Xubin
Wang
a,
Xiaoming
Wang
a,
Zhaobin
Han
a,
Zheng
Wang
*a and
Kuiling
Ding
*abc
aState Key Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 345 Lingling Road, Shanghai 200032, China. E-mail: kding@mail.sioc.ac.cn; Fax: +(21)-6416-6128
bUniversity of Chinese Academy of Sciences, Beijing 100049, China
cCollaborative Innovation Center of Chemical Science and Engineering, Tianjin 300071, China
First published on 29th November 2016
Spiroketal backbone based diphosphine ligands (SKP) were disclosed to be highly efficient and enantioselective (94 → 99% ee) in the palladium catalyzed asymmetric allylic amination of 2-diethylphosphonate-substituted allylic acetates, affording a series of chiral β-aminophosphonates bearing an α-methylene functionality in high yields with excellent regioselectivities.
Recently, our group has reported the development of spiroketal-based chiral diphosphine ligands (SKP),15 a new class of diphosphines with sterically well-defined spiro backbones.16 The SKP ligands were found to be highly efficient in the Pd catalyzed asymmetric allylic amination17 of racemic ethyl 2-(acetoxy(phenyl)methyl)acrylates, a type of Morita–Baylis–Hillman (MBH) adduct,18 to give the corresponding β-arylamino acid esters with high regio- and enantioselectivities.19 Kinetic and mechanistic studies indicated that the unusual long distance of the two P atoms in the SKP ligand allows for its unique role in the reaction, i.e. the ligand adopts an organo- and organometallic bifunctional mode in the cooperative catalysis.20 Encouraged by these results, we sought to extend the catalytic system to the asymmetric allylic amination of 2-(diethylphosphonyl)-substituted allylic acetates. The expected amination products would be enantioenriched β-aminophosphoric acid derivatives which can be viewed as the bioisosteres of α-methylene-β-amino acids that have very recently been found to be a key unnatural amino acid unit in a new class of endomorphin-1 analogues with potent antinociceptive activity.21 Furthermore, the olefin functionality present in the amination products may constitute a useful handle for further synthetic manipulation, thus giving ready access to a wider array of β-amino phosphoric acid derivatives.
The study was initiated by a survey of the reaction conditions, including variations in palladium sources and SKP ligands, catalyst loadings, solvents, and bases, for the amination of 2-(diethylphosphonyl)-substituted allylic acetate (2a) with aniline (3a) as the nucleophile. The reactions were generally conducted at room temperature for 0.5 h, using the complex generated in situ from a SKP ligand [(S,S,S)-1a–e] and a palladium precursor as the catalyst. The effects of solvents and bases on the reaction of 2a and 3a were examined in the presence of the [Pd(η3-C3H5)Cl]2 (1.0 mol%)/(S,S,S)-1a (2.5 mol%) catalyst, indicating that both parameters have a significant impact on the reactivity as well as the chemo-, regio-, and enantioselectivities (for details, see Table S1 in the ESI†). In these cases, the reaction was found to be best performed in dichloromethane in the presence of two equivalents of anhydrous K2CO3 as the base, to afford the chiral allylic amination product 4a in 95% yield with excellent chemo-, regio-, and enantioselectivities (4a/5a/6a = >98/<2/0, >99% ee, entry 1 in Table 1). Under these optimized reaction conditions, the effects of catalyst compositions and loadings on the reaction of 2a and 3a were further evaluated, using SKP ligands (S,S,S)-1a–e with subtle variations in their aryl substituents at the P atoms and a couple of Pd precursors. The results are summarized in Table 1. With [Pd(η3-C3H5)Cl]2 (1.0 mol%) as the palladium precursor, a sharp difference in catalytic behavior was observed among the SKP ligands (S,S,S)-1a–e (entries 1–5). For example, high activity and excellent regio-/enantioselectivities were obtained using ligand 1a, 1c, or 1e with phenyl, 3,5-xylyl or 4-methoxyphenyl substituents, respectively, affording the targeted product 4a in high yields (92–95%) with 98–99% ee values (entries 1, 3, and 5). In contrast, ligand 1b possessing 2-tolyl moieties on the P atoms obviously deteriorates the reactivity and regioselectivity, leading to only very poor conversion (5%) and a modest branched/linear regioselectivity (4a/5a = 2/3) under otherwise identical conditions (entry 2). Intriguingly, 4-tolyl-bearing ligand 1d, with structural features analogous to both 1a and 1e, afforded much inferior chemoselectivity albeit with a 98% ee for 4a (entry 4), presumably as a result of incomplete amination of the isomerization product 6a within 0.5 h. Intriguingly, the reaction results with some privileged chiral ligands,16e.g., (R)-BINAP, (R)-SDP or (R,R)-Trost ligand, were less satisfactory under the otherwise identical conditions, affording incomplete conversions and moderate chemo-, regio- and enantioselectivities (entries 6–8). These facts clearly indicated that SKP ligands demonstrate unique performance in the catalysis of this type of asymmetric transformation. With 1a as the ligand, the use of different palladium precursors also resulted in distinct catalytic activities and selectivities (entries 1 and 9–11). While the use of Pd2(dba)3 delivers excellent results nearly identical to those of [Pd(η3-C3H5)Cl]2 (entries 9 vs. 1), Pd(OAc)2 or Pd(CH3CN)2Cl2 turns out to be much less efficient, realizing only partial conversion of 2a (entry 7) and lower yields of 4a (entries 10 and 11), or a substantial amount of the unreacted isomerization product 6a (entry 11). Further trials to lower the catalyst loadings were thus performed using either [Pd(η3-C3H5)Cl]2 or Pd2(dba)3 along with ligand 1a as the catalyst, and the reaction times were prolonged to 3 h (entries 12–16). Under these conditions, the loading of [Pd(η3-C3H5)Cl]2 was lowered to 0.5 mol% without loss of either yield of selectivities (entry 12), whereas further decreasing the loading to 0.1 mol% resulted in partial conversion and declined yield (entry 13). In this context, Pd2(dba)3 seems to be superior as the palladium precursor, and its loading can be lowered all the way to 0.1 mol% with essentially no changes in yields or ee values of 4a (entries 14 and 15 vs. 9). Further lowering of the Pd2(dba)3 loading to 0.05 mol%, however, led to a significant decrease in the reactivity albeit still with a 98% ee value for 4a (entry 16).
Entry | [Pd]b (X mol%) | Ligandb | Conv.c (%) | 4a/5a/6ac | Yieldd (%) | eee (%) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
a Unless otherwise noted, the reaction was performed with 2a (0.2 mmol) and 3a (0.4 mmol), K2CO3 (0.4 mmol) in CH2Cl2 (2 mL) at rt for 0.5 h. b The molar percent of the Pd salt relative to that of 2a. In each case, the loading of the SKP ligand was 1.25 equiv. relative to that of Pd. c Determined by 1H NMR spectroscopy. d Yield of the isolated 4a. e The ee value of 4a was determined by chiral HPLC. f The loading of 1a was 2.5 mol% relative to that of 2a. g The reactions were run for 3 h. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 | [Pd(C3H5)Cl]2 (1) | 1a | >99 | >98/<2/0 | 95 | >99 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
2 | [Pd(C3H5)Cl]2 (1) | 1b | 5 | 2/3/0 | — | — | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
3 | [Pd(C3H5)Cl]2 (1) | 1c | >99 | 93/2/5 | 92 | >99 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
4 | [Pd(C3H5)Cl]2 (1) | 1d | >99 | 68/2/30 | 63 | 98 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
5 | [Pd(C3H5)Cl]2 (1) | 1e | >99 | >98/<2/0 | 94 | 98 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
6 | [Pd(C3H5)Cl]2 (1) | 1f | 34 | 31/69/0 | 9 | 43 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
7 | [Pd(C3H5)Cl]2 (1) | 1g | 16 | 70/30/0 | 10 | 4 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
8 | [Pd(C3H5)Cl]2 (1) | 1h | 57 | 56/34/10 | 28 | 10 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
9 | Pd2(dba)3 (1) | 1a | >99 | >98/<2/0 | 95 | 98 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
10f | Pd(OAc)2 (2) | 1a | 36 | 36/0/64 | 12 | 88 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
11f | Pd(CH3CN)2Cl2 (2) | 1a | >99 | 44/6/50 | 43 | 95 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
12g | [Pd(C3H5)Cl]2 (0.5) | 1a | >99 | >98/<2/0 | 95 | 98 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
13g | [Pd(C3H5)Cl]2 (0.1) | 1a | 87 | 89/0/11 | 82 | 97 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
14g | Pd2(dba)3 (0.2) | 1a | >99 | >98/<2/0 | 95 | 98 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
15g | Pd2(dba)3 (0.1) | 1a | >99 | >98/<2/0 | 94 | 98 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
16g | Pd2(dba)3 (0.05) | 1a | 64 | 94/0/6 | 57 | 98 |
Subsequently, we proceeded to examine the substrate scope of the catalysis by variation in both the 2-diethylphosphonate-substituted allylic acetates (2a–h) and nucleophilic amines (3a–j). The reactions were run under the optimized conditions with a low loading of Pd2(dba)3 (0.1–0.5 mol%) and (S,S,S)-1a, and the results are summarized in Table 2. Gratifyingly, excellent enantioselectivities (94 → 99% ee) were observed in the resultant β-aminophosphonates 4a–q (entries 1–17). Both electron-donating and electron-withdrawing groups on the phenyl rings, located whether on the aromatic amine or on the allylic acetate, are well tolerated. The regioselectivities for the amination products (4/5) are also generally high, ranging from 90/10 to >98/2 (entries 1–16). The reaction involving substrate 2h was an exception (entry 17), however, giving a much higher content of the linear amination product (4q/5q = 61/39) and a moderate yield (40%) of 4q even at a relatively high loading of the catalyst (1.0 mol%), probably as a result of unfavorable interaction with the Pd catalyst caused by the sterically congested o-tolyl group in 2h. It is also noteworthy that the stereoelectronic properties of the aromatic amines have no obvious influence on the catalysis, as reactions of 2a with a range of anilines (3a–i) gave the corresponding products 4a–i in comparable good yields, high regioselectivities and excellent enantioselectivities (entries 1–9). The amination of 2a also proceeded smoothly with benzylamine 3j, an aliphatic nucleophile, to furnish β-aminophosphonate 4j in 84% yield with a 90:
10 branched/linear ratio and 98% ee (entry 10). Finally, the absolute configuration of 4d was unambiguously established to be R by the X-ray crystal diffraction analysis (Fig. 2), while those for other products were deduced to be all R by comparison of their Cotton effects with that of (R)-(−)-4d as shown in the CD spectra (Fig. S2, ESI†).
Entry | 4 | X | 4/5b | Yieldc (%) | eed (%) | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
a Unless otherwise noted, the reactions were typically performed at rt with 2 (2.0 mmol) and 3 (4.0 mmol), K2CO3 (4.0 mmol) in CH2Cl2 (20 mL) for 3 h, in the presence of a specified amount of catalysts Pd2(dba)3 and (S,S,S)-1a. b Determined by 1H NMR spectroscopy. c Yield of the isolated 4a–q. d The ee values of 4a–q were determined by chiral HPLC. The absolute configurations for 4a–q were all determined to be R (see text). | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 |
![]() |
0.1 | >98/<2 | 94 | 98 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
2 |
![]() |
0.1 | 95/5 | 91 | 95 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
3 |
![]() |
0.1 | >98/2 | 94 | 96 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
4 |
![]() |
0.1 | 91/9 | 87 | 98 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
5 |
![]() |
0.1 | 95/5 | 83 | 94 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
6 |
![]() |
0.1 | 93/7 | 88 | 96 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
7 |
![]() |
0.1 | 96/4 | 70 | 98 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
8 |
![]() |
0.1 | 96/4 | 84 | 98 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
9 |
![]() |
0.1 | >98/2 | 89 | 98 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
10 |
![]() |
0.1 | 90/10 | 84 | 98 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
11 |
![]() |
0.5 | 93/7 | 75 | 94 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
12 |
![]() |
0.5 | 96/4 | 92 | 97 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
13 |
![]() |
0.5 | 94/6 | 80 | 94 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
14 |
![]() |
0.5 | 92/8 | 70 | 96 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
15 |
![]() |
0.5 | 97/3 | 84 | 98 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
16 |
![]() |
0.5 | 98/2 | 75 | >99 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
17 |
![]() |
1 | 61/39 | 40 | 94 |
The method can be applied in the Gram-scale preparation of β-aminophosphonates 4a under a reduced catalyst loading. By following the above mentioned procedure, the reaction of 2a (4.0 mmol, 1.24 g) with 3a (740 mg, 8.0 mmol) proceeded smoothly at rt for 8 h in dichloromethane (38 mL) in the presence of Pd2(dba)3 (7.3 mg, 0.008 mmol), (S,S,S)-1a (13.2 mg, 0.02 mmol), and K2CO3 (1.1 g, 8.0 mmol), to give branched amination product 4a (1.17 g, 85% yield) with 98% ee.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 1012761. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6qo00597g |
This journal is © the Partner Organisations 2017 |