ForumSpringerAstron. Astrophys.
ForumWhats NewSearchOrders

Astron. Astrophys. 326, 608-613 (1997)

Previous Section Next Section Title Page Table of Contents

3. Results

3.1. Identifications

As reported in the previous section, six X-ray sources could be identified with GSC stars. In order to examine the error boxes more carefully, we inspected the Palomar red plates from the Digitized Sky Survey and cross-correlated the results with the catalogs of stars in the field of NGC 6940 by Vasilevskis & Rach (1957: VR stars) and by Larsson-Leander (1960: LL stars), who performed deep photometric studies of the region of sky around the cluster. We consider an identification if there is a star brighter than m [FORMULA] 14 (the limit of current optical studies) within the 90% error box. The proposed identifications are summarized in Table 2. In some cases (#3, 6, 7, 12 and 17) a fainter object (m [FORMULA] ) is also present in the error box, in some others (#2, 5, 16) there are few weak objects (m [FORMULA] 16-17), while for sources #1, 8, 10, 11, 14 and 15 no objects brighter than [FORMULA] are in the error box (although star VR 165 for source #1 and two stars of m [FORMULA] 15-16 for source #15 lie just outside the error boxes). Table 2 includes an additional object, HD 340540 (=SAO 88882), a bright K0 star of V=8.9 that appears neither in the catalog of Vasilevskis & Rach nor in that of Larsson-Leander. It is not a member of the cluster and its optical coordinates, as given in the SIMBAD database, differ by less than 10 [FORMULA] from the X-ray coordinates of the strong X-ray source detected at a larger off-axis angle (source #18 in Table 1). Its X-ray to optical flux ratio is consistent with coronal emission (see Stocke et al. 1991), as it is the case for all proposed identifications. We regard this star as the most likely optical counterpart of the X-ray source, although there are other objects with magnitude in the range 17-20 within 1 [FORMULA] from the X-ray position.


Table 2. Optical identifications for sources in the NGC 6940 field. The columns contain ID of the X-ray source, ID of the optical object in Vasilevskis & Rach (1957), V and [FORMULA] of the optical object from Larsson-Leander (1960) (for source #18 from SIMBAD), membership probability from Sanders (1972), distance, X-ray luminosity in the 0.1-2.4 keV band, orbital period, eccentricity, and remarks.

Four of the proposed counterparts are with high probability members of the cluster (Vasilevskis & Rach 1957, Sanders 1972, see also Table 2). A search for binaries among the red giants of NGC 6940 has been performed by Mermilliod & Mayor (1989), who studied 24 giants in this region of the sky, 20 of which should be members of the cluster. They found that six of these 20 giants are binaries with periods ranging between [FORMULA] and [FORMULA] days. Three out of four cluster members identified with our X-ray sources are among this list of binaries (see Table 2 and Fig. 1). The fourth X-ray source identified as a member of the cluster is also a giant studied by Mermilliod & Mayor, who quote a probability of 87% against binarity. The six binaries detected by Mermilliod & Mayor and the four X-ray sources identified with cluster members (three are coincident) are all red giants, with the position of the three more luminous binaries shifted toward bluer colors, within the Hertzsprung gap (see Fig. 1). This is probably due to a composite type: a red giant primary and an A- or F-type main sequence secondary, with mass ratio larger than 0.8 (Mermilliod & Mayor 1989).

[FIGURE] Fig. 1. Color-magnitude diagram for NGC 6940 (data from Larsson-Leander 1960, membership from Sanders 1972). Stars indicate binary systems (from Mermilliod & Mayor 1989), circles are X-ray detections with a member counterpart, crosses are X-ray detections with a non-member counterpart. The square is star VR114, that we suspect to be associated with the cluster (see text).

We identify source #17 with star VR114, classified as G8 III by Larsson-Leander (1960, LL407). This source is not considered member of the cluster on the basis of proper-motion measurements (Vasilevskis & Rach 1957, Sanders 1972). Assuming an absolute magnitude of Mv =0.7 (Gray 1992), zero interstellar absorption and the visual magnitude given in Table 2, we derive a distance of [FORMULA] pc, i.e. behind NGC 6940. However, given the distance obtained, the interstellar absorption cannot be ignored. In his study of NGC 6940, Larsson-Leander (1960) obtains values of E(B-V) between 0.2-0.3 for the stars of this sky region around these distances (see his Fig. 14). In particular, he obtains E(B-V)=0.23 for star LL459, that is very near star VR114 (LL407). Using this value for the interstellar absorption, we derive a distance of [FORMULA] pc (d [FORMULA]  pc for E(B-V)=0.3 and d [FORMULA]  pc for E(B-V)=0.2). This value are very similar to the distance of the cluster, d [FORMULA]  pc. Thus, although this star is not classified as member of the cluster, it could be somehow associated with it (e.g. it could be a run-a-way member). This is shown also in Fig. 1, where VR114 (square symbol) lies in the region of the giant stars belonging to the cluster. In the following, we will consider this source as a field star, possibly associated with the cluster.

NGC 6940 is located at low galactic latitude (l [FORMULA] ), with a total galactic interstellar absorption, estimated from radio data, of N [FORMULA] cm-2 (Dickey & Lockman 1990). To estimate the number of extragalactic sources expected in the field we assumed a typical power-law spectrum with photon index 2.0 and the quoted value of NH. With these parameters our limiting countrate of [FORMULA] 1 ct/ksec corresponds to a flux limit of 2.5 [FORMULA]  erg cm-2 s-1 (in the 0.4-2.5 keV band). From the log N-log S distribution derived from the ROSAT deep survey (Hasinger 1992), we estimate that roughly 10 extragalactic sources should be detected in our observation. This means that all our unidentified sources might be extragalactic.

3.2. X-ray properties

3.2.1. The cluster members

The photon distribution of all sources is rather hard, reflecting the relatively high absorption towards the cluster. The number of counts detected is however too low to produce meaningful hardness ratios. In order to estimate the X-ray luminosity in the ROSAT band for the detected members, we adopted a spectral model typical of RS CVn binaries. Following Dempsey et al. (1993a), who studied the full sample of RS CVn binaries detected in the RASS, we used a two temperature thin emission plasma model (according to Raymond & Smith 1977). The parameters used are the average of the values in Dempsey et al. (1993a): [FORMULA] = 0.175 keV, [FORMULA] =1.4 keV, [FORMULA] =6. For each star we used a value for the interstellar absorption NH between [FORMULA] and [FORMULA] cm-2, as derived from the E(B-V) reported by Larsson-Leander (1960). The derived X-ray luminosities, assuming a distance of 870 pc (Larsson-Leander 1964), are reported in Table 2. Although the detections correspond to the hard PSPC band (0.4-2.4 keV), the values are given for the full PSPC range 0.1-2.4 keV to allow comparison with other systems. We estimate that the uncertainties on the derived luminosities, due to a different value of the temperatures and/or EM ratio but still within the typical range observed for coronal sources, are up to 50%.

3.2.2. The non-member stars

Source #3 is identified with HD 334742, classified in the SIMBAD database as an F5V, source #4 is identified with HD 196244, classified in SIMBAD as A2V, while source #18 is identified with HD 340540, classified in SIMBAD as K0V. Assuming an absolute magnitude of Mv =3.5, 1.3 and 6.0 for the three stars (Gray 1992) and the visual magnitudes given in Table  2, we derived a distance of 216, 257, and 38 pc respectively. These values are obtained assuming a negligible interstellar absorption. While this is probably the case for the nearby star HD 340540, the derived distances for HD 334742 and HD 196244 should be considered as lower limits.

For source #18 we have enough counts to derive a hardness ratio (defined as HR=(H-S)/(H+S), see Sect. 2). The obtained value of HR=0.28 [FORMULA] 0.02 is high for a normal coronal source (e.g. Schmitt et al. 1995), but is commonly found in very active X-ray selected stars (Fleming et al. 1995). A two-temperature thermal fit yields N [FORMULA] cm-2, [FORMULA] keV, [FORMULA] keV, and [FORMULA] (90% confidence errors). These temperatures are in line with those derived by Dempsey et al. (1993a) for the ROSAT RASS sample of RS CVn binaries, although the dominating component in our case is the harder one, as already implied by the hardness ratio value. The resulting 0.1-2.4 keV luminosity is [FORMULA] erg s-1. This is also the only source that shows significant variability, whit a steady increase in countrate from 0.3 to 1.0 cts s-1 throughout the observation.

For the other two sources (# 3,4) we can neither perform a spectral analysis nor derive a meaningful hardness ratio. We derived their X-ray flux by adopting the conversion formula by Fleming et al. (1995), assuming HR=0. We estimate a maximum error of [FORMULA] from this procedure. The derived X-ray luminosities, reported in Table 2, should be regarded as lower limits, since by ignoring the interstellar absorption we are underestimating both the spectral distances and the X-ray fluxes.

Finally, source #9 is identified with a star listed in the catalog of Larsson-Leander (1960) that does not seem to be a member of the cluster (see Fig. 1). Not knowing its spectral type, we can neither infer the spectral parallax nor its X-ray luminosity. From its B-V (see Table 2), we suspect it to be an active K-M field star.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: October 15, 1997