Nov 4, 2009

Iron Catalysis

Recently, I have been having trouble with performing a Wittig reaction on a particularly unstable aldehyde that had been giving me all sorts of problems. While a literature procedure exists to synthesise and isolate the said compound the amount of material that remains after workup is volatile and decomposes readily whether under argon or even stored in the freezer.

But, just the other day MSG gave me an interesting paper. It shows the use of an Iron catalyst (FeCl3) to form C(sp3)-C(sp3) bonds via an alcohol and an alkene. This would be extremely handy for me if it works as I am able to make the alcohol realatively easily and it is stable. However, while it won't give me my product in one step it will provide an alternative method of formation.

Anyway, seeing as this is a chemistry blog I thought I would try my hand at going through this paper while I read through it.

Basically, the authors intended to couple an alcohol to an alkene in the following manner.

The resultant C-C bond is particularly handy as the alcohol functional group remains and provides an excellent handle for further reactions to the center as well as having the potential to be removed at a later step in further synthesis.

From previous investigations the authors had found that Lewis acids in conjunction with a metal catalyst would achieve the desired coupling. However, the Lewis acid was used in excess amounts and more to the point you are using expensive metals, which the Boss has already told us is best avoided especially on a tight Aussie chemist budget (unless you are my old boss who is a legend in academia and gets some nice grant money).

Fortunately for the authors they found an alternative reagent that provided not only a nice metal catalyst but, lo and behold, it's also a Lewis acid! ZOMFG its FeCl3!

Initially, the authors investigated a test system with several different solvents in order to gauge any solvent effects. Their initial substrates remained constant. 3-Phenylpropanol (their alchol) and 1,1-diphenylethylene (the alkene). Their results showed that dichloroethane (DCE) gave the best results although I find it strange that they didn't try DCM which is a much more common solvent in labs. :S Nevertheless it was deemed "essential" to the reaction and was shown to promote the reaction when used as an additive in other solvent systems.

A few other iron salts were tried as well as a varied amounts of catalyst along with some promoting ligands (e.g. TMEDA, NEt3 and DACH) but to no avail and it was seen that FeCl3 in these proportions was goldilocks.

A number of reations were performed on various alcohols and alkenes and all showed that the reaction did work quite well. They explained the proposed mechanism to work in the following manner.

This is a handy reaction to have if you cant form a stable aldehyde allow you to perform a Wittig reaction. Also, FeCl3 seems like a fairly inexpensive reagent especially when you can use it in catalytic amounts. I will have to give this a shot some time soon.

Aside from the second word in the article being a spelling mistake (ze Germans are slipping), the paper illustrates quite well this new application of iron to develop C-C bonds without the use of expensive metals etc. There is also another similar paper on the use of metals in a similar manner that can be found here.

Zhang, S.-Y. et al, Angew. Chem. Int. Ed. 2009, 48, ASAP


  1. DCE is on level 1 on my 'exotic solvents' list, along with DMPU and HMPA, but it's still a pretty common solvent. I didn't look at their solvent screen too closely, but vaguely from memory they did need to heat the reaction, and DCM is too low boiling for that.

  2. Good point mate, although it always makes me wonder at times like these whether they heated it up to a certain temperature because it had to be that high or just because it could be that high. It is understandable that it may allow the reaction to proceed faster but sometimes you just never know. That is the beauty of science.

  3. Tetrahedron Letters
    Volume 50, Issue 45, 11 November 2009, Pages 6110-6111

    Bear this in mind though...

  4. @Artemis: ah I see, - duly noted

  5. Very interesting article Artemis. I will have to investigate this further. However, I don't think this will fuss me too much. Either it will work due to the contaminated Fe anyway or it won't for me it is just another possibility that I haven't ruled out yet.
    Keep reading and I may post the results.