Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 322, 785-800 (1997)

Previous Section Next Section Title Page Table of Contents

7. Summary and Discussion

We have investigated the coronal activity of a sample of 86 Pop II field binaries using data from the RASS. Only 13 Pop II systems were detected; another system is a possible detection. X-ray luminosities and hardness ratios were derived for the X-ray detections, and upper limits were placed on X-ray luminosity for the non-detections. The key results of this study can be summarised as follows:

(i) The XLDF of the entire sample, taking into account both detections and upper limits, has a median luminosity [FORMULA] erg s-1, indicating that the Pop II binaries typically have rather low X-ray luminosities. For a subsample of emission Pop II systems, which may be considered as the old, metal-poor analogs to the RS CVn binaries, the median is at [FORMULA] erg s-1. For comparison, a complete sample of RS CVn binaries detected in the RASS has a median luminosity [FORMULA] erg s-1. Thus, a complete sample of emission Pop II binaries would be at least one order of magnitude less X-ray luminous than the RS CVn binaries.

(ii) The high-luminosity tail of the XLDF at [FORMULA] erg s-1 demonstrates that Pop II binaries can maintain coronal activity levels as high as those of the RS CVn binaries.

(iii) Twelve out of the 13 X-ray detections have intermediate metallicities with [FORMULA], while there is only one extreme metal-poor detection (HD 89499). This possibly indicates that extreme Pop II binaries have typically lower X-ray luminosities than intermediate Pop II. Further, HD 89499 has a higher hardness ratio, and hence higher coronal temperature, than the intermediate Pop II systems typically have. This high coronal temperature is consistent with the idea that extreme metal-poor coronae can release their energy only by thermal bremsstrahlung, which is efficient above 20 million K.

(iv) The X-ray luminosity is not very well correlated with the rotational period. Specifically, only 6 (out of the 20) SP systems and 8 LP systems were detected. Thus, Pop II binaries can be strong X-ray emitters despite having long periods, a fact well known from Pop I binaries, especially if they are evolved. On the other hand, Pop II binaries with short orbital periods are not necessarily strong X-ray emitters. In contrast, SP Pop I binaries are typically thought to be strong X-ray emitters.

As the emission Pop II systems cover the same range of spectral types, luminosity classes, and orbital periods as the RS CVn systems, but differ distinctly in metallicity and age, it is reasonable to associate their lower median X-ray luminosity with their lower values of metallicity, significantly higher ages, and probably lower fraction of evolved stars. The evolved Pop II systems in our sample have high X-ray luminosities of [FORMULA] erg s-1, while the dwarf Pop II systems have lower luminosities of [FORMULA] erg s-1. In contrast, the X-ray luminosities of the RS CVn binaries are, to first order, independent of the luminosity class; the dwarf RS CVn systems also possess X-ray luminosities of [FORMULA] erg s-1 (cf. Fig. 3 in Dempsey et al. 1993). Thus, the lower median X-ray luminosity of the emission Pop II binaries may be explained in part by their lower fraction of evolved systems.

A comparison between RS CVn and emission Pop II binaries on the one hand, and between intermediate and extreme Pop II binaries on the other hand, indicates that the median X-ray luminosity possibly decreases with photospheric metallicity and/or age (a low photospheric metallicity also indicates a high age). Nevertheless, the extreme metal-poor, old system HD89499 is as X-ray luminous as the RS CVn systems. This suggests that, if at all, there is no linear relation between activity and metallicity/age. The relation between activity and photospheric metallicity might reflect primarily the radiation efficiency, which depends on coronal metallicity and temperature. The metallicities in stellar coronae may not be the same as in the photospheres. For example, in the Sun, the low first ionization potential (FIP) elements (Fe, Ca, Mg,...) are probably enhanced by a factor 4 in the corona compared to the solar photospheric values (FIP-effect; Meyer 1985). In contrast, the FIP-effect seems to be absent in Procyon (Drake et al. 1995). In the RS CVn binaries, the photospheric Fe abundances are possibly slightly sub-solar (cf. Sect. 5). Recent EUVE and ASCA observations have shown that the coronal Fe abundances of RS CVn and Algol-type binaries appearently are below the solar photospheric values by factors 2-4 typically (e.g., Schrijver et al. 1995, Singh et al. 1995, Stern et al. 1995, Singh et al. 1996), although these abundance determinations might be affected by the non-negligible optical thickness of the lines (e.g., Schrijver et al. 1994). Finally, in the Pop II binary HD 89499, the ASCA spectrum clearly shows the absence of emission lines (Fleming and Tagliaferri 1996). These abundance determinations indicate that the photospheric and coronal abundances typically do not differ by too large a factor, and that the photospheric metallicity may be used as a rough measure for the coronal metallicity. Therefore, for the RS CVn systems with slightly sub-solar photospheric metallicities, line emission is rather efficient in releasing the coronal thermal energy. Assuming that the intermediate Pop II systems with [FORMULA] have correspondingly lower coronal metallicities than the RS CVn, then the line emission is reduced in its efficiency. This might explain the lower activity levels of the intermediate Pop II systems. Extreme metal-poor systems are expected to radiate away their energy only via thermal bremsstrahlung, which requires extremely high temperatures to be efficient. Therefore, the extreme Pop II systems may have either high activity levels if their coronal heating rates are extremely high, or low activity levels even if their heating rates are moderately high.

For single stars, the effect of stellar age is primarily to decrease angular momentum due to magnetic braking, resulting in a decline in X-ray activity. Binary systems can maintain high rotation rates despite old age due to their synchronous rotation. Therefore, SP binaries are expected to have high activity levels, no matter how old they are. The Pop I SP binaries indeed are known to be strong X-ray emitters (e.g., RS CVns, Algols). An exception to this trend may be provided by the SP binaries of the old open cluster M67, for which a rather low detection rate (i.e., 1/11), indicating low coronal activity, was found by ROSAT observations (Pasquini and Belloni 1994). In contrast, the Pop II SP binaries are not per se strong X-ray emitters. Surprisingly, even some of the emission Pop II SP systems have rather low X-ray luminosities: For the detected SP system HD 22694, an X-ray luminosity [FORMULA] erg s-1 was derived. For the non-detected SP systems CD-48 1741 and BD+53 080, upper limits of [FORMULA] erg s-1 and [FORMULA] erg s-1 were determined, indicating X-ray luminosities below [FORMULA] erg/s despite their quite high rotational velocities of [FORMULA] and 7 (cf. Table 3). These luminosities are lower than expected from the rotation-activity relation for single stars (Pallavicini et al. 1981). The low coronal activity of CD-48 1741, BD+53 080, and possibly of further non-detected Pop II binaries might be associated with their long stellar ages, over which time the magnetic field may not have survived.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 5, 1998