This work investigates Bondi accretion to a rotating magnetized star in the "propeller" regime using axisymmetric resistive, magnetohydrodynamic simulations. In this regime accreting matter tends to be expelled from the equatorial region of the magnetosphere where the centrifugal force on matter rotating with the star exceeds the gravitational force. The regime is predicted to occur if the magnetospheric radius larger than the corotation radius and less than the light cylinder radius. The simulations show that accreting matters is expelled from the equatorial region of the magnetosphere and that it moves away from the star in a supersonic, disk-shaped outflow. At larger radial distances the outflow slows down and becomes subsonic. The equatorial matter outflow is initially driven by the centrifugal force, but at larger distances the pressure gradient becomes significant. We find that the star is spun-down mainly by the magnetic torques at its surface with the rate of loss of angular momentum L proportional to −Ω 1.3 * µ 0.8 , where Ω * is the star's rotation rate and µ is its magnetic moment. Further, we find that L is approximately independent of the magnetic diffusivity of the plasma η m . The fraction of the Bondi accretion rate which accretes to the surface of the star is found to be ∝ Ω −1.0 * µ −1.7 η 0.4 m . Predictions of this work are important for the observability of isolated old neutron stars and for wind fed pulsars in X-ray binaries.

This work investigates the propagation of magnetized, isolated old neutron stars through the interstellar medium. We performed axisymmetric, non-relativistic magnetohydrodynamic (MHD) simulations of the propagation of a non-rotating star with dipole magnetic field aligned with its velocity through the interstellar medium (ISM). Effects of rotation will be discussed in a subsequent work. We consider two cases: (1) where the accretion radius R acc is comparable to the magnetic standoff distance or Alfvén radius R A and gravitational focusing is important; and (2) where R acc << R A and the magnetized star interacts with the ISM as a "magnetic plow", without significant gravitational focusing. For the first case simulations were done at a low Mach number M = 3 for a range of values of the magnetic field B * . For the second case, simulations were done for higher Mach numbers, M = 10, 30, and 50. In both cases, the magnetosphere of the star represents an obstacle for the flow, and a shock wave stands in front of the star. Magnetic field lines are stretched downwind from the star and form a hollow elongated magnetotail. Reconnection of the magnetic field is observed in the tail which may lead to acceleration of particles. Similar powers are estimated to be released in the bow shock wave and in the magnetotail. The estimated powers are, however, below present detection limits. Results of our simulations may be applied to other strongly magnetized stars, for example, white dwarfs and magnetic Ap stars. Future more sensitive observations may reveal long magnetotails of magnetized stars moving through the ISM.

Quasi-spherical supersonic (Bondi-type) accretion to a star with a dipole magnetic field is investigated using resistive magnetohydrodynamic simulations. A systematic study is made of accretion to a nonrotating star, while sample results for a rotating star are also presented. We find that an approximately spherical shock wave forms around the dipole with an essential part of the star's initial magnetic flux compressed inside the shock wave. A new stationary subsonic accretion flow is established inside the shock wave with a steady rate of accretion to the star smaller than the Bondi accretion rateṀ B . Matter accumulates between the star and the shock wave with the result that the shock wave expands. Accretion to the dipole is almost spherically symmetric at radii larger than 2R A , where R A is the Alfvén radius, but it is strongly anisotropic at distances comparable to the Alfvén radius and smaller. At these small distances matter flows along the magnetic field lines and accretes to the poles of the star along polar columns. The accretion flow becomes supersonic in the region of the polar columns. In a test case with an unmagnetized star, we observed spherically-symmetric stationary Bondi accretion without a shock wave. The accretion rate to the dipoleṀ dip is found to depend on β ∝Ṁ B /µ 2 , where µ is the star's magnetic moment, and η m the magnetic diffusivity. Specifically,Ṁ dip ∝ β 0.5 andṀ dip ∝ η 0.38 m . The equatorial Alfvén radius is found to depend on β as R A ∝ β −0.3 which is close to theoretical dependence ∝ β −2/7 . There is a weak dependence on magnetic diffusivity, R A ∝ η 0.07 m . Simulations of accretion to a rotating star with an aligned dipole magnetic field show that for slow rotation the accretion flow is similar to that in non-rotating case with somewhat smaller values ofṀ dip . In the case of fast rotation the structure of the subsonic accretion flow is fundamentally different and includes a region of "propeller" outflow. The methods and results described here are of general interest and can be applied to systems where matter accretes with low angular momentum.

Isolated old neutron stars moving through the interstellar medium capture matter gravitationally. If the star is unmagnetized, the captured matter accretes to the surface of the star. However, the stars are expected to be magnetized. Moreover, some of the stars may be in the '' propeller '' stage of evolution. Both the magnetic field and the rotation act to decrease the accretion rate to the surface of the star. Here we consider stars that are past the propeller stage, so that rotation is unimportant. The influence of the magnetic field on the accretion rate to the star's surface is investigated using axisymmetric, resistive magnetohydrodynamic (MHD) simulations. Matter is taken to inflow at the Bondi rate for a nonmagnetized star, and we verify that stationary Bondi accretion flows occur in the absence of a magnetic field. For a magnetized star we find that an outward-propagating shock wave forms and that a new stationary, subsonic accretion flow is set up inside this shock, as first pointed out by Toropin et al. in 1999. Accretion to the surface of the star _ M M occurs along two columns aligned with the magnetic axis of the star. Only a fraction of the Bondi flux _ M M B accretes to the surface of the star. The empirical dependences we find are _where R A is the Alfvén radius. In terms of the star's magnetic moment l, we find _ M M= _ M M B / l À3 . The accretion rate decreases as the magnetic diffusivity of the plasma m decreases, _ M M / m ð Þ 0:6 . We conclude that even a very small residual magnetic field, B $ 10 6 10 8 G, may significantly reduce the accretion rate to the surface of the star and thereby make the accretion luminosity undetectable. The possibility of enhanced accretion owing to three-dimensional instabilities remains to be investigated. The results presented here may also be applicable to wind-fed X-ray stars in binary systems.