Astron. Astrophys. 356, 200-208 (2000)

5. Conclusions

The overall metallicity in the atmosphere of HD 166473 is larger than for all the roAp stars analyzed spectroscopically in our series (Paper I to Paper IV and the present paper). We observe a similar abundance pattern:

Rare earth elements are overabundant and C and O are less abundant than in the solar atmosphere. Ti, Fe and Ni are relatively most deficient and Co the most enhanced elements in the group of iron peak elements.

One of the most interesting features in the iron group is the clear overabundance of Co which could be detected in all five roAp stars. However, an even higher Co overabundance was determined recently by Nishimura (1998) for the hot (non-ro)A0p star HR 9049.

A new and most striking result is the discovery of the anomalous line strengths of second ions of REE resulting in an abundance increase of up to +1.5 dex, compared to abundances obtained from lines of the first ions. The latter, however, were typically used for an analysis because of a serious lack of reliable values for second ions.

In agreement to Paper I to Paper IV the present study shows that roAp stars basically have similar abundances for up to 33 elements investigated in our sample, which is the most complete chemical investigation hitherto published for this group of stars. It is also the first one which discusses in detail the influence of the magnetic field on abundance analyses of CP2 stars. A comparison of rapidly oscillating to non-oscillating Ap stars does not reveal a large difference in the abundance pattern. However, one should keep in mind that the number of elements with published abundances for comparable non-roAp and "normal" stars is significantly smaller.

Oscillation periods of roAp stars are much shorter than for Scuti stars although they fall in the same classical instability strip. It has been speculated by Gautschy et al. (1999) that the H-He I zone at to 0.5 be responsible for driving roAp star pulsation. If true, the upper boundary comes close to the deepest line forming regions and hence one should expect some characteristic signatures in roAp star spectra (e.g., the observed 3rd REE spectrum anomaly?) which might allow the testing of Gautschy's excitation model (Gautschy et al. 1998). Spectra with high signal-to-noise ratio and high spectral resolution are needed for this purpose.

For our analyses we mostly had been using Kurucz's model atmospheres with scaled solar abundances. Strictly speaking, the abundance analysis procedure is not self consistent, because the determined abundances obviously do not follow the scaled solar abundance pattern. However, we are presently improving on this situation by calculating individual ODFs and starting to use them for computing more consistent model atmospheres. Furthermore, we intend to investigate the effects of a magnetic field on ODFs and on the structure of model atmospheres.

© European Southern Observatory (ESO) 2000

Online publication: March 28, 2000
helpdesk.link@springer.de