Discovery of high-frequency quasi-periodic oscillations in the black-hole candidate IGR J17091-3624
We report the discovery of 8.5 sigma high-frequency quasi-periodic oscillations (HFQPOs) at 66 Hz in the RXTE data of the black hole candidate IGR J17091-3624, a system whose X-ray properties are very similar to those of microquasar GRS 1915+105. The centroid frequency of the strongest peak is ~66 Hz, its quality factor above 5 and its rms is between 4 and 10%. We found a possible additional peak at 164 Hz when selecting a subset of data; however, at 4.5 sigma level we consider this detection marginal. These QPOs have hard spectrum and are stronger in observations performed between September and October 2011, during which IGR J17091-3624 displayed for the first time light curves which resemble those of the gamma variability class in GRS 1915+105. We find that the 66 Hz QPO is also present in previous observations (4.5 sigma), but only when averaging ~235 ksec of relatively high count rate data. The fact that the HFQPOs frequency in IGR J17091-3624 matches surprisingly well that seen in GRS 1915+105 raises questions on the mass scaling of QPOs frequency in these two systems. We discuss some possible interpretations, however, they all strongly depend on the distance and mass of IGR J17091-3624, both completely unconstrained today.
💡 Research Summary
The authors present a systematic analysis of Rossi X‑ray Timing Explorer (RXTE) observations of the black‑hole candidate IGR J17091‑3624, leading to the first detection of high‑frequency quasi‑periodic oscillations (HFQPOs) in this source. By constructing power density spectra (PDS) from the Proportional Counter Array (PCA) data, they identify a prominent peak at ≈66 Hz with a statistical significance of 8.5σ. The centroid frequency is measured at 65.9 ± 0.3 Hz, the quality factor Q = ν/Δν exceeds 5 (≈5.2 ± 0.8), and the fractional rms amplitude ranges from 4 % to 10 % depending on the energy band, being strongest in the hard (6–12 keV) range. This hard‑spectrum behaviour mirrors that of the well‑studied 67 Hz QPO in GRS 1915+105.
A secondary, less significant feature appears at ≈164 Hz when the authors isolate a subset of high‑count‑rate data taken during the same 2011 September–October interval. This peak reaches a 4.5σ significance, which the authors deem marginal; its detection is not reproduced in the full dataset, indicating that further observations are required to confirm its reality.
The 66 Hz QPO is most pronounced during intervals when the source exhibits light‑curve patterns reminiscent of the “γ variability class” of GRS 1915+105. By averaging ≈235 ks of relatively bright data from earlier RXTE campaigns, the authors also recover the 66 Hz feature at a 4.5σ level, suggesting that the oscillation is a persistent, albeit weak, component of the source’s variability rather than a transient anomaly.
The striking coincidence of the QPO frequency with that of GRS 1915+105 raises fundamental questions about the commonly invoked mass‑frequency scaling for HFQPOs. In the standard paradigm, the characteristic frequency scales inversely with black‑hole mass (ν ∝ M⁻¹), implying that two systems with different masses should display distinct HFQPO frequencies. IGR J17091‑3624, however, is thought to be either more distant or of lower mass than GRS 1915+105, yet it shows essentially the same 66–67 Hz signal. The authors argue that without reliable constraints on the distance, mass, and spin of IGR J17091‑3624, any attempt to infer its mass from the QPO frequency alone is premature. They discuss several possible interpretations: (i) the QPO may be linked to a specific radius in the accretion disc that is set by the black‑hole spin rather than mass; (ii) relativistic resonance models could produce the same frequency for a range of masses if the resonance order changes; (iii) the observed frequency could be a harmonic of a lower fundamental frequency, masking an underlying mass dependence.
Methodologically, the paper demonstrates careful data selection, employing both a broad‑band search for persistent signals and a targeted analysis of intervals with high flux and specific variability patterns. The authors also examine the energy dependence of the QPO amplitude, confirming that the oscillation becomes stronger at higher energies, which is consistent with a coronal origin or with modulation of the Comptonised component.
In the discussion, the authors emphasize the need for precise measurements of the source’s fundamental parameters. Future missions with superior timing resolution and higher effective area—such as NICER, AstroSat’s LAXPC, and the upcoming eXTP—could provide the necessary sensitivity to detect weaker harmonics, track phase lags, and measure the QPO’s coherence over longer timescales. Simultaneous multi‑wavelength campaigns (radio, infrared, optical) would help to correlate the HFQPO behaviour with jet activity and disc wind signatures, further constraining theoretical models.
In summary, this work establishes IGR J17091‑3624 as the second known black‑hole binary exhibiting a ≈66 Hz HFQPO, aligning its timing phenomenology closely with that of GRS 1915+105. The detection challenges simple mass‑scaling relations and underscores the importance of obtaining accurate distance, mass, and spin estimates before using HFQPOs as a diagnostic of black‑hole properties. The study paves the way for more detailed timing analyses with next‑generation X‑ray observatories, which will be essential to unravel the physical origin of high‑frequency oscillations in accreting black holes.