IGR J17191-2821

• 20-40 keV flux of ~9mCrab, marginal det. in 40-80 keV (5.8 mCrab)
• Mention 2 nearby sources: 1RXS J171909.4-282041 (at 1.4') and IRAS 17159-2819 (@ 1.6') and indicate presence of the unidentified EGRET source 3EG J1717-2737 located at RA=17h 17m 12s, dec=-27deg 37' 48" (+-0.64 deg @95% confidence).

• RXTE detection of source with position coincident
• 2-10 keV flux of 10 mCrab.
• Nothing detected with < 0.5 mCrab in following obs
• Source not detected the following day, and no flares brighter than 2 mCrab seen in a reanalysis of all scans from this region since 1999, Feb 5.
• Maximum duration of 1.7d of its activity and its hard spectrum.
• IGR J17191-2821 is a binary with a compact object and a high mass companion.
• Low-mass companion cannot be excluded, but very short outburst and an unusually hard spectrum then.
• A stellar flare from a near-by star or an outburst from a magnetar cannot be ruled out either.

• No source seen with Swift in INTEGRAL error circle with an upper limit of 0.0021 cnts/s.
• Translates into an 2-10 keV unabsorbed flux of 8.7*10E-14 erg/s/cm^2 or 0.004 mCrab (Gamma=1.8 and Nh=3.4e21 cm-2)
• In 1 pointing detection of source at ~3.5' from the INTEGRAL position.
• The flux is 3.6*10E-13 erg/s/cm^2 (unasorbed) or 0.02 mCrab (same Nh and photon index as above)
• Source not visible any more in the following pointing.
• Although the formal INTEGRAL error circle on the position of IGR J17191-2821 indicate that the sources are unrelated, still a small possibility that both sources are one and the same
• Over the span of 5 days (from March 3 to March 8) IGR J17191-2821 has decreased by a factor 500 or more.

• Swift XRT spectrum absorbed powerlaw => N_H=5.30 x 10²¹ cm^-2, Γ = 1.75, ~1.2 x 10^-9 erg/sec/cm² (absorbed 2-10 keV flux)
• Type-I X-ray burst that lasted for about 50 seconds and during which the Swift count rate increased by a factor ~3
• Burst spectrum fitted with a black body of temperature 0.99 +/- 0.08 keV.
• Upper limit for the distance of 23 kpc (likely to be closer)
• Type I burst => NS probably, in a low mass system.

• RXTE/PCA detection of an X-ray burst and high frequency oscillations during the burst.
• The oscillation frequency drifted from about 292 Hz to 294 Hz.
• Neutron star spin frequency is probably near the maximum oscillation frequency, or 294 Hz.
• Refine a new distance upper limit of ~11 kpc.

• Refinement of the X-ray position
• V, I and H-band images of the field of IGR J17191-2821 using the SMARTS 1.3m telescope and ANDICAM instrument on May 4.295 UT.
• Two sources at (RA,DEC) = (259.811453,-28.297926) and (259.812253 -28.297579) +- 0.4", in all three bands.
• Those sources were previsously indetected in USNO B1.0 and 2MASS catalogues.

• kHz QPO in the range between 700 and 900 Hz in RXTE observations
• Detection of an additional kHz QPO at a frequency of ~1030 Hz on several occasions.
• Latter with the upper kHz QPO and the one at lower frequencies with the lower kHz QPO. The frequency separation is about 330 Hz and is consistent within 3 sigma with the spin frequency
• Behaves during its outburst as a typical neutron star atoll sources in outburst.

• Chandra obs do not detect any source within 30" from the best position
• With Γ=1.71 and N_H=5.7 x10²¹ cm^-2 => 2-10 keV source flux <6.9 x10^-14 erg/cm²/s (<3.2E-3 mCrab) <=> 0.3-10 keV luminosity <1.2 x 10³³ ergs/s(11 kpc upper limit)

• X-ray spectral and temporal analysis during outburst (Swift and RXTE)
• 5 type I X-ray bursts => similar profile, radius and temp. 2 keV < kT_max <3 keV
• High frequency (605-1185 Hz) QPO => constant Δν when a pair is detected
• Precursor to the main outburst
• Spectral transition from soft Banana to hard/extreme island, and back to the soft state => Source is an atoll
• Nearly coherent burst oscillations at 291-295 Hz => spin of NS ~ 294 Hz

• Estimate mass and compactness from differences between kHz QPOs
• M=0.843 Msun, and R=6.762 km.