Astron. Astrophys. 357, 920-930 (2000)

7. Discussion

The abundance analysis of the roAp star HD 83368 was carried out for the first time. We note again that our spectra are limited in spectral range and in number of rotational phases, therefore our abundance estimates can only have a preliminary character.

• Our calculation of the surface magnetic field shows that the Li I line intensity changes in phase with (Fig. 2 and 3b) and the profile depends on the position of the magnetic poles on the visible hemisphere. The insignificant offset between and radial velocity of the Li line can be explained by a small offset between the centre of Li spots and magnetic poles (see Polosukhina et al., 1999).

• Analysis of the behaviour of the Li I line with the rotational phase has allowed us to compute the size of the spots, their position on the star as well as their lithium abundance. The remaining disagreements between model and observed spectra can probably be explained by more complicated spot structure than two homogeneous circular (not necessarily opposite) spots, as we have assumed.

• Special attention was given to search for the third REE spectrum - Pr III and Nd III . It appears that these lines are rather strong (in comparison with Pr II and Nd II lines) and have a more complicated surface distribution than lithium. Possibly they form rings or groups of spots. For studying them new spectra and calculations including magnetic splitting of lines of rare earth ions are necessary.

• We are forced to recognize that some of our abundance estimates (in particular for La II , Ce II , Gd II , Er II ) are only upper limits due to the line weakness of these elements and the possibility of blending with other weak unidentified lines. The great uncertainty is also due to the complex spot distribution of rare earth elements, as we see from our preliminary analysis.

• The main problem of our research was initially the identification of the variable spectral feature at 6708 Å with the Li I line. New results from the observations of some roAp stars give additional indirect evidence for the identification of the line 6708 Å with the resonance doublet of Li I (Polosukhina et al., 1999).

• The research of Li problem in roAp stars is continuing. The work for reproducing the profile of the lithium blend with more thorough consideration of the lines of rare earth elements (probably their oxide's molecular lines also) and determination of the ratio ⁶Li/⁷Li is planned.

© European Southern Observatory (ESO) 2000

Online publication: June 5, 2000
helpdesk.link@springer.de