Search for a command to run...
Aims. We determine the temporal evolution of the luminosity (LWD), radius (RWD) and effective temperature (Teff) of the white dwarf (WD) pseudophotosphere of V339 Del from its discovery to around day 40. Another main objective was studying the ionization structure of the ejecta. Methods. These aims were achieved by modelling the optical/near-IR spectral energy distribution (SED) using low-resolution spectroscopy (3500–9200 Å), UBVRCIC and JHKLM photometry. Important insights in the physical conditions of the ejecta were gained from an analysis of the evolution of the Hα and Raman-scattered 6825 Å O <ns0:sc>vi </ns0:sc>line using medium-resolution spectroscopy (R ~ 10 000). Results. During the fireball stage (Aug. 14.8–19.9, 2013), Teff was in the range of 6000–12 000 K, RWD was expanding non-uniformly in time from ~66 to ~300 (d/ 3 kpc) R⊙, and LWD was super-Eddington, but not constant. Its maximum of ~9 × 1038 (d/ 3 kpc)2 erg s-1 occurred around Aug. 16.0, at the maximum of Teff, half a day before the visual maximum. After the fireball stage, a large emission measure of 1.0−2.0 × 1062 (d/ 3 kpc)2 cm-3 constrained the lower limit of LWD to be well above the super-Eddington value. The mass of the ionized region was a few × 10-4 M⊙, and the mass-loss rate was decreasing from ~5.7 (Aug. 22) to ~0.71 × 10-4 M⊙ yr-1 (Sept. 20). The evolution of the Hα line and mainly the transient emergence of the Raman-scattered O <ns0:sc>vi </ns0:sc>1032 Å line suggested a biconical ionization structure of the ejecta with a disk-like H <ns0:sc>i </ns0:sc>region persisting around the WD until its total ionization, around day 40. On Sept. 20 (day 35), the model SED indicated a dust emission component in the spectrum. The dust was located beyond the H <ns0:sc>i </ns0:sc>zone, where it was shielded from the hard, ≳105 K, radiation of the burning WD at that time. Conclusions. Our extensive spectroscopic observations of the classical nova V339 Del allowed us to map its evolution from the very early phase after its explosion. It is evident that the nova was not evolving according to the current theoretical prediction. The unusual non-spherically symmetric ejecta of nova V339 Del and its extreme physical conditions and evolution during and after the fireball stage represent interesting new challenges for the theoretical modelling of the nova phenomenon.