A final more speculative potential source of diffuse extragalactic UV emission is the radiative decay of exotic particles of cosmological origin. Since Big Bang theory predicts the existence of a cosmological "sea" of neutrinos having a particle density similar to the photon density of the 2.7 K microwave background radiation, massive neutrinos and similar "inos" of various flavors have long been considered as candidate sources for missing matter. In particular, any type of exotic particle having a present day cosmological particle density of n0 100 cm-3 will be capable of closing the Universe if its mass is of order m 100h2 eV.
In some theories, such massive exotic particles are not stable, but decay into lighter particles under the emission of photons. It is easy to show that the energy of such decay photons is given by
(17) |
where mH and mL are the masses of the heavy and light particles, and the last approximation is valid in the limit where mH >> mL. From this equation it follows that any particle massive enough to provide closure density will have emit its decay radiation at E 50 h2 eV, i.e. at ionizing extreme-UV wavelengths.
Depending on the rate of decay of the particles in question, the accumulated redshift-smeared emission from such particles could give rise to observable UV background radiation. Conversely, the UV background observations severely constrain the exponential decay time of any radiatively decaying cosmologically produced particle with a mass in the range m 10 - 100 eV to be much larger than the age of the Universe, or > 1023 s (Kimble, Bowyer & Jakobsen 1981; Overduin, Wesson & Bowyer 1993).
The intensity and spectral shape of the decay background is easily calculated from equation (1) through insertion of the appropriate line emissivity
(18) |
where n / is the particle density decay rate. This leads to
(19) |
where 0 and l are the observed and decay line wavelengths. The redshifted decay spectrum displays a characteristic jump at 0 = l and drops steeply toward the red as I -2.5 ( 1).
Following the suggestions of Cowsik & McClelland (1972) and De Rújula & Glashow (1980), much attention has in recent years focussed on the concept of decaying massive neutrinos as possible carriers of the missing mass. The astrophysical consequences of this idea have been investigated in some detail by Sciama (1990) who in a series of papers has argued that through suitable tuning of the parameters, massive decaying neutrinos are not only capable of explaining the missing mass problem, but also provide a convenient and omnipresent in situ source of ionizing radiation that is capable of explaining the ionization structure of both the interstellar medium of the galaxy (Sciama 1993) and the intergalactic Lyman forest clouds discussed in Section 2.1 (Sciama 1991). In order to accomplish all this, the neutrino properties need to be rather tightly constrained. Sciama's hypothetical neutrinos have a mass of m 28 eV and decay under emission of photons at a wavelength of 890 Å just below the Lyman limit. The neutrino exponential decay time is 1 - 2 1023 s.
Figure 6 shows the resulting redshift smeared far-UV background calculated for these parameters. Although the edge of the decay spectrum is conveniently hidden by interstellar neutral hydrogen absorption, and therefore un-observable, both the absolute intensity (I (100 - 600) h-1 photons s-1 cm-2 sr-1 Å-1) and the steep blue color of the decay spectrum are at best barely consistent with the existing observations of the UV background (cf. figure 9 of Martin et al. 1991), but can probably not be definitively excluded at this point given the uncertainties in the data (Overduin et al. 1993).
Figure 6. The redshift-smeared UV background predicted for the decaying neutrino model of Sciama (1990). The step in the spectrum at the decay wavelength of 890 Å falls below the Lyman limit (vertical dashed line) and is therefore not observable due to interstellar absorption. |
Other and more persuasive observations that point against the Sciama model have, however, been reported by Davidsen et al. (1991) and Dettmar & Schultz (1992) (and countered in Sciama 1993).