The Apparent Asymmetric Outflows of TeV Particles from Pulsar Winds

Abstract

Observations of X-ray filaments attached to a couple of powerful pulsars suggest escape of TeV electrons and/or positrons (e$^{\pm}$) from pulsar bow shocks into surrounding large scale magnetic fields. These filaments are usually asymmetric with very weak emission from the other side of the main filaments, and no significant spectral variation has been detected across these filaments, implying inefficient energy loss of emitting particles. We develop a Monte Carlo code to simulate particle transport in a large scale magnetic field and apply the model to PSR B2224+4415 (Guitar). It is shown that, with an injection power of a few tens of percent of the pulsar spin down luminosity, TeV e$^{\pm}$ can explain the observed filament properties with a scattering mean free path along the magnetic field comparable to the length of the observed filament. The model predicts a dim diffuse symmetric X-ray background aligned with the filament on a larger scale, whose flux is proportional to the X-ray emitting e$^{\pm}$ energy loss time for a stable e$^{\pm}$ injection power comparable to the luminosity of this diffuse background. Observations with a large field of view and good sensitivity should be able to detect such a component.

Figures (7)

• Figure 1: Comparison of theoretical formulas with simulations. The purple curves correspond to isotropic scattering. The remaining four curves represent Gaussian scatterings with the standard deviation $\sigma=\pi/q$ with $q$ indicated. The left panel is for $n = 100$ and the right for $n = 600$. After sufficient scatterings with $n>q^2$, the numerical results agree to the theory.
• Figure 2: Projected particle positions in the X–Z plane for $n = 4$ (left) and $n = 100$ (right). Blue dots are for all particles. Red dots mark those with $\theta\approx 30^\circ$.
• Figure 3: Evolution of pitch angle distribution of all particles in the upper (left $Z\ge0$) and lower (right $Z<0$) regions. From top to bottom, $n$ = 0, 1, 4, 100, respectively.
• Figure 4: Dependence of the flux ratio on the scattering times $n$ and the view angles $\Theta$ indicated on the figure. The lines are all for the empirical formula given by equation (\ref{['equ：equ3']}).
• Figure 5: The same as Fig. \ref{['fig:3']} but for stable continuous injection with $n = 20$ (left), and $100$ (right).