The observation of predissociations in the oxygen molecular ion by low-energy electron impact

Abstract

The kinetic energy distribution of the O + ions formed by dissociative ionization of O2 has been carefully ex-amined in the range of 0.0 to 1.3 eV. In the peak at 600 meV fine structure is observed and interpreted by predissociation. The high energy side of the peak has been ascribed to the predissociation of the 2 ∆ g , B 2 Σ g -and the c 4 Σ u -states of O 2 + . The low energy side of the peak is presumably due to the predissociation of the b 4 Σ g -state or the Φ u state of O 2 + or to autoionizing predissociation.During the last few years much attention has been paid to the dissociation of diatomic molecules. One of the dissociation channels a molecule can choose is the predissociation observed in several neutral and ionized species. Besides the optical spectroscopy methods, predissociations have been investigated by a variety of techniques such as "metastable ion" mass spectrometry [1,2], high resolution translational spectroscopy [3,4] as well as photoelectron spectroscopy [5,6].The dissociation of the oxygen molecular ion has been the subject of several papers. Doolittle et al. [7] obtained the kinetic energy distribution curve of O + from O 2 by using photoionization mass spectrometry. Freund [8] and Kieffer et al. [9] studied the kinetic energy carried away by the 0 ions formed by electron impact on O 2 . These authors ascribed the first broad and structureless peak, with a maximum at 850 meV, to O + fragment ions formed by predissociation, although no direct experimental evidence was found for this mechanism. Recently Danby and Eland [5] showed experimentally the existence of a predissociation of O 2 by using the photoelectronphotoion coincidence technique.The purpose of this paper is to report our results obtained for the kinetic energy analysis of the O + ions formed by electron impact on molecular oxygen;The instrument used, as well as the experimental conditions will be described elsewhere [10]. Briefly, the ions formed in a conventional Nier-type ion source pass through a retarding potential system before entering into a quadrupole mass spectrometer. The ion current is, at fixed electron energy, continuously scanned with respect to the retarding potential or with respect to electron energy at fixed retarding potential. The first derivative of either the retarding curves or the ionization efficiency curves is stored in a multichannel analyser and plotted on an X-Y recorder.A preliminary report [11] has been devoted to the kinetic energy distribution curve of O + from O 2 , spread over a broad energy range (0 to 12 eV). These results will be published elsewhere [10]. In this letter we will limit our discussion to the results obtained in the energy range of 0 to 1.3 eV.An example of a kinetic energy distribution of O + from O 2 , as observed by the impact of 25 eV electrons, is shown in fig. 1.The first peak maximum is measured at 0.0 eV coinciding with the maximum of the thermal distribution of the molecular ion O 2 , taken as the origin of the kinetic energy scale. The second maximum, meas...

Dissociative ionization ofH2,N2, andZyl1, Stephen2 1994Phys. Rev. A

“…[15] presents N+-energy-distribution data for electron energies of 26 been seen earlier by Kollmann [11],although with much reduced resolution. ) For 02 molecules, a similar situation occurs, as shown by the work of Locht and Schopman [16] at 60and 35-eV electron energy, and at 25 eV by Schopman and Locht [31]. While the above data of Locht and colleagues [15,16,31] form the basis for our models of the ion-energy distributions at low ion energies, Locht et at.…”

“…For 02 molecules, a similar situation occurs, as shown by the work of Locht and Schopman [16] at 60and 35-eV electron energy, and at 25 eV by Schopman and Locht [31]. While the above data of Locht and colleagues [15,16,31] form the basis for our models of the ion-energy distributions at low ion energies, Locht et at. [15] note that the "closed structure of the ion source" is "responsible for a discrimination against highly energetic ions.…”

“…5 have been shifted to lower 0+ energies by 0.15 eV. Such shifts, generally between 0.05and 0.15-eV ion energy, usually resulted in substantially better agreement with the data of Locht and colleagues [15,16,31] (and are within the +0.2-eV energy-scale uncertainty cited by Van Brunt and co-workers [28,29]). Furthermore, the influence of the energy-dependent resolution of Van Brunt's ion detector (wherein b, E; =0.…”

“…The solid-line curve is the Locht and Schopman [16] result, and the dashed-line curve shows our estimate of how much this result must be increased (for ion energies above a few eV) to bring it into agreement with the data points of Van Brunt et al [29]. The difference between these two curves can be explained by assuming the apparatus used by Locht and his colleagues [15,16,31] Unfortunately, this comparison of these 0 data is not as straightforward as the above discussion would suggest, because some of the data points of Van Brunt et al [29] plotted in Fig. 5 have also been adjusted.…”

“…5 have also been adjusted. Their apparatus, in contrast to that used by Locht and colleagues [15,16,31], discriminated against low-energy ions (the reason why such H+ energy distributions as shown in Fig. 4, for example, were not extended to ion energies below about 3 eV}.…”