The VLT-FLAMES survey of massive stars: atmospheric parameters and rotational velocity distributions for B-type stars in the Magellanic Clouds

Aims.We aim to provide the atmospheric parameters and rotational velocities for a large sample of O- and early B-type stars, analysed in a homogeneous and consistent manner, for use in constraining theoretical models. Methods: Atmospheric parameters, stellar masses, and rotational velocities have been estimated for approximately 250 early B-type stars in the Large (LMC) and Small (SMC) Magellanic Clouds from high-resolution VLT-FLAMES data using the non-LTE TLUSTY model atmosphere code. This data set has been supplemented with our previous analyses of some 50 O-type stars (Mokiem et al. 2006, 2007) and 100 narrow-lined early B-type stars (Hunter et al. 2006; Trundle et al. 2007) from the same survey, providing a sample of ~400 early-type objects. Results: Comparison of the rotational velocities with evolutionary tracks suggests that the end of core hydrogen burning occurs later than currently predicted and we argue for an extension of the evolutionary tracks. We also show that the large number of the luminous blue supergiants observed in the fields are unlikely to have directly evolved from main-sequence massive O-type stars as neither their low rotational velocities nor their position on the H-R diagram are predicted. We suggest that blue loops or mass-transfer binary systems may populate the blue supergiant regime. By comparing the rotational velocity distributions of the Magellanic Cloud stars to a similar Galactic sample, we find that (at 3s confidence level) massive stars (above 8 M?) in the SMC rotate faster than those in the solar neighbourhood. However there appears to be no significant difference between the rotational velocity distributions in the Galaxy and the LMC. We find that the v sin i distributions in the SMC and LMC can modelled with an intrinsic rotational velocity distribution that is a Gaussian peaking at 175 km s-1 (SMC) and 100 km s-1 (LMC) with a 1/e half width of 150 km s-1. We find that in NGC 346 in the SMC, the 10-25 M? main-sequence stars appear to rotate faster than their higher mass counterparts. It is not expected that O-type stars spin down significantly through angular momentum loss via stellar winds at SMC metallicity, hence this could be a reflection of mass dependent birth spin rates. Recently Yoon et al. (2006) have determined rates of GRBs by modelling rapidly rotating massive star progenitors. Our measured rotational velocity distribution for the 10-25 M? stars is peaked at slightly higher velocities than they assume, supporting the idea that GRBs could come from rapid rotators with initial masses as low as 14 M? at low metallicities.

SiO in G34.26: Outflows and shocks in a high mass star forming region

We have looked for SiO emission as evidence of shocks in the high mass star formation region G34.26+0.15. JCMT, VLA and FCRAO observations show that SiO emission is widespread across the region. The SiO emission highlights a massive, collimated out ow and other regions where stellar winds are interacting with molecular clumps. As in other star forming regions, there is also SiO at ambient velocities which is related to the out ow activity. No strong SiO abundance enhancement was measured in either the out ow or the low velocity gas, though abundances up to 10(-8) are possible if the SiO is locally enhanced in clumps and optically thick. SiO emission is not detected from the hot core itself, indicating either that SiO is not strongly enhanced in the hot core or that column densities in the region where grain mantle evaporation has taken place are low. In line of sight spiral arm clouds, we measure a SiO abundance of 0.4-2 x 10(-10), consistent with previous estimates for quiescent clouds.

A low-energy core-collapse supernova without a hydrogen envelope

The final fate of massive stars depends on many factors. Theory suggests that some with initial masses greater than 25 to 30 solar masses end up as Wolf-Rayet stars, which are deficient in hydrogen in their outer layers because of mass loss through strong stellar winds. The most massive of these stars have cores which may form a black hole and theory predicts that the resulting explosion of some of them produces ejecta of low kinetic energy, a faint optical luminosity and a small mass fraction of radioactive nickel. An alternative origin for low-energy supernovae is the collapse of the oxygen-neon core of a star of 7-9 solar masses. No weak, hydrogen-deficient, core-collapse supernovae have hitherto been seen. Here we report that SN 2008ha is a faint hydrogen-poor supernova. We propose that other similar events have been observed but have been misclassified as peculiar thermonuclear supernovae (sometimes labelled SN 2002cx-like events). This discovery could link these faint supernovae to some long-duration gamma-ray bursts, because extremely faint, hydrogen-stripped core-collapse supernovae have been proposed to produce such long gamma-ray bursts, the afterglows of which do not show evidence of associated supernovae.

CHEMISTRY IN A PROTOPLANETARY NEBULA

In this paper we investigate gas-phase chemistry in the remnant 'superwind' of a carbon-rich red giant star, during its transition to a planetary nebula. The interacting stellar winds model is used. It is found that during the first few hundred years of transition, significant abundances of a few small molecules and ions (e.g. CH+, CH2+, CH3+, CH, CH2, NH) may occur in the thin, dense, shocked shell of gas predicted by thiS model, but that most molecules observed in protoplanetary nebulae will be rapidly destroyed, through photodissociation by strong UV from the central star. If dense clumps are present during transition, they may allow the gas-phase formation and/or survival of small amounts of some molecules, such as HCN, CN, C2H2, and HC3N, until about 2000 yr after termination of the superwind; and young, fully developed planetary nebulae may show observable amounts of polyatomic molecules by this means. Such clumping may explain the existence of, e.g., HCN in NGC 7027.

The empirical metallicity dependence of the mass-loss rate of O- and early B-type stars

We present a comprehensive study of the observational dependence of the mass-loss rate in stationary stellar winds of hot massive stars on the metal content of their atmospheres. The metal content of stars in the Magellanic Clouds is discussed, and a critical assessment is given of state-of-the-art mass-loss determinations of OB stars in these two satellite systems and the Milky-Way. Assuming a power-law dependence of mass loss on metal content,. M. Z(m), and adopting a theoretical relation between the terminal flow velocity and metal content, v(infinity). Z(0.13) (Leitherer et al. 1992, ApJ, 401, 596), we find m = 0.83 +/- 0.16 for non-clumped outflows from an analysis of the wind momentum luminosity relation (WLR) for stars more luminous than 105.2 L circle dot. Within the errors, this result is in agreement with the prediction m = 0.69 +/- 0.10 by Vink et al. (2001, A& A, 369, 574). Absolute empirical values for the mass loss, based on Ha and ultraviolet (UV) wind lines, are found to be a factor of two higher than predictions in this high luminosity regime. If this difference is attributed to inhomogeneities in the wind, and this clumping does not impact the predictions, this would imply that luminous O and early-B stars have clumping factors in their Ha and UV line forming regions of about a factor of four. For lower luminosity stars, the winds are so weak that their strengths can generally no longer be derived from optical spectral lines (essentially Ha) and one must currently rely on the analysis of UV lines. We confirm that in this low-luminosity domain the observed Galactic WLR is found to be much steeper than expected from theory (although the specific sample is rather small), leading to a discrepancy between UV mass-loss rates and the predictions by a factor 100 at luminosities of L similar to 10(4.75) L circle dot, the origin of which is unknown. We emphasize that even if the current mass-loss rates of hot luminous stars would turn out to be overestimated as a result of wind clumping, but the degree of clumping would be rather independent of metallicity, the scalings derived in this study are expected to remain correct.

Blue luminous stars in nearby galaxies: Quantitative spectral analysis of M33 B-type supergiant stars

We present the detailed spectral analysis of a sample of M33 B-type supergiant stars, aimed at the determination of their fundamental parameters and chemical composition. The analysis is based on a grid of non-LTE metal line-blanketed model atmospheres including the effects of stellar winds and spherical extension computed with the code FASTWIND. Surface abundance ratios of C, N, and O are used to discuss the chemical evolutionary status of each individual star. The comparison of observed stellar properties with theoretical predictions of massive star evolutionary models shows good agreement within the uncertainties of the analysis. The spatial distribution of the sample allows us to investigate the existence of radial abundance gradients in the disk of M33. The comparison of stellar and H II region O abundances ( based on direct determinations of the electron temperature of the nebulae) shows good agreement. Using a simple linear radial representation, the stellar oxygen abundances result in a gradient of -0.0145 +/- 0.005 dex arcmin(-1) (or -0.06 +/- 0.02 dex kpc(-1)) up to a distance equal to similar to 1.1 times the isophotal radius of the galaxy. A more complex representation cannot be completely discarded by our stellar sample. The stellar Mg and Si abundances follow the trend displayed by O abundances, although with shallower gradients. These differences in gradient slope cannot be explained at this point. The derived abundances of the three alpha-elements yield solar metallicity in the central regions of the disk of M33. A comparison with recent planetary nebula data from Magrini and coworkers indicates that the disk of M33 has not suffered from a significant O enrichment in the last 3 Gyr.

Self-regulated Shocks in Massive Star Binary Systems

In an early-type, massive star binary system, X-ray bright shocks result from the powerful collision of stellar winds driven by radiation pressure on spectral line transitions. We examine the influence of the X-rays from the wind-wind collision shocks on the radiative driving of the stellar winds using steady-state models that include a parameterized line force with X-ray ionization dependence. Our primary result is that X-ray radiation from the shocks inhibits wind acceleration and can lead to a lower pre-shock velocity, and a correspondingly lower shocked plasma temperature, yet the intrinsic X-ray luminosity of the shocks, L X, remains largely unaltered, with the exception of a modest increase at small binary separations. Due to the feedback loop between the ionizing X-rays from the shocks and the wind driving, we term this scenario as self-regulated shocks. This effect is found to greatly increase the range of binary separations at which a wind-photosphere collision is likely to occur in systems where the momenta of the two winds are significantly different. Furthermore, the excessive levels of X-ray ionization close to the shocks completely suppress the line force, and we suggest that this may render radiative braking less effective. Comparisons of model results against observations reveal reasonable agreement in terms of log (L X/L bol). The inclusion of self-regulated shocks improves the match for kT values in roughly equal wind momenta systems, but there is a systematic offset for systems with unequal wind momenta (if considered to be a wind-photosphere collision).

Étude du système binaire CV Ser à l'aide du satellite MOST

Ce mémoire s’intéresse au système binaire massif CV Serpentis, composé d’une Wolf- Rayet riche en carbone et d’une étoile de la séquence principale, de type spectral O (WC8d + O8-9IV). D’abord, certains phénomènes affectant les étoiles massives sont mentionnés, de leur passage sur la séquence principale à leur mort (supernova). Au cours du premier cha- pitre, un rappel est fait concernant certaines bases de l’astrophysique stellaire observa- tionnelle (diagramme Hertzsprung-Russell, phases évolutives, etc...). Au chapitre suivant, un des aspects les plus importants de la vie des étoiles massives est abordé : la perte de masse sous forme de vents stellaires. Un historique de la découverte des vents ouvre le chapitre, suivi des fondements théoriques permettant d’expliquer ce phénomène. Ensuite, différents aspects propres aux vents stellaires sont présentés. Au troisième chapitre, un historique détaillé de CV Ser est présenté en guise d’introduc- tion à cet objet singulier. Ses principales caractéristiques connues y sont mentionnées. Finalement, le cœur de ce mémoire se retrouve au chapitre 4. Des courbes de lumière ultra précises du satellite MOST (2009 et 2010) montrent une variation apparente du taux de perte de masse de la WR de l’ordre de 62% sur une période orbitale de 29.701 jours. L’analyse des résidus permet de trouver une signature suggérant la présence de régions d’interaction en corotation (en anglais corotating interaction regions, ou CIR) dans le vent WR. Une nouvelle solution orbitale est présentée ainsi que les paramètres de la région de collision des vents et les types spectraux sont confirmés.

Étude de la variabilité spectroscopique d’un échantillon d’étoiles Wolf-Rayet de type WC9

Nous savons que la grande majorité des étoiles WC9 produit de la poussière à base de carbone. Cette dernière doit se former dans des zones de très haute densité afin de survivre à l’environnement hostile qu’est celui du vent d’une étoile WR. Les étoiles WC appartenant à un système binaire WR + O produisent de la poussière quand les vents des deux étoiles entrent en collision et forment une zone de choc pouvant augmenter la densité du gaz d’un facteur 1000. Par contre, plusieurs étoiles WC9 n’ont, à ce jour, montré aucun signe de la présence d’un compagnon. Le but du projet est de tenter d’identifier un mécanisme alternatif responsable de la formation de poussière dans les étoiles WC9 n’appartenant pas à un système binaire. Nous présentons les résultats d’une campagne d’observation visant à caractériser la variabilité spectroscopique d’un échantillon de huit étoiles WC9 et une étoile WC8d. Nos résultats indiquent que la majorité des étoiles montrent des variations à grande échelle dans la raie d’émission C III 5696, soit à un niveau d’au moins 5% du flux de la raie et que les structures dans le vent ont une dispersion de vitesses de l’ordre de 150-300 km/s. De manière générale, les variations de vitesse radiales sont anti-corrélées avec le coefficient d’asymétrie de la raie, ce qui semble infirmer la présence d’un compagnon. Des observations en photométrie de l’étoile WR103 montrent une période de 9.1 ± 0.6 jours qui s’accorde avec les variations spectroscopiques et qui ne semble pas, de manière évidente, d’origine binaire.

Recherche de champs magnétiques chez les étoiles Wolf-Rayet par l'analyse d'observations spectropolarimétriques.

Cette thèse de doctorat présente les résultats d'un relevé spectropolarimétrique visant la détection directe de champs magnétiques dans le vent d'étoiles Wolf-Rayet (WR). Les observations furent entièrement obtenues à partir du spectropolarimètre ESPaDOnS, installé sur le télescope de l'observatoire Canada-France-Hawaii. Ce projet débuta par l'observation d'un étoile très variable de type WN4 appelée EZ CMa = WR6 = HD 50896 et se poursuivit par l'observation de 11 autres étoiles WR de notre galaxie. La méthode analytique utilisée dans cette étude vise à examiner les spectres de polarisation circulaire (Stokes V) et à identifier, au travers des raies d'émission, les signatures spectrales engendrées par la présence de champs magnétiques de type split monopole dans les vents des étoiles observées. Afin de pallier à la présence de polarisation linéaire dans les données de polarisation circulaire, le cross-talk entre les spectres Stokes Q et U et le spectre Stokes V fut modélisé et éliminé avant de procéder à l'analyse magnétique. En somme, aucun champ magnétique n'est détecté de manière significative dans les 12 étoiles observées. Toutefois, une détection marginale est signalée pour les étoiles WR134, WR137 et WR138 puisque quelques-unes de leur raies spectrales semblent indiquer la présence d'une signature magnétique. Pour chacune de ces trois étoiles, la valeur la plus probable du champ magnétique présent dans le vent stellaire est respectivement de B ~ 200, 130 et 80 G. En ce qui concerne les autres étoiles pour lesquelles aucune détection magnétique ne fut obtenue, la limite supérieure moyenne de l'intensité du champ qui pourrait être présent dans les données, sans toutefois être détecté, est évaluée à 500 G. Finalement, les résultats de cette étude ne peuvent confirmer l'origine magnétique des régions d'interaction en co-rotation (CIR) observées chez plusieurs étoiles WR. En effet, aucun champ magnétique n'est détecté de façon convaincante chez les quatre étoiles pour lesquelles la présence de CIR est soupçonnée.

Modelling the line variations from the wind-wind shock emissions of WR 30a

The study of Wolf-Rayet stars plays an important role in evolutionary theories of massive stars. Among these objects, similar to 20 per cent are known to be in binary systems and can therefore be used for the mass determination of these stars. Most of these systems are not spatially resolved and spectral lines can be used to constrain the orbital parameters. However, part of the emission may originate in the interaction zone between the stellar winds, modifying the line profiles and thus challenging us to use different models to interpret them. In this work, we analysed the He II lambda 4686 angstrom + C IV lambda 4658 angstrom blended lines of WR 30a (WO4+O5) assuming that part of the emission originate in the wind-wind interaction zone. In fact, this line presents a quiescent base profile, attributed to the WO wind, and a superposed excess, which varies with the orbital phase along the 4.6-d period. Under these assumptions, we were able to fit the excess spectral line profile and central velocity for all phases, except for the longest wavelengths, where a spectral line with constant velocity seems to be present. The fit parameters provide the eccentricity and inclination of the binary orbit, from which it is possible to constrain the stellar masses.

TOWARD UNDERSTANDING THE B[e] PHENOMENON. IV. MODELING OF IRAS 00470+6429

FS CMa type stars are a recently described group of objects with the B[e] phenomenon which exhibits strong emission-line spectra and strong IR excesses. In this paper, we report the first attempt for a detailed modeling of IRAS 00470+6429, for which we have the best set of observations. Our modeling is based on two key assumptions: the star has a main-sequence luminosity for its spectral type (B2) and the circumstellar (CS) envelope is bimodal, composed of a slowly outflowing disklike wind and a fast polar wind. Both outflows are assumed to be purely radial. We adopt a novel approach to describe the dust formation site in the wind that employs timescale arguments for grain condensation and a self-consistent solution for the dust destruction surface. With the above assumptions we were able to satisfactorily reproduce many observational properties of IRAS 00470+6429, including the Hi line profiles and the overall shape of the spectral energy distribution. Our adopted recipe for dust formation proved successful in reproducing the correct amount of dust formed in the CS envelope. Possible shortcomings of our model, as well as suggestions for future improvements, are discussed.

Revisiting 2D numerical models for the 19th century outbursts of eta Carinae

We present here new results of two-dimensional hydrodynamical simulations of the eruptive events of the 1840s (the great) and the 1890s (the minor) eruptions suffered by the massive star eta Carinae (Car). The two bipolar nebulae commonly known as the Homunculus and the little Homunculus (LH) were formed from the interaction of these eruptive events with the underlying stellar wind. We assume here an interacting, non-spherical multiple-phase wind scenario to explain the shape and the kinematics of both Homunculi, but adopt a more realistic parametrization of the phases of the wind. During the 1890s eruptive event, the outflow speed decreased for a short period of time. This fact suggests that the LH is formed when the eruption ends, from the impact of the post-outburst eta Car wind (that follows the 1890s event) with the eruptive flow (rather than by the collision of the eruptive flow with the pre-outburst wind, as claimed in previous models; Gonzalez et al.). Our simulations reproduce quite well the shape and the observed expansion speed of the large Homunculus. The LH (which is embedded within the large Homunculus) becomes Rayleigh-Taylor unstable and develop filamentary structures that resemble the spatial features observed in the polar caps. In addition, we find that the interior cavity between the two Homunculi is partially filled by material that is expelled during the decades following the great eruption. This result may be connected with the observed double-shell structure in the polar lobes of the eta Car nebula. Finally, as in previous work, we find the formation of tenuous, equatorial, high-speed features that seem to be related to the observed equatorial skirt of eta Car.

ON THE NATURE OF THE PROTOTYPE LUMINOUS BLUE VARIABLE AG CARINAE. II. WITNESSING A MASSIVE STAR EVOLVING CLOSE TO THE EDDINGTON AND BISTABILITY LIMITS

We show that the significantly different effective temperatures (T(eff)) achieved by the luminous blue variable AG Carinae during the consecutive visual minima of 1985-1990 (T(eff) similar or equal to 22,800 K) and 2000-2001 (T(eff) similar or equal to 17,000 K) place the star on different sides of the bistability limit, which occurs in line-driven stellar winds around T(eff) similar to 21,000 K. Decisive evidence is provided by huge changes in the optical depth of the Lyman continuum in the inner wind as T(eff) changes during the S Dor cycle. These changes cause different Fe ionization structures in the inner wind. The bistability mechanism is also related to the different wind parameters during visual minima: the wind terminal velocity was 2-3 times higher and the mass-loss rate roughly two times smaller in 1985-1990 than in 2000-2003. We obtain a projected rotational velocity of 220 +/- 50 km s(-1) during 1985-1990 which, combined with the high luminosity (L(star) = 1.5 x 10(6) L(circle dot)), puts AG Car extremely close to the Eddington limit modified by rotation (Omega Gamma limit): for an inclination angle of 90 degrees, Gamma(Omega) greater than or similar to 1.0 for M(circle dot) less than or similar to 60. Based on evolutionary models and mass budget, we obtain an initial mass of similar to 100 M(circle dot) and a current mass of similar to 60-70 M(circle dot) for AG Car. Therefore, AG Car is close to, if not at, the Omega Gamma limit during visual minimum. Assuming M = 70 M(circle dot), we find that Gamma(Omega) decreases from 0.93 to 0.72 as AG Car expands toward visual maximum, suggesting that the star is not above the Eddington limit during maximum phases.

Strukturelle, chemische und isotopische Analysen nanoskaliger präsolarer und früher solarer Kondensate

Neben astronomischen Beobachtungen mittels boden- und satellitengestützer Instrumente existiert ein weiterer experimenteller Zugang zu astrophysikalischen Fragestellungen in Form einer Auswahl extraterrestrischen Materials, das für Laboruntersuchungen zur Verfügung steht. Hierzu zählen interplanetare Staubpartikel, Proben, die von Raumfahrzeugen zur Erde zurückgebracht wurden und primitive Meteorite. Von besonderem Interesse sind sog. primitive kohlige Chondrite, eine Klasse von Meteoriten, die seit ihrer Entstehung im frühen Sonnensystem kaum verändert wurden. Sie enthalten neben frühem solarem Material präsolare Minerale, die in Sternwinden von Supernovae und roten Riesensternen kondensiert sind und die Bildung unseres Sonnensystems weitgehend unverändert überstanden haben. Strukturelle, chemische und isotopische Analysen dieser Proben besitzen demnach eine große Relevanz für eine Vielzahl astrophysikalischer Forschungsgebiete. Im Rahmen der vorliegenden Arbeit wurden Laboranalysen mittels modernster physikalischer Methoden an Bestandteilen primitiver Meteorite durchgeführt. Aufgrund der Vielfalt der zu untersuchenden Eigenschaften und der geringen Größen der analysierten Partikel zwischen wenigen Nanometern und einigen Mikrometern mussten hierbei hohe Anforderungen an Nachweiseffizienz und Ortsauflösung gestellt werden. Durch die Kombination verschiedener Methoden wurde ein neuer methodologischer Ansatz zur Analyse präsolarer Minerale (beispielsweise SiC) entwickelt. Aufgrund geringer Mengen verfügbaren Materials basiert dieses Konzept auf der parallelen nichtdestruktiven Vorcharakterisierung einer Vielzahl präsolarer Partikel im Hinblick auf ihren Gehalt diagnostischer Spurenelemente. Eine anschließende massenspektrometrische Untersuchung identifizierter Partikel mit hohen Konzentrationen interessanter Elemente ist in der Lage, Informationen zu nukleosynthetischen Bedingungen in ihren stellaren Quellen zu liefern. Weiterhin wurden Analysen meteoritischer Nanodiamanten durchgeführt, deren geringe Größen von wenigen Nanometern zu stark modifizierten Festkörpereigenschaften führen. Im Rahmen dieser Arbeit wurde eine quantitative Beschreibung von Quanteneinschluss-Effekten entwickelt, wie sie in diesen größenverteilten Halbleiter-Nanopartikeln auftreten. Die abgeleiteten Ergebnisse besitzen Relevanz für nanotechnologische Forschungen. Den Kern der vorliegenden Arbeit bilden Untersuchungen an frühen solaren Partikeln, sog. refraktären Metall Nuggets (RMN). Mit Hilfe struktureller, chemischer und isotopischer Analysen, sowie dem Vergleich der Ergebnisse mit thermodynamischen Rechnungen, konnte zum ersten Mal ein direkter Nachweis von Kondensationsprozessen im frühen solaren Nebel erbracht werden. Die analysierten RMN gehören zu den ersten Festkörperkondensaten, die im frühen Sonnensystem gebildet wurden und scheinen seit ihrer Entstehung nicht durch sekundäre Prozesse verändert worden zu sein. Weiterhin konnte erstmals die Abkühlrate des Gases des lokalen solaren Nebels, in dem die ersten Kondensationsprozesse stattfanden, zu 0.5 K/Jahr bestimmt werden, wodurch ein detaillierter Blick in die thermodynamische Geschichte des frühen Sonnensystems möglich wird. Die extrahierten Parameter haben weitreichende Auswirkungen auf die Modelle der Entstehung erster solarer Festkörper, welche die Grundbausteine der Planetenbildung darstellen.