Polarimetry is a particularly powerful technique when imaging circumstellar environments. Currently most telescopes include more or less advanced polarimetric facilities and large telescopes count on it for their planet-finder instruments like SPHERE-ZIMPOL on the VLT or EPICS on the future E-ELT. One of the biggest limitations of this technique is the instrumental polarization (IP) generated in the telescope optical path, which can often be larger than the signal to be measured. In most cases this instrumental polarization changes over time and is dependent on the errors affecting the optical elements of the system. We have modeled the VLT and E-ELT telescope layouts to characterize the instrumental polarization generated on their optical paths using the M&m's code, an error budget and performance simulator for polarimetric systems. In this study we present the realistic Mueller matrices calculated with M&m's for both systems, with and without the setups to correct for the IP, showing that correction can be achieved, allowing for an accurate polarimetric performance.
Well over 700 exoplanets have been detected to date. Only a handful of these have been observed directly. Direct observation is extremely challenging due to the small separation and very large contrast involved. Imaging polarimetry offers a way to decrease the contrast between the unpolarized starlight and the light that has become linearly polarized after scattering by circumstellar material. This material can be the dust and debris found in circumstellar disks, but also the atmosphere or surface of an exoplanet.
We present the design, calibration approach, polarimetric performance and sample observation results of the Extreme Polarimeter, an imaging polarimeter for the study of circumstellar environments in scattered light at visible wavelengths.
The polarimeter uses the beam-exchange technique, in which the two orthogonal polarization states are imaged simultaneously and a polarization modulator is swaps the polarization states of the two beams before the next image is taken. The instrument currently operates without the aid of Adaptive Optics. To reduce the effects of atmospheric seeing on the polarimetry, the images are taken at a frame rate of 35 fps, and large numbers of frames are combined to obtain the polarization images.
Four successful observing runs have been performed using this instrument at the 4.2 m William Herschel Telescope on La Palma, targeting young stars with protoplanetary disks as well as evolved stars surrounded by dusty envelopes. In terms of fractional polarization, the instrument sensitivity is better than 10<sup>−4</sup>. The contrast achieved between the central star and the circumstellar source is of the order 10<sup>−6</sup>. We show that our calibration approach yields absolute polarization errors below 1%.
Although different approaches to model a polarimeter's accuracy have been described before, a complete error
budgeting tool for polarimetric systems has not been yet developed. Based on the framework introduced by
Keller & Snik, in 2009, we have developed the M&m's code as a first attempt to obtain a generic tool to model
the performance and accuracy of a given polarimeter, including all the potential error contributions and their
dependencies on physical parameters. The main goal of the code is to provide insight on the combined influence
of many polarization errors on the polarimetric accuracy of any polarimetric instrument. In this work we present
the mathematics and physics based on which the code is developed as well as its general structure and operational
scheme. Discussion of the advantages of the M&m's approach to error budgeting and polarimetric performance
simulation is carried out and a brief outlook of further development of the code is also given.