Ruthenium-catalyzed cycloaddition of nitrile oxides and alkynes: Practical synthesis of isoxazoles

Article abstract of DOI:10.1002/anie.200801920

(Chemical Equation Presented) 3,4-Disubstituted and 3,4,5-trisubstituted isoxazoles have been formed from alkynes and nitrile oxides in a ruthenium(II)-catalyzed process (see scheme; cod=cycloocta-l,5-diene, Cp=C₅Me₅). These reaction

Full text of DOI:10.1002/anie.200801920

Angewandte
Chemie
DOI: 10.1002/anie.200801920
Ruthenium Catalysis
Ruthenium-Catalyzed Cycloaddition of Nitrile Oxides and Alkynes:
Practical Synthesis of Isoxazoles**
Scott Grecian and Valery V. Fokin*
Isoxazoles, a major class of five-membered nitrogen hetero-
cycles, are embedded in a number of pharmaceutically
important compounds that have been the focus of numerous
biological studies during the last several years.[1] Although a
variety of methods for their synthesis have been reported, few
are general, regioselective, and high yielding.[2] The cyclo-
addition of alkynes and nitrile oxides, which is probably the
most direct route to access these heterocycles, is rarely used.[3]
The reasons for this are simple: in contrast to the reaction
with olefins, the uncatalyzed, thermal cycloaddition reactions
of nitrile oxides with alkynes are neither chemo- nor
regioselective and, as a consequence, are plagued by low
Scheme 1. Cycloaddition reactions of nitrile oxides and alkynes.
EWG=electron-withdrawing group.
yields and the formation of multiple products. These short-
comings are not surprising considering the relatively high
reactivity of nitrile oxides, their propensity to dimerize, and
the general inert character of alkynes.
Initially, we screened several ruthenium(II) complexes
and common solvents for their catalytic activity (Table 1).
When phenylacetylene and 4-chloro-N-hydroxybenzimidoyl
chloride were combined with Et3N in the absence of a
ruthenium catalyst at room temperature, the 3,5-disubstituted
regioisomer 1b was formed exclusively, as determined by GC
analysis of the crude reaction mixture (Table 1, entry 1). In
contrast, when [Cp*RuCl(cod)] (cod = cycloocta-l,5-diene,
Cp* = C5Me5) was used, 1a was formed preferentially, but
incomplete conversion was observed in DMF, THF, and
CHCl3 (Table 1, entries 2–4) With 5 mol% of [Cp*RuCl-
(cod)] in 1,2-dichloroethane (1,2-DCE), the starting materials
were converted into the 3,4-disubstituted isoxazole 1a
(95:3 1a/1b; Table 1, entry 5). Other Ru complexes, lower
Copper(I) acetylides have been shown to react regiose-
lectively with nitrile oxides to generate 3,5-disubstituted
isoxazoles.[4] However, there are no reported methods for
generating the regiocomplementary 3,4-disubstituted isomers.
In fact, even when thermal cycloaddition reactions of nitrile
oxides with alkynes are successful, they favor the formation of
the 3,5-disubstituted isomer (Scheme 1). Furthermore, exam-
ples of reactions of nitrile oxides with internal alkynes are
limited to a handful of highly activated alkynes (e.g. acetylene
dicarboxylate and related electron-deficient acetylenes).
Unactivated, electron-rich, or sterically hindered acetylenes
usually fail to react altogether.[5]
The recent discovery of the ruthenium(II)-catalyzed azide–
alkyne cycloaddition reaction,[6] which produces 1,5-disubsti-
tuted and 1,4,5-trisubstituted 1,2,3-triazoles, prompted us to
explore the catalytic activity of ruthenium complexes in nitrile
oxide–alkyne cycloaddition reactions. Herein, we report that
3,5-di- and 3,4,5-trisubstituted isoxazoles can be obtained with
excellent regioselectivity at room temperature by a rutheniu-
m(II)-catalyzed cycloaddition reaction of nitrile oxides (gen-
erated in situ from hydroximoyl chlorides by treatment with
Et3N) and terminal or internal alkynes, respectively.
Table 1: Optimization of ruthenium(II)-catalyzed reactions of nitrile
oxides and alkynes.
Entry
Reaction conditions[a]
Conversion [%][b]
1a
1b
1
2
3
4
5
6
7
8
9
10
no catalyst, 1,2-DCE
0
82
81
78
95
80
97
53
69
1
72
5
3
6
3
6
3
30
17
74
5 mol% [Cp*RuCl(cod)], DMF
5 mol% [Cp*RuCl(cod)], THF
5 mol% [Cp*RuCl(cod)], CHCl3
5 mol% [Cp*RuCl(cod)], 1,2-DCE
5 mol% [Cp*RuCl(cod)], 1,2-DCE, 08C
10 mol% [Cp*RuCl(cod)], 1,2-DCE
2 mol% [Cp*RuCl(cod)], 1,2-DCE
5 mol% [Cp*RuCl(PPh3)2], 1,2-DCE
5 mol% [CpRuCl(PPh3)2], 1,2-DCE
[*] Dr. S. Grecian, Prof. Dr. V. V. Fokin
Department of Chemistry
The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax: (+1)858-784-7562
E-mail: fokin@scripps.edu
Homepage: http://www.scripps.edu/chem/fokin
[**] Financial support from the Skaggs Institute for Chemical Biology
and Pfizer, Inc. is gratefully acknowledged. We thank Prof. K. B.
Sharpless for stimulating discussions and advice.
[a] Reaction conditions: phenylacetylene (1.0 mmol), hydroximoyl chlo-
ride (1.1 mmol), Et3N (1.25 mmol) in solvent (5 mL) at room temper-
ature for 10 h. [b] Determined by GC analysis.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200801920.
Angew. Chem. Int. Ed. 2008, 47, 8285 –8287
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8285

Communications
catalyst loadings, and different temperature profiles gave
Importantly, these ruthenium-mediated processes are not
restricted to terminal alkynes. In contrast to the copper-
catalyzed isoxazole synthesis, where copper(I) acetylides are
bona fide intermediates, internal alkynes were effective
cycloaddition partners as well. The major regioisomers
obtained when [Cp*RuCl(cod)] was used with internal
alkynes were opposite to those observed in the analogous
thermal reactions, as was the case with the terminal alkynes
(Table 3).[7] Notably, when alkynes containing a hydrogen-
bond donor were employed, the cycloaddition reactions were
especially regioselective and efficient (Table 3, entries 4–6).
In fact, the minor regioisomer was not detectable by GC-MS
or LC-MS analysis. X-ray crystallographic analysis of 15a
unambiguously confirmed its regiochemistry. The corre-
sponding thermal cycloaddition reactions, even after
extended heating with such alkynes, failed to give any
isoxazole products. The regioselectivity of these processes is
predictable: the more electronegative carbon center of the
alkyne becomes C4 of the isoxazole ring unless a hydrogen-
bond donor is present, in which cases the hydrogen-bond
donor always ends up at C4.
lower conversion and regioselectivity (Table 1, entries 6–10).
To determine the scope of this new reaction, a series of
hydroximoyl chlorides were combined with various terminal
alkynes in the presence of 5 mol% of [Cp*RuCl(cod)] and
1.25 equivalents of Et3N and stirred for 2–10 h at room
temperature (Table 2). Typically, complete consumption of
the alkyne was observed after less than 2 h, very little by-
products were formed, and usually only the 3,4-disubstituted
1
isoxazole was observed by GC and H NMR analysis. These
reactions were relatively insensitive to the electronic and
steric properties of the nitrile oxide. Terminal alkynes bearing
a wide variety of functional groups (Table 2), as well as
sterically encumbered substrates (Table 2, entries 3 and 6),
were competent reactants. When performed on a 20 mmol
scale with 2 mol% [Cp*RuCl(cod)] at 08C, regioisomerically
pure isoxazoles were obtained in high yields (Table 2,
entry 9).
Table 2: Ruthenium-catalyzed reactions of nitrile oxides with terminal
alkynes.
Currently we do not have complete understanding of the
mechanism of this new isoxazole synthesis. We hypothesize
Entry
1
Product
Compound
Yield [%]
86
Table 3: Isoxazoles produced from internal alkynes.
1a
Entry Products
Yield [%]
a
b
2
3
2a
3a
87[a]
1
71
5
2
89 (6.6:1 a/b)
67
3
4
5
68
16
4
5
6
4a
5a
6a
77
87
93
78
83
99
39
–
–
–
7
8
9
7a
8a
9a
72
6
7
85
93[b]
[a] Isolated as a 15:1 regioisomeric mixture. [b] Prepared on 20 mmol
scale (alkyne) with 2 mol% [Cp*RuCl(cod)] (see the Supporting
Information for details). Ts=4-toluenesulfonyl.
8286 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 8285 –8287

Angewandte
Chemie
that like the ruthenium-catalyzed azide–alkyne cycloaddition,
these processes may be related to the catalytic cyclotrimeri-
zation of alkynes.[8] However, there are several significant
differences between the two reactions, which stem from the
that different dipoles can be activated and engaged in
catalysis by ruthenium complexes.
intrinsically higher reactivity of nitrile oxides and, as a Experimental Section
Ruthenium(II)-catalyzed synthesis of isoxazoles as exemplified for
consequence, a tendency to give by-products. In the first
step of the proposed mechanism (Scheme 2), displacement of
the spectator cyclooctadiene ligand from the [Cp*RuCl(cod)]
catalyst by an alkyne and nitrile oxide produces the activated
complex A, which mediates the oxidative coupling of a nitrile
oxide and alkyne, resulting in ruthenacycle B.[9]
the preparation of 3-(4-chlorophenyl)-4-phenylisoxazole (1a). A
screw top vial (20 mL) purged with dry nitrogen was charged with
4-chloro-N-hydroxybenzimidoyl chloride (208 mg, 1.1 mmol) and
phenylacetylene (110 mL, 1.0 mmol). At room temperature, degassed
1,2-dichloroethane (10 mL) was added followed by [Cp*RuCl(cod)]
(19 mg, 0.05 mmol) and triethylamine (176 mL, 1.25 mmol) and the
vial was capped. After 10 h, the reaction mixture was passed through
a plug of silica gel (CH2Cl2). The resulting solution was concentrated
and the residue was purified by column chromatography on silica gel
(pure hexanes to 10:1 hexanes/EtOAc) to provide the indicated
compound as a yellow oil (219 mg, 86%). Rf = 0.41 (10:1 hexanes/
The oxidative coupling step controls the regioselectivity of
the overall process. It appears that the new carbon–oxygen
bond is formed between the more electronegative carbon
center of the alkyne and the oxygen atom of the nitrile oxide,
which represents an unexpected mode of activation of nitrile
oxides—normally, their carbon center is electrophilic and
readily reacts with nucleophiles.[10] Thus, coordination to the
ruthenium atom effectively changes the polarity of the nitrile
oxide. Ruthenacycle B undergoes reductive elimination
giving C, and release of the isoxazole product, then completes
the catalytic cycle. The observation that Cp*-based catalysts
are especially catalytically active is consistent with the lability
of the spectator ligands in such ruthenium complexes.[11]
Notably, suppression of common side reactions, such as
dimerization of nitrile oxides to form furoxans, and the
generally high reactivity of nitrile oxides observed in the
ruthenium(II)-catalyzed cycloaddition indicate that the cata-
lytic cycle turns over at least as fast as the nitrile oxide is formed
from the hydroximoyl chloride precursor. Detailed kinetic
studies of this novel ruthenium(II)-catalyzed cycloaddition
reaction are currently underway and should provide further
insight into the mechanistic underpinnings of this process.
In summary, 3,4- and 3,4,5-subtituted isoxazoles are now
readily accessible from alkynes and hydroximoyl chlorides by
an experimentally simple and general catalytic method.
Together with the copper(I)-catalyzed process,[4] the ruthe-
nium(II)-catalyzed synthesis allows regioselective and effi-
cient preparation of all isomers of isoxazoles. In addition to
the immediate practical benefits, this transformation suggests
1
EtOAc); IR (neat): n˜ = 3060, 1601, 1094 cmÀ1; H NMR (400 MHz,
CDCl3): d = 8.52 (s, 1H), 7.47–7.45 (m, 3H), 7.38–7.34 (m, 4H), 7.27–
7.24 ppm (m, 2H); 13C NMR (100 MHz, CDCl3): d = 159.4, 156.72,
156.70, 136.0, 130.2, 129.2, 129.12, 129.09, 128.9, 127.3, 120.5 ppm;
LCMS (ES): m/z 256 [M++H]; HRMS calcd for C15H11ClNO
[M++H]: 256.0529; found: 256.0525.
Received: April 24, 2008
Published online: September 22, 2008
Keywords: cycloaddition · homogeneous catalysis · isoxazoles ·
.
nitrile oxides · ruthenium
[1] J. Sperry, D. Wright, B, Curr. Opin. Drug Discov. Devel. 2005, 8,
723.
[2] a) B. J. Wakefield in Science of Synthesis: Houben-Weyl Methods
of Molecular Transformations, Vol. 11 (Ed.: E. Schaumann),
Georg Thieme, Stuttgart, 2001, pp. 229 – 288; b) T. M. V. D.
Pinho e Melo, Curr. Org. Chem. 2005, 9, 925.
[3] V. Jꢀger, P. A. Colinas in Synthetic Applications of 1,3-Dipolar
Cycloaddition Chemistry Toward Heterocycles and Natural
Products, Vol. 59 (Ed.: A. Padwa), Wiley, Hoboken, 2002,
pp. 361 – 472.
[4] a) F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtsev, L. Noodle-
man, K. B. Sharpless, V. V. Fokin, J. Am. Chem. Soc. 2005, 127,
210; b) T. V. Hansen, P. Wu, V. V. Fokin, J. Org. Chem. 2005, 70,
7761.
[5] F. A. Fouli, M. M. Habashy, A. F. El-Kafrawy, A. S. A. Youseef,
M. M. El-Adly, J. Prakt. Chem. 1987, 329, 1116.
[6] a) L. Zhang, X. Chen, P. Xue, H. H. Y. Sun, I. D. Williams, K. B.
Sharpless, V. V. Fokin, G. Jia, J. Am. Chem. Soc. 2005, 127,
15998; b) L. K. Rasmussen, B. C. Boren, V. Fokin, Org. Lett.
2007, 9, 5337; c) B. C. Boren, S. Narayan, L. K. Rasmussen, L.
Zhang, H. Zhao, Z. Lin, G. Jia, V. V. Fokin, J. Am. Chem. Soc.
2008, 130, 8923.
[7] The corresponding thermal reactions were performed for each of
the examples shown.
[8] a) T. Naota, H. Takaya, S.-I. Murahashi, Chem. Rev. 1998, 98,
2599; b) Y. Yamamoto, T. Arakawa, R. Ogawa, K. Itoh, J. Am.
Chem. Soc. 2003, 125, 12143; c) K. Kirchner, M. J. Calhorda, R.
Schmid, L. F. Veiros, J. Am. Chem. Soc. 2003, 125, 11721.
[9] a) C. Ernst, O. Walter, E. Dinjus, S. Arzberger, H. Gorls, J. Prakt.
Chem. 1999, 341, 801; b) C. Ernst, O. Walter, E. Dinjus, J.
Organomet. Chem. 2001, 627, 249.
[10] K. B. G. Torssell, Nitrile Oxides, Nitrones, and Nitronates in
Organic Synthesis, VCH, Weinheim, 1988.
[11] Y. Yamamoto, K. Kinpara, T. Saigoku, H. Takagishi, S. Okuda,
H. Nishiyama, K. Itoh, J. Am. Chem. Soc. 2005, 127, 605.
Scheme 2. Proposed catalytic cycle.
Angew. Chem. Int. Ed. 2008, 47, 8285 –8287
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8287