Pre-outburstChandraobservations of the recurrent nova T Pyxidis

Abstract

I present a total of 98.8 ksec ($\sim$ 3$\times$30 ksec) observation of T Pyx with the ACIS-S3 detector on-board the $Chandra$ Observatory obtained during the quiescent phase, about 2-3 months before its outburst in April 2011. The total $Chandra$ spectrum of the source T Pyx gives a maximum temperature kT$_{max}$$>$ 37.0 keV with (0.9-1.5)$\times 10^{-13}$ erg s$^{-1}$ cm$^{-2}$ and (1.3-2.2)$\times$10$^{32}$ erg s$^{-1}$ (at 3.5 kpc) in the 0.1-50 keV range using a multi-temperature plasma emission model (i.e., CEVMKL in XSPEC). I find a ratio of (L$_{x}$/L$_{disk}$)$\simeq$(2-7)$\times$10$^{-4}$ indicating considerable inefficiency of emission in the boundary layer. There is no blackbody emission with 2$\sigma$ upper limits kT$_{BB}$$<$ 25 eV and L$_{soft}$$<$ 2.0$\times$10$^{33}$ erg s$^{-1}$ in the 0.1-10.0 keV band. All fits yield only interstellar N${\rm _H}$ during quiescence. I suggest that T Pyx has an optically thin boundary layer (BL) merged with an ADAF-like flow (Advection-Dominated Flow) and/or X-ray corona in the inner disk indicating ongoing quasi-spherical accretion at (very) high rates during quiescent phases. Such a BL structure may be excessively heating the WD, influencing the thermonuclear runaway leading to the recurrent nova events. The central source (i.e., the binary system) emission and its spectrum is deconvolved with a detailed procedure at the sub-pixel level revealing an extended emission with S/N $\sim$6-10. The derived shape looks like an elliptical nebula with a semi-major axis $\sim$1.0 arc sec and a semi-minor axis $\sim$0.5 arc sec. The calculated approximate count rate of the extended emission is 0.0013-0.0025 c s$^{-1}$. The luminosity of the nebula is $\sim$(0.6-30.0)$\times$10$^{31}$ erg s$^{-1}$. The nebulosity seems consistent with an interaction of the outflow/ejecta from the 1966 outburst.