Cu (II)-Catalyzed Asymmetric Henry Reaction with a Novel C1-Symmetric Aminopinane-Derived Ligand

A novel C1-symmetric dinitrogen ligand was synthesized in high yield from commercially available (1R,2R,3R,5S)-(−)-isopinocampheylamine and 1-methyl-2-imidazolecarboxaldehyde. In combination with Cu(OAc)2∙H2O, this new ligand promote the reaction between nitromethane and aliphatic aldehydes with high yields (up to 97%) and moderate enantioselectivities (up to 67% ee). The reactions with benzaldehyde required prolonged reaction time that resulted in diminished yields, but accompanied with ee-values in the 55%–76% range.


OPEN ACCESS
the β-amino alcohol functionality is present in long-chain lipids such as sphingosines [8,9]. Thus, the development of an efficient asymmetric protocol for this type of reaction is of current interest [10,11].

Scheme 1. Synthesis of the ligand 4.
Initially, in connection with our interest in the synthesis of aliphatic nitroalcohols, we examined the ability of 4 to promote the nitroaldol reaction between valeraldehyde (5a) and nitromethane in the presence of CuCl2•2H2O (5 mol %) at 0 °С with DIPEA (1 equiv.) as base additive. The results are summarized in Table 1. When THF was used as solvent, the reaction proceeded smoothly resulting in high yield (92%), but moderate enantioselectivity was observed ( Table 1, entry 1). Then Cu(OAc)2•H2O was tested to see if there were any influence of the counterions on the reaction. Both copper salts gave similar results in terms of enantioselectivity, but the yields varied significantly depending on the solvent ( Table 1, entries 1-4). Most likely, this is due to differences in the solubility of the formed complexes. Hence, the use of other solvents was investigated. Of the solvents tested, i-PrOH was the best choice (Table 1, entry 6, 97% yield, 55% ee). Lowering the temperature of the reaction to −20 °С resulted in a significant decrease of the reaction rate with no improvements in enantioselectivity ( Table 1, entries 8 and 9). The absolute configuration in the product 6a was determined based on comparison of the specific optical rotation value with literature [24][25][26][27][28][29][30][31][32]. The Henry reaction is reversible [1]. The enantiomeric excess of the nitroaldol adducts can decrease as a result of the retro-nitroaldol reaction. Hence, a time-course study was performed to investigate the possibility of reversibility. The reaction was almost complete within four hours, while the ee-values of the product 6a remained constant in the conditions explored (Table 1, entry 10-13), indicating the absence of reversibility.
An external base is usually added to promote the reaction. Considering the still low ee of the product observed, and the fact that the excess of additional base can enhance the non-enantioselective reaction pathway [26], the effect of base loading was examined. The addition of even small amounts of external base noticeably improved the reaction rate, while the ee-values of the products were unchanged with increasing amounts of base added. These experiments verified that inhibition of an unselective pathway occurred (Table 1, entries [15][16][17]. The loading of 10 mol % of external base was sufficient to promote the reaction until completion. Based on the conditions investigated, different aliphatic aldehydes were tested. As shown in Table 2, all of the aldehydes smoothly converted into the nitroaldol products with good to excellent yields. Moderate ee-values ranging from 49% to 67% were observed; the highest ee-value was observed with 2,2-dimethylpropanal (5c) ( Table 2, entry 3, 67% ee). Again, the absolute configuration in the products 6a-6h was assigned based on specific optical rotation values.
Then we tested the catalyst system with benzaldehyde (5h) ( Table 2, entries 8-10). The reaction between 5h and nitromethane in the presence of ligand 4 and CuCl2•2H2O did not reach completion even after 20 hours at 4 °С ( Table 2, entry 8). Lowering the temperature of the reaction improved the level of enantioinduction to 76% ee, but drastically diminished the yields to impractical values ( Table 2, entries 9-10). Hence, other aromatic aldehydes were not investigated.
Finally, in view to explore the nitroalkane substrate scope, nitroethane was reacted with aliphatic aldehydes employing optimized catalytic procedure ( Table 3). The corresponding nitroaldol adducts bearing two stereogenic centers were obtained in high yields (up to 94%) albeit moderate diastereoselectivities favoring syn-product and enantioselectivities were observed. Diastereoselectivity was improved when α-branched aldehyde was used (Table 3, entry 2).

Experimental Section
All starting reagents and solvents were obtained from Sigma Aldrich and used as purchased without further purification. Analytical TLC was performed using silica gel 60 F254 plates (Merck). Flash column chromatography was performed on silica gel 60 (40-63 nm). NMR spectra were recorded on Varian Gemini and Bruker Ascend TM 400 spectrometers 300 MHz and 400 MHz for 1 H-NMR and 75 MHz and 100 MHz for 13 C-NMR, respectively. Chemical shifts are reported in ppm downfield from tetramethylsilane relative to CDCl3 as internal standard (7.24 ppm for 1 H and 77.00 for 13 C). IR spectra (4000-600 cm −1 ) were recorded on a Perkin-Elmer Spectrum BX series FT-IR spectrophotometer using a reflectance cell (HATR). Optical rotations were measured using a 1 mL cell with 1 on Perkin Elmer 341 polarimeter in dedicated solvent. HPLC analyzes were performed on Agilent 1200 Series instrument using chiral OD-H or AD-H columns.

General Procedure for the Enantioselective Henry Reaction
Ligand 4 (6.2 mg, 5 mol %) and Cu(OAc)2•H2O (5 mg, 5 mol %) were added to the test tube containing 2 mL of i-PrOH and stirred for an hour to obtain a blue solution. Then the test tubes were transferred to a bath at the given reaction temperature and 0.5 mmol of aldehyde, 10 equivalents of nitroalkane and 0.1 equivalent of DIPEA (9 μL) were added. The reaction mixture was left stirring for the 24 h, after that the volatile compounds were removed under reduced pressure and the residue was directly purified on silica gel column eluting with hexane:EtOAc to afford the corresponding product.

Conclusions
In conclusion, a new pinane-based dinitrogen C1-symmetric ligand has been synthesized via an experimental simple one pot procedure. This ligand was found to be reasonably selective for the copper catalyzed asymmetric Henry reaction between nitromethane and aliphatic aldehydes. Of merit, high yields were observed for this class of aldehydes. The ease of availability of ligand 4 in both enantiomeric forms, together with the simple and scalable procedure for its preparation, is of advantage for the present catalytic system. Currently, this catalyst system is investigated towards the synthesis of vicinal amino-alcohol derived lipids.