Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Am Chem Soc. Author manuscript; available in PMC 2010 August 5.
Published in final edited form as:
PMCID: PMC2760382

Allenes in Asymmetric Catalysis. Asymmetric Ring-Opening of Meso-Epoxides Catalyzed by Allene-Containing Phosphine Oxides


Unsymmetrically substituted allenes (1,2 dienes) are inherently chiral and can be prepared in optically pure form. Nonetheless, to date the allene framework has not been incorporated into ligands for asymmetric catalysis. Since allenes project functionality differently than either tetrahedral carbon or chiral biaryls, they may create complementary chiral environments. This study demonstrates that optically active C2 symmetric allene-containing bisphosphine oxides can catalyze the addition of SiCl4 to meso epoxides with high enantioselectivity. The epoxide-opening likely involves generation of a Lewis acidic, cationic (bisphosphine oxide)SiCl3 complex. The fact that high asymmetric induction is observed suggests that allenes may represent a new platform for the development of ligands and catalysts for asymmetric synthesis.

Asymmetric catalysis has revolutionized the synthesis of chiral, non-racemic materials. It has been the subject of academic inquiry and industrial implementation. The last several decades have witnessed continued introduction and development of chiral organic catalysts and ligands for transition and main-group metals.1 Most existing ligands and organic catalysts are characterized by either of two types of chirality: some species owe their chirality to stereogenic atoms, usually tetrahedral carbon or phosphorus. Many others are chiral by virtue of hindered rotation around a carbon-carbon single bond, the exemplar of which is the binaphthyl backbone. Some of the most successful catalysts combine the elements of both central and axial chirality.2,3

Allenes can be chiral but have made only limited contributions to asymmetric catalysis to date. They were predicted to display chirality by van’t Hoff over a century ago,4 but, to the best of our knowledge, allenes have been used only once in catalytic quantities to control the stereochemical outcome of a reaction. In that report, Soai and coworkers demonstrated that optically active disubstituted allenes induced asymmetry in the addition of diisopropyl zinc to a pyrimidine carboxaldehyde.5 The actual effectiveness of the allene as a chiral modulator was difficult to gauge because other experiments with the same system indicate that a strong non-linear effect coupled with auto-catalysis conspire to amplify even unmeasurable enantiomeric excesses.6

We hypothesized that allenes of the general structure A or B might represent attractive frameworks for developing ligands for asymmetric catalysis.7 The unique architecture of the allene projects functionality differently than either tetrahedral carbon or chiral biaryls and, therefore, may create a complementary chiral environment. Additionally, the ortho-relationship between the donor group and the allene might facilitate asymmetric induction by orienting metals or sites of reactivity towards the source of chirality. Finally, this basic platform could be amenable to introduction of a variety of donor groups and could find utility in main group and transition-metal catalysis or organocatalysis. Here we demonstrate that the chirality of an allene can indeed be communicated effectively to a transition state in the context of an enantioselective transformation. This observation suggests the possibility of a new class of ligands and catalysts for use in asymmetric synthesis.

An external file that holds a picture, illustration, etc.
Object name is nihms132637f3.jpg

Since phosphine oxides have shown promise as chiral Lewis bases,8 we prepared a series of optically active mono- and bisphosphine oxides that contain an allene backbone (Scheme 1). Thus, proparygylic acetates 1 were prepared from the corresponding alcohols which, in turn, were generated in optically active form through either asymmetric addition of terminal alkynes to the corresponding ketone9 or resolution of the racemic tertiary alcohol.10,11 Propargylic substitution with diorganocuprate reagents or aryl zinc halides then generated the tetrasubstituted allenes 2. The aryl bromides were treated with tert-butyl lithium, and the corresponding aryl lithium intermediates were trapped with diaryl chloro-phosphine oxides. In this way, we synthesized several mono- and bisphosphine oxides displaying diverse substitution patterns on both the allene backbone and the phosphorus. Most of the prospective catalysts were prepared in optically pure form through recrystallization of the bisphosphine oxide (3b, 3c) or the dibromoallene (3d3h).12

In the solid-state bisphosphine oxide 3d adopts a conformation characterized by π-stacking interactions involving two phenyl rings of the phosphine oxides and one of the backbone phenyl rings (Fig 1). A consequence of this arrangement is that the two oxygen atoms project in roughly the same direction. The significance of this conformation is not clear however, as the bisphosphine oxide appears C2 symmetric on the NMR time scale.

Figure 1
X-ray crystal structure of 3d. Hydrogens, an identical molecule of 3d and a molecule of water have been removed for clarity.

We initiated our studies of catalysis by considering the asymmetric addition of SiCl4 to meso epoxides. The Denmark group demonstrated that optically active phosphorus triamides could catalyze this transformation;13 subsequent reports revealed the utility of pyridine N-oxides and phosphine oxides.14 Mechanistically, these reactions are thought to involve coordination of one or more Lewis bases to SiCl4 to generate a more Lewis acidic species (LB)nSiCl3+ (LB = Lewis base) capable of activating the epoxide toward ring opening.13b

Cis-stilbene oxide was exposed to SiCl4 and iPr2NEt in the presence of catalytic allene-containing mono- or bisphosphine oxides 3 (AllenePO, Table 1). These experiments revealed several aspects of the ring-opening reaction. 1) Allene-containing ligands can induce enantioselectivity in catalytic reactions. 2) In general, the bisphosphine oxides displayed higher reactivity and enantioselectivity than the monphosphine oxide (3a vs. 3b). 3) Substantial variation was tolerated on the allene itself, as both the methyl and phenyl substituted catalysts (3c, 3d) were reactive and enantioselective. 4) The ring-opening is sensitive to the aryl rings on the phosphine oxide. Both electron-donating groups (entries 6 and 8) and, more profoundly, electron withdrawing groups (entries 7 and 9) decreased reactivity and, where measurable, selectivity. 5) The diphenyl substituted catalyst 3d is a highly reactive and enantioselective catalyst, displaying nearly 1000 turnovers and generating the chlorohydrin in 94% ee (entry 10).

Table 1
Evaluation of phosphine oxides as catalysts for the addition of SiCl4 to cis-stilbene oxide.a

A sampling of meso epoxides was opened with SiCl4 in the presence of catalyst 3d (Table 2). In general, the reaction is effective for substituted stilbene oxides. Both electron-releasing and electron-withdrawing substituents are tolerated in the meta or para position, although the electron poor substrates tended to require increased catalyst loadings (Entries 4, 7, 8). In contrast, ortho substituted stilbene oxides were not reactive (Entry 10). Finally, the chlorohydrins derived from linear, cyclic or bicyclic aliphatic epoxides were isolated in high yield but with rather unimpressive enantiomeric excesses (Entries 11–13).

Table 2
Asymmetric ring-opening of meso-epoxides.a

Allenes have attracted attention from the synthetic community because, among other reasons, they react with both nucleophiles and electrophiles, often under mild conditions. In the present circumstance, however, this reactivity profile represented a potential liability inasmuch as catalyst stability would be critical for a practical synthetic method. In this regard, we were encouraged that we could recover catalyst 3d in 94% yield from a reaction involving the ring-opening of epoxide 4a.11

An external file that holds a picture, illustration, etc.
Object name is nihms132637f4.jpg

More than a century after van’t Hoff recognized that allenes could be chiral, we have demonstrated that their chirality can be harnessed in the service of asymmetric catalysis. In the present example we have prepared organic Lewis bases and have shown that they can activate SiCl4. However, this design principle may extend equally well to other classes of organic catalysts or ligands for transition metals.7 Their utility in these contexts is the subject of current investigations.

Supplementary Material





X-ray crystallography performed by Vincent Lynch (UT Austin) and Radha Akella (UTSW). We are grateful to the NIGMS (R01-GM074822), the Welch Foundation, and Amgen. JMR is a fellow of the Alfred P. Sloan Foundation.


Supporting Information Available. Complete experimental details and characterization data including cif files. This material is available free of charge via the Internet at


1. Jacobsen EN, Pfaltz A, Yamamoto H, editors. Comprehensive Asymmetric Catalysis. Berlin: Springer-Verlag; 1999.
2. Feringa BL. Acc. Chem. Res. 2000;33:346. [PubMed]
3. For the use of axially chiral ligands based on a spirobiindane backbone see: Fu Y, Xie J-H, Hu A-G, Zhao H, Wang L-X, Zhou Q-L. Chem. Commun. 2002:480.
4. Hoff JH van’t, Eiloart A., translators. The Arrangement of Atoms in Space. London: Green and Co; 1898. pp. 103–104.
5. Sato I, Matsueda Y, Kadowaki K, Yonekubo S, Shibata T, Soai K. Helv. Chim. Acta. 2002;85:3383.
6. Kawasaki T, Sato M, Ishiguro S, Takahiro S, Yosuke M, Sato I, Nishino H, Inoue Y, Saoi K. J. Am. Chem. Soc. 2005;127:3274. [PubMed]
7. Krause and co-workers reported an allene-containing (bispyridine) Ag+ complex, although no catalysis was demonstrated. Lohr S, Averbeck J, Schurmann M, Krause N. Eur. J. Inorg. Chem. 2008:552.
8. Denmark SE, Beutner GL. Angew. Chem. Int. Ed. 2008;47:1560. [PubMed]
9. Cozzi PG, Alesi S. Chem. Commun. 2004:2448. [PubMed]
10. Toda F, Tanaka K. Tetrahedron. Lett. 1981;22:4669.
11. See supporting information for details.
12. 3a and 3b were prepared and used in 70 and 79% ee, respectively.
13. (a) Denmark SE, Barsanti PA, Wong K-T, Stavenger RA. J. Org. Chem. 1998;63:2428. [PubMed] (b) Denmark SE, Barsanti PA, Beutner GL, Wilson TW. Adv. Syn. Catal. 2007;348:567.
14. (a) Tao B, Lo MM-C, Fu GC. J. Am. Chem. Soc. 2001;123:353. [PubMed] (b) Nakajima M, Saito M, Uemura M, Hashimoto S. Tetrahedron Lett. 2002;43:8827. (c) Tokuoka E, Kotani S, Matsunaga H, Ishizuka T, Hashimoto S, Nakajima M. Tetrahedron: Asymm. 2005;16:2391. (d) Takenaka N, Sarangthem RS, Captain B. Angew. Chem. Int. Ed. 2008;47:9708. [PubMed]