There’s nothing more frustrating to a synthetic chemist than a reversible reaction. We incur great expense to prepare reactive reagents (MeMgBr, LiAlH4, Br2, PhI(OAc)2, etc.) that do their thing and then just stand by and watch after they’re done. I think this is in part why RCM was so quickly adopted by the synthetic community. You drive off ethylene and the reaction is irreversible.
Things are not nearly so clean-cut with cross metathesis (CM). Even in the simple case of the CM between 1-hexene and 1-octene, where you have the luxury of wafting away ethylene, you’ll still end up with a 1:2:1 ratio of C10:C12:C14 products. It’s hard to get excited about a reaction with a 50% theoretical yield and an annoying separation waiting at the end. Fortunately there are some guidelines for CM that can make your time in the lab much more efficient…
The seminal paper in this area came out of the Grubbs group (J. Am. Chem. Soc., 2003, 125, 11360), describing a method to think about cross metathesis reactions.
Type I – Fast homodimerization; the homodimerization product is reactive
Type II – Slow homodimerization; the homodimerization product is modestly reactive
Type III – No homodimerization
Type IV – Metathesis spectator
…olefins of type I = non-selective (equilibrium product mixture)
…olefins of the same type (non-Type I) = non-selective (kinetic product mixture)
…olefins of different types = selective CM
Everything you need to know comes down to the Olefin Types and The Rules, but they both have their share of subtleties.
- The Olefin Type is determined by the sterics and electronics of the olefin in question. At the boundaries, a simple terminal olefin is Type I, and a hindered electron-poor olefin is Type IV. For more specific (but still generic) categorizations, see the Grubbs 2003 paper.
- However, the Olefin Type is also dependent on what kind of catalyst you use. For example, 1,1-disubstituted olefins are considered type IV with the 1st gen. Grubbs catalyst, but with the 2nd gen. version, which can make trisubstituted olefins, they are Type III. Therefore, catalyst selection plays a key role.
After you’ve got that down, there are a few tricks to keep in mind for optimization. None of this is rocket science…
- Use an excess of one substrate (especially useful if it’s volatile and cheap, i.e. isobutylene).
- Remove the desired product (or the undesired byproduct) during the course of the reaction. This works if it’s volatile (ethylene), or if you know that the product is insoluble.
- If one of the CM partners is terminal, a slow addition can help minimize self-metathesis.
- Proper protecting group strategy can change the olefin category (sterics impact the Olefin Type).
- Don’t be afraid to run reactions neat. This is especially true if your CM partner is a Type III olefin like isobutylene, where you can use a large excess without worrying about self-metathesis.
A representative example (also from the Grubbs 2003 paper):
When a 1:1 ratio of the two olefins is used, they see an 80% reaction yield. Considering all the possibilities, that’s pretty good! Let’s think it through. The hexenyl acetate (Type I) is free to react with itself, but the product 1,2-dibustituted olefin can also re-enter the catalytic cycle, so it doesn’t matter. 2-Bromostyrene (Type II) is slow to react with itself, and faster to react with the hexenyl acetate (or the dimer of hexenyl acetate). The desired product is the most hindered and therefore slowest olefin in the system to react, so the net result is that essentially all of the material is siphoned to the product. Some of the 2-bromostyrene homocouples, accounting for the mass balance. When they use 3 equiv. of 2-bromostyrene, a little more homodimerization doesn’t matter and the yield is 98%!