Highly enantioselective Pd-catalyzed indole allylic alkylation using binaphthyl-based phosphoramidite-thioether ligands

Abstract

A novel type of binaphthyl-based phosphoramidite-thioether ligand has been developed and successfully applied to Pd-catalysed asymmetric C-3 allylic alkylation of indoles, giving the desired products with up to 99% yield and 98% ee.

---

The privileged chiral indole scaffolds occur frequently in a wide variety of biologically active natural products and medicinally significant compounds.¹ As a result, enormous efforts have been directed toward the development of more practical and efficient protocols for their preparation and direct functionalization.² In this context, the asymmetric Friedel–Crafts alkylation reactions³ and allylic alkylation⁴ of indoles by using metal or organic catalysts have been established as the predominant approaches for regio- and enantioselective alkylation of indoles at the C-3 position.

In 2006, the group of Bandini and Umani-Ronchi reported an elegant Pd-catalysed intramolecular enantioselective indole alkylation by allylic carbonates using the Trost ligand.⁵ Since then, the design of chiral ligands for Pd-catalyzed intermolecular asymmetric allylic alkylation of indoles has attracted considerable attention from the synthetic community (Scheme 1). For example, the Chan research group in 2007 developed a novel class of chiral ferrocenyl P,S-ligands and disclosed the first example of highly enantioselective Pd-catalysed allylic alkylation of indoles with 1,3-diphenyl-2-propenyl acetate.⁶ Stimulated by this pioneering work, the research groups of Hoshi and Hagiwara,⁷ Du,⁸ and Mino⁹ respectively reported a series of structurally diverse phosphine- or phosphoramidite-based hybrid ligands for this benchmark reaction. Most of these ligands demonstrated generally high reaction efficiency and enantioselectivity. In particular, thioether-containing ligands are of great potential because an additional chirality on the sulfur atom would be generated upon its coordination to the metal, and proved to be beneficial for stereocontrol.¹⁰ Therefore, the development of other new, readily tunable and air-stable P,S-type ligands with high stereoselectivity and a broad substrate scope for the indole allylic alkylation is still of great interest.

Scheme 1 Representative hybrid ligands for Pd-catalysed allylic alkylation of indoles.

We have recently launched a research program aimed at exploring sulfur-containing chiral ligands based on the strategy of rational incorporation of two privileged backbones into one chiral scaffold.¹¹ Employing the chiral amino alcohol-derived thioether-amines of type 4 as the key intermediates, we have successfully developed a range of chiral ligands, sulfoxide–Schiff bases L1 and sulfoxide–phosphines L2 (Scheme 2). These ligands were found to be highly efficient and enantioselective in the Cu-catalysed Henry reaction¹² and Pd-catalysed allylic alkylation.¹³ In particular, the ligands L2 enabled an efficient Pd-catalyzed direct asymmetric N-allylic alkylation of indoles, giving the corresponding products with generally excellent regio- and enantioselectivities.^13c Encouraged by these results, we further explored the expansion of these catalytic systems to the C-3 allylic alkylation of indoles. However, only moderate yields and poor regioselectivity were observed when using ligands L2 (not shown). Inspired by the versatile applications of binaphthyl-based phosphoramidite ligands in asymmetric catalysis,¹⁴ we envisioned the development of a new class of phosphoramidite-thioether ligands using thioether-amines 4 and BINOL as the starting materials (Scheme 2). Compared to our previously reported ligands, their modular structure would enable facile construction of a ligand library and easy fine-tuning for improving the regio- and stereoselectivity of indole C3-allylic alkylation reaction. Herein, we wish to report our preliminary results on this subject.

Scheme 2 Development of sulfur-containing chiral ligands for asymmetric catalysis.

As shown in Scheme 3, the target ligands L3–L7 can be synthesized through two simple steps with good overall yield starting from thioether-amines 4, which are readily prepared on a gram-scale from commercially available (1R,2S)-2-amino-1,2-diphenylethan-1-ol according to our previous procedure.¹² Reductive amination of the aldehydes 5 with thioether-amines 4 gave rise to the corresponding secondary amines 6. Then, the reaction of (R)-BINOLs 7 with dichloroaminophosphine, which were in situ-formed by treatment of the amines 6 with phosphorus trichloride in the presence of a base, provided the final binaphthyl-based ligands L3–L7.¹⁵ Moreover, we also prepared the ligand L8 without the binaphthyl scaffold.¹⁶ Notably, all of these ligands are air- and moisture-stable, and could be manipulated in air. With these ligands in hand, we then attempted to apply them to the Pd-catalysed asymmetric allylic alkylation reaction of N-free indoles with allylic acetates.

Scheme 3 Modular synthesis of phosphoramidite-thioether ligands.

Based on our previous studies,¹³ we initially screened a series of commonly used bases with ligand L3 in the model reaction of indole 1a and rac-(E)-1,3-diphenylallyl acetate 2a in CH₂Cl₂ at 40 °C (Table 1).¹⁵ To our delight, the reaction with all of the bases gave rise to the corresponding product 3a with the same enantioselectivities and comparable yields; and Cs₂CO₃ demonstrated somewhat superior reaction efficiency (entries 1–4). With Cs₂CO₃ as the base, we subsequently examined ligands L4–L8 to investigate the effect of substitution patterns of the ligands on the activity and enantioselectivity. The results of entries 5–9 exhibited that the binaphthyl scaffold, N-substituent, and thioether moiety are all essential for the catalytic performance, with product 3aa being obtained with 60–97% yield and 30–94% ee. Among them, L6 was identified as the optimal ligand in terms of both yield and enantioselectivity (entry 7). With ligand L6, we briefly screened an array of typical solvents to further improve the enantioselectivity. It was found that the reaction proceeded more efficiently in aromatic solvents than that in halogen solvent, with 90% yield and 97% ee being obtained in mesitylene after 5 h (entry 12). By changing the molar ratio of 1a to 2a from 3 : 1 to 1 : 1.5, a slight increase of yield and the same enantioselectivity were observed (entry 13). Accordingly, the optimal reaction conditions were established as follows: 2 mol% of [Pd(C₃H₅Cl)]₂, 4 mol% of ligand L6, and 2.0 equivalents of Cs₂CO₃ in mesitylene at 40 °C.

Table 1 Condition optimization^a

Entry	Ligand	Solvent	Base	Time (h)	Yield^b (%)	ee^c (%)
a Unless otherwise noted, reactions were carried out with 1a (0.9 mmol), 2a (0.3 mmol), [Pd(C3H5)Cl]2 (2.0 mol%), ligand (4.0 mol%), base (0.6 mmol) in solvent (2.0 mL) at 40 °C. b Isolated yield. c The ee values were determined by chiral HPLC analysis, the absolute configuration was established to be S according to ref. 6. d A mixture of BSA (2.0 equiv.) and LiOAc (0.1 equiv.) was added as the base. e A mixture of BSA (2.0 equiv.) and KOAc (0.1 equiv.) was added as the base. f Performed with 1a (0.3 mmol) and 2a (0.45 mmol).
1	L3	CH2Cl2	Cs2CO3	10	87	71
2	L3	CH2Cl2	K2CO3	18	82	71
3^d	L3	CH2Cl2	BSA + LiOAc	24	85	71
4^e	L3	CH2Cl2	BSA + KOAc	24	88	71
5	L4	CH2Cl2	Cs2CO3	24	97	89
6	L5	CH2Cl2	Cs2CO3	24	84	83
7	L6	CH2Cl2	Cs2CO3	24	90	94
8	L7	CH2Cl2	Cs2CO3	24	60	30
9	L8	CH2Cl2	Cs2CO3	10	85	35
10	L6	Toluene	Cs2CO3	5	88	96
11	L6	Xylenes	Cs2CO3	5	89	97
12	L6	Mesitylene	Cs2CO3	5	90	97
13	L6	Mesitylene	Cs 2 CO 3	5	92	97

---

Next, we explored the substrate scope with a variety of indoles 1b–m and several allylic acetates 2 (Table 2). The reaction with 2-substituted indoles 1b–c proceeded smoothly to give the expected products 3b and 3c with 70–82% yield and 82–95% ee. In contrast to the Hoshi and Hagiwara catalytic system,⁷ it turned out that the electronic properties of the substituents at the phenyl ring of the indoles have no obvious effect on the results. For example, C4-, C5-, C6-, and C7-substituted indoles 1d–l bearing electron-donating (Me, BnO, MeO) or electron-withdrawing (Br, Cl) functional groups were all well tolerated, and the corresponding products 3d–l were obtained in the range of 72–99% yield and 86–98% ee. The multisubstituted indole 1m also underwent the allylic alkylation reaction efficiently to furnish product 3m in 99% yield with 98% ee. Moreover, rac-(E)-1,3-diaryl-2-propenyl acetate 2b could also react with 1a efficiently to give product 3n in 73% yield with 96% ee. In the case of rac-(E)-4-phenylbut-3-en-2-yl acetate 2c and rac-(E)-4-(pyridin-4-yl)but-3-en-2-yl acetate 2d, the reaction still proceeded smoothly to afford products 3o and 3p with good yield and regioselectivity, while with only moderate enantioselectivity (up to 52% ee). In the current catalytic system, however, N-Me- and N-Boc-substituted indoles proved to be not suitable for the reaction (not shown). Remarkably, in all of these reactions, no undesired N-allylic alkylation was observed.¹⁷

Table 2 Substrate scope^a,b,c

a Unless otherwise noted, reactions were carried out with 1 (0.3 mmol), 2 (0.45 mmol), [Pd(C₃H₅)Cl]₂ (2.0 mol%), ligand L6 (4.0 mol%), Cs₂CO₃ (0.6 mmol) in mesitylene (2.0 mL) at 40 °C. b Isolated yield. c The ee values were determined by chiral HPLC analysis.

---

In addition to indole nucleophiles, the current catalytic system could also be successfully extended to enantioselective allylic etherification and amination (Scheme 4). For instance, in the case of benzyl alcohol and benzyl amine as nucleophiles, the reactions completed within 45 minutes to give the corresponding allylic ether 9 and allylic amine 11 with 98% and 97% ee, respectively. These results also demonstrated the importance of incorporation of the privileged BINOL scaffold as an additional stereocontrol element in our newly developed phosphoramidite-thioether ligands.¹³

Scheme 4 Pd-Catalysed enantioselective allylic etherification and amination.

In conclusion, we have developed a new class of readily tunable and air-stable binaphthyl-based phosphoramidite-thioether ligands. These ligands can be successfully applied to the Pd-catalysed asymmetric allylic alkylation reaction of indoles at the C3-position, giving the correspondingly products with generally good yields (up to 99%) and enantioselectivities (up to 98% ee). Further exploration of these ligands in other catalytic asymmetric reactions for the formation of carbon–carbon and carbon–heteroatom bonds is ongoing in our laboratory.

We are grateful to the National Natural Science Foundation of China (no. 21272087, 21202053, 21472058, 21472057, and 21232003) and the Youth Chen-Guang Project of Wuhan (no. 2015070404010180) for support of this research.

Notes and references

1. For selected reviews, see: (a) J. E. Saxton, Nat. Prod. Rep., 1997, 14, 559; (b) S. E. O'Connor and J. Maresh, Nat. Prod. Rep., 2006, 23, 532; (c) M. Ishikura, K. Yamada and T. Abe, Nat. Prod. Rep., 2010, 27, 1630; (d) A. J. Kochanowska-Karamyan and M. T. Hamann, Chem. Rev., 2010, 110, 4489; (e) S. Lancianesi, A. Palmieri and M. Petrini, Chem. Rev., 2014, 114, 7108; (f) L. Wang, Y.-Y. Chen and J. Xiao, Asian J. Org. Chem., 2014, 3, 1036.
2. For selected reviews, see: (a) S. Cacchi and G. Fabrizi, Chem. Rev., 2005, 105, 2873; (b) M. Bandini and A. Eichholzer, Angew. Chem., Int. Ed., 2009, 48, 9608; (c) G. Bartoli, G. Bencivenni and R. Dalpozzo, Chem. Soc. Rev., 2010, 39, 4449; (d) M. Shiri, Chem. Rev., 2012, 112, 3508; (e) R. Dalpozzo, Chem. Soc. Rev., 2015, 44, 742.
3. For selected reviews, see: (a) M. Bandini, A. Melloni and A. Umani-Ronchi, Angew. Chem., Int. Ed., 2004, 43, 550; (b) A. Umani-Ronchi, M. Bandini, A. Melloni and S. Tommasi, Synlett, 2005, 1199; (c) T. B. Poulsen and K. A. Jørgensen, Chem. Rev., 2008, 108, 2903; (d) S.-L. You, Q. Cai and M. Zeng, Chem. Soc. Rev., 2009, 38, 2190.
4. For selected reviews, see: (a) B. M. Trost and D. L. Van Vranken, Chem. Rev., 1996, 96, 395; (b) B. M. Trost and M. L. Crawley, Chem. Rev., 2003, 103, 2921; (c) Z. Lu and S.-M. Ma, Angew. Chem., Int. Ed., 2008, 47, 258; (d) B. M. Trost, T. Zhang and J. Sieber, Chem. Sci., 2010, 1, 427.
5. M. Bandini, A. Melloni, F. Piccinelli, R. Sinisi, S. Tommasi and A. Umani-Ronchi, J. Am. Chem. Soc., 2006, 128, 1424.
6. H. Y. Cheung, W.-Y. Yu, F. L. Lam, T.-T. Au-Yeung, Z.-Y. Zhou, T. H. Chan and A. S. Chan, Org. Lett., 2007, 9, 4295.
7. T. Hoshi, K. Sasaki, S. Sato, Y. Ishii, T. Suzuki and H. Hagiwara, Org. Lett., 2011, 13, 932.
8. (a) Z. Liu, Z. Cao and H.-F. Du, Org. Biomol. Chem., 2011, 9, 5369; (b) Y. Liu and H.-F. Du, Org. Lett., 2013, 15, 740; (c) Z. Cao, Y. Liu, Z. Liu, X. Feng, M. Zhuang and H.-F. Du, Org. Lett., 2011, 13, 2164.
9. T. Mino, M. Ishikawa, K. Nishikawa, K. Wakui and M. Sakamoto, Tetrahedron: Asymmetry, 2013, 24, 499.
10. (a) T. Toru and C. Bolm, Organosulfur Chemistry in Asymmetric Synthesis, Wiley-VCH, Weinheim, 2008; (b) H. Pellissier, Chiral Sulfur Ligands: Asymmetric Catalysis, Royal Society of Chemistry, Cambridge, U.K., 2009; (c) M. Mellah, A. Voituriez and E. Schulz, Chem. Rev., 2007, 107, 5133; (d) F. L. Lam, F. Y. Kwong and A. S. Chan, Chem. Commun., 2010, 46, 4649.
11. (a) T. P. Yoon and E. N. Jacobsen, Science, 2003, 299, 1691; (b) Q.-L. Zhou, Privileged Chiral Ligands and Catalysts, Wiley-VCH, Weinheim, 2011.
12. (a) H.-G. Cheng, L.-Q. Lu, T. Wang, J.-R. Chen and W.-J. Xiao, Chem. Commun., 2012, 48, 5596; (b) L.-Y. Chen, J.-R. Chen, H.-G. Cheng, L.-Q. Lu and W.-J. Xiao, Eur. J. Org. Chem., 2014, 4714.
13. (a) H.-G. Cheng, B. Feng, L.-Y. Chen, W. Guo, X.-Y. Yu, L.-Q. Lu, J.-R. Chen and W.-J. Xiao, Chem. Commun., 2014, 50, 2873; (b) B. Feng, H.-G. Cheng, J.-R. Chen, Q.-H. Deng, L.-Q. Lu and W.-J. Xiao, Chem. Commun., 2014, 50, 9550; (c) L.-Y. Chen, X.-Y. Yu, J.-R. Chen, B. Feng, H. Zhang, Y.-H. Qi and W.-J. Xiao, Org. Lett., 2015, 17, 1381.
14. For selected reviews, see: (a) B. L. Feringa, Acc. Chem. Soc., 2000, 33, 346; (b) L. Eberhardt, D. Armspach, J. Harrowfield and D. Matt, Chem. Soc. Rev., 2008, 37, 839; (c) J. F. Teichert and B. L. Feringa, Angew. Chem., Int. Ed., 2010, 49, 2486; (d) W.-H. Zhang, S. W. Chien and T. S. A. Hor, Coord. Chem. Rev., 2011, 255, 1991.
15. See the ESI† for more details. We also tried to prepare the diastereomer of ligand L3 using (S)-BINOL and thioether-amines 4a (R¹ = Me) as the starting materials. Unfortunately, all the attempts met failure using our own method or previous literature procedures, which was probably because of the steric hindrance.
16. Y. Wei, L.-Q. Lu, T.-R. Li, B. Feng, Q. Wang, W.-J. Xiao and H. Alper, Angew. Chem., Int. Ed., 2016, 55, 2200.
17. M. Bandini, A. Melloni and A. Umani-Ronchi, Org. Lett., 2004, 6, 3199.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6qo00227g

---