Cost-Effective Catalysts Could Replace Traditional Multi-Step Processes

A small-molecule APIs become increasingly complex, designing concise, practical, and economical synthetic routes for their commercial production becomes increasingly challenging. There is, consequently, growing demand for cost-effective, highly efficient catalysts that mediate atom-economical, environmentally friendly transformations that can replace traditional, non-catalytic processes that require multiple steps. A summary of selected new catalysts that show promise for pharmaceutical intermediate and API synthesis is presented in the following.

A fortuitous discovery

Silicon-containing compounds are of interest as potential drug candidates because silicon can impart greater stability and lead to compounds with improved solubilities and pharmacokinetic properties. There are two common routes to heteroarylsilanes: stoichiometric reaction of a silicon electrophile with a heteroaromatic organometallic intermediate (prepared via a Grignard reaction) or rhodium- or iridium-catalyzed C-H silylation in the presence of an excess of a hydrogen acceptor. Neither approach is practical on an industrial scale, and both suffer from limited functional group tolerance.

Researchers in the Grubbs and Stoltz groups at the California Institute of Technology fortuitously discovered an attractive alternative method based on the inexpensive base potassium tert-butoxide (KOtBu) (1). When investigating a reaction to convert biomass to chemicals via the breakage of C-O bonds using an iron catalyst, KOtBu, and a hydrosilane as a hydride equivalent, they observed that silylated heteroaromatics were formed in minor quantities in a control reaction using only KOtBu and the hydrosilane. In addition, they noted that the yield of silylated products increased as the reaction temperature decreased.

The cross-dehydrogenative heteroaromatic C-H silylation reaction with hydrosilanes was found to proceed best using a catalytic amount of KOtBu (1-20 mol%) in the absence of an acceptor, with only hydrogen gas as the byproduct. The fact that a common base catalyzes this important reaction was quite surprising, and thus the researchers went to great lengths to confirm that the KOtBu was indeed the catalyst and not some unknown impurity.

The reaction is believed to proceed via a radical mechanism that is completely different from the mechanisms observed with transition-metal catalysts. Its scope is also very broad. Indoles with a variety of substituents on the nitrogen and various positions on the arene ring and many different electron-neutral and electron-rich N-, O-, and S-containing heterocyclic compounds are suitable substrates. Carbonyl groups are generally not tolerated except when protected as acetals. Some groups (bromide, iodide, cyano, and nitro substituents) do hinder the reaction. Interestingly, fluoride, chloride, trifluoromethyl, epoxide, N-alkyl aziridine, pyridine, and tertiary amine and phosphine groups do not.

Notably, even simple arenes serve as good substrates. Substitution affects the regioselectivity; however, ortho-substitution occurs with anisole, while directing-group-free C(sp3)-H silylation leads to silylated benzyl derivatives with toluene and similar compounds. The researchers are also exploring the reaction of non-aromatic compounds (e.g., aliphatic compounds, alkenes, and alkynes).

The scientists have also shown that the cross-dehydrogenative heteroaromatic C-H silylation reaction is scalable. When run neat using N-methyl indole and 1.5 equivalents of triethylsilane at 45 °C on a 100-gm scale, desired C2 silylated product was obtained in 76% yield with a greater than 20:1 regioselectivity after simple filtration and distillation.

The researchers also demonstrated the utility of the reaction by silylating the antihistamine thenalidine and the antiplatelet drug ticlopidine in 58-68% yield with high chemo- and regioselectivity, indicating the applicability of the transformation for late-stage modification of pharmaceutically relevant compounds.

Cooperative catalyst system

Sites of unsaturation (alkenes in particular) are widely present in pharmaceutical compounds and their introduction can be achieved using a variety of methods, such as the elimination of alcohols, halides or other leaving groups, the reduction of acetylenes, and Wittig and Diels-Alder reactions. Dehydrogenation of alkanes is highly attractive because it allows for the flexible placement of alkenes. In addition, when performed under acceptor-less or terminal oxidant-free conditions, only hydrogen is generated as the byproduct. It is, however, a highly endothermic reaction traditionally mediated by expensive noble-metal catalysts at high temperature. Enzyme-based systems do proceed at room temperature, but require the use of a full equivalent of a terminal oxidant, and thus hydrogen recovery is not possible.

Scientists at Princeton University have overcome these issues with the development of a reaction that is noble-metal-free, does not require the use of a terminal oxidant (and thus generates hydrogen), and proceeds at mild temperatures (2). Their approach is based on key aspects of the iron-based desaturase system, which catalyzes dehydrogenation reactions via step-wise hydrogen atom removal. The first hydrogen atom transfer (HAT) occurs from the alkane substrate to a high-valent oxo species, which is much more difficult than the second, which can occur via HAT or electron transfer followed by proton transfer.

In their new approach, the researchers use cooperative catalysis with a dual-catalyst system to achieve the two steps. The polyoxometalate tetra-n-butylammonium decatungstate (TBADT) is easy to synthesize and has been shown to mediate HAT reactions of unactivated alkanes under near-UV light irradiation. For the second catalyst, the researchers were interested in cobaloximes because they are known to catalyze radical reactions including the reversible formation of alkenes and cobalt hydrides and have been shown to mediate water reduction under photo-irradiation conditions. Cobaloxime pyridine chloride (COPC) was selected because it is also readily synthesized.

When used together (1.5 mol% TBADT and 0.75 mol% COPC) under irradiation at 323 nm, which is the absorbance maximum of TBADT, cyclooctane was converted to cyclooctene in 23% yield. The researchers also showed that cyclopentane was converted to cyclopentene and cyclopentadiene, while cyclohexane was dehydrogenated to cyclohexane rather than benzene, which commonly occurs under noble-metal catalysis.

The reactivity of ethyl isovalerate was also interesting. Rather than produce the thermodynamically most stable product, which is generally the case with noble-metal catalysts, the dehydrogenation proceeded to form the skipped-enone product. The researchers are also exploring the tolerance for functional groups other than esters.

While the initial results of the cooperative catalysis approach with simple alkanes are not as efficient as reactions with noble-metal catalysts, the researchers believe there is potential to improve the reaction. In fact, when secondary alcohols are used as substrates, the yields increase (i.e., 83% for 1-phenylethanol).

New approaches to C-H activation

Researchers at the Scripps Research Institute (TSRI) developed a catalyst that mediates C-H activation in the synthesis of heterocycles (3). Using an N-methoxy amide group as both a directing group and an anionic ligand, the PdX2 (X = ArCONOMe) catalyst is generated in situ from a Pd(0) source using air as the oxidant. With the PdX2 species localized near the target C-H bond, the natural preference of the metal to coordinate to the heteroatom is avoided. The catalyst has been used to synthesize a number of different compound types, including furans, benzofurans, benzothiophenes, indoles, pyrroles, thiazoles, pyrazoles, imidazoles, pyridines, quinolines, pyrazines, pyrimidines, pyrazoles, and thiazoles.

Catalysts that mediate other types of C-H activation reactions have also been developed by TSRI researchers. In one case, sequential diarylation is achieved to afford non-natural β -Ar-β-Ar’-β-α-amino acids with diastereomeric ratios of greater than 20:1 (4). These compounds are building blocks for peptide drugs with structural diversity that may make them resistant to enzyme degradation. In the second reaction, palladium-catalyzed enantioselective C-H iodination is used to achieve the kinetic resolution of chiral amines (5). Notably, the reaction proceeds at room temperature and involves only inexpensive and commercially available reagents.

Coupling of heteroatom-substituted olefins

Carbon-carbon coupling reactions have become, in recent years, fundamental transformations for the commercial production of small-molecule APIs. Researchers at TSRI recently developed a new coupling reaction for the preparation of highly substituted, uniquely functionalized compounds catalyzed by an inexpensive iron complex (6). The reaction is also attractive because it is performed open to the air at room temperature.

While typically run in ethanol, the researchers showed that reactions run in vodka, gin, whiskey, tequila, beer, and wine also afforded the desired coupling products. In addition, they demonstrated the broad scope of the reaction by synthesizing more than 60 different compounds, most of which were new chemical entities not previously reported.

References

1. A.A. Toutov, et al., Nature, 518, 80-84 (2015).
2. J. G. West, D. Huang, and E.J. Sorensen, Nature Communications 6, 10093 (2015).
3. J-Q Yu, et al., Nature, 515, 389-393 (2014).
4. J-Q Yu, et al., Science, 343 (6176), 1216-1220 (2014).
5. J-Q. Yu, et al., Science, 346 (6208), 451-455 (2014).
6. P.S. Baran, et al., Nature, 516, 343-348 (2014).