Regio- and enantioselective iridium-catalysed allylic aminations and alkylations of dienyl esters

Abstract

Regio- and enantioselective iridium-catalysed allylic aminations and alkylations of dienyl substrates are presented; using phosphorus amidite L1 as ligand, aminations provided ee values of up to 97% and alkylations of up to 90%.

---

Pd-catalysed allylic substitutions are of great importance for organic synthesis. The asymmetric variant is highly developed for symmetrically substituted substrates.¹ Except for a small number of examples,² the particularly easily accessible monosubstituted allylic derivatives 1 (Scheme 1) preferentially yield the achiral linear products 3 in Pd-catalysed substitutions. This is different for reactions catalysed by complexes of Ir, Mo, and Ru, which mainly give the branched product 2.^3–5

Scheme 1

In 1997 our group developed the first enantioselective Ir-catalysed allylic substitutions, making use of phosphinooxazoline L4 as ligand, to give branched products with very high regio- and enantioselectivity in the case 1, R = aryl.⁶ Later we discovered that phosphorus amidites (cf. Fig. 1) derived from BINOL are particularly effective ligands in enantioselective alkylations and aminations of both arylated and alkylated allylic derivatives.^7,8 Particularly high degrees of enantio- and regioselectivity were very recently achieved by Hartwig et al. for aminations⁹ and etherifications¹⁰ of substrates 1, R = aryl. Alkylations using chiral phosphites, e.g.L3, as ligands were reported by Fuji¹¹ and Takemoto.¹² In general, high degrees of regioselectivity have so far only been obtained with arylated allylic substrates.

Fig. 1 Chiral ligands used in allylic substitutions.†

We have now explored the asymmetric Ir-catalysed substitutions of dienyl esters (Scheme 2).‡^,13 Only alkylations have been reported for these substrates, yielding mainly branched products 8 using either achiral Ir complexes³ or chiral Mo complexes as catalysts, which allowed alkylations to be accomplished with excellent regio- and enantioselectivity.¹⁴ It was a challenge to devise the first asymmetric, regioselective amination of dienyl substrates. The value of the highly functionalised products 6 for the synthesis of biologically active compounds is obvious.

Scheme 2

Our work began with the readjustment of reaction conditions. In previous work^8b we used sodiomalonates as nucleophiles in conjunction with acetates and with LiCl as an additive at relatively low concentrations (0.12 M) in THF. We have now found that with lithiomalonates in combination with carbonates at high concentration (0.29 M)§ a higher level of regioselectivity can be reached. Thus, the reaction of 1b with dimethyl lithiomalonate, using ligand L1, gave 2a with a regioselectivity of 2a ∶ 3a = 95 ∶ 5 and ee of 86% for (R)-2a, while previously, with 1a, the result was 2a ∶ 3a = 90 ∶ 10 and 86% ee.^8b

Table 1 Ir-catalysed allylic amination and alkylation of carbonates 4 and acetates 5 according to Scheme 2 (E = COOMe)³

Entry	Substrate	Ligand	Nucleophile	t ^a/h	Yield (%)^b of 6 + 7 or 8 + 9	Ratio^c^,^d6 ∶ 7 or 8 ∶ 9	ee (%)^e (Confign.)
a Reaction time. b Yield of isolated product. c The product of ε-attack was not observed. d Determined by ^1H NMR of the crude products. e Determined by HPLC (Daicel columns, 250 × 4.6 mm, 5 µm, + guard cartridge 10 × 4 mm, 5 µm, flow: 0.5 ml min^−1); 6a: (Daicel Chiralcel OD-H, eluent: n-hexane–i-PrOH 99 ∶ 1 + 0.1% HNEt2): tR[(R)-6a] = 21 min, tR[(S)-6a)] = 27 min; 6b: determination after transformation to 6c; 6c: (Daicel Chiralpak AD-H, 20 °C, eluent: n-hexane–i-PrOH 99.5 ∶ 0.5 + 0.1% HNEt2): tR[(−)-6c] = 79 min, tR(+) = 111 min; 8a: (Daicel Chiralcel OJ-H, eluent: n-hexane–i-PrOH 90 ∶ 10): tR[(−)-8a] = 26 min, tR[(+)-8a] = 29 min; 8b: determination after transformation to 8c; 8c: (Daicel Chiralcel AD-H, eluent: n-hexane–i-PrOH 98 ∶ 2): tR[(+)-8b] = 45 min, tR[(−)-8b] = 53 min. f Ratio Ir ∶ ligand = 1 ∶ 2. g Reaction temperature: 50 °C. h Addition of 1 eq. of LiCl or ZnCl2 to the catalyst solution.
1	4a	L1	BnNH2	24	61	99 ∶ 1	6a: 97 (+)(S)
2	4b	L1	BnNH2	48	72	94 ∶ 6	6b: 97 (−)
3	4c	L1	BnNH2	14	71	96 ∶ 4	6c: 97 (−)
4	4a	L2	LiCHE2	23	76	95 ∶ 5	8a: 0
5	5a	L2 ^f	LiCHE2	40	40	99 ∶ 1	8a: 80 (+)
6	5a	L2 ^f	NaCHE2	12	88	99 ∶ 1	8a: 50 (+)
7	4a	L1	LiCHE2	24^g	62	99 ∶ 1	8a: 76 (−)
8	4a	L1	LiCHE2–LiCl^h	43^g	63	79 ∶ 21	8a: 66 (−)
9	4a	L1	LiCHE2–ZnCl2^h	43^g	77	95 ∶ 5	8a: 78 (−)
10	4a	L1	NaCHE2–LiCl^h	22^g	87	94 ∶ 6	8a: 90 (−)
11	5a	L1	LiCHE2	43^g	49	91 ∶ 9	8a: 84 (−)
12	5a	L1	NaCHE2–LiCl^h	23^g	76	92 ∶ 8	8a: 89 (−)
13	4a	L3	LiCHE2	17	64	94 ∶ 6	8a: 58 (+)
14	4a	L3	NaCHE2	72	84	95 ∶ 5	8a: 10 (−)
15	4b	L1	LiCHE2	24^g	84	94 ∶ 6	8b: 87 (−)

---

The results obtained for both arylated and alkylated dienyl esters as substrates are presented in Table 1.¹⁵ In general, ligand L1 furnished better results than L2–L4. While L2 had previously given good results in the alkylations of acetates, surprisingly, it was completely ineffective in the substitutions of carbonates 4 (entry 4). Similarly, acetates 5 did not undergo aminations with any of the ligands L1–L4. However, the aminations of carbonates 4a–4c gave excellent results: 97% ee and regioselectivities in the range of >94 ∶ 6 (entries 1–3). Aminations were carried out at a concentration of ca. 2.0 M in THF. Rates were higher in toluene, CH₂Cl₂ or benzylamine (reaction time 10 min), however, enantioselectivities were lower in these solvents than in THF.

Alkylations were less sensitive to the choice of leaving group and ligand. Thus, under the control of ligand L2 acetate 5a reacted with a regioselectivity of 99 ∶ 1 but only moderate enantioselectivity (entries 5,6). For reactions in the presence of ligand L1, additives and the counter cation of malonate were found to be important (see above). Addition of LiCl to the catalyst and use of sodium dimethyl malonate as nucleophile proved particularly effective (cf. entries 10,12). The phosphite L3 gave somewhat disappointing results (entries 13,14).

The absolute configuration was determined for amine 6a as (+)-(S) by transformation to sulfonamide 10 (Scheme 3), which yielded crystals suitable for X-ray crystal structure analysis.¹⁶ The steric course of the Ir-catalysed substitution reaction at 4a using ligand L1 is the same as previously found for a variety of substitutions carried out with substrates 1.^8,9 In conjunction with the lack of substitution at the ε-position, this indicated that reactions of the dienyl substrates involve (allyl)Ir complexes centred at Cα–Cγ.

Scheme 3 Reagents and conditions: (i) p-Br(C₆H₄)SO₂Cl, NEt₃, CH₂Cl₂, 0 °C, 78%.

In conclusion, we have shown that dienyl carbonates are suitable substrates in enantioselective Ir-catalysed allylic aminations and alkylations with phosphorus amidites as ligands. For both aryl- and alkyl-substituted substrates very high regioselectivities in favour of the desired internal substitution products were obtained. Aminations afforded up to 97% ee and alkylations up to 90% ee. The absolute configuration of the reaction product 6a was determined.

This work was supported by the Deutsche Forschungsgemeinschaft (SFB 623) and the Fonds der Chemischen Industrie. We thank Michael Huwe and Markus Krauter for the preparation of substrates 4a, 4c and 5a, Dr F. Rominger for the crystal structure and Degussa AG for the iridium salts.

Notes and references

1. (a) B. M. Trost and C. Lee, in Catalytic Asymmetric Synthesis, 2nd edn., ed. I. Ojima, Wiley-VCH, New York, 2000, p. 593; (b) A. Pfaltz and M. Lautens, in Comprehensive Asymmetric Catalysis I-III, eds. E. N. Jacobsen, A. Pfaltz and H. Yamamoto, Springer, Berlin, 1999, p. 833.
2. S.-L. You, X.-Z. Zhu, Y.-M. Luo, X.-L. Hou and L.-X. Dai, J. Am. Chem. Soc., 2001, 123, 7471 and refs. therein.
3. For a review on Ir, see: R. Takeuchi, Synlett, 2002, 1954.
4. For literature on Mo, see: B. M. Trost and I. Hachiya, J. Am. Chem. Soc., 1998, 120, 1104.
5. For literature on Ru, see: B. M. Trost, P. L. Fraisse and Z. T. Ball, Angew. Chem., Int. Ed., 2002, 41, 1059.
6. J. P. Janssen and G. Helmchen, Tetrahedron Lett., 1997, 38, 8025.
7. B. Bartels and G. Helmchen, Chem. Commun., 1999, 741.
8. (a) B. Bartels, C. Garcia-Yebra, F. Rominger and G. Helmchen, Eur. J. Inorg. Chem., 2002, 2569; (b) B. Bartels, C. Garcia-Yebra and G. Helmchen, Eur. J. Org. Chem., 2003, 1097.
9. T. Ohmura and J. F. Hartwig, J. Am. Chem. Soc., 2002, 124, 15164.
10. F. Lopez, T. Ohmura and J. F. Hartwig, J. Am. Chem. Soc., 2003, 125, 3426.
11. K. Fuji, N. Kinoshita, K. Tanaka and T. Kawabata, Chem. Commun., 1999, 2289.
12. (a) T. Kanayama, K. Yoshida, H. Miyabe and Y. Takemoto, Angew. Chem., Int. Ed., 2003, 42, 2054; (b) T. Kanayama, K. Yoshida, H. Miyabe, T. Kimachi and Y. Takemoto, J. Org. Chem., 2003, 68, 6197.
13. Results were presented as a poster at the Heidelberg Forum of Molecular Catalysis, Heidelberg, June 27, 2003, Book of Abstracts, p 51.
14. B. M. Trost, S. Hildbrand and K. Dogra, J. Am. Chem. Soc., 1999, 121, 10416–10417.
15. Enantiomeric excess of 6b/8b was determined by HPLC-analysis of 6c/8c. Deprotection of 6b/8b followed by reaction with p-methoxybenzyl chloride gave 6c/8c in 47/79% yield over two steps.
16. Crystal data for 10: C₂₄H₂₂BrNO₂S, M = 468.40, monoclinic, space group P2₁, a = 11.6234(17) Å, α = 90°, b = 5.8505(9) Å, β = 109.524(3)°, c = 16.290(2) Å, γ = 90°, U = 1043.4(3) ų, Z = 2, µ = 2.09 mm⁻¹. 10870 reflections measured, 5045 unique (R(int) = 0.0423), 4401 observed [I > 2σ(I)]. R1 = 0.052, wR2 = 0.106 [I > 2σ(I)], flack 0.013(10). CCDC 220048. See http://www.rsc.org/suppdata/cc/b3/b311502j/ for crystallographic data in .cif format.
17. L. A. Arnold, R. Imbos, A. Mandoli, A. H. M. de Vries, R. Naasz and B. L. Feringa, Tetrahedron, 2000, 56, 2865.
18. B. L. Feringa, M. Pineschi, L. A. Arnold, R. Imbos and A. H. M. de Vries, Angew. Chem., Int. Ed. Engl., 1997, 36, 2620.
19. M. T. Reetz and G. Mehler, Angew. Chem., Int. Ed. Engl., 2000, 39, 3889.

---

Footnotes

† Ligands were prepared as described: L1,¹⁷L2,¹⁸L3,¹⁹L4.⁶

‡ Substrates 4 and 5 were prepared according to published procedures.¹⁴

§ General procedure for amination. Under argon, a solution of [Ir(COD)Cl]₂ (13.4 mg, 0.02 mmol), ligand (0.04 mmol) (ratio Ir ∶ ligand = 1 ∶ 1), substrate (1 mmol) and amine (1.3 mmol) in dry THF (0.5 ml) was stirred at rt. The solvent was removed under reduced pressure and the residue subjected to flash chromatography (silica, petroleum ether–ethyl acetate 5 ∶ 1) to give mixtures of 6 and 7.General procedure for alkylation. Under argon, a solution of [Ir(COD)Cl]₂ (13.4 mg, 0.02 mmol), ligand (0.04 mmol) (ratio Ir ∶ L* = 1 ∶ 1) and substrate (1.0 mmol) in dry THF (0.5 ml) was treated with a solution of lithium dimethylmalonate (2.0 mmol) in dry THF (3 ml) at room temperature and stirred for the time stated in Table 1. Then Et₂O (5 ml) and aqueous NH₄Cl solution (5 ml) were added. After standard extractive work-up, including washing of the organic phase with brine, drying and concentration in vacuo, the crude product was subjected to flash chromatography (silica, petroleum ether–ethyl acetate 12 ∶ 1).

---