COMPLETING THE EVOLUTION OF SUPERNOVA-REMNANTS AND THEIR BUBBLES

We begin a reexamination of the filling fraction of hot gas in the ISM with new calculations of the very long term evolution of supernova remnants and their fossil hot bubbles. We present the results of a one-dimensional (spherically symmetric) numerical simulation of the evolution of a supernova remnant in a homogeneous medium with a nonthermal pressure corresponding to a 5-mu-G magnetic field and density of 0.2 cm-3. Comparison is made with a control simulation having no magnetic field pressure. The evolution was followed from the early Sedov-Taylor phase through the final collapse of the hot bubble after 5.5 x 10(6) yr (B = 5-mu-G) or 7.2 x 10(6) yr (B = 0). We find that the evolutions, once they have become radiative, differ in several significant ways, while both differ appreciably from qualitative pictures presented in the past. Most importantly, over most of the evolution of either case, the hot bubble in the interior occupies only a small fraction of the shocked volume, the remainder in a thick shell of slightly compressed material. The magnetic field exaggerates this aspect, forcing the shell to rebound rapidly as the internal pressure decreases. By following the nonequilibrium ionization in the remnant, we find at late times that bubble gas condensing onto the interior shell wall contains significant quantities of the high-stage ions, O VI, N v, C IV, and Si IV. Column densities and radial distributions of these ions as well as examples of absorption profiles for their strong UV lines are presented. The ratios of the ion column densities in the bubble 0 VI/N V and particularly C IV/N V are close to the observed interstellar values, while Si IV is deficient relative to the other ions. The inclusion of photoionization due to remnant radiation would increase Si IV relative to the other ions, possibly improving the agreement with observations. In addition examples of calculated EUV and FUV emission spectra from the hot bubble are presented and their observability discussed.