X-ray, UV and optical analysis of supergiants: ϵ Ori

Abstract

We present a multi-wavelength (X-ray to optical) analysis, based on non-local thermodynamic equilibrium photospheric+wind models, of the B0 Ia-supergiant: ǫ Ori.The aim is to test the consistency of physical parameters, such as the mass-loss rate and CNO abundances, derived from different spectral bands. The derived mass-losswhere f ∞ is the volume filling factor. However, the S iv λλ1062,1073 profiles are too strong in the models; to fit the observed profiles it is necessary to use f ∞ <0.01. This value is a factor of 5 to 10 lower than inferred from other diagnostics, and impliesṀ 1×10 −7 M ⊙ yr −1 . The discrepancy could be related to porosity-vorosity effects or a problem with the ionization of sulfur in the wind. To fit the UV profiles of N v and O vi it was necessary to include emission from an interclump medium with a density contrast (ρ cl /ρ ICM ) of ∼100. X-ray emission in H-He like and Fe L lines was modeled using four plasma components located within the wind. We derive plasma temperatures from 1×10 6 to 7×10 6 K, with lower temperatures starting in the outer regions (R 0 ∼3-6 R * ), and a hot component starting closer to the star (R 0 2.9 R * ). From X-ray line profiles we inferṀ < 4.9×10 −7 M ⊙ yr −1 . The X-ray spectrum ( 0.1 kev) yields an X-ray luminosity L X ∼ 2.0 × 10 −7 L bol , consistent with the superion line profiles. X-ray abundances are in agreement with those derived from the UV and optical analysis: ǫ Ori is slightly enhanced in nitrogen and depleted in carbon and oxygen, evidence for CNO processed material.

The First Candidate Colliding-wind Binary in M33

“…The systematic biases in observed mass loss rates depend on choice of line diagnostic, as ρ 2 diagnostics such as Hα are especially sensitive to wind "clumpiness" (Kudritzki & Puls 2000;Smith 2014). This highlights the need for a multiwavelength approach to empirically constraining massive star wind properties (e.g., Cohen et al 2011Cohen et al , 2014Puebla et al 2016), and therefore the effects of mass loss on massive star evolution and the surrounding environment.…”

Moving inhomogeneous envelopes of stars

“…Puls et al (2006) estimated that the clumping factor D is about four times larger in the line-forming region, compared to the radio-emitting region far away from the star. From their multiwavelength analysis of a B supergiant spectrum Puebla et al (2016) estimated that f ∞ , i.e. the smallest filling factor approached at large wind velocities (v(r) → v ∞ ), is 0.01, while Bouret et al (2012) derived f ∞ = 0.03 .. 0.06 for their sample of O stars.…”

“…However, all recent studies of X-ray spectra of O stars by means of sophisticated non-LTE models show that simple smooth wind models with constant filling factors are incapable to explain the multiwavelength spectroscopic observations. Either macroclumping or radially dependent X-ray filling factors have to be included in the models to reproduce the observations adequately (Hervé et al, 2012;Shenar et al, 2015;Rauw et al, 2015;Puebla et al, 2016).…”

“…The corresponding mass-loss rate is at least one order of magnitude lower than prescribed by the Vink et al (2001) recipe. (As an alternative Puebla et al (2016) discuss possible problems with their model wind ionization structure. )…”

Ultraviolet spectroscopy of the blue supergiant SBW1: the remarkably weak wind of a SN 1987A analogue

“…If low density inhibits cooling, we can make a qualitative prediction that N V and O VI may be observed (see e.g. Zsargo et al 2008;Bouret et al 2015;Puebla et al 2016), but a more detailed model beyond CMFGEN's current capabilities would be needed to derive a specific line strength for a quantitative mass-loss rate.…”