Konische Durchdringungen und der Mechanismus von Singulett-Photoreaktionen

A theoretical investigation of photochemical reactions of an isolable silylene with benzene

The mechanisms of photochemical insertion reactions are investigated theoretically using the model system, an isolable dialkylsilylene and benzene, using the CAS(10,10)/6-31G(d) and MP2-CAS-(10,10)/6-311++G(df,pd)//CAS(10,10)/6-31G(d) methods. The structures of the conical intersections, which play a key role in such photoinsertion reactions, are determined to provide a qualitative explanation of the reaction pathways. The model investigation demonstrates that the preferred reaction route for the isolable dialkylsilylene with benzene is as follows: reactants → Franck-Condon region → conical intersection → seven-membered-ring photoproduct. The theoretical findings suggest that the singlet excited dialkylsilylene should attack benzene in the perpendicular conformation, and that no silyl radicals should exist during these photoinsertion reactions. The results obtained allow a number of predictions to be made.

A theoretical characterization of the photoisomerization channels of 1,2-cyclononadienes on both singlet and triplet potential-energy surfaces

The ground-, (1)(pipi*)-, and (3)(pipi*)-state potential-energy surfaces of 1,2-cyclononadiene and isomeric C(9)H(14) species, as well as 1-methyl-1,2-cyclononadiene and isomeric C(10)H(16) species were all mapped using CASSCF and the 6-31G(d) basis set. Theoretical results were found to be in good agreement with the available experimental observations for both 1,2-cyclononadiene and 1-methyl-1,2-cyclononadiene isomerization reactions under singlet and triplet direct or sensitized irradiation. Extremely efficient decay occurs from the first singlet excited state to the ground state through at least three different conical intersections (surface crossings). The first of these crossing points is accessed by a one-bond ring closure. From this conical intersection point (CI-A or CI-C), some possible subsequent ground-state reaction paths have been identified: 1) intramolecular C--H bond insertion to form the bicyclic photoproduct and 2) intramolecular C--H bond insertion to form tricyclic photoproducts. An excited state [1,3]-sigmatropic shift leads to the second conical intersection (CI-B or CI-E), which can give a three-bond cyclononyne species. Besides these, in the singlet photochemical reactions of 1-methyl-1,2-cyclononadiene, excited-state, one allenic C--H bond insertion leads to a third conical intersection (CI-D). Possible ground-state reaction pathways from this structure lead to the formation of a diene photoproduct or to transannular insertion photoproducts. Moreover, in the case of triplet 1,2-cyclononadiene and 1-methyl-1,2-cyclononadiene photoisomerization reactions, both chemical reactions will adopt a 1,3-biradical (T(1)/S(0)-1, T(1)/S(0)-2, and T(1)/S(0)-3), which may undergo intersystem crossings leading to the formation of tricyclic or bicyclic photoproducts. The results obtained allow a number of predictions to be made.

Conical intersections and semiclassical trajectories: Comparison to accurate quantum dynamics and analyses of the trajectories

Semiclassical trajectory methods are tested for electronically nonadiabatic systems with conical intersections. Five triatomic model systems are presented, and each system features two electronic states that intersect via a seam of conical intersections (CIs). Fully converged, full-dimensional quantum mechanical scattering calculations are carried out for all five systems at energies that allow for electronic de-excitation via the seam of CIs. Several semiclassical trajectory methods are tested against the accurate quantum mechanical results. For four of the five model systems, the diabatic representation is the preferred (most accurate) representation for semiclassical trajectories, as correctly predicted by the Calaveras County criterion. Four surface hopping methods are tested and have overall relative errors of 40%-60%. The semiclassical Ehrenfest method has an overall error of 66%, and the self-consistent decay of mixing (SCDM) and coherent switches with decay of mixing (CSDM) methods are the most accurate methods overall with relative errors of approximately 32%. Furthermore, the CSDM method is less representation dependent than both the SCDM and the surface hopping methods, making it the preferred semiclassical trajectory method. Finally, the behavior of semiclassical trajectories near conical intersections is discussed.

Femtochemistry of Norrish type-I reactions: III. Highly excited ketones--theoretical

Time-dependant density functional theory (TDDFT) and ab initio methods (CASSCF and CASMP2) are applied here for the investigation of the excited-state potential energy surfaces of ketones studied experimentally in the accompanying paper, number IV in the series. The aim is to provide a general and detailed physical picture of the Norrish type-I reaction from S0 and S1 potentials (papers I and II) and from higher-energy potentials (papers III and IV). Particular focus here is on reactions following excitation to the 3s, 3p, and 3d Rydberg state and to the (nz-->pi*) and (pi-->pi*) valence states. It is shown that the active orbitals in the CASSCF calculations can be chosen so that accurate results are obtained with a small active space. Dynamic corrections of the state-specific CASSCF energies at the multireference MP2 level do not improve the results for the Rydberg states but are significant for the valence states. The geometries of the Rydberg states are similar to the ground state; the S1 and other valence states are not. A common property of the valence states is the elongated CO bond and the pyramidalization of the carbonyl carbon atom. As a consequence, these valence states cross all Rydberg states along the CO stretching coordinate and provide an efficient pathway down to the 3s Rydberg states (S2) through a series of conical intersections (CIs). The nonadiabatic coupling vector of the CI between the (pi-->pi*) and the 3s Rydberg states guides energy channeling into the asymmetric CC-stretching mode. The energy demand for the CC bond breakage (Norrish type-I) on the S2 surface is lower than that of the CI leading to the S1 state. This CC bond breakage leads to a linear excited state acetyl radical (3s Rydberg). Crossing a small barrier the 3s acyl radical can access a CI leading either to a second CC bond breakage or to a hot ground-state acetyl radical. The barriers for the Norrish type-I reaction on the various excited-state surfaces can be rationalized within the framework of valence-bond theory. The dynamic picture of the Norrish type-I reactions is now clear: The excitation to high-energy states leads to the nonconcerted breakage of the alpha-CC bonds by an "effective downhill" potential in space involving the active excitation center CO, CC stretching, and CCO bending nuclear motions, but not, as usually thought, a direct repulsive potential along the CC bond. In our accompanying paper (part IV), it is shown that the results from the experimental investigations of Norrish type-I reactions on the femtosecond timescale are consistent with these theoretical results.