A catalytic asymmetric chlorocyclization of unsaturated amides

Article abstract of DOI:10.1002/anie.201006910

The asymmetric chlorocyclization of unsaturated amides catalyzed by (DHQD)₂PHAL yields oxazoline and dihydrooxazine derivatives (see scheme). The reaction is operationally simple and employs 1-2 mol % of the commercially available (DHQD)₂PHAL (hydroquinidine 1,4-phthalazinediyl diether) catalyst. Different substitution patterns of the olefin as well as aromatic and aliphatic olefin substituents are well tolerated. DCDPH=N,N-dichloro-5,5-diphenylhydantoin. Copyright

Full text of DOI:10.1002/anie.201006910

DOI: 10.1002/anie.201006910
Asymmetric Synthesis
A Catalytic Asymmetric Chlorocyclization of Unsaturated Amides**
Arvind Jaganathan, Atefeh Garzan, Daniel C. Whitehead, Richard J. Staples, and Babak Borhan*
Stereodefined carbon–halogen bonds are ubiquitous in nature
with several natural products exhibiting this motif.[1] While
the biogenetic origins of this unique chiral functionality has
been a subject of several investigations in the past,[2] attempts
by organic chemists to forge the carbon–halogen bond
stereoselectively have largely been unsuccessful. This prob-
lem has come into focus only recently. Several elegant reports
of asymmetric halogenations of alkenes and alkynes followed
by an intramolecular attack of a pendant nucleophile have
appeared in the literature in the last decade. Kang et al.
reported a cobalt–salen catalyzed iodoetherification reac-
tion.[3a] An asymmetric fluorocyclization of allyl silanes
mediated by a cinchona alkaloid dimer was reported by
Gouverneur and co-workers.[3b] Tang and co-workers dis-
closed an asymmetric bromolactonization of enynes catalyzed
by a cinchona alkaloid derived urea; other bromolactoniza-
tions have also appeared following the disclosure of their
report.[3c–e] More recently, Veitch and Jacobsen reported an
asymmetric iodolactonization reaction mediated by chiral
thiourea catalysts.[3f] Polyene cyclizations induced by chiral
halonium ions have also been realized as reported by the
research groups of Ishihara and Snyder.[4a–c] However, given
the fledgling nature of this research area, one may find it easy
to highlight the many drawbacks and limitations even in the
present state of the art—for example, the relatively large
catalyst loadings (superstoichiometric quantities in some
cases) to achieve meaningful levels of enantioselectivity or
the lack of a robust catalytic system that can catalyze a
number of diverse reactions rather than one specific reaction.
Moreover, efficient asymmetric chlorocyclizations have
remained underdeveloped. This situation is attributable, at
least in part, to the highly reactive nature of chloronium ions,
which are known to exist in equilibrium with the correspond-
ing carbocation rather than exclusively as cyclic chloronium
ions,[5a–c] thus making the development of chlorocyclizations a
formidable challenge.
amides to furnish chiral heterocycles. Furthermore, these
heterocycles have been transformed into useful chiral build-
ing blocks such as amino alcohols.
Chiral heterocycles such as oxazolines and dihydrooxa-
zines are commonly encountered motifs in natural pro-
ducts,[8a] molecules of pharmaceutical interest,[8b] and in
several chiral ligands.[8c] Their syntheses, however, usually
employ stoichiometric quantities of chiral amino alcohols.
With only one precedented method to access these molecules
in a catalytic asymmetric fashion,[6] we were intrigued by the
possibility of one-step access to these versatile chiral hetero-
cycles by a catalytic asymmetric halocyclization of easily
accessed unsaturated amides.
We chose the conversion of benzamide 1 into oxazoline 2
as our initial test reaction. Among the several ligands
screened for the test reaction, (DHQD)2PHAL emerged as
the best candidate, thus affording the desired oxazoline 2 in
57% ee (Table 1, entry 1) with DCDMH as the terminal
chlorine source.[9] Reactions with other chlorenium sources
Table 1: Halocyclization of unsaturated amides mediated by
(DHQD)2PHAL.[a,b]
Entry X+ source
Solvent, T [8C]
Cat. [mol%] ee [%][c]
Our research group has recently reported the catalytic
asymmetric chlorolactonization of alkenoic acids.[7] Herein,
we disclose the efficient halocyclization of unsaturated
1
2
DCDMH
TCCA
NCS
CHCl3, À20
10
10
10
10
57
18
67
43[e]
28
5
63
80
90
90
CHCl3, À20
3[d]
4
MeCN/CCl4, RT
ChloramineT·3H2O CHCl3, À20
5[d]
6[d]
7
NBS
MeCN/CCl4, À20 10
MeCN/CCl4, À20 10
[*] A. Jaganathan, A. Garzan, Dr. D. C. Whitehead, Dr. R. J. Staples,
Prof. B. Borhan
DBDMH
DCDPH
DCDPH
DCDPH
DCDPH
CHCl3, À40
10
Department of Chemistry
Michigan State University
East Lansing, MI 48824 (USA)
Fax: (+1)517-353-1793
E-mail: babak@chemistry.msu.edu
8[d]
9
MeCN/CCl4 À40 10
CF3CH2OH, À40 10
10
CF3CH2OH, À30
2
[a] All reactions were run on a 0.04 mmol scale. Complete conversion
was observed for all reactions within 8 h. [b] See Supporting Information
for the determination of absolute configuration. [c] Determined by HPLC
on a chiral stationary phase. [d] Solvent ratio was 1:1. [e] Conversion was
less than 20% as judged by NMR analysis of the crude product.
[**] We acknowledge the ACS PRF (No. 47272-AC) and the NSF
(No. CHE-0957462) for generous funding. We also thank Adam
Ryske for assistance in the synthesis of some of the substrates in
Table 3.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201006910.
DBDMH=N,N-dibromo-5,5-dimethylhydantoin,
succinimide, NCS=N-chlorosuccinimide.
NBS=N-bromo-
Angew. Chem. Int. Ed. 2011, 50, 2593 –2596
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2593

Communications
were marred with poor enantioselectivities and/or poor
conversions. For example, TCCA gave the desired oxazoline
with full conversion of the starting material into product
albeit in only 18% ee (Table 1, entry 2). Employing NCS or
Chloramine-T·3H2O resulted in very sluggish reactions (less
than 20% conversion after 8 h). Running the reaction at
elevated temperatures with NCS in a MeCN/CCl4 solvent
system gave the desired product in 67% ee (Table 1, entry 3),
while Chloramine-T·3H2O gave the desired product in a
modest 43% ee (Table 1, entry 4). Resorting to DCDPH[10] as
the terminal chlorine source led to a more promising 63% ee
(Table 1, entry 7) in CHCl3 and 80% ee in a MeCN/CCl4
solvent system. Analogous bromocyclizations were poorly
selective. NBS returned the corresponding product in 28% ee
(Table 1, entry 5), while the significantly more reactive
DBDMH led to only 5% ee (Table 1, entry 6).
An exhaustive solvent-screening process revealed that
trifluoroethanol (TFE) is the optimal solvent for the chloro-
cyclization, thus giving the desired product in 90% ee
(Table 1, entry 9). The effect of catalyst loading, concentra-
tion, and temperature was then studied. It was observed that
the stereoselectivity of this reaction was not diminished even
at a catalyst loading of 2 mol% (90% ee; Table 1, entry 10). It
also emerged that a concentration of 0.04m and a temperature
of À308C was optimal.
Scheme 1. Modulation of the substrate-catalyst interaction by variation
of amide end functionality. Yields refer to the product isolated after
column chromatography; the ee values were determined by HPLC
analysis.
All further attempts at tweaking the reaction parameters
to increase stereoselectivity (order of addition, slow addition,
catalyst aging, etc.) were infructuous. Nonetheless, the aryl
group at C2, which ultimately is revealed as the acid-labile
functionality of the oxazoline product, could be viewed as a
sacrificial entity (see Scheme 2). As such, optimization of the
reaction through alteration of the aryl group at C2 could
provide the opportunity for electronic and steric fine-tuning.
The results of optimization studies for the aryl substituent at
C2 are summarized in Scheme 1.
only a 55% ee. It merits mention that the yields for all these
reactions were excellent (79–97%) with no significant quan-
tities of side products. The 4-BrC6H4 substituent was ulti-
mately chosen as the optimal aryl group at C2 for further
studies.[11]
The precise nature of the substrate–catalyst interaction is
still under investigation, however, we believe that para sub-
stituents on the C2 aryl ring provide a steric (rather than
electronic) bias for better catalyst–substrate interaction.
Substituents at the para position of the aryl ring at C2
increased the ee value regardless of their electron-donating/-
withdrawing properties. For example, when the C2 substituent
was an unsubstituted phenyl ring, the product 4a was formed
in 90% ee. However, the 4-NO2C6H4 and the 4-OMeC6H4
substituted products, 4b and 4c, were both formed in a
slightly enhanced 93% ee. Likewise, comparison of 4d and 4i
is also indicative of the crucial role of the para substituent of
the aryl ring. The 3,5-dinitrophenyl substituent (86% ee) is
clearly inferior to the 3,5-dinitro-4-methylphenyl substituent
(98% ee), thus indicating that the methyl group at the para
position is essential for good stereoselectivity in the latter
case. The 4-BrC6H4 substituted product 4g was also formed
with an excellent 98% ee. The significantly more bulky tBu
group gave the corresponding product 4h with a lower
ee value (88%). Heterocyclic rings such as 2-pyridyl (4e,
79% ee) and 3-pyridyl (4 f, 92% ee) were also well tolerated
as the C2 substituent in this reaction. The sterically demand-
ing 2,4,6-triethylphenyl substituent significantly diminished
the stereoselectivity of the reaction to give the product 4j in
In order to probe the substrate scope for this reaction, a
series of 1,1-disubstituted olefins were subjected to the
optimized reaction conditions (Table 2). The strongly elec-
tron-withdrawing NO2 group at the meta position significantly
decreased the enantioselectivity of the reaction. The desired
product 6b was isolated in 75% yield and a modest 68% ee
(Table 2, entry 2). Interestingly, switching the NO2 group with
the electron-donating OMe group at the meta position
restored the stereoselectivity of the reaction to give the
product 6c in 93% ee (Table 2, entry 3). Halogenated aryl
rings were well tolerated (6d–6 f; Table 2, entries 4–6).
We were delighted to discover that the same reaction
conditions could be extended to trans-disubstituted and
trisubstituted olefin substrates, which yielded the correspond-
ing dihydro-4-H-1,3-oxazines (Table 3). These reactions were
inherently more stereoselective and required no steric or
electronic fine-tuning of the C2 substituent. Almost all of the
substituted phenyl rings evaluated as the C2 substituent gave
99% or better enantioselectivity when R1 was a Ph ring (these
results are not shown in Table 3; see the Supporting
Information for a short list). Having already determined
that the 4-BrC6H4 was the optimal amide end functionality for
the 1,1-disubstituted olefin substrates, we retained the same
functionality for these substrates as well. Electron-deficient
2594
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 2593 –2596

Table 2: Substrate scope: 1,1-disubstituted olefin substrates.[a,b]
group was better tolerated and returned the corresponding
product 8h in excellent yield (99%) and good enantioselec-
tivity (87% ee; Table 3, entry 8). Interestingly, the product 8i
with the 2-MeC6H4 as the C6 substituent but with the 4-
OMeC6H4 as the C2 substituent was formed with a slightly
enhanced ee value (91%; Table 3, entry 9). This result indi-
cates that the amide end functionality is an invaluable handle
for electronic and steric fine-tuning, thus allowing the
systematic optimization of the enantioselectivity for a given
substrate. Trisubstituted olefin substrates were also compat-
ible with this chemistry (8j and 8k; Table 3, entries 10 and
11).
Substrates with aliphatic olefin substituents gave poor
yields of isolated product when TFE was employed as the
solvent. Fortunately, the use of 1-nitropropane in the presence
of molecular sieves (4 ꢀ) alleviated this problem. The
cyclohexyl-substituted product 8l was isolated in 77% yield
and near complete enantioselectivity (> 99% ee; Table 3,
entry 12). Likewise, the conformationally more flexible n-
pentyl substituent was also well tolerated and returned the
corresponding product 8m in excellent yield (90%) and ee
(>99%; Table 3, entry 13). It is important to note that no
regioisomeric five-membered rings were detected under the
optimized reaction conditions for the aliphatic substrates.
Also, the absolute configuration of the products with aliphatic
substituents was the same as that for products with aryl
substituents (as verified by the X-ray crystal structure of
8l[13]). In fact, the face selectivity for the formation of the
choloronium ion intermediate with (DHQD)2PHAL is the
same for all the amide substrates examined thus far.
Substrates with Z olefins did not undergo halocyclization to
the desired products presumably as a result of the stereo-
electronic constraints that are well precedented in halolacto-
nization reactions of similar substrates.[14]
Entry
R
Product
Yield [%][a]
ee [%][b]
1
2
3
4
5
6
C6H5
6a
6b
6c
6d
6e
6 f
93
75
72
65
89
94
98
68
93
87
84
87
3-NO2C6H4
3-OMeC6H4
4-FC6H4
4-BrC6H4
4-ClC6H4
[a] Yields of products isolated after column chromatography. [b] Deter-
mined by HPLC on a chiral stationary phase.
Table 3: Substrate scope: trans-disubstituted and trisubstituted olefin
substrates.[a,b]
Entry R1
R2
Ar
Prod. Yield [%][a] ee [%][b]
1
2
3
4
5
6
7
8
C6H5
C6H5
H
H
H
H
H
H
H
H
H
4-BrC6H4
4-OMeC6H4 8b
8a
91
93 (81)
99
85
94
84
93
99
64
99
>99 (>99)
95
93
95
4-FC6H4
4-BrC6H4
4-CF3C6H4
4-OMeC6H4
4-MeC6H4
2-MeC6H4
2-MeC6H4
C6H5
C6H5
Cy
n-C5H11
CH2Cy
4-BrC6H4
4-BrC6H4
4-BrC6H4
4-BrC6H4
4-BrC6H4
4-BrC6H4
8c
8d
8e
8 f
8g
8h
20
60[d]
87
91
86
91
We have also been able to demonstrate that this reaction
can lend itself to preparative scale synthesis by synthesizing
the dihydro-4-H-1,3-oxazines 8b and 8m on a gram scale
(Table 3, entries 2 and 13). Both 8b and 8m were synthesized
in excellent enantioselectivities (> 99% ee) by employing just
1 mol% of the catalyst.
9
4-OMeC6H4 8i
10
11
12
13
14
C6H5 4-BrC6H4
Me 4-BrC6H4
H
H
H
8j
8k
8l
8m
8n
92
52[c]
77[c]
90 (77)[c]
80[c]
4-BrC6H4
4-BrC6H4
4-BrC6H4
>99
>99 (>99)
88
While the products shown in Scheme 1, Table 2, and
Table 3 are of interest in their own right, they are adorned
with several functional handles which can be manipulated and
converted into useful synthetic building blocks. Compounds
6a and 8b were efficiently transformed into their correspond-
ing 1,2- and 1,3- chiral amino alcohols 9 and 10, respectively,
by a simple acid hydrolysis of the cyclic imidate functionality
(Scheme 2). These reactions proceed with complete stereo-
chemical fidelity.
In summary, we have developed a highly stereoselective
chlorocyclization of unsaturated amides to chiral heterocycles
mediated by catalytic amounts (1–2 mol%) of the commer-
cially available (DHQD)2PHAL. The reaction is operation-
ally simple with no need to resort to strictly anhydrous or inert
reaction conditions. The reaction scope is fairly general with
regards to the substitution pattern of the olefin. Both aliphatic
and aromatic residues on the olefin are well tolerated. The
mechanistic underpinnings of this transformation are cur-
rently under investigation.
[a] Yields of product isolated after column chromatography. Yields and
ee values in parentheses refer to reactions carried out on a 1 g scale.
[b] Determined by HPLC on a chiral stationary phase. [c] Reaction was run
in 1-nitropropane in the presence of 300 wt% M.S. (4 ꢀ). [d] Substrate
was added over a period of 2 h. Cy=cyclohexyl.
aryl rings as R1 presented no problems and resulted in
excellent stereoselectivities (products 8a–8e; Table 3,
entries 1–5). The absolute configuration of 8a was verified
by the X-ray crystal structure (see the Supporting Informa-
tion) and was inferred by analogy with other substrates.
Electron-rich R1 substituents led to a significant erosion of the
stereoselectivity. For example, the 4-OMeC6H4 substituted
amide gave the corresponding product 8 f in only 20% ee and
84% yield (Table 3, entry 6).[12] The 4-MeC6H4 substituent
also led to only moderate levels of stereoinduction, thus
affording 8g in 60% ee (Table 3, entry 7). The 2-MeC6H4
Angew. Chem. Int. Ed. 2011, 50, 2593 –2596
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2595

Communications
Werness, I. A. Guzei, W. Tang, J. Am. Chem. Soc. 2010, 132,
3664. For a brief review of recent developments in this field, see
G. Chen, S. Ma, Angew. Chem. 2010, 122, 8484; Angew. Chem.
Int. Ed. 2010, 49, 8306. During the course of preparation of this
manuscript, two elegant reports of asymmetric bromolactoniza-
tions were reported: d) L. Zhou, C. K. Tan, X. Jiang, F. Chen, Y.-
Y. Yeung, J. Am. Chem. Soc. 2010, 132, 15474; e) K. Murai, T.
Matsushita, A. Nakamura, S. Fukushima, M. Shimura, H.
Fujioka, Angew. Chem. 2010, 122, 9360; Angew. Chem. Int. Ed.
2010, 49, 9174; for an organocatalytic iodolactonizatin, see:
f) G. E. Veitch, E. N. Jacobsen, Angew. Chem. 2010, 122, 7490;
Angew. Chem. Int. Ed. 2010, 49, 7332.
[4] For polyene cyclizations induced by chiral halonium ions, see:
a) A. Sakakura, A. Ukai, K. Ishihara, Nature 2007, 445, 900; b) S.
Snyder, D. Treitler, Angew. Chem. 2009, 121, 8039; Angew.
Chem. Int. Ed. 2009, 48, 7899; c) S. A. Snyder, D. S. Treitler, A. P.
Brucks, J. Am. Chem. Soc. 2010, 123, 14303.
[5] a) G. A. Olah, J. M. Bollinger, J. Brinich, J. Am. Chem. Soc.
1968, 90, 2587; b) G. A. Olah, J. M. Bollinger, Y. K. Mo, J. M.
Brinich, J. Am. Chem. Soc. 1972, 94, 1164; c) B. K. Ohta, R. E.
Hough, J. W. Schubert, Org. Lett. 2007, 9, 2317.
[6] A solitary example of a catalytic asymmetric synthesis of
dihydrooxazoles from olefins is known in literature: S. Minakata,
M. Nishimura, T. Takahashi, Y. Oderaotoshi, M. Komatsu,
Tetrahedron Lett. 2001, 42, 9019.
[7] D. C. Whitehead, R. Yousefi, A. Jaganathan, B. Borhan, J. Am.
Chem. Soc. 2010, 132, 3298.
Scheme 2. Transformation of dihydrooxazolines and dihydro-4H-1,3-
oxazines into aminoalcohols.
Experimental Section
DCDPH (35 mg, 0.11 mmol, 1.1 equiv) was suspended in trifluoroe-
thanol (2.2 mL) in a screw capped vial equiped with a stir bar. The
resulting suspension was cooled to À308C with an immersion cooler.
(DHQD)2PHAL (1.56 mg, 312 mL of a 5 mgmLÀ1 solution in TFE,
2 mol%) was then introduced into the reaction mixture. After stirring
vigorously for 10 min, the substrate (0.10 mmol, 1.0 equiv) was added
in a single portion. The vial was capped and the stirring was continued
at À308C until the reaction was complete (as evident by TLC). The
reaction was quenched by the addition of 10% aq. Na2SO3 (3 mL) and
diluted with CH2Cl2 (3 mL). The organics were seperated and the
aqueous layer was extracted with CH2Cl2 (3 ꢁ 3 mL). The combined
organic extracts were dried over Na2SO4 and concentrated in the
presence of a small quantity of silica gel. Pure products were isolated
by column chromatography on silica gel using EtOAc/hexanes (1:19)
as the eluent.
[8] a) B. S. Davidson, Chem. Rev. 1993, 93, 1771; b) S. Castellano, D.
Kuck, M. Sala, E. Novellino, F. Lyko, G. Sbardella, J. Med.
Chem. 2008, 51, 2321; c) For a review on metal-bisoxazoline
catalysis, see A. K. Ghosh, P. Mathivanan, J. Cappiello, Tetrahe-
dron: Asymmetry 1998, 9, 1.
[9] See the Supporting Information for determination of the
absolute configuration.
[10] We have previously reported an expedient synthesis of a variety
of chlorinated hydantoins including DCDPH: D. C. Whitehead,
R. J. Staples, B. Borhan, Tetrahedron Lett. 2009, 50, 656.
[11] Aliphatic C2 substituents gave only moderate levels of stereo-
induction. For example, when the C2 substituent was a CH3
group, the product was formed in only 54% ee.
[12] The erosion in stereoselectivity is likely attributable to the
nonselective background reaction: ca. 80% conversion to prod-
uct in the absence of the catalyst under otherwise identical
conditions.
[13] CCDC 799027, 799028, 799029 and 799030 contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[14] S. E. Denmark, M. T. Burk, Proc. Natl. Acad. Sci. USA 2010, 107,
20655, and references therein.
Received: November 3, 2010
Published online: February 17, 2011
Keywords: amides · enantioselectivity · halocyclization ·
.
organocatalysis · oxazolines
[1] G. W. Gribble, Acc. Chem. Res. 1998, 31, 141.
[2] a) L. C. Blasiak, C. L. Drennan, Acc. Chem. Res. 2009, 42, 147;
b) C. S. Neumann, D. G. Fujimori, C. T. Walsh, Chem. Biol. 2008,
15, 99; c) C. Wagner, M. El Omari, G. M. Kꢂnig, J. Nat. Prod.
2009, 72, 540.
[3] a) S. H. Kang, S. B. Lee, C. M. Park, J. Am. Chem. Soc. 2003, 125,
15748; b) S. Wilkinson, O. Lozano, M. Schuler, M. Pacheco, R.
Salmon, V. Gouverneur, Angew. Chem. 2009, 121, 7217; Angew.
Chem. Int. Ed. 2009, 48, 7083; c) W. Zhang, S. Zheng, N. Liu, J. B.
2596 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 2593 –2596