EUVE Finds Evidence for Stratification of N Abundance in White Dwarf REJ 1032+532 -------------------------------------------------------- EUVE spectrum of RE J1032+532 showing the SW and MW spectra. The stratified model (upper curve) fits very well the data while the homogeneous model computed with a nitrogen abundance (log N/H = -4.31) measured from UV NV lines don't (lower curve). EUVE spectrum do not show any spectral features from nitrogen and its flux level is nearly consistent with that of a pure hydrogen atmosphere. RE J1032+532 exhibits strong NV resonance lines in the HST/STIS spectrum implying a large abundance of log (N/H) = -4.3. EUVE and HST/STIS spectra can be fit simultaneously by assuming a stratification of nitrogen in the atmosphere for which nitrogen is concentrated only in the upper layers (< log ?M/M = -15.5). Such a peculiar profile may be the result on an interplay between mass loss and diffusion. Notes: Dr. Jay Holberg (University of Arizona) and his collaborators are about to publish in the Astrophysical Journal an important discovery concerning REJ 1032+532, a hot white dwarf (Teff=46,330 K, log g = 7.781) first detected during the ROSAT survey and subsequently observed by HST and EUVE. The HST/STIS spectrum revealed the presence of strong NV resonance lines (observed at 1238.9785 and 1242.9688 Å) corresponding to an abundance of log N/H = -4.3, which is unusually large for a white dwarf of this temperature. Holberg et al. computed an extreme ultraviolet spectrum assuming this abundance and find the flux would be strongly suppressed below 260 Å by nitrogen EUV opacity. Instead, the EUVE spectrum looks very much like a pure hydrogen spectrum and is not consistent with such a large abundance of nitrogen. How could the EUVE spectrum be so different from the prediction? The answer may reside in nitrogen abundance stratification according to Holberg et al. They have demonstrated that a model in which the abundance of nitrogen is set to log N/H = -4.3 only in the upper atmospheric layers starting at a mass fraction of (M/M = 3.1x10-16 will reconcile the EUV and UV spectra. The authors state this profile is not an unique solution or is a definitive proof of stratification of nitrogen abundance in RE J1032+532. However, their model also provides a better fit of the N V line profiles observed with STIS. Such a peculiar abundance profile may be explained by the interplay of mass loss and diffusion in the atmospheric layers of a white dwarf. Exploratory calculations by Chayer et al. (1997) on silicon indeed produce similar abundance stratification for moderate mass loss rate on the order of 10-14 solar mass/year. Further calculations are required for nitrogen to confirm this scenario. A stratified abundance profile of iron also appears to provide a better fit of the EUVE spectrum of G191-B2B (Barstow et al. 1999, MNRAS in press).