Efficient aziridination of olefins catalyzed by mixed-valent dirhodium(II,III) caprolactamate

Article abstract of DOI:10.1021/ol0510973

(Chemical Equation Presented) A mild, efficient, and selective aziridination of olefins catalyzed by dirhodium(II) caprolactamate [Rh ₂(cap)₄·2CH₃CN] is described. Use of p-toluenesulfonamide (TsNH₂), N-bromosuc

Full text of DOI:10.1021/ol0510973

ORGANIC
LETTERS
2005
Vol. 7, No. 13
2787-2790
Efficient Aziridination of Olefins
Catalyzed by Mixed-Valent
Dirhodium(II,III) Caprolactamate
Arthur J. Catino, Jason M. Nichols, Raymond E. Forslund, and Michael P. Doyle*
Department of Chemistry and Biochemistry, UniVersity of Maryland,
College Park, Maryland 20742
mdoyle3@umd.edu
Received May 11, 2005
ABSTRACT
A mild, efficient, and selective aziridination of olefins catalyzed by dirhodium(II) caprolactamate [Rh2(cap)4‚2CH3CN] is described. Use of
p-toluenesulfonamide (TsNH2), N-bromosuccinimide (NBS), and potassium carbonate readily affords aziridines in isolated yields of up to 95%
5+
under extremely mild conditions with as little as 0.01 mol % Rh2(cap)4. Aziridine formation occurs through Rh2 -catalyzed aminobromination
5+
and subsequent base-induced ring closure. An X-ray crystal structure of a Rh2 halide complex, formed from the reaction between Rh2(cap)4
and N-chlorosuccinimide, has been obtained.
Olefin aziridination is a powerful approach for the incorpora-
tion of nitrogen into organic compounds.1,2 Largely regarded
for their synthetic versatility, aziridines are well suited for
ring opening with an assortment of nucleophiles, yielding
functionalized amines.3 Despite their value and utility,
however, methods for the direct preparation of aziridines
remain limited. Transition metal-catalyzed processes in
conjunction with an appropriate nitrene precursor (e.g.,
iminophenyliodinanes such as TsNdIPh, or in situ variants)
have received considerable attention,4,5 for which catalysis
via dirhodium(II,II) complexes (Rh24+) holds a prominent
position.6 However, drawbacks in the uses of this methodol-
ogy arise from high catalyst loadings, limited shelf life of
TsNdIPh, competing C-H insertion, and/or poor selectivity.
In this communication, we describe a mild, selective, and
efficient aziridination protocol that involves catalysis by a
mixed-valent dirhodium(II,III) catalyst (Rh25+).
We have recently reported that dirhodium(II,II) caprolac-
tamate [Rh2(cap)4] performs admirably as a catalyst for allylic
oxidation.7 Its effectiveness is derived from its ability to
(4) Selected recent synthesis of aziridines via nitrenes: (a) Dauban, P.;
Sanie´re, L.; Tarrade, A.; Dodd, R. H. J. Am. Chem. Soc. 2001, 123, 7707.
(b) Duran, F.; Leman, L.; Ghini, A.; Burton, G.; Dauban, P.; Dodd, R. H.
Org. Lett. 2002, 4, 2481. (c) Siu, T.; Yudin, A. K. J. Am. Chem. Soc. 2002,
124, 530.
(5) Selected enantioselective aziridinations, see: (a) Li, Z.; Conser, K.
R.; Jacobsen, E. N. J. Am. Chem. Soc. 1993, 115, 5326. (b) Evans, D. A.;
Bilodeau, M. T.; Faul, M. M. J. Am. Chem. Soc. 1994, 116, 2742. (c)
Sanders, C. J.; Gillespie, K. M.; Bell, D.; Scott, P. J. Am. Chem. Soc. 2000,
122, 7132. (d) Liang, J.-L.; Huang, J.-S.; Yu, X.-Q.; Zhu, N.; Che, C.-M.
Chem. Eur. J. 2002, 8, 1563.
(1) For a general review of carbon-nitrogen bond formation, see: (a)
Kemp, J. E. G. In ComprehensiVe Organic Synthesis; Trost, B. M., Ed.;
Pergamon: Oxford, UK, 1991; Vol. 7, p 469.
(2) Selected olefin aziridination reviews, see: (a) Mu¨ller, P.; Fruit, C.
Chem. ReV. 2003, 103, 2905. (b) Dauban, P.; Dodd, R. H. Synlett 2003,
1571. (c) Jacobsen, E. N. In ComprehensiVe Asymmetric Catalysis; Jacobsen,
E. N., Phaltz, A., Yamamoto, H., Eds.; Springer-Verlag: Berlin, 1999; Vol.
2, p 607. (d) Mu¨ller, P. In AdVances in Catalytic Processes; Doyle, M. P.,
Ed; JAI Press, Inc.: Greenwich, 1997; Vol. 2, p 113.
(6) For Rh24+-catalyzed aziridination, see: (a) Mu¨ller, P.; Baud, C.;
Jacquier, Y. Tetrahedron 1996, 52, 1543. (b) Mu¨ller, P.; Baud, C.; Jacquier,
Y. Can. J. Chem. 1998, 76, 738. (c) Guthikonda, K.; DuBois, J. J. Am.
Chem. Soc. 2002, 124, 13672. (d) Liang, J.-L.; Yuan, S.-X.; Chan, P. W.
H.; Che, C.-M. Org. Lett. 2002, 4, 4507. (e) Liang, J. L.; Yuan, S. X.;
Chan, P. W. H.; Che, C. M. Tetrahedron Lett. 2003, 44, 5917. (f) Fruit, C.;
Mu¨ller, P. Tetrahedron: Asymmetry 2004, 15, 1019.
(3) For a comprehensive review of aziridine ring opening, see: (a) Hu,
X. E. Tetrahedron 2004, 60, 2701.
(7) Catino, A. J.; Forslund, R. E.; Doyle, M. P. J. Am. Chem. Soc. 2004,
126, 13622.
10.1021/ol0510973 CCC: $30.25
© 2005 American Chemical Society
Published on Web 05/26/2005

5+
undergo facile atom-transfer redox chemistry (Rh24+/Rh2
)
tages over nitrene delivery; however, the catalytic efficiency
of phenyltrimethylammonium tribromide and the formation
of 1,2-dibromide byproducts were noted limitations.
because of its low one-electron oxidation potential.8 With
similar considerations, we examined the potential of Rh2-
(cap)4 as a bromine atom-transfer redox catalyst.
Initial studies showed that Rh2(cap)4 undergoes a one-
electron oxidation in the presence of N-bromosuccinimide
(NBS) to yield paramagnetic complex 1 (Scheme 1).
Efforts using Chloramine-T as a nitrogen source with 1
as a catalyst yielded only trace amounts of aziridine due to
catalyst decomposition under the reaction conditions. We
then considered the feasibility of a less basic amine derivative
to mitigate catalyst destruction. Toward this end, treating
4-methylstyrene (1.0 equiv) in CH2Cl2 (0.27 M/olefin) with
p-toluenesulfonamide (1.1 equiv) and 0.1 mol % Rh2(cap)4
followed by NBS (1.1 equiv) rapidly gave â-bromosulfon-
amide 3 in 95% isolated yield (Scheme 3).13 This result was
Scheme 1
Scheme 3
Evidence includes an oxidative color change (light blue f
deep red) in CH2Cl2: the UV/visible spectrum of the rhodium
complex upon addition of NBS contains a low-energy
absorption (δ-δ* transition) at 971 nm (ꢀ ) 930 M-1 cm-1)
indicating a Rh2 species.9
5+
complementary to a study by Sudalai and co-workers who
observed bromoamidation of olefins with TsNH2 and NBS
using 5 mol % of various Lewis acids.14
We were encouraged by the comparative efficiency of Rh2-
(cap)4 for bromoamidation and, looking to convert the
product directly to aziridines, conducted the same reaction
in the presence of K2CO3 (2.1 equiv). Aziridine 4 was
produced in 88% isolated yield after 12 h (Scheme 4).
Numerous attempts to obtain crystals of 1 were unsuc-
cessful. However, by replacing NBS with N-chlorosuccin-
imide (NCS), suitable crystals were obtained for X-ray
analysis,10 revealing the dirhodium complex 2 containing an
axially bound chlorine (Scheme 2). The spectral properties
Scheme 2
Scheme 4
of 2 are consistent with a dirhodium(II,III) complex and thus
provide indirect support for 1.11
To determine if Rh2(cap)4 and NBS could be synthetically
useful, we considered bromine-catalyzed aziridination as first
advanced by Sharpless.12 His protocol offers unique advan-
Further, reducing the amount of catalyst to only 0.01 mol %
Rh2(cap)4 (substrate:catalyst ) 10 000) gave 4 in 73% yield
in 12 h.
This operationally straightforward reaction was readily
extended to a variety of olefins (Table 1).15,16 Aryl- and alkyl-
substituted alkenes underwent inter- and intramolecular
+
(8) Rh2(cap)4 f Rh2(cap)4 < 1 kcal/mol: Doyle, M. P.; Ren, T.
Progress in Inorganic Chemistry; Karlin, K., Ed; Wiley: New York, 2001;
Vol. 49, p 113.
(9) (a) Cotton, F. A.; Walton, R. A. Multiple Bonds Between Metal Atoms;
Wiley: New York, 1982; p 390. (b) Cotton, F. A.; Walton, R. A. Multiple
Bonds Between Metal Atoms, 2nd ed.; Oxford: New York, 1993; p 475.
(10) See Supporting Information for X-ray crystal data.
(13) By 1H NMR, 70% conversion (from 4-methylstyrene into 3) was
observed in only 3 min at 1 mol % Rh2(cap)4.
(14) Thakur, V. V.; Talluri, S. K.; Sudalai, A. Org. Lett. 2003, 5, 861.
(15) Electron-deficient (methyl-trans-cinnamate), trisubstituted (1-me-
thylcyclohexene), and R,R-disubstituted (R-methylstyrene) olefins were not
reactive substrates for this protocol.
(16) Representative Procedure. A 25 mL flask equipped with a stir
bar was charged with olefin (2.72 mmol, 100 mol %), CH2Cl2 (10 mL),
TsNH2 (2.99 mmol, 110 mol %), K2CO3 (5.71 mmol, 210 mol %), and
Rh2(cap)4 (0.0027 mmol, 0.1 mol %). To the mixture was added NBS (2.99
mmol, 110 mol %) in one portion, and the color of the solution immediately
turned from light blue to red. The flask was sealed with a septum allowing
inclusion of air. After 12 h, silica gel was added to the reaction mixture
and the solvent was evaporated. Column chromatography yielded the
analytically pure compound.
(11) For Rh25+ halide complexes containing N-C-N dinuclear bridging
ligands, see: (a) Kadish, K. M.; Phan, T. D.; Giribabu, L.; Van Caemel-
becke, E.; Bear, J. L. Inorg. Chem. 2003, 42, 8663. (b) Bear, J. L.; Yao, C.
L.; Liu, L. M.; Capdevielle, F. J.; Korp, J. D.; Albright, T. A.; Kang, S. K.;
Kadish, K. M. Inorg. Chem. 1989, 28, 1254.
(12) For bromine-catalyzed aziridination, see: Jeong, J. U.; Tao, B.;
Sagasser, I.; Henniges, H.; Sharpless, K. B. J. Am. Chem. Soc. 1998, 120,
6844. For subsequent studies by other groups, see: (a) Ali, S. I.; Nikalje,
M. D.; Sudalai, A. Org. Lett. 1999, 1, 705. (b) Dauban, P.; Dodd, R. H.
Tetrahedron Lett. 2001, 42, 1037.(c) Thakur, V. V.; Sudalai, A. Tetrahedron
Lett. 2003, 44, 989. (d) Jain, S. L.; Sharma, V. B.; Sain, B. Tetrahedron
Lett. 2004, 45, 8731.
2788
Org. Lett., Vol. 7, No. 13, 2005

formed concomitant with the disappearance of both NBS and
1
5 by H NMR analysis (eq 2). Because of the low pKa of
Table 1. Rh2(cap)4-Catalyzed Aziridination of Olefins
5,18 deprotonation shifts the equilibrium toward 6. Isolation
of the precipitate and subsequent 1H NMR analysis in
d-DMSO indicated that the precipitate was indeed 6.
We next considered the role of the dirhodium catalyst in
the reaction. The observed regioselectivity of 3 is consistent
with an ionic addition mechanism (i.e., a bromonium ion
intermediate).19 Evidence against the intermediacy of a
nitrene under the conditions described in Scheme 4 and Table
1 was provided by the failure of 7 to undergo C-H insertion
under the reaction conditions (eq 3).
A bromonium ion intermediate was further implicated by
the use of a radical (and cation) probe 920 that gave only
ring-opened product 10 under the reaction conditions with
and without K2CO3 (eq 4).
These experiments suggest that an ionic mechanism is
operative, as opposed to a nitrene process.21 Further, that a
mixed-valent dirhodium(II,III) complex such as 1 is a Lewis
acid akin to dirhodium(II,II) carboxamidates22 is suggested
by reaction inhibition in Lewis base solvents such as
acetonitrile and THF. Moreover, dirhodium(II,III) methanol
complex 2 is capable of catalyzing the hetero-Diels-Alder
(HDA) reaction of p-nitrobenzaldehyde and 1-methoxy-3-
[(trimethylsilyl)oxy]-butadiene (Danishefsky diene). There-
fore, we propose that 1 activates residual amounts of 5 and/
or NBS, catalyzing electrophilic bromonium ion transfer to
an olefin to yield 11 (Scheme 6). Capture with TsNH2 or 6,
gives bromoamide 12,23 which can undergo ring closure to
give the aziridine.
a Isolated yield after purification. b Under these reaction conditions,
aziridine diastereoselectivity was determined by 1H NMR prior to silica
purification (entry 2 (trans/cis ) 4:1), entry 3 (cis/trans ) 7:1)). c Using 5
equiv of olefin, yield based on p-TsNH2.
aziridination in high yield under these mild conditions. Trans
aminobromination occurred exclusively for cycloalkenes
prior to aziridine formation, and C-H insertion products
were not observed for aliphatic olefins (e.g., entry 8).
A mechanistic proposal for aziridination is presented in
Scheme 5. From analysis of a stoichiometric mixture of NBS
(17) A small amount of N,N-dibromo-p-toluenesulfonamide (TsNBr2) was
observed after 24 h.
(18) TsNHCl pKa ) 4.55: Morris, J. C.; Salazar, A. S.; Wineman, M.
A. J. Am. Chem. Soc. 1948, 70, 2036. Rangappa, K. S. J. Phys. Org. Chem.
2001, 14, 684.
Scheme 5
(19) Hassner, A.; Boerwinkle, F. J. Am. Chem. Soc. 1968, 90, 216.
(20) Newcomb, M.; Johnson, C. C.; Manek, M. B.; Varick, T. R. J. Am.
Chem. Soc. 1992, 114, 10915.
(21) Bromine atom-transfer via a radical process is not operative, as no
bromine addition products were observed when cyclohexene or p-methyl-
styrene was treated with NBS and Rh2(cap)4 in CH2Cl2.
(22) For dirhodium carboxamidates as Lewis acids, see: (a) Doyle, M.
P.; Phillips, I. M.; Hu, W. H. J. Am. Chem. Soc. 2001, 123, 5366. (b) Anada,
M.; Washio, T.; Shimada, N.; Kitagaki, S.; Nakajima, M.; Shiro, M.;
Hashimoto, S. Angew. Chem., Int. Ed. 2004, 43, 2665. (c) Doyle, M. P.;
Valenzuela, M.; Huang, P. L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5391.
(d) Valenzuela, M.; Doyle, M. P.; Hedberg, C.; Hu, W.; Holmstrom, A.
Synlett 2004, 13, 2422. (e) Forslund, R. E.; Cain, J.; Colyer, J.; Doyle, M.
P. AdV. Synth. Catal. 2005, 347, 87.
(23) 12 (R ) 4-methylphenyl, X ) Br) was independently synthesized
by treatment of 4-methylstyrene with N,N-dibromo-p-toluenesulfonamide
in CH2Cl2. It was found that 12 acts as an electrophilic source of bromine,
as treatment with succinimide yields NBS and aminobromide 3 in an
equilibrium mixture (see Supporting Information).
and TsNH2 in solution, we concluded that an equilibrium
mixture of N-bromo-p-toluensulfonamide (TsNHBr, 5) and
succinimide (eq 1) existed. Complete conversion of 5 to NBS
was observed by addition of excess succinimide, thereby
confirming an equilibrium process. Moreover, addition of
Rh2(cap)4 did not change the equilibrium position.17
When excess K2CO3 was added to the equilibrium mixture
(NBS, TsNHBr/TsNH2, and succinimide), a precipitate was
Org. Lett., Vol. 7, No. 13, 2005
2789

Scheme 6
Scheme 7
styrene were unreactive). Whether or not there is an extended
role for 1 beyond its capacity as a Lewis acid, however,
remains uncertain at this time and is currently under
investigation.
In summary, we have developed a catalytic olefin aziri-
dination protocol using a multivalent dirhodium catalyst. A
selection of olefins has been converted to aziridines in
moderate to high yields under extremely mild conditions with
as little as 0.01 mol % catalyst. A mechanism has been
advanced that suggests that dirhodium(II,III) caprolactamate
operates as a Lewis acid catalyst and is capable of generating
other potentially useful intermediates. Efforts are underway
to develop new applications as well as to probe the nature
of catalytic intermediates in mixed-valent dirhodium(II,III)
catalysis.
The metal-based Lewis-acid catalysts for aminobromina-
tion reported by Sudalai gave a moderate enhancement in
yield for aziridination over a measurable background reac-
tion.24 Presumably, this is due to the incompatibility of these
Lewis acid catalysts and potassium carbonate under the
reaction conditions. With the dirhodium(II,II) carboxylates,
which do not undergo one-electron oxidation under the
reaction conditions, moderate yields of aziridination products
were obtained with very low catalyst loadings. Of all the
catalysts examined, Rh2(cap)4 was the most effective.
In addition to the proposed Lewis acid activation, we
examined another mechanistic scenario that may be operative
in light of the observed substrate reactivity.15 Due to the
known nucleophilicity of Chloramine-T analogues such as
6, as well as the “through dirhodium” displacement of a
halide,25 we envisioned the possible formation of 13 (Scheme
7). Displacement of bromine from 1 by 6 would give 13 as
a metal-bound bromine atom-transfer source. Silver(I) titra-
tion experiments identified the presence of chloride ion when
1 was treated with Chloramine-T.26 In addition, the stoichio-
metric reaction of 1 with Chloramine-T (2 equiv) and styrene
in the absence of potassium carbonate rapidly gave both
aminobromination and aziridination products, consistent with
the intermediate formation of 13. Furthermore, bromine
atom-transfer from 13 would be a sterically demanding
process (as both 1-methylcyclohexene and R-methyl-
Acknowledgment. This work was supported by the
National Science Foundation and the National Institutes of
Health (GM 46503). We thank Dr. James Fettinger for X-ray
determination and Marcela Valenzuela for catalysis evidence
for 2 in the HDA reaction.
Supporting Information Available: X-ray crystallo-
graphic data and NMR analysis of reaction intermediates
(CIF). This material is available free of charge via the Internet
at http://pubs.acs.org.
OL0510973
(26) Silver(I) halide test was performed to detect Br- as the displacement
product of 1 and 6. Treatment of 1 with Ag(I)BF4 did not yield a precipitate.
Treatment of NBS with Chloramine-T in the absence of 1 with Ag(I)BF4
did not yield a precipitate. However, in the presence of Chloramine-T, 1
immediately reacted with Ag(I)BF4 to give a white precipitate determined
to be AgCl by its solubility in ammonium hydroxide.
(24) Without catalyst, arizidine 3 was obtained in 19% yield. Under the
same conditions, other transition metals were examined, giving 3 in moderate
yield: CuI (5 mol %, 55%), Mn(II)-salen (5 mol %, 41%), Rh2(OAc)4 (1
mol %, 42%), Rh2(pfb)4 (1 mol %, 49%).
(25) For displacement reactions of Rh2-halides, see: Bear, J. L.; Han,
B.; Wu, Z.; Van Caemelbecke, E.; Kadish, K. M. Inorg. Chem. 2001, 40,
2275.
2790
Org. Lett., Vol. 7, No. 13, 2005