Quantum optimal control of molecular isomerization

Quantum optimal control is applied to the problem of isomerization of small molecules with the goal of guiding the design of possible experiments as well as developing new computational tools. Three illustrative cases are presented: a reduced-dimensionality model study of ozone isomerization, a two-dimensional model study of optimal control of isomerization in the presence of a competing dissociation channel, and a full vibrational degrees of freedom treatment of hydrogen cyanide-isocyanide isomerization.;A feasibility study of ozone isomerization utilizes a model Hamiltonian constructed by holding the bond lengths constant and using the valence angle as the isomerization coordinate. A post facto analysis is performed showing a degree of inherent robustness of the isomerization yield to field noise.;Optimal control of isomerization in the presence of a competing dissociation channel is simulated on a two-dimensional model. The control is achieved through vibrational transitions on the ground-state surface as well as with the aid of an excited-state surface. The effects of different competing dissociation channel configurations on the isomerization control are explored.;A method for incorporating strong electric field polarization effects into optimal control calculations is presented. A two stage toolkit implemented to incorporate the polarization effects and reduce the cost of the optimal control dynamics calculations. As an illustration, the method is applied to optimal control of vibrational excitation in a hydrogen molecule.;Quantum optimal control of HCN isomerization is studied including all vibrational degrees of freedom and with rotation treated in the sudden approximation. The nuclear dynamics toolkit technique is used for solving the control equations. The C-N stretching mode is found to be important in the isomerization dynamics.;Sensitivity analysis of the wavefunction is shown to be useful for identifying the key variables and states in optimal control calculations. The analysis calls for little extra effort beyond that of performing the original control calculation. An application of the analysis to a model multilevel control problem is presented.