Abstract. A number of disparate observational and theoretical pieces of evidence indicate that, contrary to the conventional wisdom, neutron stars' closed field lines are populated by dense, hot plasma and may be responsible for producing some radio and high energy emission. This conclusion is based on eclipse modeling of the binary pulsar system PSR J0737-3039A/B (Lyutikov & Thompson 2005), a quantitative theory of Crab giant pulses (Lyutikov 2007) and a number of theoretical works related to production of non-thermal spectra in magnetars through resonant scattering. In magnetars, dense pair plasma is produced by twisting magnetic field lines and associated electric fields required to lift the particles from the surface. In long period pulsars, hot particles on closed field lines can be efficiently trapped by magnetic mirroring, so that relatively low supply rate, e.g. due to a drift from open field lines, may result in high density. In short period pulsars, magnetic mirroring does not work; large densities may still be expected at the magnetic equator near the Y-point.
SECRET LIVES OF CLOSED FIELD LINESOur understanding of pulsars is based on the Goldreich-Julian model [1], which postulates that pulsars are endowed with dipolar magnetic field; most magnetic field lines close back to the star, while those originating near the magnetic poles are open to infinity. All the action, like generation of radio and high energy emission, is assumed to take place on open field lines. Closed lines are assumed to be nearly dead, only carrying a particle density equal to the minimal charge density required by the electrodynamic conditions.A number of recent observational and theoretical developments indicate that (at least in some parts of magnetosphere) the real particle density on closed field lines exceeds this minimal value by a large factor, of the order of 10 4 − 10 5 and that these particles can produce radio and high energy emission, as we describe in these proceedings.
Eclipses in the double pulsar PSR J0737-3039A/BIn this system a fast recycled Pulsar A with period P A = 22.7 msec orbits a slower but younger Pulsar B which has a period P B = 2.77 sec in tightest binary neutron star orbit of 2.4 hours [2]. In addition to testing general relativity, this system provides a truly golden opportunity to verify and advance our models of pulsars magnetospheres, mechanisms of generation of radio emission and properties of their relativistic winds. This is made possible by a lucky fact that the line of sight lies almost in the orbital plane, with inclination less than half degree [3,4]. Most strikingly, Pulsar A is eclipsed