Apparent copper(II)-accelerated azide - alkyne cycloaddition

Article abstract of DOI:10.1021/ol9021113

Cu(II) salts accelerate azide-alkyne cycloaddition reactions in alcoholic solvents without reductants such as sodium ascorbate. Spectroscopic observations suggest that Cu(II) undergoes reduction to catalytic Cu(I) species via either alcohol oxidation or a

Full text of DOI:10.1021/ol9021113

ORGANIC
LETTERS
2009
Vol. 11, No. 21
4954-4957
Apparent Copper(II)-Accelerated
Azide-Alkyne Cycloaddition
Wendy S. Brotherton, Heather A. Michaels,† J. Tyler Simmons,† Ronald J. Clark,
Naresh S. Dalal, and Lei Zhu*
Department of Chemistry and Biochemistry, Florida State UniVersity, Tallahassee,
Florida 32306-4390
lzhu@chem.fsu.edu
Received September 11, 2009
ABSTRACT
Cu(II) salts accelerate azide-alkyne cycloaddition reactions in alcoholic solvents without reductants such as sodium ascorbate. Spectroscopic
observations suggest that Cu(II) undergoes reduction to catalytic Cu(I) species via either alcohol oxidation or alkyne homocoupling, or both,
during an induction period. The reactions involving 2-picolylazide are likely facilitated by its chelation to Cu(II). The highly exothermic reaction
between 2-picolylazide and propargyl alcohol completes within 1-2 min in the presence of as low as 1 mol % Cu(OAc)2.
The discovery of the Cu(I)-catalyzed azide-alkyne cycload-
dition (AAC) reaction by groups of Meldal1 and Sharpless2
has accelerated advances in areas ranging from drug dis-
covery3 to material sciences.4 The most common experi-
mental procedure involves the in situ generation of Cu(I)
by reducing CuSO4 with sodium ascorbate in aqueous
media.2 The physiological compatibility of this procedure
is imperative for its success in bioconjugation applications.
On the other hand, the preparative syntheses of 1,4-
substituted-1,2,3-triazoles outside of the bioconjugation realm
could be more efficient without additives such as sodium
ascorbate. A number of procedures have been developed
including the use of (1) Cu turning/CuSO4 combinations,5
(2) Cu(I) salts1 directly or as carbene complexes,6 (3)
electrochemically generated Cu(I),7 (4) Cu(I) on immobile
phases,8 and (5) copper-containing nanoparticles.9 These
methodological advances10 have addressed challenges posed
by specific substrate classes (e.g., reduction-prone biomol-
ecules)7 or other special needs (e.g., catalyst recycling)
involved in respective projects.
In our investigations of triazole-containing metal coordina-
tion ligands,11 we started to develop procedures for the AAC
reaction without reducing additives to simplify the isolation
of highly polar polyaza compounds. On the basis of the
observation by Yamamoto et al. that Cu(II) can be reduced
(6) Die´z-Gonza´lez, S.; Nolan, S. P. Angew. Chem., Int. Ed. 2008, 47,
8881–8884.
(7) Hong, V.; Udit, A. K.; Evans, R. A.; Finn, M. G. ChemBioChem
2008, 9, 1481–1486.
(8) (a) Chassaing, S.; Kumarraja, M.; Sido, A. S. S.; Pale, P.; Sommer,
J. Org. Lett. 2007, 9, 883–886. (b) Girard, C.; o¨nen, E.; Aufort, M.;
Beauvie`re, S.; Samson, E.; Herscovici, J. Org. Lett. 2006, 8, 1689–1692.
(c) Lipshutz, B. H.; Taft, B. R. Angew. Chem., Int. Ed. 2006, 45, 8235–
8238. (d) Chtchigrovsky, M.; Primo, A.; Gonzalez, P.; Molvinger, K.;
Robitzer, M.; Quignard, F.; Taran, F. Angew. Chem., Int. Ed. 2009, 48,
5916–5920.
† These two authors contributed equally.
(1) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002,
67, 3057–3064.
(2) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596–2599.
(3) Kolb, H. C.; Sharpless, K. B. Drug DiscoVery Today 2003, 8, 1128–
1137.
(9) (a) Pacho´n, L. D.; van Maarseveen, J. H.; Rothenberg, G. AdV. Synth.
Catal. 2005, 347, 811–815. (b) Molteni, G.; Bianchi, C. L.; Marinoni, G.;
Santo, N.; Ponti, A. New J. Chem. 2006, 30, 1137–1139. (c) Park, I. S.;
Kwon, M. S.; Kim, Y.; Lee, J. S.; Park, J. Org. Lett. 2008, 10, 497–500.
(10) For a relatively complete compilation of procedures, see:Meldal,
M.; Tornøe, C. W. Chem. ReV. 2008, 108, 2952–3015.
(4) Lutz, J.-F. Angew. Chem., Int. Ed. 2007, 46, 1018–1025.
(5) (a) Brik, A.; Muldoon, J.; Lin, Y.-C.; Elder, J. H.; Goodsell, D. S.;
Olson, A. J.; Fokin, V. V.; Sharpless, K. B.; Wong, C.-H. ChemBioChem
2003, 4, 1246–1248. (b) Deiters, A.; Cropp, T. A.; Mukherji, M.; Chin,
J. W.; Anderson, J. C.; Schultz, P. G. J. Am. Chem. Soc. 2003, 125, 11782–
11783.
(11) Huang, S.; Clark, R. J.; Zhu, L. Org. Lett. 2007, 9, 4999–5002.
10.1021/ol9021113 CCC: $40.75
Published on Web 10/07/2009
 2009 American Chemical Society

to Cu(I) in iPrOH,12 we carried out the AAC reaction
between 2-picolylazide (1) and phenylacetylene (2) in
tBuOH, iPrOH, EtOH, and MeOH (Table 1, entries 1-4)
with 5 mol % CuCl2 loading. The reactions in iPrOH and
EtOH proceeded, albeit sluggishly. The isolated yield reached
79% in MeOH after 18 h. In tBuOH, the reaction failed to
afford an identifiable amount of 3 by thin-layer chromatog-
raphy (TLC). The fact that only readily oxidizable alcohols
(MeOH, EtOH, and iPrOH) allowed the formation of a
detectable amount of the product supported the hypothesis
that a catalytic Cu(I) species was generated via the reduction
of CuCl2 (Scheme 1A).
the catalytic Cu(I) species that allows the AAC reaction to
proceed (Scheme 1B) in tBuOH. The Glaser induction period
to generate Cu(I) species was observed by Mizuno et al. in
their catalytic AAC processes using silicotungstate- and
alumina-supported Cu(II).16
Scheme 1
.
Two Processes by Which Catalytic Cu(I) Species
Might Have Been Generated
Table 1. Effects of Copper Source and Solvent on Reaction
Yield After 18 h at rta
It should be noted that the “catalytic” effect of Cu(OAc)2
on AAC reactions was reported by Kantam et al. in 2006.17
In their procedure, satisfactory yields were achieved with
20 mol % catalyst loading in aqueous solutions for 20 h.
Kantam et al. postulated a direct participation of Cu(II) in
the catalysis. We, on the other hand, hypothesize that under
the reported conditions at least two processes, alcohol
oxidation and homocoupling of terminal alkyne by Cu(OAc)2
(Scheme 1), may generate the needed Cu(I) to complete the
catalytic cycle of the AAC reaction proposed and substanti-
ated by Sharpless, Meldal, and others.10,18 It is conceivable
that the efficiencies of these “autoreduction”19 processes are
highly condition-dependent. For instance, in the presence of
ligands that favor the +1 oxidation state of copper, the
reduction of Cu(II) is known to proceed in a deceptively
effortless manner.20
entry
copper source
copper loading
solvent
yield
1
2
3
4
5
6
7
8
9
CuCl2
CuCl2
CuCl2
CuCl2
CuSO4
CuSO4
Cu(OAc)2
Cu(OAc)2
CuCl2 + NaOAc
5 mol %
5 mol %
5 mol %
5 mol %
5 mol %
5 mol %
5 mol %
5 mol %
5 mol %
tBuOH
iPrOH
EtOH
MeOH
MeOH
tBuOH
MeOH
tBuOH
tBuOH
NDb
4%
17%
79%
81%
6%
90%
>95%
88%
a 0.2-0.25 mmol of 1 and 0.3 mmol of 2 in 0.5 mL of solvent. b ND:
Not detected by thin-layer chromatography (TLC).
On the basis of the postulate that either processes A or B
or both may afford the catalytic Cu(I) species, two reaction
conditions that favor either process were selected to evaluate
various azide and alkyne substrates. Condition A entails 5
mol % CuSO4 in MeOH where Cu(II) reduction by MeOH
is presumably the major process to generate the Cu(I) species.
The effect of counterion was studied using three Cu(II)
salts, CuCl2 (entry 4), CuSO4 (entry 5), and Cu(OAc)2 (entry
7), in MeOH. All three counterions gave satisfactory yields
after 18 h. However, unlike CuSO4 and CuCl2, Cu(OAc)2
enabled a highly efficient reaction in tBuOH (>95%, entry
8). To eliminate the possibility that Cu(OAc)2 may have been
contaminated by a miniscule amount of Cu(I) species, the
reaction was run using 5 mol % of CuCl2 (99.999% pure),
which is inactive in tBuOH (Table 1, entry 1), and NaOAc
each in tBuOH (entry 9). An 88% yield was achieved. This
observation forced us to formulate an alternative hypothesis
accounting for the reaction in tBuOH, which is not prone to
oxidation (CuSO4 enabled a 6% yield in tBuOH, entry 6).13
It is known that terminal alkynes may undergo Cu(II)-
catalyzed oxidative homocouping reactions to afford diynes
(the Glaser reaction).14 In the classical Eglinton protocol,15
Cu(OAc)2 is the most active Cu(II) agent. We hypothesize
that a Glaser-type reaction may be taking place to produce
(16) (a) Kamata, K.; Nakagawa, Y.; Yamaguchi, K.; Mizuno, N. J. Am.
Chem. Soc. 2008, 130, 15304–15310. (b) Katayama, T.; Kamata, K.;
Yamaguchi, K.; Mizuno, N. ChemSusChem 2009, 2, 59–62.
(17) Reddy, K. R.; Rajgopal, K.; Kantam, M. L. Synlett 2006, 957–
959. For other examples appearing later where Cu(II) was postulated as
the catalytic species in AAC reactions, see:(a) Fukuzawa, S.-I.; Shimizu,
E.; Kikuchi, S. Synlett 2007, 2436–2438. (b) Reddy, K. R.; Rajgopal, K.;
Kantam, M. L. Catal. Lett. 2007, 114, 36–40. (c) Song, Y.-J.; Yoo, C.;
Hong, J.-T.; Kim, S.-J.; Son, S. U.; Jang, H.-Y. Bull. Korean Chem. Soc.
2008, 29, 1561–1564. (d) Namitharan, K.; Kumaaraja, M.; Pitchumani, K.
Chem.sEur. J. 2009, 15, 2755–2758.
(18) (a) Himo, F.; Lovell, T.; Hilgraf, R.; Rostovtsev, V. V.; Noodleman,
L.; Sharpless, K. B.; Fokin, V. V. J. Am. Chem. Soc. 2005, 127, 210–216.
(b) Rodionov, V. O.; Fokin, V. V.; Finn, M. G. Angew. Chem., Int. Ed.
2005, 44, 2210–2215. (c) Straub, B. F. Chem. Commun. 2007, 3868–3870.
(d) Nolte, C.; Mayer, P.; Straub, B. F. Angew. Chem., Int. Ed. 2007, 46,
2101–2103. (e) Ahlquist, M.; Fokin, V. V. Organometallics 2007, 26, 4380–
4391
(19) Kitagawa, S.; Munakata, M.; Higashie, A. Inorg. Chim. Acta 1984,
84, 79–84.
.
(12) Kirai, N.; Yamamoto, Y. Eur. J. Org. Chem. 2009, 1864–1867.
(13) The same reaction was observed to proceed to a small extent in
CH3CN. A comprehensive solvent scope study is planned in the future.
(14) Siemsen, P.; Livingston, R. C.; Diederich, F. Angew. Chem., Int.
Ed. 2000, 39, 2632–2657.
(20) (a) Musker, W. K.; Olmstead, M. M.; Kessler, R. M. Inorg. Chem.
1984, 23, 1764–1768. (b) Benzekri, A.; Cartier, C.; Latour, J.-M.; Limosin,
D.; Rey, P.; Verdaguer, M. Inorg. Chim. Acta 1996, 252, 413–420. (c)
Malachowski, M. R.; Adams, M.; Elia, N.; Rheingold, A. L.; Kelly, R. S.
J. Chem. Soc., Dalton Trans. 1999, 2177–2182. (d) Sarkar, S.; Patra, A.;
Drew, M. G. B.; Zangrando, E.; Chattopadhyay, P. Polyhedron 2009, 28,
1–6.
(15) Eglinton, G.; Galbraith, A. R. Chem. Ind. (London) 1956, 737–
738.
Org. Lett., Vol. 11, No. 21, 2009
4955

An amount of 5 mol % Cu(OAc)2 in tBuOH completes
Condition B where the Glaser reaction may result in the
reduction of Cu(OAc)2. The results after 18 h reaction time
are summarized in Table 2.
discussion on the structure and bonding of [Cu2(1)2Cl4] will
appear in a later report.
Table 2. Triazole Products Generated under Conditions A and
Ba f
,
Figure 1. Single-crystal structure of [Cu2(1)2Cl4]. Selected bond
lengths (Å): Cu1-N1 ) 2.027; Cu1-N2 ) 2.033; Cu1-Cl1 )
2.232; Cu1-Cl2a ) 2.266; Cu1-Cl2 ) 2.718; N2-N3 ) 1.256;
N3-N4 ) 1.115.
Table 3. Triazole Products Generated under Condition Ca d
,
a A: CuSO4, 5 mol % in MeOH, B: Cu(OAc)2, 5 mol % in tBuOH.
Other conditions: 0.2 mmol azide (0.6 mmol for entry 4), 0.22-0.3 mmol
alkyne, rt, 18 h. b 36% of monotriazole product (7b) was isolated. c py )
2-pyridylmethyl. d Reaction was run for 40 h. e ND: Not detected. f Azide
and alkyne components are coded blue and red, respectively.
a C: 0.2 mmol azide and alkyne in 0.5 mL of MeOH with 1 mol %
Cu(OAc)2 at rt for 18 h. b py ) 2-pyridylmethyl. c Reaction was run for
40 h. d Azide and alkyne components are coded blue and red, respectively.
Overall, the reaction yields were higher under Condition
B than those under Condition A, which suggests that
Cu(OAc)2 is a very potent catalytic Cu(I) source. 2-Picoly-
lazide (1, Table 2, entries 1-4) and 2-azidomethylquinoline
(entries 5-9) are superb substrates under these conditions.
The facile AAC reactions involving 1 under the normal
Cu(I)-catalyzed conditions was also observed by Komˇ rlj et
al.21 Other azide substrates are less ideal (entries 10-13).
The crystal structure of the complex between 1 and CuCl2
(Figure 1) revealed the association between Cu(II) and the
alkylated nitrogen of the azido group. This coordination mode
renders the azido group highly electrophilic, which likely
enables a facile reaction with the acetylide which is expected
to bind at the same copper center in the transition state
structure. The other example of a Cu(II) complex with an
organic azide that we are aware of also features coordination
of the alkylated nitrogen and a [Cu2Cl2] core.22 The in-depth
Di(2-picolyl)propargylamine was a relatively poor sub-
strate under Conditions A and B (data not shown). However,
when the reactions shown in Table 3 were run in MeOH in
the presence of 1 mol % of Cu(OAc)2 (Condition C), high
yields were achieved. The isolation of the highly polar
triazole products requires merely a short alumina column
without aqueous workup. Most compounds listed in Tables
2 and 3 are multidentate ligands for Cu(II), which is evident
in the crystal structures of the Cu(II) complexes of 18 (Figure
2A) and 3 (Figure 2B), where N3 and N2 of the 1,2,3-triazole
ring are coordinating, respectively. Cu(II) in both structures
displays the typical square planar geometry with distant
ligand(s) at axial position(s). Despite the affinity to Cu(II),
product inhibition in the AAC reactions was not observed.
(22) Barz, M.; Herdtweck, E.; Thiel, W. R. Angew. Chem., Int. Ed. 1998,
37, 2262–2265.
(21) Urankar, D.; Komˇ rlj, J. J. Comb. Chem. 2008, 10, 981–985.
4956
Org. Lett., Vol. 11, No. 21, 2009

Figure 3. (A) Absorption spectra of the mixture of 1 (1.0 mmol)
and propargyl alcohol (1.5 mmol) in the presence of 5 mol %
Cu(OAc)2 in tBuOH (2.5 mL) during the course of the reaction
over 4 min. Blue and orange traces were collected at the inception
of and after the reaction, respectively. Large absorbance values were
due to the heterogeneity of the samples. (B) EPR spectra of
Cu(OAc)2 in tBuOH at room temperature. Black: Cu(OAc)2 alone.
Blue: Cu(OAc)2 and propargyl alcohol. Orange: Mixture of
Cu(OAc)2, propargyl alcohol, and 1 after the reaction is completed.
Figure 2.
Single crystal structures of (A) [Cu(18)Cl]+ and (B)
[Cu(3)2(CH3CN)(ClO4)]+. Selected bond lengths (Å): (A) N1-Cu1
) 1.993; N2-Cu1 ) 1.997; N3-Cu1 ) 2.067; N4-Cu1 ) 2.321;
Cl1-Cu1 ) 2.238. (B) N2-Cu1 ) 2.007; N4-Cu1 ) 2.057;
N22-Cu1 ) 2.003; N24-Cu1 ) 2.056; N41-Cu1 ) 2.341;
O4-Cu1 ) 2.675.
In summary, we find that Cu(II) salts, especially Cu(OAc)2,
accelerate the AAC reactions without deliberately adding
reducing agents. 2-Picolylazide (1) and 2-azidomethylquino-
line are superb substrates for the AAC reactions under the
discovered conditions. The high reactivity of these two azides
is attributed to their abilities to chelate the alkylated azido
nitrogen, therefore facilitating the AAC reactions both
electronically and sterically. At a low catalyst loading (1 mol
%), pyridyl-containing multidentate ligands for Cu(II) were
prepared with high yields and no product inhibition. The
reaction between 1 and propargyl alcohol in the presence of
1-5 mol % Cu(OAc)2 in tBuOH completes within 60-120
s, which to our knowledge is the fastest AAC reaction under
ambient temperature reported thus far. Both absorption and
EPR data suggest that catalytic Cu(I) species is generated
in an induction period. Further mechanistic studies including
solvent and substrate scopes and capture of intermediate
species will be carried out in the near future.
All reactions listed in Tables 2 and 3 were analyzed after
18 h reaction time. Some may have taken much less time to
complete. For example, the reaction between 1 and propargyl
alcohol in the presence of 1-5 mol % Cu(OAc)2 in tBuOH
is over within 60-120 s with high exothermicity. The
reaction displays dramatic change of color. After initial
mixing, the blue tint of the reaction mixture persists until it
abruptly disappears. In the ensuing few seconds, the reaction
concludes with a spurt of heat release, and the color turns
bright yellow. The reaction progress was followed with
absorption spectroscopy (Figure 3A). The d-d band of Cu(II)
is unaltered until the end of the induction period. The
disappearance of the d-d band at the 65th second, which
suggests the reduction of Cu(II), accompanies the reaction.
More direct evidence of the reduction of Cu(II) was
obtained by the EPR measurements since Cu(II) exhibits a
characteristic EPR signal at a g-value of about 2.2, while
Cu(I) is EPR-inactive and organic radicals have a g-value
around 2.0. Cu(OAc)2 in tBuOH in the presence and absence
of propargyl alcohol yielded similarly strong EPR signals
that could be assigned to Cu(II) (black and blue curves in
Figure 3B). The spectra are rhombic, with gZ ) 2.1933, gY
) 2.1306, and gX ) 2.0732 (Figure S3, Supporting Informa-
tion). After the reaction was completed following the addition
of compound 1, the EPR signal (orange curve) was greatly
attenuated, coinciding with the color change. Both the
absorption and the EPR data are consistent with the
hypothesis that Cu(II) is converted to Cu(I) which enables
the rapid reaction between 1 and propargyl alcohol.
Acknowledgment. This work was supported by FSU and
National Science Foundation (CHE 0809201). We thank Jean
Chamoun at the Institute of Molecular Biophysics (FSU) for
assistance in EPR measurements.
Supporting Information Available: Experimental pro-
cedures, characterization of new compounds, photograph of
TLC plate for solvent screening, and .cif files for
[Cu2(1)2Cl4], [Cu(18)Cl](ClO4), and [Cu(3)2(CH3CN)(ClO4)]
(ClO4). This material is available free of charge via the
Internet at http://pubs.acs.org.
OL9021113
Org. Lett., Vol. 11, No. 21, 2009
4957