It is envisaged that chiral phosphonates will become increasingly useful chiral building blocks in organic synthesis [1, 2, 3, 4, 5, 6, 7, 8]. Furthermore, there has been considerable interest in the biological properties of the phosphorus analogues of carboxylic acids [9, 10]. Although several efficient processes have been developed during the past few decades for synthesis of chiral 1-arylethylphosphonates [11, 12, 13, 14, 15, 16, 17, 18], methods for the catalytic enantioselective synthesis of chiral alkylphosphonates, especially those bearing a β-stereogenic center, remain scarce [19, 20, 21]. Given its inherent efficiency and atom economy, it was envisaged that catalytic asymmetric hydrogenation [22, 23, 24, 25, 26, 27, 28, 29] would provide an ideal approach for the preparation of chiral phosphonate derivatives. We recently developed several new synthetic methods involving the Rh-catalyzed asymmetric hydrogenation of β-substituted α,β-unsaturated phosphonates [30, 31] and β-substituted β,γ-unsaturated phosphonates [32, 33]. Although there have already been several reports in the literature pertaining to the asymmetric hydrogenation of unsaturated phosphonates, the number of effective catalysts available for this transformation is limited, and the research towards the development of new catalysts remains a great challenge.
In our previous study, we developed a series of chiral ImiFerroPhos ligands and used the rhodium complexes of the ligand to affect the highly enantioselective hydrogenation of a series of 3-aryl-substituted 2-phosphonomethylpropenoates [34]. The structure of the ImiFerroPhos ligands was similar to that of the TaniaPhos ligand, which showed high levels of activity towards the hydrogenation of β-substituted α,β-unsaturated phosphonates but provided poor enantioselectivity. The ImiFerroPhos ligands, which contain a heterocycle, are more amenable to asymmetric catalysis than other chiral ligands because they can be readily constructed and derivatized, and possess unique electronic and steric properties. In the current study, we investigated the application of ImiFerroPhos ligands to the Rh-catalyzed asymmetric hydrogenation of β-substituted α,β-unsaturated phosphonates. The ImiFerroPhos ligands performed effectively in this regard to give the corresponding chiral phosphonates in good to high enantioselectivity under mild hydrogenation condition.
All of the reactions and manipulations were performed in a nitrogen-filled glove box or under a nitrogen atmosphere using Schlenk techniques unless otherwise noted. All of the solvents were distilled under argon in the presence of the following desiccants: sodium and benzophenone for toluene and tetrahydrofuran (THF), and CaH2 for dichloromethane (CH2Cl2). The NMR spectra were obtained on a Bruker DRX 400 spectrometer. 31P NMR shifts were referenced to external 85% H3PO4, while 13C and 1H NMR shifts were referenced to the residual signals of deuterated solvents.
All of the ligands used in the current study were prepared according to previously reported procedures [34].
1-{(R)-1-[(S)-2-(diphenylphosphino)ferrocenyl]lpropyl}-2-(diphenylphosphino)benzimidazole (Le). Orange crystals; 1H NMR (400 MHz, CDCl3): δ 0.63 (t, J = 8.0 Hz, 3H), 2.35 (m, 1H), 2.54 (m, 1H), 3.87 (s, 1H), 4.07 (m, 4H), 4.12 (s, 1H), 4.43 (s, 1H), 4.90 (s, 1H), 5.82 (m, 2H), 6.39 (m, 2H), 6.70 (m, 1H), 6.87 (m, 1H), 7.04 (m, 2H), 7.25 (m, 8H), 7.32 (m, 3H), 7.43 (m, 2H), 7.65 (m, 2H), 7.75 (m, 2H); 13C NMR (100 MHz CDCl3): δ 11.3, 28.9, 60.8 (d, J = 20 Hz), 68.3, 69.4, 70.1, 71.5, 74.4, 91.9, 111.7, 118.6, 120.0, 120.8, 125.3, 125.6, 126.2, 126.9, 127.0, 127.7, 128.2, 129.8, 133.1, 134.6, 135.6, 136.3, 136.8, 137.8; 31P NMR (162 MHz, CDCl3): δ -25.5 (d, J = 59.9 Hz), -34.5 (d, J = 55.1 Hz); HRMS (ESI): calcd for C44H38FeN2P2 [M+H]+, 713.1938; found, 713.1922.
A solution of (EtO)2P(O)CH2P(O)(OEt)2 (1.44 g, 5 mmol) in THF (2 ml) was slowly added to a slurry of NaH (0.19 g, 5.5 mmol, 70% in oil) in THF (10 ml) at 0 °C, and the resulting mixture was stirred at room temperature for 30 min. A solution of ketone (4.25 mmol) in THF (3 ml) was then added to the reaction, and the resulting mixture was stirred at room temperature until the ketone could no longer be detected by TLC. The mixture was then diluted with ether and washed sequentially with an aqueous solution of saturated NH4Cl (25 ml) and brine (25 ml) before being dried over anhydrous Na2SO4 and distilled to dryness in vacuo to give the crude product as a residue, which was purified by column chromatography (silica gel, EtOAc:hexane = 1:1).
In a nitrogen-filled glove box, [Rh(COD)2]BF4 (1.0 mg, 0.0025 mmol) and (Rc,SFc)-ImiFerroPhos (0.0028 mmol) were dissolved in degassed CH2Cl2 (1 ml) in a 5 ml vial. Following a 15-min period of stirring at room temperature, a solution of α,β-unsaturated phosphonate (1, 0.25 mmol, S/C = 100) in degassed CH2Cl2 (1 ml) was added to the reaction vial, and the resulting mixture was transferred to an autoclave. The autoclave was then charged with H2 (≈ 1 MPa), and the reaction mixture was hydrogenated at room temperature for 24 h. The H2 gas was carefully released from the autoclave, and the reaction mixture was passed through a plug of silica gel (eluting with a mixture of hexanes/EtOAc) to afford β-substituted alkylphosphonates (2). The enantiomeric excess was determined by HPLC on a chiral stationary phase.
Diethyl (E)-(2-phenyl-1-propene) phosphonate (E)-1a was initially selected as a model substrate for the ligand screening and optimization experiments for the Rh-catalyzed asymmetric hydrogenation reaction. The reactions were typically conducted at room temperature for 24 h in CH2Cl2 under 1 MPa of H2 pressure using 1% of [Rh(COD)2]BF4 and 1.1% of ligand. The results are summarized in Table 1. When (Rc,SFc)-ImiFerroPhos La (Fig. 1) bearing a methyl substituent in the ferrocenylmethyl position was used as the ligand, 1a was hydrogenated to give 2a in 82% ee (Table 1, entry 2). This result showed that ImiFerroPhos, which contained a heterocycle, was more efficient than TaniaPhos (Table 1, entry 2 vs entry 1). The introduction of an ethyl group, as in (Rc,SFc)-ImiFerroPhos Lb, increased the enantioselectivity of the transformation very slightly to 83% ee (Table 1, entry 3). Further increases in the size of the substituent at the ferrocenyl position to a phenyl (Lc) or i-Pr (Ld) group, however, resulted in a reduction in the enantioselectivity to 81% and 75% ee (Table 1, entries 4 and 5), respectively. The conversion of the imidazole heterocycle in Lb to a benzimidazole (ligand Le) led to an increase in the ee value to 90% (Table 1, entry 6).
It is noteworthy that both the catalytic activity and enantioselectivity of the Rh/(Rc,SFc)-ImiFerroPhos Le complex were sensitive to the solvent used. The use of protonic solvents such as MeOH and i-PrOH provided mixed results. For example, the MeOH provided a much lower conversion (Table 1, entry 7), whereas the i-PrOH gave complete conversion, albeit with a slightly lower enantioselectivity (Table 1, entry 8). THF and toluene proved to be inferior solvents and gave lower conversions, with toluene providing the lowest conversion of all of the solvents tested in the current study (Table 1, entries 9 and 10).
Encouraged by the positive results obtained for the hydrogenation of (E)-1a, we proceeded to investigate the scope of this transformation towards a variety of different β-substituted α,β-unsaturated phosphonates 1a-1k, using (Rc,SFc)- ImiFerroPhos Le as the ligand and CH2Cl2 as the solvent under H2 pressure of 1 MPa at room temperature for 24 h. As shown in Table 2, a range of β-aryl-α,β-unsaturated phosphonates (E)-1a-1h were successfully hydrogenated under the optimized conditions to give the corresponding 2-arylpropylphosphonates in good yields and good to high enantioselectivity (up to 92% ee). These results indicated that the current catalytic system was tolerant to the electronic properties of the substituents on the phenyl ring of the substrate in terms of the impact of these properties on the reactivity and enantioselectivity of the transformation. The position of the substituent on the phenyl ring appeared to have very little impact on the outcome of the reaction. Three substrates bearing a methoxy group at the ortho, meta, or para position of the phenyl ring were hydrogenated in good enantioselectivity (Table 2, entries 2-4). The electronic properties of the substituents at the para position of the phenyl ring also appeared to have very little impact on the enantioselectivity, with electron withdrawing and electron donating groups at this position providing similar results. Diethyl (E)-[2-(2-naphthyl)-1-propenyl] phosphonate (1g) and the (S)-heteroaromatic substrate 1h were also successfully hydrogenated under the optimized conditions to give the desired products in 84% and 86% ee, respectively (Table 2, entries 7 and 8).
To expand the synthetic scope and application of this enantioselective procedure, we also applied the optimized conditions to a variety of other α,β-unsaturated phosphonates (Fig. 2). As shown in Table 3, the β-alkyl substituted α,β-unsaturated phosphonate (E)-1i provided the hydrogenation product in good yield with 75% ee (Table 3, entry 1). In contrast, the hydrogenation of the corresponding (Z)-isomer ((Z)-1i), gave the opposite enantiomer with a lower ee value (Table 3, entry 2). The enantioselectivity results for substrates (E)-1j and (E)-1k were lower than that of their methyl analogue (E)-1a.
We have demonstrated that (Rc,SFc)-ImiFerroPhos Le is a suitable ligand for the Rh-catalyzed asymmetric hydrogenation of β-substituted α,β-unsaturated phosphonates, with good yield and ee value of up to 92% being obtained for a range of substrates under mild hydrogenation conditions (i.e., 1 MPa of H2 pressure and room temperature). This catalytic system could therefore be used in a practical sense for the synthesis of optically active β-substituted alkylphosphonates. The development of new catalytic methods for the synthesis of chiral alkylphosphonates is currently underway in our laboratory.
2014, Vol. 35 Issue (2): 227-231
PDF (481KB)
2014, Vol. 35 Issue (2): 227-231
PDF (481KB)