Astron. Astrophys. 362, 984-1003 (2000)

2. Observations

2.1. Imaging

2.1.1. N -band

Mid-infrared images in the N -band were obtained with the Thermal Infrared Multi Mode Instrument (TIMMI ) at the European Southern Observatory 10-12 May 1995. The camera featured a 64  64 element Gallium doped Silicon array with good cosmetic quality and a quantum efficiency of 25%. We used the N -band filter:  = 9.70 µm and  = 5.10 µm with a setup yielding a pixel size of , and a FOV of 40".

For the cancelation of the strong background radiation, TIMMI is operated in chopping and nodding mode. The chopping frequency is several hertz and nodding is done once or twice per minute. In chopping mode a pair of exposures is obtained: one exposure contains object plus sky and the second exposure, 20" away, only sky. To obtain the images in nodding mode the telescope was moved by an angle exactly matching the chopper throw on the sky and the observation in chopping mode was repeated. Because this observation is exactly 180 degrees out of phase with the first one, the observed object shows up as a negative measurement. Subtracting both images resulted in a clean image corrected for first and second order effects introduced by the strong and inhomogeneous background radiation. To eliminate erratic pixels, the multiple images of each object were median filtered with a high threshold. Finally the images were Fourier transformed and the MIDAS standard low bandpass Butterworth filter was applied to better reveal the sources.

A log of the observations can be found in Table 1. The weather was poor during both nights. Consequently, only the brighter IRAS sources could be detected, while under favorable weather conditions TIMMI would be ten times more sensitive than IRAS . All but one of the undetected sources have an IRAS 12-µm flux below 4 Jy. Six sources having a 12-µm flux below 4 Jy were detected when the cloud conditions improved temporarily. The other 14 detected sources have a larger IRAS 12-µm flux. We think that non-detections are due to the cloudy weather rather than faulty IRAS positions or inaccurate telescope pointing.

Table 1. Observation log for the TIMMI observations. The first column gives the object name, the second column the observation date, the third column the total integration time in seconds, the fourth column gives the IRAS 12-µm band flux in jansky, and the last column indicates whether the object was detected or not.

2.1.2. Near-infrared images

J H K L -band images of the 20 post-AGB candidates detected in the N -band were obtained with CASPIR at the 2.3-m telescope at Siding Spring Observatory in Australia on 11 and 12 March 1998. The detector is a SBRC CRC463 InSb array which is sensitive from approximately 0.9 µm to 5.5 µm. The pixel size was and the total field of view 1´. Each object was observed 60 times for 0.3 s at 12" north and 12" south of its nominal position. The objects were observed in order of right ascension: IRAS 10256-5628 to 15144-5812 on 11 March 1998 and IRAS 15544-5332 to 17088-4221 on 12 March 1998. Both nights were photometric. Standard stars were observed at the beginning of the observations each night. The seeing was sub-arcsecond.

2.2. Observing the Br line

Using the improved positions from the TIMMI images, we obtained infrared spectra with IRSPEC at the NTT at ESO 13 and 14 May 1995.

The spectra are centered at Br and have a resolution of . The SBRC array gives a total wavelength coverage from 2.15 µm to 2.18 µm. The resolution in the spatial direction is /pixel. Since the outer edges of the array are vignetted, the total field of view is approximately . The slit width is 2 pixels or . The slit orientation was north-south. First we pointed the telescope to a bright star, and subsequently offset the telescope to the TIMMI position. The pointing accuracy of the NTT has an rms of about 1". Subsequently we moved the slit east- and westwards a few arcseconds, to `peak-up' the strongest signal in order to center the object in the slit. The resulting offsets were never larger than 6".

To subtract the sky emission the beam-switching technique was used. Because all but one (IRAS 13428-6232) of the sources are much smaller than 20", two spectra were taken with the source at different positions on the array, such that when the exposures were subtracted, the spectra did not overlap. For IRAS 13428-6232 one integration was taken on source and one on the sky and these images were subsequently subtracted

The weather was very good, and the humidity was around 15%. Within the accuracy limits of spectro-photometry the night was photometric, even if the conditions slightly deteriorated at the end of the night. Every 30 to 60 min a standard star was observed. A complete log of the observations can be found in Table 2. Note that the observations include four PNe from Van de Steene & Pottasch (1995).

Table 2. Observation log for the IRSPEC observations. The first column gives the object name, the second column the total on-source integration time in seconds, the third column the estimated signal-to-noise ratio, the fourth column the airmass.

2.3. Radio continuum observations

The 6 objects showing Br in emission were observed with the Australia Telescope Compact Array for 12 h on 14 February 1997 and for 2 h on 15 February 1997. The array configuration was 6A with the antennae at stations 3, 11, 16, 30, 34, and 37. The shortest baseline was 337 m and the longest 5939 m. The bandwidth was 128 MHz divided in 32 channels centered at 4800 MHz and 8640 MHz corresponding to 6 cm and 3 cm respectively. The sources were observed simultaneously at both frequencies. To obtain adequate spatial coverage we cycled through the sample of 6 IRAS objects and 3 calibrators once an hour: 1338-59 C, 13428-6232, 14488-5405, 1511-55 C, 15144-5812, 15544-5332, 1600-48 C, 16594-4656, 17009-4154 (with C denoting the calibrators). Each object was observed for 8 min and every phase calibrator for 3 min. To avoid artifacts in the field center we used an offset of 30" in declination. At the beginning and the end of each shift the primary flux density calibrator 1934-638 was observed for 5 min to 10 min. The weather was unstable and humid.

© European Southern Observatory (ESO) 2000

Online publication: October 30, 2000
helpdesk.link@springer.de