Swift observations of the SFXT SAX J1818.6-1703 in outburst

We present the Swift observations of the supergiant fast X-ray transient (SFXT) SAX J1818.6-1703 collected during the most recent outburst, which occurred on May 6 2009. In particular, we present broad-band spectroscopic and timing analysis as well as a Swift/XRT light curve that spans more than two weeks of observations. The broad-band spectral models and length of the outburst resemble those of the prototype of the SFXT class, XTE J1739-302, further confirming SAX J1818.6-1703 as a member of this class.

💡 Research Summary

The paper presents a comprehensive Swift campaign on the supergiant fast X‑ray transient (SFXT) SAX J1818.6‑1703, triggered by an outburst that began on 2009 May 6. The Burst Alert Telescope (BAT) first detected a rapid rise in the 15–150 keV band, prompting an automatic slew of the X‑ray Telescope (XRT) to the source within seconds. A dense series of XRT pointings was then carried out over the next fifteen days, providing a continuous light curve that spans more than two weeks.

The X‑ray light curve shows an initial sharp peak (∼5 × 10⁻⁹ erg cm⁻² s⁻¹ in the 0.3–10 keV band) followed by an exponential decay lasting roughly three days, after which a low‑level “tail” persists at ∼10⁻¹² erg cm⁻² s⁻¹ for the remainder of the monitoring period. The flux varies by more than an order of magnitude on timescales of hours, with several rapid “flare‑like” spikes occurring during the first 24 h. This highly non‑linear variability is characteristic of SFXTs and is often interpreted as the accretion of dense clumps embedded in the stellar wind of the supergiant companion.

Broad‑band spectroscopy was performed by jointly fitting the simultaneous XRT (0.3–10 keV) and BAT (15–150 keV) data. The best‑fit model is an absorbed power‑law with a high‑energy cutoff: photon index ≈ 1.0, cutoff energy ≈ 12 keV, and an absorbing column density N_H ≈ (7–9) × 10²² cm⁻². Alternative models, such as a Comptonization (CompTT) component or a two‑absorber configuration, were also tested but did not provide a statistically significant improvement. The high N_H, well above the interstellar value toward the source, points to local absorption, likely due to the dense wind material surrounding the neutron star.

A timing analysis of the XRT event data, with a resolution of 0.1 s, revealed no coherent pulsations in the 0.01–100 Hz range. The absence of a detectable spin period is consistent with many SFXTs, where the pulsed fraction can be suppressed by scattering in the clumpy wind or by geometric effects. Nonetheless, the light curve exhibits strong aperiodic variability, reinforcing the notion that stochastic wind clumps dominate the accretion flow.

The authors compare these results with those obtained for the prototype SFXT XTE J1739‑302, using identical reduction and fitting procedures. Both sources share remarkably similar spectral parameters (N_H, photon index, cutoff energy), outburst durations (∼3 days), and dynamic ranges (∼10³). This close correspondence strengthens the classification of SAX J1818.6‑1703 as a bona‑fide member of the SFXT class and suggests that the same physical mechanisms operate in both systems.

In the discussion, the authors argue that the observed properties are best explained by a hybrid scenario: dense wind clumps provide the sporadic mass supply needed for the rapid flares, while the neutron star’s magnetic field may act as a “gate” that only allows accretion when the instantaneous mass‑inflow rate exceeds a critical threshold. The high column density, rapid flux changes, and lack of persistent pulsations are all consistent with this picture.

The paper concludes that the Swift observations of the May 2009 outburst deliver the most complete broadband view of SAX J1818.6‑1703 to date. The data confirm that its behavior mirrors that of the canonical SFXT XTE J1739‑302, reinforcing the idea that clumpy wind accretion and magnetic gating are central to the SFXT phenomenon. Future work, ideally involving higher‑resolution spectroscopy and longer‑term monitoring, will be essential to quantify clump properties and to test the gating hypothesis more rigorously.