Polarimetry at the U.K. Infrared Telescope

1. Polarimetry at the U.K. Infrared Telescope Chris Davis1, Antonio Chrysostomou2 & Jim Hough2 Polarimetry Data Acquisition and Reduction An Overview of Near-Infrared Polarimetry at UKIRT Polarimetry has been available at UKIRT for over ten years. Like all previous polarimetry instrumentation, our current module, IRPOL2, was built by the University of Hertfordshire in England. At the present time, both imaging and spectro-polarimetry are offered with the facility instruments UIST (our workhorse 1-5 mm imager-spectrometer) and UFTI (a 1-2.5 mm imager). Both UIST and UFTI were designed to utilise the excellent image quality routinely experienced at UKIRT; thus, their pixel scales for imaging are fine (0.12 arcsec and 0.09 arcsec respectively). 0.24 arcsec and 0.60 arcsec slit-masks are available for spectro-polarimetry. Figure 4: Top - results from the Orac-DR pipeline. Far left - the same data rotated, binned and redisplayed in GAIA. Left - published results for the same target (Whitney et al. 1997, ApJ, 485, 703). The IRPOL2 module is located above the tertiary mirror, inside the instrument support unit beneath the UKIRT primary mirror (UIST and UFTI are mounted at the folded cassegrain focus). Consequently the half-wave retarder is the first optical element in the telescope beam after the primary and secondary mirror, thus minimising instrumental polarisation. The waveplate and its holder can be seen (in position for polarimetry) in the photo in Figure 2. For linear polarimetry the waveplate is stepped between four angles, 0o, 45o, 22.5o and 67.5o. Positional accuracy is achieved by way of toothed gears, a constantly-tensioned belt and opto-switches mounted in the waveplate holder. Movements between angles are rapid - taking only a few seconds - so polarimetric observing is very efficient at UKIRT (overheads are typically less than 100%, depending on exposure time). Figure 3: An observing script or “MSB” for imaging polarimetry, prepared using the UKIRT Observing Tool. As with all observing at UKIRT, polarimetry observations are prepared using the UKIRT Observing Tool (or OT). Observing “scripts” (or Minimum Schedulable Blocks - MSBs) include not only target coordinates and instrument parameters, but also offsets on the sky and waveplate moves (see e.g. Figure 3). In this way, a complete set of observations can be prepared at the researcher’s home institute - before arriving at the telescope. This level of pre-observing preparation is needed to facilitate flexible scheduling, which is the adopted mode of observing at UKIRT. Pipeline data reduction (DR) is also available at the telescope, for both imaging and spectro-polarimetry; these tools are described in a separate poster (Cavanagh, Currie and Jenness). Example results from the pipeline DR are shown in Figure 4. Figure 1: The U.K. Infrared Telescope atop Mauna Kea, Hawaii Imaging Polarimetry Imaging polarimetry is available with both UIST and UFTI. The former offers a larger field of view north-south, the latter a larger field east-west. Focal-plane masks are used with both instruments (with UIST the mask is cold) to isolate two adjacent regions, roughly 20”x100” in size, on the sky. The two regions are separated by about 30 arcsec so that, after passage through the prism, the e- and o-beam images fit onto the array with essentially no overlap. An example point-source exposure is shown in Fig.5. Because the beams don’t overlap, observations of extended sources are possible. For very bright sources, images using a sub-array can also be obtained (without using the focal plane mask). Normally, images are obtained at the four nominal waveplate angles (noted earlier) and at dithered positions on the sky. Typically 12 frames (4 WP angles x 3 dither positions) would constitute a polarimetry observation of a target, like those shown in Figure 4 above. Figure 5: A raw UIST polarimetry frame; the star is placed in the “top” (eastern) aperture; blank sky is simultaneously observed in the bottom (western) aperture. e- and o-beam images of both apertures are projected onto the array. Spectro-Polarimetry The Wollaston prism in UIST is located inside the cryostat in one of the two grism wheels. Consequently, spectropolarimetry is only available with half of the installed grisms. Much like imaging polarimetry, a slit mask is used to block part of the focal plane. Images through two 20-arcsec-long slit sections are transmitted to the prism and then the spectroscopy grism (which disperses in a direction orthogonal to the e-/o- beam splitting; see the cartoon in Fig. 6). Point sources can thus be observed at each waveplate angle, Figure 2: A birds-eye view of the IRPOL2 waveplate and its holder, photographed through the hole in the UKIRT secondary mirror. The dichroic tertiary, which splits the optical and infrared beams for guiding and IR observing, respectively, can be seen below the waveplate. Figure 6: A very rough cartoon showing the layout of the slit mask, wollaston prism and spectroscopic grisms in UIST. IRPOL2 Characteristics with the target offset by a few arcsecs up and down one of the two slit masks (to facilitate sky subtraction, as in the example in Fig.7). More extended sources can be nodded between the upper and lower slit apertures. Finally, pipeline DR is now available at the telescope: again, see the poster by Cavanagh et al. Three waveplates, all half-wave retarders, are available to UKIRT users. Together they offer a linear polarimetric capability between 0.9 mm and 5.0 mm. The most-used waveplate is a quartz and magnesium fluoride achromat that covers the J, H and K bands. Magnesium Fluoride zero-order waveplates are also available for the L and M bands (optimised for use at 3.5 mm and 4.75 mm respectively). All three waveplates have an unobscured aperture of 95 mm, which is well-suited to the 1.5 arcmin and 2 arcmin fields of view of UFTI and UIST respectively. In both UIST and UFTI the polarising analyser is a Wollaston prism; each prism splits the incoming radiation into two orthogonally-polarised “ordinary” and “extra-ordinary” beams. This dual-beam capability has the advantage that both beams are measured simultaneously, and since the degree of polarisation is measured from the ratio of the two beams, any variation in transparency (due to, e.g., seeing or cloud cover) should not affect the measured polarisation. The UIST prism is made from Magnesium Fluoride, the UFTI prism from Beta Barium Borate (both are uncoated). Both prisms yield a beam divergence (between the e- and o-beams) of ~20 arcsec. Figure 7: Polarimetry observations of an arc lamp, a flat field and a star (the latter is sky-subtracted, with the source offset a few arcseconds up and down the slit mask). Note to prospective UKIRT/IRPOL2 Users: UKIRT is available to all researchers, regardless of nationality. Application details are available on the UKIRT web pages. For further details specific to IRPOL2, please see: http://www.jach.hawaii.edu/JACpublic/UKIRT/instruments/. Affiliations: 1. UKIRT/Joint Astronomy Centre, 660 N. A’ohoku Place, Hilo, HI 96720, USA (c.davis@jach.hawaii.edu); 2. Centre for Astrophysics Research, University of Hertfordshire, College Lane, Hatfield, Herts AL10 9AB, U.K.