Hubble Space Telescope spectroscopy of V471 Tauri:: Oversized K star, paradoxical white dwarf

We have used the Goddard High Resolution Spectrograph on board the Hubble Space Telescope to obtain Ly alpha spectra of the hot white dwarf (WD) component of the short-period eclipsing DA+dK2 precataclysmic binary V471 Tauri, a member of the Hyades star cluster. Radial velocities of the WD were determined from eight post-COSTAR spectra obtained near the two quadratures of the orbit. When combined with ground-based measurements of the dK velocities, eclipse timings, and a determination of the dK star's rotational velocity, the data constrain the orbital inclination to be i = 77 degrees and yield dynamical masses for the components of M(WD) = 0.84 and M(dK) = 0.93 M(circle dot). Model atmosphere fitting of the Ly alpha profile provides the effective temperature (34,500 K) and surface gravity (log g = 8.3) of the WD. The radius of the dK component is about 18% larger than that of a normal Hyades dwarf of the same mass. This expansion is attributed to the large degree of coverage of the stellar surface by starspots, which is indicated by both radiometric measurements and ground-based Doppler imaging; in response, the star has expanded in order to maintain the luminosity of a 0.93 M(circle dot) dwarf M(circle dot). The radius of the WD, determined from a radiometric analysis and from eclipse ingress timings, is 0.0107 The R(circle dot). The position of the star in the mass-radius plane is in full accord with theoretical predictions for a degenerate carbon-oxygen WD with a surface temperature equal to that observed. The position of the WD in the H-R diagram is also fully consistent with that expected for a WD with our dynamically measured mass. Both comparisons with theory are probably the most stringent yet made for any WD. The theoretical cooling age of the WD is 10(7) yr. The high effective temperature and high mass of the WD present an evolutionary paradox. The WD is the most massive one known in the Hyades but also the hottest and youngest, in direct conflict with expectation. We examine possible resolutions of the paradox, including the possibility of a nova outburst in the recent past, but conclude that the most likely explanation is that the WD is indeed very young and is descended from a blue straggler. A plausible scenario is that the progenitor system was a triple, with a close inner pair of main-sequence stars whose masses were both similar to that of the present cluster turnoff. These stars became an Algol-type binary, which merged after several hundred million years to produce a single blue straggler of about twice the turnoff mass. When this star evolved to the asymptotic giant branch phase, it underwent a common envelope interaction with a distant dK companion, which spiraled down to its present separation, and ejected the envelope. We estimate that the common envelope efficiency alpha (CE) parameter was on the order of 0.3-1.0, in good agreement with recent hydrodynamical simulations.