Spitzer Space Telescope Observations of Magnetic Cataclysmic Variables: Possibilities for the Presence of Dust in Polars

• Published in: The Astrophysical journal Vol. 659; no. 2; pp. 1541 - 1562
• Main Authors: Brinkworth, C. S, Hoard, D. W, Wachter, S, Howell, S. B, Ciardi, David R, Szkody, P, Harrison, T. E, van Belle, G. T, Esin, A. A
• Format: Journal Article
• Language: English
• Published: Chicago, IL IOP Publishing 20.04.2007; University of Chicago Press
• Subjects: Astronomy; Earth, ocean, space; Exact sciences and technology; Low-mass stars; Brown dwarf stars; stars: magnetic fields; infrared: stars; Magnetic stars; Cataclysmic variable stars; Magnetic fields; Infrared star; stars: low-mass, brown dwarfs
• ISSN: 0004-637X; 1538-4357
• DOI: 10.1086/512797

We present Spitzer photometry of six short-period polars, EF Eri, V347 Pav, VV Pup, V834 Cen, GG Leo, and MR Ser. We have combined the Spitzer IRAC (3.6-8.0 mu m) data with the 2MASS JHK sub(S) photometry to construct the SEDs of these systems from the tear- to mid-IR (1.235-8 mu m). We find that five out of the six polars have flux densities in the mid-IR that are substantially in excess of the values expected from the stellar components alone. We have modeled the observed SEDs with a combination of contributions from the white dwarf, secondary star, and either cyclotron emission or a cool, circumbinary dust disk to fill in the long-wavelength excess. We find that a circumbinary dust disk is the most likely cause of the 8 mu m excess in all cases, but we have been unable to rule out the specific (but unlikely) case of completely optically thin cyclotron emission as the source of the observed 8 mu m flux density. While both model components can generate enough flux at 8 mu m, neither dust nor cyclotron emission alone can match the excess above the stellar components at all wavelengths. A model combining both cyclotron and dust contributions, possibly with some accretion-generated flux in the near-IR, is probably required, but our observed SEDs are not sufficiently well sampled to constrain such a complicated model. If the 8 mu m flux density is caused by the presence of a circumbinary dust disk, then our estimates of the masses of these disks are many orders of magnitude below the mass required to affect CV evolution.