Search

Transmission losses noted in cooled UV-quality synthetic fused silica windows were investigated. A deuterium lamp was employed as a source of continuum emission over the 400-190-nm range considered. Even slight cooling was found to produce a significant transmission change at wavelengths shorter than about 320 nm. Attempts to clean the windows produced little change in the results, indicating that surface contamination was not responsible for the effect.

Recent sounding rocket and satellite studies suggest that simultaneous measurements of the O I 989-A and 1304-A resonance lines and of the forbidden 1172.6-A and 1641.3-A transitions would form the basis of a useful remote sensing technique for measuring the O I density and optical opacity of a planetary or stellar atmosphere. Because the 1172.6-A and 1641.3-A emissions are weak lines and are emitted in a wavelength region rich in spectral features, it is important to determine whether typical flight instruments can make measurements with sufficient spectral purity so that the remote sensing observations will yield accurate results. A detailed, high-resolution study of the far UV emission features in the regions surrounding the atomic oxygen transitions at 1172.6 and 1641.3 A was made. These spectra, which were excited by electron impact on O2 and N2, are presented in an attempt to display some potential sources of interference in aeronomical measurements of these O I lines. Both atomic and molecular emissions are found, and the spectral resolution necessary to make unambiguous measurements is discussed.

The cross section for the dissociative excitation of the O I 1172.6 A multiplet is measured by the bombardment of O2 by 100 eV electrons. A cross section value of about 1.56 x 10 to the -22nd sq cm + or - 30 percent and an O I 1173 A/O I 989 A branching ratio of about 0.00013 + or - 35 percent are obtained. The branching ratio values are compared to the data of Morrison (1985) and Garstang (1961). The data reveal that the 1173 A branch from the O I (3s prime 3D0) state is not large enough to account for the discrepancy between the predicted and observed O I 989 A intensity in the earth's upper atmosphere.

Metastable N(+)(5S) ions were produced in the laboratory by dissociative excitation of N2 with energetic electrons. The resulting radiative decay of the N(+)(5S) state was observed with sufficient resolution to completely resolve the doublet from the nearby N2 molecular radiation. The excitation function was measured from threshold to 500 eV. The cross section peaks at a high electron energy and also exhibits a high threshold energy both of which are typical of dissociative excitation-ionization processes. This finding complicates the explanation of electron impact on N2 as the mechanism for the source of the 2145 A 'auroral mystery feature' by further increasing the required peak cross section. It is suggested that the apparent N(+)(5S) quenching in auroras may be an artifact due to the softening of the electron energy spectrum in the auroral E region.

The ratio of the cross sections for the direct and dissociative excitation of the OI(3s 3S0-2p 3P; 1304 A wavelength) transition, sigma A/sigma D, are accurately determined, and the sigma A/sigma D ratio is directly normalized to the ratio of the O(+) and O2(+) ionization cross sections using a high-density diffuse gas source, an electrostatically focused electron gun, a vacuum-ultraviolet monochromater, and a quadrupole mass spectrometer for simultaneous optical and composition measurements. Using revised sigma A(1304 A) values calculated with new calibration standards, the shape of the cross section for the excitation of the O(3s 3S0) state agrees well with previous results, though the absolute magnitude of sigma A(1304 A) is smaller than the results of Stone and Zipf (1974) by a factor of 2.8. The revised cross sections agree well with recent quantum calculations when cascade excitation of the 3s 3S0 state is taken into account.

The cross section for the dissociative excitation of the forbidden OI(2p41D-3s3S, 1641.3 A) transition by electron impact on O2 is measured. At 100 eV, a cross section value of 1.22 x 10 to the -23/sq cm + 2 - 35 percent is found. This result is consistent with recent estimates of this branching ratio from theoretical calculations and aeronomical observations and supports the suggestion by Meier et al. (1985) that observation of the OI 1641.3 A line would be useful for remote monitoring of atomic oxygen in the upper atmosphere.

Single and double ionization of atomic oxygen by electron impact has been studied experimentally. The energy dependence of the specific ionization cross sections sigma(O+) and sigma(O2+) has been measured in the energy range 40-300 eV using a high-density atomic oxygen source not contaminated by metastable O(D-1) or O(S-1) atoms. The ratios of the measured cross sections to absolute values are given in a table.

Revised cross-section values for the excitation of three O I resonance transitions at 1304, 1027, and 989 A, by electron impact on atomic oxygen are presented from threshold to 300 eV. These results are smaller than the excitation cross sections used in some airglow models by a factor of about 2.8. The revised values are in good agreement with recent quantum-scattering calculations. The downward revision is required by new laboratory studies in which the direct and dissociative cross sections for 1304 A excitation were normalized with small probable error to the O and O2 ionization cross sections. The results also reflect new advances in VUV optical calibration techniques. A number of outstanding airglow problems are simplified by these revisions.

Doppler broadening of the 8691, 8212, and 1200-A multiplet lines of N I is investigated experimentally, and its implications for the interpretation of the earth's 1200-A UV dayglow are considered. A regulated 100-eV, 1-mA electron beam is passed through N2 at 300 K and about 0.0005 torr flowing through a collision chamber within a UHV system, and the radiation emitted is observed with a temperature-stabilized short-focal length monochromator with a bandpass of 0.2 A in the IR and an effective UV resolution (in second-order operation with a 3600-groove/mm plane grating) of about 0.04 A. Both the IR and VUV lines are found to be broadened to about 25 times the thermal Doppler linewidth, with the IR transitions accounting for more than half of the total N(4P) cross section at 100 eV. The kinetic energy of the N(4P) atoms produced by dissociative excitation is such that their 1200-A resonance radiation (2p2 3s4P - 2p3 4SO) would be optically thin in the upper atmosphere, contrary to what has been observed. A need to revise some aspects of current UV-dayglow models is identified.

The results of experimental observations of the Cameron bands of CO produced by electron impacts with CO and CO2 are presented, noting that the bands have been detected in the atmospheres of both Mars and Venus. The study was initiated to account for Conway's (1981) data that the Martian airglow displayed cross-sectional dissociative excitation of the Cameron bands seven time larger than laboratory measurements by Ajello (1971). Spectrometer measurements were obtained of processes occurring in a chamber filled with CO or CO2 gas being bombarded by an electron beam. A value three times higher than the previous lab oratory estimate was obtained. Reasons for the discrepancy are discussed, including a present factor of two error in laboratory estimates.

We have successfully used a closed-cycle cryogenic cooling system as a high-speed UHV pump for instruments carried into the upper atmosphere on board sounding rockets. The device utilizes a commercially available, Sterling-cycle refrigerator that has proven to be rugged and simple to use under field conditions.

In the terrestrial atmosphere, nitrous oxide (N2O) has a major role in the chemistry of ozone. Current atmospheric models assume that N2O is produced only by fixation at the earth's surface and that there are no local sources in the stratosphere or mesosphere. It is pointed out here that a significant in situ N2O source does exist above 20 km due to the excitation of the metastable N2(A 3Sigma u +) state by resonance absorption of solar UV photons that penetrate deeply into the atmosphere through the 1,800-2,200 A O2-O3 window. This source significantly affects the NO altitude distribution in the mesosphere and, in the earth's prebiological atmosphere, made N2O an important stratospheric constituent.

A mechanism by which solar proton (SP) events and relativistic electron precipitation (REP) events may lead to the production of stratospheric NO is described. The process comprises the production of N2O in the mesosphere, its downward migration, and conversion in the stratosphere to NO by the reaction N2O + O(D) yields 2NO. This process would amplify direct NO production by more than 10%.

The prominent emission feature near 2145 A in the ultraviolet spectrum of an aurora has been tentatively identified by Dick (1978) as the doublet lines, 2139.68 A and 2143.55 A, emitted by metastable N+(5S) ions, and dissociative excitation of N2 by electron impact with a cross section greater than or equal to 2 x 10 to the -18th sq cm has been proposed as the source of this species. A detailed laboratory study of dissociative excitation is described that suggests two alternative viewpoints of this process: (1) If the calculated radiative lifetime for the N+(5S) state (4.4 microsec) is correct, then the N+(5S) dissociative excitation cross section is less than 3 x 10 to the -21st sq cm. Thus, a new N+(5S) source mechanism would have to be found in order to account for the auroral data. (2) If dissociative excitation does form N+(5S) ions efficiently, then the laboratory and field observations imply a radiative lifetime for this state of more than 10 msec, thus suggesting that there are major errors in the lifetime computation.

A study of nitrous oxide formation mechanisms indicates that N2O concentrations greater than 10 to the 9th per cu cm could be produced in IBC III aurora or by lower-level activity lasting for many hours, and, in favorable conditions, the N2O concentration could exceed the local nitric oxide density. An upper limit on the globally averaged N2O production rate from auroral activity is estimated at 2 x 10 to the 27th per second.

The dissociative recombination of vibrationally excited O2(+) ions is studied in light of the possible importance of this reaction in upper atmospheric chemistry. A plasma spectroscopy experiment was performed in a microwave cavity filled by an argon-oxygen mixture, with O(1S) production monitored by measurements of the 5577-A afterglow, the O2(+) density and the electron concentration. Plasma and optical data reveal the predominant afterglow ions to be Ar2(+) and O2(+), with an effective O(1S) dissociative recombination coefficient of 2.1 x 10 to the -8th cu cm/sec, corresponding to a quantum yield of 10%. Experiments with an argon-krypton-oxygen mixture reveal that vibrationally excited O2(+) ions are the chief source of the O(1S) atoms, with a specific recombination coefficient for the dissociation of O2(+)(2 pi g) into O(1S) and O(1D) of 4.2 x 10 to the -9th cu cm/sec. A comparison of the laboratory results with Atmospheric Explorer data on the 5577-A airglow implies that O2(+) ions in the sunlit ionosphere are vibrationally excited to the same degree as in the laboratory, with the vibrational relaxation of these ions much slower than dissociative recombination. Results also predict a dawn-twilight asymmetry in the effective O(1S) yield due to the normal variation of electron content.

Measurements were made of the rotational profiles of specific bands of the CO fourth-positive group (4PG). The CO 4PG bands were excited by electron impact dissociative excitation of CO2. The results are applicable to analysis of the Mariner observations of the CO 4PG in the dayglow of Mars. The results indicate that dissociative excitation of CO2 by electron impact leads to CO(A 1Pi) fragments with a rotational distribution that is highly nonthermal. The parent CO2 temperature was about 300 K in the experiment, while the fragment CO(A 1Pi) showed emission band profiles consistent with a rotational temperature greater than about 1500 K. Laboratory measurement of the reduced transmission of the hot bands by thermal CO appears to be the most direct way of determining the column density responsible for the CO(v',0) absorption of Mars.

Results obtained by rocket-borne optical spectrometry are presented. Composition measurements and auroral studies are reported. The production of N (D-2) atoms by photo-absorption processes, and by electron impact excitation of N2 are discussed along with vibrationally excited CO2(+) ions in planetary atmospheres.

Metastable fragments produced by electron impact excitation of CH4 have been investigated for incident electron energies from threshold to 300 eV. Only metastable hydrogen and carbon atoms were observed. Onset energies for the production of metastable hydrogen atoms were observed at electron impact energies of 22.0 + or - .5 eV, 25.5 + or - .6 eV, 36.7 + or - .6 eV and 66 + or - 3 eV, and at 26.6 + or - .6 eV for the production of metastable carbon atoms. Most of the fragments appear to have been formed in high-lying Rydberg states. The total metastable hydrogen cross section reaches a maximum value of approximately 1 X 10 to the minus 18th power sq cm at 100 eV. At the same energy, the metastable carbon cross section is 2 x 10 to the minus 19th power sq cm.

The spectral line shapes of the radiation emitted from O atoms produced by the dissociative excitation of O2, CO, CO2 and NO are investigated. Doppler line shapes are derived from time-of-flight spectra of O (5S0) atoms decaying by the emission of 1356-A radiation after being produced in electron impact experiments at incident electron energies from 25 to 300 eV. It is shown that the effective line width of the radiation is large compared with the Doppler absorption widths of ambient O atoms in both photoelectron and auroral excitation, and thus the dissociatively excited component of the O I 1304-A airglow will behave as though it were optically thin, exhibiting pronounced limb brightening effects and a scale height characteristic of the initial, local source function. It is found that the average kinetic energy of the dissociation fragments inferred from O I (5S) time-of-flight spectra is in good agreement with that of O I (3S) atoms in the electron impact dissociation of CO2, although not for O2. Finally, it is suggested that although electron impact dissociation of CO and CO2 contributes to the 1304-A emission in the upper atmosphere of Venus, it cannot be the dominant source of this radiation since the absolute cross sections for the reaction are too small.

Results of a study of the electron-impact excitation of the b 1 pi u state of N2, one of the singlet states thought to be excited by precipitating electrons in the auroral zones, and of its predissociation and radiative relaxation through the emission of the Birge-Hopfield band system (b 1 pi u - X 1 Sigma g +) are presented. A collimated electron beam was passed through N2 gas producing a variety of atomic states through dissociative excitation, and the radiation resulting from relaxation of these states was observed by VUV and visible-IR monochromators. Absolute emission cross sections for 11 Birge-Hopfield bands are obtained for energies from threshold to 500 eV, and used to calculate the absolute transition probabilities for BH(1, v-prime) bands and the variation of the electric dipole moment with internuclear distance. With the exception of the v-prime equals 1, 5 and 6 vibrational levels, all b 1 pi u levels are found to predissociate with a specific predissociation branching ratio greater than 0.99, representing a major source of nitrogen atoms.

The concentration and altitude distribution of metastable N(2P) atoms was measured in a diffuse IBC II(+) auroral arc. The dominant N(2P) source is shown to be the dissociative excitation of N2 by electron impact with a minor contribution from the dissociative recombination of N2(+) ions. The possibility that an ion-molecule process involving atomic oxygen and vibrationally excited N2(+) ions is a significant N(2P) source is examined. Values for the proportional yield of N(+), N(2P), N(2D), and N(4S) atoms from electron-impact dissociation of N2 under optically thick conditions are given.

The rate coefficient for the quenching of metastable O(1S) atoms by O2 was measured as a function of temperature from 250 to 550 K. The resulting Arrhenius expression correlates well with previous laboratory work. It is suggested that the much larger value of the rate coefficient inferred from an analysis of artificial auroral experiment, Precede, may be explained by overestimation of the contribution of O(1S) production from O2(+) dissociative recombination. The possibility that O(1S) atoms are produced only by the dissociative recombination of vibrationally excited O2(+) ions is examined; such excited ions would not exist in the Precede experiment because of the rapid cooling of the ions by resonant charge transfer processes.

The absolute cross-sections for the excitation of the 989 A, 1027 A, 7990 A, 8446 A, 1.1287 micron and 1.3164 micron 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 O I(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 lambda 989 A and lambda 1027 A 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 EUV emission spectrum in auroras and in the sunlit F-region at high exospheric temperatures.

The dissociation of N2 by electron impact and by absorption of EUV photons was studied experimentally. It was shown that most of the N2 molecules excited to singlet states in the 12.5-14.86 eV range are depopulated by predissociation and not by the emission of EUV photons and that this is the principal mechanism by which N2 is dissociated by solar EUV absorption and by electron impact. The experiments provide a physical explanation for the near absence of N2 band radiation in airglow and auroral EUV spectra, and rule out the excitation of EUV radiation as a major factor in the overall energy economy of an auroral substorm.

The cross sections for the excitation of Ly-beta and H-alpha when methane is dissociated by electron impact have values of 17.1 by 10 to the -19th power sq cm and 26.0 by 10 to the -19th power sq cm, respectively, at an electron impact energy of 100 eV. These results are in disagreement with the implications of recent polarization measurements of H-alpha radiation that suggest negligible H(3p) excitation in the dissociation of CH4 by electron impact.

Time-of-flight techniques have been used to investigate the electron-impact dissociation of H2, HD, and D2 in order to determine the effect of isotopic mass variation in the target molecule on the dissociative excitation process. At incident electron energies near 100 eV, the time-of-flight spectrum produced from each molecule consists of atoms in the metastable 2s state and in high-lying long-lived Rydberg levels. The individual time-of-flight distributions, kinetic-energy spectra, and relative differential cross sections for these two species resulting from each molecule have been measured. The kinetic-energy spectrum of the Rydberg atoms produced from dissociative excitation of H2 was notably dissimilar in shape from the corresponding distributions produced from HD and D2. Also the 2s and Rydberg production cross sections differed between the three molecules. In the dissociation of the heteronuclear HD molecule, the ratio of fast H(2s) atoms to D(2s) atoms was about 1 to 1, while the same ratio comparing the Rydberg atoms was nearly 2 to 1. These differences indicate the influence of the mass variation on the position of the Franck-Condon region in the production of 2s atoms and on the competition between autoionization and dissociation in the formation of Rydberg fragments.

Absolute cross sections for the excitation of the H2 Werner band system were measured from energy threshold to 300 eV for electron impact on H2. The bands were observed in emission in the wavelength region 1100A to 1250A. The measured cross sections were compared with published transition probabilities, leading to the conclusion that the Werner bands are suitable as the basis for a relative spectral response calibration only when the bands are observed under sufficiently high resolution. The effect of the perturbation between the C 1Pi u and B 1 Sigma-u states of the hydrogen molecule was clearly observed in anomalies in the rotational intensity distribution in bands of the (3 v '') progression.

The dissociative excitation of CO (a 3Pi) and other metastable fragments produced by electron impact on CO2 has been investigated from threshold to 50 eV. The observed threshold for CO (a 3Pi) production at 11.9 (plus or minus 0.5) eV was near the minimum required energy of 11.5 eV.

The contribution of the N2(E(3 Sigma g plus)) state to the total metastable excitation function of N2 assessed on the basis of time-of-flight studies of metastable nitrogen molecules. The cross section for electron impact excitation state was determined in the domain of the resonance form threshold (11.87 eV) to an energy of about 13 eV. The maximum value of the cross section was found to be (7.0 + or - 4.0) x 10 to the -18th power sq cm at an energy of 12.2 eV. The measurement was made absolute by using the previously determined yield of the metastable detector, the lifetime of the E state, and by eliminating the energy spread in the electron beam from the raw data. The half-width (FWHM) of the resonance-like excitation function near threshold was found to be about 0.4 eV. No substantial evidence was obtained from the present data for the presence of the nonresonant part of the excitation function for the state studied.

Vacuum ultraviolet multiplets of C I, C II, and O I were produced by electron impact of CO2. Absolute emission cross sections for these multiplets were measured from threshold to 350 eV. The electrostatically focussed electron gun used in this series of experiments is described in detail. The atomic multiplets which were produced by dissociative excitation of CO2 and the cross sections at 100 eV are given. The dependence of the excitation functions on electron energy shows that these multiplets are produced by electric-dipole-allowed transitions in CO2.

Recent rocket observations of the molecular nitrogen Vegard-Kaplan system in the aurora have been reinterpreted using an atmospheric model based on mass spectrometer measurements in an aurora of similar intensity at the same time of year. It is found that the population rates of the considered levels in the aurora are accurately determined by radiative cascade from two other states excited by direct electron impact. In some bright auroras the role of NO in quenching the lower vibrational levels of the A state is significant. The conclusions are based on a number of relevant auroral observations in combination with calculations using electron cross section and transition probability measurements by Shemansky and Broadfoot (1971).

Estimation of the aeronomically important cross section for the production of O(super 5 S super 0) by electron impact dissociation of O2 on the basis of recently published work of several groups. The cross section reaches a maximum value of 7 x 10 to the minus 18th sq cm plus or minus 50% at about 100 eV. The assumptions made in arriving at this cross section suggest several experiments which might be performed to further refine this estimate.