This project by Paul Young of the University of Illinois, Chicago, is within the Organic Dynamics Program, and is aimed at a study of the reactions of organic compounds containing tricoordinate (i.e., three bonds) sulfur atoms. The detailed pathway by which these compounds react with negatively charged species will be investigated with special attention given to the competition between single and multiple step routes. This work should allow chemists to better predict and control complex reactions involving sulfur atoms. There is a significant body of data suggesting that, under certain conditions, alkyl transfer reactions from simple sulfonium salts may occur through a ligand coupling mechanism involving reductive elimination from an intermediate tetracoordinate sulfurane. This suggestion is easily testable by monitoring the product distribution and stereochemistry of selected reactions. A series of chiral S-methyl-S-(substituted alpha-methylbenzyl)-S-(substituted phenyl)sulfonium salts with known chirality around the benzyl group will be synthesized, and reacted with the following nucleophiles: (1) substituted ethyl alcohols (solvolysis conditions will be utilized and the progress of the reaction followed by NMR), (2) substituted pyridines and primary, secondary and tertiary amines in aprotic solvents and protic solvents found to support minimal solvolysis activity, (3) strong nucleophiles including phosphines and carbanions, (4) high concentrations of substituted acetates and benzoates under conditions where a second-order dependence on nucleophile concentration is observed, and (5) bifunctional nucleophiles including alpha- pyridones and alpha-pyridinethiones. For each class of nucleophile, product distributions will be obtained with special emphasis on the chirality of the transferred benzyl group. For each of the observed pathways, the effects of structure on reactivity will be investigated using classical Hammett and Bronsted correlations. Secondary deuterium isotope effects and C-13 isotope effects on the nucleophilic rate constants will be determined for comparison with isotope effects observed for methyl transfers catalyzed by the enzyme, catechol-O-methyltransferase. The effect of nucleophile strength on the partitioning pathways will be investigated, both within a given series (Bronsted correlations) and across the array of nucleophiles investigated (N+-type correlations). These data, taken as a complete set, will define the limits of the "ligand coupling" pathway in substitution reactions occurring at tricoordinate sulfur and will allow a more complete understanding of the role of nucleophile structure on each of the potential pathways for the alkyl transfer process.
