A bit of background We know this system since 1966 ApJ 150, 57 X-RAY INTENSITIES AND SPECTRA FROM SEVERAL COSMIC SOURCES* G. Chodil, Hans Mark, R. Rodrigues, F. D. Seward, and C. D. Swift Lawrence Radiation Laboratory, University of California, Livermore Received March 22, 1967 ABSTRACT This paper reports the results of X-ray spectrum and intensity measurements for several cosmic X-ray sources. Two flights were conducted, one from Kauai, Hawaii on July 28, 1%6, and the other from Johnston Atoll on September 20, 1966. Proportional counters with anticoincidence shields to eliminate charged-particle background counts were used to detect the X-rays. Four known sources were observed: Sco XR-1, Tau XR-1, Cyg XR-1, and Cyg XR-2. Total intensity determinations were made for all of these sources, and spectra were obtainec] for Sco XR-1 and Cyg XR-2. A search was made for X-rays from the Large and Small Magellanic Clouds, but no X-rays above background were found in that region of the sky. An upper limit of the X-ray intensity from the Magellanic Clouds has been determined from these data. Ajve^k^^ra^swircejio^ 1200 1000 I 800 < x u 600 m > Z 400 5 O U 200 "T I T 1 I r SMC---v -. - • LMC> * * VEL XR-1 J_ 1 ± CYG XR-2 ' ■ ' ■ 20 4cT 60 80 I 100 120 1140 160 1180 264° SOUTH EAST NORTH 280° CHANNEL .. in other words, since the early times of X-ray astronomy Sounding BaMoons rockets is ♦I GINGA SWIFT Uhuru HEAO-1 ve a bti M • || v -h -1- —i— Data from almost 50 years ago can be found in archives! KuSTAR eROSITA Athena 4? I Sun Sco X-l t ft f GRB Pu sars AGN | Transients X-ray bursts I First all-sky survey SNR Clusters QPOs CVs blazars ULXs TDEs Galactic ridge Imaging telescope Microquasars CCDs GRB fo Black hole spin , 150000 objects X ray background Comets Crab nebula Iron lines Cyclotron lines |50|atcd neutron stars Stars Jets QPOs from AGN IMBHs? Cluster mergers Cosmology High-z quasars Picture © Ole König My acquaintance with Vela X-1 is not quote as old, but still Hard X-Ray Observations of Vela X-1 and A0535+26 With HEXE: Discovery of Cyclotron Lines ^ Q Q 2 E. Kendziorra1, B. Mony1, P. Kretschmar1, M. Maisack4, R. Staubert1, S. Dobereiner2, J. Englhauser2, W. Pietsch2, C. Reppin2, J. Trumper2, V. E fremov3, s. Kaniovsky3, R. sunyaev3 The Astrophysical Journal Supplement Series, 92:448-450, 1994 June © 1994, The American Astronomical Society. All rijhis reserved. Printed in U.S.A. 1994 VARIABLE SOFT X-RAY ABSORPTION AND EXCESS OF VELA X-1 H. C. Pan, '■' P. Kretschmar,2 G. K. Skinner, 1 E. Kendziorra,2 R. A. Sunyaev,3 and K. N. Borozdin3 Received 1993 May 4; accepted 1993 August 12 ASTRONOMY k ASTROPHYSICS SUPPLEMENT SERIES Astern. Astrophys. Suppl. Ser. 120, 175-178 (1996) DECEMBER III 1996, PAGE 175 1996 Absorption features in the hard X-ray spectra of PSR A 0535+26 and Vela X-1 Astron. Astrophys. 325, 623-630 (1997) ASTRONOMY AND ASTROPHYSICS 1997 Phase resolved X-ray spectra of Vela X-1 P. Kretschmar1 \ H. C. Pan2, E. Kendziorra1, M. Maisack1, R. Staubert1, G. K. Skinner2, W. Pietsch', J. Trumper1, V. Efremov4, and R. Sunyaev4 Astron. Astrophys. 341. 141-150 (1999) Vela X-1 as seen by RXTE ASTRONOMY AND ASTROPHYSICS 1999 I. Kreykenbohm1, P. Kretschmar' J. Wilms1, R. Staubert1, E. Kendziorra1, D.E. Gruber2, W.A. Heindl2, and R.E. Rothschild2 A&A 563, A70 (2014) DOI: 10.1051/0004-6361/201322404 ©ESO 2014 2014 Astronomy Astrophysics The accretion environment in Vela X-1 during a flaring period using XMM-Newton S. Martinez-Nunez', J. M. Torrejon' 2, M. Kiihnel3, P. Kretschmar4, M. Stuhlinger4, J. J. Rodes-Roca12, F. Fiirst5, I. Kreykenbohm3, A. Martin-Carrillo6, A. M. T. Pollock4, and J. Wilms3 A&A 608, AI43 (2017) DOI: 10.1051/0004-6361/201731843 © ESO 2017 2017 Astronomy Astrophysics The clumpy absorber in the high-mass X-ray binary Vela X-1 V. Grinberg', N. Hell2, I. El Mellah3, J. Neusen4, A. A. C. Sander1, M. Leutenegger6'7, F. Fürst8, D. P. Huenemoerder9, P. Kretschmar8, M. KUhnel10, S. Martinez-Nünez", S. Niu (4^)'°'l2, K. Pottschmidt6-N. S. Schulz9, J. Wilms1", and M. A. Nowak9 Mem. S.A.II. Vol. 90, 221 ©SAIt 2019 2019 Vela X-1 as a laboratory for accretion in high-mass X-ray binaries P. Kretschmar1, S. Martinez-Nünez2, F. Fürst1, V. Grinberg3, M. Lomaeva4, I. El Mellah5, A. Manousakis6, A. A. C. Sander7, N. Degenaar8, and J. van den Eijnden8 A&A 641, A144 (2020) https://doi.org/10.1051/0004-6361/202037807 © ESO 2020 2020 Astronomy Astrophysics High-resolution X-ray spectroscopy of the stellar wind in Vela X-1 during a flare M. Lomaeva'2, V. Grinberg3, M. Guainazzi2, N. Hell4, S. Bianchis, M. Bissinger ne Ktihnel6, F. Fürst7, P. Kretschmar7, M. Martinez-ChicharroK, S. Martinez-Nünez9, and J. M. Torrejon8 It seemed like a straightforward idea - back in February 2017 ISSI TEAM Feb. 2016 & 2017 A Comprehensive View of Stellar Winds in Massive X-ray Binaries A COLLABORATION TO FURTHER OUR UNDERSTANDING OF THE INTERACTION BETWEEN THE COMPANION, ITS WIND, AND THE COMPACT OBJECT IN MASSIVE X-RAY BINARIES. We talked so much about Vela X-1. But different people use different assumptions based on different published result Two unequal partners - a blue supergiant and a neutron star X-ray Binaries: a lot of physics on many different scales 0.1 1 10 Earth's atmosphere 1000 Size (km) 10000 X Athena @ L2 © Felix Fürst Essential length scales in the Vela X-1 system • Roche lobe: bound to donor star. • Bondi-Hoyle-Lyttleton: gravitational capture from wind. • Ionization: X-rays may ionize inflowing gas. • Corotation: Keplerian orbit at angular speed of neutron star rotation. • Magnetosphere: neutron stars magnetic field dominates. • Circularization: Keplerian orbit with angular momentum of accreted flow. length scale [cm] A simplified model picture of the system Slightly eccentric, not quite circular orbit. Supergiant somewhat distorted towards neutron star focused wind Accretion wake from hydrodynamical interaction. Photoionization wake from stalled wind close to neutron star. 250 200 150 100 X 50 0) 0 m -50 -100 -150 -200 -250 i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i i i i i accretion wake J_I_I_L J_I_I_I_I_I_I_I_I I I I I I I I- -400 -300 -200 -100 0 Distance (lt-s) 100 Diagnostics Different diagnostics (obs. & models) covering different scales UVOIR spectroscopy J f X-ray line spectroscopy Wind Pulse Period Structure V Evolution Flow near Magnetosphere Pulse Profiles Accretion Column Continuum Cyclotron Lines spectroscopy J \ Different diagnostics (obs. & models) covering different scales Overall flux variations Accretion Column Continuum Cyclotron Lines spectroscopy J \ The Vela X-1 system has been also detected in the radio! Recent result (van den Eijnden, et al., in prep.): • Highly significant (-100 |jJy) radio detection of Vela X-1 with ATCA. • Observation done by chance at mid eclipse. More observations done recently. • Flat radio spectrum, like for a compact jet. • Cannot exclude donor star as radio source yet, but this would also be interesting. Line spectroscopy to derive wind parameters 37 _ 1 fi«7 6.31 fT" H r ^1 T X-ray fluorescence lines can yield information on state of matter Plasmas of different densities, temperatures, velocities, and ionization states reprocess the radiation from the neutron star, imprinting characteristic features. But interpretation complex and different model codes can yield quite different plasma parameters (Lomaeva et al. 2020). 0.22 orbital phasi c 10 The flux can change from one pulse to next M o Inoue et al. (1984) Tenma 1983 5:46:44 5:54:22 6:01:59 March 12, 1983 c 3 C 0 u 1 c =i c U Börner et al. (1984) 7e/7ma 1983 500 o 400 to *|300 S 200 o ■■-> 100 -I-1-r T-1 I ■a; 800 1200 Time (sec) : [b) 1600 2000 8:30 8:40 8:50 2003 December 8 Kreykenbohm et al. (2008) INTEGRAL ISGRI 2003 Pulse-averaged flux shows log-normal distribution Fürst et al. (2010): Bins of 283.5 s (-average over pulse), filtered to avoid eclipse. "Shock fronts and turbulence breaking up clumps can transfer any given distribution into a log-normal like distribution." 1000 Countrate [cps] (20-60keV) Modelling efforts Modelling the right amount of variation from clumps can be difficult 'Naive' 1-D modelling of accreting clumps (shells) by BHL accretion over-predicts observed variability strongly. Simulated clump distribution gives more realistic light curve (Ducci et al. 2009), but clump sizes required uncomfortably large. _ a _38 (ft r,. 32 o ^ 30 — 38 (A J? 36 u ~ 34 ' J 1 Lt = —"i .-! -a=2.5R» 30 - Ij = 7x 10'6crg s 1 .a=2R. i i i 'Realistic' clump model for Vela X-1 under-predicts observed absorption variations, if assumed to be caused by clumps (Grinberg et al. 2017) i 1 i 1 i 1—1—1—1—.....i 1 i 1 i observed distribution" 1 i 1 i 1 i ■ i i _i_i_i_i_i_ • • • The X-ray radiation may self-regulate the wind 1.5 Photoionization of the wind destroys ions responsible for acceleration bubble of stagnant flow around neutron star Krticka et al. (2012, 2015, 2016, 2018): photionization may lead to self regulated winds with HMXB close to forbidden area, where X-rays would fully stop wind. Radial solution of wind equation. Latest studies include wind clumping (favouring recombination) 39 ,.' 38 £ 37 > CO X 34 33 32 31 30 —•-1-■-1— wind distruption forbidden area V662*Cas * x x x*at 'g X-1 • xv x x x xxx*- TT -Vela X-1 IGR J18029-2016 * •-> IGR J18483-0311 x xx XXX X+Ht- IGR J18450-0435 x x + ■++«- ,IGR J16207-5129 IGR J 164^9-4511 x + + + +H- XTEX1739-302 + + + + +H- + + + + + + + + 1.0 0.5 Q 0.0 -0.5 -1.0 0.2 0.4 0.6 0.8 1 1.2 optical depth parameter fx 1.4 1.6 -1.5 •1.0 -0.5 0.0 0.5 1 x/D 700 0 1.5 2.0 600 500 400 r 3oo 200 100 X-rays no X-rays 1 1.5 2 radius in stellar 2.5 radii r/ff* Wind driving can become very complex Sander et al. (2018): Simulations of wind acceleration using updated Potsdam Wolf-Rayet (PoWR) code including impact of X-rays (but in 1D treatment). Hydrodynamically self-consistent solution for wind structure, accounting for 16 elements and -5000 lines in calculations. Different ions dominate acceleration at different distances. Wind velocity profile differs strongly from a "beta law". Wind speed very low in inner zone --► impact on accretion (see later slides). Weak X-rays can increase wind driving in outer zone and terminal velocity. Strong X-rays disrupt wind. i—i—i—i—I—i—i—i—i—I—i—i—i—i—I—i i i i moderate X-rays 0.0 0.5 log(r/Ä. - 1) v(r) = v Hydrodynamic models also predict variations 10 8-. 4- 2- -10 if it? 4° j4e O 2D hydrodynamic models by Blondin et al. (1990, 1991, 1994) later picked up and enhanced by Manousakis et al. (2011, 2012, 2013, 2014) also yield clear variations. Radiative transfer not handled in detail, relying on critical ionization parameter as "on-off" switch. Wind clumping not (yet) included. Orbit approximated as circular. See also Cechura &Hadrava(2015) for Cyg X-1. -5 Manousakis PhD 201 trailing accretion structures 10 2D/3D Models of local accretion remain a challenge • El Mellah et al. (2018): 3D hydrodynamic simulations of the wind in the vicinity of the accretor. Several spatial orders of magnitude, down to the NS magnetosphere within spherical stretched adaptive mesh. • Inflow 'extruded' from realistic 2D simulation of clump formation close to star (Sundqvist et al.2017). The knowns and unknowns of the Vela X-1 system Distance and origin of this runaway HMXB system -1—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i- Hiltneret al. (1972) $ 2000 h Nandy et al. (1975) Zuiderwijk et al. (1974) Conti (1978) Sadakane et al. (1985) Vanbeveren (1993) Kaperet al. (1997) Vela 0B1) Vela OB1 with Gaia DR2 Coleiro & Chaty (2013) Giménez-García et al. (2016) A Bailer-Jones * etal. (2018) this work —■— J_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_L Update with Gaia EDR3 results! 1.5 2 2.5 3 estimated distance (kpc) -i—i—i—i—J—i—i—i—i—J—i—r 1.5 ]—i—i—i—i—J—i—i—i—r (Gaia DR2) -40- -42 -44 cd cd I "46 CO _c Ü cd Q -48 -50 ~i-1-r HIC *t Sahu etal. 1992 * HD77581 i • i ~i—i—1~- ~i-1-r~ n-1-r~ Maíz-Apellániz et al. 2018 t3 CD CD £2! I" co ^! --I 3i _l_I_l_ < cu —^ cu 3 cu Q-N CD CD ro o co _i_i_i_ V \* VelOBI ^ N c^ \ _i_i_i_ _i_i_i_ _i_i_i_ j_v. 138 136 134 132 130 Right Ascension (deg) 128 126 Orbital parameters The orbital period is extremely well known (8.964357±0.000029 d), due to eclipses, but slight tension between last two determinations. Small but significant difference between zero points of orbital phase. Eccentricity very well determined from X-ray pulse timing (0.0898±0.0012). Inclination (73-90 deg) is major unknown factor for orbit scale --► impacts mass & radius determination! 2.0 1.5 Ifl 1.0 3 re 0.5 L. 0.0 re in -0.5 > -1.0 -1.5 -2.0 a= 1.8R. 1 = 80.0° e = 0.0898 O = 0.0" u= 152.59° -2-10 1 x / stellar radius 1 1 1 1 1 1 1 1 1 1 1 . 1 Hutchings (1974) Watson & Griffiths (1^77) Ogelman (1977) van Paradijs (1977) Rappaport (1980)^ Nagase (1983) Nagase (1984) Sato (1986)_^_ Boynton (1986) „ Deeter (1987a)^ i i i i I i i i i I i i i Deeter (1987a) ^ Deeter (1987b)^ Bildsten (1997) Kreykenbohm (2008)^ Deeter (1987b)_^ Bildsten (1997) Kreykenbohm (2008)^ Falanga (2015) ecl.^ Falanga (2015) Trt2 Falanga (2015) eel Falanga (2015) Trt2^ + I , . . i T . . ... I .... i*. . . 8.96430 8.96440 i , , , i 8.962 8.964 8.966 Orbital period [d] 8.968 8.970 I I I I 1 1 1 I -1-1-1-1-1-1-1-1-1-1-1-1 Deeter (1987b)^ Bildsten (1997)^ Kreykenbohm (2008)^ . . i , , , , i Falanga (2015) -e^0 , , , , i , , . , i . . 0.17 0.18 0.19 0.2 predicted orbital phase on 2021-Jan-01T00:00 (MJD 59215) New spectral classification of mass donor Different spectral classifications listed in SIMBAD: BO to B0.5 and in luminosity class from lb to la. New spectral classification based on Galactic O-Star Spectroscopic Survey (GOSSS) and Gaia DR2 distance: spectral type & class: B0.2 la Stellar parameters to be redone with Gaia EDR3 distances. Ongoing, results maybe this Friday. X (Angstroms) 10 11 12 5-103 Update with Gaia EDR3 results! 1-10" Iff1 2-104 1-10= i-io-'• 5-10- 2-10-'• 1-10 ' 5-10- 2-10 1 -10 ' 5-10- 1 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 3.0 2.5 2.0 1.5 1.0 0.5 MX (urn-1) Blowing in the wind - at very different velocities Terminal wind speeds and velocity profiles derived very differently over the years. Major impact on accretion flow close to neutron star! o o CD > c 1750- 1500 1250- 1000- 750 500- Dupree et al. (1980) ^0. orbital velocity Watanabe et al. (2006) ionized wind -1.5 -0.5 0 log (r/Rstar-1) Between wind and disk accretion? A variable mass transfer? Filling factor (ratio between stellar and Roche Lobe radius) varying along orbit due to eccentricity and often >1. Either the inclination, and thus mass ratio between giant star and neutron star is on upper end of assumed distribution. Or have intermittent Roche-Lobe overflow at some orbital phases. Mass transfer may be more complicated than basic wind acceleration. 2.0 1.5 in 1.0 3 ID 0.5 c 1. 25 0.0 "3 *> in -0.5 '— > -1.0 -1.5 -2.0 Quaintrell+03 Rawls+11 Falanga+15 0.2 0.4 0.6 0.8 Orbital phase (w.r.t. mid-eclipse) -10 1 x / stellar radius How random are the torques on the neutron star? 283.6 - 283.4 - co ■° 283.2 CD Q_ CD _CO t£ 283.0 282.8 - 282.6 -i—i—i—r i—i—i—r ~i—i—i—r ~i—i—i—r ~i—i—i—r i—i—i—r □ Long-term pulse period evolution usually described as random walk. Caveat: period changes are 'measured' between data points at least days apart, much longer time scales than flux variations. A convincing theory for wind-accreting systems is lacking. Some spectral evidence for temporary accretion disk formation (Liao et al. 2020). Vela X-1 is not in spin equilibrium. The corotation radius is much larger than the magneto-spheric radius! BATSE Fermi GBM ° B _L J_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_L _L _L _L 42500 45000 47500 50000 52500 55000 57500 MJD Probably a massive neutron star • Vela X-1 is often quoted as example of massive (clearly > 1.4 M®, maybe > 2 M®) neutron star. • Full picture, taking into account inclination uncertainty is less clear, but leaning towards heavy solutions. • Mostly in mass range where radius is almost stable according to theoretical equations of state (EOS) '-► probable radius 11-12.5 km. • For highest possible masses Maselli et al (202°) interesting area of EOS would be sampled. Joss & Rappaport 1984 (i>73) -1-1-1-f~ I-1-1-1- c Bulik et al. (1995) van Kerkwijk et al. 1995 (#1) ^van Kerkwijk et al. 1995 (#2) "•-van Kerkwijk et al. 1995 (#3) Stickland etal. 1997 Stickland etal. 1997 (i=90) Barzivetal. (2001) (i>73) Rawls et al. 2011, analytical (i=83.6±3.1) _Rawls et al. 2011, numerical (i=78.8±1.2) Falanga et al. 2015 (i=72.8±0.4) _ neutron star mass Pulse profiles should allow to disentangle the emission geometry Vi n n 0.5 1.0 1.5 PULSE PHASE .7-2 keV 2-6 keV 6-11 keV 11-17 keV O 17-22 keV §5 cd E 'cd _c 22-28 keV £= er The pulse profile is complex at lower energies and overall usually rather stable. Doroshenko et al. (2011) found changed pulse pattern in "off-state". In principle able to derive information on emission geometry. But complicated analysis if general relativity and realistic emission geometries are taken into account! Still quite a bit of work on models and comparison. Falkneratal. (2016) Cyclotron lines maybe more puzzling than enlightening • Cyclotron Resonant Scattering Features found in 36 sources so far (Staubert et al. 2019). • Most direct measure of magnetic field strength. Variations in observed centre energy changes in (height of) emission region. • Fürst et al. (2014): harmonic line varies with luminosity. No clear picture for fundamental. • Ji et al. (2019, submitted): possible long-term trend in energy (Swift BAT). ^ Will need improved accretion column models to better interpret the data. Time (MJD) Further progress More observational data is available and being studied Major observational X-ray campaign in January 2019 motivated by planned X-Calibur balloon observations (polarisation). The balloon failed early, but the X-ray data is being analysed. Radio observations at 4 orbital phases end Sep 2020. Could still use: o o o o 250 200 150 100 - 50 -50 -100 -150 -200 -250 More multiple high-resolution spectra in optical and near bands. Newer UV spectra - we still rely on IUE (1978-1996). High-resolution X-ray spectra on shorter time scales INTEGRAL (XRISM, Athena). NuS™ X-ray polarization data (IXPE). ATCA ~~1 r -i—r—i—r—r—i—r—i—r—i I I I I I I I I INTEGRAL/ISGRI (20-40 keV) NuSTAR (3-79 keV) I I I I I y To Earth I I I I I I I -200 0.00 -100 0.20 0 [lt-sec] orbital phase 0.40 100 200 0.60 0.80 .....1 1 1 1 1 1 1 1 1 L ■HI 1 1 1 1 1 1 Swift BAT * —c— MAXI (2-20 keV) ..-V .....9*'_Ji^f*mK^W 1 i i l*-T»S- -1.5 1 pattern & //light spectrum^ bending^ radiation transport JL absorption ^scattering fluorescenc ionisation ind feedback/ JL stellar surface heating lunz 0.1 1 0 ■ 1000 Size (km) ■ 111_I_I_I_.....I_ 10000 10s Athena @ L2