Chikako Ogawa

Lewis Acid Catalysis in Water with a Hydrophilic Substrate: Scandium‐Catalyzed Hydroxymethylation with Aqueous Formaldehyde in Water

The use of Lewis acids as catalysts is one of the most powerful strategies in organic chemistry. In general, conventional Lewis acids such as AlCl3, TiCl4, BF3·OEt2, SnCl4, etc., require strictly anhydrous conditions because they immediately react with water in preference to the substrates, resulting in serious decomposition of the catalysts or the substrates. To address this issue, water-compatible Lewis acid catalyzed organic reactions have been intensively studied in the past decade. Considering the growing concern over environmental pollution, water is an ever more desirable alternative to organic solvents, as water is a clean, safe, and inexpensive solvent. Several surfactant-type catalysts derived from water-compatible Lewis acids have been developed, and catalytic asymmetric reactions have been achieved in water without requiring any organic cosolvents. These reactions proceeded smoothly by creating hydrophobic domains in the water to stabilize and concentrate the organic substrates, or by suppressing undesired reaction pathways that may occur in water. One of the key factors for these successes is the hydrophobicity of the substrates. Aqueous formaldehyde solutions (i.e. formalin) are one of the most important single carbon electrophiles and are representative of a hydrophilic substrate. As mentioned above, the hydrophobicity of the substrates is very important for organic reactions in water and hydrophilic substrates are often difficult to handle in water. Herein, we address this issue and describe a catalytic hydroxymethylation reaction with aqueous formaldehyde in which water is the sole solvent. First, we carried out the hydroxymethylation of silicon enolate 1 with 1 equivalent of formaldehyde (36% aqueous solution; aq. HCHO) in the presence of 2 mol% of scandium tris(dodecyl sulfate) (Sc(DS)3) [4] at a 1.0m concentration at 20 8C for 1 hour in water. The reaction proceeded sluggishly and afforded desired product 2 in poor yield. Increasing the amount of aq. HCHO to 3 equivalents improved the yield to 25%, but addition of more aq. HCHO did not result in increased amounts of the desired product (Figure 1a, navy blue). Gratifyingly, while optimizing the reaction conditions, we found that desired product 2 was obtained in 82% yield by using 5 equivalents of aq. HCHO and by extending the reaction time to 8 hours. Notably, under these conditions the competitive hydrolysis of silicon enolate 1 was observed. The loading levels of the catalyst also has an effect on the yield; the yield of 2 showed good correlations to the amount of Sc(DS)3 used, which ranged from 2, 5, 10, to 20 mol%. To improve the yield, we also investigated the concentration effect of aq. HCHO for each of the different Sc(DS)3 loadings (2, 5, 10, or 20 mol%) in the presence of 5 equivalents of aq. HCHO (Figure 1b). In the presence of 10 and 20 mol% of Sc(DS)3, the yields were improved to more than 80% as the concentrations increased to 2.0m, however, no additional improvement was observed when the concentration was increased to more than 2.0m. In the cases of 2 and 5 mol% of Sc(DS)3, the yields leveled off at much lower concentrations of 0.5m and 1.0m, respectively. These results indicated that Sc(DS)3 might be saturated by aq. HCHO. On the basis of the experiments described above, it can be said that despite the good solubility of HCHO in water, the amount of HCHO in the hydrophobic environment increases in the presence of Sc(DS)3 because of Lewis acid–Lewis base interactions between Sc(DS)3 and HCHO, therefore, allowing the reaction of HCHOwith silicon enolate 1 to proceed smoothly in water. After tuning the reaction conditions, we found that several silicon enolates reacted with aq. HCHO (5.0 equiv) in the presence of 5 mol% of Sc(DS)3 in water (1.0m) at 20 8C Figure 1. Hydroxymethylation of 1. a) Changes in yield as a function of the number of equivalents of HCHO at different catalyst loadings for a 1.0m solution monitored over 1 hour. b) Changes in yield as a function of concentration at different catalyst loadings for reaction containing 5.0 equivalents of HCHO monitored over 8 hours.

Zn-catalyzed asymmetric allylation for the synthesis of optically active allylglycine derivatives. Regio- and stereoselective formal alpha-addition of allylboronates to hydrazono esters.

We have developed Zn-catalyzed asymmetric allylation of hydrazono esters with allylboronates. The reactions proceeded smoothly in high yields and high stereoselectivities. Remarkably, formal α-addition occurred for α-substituted allylboronates exclusively, and excellent stereoselectivities were observed. This is the first example of catalytic regio- and stereoselective allylations with formal α-addition. In addition, the reaction proceeded in aqueous media, and the use of water is essential. Zn(OH)2 might be a catalyst in this asymmetric allylation, and the catalytic activity of Zn(OH)2 was confirmed. This is also the first case of chiral metal hydroxide-catalyzed asymmetric reactions.

Scandium-catalyzed ring-opening desymmetrization of meso-epoxides

In the presence of catalytic amounts of Sc(DS) 3 and chiral bipyridine ligand 1, ring-opening desymmetrization of mesa-epoxides with aromatic amines and indole derivatives proceeded smoothly in water without using any organic solvents to afford the corresponding adducts in high yields with high enatioselectivities. Interestingly, the Sc-caytalyzed reactions proceeded much faster in water than in dichloromethane.