Direct synthesis of 1,5-disubstituted-4-magnesio-1,2,3-triazoles, revisited

Article abstract of DOI:10.1021/ol0499203

After revisiting earlier works reporting the regioselective synthesis of 1,5-disubstituted-1,2,3-triazoles via the addition of bromomagnesium acetylides to azides, much improved yields of the products were obtained for a wide array of azides and alkynes. The intermediates of that reaction can be trapped with different electrophiles to regioselectively form 1,4,5-trisubstituted 1,2,3-triazoles.

Full text of DOI:10.1021/ol0499203

ORGANIC
LETTERS
2004
Vol. 6, No. 8
1237-1240
Direct Synthesis of
1,5-Disubstituted-4-magnesio-1,2,3-triazoles,
Revisited
Antoni Krasin´ski,* Valery V. Fokin, and K. Barry Sharpless
Department of Chemistry and the Skaggs Institute for Chemical Biology, The Scripps
Research Institute, 10550 N. Torrey Pines Road, La Jolla, California 92037
kranik@scripps.edu
Received January 13, 2004
ABSTRACT
After revisiting earlier works reporting the regioselective synthesis of 1,5-disubstituted-1,2,3-triazoles via the addition of bromomagnesium
acetylides to azides, much improved yields of the products were obtained for a wide array of azides and alkynes. The intermediates of that
reaction can be trapped with different electrophiles to regioselectively form 1,4,5-trisubstituted 1,2,3-triazoles.
1,2,3-Triazoles are attractive constructs, which because of
their unique chemical properties and structure should find
many applications in organic, organometallic, and medicinal
chemistry, as well as in materials chemistry. Not present in
natural products, they are remarkably stable to metabolic
transformations, such as oxidation, reduction, and both basic
and acidic hydrolysis. Furthermore, 1,2,3-triazole moieties
are emerging as powerful pharmacophores in their own right.1
However, probably because of the lack of convenient direct
methods for their synthesis, these aromatic heterocycles have
not received as much attention as they deserve.
Known methods for the regioselective synthesis of 1,5-
disubstituted- and 1,4,5-trisubstituted-1,2,3-triazoles include
reactions of azides with active methylene compounds.2
Functional groups may also be introduced to the existing
1,2,3-triazole ring by lithiation of the heterocycle followed
by reaction with an electrophile.3 One of the most attractive
approaches to the synthesis of 1,2,3-triazoles are 1,3-dipolar
cycloadditions of azides and alkynes.4 Although there are
known examples of influencing the regiochemistry of the
addition by the electronic properties of the substrate, they
are neither general nor reliable, usually requiring a strong
electron-withdrawing substituent on the alkyne.5 Metals, such
as tin, germanium, or silicon, attached to the acetylenic
carbon atom have been shown to give mainly 4-metalated
1,5-disubstituted triazoles.6 Reactions of sodium, lithium, or
magnesium acetylides with organic azides have also been
reported.7,8 1,5-Disubstituted 1,2,3-triazoles are the major
products of these reactions. The scope of this transformation
was investigated by Akimova et al. in the late 1960s.8
Extensive studies were presented, in which a wide variety
of azides and lithium or magnesium acetylides were em-
(4) Huisgen, R. In 1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.;
Wiley: New York, 1984; pp 1-176.
(5) (a) Padwa, A. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Ed.; Pergamon: Oxford, 1991; Vol. 4, pp 1069-1109. (b) Fan, W.-Q.;
Katritzky, A. R. In ComprehensiVe Heterocyclic Chemistry II; Katritzky,
A. R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon: Oxford, 1996; Vol.
4, pp 101-126.
(6) Himbert, G.; Frank, D.; Regitz, M. Chem. Ber. 1976, 109, 370.
(7) (a) Fridman, S. G.; Lisovska, N. M. Zap. Inst. Khim., Akad. Nauk
U.R.S.R., Inst. Khim. 1940, 6, 353. (b) Boyer, N. M.; Mack, C. H.; Goebel,
N.; Morgan, L. R., Jr. J. Org. Chem. 1958, 23, 1051. (c) Akimova, G. S.;
Chistokletov, V. N.; Petrov, A. A. Zh. Org. Khim. 1965, 1, 2077.
(8) (a) Akimova, G. S.; Chistokletov, V. N.; Petrov, A. A. Zh. Org. Khim.
1967, 3, 968. (b) Akimova, G. S.; Chistokletov, V. N.; Petrov, A. A. Zh.
Org. Khim. 1967, 3, 2241. (c) Akimova, G. S.; Chistokletov, V. N.; Petrov,
A. A. Zh. Org. Khim. 1968, 4, 389.
(1) (a) Bourne, Y.; Kolb, H. C.; Radic´, Z.; Sharpless, K. B.; Taylor, P.;
Marchot, P. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 1449. (b) Lewis, W.
G.; Green, L. G.; Grynszpan, F.; Radic´, Z.; Carlier, P. R.; Taylor, P.; Finn,
M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2002, 41, 1053.
(2) L’Abbe`, G. Ind. Chim. Belg. 1969, 34, 519 and references therein.
(3) (a) Holzer, W.; Ruso, K. J. Heterocycl. Chem. 1992, 29, 1203. (b)
Ohta, S.; Kawasaki, I.; Uemura, T.; Yamashita, M.; Yoshioka, T.;
Yamaguchi, S. Chem. Pharm. Bull. 1997, 45, 1140. (c) Uhlmann, P.;
Felding, J.; Vedsø, P.; Begtrup, M. J. Org. Chem. 1997, 9177. (d) Felding,
J.; Uhlmann, P.; Kristensen, J.; Vedsø, P.; Begtrup, M. Synthesis 1998,
1181.
10.1021/ol0499203 CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/12/2004

Table 1. Synthesis of 1,5-Disubstituted-1,2,3-triazoles
a Unless stated otherwise, the yields are crude; yields in parentheses are taken from the original papers.8 b After flash chromatography. c After recrystallization.
ployed. Their proposed mechanism (Scheme 1) begins with
the nucleophilic attack of the acetylide 1 on the terminal
nitrogen atom of the azide 2, followed by spontaneous
closure of the linear intermediate 3 to the 4-metallotriazole
species 4. The reaction of the bromomagnesium acetylides
with azides gave, after hydrolysis, preferentially the 1,5-
disubstituted triazoles 5; however, the yields were low.
Formation of byproduct 7 was reported in some cases along
with unreacted starting materials. On the other hand, the
use of lithium acetylides favored further attack by inter-
mediate 4 on a second molecule of azide, resulting upon
hydrolysis, in the 4-triazene-substituted triazoles 7, often in
high yields.8
Finding the copper(I)-catalyzed route to 1,4-disubstituted-
1,2,3-triazoles from azides and terminal alkynes9 highlighted
the need for a direct way to achieve the complementary union
of the same reactants, namely, one giving the regioisomeric
1,5-triazole analogues. Although Akimova, et al. had studied
such a process over 30 years ago,8 a citation search turned
(9) Rostovtsev, V. V.; Green, L. G.; Fokin V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596.
1238
Org. Lett., Vol. 6, No. 8, 2004

Scheme 1. Proposed Mechanism for the Synthesis of
1,5-Disubstituted-1,2,3-triazoles
Figure 1. Reaction byproducts.13
Although the reaction is usually fast and clean, several
types of trace byproducts can be observed (Figure 1). Long
reaction times without strict exclusion of oxygen promote
oxidative couplings, which result in the formation of
structures such as i, ii, or iii. Byproducts can also include 7
(vide supra), but an excess of the azide over the acetylide as
well as extended reaction times are required to form
significant amounts of 7. When bromomagnesium acetylides
are used as substrates, trace amounts of 1,5-disubstituted-
4-bromotriazoles are observed.12 The presence of the 4-halo-
Table 2. Synthesis of Fully Substituted 1,2,3-Triazoles
up no further uses of their nice one-step procedure, likely
because of the poor to moderate yields they reported. In any
case, we find that Akimova’s procedure actually gives 1,5-
disubstituted-1,2,3-triazoles in much better yields than orig-
inally reported (Table 1).8
Most reactions furnished 1,5-disubstituted-1,2,3-triazoles
at room temperature overnight in almost quantitative yields,
and the crude products required no purification.10 Aliphatic
acetylides are less reactive than their aryl counterparts.
Electron-poor azides react much faster than the electron-
rich ones, an observation consistent with the proposed
nucleophilic attack of the acetylide on the azide (Scheme
1). For example, products 5c and 5g were formed almost
instantly after mixing the reagents (concentration of 0.4 M),
whereas the formation of product 5s, from two of the least
reactive components, required several days under the same
conditions to reach completion.11
(10) The reactions are carried out at concentrations ranging from 0.4 to
1.0 M on a 2-10 mmol scale. Experimental Procedure. To the dried flask
containing a solution of EtMgBr or EtMgCl in dry THF under a nitrogen
atmosphere, the terminal alkyne is added dropwise at room temperature.
Good results are usually obtained using commercially available 1 M EtMgBr
in THF and no extra solvent. After the alkyne is added, the solution is
heated to 50 °C for 15 min and cooled to room temperature (if necessary
can be diluted with dry THF). Neat azide (or a concentrated THF solution
thereof) is added dropwise. The reaction is exothermic, and for the reactive,
electron-poor azides (e.g., entries c and g, Table 1) over within 1 h as a
result of self-heating or after 24 h in any case. Since there are no significant
side reactions, elevated temperature (50 °C) can be applied in most cases,
whereupon reactions are complete within 1 h. After quenching with aqueous
NH4Cl, the products are extracted using ethyl acetate or chloroform.
Provided the azide-acetylide couple is reactive and there is no intention of
in situ derivatization of the 4-magnesio-triazole, the reactions can even be
performed on 2 mmol or larger scale with no special precautions concerning
moisture or oxygen exclusion, in screw-capped vials. In the case of less
reactive (electron-rich aromatic) azides and/or when further functionalization
of the intermediate 4-halomagnesio triazoles is intended, maintaining
complete dryness is crucial.
a Yields after recrystallization. b Yield taken from the original paper.8a
c Yields after flash chromatography; substantial amounts of 1,5-diphenyl-
1,2,3-triazole were recovered. d Yield after trituration with Et2O.
Org. Lett., Vol. 6, No. 8, 2004
1239

byproducts is eliminated by using chloromagnesium Grignard
reagents to generate the magnesium acetylide reactant.
Of course, the 4-halomagnesiotriazole intermediates, in-
stead of being hydrolyzed, can be trapped with electrophiles
other than protons,14 resulting in a regioselective formation
of 1,4,5-trisubstituted 1,2,3-triazoles (Table 2). This option,
noted previously but unexplored,8a becomes more attractive
because of the high yields of the 4-magnesiotriazole inter-
mediates. Table 2 highlights the results of quenching of the
4-chloromagnesio-1,5-diphenyl-1,2,3-triazole with nine dif-
ferent electrophiles.15 The yields are usually good, but in
cases where trace acid impurities are hard to avoid (e.g., entry
9), the product of protonation at C-4 is also formed.
Not all electrophilic compounds are suitable for this
capture process. For example, the use of sulfamoyl and
sulfonyl chlorides results in partial chlorination of the triazole
ring at C-4. Although iodination with elemental iodine is an
efficient process, the use of bromine promotes the formation
of oxidatively coupled 4,4′-bis-triazoles of type iii in
significant amounts, along with the 4-bromotriazole ana-
logues. Because of the very strong basicity of 4-magnesio-
triazoles, reactions with electrophiles, which also possess
acidic C-H bonds usually fail as a result of competing
protonation at C-4. Many of these drawbacks can be
overcome by transmetalation prior to the coupling.3d
(11) This particular compound was synthesized at the concentration of
2.0 M at room temperature in 4 h. The reaction times for less active azides
(e.g., p-methoxyphenyl azide) can be shortened to about 1 h at 50 °C.
(12) Similar observations were made in the copper(I)-catalyzed reaction,
described in ref 9.
In summary, upon reexamination, the old process of
Akimova et al.8 is much better than was thought and should
find many uses in the rapidly growing field of 1,2,3-triazole
chemistry.
(13) Structures tentatively assigned by mass spectrometry.
(14) However, it is important to note that labeling of the 4-position with
the other isotopes of hydrogen, namely, deuterium (98% D incorporation,
Table 2, entry 1) and tritium can be very useful for mechanistic and
biological studies. Once attached to the carbon at the 4-position of the 1,5-
disubstituted triazoles, the hydrogen atom is very difficult to remove
(Begtrup, M., personal communication; ref 3b suggests that 5-lithiation of
triazoles is strongly favored over 4-lithiation, even in the absence of a
directing group at the 1-position).
(15) Reactions between phenyl azide and chloromagnesium phenylacetyl-
ide were performed on 6 mmol scale at 50 °C for 1 h, and then the
electrophile (neat or as a concentrated THF solution) was added. For
experimental procedure see ref 10.
Acknowledgment. We thank the National Institute of
General Medical Sciences, the National Institutes of Health
(GM-28384), National Science Foundation (CHE-9985553),
and the W. M. Keck Foundation for financial support.
OL0499203
1240
Org. Lett., Vol. 6, No. 8, 2004