Astron. Astrophys. 334, 873-894 (1998)

10. R- and K-band imaging

Our spectroscopic observations (Sect. 3) show no evidence for companions to P1724 with periods up to days. In order to search for much wider companions, we have performed R -band imaging with the WIYN 3.5m telescope at KPNO. We used the S2KB instrument with a STIS pixel CCD, and a resolution of 0.195 arc sec per pixel. Data reduction was carried out with the GaussFit package (Jeffries et al. 1991) following Brandner et al. (1996). For details on the observations and data reductions, see Sterzik et al. (1997).

We obtained four exposures of P1724 on 19 Nov 1995 with 1 to 10 sec integrations, and a seeing of arc sec. One of our images is displayed in Fig. 17, and clearly shows a star arc sec south of P1724. This object is known as JW 242, star number 242 in Jones & Walker (1988), for which those authors give mag. The brightness difference between P1724 and JW 242 is mag in our WIYN image. Jones & Walker (1988) give .

Fig. 17. R-band image of P1724. One of our WIYN images in the R band showing P1724 and the faint star JW 242, about 4 arc sec to the south. The exposure time was 5 seconds. The faint annulus seen in our WIYN R band image around P1724 is due to the PSF of the instrument

The proper motion study by Jones & Walker (1988) is the deepest done so far, and their proper motion of JW 242 is the only one published. They list and for JW 242, which is deviant from the proper motion they list for P1724, namely and (all in mas/yr). From the positions of these two stars given by Jones & Walker (1988), we estimate the position angle to be and the separation to be arc sec. Since the epoch difference between the Jones & Walker image and our WIYN image is about 45 years, and given the proper motions, the position angle of this pair should now (i.e. at the time of our WIYN observation) be , unless there is orbital motion. The measured position angle in our WIYN image is . The excellent agreement with the prediction from proper motions further supports the conclusion that the pair is only a chance projection, and not a physical system.

In order to search for companions that are even closer, we used the speckle interferometry technique. The observations were carried out on 22 Nov 1997 at the 3.5m telescope on Calar Alto, using the near-IR camera MAGIC (Herbst et al. 1993) in the K-band at . The modulus of the complex visibility (i.e. the Fourier transform of the object brightness distribution) was determined from power spectrum analysis, the phase was computed using the Knox-Thompson algorithm (Knox & Thompson 1974) and from the bispectrum (Lohmann 1983).

We found no evidence for a companion. To quantify this statement, we computed the maximum brightness ratio of a companion that could be hidden in the noise of the data to obtain upper limits for the brightness of an undetected companion as function of the separation. For details of this procedure, see Leinert et al. (1996). We can exclude companions brighter than of the primary in K, corresponding to a magnitude difference of mag, at all separations larger than 0.13 arc sec, the diffraction limit of the telescope. For separations larger than 0.2 arc sec, we can even exclude companions brighter then of the primary, i.e. mag.

© European Southern Observatory (ESO) 1998

Online publication: June 2, 1998

helpdesk.link@springer.de