Detection of a 924s Modulation from Swift J1808.4-1754

ATel #6170; John A. Tomsick (SSL/UCB), Matteo Bachetti (Univ. de Toulouse and CNRS), Jamie A. Kennea (PSU), Hans A. Krimm (CRESST/GSFC/USRA)
on 26 May 2014; 01:09 UT
Credential Certification: John A. Tomsick (jtomsick@ssl.berkeley.edu)

Tweet

Swift J1808.4-1754 is an X-ray transient that was first detected on 2014 May 9 (ATEL#6138). A follow-up Swift/XRT observation on 2014 May 19 showed that the source has a hard spectrum and a likely near-IR counterpart, and it was suggested that it may be an accreting neutron star (ATEL#6155).

Here, we report on two more Swift/XRT observations that took place on 2014 May 22 (starting at 12.9h UT with an exposure time of 1946s) and 2014 May 25 (starting at 9.9h UT with an exposure time of 1273s). The May 22 observation shows a strong modulation in the 0.5-10 keV light curve. Although we only detect two cycles of the modulation, it is well-described (reduced-chi2 = 1.01 for 95 degrees of freedom) by a sinusoid with a period of 924+/-3 seconds. The fractional amplitude of the signal (half-amplitude of the sinusoid in counts/s divided by the mean count rate) is 39+/-2% (1-sigma errors). The modulation is also apparent in the light curve of the May 25 observation.

The mean 0.5-10 keV XRT count rates for the two observations are 1.94+/-0.04 c/s on May 22 and 1.65+/-0.04 c/s on May 25. On May 22, the 1-10 keV spectrum is well-described (reduced-chi2 = 1.00 for 68 degrees of freedom) by an absorbed power-law with nH = (4.4+/-0.6)e22 cm-2, Gamma = 0.98+/-0.14, and 2-10 keV unabsorbed flux = (2.75+/-0.10)e-10 erg/cm2/s. The nH is calculated with Wilms et al. (2000) abundances, and the errors quoted are 90% confidence. On May 25, we measure nH = (3.8+/-0.7)e22 cm-2, Gamma = 0.65+/-0.20, and 2-10 keV unabsorbed flux = (2.66+/-0.15)e-10 erg/cm2/s. Also, for both observations, there is a strong excess in the spectrum below 0.6 keV.

The light curve in the BAT 15-50 keV band started decreasing on 2014 May 23, falling to 0.008+/-0.002 ct/s/cm-2 (~35 mCrab) from its peak on May 20 of 0.011+/-0.0002 ct/s/cm-2.

The 924s period could be interpreted as the spin period of either a neutron star with a high magnetic field or a white dwarf. The hard power-law above 1 keV is typical for an accreting pulsar. If the system harbors a white dwarf, then the identification would be a magnetic CV (and probably an intermediate polar). The transient nature (ATEL#6138) and the fact that the source is seen in the near-IR but not in the optical (ATEL#6155), which argues for a larger distance and luminosity, suggest that a neutron star High-Mass X-ray Binary may be the most likely identification.