IGR J17391-3021


Other name = XTE J1739-302, AX J1739.1-3920

Type Orb. P/Pulse P. Optical Counterpart Infrared Counterpart
R.A. (J2000) Dec. (J2000) References (position) R.A. (J2000) Dec. (J2000) References (position)
Supergiant Fast X-ray Transient
O8 Iab(f)
51.47d / ? USNO B1.0 0596-0585865 Smith et al. 2005 2MASS J17391155-3020380 Smith et al. 2005
Published Papers
Miscellaneous :

Smith 2004 (Atel 338): Optical magnitudes (USNO B1.0): B2=17, R2=12.9, possible distance=3.3 kpc
Sguera et al. 2005 :
  • Fast transient
  • Recurrent outburst
  • Flares lasting between 30 minand 3 hours
  • The source spectrum is typical of a NS binary
  • Most likely a HMXB given counterpart but outbursts shorter than expected for HMXBS or Be/NS binaries.
Smith et al. 2005 :
  • During both low level and outburst states high energy spectra are hard.
  • absorption column density seen to vary from one outburst to another.
  • Class of Supergiant Fast Xray Transient
Negueruela et al. 2005 :
  • Companion is O8 Iab(f) supergiant
  • distance 2.3 kpc
Sguera et al. 2006 :
  • 20-60 keV peak flux of 480 mCrab
  • outburst from 30 min to 14 hours duration
Türler 2007 (Atel 1019):
  • one of the brightest outbursts ever recorded occured at UT:13:30 on March 2, 2007.
  • ~1-hour long outburst with peak fluxes of 330 and 250 mCrab in the 20-40 and 40-80 keV bands, respectively.
  • thermal bremsstrahlung model with kT ~ 22.5 keV fits the 20-100 keV spectrum.
Chaty et al. 2008:
  • Optical and infrared observations
  • 2MASS source is a blended source
  • B>21.2, V=14.97, R=12.94, I=11.28, Z=10.458
  • Discrepancy of IR mag with 2MASS cat due to domain of non linearity of SOFI detector on NTT.
  • Hα, Paschen H lines, HeI, HeII, NI lines in optical, and IR spectra very rich in H lines.
  • P-Cygni profile=> O sg, not a Be as stated in Bird et al. 2006
Rahoui et al. 2008:
  • Medium infrared observations, GLIMPSE data
  • SED fitting => AV=9.2, Tstar=31400 K
  • No need for an additional component to fit MIR spectra
  • Suggested distance 2.7 kpc
Romano et al. 2008: Atel 1466
  • New outburst detected by Swift/BAT on 2008-04-08 at 21:28:15
  • 15-150 keV BAT lightcurve nearly constant from T-119 to T+950 s
  • BAT spectra simple power laws with photon indices ~constant (2.7 and 3.0).
  • Swift/XRT light curve <=> 2 distinct flares the 2nd half bright as the first. Photon indices harder than those from BAT spectra.
Chenevez et al. 2008: Atel 1471
  • INTEGRAL: outburst detected by Swift started in reality 5 hours earlier than the BAT trigger (2008-04-08)
  • JEM-X detected a flare starting at 16:18
  • 3-10 keV (resp 10-25 keV) mean flux averaged over ~2 hours was 42 mCrab (resp. 80 mCrab).
  • 3-10 keV (10-25 keV) upper limits before the flare were 4 mCrab (resp. 8 mCrab).
  • Detection by ISGRI with 20-40 keV (40-80 keV) fluxes of 31 mCrab (13 mCrab) over a 8 ksec observation.
Sidoli et al. 2008:
  • Swift study of the "out-of-outburst" periods
  • Absorbed pl or bbody fits the data well
  • Γ= 1.4 and NH=3.3x1022cm-2 or kT=1.5 keV and NH=1.6x1022cm-2
Blay et al. 2008:
  • Analysis of INTEGRAL data (GPS, GCDE and data from the Galactic centre key project from AO 4 and 5)
  • Long term light curves of the source=>two different types of activity (outbursts):
    • the most numerous ones are the faintest, with a 20-40 keV rate of ~10 cts/s, that seem to last ~0.6h
    • the brightest are usually found at ~60 cts/s, and last for longer than ~1h
  • No coherent modulation up to 100 s.
  • Spectra: in addition to an absorbed exponentially cut-off power law continuum, presence of broad absorption at 30 and 60 keV
  • Cyclotron Resonant scatering feature to the continuum does not improve the fit significantly.
  • Clumpy wind models explain well the properties of the source => predict an orbital period of ~8 days.
Romano et al. 2008: Atel 1659
  • New outburst seen with swift
  • BAT triggered twice on 2008-08-13 23:49:17 UT and 2008-08-14 00:12:53 UT.
  • The swift BAT and XRT spectral analysis showed consistent parameters with those obtained during the previous outburst of this source
Sidoli et al. 2008:
  • Swift monitoring => observation of 1 flare
  • Spectrum has Γ=1.5 and is higly absorbed (NH=13 x1022 cm-2).
  • Compared to the out-of-outburst periods, the spectrum has similar continuum but is more absorbed
  • Broad band spectrum <=> power law
  • Possible cut-off at 13 keV <=> Comptonization with kT=8 keV and τ=6.8 but same χ2 as for simple power law
  • B comptible with 2 x1012 G => difficut to reconcile with high B required by Bozzo et al.'s model
Sidoli et al. 2009 (MNRAS):
  • Time resolved spectroscopy of the flaring event of 2008 August 13
  • Spectra are well fitted with either an absorbed power law or absorbed black body
  • Values of the photon index or the black body temperature remain constant
  • Clear evolution of the absorption column density: NH higher during the rising portion of the flare
  • bmc also used => bb temperature (of the seed photons) of 1.6 keV for a black body size of 1.6 km => consistent with the size of the polar cap
Romano et al. 2009 (MNRAS):
  • Swift analysis of the first year of monitoring
  • Complete spectral analysis of the data => spectra well fitted with either an absorbed power law or an absorbed black body
  • Study of UVOT light curves
  • duty cycle of inactivity = 39% => true quiescence is a rare state
Bozzo et al. 2010 (A&A):
  • XMM analysis during quiescence
  • Quiescent emission composed of multiple flares occuring after period of faint X-ray emission.
  • Hardness ratio and the intensity are positively correlated
  • Spectral analysis => soft component below 2 keV, in addition to a standard (cutoff) power law with absorption
  • Excess well represented by either a black body or thermal plasma model
  • Thermal plasma could be the X-ray emission due to shocks in the wind of the supergiant companion.
Drave et al. 2010 (MNRAS):
  • 51.47 d period in analysis of 12.4Ms of INTEGRAL data
  • Interpreted as the orbital period of the system
  • Observation of 35 epochs of outburst throughout orbit
  • Source is not powered by Roche lobe overflow
  • Semi major axis within 173.4-179.8 Rsun => eccentricities up to 0.8
  • Neutron star probably orbits in an inhomogeneous clumpy wind
  • Possibility of enhanced equatorial density region
Romano et al. 2011:
  • 2 years of Swift monitoring
  • duty cycle of inactivity ~ 39%
  • X-ray spectroscopy in different luminosity phases of activity
Bodaghee et al. 2011:
  • Suzaku observation
  • First half in a dormant state <=> quiescence, absorbed luminosity of 1.3e33 erg/s (d/2.7 kpc)^2
  • Second half in low-activity state with weak flares. Peak luminosity of 7.4e33 erg/s
  • During flaring phase NH ~4.1e22 /cm2 <=> increase of a factor of 2--4 (compared toquiescence) => accretion of obscuring clumps of wind.
  • Low state represents ~60% of all observations => most common behavior in this source.
Farinelli et al. 2012 (MNRAS):
  • Swift observations
  • Spectral fits and interpretation within framework of comptonisation
Wang & Chang 2012 (A&A):
  • Characteristic time scale of variability 1550-2000 s
  • Study retrograde wind accretion scenario
  • asuming variability ~spin and orbital period => R corrotation ~ 1.3-1.6e10 cm.
  • Uncertain nature of accretion regime

Last updated 7 Nov. 2012

Jerome Rodriguez