(Bicyclo[2.2.1]hepta-2,5-diene)[1,4-bis(diphenylphosphino)butane]rhodium(I) Tetrafluoroborate

[82499-43-2]  · C35H36BF4P2Rh  · (Bicyclo[2.2.1]hepta-2,5-diene)[1,4-bis(diphenylphosphino)butane]rhodium(I) Tetrafluoroborate  · (MW 708.4)

(reagent for catalytic hydrogenation;9-11 hydrosilylation,26 hydroboration,27,28 and aldol condensation32)

Physical Data: mp 211-212 °C (dec).

Solubility: insol Et2O, pentane; sol CH2Cl2, MeOH, etc.

Form Supplied in: orange crystalline powder. Drying: used as supplied in anhydrous solvent.

Analysis of Reagent Purity: 31P NMR indicates presence of free phosphine ligand and phosphine oxides; the characteristic 103Rh coupled doublet (JRhP = 155 Hz) demonstrates pure material.

Preparative Methods: a solution of [Rh(nbd)acac] in THF is treated with fluoroboric acid, followed by addition of 1,4-bis(diphenylphosphino)butane (dppb). The solution becomes deep red and ether is added to precipitate the catalyst, which is then isolated by filtration.3a The unpurified product is generally adequately pure for most applications. Recrystallization from methanol can be performed to obtain orange needles of [Rh(nbd)(dppb)]BF4. Although air-sensitive in solution, the crystalline complex is indefinitely stable when prepared pure and stored under N2 or Ar below 0 °C. A convenient preparation of the corresponding trifluoromethanesulfonate has been described.3b The cyclooctadiene (cod) analog has been characterized by X-ray crystallography,4 and the dynamic solution structure of related systems has been studied by multinuclear NMR techniques.5

Handling, Storage, and Precautions: store under argon in freezer; stable as solid; solutions are air sensitive. Literature describes the use of perchlorate salts but their use cannot be recommended due to risk of detonation. Use in a fume hood.


Homogeneous catalytic hydrogenation with cationic rhodium catalysts has been extensively explored by Schrock and Osborn.1 Use of these complexes in stereoselective organic synthesis has been a topic of more recent interest, and has been recently reviewed.2 The reagent of choice for many of these directed hydrogenations has continued to be [Rh(nbd)(dppb)]BF4 (1).

Directed Hydrogenation.

By far the most significant application of (1) has been in diastereoselective hydrogenation reactions. The ability of (1) to retain a significant level of Lewis acidity under conventional hydrogenation conditions has facilitated the development of directed hydrogenations in which (1) engages in prior coordination to heteroatoms situated proximal to the alkene functionality. Scheme 1 shows a general mechanistic scheme by which (1) may catalyze hydrogenation.6 This scheme is derived by analogy to mechanistic steps elucidated in the context of the hydrogenation of N-acyl dehydroamino acids, which have been demonstrated to coordinate to (1) in a bidentate fashion.

Catalyst (1) readily absorbs 2 moles of H2 to form complex A which may coordinate two solvent molecules.7 The reversible binding of the substrate then occurs in a bidentate fashion to give complex B. Oxidative addition of dihydrogen then occurs to form the dihydride species C, which then undergoes migratory insertion to form the rhodium alkyl D. Although a primary Rh alkyl is shown in Scheme 1, a secondary Rh alkyl is possible as well. Finally, reductive elimination occurs to give the reduced product with concomitant regeneration of the active catalyst.

At low pressures the addition of hydrogen to a rhodium complex (schematically shown as B) is probably rate-determining, but at higher pressure, pre-equilibria (A to B) can contribute to the rate law. The effect of pressure on mechanism has important implications in the directed hydrogenation of substituted alkenes which can undergo double bond isomerization. This issue has been addressed in synthetic and mechanistic studies employing (1) (see below).

An early observation of directed hydrogenation was made by Thompson8a using ClRh(PPh3)3 with a cyclic homoallylic alcohol as reactant; the stereoselectivity of hydrogenation could be enhanced by base so that delivery of hydrogen to the alkene from an alkoxide-coordinated rhodium was postulated. Acyclic diastereoselection was observed in hydrogenation of allylic alcohols with complex (1)8b and a high level of stereochemical control observed for hydrogenations catalyzed by (1,5-Cyclooctadiene)(tricyclohexylphosphine)(pyridine)iridium(I) Hexafluorophosphate (2).8c,8d The ensuing discussion will concentrate on catalysis by complex (1).

Cyclic Alkenes.

The directing effect of alcohol substituents is perhaps most dramatically demonstrated with cyclic alkenes in which products are formed via hydrogenation from the more hindered face (Table 1). Entries 1-3 serve to illustrate that the directing alcohol may reside in either the allylic, homoallylic, or bis(homoallylic) position relative to the alkene undergoing hydrogenation, while synthetically useful levels of selectivity are retained.9 Entry 4 is included to illustrate the dramatic steric congestion which can be overcome in directed hydrogenation reactions employing (1). Entries 510 and 611 illustrate the compatibility of (1) with protected amine functionality. In addition, when the alkenes in entry 5 or 6 are reduced under heterogeneous, nondirecting conditions, the other face of each alkene is reduced with high selectivity. Finally, entry 7 demonstrates that hydroxyl directivity in the [2.2.2] bicyclic system is also possible.12

Other Oxygen Directing Groups.

The Lewis acidity of (1) is manifested in coordination to other heteroatom-based functional groups which can direct hydrogenation. Table 2 demonstrates that (1) has a sufficient affinity for ethers11 and esters13 so that directed hydrogenations can be achieved when these functional groups are proximal to alkenes, although the rates and selectivities are somewhat lower than with the corresponding alcohols. While ketones, acetals, carboxylic acids, and amides have been demonstrated to direct catalytic hydrogenation with the corresponding Ir+ catalyst (2),14 these functional groups do not provide good results when (1) is employed with cyclic alkenes. The lack of directivity from amides with (1) is particularly intriguing given the dramatic coordination properties of this moiety in the RhI catalyzed hydrogenation of N-acyl dehydroamino acids.15

Acyclic Alkenes.

Acyclic alkenes also undergo stereoselective hydrogenation with catalysis by (1) (Table 3). It is significant to note that, in entry 1,8b although high selectivity is observed for the anti isomer, about 20% of the product is obtained as the corresponding methyl ketone, indicating that alkene isomerization is a competitive process. Entries 2 and 3 illustrate that either the syn or anti isomer can be obtained depending on the substitution pattern on the starting alkene.8 In addition, this study established that alkene isomerization can be effectively suppressed at higher pressures. Entries 4-6 establish the efficiency of (1) for the reduction of unsaturated hydroxy esters,16,17 which are not effectively reduced by (2). A model rationalizing the stereochemical outcome of these reductions has been proposed.18 In this model, simultaneous coordination of the hydroxyl group and the alkene differentiates the diastereotopic faces. This analysis also explains the fact that alkene isomerization at low H2 pressures contributes to diminished diastereoselectivities. This explanation has been confirmed by independent mechanistic experiments in which entries 2 and 3 were studied employing (1) and D2.19

Catalyst (1) is effective in the directed reduction of homoallylic alcohols as well (Table 4). Entries 1-3 indicate that 1,1-disubstituted alkenes exhibit reasonable levels of diastereoselectivity in directed hydrogenations, although this level depends greatly on substitution pattern.15,20 Entries 4-8 illustrate homoallylic trisubstituted alkenes undergo directed hydrogenation under catalysis by (1) with very high levels of selectivity, and that the configuration of the allylic substituent plays a significant role in modulating the level of selectivity. This study includes a rationale to explain the observed diastereoselectivities based on the principles of allylic strain.1a,21

As with cyclic alkenes, nonhydroxylic directing groups can be used in the directed hydrogenation of acyclic alkenes. The selective reductions of 3-substituted itaconate esters illustrate the directing capacity of esters (Table 5). It appears that the presence of coordinating allylic substituents can effect the level of selectivity.22

Applications in Total Synthesis.

Two recent examples of directed hydrogenations employing (1) in the total synthesis of complex molecules are illustrated. In eq 1 a simultaneous diastereoselective reduction of the trisubstituted alkene and the a,b-unsaturated ester afforded the illustrated advanced intermediate in the asymmetric total synthesis of ionomycin.23 In addition, a two-directional application has been utilized in an asymmetric synthesis FK-506 (eq 2).24

Miscellaneous Applications.

The hydroxyl directed hydrogenation of vinylstannanes and -silanes has been demonstrated to proceed efficiently.25 The authors present a transition state model which rationalizes the observed results (eqs 3 and 4).


Rhodium(I) complexes catalyze the asymmetric hydrosilylation of prochiral ketones (eq 5), in the presence of (-)-sparteine.26 Secondary alcohols are obtained in up to 30% optical yield by this method.


[Rh(cod)(dppb)]BF4 is an efficient catalyst for the hydroboration of a range of vinylarenes.27,28 Addition of Catecholborane to various styrene derivatives in the presence of 2 mol % catalyst gives, after oxidative workup, >99% secondary alcohol; in contrast, the corresponding uncatalyzed reaction gives the primary alcohol as the major product (eq 6).

This high regioselectivity is restricted to monosubstituted alkenes, indene, and (E)-1-phenylpropene. Analogous reactions with terminal aliphatic alkenes generally lead to primary alcohols as the major product, although alkene isomerization and BH3-derived products arising from catalyst degradation can be problematic. 1,1-Disubstituted alkenes require longer reaction times, 1,2-disubstituted alkenes are still less reactive, and trisubstituted alkenes are essentially unreactive. This allows preferential hydroboration at the less hindered double bond in limonene (eq 7).29

Asymmetric hydroboration of styrenes with the boranes derived from ephedrine and pseudoephedrine catalyzed by [Rh(nbd)(dppb)]OTf gives excellent regioselectivity, but poor enantioselectivity (eq 8). Better optical yields are obtained using rhodium complexes with more rigid [ferrocenyl]diphosphines.30

Diastereoselective rhodium-catalyzed hydroborations of allylic alcohol derivatives give results complementary to those observed in the uncatalyzed reaction with 9-Borabicyclo[3.3.1]nonane. The syn selectivity of the catalyzed reaction increases as the bulk of the R group increases (syn:anti = 79:21 for R = TBDPS) (eq 9).29

Exocyclic 1,1-disubstituted alkenes also give complementary selectivity (syn:anti = 93:7 for R = TBDMS) (eq 10) to that observed with 9-BBN.29

Aldol Condensations.

The rhodium complex has been utilized as a catalyst in aldol condensation of silyl enol ethers and aldehydes31 or aldehyde equivalents (eqs 11 and 12).32

Although this rhodium complex has been studied in hydrocarbonylation and alkene isomerization, other rhodium catalysts give much higher yields and/or offer greater selectivity.

Finally, the ability of (1) to induce the isomerization of alkenes has been exploited in synthesis.33 In the absence of H2, the dominant product of the isomerization of the illustrated homoallylic silane is the allylic silane shown (eq 13). The reaction is presumably driven by the b-silicon effect.34

1. (a) Schrock, R. R.; Osborn, J. A. JACS 1976, 98, 2134. (b) Schrock, R. R.; Osborn, J. A. JACS 1976, 98, 2143. (c) Schrock, R. R.; Osborn, J. A. JACS 1976, 98, 4450.
2. (a) Hoveyda, A. H.; Evans, D. A.; Fu, G. C. CRV 1993, 93, 1307. (b) Brown, J. M. AG(E) 1987, 26, 190.
3. (a) Brown, J. M.; Chaloner, P. A. JACS 1980, 102, 3040. (b) Brown, J. M.; Evans, P. L.; James, A. P. OS 1989, 68, 64.
4. Anderson, M. P.; Pignolet, L. H. IC 1981, 20, 4101.
5. Chaloner, P. JOM 1984, 266, 191.
6. Halpern, J. Science 1982, 217, 401, and references cited therein.
7. Halpern, J.; Riley, D. P.; Chan, A. S. C.; Pluth, J. J. JACS 1977, 99, 8055.
8. (a) Thompson, H. W.; McPherson, E. JACS 1974, 96, 6232. (b) Brown, J. M.; Naik, R. G. CC 1982, 348. (c) Crabtree, R. H.; Davis, M. W. OM 1983, 2, 681. (d) Stork, G.; Kahne, D. E. JACS 1983, 105, 1072.
9. Evans, D. A.; Morrissey, M. M. JACS 1984, 106, 3866.
10. Machado, A. S.; Olesker, A.; Castillon, S.; Lukacs, G. CC 1985, 330.
11. Hamada, Y.; Kawai, A.; Matsui, T.; Hara, O.; Shioiri, T. T 1990, 46, 4823.
12. Brown, J. M.; Hall, S. A. T 1985, 41, 4639.
13. Brown, J. M.; Hall, S. A. JOM 1985, 285, 333.
14. See for example (a) Schultz, A. G.; McCloskey, P. J. JOC 1985, 50, 5905. (b) Takagi, M.; Yamamoto, K. T 1991, 47, 8869.
15. Landis, C. R.; Halpern, J. JACS 1987, 109, 1746.
16. Brown, J. M.; Cutting, I. J. CC 1985, 578.
17. Sato, S.; Matsuda, I.; Shibata, M. JOM 1989, 377, 347.
18. (a) Hoffman, R. W. CRV 1989, 89, 1841. (b) Johnson, F. CRV 1968, 68, 375.
19. (a) Morrissey, M. M. Ph.D. Thesis, Harvard University, 1986 (Diss. Abstr. Int. B 1987, 48, 444). (b) Hoveyda, A. H.; Evans, D. A.; Fu, G. C. CRV 1993, 93, 1336.
20. Birtwistle, D. H.; Brown, J. M.; Herbert, R. H.; James, A. P.; Lee, K.-F.; Taylor, R. J. CC 1989, 194.
21. Evans, D. A.; Morrissey, M. M.; Dow, R. L. TL 1985, 26, 6005.
22. (a) Brown, J. M.; James, A. P. CC 1987, 181. (b) Brown, J. M.; Cutting, I.; James, A. P. BSF(2) 1988, 211.
23. Evans, D. A.; Dow, R. L.; Shih, T. L.; Takasc, J. M.; Zahler, R. JACS 1990, 112, 5290.
24. Villalobos, A.; Danishefsky, S. J. JOC 1990, 55, 2776.
25. Lautens, M.; Zhang, C.; Crudden, C. M. AG(E) 1992, 31, 232.
26. Goldberg, Y.; Alper, H. TA 1992, 3, 1055.
27. Westcott, S. A.; Blom, H. P.; Marder, T. B.; Baker, R. T. JACS 1992, 114, 8863.
28. Hayashi, T.; Matsumoto, Y.; Ito, Y. TA 1991, 2, 601.
29. Evans, D. A.; Fu, G. C.; Hoveyda, A. H. JACS 1992, 114, 6671.
30. Brown, J. M.; Lloyd-Jones, G. C. TA 1990, 1, 869.
31. (a) Sato, S.; Matsuda, I;. Izumi, Y. TL 1987, 28, 6657. (b) Sato, S.; Matsuda, I.; Izumi, Y. TL 1986, 27, 5517.
32. Reetz, M. T.; Vougioukas, A. E. TL 1987, 28, 793.
33. Matsuda, I.; Kato, T.; Sato, S.; Izumi, Y. TL 1986, 27, 5747.
34. Lambert, J. B.; Emblidge, R. W.; Malany, S. JACS 1993, 115, 1317, and references therein.

David A. Evans & Scott J. Miller

Harvard University, Cambridge, MA, USA

John M. Brown, Timothy P. Layzell, & James A. Ramsden

Oxford University, UK

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