Astron. Astrophys. 318, 134-139 (1997)

3. Discussion

RX J2353.0-3852 is a puzzling object. The presence of two independent periodicities in the optical data from 1993 implies that the primary may be an asynchronously rotating, magnetic white dwarf, and the object is an intermediate polar (IP). However, neither of these two periodicities are present in the 1994 optical light curves and only incoherent variations are observed. Likewise, neither periodicity is present in the X-ray data. Furthermore, He II 4686, usually considered an indicator of a magnetic primary, is not present.

The optical spectrum incorporates strong, broad, double-peaked emission lines and wide Balmer absorption troughs from H onwards. The double-peaked lines imply a disk viewed at high inclination but there is no evidence of eclipses in the light curves (the few low points observed do not appear on a repeating cycle). Since the star was observed at the quiescent brightness of V = 16.5 shortly before the spectroscopic observations, we suggest that the absorption troughs are not due to an outburst. The more likely source is the white dwarf atmosphere (Szkody 1985); this implies the presence of a faint accretion disk and therefore a low . Patterson & Raymond (1985a) find a correlation between and which predicts g s^-1 for RX J2353.0-3852; this is low for a CV, but not exceptionally so. However, the correlation of with He II 4686 noted by Patterson & Raymond (1985b) implies that we should expect to see this line in RX J2353.0-3852. Vrtilek et al. (1994) find similar X-ray spectra for low CVs, although our spectrum of RX J2353.0-3852 is a little softer than the examples given. RX J2353.0-3852 is also consistent with Patterson & Raymond's (1985a) correlation between and H equivalent width.

Augusteijn, Wisotski & Hazen (1996) report spectroscopic observations of RX J2353.0-3852 in which they independently discovered a 78 min periodicity suggesting that the 2282 s periodicity we observe is actually one-half the true orbital period. Indeed, twice the one-day alias of the 2282 s periodicity is close to the 12-hour alias of the 5247 s periodicity.

We note the similarity of RX J2353.0-3852 to WZ Sge, which has a short orbital period (4898 s), a variable, double-humped light-curve, a spectrum dominated by broad, double-peaked Balmer lines which lie in wide absorption troughs from H onwards, no He II 4686 emission (e.g. Gilliland, Kemper & Suntzeff 1986), and a low mass transfer rate.

Alternatively, if our two photometric periods do reflect the orbital and spin periods of a magnetic system, the X-ray properties of RX J2353.0-3852 are somewhat unusual. Before ROSAT, IPs were known for intrinsically absorbed thermal bremsstrahlung emission with temperatures (kT) of tens of keV. However, Haberl & Motch (1995) identify three ROSAT IPs with X-ray spectra more typical of polars, being dominated by soft, blackbody-like radiation with little intrinsic absorption ( tens of eV). These objects are suggested to be the progenitors of polars. The X-ray spectrum of RX J2353.0-3852 appears to lie between these two regimes, peaking at about 0.2-0.3 keV - too hard for a polar or soft IP, and too soft for a hard IP. If RX J2353.0-3852 is an IP, then it has the shortest orbital period of any IP, with the single exception of the suspected IP SW UMa ( = 4909 s, = 954 s, Shafter et al. 1986). Nevertheless, the overall weight of evidence seems to suggest that the nonmagnetic model is more likely.

© European Southern Observatory (ESO) 1997

Online publication: July 8, 1998
helpdesk.link@springer.de