Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Nov 24.
Published in final edited form as: J Am Chem Soc. 2010 Oct 29;132(46):16365–16367. doi: 10.1021/ja107968c

Cobalt-Mediated, Enantioselective Synthesis of C2 and C1 Dienes

W Christopher Boyd 1, Mark R Crimmin 1, Lauren E Rosebrugh 1, Jennifer M Schomaker 1, Robert G Bergman 1,*, F Dean Toste 1,*
PMCID: PMC3058246  NIHMSID: NIHMS249611  PMID: 21033667

In recent years, enantiopure C2- and C1-symmetric dienes have been established as authoritative ligands for asymmetric reactions mediated by late transition metals.1 Despite their utility, current synthetic routes to these ligands remain unsatisfying and often rely upon preparative chiral HPLC, chiral auxiliaries, chiral pool starting materials, or chemoenzymatic reactions to access enantiopure material.26 Hayashi and co-workers reported an effective enantioselective synthesis of C2-symmetric dienes based upon the desymmetrization of norbornadiene.2a-b In the first step in the synthesis of the chiral ligand, a double asymmetric hydrosilylation of norbornadiene with HSiCl3 catalyzed by [PdCl(η3-C3H5)]2 and a chiral phosphine ligand was reported to yield an enantiopure di(silane). However, a multi-step reaction sequence was required to re-establish the diene moiety and build steric bulk into the chiral scaffold.2a-b,3

If the synthesis of C2-dienes could be achieved via a direct desymmetrization of the parent diene by an asymmetric C-H functionalization of the sp2-carbons, it would represent a concise approach to this important class of chiral ligands (Scheme 1). In addition, the use of a prochiral electrophile in the reaction would allow for the preparation of enantiopure ligands that bear additional stereocenters, provided these reactions proceed with high stereoselectivities.

Scheme 1.

Scheme 1

Synthesis of C2-dienes by desymmetrization of norbornadiene.

Recently, we reported a stepwise C-H functionalization of the sp2 carbon of simple alkenes mediated by the reactive intermediate [CpCo(NO)2] (Scheme 2).79 Herein we report that N-benzylated ammonium chloride salts (Chart 1, 1–4) combined with sodium hexamethyldisilazide (NaHMDS) can promote this reaction, while serving as a source of chirality to desymmetrize norbornene and norbornadiene. This methodology has been applied to a modular asymmetric synthesis of C1- and C2-symmetric diene ligands.

Scheme 2.

Scheme 2

C-H functionalization mediated by [CpCo(NO)2].

Chart 1.

Chart 1

N-benzylated ammonium chlorides.

Initial experiments demonstrated that addition of a mixture of NaHMDS and the quininium salt 1 to a mixture of 2- cyclohexen-1-one and cobalt complex 5 gave adduct 6 in 87% d.e. and high yield. Subsequent cycloreversion of 6 in the presence of norbornene yielded 5 and the diastereomerically impure alkene 7 which was obtained in a low but significant 18% e.e. (Table 1, entry 1). The order of reagent addition proved to be critical to the enantioinduction; in fact, adding NaHMDS to a mixture containing the remaining reagents gave the product with only 2% e.e. (entry 2). As NaHMDS alone can promote a racemic background reaction, we suspect a chiral base formed upon mixing 1 with NaHMDS is responsible for the observed enantioinduction.

Table 1.

Screening of reaction conditions for addition of norbornene to 2- cyclohexen-1-one.

graphic file with name nihms249611u1.jpg
Entry Temp.(°C) Solvent Cp' (R*)4NCI time Yield of 6 d.e. of 6 e.e. of 7a
1 25 THF/HMPAc Me4Cp 1 3h 87% 87% 18%
2b 25 THF/HMPAc Me4Cp 1 4h 81% 72% 2%
3 25 THF Me4Cp 1 17.5h 75% 94% 36%
4 −60 to 0 THF Me4Cp 1 6 days (14 h at 0 °C) 74% >99% 58%
5 0 THF tBuCp 1 22 h 35% >99% 72%
6 −45 THF tBuCp 2 17 h 53% >99% 74%
7 −58 THF TBDMSCp 3 15 h 73% >99% 83%
a

NaHMDS and (R*)4NCl premixed and filtered, e.e. of major diastereomer only,

b

NaHMDS added to mixture of substrates and (R*)4NCl,

c

5.3:1 THF to HMPA.

Abbreviations: Me4Cp = η5-Me4C5H, tBuCp = η5-tBuC5H4, TBDMSCp = η5- tBuMe2SiC5H4.

Several alternative solvents gave comparable or inferior results compared to those observed with THF/hexamethylphosphoramide (HMPA). Performing reactions in THF, however, gave significantly higher e.e. at room temperature (entry 3), while lowering the reaction temperature to −60 °C increased the e.e. to 58% (entry 4). It is noteworthy that at lower temperatures, commensurate with the improvement of enantioselectivity, there was an improvement in the diastereoselectivity of the reaction. Employing the bulkier and more directional ligand [η5-(tBu)C5H4] on cobalt allowed optimization to 72% e.e. (entry 5). Comparable results were obtained with [η5-(tBuMe2Si)C5H4] in place of the [η5- (tBu)C5H4] ligand, and using quininium salts bearing trifluoromethyl groups on their N-benzyl moieties allowed further optimization, with the bis(trifluoromethyl) substituted salt 3 yielding the Michael adduct 6 in 73% yield and >99% d.e and the corresponding alkene in 83% e.e. (entry 7).

Using these conditions, we examined the asymmetric C-H functionalization of both norbornene and norbornadiene with 5, 6, and 7-membered cyclic enones. The details of these experiments are provided in the Supporting Information (Table S3). Michael addition reactions proceeded in good yields (66–90%) to give enantioenriched cobalt complexes with high diastereoselectivity (>99% d.e.) with subsequent retrocycloaddition yielding alkene products with moderate enantioselectivity (43–85% e.e.). The enantioselective functionalization of norbornadiene prompted the application of this methodology to asymmetric ligand synthesis (Scheme 3). If two sequential C-H activation/Michael addition steps could be carried out efficiently and benefit from double stereodifferentiation, it could not only result in overall higher enantioselectivities, but would obviate the requirement for enantiopure precursors in the synthesis of chiral diene ligands.

Scheme 3.

Scheme 3

Synthesis of C-H functionalized norbornadiene complexes

The first step of this preparation was the isomerization of enantioenriched complexes 8a-c to 10a-c (Scheme 4). Despite the fact that the alkene that accepts the [Cp’Co(NO)2] fragment is not present in excess, these reactions proceeded cleanly and in high yield (80–93 %) by simply heating samples of 8a-c in toluene or THF solution. A second asymmetric C-H functionalization of 10a-c yielded a mixture of cobalt complexes anti-11a-d and syn-11a-d which could be converted, via alkene exchange with norbornadiene, to a mixture of the corresponding dienes anti-12a-d and syn-12a-d respectively (Table 2).

Scheme 4.

Scheme 4

Isomerization of 8a-c to 10a-c

Table 2.

Asymmetric synthesis of C2- and C1-dienes

graphic file with name nihms249611u2.jpg
n m complex yield of 11a diene anti : syn e.e. of anti-12
1 1 11a 42% 12a 3.7:1 90% (R,R,R,R)
2 2 11b 67% 12b 20:1 96% (R,R,R,R)
2 2 11b 73%b 12b 10:1 94% (S,S,S,S)c
3 3 11c 87% 12c 11:1 91%(R,R,R,R)
2 3 11d 84%b 12d 11:1 96%(R,R,R,R)
a

chiral conditions; 3/NaHMDS pre-mixed, filtered and then added to reaction mixture at −58 °C,

b

reaction conducted at −75 °C.

c

salt 4 used in place of 3.

While poor anti:syn selectivity (1.3–2.7:1) was observed using achiral starting materials and a non-chiral base (see supporting information, Figure S6),10 using enantioenriched starting materials and the asymmetric base mixture (vide supra) anti:syn ratios were 3.7–20:1 across the series of substrates. Under these conditions only one diastereomer of each regioisomeric product was observed, and following retrocycloaddition, dienes anti-12a-d were isolated in good yield and 90–96% e.e.

Recrystallization of a mixture of anti- and syn-12b from n- pentane removed the minor regioisomer and increased the e.e. to 99%. The absolute configuration of anti-12b was determined as (R,R,R,R)-12b by single crystal X-ray diffraction analysis (Figure 1). The enantiomer, (S,S,S,S)-12b, is available in 94 % e.e. via use of chiral ammonium salt 4 in the synthetic sequence.

Figure 1.

Figure 1

ORTEP representation of (R,R,R,R)-12b. H-atoms omitted for clarity. Selected bond lengths (Å): C(5)-C(6) 1.332(2), C(5)-C(4) 1.502(2), C(2)-O(2) 1.216(2). Flack parameter = 0.1451(0.1655).

We propose that the high selectivity in this C-H functionalization reaction sequence occurs because desymmetrization of the nucleophile controls enantioselectivity while approach of the electrophile controls diastereoselectivity (Figure 2). The two stereoselection events are orthogonal and two pairs of non-contiguous stereocenters are set with near perfect control in the enantioselective synthesis of the anti-regioisomer of these diene ligands. The minor regioisomeric products, syn-12a-c, are meso-symmetric and arise from imperfect regiocontrol in the second carboncarbon bond forming step.

Figure 2.

Figure 2

Proposed origin of stereoselectivity.

In a further demonstration of utility, the ease of ligand derivatization by functional group manipulation of the ketone moieties of (R,R,R,R)-12b was demonstrated. Thus, reduction of the carbonyl groups with NaBH4 followed by xanthate formation and Barton-McCombie reduction yielded (1R,4R)- 2,5-dicyclohexylbicyclo[2.2.1]hepta-2,5-diene, an effective ligand for rhodium-catalyzed asymmetric conjugate addition (see supporting information for experimental details).3

In short, we have demonstrated an asymmetric synthesis of C2- and C1-symmetric dienes, based upon a C-H functionalization reaction mediated by cobalt di(nitrosyl) complexes that proceeds with excellent regio-, diastereo- and enantiocontrol. We are continuing to study the nature of the active base derived from quininium salt/metal amide mixtures and the applications of the enantiopure products described herein as ligands in asymmetric catalysis.

Supplementary Material

1_si_001
2_si_002

Acknowledgments

We are grateful for financial support from the NSF in the form of a research grant to R.G.B. (CHE-0841786) and a predoctoral fellowship to W.C.B. M.R.C. acknowledges the Royal Commission for the Exhibition of 1851 for a postdoctoral fellowship. J.M.S. acknowledges support from a National Institutes of Health NRSA Ruth Kirschstein postdoctoral fellowship. We are also grateful to Melanie Chiu, Jerome Volkman, Yiming Wang, and Chen Zhao for helpful discussions.

Footnotes

Supporting Information Available: Experimental procedures and spectroscopic data for all new compounds; conditions for chiral GC and HPLC methods and CIF file for 12b. This material is available free of charge via the Internet at http://pubs.acs.org.

References

  • 1.For review articles see Glorius F. Angew Chem, Int Ed. 2004;43:3364. doi: 10.1002/anie.200301752.Defieber C, Grutzmacher H, Carreira EM. Angew Chem, Int Ed. 2008;47:4482. doi: 10.1002/anie.200703612.Shintani R, Hayashi T. Aldrichimica Acta. 2009;42:31.
  • 2.(a) Uozumi Y, Lee SY, Hayashi T. Tet Lett. 1992;33:7185. [Google Scholar]; (b) Hayashi T, Ueyama K, Tokunaga N, Yoshida K. J Am Chem Soc. 2003;125:11508. doi: 10.1021/ja037367z. [DOI] [PubMed] [Google Scholar]; (c) Berthon-Gelloz G, Hayashi T. J Org Chem. 2006;71:8957. doi: 10.1021/jo061385s. [DOI] [PubMed] [Google Scholar]; (d) Tokunaga N, Otomaru Y, Okamoto K, Ueyama K, Shintani R, Hayashi T. J Am Chem Soc. 2004;126:13584. doi: 10.1021/ja044790e. [DOI] [PubMed] [Google Scholar]; (e) Shintani R, Okamoto K, Otomaru Y, Ueyama K, Hayashi T. J Am Chem Soc. 2005;127:54. doi: 10.1021/ja044021v. [DOI] [PubMed] [Google Scholar]; (f) Otomaru Y, Okamoto K, Shintani R, Hayashi T. J Org Chem. 2005;70:2503. doi: 10.1021/jo047831y. [DOI] [PubMed] [Google Scholar]; (g) Okamoto K, Hayashi T, Rawal VH. Chem Commun. 2009:4815. doi: 10.1039/b904624k. [DOI] [PubMed] [Google Scholar]
  • 3.Noël T, Vandyck K, Van der Eycken J. Tetrahedron. 2007;63:12961. [Google Scholar]
  • 4.(a) Paquin JF, Defieber C, Stephenson CRJ, Carreira EM. J Am Chem Soc. 2005;127:10850. doi: 10.1021/ja053270w. [DOI] [PubMed] [Google Scholar]; (b) Fischer C, Defieber C, Suzuki T, Carreira EM. J Am Chem Soc. 2004;126:1628. doi: 10.1021/ja0390707. [DOI] [PubMed] [Google Scholar]; (c) Defieber C, Paquin JF, Serna S, Carreira EM. Org Lett. 2004;6:3873. doi: 10.1021/ol048240x. [DOI] [PubMed] [Google Scholar]; (d) Paquin JF, Stephenson CRJ, Defieber C, Carreira EM. Org Lett. 2005;2:3821. doi: 10.1021/ol051533l. [DOI] [PubMed] [Google Scholar]
  • 5.Luo Y, Carnell AJ. Angew Chem, Int Ed. 2010;49:2750. doi: 10.1002/anie.200907033. [DOI] [PubMed] [Google Scholar]
  • 6.Brown MK, Corey EJ. Org Lett. 2010;12:172. doi: 10.1021/ol9025793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.(a) Brunner H. J Organomet Chem. 1968;12:517. [Google Scholar]; (b) Brunner H, Loskot S. Angew Chem, Int Ed Engl. 1971;10:515. [Google Scholar]; (c) Brunner H, Loskot S. J Organomet Chem. 1973;61:401. [Google Scholar]
  • 8.(a) Becker PN, Bergman RG. Organometallics. 1983;2:787. [Google Scholar]; (b) Becker PN, Bergman RG. J Am Chem Soc. 1983;105:2985. [Google Scholar]
  • 9.(a) Schomaker JM, Boyd WC, Stewart IC, Toste FD, Bergman RG. J Am Chem Soc. 2008;130:3777. doi: 10.1021/ja800738d. [DOI] [PubMed] [Google Scholar]; (b) Schomaker JM, Toste FD, Bergman RG. Org Lett. 2009;11:3698. doi: 10.1021/ol901542n. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Lensink C, Xi SK, Daniels LM, Verkade JG. J Am Chem Soc. 1989;111:3478. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1_si_001
2_si_002

RESOURCES