Energy distribution function of translationally hot O(3P) atoms in the atmosphere of Earth

Translationally hot O(³P) atoms are produced in the atmosphere of Earth by photolysis of O₂ and O₃ and quenching of O(¹D). A rigorous kinetic theory analysis of this problem is developed and compared with the approach previously employed by Logan and McElroy [Planet. Space Sci.25, 117 (1977)]. It is shown that the kinetic theory employed by the previous workers is somewhat deficient. With the line-of-centers cross-section, the rates of reactions of the translationally hot O(³P) atoms with other atmospheric gases are calculated and found to be in some instances many orders of magnitude larger than the equilibrium rates. Though the non-equilibrium reaction rates with O(³P) are substantially increased, they are still not competitive with the corresponding reaction rates with O(¹D).     

Linewidth studies on the NI(4S−4P) resonance multiplet

We have measured the linewidths of the NI multiplets $\text{[2p}{}^2\text{3p}{}^4\text{D}{}^0\text{?2p}{}^2\text{3s}{}^4\text{P, λ8691}\text{A}\text{?}\text{; 2p}{}^2\text{3p}{}^4\text{P}{}^0\text{?2p}{}^2\text{3s}{}^4\text{P, λ8212}\text{A}\text{?}\text{; 2p}{}^2\text{3s}{}^4\text{P?2p}{}^3{}^4\text{S}{}^0\text{, λ1200}\text{A}\text{?}\text{]}$ produced in the dissociative excitation of N₂ by energy electrons. The infrared transitions excite the N(⁴P) resonance state by cascade and they account for > 50% of the total N(⁴P) cross section at 100 eV. Both the i.r. and v.u.v. lines are found to be highly Doppler broadened ( ～ 25 times the thermal Doppler line width). These results indicate that dissociative excitation of N₂ produces N (⁴P) atoms with sufficient kinetic energy so that the $\text{λ1200}\text{A}\text{?}$ resonance radiation $\text{[2p}{}^2\text{3s}{}^4\text{P ?2p}{}^3{}^4\text{S}{}^0\text{]}$ emitted by these excited atoms would be optically thin in the Earth's upper atmosphere. We also found that the line strength ratios for the resolved components of the $\text{λ1200}\text{A}\text{?}$ triplet excited by dissociative excitation differ from those predicted by the multiplicities of the states involved and used in current entrapment models; the intensity ratios also vary with the energy of the incident electron. These developments introduce new complications into the analysis of the terrestrial ultraviolet dayglow.     

Energy transfer from excited N2 and O2 as a source of O(1S) in the Aurora

It is proposed that energy transfer from excited O₂ contributes to the production of O(¹S) in aurora. An analysis is presented of the OI5577 Å emission in an IBC II⁺ aurora between 90 and 130 km. The volume emission rate of the emission at these altitudes is consistent with the production rate of O(¹S) by energy transfer to O(³P) from N₂ in the A³Σ₂⁺ state and O₂ in the A³Σ_u⁺, C³Δc¹Σ_u^? states, the N₂A state being populated by direct electron impact excitation and B → A cascade and the excited O₂ states by direct excitation. Above the peak emission altitude (~105 km), energy transfer from N₂A is the predominant production mechanism for O(¹S). Below it, the contribution from quenching of the O₂ states becomes significant.     

O(1S) quenching profile between 75 and 115 km

The rate at which O(¹S) is quenched in the atmosphere has been calculated as a function of altitude in the 75–115 km region. Recent measurements of the temperature-dependent O ₂ quenching rate coefficient have been used, while for quenching by O(³P), an expression combining new theoretical and experimental results is employed. For the O(³P) altitude profile, the Jacchia (1971) model is chosen. The quenching profile shows a pronounced minimum quenching rate at about 87 km. It is concluded that different studies carried out on pulsating Type-B red aurorae, which extract an O(¹S) quenching rate from the time lag between N ₂⁺(B?X) emission and 5577-Åemission, can now be interpreted as indicating an altitude range for these aurorae of 84–89 km. This conclusion is in accord with observations made on artificial aurorae.     

Excitation of O(1S) and emission of 5577 Å radiation in aurora

The results of recent laboratory experiments suggest that the reaction N⁺ + O₂ → NO⁺ + O(¹S) is the principal source of O(¹S) in aurora. A negligible time delay between auroral ionization and O(¹S) production is associated with this indirect process, which is a necessary condition for a viable mechanism. The $\text{5577}\text{A}\text{?}\text{3914}\text{A}\text{?}$ volume emission rate ratio associated with this production source remains constant with altitude. The problems encountered by the currently accepted source of O(¹S), the reaction of N₂(A³Σ) molecules with atomic oxygen are explored, and the contributions of this and other reactions to the auroral green line emission are reevaluated.     

An indirect mechanism for the production of O(1S) in the aurora

Recent laboratory measurements of the deactivation rate constants for O(¹S) have suggested that the dominant production mechanism for the green line in the nightglow is a two-step process. A similar mechanism involving energy transfer from an excited state of molecular oxygen is considered as a potential source of the OI (5577 Å) emission in the aurora. It is shown that the mechanism, $\text{O}{}_2\text{+ e → O}{}_2{}^1\text{+ e O}{}_2{}^1\text{+ O → O}{}_2\text{+ O(}{}^1\text{S}\text{)}$, is consistent with auroral observations; the intermediate excited state has been tentatively identified as the O₂(c¹∑^?_u) state. For the proposed energy transfer mechanism to be the primary source of the auroral green line, the peak electron impact cross-section for O₂¹ production must be approximately 2 × 10^?17 cm².     

Changes in thermospheric composition inferred from twilight O+(2P) emission

Measurements of the twilight enhancement of airglow emission from O⁺(²P) near 7325 Å reveal major changes which accompany geomagnetic activity, no significant distance between evening and morning and an increase in brightness paralleling the approach to solar maximum. The principal source for O⁺(²P) is direct photoionization from O(³P) but at low solar activity there appears to be a contribution from another source in early twilight which may be local photoelectron ionization into O⁺(²P). The geomagnetic and solar effects appear to reflect changes in the O and N₂ density in the thermosphere; ground based twilight measurements of O⁺ emissions thus provide a simple means for monitoring thermospheric structure from 300 km to ～ 500 km at solar minimum and to ～600 km at solar maximum.     

Line shape of the non-thermal 6300 Å O(1D) emission

The line shape of the non-thermal O(¹D) 6300 Å emission is calculated using the two population model of Schmitt, Abreu and Hays (Planet. Space Sci.29, 1095, 1981). The calculated line shapes simulate observations made from a space platform at different zenith angles and altitudes. The non-thermal line shapes observed at zenith angles other than the local vertical have been obtained by using the Addition theorem for spherical harmonics of a Legendre polynomial expansion of the non-thermal population distribution function.     

Observations of horizontal transport effects on high latitude metastable O(1D), N(2D) auroral emissions

An analysis is presented of photometric measurements of the NI (λ = 520nm),OI(λ = 630nm)and other emissions made at Nord, where the invariant latitude is Λ = 80°4. The time variations of the intensities are interpreted in the following way by comparison with simultaneous ground based or satellite measurements.The N(²D) atoms formed in the dayside cleft are carried by the neutral wind in a plume across the polar cap, so that the ratio of λ(630 nm) to λ(520 nm) intensities decreases along the plume with increasing distance from the source region.In the polar cap, but outside the plume region, 630 nm emission is produced by electron impact of polar rain and by substorms that reach high latitudes. Ionization produced at the same time, especially by the substorms, will produce further 630 nm emission through dissociative recombination. In any case, the region outside the plume may be regarded as a source region, with a high value of the ratio $\text{I(630)}\text{I(520)}$. This explains in part the diurnal variations, since this ratio is depressed as Nord crosses the dayside plume.The electron energy along the oval increases progressively from the dayside to the nightside. The intensity ratio increases with increasing electron energy because N(²D) is quenched more rapidly than O(¹D). Thus the ratio rises progressively from noon to midnight.An effect of the interplanetary magnetic field is superimposed on this pattern : as its North-South component B_z increases, the oval contracts so that Nord becomes nearer from the cleft source and the intensity ratio increases on the dayside. The inverse effect is also observed. On the nightside, negative B_z is associated with substorms that produce poleward expansions of the poleward oval boundary, that brings more energetic precipitation to Nord. This causes the intensity ratio to increase with decreasing B_z in a way that is opposite to that for the dayside.     

Associative ionization of N(2D) and O

Recent laboratory measurements have shown that N(²P) atoms, and thus probably hot N(²D) atoms, will recombine with atomic oxygen via an associative ionization process at the gas kinetic rate. While the reaction is endothermic, it has been suggested that this has interesting implications for the upper atmosphere in that N(²D) atoms in the tail of the velocity distribution could provide an additional source of NO⁺ through the reverse of the dissociative recombination reaction

$\text{NO}{}^{\mathrm{+}}\text{+ e ?}\text{N}\text{(}{}^2\text{D}\text{) +}\text{O}$

. It has also been suggested that this process might account for the difference between a laboratory determination of the rate coefficient and that determined from the Atmospheric Explorer Satellite data. In this paper we investigate further the likelihood of the associative ionization of N(²D) and O playing a significant role in the normal ionosphere, in the light of several recent relevant studies. We conclude that the associative ionization process is not an important factor and that a more probable cause for disagreements in the various determinations of the recombination coefficient, is the difference in excited states of the ions in the various experiments.     

Quenching of O+(2D) by electrons in the thermosphere

A major loss process for the metastable species, O⁺(²D), in the thermosphere is quenching by electrons

$\text{O}{}^{\mathrm{+}}\text{(}{}^2\text{D) + e → O}{}^{\mathrm{+}}\text{(}{}^4\text{S) + e}$

.To date no laboratory measurement exists for the rate coefficient of this reaction. Thermospheric models involving this process have thus depended on a theoretically calculated value for the rate coefficient and its variation with electron temperature. Earlier studies of the O⁺(²D) ion based on the Atmosphere Explorer data gathered near solar minimum, could not quantify this process. However, Atmosphere Explorer measurements made during 1978 exhibit electron densities that are significantly enhanced over those occurring in 1974, due to the large increases that have occurred in the solar extreme ultraviolet flux. Under such conditions, for altitudes ? 280 km, the electron quenching process becomes the major loss mechanism for O⁺(²D), and the chemistry of the N⁺₂ ion, from which the O⁺(²D) density is deduced, simplifies to well determined processes. We are thus able to use the in situ satellite measurements made during 1978 to derive the electron quenching rate coefficient. The results confirm the absolute magnitude of the theoretical calculation of the rate coefficient, given by the analytical expression k(T_e) = 7.8 × 10^?8 (T_e/300)^?0.5cm³s^?1. There is an indication of a stronger temperature dependence, but the agreement is within the error of measurement.     

A study of the excitation and radiative decay of the 3s′ 3D0 and 3d 3D0 levels of atomic oxygen

The absolute cross-sections for the excitation of the 989 Å, 1027 Å, 7990 Å, 8446 Å, 1.1287 μm and 1.3164 μm multiplets of atomic oxygen by electron impact dissociation of O₂ are reported. The radiative branching ratios for these transitions are calculated from these results and compared with the NBS compilation of Wiese et al. (1966) and the recent theoretical calculations of Pradhan and Saraph (1977). The cascade models of O⁺ radiative recombination and of electron-impact excitation of the OI(³S) state in the terrestrial airglow are discussed in the light of the laboratory measurements, and the effects of the resonant absorption of components of the λ 989 Å and λ. 1027 Å multiplets by the Birge-Hopfield band system of N₂ are investigated. This process is shown to depend sensitively on the N₂ vibrational temperature and to cause characteristic changes in the OI e.u.v. emission spectrum in auroras and in the sunlit F-region at high exospheric temperatures. It is also suggested that the λ 1027 Å radiation observed in auroral spectra is actually due to molecular nitrogen band emission that has been enhanced by entrapment effects and not to the excitation of the 2p ³P-3d ³D⁰ transition of atomic oxygen as believed previously.     

Electron impact excitation of the ground-state 3P fine-structure levels in atomic oxygen

Fine-structure collision strengths are calculated for transitions between the ground-state levels of atomic oxygen. The R -matrix method is used in which the (2p⁴) ³P, ¹D and ¹S terms are included as well as three pseudo-states chosen to represent the dipole polarizability of the ³P ground state. Very sophisticated configuration–interaction wavefunctions are used for the target states and a recoupling transformation to pair coupling is applied to the inner-region Hamiltonian matrix, with the ³P fine-structure energy levels being adjusted to match the observed splitting prior to diagonalization. Effective collision strengths are obtained by integrating over a Maxwellian distribution and the rate coefficient of the cooling of electron gas is determined. The cooling rates are significantly lower than those currently available, in confirmation of the result deduced from measurements of the Earth's ionosphere electron gas.     

The quenching rate of O(1D) by O(3P)

The rate coefficient for the quenching ofO(¹D) by O(³P) has recently been calculated by Yee et al. (1985). Their results indicate that quenching by atomic oxygen should not be ignored in the analysis of the 6300 Å emission airglow. Data obtained by the Visible Airglow Experiment (VAE) on board the AE satellites have been reanalyzed to determine the quenching rate of O(¹D) by atomic oxygen. The results of this analysis are presented.     

Further quantification of the sources and sinks of thermospheric O1D) atoms

In this paper we confirm an earlier finding that the reaction

constitutes a major source of OI 6300 Å dayglow. The rate coefficient for this reaction is found to be consistent with an auroral result, namely k₁ ≈ 6 × 10^?12cm³s^?1. We correct an error in an earlier publication and demonstrate that reaction (1) is consistent with the laboratory determined quenching rate for the reaction

where $\text{k}{}_2\text{= 2.3 × 10}{}^{\mathrm{?11}}\text{cm}{}^3\text{s}{}^{\mathrm{?1}}$. Dissociative recombination of O⁺₂ with electrons is found to be a major daytime source in summer above ～220 km.     

A discussion of the energy transfer from N2(A3Σu+) to O(1S) in pulsating aurora

In a recent paper, Brekke and Pettersen (1972) have introduced a method for estimating any indirect process in the production of the $\text{O}\text{(}{}^1\text{S}$) atoms in pulsating aurora; for 38 per cent of their data they found that the decay time for the indirect mechanism was shorter than the effective lifetime of the ${}^1\text{S}$ state. These data are related to the energy transfer from the N₂(A³Σ) molecules to the $\text{O}\text{(}{}^1\text{S)}$ state, and evidence is found for this process to contribute in the altitude range below 125 km.     

Collisional excitation of metastable oxygen O(D) atoms through the B∑u channel of O2

Pectroscopic data on the shifts and widths of the energy levels of molecular oxygen have been used in the empirical construction of a diabatic potential matrix that characterizes the interactions of the B³∑⁻_u state with the ⁵Π_u, 2³∑⁺_u, ³Π_u and ¹Π_u states. The diabatic potential matrix is u theory formulation to calculate the cross-sections for the excitation of O(¹D) atoms in collisions of two O(³P) atoms. Total cross-sections are obtained by adding the excitation from the ³Π_g, channel. The rate coefficient for quenching of O(¹D) by O(³P) is evaluated as a function of temperature. The values conflict with a recent analysis of the emission of the oxygen red line in the upper atmosphere.     

The dissociative recombination of N2+ (v = 0, 1) as a source of metastable atoms in planetary atmospheres

The yield of metastable nitrogen atoms in dissociative recombination of N₂⁺ (v = 0, 1)ions has been tudied for different experimental conditions. In a first experiment, the branching ratio for N(²D) production was directly measured as being higher than 1.85; for N₂⁺ (v = 0) this implies that ²D + ²D is the main reaction channel; for N₂⁺ (v = 1) a minor channel could be ²P + ²D, ²P being then quenched toward ²D by electrons. In a second experiment, at higher electron densities, the influence of superelastic collisions was studied; a steady state analysis yields the quenching rate coefficient k₄, of ²D towards ⁴S equal to 2.4 × 10^?10 cm³ s^?1for T_e = 3900 K and shows that ²D + ²D is always the major channel of the reaction for N₂⁺ (v = 1), ²D + ²P being a minor channel. All these results are in good agreement with thermospheric models but imply that N₂⁺ dissociative recombination is a less important source for nitrogen escape of Mars.