Spectroscopy aims at measuring the flux distribution of a light source (a star for instance) as a function of the radiation wavelength (i.e. colour in the particular case of visible light). These flux distributions are called spectra. Here is for instance what the spectrum of the Sun looks like at optical wavelength, both globally and locally (around a green wavelength of 500 nm):
The local sharp dips we see all around the place are called
spectral lines
and are due to the absorption of the solar radiation from the
solar inner by the atoms of the outer layers (the
photosphere).
The three deeper spectral lines seen above are from neutral iron. Their shape
tells us physical information on the photospheric plasma
(e.g. temperature, turbulence). Nowadays, stellar spectroscopy often involves a
a high-resolution echelle spectrograph comprising at least a diffraction grating,
a cross-dispersing prism, a collimator, a camera and a digital bidimensional CCD
detector. Such an instrument can record a very wide spectral domain (as much as
the whole optical domain from 370 to 900 nm, i.e. from the near ultraviolet to the
near infrared) in a single exposure, like for instance the Fiber-fed Extended Range
Optical Spectrograph (or FEROS)
built by ESO for their 1.5m telescope in Chile.
Radiation from a light source can be polarised. When the electric vector of the
electromagnetic radiation vibrates in a fixed plane, the radiation
is called linearly polarised. When the electric vector describes
an helix about the direction of propagation, the radiation is said
to be circularly polarised. A combination of both usually
yields elliptical polarisation. Polarimetry aims at measuring the
degree to which a radiation from a light source is polarised, as
well the polarisation state of the corresponding light. A polarimeter usually involves
retardation components (cristalline plates or Fresnel rhombs,
retarding one component of the electric vibration with respect to the other by a fixed
amount) in association with a birefringent cristal (which
splits the two orthogonal states of linear polarisation of the incoming beam into
two separate beams). To perform a circular polarisation light analysis for instance,
one normally uses at least a quarter-wave plate (changing the incoming circular
polarisation into linear polarisation) and a birefringent cristal (to search for
potential linear polarisation emerging from the quarter-wave plate).
Polarimetric observations of stellar sources can inform us on the geometry and chemical
composition of circumstellar environments (winds, discs, dust
envelopes) through the continuum polarisation scattering processes
produce. It can also tell us about stellar surface magnetic fields
through the line profile polarisation that the
Zeeman effect
generates in photospheric spectral lines.
Although spectroscopy of unpolarised light and broadband photopolarimetry are quite
common tools for modern astronomers, the combination of both, called
spectropolarimetry, is much more unusual. The coupling of
both instruments is indeed not totally trivial. While the polarisation analysis is
usually performed at the primary or Cassegrain focus of a telescope (to minimise
instrumental polarisation produced by oblique reflections in the telescope),
high-resolution spectroscopy is often done at Coudé focus (for better
spectrograph stability). A double-fibre feed (one fibre
for each orthogonal polarisation state) must therefore be used to convey the light
from the Cassegrain focus down to the Coudé room and couple both foci with
no compromise either on the accuracy of the polarisation analysis or on the
spectrograph stability.
Only very few such instruments exist worldwide. Those we currently use are the
polarimeter of the
MuSiCoS Echelle
spectrograph (on the 2m TBL
telescope of Pic du Midi), or with
Semel's visitor polarimeter on the
UCL Echelle
Spectrograph of the 3.9m
Anglo-Australian Telescope.
I am presently involved in the development of a new generation instrument for the
Canada-France-Hawaii Telescope, named
ESPaDOnS.
Donati J.-F. ,
``ESPaDOnS: An Echelle SpectroPolarimetric Device for the
Observation of Stars at CFHT'',
in: Trujillo Bueno J., Sanchez Almeida J. (eds.),
proceedings of the Solar Polarisation Workshop #3 (2003).
ASP Conf. Series (in press).
Donati J.-F. , Catala C., Mathys G., et al.,
``High resolution spectropolarimetry on the VLT'',
in: Monnet G., Bergeron J. (eds.), ``Scientific Drivers for ESO Future
VLT/VLTI Instrumentation'' (2001). Springer, Berlin (in press)
Donati J.-F. ,
``Spectropolarimetry on giant telescopes'',
in: Mathys G., Solanki S.K., Wickramasinghe D.T. (eds.),
``Magnetic fields across the HR diagram'' (2001). ASP Conf. Series 248, 563
Donati J.-F., Catala C., Wade G.A., Gallou G., Delaigue G.,
Rabou P., ``A dedicated polarimeter for the MuSiCoS
échelle spectrograph''
(1999) A&AS 134, 149
Donati J.-F. , Catala C., Landstreet J.D.,
``ESPaDOnS: An Echelle SpectroPolarimetric
Device for the Observation of Stars at CFHT'',
in: Martin P., Rucinski S. (eds.), `proceedings of the ``fifth CFHT users'
meeting'' (1998). . Springer, Berlin, p. 212
Donati J.-F., Semel M., Carter B.D., Rees D.E., Cameron A.C.,
``Spectropolarimetric observations of active stars''
(1997) MNRAS 291, 658
Semel M., Donati J.-F., Rees D.E.,
``Zeeman-Doppler Imaging of active stars III. Instrumental and technical
considerations''
(1993) A&A 278, 231
© Jean-François Donati, last update on 2003 Jun. 16