Using Olefin Metathesis To Produce Macrocyclic Products

Molecules containing twelve or more atoms within at least one large ring are called macrocyclic compounds or macrocycles. At the beginning of the 20th century, the synthesis or even existence of macrocycles was questioned by many chemists.

One of the pioneers, a Croatian-Swiss scientist Leopold Ružička[1] in his Nobel lecture stated: “I was hindered (…) by the general prejudice, shared by myself, against the probability of the existence of a 15- and a 17-membered ring”. But despite this bias, he is today famous for proving in 1923 that muscone and civetone—valuable perfume ingredients—obtained from the musk deer (Moschus moschiferus)[2] and the civet cat[3] (Viverra civetta) are in fact macrocyclic compounds.

Actually, these were the very first natural products shown to have macrocyclic rings in their structure. A short time later Ružička accomplished a groundbreaking, although low yielding, first chemical synthesis of such macrocycles by thermolysis of cerium or thorium dicarboxylic acids salts.

Fig. 1. Siberian musk deer and African civet cat. Credit: “Siberian Musk Deer” by ErikAdamsson licensed under CC0. “Civettictis civetta” by Николай Усик / http://paradoxusik.livejournal.com/ licensed under CC-BY-SA 3.0. Both images via WikiCommons.

In recent years an enormous progress in the synthesis of macrocyclic compounds has been made, however, macrocyclizations are still considered as challenging transformations. According to the commonly accepted paradigm (adopted from Jacobson-Stockmayer polymer theory), the balance between the desired macrocycle and undesired oligomers and polymers is ultimately controlled by the concentration, and high concentrations favor intermolecular reactions, while high dilutions favor the macrocyclization.

All known synthetic methods, including the Nobel-winning olefin metathesis reaction[1] suffer from this limitation. Since the first reports in 1980, where chemists from International Flavors & Fragrances, Inc.[2] claimed production of macrocyclic musks by so-called ring-closing olefin metathesis(RCM) till the most recently published examples, preparation of medium and large rings by RCM was always performed in very diluted solutions, under so-called high dilution conditions. When the concentration is higher than recommended (ca. 5 mmol/dm³), a large amount of polymeric products is formed, making the yield of the desired macrocyclic product low. This makes, especially from the industrial perspective, large-scale production of macrocycles problematic, due to the environmental and economic costs of purchasing, storing, transferring, and then at the end separating and disposing of large volumes of organic solvents.

The goal of a research conducted by our team[3] was to utilize the inherent reversibility of olefin metathesis to produce macrocyclic products relevant to the flavor and fragrance industry (F+F) at concentrations much higher than normally used for similar RCM macrocyclizations. We assumed that under carefully selected conditions oligomers formed during a process can be effectively recycled by a catalyst (“backbit”) to yield the expected macrocyclic product in high yield and selectivity. Once the macrocycle was formed, it was planned to distill it out of the reaction mixture under vacuum.[4]

Fig. 2. Planned preparation of macrocycles at high concentration. Credit: Karol Grela

A conceptually simple process proved to be extremely difficult to carry out. A series of initial failures demonstrated distinctly that the task was more challenging than we naïvly assumed. The results obtained varied from no reaction at all, through a complete burn of the substrate during heating.

In other cases, even if some macrocycle was distilled out, its formation was disappointingly nonselective, leading to complicated mixtures composed of dozens of products. We blamed the insufficient stability of commercial metathesis catalysts, that under such conditions led to severe isomerization of the substrate and products (mostly via the C-C double bond shift) and to other processes.

Fig. 3. Three catalysts exhibiting the best selectivity in macrocyclic musk production. Credit: Karol Grela

The failure of the above experiments explains why, despite all conceptual elements, like the analogy to famous Carothers’ acid-catalyzed depolymerization were known, there is no single example of a metathesis-based synthesis of macrocycles at very high concentrations with documented high selectivity.

Evidently, the key was to find an adequate catalyst, which merges high activity with durability. Luckily, having a rich in-house collection of various proprietary olefin metathesis catalysts, we screened a number of them. We didn’t have to wait long for reward, and with three catalysts originally tailored to take up the challenge in the self-metathesis of α-olefins,[1]  were able, for the first time, to obtain a macrocycle with nearly perfect selectivity. This method was quite general, and we were able to produce various cyclic lactones, ethers, and ketones of the ring size 7-19 in good yields and with high selectivity at a concentration of 0.25 mol/kg. That means we used 40 times less solvent than recommended.

Fig. 4. Two musk scents made from a renewable source using the HC-RCM method. Credit: Karol Grela

As this new high-concentration RCM method (HC-RCM) was found to work very well for non-terminal dienes, so we used oleic acid from a renewable plant oil as a substrate to obtain two musks: an unsaturated analog of elegant 16-membered Firmenich Exaltolide® scent and the African civet derived (E/Z)-civetone.

Contrary to traditional solution-based RCM which uses often hazardous aromatic or chlorinated solvents, in the new HC-RCM method we applied inexpensive and nonvolatile diluents, such as common paraffin oil or polyalfaolefin (PAO) based synthetic motor oils.

Fig. 6. Comparison of classical RCM “in solution” with the HC-RCM method. Credit: Karol Grela

In the final stage of the present study, we tested HC-RCM at an even higher concentration of 33-90 wt % (so 280–380 times higher than those used typically today) to show that the newly developed method has a real prospect for industrial applications. However, to transform this invention[1] into industrially matured technology, further optimization related to chemical engineering and deeper catalysis studies is definitely needed.

These findings are described in the article entitled At Long Last: Olefin Metathesis Macrocyclization at High Concentration, recently published in the Journal of the American Chemical Society. This work was conducted by Adrian Sytniczuk, Michał Dąbrowski, Łukasz Banach, Mateusz Urban, Sylwia Czarnocka-Śniadała, Mariusz Milewski, Anna Kajetanowicz, and Karol Grela from the University of Warsaw

References:

  1. https://en.wikipedia.org/wiki/Leopold_Ru%C5%BEi%C4%8Dka
  2. https://en.wikipedia.org/wiki/Musk_deer
  3. https://en.wikipedia.org/wiki/Civet
  4. https://en.wikipedia.org/wiki/Olefin_metathesis
  5. https://en.wikipedia.org/wiki/International_Flavors_%26_Fragrances
  6. http://www.karolgrela.eu/
  7. A. Sytniczuk, M. Dąbrowski, Ł. Banach, M. Urban, S. Czarnocka-Śniadała, M. Milewski, A. Kajetanowicz, and K. Grela. At Long Last: Olefin Metathesis Macrocyclization at High Concentration, Journal of the American Chemical Society Article ASAP, DOI: 10.1021/jacs.8b04820
  8. a) Patent EP3294747, 2018. b) WO2018100515, 2018, c) Chołuj, A.; Zielinski, A.; Grela, K.; Chmielewski, M. J. Metathesis@MOF: Simple and Robust Immobilization of Olefin Metathesis Catalysts inside (Al)MIL-101-NH2. ACS Catal. 2016, 6, 6343– 6349, DOI: 10.1021/acscatal.6b01048.
  9. Patent Appl. P.421462 (International Patent Appl. PCT/IB2018/051566)