First magnetic detection in the innermost regions of a protostellar accretion disc

According to scientists, the Sun and stars, as well as their surrounding planets, are formed from a collapsing cloud of cosmic gas and dust; the flat rotating structure resulting from the collapse is refered to as an 'accretion disc' in the astrophysical jargon. Observations indicate that these accretion discs often produce plasma jets, through a mechanism whose impact on the formation of the central star and surrounding planetary system is not well known. Theoretical models suggest that the magnetic field play a crucial role in this process; yet, no observational constrain existed about these fields. A team of french scientists[1] from CNRS has observed the innermost regions of the protostellar accretion disc FU Orionis using the ESPaDOnS spectropolarimeter[2], recently installed on the Canada-France-Hawaii Telescope[3]. They demonstrated for the first time that a strong magnetic field is present in the accretion disc core, and that the field topology is compatible with model predictions. The field is apparently able to slow down the disc plasma more than what models predict, which may explain why some discs fail at forming jets. These results are published in Nature (2005 Nov 24 issue).

Accretion discs are one of the basic building blocks of modern astrophysics: for instance, scientists think that stars are formed from a rotating gas cloud that collapses into a disc; in a second step, the disc progressively dissipates to become a newly born star with its surrounding planetary system. This scenario is inspired from that proposed 200 years ago by the french mathematician Pierre Simon Laplace to explain the formation of the solar system. Understanding in detail the physics of accretion discs is thus crucial for disclosing the secrets of how the Sun and the planets of the solar system are born.
Accretion discs can also play a role in the innermost regions of some galaxies (called active galaxies) whose cores presumably host black holes billions of times more massive than our Sun. These accretion discs are physically similar to, although much larger in size than, those found around forming stars; they apparently witness the collapse of the whole galaxy towards the central black hole.
Artist view of a protostellar accretion disc. (©David Darling)

What was not predicted by Laplace however are the very thin plasma beams (collimated jets) that apparently escape from accretion disc cores, in a direction perpendicular to the plane of the disc. These jets are observed around forming stars and active galaxies, and can reach incredible lengths, of up to several light-years in the particular case of forming stars. Scientists think that these jets are how the disc succeed at dissipating most of its mass and angular momentum before starting to form the protoplanetary clumps that will eventually produce planets. To produce such collimated jets, theoretical models all require the presence of magnetic fields; however, no observational constraint was yet available on the magnetic field in the innermost disc regions, from which jets presumably originate.
Example of an accretion disc and jet in a protostar (the one showed here is HH30 and not FU Ori) as observed by the Hubble Space Telescope: the jet (in red) is perpendicular to the accretion disc, seen edge-on (and appearing on the bottom of the image, as a dark region between two bright lobes, ©Burrows, STSci/ESA, WFPC2, NASA)

In some models (called magnetocentrifugal models, and initially proposed in 1976), the rotation of the accretion disc twists the initial magnetic field, assumed to be the large-scale primordial (interstellar) magnetic field, oriented perpendicularly to the disc. The field responds by slowing down the disc plasma and causing it to fall towards the disc central regions. The energy flux produced in this process points away from the disc surface, pushing the surface plasma outwards, leading to a wind from the disc and sometimes a collimated jet. Other models (eg dynamo models) suggest that the field is produced within the disc itself, through processes similar to those generating the magnetic field of the Sun.
The rotation of the disc (the flat structure in the centre) twists the initially vertical magnetic field (shown here as yellow ropes), leading to the ejection of plasma (shown here as a blue cylinder) perpendicularly to the disc surface, and to the formation of a collimated jet. This result was obtained through a numerical simulation (©Casse & Keppens 2004).

By detecting the magnetic signatures (through the Zeeman effect) on thousands of spectral absorption lines formed in the inner disc regions (within less than 0.2 astronomical units from disc centre), a team of scientists[1] from the Laboratoire d'Astrophysique de Toulouse-Tarbes (LATT: UMR CNRS, Université Paul Sabatier, Observatoire Midi-Pyrénées) and the Laboratoire d'Astrophysique de Grenoble (LAOG: UMR CNRS, Université Joseph Fourier, Observatoire des Sciences de l'Univers de Grenoble) demonstrated that a strong magnetic field is present, whose intensity is comparable to that emerging from the spots at the surface of the Sun. Moreover, they could establish that the field hosts both a vertical component (perpendicular to the disc) and an azimuthal component (within the disc plane and perpendicular to the disc radius), in agreement with magnetocentrifugal models (and in contradiction with dynamo models). Finally, they find that the field slows down the disc much more than models predict, which may explain why some accretion discs fail at forming collimated jets.
The polarimetric signal from FU Orionis (V/Ic, upper curve, expanded by 100) detected in absorption lines (I/Ic, lower curve) is 2000 times weaker than the radiation energy (Ic) emitted by the disc (©Donati).

This discovery could be achieved thanks to ESPaDOnS[2], the new spectropolarimeter built at Observatoire Midi-Pyrénées (by Groupe d'Instrumentation Grands Télescopes of LATT) and recently installed at Canada-France-Hawaii Telescope (CFHT[3]). The technique of spectropolarimetry consists in measuring the polarisation in the light emitted by an astrophysical object, and in particular the variation of the polarisation through the spectral lines of this object.
The Canada-France-Hawaii Telescope is located atop the Mauna Kea volcano, in the big island of Hawaii (©Cuillandre, CFHT).
This technique, frequently used in solar physics (and in particular for studying the magnetic field of the Sun) is relatively new and thus very promising in other domains of astrophysics. ESPaDOnS is the most powerful instrument worldwide for this kind of studies, and the only one able to detect the very weak polarisation signals from accretion discs.
ESPaDOnS consists of two modules, a polarimeter mounted at the telescope focus (left), fibre-feeding a bench-mounted high-resolution spectrograph (right, ©CFHT).

French scientists, and in particular those from the Toulouse team, are worldwide leaders in the field of stellar spectropolarimetry, both on instrumental and scientific aspects. A copy of ESPaDOnS, dubbed NARVAL[4], will soon be installed on the Bernard Lyot Telescope atop Pic du Midi and will be used in conjunction with ESPaDOnS. An industrial partnership is being negotiated at the moment for building additional copies of ESPaDOnS (eg for China and India).

Scientific contact:

Jean-François Donati, Laboratoire d'Astrophysique de Toulouse-Tarbes, Observatoire Midi-Pyrénées, 14 avenue E. Belin, 31400 Toulouse. Tel: (33) 561332917, Fax: (33) 561332840, email: donati[AT]

Related links:

[1] This team includes JF Donati and F Paletou (LATT), as well as J Bouvier and J Ferreira (LAOG)
[2] ESPaDOnS was cofunded by France (CNRS/INSU, Ministère de la Recherche, LATT, Observatoire Midi-Pyrénées, Laboratoire d'Etudes Spatiales et d'Instrumentation en Astrophysique, Observatoire de Paris-Meudon), Canada (NSERC), CFHT and ESA (ESTEC/RSSD). First light occured at CFHT on 2004 Sept 2.
[3] CFHT operation is funded by Canada (NSERC), France (CNRS/INSU) and the University of Hawaii.
[4] NARVAL is cofunded by Région Midi-Pyrénées (contrat de plan Etat-Région), Conseil Général des Hautes-Pyrénées, European Union (FEDER), CNRS (INSU) and Ministère de la Recherche. First light is planned for mid-2006.
© Jean-François Donati (2005 Nov 28)