Sodium in the Stratosphere¹

For a long time the nature of the yellow emission line observed in the spectrum of the night sky remained unknown. Numerous determinations of the wavelength of this line, made beginning in 1929 by Slipher, Dufay, Sommer, Cabannes, Vegard, and Tönsberg, gave results very close to the sodium \(D\) line, but all differing by several angstroms. New determinations made in 1935–1937 by Monnellen, Cabannes, Dufay, and Gauzit gave the value \(\lambda 5894 \pm 1\,\text{Å}\). A careful study of spectrograms of the night sky, including those taken by Dufay in 1932 with a thin wire stretched along the slit, led to the conclusion that the yellow line actually consists of two close lines. All this indicated that it belongs to sodium.

The definitive verification of this hypothesis was carried out by Cabannes, Dufay, and Gauzit in 1938 with a Fabry–Perot interferometer. In this interferometer, with an air gap thickness of \(0.15\,\text{mm}\), the band from the sodium \(D\) line lies between the bands from \(D_2\), while at a thickness of \(0.30\,\text{mm}\) the two systems of bands are superposed. Photographs of the yellow radiation made at both thicknesses gave a system of bands exactly coinciding with the bands from \(D_1\) and \(D_2\). Such coincidence was obtained both for night photographs and for photographs taken at twilight, in the morning and evening, when the intensity of the yellow radiation increases greatly.

The yellow radiation cannot be ascribed to interstellar sodium—the strengthening of it toward the horizon, as well as diurnal and annual fluctuations of its intensity, coinciding with fluctuations in the intensity of the green line (\(\lambda 5577\)), prove its terrestrial origin. Using the ratio of intensities at the horizon and at the zenith, measured by Garrag, it was possible to estimate the height of the luminous layer (regarding it as thin). It proved to be about \(130\,\text{km}\).

Cabannes, Dufay, and Gauzit believe that sodium in the stratosphere is of cosmic origin. The amount of matter delivered to the Earth by meteors is \(4\,\text{g}\) per \(1\,\text{km}^2\) per year, the mass fraction of sodium being about \(0.7\%\). It follows from this that the Earth receives \(2.5 \cdot 10^7\) sodium atoms per \(1\,\text{m}^2\) per second. The intensity of the yellow radiation shows that the number of transitions of the type \({}^2P \to {}^2S\) (yielding the \(D_1\) and \(D_2\) lines) is \(2 \cdot 10^{11}\) per second per \(1\,\text{m}^2\). Thus each atom accounts for about 8,000 transitions which it must undergo before it “dies.” Of the other elements present in sufficient quantity in meteorites, only calcium and aluminum can give bright lines. And indeed, in the spectrum of the night sky there is observed a line \(\lambda 4226\), coinciding with the resonance line of neutral calcium \(\lambda 4226.73\).

It should be noted that the main evaporation of meteors usually occurs at altitudes lower than \(130\,\text{km}\), and therefore this value for the height of the luminous layer may be an overestimate.

B. Levin, Moscow

Literature

1. J. Cabannes, J. Dufay a. J. Gauzit, Astrophys. J., 88, 164, 1938.