Nova Mon 2012: High soft X-ray variability and probable 7.1 hour UV/X-ray period

ATel #4727; Julian P Osborne, Andy Beardmore, Kim Page (University of Leicester)
on 11 Jan 2013; 13:25 UT
Credential Certification: Julian P Osborne (julo@star.le.ac.uk)

Tweet

Nova Mon 2012, a neon nova (ATEL #4709) discovered as a Fermi LAT source at E > 100 MeV on 22 Jun 2012 (ATEL #4224, ATEL #4310), has been the subject of a daily Swift observing campaign since it entered its super-soft phase (ATEL #4590) on 18 Nov 2012.

Since that date (101 days after optical discovery on 9 Aug, CBET #3202, our T_0), the soft X-ray flux has increased substantially, while at the same time becoming extremely variable. The soft X-ray count rate peaked at 3.0 counts/sec on 10 Jan 2013 (day 153), but within 3 hours it was back down at 0.8 counts/sec.

Similar early-phase super-soft extreme variability has also been reported in RS Oph (Osborne et al 2011 ApJ 727 124), and other novae (see e.g. Schwarz et al 2011 ApJS 197 31). It has also been seen in Nova Mon 2012 by Orio et al (ATEL #4633). The origin of this variability is unclear, it may be due to clumpy ejecta passing through the line of sight.

Periodogram analysis of the UVOT image mode data after day 100 reveals a strong period at 0.2957 days (with an estimated error of +/- 0.0007 days) after 3rd order polynomial detrending. The same period is present in all of the 3 filters in use (uvw1, uvm2, uvw2; central wavelengths 2600, 2246, 1928 Angstroms respectively). The modulation amplitude is approximately +/- 0.05 magnitudes. Additional peaks are also present in the periodogram at aliases of the Swift orbital period (0.0862 & 0.0545 days), however folding analysis suggests that these are not the intrinsic period. Observations before day 100 were taken at a lower cadence, and do not provide evidence for the period. We emphasise that observations to date have been mostly taken at daily intervals (sometimes spread over 3 snapshots in 3 hours), and are not optimal for the detection of this 7.1 hour period; we aim to verify the period detection with higher cadence observations shortly.

A similar analysis of the bright Swift XRT data (which is dominated by counts at energies below 0.7 keV) in 100 sec bins also reveals the same period (i.e. P = 0.2955 +/- 0.0015 days). Folding the XRT and UVOT data shows a roughly sine-like modulation, which has a common phasing in the two wavebands.

The apparent detection of a 7.1 hour common photometric modulation in the UV and X-rays most likely implies that this is the orbital period of Nova Mon 2012. The shape of the modulation, and its shallow UV depth, suggests that we are not seeing a total eclipse of the central region by the secondary star, but rather a partial eclipse of extended emission by an accretion disk rim which is raised at the point of impact of the stream from the secondary star. An accretion disk thus exists at this time. A similar in-phase UV-X-ray modulation was seen in the nova HV Cet (Beardmore et al 2012 A&A 545 A116).

A 7.1 hour orbital period is longer than that of most classical novae, but many are known around this period; it implies a near main sequence binary companion.

Spectroscopic observations are encouraged to confirm this candidate orbital period.

We thank the Swift PI and operations team for supporting these observations.