Astron. Astrophys. 329, 613-623 (1998)

5. Results

Our analysis procedure was as follows: we have determined effective temperatures and gravities for each supergiant for our two model helium abundances. These are listed in Table 1 - the number of significant figures provided are to illustrate the dependence of these quantities on the helium fraction and are not representative of the likely errors. For both stars, increasing the helium abundance reduces our estimate of the gravity which in turn reduces that for the effective temperature. As was implied by the synthetic profiles, our temperature and gravity estimates were effectively independent of the value adopted for the microturbulent velocity.

Table 1. Dependence of derived atmospheric parameters of our target stars on assumed helium fraction.

The He I lines have been calculated using the appropriate values of the atmospheric parameters for each value of y. In Table 2 we show the estimates of helium fraction, while in Figs. 3 and 4 we present our spectral line fits. The extrapolated values were obtained by linearly extrapolating the helium line profiles beyond the limits of our grid. In some cases (marked with a dagger) the extrapolation was relatively small and the estimates should be useful; in other cases (marked by colons), such an approach is clearly over-simplistic and the values are only intended to be illustrative. Estimates of the helium fraction from the line at 5015Å in both stars and from the line at 4713Å for  Ori (both for zero microturbulence) are not given, as the strength of these lines implied unphysically large values for a strict linear extrapolation, while allowing for line saturation (where W ) yielded typically 0.5.

Table 2. Helium fractions, y, implied by different He I lines. For each target, we give two estimates for y ; viz. that determined from an analysis performed with =0 kms^-1, and that determined using a value determined from the metal lines. In some cases the strength of the line led to an unphysically large value for the helium fraction, and y -estimates are not given - see text for further discussion.

Fig. 3a. Observed profiles and non-LTE calculations for singlet He I in Ori. The latter are for atmospheric parameters =29000 K, =3.06, y =0.1 and =28500 K, =3.04, y =0.2. Theoretical profiles for =0 kms-1 and =12 kms-1 are shown on the left and right hand side, respectively and are in order of increasing oscillator strength. The absorption in the red wing of 4922 is due to O II and there is also a diffuse interstellar band bluewards of the He I line at 4437Å.

Fig. 3b. As for Fig. 3a. - for the triplet lines of He I in Ori.

Fig. 4a. Observed profiles and non-LTE calculations for singlet He I in Ori. The latter are for atmospheric parameters =27500 K, =3.00, y =0.1 and =27000 K, =2.92, y =0.2. Theoretical profiles for =0 kms-1 and =12 kms-1 are shown on the left and right hand side, respectively and are in order of increasing oscillator strength.

Fig. 4b. As for Fig. 4a. - for the triplet lines of He I in Ori.

© European Southern Observatory (ESO) 1998

Online publication: December 8, 1997
helpdesk.link@springer.de