Principal Investigator Brian Yates Project g29

Chemistry Department, Machine VP

University of Tasmania

Co-Investigator George Heard

Chemistry Department, University of Tasmania

Theoretical Studies on the Mechanism of the Stevens Rearrangement and Related Reactions

We have used computational chemistry to calculate the structures and energies of organic molecules, reactive intermediates and transition structures, with a view to investigating the mechanism of a specific organic reaction, the Stevens rearrangement.

The Stevens rearrangement of an alkylammonium ylide to an amine was discovered in 1928 and has found recent application in organic synthesis in the preparation of new heterocyclic compounds. This reaction and related rearrangements have been used in our Department in various ring-expansion procedures. With the aid of the VP2200 we have carried out a thorough investigation of the mechanism of this reaction, including concerted, stepwise, ionic, and metal-catalysed pathways. The use of the VP2200 has made it possible to include solvation effects and to study systems of a realistic experimental size. All of this information has provided us with a broader understanding of the mechanism and enabled us to better interpret the experimental studies.

This project was supported by a large ARC grant.

What are the basic questions addressed?

For the Stevens rearrangement, which is lower in energy: the concerted or the free-radical reaction path?

How does this answer change when the size of the molecule is increased?

How does this answer change as a function of different steric and electronic substituent effects?

How does this answer change as a function of solvation?

What are the results to date and the future of the work?

On the basis of extensive calculations, we have concluded that a radical pathway is preferred for the Stevens rearrangement reaction in organic chemistry. Further calculations on solvation effects and alternative pathways (such as the ion pair and lithium catalysed mechanisms) have not altered our basic conclusion. Although the inclusion of various substituents has some minor effects on the reaction profile, our study of twelve systems (including experimentally observed reactions) showed a remarkable similarity in the relative energies of the reactants, products and intermediates.

Further work has begun to study a set of related competing rearrangement reactions. Experimental evidence suggests that a number of products can arise from the same starting material as shown in the reactions below:

Ylide Stevens Sommelet-Hauser Hofmann

Initially a prototype system has been chosen for study and the competing pathways have been explored. Further calculations are planned to study larger systems and to determine the factors which control the experimental preference for one pathway over another.

What computational techniques are used and why is a supercomputer required?

We have used quantum chemistry techniques (mainly conventional ab initio SCF and MP2 methods together with density functional methods) with the Gaussian 92 program. These calculations make good use of the vector facility. We require a supercomputer to enable us to study large molecules with high levels of theory. Such big calculations are beyond our local workstation resources.


Theoretical Studies of the Stevens Rearrangement of Alkylammonium Ylides, G L Heard and B F Yates, J. Mol. Struct. (Theochem), 310, 197-204 (1994).

Steric and Electronic effects on the Mechanism of the Stevens Rearrangement - Large Organic Ylides of Unusually High Symmetry, G L Heard and B F Yates, Aust. J. Chem., 47, 1685-1694 (1994).

Theoretical Evaluation of Alternative Pathways in the Stevens Rearrangement, G L Heard and B F Yates, Aust. J. Chem., submitted.