Synthesis of 3-hydroxyoxindoles by Pd-catalysed intramolecular nucleophilic addition of aryl halides to α-ketoamides

Abstract

Pd/PtBu₃-catalysed intramolecular nucleophilic addition of aryl halides to α-ketoamides in the presence of nBuOH and base has been realized with high yields, providing a new, direct, and efficient synthetic strategy to obtain 3-hydroxyoxindoles.

---

The predominantly electrophilic reactivity of aryl and vinyl organopalladium intermediates derived from the corresponding halides by an oxidative addition process with a Pd(0) catalyst is well established.¹ Conversely their use as nucleophiles has received much less attention although literature precedent exists and includes reactions with aldehydes,² ketones,³ imines,⁴ esters and amides,⁵ anhydrides⁶ and nitriles.⁷ The direct catalytic addition reaction of aryl and vinyl halides to electrophilic partners is attractive since no extra organometallic reagents are needed. In the context of our interest in the synthesis of oxindoles⁸ we envisioned that it might be possible to obtain 3-hydroxyoxindoles by an intramolecular addition of aryl halides to α-ketoamides using a Pd catalyst. This would provide a direct access to these compounds avoiding the use of Grignard-, organoborane-, or organolithium-reagents (Scheme 1).⁹ Herein, we report the preliminary results of this new synthetic strategy to oxindoles.

Scheme 1 New route to 3-hydroxyoxindoles.

The model substrate 1a was prepared by condensation of N-methyl o-bromoaniline and 2-oxo-2-phenylacetyl chloride. Initial attempts to use Pd(PPh₃)₄ as a catalyst were not efficient for the transformation 1a → 2a and attempts to improve results by modifications of base, solvent, and additives were not successful. The best result was a 17% yield of 2a with 5 mol% Pd(PPh₃)₄ in toluene at 110 °C for 24 h in the presence of 2 eq. Na₂CO₃ and 10 eq. NEt₃. Using nBuOH as solvent afforded 3a and only trace amounts of 2a. We then focused on the use of electron-rich ligands,¹⁰ which should favour oxidative addition of the catalyst to the aryl bromide and, more importantly, facilitate the nucleophilic attack of the Pd–aryl intermediate on the ketone. Indeed, when 10 mol% PtBu₃·HBF₄ was used as a ligand precursor, the reaction of 1a proceeded smoothly in the presence of 5 mol% Pd(dba)₂, 2 eq. Cs₂CO₃, and 2 eq. nBuOH in toluene at 50 °C giving oxindole 2a in 76% isolated yield along with 19% of the corresponding hydrodehalogenation side product 3a (Table 1, entry 1). We note that these conditions are milder than those reported previously for this type of reaction—e.g. 100–150 °C for intramolecular additions to ketones.³ Solvent examination revealed toluene to be the best choice as low yields (<15%) were observed in ether solvents such as THF, DME (1,2-dimethoxy ethane), and 1,4-dioxane. In DCE (1,2-dichloroethane) product yield was 54% after 20 h, while in DMF the reaction was completely suppressed. Although the best result was obtained when nBuOH was used as an additive, MeOH and EtOH afforded slightly lower yields whereas iPrOH gave a poor product yield (entries 4–6). In the absence of an additive, the reaction fails, as has been also reported by the Yamamoto group.^3b Interestingly, instead of an alcohol H₂ (1 bar) could be used to promote this reaction with comparable yield (entry 7). Moderate yields of 2a were observed when nBu₃SnH and Et₃SiH were used and for the former no side product 3a was detected (entries 8, 9).

Table 1 Optimization of reaction conditions^a

Entry	Base	Solvent	Reductant	Temp./°C	Time/h	Y^b (%) (2a/3a)
a 0.4 mmol scale in 8 mL solvent. b Isolated yield. nd = not determined. c With a hydrogen balloon. d 10 mol% PCy3 was used instead of PtBu3·HBF4.
1	Cs2CO3	Toluene	nBuOH	50	7	76/19
2	Cs2CO3	DCE	nBuOH	50	20	54/nd
3	Cs2CO3	tBuOMe	nBuOH	50	20	29/nd
4	Cs2CO3	Toluene	MeOH	50	7	69/nd
5	Cs2CO3	Toluene	EtOH	50	7	71/nd
6	Cs2CO3	Toluene	iPrOH	50	16	21/nd
7^c	Cs2CO3	Toluene	H2	100	12	62/28
8	Cs2CO3	Toluene	nBu3SnH	100	24	43/0
9	Cs2CO3	Toluene	Et3SiH	50	12	58/23
10	K2CO3	Toluene	nBuOH	80	24	23/nd
11	K3PO4·H2O	Toluene	nBuOH	50	30	85/9
12	K3PO4·H2O	Toluene	nBuOH	80	7	88/7
13^d	K3PO4·H2O	Toluene	nBuOH	80	9	85/7

---

Variation of the base revealed this to be an important factor. No reactions took place in the presence of Cy₂NMe, KF, KOAc, Li₂CO₃, and Na₂CO₃. With K₂CO₃ as base, oxindole 2a could be isolated in 23% yield after 24 hours (entry 10). Finally, K₃PO₄·H₂O was found to be the best base giving the highest yield of the product albeit a longer reaction time was needed to achieve full conversion at 50 °C (entry 11). Increasing the temperature to 80 °C resulted in complete reaction in 7 hours and 2a was isolated in 88% yield along with 7% side product 3a (entry 12). Further investigation of the ligand showed that PCy₃ was also suitable (entry 13), while dppe (1,2-bis(diphenylphosphino)ethane) and 1,3-bis(2,6-diisopropyl-phenyl)-4,5-dihydroimidazolium chloride (SIPr·HCl) as ligand or ligand precursor were inefficient in this reaction.

The optimum reaction conditions were then applied to the substrates shown in Table 2. Yields are generally good (81–92%). It is noteworthy that the reaction to oxindole 2d with CF₃ as an electron-withdrawing group was complete in 3 h, while 14 h were required for the reaction of substrate 1e with an o-MeO group. Not only N-Me substrates but also an N-benzyl amide (2i) could be used, but no reaction took place for the free N-H substrate (R¹ = R² = H, Ar = Ph).

Table 2 Substrate scope^a

a 0.4 mmol scale in 8 mL toluene.

---

The reaction is sensitive to the aryl halide used (eqn (1)). In comparison with 1a, no reaction took place for the analogous chloride substrate 1m. Noteworthy, the reaction of the iodo substrate 1n was slower than that of the bromo substrate 1a and 48 h were needed to achieve full conversion. Since the reaction of 1a with Cs₂CO₃ as base was faster than with K₃PO₄·H₂O, we introduced Cs₂CO₃ to the reactions of these two substrates. Product 2a was isolated in 72% yield from 1n after 6 h at 80 °C and in 33% yield from 1m after 72 h at 110 °C. Interestingly, the reaction of 1a under standard conditions was suppressed in the presence of 3.0 eq. LiI as additive and only traces of oxindole 2a were detected after 48 h.

Vinyl bromide substrate 1o was synthesized and applied in the reaction (eqn (2)). Pyrrolidinone product 2m could be obtained; yields were low (16% for K₃PO₄·H₂O as base, 29% for Cs₂CO₃).Next, experiments designed to shed light on mechanistic aspects of the reaction were carried out. As shown in eqn (3), oxindole 2a could be isolated in 92% yield with stoichiometric Pd(0)/PtBu₃ complex in the absence of nBuOH, thus demonstrating that Pd(0) can perform the same role as Mg in Grignard reactions. As mentioned above, catalytic use of Pd requires an additive and primary alcohols performed best. An earlier study of an intramolecular aryl addition to ketones, catalysed by Pd with added 1-hexanol,^3b reported that the alcohol was recovered. Our findings are different. Using piperonyl alcohol, as shown in eqn (4), piperonyl aldehyde was isolated in 95% yield. This confirms the role of the alcohol as reductant in the reaction and thus closing the catalytic cycle. Also, the deuterated dehalogenation side product was detected when MeOH-d₄ was used (eqn (5)).Based on the above, we propose the mechanism shown in Fig. 1. The oxindole is formed by intramolecular addition of the oxidative addition intermediate I to the ketone. Alkoxyligand exchange II (cycle A) liberates product 2a and the Pd(0) catalyst is regenerated by β-H elimination (aldehyde formation) followed by base assisted reductive elimination. For the hydrodehalogenation side product 3a we propose halide–alcoholate ligand exchange in I to give III (cycle B). β-Elimination followed by reductive elimination affords 3a (cycle B). The observation of the deuterated side product shown in eqn (5) supports this pathway.

Fig. 1 Proposed mechanism.

In summary, we have developed a new and efficient procedure to synthesize 3-hydroxyoxindole compounds in high yields by applying palladium-catalysed direct addition of aryl halides to α-ketoamides. Both, electron-rich phosphine ligands and a reductant (alcohol, H₂) are crucial. The formation of a new stereogenic center in 2 is another important aspect that is receiving attention in ongoing research.

Notes and references

1. J. Tsuji, Palladium Reagents and Catalysis—New Perspectives for the 21st Century, Wiley, Weinheim, 2004.
2. (a) R. C. Larock, M. J. Doty and S. Cacchi, J. Org. Chem., 1993, 58, 4579–4583; (b) V. Gevorgyan, L. G. Quan and Y. Yamamoto, Tetrahedron Lett., 1999, 40, 4089–4092; (c) J. Vicente, J. A. Abad, B. Lopez-Pelaez and E. Martinez-Viviente, Organometallics, 2002, 21, 58–67; (d) Y. B. Zhao, B. Mariampillai, D. A. Candito, B. Laleu, M. Z. Li and M. Lautens, Angew. Chem., Int. Ed., 2009, 48, 1849–1852.
3. (a) L. G. Quan, V. Gevorgyan and Y. Yamamoto, J. Am. Chem. Soc., 1999, 121, 3545–3546; (b) L. G. Quan, M. Lamrani and Y. Yamamoto, J. Am. Chem. Soc., 2000, 122, 4827–4828; (c) D. Sole, L. Vallverdu, X. Solans, M. Font-Bardia and J. Bonjoch, J. Am. Chem. Soc., 2003, 125, 1587–1594.
4. (a) K. R. Roesch and R. C. Larock, Org. Lett., 1999, 1, 553–556; (b) H. Ohno, A. Aso, Y. Kadoh, N. Fujii and T. Tanaka, Angew. Chem., Int. Ed., 2007, 46, 6325–6328.
5. (a) D. Sole and O. Serrano, Angew. Chem., Int. Ed., 2007, 46, 7270–7272; (b) D. Sole and O. Serrano, J. Org. Chem., 2008, 73, 9372–9378.
6. S. Cacchi, G. Fabrizi, F. Gavazza and A. Goggiamani, Org. Lett., 2003, 5, 289–291.
7. (a) C. C. Yang, P. J. Sun and J. M. Fang, J. Chem. Soc., Chem. Commun., 1994, 2629–2630; (b) R. C. Larock, Q. P. Tian and A. A. Pletnev, J. Am. Chem. Soc., 1999, 121, 3238–3239; (c) A. A. Pletnev and R. C. Larock, J. Org. Chem., 2002, 67, 9428–9438; (d) A. A. Pletnev, Q. P. Tian and R. C. Larock, J. Org. Chem., 2002, 67, 9276–9287; (e) Q. P. Tian, A. A. Pletnev and R. C. Larock, J. Org. Chem., 2003, 68, 339–347; (f) C. X. Zhou and R. C. Larock, J. Org. Chem., 2006, 71, 3551–3558.
8. (a) E. P. Kündig, T. M. Seidel, Y.-X. Jia and G. Bernardinelli, Angew. Chem., Int. Ed., 2007, 46, 8484–8487; (b) Y.-X. Jia, J. M. Hillgren, E. L. Watson, S. P. Marsden and E. P. Kündig, Chem. Commun., 2008, 4040–4042; (c) Y.-X. Jia and E. P. Kündig, Angew. Chem., Int. Ed., 2009, 48, 1636–1639.
9. 3-Hydroxyoxindoles can be obtained by addition of aryl Grignard reagents or aryl lithium reagents (both are prepared from aryl halides) to isatin: (a) M. A. Sukari and J. M. Vernon, J. Chem. Soc., Perkin Trans. 1, 1983, 2219–2223; (b) S. J. Angyal, E. Bullock, W. G. Hanger, W. C. Howell and A. W. Johnson, J. Chem. Soc., 1957, 1592–1602 and ref. cit. Catalytic asymmetric synthesis of 3-hydroxyoxindole: Pd catalysed intramolecular arylation of amides; (c) ref. 8b; M-catalysed addition of aryl boronic acids to isatin; (d) R. Shintani, M. Inoue and T. Hayashi, Angew. Chem., Int. Ed., 2006, 45, 3353–3356; (e) H. S. Lai, Z. Y. Huang, Q. Wu and Y. Qin, J. Org. Chem., 2009, 74, 283–288; Lewis acid-catalysed asymmetric hydroxylation of oxindoles; (f) T. Ishimaru, N. Shibata, J. Nagai, S. Nakamura, T. Toru and S. Kanemasa, J. Am. Chem. Soc., 2006, 128, 16488–16489.
10. Although PCy₃ has been used as a ligand in a similar addition reaction of simple ketone substrates (ref. 3b), no reaction took place from 1a to 2a using the reported conditions (5 mol% Pd(OAc)₂, 10 mol% PCy₃, 2 eq. Na₂CO₃, 5 eq. 1-hexanol in DMF).

---

Footnotes

† This article is part of a ChemComm ‘Catalysis in Organic Synthesis’™ web-theme issue showcasing high quality research in organic chemistry. Please see our website (http://www.rsc.org/chemcomm/organicwebtheme2009) to access the other papers in this issue.

‡ Electronic supplementary information (ESI) available: Experimental details, spectral characterisation and analytical data of reported compounds. See DOI: 10.1039/b917958e

---