目录
图像
上传图像
DSS Images Other Images
相关文章
A multiwavelength investigation of the temperature of the cold neutral medium
We present measurements of the HI spin temperatures (Ts) ofthe cold neutral medium (CNM) towards radio sources that are closelyaligned with stars for which published H2 ortho-paratemperatures (T01) are available from ultraviolet (UV)observations. Our sample consists of 18 radio sources close to 16 nearbystars. The transverse separation of the lines of sight of thecorresponding UV and radio observations varies from 0.1 to 12.0 pc atthe distance of the star. The UV measurements do not have velocityinformation, so we use the velocities of low ionization species (e.g.NaI/KI/CI) observed towards these same stars to make a plausibleidentification of the CNM corresponding to the H2 absorption.We then find that T01 and Ts match withinobservational uncertainties for lines of sight with H2 columndensity above 1015.8cm-2, but deviate from eachother below this threshold. This is consistent with the expectation thatin the CNM Ts tracks the kinetic temperature due tocollisions and that T01 is driven towards the kinetictemperature by proton exchange reactions.
| The Discordance of Mass-Loss Estimates for Galactic O-Type Stars
We have determined accurate values of the product of the mass-loss rateand the ion fraction of P+4, M˙q(P+4), for asample of 40 Galactic O-type stars by fitting stellar wind profiles toobservations of the P V resonance doublet obtained with FUSE, ORFEUSBEFS, and Copernicus. When P+4 is the dominant ion in thewind [i.e., 0.5<~q(P+4)<=1], M˙q(P+4)approximates the mass-loss rate to within a factor of <~2. Theorypredicts that P+4 is the dominant ion in the winds of O7-O9.7stars, although an empirical estimator suggests that the range O4-O7 maybe more appropriate. However, we find that the mass-loss rates obtainedfrom P V wind profiles are systematically smaller than those obtainedfrom fits to Hα emission profiles or radio free-free emission bymedian factors of ~130 (if P+4 is dominant between O7 andO9.7) or ~20 (if P+4 is dominant between O4 and O7). Thesediscordant measurements can be reconciled if the winds of O stars in therelevant temperature range are strongly clumped on small spatial scales.We use a simplified two-component model to investigate the volumefilling factors of the denser regions. This clumping implies thatmass-loss rates determined from ``ρ2'' diagnostics havebeen systematically overestimated by factors of 10 or more, at least fora subset of O stars. Reductions in the mass-loss rates of this size haveimportant implications for the evolution of massive stars andquantitative estimates of the feedback that hot-star winds provide totheir interstellar environments.
| Bright OB stars in the Galaxy. III. Constraints on the radial stratification of the clumping factor in hot star winds from a combined Hα, IR and radio analysis
Context: .Recent results strongly challenge the canonical picture ofmassive star winds: various evidence indicates that currently acceptedmass-loss rates, {dot M}, may need to be revised downwards, by factorsextending to one magnitude or even more. This is because the mostcommonly used mass-loss diagnostics are affected by "clumping"(small-scale density inhomogeneities), influencing our interpretation ofobserved spectra and fluxes. Aims: .Such downward revisions wouldhave dramatic consequences for the evolution of, and feedback from,massive stars, and thus robust determinations of the clumping propertiesand mass-loss rates are urgently needed. We present a first attemptconcerning this objective, by means of constraining the radialstratification of the so-called clumping factor. Methods: .To thisend, we have analyzed a sample of 19 Galactic O-type supergiants/giants,by combining our own and archival data for Hα, IR, mm and radiofluxes, and using approximate methods, calibrated to more sophisticatedmodels. Clumping has been included into our analysis in the"conventional" way, by assuming the inter-clump matter to be void.Because (almost) all our diagnostics depends on the square of density,we cannot derive absolute clumping factors, but only factors normalizedto a certain minimum. Results: .This minimum was usually found tobe located in the outermost, radio-emitting region, i.e., the radiomass-loss rates are the lowest ones, compared to {dot M} derived fromHα and the IR. The radio rates agree well with those predicted bytheory, but are only upper limits, due to unknown clumping in the outerwind. Hα turned out to be a useful tool to derive the clumpingproperties inside r < 3{ldots}5 Rstar. Our most importantresult concerns a (physical) difference between denser and thinnerwinds: for denser winds, the innermost region is more strongly clumpedthan the outermost one (with a normalized clumping factor of 4.1± 1.4), whereas thinner winds have similar clumping properties inthe inner and outer regions. Conclusions: .Our findings arecompared with theoretical predictions, and the implications arediscussed in detail, by assuming different scenarios regarding the stillunknown clumping properties of the outer wind.
| Forty Years of Spectroscopic Stellar Astrophysics in Japan
The development of Japanese spectroscopic stellar astrophysics in therecent 40 years is reviewed from an observational point of view. In thisarticle, the research activities are provisionally divided into fourfields: hot stars, hot emission-line (Be) stars, cool stars, and otherstars. Historical developments of the observational facilities atOkayama Astrophysical Observatory (spectrographs and detectors) are alsosummarized in connection with the progress in scientific researchactivities.
| Correlation patterns between 11 diffuse interstellar bands and ultraviolet extinction
We relate the equivalent widths of 11 diffuse interstellar bands,measured in the spectra of 49 stars, to different colour excesses in theultraviolet. We find that most of the observed bands correlatepositively with the extinction in the neighbourhood of the2175-Åbump. Correlation with colour excesses in other parts of theextinction curve is more variable from one diffuse interstellar band toanother; we find that some diffuse bands (5797, 5850 and 6376 Å)correlate positively with the overall slope of the extinction curve,while others (5780 and 6284 Å) exhibit negative correlation. Wediscuss the implications of these results on the links between thediffuse interstellar band carriers and the properties of theinterstellar grains.
| A Medium Resolution Near-Infrared Spectral Atlas of O and Early-B Stars
We present intermediate-resolution (R~8000-12,000) high signal-to-noise(S/N) H- and K-band spectroscopy of a sample of 37 optically visiblestars, ranging in spectral type from O3 to B3 and representing mostluminosity classes. Spectra of this quality can be used to constrain thetemperature, luminosity, and general wind properties of OB stars, whenused in conjunction with sophisticated atmospheric model codes. Mostimportant is the need for moderately high resolutions (R>=5000) andvery high signal-to-noise (S/N>=150) spectra for a meaningful profileanalysis. When using near-infrared spectra for a classification system,moderately high signal-to-noise (S/N~100) is still required, though theresolution can be relaxed to just a thousand or two. In the Appendix weprovide a set of very high-quality near-infrared spectra of Brackettlines in six early-A dwarfs. These can be used to aid in the modelingand removal of such lines when early-A dwarfs are used for telluricspectroscopic standards.
| Cloud Structure and Physical Conditions in Star-forming Regions from Optical Observations. II. Analysis
To complement the optical absorption line survey of diffuse moleculargas in Paper I, we obtained and analyzed far-ultraviolet H2and CO data on lines of sight toward stars in Cep OB2 and Cep OB3.Possible correlations between column densities of different species forindividual velocity components, not total columns along a line of sightas in the past, were examined and were interpreted in terms of cloudstructure. The analysis reveals that there are two kinds of CH indiffuse molecular gas: CN-like CH and CH+-like CH. Evidenceis provided that CO is also associated with CN in diffuse molecularclouds. Different species are distributed according to gas density inthe diffuse molecular gas. Both calcium and potassium may be depletedonto grains in high-density gas, but with different dependencies onlocal gas density. Gas densities for components where CN was detectedwere inferred from a chemical model. Analysis of cloud structureindicates that our data are generally consistent with the large-scalestructure suggested by maps of CO millimeter-wave emission. On smallscales, the gas density is seen to vary by factors greater than 5.0 overscales of ~10,000 AU. The relationships between column densities of COand CH with that of H2 along a line of sight show similarslopes for the gas toward Cep OB2 and Cep OB3, but the CO/H2and CH/H2 ratios tend to differ, which we ascribe tovariation in average density along the line of sight.
| Bright OB stars in the Galaxy. II. Wind variability in O supergiants as traced by Hα
We investigate the line-profile variability (lpv) of Hα for alarge sample of O-type supergiants (15 objects between O4 and O9.7), inan objective, statistically rigorous manner. We employed the TemporalVariance Spectrum (TVS) analysis, developed for the case of photosphericabsorption lines and modified by us to take into account the effects ofwind emission. By means of a comparative analysis we place constraintson the properties of this variability - quantified in terms of a meanand a newly defined fractional amplitude of deviations - as a functionof stellar and wind parameters. The results of our analysis show thatall the stars in the sample show evidence of significant lpv inHα, mostly dominated by processes in the wind. The variationsoccur between zero and 0.3 v_&infy; (i.e., below 1.5 R_star ), in goodagreement with results from similar studies. A comparison between theobservations and corresponding line-profile simulations indicates thatfor stars with intermediate wind densities the properties of theHα variability can be explained by simple models consisting ofcoherent or broken shells (blobs) uniformly distributed over the windvolume, with an intrinsic scatter in the maximum density contrast ofabout a factor of two. For stars at lower and higher wind densities, onthe other hand, we found certain inconsistencies between theobservations and our predictions, most importantly concerning the meanamplitude and the symmetry properties of the TVS. This disagreementmight be explained by the presence of coherent large-scale structures,partly confined in a volume close to the star. Interpreted in terms of avariable mass-loss rate, the observed variations of Hα indicatechanges of ±4% with respect to the mean value of dot M for starswith stronger winds and of ± 16% for stars with weaker winds. Theeffect of these variations on the corresponding wind momenta is ratherinsignificant (less than 0.16 dex), increasing only the local scatterwithout affecting the Wind Momentum Luminosity Relationship.
| Quantitative H and K band spectroscopy of Galactic OB-stars at medium resolution
In this paper we have analyzed 25 Galactic O and early B-stars by meansof H and K band spectroscopy, with the primary goal to investigate towhat extent a lone near-IR spectroscopy is able to recover stellar andwind parameters derived in the optical. Most of the spectra have beentaken with subaru-ircs, at an intermediate resolution of 12 000, andwith a very high S/N, mostly on the order of 200 or better. In order tosynthesize the strategic H/He lines, we have used our recent,line-blanketed version of fastwind (Puls et al. 2005, A&A, 435,669). In total, seven lines have been investigated, where for two starswe could make additional use of the Hei2.05 singlet which has beenobserved with irtf-cshell. Apart from Brγ and Heii2.18, the otherlines are predominately formed in the stellar photosphere, and thusremain fairly uncontaminated from more complex physical processes,particularly clumping. First we investigated the predicted behaviour ofthe strategic lines. In contradiction to what one expects from theoptical in the O-star regime, almost all photospheric H/Hei/Heii H/Kband lines become stronger if the gravity decreases. Concerning H andHeii, this finding is related to the behaviour of Stark broadening as afunction of electron density, which in the line cores is different formembers of lower (optical) and higher (IR) series. Regarding Hei, thepredicted behaviour is due to some subtle NLTE effects resulting in astronger overpopulation of the lower level when the gravity decreases.We have compared our calculations with results from the alternative NLTEmodel atmosphere code cmfgen (Hillier & Miller 1998, ApJ, 496, 407).In most cases, we found reasonable or nearly perfect agreement. Only theHei2.05 singlet for mid O-types suffers from some discrepancy, analogouswith findings for the optical Hei singlets. For most of our objects, weobtained good fits, except for the line cores of Brγ in earlyO-stars with significant mass-loss. Whereas the observations showBrγ mostly as rather symmetric emission lines, the models predicta P Cygni type profile with strong absorption. This discrepancy (whichalso appears in lines synthesized by cmfgen) might be an indirect effectof clumping. After having derived the stellar and wind parameters fromthe IR, we have compared them to results from previous optical analyses.Overall, the IR results coincide in most cases with the optical oneswithin the typical errors usually quoted for the correspondingparameters, i.e., an uncertainty in T_eff of 5%, in log g of 0.1 dex andin {dot M} of 0.2 dex, with lower errors at higher wind densities.Outliers above the 1-σ level where found in four cases withrespect to log g and in two cases for {dot M}.
| On the massive stellar population of the super star cluster Westerlund 1
We present new spectroscopic and photometric observations of the youngGalactic open cluster Westerlund 1 (Wd 1) that reveala unique population of massive evolved stars. We identify ~200 clustermembers and present spectroscopic classifications for ~25% of these. Wefind that all stars so classified are unambiguously post-Main Sequenceobjects, consistent with an apparent lack of an identifiable MainSequence in our photometric data to V 20. We are able to identifyrich populations of Wolf Rayet stars, OB supergiants and short livedtransitional objects. Of these, the latter group consists of both hot(Luminous Blue Variable and extreme B supergiant) and cool (YellowHypergiant and Red Supergiant) objects - we find that half the knownGalactic population of YHGs resides within Wd 1. We obtain a meanV-MV ~ 25 mag from the cluster Yellow Hypergiants, implying aMain Sequence turnoff at or below MV =-5 (O7 V or later).Based solely on the masses inferred for the 53 spectroscopicallyclassified stars, we determine an absolute minimum mass of ~1.5 ×10^3~Mȯ for Wd 1. However, considering the completephotometrically and spectroscopically selected cluster population andadopting a Kroupa IMF we infer a likely mass for Wd 1 of~10^5~Mȯ, noting that inevitable source confusion andincompleteness are likely to render this an underestimate. As such, Wd 1is the most massive compact young cluster yet identified in the LocalGroup, with a mass exceeding that of Galactic Centre clusters such asthe Arches and Quintuplet. Indeed, the luminosity, inferred mass andcompact nature of Wd 1 are comparable with those of Super Star Clusters- previously identified only in external galaxies - and is consistentwith expectations for a Globular Cluster progenitor.
| Correlations between diffuse interstellar bands and atomic lines
We present and discuss correlations between strengths of the well-known,strong interstellar atomic lines of KI and CaII, and four selected,strong unidentified diffuse interstellar bands (DIBs): 5780, 5797, 5850and 6614. In order to analyse a homogeneous sample of echellehigh-resolution spectra it has been chosen to use measurements fromTerskol Observatory in Northern Caucasus plus a selected number ofhigher resolution observations performed using other instruments. Wedemonstrate that the strength of certain DIBs correlate well withneutral potassium lines and to a much lower degree with ionized calciumlines. This fact suggests that the degree of irradiation of a cloud withUV photons, capable to ionize interstellar atoms, plays a crucial rolein the formation/maintenance of certain molecular species: possiblecarriers of DIBs.
| Anomalous dust-to-gas ratios in the Galaxy
Lines of sight with E(B-V)/N(HI) considerably smaller than the averagevalue for the solar neighbourhood have been selected from the catalogueof Diplas & Savage. In order to develop quantitative considerations,estimates of the molecular hydrogen column density were obtained usingthe relation of Savage et al. extended at E(B-V) > 0.4 with therecent data of Rachford et al. Contrary to the prevailing opinion in theliterature for sightlines with similar behaviour, we found that only 22per cent of our sample was characterized by both an average gas densitylarger than 1 cm-3 and a value of RV larger thanthat in the diffuse interstellar medium. By computing extinction models,we were able to fit the E(B-V)/N(HI) by changing the value ofRV only for some sightlines. For the remaining ones, aρd/ρH ratio different from the averageGalactic value must be invoked. The application of the Kramers-Kronigrelation to the observed extinction curves confirmed this possibility.Moreover, attempts to fit such curves with models having grain volumescorresponding to the standard ρd/ρH ratiofailed.We find a linear relation between ρd/ρHand E(B-V)/N(H) for our sightlines. The average Galactic value marks theseparation into two groups characterized by lower abundances of C and Sitrapped into the grains when E(B-V)/N(H) is smaller than the Galacticvalue, and by larger abundances when E(B-V)/N(H) is greater.
| Line Variability in the Spectrum of Supergiant alpha Cam
CCD spectra taken with the PFES and CEGS echelle spectrographs attachedto the 6-m Special Astrophysical Observatory (Russian Academy ofSciences) telescope and the 2-m Shamakha Astrophysical Observatory(National Academy of Sciences of Azerbaijan) telescope, respectively,were used to study the line-profile variations in the spectrum of thehot supergiant alpha Cam. No fast (<1.5 h) line-profile andradial-velocity variations were found. Some of the systematic effectsthat cause spurious variability are considered. The H-alpha profilevariability appears symmetric relative to the radial velocity of thestar's center of mass and is attributable to variable blueshifted andredshifted emission and/or absorption components superimposed on avariable photospheric profile. The H line shows evidence of alarge-scale mass ejection from the stellar surface, which is alsotraceable in other spectral lines. The He II 4686 line exhibits aninverse P Cyg profile, while the red wing of the He I 5876 line showsweak and variable emission. The fast (on characteristic time scales ofshorter than an hour) variability of the He II 4686 profile that waspreviously revealed by our observations (Kholtygin et al. 2000) iscalled into question. A comparison of the observational data on thevariability of ultraviolet and optical line profiles for the supergiantalpha Cam suggests that nonradial motions are mainly responsible for theradial-velocity and line-profile variability.
| Cloud Structure and Physical Conditions in Star-forming Regions from Optical Observations. I. Data and Component Structure
We present high-resolution optical spectra (at ~0.6-1.8 kms-1) of interstellar CN, CH, CH+, Ca I, K I, andCa II absorption toward 29 lines of sight in three star-forming regions,ρ Oph, Cep OB2, and Cep OB3. The observations and data reduction aredescribed. The agreement between earlier measurements of the totalequivalent widths and our results is quite good. However, our higherresolution spectra reveal complex structure and closely blendedcomponents in most lines of sight. The velocity component structure ofeach species is obtained by analyzing the spectra of the six species fora given sight line together. The tabulated column densities and Dopplerparameters of individual components are determined by using the methodof profile fitting. Total column densities along lines of sight arecomputed by summing results from profile fitting for individualcomponents and are compared with column densities from the apparentoptical depth method. A more detailed analysis of these data and theirimplications will be presented in a companion paper.
| A Galactic O Star Catalog
We have produced a catalog of 378 Galactic O stars with accuratespectral classifications that is complete for V<8 but includes manyfainter stars. The catalog provides cross-identifications with othersources; coordinates (obtained in most cases from Tycho-2 data);astrometric distances for 24 of the nearest stars; optical (Tycho-2,Johnson, and Strömgren) and NIR photometry; group membership,runaway character, and multiplicity information; and a Web-based versionwith links to on-line services.
| Terminal Velocities of Luminous, Early-Type Stars in the Small Magellanic Cloud
Ultraviolet spectra from the Space Telescope Imaging Spectrograph (STIS)are used to determine terminal velocities for 11 O and B-type giants andsupergiants in the Small Magellanic Cloud (SMC) from the Si IV and C IVresonance lines. Using archival data from observations with the GoddardHigh-Resolution Spectrograph and the International Ultraviolet Explorertelescope, terminal velocities are obtained for a further five B-typesupergiants. We discuss the metallicity dependence of stellar terminalvelocities for supergiants, finding no evidence for a significantscaling between Galactic and SMC metallicities forTeff<30,000 K, consistent with the predictions ofradiation-driven wind theory. A comparison of thev&infy;/vesc ratio between the SMC and Galacticsamples, while consistent with the above statement, emphasizes that theuncertainties in the distances to galactic OB-type stars are a seriousobstacle to a detailed comparison with theory. For the SMC sample thereis considerable scatter in v&infy;/vesc at agiven effective temperature, perhaps indicative of uncertainties instellar masses.
| On the Hipparcos parallaxes of O stars
We compare the absolute visual magnitude of the majority of bright Ostars in the sky as predicted from their spectral type with the absolutemagnitude calculated from their apparent magnitude and the Hipparcosparallax. We find that many stars appear to be much fainter thanexpected, up to five magnitudes. We find no evidence for a correlationbetween magnitude differences and the stellar rotational velocity assuggested for OB stars by Lamers et al. (1997, A&A, 325, L25), whosesmall sample of stars is partly included in ours. Instead, by means of asimulation we show how these differences arise naturally from the largedistances at which O stars are located, and the level of precision ofthe parallax measurements achieved by Hipparcos. Straightforwardlyderiving a distance from the Hipparcos parallax yields reliable resultsfor one or two O stars only. We discuss several types of bias reportedin the literature in connection with parallax samples (Lutz-Kelker,Malmquist) and investigate how they affect the O star sample. Inaddition, we test three absolute magnitude calibrations from theliterature (Schmidt-Kaler et al. 1982, Landolt-Börnstein; Howarth& Prinja 1989, ApJS, 69, 527; Vacca et al. 1996, ApJ, 460, 914) andfind that they are consistent with the Hipparcos measurements. AlthoughO stars conform nicely to the simulation, we notice that some B stars inthe sample of \citeauthor{La97} have a magnitude difference larger thanexpected.
| Stellar and wind parameters of Galactic O-stars. The influence of line-blocking/blanketing
We have re-analyzed the Galactic O-star sample from \citet{puls96} bymeans of line-blanketed NLTE model atmospheres in order to investigatethe influence of line-blocking/blanketing on the derived parameters. Theanalysis has been carried out by fitting the photospheric and wind linesfrom H and He. In most cases we obtained a good fit, but we have alsofound certain inconsistencies which are probably related to a stillinadequate treatment of the wind structure. These inconsistenciescomprise the line cores of Hγ and Hβ insupergiants (the synthetic profiles are too weak when the mass-loss rateis determined by matching Hα) and the ``generalizeddilution effect'' (cf. \citealt{vo89}) which is still present in He I4471 of cooler supergiants and giants.Compared to pure H/He plane-parallel models we found a decrease ineffective temperatures which is largest at earliest spectral types andfor supergiants (with a maximum shift of roughly 8000 K). This findingis explained by the fact that line-blanketed models of hot stars havephotospheric He ionization fractions similar to those from unblanketedmodels at higher Teff and higher log g. Consequently, anyline-blanketed analysis based on the He ionization equilibrium resultsin lower Teff-values along with a reduction of either log gor helium abundance (if the reduction of log g is prohibited by theBalmer line wings). Stellar radii and mass-loss rates, on the otherhand, remain more or less unaffected by line-blanketing.We have calculated ``new'' spectroscopic masses and compared them withprevious results. Although the former mass discrepancy \citep{h92}becomes significantly reduced, a systematic trend for masses below 50Msun seems to remain: The spectroscopically derived valuesare smaller than the ``evolutionary masses'' by roughly 10Msun. Additionally, a significant fraction of our samplestars stays over-abundant in He, although the actual values were foundto be lower than previously determined.Also the wind-momentum luminosity relation (WLR) changes because oflower luminosities and almost unmodified wind-momentum rates. Comparedto previous results, the separation of the WLR as a function ofluminosity class is still present but now the WLR for giants/dwarfs isconsistent with theoretical predictions.We argue that the derived mass-loss rates of stars withHα in emission are affected by clumping in the lowerwind region. If the predictions from different and independenttheoretical simulations (\citealt {Vink00, Paul03, puls03a}) that theWLR should be independent of luminosity class were correct, a typicalclumping factor <ρ2>/<ρ>2 ≈5 should be derived by ``unifying'' the different WLRs.Based upon observations obtained at the INT and the European SouthernObservatory, La Silla, Chile. The INT is operated on the island of LaPalma by the ING in the Spanish Observatorio de El Roque de losMuchachos of the Instituto de Astrofísica de Canarias.Appendix A in only available in electronic form athttp://www.edpsciences.org
| On the relation between diffuse bands and column densities of H2, CH and CO molecules
Mutual relations between column densities of H2, CH and COmolecules as well as between the latter and strengths of the major 5780and 5797 diffuse bands are presented and discussed. The CH radical seemsto be a good H2 tracer, possibly better than CO. It is alsodemonstrated that the molecular fraction of the H2 moleculeis correlated with an intensity ratio of 5797 and 5780 DIBs, suggestingthe possible formation of narrow DIB carriers in denser clouds,dominated by molecular hydrogen and reasonably shielded from ionizing UVradiation by small dust grains.Tables 1 and 2 are only available in electronic form at the CDS viaanonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or viahttp://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/414/949
| Bright OB stars in the Galaxy. I. Mass-loss and wind-momentum rates of O-type stars: A pure H\alpha analysis accounting for line-blanketing
We study mass-loss and wind momentum rates of 29 Galactic O-type starswith luminosity classes I, III and V by means of a pure H\alpha profileanalysis and investigate to what extent the results compare to thoseoriginating from a state-of-the-art, complete spectral analysis. Ourinvestigation relies on the approximate method developed by\citet{Puls96} which we have modified to account for the effects ofline-blanketing. Effective temperatures and gravities needed to obtainquantitative results from such a simplified approach have been derivedby means of calibrations based on most recent spectroscopic NLTEanalyses and models of Galactic stars by \citet{Repo03} and\citet{Martins02}. Comparing (i) the derived wind-densities to thosedetermined by \citet{Repo03} for eleven stars in common and (ii) theWind-momentum Luminosity Relationship (WLR) for our sample stars tothose derived by other investigations, we conclude that our approximateapproach is actually able to provide consistent results. Additionally,we studied the consequences of ``fine tuning'' some of the direct andindirect parameters entering the WLR, especially by accounting fordifferent possible values of stellar reddening and distances. Combiningour data set with the corresponding data provided by \citet{Herrero02}and \citet{Repo03} we finally study the WLR for the largest sample ofGalactic O-type stars gathered so far, including an elaborate errortreatment. The established disagreement between the theoreticalpredictions and the ``observed'' WLRs being a function of luminosityclass is suggested to be a result of wind clumping. Different strategiesto check this hypothesis are discussed, particularly by comparing theH\alpha mass-loss rates with the ones derived from radio observations.
| Toward an adequate method to isolate spectroscopic families of diffuse interstellar bands
We divide some of the observed diffuse interstellar bands (DIBs) intofamilies that appear to have the spectral structure of single species.Three different methods are applied to separate such families, exploringthe best approach for future investigations of this type. Starting witha statistical treatment of the data, we found that statistical methodsby themselves give insufficient results. Two other methods of dataanalysis (`averaging equivalent widths' and `investigating the figureswith arranged spectrograms') were found to be more useful as tools forfinding the spectroscopic families of DIBs. On the basis of thesemethods, we suggest some candidates as `relatives' of 5780- and5797-Å bands.
| Fast Line-Profile Variability in the Spectra of O Stars
A program of the search for and analysis of profile variability in thespectra of bright O supergiants with a time resolution of 5-30 min isdescribed. Preliminary results of the spectroscopic observations of thestars lambda Ori, alpha Cam, and 19 Cep with the 1-m SpecialAstrophysical Observatory telescope in 2001 are presented. Line-profilevariability was detected in the spectra of all the stars studied;variability in the H_alpha and C III lambda5696 lines in the spectrum oflambda Ori has been found for the first time. The variability amplitudeis 4-5% for 19 Cep and 1-2% for alpha Cam and lambda Ori on time scalesfrom several hours to 3 or 4 days, and the width of the variablefeatures reaches 2 Angstroms (100 km/s). We detected cyclical variationsin the He II lambda4686 and C III lambda5696 line profiles in thespectrum of lambda Ori on time scales of 1.3-1.6 days. Rapid profilevariations on time scales of 3.5-7 h were found in the violet parts ofthe H_alpha and He I lambda4715 line profiles in the spectrum of lambdaOri A.
| High-Resolution Observations of Interstellar Ca I Absorption-Implications for Depletions and Electron Densities in Diffuse Clouds
We present high-resolution (FWHM~0.3-1.5 km s-1) spectra,obtained with the AAT UHRF, the McDonald Observatory 2.7 m coudéspectrograph, and/or the KPNO coudé feed, of interstellar Ca Iabsorption toward 30 Galactic stars. Comparisons of the column densitiesof Ca I, Ca II, K I, and other species-for individual componentsidentified in the line profiles and also when integrated over entirelines of sight-yield information on relative electron densities anddepletions (dependent on assumptions regarding the ionizationequilibrium). There is no obvious relationship between the ratio N(CaI)/N(Ca II) [equal to ne/(Γ/αr) forphotoionization equilibrium] and the fraction of hydrogen in molecularform f(H2) (often taken to be indicative of the local densitynH). For a smaller sample of sight lines for which thethermal pressure (nHT) and local density can be estimated viaanalysis of the C I fine-structure excitation, the average electrondensity inferred from C, Na, and K (assuming photoionizationequilibrium) seems to be independent of nH andnHT. While the electron density (ne) obtained fromthe ratio N(Ca I)/N(Ca II) is often significantly higher than the valuesderived from other elements, the patterns of relative nederived from different elements show both similarities and differencesfor different lines of sight-suggesting that additional processesbesides photoionization and radiative recombination commonly andsignificantly affect the ionization balance of heavy elements in diffuseinterstellar clouds. Such additional processes may also contribute tothe (apparently) larger than expected fractional ionizations(ne/nH) found for some lines of sight withindependent determinations of nH. In general, inclusion of``grain-assisted'' recombination does reduce the inferred ne,but it does not reconcile the ne estimated from differentelements; it may, however, suggest some dependence of ne onnH. The depletion of calcium may have a much weakerdependence on density than was suggested by earlier comparisons with CHand CN. Two appendices present similar high-resolution spectra of Fe Ifor a few stars and give a compilation of column density data for Ca I,Ca II, Fe I, and S I.
| Quantitative Spectral Analysis of Early B-Type Supergiants in the Sculptor Galaxy NGC 300
The spectra of two early B-type supergiant stars in the Sculptor spiralgalaxy NGC 300 are analyzed by means of non-LTE line-blanketed unifiedmodel atmospheres, aimed at determining their chemical composition andthe fundamental stellar and wind parameters. For the first time, adetailed chemical abundance pattern (He, C, N, O, Mg, and Si) isobtained for a B-type supergiant beyond the Local Group. The derivedstellar properties are consistent with those of other Local Group B-typesupergiants of similar types and metallicities. One of the stars shows anear solar metallicity, while the other one resembles more a SMC Bsupergiant. The effects of the lower metallicity can be detected in thederived wind momentum.Based on observations obtained at the ESO Very Large Telescope.
| The total-to-selective extinction ratio determined from near IR photometry of OB stars
The paper presents an extensive list of the total to selectiveextinction ratios R calculated from the infrared magnitudes of 597 O andB stars using the extrapolation method. The IR magnitudes of these starswere taken from the literature. The IR colour excesses are determinedwith the aid of "artificial standards" - Wegner (1994). The individualand mean values of total to selective extinction ratios R differ in mostcases from the average value R=3.10 +/-0.05 - Wegner (1993) in differentOB associations. The relation between total to selective extinctionratios R determined in this paper and those calculated using the "methodof variable extinction" and the Cardelli et al. (1989) formulae isdiscussed. The R values presented in this paper can be used to determineindividual absolute magnitudes of reddened OB stars with knowntrigonometric parallaxes.
| The Lack of Blue Supergiants in NGC 7419, a Red Supergiant-rich Galactic Open Cluster with Rapidly Rotating Stars
According to previous studies based on photometry alone, NGC 7419reveals a surprisingly low ratio of blue to red supergiants: only oneblue supergiant (BSG) along with a record number of five red supergiants(RSGs). However, for a cluster like NGC 7419 with solar metallicity, oneexpects twice as many BSGs as RSGs. To verify the small ratio of BSGs toRSGs, we have obtained spectroscopic observations of the seven mostluminous blue member stars using the 1.6 m telescope of the MountMégantic Observatory. (The RSGs have already been studiedspectroscopically.) To classify the stars, we have developed a systemespecially adapted for these heavily reddened stars in a spectral regionfrom 8400 to 8920 Å, near the hydrogen Paschen series limit. Thisclassification system is based on standard stars of known MK spectraltype extending over O9-B5 and all luminosity classes and is linkedthrough a grid of synthetic spectra to the atmospheric physicalparameters Teff and logg. We also include Be stars. Among theseven blue stars observed in NGC 7419, four have red spectra that aredominated by absorption lines and three by emission lines. The spectraltypes for the former are B2.5 II-III, B2.5 III, B0 III, and B4 III (e),while those for the latter are Be, B1 III-Ve, and Be, respectively. Theaverage heliocentric radial velocity of these stars is -66+/-6 kms-1, compatible with the value of -74+/-9 km s-1measured for the five RSG members. A distance of 1.7+/-0.4 kpc for thiscluster was estimated using the blue-star spectral types, in agreementwith the value of 2.3+/-0.3 kpc found by Beauchamp and coworkers, basedon isochrone fitting in the color-magnitude diagram. With no BSG starsdetected spectroscopically, we confirm the low number, in this caseabsence, of BSGs in this cluster. The high fraction of Be stars detectedby us and others among the bright blue member stars could be explainedby an average rotational velocity for the stars in NGC 7419, which issignificantly higher than in other clusters of similar age andmetallicity. Since higher stellar rotation rates shorten the BSG phase,we suggest that this explains why the evolved stars in NGC 7419 havebecome RSGs. Thus, NGC 7419 is an exceptional case, since high stellarrotation normally tends to occur at lower metallicity.
| The association of IRAS sources and 12CO emission in the outer Galaxy
We have revisited the question of the association of CO emission withIRAS sources in the outer Galaxy using data from the FCRAO Outer GalaxySurvey (OGS). The availability of a large-scale high-resolution COsurvey allows us to approach the question of IRAS-CO associations from anew direction - namely we examined all of the IRAS sources within theOGS region for associated molecular material. By investigating theassociation of molecular material with random lines of sight in the OGSregion we were able to construct a quantitative means to judge thelikelihood that any given IRAS-CO association is valid and todisentangle multiple emission components along the line of sight. Thepaper presents a list of all of the IRAS-CO associations in the OGSregion. We show that, within the OGS region, there is a significantincrease ( ~ 22%) in the number of probable star forming regions overprevious targeted CO surveys towards IRAS sources. As a demonstration ofthe utility of the IRAS-CO association table we present the results ofthree brief studies on candidate zone-of-avoidance galaxies with IRAScounterparts, far outer Galaxy CO clouds, and very bright CO clouds withno associated IRAS sources. We find that ~ 25% of such candidate ZOAGsare Galactic objects. We have discovered two new far outer Galaxystar-forming regions, and have discovered six bright molecular cloudsthat we believe are ideal targets for the investigation of the earlieststages of sequential star formation around HII regions. Finally, thispaper provides readers with the necessary data to compare othercatalogued data sets with the OGS data.Tables 1, 2 and A1 are only available in electronic form at the CDS viaanonymous ftp to\ cdsarc.u-strasbg.fr (130.79.128.5) or via\http://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/399/1083
| Revised Stellar Temperatures for Magellanic Cloud O Supergiants from Far Ultraviolet Spectroscopic Explorer and Very Large Telescope UV-Visual Echelle Spectrograph Spectroscopy
We have undertaken quantitative analysis of four LMC and SMC O4-9.7extreme supergiants using far-ultraviolet FUSE, ultraviolet IUE/HubbleSpace Telescope, and optical Very Large Telescope UV-Visual EchelleSpectrograph spectroscopy. Extended, non-LTE model atmospheres thatallow for the consistent treatment of line blanketing, developed byHillier & Miller, are used to analyze wind and photosphericsspectral features simultaneously. Using Hα to constrain mass-lossrates, He I-He II photospheric lines reveal stellar temperatures thatare systematically (5-7.5 kK) and substantially (15%-20%) lower thanpreviously derived from unblanketed, plane-parallel, non-LTEphotospheric studies. We have confidence in these revisions sincederived temperatures generally yield consistent fits across the entire912-7000 Å observed spectral range. In particular, we are able toresolve the UV-optical temperature discrepancy identified for AzV 232(O7 Iaf+) in the SMC by Fullerton and coworkers. Thetemperature and abundance sensitivity of far-ultraviolet, UV, andoptical lines are discussed. ``Of'' classification criteria are directlylinked to (strong) nitrogen enrichment (via N III λ4097) and(weak) carbon depletion (via C III λλ4647-4651), providingevidence for mixing of unprocessed and CNO-processed material at theirstellar surfaces. Oxygen abundances are more difficult to constrain,except via O II lines in the O9.7 supergiant, for which it is also foundto be somewhat depleted. Unfortunately, He/H is very difficult todetermine in individual O supergiants because of uncertainties inmicroturbulence and the atmospheric scale height. The effect of windclumping is also investigated, for which P V λλ1118-1128potentially provides a useful diagnostic in O star winds, unlessphosphorus can be independently demonstrated to be underabundantrelative to other heavy elements. Revised stellar properties affectexisting calibrations of (1) Lyman continuum photons-a factor of 2 lowerfor the O4 supergiant-and (2) kinetic energy released into the ISM by Osupergiants. Our results also have importance for the calibration of thewind momentum-luminosity relationship for OB stars, particularly sincethe stars studied here are among the visually brightest OB stars inexternal galaxies. Based on observations made with the NASA-CNES-CSA FarUltraviolet Spectroscopic Explorer. FUSE is operated for NASA by JohnsHopkins University under NASA contract NAS 5-32985. Also based in parton observations collected at the European Southern Observatory VeryLarge Telescopes in programs 65.H-0705 and 67.D-0238, plus archival dataobtained with the NASA-ESA Hubble Space Telescope and NASA-ESA-PPARCInternational Ultraviolet Explorer.
| Tomographic Separation of Composite Spectra. VIII. The Physical Properties of the Massive Compact Binary in the Triple Star System HD 36486 (δ Orionis A)
We present the first double-lined spectroscopic orbital elements for thecentral binary in the massive triple δ Orionis A. The solutionsare based on fits of cross-correlation functions of IUE high-dispersionUV spectra and He I λ6678 profiles. The orbital elements for theprimary agree well with previous results, and in particular, we confirmthe apsidal advance with a period of 224.5+/-4.5 yr. We also presenttomographic reconstructions of the primary and secondary stars' spectrathat confirm the O9.5 II classification of the primary and indicate aB0.5 III type for the secondary. The relative line strengths between thereconstructed spectra suggest magnitude differences ofΔm=-2.5log(Fs/Fp)=2.6+/-0.2 in the UV andΔm=2.5+/-0.3 at 6678 Å. The widths of the UVcross-correlation functions are used to estimate the projectedrotational velocities, Vsini=157+/-6 and 138+/-16 km s-1 forthe primary and secondary, respectively (which implies that both theprimary and the secondary rotate faster than the orbital motion). Weused the spectroscopic results to make a constrained fit of theHipparcos light curve of this eclipsing binary, and the model fits limitthe inclination to the range i=67deg-77deg. Thelower limit corresponds to a near Roche-filling configuration that hasan absolute magnitude that is consistent with the photometricallydetermined distance to Ori OB1b, the Orion Belt cluster in which δOri resides. The i=67deg solution results in masses ofMp=11.2 and Ms=5.6 Msolar, both ofwhich are substantially below the expected masses for stars of theirluminosity. The binary may have experienced a mass ratio reversal causedby case A Roche lobe overflow or the system may have suffered extensivemass loss through a binary interaction (perhaps during a common envelopephase) in which most of the primary's mass was lost from the systemrather than transferred to the secondary. We also made three-componentreconstructions to search for the presumed stationary spectrum of theclose visual companion δ Ori Ab (Hei 42 Ab). There is noindication of the spectral lines of this tertiary in the UV spectrum,but a broad and shallow feature is apparent in the reconstruction of HeI λ6678 indicative of an early B-type star. The tertiary may be arapid rotator (Vsini~300 km s-1) or a spectroscopic binary.
| On the Absolute Magnitudes of the O Stars
The conclusion published in 1992 by Garmany & Stencel from a studyof northern OB associations, that the absolute magnitudes of the O starsshow ``a large scatter ... intrinsic to the classification system,'' iscritically examined. It is found that the differences between theirderived absolute magnitudes of O stars and this author's 1973calibration exhibit large systematic effects in several associations,ranging from -0.74 to +1.02 mag with substantially smaller dispersions.Of course, when these results are combined, the scatter equals the fullrange of the systematic effects. To investigate the possibility ofdistance errors, the Garmany & Stencel B0-B2.5 stars in the sameassociations are subjected to the same analysis. The results for the Bstars show no significant systematic differences, eliminating errors inthe association distances derived by Garmany & Stencel from the Bstars as the source of the differences found for the O stars. It isnoteworthy that the dispersions in the absolute magnitudes of the Bstars within a given association are similar to or larger than those ofthe O stars. An examination of the distribution on the sky of the starsshows that the O and B stars in the discrepant associations aregenerally not colocated; such was already known to be the case for theimportant Perseus OB1 association. It is suggested that despite theirefforts to improve them, significant problems remain with theassociation memberships adopted by Garmany & Stencel; the relativelysmall dispersions of the O star absolute magnitudes even in thediscrepant cases indicate that they belong to different, usually moredistant associations near the lines of sight to the B associations withwhich they have been mistakenly connected. Several individual cases ofunrecognized multiple systems and classification errors are also foundin the Garmany & Stencel sample. It is concluded that the scatter inthe absolute magnitudes of the O stars is not as large as found byGarmany & Stencel, and not larger than that of the B stars.
|
提交文章
相关链接
提交链接
下列团体成员
|
观测天体数据
目录:
|
INHABITED SKY
