Annu. Rev. Astron. Astrophys. 1991. 29: 59-88
Copyright © 1991 by . All rights reserved

Next Contents Previous

4. THE SPECTRUM OF THE FAR ULTRAVIOLET BACKGROUND

4.1 Theoretical Suggestions

A wide-ranging set of processes that would produce spectral features in the far ultraviolet background has been suggested. Early on, when data were limited and the flux was thought to be extragalactic, speculation centered on processes that might be associated with galaxy formation in the early universe, or processes in a possible intergalactic medium. For a discussion of these early suggestions, the reader is referred to the reviews by Davidsen, Bowyer, and Lampton (1974) and Paresce and Jakobsen (1980).

With the realization that the far ultraviolet background might well be dominated by Galactic radiation, entirely different classes of source mechanisms were advanced. Duley and Williams (1980) suggested H2 fluorescence as a contributor to this background. The ratio of molecular to atomic hydrogen is determined primarily by the balance of the formation of molecules on dust grains and their destruction by photoabsorption of ultraviolet light in the interstellar radiation field. Duley and Williams noted that a byproduct of the destruction process which was observable, in principle, was emission in the far ultraviolet. This emission consists of two components. The first is due to transitions between various rotational and vibrational levels within different electronic states of the H2 molecule when incident photons in the 845-1108 Å range are absorbed in the Lyman and Werner bands from the ground state, which are then reemitted in far ultraviolet bands with lambda leq 1845 Å through decays to excited vibrational levels of the ground electronic state. The second process involves continuum emission that arises during decays to unbound levels of the ground state. This process leads to dissociation of the molecule, with a probability per absorption of ~ 10%. Jakobsen (1982) computed a detailed spectrum of this radiation and showed that, in reasonable conditions in the Galaxy, this process could contribute up to 30% of the observed diffuse background.

Jakobsen and Paresce (1981) suggested that a hot (~ 105 K) Galactic corona would produce emission lines in the far ultraviolet. Multiply ionized carbon, nitrogen, oxygen, and silicon were predicted to produce the strongest lines in this gas. Although the intensities derived by these authors were below the sensitivity of instrumentation available at the time, they showed that this flux might be detectable with future instrumentation. Edgar and Chevalier (1986) refined these calculations to account for the fact that interstellar gas at ~ 105 K is at the peak of its cooling curve, and hence a more realistic emission spectrum would result from time-dependent ionization calculations in a gas cooling from a higher temperature. The intensities they obtained were substantially different from those obtained by Jakobsen and Paresce for some lines because of the persistence of highly ionized species as cooling occurs. They also showed that by measurement of an appropriate set of lines, one could, in principle, determine both the character of the cooling and the mass inflow rate of the cooling material.

Deharveng, Joubert, and Barge (1982) investigated the possible contribution of a warm (~ 104 K) intercloud medium to the far ultraviolet background. They concluded that for lambda < 2000 Å, an intercombination line of CIII is a potential emission line. They also showed that a potential source of continuum emission is hydrogen two-photon emission. Recombination of ionized hydrogen populates the 22 S level both by direct recombination and by cascades following recombination to higher levels. This emission becomes stronger than the decreasing HI free-bound and free-free continua below 2700 Å, and it reaches a peak at about 1400 Å, where it exceeds these continua by more than two orders of magnitude.

Next Contents Previous