In today’s chemical industry, catalytic processes are of paramount importance. In particular, the application of asymmetric organocatalysts is receiving increased attention . This is illustrated by the fact that in 2021 the Nobel Prize in Chemistry was awarded for the discovery of asymmetric organocatalysis . The use of organocatalysts has been a major breakthrough in the realization of enantioselective transformations. Stereoselective synthesis is essential in the pharmaceutical industry, as the development of drugs often requires the production of enantiomerically pure chiral compounds .

The application of organocatalysts is well-established in several organic reactions, including but not limited to aldol reactions , Michael additions , Mannich reactions , aza-Henry reactions , and Diels–Alder cycloadditions . Although the benefits of organocatalysts are undoubted, their synthesis is often a long and expensive process. Therefore, for sustainable application, the cost-efficient recovery and reuse of organocatalysts are critical issues. Fortunately, a wide range of recycling options are known in the literature, often based on liquid–solid phase separation .

Catalyst recycling can be achieved, for example, by immobilizing the catalysts to a solid support , e.g., silica gel , organic polymers , magnetic nanoparticles , or by membrane separation, e.g., using organic solvent nanofiltration (OSN) , which methods can be easily implemented in flow systems. Accordingly, the main recycling methods rely on the immobilization of catalysts on heterogeneous supports, however, this could often lead to the deterioration of activity and/or selectivity . A possible solution to avoid these drawbacks is the heterogenization of the catalyst after a homogeneous reaction.

For example, by incorporating a lipophilic side chain on the organocatalyst that does not interfere with its catalytic activity thanks to a linker between the catalyst and lipophilic units. In this way, a significant difference in polarity can be achieved between the catalyst and the other components of the reaction mixture. The lipophilic O-alkylated gallic acid unit increases the solubility of the organocatalyst in less polar solvents, such as DCM or toluene but leads to the precipitation of the organocatalyst in polar solvents, including MeOH or MeCN. As a result, the recycling of the organocatalysts can be achieved in a simple step by centrifugation or filtration.

Figure 1: Application of cinchona squaramide 1 and recyclable, lipophilic cinchona squaramide organocatalysts 2 in a new, gram-scale stereoselective synthesis of baclofen precursor.

Scheme 1: Synthesis of demethylated cinchona squaramide organocatalyst and the incorporation of the flexible 2-aminoethylene linker.

Scheme 2: Synthesis of the lipophilic tag from methyl gallate (8) and attachment to the cinchona squaramide.

Table 1: Catalyst 2 solubility in solvents with low polarity.a

aTo the lipophilic catalyst (2 mg) in a vial the appropriate solvent (200 µL) was added. Then, to check the precipitability of the catalyst, MeOH (800 µL) was added. bLarger amounts of cyclohexane (1 mL) was needed to dissolve the catalyst.

The main requirement for the polar solvent is that it should not dissolve the catalyst while it should completely dissolve the product. Therefore, we investigated the solubility of the Michael adduct 14 in methanol, ethanol, propan-2-ol, Patosolv® (a mixture of 85% of ethanol and 15% of propan-2-ol), and acetonitrile. The highest solubility of 14 was found in acetonitrile (63 mg mL−1) and methanol (17 mg mL−1). In both of these solvents, a low solubility of the lipophilic catalyst 2 was measured (<0.5 mg mL−1). Based on these results, acetonitrile was chosen as a precipitating solvent for the catalyst recycling.

Table 2: Solvent screening in the Michael addition of acetylacetone (13) to trans-β-nitrostyrene (12).a

aReaction conditions: acetylacetone (13, 0.21 mmol) was added to the solution of trans-β-nitrostyrene (12, 0.08 mmol) and 5 mol % of catalyst 2 in 0.5 mL of solvent, then, the resulting mixture was stirred at room temperature for 24 hours. bIsolated yields. cDetermined by chiral HPLC ((S)-enantiomer).

Figure 2: Classification of the tested non-polar solvents according to the GSK’s solvent sustainability guide .

Figure 3: Recycling of the lipophilic organocatalyst in the stereoselective Michael addition by replacing the solvent.

Table 3: Recycling of lipophilic organocatalyst 2 in the Michael addition of acetylacetone (13) to trans-β-nitrostyrene (12).a

aReaction conditions: acetylacetone (13, 0.96 mmol) was added to the solution of trans-β-nitrostyrene (12, 0.38 mmol) and 5 mol % of catalyst 2 in 2 mL of toluene, then, the resulting mixture was stirred at room temperature for 24 hours. After the reaction was completed, the volatile components were evaporated, and acetonitrile was added for the recycling of the catalyst 2. bIsolated yields. cDetermined by chiral HPLC ((S)-enantiomer).

Scheme 3: A new, stereoselective synthetic route for baclofen.

Table 4: Solvent and catalyst amount screening in the Michael addition of the cyclohexyl derivative of Meldrum’s acid 15 to 4-chloro-trans-β-nitrostyrene (16).a

aReaction conditions: Meldrum’s acid derivative 15 (0.064 mmol) was added to the solution of 4-chloro-trans-β-nitrostyrene (16, 0.096 mmol) and 1, 5 or 10 mol % of catalyst 2 in 470 µL of solvent, then, the resulting mixture was stirred at room temperature for 3 hours. bIsolated yields. cDetermined by chiral HPLC ((S)-enantiomer). dReaction time was 5 hours to achieve full conversion.

Table 5: Recycling of lipophilic organocatalyst 2 in the Michael addition of the cyclohexyl derivative of Meldrum’s acid 15 to 4-chloro-trans-β-nitrostyrene (16).a

aReaction conditions: Meldrum’s acid derivative 15 (0.19 mmol) was added to the solution of 4-chloro-trans-β-nitrostyrene (16, 0.29 mmol) and 10 mol % of catalyst 2 in 1.4 mL of solvent, then, the resulting mixture was stirred at room temperature for 3 hours. After the reaction was completed, the volatile components were evaporated, and acetonitrile was added for the recycling of the catalyst 2. bIsolated yields. cDetermined by chiral HPLC ((S)-enantiomer).

Finally, we planned to demonstrate the recyclability of our lipophilic organocatalyst 2 in the gram-scale synthesis of baclofen precursor (S)-17. The catalyst loading was set to 1 mol % to reduce the needed catalyst amount and the reaction time was increased to 5 hours to achieve full conversion. After the organocatalytic reaction in toluene, the volatile components were evaporated, and acetonitrile was added to precipitate the catalyst. In contrast to the small-scale recycling, in this case, we used filtration instead of centrifugation to recover the catalyst without significant loss (<5%). This demonstrates that we developed an organocatalytic reaction that can be easily scaled-up and the novel lipophilic catalyst can be recovered not only by centrifugation but also by filtration.

Scheme 4: Gram-scale synthesis of (S)-baclofen hydrochloride.

In conclusion, we have prepared a new lipophilic cinchona squaramide organocatalyst 2 modified with octadecyl side chains. Thanks to the lipophilic unit, the catalyst can be easily precipitated by exchanging the non-polar solvent with a more polar one, and then its separation can be achieved by centrifugation. The lipophilic catalyst 2 demonstrated its excellent catalytic activity in two stereoselective Michael addition reactions. Homogeneous catalysis was carried out in non-polar solvents (i.e., toluene), which allows the high performance of the lipophilic organocatalyst in terms of yield and stereoselectivity. To facilitate the pharmaceutical use of the lipophilic organocatalyst, we investigated a new, industry-relevant synthesis route of baclofen in the gram-scale. The chiral precursor 17 of baclofen was obtained with quantitative yield and excellent enantiomeric excess (up to 96%). The catalyst was applied in five consecutive runs without a decrease in catalytic activity, moreover, the catalyst loss was also negligible (<10%). Overall, it can be concluded that the incorporation of the lipophilic unit does not affect the catalytic activity and selectivity but enables the facile recycling of the catalyst.