Isomerizing metathesis

Olefin metathesis reactions are applied in almost any field of chemical synthesis. The synthesis of ,-functionalized monomers via metathesis reaction was already described in section 1.1 of this thesis. However, in these transformations stoichiometric amounts of less valuable coupling products are formed. In addition, olefin metathesis is often an equilibrium reaction and only 50 % conversion can be attained unless the product can be removed from the reaction mixture selectively.

Figure 1-22: Isomerizing cross-metathesis between methyl oleate and a functionalized olefinic substrate.

Thus, a one-pot transformation starting from an internal olefin (e.g. methyl oleate) with the first step being an isomerization of the internal double bond to the terminal position and the second step being an olefin metathesis reaction is desirable. This metathesis can be either cross-metathesis between the terminal unsaturated fatty acid substrate and a functionalized olefinic substrate (Figure 1-22) or self-metathesis between two of the terminal unsaturated fatty acid substrates (Figure 1-23). The challenge of both reactions is to choose appropriate isomerization and metathesis catalysts which selectively catalyze the desired reactions. Both reactions generate a linear long-chain unsaturated ,-diester as the desired product and stoichiometric amounts of a coupling product. In the self-metathesis reaction, ethylene forms, which can be removed from the

reaction mixture selectively and thus the equilibrium reaction may be shifted to conversions > 50 %. In the cross-metathesis reaction, the coupling product which is formed depends on the choice of the functionalized olefinic substrate used as reaction partner for the fatty acid substrate. By appropriate choice of this substrate the desired linear long-chain unsaturated ,-diester which is formed in this metathesis reaction is less reactive than the starting materials.¹¹⁶ In addition coupling products that can be removed from the reaction mixture selectively are desirable, to shift the equilibrium to conversions > 50 % (vide infra).

Figure 1-23: Isomerizing self-metathesis of methyl oleate.

Some approaches in combining isomerization and self-metathesis were already presented. Porri and-co-workers reported the generation of a mixture of linear butenes, pentenes, hexenes and olefins higher than C7, from 1-pentene by isomerizing self-metathesis using an [IrCl(coe)2]2 catalyst that was activated with AgOTf (2 equiv.) in the presence of trifluoromethanesulfonic acid (14 equiv.).¹¹⁷ Grubbs and co-workers used a similar system of 4 mol-% [IrCl(coe)2]2 activated with AgOTf (4 equiv.) in isomerizing self-metathesis of methyl oleate. After hydrogenation of the reaction mixture, a mixture of linear C9-C26 alkanes, linear C8-C28 monoesters and linear C11-C26 ,-diesters was obtained.¹¹⁸ Consorti and co-workers used a biphasic systems of toluene and an ionic liquid to combine a Ruthenium-hydride isomerization catalyst and a modified Grubbs-Hoveyda 2^nd generation metathesis catalyst to convert trans-3-hexene into a mixture of linear C4-C17 olefins.¹¹⁹ More recently, Gooßen and co-workers combined a Palladium isomerization catalyst (0.6 mol-%) and a NHC-indenylidene Ruthenium metathesis catalyst (0.5 mol-%) to convert oleic acid into an equilibrium mixture of C8-C32 olefins, C13-C25 monocarboxylates and C13-C22 ,-dicarboxylates.¹²⁰ The same group reported the isomerizing cross-metathesis between oleic acid and trans-3-hexenedioic acid (2 equiv.) with a similar catalysts system (2.5 mol-% Pd, 5.0 mol-% Ru) yielding a mixture

of olefins, monocarboxylates and ,-dicarboxylates. By using a high boiling solvent, the olefin fraction was continuously removed by distillation during the isomerizing metathesis, which resulted in a shift of the equilibrium towards the dicarboxylic acid fraction. It is important to note that: a) the ratio of oleic acid / trans-3-hexenedioic acid influences the relative chain-length distribution and b) successful isomerizing metathesis was not observed with acrylic acid or maleic acid as the coupling partner. Nevertheless, all these approaches do not result in the formation of a single compound in high selectivity, but in mixtures of compounds with different chain lengths. Moreover, when methyl oleate or oleic acid are used as the substrate, mixtures of olefins, monocarboxylates and ,-dicarboxylates are formed.

Schrock and co-workers recently presented the only approach that yielded a single product in high selectivity from an isomerizing metathesis reaction.¹²¹ By combination of a Ruthenium based ‘alkene zipper’ catalyst that selectively isomerizes trans-olefins into a thermodynamic mixture of trans-olefins and a Tungsten based Z-selective metathesis catalyst, they were able to transform trans-3-hexene into cis-5-decene and ethylene in a selectivity of up to 64 % (Figure 1-24). In a typical experiment, they used 0.16 mol-% of the Tungsten metathesis catalyst and 0.05 mol-% of the Ruthenium isomerization catalyst in refluxing methylene chloride (40 °C), at a reaction time of 6 hours. Conversion is 12.8 % and thus relatively low, however the selectivity of 64 % is remarkable (this corresponds to an average TOF to the desired product of 11.2 h^-1).

Figure 1-24: Isomerizing metathesis of trans-3-hexene to cis-5-decene.

A general problem in isomerizing metathesis reactions are further reaction cycles of the products initially formed (secondary metathesis). In the example of Schrock and co-workers this translates into further isomerizing metathesis of cis-5-decene. They

overcame this issue by application of a Z-selective metathesis catalyst that produces internal cis-olefins and an isomerization catalyst that exclusively isomerizes trans-olefins. Thus a second isomerizing metathesis cycle is not possible as the initial product cis-5-decene is unreactive towards both catalysts. Aiming at the isomerizing metathesis of unsaturated fatty acids, secondary metathesis of the desired unsaturated ,-dicarboxylate products is less problematic, as formation of terminal olefins by double bond isomerization is not possible from these compounds. However, a metathesis catalyst is needed that a) tolerates the carboxylate groups of the substrate and b) is even more selective to terminal olefins than the Tungsten based catalyst used by Schrock and co-workers. The latter is due to the concentration of terminal olefins, which is much lower for a longer chain substrate (e.g. < 0.2 % for methyl oleate) versus the trans-hexenes (ca.

1 %) used by the authors.