A Friedel-Crafts Goes Awry…

Dolbier, W.R. Jr.; Cornett, E.; Martinez, H.; Xu, W. Friedel-Crafts Reactions of 2,2-Difluorocyclopropanecarbonyl Chloride: Unexpected Ring-Opening Chemistry J. Org. Chem. ASAP April 1st, 2011

Another long week in the lab. Research continues (knock on wood) to go quite well. Me and my lab-mates certainly have enough chemistry to get done to keep us busy. The odd thing to me about grad school is the lack of “school”. I really don’t feel the same sort of pressure to perform at 110% in class. I mean don’t get me wrong I sure as hell still try and still do the assignments early, but its different (and clearly hard to explain). I treat grad school more like I would a transitional job (where I’m there pretty much to pump out good research) rather than an extension of my undergraduate career. Hence why I work 12+ hour days because its my job to make chemistry (and make a lot of it ) happen. Because I’m always there late, I’ve made a friend in one of the other groups (a 5th year doing some pretty cool organic synthesis). We hope to do a collaboratory project (adding to my ever-long list of things to do) at some point on a project he has been developing. In other news Dr. Ronald Breslow paid UConn a visit as part of our R.T. Major symposium. I had met Breslow before when I was down at Columbia so I had an idea of what he did and the kind of guy he is. Nonetheless, I was very impressed with his talks, in particular, one regarding the origins of why natural amino acids are (L) and natural sugars are (D). I also read up on it following the talk in a recent article published by his group. While one can never really “prove” with absolute certainty that the process proposed by Breslow is truly what lead to all amino acids (natural) being (L) it does at least give us a theory to possibly explaining it. But enough about all that, you’re here for the chemistry aren’t ya? So let’s talk chemistry!

So, this article also has a bit of irony to it (though not the element). My wonderful girlfriend recently got me a book I’ve been dying to have: A Guide to Fluorine NMR for Organic Chemists. It is by far one of the best organic chemistry books I’ve gotten besides the holy bible of named reactions, Strategic Applications of the Named Reactions in Organic Synthesis. The book only has one author and he has somewhat of an unusual last name. So, unknowingly, when I was scrolling through the ASAPs, I stumbled onto this gem of an article in JOC. Low and behold, it was written by the same chemist that wrote my F-NMR book (therefore I have to review it right?)!

The article starts out by outlining how the synthesis of 2,2, difluorocyclopropyl ketones has been realized by the discovery of a difluorocarbene source, trimethylsilyl fluorosulfonyldifluoroacetate (TFDA). The Dolbier group pretty much has been pioneering this reagent and exploring its applications/reactivity. I wouldn’t say it’s terribly well-known simply because there often isn’t a call for geminal difluorocyclopropanes. However, Dolbier (who is the editor of J. Fluorine Chem. which I found pretty awesome) has been exploring the reactions of 2,2, difluorocyclopropyl ketones and found some interesting reactions.

One of the main drawbacks to his sort of carbene chemistry is that aryl alpha, beta unsaturated ketones make poor coupling partners because of competitive polymerization. Hence, they wanted to see if you could reach these aryl substrates via standard sort of Friedel -Crafts chemistry. Based on the fact that dichlorocycloproylcarbonyl chlorides had been utilized to this effect with no adverse effect to the cyclopropyl moiety, it stood to reason that the fluorine analog should react the same (if not better due to the inert nature of organofluoro compounds). But, as usual, fluorine doesn’t play by the rules…

Instead of giving the normal, unrearranged acylation product, it gave the straight chain compound via a ring-opening reaction. Not only that, it gave a mixed halogen ketone which is until this paper would be near impossible to synthesize. Since these are new compounds, their synthetic potential has yet to be explored so the authors really couldn’t give too much in the way of an application. But they certainly did decide to determine the scope of the reaction. The reaction seemed to work best with alkyl benzene derivatives (toluene, ethylbenzene, p-xylene). Moreover they were sensitive to sterics: the higher the substitution on the benzene ring the slower the reaction (discovered via a competition reaction between toluene and p-xylene).

When switching to what I would call more activated aromatic systems (anisole, thiophene) the preferred product ended up being the “normal” Friedel-Crafts product. Even more unusual was that if 5 equivalents of the starting arene were used, a reversal in the product was seen (favoring the cyclopropyl-containing product). No obvious explanation could be give as to why the difluorocyclopropyl system behaved so much differently than the dichlorocycloproyl system nor could they explain the trends seen in their substrates. Hence the authors turned to computation. The results were astounding. First, they noted that alpha-fluorine substituents are known to stabilize carbocations substantially as compared to chlorines. I found that quite surprising due to the fact that fluorine is such a strongly destabilizing substituent in all other cases. My guess is that, since it is in the same row as carbon, that it possesses the proper orbital overlap to participate in stabilization as compared to the orbital mismatch seen with chlorine. The authors also note that geminal difluorocycloprpanes are under a substantial degree of strain as compared to their fluorine analogs (which I assume relates to inductive and hybridization effects). Therefore they proposed the following reaction intermediate:

They argue that this intermediate explains the reactivity they were observing in that the chloride anion competes with acylation of the arene. Since the more activated arenes react faster, it’s no surprise that these give more of the non-ring-opened product as they can actually compete with the chloride ion. Likewise, having additional amounts of the arene (5 equiv) leads to out competition of the chloride ion from a statistical standpoint (and why it never fully gives 100 percent of the unrearragned product).

What I really loved about this article is the thoroughness. Not only did they back up their hypotheses computationally, they were sure to do several control study to ensure that their proposed mechanism wasn’t an artifact of decomposition of the unrearranged product. First, they extended the reaction times and noted no increase in the quantity of rearranged product. Also they took the unrearranged product and treated it with AlCl3 under the normal reaction conditions and noted no reactions. To further verify their findings, they then compared the Friedel-Crafts reaction to other reactions done by the group.

One such reaction, the solvolysis of 2,2-difluorocyclopropyl-methyl tosylate (the nostalgia to my undergrad work is ever present :P) leads to the ring opening reaction to produce O-alkylation by the solvent or by the tosyl anion. However, this also proceeds through the stabilized difluorocarbocation. Moreover, a drastic difference in reactivity is observed when the mechanism becomes more SN2 like. Dolbier’s group treated these same sorts of difluorocyclopropyl ketones with HBR and observed the following:

The nucleophile actually ends up substituting the “less” hindered and more reactive CH2 group over the CF2 group. Therefore, if the reaction is truly more SN2 like (which its not) we would expect to see more substitution at the CH2 group. Since we see a more analogous reactivity to the solvolysis studies, one must conclude that the difluorocarbocation is the reaction intermediate!

Overall, this is one of the best papers I’ve read in some time and look forward to hearing more about the crazy fluorine chemistry done by Dolbier’s group. Congrats!!! Ckellz…Signing off…



  1. Awesome write-up on so many levels! Thank you. You’re a talented story teller. Too. Funny, I was just reading up on the origin of chirality this morning.

  2. Thanks!!! I’m glad you enjoyed it, I tried searching for that article you were looking for for like an hour but to no avail. Let me know if you find it!!! And I love hearing comments on my post/discussing chem so yeah comment as much as you want!

  3. […] Prakash and Olah decided to tackle a somewhat different problem in this latest article. A few posts back, I mentioned a reagent developed by the Dolbier group known as trimethylsilyl […]

  4. […] molecules. The first is very reminiscent of two articles I’ve reviewed in the past (link 1, link 2) The author? None other than William R. Dolbier, Jr., one of the premier organofluorine chemists. […]

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 )

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