Technical Report 34 K1hetic Energy Distribution of Negative Ions Formed by Dissociative Attachment and the Asurem1ht of the Electbon Affinity of Oxygen

The kinetic energy distribution of ions produced by a dissociative ionization process is derived, taking into account the effect of thermal uotion of the target molecule. In the case of dissociative attachment of monoenergetic electrons to a diatomic molecule the width at half maximum of the negative ion energy distribution is 1/2 given by (ll^kTE ) ' where ß is the ratio of the mass of the ion to that of the parent molecule, T is the target gas temperature, and E is the most probable ion energy. Using a crossed field velocity filter 0 ion energy distributions arising from tne attachment of essentially monoenergetic electrons to 0., are studied as a function of electron energy at two gas temperatures. The measured widths of the distributions are consistent with the above relationship. Measurements of E o as a function of the electron energy allow a determination of the electron affinity, A, of atomic oxygen. The result, A = 1.5+0.1 eV, is in excellent agreement with photodetachment threshold determinations. * This research was supported in part by the Advanced Research Projects Agency through the U. S. Office of Naval Research. ** Present address: Yale University, New Hiven, Connecticut, "9««w« i^l.JJ "l I FLi■''T Electron beam experiments have been used repeatedly for a study of negative ion forsntion resulting from dissociative attachment. In particular, measurements of the magnitude of the cross section and the kinetic energy of the fragment ions are of interest. The electron I energy dependence of the cross-section is of value in determining the potential energy curve of Jie molecular negative ion compound state ■ along which dissociation occurs. The position of this curve at I infinite internuclear separation can be determined from the electron energy dependence of the fragment ion kinetic energy. If the disso1 ciation energy of the neutral molecule is known, this provides a determination of the electron affinity involved. The present paper is I concerned primarily with this second aspect of the problem. 1 1. In the study of dissociative ionization processes the position of •* the relevant potential energy curve in the Franck-Condon region is often determined by the reflection method. This consists of drawIing the curve so that the distribution in kinetic energy of the fragments is given by the reflection of the square of Che ground state vibrational wave-function in the potential energy curve onto I the energy axis. (See for example H. D. Hagstrum and J. T. Täte, Phys, Rev. 59354 (1941)). Because of the resonant nature of dissociative attachment this pre ^dure may be carried through without a knowledge of the kinetic energy of the fragments, requiring instead that the reflection method reproduce the electron «» energy dependence of the cross-section on the energy axis. The potential energy curve so derived is however likely to be seriously ^ in error, since the method implicitly assumes that the survival probability against autodetachment of the compound state to complete dissociation is independent of initial internuclear separa„, tion, i.e. electron energy. This is unlikely to be the case, and the method gives only a first approximation to the compound state potential energy curve. Determination of its true position must . also involve a determination of the probability of autodetachment as a function of internuclear separation. For a detailed applicaJ tion of these considerations see T, F. O'Malley, (submitted to Phys. Rev. Letters). I In some cases the electron affinity has been determined by other methods, such as photodetachment, and from such comparisons it has become apparent that a serious discrepancy exists between the electron affinity of atomic oxygen as determined from photodetachment experiments and the value from electron beam experiments. An attempt to resolve this discrepancy by improving the procedure used in electron beam experiments, (more reliable electron and ion energy scale calibration and improved ion collection efficiency) has not lessened the 2 discrepancy. Whereas the value of the electron affinity of atomic oxygen obtained from photodetachment experiments was 1.465 eV, the 11 values from previous electron beam experiments centered about 2.0 eV. 4 We have recently pointed out that this discrepancy in the values of the electron affinity resulted from an Incorrect interpretation of ion retarding curves which are often used to determine the ion kinetic energy in electron beam experiments. In this pap?' we present a more detailed treatment of the theory involved in the interpretation of such experiments. The experimental work reported In this paper was undertaken in order to demonstrate certain features of the theory, and also to develop techniques for the proper determination of 2. For a recent review, and references regarding this problem, see G. J. Schulz, Phys. Rev. 128, 178 (1962). 3. L. M. Branscomb, D, S. Burch, S. J, Smith, and S. Geltman, Phys. Rev. Ill, 504 (1958); for a review see L. M. Branscomb, Chap. 4, Atomic and Molecular Processes, edited by D. R. Bates (Academic Press, New York, 1962). 4. P. J. Chantr> and G. J. Schulz, Phys. Rev. Leiters 12, 449 (1964). I II II II li I II I J I I 1 1 1