Olefin Metathesis - Historical Overview

Historical Overview

Known chemistry prior to the advent of olefin metathesis was introduced by Karl Ziegler in the 1950s who as part of ongoing work in what would later become known as Ziegler-Natta catalysis studied ethylene polymerization, which on addition of certain metals resulted in 1-butene instead of a saturated long-chain hydrocarbon (see nickel effect).

In 1960 a Du Pont research group polymerized norbornene to polynorbornene using lithium aluminum tetraheptyl and titanium tetrachloride (a patent by this company on this topic dates back to 1955),

a reaction then classified as a so-called coordination polymerization. According to the then proposed reaction mechanism a RTiX titanium intermediate first coordinates to the double bond in a pi complex. The second step then is a concerted SNi reaction breaking a CC bond and forming a new alkylidene-titanium bond, the process then repeats itself with a second monomer:

Only much later the polynorbornene was going to be produced through ring opening metathesis polymerisation. Giulio Natta in 1964 also observed the formation of an unsaturated polymer when polymerizing cyclopentene with tungsten and molybdenum halides.

In a third development leading up to olefin metathesis researchers at Phillips Petroleum Company in 1964 described olefin disproportionation with catalysts molybdenum hexacarbonyl, tungsten hexacarbonyl, and molybdenum oxide supported on alumina for example converting propylene to an equal mixture of ethylene and 2-butene for which they proposed a reaction mechanism involving a cyclobutane (they called it a quasicyclobutane) - metal complex:

This particular mechanism is symmetry forbidden based on the Woodward-Hoffmann rules first formulated two years earlier. Cyclobutanes have also never been identified in metathesis reactions another reason why it was quickly abandoned.

Then in 1967 researchers at the Goodyear Tire and Rubber Company described a novel catalyst system for the metathesis of 2-pentene based on tungsten hexachloride, ethanol the organoaluminum compound EtAlMe₂ and also proposed a name for this reaction type: olefin metathesis. Formerly the reaction had been called "olefin disproportionation."

In this reaction 2-pentene forms a rapid (a matter of seconds) chemical equilibrium with 2-butene and 3-hexene. No double bond migrations are observed, the reaction can be started with the butene and hexene as well and the reaction can be stopped by addition of methanol.

The Goodyear group demonstrated that the reaction of regular 2-butene with its all-deuterated isotopologue yielded C₄H₄D₄ with deuterium evenly distributed. In this way they were able to differentiate between a transalkylidenation mechanism and a transalkylation mechanism (ruled out):

In 1971 Chauvin proposed a 4-membered metallacycle intermediate to explain the statistical distribution of products found in certain metathesis reactions . This mechanism is today considered the actual mechanism taking place in olefin metathesis.

The active catalyst, a metallocarbene., was discovered by in 1964 by E. O. Fischer. Chauvins experimental evidence was based on the reaction of cyclopentene and 2-pentene with the homogeneous catalyst tungsten(VI) oxytetrachloride and tetrabutyltin:

The three principal products C9, C10 and C11 are found in a 1:2:1 regardless of conversion. the same ratio is found with the higher oligomers. Chauvin also explained how the carbene forms in the first place: by alpha-hydride elimination from a carbon metal single bond. For example propylene (C3) forms in a reaction of 2-butene (C4) with tungsten hexachloride and tetramethyltin (C1).

In the same year Pettit who synthesised cyclobutadiene a few years earlier independently came up with a competing mechanism. It consisted of a tetramethylene intermediate with sp3 hybridized carbon atoms linked to a central metal atom with multiple three-center two-electron bonds.

Experimental support offered by Pettit for this mechanism was based on an observed reaction inhibition by carbon monoxide in certain metathesis reactions of 4-nonene with a tungsten metal carbonyl

Robert H. Grubbs got involved in metathesis in 1972 and also proposed a metallacycle intermediate but one with 4 carbon atoms in the ring. The group he worked in reacted 1,4-dilithiobutane with tungsten hexachloride in an attempt to directly produce a cyclomethylenemetallacycle producing an intermediate, which yielded products identical with those produced by the intermediate in the olefin metathesis reaction. This mechanism is pairwise:

In 1973 Grubbs found further evidence for this mechanism by isolating one such metallacycle not with tungsten but with platinum by reaction of the dilithiobutane with cis-bis(triphenylphosphine)dichloroplatinum(II)

In 1975 Katz also arrived at a metallacyclobutane intermediate consistent with the one proposed by Chauvin He reacted a mixture of cyclooctene, 2-butene and 4-octene with a molybdenum catalyst and observed that the unsymmetrical C14 hydrocarbon reaction product is present right from the start at low conversion.

In any of the pairwise mechanisms with olefin pairing as rate-determining step this compound, a secondary reaction product of C12 with C6, would form well after formation of the two primary reaction products C12 and C16.

In 1974 Casey was the first to implement carbenes into the metathesis reaction mechanism:

Grubbs in 1976 provided evidence against his own updated pairwise mechanism:

with a 5-membered cycle in another round of isotope labeling studies in favor of the 4-membered cycle Chauvin mechanism:

In this reaction the ethylene product distribution (d₄,d₂,d₀) at low conversion was found to be consistent with the carbene mechanism. On the other hand Grubbs did not rule out the possibility of a tetramethylene intermediate.

The first practical metathesis system was introduced in 1978 by Tebbe based on the (what later became known as the) Tebbe reagent. In a model reaction isotopically labeled carbon atoms in isobutene and methylenecyclohexane switched places:

The Grubbs group then isolated the proposed metallacyclobutane intermediate in 1980 also with this reagent together with 3-methyl-1-butene:

They isolated a similar compound in the total synthesis of capnellene in 1986:

In that same year the Grubbs group proved that metathesis polymerization of norbornene by Tebbe's reagent is a living polymerization system and a year later Grubbs and Schrock co-published an article describing living polymerization with a tungsten carbene complex While Schrock focussed his research on tungsten and molybdenum catalysts for olefin metathesis, Grubbs started the development of catalysts based on ruthenium, which proved to be less sensitive to oxygen and water and therefore more functional group tolerant.