Large scale magnetic fields in viscous resistive accretion disks

I. Ejection from weakly magnetized disks

¹ Dublin Institute for Advanced Studies, 31 Fitzwilliam Place, Dublin 2, Ireland e-mail: gmurphy@cp.dias.ie
² Laboratoire d'Astrophysique de Grenoble, CNRS, Université Joseph Fourier, BP 53, 38041 Grenoble, France e-mail: Jonathan.Ferreira@obs.ujf-grenoble.fr
³ INAF – Osservatorio Astronomico di Torino, Strada Osservatorio 20, Pino Torinese, Italy e-mail: zanni@oato.inaf.it

Received: 4 June 2009
Accepted: 21 January 2010

Abstract

Aims. Cold steady-state disk wind theory from near Keplerian accretion disks requires a large scale magnetic field at near equipartition strength. However the minimum magnetization has never been tested with time dependent simulations. We investigate the time evolution of a Shakura-Sunyaev accretion disk threaded by a weak vertical magnetic field. The strength of the field is such that the disk magnetization falls off rapidly with radius.

Methods. Four 2.5D numerical simulations of viscous resistive accretion disk are performed using the magnetohydrodynamic code PLUTO. In these simulations, a mean field approach is used and turbulence is assumed to give rise to anomalous transport coefficients (alpha prescription).

Results. The large scale magnetic field introduces only a small perturbation to the disk structure, with accretion driven by the dominant viscous torque. However, a super fast magnetosonic jet is observed to be launched from the innermost regions and remains stationary over more than 953 Keplerian orbits. This is the longest accretion-ejection simulation ever carried out. The self-confined jet is launched from a finite radial zone in the disk which remains constant over time. Ejection is made possible because the magnetization reaches unity at the disk surface, due to the steep density decrease. However, no ejection is reported when the midplane magnetization becomes too small. The asymptotic jet velocity remains nevertheless too low to explain observed jets. This is because of the negligible power carried away by the jet.

Conclusions. Astrophysical disks with superheated surface layers could drive analogous outflows even if their midplane magnetization is low. Sufficient angular momentum would be extracted by the turbulent viscosity to allow the accretion process to continue. The magnetized outflows would be no more than byproducts, rather than a fundamental driver of accretion. However, if the midplane magnetization increases towards the center, a natural transition to an inner jet dominated disk could be achieved.