1-(4,5-Dihydro-4-phenyl-2-oxazolyl)-2-(diphenylphosphino) ferrocene

[163169-12-8]  · C31H26NOPFe  · (MW 515.39)

(chiral ligand for asymmetric transition metal-catalyzed reactions including ketone and imine reductions, cuprate additions to enones, allylic cross-coupling reactions with Grignard reagents, and the enantioselective ring opening of oxabicyclic alkenes)

Physical Data: dark brown solid, mp 184-185 °C (decomposed)1 [a]D +36.8 (c = 0.005 g mL-1, CHCl3).2

Solubility: soluble in most common organic solvents. Not soluble in water.

Form Supplied in: not currently commercially available.

Analysis of Reagent Purity: 1H NMR, 13C NMR.

Preparative Methods: this compound can be prepared from the corresponding ferrocenyl oxazoline via a diastereoselective metalation with an alkyl lithium base followed by trapping with chlorodiphenylphosphine.3,4 The diastereoselectivity of the metalation can vary significantly depending on the base, and the greatest selectivity is observed using sec-butyllithium/TMEDA in diethyl ether or hexanes (but not in THF).5,6

Purification: flash chromatography.

Handling, Storage, and Precautions: no special instructions for storage and handling are mentioned in the literature. Use in a fume hood.

Asymmetric Reduction of Ketones and Imines

In 1991, Chowdhury and Backvall showed that RuCl2(PPh3)3 (2) is an effective catalyst for the transfer hydrogenation (Meerwein-Ponndorf-Verley reduction) of ketones provided that 2% NaOH is present as a co-catalyst.7 This result prompted others to study the use of chiral ligands in this process, and the groups of Sammakia and of Uemura and Hidai have investigated the use of ligands 1 and 3. Sammakia and Stangeland examined an in situ preparation of a catalyst mixture consisting of 0.26% 1, 0.2% 2, and 2.5% i-PrOK. This preparation provides an approximately 5:1 mixture of two diastereomers which are presumably epimeric at the metal, and is effective for the reduction of aryl-alkyl ketones (1 and 2). Typical enantioselectivities range between 91% and 96%; however, electron-rich and bulky substrates provide lower selectivities (para-methoxy acetophenone, 84% ee; phenyl-isopropyl ketone, 88% ee). Because the reaction is reversible, the enantioselectivity erodes at long reaction times, and the reactions must be monitored carefully and worked up at approximately 95% conversion. Typical isolated yields are in the range of 80% to 92%.

In an elegant study on the effects of structure on the selectivity of the catalyst, Nishibayashi et al. showed that by isolating a single diastereomer of the ruthenium catalyst, the selectivities in the reaction could be greatly enhanced.8 Thus, recrystallization of the 5:1 mixture of diastereomers from dichloromethane-diethyl ether provides a diastereomerically pure catalyst which was characterized by X-ray crystallography (3). Using this pure catalyst, selectivities in excess of 99% ee are observed for many aryl-alkyl ketones and for pinacolone (4). In addition to 1, the authors used ligand 3 in which the substituent on the oxazoline is an isopropyl group rather than a phenyl group, and the two ligands appear to behave similarly. Unfortunately, reactions of dialkyl ketones require a large steric differentiation such as is found in pinacolone, and other dialkyl ketones (such as cyclohexyl methyl ketone or 2-octanone) react with low enantioselectivities.

This catalyst, after treatment with either Cu(OTf)2 or AgOTf, has also been studied for the asymmetric hydrosilylation of ketones by Nishibayashi et al.; however, the yields and selectivities are not as high as in the transfer hydrogenation reactions.9

The hydrosilylation of imines has also been studied using either iridium, ruthenium, or rhodium complexes of ligand 1.10 The ruthenium- and iridium-derived catalysts {prepared using ligand 1 and either 2 or [Ir(COD)Cl]2, respectively} provide similar results and are more effective than the rhodium-derived catalysts {prepared from [Rh(COD)Cl]2}. These catalysts show good selectivities with selected examples, but the method is not general. For example, the five-membered ring imine 4 can be reduced to the corresponding amine in >95% yield (as determined by GC) and 85% ee; however, the related six-membered ring imine 5 provided low yields (10%, and 18% with the ruthenium- and iridium-derived catalysts, respectively) and low ees (25% and 7% with the ruthenium- and iridium-derived catalysts, respectively, 5). Furthermore, the N-methylimine of acetophenone reacted selectively (56% conversion, 89% ee using the iridium-derived catalyst), while the N-phenyl imine did not (25% yield, 23% ee using the iridium-derived catalyst).

Copper-Catalyzed Conjugate Addition of Grignard Reagents to Enones

A system composed of ligand 1 and CuI has been shown to catalyze the conjugate addition of Grignard reagents to enones.2 The reaction was optimized using n-BuMgCl as the nucleophile, and cyclohexenone as the electrophile, and the overall reaction was found to depend strongly on solvent and additives. Optimal conditions consist of using ether as the solvent with no additives, a 1.2:1 ratio of ligand to CuI, and a 10% loading of CuI. These conditions provide a 1,4- to 1,2-addition ratio of greater than 100:1, a 97% yield, and 83% ee. Consistent with other asymmetric conjugate addition catalysts,11 cyclopentenone provides slightly lower enantiomeric excess (65% ee, 6) than cyclohexenone, and the opposite sense of asymmetric induction. Cycloheptenone provides enhanced selectivities and the same sense of asymmetric induction as cyclohexanone (92% ee, 7). A single acyclic example has been studied, and it provided the conjugate addition product in 81% ee and 61% yield (8).

Nickel-Catalyzed Allylic Cross-Coupling with Hard Nucleophiles

The asymmetric nickel-catalyzed allylic cross-coupling of aryl Grignard reagents has been described by Chung et al. using 1.12 The reaction works well with selected partners but is not general. Cyclohexene derivatives provide enantioselectivities between 80% and 95%, and, if the aryl Grignard is unhindered, provide good-to-excellent yields (72% to 98% by GC, 9). Unfortunately, the reported yields are based on GC analysis, and are not of isolated, purified material. However, coupling with the somewhat hindered substrate, 1-naphthylmagnesium bromide, proceeded in only 39% yield and 80% enantioselectivity. An example of a coupling with a cyclopentenyl system provided low enantioselectivity (47%ee, 10), as did two acyclic systems (16% ee and 33% ee). The reaction can also be accomplished with arylboronic acids, but is less effective than the reaction using Grignard reagents.13

Ring-Opening of Oxabicyclic Alkenes

The palladium-catalyzed enantioselective ring opening of oxabicyclic alkenes has also been studied with 1 and related ligands.14 Lautens et al.14 recently reported the use of Pd(CH3CN)2Cl2 and chiral ferrocenyl phosphinoxazoline ligands for the asymmetric ring opening of [2.2.1]- and [3.2.1]oxabicyclic systems using dimethyl zinc as the nucleophile. The reaction was optimized using substrate 6, and the effect of varying the substituent on the oxazoline of the ligand was studied. It was found that the enantioselectivities increased with increasing bulk of the substituent and that the tert-butyl-substituted ligand 7 provided the highest enantioselectivities (11).

Related Reagents.

1-(4,5-Dihydro-4-tert-butyl-2-oxazolyl)-2-(diphenylphosphino) ferrocene; 1-(4,5-dihydro-4-isopropyl-2-oxazolyl)-2-(diphenylphosphino) ferrocene.

1. Nishibayashi, Y.; Segawa, K.; Arikawa, Y.; Ohe, K.; Hidai, M.; Uemura, S., J. Organomet. Chem. 1997, 546, 381.
2. Stangeland, E. L.; Sammakia, T., Tetrahedron 1997, 53, 16503.
3. Richards, C. J.; Damalidis, T.; Hibbs, D. E.; Hursthouse, M. B., Synlett 1995, 74.
4. Uemura, S.; Nishibayashi, Y., Synlett 1995, 79.
5. Sammakia, T.; Latham, H. A., J. Org. Chem. 1995, 60, 6002.
6. Sammakia, T.; Stangeland, E. L., J. Org. Chem. 1997, 62, 6104.
7. Chowdhury, R. L.; Backvall, J.-E., J. Chem. Soc., Chem. Commun. 1991, 1063.
8. Nishibayashi, Y.; Takei, I.; Uemura, S.; Hidai, M., Organometallics 1999, 18, 2291.
9. Nishibayashi, Y.; Takei, I.; Uemura, S.; Hidai, M., Organometallics 1998, 17, 3420.
10. Takei, I.; Nishibayashi, Y.; Arikawa, Y.; Uemura, S.; Hidai, M., Organometallics 1999, 18, 2271.
11. Krause, N.; Hoffmann-Roder, A., Synthesis 2001, 171.
12. Chung, K.-G.; Miyake, Y.; Uemura, S., J. Chem Soc., Perkin 1 2000, 2725.
13. Chung, K.-G.; Miyake, Y.; Uemura, S., J. Chem Soc., Perkin 1 2000, 15.
14. Lautens, M.; Hiebert, S.; Renaud, J.-L., Org. Lett. 2000, 2, 1971.

Tarek Sammakia

University of Colorado, Boulder, Colorado, USA

Copyright 1995-2000 by John Wiley & Sons, Ltd. All rights reserved.