next up previous contents index
Next: Collaborations Up: Observing modes Previous: Integral light mode   Contents   Index

Polarimetric mode

Methods to measure the polarisation status of light are based on intensity measurements at different relative orientation of polarimetric components or at different orientations of the equippement relative to the light source. In a generalised polarimeter, a Retarder, with retardance $\Delta$ and reference axis forming and angle $\beta$ with respect a reference system, is combined with a Polariser whose transmission axis forms an angle $\alpha$ with the reference system. Clarke & Grainger() have shown the advantage of double-beams techniques with respect to single-beam ones, so we discarded Polariser, e.g.the Nicol prism, where one of the linearly polarised beam is extinguished.

Two methods are commonly used to perform spectropolarimetry, one is based on a $\lambda/2$ and a $\lambda/4$ retarder that can be alternatively inserted along the optical path and rotated with reference to a beam displacer to measure the linear and the circular polarisation respectively. The other method to measure the light polarisation, often called dual-waveplate method, is based on a $\lambda/2$ and a $\lambda/4$ retarders both inserted along the optical beam. The optical axes of these two retarders are opportunely rotated with respect to a beam displacer to measures the linear and the circular polarisation.

It has to be noted that both methods need the same number of exposures to measure the linear and the circular polarisation. The advantage of the dual-waveplate method is the absence of mechanisms necessary to exchange the retarders, however this method adds all the defects of the two retarders and important, for us, it reduces the overall efficiency of the polarimeter. Thus we decide not to follow the dual-waveplate method.

We use achromatic lambda retarders supplied by the Bernhard Halle Nachfl. GmbH (Germany). Two-part combination of quartz and MgF$_2$ plates cemented between plane parallel plates of crown glass BK7 with antireflection coating. Accuracy of path difference is $\pm$3%.

To spatially separate the ordinary and extraordinary beams, we have adopted a Savart plate. This device comprises two beam displacing plates with their optic axes orthogonal. An incident beam propagating through the first plate is resolved into ordinary and extraordinary beams which are displaced from each other in a first principal plane. Upon entering the second plate, the ordinary beam becomes an extraordinary beam, and vice-versa. Displacing again takes place in a second principal plane, which is orthogonal to the first one. The result is two emerging beams displaced along a diagonal. The displacement is $\sqrt{2}$ times that of each plate.

Savart length was chosen to have a 2 mm separation to feed the two fibes. These mounted on the same mechanic support and coupled to the telescope by micro lenses. The optical path difference between the two beams emerging from the Savart plate is zero, a very important property to reduce errors in producing the mechanical parts.


next up previous contents index
Next: Collaborations Up: Observing modes Previous: Integral light mode   Contents   Index
Innocenza Busa' 2005-11-14