Columbia Strikes Again?

Brendan D. Kelly, Tristan H. Lambert Cyclopropenium-Activated Cyclodehydration of Diols Organic Letters, ASAP January 19, 2011

Lambert is certainly a beast. He churned out not one but two ACS-grade papers in a mere week of each other. Not only that, but on completely different projects! That’s quite a feat. Now before you all go and start saying that I’m a Lambert fanboy, I’d like to let you know the only reason I found this one was by visiting the Lambert group website to see how big his group is and if he was taking on post-docs. Because I would certainly consider working for him. His work may not be on the level of a Corey synthesis or as impactful say as some of the work done by Doc Brown, but it is certainly interesting, weird, and ingenuitive. And practical. I love practical. And carbocations, I love ’em too. Ponders Maybe I am becoming a Lambert fanboy…Moving on…
I am very impressed with this article, less so than the tropylium article but not by much. I have some quips with this paper such as a lack of a complete mechanism amongst others but I will get to those towards the end of the review. Lambert begins by asserting that one of the most important reaction “types” in an organic chemist’s arsenal is the dehydration reaction. And, if you think about it, he’s certainly at least partially right. Reactions of this type range from simple (such as a SN1 or E1 reaction) to more elaborate transformations such as the Mitsunobu reaction. However, Lambert states that there are a lot of limits on how we can employ this weapon effectively (including harshness of conditions or limited substrate scope). So, his group sought to fix that by developing a reagent to facilitate dehydration. Using their already established methodology dubbed by Lambert as “cyclopropenium activation”, the Lambert group investigated one type of dehydration reactions, namely the formation of cyclic ethers from diols. So why cyclic ethers? Well they appear in MANY natural products. If you don’t believe me, just pick out a total synthesis from the literature. I’d say there is a 75% chance there is a cyclic ether in it. So we are actually talking about some synthetically useful stuff here.

They started off with the above transformation as a model, basing it off their previously published work with the cyclopropenium ion (2009 and 2010 JACS communications). They hoped to generate their well known dichlorocyclopropene in situ and using that to facilitate cyclization (since in their previous article they showed that this cyclic moiety can be used to convert hydroxyl groups into chlorides much in the same way that SOCl2 operates). Pretty logical step, go from a system where there are no intramolecular nucleophiles to one where there is and see what happens. They tracked the reaction by NMR and were mostly pleased with the results. Their commercially-available enantiopure diol cyclized just the way they wanted to. Not only that but it was fast, high yielding, and pretty much free of side reactions. On the downside, they killed the d.r. Not terribly so, but going from 20:1 to 12:1 dr is a big hit considering that most synthetic chemists rarely have a 20:1 pure intermediate.

They sought to fix this problem by switching their “activating agent” to various anhydrides. The one they found worked the best (retaining yield and now with complete conservation of d.r.) was methanesulfonic anhydride (Ms2O). They also screened the bis iso-propyl derivative of their cyclopropene and found it gave comparable results to the bis-phenyl compound. Since it seems like the bis phenyl is easier to make, they stuck with that. With the optimal conditions in hand, they explored the scope. The reaction proved to be quite broad (10 examples). Notably, they were phenolic hydroxyl groups giving rise to a chroman (benzene fused to a tetrahydropyran) derivative and from a stereochemically intense benzylated mannitol straight to its THF analog (as a single diastereomer!). Finally, Lambert finished with the synthesis of a potential intermediate in a total synthesis. to illustrate the broad impact of his method and also to show scalability. Starting from the hydroboration/oxidation product of (-)-Isopulegol , the Lambert group (aka Brendan) was able to convert it into the cooresponding ether in great yield and completely conserved d.r. (see above). Not only that but this was done on A GRAM scale. Quite nice in my opinion.

Now you may wonder how this reaction works…and so did I. I mean I got the general idea from some of their previous articles. But I really appreciate when papers put in their mechanism so I was kind disappointed they didn’t (or at least not to my satisfaction, they did show the “mechanistic design” but I like specifics not generalities in mechanisms). Even a plausible mechanism would be nice. Hence why I put one above. The other thing that bugged me was if you look at the net equation at the top, the cyclopropenone is regenerated at the end of the reaction. I really think this methodology would be quite amazing if you could turn it truly catalytic (5% or 10% mol of the cyclopropenone). I mean regardless it is recoverable so it’s still technically catalytic. But it’s a lot more practical on a large scale to cut down on what ever you are throwing in…because somewhere down the line you’ll need to get it back out. Another issue is, if you consider my mechanism (which may be wrong but is probable) there should be the formation of methanesulfonic acid (2 moles per mole of substrate). This would get awfully acidic very fast. I wonder if they ran into these problems or their ethers were simply resistant to degradation. Regardless, overall another excellent article by Lambert! Ckellz…signing off…


  1. In your mechanism, that activation pathway for the carbonyl looks remarkably unstable after initial mesylation of the carbonyl. I am not an organic graduate student but an undergrad. However, I didn’t think the orbitals could align for resonance stabilization as you drew.

    I was thinking that the alkene existed as a diradical form and reacted through such a pathway, but I didn’t actually figure out a potential pathway from that.

  2. Oh very very interesting point, I can see the diradical thing (where maybe it can make the whole reaction diradical, and that the carbonyl oxygen itself is in fact reacting as a radical. However I’m thinking despite the apparent instability, the driving force is making that cyclopropene aromatic. I mean it would have a huckel # of 4n+2=2 and if you consider 0 a even number then technically the system would be aromatic. I believe it works the same way that the tropylium ion work…oh theres an idea, use tropone to do a similar kind of reaction….would probably be easier too cause tropone is commercially-available….youve given me something to look into… Their “mechanism” proceeds through the formation of an acetal. The reason I showed activation the way I did was because they cite that when using Oxalyl chloride they get the dichloride. I just assumed that it would react with Ms2O in the same way (they also say something to that effect). But i’d be curious to see a diradical pathway 🙂

Comments RSS TrackBack Identifier URI

Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s