Circumstellar Disks from Rotating Stars with and without Magnetic Fields
Presented at the meeting of the Working Group on Active B
Stars during the
26th IAU General Assembly in Prague, Czech Republic on 2006 August 18
Stan Owocki, Rich Townsend, & Asif ud-Doula
Bartol Research Institute, Department of Physics &
Astronomy, University of Delaware, Newark, DE 19711 USA;
owocki,rhdt,asif@bartol.udel.edu
Received: 2007 June 29; Accepted: 2007 July 2.
This talk reviewed recent efforts to develop dynamical models for the effects
of a surface dipole field on radiatively driven wind outflows.
One particular project applies magnetohydrodynamic (MHD) simulations of a
Magnetically Confined Wind Shock (MCWS)
model originally developed by Babel & Montmerle (1997, ApJ L485, 29)
to explain X-ray emission observed by Rosat
(Gagne et al. 1997, ApJ 478, L87)
from the magnetic O7V star θ1 Ori C.
Results are summarized in Figure 1.
We also consider the role of magnetic fields in spinning up the wind
outflow from a rotating star, emphasizing that this does not produce the
Magnetically Torqued Disk (MTD)
proposed by
Cassinelli et al. (2002, ApJ 578, 951)
as a mechanism for producing the orbiting, Keplerian
disks inferred from the characteristic Balmer line emission in Be stars.
Rather, as illustrated in Figure 2, material in the equatorial
compression tends either to fall back on the star, or be ejected
outward.
However, the very strong magnetic fields of Bp stars
can lead to a
Rigidly Rotating Magnetosphere (RRM)
(Townsend & Owocki 2004, MNRAS, 357, 251)
with rigid-body disks or clouds
(see Figure 3) that can explain quite well the observed
emission in Bp stars like σ Ori E (Figure 4).
Moreover, the eventual centrifugal
breakout of such material can lead to strong heating from magnetic
reconnection, which thus could explain the very hard X-ray flares seen from
this star
(ud-Doula et al. 2006, ApJ 640, L191), (Figure 5).
A more complete summary of these results will appear in
Owocki et al. (2007, Phys. Plasma 14, 056502, in press).
Further information, including animations of the simulations, can be
accessed on the web at:
http://shayol.bartol.udel.edu/massivewiki/index.php/Category:Magnetic_wind_confinement .
Figure 1. MHD simulations of the MCWS model for
θ1 Ori C,
showing the
logarithmic density ρ and temperature T in a meridional
plane. Left: at a time 80 ksecs after the initial condition, the
magnetic field has channeled wind material into a compressed, hot disk
at the magnetic equator. Right: at a time 180 ksecs, the cooled
equatorial material is falling back toward the star along field lines,
in a complex `snake' pattern. The darkest areas of the temperature
plots represent gas at T ~ 107 K,
hot enough to produce relatively hard X-ray emission of a few keV.
Figure 2. Density of a 2D MHD simulation for
a star rotating at half the critical with moderate magnetic
confinement,
shown at time snapshots of 90 ksec (left) and 390 ksec (right)
after a dipole field is introduced into an
initially steady-state, unmagnetized, line-driven stellar wind.
The curves denote magnetic field lines, and the vertical
dashed lines indicate the equatorial location of the Kepler, Alvén, and
Escape radii. The arrows denote the upward and downward flow above and below the
Kepler radius, emphasizing that the material never forms a stable,
orbiting disk.
Figure 3. Maps of the optically-thick Hα emission from
circumstellar plasma in an RRM model for σ Ori E,
plotted at five consecutive phases of the stellar rotation cycle
(indicated at the top left of each panel). Note that
the circumstellar emission is dominated by two clouds, edge-on at
phase 0.25 and face-on at phase 0.75.
Figure 4. Time-series spectra of the varying circumstellar Hα
emission observed from σ Ori E (left),
phased on the 1.19-day rotation period of the star, compared against
the corresponding synthetic data predicted by the RRM model (right;
see Fig. 3); white indicates emission relative to the
background photospheric Hα profile, and black indicates
absorption.
Note in particular the model's reproduction of the
observed double S-wave variability, including the blue/red and
temporal asymmetries, and the correct positioning of the eclipse-like
absorptions at phases 0.05 and 0.45.
Figure 5. MHD simulations of a Centrifugal Breakout model for X-ray
flaring, showing the logarithmic temperature T in a meridional
plane. Left: at a time 190 ksecs, the centrifugal force acting on
dense material in the equatorial plane has drawn the magnetic field
out into a long, narrow neck. Middle: at a time 220 ksecs, the
stressed magnetic field has reconnected, heating material in the outer
regions of the equatorial plane to T ~ 108 K.
Right: at a time 240 ksecs, the reconnected field has snapped back toward the
star, producing further heating.
|
![[BULLET]](dingbak.gif)