RuCl₂(PPh₃)₃-catalyzed chemoselective hydrogenation of β, δ-diketo acid derivatives at the β-carbonyls

Abstract

Chemoselective reduction of the β-carbonyls in β, δ-diketo acid derivatives was achieved through RuCl₂(PPh₃)₃ catalyzed homogeneous hydrogenation. Tetrahydrofuran (THF) played a key role in the chemoselectivity control. This atom-economical protocol provided β-hydroxy-δ-keto esters and amides as useful intermediates in good to excellent yields.

---

The reduction of ketones to alcohols is a very fundamental organic transformation,¹ which has been routinely accomplished by various alumino- or borohydrides.² However, large-scale use of these stoichiometric reagents often suffers from safety concerns, high cost and tedious work-up. Besides, common reductants like NaBH₄ and LiAlH₄ show hardly any chemoselectivity for poly-carbonyl substrates,³ especially when the carbonyls are in similar steric and electronic environment.⁴

For example, the selective reduction of the β, δ-diketo acid derivatives at the desired position is important but difficult to control. As illustrated in Scheme 1, metal hydrides and their modifiers tend to reduce both β- and δ- carbonyls and semi-reduction products can rarely be attained except for a few sterically hindered substrates.⁵ Even many ketoreductases were unable to give an ideal chemoselectivity.⁶

Scheme 1 Various methods for reduction of 1.

Several recently-developed biocatalysts reduced the δ-carbonyls with spectacular chemo- and enantioselectivity,^6c,7 although few of them can reduce β-carbonyls equally well.^6c Moreover, the high costs and narrow substrate scope limited their applications. Until now, no practical reduction methods with good β-selectivity have been reported.

Homogeneous hydrogenation has been increasingly employed in the reduction of unsaturated functionalities due to its unique merits.⁸ Earlier attempts on the β-selective hydrogenation of β, δ-diketo esters failed because the initial hydrogenation products easily underwent further hydrogenation in alcohols.⁹ In 1999, Carpentier et al.^9a implemented a β-selective hydrogenation of methyl 3,5-dioxohexanoate with a Ru-BINAP system under the specified conditions (20 to 50 °C, 100 atm of H₂, CH₂Cl₂). However, the reactions must be carefully monitored by GLC during the course to avoid over hydrogenation.

In our earlier studies on the asymmetric hydrogenation of 3-oxoglutaric acid derivatives,¹⁰ we discovered that a coordinative solvent like THF helped to modulate the coordination ability of carbonyls in β-keto esters and β-keto amides. The latter were hydrogenated much more quickly than the former.

Based on these conclusions, we tried the hydrogenation of various β, δ-diketo acid derivatives in THF, which was an ineffective solvent for the catalytic hydrogenation of many functionalized ketones.¹¹ To our pleasure, good to excellent yields were obtained with RuCl₂(PPh₃)₃ as the catalyst under moderate H₂ pressure (20 bar).

Simple optimization of the reaction conditions with 1b and 1m was performed. It took 30 h for 90% conversion of 1m at 50 °C under 5 bar of H₂ with S/C = 100. At 90 °C, the reaction became complex for esters but was still good for amides. Thus we performed hydrogenation reactions at 70 °C and at 20 bar of H₂. Under such relatively mild conditions, all the tested substrates converted to the desired products in 8 to 15 h. A prolonged reaction time (especially for the amides) after full conversion (8 h for most amides) lead to no over reduction. The resistance to further hydrogenation of the β-hydroxy amides was illustrated by the fact that 2m remained untouched when subjected to harsher conditions (1% mmol of catalyst, 20 bar of H₂, 95 °C, 8 h). Therefore, cautious monitoring of the hydrogenation process is unnecessary.

As listed in Table 1, the esters 1a–e gave the desired products in good yields. The (hetero)aryl substituted esters (entries 3–5, Table 1) gave somewhat higher yields than the alkyl substituted esters (entries 1–2, Table 1), as the latter was partially converted to some byproducts.^9a Various amides, especially the aryl substituted ones, were cleanly hydrogenated in almost quantitative yields. Diethyl amides reacted faster than tert-butyl esters. For example, 1m, 1n and 1o underwent full conversion in 8 h while 1b and 1d did not. This was also consistent with the hydrogenation rate sequence reported by Brückner et al.¹² Electron withdrawing or donating substituents on different positions of phenyl rings had little influence on the reactions and the heteroatoms, such as sulfur in the thiophene did not affect the activity of the catalyst in this reaction (entry 17, Table 1).¹³ Substrates with geminal dimethyl substituents on the γ-methylene (1t) can also be hydrogenated smoothly. The asymmetric version of this type of substrates were used in the synthesis of epothilone and its congeners.¹⁴

Table 1 Hydrogenation of various β, δ-diketo acid derivatives^a

Entry	1	R	Het	Yield(%)^b
a All reactions were carried out with 1 mmol of substrate in 5 mL of THF at 70 °C under 20 bar of H2 for 8–15 h. S/C = 200. Conversion = 100%. 8 h is enough for most amide substrates. b Isolated yields.
1	1a	Me	OMe	85
2	1b	Me	OtBu	86
3	1c	Ph	OtBu	87
4	1d	2-Naphthyl	OtBu	90
5	1e	2-Furanyl	OtBu	92
6	1f	Et	NHtBu	95
7	1g	Cy	NHtBu	91
8	1h	Me	N-Morpholinyl	91
9	1i	tBu	N-Morpholinyl	93
10	1j	C6H5	N-Morpholinyl	95
11	1k	Me	NBn2	94
12	1l	Ph(CH2)2	NEt2	95
13	1m	C6H5	NEt2	96
14	1n	2-MeC6H5	NEt2	94
15	1o	3-ClC6H5	NEt2	96
16	1p	4-MeOC6H5	NEt2	96
17	1q	4-CF3C6H5	NEt2	95
18	1r	1-Naphthyl	NEt2	95
19	1s	2-Thienyl	NEt2	96
20	1t	Et	NEt2	91

---

To investigate which carbonyl was the directing group in the β-selective hydrogenation, competitive reaction tests with a, b and c were performed (Scheme 2). Both β-keto ester (a) and β-diketone (b) were inert in the same reaction conditions, while the β-keto amide (c) can be hydrogenated smoothly to give the hydroxy product. One can envision that the 1,3-dicarbonyls in β-keto amides chelated with the ruthenium center, an essential step for the hydrogenation to proceed.¹⁵ In coordinative solvents like THF, which may act as a competitive ligand,¹⁶ the chelation ability of 1,3-dicarbonyl system in β-diketones and β-keto esters were comparably depleted. Intriguingly, the monofunctionalized ketones a and b were both inert to hydrogenation in THF, whereas 1a, a combined form of these two components was hydrogenated smoothly at the β-position. It was likely that the ruthenium coordinated to the enol forms of the substrates.^9a As for the amide substrates, we suspected that the 1-carbonyls were the dominant directing groups, as depicted in Scheme 2, but further evidence is needed to determine if the coordination was related to the enol form at the β-carbon.

Scheme 2 Proposed chelation modes in the hydrogenation reaction.

In summary, we have developed a general method for the highly chemoselective reduction of the β, δ-diketo acid derivatives at the β-positions. The resultant products are very useful building blocks for organic synthesis. More arrestingly, this method employed the readily available RuCl₂(PPh₃)₃ as the catalyst¹⁷ and inexpensive hydrogen as the green reductant. The asymmetric version of this reaction with chiral catalysis for simultaneous high chemo- and enantioselectivity is more challenging and is ongoing in our lab.

This work was financially supported by National Natural Science Foundation of China and Science and Technology Commission of Shanghai Municipality. We thank Johnson Matthey (Shanghai) Chemicals Limited for generous loan of precious metal catalysts.

References

1. (a) N. Greeves, in Comprehensive Organic Synthesis, ed. B. M. Trost, Pergamon, 1991, vol. 8, pp. 1–30; (b) P. G. Andersson and I. J. Munslow, ed. Modern Reduction Methods, Wiley VCH, Weinheilm, 2008.
2. (a) H. C. Brown and S. Krishnamurthy, Tetrahedron, 1979, 35, 567; (b) J. Seyden-Pennein, Reductions by the Alumino- and Borohydrides in Organic Synthesis, Wiley-VCH, 1997; (c) W. G. Brown, in Organic Reactions, John Wiley & Sons, Inc., 1951, vol. 6.
3. Z. Guo, Y. Chen, A. Goswami, R. L. Hanson and R. N. Patel, Tetrahedron: Asymmetry, 2006, 17, 1589.
4. B. M. Trost, Science, 1983, 219, 245.
5. (a) G. B. Reddy, T. Minami, T. Hanamoto and T. Hiyama, J. Org. Chem., 1991, 56, 5752; (b) T. Hiyama, G. B. Reddy, T. Minami and T. Hanamoto, Bull. Chem. Soc. Jpn., 1995, 68, 350; (c) T. Hiyama, T. Minami and K. Takahashi, Bull. Chem. Soc. Jpn., 1995, 68, 364.
6. (a) M. Wolberg, I. A. Kaluzna, M. Müller and J. D. Stewart, Tetrahedron: Asymmetry, 2004, 15, 2825; (b) R. Patel, A. Banerjee and L. Szarka, J. Am. Oil Chem. Soc., 1995, 72, 1247; (c) R. N. Patel, A. Banerjee, C. G. McNamee, D. Brzozowski, R. L. Hanson and L. J. Szarka, Enzyme Microb. Technol., 1993, 15, 1014.
7. (a) M. Ernst, B. Kaup, M. Müller, S. Bringer-Meyer and H. Sahm, Appl. Microbiol. Biotechnol., 2005, 66, 629; (b) S. Lüdeke, M. Richter and M. Müller, Adv. Synth. Catal., 2009, 351, 253; (c) M. Müller, Angew. Chem., Int. Ed., 2005, 44, 362; (d) M. Müller, M. Wolberg, T. Schubert and W. Hummel, Adv. Biochem. Engin./Biotechnol., 2005, 92, 261; (e) M. Wolberg, W. Hummel, C. Wandrey and M. Müller, Angew. Chem., Int. Ed., 2000, 39, 4306; (f) M. Wolberg, A. Ji, W. Hummel and M. Müller, Synthesis, 2001, 937.
8. (a) R. E. Harmon, S. K. Gupta and D. J. Brown, Chem. Rev., 1973, 73, 21; (b) G. Dolcetti and N. W. Hoffman, Inorg. Chim. Acta, 1974, 9, 269; (c) G. Shang, W. Li and X. Zhang, in Catalytic Asymmetric Synthesis, ed. I. Ojima, John Wiley & Sons, Hoboken, New Jersey, 3rd edn, 2010; (d) B. Cornils and W. A. Herrmann, ed. Applied Homogeneous Catalysis with Organometallic Compounds, Second, Completely Revised and Enlarged edn, Wiley-VCH, Weinheim, 2002; (e) J. G. de Vries and C. J. Elsevier, ed., The Handbook of Homogeneous Hydrogenation, Wiley-VCHWeinheim, 2007.
9. (a) V. Blandin, J. F. Carpentier and A. Mortreux, Eur. J. Org. Chem., 1999, 3421; (b) O. Labeeuw, C. Roche, P. Phansavath and J.-P. Genêt, Org. Lett., 2006, 9, 105.
10. W. F. Li, X. Ma, W. Z. Fan, X. M. Tao, X. M. Li, X. M. Xie and Z. G. Zhang, Org. Lett., 2011, 13, 3876.
11. (a) A. Wolfson, I. F. J. Vankelecom, S. Geresh and P. A. Jacobs, J. Mol. Catal. A: Chem., 2003, 198, 39; (b) E. Starodubtseva, O. Turova, M. Vinogradov, L. Gorshkova and V. Ferapontov, Russ. Chem. Bull., 2007, 56, 552.
12. (a) R. Kramer and R. Brückner, Angew. Chem., Int. Ed., 2007, 46, 6537; (b) R. Kramer and R. Brückner, Chem.–Eur. J., 2007, 13, 9076.
13. The nitrogen atom in the pyridine ring inhibited the catalyst. For example, when R = 2-pyridinyl or 3-pyridinyl, the substrates became totally unreactive toward the catalytic reaction. A sharply different dark red solution was formed.
14. (a) A. Rivkin, F. Yoshimura, A. E. Gabarda, Y. S. Cho, T.-C. Chou, H. Dong and S. J. Danishefsky, J. Am. Chem. Soc., 2004, 126, 10913; (b) T. C. Chou, H. Dong, A. Rivkin, F. Yoshimura, A. E. Gabarda, Y. S. Cho, W. P. Tong and S. J. Danishefsky, Angew. Chem., Int. Ed., 2003, 42, 4761.
15. (a) R. Noyori, M. Kitamura and T. Ohkuma, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 5356; (b) In Organotransition metal chemistry - from bonding to catalysis, ed. J. F. Hartwig, University Science Books, Sausalito, California, 2010, pp. 599.
16. C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2003.
17. Other complexes with readily available ligands such as rac-BINAP could also furnish the reaction with comparable results under the same conditions. RuCl₂(PPh₃)₃ was chosen for cost consideration.

---

Footnote

† Electronic Supplementary Information (ESI) available. See DOI: 10.1039/c2ra20114c/

---