Crossed molecular beam experiments and accurate quantum scattering calculations have been carried out for the polyatomic H + CD 4  → HD + CD 3 reaction. Unprecedented agreement has been achieved between theory and experiments on the energy dependence of the integral cross section in a wide collision energy region that first rises and then falls considerably as the collision energy increases far over the reaction barrier for this simple hydrogen abstraction reaction. Detailed theoretical analysis shows that at collision energies far above the barrier the incoming H-atom moves so quickly that the heavier D-atom on CD 4 cannot concertedly follow it to form the HD product, resulting in the decline of reactivity with the increase of collision energy. We propose that this is also the very mechanism, operating in many abstraction reactions, which causes the differential cross section in the backward direction to decrease substantially or even vanish at collision energies far above the barrier height.

Theoretical studies of the potential energy surface and bound states were performed for the N(2)O dimer. A four-dimensional intermolecular potential energy surface (PES) was constructed at the CCSD(T) level with aug-cc-pVTZ basis set supplemented with bond functions. Three co-planar local minima were found on this surface. They correspond to a nonpolar isomer with slipped-antiparallel planar structure and two equivalent polar isomers with slipped-parallel planar structures. The nonpolar isomer is energetically more stable than the polar ones by 162 cm(-1). To assign the fundamental vibrational frequencies for both isomers, more than 150 vibrational bound states were calculated based on this PES. The orientation of the nodal surface of the wave functions plays an important role in the assignment of disrotation and conrotation vibrational modes. The calculated vibrational frequencies are in good agreement with the available experimental data. We have also found a quantum tunneling effect between the two equivalent polar structures in the higher vibrational excited states. Rotational transition frequencies of the polar structure were also calculated. The accuracy of the PES is validated by the good agreement between theoretical and experimental results for the transition frequencies and spectroscopic parameters.

Full-dimensional quantum calculations of the vibrational states of H5(+) have been performed on the accurate potential energy surface developed by Xie et al. [J. Chem. Phys. 122, 224307 (2005)]. The zero point energies of H5(+), H4D(+), D4H(+), and D5(+) and their ground-state geometries are presented and compared with earlier theoretical results. The first 10 low-lying excited states of H5(+) are assigned to the fundamental, overtone, and combination of the H2-H3(+) stretch, the shared proton hopping and the out-of-plane torsion. The ground-state torsional tunneling splitting, the fundamental of the photon hopping mode and the first overtone of the torsion mode are 87.3 cm(-1), 354.4 cm(-1), and 444.0 cm(-1), respectively. All of these values agree well with the diffusion Monte Carlo and multi-configuration time-dependent Hartree results where available.

Femtosecond stimulated Raman spectroscopy (FSRS), using an overlapping pair of narrow band Raman pump and broadband probe pulses with heterodyne detection along the probe pulse direction, is a new nonlinear spectroscopic technique to record vibrational spectra of even highly fluorescent molecules and to study vibrational dynamics on excited electronic states of molecules, as in photoisomerization. FSRS is described by diagrammatic third-order perturbation theory with wave packet analysis. The phase matching condition gives rise to forty-eight terms for FSRS, but the resonant condition reduces it to just eight terms, which can be depicted by Feynman dual time-line diagrams, or closed time path loop diagrams, or the complementary four-wave mixing energy level diagrams. The eight terms fall into four sets-SRS(I), SRS(II), IRS(I), IRS(II)-where SRS stands for stimulated Raman scattering and IRS stands for inverse Raman scattering. The SRS(I) set can also account for spontaneous Raman scattering, but the remaining SRS(II), IRS(I) and IRS(II) terms are only present in stimulated scattering with the presence of a probe field. The SRS(I) set accounts for the Stokes Raman lines while the IRS(I) term accounts for the anti-Stokes lines, relative to the Raman pump frequency, in the FSRS spectrum. The remaining SRS(II) and IRS(II) terms give rise to broad baselines. Using a harmonic oscillator model, analytic results are obtained for the four-time correlation functions in the third-order polarizations. The issue of high time and high frequency resolution in time-resolved FSRS spectra is discussed. Calculations are made with the theory to compare with experimental results for: (a) resonance FSRS of fluorescent Rhodamine 6G and (b) 2D-FSRS from a coherent vibrational state that has been prepared by an impulsive, off-resonant pump pulse on CDCl 3 . The calculated results compared well with experimental results, and in the case of 2D-FSRS on CDCl 3 there is a dominant cascade effect contributing to the FSRS spectra.femtosecond stimulated Raman spectroscopy, temporal and spectral resolution, cascade effect

The initial state selected time-dependent wave packet method was employed to calculate the integral cross sections for the H(2) + D(2) reaction with and without the centrifugal sudden (CS) approximation by including all important K (the projection of the total angular momentum on the body-fixed axis) blocks. With a full-dimensional model, the first fully converged coupled-channel (CC) cross sections for different competitive processes from the ground rotational state were obtained: collision induced dissociation (CID), four-center (4C) reaction and single exchange (SE) reaction. The effect of the total angular momentum J on the reaction dynamics of H(2) + D(2) and the accuracy of the CS approximation have also been studied. It was found that the CID and SE processes occur in a wide range of J values while the 4C process can only take place in a narrow window of J values. For this reason, the CC cross section for the 4C channel is merely comparable to the SE channel. A comparison of the integral cross sections from CC and CS calculations showed that the CS approximation works well for the CID process but not for the 4C and SE processes, and the discrepancy between the CC and CS cross sections grows larger as the translational energy and/or the vibrational energy increase(s).