PHYSICAL REVIEW LETTERS

VOLUME 29 4 SEPTEMBER 1972 NUMBER 10

Superelastic Collisions of Vibrationally Excited H2+ with Atoms and Molecules*

F. A. Herrero and J. P. Doering Department of Chemistry, The Johns Hopkins University, Baltimore, Maryland 21218

(Received 20 July 1972)

We report the direct experimental observation of superelastic collisions of vibrational­ly excited H2

+ ions with atomic and molecular targets. Energy-change spectra taken with H2

+ incident on He, Ne, Ar, Kr, and H2 targets show a strong dependence of the cross section on the atomic number of the target. The cross sections for superelastic colli­sions were observed throughout the kinetic-energy range 100-1500 eV and they appear to be largest in the region 100-500 eV, decreasing slowly at higher energies.

In previous papers,1 '2 we have reported the cross sections for vibrational excitation process­es in ion-molecule collisions. These studies were principally concerned with proton excitation of vibrational states within the ground electronic state of a molecular target such as H2. Vibration­al excitation was detected by observation of peaks in the energy-loss spectrum at energies corre­sponding to the vibrational energy levels of the target molecule.

Recently, we have undertaken similar inelastic scattering experiments using H2

+ projectiles in­stead of protons. We have found that the "energy-loss" spectra produced by collisions of H2

+ ions with various atomic and molecular target ions contain three distinct components. The first two components are inelastic processes correspond­ing to excitation of the vibrational energy levels within the ground electronic state of the H2

+ pro­jectiles and, in the case of molecular targets, excitation of the target ground-state vibrational energy levels. The third component of these spectra consists of a very intense superelastic "energy-gain" spectrum corresponding to colli-sional conversion of quanta of vibrational energy in initially excited H2

+ projectile ions into transla­tion. It appears that the duoplasmatron ion source used in our experiments produces a beam of H2

+

ions with a high degree of vibrational excitation

--up to at least the v' = 4 vibrational state of the X 2 S g

+ ground electronic state of H2+—and that the

cross section for vibration-translation conversion is so large that an intense superelastic spectrum is produced. This Letter reports measurements of the cross sections for vibration-translation conversion for H2

+ ions incident on a variety of atomic and molecular targets.

Figure 1 shows an "energy-change" spectrum produced by collision of our vibrationally excited H2

+ beam with Kr atoms. The abscissa indicates the energy change observed by the scattered-ion analyzer. The apparatus used has been described previously.2 All experiments reported in this Letter were done at a scattering angle of 0° (angu­lar resolution, ±1.9°). Energy resolution was typically 100 meV (full width at half-maximum). The intense peak at zero energy change corre­sponds to the transmitted, unscattered H2

+ ion beam. Positive values of energy change corre­spond to superelastic collisions with an increase in translational energy, while the usual inelastic spectrum occurs at negative values of energy change. The spacing of the observed peaks in both the superelastic and inelastic energy regions corresponds closely to the known vibrational state spacing in the ground electronic state of H2

+.3 We conclude, therefore, that the peaks at positive values of energy change arise from superelastic

609

VOLUME 29, NUMBER 10 PHYSICAL REVIEW LETTERS 4 SEPTEMBER 1972

-i.o -0.5 0 0.5 ENERGY CHANGE ( e V )

FIG. 1. Energy-change spectrum produced by 500-eV H2

+ on Kr. The collision chamber pressure was 0.7 mTorr. The transmitted unscattered primary beam appears at 0 eV. The positions in energy of the various vibrational transitions within the H2

+ ground electronic state (Ref. 3) are shown above the peaks.

collisions in which one or more quanta of vibra­tional energy in the initially excited H2

+ ions is converted into translation, and the peaks at neg­ative energy-change values correspond to colli-sionally excited vibrational transitions between the vibrational states of the H2

+projectile ions. Experiments in which the hydrogen in the ion

source was replaced by deuterium showed the ex­pected result. The peaks in the superelastic and inelastic energy-change spectrum contracted to a spacing corresponding to the vibrational energy-level spacing in D2

+. Because of the small energy spacing between the D2

+ vibrational energy levels, we were unable to resolve completely the D2

+ en­ergy-change spectrum and the inelastic process­es appeared as shoulders on the transmitted beam peaks. However, these results confirm the a s ­signment of the observed peaks in the H2

+ energy-change spectrum to vibrationally inelastic and superelastic collisions.

Since the distribution among vibrational states of the excited H2

+ ions produced in the ion source is not known, we must assume that each of the in­elastic and superelastic peaks in the energy-change spectrum contains contributions from a number of different vibrational transitions. For

X30 X300 (a)

JL—1 I I I I t • •* I »

-0.5 0 0.5 1.0 ENERGY CHANGE { e V )

FIG. 2. (a) Energy-change spectrum produced by 600-eV H2

+ on H2. The collision chamber pressure was 1.0 mTorr. (b) Energy-change spectrum produced by 300-eV H on H2 with collision chamber pressure 1 mTorr. Note absence of superelastic features.

example, the first peak in the superelastic spec­trum in Fig. 1 at an energy gain of 0.25 eV must correspond to a sum of contributions of the type v> +1 - v \ The range of initial v' +1 states which contribute measurably to the observed intensity is determined by the extent of vibrational excita­tion produced in the ion source. Similarly, the second and third peaks in the superelastic energy-change spectrum correspond to the transitions At;' = - 2 and Avf = - 3. The known vibrational spacings corresponding to these transitions are also shown in Fig. I.3 Comparison of these spac­ings with the observed peak envelopes suggests that most of the observed intensity arises from the states t>'=l, 2, 3, and 4. If we assume that the efficiency for the inelastic or superelastic process is independent of the initial vibrational state, but depends solely on the magnitude of the change in vibrational quantum number At;', it follows that the population of excited vibrational states in the H2

+ projectile beam is highest around v' = 2. This result is consistent with the known Frank-Condon factors4"*6 for ionization of EL

610

VOLUME29,NUMBER 10 P H Y S I C A L R E V I E W L E T T E R S 4SEPTEMBER 1972

which predict that for the transition H2(X1Sg+)

-H2+(X2Sg

+) +£, the maximum population of H2 +

ions is in the v' = 2 state. The influence of the nature of the target atom

or molecule is also of interest. Figure 2(a) shows a spectrum taken with an H2 target. The phenomena are identical to those observed in the H2

++Kr case with two exceptions. First, the re l ­ative cross section for superelastic scattering is approximately 4 that for Kr, and so the energy-change spectrum is somewhat weaker in this case. Second, the inelastic spectrum shows the expect­ed strong inelastic peak corresponding to the v' = 0 - v' = 1 transition in the H2 target. Figure 2(b) shows a spectrum taken with H+ projectile ions of nearly the same velocity. As expected, this spectrum shows only the inelastic process cor­responding to the v' = 0 - v' = 1 transition of H2.

As we have previously shown,2 the apparent cross sections for the Avf = ± 1, ± 2, ± 3 transitions can be obtained directly from the spectrum. For the H2

+ +Kr superelastic system at 500 eV H2 +

kinetic energy, we obtain the following: o(Av' = - l ) = 1.08 A2, a(At;' = - 2 ) = 0.22 A2, o(&v' = -3) = 0.047 A2. Similarly, for the H2

++H2 superelas­tic system at 600 eV kinetic energy, we have the following: o(Avf = - 1) = 0.23 A2, o(&v' = - 2) = 0.081 A2, and o(Avf = - 3) = 0.025 A2. Similar studies with He, Ne, and Ar target atoms clearly show a

Pioneering experiments1 have established con­clusively that the radiative decay 2S to IS in He+

proceeds via spontaneous two-photon emission. This is in strict accord with theory2 which for­bids single-photon electric dipole (El) radiation by parity conservation, and predicts a single-photon magnetic dipole (Ml) rate » 5x 10"6 times smaller than that for emission of two E l photons, Recently, Marrus and Schmieder have reported3

measurements of the 2S lifetime, T 2 S , in hydro­genlike Ar+17 and S+15 by a beam-foil time-of-

strong increase of the cross sections with in­creasing atomic number of the target atom. The relative cross sections for superelastic scatter­ing decrease slowly with increasing kinetic ener­gy, being largest in the 100-500-eV region. It is interesting to note that superelastic collision cross sections for the conversion of vibrational to translational energy are comparable in magni­tude and energy dependence to the cross sections for the inverse inelastic process of direct vibra­tional excitation.

Further measurement in the H2+ projectile en­

ergy range 100-1500 eV with He, Ne, Ar, Kr, Xe, and H2 targets are in progress and complete results will be reported later.

*Work supported by a grant from the National Science Foundation.

^J, H. Moore, J r . , and J. P. Doering, Phys. Rev. Lett. 23, 564 (1969).

2 F. A. Herrero and J. P. Doering, Phys. Rev. A j>, 702 (1972).

3T. E. Sharp, Lockheed Palo Alto Research Laborato­ry Report No. LMSC5-10-59-9 (unpublished).

4G. H. Dunn, J . Chero. Phys. 44, 2592 (1966). 5J. Berkowitz, H. Ehrhardt, and T. Tekaat, Z. Phys.

200, 69 (1967). 6R. Spohr and E. von Puttkamer, Z. Naturforsch. 22a,

705 (1967).

flight technique. However, except for the estab­lishment of lower limits in H4 and He + ,5 an ex­perimental value for r2S in a low-Z ion has not been reported.

This Letter describes work carried out to de­termine r2S in He+. The results are in agree­ment with the theoretical value6 of 1.899 msec and lead to a new upper limit on the amount of parity impurity in the 2S-state wave function in H e \

The 2S state lies 40.8 above the He+ IS ground

Lifetime of the 2S State of He*f

M. H. P r i o r Lawrence Berkeley Laboratory, Berkeley, California 94720

(Received 26 June 1972)

The lifetime of the 2S state of He+ has been measured by counting decay photons ver ­sus time from an ensemble of excited He+ ions stored in an electromagnetic trap of the Penning type. The result is T2 S = 1.922(82) msec. This is in agreement with the theoret­ical value of 1.899 msec. The agreement between theory and experiment implies a new upper limit on the amount of parity impurity in the ZS wave function.

611