Regioselective synthesis of [1,2,3]-triazoles catalyzed by Cu(I) generated in situ from Cu(0) nanosize activated powder and amine hydrochloride salts

Article abstract of DOI:10.1016/j.tetlet.2005.02.127

A straightforward and efficient method for the regioselective synthesis of functionalized 1,4-disubstituted [1,2,3]-triazoles, from terminal alkynes and azides, has been established utilizing Cu(0) as the source of the catalytic species. The presumed cata

Full text of DOI:10.1016/j.tetlet.2005.02.127

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 2911–2914
Regioselective synthesis of [1,2,3]-triazoles catalyzed by
Cu(I) generated in situ fromCu(0) nanosize activated
powder and amine hydrochloride salts
*
Hernan A. Orgueira, Demosthenes Fokas, Yuko Isome, Philip C.-M. Chan
´
and Carmen M. Baldino
Department of Chemistry, ArQule, Inc., 19 Presidential Way, Woburn, MA 01801, USA
Received 30 November 2004; revised 14 February 2005; accepted 18 February 2005
Abstract—A straightforward and efficient method for the regioselective synthesis of functionalized 1,4-disubstituted [1,2,3]-triazoles,
from terminal alkynes and azides, has been established utilizing Cu(0) as the source of the catalytic species. The presumed catalytic
Cu(I) species is generated by the combination of 10 mol% copper nanosize activated powder and 1 equiv of an amine hydrochloride
salt. The addition of an amine hydrochloride salt into the reaction mixture enhanced the dissolution of copper metal, and subse-
quently facilitated the formation of the Cu(I)-acetylide intermediate required for the regioselective cycloaddition.
Ó 2005 Elsevier Ltd. All rights reserved.
[1,2,3]-Triazoles with general structures 3 and 4 are
important five-membered nitrogen heterocycles, in-
volved in a wide range of industrial applications such
as agrochemicals, corrosion inhibitors, dyes, optical
brighteners as well as biologically active agents.1 The
well established approach utilized thus far for the syn-
thesis of the [1,2,3]-triazole ring system relies on the
thermal 1,3-dipolar Huisgen cycloaddition between
alkynes 1 and azides 2 (Scheme 1, a).2 However, this
non-catalyzed process exhibits several disadvantages,
including: (i) the requirement for high temperature con-
ditions with the potential for the decomposition of labile
products, (ii) the synthesis of the desired [1,2,3]-triazoles
generally in low yields, and (iii) poor regioselectivity,
given that the non-catalyzed cycloaddition affords a
mixture of 1,4- and 1,5-disubstituted triazoles, unless
the alkyne is substituted with an electron withdrawing
group.3
lyzed [3+2] cycloaddition of terminal alkynes and azides
has only recently been described (Scheme 1, b).5
It is postulated that the reaction proceeds via a copper-
acetylide intermediate, generated from Cu(I) and the ter-
minal alkyne, which then participates in an annealing
process upon its coordination with the reacting azide.5b
Although Cu(I) could be introduced directly in the form
of different copper salts, the presence of a nitrogen con-
taining base as well as prior exclusion of oxygen from
the reaction are usually required in order to minimize
the formation of undesired by-products, primarily
diacetylenes.5a Alternatively, the catalytic Cu(I) species
could be generated in situ from CuSO4 and sodium
ascorbate.5b The latter method eliminates the problem
of by-product formation and has been used successfully
in a H2O/t-BuOH solvent system, without the need for
prior exclusion of oxygen or the presence of a nitrogen
base.
Over the years, several efforts to control the 1,4- versus
1,5-regioselectivity have been reported.4 However, the
regioselective and high yielding synthesis of 1,4-substi-
tuted triazoles with general structure 3 via a Cu(I) cata-
The observation by Sharpless and Fokin suggesting that
even Cu(0) coiled metal turnings, albeit in stoichiometric
quantity,5b can be used as a source of the catalytic spe-
cies for the regioselective formation of [1,2,3]-triazoles
sparked our interest for further investigation. Indeed,
although longer reaction times are required when
Cu(0) is used, we felt these findings had substantial
implications and the potential for further development.
Keywords: Cu(I) generated in situ; Triazoles; Regioselective synthesis.
*
Corresponding author. Tel.: +1 781 994 0474; fax: +1 781 376
6019; e-mail: horgueira@arqule.com
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.02.127

2912
H. A. Orgueira et al. / Tetrahedron Letters 46 (2005) 2911–2914
R2
R2
N
R1
R1
N
N
3
N
1
5
∆
N
N
N
+
N
N
a
R2
4
2
+
ca. 1:1
via
[3+2]
R1
4
N
R1
R1
Cu
Cu(I)
b
N
1
N
1
N
N
R2
N
R2
3
Scheme 1.
Compelled to conduct additional studies in search of
means to enhance the catalytic process for the regio-
selective synthesis of this ring system, we surmised that
the use of a catalytic amount of Cu(0) nanosize activated
powder, with a larger surface area relative to the coiled
metal turnings, could enhance the formation of copper-
acetylide and subsequently facilitate the cycloaddition
process. Our initial attempts using 10 mol% of activated
copper powder, with different azides and terminal alky-
nes in a H2O/t-BuOH solvent system for 2 h, provided
only trace amounts of the desired product or recovered
starting materials. However, when an equimolar mixture
of propargyl amine hydrochloride and benzyl azide was
subjected to the same reaction conditions, efficient disso-
lution of copper powder was observed along with the
regioselective formation of the corresponding 1,4-disub-
stituted triazole 3a6a (Scheme 2). Given that no reaction
occurred when propargyl amine was utilized as the free
base, we speculated that the presence of an amine ligand
and a slightly acidic environment,7 which would pre-
sumably be ensured with the addition of an amine
hydrochloride salt, might be required to induce the dis-
solution of copper and subsequently, trigger the genera-
tion of the Cu(I) catalytic species.
The reaction worked equally well, albeit required longer
reaction times (12–24 h), even when less than 1 mol%
copper was used as the catalyst.9 However, trace
amount of product was observed under the optimized
reaction conditions and within the same reaction times
when Et3N was utilized as the free base. It is noteworthy
to mention that the reaction worked well with a second-
Table 1.
Entry
Producta
% Yieldb
N
N
N
1
91 (3b)
S
O
N
N
N
2
3
4
92 (3c)
92 (3d)
95 (3e)
MeO
N
O
N
N
HO
N
N
N
N
N
H
N
Boc
5
6
94 (3f)
96 (3g)
In order to test our hypothesis, the cycloaddition reac-
tion between several azides and terminal alkynes, in
the presence of 10 mol% Cu(0) powder and 1 equiv of
NEt3ÆHCl salt, was attempted. Indeed, as we had ex-
pected, the reaction proceeded smoothly at room tem-
perature and led to the facile and regioselective
formation of the desired 1,4-disubstituted [1,2,3]-triaz-
oles (Table 1). The chemistry works well with aliphatic
(entries 6–8), benzylic (entries 1–5) and aryl azides (entry
9), and tolerates a wide spectrum of electron donating
and electron withdrawing functional groups in both
the alkyne and azide starting material.
N
N
N
N
NBoc
N
N
N
7
93 (3h)
NBoc
O
O
8a R = Et
8b R = Me
8c R = H
88 (3i)
91 (3j)
95 (3k)
N
N
N
O
RO
N
N
N
OH
OH
9
83 (3l)
NH .HCl
2
NH .HCl
2
N
0
N
3
O
Cu (10 mol%)
N
+
a All reactions were completed within 2 h and carried out in a H2O/
t-BuOH solution containing 10 mol% of activated Cu nanosize
powder and 1 equiv of Et3NÆHCl salt.
N
H O/t-BuOH
2
Ph
rt, 2 h
Ph
3a
b Isolated yields. All compounds produced satisfactory 1H NMR, 13C
NMR, and mass spectra and matched those reported in literature.8
Scheme 2.

H. A. Orgueira et al. / Tetrahedron Letters 46 (2005) 2911–2914
2913
ary amine hydrochloride salt such as Et2NHÆHCl, but
failed with NH4Cl.
due its thermodynamic instability. Formation of the
more stable Cu(II) species eventually prevails, as it
was evidenced by the appearance of a blue coloured
solution when the reaction was completed.
Although the compounds reported in Table 1, except
3e5b and 3f,14 have previously been successfully synthe-
sized via a thermal Huisgen [3+2] cycloaddition reac-
tion, the previous methods required laborious isolation
and resulted in low yields (15–50%) from a regioisomeric
mixture of the corresponding 1,4- and 1,5-triazoles.8 In
contrast, the efficient and regioselective synthesis of
these compounds was achieved in a straightforward
manner by application of our enhanced protocol.
In summary, we have developed a mild and efficient
protocol for the copper-catalyzed 1,3-dipolar cycloaddi-
tions of terminal alkynes and azides, in which the pres-
ence of an amine hydrochloride salt enhances the
dissolution of Cu(0) resulting in the facile generation of
the catalytic Cu(I) species. This set of conditions nicely
complement13 the work developed previously by Meldal
and Sharpless, and provides facile access to functional-
ized triazoles not easily synthesized via the thermal
Huisgen conditions. The operational simplicity of this
method and the purity of the recovered products14
makes it attractive not only for the large scale synthesis
of this class of biologically active molecules, but for the
synthesis of screening libraries for drug discovery as
well.
Similarly, exploiting the chemistry shown in Scheme 2
led to the design and synthesis of several amino func-
tionalized triazoles, starting from the corresponding
hydrochloride salts of amino containing azides (entries
3 and 4) or amino containing alkynes (entries 1 and 2)
(Table 2).10 The cycloaddition reaction was found to
work comparably, even without the addition of an exter-
nal amine hydrochloride salt, regardless of whether the
amine salt resided on the alkyne or the azide component.
This class of compounds, not readily accessible by previ-
ous non-catalyzed methods, were synthesized in this
manner without the need for prior use of protecting
groups.11
References and notes
1. For reviews of [1,2,3]-triazoles, see: (a) Dehne, H. In
Methoden de Organischen Chemie (Houben-Weyl); Schu-
mann, E., Ed.; Thieme: Stuttgart, 1994; Vol. E8d, pp 305–
320; (b) Wamhoff, H. In Comprehensive Heterocyclic
Chemistry; Katritzky, A. R., Rees, C. W., Eds.; Pergamon:
Oxford, 1984; Vol. 5, pp 669–732; (c) Bo¨hm, R.; Karow,
C. Pharmazie 1981, 36, 243–247.
2. For reviews of 1,3-dipolar cycloadditions see: (a) Huisgen,
R. In 1,3-Dipolar Cycloaddition Chemistry; Padwa, A.,
Ed.; Wiley: New York, 1984; pp 1–176, Chapter 1; (b) Sha,
C.-K.; Mohanakrishan, A. K. In Synthetic Applications of
1,3-Dipolar Cycloaddition Chemistry Toward Heterocycles
and Natural Products; Padwa, A., Pearson, W. H., Eds.;
Wiley: New York, 2003; pp 623–680; (c) Karlsson, S.;
Hogberg, H. E. Org. Prep. Proced. Int. 2001, 33, 103–172;
(d) Gothelf, K. V.; Jorgensen, K. A. Chem. Rev. 1998, 98,
863–909.
3. (a) Winter, W.; Muller, E. Chem. Ber. 1974, 107, 705–709;
(b) Bastide, J.; Henri-Rousseau, O. In The Chemistry of
the Carbon–Carbon Triple Bond; Patai, S., Ed.; Inter-
science: London, 1978; pp 447–552.
4. (a) Palacios, F.; Ochoa de Retana, A. M.; Pagalday, J.;
Sanchez, J. M. Org. Prep. Proced. Int. 1995, 27, 603–612;
(b) Hlasta, D. J.; Ackerman, J. H. J. Org. Chem. 1994, 59,
6184–6189; (c) Mock, W. L.; Irra, T. A.; Wepsiec, J. P.;
Adhya, M. J. Org. Chem. 1989, 54, 5302–5308; (d) Peng,
W.; Zhu, S. Synlett 2003, 187–190.
The dissolution of copper in aqueous systems is a well
known process12a and thus the generation of Cu(I) spe-
cies might presumably proceed via a stepwise mecha-
nism. First, oxidative dissolution of copper, facilitated
by the presence of an amine hydrochloride salt,12b fol-
lowed by its coordination with a nitrogen based ligand,
could result in the production of a Cu(I)-amino complex
and posterior formation of the copper acetylide com-
plex.12c Subsequently, Cu(I) could be easily oxidized to
Cu(II) and/or disproportionated12d to Cu(0) and Cu(II)
Table 2.
Entry
Producta
% Yieldb
N
N
N
1
93 (3m)
HCl.N
N.HCl
N
2
95 (3n)
N
N
5. (a) Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org.
Chem. 2002, 67, 3057–3064; (b) Rostovtsev, V. V.; Green,
L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem., Int.
Ed. 2002, 41, 2596–2599; For the Pd(0)–Cu(I) catalyzed
synthesis of triazoles from nonactivated terminal alkynes
via a three component reaction see: (c) Kamijo, S.; Jin, T.;
Huo, Z.; Yamamoto, Y. J. Am. Chem. Soc. 2003, 125,
7786–7787; While this manuscript was in preparation, a
microwave-assisted click chemistry synthesis of 1,4-disub-
stituted 1,2,3-triazoles via a three component reaction was
reported: (d) Appukkuttan, P.; Dehaen, W.; Fokin, V. V.;
Van der Eycken, E. Org. Lett. 2004, 6, 4223–4225.
6. (a) Spectral data for compound 3a: 1H NMR (D2O) d 7.96
(s, 1H), 7.27–7.20 (m, 5H), 5.46 (s, 2H), 4.64 (s, 2H); 13C
NMR (D2O) d 134.7, 129.3, 129.0, 128.3, 125.6, 54.1, 34.2.
MS (ES+) for C10H12N4 (M+H) 189.
N
N
N
N
3
4
90 (3o)
94 (3p)
NH.HCl
NH.HCl
N
N
N
a All reactions were completed within 2–5 h and carried out in a H2O/
t-BuOH solution containing 10 mol% of Cu nanosize activated
powder.
b Isolated yields as free bases. All compounds produced satisfactory
1H NMR, 13C NMR, and mass spectra.

2914
H. A. Orgueira et al. / Tetrahedron Letters 46 (2005) 2911–2914
7. The pH of the reaction mixture was slightly acidic
3.21–3.18 (m, 2H), 2.77–2.70 (m, 2H), 2.15–2.11 (m, 2H),
1.90–1.80 (m, 3H); 13C NMR (CDCl3) d 147.6, 139.3,
128.9, 128.8, 126.7, 119.2, 58.7, 45.7, 34.1, 32.6. MS (ES+)
for C14H18N4 (M+H) 243.
(pH = 5) upon the addition of Et3NÆHCl. After the
reaction was completed, it turned slightly basic (pH = 8).
8. For compound 3b, see: Biagi, G.; Livi, O.; Ramacciotti, G.
L.; Scartoni, V.; Bazzichi, L.; Mazzoni, M. R.; Lucacch-
inni, A. Il Farmaco 1990, 45, 49–57; For compound 3c,
see: Blass, B. E.; Coburn, K. R.; Falukner, A. L.; Seibel,
W. L.; Srivastava, A. Tetrahedron Lett. 2003, 44, 2153–
2155; For compound 3d, see: De las Heras, F. G.; Alonso,
R.; Alonso, G. J. Med. Chem. 1979, 22, 496–501; For
compound 3e, see Ref. 6b. For compounds 3g and 3h, see:
William, R.; Leeson, P. D.; Moore, K. W. PCT Int. Appl.
1996, 24–27; For compound 3i, see: Palacios, F.; Ochoa de
Retana, A. M.; Pagalday, J.; Sanchez, J. M. Org. Prep.
Proced. Int. 1995, 27, 603–612; For compound 3l, see: Da
Settimo, A.; Livi, O.; Biagi, G.; Lucacchinii, A.; Caselli, S.
11. For the use of aminotriazoles as Cu(I)-stabilizing ligands
in catalysis, see: Chan, T. R.; Hilgraf, R.; Sharpless, K. B.;
Fokin, V. V. Org. Lett. 2004, 6, 2853–2855.
12. (a) Thayer, J. S. Adv. Organomet. Chem. 1995, 38, 71–74;
(b) Timms, P. L.; Turney, T. W. Adv. Organomet. Chem.
1977, 15, 84–87; (c) Jukes, A. E. Adv. Organomet. Chem.
1974, 12, 228–229; (d) Crivelli, I. G.; Andrade, C.;
Francois, M. A.; Boys, D.; Haberland, A.; Segura, R.;
Leiva, A. M.; Loeb, B. Polyhedron 2000, 19, 2289–2295.
13. The new protocol could have a potential use in biocon-
jugation since sodium ascorbate was reported to induce
subtantial disassembly of proteins: (a) Wang, Q.; Chan, T.
R.; Hilgraf, R.; Fokin, V. V.; Sharpless, K. B.; Finn, M.
G. J. Am. Chem. Soc. 2003, 125, 3192–3193; (b) Link, A.
J.; Vink, M. K. S.; Tirrell, D. A. J. Am. Chem. Soc. 2004,
126, 10598–10602.
´
Farmaco Edizione Scientifica 1983, 38, 725–737; The
requisite 4-azido salicylic acid was prepared according to
Gano, K. W.; Monbouquette, H. G.; Myles, D. C.
Tetrahedron Lett. 2001, 42, 2249–2251.
9. The reaction worked successfully even with 0.1 mol%
copper. The latter was the lowest stoichiometry tested
during this study.
14. Typical experimental procedure for the synthesis of
triazole 3f: N-Boc propargyl amine (15.5 g, 0.1 mol) and
benzyl azide (8.4 g, 0.1 mol) were suspended in a mixture
(1:1) of H2O/t-BuOH (50 mL). Cu(0) nanosize activated
powder (Aldrich) (0.6 g, 0.0094 mol) and Et3NÆHCl
(13.7 g, 0.1 mol) were added, and the heterogeneous
mixture was stirred vigorously for over 2 h. The reaction
mixture was diluted with water and a white precipitate was
formed, which was then collected by filtration. The filtrate
was washed with water, and then dried under reduced
pressure to afford the desired triazole as a white crystalline
10. For compound 3m, see Ref. 6b. For compound 3n, see:
Stanslas, J.; Hagan, D. J.; Ellis, M. J.; Turner, C.;
Carmichael, J.; Ward, W.; Hammonds, T. R.; Stevens,
M. F. G. J. Med. Chem. 2000, 43, 1563–1572; The
9-azidoacridine was prepared according to Mair, A. C.;
Stevens, F. G. J. Chem. Soc., Perkin Trans. 1 1972, 161–
1
165. Spectral data for compound 3o: H NMR (CDCl3) d
7.82–7.7 (m, 3H), 7.41–7.28 (m, 3H), 4.60–4.54 (m, 1H),
3.24 (d, J = 11.5 Hz, 2H), 2.80–2.75 (m, 2H), 2.21 (d, J =
11.5 Hz, 2H), 1.99–1.89 (m, 2H), 1.76 (br s, 1H); 13C
NMR (CDCl3) d 147.7, 130.9, 129.0, 128.3, 125.9, 117.5,
58.7, 45.7, 34.2. MS (ES+) for C13H16N4 (M+H) 229.
Spectral data for compound 3p: 1H NMR (CDCl3) d 7.31–
7.19 (m, 5H), 7.17 (s, 1H), 4.52–4.44 (m, 1H), 4.07 (s, 2H),
1
solid 3f (27.2 g, 94%). H NMR (CDCl3) d 7.41 (s, 1H),
7.31–7.26 (m, 3H), 7.21–7.17 (m, 2H), 5.42 (s, 2H), 5.35
(br s, 1H), 4.28 (d, J = 5.8 Hz, 2H), 1.34 (s, 9H); 13C NMR
(CDCl3) d 156.1, 146.1, 134.9, 129.3, 128.9, 128.2, 122.1,
79.7, 54.3, 36.3, 28.5. MS (ES+) for C15H20N4O2 (M+H)
289.