MassiveMassive starsstars inin thethe X-rayX-ray lightlight Lidia OskinovaDublin The X-ray Universe 2014 01 Constellation Massive: Minitial > 8 M Orion 4 full moon Luminous: Lbol >10 L Hot: T > 20000 K in visual light eff Young: 1Myr - 100 Myr Final evolutionary stage: neutron star or black hole Only 0.4% of all stars 02 Constellation Orion Hot: Teff > 20000 K in X-rays Massive stars emit X-rays on nearly all stages of their lives NGC 602 in the SMC (Optical/X-ray:XMM-Newton & Chandra): 03 Born in a nebula Evolution of massive stars NGC 602 in the SMC (Optical/X-ray:XMM-Newton & Chandra): 04 Born in a nebula OB-type stars on the main sequence Evolution of massive stars NGC 602 in the SMC (Optical/X-ray:XMM-Newton & Chandra): 05 Evolve to Red Supergiant, LBV, or (and) Wolf-Rayet (WR) stage Born in a nebula OB-type stars on the main sequence Evolution of massive stars NGC 602 in the SMC (Optical/X-ray:XMM-Newton & Chandra): 06 Evolve to Red Supergiant, LBV, and Wolf-Rayet (WR) stage Explode as supernovae Born in a nebula OB-type stars on the main sequence Evolution of massive stars NGC 602 in the SMC (Optical/X-ray:XMM-Newton & Chandra): 07 Evolve to Red Supergiant, LBV, and Wolf-Rayet (WR) stage Explode as supernovae Born in a nebula End up as a NS or BH OB-type stars on the main sequence Evolution of massive stars 08 Star formation Massive star clusters Hot superbubbles Stellar winds and feedback X-ray diagnostics: O and WR stars Colliding wind biaraies Magnetic stars γ Cas-type stars Final evolutionary stages: HMXBs R. Willatt and ESA: 3 color XMM image of NGC 2264 Eagle Nebula (Herschel IR/X-ray XMM-Newton: ESA/Boulanger) 09 Massive star formation X-rays: Herbig-Haro objects (Lopez-Santiago+ 2013; Bonito+ 2010, 2011; Schneider+ 2011 ) X-rays: massive protostars (Pravdo+ 2009; Anderson+ 2011; Munar-Adrover+ 2013) north and east fields in λ Orionis cluster (XMM-Newton, Barrado+ 2011) 10 X-rays: studies of low-mass star populations in vicinity of massive stars X-ray T Tau (Jeffries+ 2009; Naylor+ 2009; Franciosini+ 2011) Do low-mass stars form earlier or later than the massive ones? Clusters of massive stars 11 Most massive stars are in the most massive young clusters -7 For an OB star LX ≈ 10 Lbol (Pallavicini+ 1981; Berghoefer+ 1997;Sana+ 2006; Naze 2009...) OB2 massive star association 12 XMM-Newton (Rauw+ 2011) O, WR, LBV, SNR in the same cluster (W33, Messineo+ 2011) X-rays: massive stars in the Galactic plane (Anderson+ 2011, Warwick+ 2011) Million-Degree Plasma Pervading the Orion Nebula 13 GuÈdel+ 2008 Stellar winds and SNe feedback: superbubbles & cluster winds Townsley+ 2011; Rodriguez-Gonzalez+ 2011; Arthur 2012, Silich+ 2013; Roger & Pittard 2014; Krause+ 2014 Diffuse X-rays are well observed in the LMC 14 SNR and a superbubble in the HII region N206 in the LMC (Kavanagh+ 2012) Type of equilibrium CIE vs NEI? The heating and cooling processes? Does steller feedback provide enough energy to heat superbubbles? 16 Line-driven stellar winds (Castor, Abbott & Klein 1975, ``CAK'') Atmosphere transparent in continuum, opaque in many spectral lines UV photon comes from radial direction but scattered isotropically Momentum transfer Acceleration Velocity Acceleration 17a Theories of stellar wind instability Lucy & Solomon (1970): stellar winds are radiatively driven; this mechanism is unstable. Radiative Hydrodynamics: Wind shocks (Owocki+ 1983, Feldmeier+ 1997) X-ray 3000 2000 1000 Velocity [km/s] 8 7 6 Log(T [K]) 5 4 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Feldmeier+ 1997 Distance (R ) 17 Mass removal by wind drives the evolution Teff / kK 100 50 20 10 5 7 M˙ is a critical factor of stellar evolution 6 WR OB LBV L / 40 M L log 5 ZAMS −6 −5 −1 M˙ CAK ≈ 10 − 10 M⊙yr 4 5 4 log Teff / K 18 Stellar wind feedback Combined Spitzer (red+green) and X-ray EPIC (blue) image of the SFR ON 2 Mechanical energy of a wind: up to 1038 erg/s Stellar life-time: few Myr A star deposits ~ 1052 erg during its life-time! Comparable to to SN output WR 142 (WO-type) Oskinova+ 2010 19 Stellar Winds Ubiquitous in hot non-degenerate stars Strongly affect the ISM Spitzer IRAC image of ζ Oph (Hubrig, Oskinova, SchoÈller 2011) 20 Stellar Winds Ubiquitous in hot non-degenerate stars Strongly affect the ISM AE Aur: First detection of non-thermal X-rays from a bow shock produced by a runaway star (XMM-Newton EPIC, Lopez-Santiago+ 2012) Are bow shocks and wind-blown bubbles sites of particle acceleration? Schulz+ 2014; De Becker 2014; del Valle & Romero 2013, 2014; Benaglia+ 2012 Spitzer IRAC image of ζ Oph (Hubrig, Oskinova, SchoÈller 2011) Wind Blown Bubble S308 around the WR star EZ CMa 21 Toala+ 2012 (Soft X-rays 1MK, limb brighten morphology) van Marle & Keppens 2012; Dwarkadas+ 2007; Toala & Arthur 2011; Zhekov 2014 23 X-ray diagnostics of stellar winds: RGS SiXIII MgXII MgXI NeX NeX NeIX FeXVIII FeXVII OVIII FeXVII FeXVII OVIII OVII NVII NVI CVI 0.2 XMM RGS ζ Pup O4I 0.1 0.0 0.3 XMM RGS ζ Ori 0.2 O9.7I 0.1 0.0 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 o Oskinova etal. (2006) Wavelength (A) First high-resolution X-ray spectra of O-stars - 2001 (Chandra and XMM-Newton) Emission of a hot 106 K plasma in stellar wind 24 Common X-ray diagnostics: line shapes Line shapes Emission: opticaly thin lines Standard model vs. observed line Broad: wind velocity Blueshifted: wind opacity O VIII ζ Ori 0.10 ) o Marrying the model and observations: 0.05 Severely reduced wind density Flux (counts/sec/A Wind inhomogeneity λ 0 X-ray production far out in the wind 0.00 18.8 18.9 19.0 19.1 o Wavelength (A) 25 Common X-ray diagnostics: lines of He-like ions Ratio of forbidden to intercombination line flux depends on the UV radiation field resonance UV flux dilutes with r-2 intercombination f/i ratio estimator for UV distance where the forbidden hot gas is located Requires knowledge of stellar UV field Gabriel & Jordan 1969 Analysis of 0.7Ms RGS spectrum of O-supergiant ζ Puppis 0.6 NLTE wind model 0.4 APEC 0.2 0 0.1 0 -0.1 10 15 20 0.6 0.4 Hot gas spread out 0.2 Reasonable MÇ 0 T distribution 0.1 X 0 -0.1 20 25 30 35 (similar Leutenegger+ 2013) Herve+ 2013 27 Common X-ray diagnostics: time variability 10 years of EPIC observations: O-supergiant ζ Pup: the quality of the data will not be surpassed for decades. No short term (~hours) variability is detected Variability on time-scale of days is detected Plausibly associated with rotation Berghoefer+ 1996; Oskinova+ 2001; Naze+ 2013; Massa+ 2014; Pollock+ 2013; Ignace etal. '13 Similar time-scale: UV & Hα lines Corotating intraction regions UV MEGA compaign; Mullan '84; Cranmer & Owocki '96; Lobel+ 2008 28 Emerging concept of stellar wind structure Small scale wind clumping triggered by sub-photospheric convection (Cantiello & Braithwaite 2011) Large scale structures associated with rotation (Mullan+ 1984; ...;Ramiaramanantsoa+ 2014) Pockets of hot 1-10 MK plasma permeated with the cool 10kK wind 29 First high-resolution X-ray spectrum of a single WR Star 6400 RGS XMM-Newton 5600 Fe XVII spectrum of WR6 Ne X 4800 Ne IX Fe XVII 4000 N VII -1 Mg XI o 3200 Fe XVII Al XII N VI Counts A 2400 N VII N VII 1600 800 0 8 10 12 14 16 18 20 22 24 26 28 30 o Oskinova+ 2012 Wavelength (A) X-ray spectra of WN stars are harder than O-stars (Skinner+ 2010, 12) Lines are broad, strongly blueshifted, forbidden lines are detected EPIC: X-ray variability (Ignace+ 2013) - corotating interaction regions X-rays + IR a new way to find new WR stars (Mauerhan+ 2009). X-ray bright WR stars are binaries (Pollock + 2009, Pandey+ 2014;...) 30 Colliding Wind Binaries >80% of massive stars are binary or multiple (Chini+ 2012; Sana+ 2012) Russell+ 2013 CWB O3.5If+O6V Theory: Stevens et al.; Walder & Follini, Antokhin et al; Pittard et al.; Russell et al; Madura et al; Parkin et al.; 31 Colliding Wind Binaries: observations Large XMM compaigns: η Car, WR 140, V444 Cyg Comparison to models key Example of X-ray light curve: Bhatt+ 2009 information about stellar properties CWB sources of non-thermal radiation (de Becker 2011, 13; Ohm+ 2011; van Loo+ 2008, Blomme+ 2010) WR and LBV binaries are significantly X-ray brighter than single stars. O+OB binaries are not. Why? Cazorla+ 2014; Corcoran+; Williams+ 2011; Grunhut+ 2013; Zhekov+ 2012; Gosset+ 2012; Liege group;... 34 Magnetic massive stars <10% of massive stars: large scale magnetc fields (Grunhut+ 2012) X-rays reveal magnetically confined stellar winds Wade+ 2012; Naze+ 2010; Oskinova+ 2012; Ignace+ 2013; Petit+ 2014; ud Doula+ 2014 35 X-ray pulsations in a non-degenerate massive star ξ1 CMa: stellar radial pulsations X-ray pulses in phase with optical Ne X Ne IX Fe XVII Fe XVII O VIII OVII N VII N VI CVI X-ray Image 0.04 -1 o A -1 0.03 MINIMUM NVII 0.8 0.02 0.01 Normalized counts s 0.7 0.00 12 14 16 18 20 22 24 26 28 30 32 34 o X-ray count rate Wavelength [A] XMM-Newton observations: 0 5 10 15 20 25 30 light curves Time [hr] phase-resolved spectroscopy Oskinova+ 2014 (Nature Com.) 36 X-ray pulsations in a non-degenerate massive star ξ1 CMa: stellar radial pulsations X-ray pulses in phase with optical Ne X Ne IX Fe XVII Fe XVII O VIII OVII N VII N VI CVI X-ray Image 0.04 -1 o A -1 0.03 MAXIMUM NVII 0.8 0.02 0.01 Normalized counts s 0.7 0.00 12 14 16 18 20 22 24 26 28 30 32 34 o X-ray count rate Wavelength [A] XMM-Newton observations: 0 5 10 15 20 25 30 light curves Time [hr] phase-resolved spectroscopy Oskinova+ 2014 (Nature Com.) 37 γ Cassiopeiae class of X-ray sources γ Cas - Be-type star (B0.5 IVe), very fast rotator, binary Variable hard X-ray emission: magnetic or accretion ? Ten γ Cas-class objects: -6 LX /Lbol =10 (<< BeXRB) Variable X-ray light curve Thermal spectrum with kT=12-14 keV Smith+ 2006, 2012a,b; Motch+ 2007, 2010; Lopes de Oliveira+ 2006, 2011; Rakowski+ 2006; Safi-Harb+ 2006; Rauw+ 2013; Torrejon+ 2013 High-mass X-ray binaries 38 Accretion of OB star matter by a BH or a NS Massive star evolution, structure, winds are parts of this story Bozzo+ 2011; Sidoli+ 2012; Manousakis+ 2012; Walter & Zurita Heras 2007; Doroshenko+ 2011; Ducci+ 2009; FuÈrst+ 2010; Negueruela+ 2008; Martinez- Nunez+ 2014; Duro+ 2011; ..