Astron. Astrophys. 319, 637-647 (1997)

6. Conclusion

The energy distribution longward 1650 Å  is well fitted by the model with parameters = K, , and solar scaled abundances for all the elements. However, the observed UV excess shortward 1650 Å , where the Si discontinuities occur, cannot be explained with Si underabundance, because abundance analyses by Adelman (1973) and Savanov & Malanushenko (1990) yielded about solar abundance for it. Our abundance analysis has given solar abundance as upper limit for Si (Table 4).

A more plausible possibility is that the companion is a  Boo star about visual magnitude fainter than   , metal poor () and fast rotating, in order to explain the fact that no spectral lines of   are observable. From the mass function , (Kamper et al., 1990), assuming a mass ratio R = 1.7, as suggested by the magnitude difference of about , we find a mass of 2.16 solar masses for   , and of 1.27 for   , with a total mass of the system of 3.43 in agreement with the value given by Kamper et al. of 4.3    2.0  .

From Table 4, we note that there is generally good agreement among the abundances derived by Hack (1958), Adelman (1973), Savanov & Malanushenko (1990), and the present work. Some relatively small discrepancies must be partly imputed to the variability of the equivalent widths of several elements and partly to different atmospheric models and different values of the used in the determination of the abundances. One can synthesize the chemical peculiarities of  CrB as follows: a defect of light elements like C , N , Al , possibly Si , an excess by a factor of about 10 for the iron group, and excesses by factors ranging from 10² to 10⁴ for heavy elements and Rare Earths.

A possible explanation for the behavior of the Li blend is the spotted distribution of lithium and other components of the blend. The lists of lines in the Li region given by Gerbaldi et al. (1995) and by Burkhart & Coupry (1991) suggest that the influence on lithium doublet of blending lines should be minor. However, in the case of  CrB the feature at 6708 Å  cannot be explained by assuming an anomalous large Li⁶ /Li⁷ ratio, but rather by assuming that Li is blended with some unidentified line, or possibly with the V I line at . However, in this last case, a very unlikely vanadium overabundance by a factor has to be proved. Another source of uncertainty is the not sufficient precision in the wavelength scale. We have made correction to the wavelength scale after a critical analysis of the binary radial velocity curve. The accuracy of the value of the - velocity is of 1.4 - 2.5 km s^-1. New more precise determinations of the orbital parameters should be necessary.

The complex relations between the equivalent widths, the FWHM, and the radial rotational velocity versus the rotational phase suggest that the various elements are concentrated in one or several spots. However much more observations are necessary, in order to cover completely the rotational period. Moreover a better knowledge of the law of variation of the equivalent widths of several elements is needed for a more accurate determination of the abundance peculiarities. However this is the first indication and observational evidence of "spotness" for a slowly rotator cold SrCrEu star with a large magnetic field.

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998
helpdesk.link@springer.de