Discovery of X-Ray Emission from the Wolf-Rayet Star WR142 of Oxygen Subtype
Total Page:16
File Type:pdf, Size:1020Kb
Discovery of X-ray Emission from the Wolf-Rayet star WR 142 of oxygen subtype L. M. Oskinova, W.-R. Hamann, A. Feldmeier Institute for Physics and Astronomy, University Potsdam, 14476 Potsdam, Germany [email protected] R. Ignace Department of Physics and Astronomy, East Tennessee State University, Johnson City, TN 37614, USA Y.-H. Chu Department of Astronomy, University of Illinois, 1002 West Green Street, Urbana, IL 61801, USA ABSTRACT We report the discovery of weak yet hard X-ray emission from the Wolf-Rayet (WR) star WR 142 with the XMM-Newton X-ray telescope. Being of spectral subtype WO2, WR142 is a massive star in a very advanced evolutionary stage, short before its explosion as a supernova or γ-ray burst. This is the first detection of X-ray emission from a WO-type star. We rule out any serendipitous X-ray sources within ≈ 1′′ of WR142. WR142 has an X-ray luminosity of 30 −1 < −8 LX = 7 × 10 ergs , which constitutes only ∼10 of its bolometric luminosity. The hard X- ray spectrum suggests a plasma temperature of about 100MK. Commonly, X-ray emission from stellar winds is attributed to embedded shocks due to the intrinsic instability of the radiation driving. From qualitative considerations we conclude that this mechanism cannot account for the hardness of the observed radiation. There are no hints for a binary companion. Therefore the only remaining, albeit speculative explanation must refer to magnetic activity. Possibly related, WR 142 seems to rotate extremely fast, as indicated by the unusually round profiles of its optical emission lines. Our detection implies that the wind of WR 142 must be relatively transparent to X-rays, which can be due to strong wind ionization, wind clumping, or non-spherical geometry from rapid rotation. Subject headings: stars: winds, outflows — stars: Wolf-Rayet — stars: individual (WR 142) — X-rays: arXiv:0901.4553v1 [astro-ph.SR] 28 Jan 2009 stars 1. Introduction type stars, that represent the final evolutionary stage of a massive star prior to its explosion as Stars of the Wolf-Rayet (WR) type have a a type Ic supernova or γ-ray burst (Hirschi et al. highly peculiar chemical composition and very 2005). strong stellar winds. The WR spectra are sorted The WR stars continue to challenge our un- into three subclasses: WN, WC, and WO, ac- derstanding of line-driven winds. Schaerer (1996) cording to the dominance of nitrogen, carbon, pointed out the importance of the iron opacity for and oxygen emission lines, respectively. These the acceleration of WR winds. The first hydrody- spectral types correspond to evolutionary stages namical model for a WR wind was presented by (Conti et al. 1983). The largest C+O abundance Gr¨afener & Hamann (2005). In their simulation and the fastest stellar winds are observed in WO 1 the mass loss is initiated at high optical depth by cations of the XMM-Newton detection of WR142 the so-called “iron bump” in the opacity. It was are discussed in Sect. 4. thus demonstrated that WR-type winds can be driven by radiation pressure. 2. The WO-type Star WR142 It has long been known that line-driven winds WR142, also named Sand5 and St3, has a are subject to an instability that can lead to spectrum characteristic for the spectral subtype strong shocks (Lucy & White 1980). These WO2 (Barlow & Hummer 1982; Kingsburgh et al. shocks are thought to explain the X-ray emis- 1994). The optical spectrum of WR142 was dis- sions from O star winds, as predicted by time- cussed by Polcaro et al. (1997), who also noticed dependent hydrodynamic modeling (Owocki et al. some line variability. 1988; Feldmeier et al. 1997) and largely confirmed by observations (Kramer et al. 2003; Oskinova et al. Figure 1 shows the spectral energy distribution 2006; Zhekov & Palla 2007; Waldron & Cassinelli (SED) of WR 142 together with model calculated 2007). The growth of instability in WR winds was with the Potsdam Wolf-Rayet (PoWR) model at- investigated by Gayley & Owocki (1995). They mosphere code (Gr¨afener et al. 2002). Photomet- found that despite of damping effects due to the ric IR measurements are plotted together with multi-line scattering, the instability remains effec- our optical spectrum and the Spitzer IRS mid- tive. Therefore, X-ray emission from wind shocks IR spectrum. We have adopted parameters typ- could, in principle, be expected in WR winds, a ical for a WO star: stellar temperature T∗ = conjecture that has not been yet tested by time- 150kK, “transformed radius” (cf. Gr¨afener et al. dependent hydrodynamic simulations. 2002) Rt = 2R⊙, and a composition of 40% carbon, 30% oxygen and 30% helium (by mass). Significant observational effort has been made WR142 is assumed to be a member of the open to study the X-ray emission of WR stars. White & Long cluster Berkeley 87 at a distance of d = 1.23kpc (1986), Pollock (1987), Pollock et al. (1995), (Turner et al. 2006). Based on this preliminary Oskinova (2005) presented X-ray observations model, the fit of the photometric observations of Galactic WR stars. A survey of X-ray emis- requires an interstellar reddening of EB−V = sion from WR stars in the Magellanic Clouds 1.7mag and a stellar luminosity of log Lbol/L⊙ = was conducted by Guerrero & Chu (2008a,b). 5.35, implying a stellar radius of only R∗ = Ignace et al. (2000) and Oskinova (2005) demon- 0.6 R⊙. The corresponding mass-loss rate is about strated that X-ray properties of single WR stars −5.1 −1 10 M⊙yr for an adopted microclumping vol- differ from those of O stars. Whereas O stars dis- ume filling factor of 0.1. play a trend in which the ratio of the X-ray to the bolometric luminosity LX/Lbol has a typical value The adopted model does not provide an en- of 10−7 (Long & White 1980; Berghoefer et al. tirely satisfactory fit to the line spectra. For ex- vi 1997; Sana et al. 2006), this trend is not observed ample, the model does not match the huge O ˚ in the case of WR stars. emission at 3811, 3834 A, a problem also experi- enced by Crowther et al. (2000) while reproducing Observations with the XMM-Newton and Chan- these lines for Sand 2 with the cmfgen code. We dra X-ray telescopes established that some bona also fitted the SED shown in Fig. 1 with a model fide single WN stars are X-ray active (Skinner et al. that has a mass-loss rate lower by a factor of two, 2002a,b; Ignace et al. 2003; Oskinova 2005), while −5.4 −1 10 M⊙yr , and higher bolometric luminosity, others are apparently not (Oskinova 2005; Gosset et al. log Lbol/L⊙ = 5.65. This model fits the SED in 2005). Oskinova et al. (2003) found that no sin- Fig. 1 equally well. Figure 2 shows the radius of gle WC star had been conclusively detected at unity optical depth plotted as function of wave- X-ray energies, a result that continues to hold. length in the X-ray range for both models. The Among the WR subclasses, only the WO-type stronger wind is opaque even to hard X-rays, but stars have not been observed in X-rays so far. In the thinner wind is largely transparent, because this Letter we present the XMM-Newton observa- its higher ionization reduces the X-ray absorbing tions of the closest WO type star WR142. The ions. The same could happen in denser, but hotter star is introduced in Sect. 2, and its XMM-Newton models. Thus our ability to predict the influence observations are described in Sect. 3. The impli- 2 of wind photo-absorption is somewhat limited, ow- 12.0keV band are (1.89 ± 0.34) × 10−3 cs−1 for ing to ambiguities in the ionization state of metals EPIC MOS1+2 and (3.80 ± 0.84) × 10−3 cs−1 for and uncertainty in the mass-loss rate. EPIC PN cameras. Assuming a two-temperature The profiles of the emission lines in the spec- thermal plasma model (kT1 = 0.3 keV,kT2 = trum of WR 142 are very broad. Assuming that 10 keV), the observed X-ray flux of WR 142 is −14 −1 −2 the line widths correspond to the wind termi- FX = 4 ± 2 × 10 ergs cm . The redden- −1 nal velocity, the velocity of v∞ ≈ 5500kms ing towards WR 142 is known from the analysis would be deduced. However, the profile shapes of its optical spectrum, and the distance is known of almost all lines are much more round than the from its cluster membership. The X-ray luminos- 30 −1 roughly Gaussian shapes usually seen in WR spec- ity of WR 142 is thus LX ≈ 7 × 10 ergs , or tra. Such round profiles cannot be reproduced log LX/Lbol ≈−8. < ′′ by the standard models. It is tempting to repro- The angular resolution of XMM-Newton is ∼ 6 . duce these profiles by convolution with the semi- To exclude the potential confusion with a source ellipse for rotational broadening, albeit rotating in close vicinity of WR 142, we inspected opti- stellar winds certainly require a more sophisti- cal and infra-red images with higher angular res- cated treatment which has not been accomplished olution. According to the USNO-B1.0 catalog yet. If rotation is the cause for the round pro- (Monet et al. 2003), the closest object to WR142 files, the projected rotation speed must be com- is located 8′′ away. The optical monitor (OM) on parable to v∞, i.e.