Dispelling those undergraduate myths…

Mark Mascal, Nema Hafezi, and Michael D. Toney 1,4,7-Trimethyloxatriquinane: SN2 Reaction at Tertiary Carbon J. Am. Chem. Soc. 2010 132 10662-10664

So every once in a while an article comes along that really is an example of my kind of chemistry: pure, unapplied, organic chemistry. In the golden days of organic chemistry, the majority of publications contained chemistry that really did not have tangible application but were just synthetically sweet. For example, look at the numerous syntheses of highly strained molecules or highly symmetrical molecules. Basically, I love the days when physical organic was much more valued than it is today. So when I found this article a while ago, I was thrilled. No applications at all – just pure chemistry ownage in one of the leading journals, JACS. I was very impressed. I must warn you though, this article by Mark Mascal’s group over at UC Davis isn’t what I would call a methodology (or at least not in the traditional sense, I would call it a hybrid between a total synthesis and a methodology.) But nonetheless, this article is awesome and just by chance, I ran into it again recently. I believe it more than deserves a highlight on New Reactions.

So as budding organic chemists, we all learn that it’s impossible to do the reaction I’ve depicted in equation 1 above. Tertiary centers are simply too inaccessible to nucleophiles to get that good old backside attack to occur. However, we know these sorts of substrates are great for SN1 reactions, e.g. equation 2. Later, if we aren’t completely traumatized by organic chemistry and actually enjoy it enough to pursue a career in it, we enter grad school and, to paraphrase Yoda, ‘we must unlearn what we have learned’. We find out that it’s really the steric interactions in the pentavalent-transition state that make tertiary substrates mostly incompatible with the SN2 model. The activation energy is phenomenally high so the reaction just doesn’t go. Now you could argue that all you would have to do is heat it but the energy you would have to put in to get it to go would most likely yield products from less energetically demanding reaction paths. In short, you’d kill the starting material either through decomposition or lose it to a competitive reaction path. Notice how I said mostly and most likely. There have been isolated examples in the literature of SN2 reactions occurring under reasonable conditions at tertiary centers. Some have been explained as addition-elimination reactions, but others truly give evidence for SN2 reactions occurring on tertiary centers. However, they have never been as clear-cut as the work done by Mascal’s group.

Mascal’s group has a relatively broad range of interest, ranging from biofuels to total syntheses. However, his most well-known contributions are in the syntheses of two unusual classes of molecules: triquinanes (1) and triquinacenes(2). Lately he has been interested in the oxonium bad boys you see above (3,4,5). Now I want to draw your attention to the fact that these compounds are stable salts. 3 is highly stable (I don’t mean like glovebox stable, I mean like the AK-47: this compound can be subjected to the harshest conditions and survive). Mascal’s group was able to reflux 3 in water for 72 hours and even column chromatograph it on silica gel without adverse effects. That’s unprecedented for an oxonium ion. While 4 wasn’t as rugged, it still was isolatable and stable in acetonitrile (which you would think would be N-alkylated). Not surprisingly, both compounds reacted well toward harder nucleophiles (N3-, CN-, OH-) but 1 did not react towards weaker (softer) nucleophiles such as amines, iodide, or alkyl thiols/alcohols.

Now all that work was done in 2008. Fast-forward to 2010. In this recent article they disclose not only the synthesis of 5 but its very unique reactivity. I’ve shown the path they took to synthesize 5. Overall, not bad if we put this on a total synthesis scale (1.34% over 18 steps). Reactions up until intermediate 13 (in red) were performed in their 2008 work, so they didn’t actually show that compound. Kind of deceptive because that took 7 steps to make. But regardless, the point of this particular article wasn’t just a total synthesis; it was to construct a highly stable tertiary oxonium ion.

They did that and more. Like their previous article, they subjected 5 to a variety of harsh conditions. One would expect that, just as with any other tertiary oxonium ion(usually formed in situ ), solvolysis would proceed rapidly with 5. However, this was not the case whatsoever. That in and of itself is extremely anomalous if we considered what we learned about SN1 reactions in Organic I. Normal unreactivity to an organic chemist is a bad thing but in this case it drove the Mascal group to dig further. They attempted to react 5 under SN2 reactions with the hard nucleophiles mentioned earlier. As one would expect, the majority of these nucleophiles (CN-, OH-) acted as bases yielding E2-like elimination. However, reaction with N3- gave solely the azide product (above). While one could invoke ion-pair arguments, the next two pieces of data seemed to easily dismiss that possibility: The reaction was slowed both by polar protic solvents and by the addition of a non-nucleophilic salt, LiBF4. Now for the nail in the coffin. They performed what may seem like an undergraduate exercise: Determining the rate of reaction and rate constant by NMR and graphing it. And here’s why I said this is a clear-cut example of a tertiary SN2 reaction. The rate constant was determined to be 0.0235 M-1 S-1 . And computational data results suggest that no such equilibrium between a carbocationic form and the oxonium form exists (see below).
So I must say, while this article isn’t as impactful as the discovery of the Dess-Martin reagent or the Prakash procedure for trifluoromethylation, it is very cool to me. And the results don’t lie. If there were any SN1 or ion-pair character to the mechanism, the addition of the salt or running the reaction in a polar protic solvent should have increased the rate of the reaction. The kinetic results clearly show a 2nd order rate dependency and the computational results support that a carbocationic form is not a viable intermediate. Congrats to the Mascal group for an excellent and thorough job and proving that tertiary substrates can be used in SN2 land(and for keeping pure chemistry alive today!) .

Leave a comment

No comments yet.

Comments RSS TrackBack Identifier URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s