Asymmetric Henry Reaction of 2-Acylpyridine N-Oxides Catalyzed by a Ni-Aminophenol Sulfonamide Complex: An Unexpected Mononuclear Catalyst

Abstract

The asymmetric Henry reaction of 2-acylpyridine N-oxide remains a challenge as N-oxides generally act as competitive catalyst inhibitors or displace activating ligands. A novel variable yield (up to 99%) asymmetric Henry reaction of 2-acypyridine N-oxides catalyzed by a Ni-aminophenol sulfonamide complex with good to excellent enantioselectivity (up to 99%) has been developed. Mechanistic studies suggest that the unique properties of the electron-pairs of N-oxides for complexation with Ni makes the unexpected mononuclear complex, rather than the previously reported dinuclear complex, the active species.

1. Introduction

The asymmetric construction of chiral quaternary stereocenters represents a considerable challenge in modern organic synthesis [1,2,3,4,5,6]. The Henry (nitroaldol) reaction [7,8,9,10,11,12,13] of ketones has become one of the most important and versatile reactions for the construction of quaternary carbons containing hydroxyl and nitro groups. In recent years, considerable effort has been devoted to the development of efficient chiral catalysts for asymmetric Henry reactions of reactive ketones, such as trifluoromethyl ketones (for selected examples, see ref [14,15,16,17,18]), α-keto esters (for selected examples, see [19,20,21,22,23,24]), α-keto amides (for selected examples, see [25,26]), α-keto-phosphonates [27,28], and glyoxal hydrates [29]. Although Matsunaga and Shibasaki reported the kinetic resolution of racemic derivatives [30], the asymmetric catalytic version of simple ketones has experienced little progress. At the same time, interest in pyridine derivatives has increased dramatically with the discovery of many bioactive compounds [31,32,33] and ligands containing pyridine rings [34,35,36,37,38,39]. Pedro and Blay first extended the Henry reactions to 2-acylpyridine N-oxides, which provided a convenient way for synthesizing β-amino tert-alcohols substructure bearing a quaternary stereocenter bonded to a 2-pyridyl moiety [40]. The asymmetric Henry reaction of 2-acylpyridine N-oxide remains a challenge as N-oxides generally act as competitive catalyst inhibitors or displace activating ligands (For examples of related asymmetric Henry reaction using N-oxides as ligands, see ref [22,41,42,43]). We recently reported an asymmetric Henry reaction of 2-acylpyridine N-oxides catalyzed by a pre-prepared Ni-PyBisulidine complex, and the corresponding results are not satisfactory [23]. Herein, we describe a Ni-aminophenol sulfonamide complex for the asymmetric Henry reaction of 2-acylpyridine N-oxides.

2. Results and Discussion

2.1. Catalytic Asymmetric Henry Reaction

In the preliminary study, the complexes prepared in situ from L1 (Figure 1) and different metal salts in a 1/2 molar ratio (for examples in asymmetric bimetallic catalysis based on aminophenol sulfonamide ligands, see [44,45,46]) were used to catalyze the asymmetric Henry reaction of 2-acylpyridine N-oxide and nitromethane, and Ni(OAc)₂ gave the best results (see the supplementary materials for details). However, in the subsequent molar ratio investigation of metal/ligand, it was found that 1/1 gave better enantioselectivity than 2/1 (Table 1, entry 1 vs. entry 3).

The ratio was investigated intensively with the results summarized in Table 1 (entries 1–7). It was found that increased metal ratio could increase the reactivity (Table 1, entries 1 and 2 vs. 3–7) and excess ligands provided higher ee (Table 1, entries 4–7 vs. 1–3). The best ratio of metal/ligand was 1/1.1 with 86% yield and 85% ee. We speculate that the excess metal could increase the amount of Ni/N-oxide complexes and Ni₂/L1 complexes [45], both of which are higher active species with lower selectivity. At the ratio of 1/1.1, a 1/1 coordination complex of Ni/L1 could be generated to the greatest extent. After the screening of benzenesulfonyl moiety, L2 was found to be the most promising ligand (Table 1, entries 8–14; Figure 1). The corresponding results of L9-L11 (Table 1, entries 15–17; Figure 1) showed both of the phenolic hydroxyl group and sulphonamide group played a key role in achieving high ee.

Next, different bases were examined, with the results shown in Table 2. The tertiary and secondary amines investigated showed excellent activity, except for N-methylmorpholine. The substituent size at the nitrogen atoms plays a key role in the selectivity and N,N-dicyclohexyl-methylamine gave the best results (Table 2, entry 2). On the other hand, the addition order of 2-acylpyridine N-oxide and nitromethane had an effect on the enantioselectivity and the addition of 2-acylpyridine N-oxide first was conducive to high ee (Table 2, entries 2 vs. 8).

With the optimized reaction conditions in hand (for more detailed results of optimization studies, such as solvents effect, substrate concentration and the amount of nitromethane, see the supplementary materials), the substrate scope was explored. The results are summarized in Table 3. The presence of 4- and 5-substituents (Me or Cl) on the pyridine ring did not affect the high activity and excellent selectivity (Table 3, entries 2, 3 and 5). The substituent of 5-Br provided an unexpectedly low yield with a good ee (Table 3, entry 6). The 6-position substituent on the pyridine ring greatly impairs the ee (Table 3, entry 4). The reaction between 3-methyl substituted substrate and CH₃NO₂ did not take place. This catalytic system is still effective for ethyl and propyl ketones (Table 3, entrys 7 and 8). The aromatic ketone afforded product 2i in good ee, albeit with moderate yield (Table 3, entry 9).

2.2. Mechanistic Studies of Ni-Aminophenol Sulfonamide Complex

The control experiments (Table 1, entries 1–7) indicated that the mononuclear system is important for high stereoselectivity and the addition of 2-acylpyridine N-oxide first was conducive to high ee (Table 2, 2 and 8). To gain some insight into the mechanism, the ESI-MS studies of the mixture of Ni(OAc)₂/L2 (1:1.1) and 1a were carried out (Figure 2, for more details, see supplementary materials). The spectrum displayed ions at m/z 1179, 1316, 1453, 1590, which corresponded to C1-C4 (Figure 3). This confirms the unique properties of the electron-pairs of N-oxides for complexation with Lewis acids [47,48,49]. In addition, there was a linear relationship between the enantiomeric excess of the Ni(OAc)₂-L2 (1:1.1) catalyst and product 2a (Figure 4). These results suggested that the active species in the present reaction would be a monomeric NiOAc-L2 catalyst. The proposed working model was illustrated in Figure 5 to rationalize the asymmetric induction. The keto functionality is coordinated to Ni in the more Lewis acidic equatorial position for maximal activation [50,51], whereas the nitronate generated by the base is positioned by the hydrogen bonding.

3. Experimental Section

3.1. General Information

Commercial reagents were used as purchased. NMR spectra (600 MHz, Bruker, Karlsruhe, Germany) were recorded in the deuterated solvents as stated, using residual non-deuterated solvent signals as the internal standard. High resolution mass spectra were recorded with a Bruker Solari XFT-ICR-MS system. The enantiomeric excess (ee) was determined by HPLC analysis (LC-16, Shimadzu, Suzhou, China) using the corresponding commercial chiral column as stated in the experimental procedures at 23 °C with UV detector. Optical rotations were measured on a commercial polarimeter (Autopol I, Rudolph, Hackettstown, NJ, USA) and are reported as follows: [α]_D^T (c = g/100 mL, solvent). The absolute configuration of 2a–2d, 2f, 2g and 2i were assigned by comparison with the sign of optical rotation value found in the literature. The absolute configuration of 2e and 2h was determined by analogy.

3.2. General Procedure for Catalytic Asymmetric Reaction

The mixture of Ni(OAc)₂·4H₂O (0.02 mmol, 10 mol%) and L2 (0.022 mmol, 11 mol%) was stirred in THF (0.5 mL) at 35 °C for 1 h. Then 2-acylpyridine N-oxide (0.2 mmol) and the base (0.04 mmol, 20 mol%) were added. The mixture was cooled to 0 °C. After stirring for 10 min at 0 °C, CH₃NO₂ (0.2 mL) and THF (0.3 mL) were added. The mixture was further stirred at 0 °C for the time indicated in Table 3. The resulting solution was purified by column chromatography (EtOH/AcOEt or petroleum ether/AcOEt) on silica gel to afford the products.

1-Methyl-2-nitro-1-(1-oxido-2-pyridinyl) ethanol (2a), brown oil, 99% yield, 94% ee; ¹H-NMR (CDCl₃) δ 8.26 (d, 1H, J = 6.4), 7.45–7.42 (m, 2H), 7.37–7.35 (m, 1H), 5.35 (d, 1H, J = 11.1), 4.82 (d, 1H, J = 11.2), 1.79 (s, 3H). $[\mathsf{\alpha }]{}_D^{20}$ = +57 (c 0.9, MeOH) [lit. [40] $[\mathsf{\alpha }]{}_D^{20}$ = +48 (c 0.9, MeOH) in 86% ee]; HPLC (CHIRALPAK AD-H column, Daicel, Osaka, Japan, hexane/2-propanol = 75/25, flow 1.0 mL/min, detection at 254 nm) t_r = 8.7 min (major) and t_r = 20.7 min (minor).

1-Methyl-2-nitro-1-(4-methyl-1-oxido-2-pyridinyl) ethanol (2b), brown solid, 99% yield, 99% ee; ¹H-NMR (CDCl₃) δ 8.26 (s, 1H), 8.16 (d, 1H, J = 6.6), 7.20 (s, 1H), 7.17 (d, 1H, J = 6.7), 5.47 (d, 1H, J = 11.0), 4.73 (d, 1H, J = 10.9), 2.42 (s, 3H), 1.81 (s, 3H). $[\mathsf{\alpha }]{}_D^{20}$ = +156 (c 0.4, MeOH) [lit. [40] $[\mathsf{\alpha }]{}_D^{20}$ = +41 (c 0.9, MeOH) in 84% ee]; HPLC (CHIRALPAK AD-H column, hexane/2-propanol = 80/20, flow 1.0 mL/min, detection at 254 nm) t_r = 8.2 min (major) and t_r = 32.8 min (minor).

1-Methyl-2-nitro-1-(5-methyl-1-oxido-2-pyridinyl) ethanol (2c), brown solid, 99% yield, 97% ee; ¹H-NMR (CDCl₃) δ 8.13 (s, 1H), 8.04 (s, 1H), 7.30–7.29 (m, 1H), 7.26–7.25 (m, 1H), 5.43 (d, 1H, J = 10.9), 4.73 (d, 1H, J = 10.9), 2.37 (s, 3H), 1.80 (s, 3H). $[\mathsf{\alpha }]{}_D^{20}$ = +181 (c 0.4, MeOH) in 97% ee [ lit. [40] $[\mathsf{\alpha }]{}_D^{20}$ = +60 (c 0.6, MeOH) in 81% ee]; HPLC (CHIRALPAK AD-H column, hexane/2-propanol = 75/25, flow 1.0 mL/min, detection at 254 nm) t_r = 12.4 min (major) and t_r = 18.3 min (minor).

1-Methyl-2-nitro-1-(6-methyl-1-oxido-2-pyridinyl) ethanol (2d), brown solid, 99% yield, 17% ee; ¹H-NMR (CDCl₃) δ 8.30 (s, 1H), 7.37–7.29 (m, 3H), 5.47 (d, 1H, J = 10.9), 4.73 (d, 1H, J = 11.0), 2.58 (s, 3H), 1.80 (s, 3H). $[\mathsf{\alpha }]{}_D^{20}$ = +21 (c 0.4, MeOH) in 17% ee [ lit. [40] $[\mathsf{\alpha }]{}_D^{20}$ = +109 (c 0.9, MeOH) in 55% ee]; HPLC (CHIRALPAK AD-H column, hexane/2-propanol = 80/20, flow 1.0 mL/min, detection at 254 nm) t_r = 7.7 min (major) and t_r = 11.1 min (minor).

1-Methyl-2-nitro-1-(4-chlorine -1-oxido-2-pyridinyl) ethanol (2e), brown solid, 99% yield, 92% ee; ¹H-NMR (CDCl₃) δ 8.20 (d, 1H, J = 6.9), 7.45 (d, 1H, J = 2.9), 7.41 (s, 1H), 7.36 (dd, 1H, J₁ = 6.9, J₂ = 2.8), 5.40 (d, 1H, J = 11.5), 4.85 (d, 1H, J = 11.5), 1.80 (s, 3H). ¹³C-NMR (150 MHz, CDCl₃) δ 150.6, 141.1, 134.9, 126.0, 125.4, 80.0, 72.4, 23.0. HRMS (ESI): m/z Calcd [C₈H₁₀ClN₂O₄]⁺ [M + H]⁺: 233.0324 (Cl³⁵), 235.0300 (Cl³⁷), Found 233.0323, 235.0290. $[\mathsf{\alpha }]{}_D^{20}$ = +52 (c 0.5, MeOH); HPLC (CHIRALPAK AD-H column, hexane/2-propanol = 75/25, flow 1.0 mL/min, detection at 254 nm) t_r = 6.0 min (major) and t_r = 14.2 min (minor).

1-Methyl-2-nitro-1-(5-bromo-1-oxido-2-pyridinyl) ethanol (2f), brown solid, 26% yield, 84% ee; ¹H-NMR (CDCl₃) δ 8.42 (d, 1H, J = 1.9), 7.57 (dd, 1H, J₁ = 8.6, J₂ = 1.8), 7.32 (d, 1H, J = 8.6), 5.39 (d, 1H, J = 11.4), 4.80 (d, 1H, J = 11.3), 1.79 (s, 3H). $[\mathsf{\alpha }]{}_D^{20}$ = +48 (c 0.3, MeOH) [lit. [40] $[\mathsf{\alpha }]{}_D^{20}$ = +74 (c 0.9, MeOH) in 89% ee]; HPLC (CHIRALPAK AD-H column, hexane/2-propanol = 80/20, flow 1.0 mL/min, detection at 254 nm) t_r = 9.5 min (major) and t_r = 10.7 min (minor).

1-Nitromethyl-1-(1-oxido-2-pyridinyl)propan-1-ol (2g), brown solid, 86% yield, 92% ee; ¹H-NMR (CDCl₃) δ 8.29 (d, 1H, J = 6.4), 7.46–7.44 (m, 2H), 7.38–7.36 (m, 1H), 5.31 (d, 1H, J = 11.4), 4.97 (d, 1H, J = 11.4), 2.28–2.22 (m, 1H), 2.12–2.05 (m, 1H), 1.09 (t, 3H, J = 7.4). $[\mathsf{\alpha }]{}_D^{20}$ = +64 (c 0.4, MeOH) [lit. [40] $[\mathsf{\alpha }]{}_D^{20}$ = +63 (c 1.2, MeOH) in 81% ee]; HPLC (CHIRALPAK AD-H column, hexane/2-propanol = 80/20, flow 1.0 mL/min, detection at 254 nm) t_r = 12.6 min (major) and t_r = 31.9 min (minor).

Nitromethyl-1-(1-oxido-2-pyridinyl) but-1-ol (2h), brown solid, 67% yield, 69% ee; ¹H-NMR (CDCl₃) δ 8.28 (d, 1H, J = 6.5), 7.47–7.43 (m, 2H, J = 12.3), 7.37–7.35 (m, 1H), 5.28 (d, 1H, J = 11.5), 5.02 (d, 1H, J = 11.4), 2.20–2.15 (m, 1H), 2.03–1.98 (m, 1H), 1.65–1.59 (m, 1H), 1.46–1.41 (m, 1H), 1.0 (t, 3H, J = 7.4).¹³C-NMR (150 MHz, CDCl₃) δ 148.2, 139.7, 126.9, 124.6, 124.5, 78.2, 73.8, 36.2, 15.1, 13.2. HRMS (ESI): m/z calcd for C₁₀H₁₄N₂NaO₄⁺ [M + Na]⁺: 249.0846, found 249.0840. $[\mathsf{\alpha }]{}_D^{20}$ = +67 (c 0.3, MeOH); HPLC (CHIRALPAK IA column, hexane/2-propanol = 85/15, flow 0.8 mL/min, detection at 254 nm) t_r = 16.2 min (major) and t_r = 19.2 min (minor).

1-(4-Chlorophenyl)-2-nitro-1-(1-oxido-2-pyridinyl)ethanol (2i), brown solid, 48% yield, 79% ee; ¹H-NMR (CDCl₃) δ 8.22 (d, 1H, J = 6.3), 7.55 (dd, 1H, J₁ = 8.1, J₂ = 1.8), 7.46 (t, 1H, J = 7.7), 7.43–7.41 (m, 2H), 7.39–7.36 (m, 3H), 5.44 (d, 1H, J = 12.7), 5.12 (d, 1H, J = 12.7). $[\mathsf{\alpha }]{}_D^{20}$ = +50 (c 0.2, MeOH) [lit. [40] $[\mathsf{\alpha }]{}_D^{20}$ = +55 (c 0.7, MeOH) in 90% ee]; HPLC (CHIRALPAK AD-H column, hexane/2-propanol = 80/20, flow 1 mL/min, detection at 254 nm) t_r = 13.4 min (major) and t_r = 17.9 min (minor).

4. Conclusions

We have developed a new mononuclear Ni-aminophenol sulphonamide complex for the asymmetric Henry reaction of 2-acylpyridine N-oxides. The simple experimental protocol affords various optically active pyridine-containing β-nitro tert-alcohols in variable yield (up to 99%) with good to excellent enantioselectivity (up to 99%). Mechanistic studies suggested that the unique properties of the electron-pairs of N-oxides for complexation with Ni makes the unexpected mononuclear complex, rather than the previously reported dinuclear complex, the active species.

Supplementary Materials

The following are available online: NMR spectra of L11, 2a–2i; HPLC of 2a–2i; ESI-MS analysis of Ni(OAc)₂/L2 (1:1.1) and 1a; more detailed results of optimization studies.

Author Contributions

H.Z. outlined the research strategy and idea. M.L. and D.G. performed the experiments; H.Z. and P.D. drafted and revised the manuscript. All authors read and approved the final manuscript.

Funding

This study was supported by the National Nature Science Foundation of China (21672031), Fundamental and Advanced Research Projects of Chongqing City (cstc2017jcyjAX0352).

Acknowledgments

We are grateful for the generous financial support by the National Nature Science Foundation of China (21672031), Fundamental and Advanced Research Projects of Chongqing City (cstc2017jcyjAX0352).

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Sample Availability: All data generated and/or compound samples analyzed during this study are included in this article and are available from the corresponding author on reasonable request.

Figures and Tables

Figure 1. Structures of ligands.

Figure 2. ESI-MS of Ni(OAc)2/L2/1a = 0.1/0.11/1.

Figure 3. The speculated structures of Ni/L2/1a according to the ESI-MS analysis.

Figure 4. Linear relationship between ee of L2 and ee of product 2a.

Figure 5. The proposed working model.

Table 1

Effect of the metal/ligand ratio and the ligand structure in the asymmetric Henry reaction ^a.

[Image omitted. Please see PDF.]
Entry	Ni(OAc)2 (x)	Ligand (y)	x/y	Yield (%) b	ee (%) c
1	20	L1 (10)	2/1	94	76
2	15	L1 (10)	1.5/1	98	81
3	10	L1 (10)	1/1	81	81
4	10	L1 (11)	1/1.1	86	85
5	10	L1 (12)	1/1.2	79	83
6	10	L1 (15)	1/1.5	72	84
7	10	L1 (20)	1/2	73	83
8	10	L2 (11)	1/1.1	99	91
9	10	L3 (11)	1/1.1	90	83
10	10	L4 (11)	1/1.1	91	76
11	10	L5 (11)	1/1.1	89	89
12	10	L6 (11)	1/1.1	50	4 d
13	10	L7 (11)	1/1.1	82	76
14	10	L8 (11)	1/1.1	92	92
15	10	L9 (11)	1/1.1	88	7
16	10	L10 (11)	1/1.1	69	5 d
17	10	L11 (11)	1/1.1	98	15

^a Reactions were carried out with 2-acylpyridine N-oxides (0.2 mmol) with i-Pr₂NEt (20 mol%) in a mixture of THF (0.8 mL) and CH₃NO₂ (0.2 mL) for 20 h. ^b Isolated yield. ^c Determined by chiral HPLC. ^d The absolute configuration of the major product was inverse compared with the others by the analysis of HPLC.

Table 2

Further optimization of the reaction ^a.

[Image omitted. Please see PDF.]
Entry	Base	Yield (%) b	ee (%) c
1	iPr2NEt	99	91
2	Cyhex2NMe d	99	94
3	Et3N	99	89
4	NMM e	87	86
5	iPr2NH	99	81
6	Bu2NH	99	88
7	Cyhex2NH f	99	88
8 g	Cyhex2NMe d	99	83

^a Reactions were carried out with 2-acylpyridine N-oxides (0.2 mmol) with base (20 mol%) in a mixture of THF (0.8 mL) and CH₃NO₂ (0.2 mL) for 15–20 h. ^b Isolated yield. ^c Determined by chiral HPLC. ^d N,N-Dicyclohexylmethylamine. ^e N-methylmorpholine. ^f Dicyclohexylamine. ^g Different reaction operation: the order of addition of nitromethane and 2-acylpyridine N-oxide was reversed. In the standard operation, 2-acylpyridine N-oxide was added to the complex prepared in situ for 10 min before the addition of nitromethane. For the detailed standard operation, see the experimental section.

Table 3

Substrate scope for the asymmetric Henry reaction ^a.

[Image omitted. Please see PDF.]
Entry	R1	R2	Product	Time (h)	Yield (%) b	ee (%) c
1	H	Me	2a	15	99	94
2	4-Me	Me	2b	15	99	99
3	5-Me	Me	2c	15	99	97
4	6-Me	Me	2d	24	99	17
5	4-Cl	Me	2e	17	99	92
6	5-Br	Me	2f	72	26	84
7	H	Et	2g	42	86	92
8	H	Bu	2h	42	67	69
9	H	4-ClC6H4	2i	72	48	79

^a Reactions were carried out with 2-acylpyridine N-oxides (0.2 mmol) withN,N-dicyclohexyl-methylamine (20 mol%) in a mixture of THF (0.8 mL) and CH₃NO₂ (0.2 mL). ^b Isolated yield. ^c Determined by chiral HPLC.