Chiral anion-mediated asymmetric ring opening of meso-aziridinium and episulfonium ions

Article abstract of DOI:10.1021/ja806431d

Reactions proceeding through cationic intermediates that lack a Lewis or Bronsted basic site present a challenge for traditional asymmetric catalysis based on chiral metals or organocatalysts. We present an enantioselective ring opening of tetrasubstituted meso-aziridinium ions with alcohol nucleophiles proceeding through a chiral ion pair with a binaphthol-phosphate anion. The reaction is initiated by silver-induced ring closure of β-chloroamines using the Ag salt of the chiral anion as in situ generated catalyst. Use of insoluble Ag₂CO₃ as silver source is essential to obtain high enantioselectivity; we believe the chiral phosphate acts as a chiral anion phase transfer catalyst to bring silver ion into the organic phase. The chiral anion concept can also be extended to the related asymmetric opening of meso-episulfonium ions generated by protonation of trichloroacetimidates vicinal to sulfides. Copyright

Full text of DOI:10.1021/ja806431d

Published on Web 10/21/2008
Chiral Anion-Mediated Asymmetric Ring Opening of meso-Aziridinium and
Episulfonium Ions
Gregory L. Hamilton, Toshio Kanai, and F. Dean Toste*
Department of Chemistry, UniVersity of California, Berkeley, California 94720
Received August 13, 2008; E-mail: fdtoste@berkeley.edu
Recently, chiral anions have been uncovered as a potent means
for inducing asymmetry in catalytic reactions.1 Introduced in the
arena of cationic transition metal catalysis in 2000 by Arndtsen
and co-workers,2 the strategy has recently been demonstrated by
our group and others to be capable of providing products with
exceptional enantioselectivity, even in cases where chiral ligands
have failed.3,4 Parallel to these discoveries, work spearheaded by
List has successfully utilized chiral counteranions as supplements
or alternatives to chiral amines in iminium-based organocatalytic
reactions.5 Instrumental to all these successes has been the use of
Figure 1. Proposed chiral anion phase transfer catalysis (left) as compared
to chiral cation PTC.
axially chiral phosphate ions initially developed by Akiyama and
Terada as chiral Brønsted acids for enantioselective additions to
imines.6-8 A goal of our laboratory has been probing the stereo-
differentiating ability of these chiral phosphates beyond cases
involving a putative hydrogen bond to an imine or other electrophile.
Moving forward from the existing precedent with an eye toward
transformations without a current asymmetric variant, we reasoned
that chiral anions could provide a much-needed solution to the
problem of reactions proceeding through cationic intermediates that
do not possess a (Lewis or Brønsted) basic site capable of
coordinating to a metal or proton. One approach to this problem,
advanced quite recently by Jacobsen and colleagues, involves the
binding of a halide counteranion by a chiral thiourea catalyst,
thereby enabling asymmetric additions to iminium and oxocarbe-
nium intermediates.9,10
Table 1. Reaction Development
entry
catalyst
AgB
additive
yield (%)a
ee (%)b
1
2
3
4
5
6
7
2a
2a
2b
2b
none
2b
Ag2CO3
AgOTs
Ag2CO3
none
Ag2CO3
Ag2CO3
Ag2CO3
none
none
none
none
none
4Å ms
4Å ms
77
88
74
trace
trace
84
94
56
94
ND
ND
94
We envisioned a complementary strategy initiated by an “anion
metathesis,” wherein the silver salt of a chiral anion could abstract
a halide from an organic fragment to produce a chiral ion pair (eq
1). The resulting cationic intermediate could then be intercepted
by a nucleophile in an enantioselective manner directed by the chiral
counteranion. After proton transfer, this sequence would result in
the desired optically active substitution product and the conjugate
acid of the chiral anion.
2b (recycled)
83
94
a Conditions: 15 mol % 2, 1.2 equiv (in Ag) AgB, 4 equiv 3, 0.1 M
in toluene, 50 °C, 24 h. Yields refer to isolated material. b Determined
by 1H NMR using chiral shift reagent; see Supporting Information.
Surprisingly, examination of the literature reveals no successful
examples of chiral anion phase transfer catalysis, although we came
across one report of work aiming toward the concept.14 In it the
authors investigated the asymmetric ring opening of meso-aziri-
dinium ions generated from ring closure of ꢀ-chloro tertiary amines,
obtaining products with a maximum of 30% yield and 15% ee. No
attempt was made to achieve catalytic turnover. This transformation
sparked our interest for two reasons. First, since the meso
intermediate is a quaternary-substituted aziridinium ion it is unable
to be activated by a Brønsted or Lewis acid. Furthermore, the
reaction generates two stereocenters and allows a direct synthesis
of valuable unprotected amines.15-17
The challenge for this seemingly intuitive approach results from
reforming the chiral silver salt so that the overall transformation is
catalytic in the chiral material. Addition of an achiral Ag(I) source
to the reaction mixture in order to turn over the chiral anion
necessarily brings an accompanying achiral counteranion that can
lead to racemic product. An attractive solution came to us in the
form of establishing a phase separation between the substrate and
chiral anion catalyst (bulk solution) and the stoichiometric silver
salt (solid phase). The resulting catalytic cycle can be thought of
as chiral anion phase transfer catalysis in analogy to the established
chiral cation variety (Figure 1).11-13
We initiated our investigation with racemic chloroamine 1 and
neopentyl alcohol as nucleophile (Table 1). For chiral silver salt
we chose the binaphthol-phosphate 2 that has proven successful
for chiral counteranion-based reactions developed by our group and
others.3,5 Our confidence in this anion was further bolstered by the
9
14984 J. AM. CHEM. SOC. 2008, 130, 14984–14986
10.1021/ja806431d CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1. Nucleophile and Substrate Scopea
Secondary, tertiary, and relatively hindered primary alcohols all
react with high enantioselectivity. Notably, these classes of alcohols
generate products that would be generally difficult to prepare by a
traditional Williamson ether synthesis due to competing elimination
and sluggish reactivity of the necessary electrophiles. In addition
to structural variation, functional group tolerance was also found
to be high, with acid-labile (acetal, silyl ether) and oxidation-
sensitive (indole, electron rich benzyl alcohol) functionalities
surviving the reaction conditions without incident.
Our reaction can also be applied to substrates containing different
amine substituents. Acyclic (dimethyl) substitution, other ring sizes,
heteroatoms, and fused rings are well tolerated. Furthermore,
tetrahydroisoquinoline 15 demonstrates that the amine need not be
symmetrically substituted (R1 * R2).
We also investigated a chiral racemic alcohol as nucleophile in
an effort to obtain a product with three stereocenters. Since the
alcohol is used in excess relative to the more expensive amine
fragment, 100% theroretical yield of one diastereomer of the desired
adduct is possible assuming the catalyst can mediate an efficient
kinetic resolution. In fact, when the reaction was performed using
racemic alcohol 16 it proceeded as we hoped to furnish product 17
with exceptional de and ee (eq 2).
a Conditions: 15 mol % 2b, 0.6 equiv Ag2CO3, 4 equiv R3OH, 0.1 M in
toluene, 4Å MS, 50 °C, 24-36 h. Yields refer to isolated material; ee’s
determined by chiral HPLC or 1H NMR using chiral shift reagent, see
t
Supporting Information. b Run using 30 equiv BuOH.
high solubility of its salts in organic solvents, a property that
facilitates the phase transfer process.
Although a valuable feature of the described reaction is the direct
synthesis of substituted ꢀ-alkoxy amines, 1,2-amino alcohols can
also be prepared by a two-step procedure. We found that 2,6-
dimethyl-4-methoxybenzyl ether 5 could be efficiently cleaved by
the agency of CAN without significant overoxidation (eq 3).
As mentioned previously, we expected that the choice of
stoichiometric silver salt would prove critical. Testing several
insoluble basic silver compounds revealed that Ag2CO3 efficiently
turned over the catalyst while being mild enough to not significantly
react with starting materials or products: using 15 mol% of the
silver phosphate salt 2a and a slight excess of Ag2CO3 we obtained
the desired substitution product in good yield and excellent ee (Table
1, entry 1). In contrast, and as expected, addition of more soluble
silver compounds such as AgOTs led to significantly lower ee due
to competition from the achiral anion (entry 2). Only the shown
diastereomer resulting from double inversion was obtained; in no
cases did we observe competition from a direct SN2 substitution of
the chloride.
Importantly, since Ag2CO3 rapidly generates the silver phosphate
salt from the corresponding phosphoric acid, the more easily
prepared and stored 2b performs equally well as precatalyst (entry
3). Accordingly, control experiments demonstrated that both the
chiral phosphoric acid and Ag2CO3 are necessary for the reaction
to proceed (entries 4 and 5). The failure of Ag2CO3 to promote the
reaction by itself is consistent with the phase transfer role of the
phosphate anion; without it, there is no Ag+ in solution. By way
of yield optimization, it was found to be desirable to add powdered
4Å molecular sieves to minimize competitive nucleophilic addition
of the H2O equivalent generated by protonation of Ag2CO3 (entry
6). Finally, we wanted to establish that the catalyst could be
recycled. Simple reacidification by aqueous workup of the phos-
phate following normal chromatographic purification allows re-
covery of 2b in quantitative yield. Catalyst obtained thusly provides
product with identical yield and ee when used in another reaction
(entry 7).
During studies on the aziridinium opening reaction, we hypoth-
esized that our chiral anion strategy could be applied to a wide
variety of cationic reactions, as long as we could selectively generate
the key cationic intermediate as an ion pair with a chiral counter-
anion. We considered a reasonable extension of our chemistry to
the sulfur analogue of the ring opening reaction: namely, the
desymmetrization of meso-episulfonium ions;18 however, it was
apparent that we could not simply employ our silver-halide
abstraction method of cation generation, since the sulfide would
likely bind Ag+. Instead, we settled on trichloroacetimidate as a
leaving group that could be directly activated by a chiral phosphoric
acid (Scheme 2). After protonation of the imidate, loss of trichlo-
roacetamide and ring closure would generate the desired episulfo-
nium ion paired with the chiral phosphate. Interception of the
intermediate with a nucleophile would furnish the chiral sulfide
product and complete the catalytic cycle after subsequent proton
transfer.
In practice, we were pleased to find that the reaction worked as
designed. A small sample of substrates containing different sulfur
substituents reacted with various alcohols to provide in all cases
Next we turned to establishing the scope of the enantioselective
ring opening by varying the alcohol nucleophile (Scheme 1).
9
J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008 14985

C O M M U N I C A T I O N S
Scheme 2. Asymmetric Opening of Episulfonium Ionsa
(2) (a) Llewellyn, D. B.; Adamson, D.; Arndtsen, B. A. Org. Lett. 2000, 2,
4165. For follow-up research, see:(b) Llewellyn, D. B.; Arndtsen, B. A.
Tetrahedron: Asymmetry 2005, 16, 1789. Another study: (c) Dorta, R.;
Shimon, L.; Milstein, D. J. Organomet. Chem. 2004, 689, 751.
(3) (a) Hamilton, G. L.; Kang, E. J.; Mba, M.; Toste, F. D. Science 2007, 317,
496. (b) Mukherjee, S.; List, B. J. Am. Chem. Soc. 2007, 129, 11336.
(4) For examples of chiral Brønsted acids used in cooperation with transition
metals, see:(a) Komanduri, V.; Krische, M. J. J. Am. Chem. Soc. 2006,
128, 16448. (b) Rueping, M.; Antonchick, A. P.; Brinkmann, C. Angew.
Chem., Int. Ed. 2007, 46, 6903.
(5) (a) Mayer, S.; List, B. Angew. Chem., Int. Ed. 2006, 45, 4193. (b) Martin,
N. J. A.; List, B. J. Am. Chem. Soc. 2006, 128, 13368. (c) Wang, X.; List,
B. Angew. Chem., Int. Ed. 2008, 47, 1119. (d) Wanner, M. J.; van der
Haas, R. N. S.; de Cuba, K. R.; van Maarseveen, J. H.; Hiemstra, H. Angew.
Chem., Int. Ed. 2007, 46, 7485. For a related study using Brønsted acid
catalysis, see:(e) Storer, R. I.; Carrera, D. E.; Ni, Y.; MacMillan, D. W. C.
J. Am. Chem. Soc. 2006, 128, 84.
(6) (a) Uraguchi, D.; Terada, M. J. Am. Chem. Soc. 2004, 126, 5356. (b)
Akiyama, T.; Itoh, J.; Yokota, K.; Fuchibe, K. Angew. Chem., Int. Ed. 2004,
43, 1566.
(7) As early as ref 6b the limiting case transition state of fully ionized phosphate/
iminium ion pair was proposed, thereby foreshadowing the future develop-
ments in chiral counteranion chemistry.
(8) Reviews on chiral Brønsted acids: (a) Akiyama, T. Chem. ReV. 2007, 107,
5744. (b) Akiyama, T.; Itoh, J.; Fuchibe, K. AdV. Synth. Catal. 2006, 348,
999. (c) Connon, S. J. Angew. Chem., Int. Ed. 2006, 45, 3909.
(9) (a) Raheem, I. T.; Thiara, P. V.; Peterson, E. A.; Jacobsen, E. N. J. Am.
Chem. Soc. 2007, 129, 13404. (b) Raheem, I. T.; Thiara, P. V.; Jacobsen,
E. N. Org. Lett. 2008, 10, 1577. (c) Reisman, S. E.; Doyle, A. G.; Jacobsen,
E. N. J. Am. Chem. Soc. 2008, 130, 7198.
(10) For use of a stoichiometric chiral anion in an asymmetric Stevens
rearrangement: Gonc¸alves-Farbos, M-H.; Vial, L.; Lacour, J. Chem.
Commun. 2008, 829.
(11) Reviews on chiral phase transfer catalysis: (a) Ooi, T.; Maruoka, K. Angew.
Chem., Int. Ed. 2007, 46, 4222. (b) O’Donnell, M. J. In Catalytic
Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed.; Wiley-VCH: New York,
2000; Chapter 10, pp 727-755. (c) O’Donnell, M. J. Acc. Chem. Res. 2004,
37, 506. (d) Lygo, B.; Andrews, B. I. Acc. Chem. Res. 2004, 37, 518.
(12) For selected recent examples of chiral cation phase transfer catalysis, see:
(a) Elsner, P.; Bernardi, L.; Salla, G. D.; Overgaard, J.; Jørgensen, K. A.
J. Am. Chem. Soc. 2008, 130, 4897. (b) Moss, T. A.; Fenwick, D. R.; Dixon,
D. J. J. Am. Chem. Soc. 2008, 130, 10076. (c) Wang, X.; Kitamura, M.;
Maruoka, K. J. Am. Chem. Soc. 2007, 129, 1038. (d) Poulsen, T. B.;
Bernardi, L.; Alema´n, J.; Overgaard, J.; Jørgensen, K. A. J. Am. Chem.
Soc. 2007, 129, 441.
(13) Many chiral phase transfer reactions go by an interfacial mechanism; this
is possible for our concept also but is not shown for simplicity.
(14) Carter, C.; Fletcher, S.; Nelson, A. Tetrahedron: Asymmetry 2003, 14, 1995.
(15) For selected studies of biologically active compounds bearing a trans-ꢀ-
aminoether motif, see: (a) Plouvier, B.; Beatch, G. N.; Jung, G. L.; Zolotoy,
A.; Sheng, T.; Clohs, L.; Barrett, T. D.; Fedida, D.; Wang, W. Q.; Zhu,
J. J.; Liu, Y.; Abraham, S.; Lynn, L.; Dong, Y.; Wall, R. A.; Walker,
M. J. A. J. Med. Chem. 2007, 50, 2818. (b) Rebollo, O.; del Olmo, E.;
Ruiz, G.; Lopez-Perez, J. L.; Gimenez, A.; San Feliciano, A. Bioorg. Med.
Chem. Lett. 2008, 18, 184. (c) Witty, D. R.; Bateson, J.; Hervieu, G. J.;
Al-Barazanji, K.; Jeffrey, P.; Hamprecht, D.; Haynes, A.; Johnson, C. N.;
Muir, A. I.; O’Hanlon, P. J.; Stemp, G.; Stevens, A. J.; Thewlis, K.;
Winborn, K. Y. Bioorg. Med. Chem. Lett. 2006, 16, 4872. (d) Takeuchi,
K.; Kohn, T. J.; Sall, D. J.; Denney, M. L.; McCowan, J. R.; Smith, G. F.;
Gifford-Moore, D. S. Bioorg. Med. Chem. Lett. 1999, 9, 759.
a Conditions: 15 mol % 2b, 2 equiv R2OH, 0.1 M in toluene, room temp,
12 h. Yields refer to isolated material; ee’s determined by chiral HPLC,
see Supporting Information.
products in very high yield and good ee (Scheme 2). Although the
reaction is initiated by protonation by the phosphoric acid, it is
clearly mechanistically distinct from chiral Brønsted catalyzed
reactions where a prochiral imine is activated by an acid. Since
the trichloroacetimidate starting material is chiral while the
intermediate episulfonium is meso, the ring opening must be
enantiodetermining. Therefore the enantioselectivity likely results
from an ion pair with the chiral anion rather than a hydrogen bond
to the electrophile.
In summary, we have developed two enantioselective ring
opening reactions of cationic intermediates that could not be readily
achieved using previously established methods of asymmetric
catalysis. In doing so we have expanded the range of chiral
counteranion-mediated reactions beyond iminium and oxocarbenium
based transformations. Furthermore, we have demonstrated that a
primary barrier to designing enantioselective reactions of more types
of cationic intermediates is the selective generation of the critical
chiral ion pair. Toward that end, we have shown here the success
of two methods: one based on a unique chiral anion phase transfer
catalytic cycle and the other on activation of trichloroacetimidates.
The flexibility of the two strategies should facilitate new reaction
development since either an organohalide or an alcohol (in the form
of the imidate) can serve as precursor.
Acknowledgment. We gratefully acknowledge Merck Research
Laboratories, Bristol-Myers Squibb, and Novartis for funding.
G.L.H. thanks the NSF for a graduate fellowship. T.K. thanks
Dainippon Sumitomo Pharma for funding and research opportunity.
(16) For asymmetric ring opening of protected aziridines with TMSN3, see:(a)
Fukuta, Y.; Mita, T.; Fukuda, N.; Kanai, M.; Shibasaki, M. J. Am. Chem.
Soc. 2006, 128, 6312. (b) Li, Z.; Ferna´ndez, M.; Jacobsen, E. N. Org. Lett.
1999, 1, 1611. With TMSCN: (c) Mita, T.; Fujimori, I.; Wada, R.; Wen,
J.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2005, 127, 11252.
(17) Recently a chiral phosphoric acid has also been used for desymmetrization
of meso-aziridines with TMSN3. In this case covalent activation of
benzoylated aziridine by silylated phosphate was proposed: Rowland, E. B.;
Rowland, G. B.; Rivera-Otero, E.; Antilla, J. C. J. Am. Chem. Soc. 2007,
129, 12084.
Supporting Information Available: Experimental procedures and
compound characterization data. This material is available free of charge
via the Internet at http://pubs.acs.org.
(18) Episulfonium ions are intermediates in numerous reactions but have not
been exploited in the context of asymmetric catalysis; for a review, see:
Fox, D. J.; House, D.; Warren, S. Angew. Chem., Int. Ed. 2002, 41, 2462.
References
(1) For a review on chiral anions, see: Lacour, J.; Hebbe-Viton, V. Chem.
Soc. ReV. 2003, 32, 373.
JA806431D
9
14986 J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008