Apparent Superorbital Modulation in 4U 1538-52 at Four Times the Orbital Period

ATel #11918; Robin H. D. Corbet (UMBC/NASA GSFC), Joel B. Coley (NASA GSFC/NPP/USRA), Hans A. Krimm (USRA/NSF), Katja Pottschmidt (UMBC/NASA GSFC)
on 6 Aug 2018; 13:01 UT
Credential Certification: Robin Corbet (corbet@umbc.edu)

Tweet

4U 1538-52 is a bright eclipsing high-mass X-ray binary containing a neutron star accreting from the wind of QV Nor, a B0Iab star.

We calculated the power spectrum of the Swift BAT light curve from the transient monitor program (Krimm et al. 2013, ApJS, 209, 14) covering the energy range 15 - 50 keV, and a time range of MJD 53416 to 58318 (2005-02-15 to 2018-07-19) for periods from 0.07 days to the length of the light curve. The power spectrum is dominated by very strong peaks at the orbital period and its harmonics. However, in addition, we noted a peak near 14.9 days at a height of ~22.5 times the mean power level with a nominal false alarm probability (FAP) of 10^-5. To further investigate this modulation, we removed data within +/- 0.15 in phase of the eclipse center, using the ephemeris of Mukherjee et al. (2006, JApA, 27, 411). From a power spectrum of this modified light curve, the orbital modulation was strongly reduced, and the only significant peak was near 14.9 days at a height ~30 times the local power level with an FAP of ~10^-8. The derived period was 14.916 +/- 0.003 days. We note that this is consistent, within the errors, with four times the orbital period of 3.728382 +/- 0.000011 days.

We note that both the candidate superorbital period and the orbital period of 4U 1538-52 are very similar to those of IGR J16418-4532 (Corbet & Krimm, 2013, ApJ, 778:45) where the ratio of superorbital to orbital period is 3.94, and thus would be consistent with the possible correlation between superorbital and orbital periods in wind-accretion HMXBs.

To characterize the modulation near 14.9 days in 4U 1538-52 we also made a sine wave fit to the modified light curve. The mean count rate is 0.0055 cts/cm^2/s (approximately 25.3 mCrab) and the sine wave semi-amplitude is 0.00045 +/- 0.00004 cts/cm^2/s, approximately 2.1 mCrab. The epoch of maximum flux is MJD 56106.6 +/- 0.2, which is 0.37 +/- 0.2 days after the time of eclipse center. We note that because of the exact ratio between the candidate superorbital period and orbital period and the phasing, we are not measuring the flux at the implied maximum or minimum of the modulation.

We are not aware of any artifact that would cause modulation at four times the orbital period and aligned with the eclipse. We investigated the power spectra of BAT light curves of other eclipsing supergiant HMXBs without already known superorbital periods for similar modulation, and did not see any. Observations at other wavelengths over the proposed superorbital period are encouraged.