The Oxygen Nightglow

Abstract

A brief account is given of the history of the identification of the nightglow spectral features.

Elementary kinetic considerations show that the Chapman process is a less important source of 5577A in the nightglow than is the Barth mechanism involving transfer from a precursor. Re-examination of the 1976 afterglow measurements of Slanger and Black combined with spectroscopic observations by Degen (1972) shows that the 0(1S) precursor in the afterglow is not O₂(A ³∑ ⁺_u ) and is probably not O₂(A′ ³Δ_u). A facet of the afterglow measurements suggests that this is also true of the o(¹S) precursor of the nightglow. The altitude profile of the 5577A emission confirms that the Barth mechanism is responsible but provides no clue to the identity of the precursor.

Consideration of the Atmospheric and Infrared Atmospheric systems is facilitated by the nightglow emission originating from only the v=O vibrational level of each. This allows the quenching rates to be calculated reliably. The observed intensities of the two systems are greater than would be expected from the fractions F(b,M) and F(a,M) of termolecular associations O + O + M that produce O₂(b ¹∑ ⁺_g ) and O₂(a ¹Δ_g) respectively. This is strong evidence for a precursor (even though other processes may contribute). Measurements that purport to show that the two systems necessarily have different precursors are discounted. Should O₂(⁵Π_g) have a dissociation energy of about 0.2 eV (Saxon and Liu 1978) it would quite probably be the precursor for O₂(b ¹∑ ⁺_g ) (with O₂(A′ ³Δ_u) as a possible alternative) and would very probably be the precursor for O₂(a¹Δ_g). However should it have a dissociation energy of only about 0.1 eV (Partridge, Bauschlicher and Langhoff 1986) it could not be the precursor of either of the two states. The precursor would then be O₂(A′ ³Δ_u) and perhaps another of the close metastable trio. The altitude profile of the Atmospheric system emission shows-that there is a precursor but is not a useful guide to its identity.

The quenching of the Herzberg I and II and Chamberlain systems has not been adequately studied in the laboratory. It is reasoned that data of McNeal and Durana (1969) require that quenching of the higher v levels of O₂(A ³∑ ⁺_u ) proceeds more rapidly than the reported (Kenner and Ogryzlo 1983) rate of quenching of the v = 2 level. This is in agreement with a laser experiment by Slanger et al (1984). The same conclusion is reached from the intensity of the Herzberg I system in the nightglow. The intensities of the Herzberg II and Chamberlain systems indicate that O₂(c ¹∑ ^-_u ) and O₂(A′ ³Δ_u) are also quite readily quenched. It is likely that the close metastable trio are coupled by collisions - the collisional coupling between O₂(A ³∑ ⁺_u ) and O₂(A′ ³Δ_u) would be expected to be strong.

Some of the rate coefficients measured in the laboratory have been misinterpreted.