Abstract by Jean-Claude Bouret
Early type O supergiants (from O4 to O7) present conspicuous wind profiles. Therefore, they are excellent targets to study wind clumping and its consequences on measured mass-loss rates. These stars are also
expected to be on a rather advanced evolutionary stage.
We have analyzed the spectra from FUV to optical of 8 galactic supergiants, using the NLTE wind code CMFGEN. We derived the effective temperatures, luminosities, surface gravities and surface abundances, mass loss rates, wind terminal velocities and filling factors associated to clumping.
The supergiants define a very clear evolutionary sequence, in terms of ages and masses, from younger and more massive to older stars with lower initial masses.
The surface chemical composition are typical of evolved O supergiant (nitrogen-rich, carbon and oxygen-poor).
The observed span for carbon and nitrogen mass-fractions are compatible with those expected from the models, for the measured stellar masses but the [N/C] ratios as a function of age is inconsistent with
the theoretical predictions for the four earliest (O4 spectral type) stars of the sample. The efficiency of rotational mixing as a function of age is questioned for these stars and other mechanism might be needed to explain the observed patterns. Mass-loss rates derived using clumped models range within factors up to three with theoretical mass-loss rates. This difference is actually very reasonable given the known uncertainty on both estimates. The corresponding filling-factors associated to small-scale clumping are 0.05$\pm 0.02$. Clumping is found to start close to the photosphere for all but three stars, two of which are fast rotators.
Talk Slides
pdf
What star is the Ne X profile from?
ReplyDeleteThis is the first I've seen HD 16691 as Onfp, i.e. with He II 4686 emission "reversed" or split. Although rapid rotation is evident at classification resolution, high resolution is required to see the 4686 detail in this spectrum. (Vsini 140 km/sec vs. 220 km/sec for Zeta Pup and Lambda Cep.) I note you don't fit this structure in any of these spectra, although you do fit H-alpha in the latter two but not HD 16691.
Are the rotation rates of these luminous evolved stars consistent with expected wind braking, or might they be spun-up binary interaction products, even mergers? Both Zeta Pup and Lambda Cep are runaways. At Lac Taureau, van Beveren suggested that Zeta Pup may be a merger.
C III 5696 is a selective emission line like N III 4640, which has an entirely different behavior in the HRD from C III 4650. It is a photospheric emission line. Cardona reportedly explained 5696 in an unpublished Colorado thesis--but the 4640 mechanisms of Mihalas & Hummer have very recently been shown to be inoperative by Puls, who has explained it by other effects in line-blanketed models. 4650 comes into emission only in (some) Galactic spectra at or near O5--the ((fc)) effect, Walborn et al. (2010); but also in some SMC and LMC spectra with a broader range of types. The relationship to Fe IV is interesting and could be relevant to the latter, i.e. metallicity effects.
The range of derived N/C abundances is huge, indicating that some of these stars are further along toward the WN stage, as is also indicated by the range of optical emission-line strengths and widths. Are those properties correlated?
It is important to sort out effects of initial mass, initial rotation, mass-loss rates, and (evolutionary) age, some of which are correlated and others independent...
Great progress! This is what we need more of, throughout the entire OB HRD...
@Nolan
ReplyDelete- The star is Zeta Ori (see Oskinova et al. 2004)
- I agree, rotation seems as important in HD 16691 as in Zeta Pup or Lambda Cep. At least that's what is suggested by the shapes of the emission lines in the optical spectrum. To me, the fit of Halpha in HD 16691 as it is now, is not much worse as in the 2 fast rotators. We need 2D RT to improve the fits. We have a paper in preparation on this.
- A quick check to a stellar evolution model with 60 Msol, best suited for Zeta Pup from this work, shows that after 3.5 Myr the rotation rate should be 108 km/s only. Since Zeta Pup is likely seen equator-on, the actual rotation rate is surely close to the measured vsini =220 km/s. Therefore we have a problem! On the other hand, the evolution model was computed assuming a mass-loss from Vink et al. (2000), which is a bit more than twice as high as the mass-loss rate we derive. The spin down should be less accordingly, meaning we should be OK. Will check that for the other objects, but then we can't say much because we don't know the real inclination to the line of sight.
Note that Stone (1979) suggested that HD 16691 is a runaway.
As for the merger interpretation for Zeta Pup, it's also based on the luminosity they derive, which is about 2x the luminosity I derive. And I don't think their carbon is right. It's much too high. Anyway, it's going to be hard to prove them right or wrong.
- The issue with CIII lines formation will be investigated into more details in a next step but I agreer that this is critical for further interpretation of the evolutionary status of the stars. Note that models do produce it as an emission in the 3 later type objects. Since the intensity is incorrect (but in Lambda Cep), it means we under-estimate the decay to the lower level of the transition. Thanks for pointing me to this thesis by Cardona, I'd like to see what his solution was.
- The two stars having the most extreme N/C ratios turn out to be
HD 190429A and HD 16691, both exhibiting some WN like features in the NIR (Conti et al. 1995)
The strengths and widths do not depend on abundances only, they are also very sensitive to Teff and logg (the density in the photosphere). The mass-loss rate + velocity field also play a role. All in all, the effects leading to the formation of the diagnostic optical lines of C and N abundances are really intricate. More work to sort them out....
Thanks for your comments.