![]() | ![]() |
Astron. Astrophys. 364, 587-596 (2000) 1. IntroductionSolving the radiative transfer equation for each of the four Stokes parameters is not straightforward, and complicated geometries often lead to extremely complicated equations. Therefore numerical methods, such as the Monte Carlo method (see e.g. Cashwell & Everett 1959), are often used to approximate the solution. Monte Carlo simulations have been used to estimate the polarization levels of axisymmetric systems, such as galaxies (Bianchi et al. 1996) and disks around Be stars (Wood et al. 1996). I have used the Monte Carlo method to simulate the polarization of axisymmetric pure electron scattering accretion disks. These disks are similar to those observed in two types of low-mass X-ray binaries (LMXB:s): Z sources and soft X-ray transients (SXT:s) in outburst. These systems have high mass accretion rates and X-ray luminosities. In LMXB:s the outer accretion disk is heated by the X-rays from the compact star and the inner accretion disk. X-rays heat the disk often to temperatures above the hydrogen ionization temperature (King et al. 1997). Therefore in systems with high accretion rates the disk is completely ionized. Only very few observations of linear polarization from LMXB:s have
been published. (Dolan & Tapia 1989) have observed the transient
source A0620-00, and (Egonsson & Hakala 1991) have observed the
eclipsing system Her X-1. In both systems the polarization was
variable with amplitudes The simulations produced polarization levels of the order 1% in white light. The geometry of the radiating area was the most important factor affecting the polarization. My results indicate that if a disk model can be constructed on the basis of other observations, some on constraints vertical emissivity structure can be derived from polarization measurements. ![]() ![]() ![]() ![]() © European Southern Observatory (ESO) 2000 Online publication: January 29, 2001 ![]() |