Stenflo

Solar polarimetry with ZIMPOL:Plans for the future: 

Solar polarimetry with ZIMPOL: Plans for the future J.O. Stenflo Institute of Astronomy ETH Zurich, Switzerland ZIMPOL at IRSOL

Main sources of noise in imaging polarimetry: 

Main sources of noise in imaging polarimetry Seeing noise Gain-table noise A polarization image, e.g. Stokes Q/I, is formed from the difference of two images in orthogonal polarization states. If the two images are recorded simultaneously (spatially separated), the seeing noise will subtract out, but the gain-table noise will remain. If the same detector area is used for the two images but they are temporally separated, they will contain different seeing noise, unless the temporal separation is less than a few milliseconds. This problem may be solved if the modulation is fast enough (in the kHz range). Problem: Large CCD detectors have slow readout.

Principle of ZIMPOL: 

Principle of ZIMPOL Problem of incompatibility between fast modulation and slow readout solved by creating fast, hidden buffer storage areas within the CCD sensor, and cycle the photo charges between these storage areas in synchrony with the modulation. Four simultaneous image planes are therefore available, one exposed and three hidden. Only after many thousand modulation cycles, after the CCD has been filled , it is read out. As the polarization images are formed from the ratio difference / sum, the gain table divides out completely.

Implementation of the ZIMPOL idea: 

Implementation of the ZIMPOL idea The four image planes are created by depositing a mask on the sensor during the fabrication process, such that for each group of four pixel rows three are masked and one is open. The photo charges are cycled horizontally between the rows in each group. Since this is done in synchrony with the modulation, the image planes correspond to different polarization states. ZIMPOL can be used with any type of modulation system. Future CMOS technology would allow vertically hidden fast buffers, so that the trick of masking a CCD would become obsolete. This technology however lies at least several years in the future. In the meantime such systems are operationally emulated via the ZIMPOL solution.

Version of ZIMPOL for the simultaneous recording of all four Stokes parameters, without light losses on the masked areas: 

Version of ZIMPOL for the simultaneous recording of all four Stokes parameters, without light losses on the masked areas The microlenses, which increase the light efficiency by nearly a factor of four, were successfully implemented for the first time only last month, in an observing run at the SST (La Palma) in October 2006.

Modulator types used with ZIMPOL: 

Modulator types used with ZIMPOL Piezoelastic modulators (PEMs). Almost all our published results about the Second Solar Spectrum have been obtained with a PEM. Ferro-electric liquid crystals (FLCs). Pockels cells.

PEM: Advantages and disadvantages: 

PEM: Advantages and disadvantages Advantages Easy to handle Superior optical quality over large area Excellent transmission in the UV down to the atmospheric cut-off near 300 nm Modulation efficiency optimized electronically for any given wavelength Disadvantages Resonant device. Therefore wave form cannot be chosen (always sinusoidal) modulation frequency cannot be chosen (in our case 42 and 84 kHz) very difficult to phase-lock two modulators, as needed for the simultaneous recording of all Stokes parameters Since we have not yet succeded with phase-locking two modulators, we have to make two successive recordings, one for I,Q,V, the other for I,U,V (after mechanically rotating the modulation package 45 degrees).

FLCs and Pockels cells: 

FLCs and Pockels cells Advantages Non-resonant devices. Therefore modulation frequency can be chosen wave form can be rectangular phase-locking of two modulators is no problem Disadvantages It has not yet been possible to manufacture Pockels cells with the needed aperture size having sufficient optical quality. The optical quality and transmission of FLCs are inferior to PEMs. FLCs cannot be used in the UV below 400 nm. The Second Solar Spectrum is extremely rich in the UV. FLCs age rather fast, while PEMs live almost forever.

Scientific programs with ZIMPOL: 

Scientific programs with ZIMPOL Second Solar Spectrum in the spectrograph focal plane, at IRSOL and McMath (Kitt Peak) with narrow-band filters, at DST (Sac Peak), SST (La Palma), and IRSOL Search for extra-solar planets (the Planetfinder project for the ESO VLT next- generation instrumentation) Jupiter, recorded with ZIMPOL at McMath (Kitt Peak)

Slide10: 

Examples of Ca II K polarization in various magnetic regions Weakly magnetic region, m = 0.96 Active region, m = 0.9 Sunspot, m = 0.61

Slide11: 

Hyperfine structure components due to the odd barium isotopes (with nuclear spin 3/2), which represent 18% of the Ba abundance Central component due to the even isotopes (with zero nuclear spin) Hyperfine splitting in scandium Hyperfine structure and isotope effects Hyperfine splitting in barium

Slide12: 

Both sodium and barium have nuclear spin 3/2, so the J = 1/2 lower and upper states are split into two hyperfine states with F = 1 and 2. However, even when both hyperfine structure and optical pumping (to create ground-state polarization) are taken into account, the QM theoretical framework fails by two orders of magnitude and gives the wrong symmetry when trying to reproduce the observed core polarization. J = 1/2 F = 2 F = 1 The D1 enigma Na I D1 Ba II D1 Stenflo et al. 2000

Slide13: 

Sr I 4607 Å, a photospheric line Ca I 4227 Å, a chromospheric line Difference in Hanle signatures between photospheric and chromospheric lines Model idealizations: turbulent field (photosphere) and canopy field (chromosphere)

Slide14: 

Scattering polarization in CN lines in magnetic environments: 3771 – 3775 Å

Slide15: 

Cr I 3593.5 Å inside and outside limb faculae

Slide16: 

Spatial distribution of the scattering polarization Relation to the network and to the Zeeman effect Different filter positions in the Na D2 and D1 lines (band width 0.2 Å) Observations with ZIMPOL in combination with the UBF/DST at Sac Peak Stenflo et al. 2002

Future developments for solar applications: 

Future developments for solar applications Construction of ZIMPOL-3, which has a completely overhauled, modernized electronic design and larger CCD detectors, equipped with microlenses. Two such complete ZIMPOL systems (including optics) will be available, one for permanent use at IRSOL, the other for use in special observing campaigns at external telescopes. A tunable narrow-band filter system based on two Y-cut lithium-niobate Fabry-Perot etalons will be used in combination with ZIMPOL for monochromatic Stokes vector imaging of the scattering polarization and the Zeeman-Hanle effect. A mobile telecentric version is for use at external telescopes, a collimated version for use at IRSOL. An adaptive optics system is being implemented at IRSOL for use with ZIMPOL.

Future situation with uncertainties: 

Future situation with uncertainties Stenflo retires from ETH end of November 2007 and will no longer control funding for ZIMPOL after that. The Stenflo succession will only be known a year from now. IRSOL will remain a home base for the ZIMPOL systems, complemented by special observing campaigns at external telescopes. THEMIS is considered for the next external campaign, in 2007. It would be the first time to combine ZIMPOL with THEMIS. We are planning to use an FLC-type modulation system in the THEMIS Cassegrain focus. An observing campaign in the second half of 2007 is being considered.