Studying the Stellar Wind in the Vela X-1 System with XMM-Newton/RGS ESAC Trainee Project

Studying the stellar wind in the Vela X-1 system with XMM-Newton/RGS ESAC trainee project

Manfred Hanke (Remeis-observatory, Bamberg, Germany) (tutor: Andy Pollock) 2007, June – August Overview

• The Vela X-1 system i – known facts general concept of NS HMXB systems and evidences in the case of Vela X-1 [¡ theory → observation !]

• The Vela X-1 system ii – unresolved issues physical eﬀects in an XRB system and goals of this project (to constrain the physical parameters) [¿ observation → theory ?]

• First results - modelling of two high resolution spectra in eclipse - a ﬂare in the latest XMM-observation The high-mass X-ray binary Vela X-1 / HD 77581

Vela X-1 is an X-ray binary, consisting of a neutron star and the 23.5 M B0.5 Ib supergiant star HD 77581. The high-mass X-ray binary Vela X-1 / HD 77581

5 Helium-rich WR star T~2·10 yr, N~500 with OB-companion

−2 -1 Primary explodes as ν~10 yr type Ib Supernova and becomes a neutron star or black hole

4 Secondary is close to Roche lobe. T~10 yr, N~100 Accretion of stellar wind results in powerful X-ray emission

The evolution of a compact binary. (after Postnov & Yungelson, 2006, Fig. 4) Under certain conditions (Blaauw, 1961), the supernova explosion does not disrupt the binary system, but gives a kick velocity. A bow shock around HD 77581

The star HD 77581 has a large proper motion (& 7 mas/yr).

(distance = 1.82 kpc) ⇒ space velocity & 90 km/s ⇒ HD 77581 is a runaway star

Due to the supersonic motion through the ISM, the strong stellar wind causes shock fronts.

(corrected Hα image; Kaper et al., 1997, Fig. 1) The high-mass X-ray binary Vela X-1 / HD 77581

Vela X-1 is an X-ray binary, T~3 Myr, N~104 Two OB main-sequence consisting of a neutron star stars and the 23.5 M B0.5 Ib supergiant star HD 77581. T~104 yr, N~30 More massive star (primary) overfills Roche lobe While orbiting the companion (in 8.96 days), only 0.8 stellar radii 5 Helium-rich WR star T~2·10 yr, N~500 with OB-companion from its surface, the neutron star sweeps up part of the stellar wind. −2 -1 Primary explodes as ν~10 yr type Ib Supernova and This accretion releases becomes a neutron star or black hole gravitational energy in the form of X-rays. 4 Secondary is close to Roche lobe. T~10 yr, N~100 Accretion of stellar wind results in powerful X-ray emission

The evolution of a compact binary. (after Postnov & Yungelson, 2006, Fig. 4) The high-mass X-ray binary Vela X-1 / HD 77581

The photoionization of the stellar wind by the hard X-rays from the neutron star creates very complex structures in the system.

Fully ionized material becomes transparent to the star’s radiation pressure.

(Goldstein et al., 2004, Fig. 4) X-ray photoionization of the stellar wind

The photoionization of the stellar wind by the hard X-rays from the neutron star creates very complex structures in the system.

Fully ionized material becomes transparent to the star’s radiation pressure.

Recent simulations (Mauche et al., 2007) have shown that rather chaotic structures may emerge – just from the interaction of the X-rays with the wind (in the rotating system).

(Goldstein et al., 2004, Fig. 4) X-ray spectrum of Vela X-1 in eclipse

Due to the system’s high inclination > 73◦, the neutron star (X-ray source) is eclipsed from orbital phase φ = 0.9 to φ = 0.1. ⇒ Reprocessed X-rays from the (in parts) highly ionized wind / circumstellar material, which has been excited, becomes visible: X-ray spectrum of Vela X-1 in eclipse

• Emission lines from highly ionized ions np → 1s [H-like], 1s np → 1s2 [He-like]

• Radiative recombination continua e− + X+(n+1) → X+n

• Fluorescent Kα emission lines 1s 2s22px · · · → 1s2 2s22px−1 ... X-ray spectrum of Vela X-1 in eclipse

• Emission lines from highly ionized ions np → 1s [H-like], 1s np → 1s2 [He-like]

• Radiative recombination continua e− + X+(n+1) → X+n

• Fluorescent Kα emission lines 1s 2s22px · · · → 1s2 2s22px−1 ...

Can we perform a time-resolved high-resolution spectroscopy? Can we test the predictions of the wind-simulations by observational means? Can we put any constraints on the accretion ﬂow? High resolution X-ray observations of Vela X-1

Capable instruments currently in orbit: Chandra / High Energy Transmission Grating Spectrometer (HETGS) XMM-Newton / Reﬂection Grating Spectrometer (RGS) High resolution X-ray observations of Vela X-1

Capable instruments currently in orbit: Chandra / High Energy Transmission Grating Spectrometer (HETGS) XMM-Newton / Reﬂection Grating Spectrometer (RGS)

Chandra 102

Chandra 1926

Chandra 1928

Chandra 1927

0111030101 0406430201

0 0.2 0.4 0.6 0.8 orbital phase

There are 4 + 2 observations of Vela X-1 from 2000 & 2001 and 2000 & 2006.

The latest XMM-observation has an exposure time of almost 125 ks. First results: Modelling the spectrum of Chandra-obs. # 102 α α α α α 4 S XV S XV S XV Si XIV Si XIV S K Si XIV Si XIV Si XIII Si XIII Si XIII Si X K Si IX K Si VIII K Si VII K Mg XII Mg XII Mg XI 2 0 30 20 MEG 1 10 0 15 10 5 HEG 1 0

5 5.5 6 6.5 7 7.5 8

Wavelength [Å] 4 Mg XII Mg XII Ne X RRC Mg XI Mg XI Mg XI Ne X Ne X Ne X Ne X Ne X Ne X 2 0 15 10 5 MEG 1 0 15 10 HEG 1 5 0

8 8.5 9 9.5 10 10.5 11

Wavelength [Å] First results: Modelling the spectrum of Chandra-obs. # 1926 α α α α α 4 S XV S XV S XV Si XIV Si XIV S K Si XIII Ni XXVII Si XIV Si XIV Mg XII RRC Si XIII Si XIII Si XIII Si X K Si IX K Si VIII K Si VII K Mg XII Mg XII Mg XI Mg XI 2 0 80 60 40 MEG 1 20 0 30 20 HEG 1 10 0 5 5.5 6 6.5 7 7.5 8

Wavelength [Å] α α α α 4 Mg XII Mg XII Ne X RRC Mg XI Mg XI Mg XI Ne X Ne X Mg VII K Ne X Ne X Mg VI K Mg V K Mg K Ne X Ne X Ne IX 2 0 40 30 20 MEG 1 10 0 30 20 HEG 1 10 0 8 8.5 9 9.5 10 10.5 11

Wavelength [Å] XMM-observation # 0406430201 of Vela X-1 1 0.1 Å [counts/s] 0.01 12 − 24 10 1 0.1 Å [counts/s] 6 − 12 0.01 Å) 10 Å) / (12 − 24 1 (6 − 12

0 20 40 60 80 100 120

observation time [ks] These light curves show the count rate of RGS 2 in diﬀerent energy bands and the corresponding ‘hardness’ ratio as a function of time. XMM-observation # 0406430201 of Vela X-1 1 0.1 Å [counts/s] 0.01 12 − 24 10 1 0.1 Å [counts/s] 6 − 12 0.01 Å) 10 Å) / (12 − 24 1 (6 − 12

0 20 40 60 80 100 120

observation time [ks] 77 ks after the start of this observation, the spectrum changes signiﬁcantly: The low energy (12–24 A)˚ count rate increases by ×10, and the high energy (6–12 A)˚ count rate even by ×100.

What is the nature of this ﬂare-like event? ASM light curves during the XMM-observation # 0406430201

The All Sky Monitor onboard RXTE conﬁrms this ﬂaring behaviour:

8 6 4 2 0 1.5 keV - 3 -2 10 8 6 4 2

3 keV - 5 0 -2 25 20 15 10 5

5 keV - 12 0 -5 -1000 -500 0 500 1000 time from start of observation # 0406430201 [ks] (The orbital period is 774 ks, i.e., this shows one period before and one period after the observation.) ASM light curves before and after this XMM-observation

Such ﬂares are, however, not uncommon, as a longer view shows:

8 6 4 2 0 1.5 keV - 3 -2 10 8 6 4 2

3 keV - 5 0 -2 25 20 15 10 5

5 keV - 12 0 -5 -400 -300 -200 -100 0 100 200 300 400 time from start of observation # 0406430201 [days] (The orbital period is 8.96 days.) ASM light curves before and after this XMM-observation

In the last 2 years, such ﬂares were only seen in orbital phases 0.15. . . 0.70.

8 6 4 2 0 1.5 keV - 3 -2 10 8 6 4 2

3 keV - 5 0 -2 25 20 15 10 5

5 keV - 12 0 -5 -0.5 0 0.5 1 1.5 binary orbital phase (orbital period ≡ 1) The end

As you may guess: There is still a lot to ﬁnd out.

Thank you for your Attention!

References Barziv O., Kaper L., Van Kerkwijk M.H., et al., 2001, A&A 377, 925 Blaauw A., 1961 15, 265 Goldstein G., Huenemoerder D.P., Blank D., 2004, AJ 127, 2310 Kaper L., van Loon J.T., Augusteijn T., et al., 1997, ApJ 475, L37+ Mauche C.W., Liedahl D.A., Akiyama S., Plewa T., 2007, ArXiv e-prints 704 Postnov K.A., Yungelson L.R., 2006, Living Rev. in Relativity 9, 6 Schulz N.S., Canizares C.R., Lee J.C., Sako M., 2002, ApJ 564, L21 van Kerkwijk M.H., van Paradijs J., Zuiderwijk E.J., et al., 1995, A&A 303, 483 Watanabe S., Sako M., Ishida M., et al., 2006, ApJ 651, 421