paper

Turbulence Dissipation and Particle Injection in Non-Linear Diffusive Shock Acceleration with Magnetic Field Amplification

arXiv:0807.1321 · doi:10.1086/592240

Abstract

The highly amplified magnetic fields suggested by observations of some supernova remnant (SNR) shells are most likely an intrinsic part of efficient particle acceleration by shocks. This strong turbulence, which may result from cosmic ray driven instabilities, both resonant and non-resonant, in the shock precursor, is certain to play a critical role in self-consistent, nonlinear models of strong, cosmic ray modified shocks. Here we present a Monte Carlo model of nonlinear diffusive shock acceleration (DSA) accounting for magnetic field amplification through resonant instabilities induced by accelerated particles, and including the effects of dissipation of turbulence upstream of a shock and the subsequent precursor plasma heating. Feedback effects between the plasma heating due to turbulence dissipation and particle injection are strong, adding to the nonlinear nature of efficient DSA. Describing the turbulence damping in a parameterized way, we reach two important results: first, for conditions typical of supernova remnant shocks, even a small amount of dissipated turbulence energy (~10%) is sufficient to significantly heat the precursor plasma, and second, the heating upstream of the shock leads to an increase in the injection of thermal particles at the subshock by a factor of several. In our results, the response of the non-linear shock structure to the boost in particle injection prevented the efficiency of particle acceleration and magnetic field amplification from increasing. We argue, however, that more advanced (possibly, non-resonant) models of turbulence generation and dissipation may lead to a scenario in which particle injection boost due to turbulence dissipation results in more efficient acceleration and even stronger amplified magnetic fields than without the dissipation.

Submitted to ApJ 30 April 2008. 20 pages, 6 figures made with 8 Postscript files. Last version has minor corrections and uses emulateapj.cls to reduce page count

References in corpus (18)

Cited by in corpus (46)