The Advanced Spectroscopic and Coronagraphic Explorer (ASCE) was proposed in 2001 to NASA's Medium-Class Explorer (MIDEX) program by the Smithsonian Astrophysical Observatory in collaboration with the Naval Research Laboratory, Goddard Space Flight Center and the Italian Space Agency. It is one of four missions selected for Phase A study in 2002. ASCE is composed of three instrument units: an Advanced Ultraviolet Coronagraph Spectrometer (AUVCS), an Advanced Large Aperture visible light Spectroscopic Coronagraph (ALASCO), and an Advanced Solar Disk Spectrometer (ASDS). ASCE makes use of a 13 m long boom that is extended on orbit and positions the external occulters of AUVCS and ALASCO nearly 15 m in front of their respective telescope mirrors. The optical design concepts for the instruments will be discussed.

Although designed primarily as a hard X-ray imager and spectrometer, the Ramaty High Energy Solar Spectroscopic Imager (RHESSI) is also capable of measuring the polarization of hard X-rays (20-100 keV) from solar flares. This capability arises from the inclusion of a small unobstructed Be scattering element that is strategically located within the cryostat that houses the array of nine germanium detectors. The Ge detectors are segmented, with both a front and rear active volume. Low energy photons (below about 100 keV) can reach a rear segment of a Ge detector only indirectly, by scattering. Low energy photons from the Sun have a direct path to the Be and have a high probability of Compton scattering into a rear segment of a Ge detector. The azimuthal distribution of these scattered photons carries with it a signature of the linear polarization of the incident flux. Sensitivity estimates, based on simulations and in-flight background measurements, indicate that a 20-100 keV polarization sensitivity of less than a few percent can be achieved for X-class flares.

The description of the Imaging Magnetograph eXperiment (IMaX) is presented in this contribution. This is a magnetograph which will fly by the end of 2006 on a stratospheric balloon, together with other instruments (to be described elsewhere). Especial emphasis is put on the scientific requirements to obtain diffraction-limited visible magnetograms, on the optical design and several constraining characteristics, such as the wavelength tuning or the crosstalk between the Stokes parameters.

The Instituto de Astrofisica de Canarias (IAC), Spain, together with the Spanish company Tecdis Displays Iberica, S.A., are developing voltage tunable optical retarders using liquid crystals as phase retarding medium. The ROCLIs are built for being used in the Imaging Magnetograph eXperiment (IMaX), which is one of the instruments aboard of the SUNRISE balloon mission (details about IMaX are described in a different paper in this session). A big advantage of using voltage tuned retarder plates is that no mechanisms are needed, which reduces significantly failure risk, weight, power and cost, aspects of particular importance in the SUNRISE balloon mission and for many future space borne applications. A set of prototypes has already been fabricated by Tecdis S.A. and is being characterized in the IAC laboratories. The purpose of these prototypes is to evaluate and demonstrate conceptually the suitability of the chosen liquid crystal for our use in IMaX. First results are very promising. In this paper we will present a full technical description of the ROCLIs for IMaX together with the laboratory test and verification results.

A new Stokes polarimeter for high accuracy measurement of solar magnetic field is being developed by National Astronomical Observatories, Chinese Academy of Sciences (NAOC). This instrument will be installed in the Space Solar Telescope (SST), which has a one-meter primary mirror, and a two-dimensional spectrograph with 8-channel birefringent filter. In order to minimize the crosstalk in linear polarization from the circular polarization, a novel optical assembly has been selected. This design consists of a quarter retarder, a rotatable polarizer, and a half retarder. The half retarder will rotate with the polarizer by an angle of half quantity of the rotating angle of the polarizer. In this paper, we will introduce this polarimeter and discuss the achromatic performance of the polarization elements in the polarimeter. We will focus on comparing the orientation and retardation of different kind of achromatic waveplate in profiles and give which is the best one for SST polarimeter. In addition, we will show the calculation crosstalk value of this polarimeter at designed wavelength.

In this contribution, the main characteristics of near infrared spectropolarimetric measurements are described, putting especial emphasis on the techniques to minimize the crosstalk between the Stokes parameters.

The Mk4 K-coronameter records polarization brightness images of the solar corona from the Mauna Loa Solar Observatory, Hawaii, USA. Calibration is required to quantitatively measure coronal polarization brightness, which in turn is used to infer coronal electron density. Matrix techniques are used to map the instrument polarization response. Brightness scaling depends upon precise knowledge of properties of an opal glass attenuator and calibration polarizer, sky transmission, and telescope pointing. In addition, account must be made for polarization at the objective lens and from the terrestrial atmosphere. Calibration parameters are stable to a few percent over a day, but when coupled with uncertainties in calibration optics values, sky transmission, and pointing, the average measurement uncertainty is ±15% ±6×10^-9 pB/B_Sun.

A motorized rotating wave plate polarization analyzer has been built for modulating solar light as input to the Celeste high-resolution cryogenic grating spectrometer to record full Stokes parameter maps at 12 microns wavelength of areas around sunspot and plage regions. The instrument system was used at the McMath-Pierce (1.5 m aperture, f/54) solar telescope at Kitt Peak, where single position slit spectra are spatially 120 arc-sec long, with a double sampled spectral resolution of 0.036 cm^-1. Stepper motor driven limb guiders and synchronized action of polarization wheels are under the control of the Celeste data system computer, allowing unattended programmable scanning with 2-3 arc-sec steps for several hours. The results are spatial maps of magnetic field strength and direction, strength measured directly to the level of a few hundred gauss, with a time resolution of approx. 5 minutes per slit position. In this presentation we describe the Mid-IR Stokes polarimetry instrument system, design and performance, and discuss plans for future development.

With a new class of imaging polarimeters it is now possible to eliminate the previous main limiting factors of seeing and gain-table noise in the polarization images to allow spectro-polarimetry with a precision of 5 × 10^-6. This has opened the door to a previously unexplored world of polarization phenomena with promising diagnostic possibilities not only for the Sun but also for night-time astronomy. While illustrating examples of what has been achieved, we present an overview of the new opportunities and quantify the limitations imposed by the photon flux.

New highly sensitive polarimetric instruments and observational techniques allow to observe weak polarization signals in the visible and near ultraviolet part of the solar spectrum. Many of these signals are caused by scattering processes in the upper photosphere and lower chromosphere and thus reflect the thermodynamics of these layers. Also magnetic fields lead to polarization via the Zeeman effect or alter scattering polarization via the Hanle effect. The observation of both effects requires highest polarimetric sensitivity in combination with very high spectral resolution. In the following the instrumental and observational concepts are described. Special emphasis will be given to the Zurich Imaging Polarimeter II, which is now sensitive to the near ultraviolet part of the solar spectrum down to the atmospheric cut-off around 300 nm thanks to the use of a special CCD sensor, which for the first time combines so-called 'open electrod structure' with fast on-chip demodulation in the kHz regime.

Much progress has been made during the last years in obtaining polarimetric observations of the Sun close to the diffraction limit. Here I summarize the problems one encounters when observing close to the diffraction limit of a large solar telescope, review techniques, present examples of recent observations, and discuss the need for further developments of instruments and image reconstruction techniques.

GREGOR is the project of a high-resolution solar telescope with an aperture of 1.5m and an effective focal length of about 55m. It is designed to support ground-based accurate, high sensitive spectro-polarimetry at visible and IR wavelengths in the solar photosphere and chromosphere for studying the dynamics of the solar atmosphere and the underlying physical processes. The concept of polarimetric measurements with GREGOR is based on several unique and highly specialized post-focus polarimeters like POLIS or TIP and a polarimetric equipment (GPU) situated near the telescope's secondary focus F2 where the optical properties are still rotationally symmetric and the telescope can be regarded as polarization free at the 10^-4 level. The GPU is an integral part of the telescope, consisting of a calibration unit and a modulation unit. The calibration unit allows to calibrate the modulation unit as well as polarimeters built in any post-focal device. It consists of a linear polarizer and quarter wave retarders for the visible and the IR spectral range and is located in front of the modulation unit. The modulation unit is supposed to permit a very efficient polarimetric analysis in the spectral range from 400 nm to 700 nm. It consists of two electro-optical modulators and a prism as linear polarizer.

The design of an achromatic polarisation modulator is presented. The modulator is based on a combination of three electrically switchable non-achromatic ferroelectric liquid crystal retarders. The design follows the idea by Pancharatnam who first introduced suitable achromatic combinations of crystal retarders. We combined three ferroelectric liquid crystal retarders to create an electrically switchable achromatic halfwave plate which can be used in the spectral range from 400 nm to 750 nm. Different designs are theoretically modeled and compared under the aspects of their individual response to temperature fluctuations and useful wavelength range. First results of laboratory tests are presented to experimentally evaluate the feasibility of the concept.

A diffraction limited spectro-polarimeter is under construction at the National Solar Observatory in collaboration with the High Altitude Observatory. The scientific objective of the project is to measure the magnetic fields on the Sun up to the diffraction limit of the Dunn Solar Telescope. The same instrument would also measure the magnetic field of large sunspots or sunspot groups with reasonable spatial resolution. This requires a flexible image scale which cannot be obtained with the current Advanced Stokes Polarimeter (ASP) without loosing 50% of the light. The new spectro-polarimeter is designed in such a way that the image scale can be changed without loosing much light. It can work either in high-spatial resolution mode (0.09 arcsec per pixel) with a small field of view (FOV: 65 arcsec) or in large FOV mode (163 arcsec) with low-spatial resolution (0.25 arcsec per pixel). The phase-I of this project is to design and build the spectrograph with flexible image scale. Using the existing modulation, calibration optics of the ASP and the ASP control and data acquisition system with ASP-CHILL camera, the spectrograph was tested for its performance. This paper will concentrate on the performance of the spectrograph and will discuss some preliminary results obtained with the test runs.

For the interferometric environment, optical polarimetry may need to assimilate radio-polarimetric concepts. In particular, the Stokes parameters should be defined in terms of complex correlations rather than as differences of orthogonally-polarized fluxes. As a corollary, traditional polarization modulators may not be the cure-alls they are for single-telescope polarimetry. Polarization effects in the Coudé train and delay lines spoil the accuracy of traditional quasi-scalar interferometers. An alternative optical architecture is proposed, using traditional (i.e. single-beam) optical polarimetry in the correlator, but 'radio-type' transfer of light from telescope foci to correlator. Such a fundamental solution can eliminate errors due to inclined mirrors (phase shifts and added polarization). The architecture enables full-Stokes polarimetry at the resolution of the interferometer, but also a 'no-polarization-desired' mode which does not necessarily involve loss of signal-to-noise ratio and yet is free from polarization-induced errors of photometry. Existing polarization components permit a very wide instantaneous bandwidth.

In this paper, I discuss the need for optical interferometric polarimetry and demonstrate that it is possible with existing technology. Coherent averaging is required; it may be accomplished simply with the addition of a non-polarizing beam splitter after the beam combiner. I define three observables and their corresponding output vectors. One vector is identical to the normalized Stokes vector obtained from classical polarimetry, while the other two are related to the Stokes visibility vector. Calibration of the instrument requires only separate zero-spacing observations of calibrator stars on each of the individual arms - no interferometric measurements are necessary. For a large class of objects, it is possible to create images in all Stokes parameters with a single variable-length baseline.

An optical-wavelength polarimetry module has been in use for the past ten years at the W.M. Keck Observatory. The module is used in conjunction with the LRIS imaging spectrograph. It provides either imaging polarimetry or spectropolarimetry across the full wavelength range of the instrument (320-1100 nm). The design, performance, and limitations of the polarimetry module are described, along with a sampling of science results.

The use of CCDs as focal plane detectors has changed the design concept for the polarimetric mode in the visible and, actually, also in the near infrared spectral ranges. Providing astronomical instrumentation with polarization sensitive crystal in appropriate configuration (Wollaston, Rochon, Glann-Thomson, and so on), together with half- or quarter- waveplate retarder, it is possible to determine the linear and/or circular Stokes parameters. The polarization analyser placed at the Asiago Faint Object Spectrographic Camera (AFOSC), allows simultaneous measurements of the two linear Stokes parameters usually named Q and U without any λ/2 retarder plate. Similar devices may be designed to be placed at every instrument with an accessible pupil. The paper reports details of the instrument concept and the results of observations of polarized and unpolarized standard stars, both in broad-band and spectro-polarimetry observing modes.

We present optical linear polarimetry of 9 ultra cool field dwarfs. The linear polarisation of each L dwarf we measured is less than 0.2 percent. Three dwarfs have polarisations compatible with zero, two are marginal detections, and three have significant polarisation. Due to their small distance, an insterstellar origin for the detected polarisation can be safely ruled out. Our detections confirm that dust is present in the atmosphere of these L dwarfs and that the scattering geometry is not symmetric.

Linear spectropolarimetry of spectral lines is a neglected field in astronomy, largely because of the lack of instrumentation. Techniques that have been applied, but rarely, include investigation of the dynamics of scattering envelopes through the polarization of electron- or dust-scattered nebular light. Untried techniques include promising new magnetic diagnostics like the Hanle Effect in the far-ultraviolet and magnetic realignment in the visible. The University of Wisconsin Space Astronomy Lab is developing instrumentation for such investigations. In the visible, the Prime Focus Imaging Spectrograph (PFIS) is a first light instrument for the Southern African Large Telescope (SALT), which at an aperture of 11m will be the largest single telescope in the Southern Hemisphere. Scheduled for commissioning in late 2004, PFIS is a versatile high-throughput imaging spectrograph using volume-phase holographic gratings for spectroscopic programs from 320nm to 900nm at resolutions of R=500 to R=6000. A dual-etalon Fabry-Perot subsystem enables imaging spectroscopy at R=500 and R=3000 or 12,500. The polarization subsystem, consisting of a very large calcite polarizing beam-splitter used in conjunction with half- and quarter-wave Pancharatnam superachromatic plates, allow linear or circular polarimetric measurements in any of the spectroscopic modes. In the FUV, the Far-Ultraviolet SpectroPolarimeter (FUSP) is a sounding rocket payload, scheduled for its first flight in 2003, that will obtain the first high-precision spectropolarimetry from 105 - 150 nm, and the first astronomical polarimetry of any kind below 130 nm. The 50 cm primary mirror of the telescope is F/2.5. At the prime focus are the polarimetric optics, a stressed lithium fluoride rotating waveplate, followed by a synthetic diamond Brewster-angle mirror. The spectrometer uses an aberration-corrected spherical holographic grating and a UV-sensitized CCD detector, for a spectral resolution of R=1800.

PEPSI (Postham Echelle Polarimetric and Spectroscopic Instrument) is to use the unique feature of the LBT and its powerful double mirror configuration to provide high and extremely high spectral resolution full-Stokes four-vector spectra in the wavelength range 450-1100nm. For the given aperture of 8.4m in single mirror mode and 11.8m in double mirror mode, and at a spectral resolution of 40,000-300,000 as designed for the fiber-fed Echelle spectrograph, a polarimetric accuracy between 10^-4 and 10^-2 can be reached for targets with visual magnitudes of up to 17th magnitude. A polarimetric accuracy better than 10^-4 can only be reached for either targets brighter than approximately 10th magnitude together wiht a substantial trade-off wiht the spectral resolution or with spectrum deconvolution techniques. At 10^-2, however, we will be able to observe the brightest AGNs down to 17th magnitude.

ESPaDOnS is an ongoing project for a cross-dispersed echelle spectrograph/spectropolarimeter, for general community use at the Canada-France-Hawaii Telescope. This instrument will provide: (1) a complete optical spectrum from 370 to 1,000 nm in a single exposure, with a variable resolution between 50,000 and 75,000, (2) all polarization components of the stellar light, and (3) 20% peak total throughput. The wide spectral coverage will maximize the multiplex gain associated wtih multi-line techniques. The high resolution will give sufficient spatial resolution for Doppler and Zeeman-Doppler imaging of rapid rotators, increasing the sensitivity of the observations to small-scale structures. It will also allow accurate brightness, abundance and magnetic field mapping of moderate rotators, and help study stellar pulsations, abundances, and extrasolar planets. The possibility of recording two interleaved spectra will be very useful in polarimetric mode, and very convenient for faint object spectroscopy in non-polarimetric mode, where the spectrum of the adjacent sky will be recorded along with that of the target. The spectrographic is bench-mounted and fed by low-OH H-treated Ceram-Optec optical fibers from a Cassegran module containing all calibration and polarimetric facilities, making it possible to have extremely good wavelength stability and minimal instrumental polarization. The achromatic polarimeter includes one quarter-wave and two half-wave Fresnel rhombs coupled to a removable Wollaston prism. ESPaDOnS will be a unqiue instrument worldwide in polarimetric mode, competitive wiht similar instruments on 8-m class telescopes in non-polarimetric mode. Astronomers will be able to address with unprecedented details a broad range of important issues in stellar physics, circumstellar environments, and extrasolar planets.

We are developing a Cassegrain optical spectropolarimeter, LIPS. LIPS employs an echelle-type spectrograph to get a high spectral resolution of R > 7,000. LIPS consists of three instrumental units, a polarimeter, a spectrometer, and a detector. One serious problem was the appearance of 'spectral ripple' generated in the Pancharatnam type super-achromatic half-wave plate (PWP). We found out through laboratory works and numerical simulations that the ripple is caused by the interference among layers in the half-wave plate. In stead of PWP, we adopted a new type of super-achromatic half-wave plate composed of five polymethyl-methacrylate layers, which was manufactured by Astropribor Company in Ukraine. This wave plate greatly reduces reflection at the boundaries among layers, which results in significant reduction of ripple. Since the spring of 2001, we have carried out engineering observations with LIPS at IRS and UH88. We conclude from these observations that LIPS have achieved the accuracy of ΔP < 0.1 %.

Polarimetry is a powerful means for detecting and constraining various physical phenomena, such as scattering processes or magnetic fields, occuring in a large panel of stellar objects: extended atmospheres of hot stars, CP stars, Young Stellar Objects, Active Galaxy Nuclei, ... However, the lack of angular resolution is generally a strong handicap to drastically constrain the physical parameters and the geometry of the polarizing phenomena because of the cancelling of the polarized signal. In fact, even if stellar features are strongly polarized, the (spectro-)polarimetric signal integrated over the stellar surface rarely exceeds few percents. Coupling polarimetric and interferometric devices allows to resolve these local polarized structures and thus to constrain complex patchy stellar surfaces and/or environments such as disk topology in T Tauri stars, hot stars radiative winds or oscillations in Be star envelopes. In this article, we explain how interfero-polarimetric observables, basically the contrast and the position of the interference fringe patterns versus polarization (and even versus wavelength) are powerful to address the above scientific drivers and we emphasize on the key point of instrumental and data calibrations: since interferometric measurements are differential ones between 2 or more beams, this strongly relaxes the calibration requirements for the fringe phase observable. Prospects induced by the operation of the optical aperture synthesis arrays are also discussed.

In alt-azimuth telescopes Nasmyth foci are suitable focal planes to reduce mechanical instrument complexity and costs. However, they present some disadvantages mainly due to the field rotation. This is a particularly crucial point in polarimetric observations. Because of the folding mirror, the radiation polarization state is so modified that, to avoid systematic errors, instrumental polarization has to be removed as a function of the telescope position. A model of the polarization introduced by the Telescopio Nazionale Galileo (TNG) at its focal plane is presented. The model takes into account physical and geometrical properties of the optical system, complex refraction index of the mirrors and their relative position, deriving instrumental polarization as a function of the pointing coordinates of the telescope. This model has been developed by means of Muller matrices calculation. Telescope instrumental polarization has been measured following some standard polarization stars at different telescope positions. The mathematical model, here discussed, was confirmed comparing the theoretical results and the experimental measurements at the TNG instruments.

The polarimeter built for the high resolution spectrograph (SARG) of the alto-azimuthal Telescopio Nazionale Galileo is presented. This double-beam instrument, able to take into account time independent (instrumental) and time dependent (sky transparence) sensitivity, is based on a Fresnel prism (λ/2) and K-prism (λ/4) which gives an almost constant retard along the very large wavelength interval covered with new spectrographs: SARG covers the 370 - 1020 nm range and more than 300 nm in a single exposure. The two flat metallic mirrors, which are necessary to feed the spectrograph, and the alto-azimuthal mounting of the telescope are responsible of an instrumental polarisation depending on the sky position of the target. A modelling of the instrumental polarisation and a hardware correction of the sky rotation are performed to measure the polarisation across stellar-like object at R=115,000 resolution.

The Homunculus nebula, surrounding the massive star system Η Carinae, is a bipolar dust nebula which is outflowing at up to 700 km/s. The bipolar lobes display high linear polarization in the optical and near-IR, which is consistent with an origin in dust scattering from the central source. Extensive imaging and spectropolarimetric studies have not however been able to provide a consistent picture of the dust which has been ejected in the mass loss events, the most important of which occurred in the 1840's. The magnitude of the linear polarization shows very little change with wavelength, suggesting very small grains. On the other hand, models of the IR emission suggest a mixed grain population. The scattering properties of feasible dust mixtures do not however well match the observed optical and near-infrared polarization behaviour. Three possibilities are advanced to explain the dust grain population in the Homunculus: optical depth effects within a clumped distribution; the presence of many small clouds with grain size dependent on depth into the cloud; large-scale grain alignment. The last suggestion is supported by observation of 10μ polarization. Visible light circular polarization observations and refined geometric dust-scattering models are presented to advance the picture of the dust in the Homunculus. Since the dust ejected from Η Carinae is several solar masses, this study is also relevant to the understanding of ISM dust.

We describe the prospective work undertaken on an interferometric technique using polarimetry called SPIN (Spectro-Polarimetric INterferometry). The polarizing phenomena described in this work have to be taken into account by any stellar interferometer in order to control the fringe signal. Adding a polarimetric device at their combined focus represents no technical difficulty. The use of SPIN can extend interferometry by an important complementary tool for locating and quantizing the mass loss from early type stars since these stars are subject to strong Thomson scattering in their vicinity. As an illustration of the potential of SPIN, we present the results of Monte-Carlo simulations showing the expected signal for realistic hot star environment. Radiative winds ranging from A supergiants to earliest O stars are considered. In particular, the results show the strong expected signal from spherical winds for which no detection of polarization is achievable by classical technics.

Four different designs for a polarising beamsplitter (BS) are compared under the aspect of their suitability for high resolution solar spectropolarimetry. The four designs are: A solution based on a Savart-plate, two air-spaced Wollaston prisms, and a glass beamsplitter cube with polarisation sensitive dielectric coating, and a single Wollaston prism inside a focal reducer. Using ray-tracing algorithms these beamsplitters are characterised with the help of spot diagrams for the two light paths of orthogonal polarisation. It is shown that with the current spectrographs employed in solar research, the differential optical aberrations introduced by the beamsplitter are negligible, thanks to the slow F/#-ratio of existing solar telescopes and the limited field of view of currently used array detectors. It will, however, be demonstrated that with the new generation of large solar telescopes care must be exercised on the design of the beamsplitter. This will be shown using an example spectrograph, as could be used in a new 1.5 m class solar telescope.

An imaging polarimeter for the visible has been built for the mont Megantic Observatory. It uses a new design with a Foster prism which allows to obtain the two orthogonal beams of polarized light onto the detector at the same time, thus alleviating calibration problems associated with variations in the sky transparency during the observations. The field of view is 2.4 arc minutes at the f/15 focus of the telescope. A rotating achromatic half-wave plate is used to get all the linear polarization components and eliminate systematic calibration effects. Circularly polarized images can also be obtained by substituting a quarter-wave plate for the rotating half-wave plate. A user-friendly software for carrying out the data reduction has been developed.

We report the development and performance of a near-IR polarimeter for the Subaru 8.2m telescope. The polarimeter is currently used with one of the Subaru instruments, CIAO, the stellar coronagraphic imager with adaptive optics. CIAO is the instrument specialized to obtain high contrast images of faint objects in the vicinity of bright objects. For achieving both high spatial resolution and high dynamic range, the instrument is used wiht the Subaru adaptive optics and has a dedicated cold coronagraphic capability. The polarimeter comprises two components. One component consists of an achromatic half-waveplate, an achromatic quarter-waveplate, and a calibration wire grid. Both half- and quarter-waveplates are rotatable and retractable, while the calibrator is only retractable. This componetn is placed upstream of any opticla components including adaptive optics system, which minimizes the effect of various mirros on instrumental polarization. The other component consists of two anlayzers, a cold wire grid and acold Wollastron prism. These are placed in the filter wheels of CIAO cryostat and can be chosen. The whole system is remotely controlled.

We have carried out KHL band high resolution imaging and H band imaging-polarimetry of the Red Rectangle nebula using CIAO and 36 element AO mounted on the 8.2m Subaru telescope. HK band images show a X-shape structure close to 0.1 inch and 2 lobes with separation of 0.15 inch at the north and the south. Our L band image show a small clump and its position is 0.1 arcsec east from the center of the southern lobe. The polarization map shows roughly centrosymmetric vector pattern and the center of the pattern is consistent with that of 2 lobes. There is scatter of the vector pattern at approximately 0.1 inch east from the southern lobe and a local minimum in the degree of polarization. These results can explain that the primary star HD44179 is at 0.1 inch east from the southern lobes and the dominant illumination source is a M type star at the center of the nebula.

Gemini polarimetry is based on a waveplate module located in the base of the A&G unit that can be used by all instruments operating in the optical and near-infrared. Because of space limitations and limited access to the module, a single half-wave retarder covering 0.34 to 2.5μm is used for linear polarimetry. A composite zero-order half-wave retarder is used for the L-band (mid-IR instruments have their own waveplate module). The plates have a clear aperture of 95mm and are surrounded by a transparent annulus to increase the field of view for the on-instrument wavefront sensors. Each instrument includes, or will include, a 2-beam polarising prism, usually in the form of a Wollaston prism. Provision for circular polarimetry has been included but not yet implemented. The design of the waveplate module and the techniques employed to provide high precision are described. The materials available for the Wollaston prisms, including those used in the mid-IR, are also discussed. Techniques to avoid ripple in the polarisation spectrum observed with some spectrometers are presented. Unfortunately at present observational results are not available to include in this paper.

We present and discuss the capabilities of the infrared polarimetric modes of the ESO-VLT adaptive optics system NAOS-CONICA. Commissioning results obtained both with wire-grids and Wollaston prisms are shown. In particular, NACO observations of the Calabash reflection nebula are compared with earlier, seeing limited, results obtained at ESO to illustrate the new potential offered by adaptive optics assisted polarimetry on an 8m class telescope.

ESO is presently building an adaptive-optics fed Cryogenic Infrared Echelle Spectrograph (CRIRES) for the VLT-observatory operating in the wavelength range from 1-5μm. Spectro-polarimetry with a focus on circular polarization in the infrared is particularly interesting as the ratio of Zeeman-splitting to intrinsic line widths improves linearly with wavelength. Also the contrast between absorption lines in starspots and the surrounding photosphere becomes more favourable when observing a longer wavelengths (i.e. closer to the Jeans-case). Moreover it is well known that even extremely red objects such as Brown Dwarf candidates show X-ray emission and hence must have magnetic activity. CRIRES shall be equipped with a reflective phase retarder and a Wollaston-prism allowing nearly un-compromised measurements of circular polarization at a spectral resolution of 100000. Linear polarization measurements are also possible, but most likely with compromised performance. We show preview spectra of Zeeman sensitive transitions in the infrared based on Fourier-transform spectra of sunspots from literature.

ESO's Thermal Infrared Multimode Instrument, TIMMI2,in regular operation at the 3.6m telescope on La Silla, Chile,since January 2001 is equipped with a linear polarization mode which can be used in conjunction with all scientific observing modes available. A description of the polarimeter, working between 5 and 24mu m in imaging and low-resolution grism spectroscopy is given. Calibration issues and other operational aspects are described. We report first results from the final astronomical commissioning.

We present the results of lab comparison tests of performance for several commercially available grating polarizers for use at mid-infrared wavelengths. The tests were done using a polarized laser diode source (9.2 $\mu$m) and photoconductive HgCdTe single pixel detector. We describe some basic equations governing quantification of polarization performance, our instrumental test setup and our results. There is a large difference in the contrast produced by polarizers from different companies. Availability of high-transmission, high-contrast polarizers for use at near and mid-infrared wavelengths will make it possible to routinely characterize polarization of astronomical sources, such as physical properties of dust grains and magnetic field lines around sources of interest. Such polarizers will also be valuable in the development of the instrumentation needed for infrared nulling interferometry.

We are constructing a high sensitivity optical polarimeter capable of detecting fractional polarization levels below 10^-6. The science goal is to directly detect extra-solar planets (ESP), in contrast to the indirect methods such as radial velocity measurements. The polarimeter will detect starlight scattered from the atmosphere of the planet as a polarisation signal thereby giving information on the planetary atmospheres. The radius of the planet and the planet temperature can be determined from the measured albedo. The position angle of polarisation will enable the mass of planets, detected through radial velocity measurements, to be determined without the uncertainty of the orbit inclination (Msini). The polarimeter has an essentially simple and classical design but is able to take advantage, inter alia, of modern detector technology.

MC3D is a three-dimensional, self-consistent continuum radiative transfer code. It can be used to determine the spatial temperature distribution in arbitrary dust/electron configurations, such as around young stellar objects or active galactic nuclei. Based on this temperature distribution, MC3D allows calculation of polarization maps, images, and spectral energy distributions. We present the numerical techniques applied in this code, describe its capabilities, and show instructive examples of previous applications.

We present high-resolution polarization maps, obtained with near-infrared instruments such as ISAAC at the VLT, SOFI at the NTT, and SCUBA at the JCMT. While we use the near-infrared polarization maps to determine the structure of the optical reflection nebula Cederblad 110 IRS 4 and to investigate the alignment of circumstellar disks around T Tauri binary stars, submillimeter polarization maps are used to derive the magnetic field structure and strength in Bok globules. Furthermore we show that near-infrared polarimetry represents a powerful tool to distinguish between different polarization models developed for active galactic nuclei.

After a short analysis of the main problems involved in the construction of a Far Infrared polarimeter with very low instrumental noise, we describe the instrument that will be employed at MITO telescope to search for calibration sources and investigate polarization near the CMB anisotropy peaks in the next campaign (Winter 2002-03).

A far-infrared polarimeter, Hale, will be proposed for the next round of instruments for SOFIA. Key features are: simultaneous detection of two components of polarization; detector arrays providing >4000 pixels on the sky; and four passbands between 53 μm and 215 μm, a range characterized by strong dependence of polarization on wavelength. At 53 μm the diffraction-limited resolution, 1.2 λ/D, will be 5.2 arcsec. In all passbands the systematic errors in polarization will be Δ(P) < 0.2%, Δθ< 2 °.

We discuss the conceptual and practical guidelines of a method to calculate the cross-polarization of a telescope, including its relay optics, using a commercial optical design software, without the need to use complex, slow and expensive Physical Optics programs. These effects are usually negligible at visible and infrared wavelengths but may be of considerable importance at radio wavelengths. Offset reflector antenna configurations, common in the telecommunication industry, and antenna relay optics consisting of offset mirrors, common in millimeter and submillimeter-wave telescopes, result in an increased contribution to the cross-polarization. Polarization measurements are also becoming very important in Radio Astronomy. In fact, dust emission polarimetry and the study of linearly polarized, nonmasing, rotational lines (e.g., CO) with submm telescopes are both powerful diagnostic of magnetic fields in molecular clouds. However, the low average source polarization requires a careful optimization of the optical design to keep any instrumental polarization contribution from both telescope and relay optics as low as possible in astronomical polarimetry experiments. Likewise, in telecommunications applications polarization separation can be used to effectively double the available bandwidth provided the isolation between the two orthogonal polarization states is sufficient.

We report on the commissioning of a polarimeter operating at the intermediate frequency (150 MHz) of the IRAM 30m telescope. The polarimeter adds, with suitable phase shifts, the amplitudes of two orthogonally linearly polarized heterodyne receivers, and determines from these signals all four Stokes parameters. The polarimeter allows continuum as well as spectral line observations in all millimeter atmospheric windows accessible with the 30m telescope. The instrumental phase between the two receivers, controlled through a phase shifter, does not exceed one degree (rms) at 3mm. Instrumental polarization on boresight is found to be small, below one percent for Stokes U and V. Limitations due to beam polarization are discussed. We present results from the first large scale 3mm polarization survey started in 1999. About 100 Active Galactic Nuclei with flux densities stronger than approximately 0.8 Jy were observed, many of them repeatedly. Linear polarization larger than 10 percent is found in several sources at several epochs. We briefly describe results from two spectro--polarimetric surveys: SiO maser stars and methanol masers. Polarization is easily detected in both groups.

We describe the design and expected performance of BICEP, a millimeter wave receiver designed to measure the polarization of the cosmic microwave background. BICEP uses an array of polarization sensitive bolometers operating at 100 and 150 GHz to measure polarized signals over a 20 degree field of view with 1 degree resolution. BICEP is designed with particular attention to systematic effects which can potentially degrade the polarimetric fidelity of the observations. BICEP is optimized to detect the faint signature of a primordial gravitational wave background which is a generic prediction of inflationary cosmologies.

Coherent radiometers have been used for decades in the detection of polarization of astronomical signals. Recent success in detection of intensity fluctuations of the Cosmic Microwave Background (CMB) has created great interest in the potential measurement of CMB polarization. Detection of polarization at predicted levels may stretch the sensitivity of today's radiometers. I will review the state-of-the-art of receiver front-ends, discuss their application to polarimeter design in experiments, both existing and under development and finally describe the prospects for massive arrays for the detection of CMB polarization.

SPOrt (Sky Polarization Observatory) is a space experiment to be flown on the International Space Station during Early Utilization Phase aimed at measuring the microwave polarized emission with FWHM = 7 deg, in the frequency range 22-90 GHz. The Galactic polarized emission can be observed at the lower frequencies and the polarization of Cosmic Microwave Background (CMB) at 90 GHz, where contaminants are expected to be less important. The extremely low level of the CMB Polarization signal calls for intrinsically stable radiometers. The SPOrt instrument is expressly devoted to CMB polarization measurements and the whole design has been optimized for minimizing instrumental polarization effects. In this contribution we present the receiver architecture based on correlation techniques, the analysis showing its intrinsic stability and the custom hardware development carried out to detect such a low signal.

Bolometers currently offer the best sensitivity for measuring the anisotropy and polarization of the Cosmic Microwave Background (CMB). The next generation of CMB instruments intended to search for faint 'curl-mode' polarization require bolometer focal plane arrays with significantly higher sensitivity than current temperature anisotropy receivers, and unprecedented control of systematic errors. Bolometers for the ESA/NASA Planck experiment, thermally optimized for the photon background from the sky and instrument, approach the fundamental photon noise from the CMB with defined allocations for systems level noise contributions. Ground-based CMB polarimeters will soon field focal planes with approximately the instantaneous sensitivity of Planck HFI to deeply probe limited regions of sky. Future CMB polarimeters require large-format arrays of bolometers. Antenna-coupled bolometers with superconducting transition-edge readouts promise large-format arrays with well-controlled beam patterns and integral lithographed transmission-line filters.

BaR-SPOrt (Balloon-borne Radiometers for Sky Polarisation Observations) is an experiment to measure the linearly polarized emission of sky patches at 32 and 90 GHz with sub-degree angular resolution. It is equipped with high sensitivity correlation polarimeters for simultaneous detection of both the U and Q stokes parameters of the incident radiation. On-axis telescope is used to observe angular scales where the expected polarization of the Cosmic Microwave Background (CMBP) peaks. This project shares most of the know-how and sophisticated technology developed for the SPOrt experiment onboard the International Space Station. The payload is designed to flight onboard long duration stratospheric balloons both in the Northern and Southern hemispheres where low foreground emission sky patches are accessible. Due to the weakness of the expected CMBP signal (in the range of microK), much care has been spent to optimize the instrument design with respect to the systematics generation, observing time efficiency and long term stability. In this contribution we present the instrument design, and first tests on some components of the 32 GHz radiometer.

We have developed a correlation radiometer at 33 GHz devoted to the search for the residual polarization of the Cosmic Microwave Background (CMB). The two instrument's outputs are a linear combination of two Stokes parameters. The instrument is therefore directly sensitive to the polarized component of the radiation (rispectively linear and circular). The radiometer has a beamwidth of 7 or 14 degree, but it can be coupled to a telescope increasing the resolution. The expected CMB polarization is at most a part per million. The polarimeter has been designed to be sensitive to this faint signal, and it has been optimized to improve its long term stability, observing from the ground. In this contribution the performances of the instrument are presented, together with the preliminary tests and observations.

A lot of effort is currently devoted to the design of ground based experiments dedicated to the detection of the Cosmic Microwave Background Radiation Polarisation. The very low level of the expected signals requires a detailed study of the spurious effects introduced by each element of the optics. We use the well-known general formulae for the reflection and transmission coefficients in isotropic medium-uniaxial birefringent crystal interfaces for any angle of incidence and arbitrary birefringence optical axis orientation. Ordinary and extraordinary ray tracings are carried out and arbitrary incoming polarisation is used to calculate instrumental polarisation and depolarisation effects.

Using multilayer coated mirrors to provide high reflectivity at large graze angles, we have proposed to launch a small telescope that is capable of measuring the linear polarization of the soft x-ray fluxes from many astronomical sources. Three identical mirror-detectoer assemblies are designed for maximum efficiency at 0.25 keV, where the photon spectra of many celestial targets peak. In observations lasting 1-3 days using this low risk instrument with proven heritage, we can detect polarizations of 5-10% at 5σ due to Compton scattering or synchrotron processes in the relativistic jets of BL Lac objects, accretion disks or jets in active galactic nuclei and atmospheres of isolated pulsars. Pulsar data can be binned by pulse phase to measure the orientation of the neutron star rotation and magnetic field axes and constrain the mass to radius ratio. This project has been selected for technology development funding by the NASA Explorer Program.

We report on a new instrument that brings high efficiency to x-ray polarimetry, which is the last unexplored field of x-ray astronomy. It derives the polarization information from the tracks of the photoelectrons imaged by a finely subdivided gas pixel detector. The device can also do simultaneously good imaging, moderate spectroscopy and fast, high rate timing down to 150 eV. Moreover, being truly 2D, it is non dispersive and does not require rotation. The great immprovement of sensitivity will allow direct exploration of the most dramatic objects of the x-ray sky; with integrations of the order of one day we could perform polarimetry of Active Galactic Nuclei at the percent level, a breakthrough in this fascinating window of high energy astrophysics.

We report on the development of a new higly efficient polarimeter, based on the photoelectric effect in gas, for the 2-10 keV energy range, a particularly interesting band for x-ray astronomy. We derive the polarization information by reconstructing the direction of photoelectron emission with a pixel gas detector. Attention is focused on the algorithms used in data analysis in order to maximize the sensitivity of the instrument. Monte Carlo simulation is also discussed in details.

A Micropattern detector in the focus of a grazing incidence telescope is nowadays the most powerful tool to perform a sensitive and reliable measurement of the linear polarization of celestial X-ray sources. The actual implementation of such a completely new device results from a trade-off of various factors and can provide a break-through increase of sensitivity with respect to traditional instrumental approaches. The sensitivity depends on the effective area of the optics and the modulation factor and efficiency of the detector. The latter strongly depends on the filling gas through various factors, including the absorption probability, the length of track versus the pixel size, the blurring introduced by the lateral diffusion during the drift. We discuss the impact of the choice of the filling gas on the sensitivity and on the operative band of the instrument, while the noble gases drive the efficiency, the organic quenching gases impact both in reducing the scattering and producing most straight tracks and on reducing diffusion. Some design solution are discussed both for a low energy oriented and high energy oriented polarimeters.

To date polarimetry in astrophysics in the energy domain from hard X-rays up to soft gamma-rays has not been pursued due to the difficulties involved in obtaining sufficient sensitivity. Indeed for those few instruments which are capable of performing this type of measurement, polarimetry itself plays a secondary role in the mission schedule, as the efficiencies and polarimetric Q factors are relatively limited. In order to perform efficient polarimetric measurements for hard X and soft gamma-ray sources, with an instrument of relatively robust and simple design, a CdTe based telescope (CIPHER: Coded Imager and Polarimeter for High Energy Radiation) is under study. This instrument is based on a thick (10 mm) CdTe position sensitive spectrometer comprising four modules of 32×32 individual pixels, each with a surface area of 2×2 mm². The polarimetric performance and design optimisation of the CIPHER detection surface have been studied by use of a Monte Carlo code based on GEANT4 modules. Simulations show that we can achieve a Q factor better than 0.5 and Compton double event efficiency better than 11% between 100 keV and 1 MeV. Herein we will present and discuss the general problems that affect polarimetric measurements in space such as the inclination of the source with respect to the telescope optical axis and space background radiation. Q factor calculations for several beam inclinations as well as for space background together with simulated astronomical sources will be presented and discussed.