spectra: data ID, observation date and time, and exposure time. In addition, we provide parameters needed for the analysis of the FUSE spectra such as the reddening E(B-V), distance, and state (high, low, intermediate) of the system at the time it was observed. For some of these spectra we have carried out model fits to the continuum with synthetic stellar and/or disk spectra using the codes TLUSTY and SYNSPEC. We provide the parameters obtained from these model fits; this includes the white dwarf temperature, gravity, projected rotational velocity and elemental abundances of C, Si, S and N, together with the disk mass accretion rate, the resulting inclination and model-derived distance (when unknown). For each object one or more figures are provided (as gif files) with line identification and model fit(s) when available. The FUSE spectra as well as the synthetic spectra are directly available for download as ascii tables. References are provided for each object as well as for the model fits. In this article we present 36 objects, and additional ones will be added to the online catalog in the future. In addition to cataclysmic variables, we also include a few related objects, such as a wind accreting white dwarf, a pre-cataclysmic variable and some symbiotics.
We report the first detection of the Pb II line at 2203.534 Angstroms in three metal-poor stars, using ultraviolet spectra obtained with the Space Telescope Imaging Spectrograph on board the Hubble Space Telescope. We perform a standard abundance analysis assuming local thermodynamic equilibrium (LTE) to derive lead (Pb, Z=82) abundances. The Pb II line yields a higher abundance than Pb I lines by +0.36 +/- 0.34 dex and +0.49 +/- 0.28 dex in the stars HD 94028 and HD 196944, where Pb I lines had been detected previously. The Pb II line is likely formed in LTE, and these offsets affirm previous calculations showing that Pb I lines commonly used as abundance indicators underestimate the Pb abundance in LTE. Pb is enhanced in the s-process-enriched stars HD 94028 ([Pb/Fe] = +0.95 +/- 0.14) and HD 196944 ([Pb/Fe] = +2.28 +/- 0.23), and we show that Pb-208 is the dominant Pb isotope in these two stars. The log epsilon(Pb/Eu) ratio in the r-process-enhanced star HD 222925 is 0.76 +/- 0.14, which matches the Solar System r-process ratio and indicates that the Solar System r-process residuals for Pb are, in aggregate, correct. The Th/Pb chronometer in HD 222925 yields an age of 8.2 +/- 5.8 Gyr, and we highlight the potential of the Th/Pb chronometer as a relatively model-insensitive age indicator in r-process-enhanced stars.