1,3-Dipolar cycloaddition of organic azides to alkynes by a dicopper-substituted silicotungstate

Article abstract of DOI:10.1021/ja806249n

The dicopper-substituted γ-Keggin silicotungstate TBA ₄[γ-H₂SiW₁₀O₃₆Cu ₂(μ-1,1-N₃)₂] (I, TBA = tetra-n- butylammonium) could act as an efficient precatalyst for the regioselective 1,3-dipolar cycloaddition of organic azides to alkynes. Various combinations of substrates (four azides and eight alkynes) were efficiently converted to the corresponding 1,2,3-triazole derivatives in excellent yields without any additives. The present system was applicable to a larger-scale cycloaddition of benzyl azide to phenylacetylene under solvent-free conditions (100 mmol scale) in which 21.5 g of the analytically pure corresponding triazole could be isolated. In this case, the turnover frequency and the turnover number reached up to 14 800 h^-1 and 91 500, respectively, and these values were the highest among those reported for the copper-mediated systems so far. In addition, I could be applied to the one-pot synthesis of 1-benzyl-4-phenyl-1H-1, 2,3-triazole from benzyl chloride, sodium azide, and phenylacetylene. The catalyst effect, kinetic, mechanistic, and computational studies show that the reduced dicopper core plays an important role in the present 1,3-dipolar cycloaddition.

Full text of DOI:10.1021/ja806249n

Published on Web 10/25/2008
1,3-Dipolar Cycloaddition of Organic Azides to Alkynes by a
Dicopper-Substituted Silicotungstate
Keigo Kamata, Yoshinao Nakagawa, Kazuya Yamaguchi, and Noritaka Mizuno*
Department of Applied Chemistry, School of Engineering, The UniVersity of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
Received March 25, 2008; E-mail: tmizuno@mail.ecc.u-tokyo.ac.jp
Abstract: The dicopper-substituted γ-Keggin silicotungstate TBA4[γ-H2SiW10O36Cu2(µ-1,1-N3)2] (I, TBA )
tetra-n-butylammonium) could act as an efficient precatalyst for the regioselective 1,3-dipolar cycloaddition
of organic azides to alkynes. Various combinations of substrates (four azides and eight alkynes) were
efficiently converted to the corresponding 1,2,3-triazole derivatives in excellent yields without any additives.
The present system was applicable to a larger-scale cycloaddition of benzyl azide to phenylacetylene under
solvent-free conditions (100 mmol scale) in which 21.5 g of the analytically pure corresponding triazole
could be isolated. In this case, the turnover frequency and the turnover number reached up to 14 800 h-1
and 91 500, respectively, and these values were the highest among those reported for the copper-mediated
systems so far. In addition, I could be applied to the one-pot synthesis of 1-benzyl-4-phenyl-1H-1,2,3-
triazole from benzyl chloride, sodium azide, and phenylacetylene. The catalyst effect, kinetic, mechanistic,
and computational studies show that the reduced dicopper core plays an important role in the present
1,3-dipolar cycloaddition.
the basis of kinetic, mechanistic, and theoretical studies,5c,o-q,6
there are no reports on the 1,3-dipolar cycloaddition by dinuclear
Introduction
Huisgen 1,3-dipolar cycloaddition of organic azides to alkynes
is one of the most important synthetic routes to 1,2,3-triazole
derivatives,1 which have been utilized as dyes, photostabilizers,
agrochemicals, and biochemicals.2 This transformation shows
high chemoselectivity because many functional groups do not
react with azides or alkynes. However, Huisgen cycloaddition
usually requires high reaction temperature (>ca. 353 K) and
results in the formation of a mixture of 1,4- and 1,5-regioiso-
mers.1 In 2002, the groups of Sharpless3 and Meldal4 have
independently reported that copper catalysts dramatically ac-
celerate the reaction and make it totally regioselective to the
1,4-regioisomer. The copper-catalyzed regioselective 1,3-dipolar
cycloaddition (“click reaction”) has now been used for the tailor-
made syntheses of various complex materials.5
It has been reported that the 1,3-dipolar cycloaddition is
catalyzed by copper(I) acetylide species. However, it is still
controversial to whether a mononuclear or dinuclear (or more)
acetylide species is the active species for the cycloaddition.
Some catalytically active mononuclear copper(I) complexes with
N-heterocyclic carbene, polyamine, and triazole ligands have
been isolated and characterized.5f,m,n Although the dinuclear
copper species has been postulated as a key intermediate on
copper catalysts.
Interest in the catalysis of metal-substituted polyoxometalates
(POMs), which are synthesized by the introduction of substituent
(5) (a) Pacho´n, L. D.; van Maarseveen, J. H.; Rothenberg, G. AdV. Synth.
Catal. 2005, 347, 811–815. (b) Himo, F.; Lovell, T.; Hilgraf, R.;
Rostovtsev, V. V.; Noodles, L.; Sharpless, K. B.; Fokin, V. V. J. Am.
Chem. Soc. 2005, 127, 210–216. (c) Rodionov, V. O.; Fokin, V. V.;
Finn, M. G. Angew. Chem., Int. Ed. 2005, 44, 2210–2215. (d) Sen,
G. S.; Raja, K. S.; Kaltgrad, E.; Strable, E.; Finn, M. G. Chem.
Commun. 2005, 4315–4317. (e) Molteni, G.; Bianchi, C. L.; Morinoni,
G.; Santo, N.; Ponti, A. New J. Chem. 2006, 30, 1137–1139. (f) D´ıez-
Gonza´lez, S.; Correa, A.; Cavallo, L.; Nolan, S. P. Chem.sEur. J.
2006, 12, 7558–7564. (g) Lipshutz, B. H.; Taft, B. R. Angew. Chem.,
Int. Ed. 2006, 45, 8235–8238. (h) Ladmiral, V.; Mantovani, G.;
Clarkson, G. J.; Cauet, S.; Irwin, J. L.; Haddleton, D. M. J. Am. Chem.
Soc. 2006, 128, 4823–4830. (i) Srinivasan, R.; Uttamchandani, M.;
Yao, S. Q. Org. Lett. 2006, 8, 713–716. (j) D´ıaz, D. D.; Rajagopal,
K.; Strable, E.; Schneider, J.; Finn, M. G. J. Am. Chem. Soc. 2006,
128, 6056–6057. (k) Bock, V. D.; Hiemstra, H.; Maarseveen, J. H.
Eur. J. Org. Chem. 2006, 51–68. (l) Angell, Y. L.; Burgess, K. Chem.
Soc. ReV. 2007, 36, 1674–1689. (m) Candelon, N.; Laste´coue`res, D.;
Diallo, A. K.; Aranzaes, J. R.; Astruc, D.; Vincent, J.-M. Chem.
Commun. 2008, 741–743. (n) Nolte, C.; Mayer, P.; Straub, B. F.
Angew. Chem., Int. Ed. 2007, 46, 2101–2103. (o) Ahlquist, M.; Fokin,
V. V. Organometallics 2007, 26, 4389–4391. (p) Straub, B. F. Chem.
Commun. 2007, 3868–3870. (q) Rodionov, V. O.; Presolski, S. I.; D´ıaz,
D. D.; Fokin, V. V.; Finn, M. G. J. Am. Chem. Soc. 2007, 129, 12705–
12712.
(1) Huisgen, R. In 1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.;
Wiley: New York, 1984; pp 1-176.
(6) Generally, copper(I) acetylide is well-known to prefer at least dinuclear
structures with bridging µ-acetylide groups. According to the quantum
chemical calculations reported by Straub, the dinuclear acetylide
species shows higher stability and reactivity for the 1,3-dipolar
cycloaddition than the mononuclear species.5p In addition, the Gibbs
free energy of the mononuclear transition state is equivalent to an
overall Gibbs free energy barrier and is significantly higher than that
of uncatalyzed cycloaddition.5p Finn and co-workers reported that the
reaction rate is second-order dependent on the concentration of the
copper species under catalytic conditions ([alkyne]/[Cu] > 2.5).5c
(2) (a) Alvarez, R.; Velazque, S.; San, F.; Aquaro, S.; De, C.; Perno, C. F.;
Karlsson, A.; Balzarini, J.; Camarasa, M. J. J. Med. Chem. 1994, 37,
4185–4194. (b) Genin, M. J.; et al. J. Med. Chem. 2000, 43, 953–
970. (c) Brik, A.; Lin, Y.-C.; Elder, J.; Wong, C.-H. Chem. Biol. 2002,
9, 891–896.
(3) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596–2599.
(4) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67,
3057–3064.
9
15304 J. AM. CHEM. SOC. 2008, 130, 15304–15310
10.1021/ja806249n CCC: $40.75
2008 American Chemical Society

1,3-Dipolar Cycloaddition of Azides to Alkynes
A R T I C L E S
Table 1. 1,3-Dipolar Cycloaddition of 1a to 2aa
entry
catalyst
solvent
yield of 3a (%)
1
2b
3
4
5
6
7
8
9
I
I
acetonitrile
acetonitrile
acetonitrile
DMSO
98
13
<1
51
38
8
6
4
1
1
without
I
I
I
I
I
I
DMF
Toluene
Methanol
Water
1,2-dichloroethane
acetonitrile
10 TBA4[R-H2SiW11CuO39]
11c TBA4[γ-SiW12O40]
12d TBA4[γ-SiW10O34(H2O)2] +
CuCl2 + NaN3
acetonitrile
acetonitrile
2
28
Figure 1. A ball-and-stick representation of the anion part of I. Blue, green,
black, purple, and red spheres show the copper, nitrogen (azido), tungsten,
silicon, and oxygen atoms, respectively.
13d TBA3[R-PW12O40] + CuCl2 +
NaN3
acetonitrile
acetonitrile
acetonitrile
7
14d TBA4[R-SiW12O40] + CuCl2 +
NaN3
10
10
<1
<1
15d TBA4[γ-SiW12O40] + CuCl2 +
NaN3
metal cations into the vacant site(s) of lacunary POMs, has been
rapidly increasing because of the unique reactivity depending
on their compositions and structures of active sites.7 Until now,
various kinds of metal-substituted POMs have been synthesized
and used as catalysts for various functional group transforma-
tions.7 Very recently, we have reported that the dicopper-
substituted γ-Keggin silicotungstate TBA4[γ-H2SiW10O36Cu2(µ-
1,1-N3)2] (I, TBA ) tetra-n-butylammonium, Figure 1) shows
high catalytic activity for the oxidative alkyne-alkyne homo-
coupling.8 In this paper, we report that complex I acts as an
efficient precatalyst for the regioselective 1,3-dipolar cycload-
dition of organic azides to alkynes. The catalyst effect, kinetic,
mechanistic, and computational studies support the possible
cooperative involvement of a reduced dicopper core in cata-
lyzing the present cycloaddition reaction.
16d TBA4H4[R-SiW11O39] + CuCl2 + acetonitrile
NaN3
17d TBA3H7[R-SiW9O34] + CuCl2 + acetonitrile
NaN3
18e CuCl2 + NaN3
19 Cu(ClO4)2 ·6H2O
20 CuCl2
21 Cu(OTf)2
22 CuSO4 ·5H2O
23 CuCl
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
9
<1
<1
<1
<1
9
8
3
<1
<1
24 CuI
25 Cu(Ct CPh)
26 [Cu(CH3CN)4]PF6
27 [Cu(OTf)]2 ·C6H6
a Reaction conditions: I (Cu: 0.2 mol%), 1a (1.0 mmol), 2a (1.0
mmol), solvent (1.5 mL), 333 K, 3 h, in 1 atm of Ar. Yields were
determined by GC analysis using an internal standard (naphthalene). b In
1 atm of air. c TBA4[γ-SiW12O40] (0.1 mol%). d Polyoxometalate (0.1
mol%), CuCl2 (0.2 mol%), NaN3 (0.2 mol%). e CuCl2 (0.2 mol%), NaN3
(0.2 mol%).
Results and Discussion
1,3-Dipolar Cycloaddition of Organic Azides to Alkynes
by I. The 1,3-dipolar cycloaddition of benzyl azide (1a) to
phenylacetylene (2a) in the presence of I efficiently proceeded
in polar aprotic solvents (I/1a/2a ) 1:1000:1000, Table 1).
Among the solvents tested, acetonitrile was the most effective
solvent and the corresponding 1,2,3-triazole 3a was obtained
in 98% yield (entry 1). The reaction rate in air was much slower
than that in Ar (entry 2). In contrast, nonpolar and polar protic
solvents gave low yields of 3a (entries 4-9). The use of
acetonitrile has a significant advantage of the catalyst/product
separation. After the reaction was completed, the analytically
pure crystals of 3a could be isolated in 98% yield (>99% purity,
no contamination of metal species) only by cooling the reaction
mixture (243 K, Figure S1). In addition, the precatalyst could
easily be recovered in 94% yield by addition of an excess
amount of toluene (precipitation method; see Experimental
Section) to the reaction solution. The recovered precatalyst could
be reused at least three times without loss of its catalytic activity
and selectivity (entries 2-4 in Table 2). As shown in Table 1,
the reaction hardly proceeded in the absence of the catalysts
(entry 3) or in the presence of simple copper(I) and copper(II)
salts under the same conditions (entries 19-27). The reactivity
of a mixture of CuCl2 and NaN3 for the cycloaddition was
enhanced by the addition of γ-Keggin-type divacant silicotung-
state TBA4[γ-SiW10O34(H2O)2] (entry 12), while a mixture of
CuCl2 and NaN3 was almost inactive (entry 18). On the other
hand, such an acceleration was not observed by addition of other
POMs (entries 13-17), suggesting that the divacant γ-Keggin
anion is a specific ligand for the present cycloaddition.
(7) (a) Pope, M. T. Heteropoly and Isopoly Oxometalates; Springer-Verlag:
Berlin, 1983. (b) Okuhara, T.; Mizuno, N.; Misono, M. AdV. Catal.
1996, 41, 113–252. (c) Hill, C. L. Chem. ReV. 1998, 98, 1–390. (d)
Neumann, R. Prog. Inorg. Chem. 1998, 47, 317–370. (e) Kozhevnikov,
I. V. Catalysis by Polyoxometalates; John Wiley & Sons, Ltd:
Chichester, U.K., 2002. (f) Mizuno, N.; Yamaguchi, K.; Kamata, K.
Coord. Chem. ReV. 2005, 249, 1944–1956.
In the presence of a reducing agent (sodium ascorbate, one
equiv with respect to I), the cycloaddition of 1a to 2a efficiently
proceeded at 303 K to give 3a in 98% yield and the turnover
frequency (TOF) reached up to 370 h-1 (eq 1). The TOF was
higher than those (0.2-120 h-1) of the copper catalysts (except
for [(SIMes)CuBr] (408 h-1 5f,9
) with reducing agents and/or
)
(8) (a) Kamata, K.; Yamaguchi, S.; Kotani, M.; Yamaguchi, K.; Mizuno,
N. Angew. Chem., Int. Ed. 2008, 47, 2407–2410. (b) Yamaguchi, K.;
Kamata, K.; Yamaguchi, S.; Kotani, M.; Mizuno, N. J. Catal. 2008,
258, 121–130.
nitrogen bases such as N-heterocyclic carbene, phenanthroline,
and tris(triazolyl)amine additives at ambient temperature (see
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15305

A R T I C L E S
Kamata et al.
Table 2. 1,3-Dipolar Cycloaddition of Various Organic Azides to
Alkynesa
entry
azide (R ))
alkyne (R′ ))
triazole time (h) yieldb (%)
1
Bn (1a)
1a
1a
1a
1a
1a
1a
1a
Ph (2a)
2a
2a
2a
4-CH3C6H4 (2b)
3-CH3C6H4 (2c)
4-CH3OC6H4 (2d)
4-FC6H4 (2e)
3-ClC6H4 (2f)
n-C6H13 (2g)
n-C8H17 (2h)
2a
3a
3a
3a
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3
2
2
2
1.5
2.5
2
3
5
10
18
2.5
1.5
5
4
2.5
2
98
98
99
99
90
97
94
91
89
93
88
90
88
95
94
80
94
91
2c
3c
4c
3
4
5
6
7
1a
1a
1a
8d
9d
10
11
4-CH3Bn (1b)
1b
2b
12d adamantyl (1c) 2a
13d 1c
2e
2a
2f
14
15
n-C8H17(1d)
1d
3m
3n
3o
16d 1d
2g
3.5
a Reaction conditions: I (0.1 mol%), 1 (1.0 mmol), 2 (1.0 mmol),
acetonitrile (1.5 mL), 333 K, in 1 atm of Ar. b Isolated yields. c Reuse
experiment: first recycle (entry 2), second recycle (entry 3), and third
recycle (entry 4). The reaction conditions were the same except that the
recovered precatalyst was used. d I (0.5 mol%).
Table S1). In addition, the present system was applicable to a
larger-scale cycloaddition of 1a to 2a under solVent-free
conditions (100 mmol scale) and the analytically pure 21.5 g
of 3a could be isolated (eq 2). In this case, the TOF and the
turnover number (TON) reached up to 14 800 h-1 and 91 500,
respectively, and these values were the highest among those
reported for the copper-mediated systems so far (TOF: <1-408
h-1, TON: 1-1720; see Table S1). The one-pot synthesis of
3a from benzyl chloride, sodium azide, and 2a could also be
achieved (eq 3).10b,c,j
Figure 2. (a) Changes of UV-vis spectra (∼70 min) for the 1,3-dipolar
cycloaddition of 1a to 2a in the presence of I and (b) time profiles of the
absorbance at 700 nm and formation of 3a. Reaction conditions: I (0.58
mM), 1a (0.58 M), 2a (0.58 M), acetonitrile (1.5 mL), 333 K, in 1 atm
of Ar.
was confirmed by the X-ray crystallographic analyses (Figure
S2) and/or NOE experiments that the 1,4-disubstituted-1,2,3-
triazoles were formed in a completely regioselective manner
without formation of 1,5-regioisomers in all cases. Not only
benzylic (1a and 1b) but also aliphatic azides (1c and 1d) gave
excellent yields of the corresponding 1,2,3-triazoles. Also,
aromatic terminal alkynes (2a-2f) as well as aliphatic ones (2g
and 2h) worked well as reaction partners of organic azides.
In the presence of I, various combinations of organic azides
and alkynes were efficiently converted into the corresponding
1,2,3-triazole derivatives in excellent yields without any addi-
tives such as reducing agents and nitrogen bases (Table 2). It
(9) The reaction rate for the [(SIMes)CuBr]-catalyzed cycloaddition of
1a to 2a in DMSO showed a first-order dependence on the concentra-
tion of [(SIMes)CuBr] (6.5-50.4 mM, see Supporting Information).
In the case of [(SIMes)CuBr], the cycloaddition of internal alkynes
efficiently proceeded. On the other hand, no cycloaddition of internal
alkynes proceeded in the presence of the precatalyst I. The discrepancy
suggests that the [(SIMes)CuBr]-catalyzed cyclization likely proceeds
in a different way in comparison with the I-mediated cycloaddition.
(10) (a) Saha, B.; Sharma, S.; Sawant, D.; Kundu, B. Synlett 2007, 1591–
1594. (b) Zhao, Y.-B.; Yan, Z.-Y.; Liang, Y.-M. Tetrahedron Lett.
2006, 47, 1545–1549. (c) Appukkuttan, P.; Dehaen, W.; Fokin, V. V.;
van der Eycken, E. Org. Lett. 2004, 6, 4223–4225. (d) Reddy, K. R.;
Rajgopal, K.; Kantam, M. L. Catal. Lett. 2007, 114, 36–40. (e) Zhang,
L.; Chen, X.; Xue, P.; Sun, H. H. Y.; Williams, I. D.; Sharpless, K. B.;
Fokin, V. V.; Jia, G. J. Am. Chem. Soc. 2005, 127, 15998–15999. (f)
Kantam, M. L.; Jaya, V. S.; Sreedhar, B.; Raoa, M. M.; Choudary,
B. M. J. Mol. Catal. A: Chem. 2006, 256, 273–277. (g) Chan, T. R.;
Hilgraf, R.; Sharpless, K. B.; Fokin, V. V. Org. Lett. 2004, 6, 2853–
2855. (h) Beckmann, H. S. G.; Wittmann, V. Org. Lett. 2007, 9, 1–4.
(i) Chassaing, S.; Kumarraja, M.; Sido, A. S. S.; Pale, P.; Sommer, J.
Org. Lett. 2007, 9, 883–886. (j) Sreedhar, B.; Reddy, P. S. Synth.
Commun. 2007, 37, 805–812. (k) Gerard, B.; Ryan, J.; Beeler, A. B.;
Porco, J. A., Jr. Tetrahedron 2006, 62, 6405–6411. (l) Lee, B.-Y.;
Park, S. R.; Jeon, H. B.; Kim, K. S. Tetrahedron Lett. 2006, 47, 5105–
5109.
Role of Dinuclear Copper Site. The 1,3-dipolar cycloaddition
of 1a to 2a under the conditions described in Table 1 showed
an induction period of ∼50 min (Figure 2b). During the
induction period, an almost equimolar amount of 2a with respect
to copper(II) species in I was converted into 1,4-diphenyl-1,3-
butadiyne and the absorption band at 700 nm, which is
assignable to the d-d transition of the copper(II) species in I
(Figure 2a),11 almost disappeared. These results suggest that
all copper(II) species in I are reduced to copper(I) species and
that the reaction is initiated after the formation of copper(I)
species. The idea was supported by the following facts: (1) Upon
pretreatment of I with 2a or a reducing agent of ascorbic acid
or sodium ascorbate, the 700-nm band intensity greatly de-
creased and the induction period disappeared (Figure 3b, d, and
f). (2) The reaction rates for the cycloaddition of 1a to 2a upon
pretreatment of I with 2a, ascorbic acid, and sodium ascorbate
(11) Hathaway, B. J.; Billing, D. E. Coord. Chem. ReV. 1970, 5, 143–207.
were 11.9 ( 1.4, 9.7 ( 1.5, and 10.0 ( 0.8 mM min-1
,
9
15306 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

1,3-Dipolar Cycloaddition of Azides to Alkynes
A R T I C L E S
Figure 3. Reaction profiles for the 1,3-dipolar cycloaddition of 1a to 2a.
Reaction conditions: I (Cu: 1.16 mM), 1a (0.58 M), 2a (0.58 M), acetonitrile
(1.5 mL), 333 K, in 1 atm of Ar atmosphere. The profiles show the
cycloaddition catalyzed by (a) I, (b) I pretreated with 2a at 333 K for 90
min before addition of 1a, (c) I pretreated with 1a at 333 K for 90 min
before addition of 2a, (d) I in the presence of ascorbic acid (0.58 mM, 0.5
equiv with respect to Cu), (e) TBA4[R-H2SiW11CuO39] in the presence of
ascorbic acid (0.58 mM, 0.5 equiv with respect to Cu), and (f) I in the
presence of sodium ascorbate (0.58 mM, 0.5 equiv with respect to Cu) in
a mixed solvent of acetonitrile/water (1.48/0.02 mL).
Figure 5. (a) Reaction profiles and (b) dependence of the reaction rate on
the concentration of 1a for 1,3-dipolar cycloaddition of 1a to 2a. Reaction
conditions: I (0.58 mM), 1a (0.32-1.59 M), 2a (0.58 M), acetonitrile (1.5
mL), 333 K, in 1 atm of Ar.
When the recovered precatalyst was used, the induction period
disappeared and the reaction proceeded at almost the same rate
as that after the induction period with fresh I (Figure S3).12
The catalytic activities of the monocopper-substituted TBA4[R-
H2SiW11CuO39] and the noncopper-substituted TBA4[γ-SiW12-
O40] were much lower than that of I (entries 10 and 11 in Table
1). Even in the presence of a reducing agent (ascorbic acid, 0.5
equiv with respect to Cu), TBA4[R-H2SiW11CuO39] with a
monocopper site was almost inactive for the 1,3-dipolar cy-
cloaddition (Figure 3e). In addition, the first-order dependence
of the reaction rate on the concentration of I was observed
(Figure 4).9 These facts suggest that the γ-Keggin structure of
I during the catalysis is stable and that the reduced dinuclear
Cu(I) core plays an important role in the present cycloaddition.12
The present 1,3-dipolar cycloaddition of organic azides to
alkynes possibly proceeds as follows (Scheme S1): First, the
(12) After the reaction was completed, the precipitate was recovered by
addition of an excess amount of toluene (precipitation method) to the
reaction solution. The copper content in the filtrate was less than 1%
of that in fresh I, indicating that the copper species hardly leaches
into the reaction solution. The elemental analysis and 1H NMR
spectrum of the recovered precipitate showed that the recovered I
consists of one molecule of dicopper-substituted silicotungstate and
two molecules of 3a. The UV-vis spectrum of the recovered I agreed
well with that of the sum of fresh I and 3a, suggesting that the
γ-Keggin structure of the recovered I is retained after the cycloaddition.
No absorption bands at 700 nm due to the d-d transition of the
copper(II) species was observed in the UV-vis spectrum of the
recovered I, suggesting the presence of the reduced copper(I) species
in the recovered I.
Figure 4. (a) Reaction profiles and (b) dependence of the reaction rate on
the concentration of I for 1,3-dipolar cycloaddition of 1a to 2a. Reaction
conditions: I (0.29-2.31 mM), 1a (0.58 M), 2a (0.58 M), acetonitrile (1.5
mL), 333 K, in 1 atm of Ar. Slope ) 0.99 (R2 ) 0.99).
respectively, and were almost the same as that (10.7 ( 1.1)
after the induction period with fresh I.
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15307

A R T I C L E S
Kamata et al.
acetylene-d1 under the conditions in Table 1 (kH/kD ) 1.0 (
0.1), showing that the formation of the acetylide species is not
included in the rate-limiting step. All the results show that the
reaction of the acetylide species with an azide is the rate-limiting
step.5c,o-q
The density functional theory calculations were carried out
to confirm the possible cooperative involvement of two copper
centers in catalyzing the present cycloaddition reaction. The
energies of the reaction steps were calculated with density
functional theory according to Scheme S1, and the results are
summarized in Figure 7. The formation of the dicopper(I)-
alkynyl species containing one σ,π-bridging acetylide unit (B)
by the reaction of the dicopper(I) species (A) with an alkyne
was calculated to be exothermic by 167 kJ mol-1. The formation
of the dicopper(I)-alkynyl species containing two σ,π-bridging
acetylide units was energetically unfavorable by 150 kJ mol-1
in comparison with that of B. The azide interacts with B to
form the monoalkynyl monoazido dicopper(I) intermediate (C).
This step was calculated to be exothermic by 63 kJ mol-1. Then,
the nucleophilic attack of the terminal nitrogen atom of an azide
on the positively charged carbon atom of the acetylide species
takes place to form the six-membered dicopper metallacycle
intermediate (D). This step is the rate-limiting step for the
present 1,3-dipolar cycloaddition, and the activation energy
(TS1) was calculated to be 59 kJ mol-1. This value was com-
parable to those reported for the 1,3-dipolar cycloaddition with
dicopper complex of Cu2L(Ct CMe)(N3Me)5o,p (L ) phenan-
throline, Cl-, Ct CMe-, and H2O, 44-78 kJ mol-1) calculated
at the B3LYP level. The activation energy (TS2) for the
formation of the triazolyl-dicopper species (E) from D was very
low. Triazole is eliminated followed by the reaction with an
alkyne and an azide, establishing the catalytic cycle. These
computational results also support that an alkyne and an organic
azide are activated cooperatively by the dinuclear copper site.
Figure 6. (a) Reaction profiles and (b) dependence of the reaction rate on
the concentration of 2a for 1,3-dipolar cycloaddition of 1a to 2a. Reaction
conditions: I (0.58 mM), 1a (0.58 M), 2a (0.32-1.59 M), acetonitrile (1.5
mL), 333 K, in 1 atm of Ar.
Conclusion
alkyne-alkyne homocoupling proceeds on the dicopper(II) site
in I to form the corresponding diyne and reduced copper(I)
species. This step corresponds to the induction period for the
present cycloaddition. Then, the copper(I) acetylide species
would be formed by the reaction of the copper(I) species with
an alkyne followed by the reaction with an azide to form the
corresponding 1,2,3-triazole. The reaction rates for the I-
mediated 1,3-dipolar cycloaddition of 1a to 2a showed a first-
order dependence on the concentrations of I (0.29-2.31 mM)
and 1a (0.32-1.04 M) and were almost independent of the
concentration of 2a (0.32-1.59 M) under the present catalytic
conditions (Figures 4-6). No kinetic isotope effects were
observed for the 1,3-dipolar cycloaddition of 1a to 2a and phenyl-
In summary, complex I exhibited high catalytic activity for
the 1,3-dipolar cycloaddition of organic azides to alkynes
without any additives such as reducing agents and nitrogen
bases. In addition, the one-pot synthesis of triazole from a halide,
sodium azide, and an alkyne could be realized. The kinetic,
spectroscopic, mechanistic, and computational investigations
showed that the reduced dicopper core plays an important role
in the present cycloaddition.
Experimental Section
General. GC analyses were performed on a Shimadzu GC-2014
with a flame ionization detector equipped with a TC-5 capillary
9
15308 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

1,3-Dipolar Cycloaddition of Azides to Alkynes
A R T I C L E S
Figure 7. Calculated energy diagram of the 1,3-dipolar cycloaddition of methylazide to methyl acetylene on the reduced precatalyst I in the gas phase
(energies and lengths in kJ mol-1 and Å, respectively). Blue, black, green, red, and light blue balls represent copper, carbon, nitrogen, oxygen, and hydrogen
atoms, respectively. Polyoxotungstate frameworks of [γ-HxSiW10O32]x+ (x ) 2 and 3 for A and the other states, respectively) are omitted for clarity.
column (internal diameter ) 0.25 mm, length ) 60 m) or an
InertCap 5 capillary column (internal diameter ) 0.25 mm, length
) 60 m). HPLC analyses were performed on an Agilent 1100 series
LC with a UV-vis detector using a CAPCELL PAK MG C18
reversed phase column (5 µm × 3 mmφ × 250 mm, SHISEIDO
FINE CHEMICALS). Mass spectra were recorded on a Shimadzu
GCMS-QP2010 equipped with a TC-5HT capillary column (internal
diameter ) 0.25 mm, length ) 30 m). NMR spectra were recorded
at 298 K on a JEOL JNM-EX-270 (1H, 270 MHz and 13C, 67.8
MHz) spectrometer. IR spectra were measured on a Jasco FT/IR-
460 Plus using KBr disks. UV-vis spectra were recorded on a
Jasco V-570 spectrometer with a Unisoku thermostatic cell holder
(USP-203). Sodium azide and copper salts were obtained from
Wako Pure Chemical Industries or Kanto Chemical (reagent grade)
and used as received. Acetonitrile, dichloromethane, and toluene
(Kanto Chemical) were purified by The Ultimate Solvent System
(GlassContour Company).13 Other solvents and alkynes were
obtained from TCI or Aldrich (reagent grade) and purified prior to
use.14 Organic azides were prepared by the nucleophilic substitution
of the corresponding alkyl chloride with sodium azide.15 The
dicopper-substituted silicotungstate I was synthesized according to
the literature procedure (see Supporting Information).8
1,3-Dipolar Cycloaddition of Organic Azides to Alkynes. A
typical procedure for the dipolar cycloaddition of azides to alkynes
was as follows: Into a glass vial were successively placed I (0.1
mol%), 1a (1.0 mmol), 2a (1.0 mmol), and acetonitrile (1.5 mL).
The reaction mixture was stirred at 333 K in 1 atm of Ar. The
yield was periodically determined by GC analysis. After the reaction
was completed, the reaction solution was cooled to 243 K and
analytically pure crystals of 3a appeared (Figure S1). The crystals
were isolated from the reaction solution by decantation (98%
isolated yield). Purities of isolated products were determined by
1H NMR (>99% in all cases). Toluene (50 mL) was added to the
filtrate, and the precipitated precatalyst was recovered by filtration
(94% recovery).
The present cycloaddition showed an induction period. Good
linearity was observed between the product amount and the reaction
time after the induction period (Figures 2-6). When the recovered
precatalyst was used, the induction period disappeared and the
reaction proceeded at almost the same rate as that after the induction
period with fresh I (Figure S3). Therefore, the reaction rates (R0)
for the kinetic analyses were determined from the slopes of the
reaction profiles (conversion vs time plots) after the induction
period.
Quantum Chemical Calculations. The calculations were carried
out at the B3LYP level theory16 with 6-31G* basis sets for H, C,
N, O, and Si atoms and the double-ꢀ quality basis sets with effective
core potentials proposed by Hay and Wadt17 for Cu and W atoms.
The entire structure of the dicopper-substituted polyoxotungstate
was used as a model in the calculations. Methyl azide and
methylacetylene were used as model substrates. To minimize the
differences in the solvent effect among the intermediates and
transition states, protons were added to the polyoxometalate and
the total charge of the system was kept -4 throughout the cal-
culations. The geometry of A was optimized within the C2 symmetry
restrictions. The geometries of B, C, D, E, TS1, and TS2 were
optimized without the symmetry restrictions. Transition state
structures were searched by numerically estimating the matrix of
second-order energy derivatives at every optimization step and by
requiring exactly one eigenvalue of this matrix to be negative. For
the transition states, the frequency analysis was conducted at the
same level at the final geometry. The optimized geometries were
shown in Table S3 and Figure S5 (Supporting Information). The
(13) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518–1520.
(14) Purification of Laboratory Chemicals, 3rd ed.; Perrin, D. D., Armarego,
W. L. F., Eds.; Pergamon Press: Oxford, U.K., 1988.
(15) Lieber, E.; Chao, T. S.; Rao, C. N. R. J. Org. Chem. 1957, 22, 238–
240.
(16) Becke, A. D. J. Chem. Phys. 1993, 98, 1372–1377.
(17) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270–283.
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15309

A R T I C L E S
Kamata et al.
zero-point vibrational energies were not included. All calculations
were performed with the Gaussian03 program package.18
Education, Culture, Science, Sports and Technology of Japan. We
are grateful to Dr. S. Yamaguchi (JST) and Ms. M. Kotani (JST)
for their help in experiments.
Acknowledgment. This work was supported by the Core
Research for Evolutional Science and Technology (CREST)
program of the Japan Science and Technology Agency (JST) and
a Grant-in-Aid for Scientific Research from the Ministry of
Supporting Information Available: Experimental details, data
of triazole derivatives 3a-3o, Tables S1-S3, Scheme S1, and
Figures S1-S5. This material is available free of charge via
the Internet at http://pubs.acs.org.
(18) Frisch, M. J. et al. Gaussian 03, revision D.02; Gaussian, Inc.:
Wallingford, CT, 2004.
JA806249N
9
15310 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008