In 1999 Francis and Jacobsen disclosed the discovery of a novel catalyst for alkene epoxidation . Using a solid-phase synthesis protocol on an aminomethyl polystyrene resin, ligands bearing potential metal-binding moieties were prepared. The ligands comprised five amino acids with donor side arms (Asp, Cys, His, Met, and Ser), three different chiral linking elements each differing in rigidity (1-amino-2-indanol, trans-1,2-diaminocyclohexane, and Ser), and 12 different capping agents with functionalities such as heterocycles, phosphanes, and salicylimines, which in turn were attached through imine and amide bonds. Initially, the metal-binding ability of the resulting 192 ligands was determined in a pooled assay with 30 different transition metal ions to yield 5760 possible metal-ligand combinations. Metal binding was visually detected with selective inorganic staining reagents in a fashion similar to previous studies. The entire pooled catalyst library was screened for the epoxidation of trans-b-methylstyrene (TBMS) to determine suitable reaction conditions. Those catalysts prepared from VOSO4 and FeCl2 were most active. Since VOSO4 also displayed significant epoxidation activity in the absence of the ligand library, FeCl2 was chosen for further investigations. The screening of 12 FeCl2-derived libraries each containing a mixture of 16 basic structures and a different end cap furnished ligands with pyridine-containing end caps as the most active epoxidation catalysts. Further deconvolution showed that the most active catalytic system comprised FeCl2, ligand structures exhibiting a pyridine-containing end cap, a serine linker, and serine or cysteine as the amino acid bound to the solid support. These systems resulted in high levels of catalytic activity, but only low enantioselectivities for the epoxidation of TBMS was observed (ee = 4-7%). A second-generation 96-member optimization library based on the identified epox-idation catalyst produced moderately enantioselective variants (ee = 15-20%) for the target reaction.
Hoshino and Yamamoto have described the vanadium-catalyzed asymmetric epoxidation of allylic alcohols using a-amino acid-based hydroxamic acid ligands . In order to optimize the ligands, again the iterative positional optimization approach was used, which involves screening one compound of a ligand structure for selectivity, while holding the other units constant (Scheme 32.17). In the first step, the source of chirality for the ligand - the amino acid moiety - was optimized, and the best result in this case was achieved using the tert-leucine-derived hydroxamic acid. In the second step, the imido group was examined and optimized to the 1,8-naphthalenedicarbonyl-protected hydroxamic acid (87% ee). Finally, the aryl groups near the metal coordination site were changed, identifying N-bis-(1-naphthyl)methyl-substituted hydroxamic acid as the most effective ligand for a range of disubstituted allylic alcohols. The asymmetric epoxidations were performed in the presence of 1 mol% of VO(OiPr)3 and 1.5 mol% of the best ligand giving rise to the corresponding chiral epoxy alcohols in 58-99% yield and 7696% ee.
Was this article helpful?