A Chiral Primary Amine Thiourea Catalyst for the Highly Enantioselective Direct Conjugate Addition of α,α‐Disubstituted Aldehydes to Nitroalkenes

Abstract

Doppelte Aktivierung: Der difunktionelle Katalysator 1 – ein primäres Amin und zugleich ein Thioharnstoffderivat – ermöglicht die hoch enantioselektive direkte konjugierte Addition von α-verzweigten Aldehyden an Nitroalkene (siehe Schema) durch kooperative Aktivierung von Nucleophil und Elektrophil. Unter milden Reaktionsbedingungen ist eine Vielzahl an Addukten mit benachbarten quartären und tertiären Stereozentren zugänglich (>90 % ee). Despite extensive studies on secondary amine catalyzed conjugate additions of carbonyl compounds to nitroalkenes,11, 12 only two reports by Barbas and co-workers have addressed α,α-disubstituted aldehydes as potential nucleophilic partners.13 We selected the challenging combination of 1-nitrohex-1-ene, a β-alkyl-substituted nitroalkene, and 2-phenylpropionaldehyde as model substrates for initial optimization studies,14 with the hope that broad reaction scope would ensue. Among the primary amine thiourea catalysts examined, derivatives 1 and 3 were found to induce particularly high diastereo- and enantioselectivities (Table 1, entries 1 and 2). The best results were obtained using only a twofold excess of aldehyde relative to nitroalkene.15, 16 Diaminocyclohexane-derived catalyst 1 proved more broadly applicable and was selected for further optimization as catalyst 3, derived from diphenylethylene diamine, was found subsequently to afford optimal results only in reactions involving 2-phenylpropionaldehyde.17 Variation of standard reaction parameters (solvent, temperature, reagent ratios, concentration, catalyst loading) failed to improve the product yield, however, inclusion of controlled amounts of water led to significant improvements (Table 1, entry 1 vs 3–5, 2 vs 8).18 Catalysts bearing a secondary amide (1 and 3) afforded higher levels of substrate conversion and product yields than their tertiary amide counterparts (e.g. 2), while derivatives lacking an amide, as in 4, were virtually inactive (Table 1, entry 7). Entry Catalyst H2O [equiv] Yield [%][a] d.r. (syn/anti)[a] ee [%][b] 1 1 0 34 >10:1 96 2 3 0 93 >10:1 99 3 1 2.0 56 >10:1 96 4 1 5.0 64 >10:1 96 5 1 10 54 >10:1 96 6 2 5.0 31 >10:1 96 7 4 5.0 <5 – – 8 3 5.0 100 >10:1 99 A wide range of α,α-disubstituted aldehyde/nitroalkene combinations were surveyed to determine the scope and limitations of the methodology. Excellent enantioselectivity and useful levels of diastereoselectivity were obtained with a variety of substrates by using catalyst 1 (Table 2). The highest levels of diastereoselectivity (>10:1 d.r.) were observed for aldehydes bearing phenyl or ethereal (R1=OPh and p-MeOC6H4CH2O) α-substituents. At the other extreme, only modest diastereoselectivities (2.1–4.7:1 d.r.) were obtained for adducts 9, 10, and 19–22, results that were deemed satisfactory nonetheless given the minimal degree of steric differentiation between the aldehyde α-substituents. A variety of β-aryl-, β-heteroaryl-, and β-alkyl-substituted nitroalkenes underwent conjugate addition in good yields regardless of their electronic properties. The highly electrophilic 3,3,3-trifluoro-1-nitroprop-1-ene afforded adduct 7 in high ee and d.r. but only modest yield, a result ascribable to its susceptibility to effect alkylation of the primary amine group of 1.19, 20 Product Yield [%] d.r. (syn/anti) ee [%] Product Yield [%] d.r. (syn/anti) ee [%] 54 28:1 96 (syn) 91 23:1 99 (syn) 34 >50:1 97 (syn) 87 >50:1 99 (syn) 61 3.3:1 99 (syn) 99 (anti) 82 3.9:1 99 (syn) 99 (anti) 87 6.3:1 99 (syn) 98 (anti) 86 6.6:1 99 (syn) 94 (anti) 85 7.1:1 99 (syn) 95 (anti) 85 6.6:1 99 (syn) 97 (anti) 79 5.4:1 99 (syn) 95 (anti) 94 5.6:1 99 (syn) 96 (anti) 78 10.4:1 94 (syn) 92 (anti) 78 13:1 96 (syn) 98 2.1:1 99 (syn) 97 (anti) 82 3.1:1 99 (syn) 99 (anti) 63 4.7:1 98 (syn) 92 (anti) 81 3.8:1 99 (syn) 99 (anti) As noted above, diphenylethylene diamine derivative 3 proved a more effective catalyst than 1 in conjugate additions involving 2-phenylpropionaldehyde as the nucleophilic partner. The effect was especially pronounced in additions to trans-β-nitrostyrene, which proceeded in substantially higher enantioselectivity and diastereoselectivity with catalyst 3 compared to 1 (Table 3, entries 1–2). Further modifications to the catalyst structure led to the observation that valine-derived catalyst 23 afforded almost identical results to tert-leucine-derived 3, a significant outcome given the substantially lower cost of the precursor amino acid and also because this equivalence has not been observed in any other tert-leucine-derived thiourea-catalyzed reactions. The significantly diminished yield and ee values obtained with mismatched catalyst 24 point to the cooperative role of the stereochemistry of both the amino acid and the diamine in simultaneously defining conjugate addition selectivity and catalyst activity. Entry Catalyst Yield [%][a] d.r. (syn/anti)[a] ee [%] (syn/anti)[b] 1 1 100 1.2:1 67:3 2 3 86 8.6:1 97:24 3 23 84 11.9:1 97:32 4 24 35 6.9:1 71:40[c] A catalytic cycle consistent with our experimental observations is depicted in Scheme 1. Tautomerization of imine A, resulting from the condensation of aldehyde and catalyst 1, leads to the formation of an E or Z enamine. Preferred reaction via the thermodynamically favorable E enamine B is proposed to account for the observed diastereoselectivities.21 Binding of the nitroalkene through only one oxygen atom allows the enamine to attain sufficiently close proximity for carbon–carbon bond formation to occur (intermediate C).22 Proton transfer (D to E) followed by imine hydrolysis yields the product and regenerates the catalyst. Zwitterionic species analogous to D have been invoked in numerous studies concerning the mechanism of conjugate addition of enamines to nitroalkenes.23 These intermediates may undergo hydrolysis to the nitroaldehyde product or collapse to 1,2-oxazine-N-oxide and cyclobutane intermediates F and G, respectively. Although 1,2-oxazine-N-oxides such as F undergo hydrolysis readily in the presence of atmospheric moisture,23l hydrolysis of cyclobutanes analogous to G requires strong aqueous acid.23i,23l The beneficial role of water may lie in increasing turnover by eliminating a potential catalyst sink (formation of F) and accelerating imine (E) hydrolysis. However, cyclobutane G is unlikely to undergo hydrolysis under the catalytic conditions, and its irreversible formation may be responsible for catalyst deactivation.24 Proposed mechanism for the asymmetric addition of α,α-disubstituted aldehydes to nitroalkenes catalyzed by 1. Bn=benzyl. We have shown that chiral primary amine thiourea catalysts are highly effective in the addition of α,α-disubstituted aldehydes to nitroalkenes, generating synthetically versatile nitroaldehyde adducts. Simultaneous activation of both nucleophile and electrophile through a combination of effects typically associated with enzymes (approximation, hydrogen bonding, and covalent nucleophilic catalysis) allows this challenging transformation to take place under mild reaction conditions and with broad substrate scope. Interestingly, and in contrast to many enzymes, these bifunctional catalysts function by sequestering substrates from hydrophobic organic solvents into hydrophilic active sites. This study adds to a growing body of evidence suggesting that dual-activation catalysis with simple bifunctional organic frameworks holds substantial promise for asymmetric synthesis.25 Our current efforts are focused on further development of conjugate addition reactions promoted by thiourea amine derivatives, as well as on the design of new bifunctional frameworks for use in asymmetric catalysis. (2S,3R)-2,3-Dimethyl-4-nitro-2-phenylbutanal (6): Under a positive pressure of nitrogen at room temperature, thiourea catalyst 1 (75.3 mg, 0.20 mmol, 20 mol %) was loaded into an oven-dried 25-mL round-bottomed flask equipped with a magnetic stir bar, rubber septum, and nitrogen inlet. The catalyst was dissolved in dichloromethane (6.7 mL). Water (90.1 μL, 5.0 mmol, 5.0 equiv) and 2-phenylpropionaldehyde (265.4 μL, 2.0 mmol, 2.0 equiv) were subsequently added by syringe. The resulting clear colorless solution was stirred for approximately 2 min. Addition of 1-nitropropene (87.1 mg, 1.0 mmol, 1.0 equiv) by syringe produced a light yellow solution. The rubber septum was quickly replaced with a yellow polyethylene stopper (to avoid absorption of dichloromethane by the septum), and the reaction mixture was stirred for 24 h at room temperature. Aqueous hydrochloric acid solution (1 M, 7 mL) was added to the reaction flask, and the resulting biphasic mixture was stirred vigorously for 5 min at room temperature. The biphasic mixture was transferred to a separating funnel, and additional portions of dichloromethane (30 mL) and 1 M HCl (30 mL) were added. The phases were separated, and the aqueous layer was washed with dichloromethane (30 mL). The organic layers were combined and washed with saturated aqueous sodium bicarbonate solution (30 mL), saturated aqueous sodium chloride solution (30 mL), dried over anhydrous sodium sulfate, filtered, and concentrated under reduced pressure. The resulting yellow residue was purified by chromatography on silica (8 % diethyl ether/hexanes), providing the title compound as a colorless/light yellow liquid in 91 % yield (201.1 mg) in 23:1 diastereomeric ratio and with 99 % ee (major diastereomer) as determined by HPLC (Chiralpak AD-H, 2.0 % propan-2-ol/hexanes, 1.0 mL min−1, 230 nm; tr(minor enantiomer, minor diastereomer)=11.83 min, tr(major enantiomer, minor diastereomer)=12.87 min, tr(minor enantiomer, major diastereomer)=13.82 min, tr(major enantiomer, major diastereomer)=15.48 min). [α]=+88.6° (c=0.0200 g/2.0 mL, chloroform); 1H NMR (400 MHz, CDCl3): δ=9.47 (1 H, s), 7.42 (2 H, t, J=7.3 Hz), 7.34 (1 H, t, J=7.3 Hz), 7.24 (2 H, d, J=7.3 Hz), 4.57 (1 H, dd, J=3.3, 12.0 Hz), 4.19 (1 H, dd, J=10.6, 12.0 Hz), 3.17 (1 H, m), 1.48 (3 H, s), 0.81 ppm (3 H, d, J=7.0 Hz); 13C NMR (100 MHz, CDCl3): δ=200.6, 137.4, 129.4, 128.2, 127.4, 78.8, 55.9, 37.1, 14.6, 13.2 ppm; IR (neat): =3060 (w), 2981 (m), 2819 (w), 2719 (w), 1722 (s), 1533 (s), 1496 (m), 1446 (m), 1377 (s), 763 (m), 702 (m); HRMS (ESI): calcd for [C12H15NO3+NH4]+: 239.1396; found: 239.1402. Supporting information for this article is available on the WWW under http://www.wiley-vch.de/contents/jc_2001/2006/z602221_s.pdf or from the author. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.