The principal endeavour of my research group is the development of new methods to study chemical dynamics. We calculate vibrational spectra, ro-vibrational spectra, photodissociation cross sections, and reaction rate constants.
The quantum description of vibrational and ro-vibrational states is a subject of great interest because of the fundamental importance of intramolecular energy transfer in many areas of chemical dynamics. To refine current theories of unimolecular decomposition, collisional energy transfer, multiphoton excitation, and mode selective processes it is imperative that we improve our understanding of highly excited states.
Variational calculations of molecular spectra are important because they enable one to test and determine potential energy surfaces, to predict spectra, and to interpret and understand experimental data. To determine accurate global potentials one must optimise a potential function by comparing calculated and experimental energy levels. The spectra of molecules with large amplitude vibrations enable us to determine much of the potential energy surface because the vibrations explore large regions of the surface. The study of highly-excited molecules is also valuable because it fosters the development of new concepts. The nature of quantum systems in the classically chaotic regime is among the most fundamental issues in modern chemical physics. The characterisation of excited states is important for the understanding of multiphoton excitation, intramolecular energy redistribution (IVR), unimolecular reactions, laser-molecule interactions, selective photodissociation, etc.
Highly excited vibrational states are interesting and important because they extend over large anharmonic portions of the potential energy surface where there is significant coupling between many modes but for these reasons it is also very difficult to calculate their energies and wavefunctions. One of the important factors that impedes the calculation of a spectrum from a matrix (basis set) representation of a Hamiltonian, H, is the difficulty of computing eigenvalues and eigenvectors of a large matrix. The size of H increases exponentially with the number of degrees of freedom. We have developed good iterative methods for computing energy levels and wavefunctions.