Over the past two years the major focus of our research has been in the area of carbon radical generation.  Radicals are unstable, highly reactive species which have an unpaired electron.  They tend to react so as to generate a two electron bond (a paired electron situation).  Often this occurs via a radical chain process, that is, a new radical at a different site is generated concomitantly with bond formation.  Our goal is to find new ways to produce radicals from highly oxidized carbon species.  We are trying to produce radicals in a reductive environment in a manner which is compatible with carbon carbon bond formation.  This is a quite important area of inquiry because C-C bonds are among the hardest to make in organic chemistry.  Many research groups around the world are engaged in similar quests.

As mentioned, we are trying to generate radicals from highly oxidized carbon atoms.  This means we are looking at functionalities which possess multiple(2 or 3) heteroatoms (a heteroatom is an atom other than a carbon, hydrogen, or metal). A general description of our approact to this is shown in Scheme 1.  We first activate the heteroatom-substituted carbon atom with an electrophile.  The resultant cation is then reduced with samarium diiodide (SmI₂, or Sm⁺² for simplicity).  Samarium diiodide is a one electron reductant thus it generates a radical or unpaired electron species.  This unstable, reactive moiety in turn reacts with an appended activated alkene elsewhere in the molecule to generate a new cyclized radical, which is further reduced with samarium diiodide and then converted to a stable species upon subsequent addition of water.
                                           ;                          Scheme 1

Up to the summer of '99 we had  had modest success with this idea starting with amides as our incipient radical site (Scheme 2).  Our best results involve treating our amide with triflic anhydride as our electrophile, then adding samarium diiodide and HMPA (a cosolvent to enhance the reactivity of the samarium species), and finally water.  Using this protocol we are able to isolate 35% of our cyclized product. Significant amounts of starting amide are also recovered.  This represented our best result after a significant amount of time spent optimizing this reaction.  We've investigated changing solvents, proportion of regents, temperatures regimens, and nitrogen substituents.
                                           ;                              Scheme 2
 
 

 
 
 
In the fall of '99 we finally got around to trying a couple of ideas that have percolating for a while.  Several literature reports have recently indicated that certain transition metal salts can function as electron transfer catalysts for SmI₂ reduction.  We had held off trying this idea because the literature examples seemed to all be in the realm of two-electron transfer reactions involving SmI₂  (or two rapid one-electron transfers,  "Barbier-like" reactions).  I did find one example involving the reductive dimerization of imines which most likely occurred via a one-electron pathway, which used NiI₂ as the electron transfer catalyst.  So finally we tried it with the hexyl ester version of our N-methyl,N-phenyl amide.  We were stunned (but pleased) to see the product yield jump into the 60% range on three successive tries!  Buoyed by these results, we also tried our other idea of trying to stabilize our cyclized product by protonation of the ester enolate which is formed in the second reduction step.  Typically in samarium reaction, t-butanol is used for this purpose.  When we tried the reaction with Tf₂O as electrophilic activator, SmI₂ as reductant, NiI₂ as electron transfer catalyst, and t-butanol as our proton sources, we were rewarded with a 87% (yes 87%) yield of the cyclized product.  After a three-year odyssey, we can finally pronounce the parent reaction of this series optimized.
 
 
 
 

We are currently attempting to generalize this reaction.  We are in the process of constructing other appropriate substrates to examine effect of substitution on the various atoms of the incipient ring, whether 6-membered rings can be formed, and whether an intermolecular version of the reaction is doable.

Over the past 12 months  investigated related compounds as to their suitability as substrates for our electrophilic activation/reduction protocol for radical generation as shown below in Scheme 3.  Note all of these substrates posess a carbon atom bound to multiple heteratoms (marked with an asterisk).  To date, none of these has yielded any identifiable cyclized product.

                                                                 Scheme 4
 

Recently our attention has been directed to 1,3-thiazolines as potential substrates.  Jennifer Kowalchick had some intruiging results during our Summer '99 research.  She showed that simple thiazolines are reduced by our electrophilic/reduction protocol in pretty good yield (this reaction hasn't even been optimized yet.  Attempts at cyclizing a styryl-substituted thiazoline resulted in simple reduction.  Our full focus then will from now on be directed at the original (and now optimized) amide cyclizations.

                                                                                 Scheme 4

Here's my summer '99 research group.  That's Amy Galka on the left, then me, then John Mazzullo, and finally Jen Kowalchick.  Below is a full list of students who, over the past three years, have worked on this project.  I am looking forward to seeing how much progress we can make on this project over the next spring and summer.
 

Student Researchers:  Ahren Green, Dave Wisnoski, Joe Keane, Amy Galka, Jennifer Kowalchick, John Mazzullo, Vanessa Richter, Zach Shiffler