Portal de Periódicos da CAPES

Homogeneous Chiral Nickel‐Catalyzed Asymmetric Hydrogenation of Substituted Aromatic α ‐Aminoketone Hydrochlorides through Dynamic Kinetic Resolution

2009; Wiley; Volume: 1; Issue: 2 Linguagem: Inglês

10.1002/cctc.200900084

1867-3899

Takuya Hibino, Kazuishi Makino, Takaya Sugiyama, Yasumasa Hamada,

Catalysis for Biomass Conversion

Vive la résolution: Homogeneous chiral nickel complexes catalyze the asymmetric hydrogenation of substituted aromatic α-aminoketone hydrochlorides through dynamic kinetic resolution to afford medicinally important β-aminoalcohols with excellent diastereo- and enantioselectivities. Precious transition metals, such as rhodium, iridium, and ruthenium, play a crucial role in homogeneous catalysis of asymmetric hydrogenation.1 However, they are expensive rare metals and are anticipated to be limited by the depletion of natural resources in the future. Hence, the investigation and development of sustainable methods using abundant and cheap base transition metals is desirable.2 Recently, we have developed of homogeneous chiral nickel catalysts for the asymmetric hydrogenation of a chiral, labile α-amino-β-keto esters through dynamic kinetic resolution (DKR).3–6 Asymmetric hydrogenation through DKR7 constitutes one of the most important carbon–hydrogen bond forming reactions, in which racemic substrates can be converted into enantiomerically pure products with two or more contiguous stereogenic centers using a single operation.8–10 As an extension of this study, we have investigated the diastereoselective hydrogenation of substituted aromatic α-aminoketone hydrochlorides by using a homogeneous achiral nickel catalyst. During the course of this study, we discovered a novel asymmetric hydrogenation and a remarkable selectivity change, based on the solvent effect. We now describe the asymmetric hydrogenation of substituted aromatic α-aminoketone hydrochlorides through DKR to afford β-aminoalcohols, which are important medicines11 and chiral modifiers,12 with anti stereochemistry and high enantio- and diasteroselectivity. Although the enantioselective hydrogenation of α-aminoketones has been extensively investigated,13, 14 studies concerning the diastereoselective hydrogenation of substituted aromatic primary α-aminoketones are still limited.15, 16 We first attempted the screening of achiral ligands for the diastereoselective hydrogenation of an α-aminoketone (S)-1 by employing our developed conditions for the synthesis of α-amino-β-ketoester hydrochlorides using nickel acetate (10 mol %) and a phosphine ligand (10 mol %) in the presence of sodium acetate (1 equiv) in trifluoroethanol (TFE) under a hydrogen pressure of 100 atm (Scheme 1). Surprisingly, no reaction took place. However, it was discovered that the presence of sodium tetrakis[bis(3,5-trifluoromethyl)phenyl]borate (NaBArF, 10 mol %) was essential for this hydrogenation. We then carried out preliminary experiments using the above conditions with 10 mol % NaBArF (Table 1). Diastereoselective hydrogenation of an aromatic α-aminoketone. Entry Ligand[h] Yield [%][b] anti/syn[c] 1 dcpm NR[d] – 2 dcpe quant. >95:5 3 dcpp 16 – 4 dcpb trace – 5 dppe NR – 6 dppf NR – 7[e] dcpe NR – 8[f] dcpe 4 – 9[g] dcpe 24 >95:5 Although most of the bidentate and monodentate phosphines resulted in little or no conversion, we were pleased to find that the hydrogenation using bis(dicyclohexylphosphino)ethane (dcpe) as the ligand smoothly proceeded at room temperature for 24 h to produce the anti β-aminoalcohol in quantitative yield with a diastereomeric ratio of 95:5 (Table 1, entry 2). The presence of the additives, NaBArF and sodium acetate, were essential for the catalytic activity and smooth reaction (Table 1, entries 7–9). However, the measurement of the optical purity of the product showed complete racemization during the homogeneous nickel-catalyzed hydrogenation. From this unexpected result, we envision that a homogeneous chiral nickel catalyst should catalyze the asymmetric hydrogenation of racemic substituted aromatic α-aminoketones to afford β-aminoalcohols with anti stereochemistry through DKR. Several commercially available chiral phosphines and solvents were then tested (Scheme 2, Table 2).17 Only the Josiphos-type phosphines were effective for this hydrogenation similar to the case of the α-amino-β-ketoesters.3 The hydrogenation of (rac)-1 using nickel acetate (10 mol %) and (R,S)-3 a in the presence of NaBArF (10 mol %) and sodium acetate (1 equiv) in toluene at 23 °C for 24 h afforded (1R,2S)-2 with complete diastereoselectivity and 78 % ee (Table 2, entry 1). The anti stereochemistry of (1R,2S)-2 was established by the NOE experiment after its transformation to 1,3-oxazolidin-2-one 4 (Scheme 2).18 When the alternate ligand (S,R)-3 a was used as the chiral ligand, the enantiomer, (1S,2R)-2, was obtained in 82 % yield and 76 % ee (Table 2, entry 2). This result clearly indicates that the nickel complex catalyzed the rapid racemization of the substrate and the hydrogenation proceeded through DKR. In the case of the bulky (R,S)-3 b with electron-donating substituents, the hydrogenation was completed in 24 h to give the product in high yield, and the enantioselectivity was slightly improved from that of (R,S)-3 a (Table 2, entry 3). An investigation of the solvent effect revealed the remarkable character of this homogeneous nickel-catalyzed hydrogenation (Table 2, entries 6-9). When acetic acid was used as the solvent, the racemization of the substrate by the nickel catalyst was suppressed; the hydrogenation of the optically active (S)-1 proceeded diastereoselectively with retention of the stereogenic center in the substrate to generate (1S,2S)-2 (syn) and (1R,2S)-2 (anti) as the major products by using (R,S)-3 a and (S,R)-3 a as the ligands, respectively (Table 2, entries 8 and 9). These results clearly provide the proof of a ketone-reduction mechanism during the hydrogenation using the homogeneous nickel catalyst. Asymmetric hydrogenation and chiral ferrocenyl ligands. Entry Ligand Solvent Yield [%][b] anti/syn[c] ee [%][d] anti syn 1 3 a PhCH3 75 >99:1 78 – 2 3 a[e] PhCH3 82 >99:1 76[f] – 3 3 b PhCH3 92 >99:1 86 – 4 3 c PhCH3 92 >99:1 6 – 5 3 d PhCH3 NR – – – 6 3 a TFE 94 90:10 17 – 7 3 a CH2Cl2 68 98:2 70 – 8[g] 3 a AcOH 42 18:82 40 97 9[g] 3 a[e] AcOH 62 86:14 92 17 10[h] (R)-BINAP CH2Cl2 NR – – – 11[i] (R)-MeO-BIPHEP AcOH 70 95:5 10 – 12[j] (S,S)-MeDuphos TFE trace – – – In contrast, the precious metal-catalyzed hydrogenation of (rac)-1 was inefficient (Table 2, entries 10–12). Surprisingly, in the case of the Ru-catalyzed hydrogenation, which employs the conditions for the α-amino-β-ketoester hydrochlorides, no reaction was detected.8b,8d The Ir-catalyzed hydrogenation, under the same conditions for the α-amino-β-ketoester hydrochlorides, proceeded sluggishly in an anti-selective manner, and gave a product with low enantioselectivity.10 The Rh-catalyzed hydrogenation using the conditions for the α-aminoacetophenones resulted in little or no reaction.13h The homogeneous nickel-catalyzed asymmetric hydrogenation was applied to various aromatic α-aminoketones20 (Scheme 3, Table 3). Although the diastereoselectivities were complete (except for entry 4) and the enantioselectivities were good, the reaction rate was significantly influenced by the electronic and steric properties of the substrate. In the case of the substrates with an electron-donating or bulky group, the enantioselectivities increased, but the reactions were sluggish (Table 3, entries 2 and 7). Raising the temperature to 50 °C resulted in a complete reaction, but with lower enantioselectivity (Table 3, entries 3 and 9). Nonetheless, the results of the above asymmetric hydrogenation using homogeneous chiral nickel catalysts are noteworthy because the nickel complexes catalyze the asymmetric hydrogenation of configurationally stable α-aminoketones to afford anti β-aminoalcohols with good enantioselectivity through dynamic kinetic resolution. Asymmetric hydrogenation of aromatic α-aminoketones. Entry Ar R Yield [%][b] anti/syn[c] ee [%][d] 1 p-MePh Me 92 >99:1 86 2 p-MeOPh Me 22 >99:1 85 3[e] p-MeOPh Me 92 >99:1 75 4 p-CF3Ph Me 82 80:20 63 5 m-MePh Me 90 >99:1 85 6 o-MePh Me 76 >99:1 61 7 p-MePh iPr 21 >99:1 96 8[e] p-MePh iPr 92 >99:1 86 9[f] p-MePh iPr 90 >99:1 69 10 p-MePh Bn 92 >99:1 84 11[g] p-MePh Bn 90 >99:1 74[h] 12 Ph Me 86 >99:1 76[i] 13 Ph Et 94 >99:1 85 14 Ph nPr 91 >99:1 82 In summary, we have succeeded in developing an asymmetric hydrogenation of aromatic α-aminoketones using homogeneous chiral nickel catalysts, which proceeds in toluene with high diastereoselectivities and excellent enantioselectivities through DKR. Using acetic acid as a solvent, the hydrogenation proceeds diastereoselectively under reagent control with the retention of the stereogenic center of the α-aminoketones to afford the syn or anti β-aminoalcohol as the major product. The asymmetric hydrogenation described above represents the first example of asymmetric hydrogenation through dynamic kinetic resolution for substituted aromatic α-aminoketones with a primary amino group. Further investigations of the mechanistic details and potential applications of the asymmetric hydrogenation using homogeneous nickel catalysts for the synthesis of optically active compounds are underway. Typical procedure for the asymmetric hydrogenation (the reaction was carried out in a glassware placed in a stainless autoclave apparatus): A dried glass test tube was charged with nickel (II) acetate tetrahydrate (2.5 mg, 0.010 mmol), the (R,S)-ferrocenyl ligand (3 a, 0.010 mmol), NaBArF (8.9 mg, 0.010 mmol), and dry CH2Cl2 (0.5 mL). After being degassed by three freeze-thaw cycles, the mixture was stirred at 50 °C for 45 min under an argon atmosphere. The resulting solution was concentrated and dried in vacuo at room temperature for 15 min. The α-aminoketone (rac)-1 (0.10 mmol), sodium acetate (8.2 mg, 0.10 mmol), and dry toluene (0.5 mL) were added to the above residue in a glove bag. After this mixture was degassed by three freeze-thaw cycles, the glass test tube was transferred to a stainless steel autoclave in a glove bag. The mixture was stirred at 23 °C under hydrogen (100 atm) for 24 h. After the hydrogen was carefully released, 1 M hydrochloric acid (0.2 mL) and methanol (0.5 mL) were added and the mixture was concentrated in vacuo below 40 °C. The crude residue was dissolved in methanol and again concentrated in vacuo. The conversion yield and diastereomeric ratio were determined by 1H NMR of the crude material in d6-DMSO. Benzoic anhydride (22.6 mg, 0.10 mmol) in THF (1 mL) and triethylamine (42 μL, 0.30 mmol) in THF (0.3 mL) were added dropwise with stirring to the above residue at 0 °C. After stirring the mixture at room temperature overnight, the reaction was quenched with saturated aqueous ammonium chloride, and the mixture was diluted with ethyl acetate (20 mL). The organic layer was washed with saturated aqueous ammonium chloride and brine, dried over sodium sulfate, filtered, and concentrated in vacuo. The residue was purified by preparative TLC to give the N-benzoyl derivative. HPLC analysis using a CHIRALCEL OD-H column and hexane/iPrOH (85:15, 0.5 mL min−1, retention times: (1R, 2S): 15.0 min; (1S, 2R): 20.8 min); m.p. 163–164 °C (ethyl acetate n-hexane); =−56.5 cm3 g−1 dm−1 (c=0.91 g cm−3 in CHCl3) (for 86 % ee); 1H NMR (400 MHz, CDCl3, 23 °C, TMS): δ=1.13 (d, 3J(H,H)=6.8 Hz, 3 H; CH3), 2.35 (s, 3 H; ArCH3), 4.56 (m, 1 H; NCH), 4.96 (d, 3J(H,H)=2.4 Hz, 1 H; OCH), 6.22 (brd, 3J(H,H)=8.0 Hz, 1 H; NH), 7.17 (d, 3J(H,H)=8.0 Hz, 2 H; ArH), 7.43–7.54 (m, 4 H; ArH), 7.75–7.77 (m, 2 H; ArH), 8.09–8.11 ppm (m, 1 H; ArH); 13C NMR (100 MHz, CDCl3, 23 °C, TMS): δ=14.6, 21.1, 51.3, 76.5, 126.2, 127.0, 128.4, 128.6, 128.9, 131.6, 137.2, 137.6, 168.0 ppm; IR (KBr): =3290, 2915, 2849, 1636, 1541 cm−1; HRMS(ESI): m/z: calcd for C17H20NO2: 270.1494 [M+H]+; found: 270.1484. This work was financially supported in part by a Grant-in-Aid for Scientific Research (B) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.