Benot de Carn-Carnavalet, Alexis Archambeau, Christophe Meyer, Janine Cossy, Benot Follas, Jean-Louis Brayer, and Jean-Pierre Demoute Copper-Free Sonogashira Coupling of Cyclopropyl Iodides with Terminal Alkynes Organic Letters ASAP January 26, 2011
So I have a confession. I’m not only a sucker for fluorinated molecules, I absolutely love strained hydrocarbons. They are just freaking cool. Like cubane. I mean how can you not like them? They are potent explosives, they have unusual properties that defy conventional reactivity, and they are a challenge for an organic chemistry to make. In short they have limited application but they are equivalent to getting all the golden skulltulas in Zelda: Orcarina of Tme (aka ownage). And certainly, the most strained cyclic system I can think off has to have a cyclopropane in it. That’s really the most distorted you can get as far as organic chemistry goes. In part because of my undergrad research, I have a fond place in my heart for the cyclopropane motif. So its no surprise this article, which enables cyclopropanes to be substituted with ease, caught my attention. Plus it invovles palladium catalysis! One of my more recent specialties!
So it starts out as most methodology articles do: “Why is the product I’m making so awesome”. I was impressed with this part. Cossy (who I will refer to as the principle author, although the work has two with Christophe Meyer, a member of her lab, being the second) really went all out with info about cyclopropanes. I didn’t realize how common it was to put a cyclopropane in a drug. She cites that they are relatively easy to prepare due to the large number of methodology known to produce them, to which I agree with some reservation. A lot of those methods are very substrate specific or involve highly reactive intermediates which are not compatible with every functional group under the sun (think carbenes
). She goes on to say that there have been a variety of methods to substitute these cyclopropanes once they are formed (which rely on the formation of some sort of organometallic cyclopropane or palladium cross coupling reactions). However, Cossy then states that are few electrophilic cyclopropanes being used in these sorts of coupling reactions. Following a report
from Charette group
in the late 90s, Cossy investigated whether such cyclopropanes could be functionalized using some sort of cross-coupling technique. Cossy picked alkynyl functionalization as their challenge because it looks like alkynylcyclopropanes have some applications from a pharma standpoint and they are somewhat of a pain to synthesize traditionally.
So they selected the model reaction above to test whether they could even get one of the standards of organometallic chemistry the Sonogashira
reaction to go. Unfortunately for them, the normal conditions just plain failed hard. I say hard, because their starting material needed to be synthesized
(3 steps in fact, I back tracked the lit for that one). So I you invest all that time in synthesizing a compound just to test your reaction and find oh no FAIL! it hurts a looooot more. So kudos to the Cossy group for not giving up right there and pursuing it. Interestingly, Buckwald played a role in their next move. Since he’s been using a lot of X-Phos lately with success with difficult substrates
, Cossy’s group tossed that into the mix and that did the trick. They got 100% conversion to the desired product in 93% isolated yield. They screened other ligands too, but none worked as good as trusty X-Phos (I must admit, that’s a kind of badass name as far as ligands go). After screening solvent, temperature, and base they were ready to go (I’ve put the optimized conditions above in the figure). It’s substrate scope time.
The scope was impressive. They had all sorts of acetylene coupling partners. Some exotic ones too like a m-fluorophenylacetylene, a TIPS (Triisopropylsilyl) protected alkyne, 1-pentyne and ethoxy acetal. Quite the range! And the yields were excellent ranging from 81% to 97% with minimal alteration to the optimal conditions (one intially gave low yield 44%). They next altered the cyclopropane coupling partners by altering the stereochemistry as well as the substitution pattern. The majority of reactions went in good to excellent yield (72-97%). Moreover they really did put alot of effort into their reactions. They didn’t just screen one generic acetylene with a test cyclopropane. They would test 2-5 alkynes with each cyclopropane in question. So again, kudos to the Cossy group. However, here’s where I encountered an issue. They relied a lot on using a cyclopropanol, not a simply cyclopropane. So they (and I) wondered if that was playing some role in the mechanism. So they PMB (p-methoxybenzyl) protected it, making the argument that it’s no longer a free hydroxyl. Granted this is true, but if you think about it, the oxygen lone pairs are still untouched; they can still participate in the reaction. Now you could come back to me and make arguments about steric restrictions (which are perfectly legitimate). Still, I would of preferred to see not a protected but a functional group free cyclopropyl iodide as a substrate. Despite that small qualm, the next line makes up for it: The reaction “proceeded stereoselectively and with retention of configuration”. That right there is powerful chemistry: high yields (with a class of substrates that are quite strained) and stereospecificity.
Now they could of just stopped there. But they didn’t, and I think that’s what really set this paper apart from others I was reading. They next investigated derivatives cyclopropyl amides and esters to see if these types of compounds were also compatible their coupling strategy. After a little bit of re-optimization (new conditions are above), they found that yes, indeed it was. Not only that but it was relatively broad and both the alkyne substitution pattern and the substitution on the nitrogen of the amide could be modulated without compromising yield. And esters are esters…so they only did one lol. Interestingly, it was extremely important to add the alkyne slow to the cyclopropane solution. Apparently they observed oligomerization mediated by the amide when added directly so they extended the addition time to 2 hours
. Adding over two hours isn’t that bad at all, especially when you have the awesomeness that is an addition funnel
or better yet a syringe pump
. Overall an excellent job on this work, it’s really well done and thorough. Great article Cossy and co-workers!!!