next previous
Up: Blue stragglers in open clusters


4 Elemental abundances

The atmosphere models of Kurucz (1991) and oscillator strengths for the lines of interest from the VALD data base were used to determine individual elemental abundances in conjunction with the LTE spectrum synthesis method (SYNSPEC, Hubeny et al. 1994). For all the program blue stragglers we adopted the microturbulent velocity $V_{\rm t}=3$ kms-1(the same value as adopted in Andrievsky et al. 2000 - Paper II), while for Vega we used the literature value $V_{\rm t}=2$ kms-1(see, e.g. Sadakane & Nishimura 1981). For the sake of completeness, we also calculated the abundances with the smaller value of $V_{\rm t}=0.6$ kms-1 recommended by Adelman & Gulliver (1990). Note that our implementation of the SYNSPEC code enables one to calculate only the spectra of chemical elements with $Z \le 30$, therefore, the lines of the heavier species (if observed) were ignored. The steps of the analysis mentioned above are described in more detail in Paper II.

There is much evidence that Vega is a mild metal deficient star, therefore, we used the atmosphere model selected from the grid with $[\rm A]=-0.5$ for its spectrum synthesis.

Elemental abundances for Vega and our program blue stragglers are given in Tables 3 and 4. For Vega we also give a comparison with the most recent study by Hill (1995) also based on the spectrum synthesis technique. It should be stressed that with this study we did not aim to enlarge the number of precise and detailed spectroscopic investigations of Vega. Our main reason was to have some external indication of the reliability of the analysis based on our homogeneous spectroscopic material. Although the number of lines used is limited, the agreement in the derived abundances with Hill (1995) is satisfactory.

All of our high-probability proper-motion members show a strong deficiency of magnesium and scandium. For example, in Figs. 1 and 2 the synthetic and observed spectra for program stars in the vicinity of the Mgii 4481 Å line are shown. The synthetic spectra were calculated with the parameters listed in Table 2 and magnesium abundances for individual stars from Tables 3 and 4.

The detected magnesium deficiency for blue straggler stars cannot be removed by any reasonable changes in the atmospheric parameters, because within the temperature region 8000 K-10000 K, the magnesium line 4481 Å appears to be practically insensitive to temperature and gravity variations. This was also mentioned by Holweger, Gigas & Steffen (1986) who performed a qualitative search for abundance indicators in early A stars which are both temperature and gravity insensitive. As an example, we have calculated the magnesium abundance for K1270 assuming temperature and gravity uncertainties of about $\pm 500$K and $\pm 0.5$ dex respectively. The resulting changes in the abundance appeared to be negligible, and even smaller for temperatures higher than 8500 K. The response of the magnesium abundance (as derived from the Mgii 4481 Å line) caused by the parameter variations is given in Table 5. The $\Delta$ means the difference between the magnesium abundance [Mg/H] derived for K1270 with its basic model parameters (see Table 2) and with varied parameters (e.g., 7750/4.0/3.0 denotes model with $T_{\rm eff}$ = 7750K, $\log g = 4.0$ and $V_{\rm t}\ = 3.0$ km s-1).


 

 
Table 3: Elemental abundances for Vega

Present Paper Hill (1995)

0.6 kms-1 2.0 kms-1  

El.
[El/H] [El/H] $\sigma$ N [El/H]

Mg
-0.10 -0.26 - 1 -0.27
Ca -0.63 -0.73 - 1 -0.47
Sc -1.27 -1.41 - 1 -
Ti -0.30 -0.64 0.09 7 -0.46
Cr -0.25 -0.34 - 1 -0.48
Fe -0.18 -0.46 0.11 7 -0.54



   
Table 4: Relative elemental abundances [El/H] for program stars in the field of NGC 7789

Star
He Mg Ca Sc Ti Cr Fe

K88
  -0.70 (-, 1) -0.25 (-, 1) -1.11 (-, 1) -0.28 (0.18, 5)   -0.12 (0.20, 5)
K316   -1.04 (-, 1) -0.53 (-, 1) -1.31 (0.18, 4) -0.41 (0.19, 3) +0.09 (-, 2) -0.23 (0.19, 11)
K371   -0.20 (-, 1)   -0.77 (-, 1)     -0.13 (0.23, 4)
K409   +0.00 (-, 1)          
K677 +0.07 (-, 1) +0.16 (0.18, 5)     -0.24 (0.22, 6)   -0.08 (0.17, 7)
K746   -0.64 (-, 1) -0.14 (0.17, 4) -0.92 (0.25, 5) -0.23 (0.19, 15) +0.03 (0.16, 4) -0.13 (0.19, 13)
K1211 -1.00 (-, 1) -1.04 (-, 1)         -0.10 (0.23, 4)
K1270   -1.14 (-, 1) -0.71 (0.29, 4) -2.03 (0.19, 5) -0.39 (0.24, 11) +0.00 (0.18, 4) -0.20 (0.18, 15)

Given in the brackets are s.d. and number of used lines respectively.


 

 
Table 5: Parameter variation and Mg abundance changes for K1270

7750/3.0/3.0 7750/4.0/3.0 8750/3.0/3.0 8750/4.0/3.0 8300/3.6/2.5 8300/3.6/3.5

$\Delta$
-0.12 +0.18 -0.05 +0.05 +0.09 -0.05


For the iron-group (Ti, Fe), the abundances for the blue stragglers do not differ from those determined for other stars in NGC 7789. Tiede et al. (1997) estimated the metallicity of NGC 7789 by IR photometry of the giant branch to be $[\rm Fe/H] = -0.62$. Recently, Vallenary et al. (2000) revised this value by an improved method using new IR photometry and found $[\rm Fe/H] = -0.25 \,\pm \,0.11$. Friel & Janes (1993) carried out low-resolution spectroscopy for several giants and obtained a mean metallicity of $[\rm Fe/H] = -0.26 \pm 0.06$. Pilachowski (1985) performed a high-resolution spectroscopic investigation of six giant stars and found $[\rm Fe/H] = -0.1 \pm 0.2$, in excellent agreement with our mean iron abundance (-0.16 dex).

As to other elements, there is only one indication in the literature that the relative-to-solar abundances of atomic species like Ca, Sc, Ti, etc. scatter only slightly about that of iron (Pilachowski 1985). Our results for titanium and chromium agree very well with those of Pilachowski, that of calcium only marginally. The only really discrepant case is for scandium where our blue stragglers appear to be deficient between a factor of 10 to 100. It is known, however, that A-type stars very often show peculiarities of certain elements (see Sect. 5.2).


  \begin{figure}
\par\includegraphics[width=6cm,clip]{f1r.eps}\end{figure} Figure 1: Observed ( thin line) and calculated ( thick line) fragment of the spectra in the vicinity of Mgii 4481 Å line


  \begin{figure}
\par\includegraphics[width=6cm,clip]{f2r.eps}\end{figure} Figure 2: Same as Fig. 1


next previous
Up: Blue stragglers in open clusters

Copyright ESO 2001