3. Results for Praesepe
The stars on the upper main sequence of Praesepe have not received much attention so far as concerns radial-velocity measurements. Raboud & Mermilliod (1998) have summarized the available data from the literature on radial velocities. Table 3 contains a summary of the orbital elements we have determined. The individual binaries are discussed in more detail in the subsections below.
Table 3. Orbital elements of Am spectroscopic binaries in Praesepe
3.1. KW 40 (HD 73174)
The observations of KW 40 (A4/A8/F2, Abt 1986) show a single line spectroscopic binary (SB1) with a period of 597. Recently Abt & Willmarth (1999) published the orbital elements for a triple stars system. However, early attempts to compute an orbit produced residuals larger (2.98 km s-1) than the measurement errors (0.52 km s-1) which could be explained by a change in the systemic velocity. Therefore, this star was continuously monitored, from the end of 1979 to 1997 to follow the variation of the systemic velocity. Only the Am primary is visible. All efforts to detect a correlation for any of the two other components were unsuccessful.
The spectroscopic orbit, represented in Figs. 1 and 2, is solved by taking into account the two periods. Thus the radial velocities of the short period () are corrected by the motion of the center of masses to compute the short solution and these corrections are used to solve for the long period (). If the short period solution agrees with Abt & Willmarth (1999), our value for the long period system is twice as large as their value. An attempt to plot our observations in phase with their period failed. Therefore the correct value is
Assuming a mass of 2.0 for the Am primary, we get a minimum mass of 0.68 for the secondary of the spectroscopic orbit. Taking a minimum total mass of 2.7 for the binary, the minimum mass of the third star is again about 0.63 .
In their speckle investigation of Praesepe, Mason et al. (1993a) have observed KW 40 in 1991, just at the phase where the two components of the long period system had a large difference in radial velocitiy. This implies the two components of the long-period system were close and no evidence of the third component was seen.
KW 40 was detected in X-ray from ROSAT PSPC pointing (Randich & Schmitt 1995) with a luminosity as strong as that of solar-type stars. The observed flux may be the sum of the contributions of the two unseen companions or may result from the effect of the short binary period on the Am companion.
3.2. KW 229 (HD 73619)
Sanford (1931) has already determined an orbit (9117, ) for the double-lined system KW 229 (A3/A7/F0, Abt 1986). The present orbital solution is in very good agreement with that found by Sanford (1931), although we obtained an eccentricity () somewhat larger than Sanford's value (), but in good agreement with Abt & Willmarth's (1999) value (). The residual dispersion ( = 0.76 km s-1) is quite satisfactory for this type of star. The mass ratio is . Because both components are observed, we infer that both are Am stars.
KW 229 and KW 224, another double-lined Am which could not be observed with CORAVEL, were detected in X-ray from ROSAT PSPC pointing (Randich & Schmitt 1995). Their luminosities ( = 29.94) place them among the three strongest sources detected in Praesepe, just behind the eclipsing binary TX Cnc (KW 244). Because both components seem to be quite similar, the observed flux cannot originate from a solar-type companion. The flux enhancement may arise from the binary short periods, although that for KW 224 is not known.
3.3. KW 279 (HD 73709)
KW 279 was classified Am (A2/A5/F0) by Gray et al. (1989), but was found photometrically Ap by Maitzen & Pavlovski (1987) according to the index (). The Geneva peculiarity index gives an ambiguous answer: = 0.001 is only a few thousands of magnitude larger than the average of normal stars.
This star is also considered as the third component of the visual quadruple system ADS 6921 (Worley 1996). We found that KW 279 is a single-lined spectroscopic binary, with a short period of 7.22 days, which is in good agreement with the determination of Abt & Willmarth (1999). Two additional radial velocities were taken during the survey for magnetic fields of Ap stars with the spectrograph Elodie (Babel et al. 1995, 1997) and were taken into account in the final solution.
KW 279 is an extremely interesting star because it shows at the same time the characteristics of both an Am and an Ap star, and presents a strong magnetic field (North 2000). It has a reliable Am classification and positive : it was generally accepted that Am stars never show enhanced values (Maitzen 1976, Maitzen et al. 1998) which are characteristic of magnetic Ap stars only. Conversely, large-scale magnetic fields are generally not found in Am stars, with the probable exception of the hot Am star o Peg (Mathys 1988; Mathys & Lanz 1990). The two spectra taken with the Elodie spectrograph consistently show a surface magnetic field of about 7.5 kG which seems very significant, in spite of a relatively large projected rotational velocity km s-1 (North 2000).
A system formed by an Ap + an Am could explain the presence of the magnetic field. Such systems have been found by Abt & Cardona (1984). However, the mass function is small: = 0.0223 and if we assume that the star rotation is synchronised with the orbital period, because 0, the inclination angle 70.5 o and the secondary mass is 0.51 . This rules out the hypothesis and rejects the presence of an Ap star as primary or secondary and implies that the magnetic field is related to the Am star.
3.4. KW 538 (HD 73045)
KW 538, classified Am (A3/A9/F3) by Abt (1986), is a single-lined spectroscopic binary, with a period of 435.571 days. The data were collected from 1979 to 1996, independently by (JCM) and by (JMC), which explains the large number of observations obtained for this star.
KW 538 belongs to the scarce sample of Am stars which have an orbital period between 50 and 800 days (Budaj 1996). The amplitude of the radial-velocity variation is still quite comfortable for CORAVEL, but may require good precision measurements to detect it with classical spectrographs and a long-term observing program.
© European Southern Observatory (ESO) 2000
Online publication: February 25, 2000