Astron. Astrophys. 358, 910-922 (2000)

5. NGC 6752

Two sources have been detected in the core and two near the core of NGC 6752 in a ROSAT HRI observation obtained in 1992 (Grindlay 1993); close to one of the core sources, two candidate cataclysmic variables have been identified on the basis of H emission and variability on (presumably orbital) periods of 5.1 and 3.7 hrs (Bailyn et al. 1996).

5.1. Data analysis and source list

Two more observations of NGC 6752 have been obtained by us. Because the three observations have comparable length, we add them into a combined image which we analyse and use as a reference for the individual observations. To add the three observations we use the method outlined by Verbunt & Hasinger (1998), as follows. First we correct the data for each observation separately for the changed pixel size (see Sect. 2), analyse the resulting images and determine the offsets between sources detected in separate observations. Averaging these offsets we find (on the basis of sources X 3, X 13, X 4, and X 6) that the X-ray coordinates of the 1996 observation have to be shifted by in right ascension and in declination to be brought in line with the 1992 data. Similarly, the 1995 data (on the basis of the same sources plus X 14) must be shifted by , . We apply these corrections to the pixel coordinates of the photons, and then add the three images into a combined image, which is analysed in the standard way. The resulting list of sources in given in Table 6.

Table 6. X-ray sources detected in NGC 6752 (, kpc, Djorgovski 1993) with the ROSAT HRI, for the standard analysis of the whole field, and separately for the multi-source analysis of the central area. Numbers up to 15 are sources from Johnston et al. (1994), higher numbers are new. All X-ray positions have been corrected for boresight. Identifications with letter on the right refer to Grindlay (1993). The positions of the center of the cluster (GC, Djorgovski & Meylan 1993), it core radius and half-mass radius (Trager et al. 1993) and the positions of some optical objects discussed in the text are also listed.

We identify two sources with stars with accurate positions: X 19 is close to TYC 9071 228 1 (CD-60 7128), a star with and , and X 11 is close to HIP 94235/HD 178085, a G0V star with and . The latter identification was suggested already by Johnston et al. (1994) on the basis of the PSPC observation. The chance probability of finding a counterpart at these optical brightnesses is small, and we consider both identifications secure. The X-ray position of X 11 is affected by its proximity to the edge of the HRI detector. For this reason we use X 19 to tie the X-ray to the optical coordinates. X 19 is not found in any of the three individual observations, showing up only in the combined frame. It is a relatively weak source and its position accordingly has an error of both in right ascension and in declination. The shift required to bring X 19 in coincidence with TYC 9071 228 1 is given in Table 1, and has been applied to all positions of the X-ray sources; the resulting positions are listed in Table 6. (The remaining offset between X 11 and HIP 94235 is within the nominal error for the right ascension, and within 2-sigma for the declination: note that the error is composed of the statistical uncertainties in the positions of both X 19 and X 11.)

The detection limit in the total observation is about . An area with radius in the ROSAT Deep Survey contains 25 sources brighter than this limit; we thus expect to find 0.6 in the region within the half-mass radius of NGC 6752, . The sources in the core thus probably belong to the cluster, and possibly the two sources X 6/A and X 14/D as well.

5.2. Sources in the center of the cluster

Analysing the central source with the method described in Sect. 2, we find four significant sources (the increase in is 29 both for the third and for the fourth source). This adds two sources to the two already described by Grindlay (1993). The position and fluxes of these sources are listed in Table 6; Fig. 7 shows the positions and X-ray contours of the center of NGC 6752, together with the positions of the two candidate cataclysmic variables found by Bailyn et al. (1996). The southern cataclysmic variable (`star 1') is at from X 7a, and therefore remains a possible counterpart (assuming an error of 1" for the optical position, and taking into account the 2" error of X 19).

Fig. 7. X-ray contours in the central area of NGC 6752 as observed with the ROSAT HRI in the combined data of 1992-1996, with the positions of the sources obtained with the multi-source fit (). The positions of two candidate cataclysmic variables found by Bailyn et al. (1996) are indicated with .

We have analysed the separate observations, keeping the position of the four central sources fixed at those of the co-added image (as listed in Table 6), but allowing their fluxes to vary. We do not find significant evidence for variation; due to the limited statistics we cannot exclude variations by a factor two.

A countrate of 1 cts ksec^-1 corresponds to a luminosity between 0.5 and 2.5 keV of erg s^-1 at the distance of NGC 6752 and for an assumed 0.6 keV bremsstrahlung spectrum. Thus, X 7a and X 7b have X-ray luminosities of about erg s^-1, and X 21 and X 22 about a quarter of this. X 6 and X 14, the two sources outside the cluster core have luminosities of erg s^-1 and erg s^-1, respectively, if they are in NGC 6752.

5.3. Sources not related to the cluster

The spectral type of TYC 9071 228 1 is not known; on the basis of its magnitude and colour (, ) the star could be a late F star at a distance of pc. At this distance and for an assumed unabsorbed 1.4 keV bremsstrahlung spectrum, the countrate of X 19 converts to an X-ray luminosity in the 0.5-2.5 keV band of erg s^-1, a reasonable value for a late F main-sequence star (see e.g. the list of ROSAT detections of bright stars by Hünsch et al. 1998). The Tycho Catalogue marks this star with `unresolved duplicity', with visual magnitude varying between 9.51 and 10.88.

HIP 94235 has a significant parallax which puts it at 57 pc. Its countrate converts to an X-ray luminosity at that distance of about erg s^-1, a normal X-ray luminosity for a G0V star.

Comparison of the ROSAT image with the USNO-A2 Catalogue gives a candidate identification for X 16, at a distance of , see Table 6. No other sources outside the cluster have been identified by us.

We have analysed the three separate HRI observations, and find no evidence for variablility, except for X 15, which in March 1992 had an X-ray flux about half of that observed in March 1995 and April 1996.

© European Southern Observatory (ESO) 2000

Online publication: June 20, 2000
helpdesk.link@springer.de