A Deeper Look at Organic Process Research & Development (OPR&D) - Part 1

Have you been reading Organic Process Research & Development?

Although I have not become a process chemist, I have learned a lot from OPR&D and have been reading it since I was a student. I would like to share with you some of the articles in the "Most Read" or "Some Items of Interest to Process R&D Chemists and Engineers", which is a compilation of articles by Dr. John Knight that I think chemists should read.

Today's article is co-authored by Mr. Masatoshi Yamada of the Pharmaceutical Research Division of Spera Pharma, Inc., members of the API team of the CMC Research Division, and members of Tohoku University.

The article describes an efficient and scalable asymmetric total synthesis of (-)-Emetine with pharmaceutical grade quality, which is the first multigram scale synthesis. The starting material is inexpensive homoveratrylamine (3,4-dimethoxyphenethylamine), which is surprising to see again at the end of the process.

According to Wikipedia, the synthesis of 3,4-dimethoxyphenethylamine is done by using vanillin or its methylated form as a starting material, condensation with acetic acid to increase carbon, hydrogenation to the double bond, and Hofmann rearrangement. Another route is the reduction from the Henry reaction using nitromethane.

The synthesis starts with the one-pot intramolecular SEAr reaction from imine formation of the terminal amine to produce 6,7-Dimethoxy-3,4-dihydroisoquinoline. This reaction is named the Pictet-Spengler reaction and uses electrophilic carbons produced by the decomposition of hexamethylenetetetramine (HMTA) under the acidic conditions used in the Duff reaction. The order of adding HMTA and TFA should be reversed if you want to proceed with the Friedel-Crafts type reaction favorably. In addition to the neutralization of TFA by the amino group of the raw material, even if HMTA reacted with TFA, there would still be a sufficient amount of TFA remaining to start heating. The remaining 0.3 equivalents of HMTA were added in three separate additions, and the final yield, including purification, was nearly 70%, so it seems safe to assume that HMTA provided the electrophilic carbon at least twice.

The asymmetric allylation of 6,7-Dimethoxy-3,4-dihydroisoquinoline shows the dark side of scale-up. The reaction went well on a scale of about 130 g. When they increased the amount of catalyst and ligand further and ran the reaction on a 2 kg scale, they encountered several issues, such as the appearance of byproducts, a decrease in enantioselectivity, and difficulty in controlling the reaction temperature. These issues were resolved by optimizing the reaction conditions and using a different catalyst and ligand.

Authors have successfully functionalized the terminal allyl group using a second-generation Grubbs catalyst for an olefin metathesis reaction with ethyl acrylate. The nitrogen functional group is useful as it can be converted into an organic salt every time it is isolated.

The benzoquinolizidine skeleton is constructed through the Michael addition of a secondary amine of tetrahydroisoquinoline to an α,β-unsaturated ketone, followed by intramolecular cyclization. In this process, the Michael addition with the easily polymerizable methyl vinyl ketone is controlled by neutralizing the hydrochloride salt of the raw material from the previous isolation process with exactly one equivalent of triethylamine. In the subsequent intramolecular cyclization, pyrrolidine is added to deprotonate the α-hydrogen of the ketone, and the reaction proceeds in a one-pot fashion.

The remaining ketone is reduced with sodium borohydride, but the reduction also competes with the intramolecular condensation of the alcohol and ester resulting from the reduction, forming a lactone. This is not surprising, especially since concentration during the post-processing process can be problematic. Therefore, tosyl alcohol is protected by adding anhydrous tosylate and then isolated as a tosylate. This is a smart approach.

After removing the tosyl group by hydrogenation, hydrolyzing the ester, and activating with pivalic acid chloride and triethylamine, condensation with homoveratrylamine, which also appeared first, forms an amide. Here, an intramolecular SEAr reaction is carried out using phosphoryl chloride to form an imine from the amide, which is famous for the Vilsmeier-Haack reaction, to produce an electrophilic carbon, resulting in the Bischler-Napieralski reaction.

Finally, the target compound is obtained by reducing the imine through the Noyori asymmetric hydrogen transfer reaction.

How was it? The last part was a bit of a rush, but I think it is good to read OPR&D in depth to learn a lot.