The photodissociation dynamics of 1 state ammonia molecules (both NH3 and ND 3 ) has been further investigated using the technique ofH(D) atom photofragment translational spectroscopy. The resulting NH2 (ND 2 ) fragments are observed to carry high levels of internal excitation, the precise disposition of which is sensitively dependent upon the parent v~ level excited. Dissociation from the v~ = 0 level of the 1 state yields ground state NH2 (ND 2 ) fragments, primarily in their zero-point level, but with high levels of rotational excitation specifically concentrated about the a-inertial axis; the population distribution over the energetically accessible product rotational levels with N <::::.Ka appears near to statistical. In contrast, dissociation from the parent v~ = 1 level yields a markedly inverted fragment internal energy distribution. These different energy disposals have been rationalized via classical traje~tory calculations employing the best available ab initio potential energy surfaces for the 1 and X states of the ammonia molecule. The energy disposal following excitation to the parent v~ = 2 and 3 levels is found to mimic that observed for, respectively, the v~ = 0 and 1 levels.These results provide clear evidence for the importance of anharmonic coupling (whereby an even number of bending quanta are redistributed into stretching motions) in promoting the fra~entation_o!parent levels ~ith v; ;;'2. The threshold energyJor producing electronically excIted NH2 (A AI) fragments IS 6.02 e V [ -6.16 e V for ND2 (A) ]. The present studies of NH3 photolysis suggest that this fragmentation channel opens at threshold and clearly indicate that branching into this channel occurs with much higher quantum yield than hitherto believed.

A new and improved version of the technique of H atom photofragment translational spectroscopy has been applied to a study ofH 2 S photodissociation at 121.6 nm. The primary fragmentation pathways leading to H + SH(A) fragments and H + H + SeD) atoms are observed to dominate the product yield; the yield of H atoms formed in conjunction with ground state SH (X) fragments is undetectably small. The majority of the SH (A) fragments are formed in their v = 0 level with a rotational state population distribution that spans all possible bound and quasibound rotational levels. The experimental determination of the energies of these hitherto unobserved high rotational states has enabled a refinement of the SH(A) potential energy function, an improved estimate ofthe SH(A) well depth (9280 ± 600 cm-I ), and thus of the SH(X) ground state bond dissociation energy Dg (S-H) = 3.71 ± 0.07 eV. All aspects ofthe observed energy disposal in the title photodissociation process may be understood, qualitatively, if it is assumed that (i) the primary fragmentations occur on the B IAI potential energy surface and (ii) Flouquet's ab initio calculations of portions of this surface [Chern. Phys. 13, 257 (1976)] correctly predict its gross topological features.

We have carried out a systematic crossed molecular beam study of the hydrogen exchange reaction in the H+D2→HD+D isotopic form at two collision energies: 0.53 and 1.28 eV. The Rydberg atom time-of-flight method was used to measure the D-atom product angle-velocity distribution. For the first time ro-vibrational quantum state resolved differential cross sections for the title reaction were measured, which can directly be compared to theoretical predictions at this detailed level. Experimental results are compared to theoretical predictions from both quasi classical and quantum mechanical calculations on different potential energy surfaces as well as to earlier experiments. A general good agreement is found for the converged quantum mechanical calculations with indications that the Boothroyd-Keogh-Martin-Peterson potential energy surface is better suited to describe the dynamics of the reaction. For the higher collision energy the quasi classical trajectory calculations reproduce the experimental data quite well, whereas they fail to describe the situation at the lower collision energy especially with respect to angular resolved differential cross sections.

The photofragmentation dynamics of ammonia molecules following pulsed laser excitation to the two lowest levels (vi = 0 and 1) of their A tA ; excited state has been investigated by monitoring the time-of-flight spectra ofthe nascent H-atom products. These spectra show well resolved structure. Analysis of this structure confirms recent revised estimates of the quantity DJ (H-NH2) (4.645 ± 0.01 eV) and reveals that the majority ofthe accompanying NH2 (X 2 B t ) fragments are formed vibrationally unexcited, but with high levels of rotational excitation specifically concentrated about the a-inertial axis. The detailed energy disposal is sensitive to the initially excited parent vibronic (and even rovibronic) level: the NH2 (X) fragments resulting from photodissociation via the vi = 1 level of NH3 (A) carry a higher level of excitation of the N = Ka rotational levels, which show an inverted population distribution.We also describe the results of trajectory calculations employing the recently reported [M. I. McC~rthy et al., J. Chem. Phys. 86, 6693 (1987)] ab initio potential energy surfaces for theA and X states of ammonia. These provide a detailed rationale for the experimentally observed energy disposal and highlight the massive influence on the eventual fragmentation dynamics of the conical intersection between these surfaces along the H-NH2 dissociation coordinate.

The H + H 2 exchange reaction constitutes an excellent benchmark with which to test dynamical theories against experiments. The H + D 2 (vibrational quantum number v = 0, rotational quantum number j = 0) reaction has been studied in crossed molecular beams at a collision energy of 1.28 electron volts, with the use of the technique of Rydberg atom time-of-flight spectroscopy. The experimental resolution achieved permits the determination of fully rovibrational state-resolved differential cross sections. The high-resolution data allow a detailed assessment of the applicability and quality of quasi-classical trajectory (QCT) and quantum mechanical (QM) calculations. The experimental results are in excellent agreement with the QM results and in slightly worse agreement with the QCT results. This theoretical reproduction of the experimental data was achieved without explicit consideration of geometric phase effects.

The technique of H(D) atom photofragment translational spectroscopy has been applied to the photodissociation of CH 4 (CD 4 ) at 121.6 nm. Contrary to the previous consensus view, we find simple C-H bond fission to be the dominant primary process following excitation at this wavelength. The resulting CH 3 fragments are formed with very high levels of internal excitation: Some (~25%) possess so much internal energy that they must undergo subsequent unimolecular decay. The present experiments do not provide a unique determination of the products of this secondary decay process, but statistical arguments presented herein suggest that they will be predominantly CH and H2 fragments. Similar considerations point to a significant role for the direct three body process yielding the same products H + H2 + CH. This overall pattern of energy disposal can be rationalized by assuming that most of the initially prepared CH 4 (A IT 2 ) molecules undergo rapid internal conversion (promoted by the Jahn-Teller distortion of this excited state) to high vibrational levels of the ground state prior to fragmentation. The realization that CH 4 photodissociation at 121.6 nm yields CH 3 (and CH) fragments, rather than methylene radicals, will necessitate some revision ofcurrent models of the hydrocarbon photochemistry prevailing in the atmospheres of the outer planets and some of their moons, notably Titan.2054

The hydrogen exchange reaction H+D2(v=0,j=0)→HD(v′,j′)+D was investigated at collision energies between 1.27 and 1.30 eV in a high resolution crossed beam experiment. The angle-velocity distribution of nascent D-atoms was measured using the technique of Rydberg atom time-of-flight spectroscopy. The resolution of this technique allows the identification of individual ro-vibrational states of the associated HD product molecule. Calculations done on the Liu–Siegbahn–Truhlar–Horowitz (LSTH) potential energy surface (PES) explicitly including the Geometric Phase effect predict a resonance in reactive scattering for collision energies close to 1.29 eV. The experimental data do not show signatures of this resonance in the energy range investigated. Instead of this a general good agreement between experiment and theory even on the basis of state-to-state differential cross sections is already found for calculations on the LSTH PES at a collision energy of 1.30 eV not including the Geometric Phase indicating that this effect does not play an important role at these collision energies.

The HϩD 2 (vϭ0,jϭ0)→HD(vЈ, jЈ)ϩD isotopic variant of the hydrogen atom exchange reaction has been studied in a crossed molecular beam experiment at a collision energy of 2.20 eV. Kinetic energy spectra of the nascent D atoms were obtained by using the Rydberg atom time-of-flight technique. The extensive set of spectra collected has permitted the derivation of rovibrationally state-resolved differential cross sections in the center-of-mass frame for most of the internal states of the HD product molecules, allowing a direct comparison with theoretical predictions. Accurate 3D quantum mechanical calculations have been carried out on the refined version of the latest Boothroyd-Keogh-Martin-Peterson potential energy surface, yielding an excellent agreement with the experimentally determined differential cross sections. The comparison of the results from quasi-classical trajectory calculations on the same potential surface reveals some discrepancies with the measured data, but shows a good global accordance. The theoretical calculations demonstrate that, at this energy, reactive encounters are predominantly noncollinear and that collinear collisions lead mostly to nonreactive recrossing. The experimental results are satisfactorily accounted for by theoretical calculations without consideration of Geometric Phase effects.