UV-optical classification of XTE J0052-723 (SXP 4.78; Swift J005139.2-721704) and evidence of substantial IR excess

ATel #12251; B. McCollum (American Univ.), S. Laine (Caltech/IPAC)
on 30 Nov 2018; 00:59 UT
Credential Certification: Bruce McCollum (mccollub@cua.edu)

Tweet

We report SED fitting of catalog data to the progenitor of the recent X-ray transient Swift J005139.2-721704 (ATel # 12209, 12219). Survey catalogs report magnitudes in GALEX NUV, U, B, V, Gaia Gbp and Grp, I, J, H, Ks, and the four WISE bands at 3.4, 4.6, and 12 um along with an upper limit at 22 um. The J, H, and Ks magnitudes differ by ~0.1 mag or less (two-sigma) between the measurements of 2MASS in 1997 and those of Kato et al. (2007, PASJ, 59, 615), and this object is flagged as not variable in the WISE catalog, so its IR SED appears to be stable across that period.

We fitted the SED to the TLUSTY O and B star grids (Hubeny & Lanz 1995, ApJ, 439, 875; Lanz & Hubeny 2003, ApJS, 146, 417; Lanz & Hubeny 2007, ApJS, 169, 83). The Teff was varied as a free parameter from 17,000 K to 31,000 K in 1000 K intervals, corresponding to sp. type O9 to B3 following the B1-2e classification of Coe et al. (ATel #12237). The metallicity was allowed to vary from 0.1 to 0.2 to reflect the SMC metallicity, and log g varied from 1.75 to 4.75. Av was varied from 0.1 to 4 to take into account possibly strong circumstellar reddening. A standard Galactic extinction law (R = 3.1) was used. The SMC extinction law can be significantly different, especially in the UV, and mainly in or near the bar, with no bump at 2175 A and rising more steeply into the UV. The GALEX NUV filter bandpass is 1771 - 2831 A. The GALEX FUV extinction for the SMC extinction law would be roughly 20% less than in the case of the Galactic law (Prevot et al. 1984, A&A, 132, 389). J005139.2-721704 appears to be outside the bar, but we compared fits using a Galactic law and fits with the GALEX FUV flux increased by 20% as a crude proxy for the SMC law.

Only a very poor fit can be obtained if the entire SED is used, because the IR segment does not match the UV-B segment. The best fit (chi-square = 0.47) is obtained from Teff = 27,000 +/- 1860 K, corresponding to a spectral type of ~B0.5 to B1.5, Av = 0.82 +/- 0.23 using the Galactic extinction law, and log g = 4.0 to 4.75 (two sigma range), while treating the emission in the I band and longward as upper limits, i.e. as an IR excess. Including the I magnitude in the photospheric fit gives a spectral type of O which is inconsistent with the spectrum of Coe et al. (ATel #12237), and doubles the chi-squared value from the value if I is omitted, so we conclude that the I band includes a significant contribution from an IR excess. A typical value for foreground extinction toward the SMC is Av ~0.12. The IR excess relative to the model SED is ~4x in the 5.8 um bandpass, and ~20x in the 12 um bandpass. The SED and model comparison plot is posted at http://spider.ipac.caltech.edu/staff/seppo/plots/J005139.2-721704-SED.png .

We note that the J-H and H-K colors are consistent with those of the well-studied Be star Delta Sco (B0-1IVe). The Delta Sco WISE W1-W3 = 1.5 +/- 0.45, and for J005139.2-721704 W1–W3 = 2.23 +/- 0.49 indicating a much larger IR excess than in Delta Sco. (Although Delta Sco is a binary, Halonen et al. 2008, PASP, 120, 498 showed that the secondary contributes much less than 10% of the IR flux in WISE bandpasses.)

In summary, we confirm the spectral class as B0.5 to B1.5 consistent with Coe et al. (ATel #12237), conclude that the star may be ascending into the subgiant stage up to log g = 4, and find a substantial IR excess from the I-band to ~22 um consistent with that of a Be star.

This publication uses data products from the Wide-field Infrared Survey Explorer (WISE), and the Two Micron All Sky Survey (2MASS). This publication makes use of VOSA, developed under the Spanish Virtual Observatory project supported from the Spanish MINECO through grant AyA2017-84089. This work uses data from the European Space Agency (ESA) space mission Gaia. This research has made use of the VizieR catalogue access tool, CDS, Strasbourg, France.