| New Theoretical Approaches to the Solution of the
Schodinger Equation for the Electronic and Vibrational States of
Polyatomic Molecules |
During the past twenty-five years, quantum chemistry has progressed to the point that computer results on the electronic properties of molecules often rival those obtained from experiment. Over this period, Professor Segal has made many contributions to this accomplishment.
His group has two primary foci- the development of new theoretical approaches to the solution of the Schodinger Equation for the electronic and vibrational states of polyatomic molecules, and the application of these methods to problems of current interest. As in all such efforts, one applies a new method until it fails, as all inexact methods must do, then moves to an improved approach capable of dealing with the problem at hand.
While he has most recently been inactive in research while serving as the Dean of the School of Letters, Arts and Sciences, he has always collaborated closely with experimentalists to (1) provide guidance towards a deeper understanding of experimental results and (2) compute accurate surfaces upon which further chemical dynamics calculations can be carried out. His group has developed highly efficient computational methods for calculating fully correlated ab initio wavefunctions and energies for molecules of some size. These techniques, which allow computations on either ground or electronically excited states, have led in a number of directions.
For example, he has studied some features of the potential energy surface of HClCN, the surface on which the scattering of H atoms from cyanogen chloride, ClCN, takes place. He has also studied the ground and electronically excited potential energy surfaces of NCNO along its dissociation coordinate to CN + NO. Experiments in both of these systems have been actively pursued.
The development of correlated techniques which are both accurate and efficient allows computations on moderately large molecules. This permits ab initio studies of circular dichroism spectra where such studies have previously been difficult or impossible, since the effect doesn't even exist for molecules small enough to be treated by previous computational methods.
Segal has also carried out calculations on the vacuum ultraviolet circular dichroism spectra of amino acid zwitterions. The zwitterionic form does not exist in the gas phase - there must be a dielectric medium to stabilize the separation of charge - and this must be taken into account in the quantum mechanical calculation of the circular dichroism spectra. To meet this problem, a method for ab initio calculations of the wave functions and energies of zwitterions in water solution has been worked out using a fractional charge model for the solvent molecules.
The initial application was to the electronic absorption spectrum of the glycine zwitterion. As expected, careful calculations on the electronic states of this molecule in the gas phase, including a rather full consideration of electron correlation, led to predictions which bore no resemblance to experimental observation. Explicit inclusion of the solvent resulted in a dramatic change in the predicted spectrum involving both large hyposochromic and bathochromic shifts for the valence excited states and the suppression of absorption by Rydberg states.
These calculations on glycine were an exploratory study on a molecule which is well understood but not optically active. Further development of this approach is needed.