Anomalous Molecular Rotation and the Temperature of the Upper Atmosphere

Abstract

This experiment verifies the prediction of Oldenberg that the spectroscopically measured rotational temperature of a diatomic gas will be lower than the translational temperature when (1) the pressure is low, (2) the gas is excited by electron impact, and (3) the excited electronic state from which the measured bands are radiated has an equilibrium nuclear separation greater than the internuclear distance in the ground state. For gas temperatures from 400 to 670\ifmmode^\circ\else\textdegree\fi{}K rotational temperatures from the second negative bands of ${\mathrm{O}}_{2}^{+}$ were found in qualitative agreement with the predicted relation ${T}_{\mathrm{rot}}=\frac{{T}_{\mathrm{trans}}{B}^{\ensuremath{'}}}{{B}^{\ensuremath{'}\ensuremath{'}}}$. Upper atmosphere temperatures derived from band profiles in night sky spectra are consistently lower than temperatures estimated from other data. The possible occurrence of anomalous rotation of the night sky molecules casts some doubt on the meaningfulness of the night sky temperature measurements. A partial rotational analysis in the course of this experiment suggests revisions of the ${B}_{0}$ and $\ensuremath{\alpha}$-values for the ${\mathrm{O}}_{2}^{+}$ molecule in the $^{2}\ensuremath{\Pi}_{{u}^{\ensuremath{-}}}$- and $^{2}\ensuremath{\Pi}_{{g}^{\ensuremath{-}}}$-states.