Volume 215, Issue 2 - 2

The unimolecular gas phase chemistry of the title ion,(CH 3 O) 2 P(H)=O • + , ( 1 • + ) and its tautomer dimethyl phosphite, (CH 3 O) 2 P-OH • + , ( 2 • + ) was investigated using mass spectrometry based experiments in conjunction with computational quantum chemistry. A facile tautomerization of the "keto" ion 1 • + into its "enol" isomer 2 • + is prevented by a high 1,2-H shift barrier. Instead, 1 • + readily isomerizes via a 1,4-H shift to the very stable distonic ion CH 2 O-(CH 3 O)P(H)OH • + ( 1a • + ) and related ion-dipole complexes which serve as precursors for the low energy loss of CH 2 =O. Loss of CH 2 =O is also the major dissociation of the enol ion 2 • + , which is more stable than 1 • + by 31 kcal/mol. The reaction involves a 1,3-H shift leading to 1a • + and, at a marginally higher energy, a competing 1,4-H shift leading to the ion-dipole complex CH 2 O-P(OH)-O(H)CH 3 • + ( 1b • + ). The resulting product ions, viz (CH 3 O)P(H)OH • + and P(OH)-O(H)CH 3 • + , are separated by a high 1,2-H shift barrier (44 kcal/mol). However, the CH 2 O moiety in 1a • + and 1b • + is calculated to reduce this barrier significantly by a mechanism coined as proton-transport catalysis. The identity of the ions was probed by tandem mass spectrometry methods. These include MI (metastable ion) or CID (collision induced dissociation) spectra, consecutive MI/CID and CID/CID spectra, NRMS (neutralization-reionization mass spectra), NR/CID and CIDI (collision induced dissociative ionization) spectra, time-resolved CID spectra and deuterium labelling. The energetics of the CH 2 =O loss from 1 • + and 2 • + was derived from ionization and appearance energies determined by VUV photoionization. The experimental results agree quite well with the computational findings. Heats of formation, isomerization barriers and dissociation energies of the various ionic and neutral species were obtained by the wavefunction-based CBS-QB3 method. Essentially identical energy profiles on the C 2 H 7 O 3 P • + surface were obtained with the computationally less demanding novel MPW1K empirical DFT method in conjunction with the aug-cc-pVTZ basis set. Theory and experiment yield a consistent potential energy profile that describes the isomerization and low energy dissociation chemistry of ions 1 • + and 2 • + (Scheme 4). In the μs timeframe ions 1 • + have completely isomerized into distonic ions 1a • + which do not significantly communicate with their more stable enol counterparts. However, ion-molecule reactions of 1 • + with benzonitrile lead to a complete enolization. This is by virtue of a dipole-assisted lowering of the 1,3-H shift barrier separating isomers 1a • + and 2 • + .

Quantum yields of photodissociation were determined for iodine in compressed liquid n -alkanes and supercritical CO 2 and xenon after laser excitation at 532 nm. The quantum yield decreases nearly linearly with increasing density. No cluster effects at low density were observed at this excitation energy. Our results are compared with molecular dynamics simulations and point at the importance of kinematic processes involved in the recombination dynamics of iodine.

For seven difficult diatomic molecules, O 2 , F 2 - , Cl 2 - , SiF, SiCl, ClO and MgO, spectroscopic constants, R e , B e , ω e , ω e x e , and D 0 , have been calculated by two computational methods, MR-ACPF and CCSD(T). The sequence of aug-cc-pVnZ+ basis sets has been used, and for O 2 , subsequent extrapolation of the total energies has been performed. In particular, new bond lengths for F 2 - and Cl 2 - are presented, and the established experimental dissociation energies for SiF and SiCl are corrected. The dissociation energy of the O 2 molecule has been calculated by different methods, overcoming the deficiencies of the aug-cc-pVnZ basis sets, at an accuracy of at least 0.3 kcal mol -1 .

The vibrational bound and resonance state structure of highly excited DCO (X ˜ 2 A´) has been investigated using an effective Hamiltonian based on the polyad approximation. The model Hamiltonian takes advantage of the accidental 1:1:2 degeneracy of the three DCO (X ˜ ) fundamental vibrational frequencies. Allowance was made for vibrational couplings by the 1:1 and 2:2 interaction terms between the DC and CO stretching modes and the 1:2 coupling terms of the two stretching vibrations with the DCO bending mode. A pleasing description of the experimentally observed vibrational term energies [2] was reached up to excitation energies of E v (X ˜ )≈15000 cm -1 , far above the D-CO (X ˜ ) bond dissociation limit. Moreover, a one-to-one state-to-state correspondence was established between the experimental energies, the effective Hamiltonian predictions, and exact quantum dynamics results [18]. The Hamiltonian was subsequently applied to explore the kinetics of the intramolecular vibrational energy redistribution (IVR) in the highly excited DCO (X ˜ ). The vibrational eigenstate compositions were investigated in terms of the respective zero-order basis functions to analyze the extent of the state mixing. 2D and 3D vibrational wavefunctions predicted by the effective Hamiltonian were compared to the wavefunctions from the quantum dynamics studies. IVR rates and IVR pathways were evaluated as a function of the vibrational excitation. Eventually, an attempt was made to model the trends of the observed state-specific DCO (X ˜ ) uni-molecular decay constants by a complex effective Hamiltonian. The results give a rationale for the two orders of magnitude difference between the experimental state-specific decay constants and statistical unimolecular rate theory predictions observed by Stöck et al.  [2].

A geometric approach to the study of multiple-time-scale kinetics is taken here. The approach to equilibrium for kinetic systems is studied via low-dimensional manifolds, with an application to a nonlinear master equation for vibrational relaxation. One of our main concerns is the asymptotic (in time) behavior of the system and whether there is a well-defined rate of approach to equilibrium. One-dimensional slow manifolds provide a good means for studying such behavior in nonlinear systems, because they are the analogue of the eigenvector with least negative eigenvalue for linear kinetics.

Photodissociation of ethyl bromide cation, C 2 H 5 Br + , as well as its isotopomers CH 3 CD 2 Br + and CD 3 CH 2 Br + has been studied at 355 nm by means of time of flight mass spectrometry (TOF-MS) and ion velocity imaging techniques. The TOF mass spectrum of fragment ions from C 2 H 5 Br + +hν 355 nm shows three peaks at m/e of 29 (C 2 H 5 + ), 28 (C 2 H 4 + ) and 27 (C 2 H 3 + ). The TOF spectra for the two isotopomers CH 3 CD 2 Br + and CD 3 CH 2 Br + clearly show that all possible H-loss patterns are operative. The observation indicates that both three- and four-center intermediates are involved for both C 2 H 4 + and C 2 H 3 + channels. Images are recorded for C 2 H 5 + , C 2 H 5 + + C 2 H 4 + and C 2 H 3 + ions, from which translational energy and angular distributions are derived using back-projection algorithm. The C 2 H 5 + + Br channel could be well described by a parallel excitation followed by fast fragmentation with β∼1.7. C 2 H 4 + +HBr channel has a very small translational energy release and a nearly isotropic angular distribution. Most of the energy went into the vibrational degrees of freedom of both C 2 H 4 + and HBr. The anisotropy parameter β∼1.0 is found for C 2 H 3 + , whose translational energy is slightly higher than C 2 H 4 + , but much lower than C 2 H 5 + . From these results, we can rule out the possibility of the secondary dissociation of C 2 H 5 + or C 2 H 4 + . Reaction C 2 H 5 Br + + h ν 355 nm → C 2 H 3 + +H 2 +Br could be considered as a concerted three-body dissociation process. Before Br atom moves out of the interaction range the H 2 molecule is already formed. Morework is being done to reveal each of the H-loss channels. By measuring negative ions from C 2 H 5 Cl and CH 3 Br we show that the previously proposed ion-pair mechanism for the formation of C 2 H 5 + from ethyl bromide at 118 nm is wrong [J. Phys. Chem. A 101 (1997) 1222].