Probes of pulsar emission physics: The double pulsar and the gamma-ray pulsar population.

摘要

The double pulsar, PSR J0737-3039A/B, is a unique binary system in which both neutron stars have been detected as radio pulsars. We analyze the evolution of the radio emission from the second born, 2.8 s pulsar (pulsar B) based on five years of Green Bank Telescope data since 2003 December. We find that the pulse profile and the flux density of pulsar B change significantly over time, culminating in its radio emission disappearance towards our line of sight since 2008 March. Over this time, the flux density decreases dramatically and the pulse profile evolves from a single to a double peak. This profile shape evolution is likely caused by relativistic spin precession. We explain the profile evolution by an elliptical beam shape geometry model based on geodetic spin precession. By fitting for the observed pulse profile widths, the model constrains the geometry angles of pulsar B, namely the magnetic misalignment from the spin axis alphaB = 61 deg and the spin misalignment from the orbit normal thetaB = 138.5 deg, which are consistent with and similar to those derived by Breton et al. (2008) with a completely different geometry framework. The elliptical beam model predicts that the radio emission reappearance from pulsar B towards our line of sight is expected to happen between 2014 and 2035, with the variation depending on assumptions of the symmetry of the beam.;The strong stellar wind produced by the high spin-down luminosity of the first born, recycled, 23 ms pulsar (pulsar A) of the double pulsar system distorts the magnetosphere of its companion pulsar B. The wind-magnetosphere interaction model determines a bow-shock around pulsar B and it is likely the boundary of its magnetosphere. With geodetic spin precession, pulsar B provides an excellent opportunity to study different emission regions in the magnetosphere. Using the distorted magnetosphere and the well-defined geometrical parameters of pulsar B, we estimate the emission altitude to be ~20 neutron star radii in the bright orbital longitude regions. We further find that the emission altitude varies across the orbit due to the change in the orientation of the bow-shock with respect to our line of sight. Moreover, the emission altitude of pulsar B changes over time due to spin precession.;We then study the pulse profile variation of pulsar A. Analyzing more than six years of data, we confirm that pulsar A does not show a significant pulse width variation over time, which is consistent with previous works. Following a similar geometry framework as for pulsar B, we determine the geometry of pulsar A based on geodetic spin precession, including subtle changes of the pulse width at lower intensity levels from pulse peaks. By knowing the complete geometry of both pulsars, we construct the full geometrical configuration of the system. We find that the relative angle between the spin axes of the two pulsars varies periodically over time. This is the first time that this relative spin angle has been estimated for a double neutron star system.;Finally, we use Fermi Gamma-ray Space Telescope results on non-recycled pulsars to study the gamma-ray pulsar population. We use pulsar detections obtained from the Large Area Telescope (LAT) to constrain how the gamma-ray luminosity L depends on the period P and the period derivative P. Using LAT-measured diffuse fluxes, we place a 2sigma upper limit on the average braking index and a 2sigma lower limit on the average surface magnetic field strength of the gamma-ray pulsar population of 3.8 and 3.2e10 G, respectively. We then predict the number of non-recycled pulsars detectable by the LAT based on our population model. Using the two-year sensitivity, we find that the LAT is capable of detecting emission from about 380 non-recycled pulsars, including 150 currently identified radio pulsars. Using the expected five-year sensitivity, about 620 non-recycled pulsars are detectable, including about 220 currently identified radio pulsars. We note that these predictions are significantly dependent on our model assumptions. (Abstract shortened by UMI.).