SPITZER OBSERVATIONS OF SUPERNOVA REMNANTS. II. PHYSICAL CONDITIONS AND COMPARISON WITH HH7 AND HH54

We have studied the shock-excited molecular regions associated with four supernova remnants (SNRs)-IC443C, W28, W44, and 3C391-and two Herbig-Haro objects-HH7 and HH54-using Spitzer's Infrared Spectrograph (IRS). The physical conditions within the observed areas (roughly similar to 1' x 1' in size) are inferred from spectroscopic data obtained from IRS and from the Short and Long Wavelength Spectrometers on board the Infrared Space Observatory (ISO), together with photometric data from Spitzer's Infrared Array Camera. Adopting a power-law distribution for the gas temperature in the observed region, with the mass of gas at temperature T to T+dT assumed proportional to T(-b)dT, the H-2 S(0) to S(7) spectral line maps obtained with IRS were used to constrain the gas density, yielding estimated densities n(H-2) in the range of similar to (2-4) x 10(3) cm(-3). The excitation of H-2 S(9) to S(12) and high-J CO pure rotational lines, however, require environments several times denser. The inconsistency among the best-fit densities estimated from different species can be explained by density fluctuations within the observed regions. The best-fit power-law index b is smaller than the value 3.8 predicted for a paraboloidal C-type bow shock, suggesting that the shock front has a flatter shape than that of a paraboloid. The best-fit parameters for SNRs and Herbig-Haro objects do not differ significantly between the two classes of sources, except that for the SNRs the ortho-to-para ratio (OPR) of hot gas (T > 1000 K) is close to the local thermal equilibrium value 3, while for HH7 and HH54 even the hottest gas exhibits an OPR smaller than 3; we interpret this difference as resulting from environmental differences between these classes of sources, molecular material near SNRs being subject to stronger photodissociation that results in faster para-to-ortho conversion. Finally, we mapped the physical parameters within the regions observed with IRS and found that the mid-lying H-2 emissions-S(3) to S(5)-tend to trace the hot component of the gas, while the intensities of S(6) and S(7) are more sensitive to the density of the gas compared to S(3) to S(5).