Next: The ``Reflection'' Effect Up: Ejectors in Massive Binary Previous: The Radiopulsar PSRB1259-63 - a
The radiopulsar PSRB1259-63 in a massive binary has an extremely high eccentricity (Johnston et al., 1994). The probability of finding a pulsar in an orbit with this eccentricity is at least 0.77. The natural questions arise - how such eccentricity could be explained within the framework of the theory of binary evolution and what is the probability of finding a radiopulsar in such an elongated orbit? To answer these questions we have numerically simulated the evolution of an ensemble of binaries with the aim to calculate the eccentricity distribution of PSR+I binaries. Three factors are known to affect the orbital motion of a pulsar in a binary after the supernova (SN) explosion: (a) the initial eccentricity of the supernova progenitor system; (b) the mass loss during SN explosion; and (c) the additional ``kick'' velocity of a NS caused by anisotropy of collapse. All these factors were taken into account in our simulations.
The first factor - the initial eccentricity - is of importance only in wide systems, because mass exchange leads to orbital circularization during of the common envelope stage, so that such systems ``forget'' the initial eccentricity. Mass loss associated with the SN explosion leads to a decrease of the gravitational binding energy of the binary, and the disruption of the binary occurs if the mass lost exceeds half of the total binary mass prior to the explosion. We have assumed that explosion occurs at a random moment on the orbit and all the mass thrown off by a star instantly leaves the system because the characteristic velocity of matter from the exploded star is much higher than the orbital velocity. The third factor - an additional ``kick'' - is the most uncertain parameter, this could dramatically change destiny of the system .
We have calculated the distribution of eccentricities of binary radiopulsars with normal stars in two cases - with an additional ``kick'' velocity 75 km s and without it. The calculation has been done for all pulsars in pairs with normal stars and separately for visible pulsars (the optical depth for free-free absorption in the stellar wind is less than 1). These distributions are presented in Figure 32.
Figure 32: Distribution of binary NS with normal components over
orbital eccentricities for ejectors (dashed line) and visible pulsars (solid
line) without (left-hand panel) and with ``kick'' velocity 75 km s
(right-hand panel) (Lipunov et al., 1994a).
In the case of the scenario without collapse anisotropy, the systems show a bimodal eccentricity distribution with two maxima - close to 0 and at 1.
The systems with highly eccentric orbit origin from essentially wide pairs. Almost all wide systems are disrupted after the first SN explosion (formal eccentricity is more than 1) and only a few (which could have weak mass exchange) remain in binaries. On the other hand, close binaries with originally high mass ratio (q > 3) remain on almost circular orbit - the primary star 10-20 loses a significant part of its mass during the mass exchange stage and before the SN explosion its mass is less than that of the secondary component. So the total loss due to explosion is less than a quarter of the total mass and the system remains in an almost circular orbit. If the mechanism of an additional ``kick'' is taken into account the distribution of eccentricities is almost uniform (Figure 32). However the distribution of systems with ``visible'' pulsars displays only one maximum at because the pulsars with high eccentricity are in systems with relatively large semi-major axes (so that ). Moreover, these pulsars spend almost all the time at the most distant part of the orbit, which increases the probability of their detection. The probability of finding a pulsar in an orbit with the eccentricity is at least 77 percent.