Astron. Astrophys. 356, 146-156 (2000)

7. Conclusions

The magnetic chemically peculiar star HR 1094 has been investigated using optical and ultraviolet spectral data from the viewpoints of its elemental abundances and the stellar properties rotation velocity and magnetic field strength. A common theme has been that the choice of spectral lines is an important factor in both the interpretation and systematic reduction of errors. By choosing spectral lines that are little affected by Zeeman broadening, we have determined the rotational velocity of HR 1094 to have an upper limit of 17 km s^-1. It has been shown that an earlier determination (Sadakane, = 10 km s^-1) does not reproduce observed line widths. This increase in rotational velocity implies that the value of the inclination, i, is underestimated by Hill & Blake.

Abundance enhancements of the elements chlorine and cobalt are confirmed to be at levels unusually high, apparently even for chemically peculiar stars. However, only a relatively small number of magnetic B-type stars have been scrutinized for their abundance distributions relative to their Ap type counterparts and it therefore remains unclear as to the prevalence of the enhancements for lines of these elements. One can speculate as to whether a strong cobalt line enhancement is related to the presence of a magnetic field. The two cobalt-strong B-type stars (HR 1094, HR 5049) are known to be magnetic, and the cooler roAp stars also show strong cobalt line enhancements (Gelbmann 1998). It may therefore be of some merit to investigate a threshold CoII line strength that identifies the presence of a measurable magnetic field.

The inclusion of additional elements in the abundance analysis, particularly the heavier elements, presents an abundance distribution that is familiar to many chemically peculiar stars: deficiencies for light elements such as helium and nitrogen, enhancements of the iron-group elements, with an still increased abundances for heavier elements. The influence of the magnetic field upon the abundances has not been accounted for in this analysis and may thus lead to overestimates for spectral lines where the Zeeman broadening is particularly large. Likewise, the mixture of spectral data from different phases may also contribute to deviations of the general abundance pattern from that at any specific phase. However, we note that only for the element gadolinium do we notice a distinct difference in the abundance at different rotational (magnetic) phases. Yet even this difference may be influenced by the goodness of the oscillator strengths of the lines employed in the analysis, as generally weaker lines were observed during 1998 than in 1995.

We have investigated the magnetic field strength of HR 1094 by including the Zeeman components of several spectral lines in a spectrum synthesis analysis. The relative depths of the FeII 6147 and 6149 lines are well fit by a field strength of 6 kG, in contrast to the value of 2 kG at the same phase based upon polarization measurements addressing the field longitudinal component. Our fitting of spectral lines is somewhat analogous to measurement of the mean magnetic field modulus (Mathys 1990) for stars presenting resolved magnetic components. In the case of HR 1094 the rotational velocity is high enough to blend the Zeeman pattern for FeII 6149 and we extract the field strength from the relative depths of the two FeII lines. Confidence in this approach is garnered from the good fit to these lines for the star HR 5049 when using the mean magnetic field value of Mathys et al. Mathys et al. (1997). Providing that the instrumental resolution and stellar rotational velocity are known, the synthetic spectrum fitting procedure can also provide plausible results for marginally resolved Zeeman components by inherently accounting for shifts of the observed component wavelengths that result from the overlapping profiles of the individual components.

© European Southern Observatory (ESO) 2000

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