Ligand-assisted, copper(II) acetate-accelerated azide-alkyne cycloaddition

Article abstract of DOI:10.1002/asia.201100426

Polytriazole ligands such as the widely used tris[(1-benzyl-1H-1,2,3- triazol-4-yl)methyl]amine (TBTA), are shown to assist copper(II) acetate-mediated azide-alkyne cycloaddition (AAC) reactions that involve nonchelating azides. Tris(2-{4-[(dimethylamino)

Full text of DOI:10.1002/asia.201100426

DOI: 10.1002/asia.201100426
Ligand-Assisted, Copper(II) Acetate-Accelerated Azide–Alkyne
Cycloaddition
Heather A. Michaels and Lei Zhu*[a]
On the occasion of the 10th anniversary of click chemistry
Abstract: Polytriazole ligands such as
ed conditions. The copper(II) acetate-
mediated formation of the three tria-
zolyl groups in a tris(triazolyl)-based
ligand occurs sequentially with an in-
hibitory effect in the last step. The ki-
netic investigations of the ligand-assist-
ed reactions reveal an interesting
mechanistic dependence on the relative
affinity of azide and alkyne to copper
(II). In addition to expanding the scope
the widely used tris[(1-benzyl-1H-1,2,3-
triazol-4-yl)methyl]amine (TBTA), are
shown to assist copper(II) acetate-
mediated azide–alkyne cycloaddition
(AAC) reactions that involve nonche-
lating azides. Tris(2-{4-[(dimethylami-
no)methyl]-1H-1,2,3-traizol-1-yl}ethyl)-
amine (DTEA) outperforms TBTA in
a number of reactions. The satisfactory
solubility of DTEA in a wide range of
polar and nonpolar solvents, including
water and toluene, renders it advanta-
geous under copper(II) acetate-mediat-
of the copper(II) acetate-mediated
AAC reactions to include nonchelating
azides, this work offers evidence for
the mechanistic synergy between the
title reaction and the alkyne oxidative
homocoupling reaction. The elucida-
tion of the structural details of the pol-
ytriazole-ligand-bound reactive species
in copperACTHNUTRGNE(NUG I/II)-mediated AAC reac-
tions, however, awaits further charac-
terization of the metal coordination
chemistry of polytriazole ligands.
Keywords: alkynes · azides · click
chemistry · copper acetate · cyclo-
addition
Introduction
elevated temperatures, thus making it ideal for conditionally
demanding biological applications.[5]
The copper(I)-catalyzed variant of Huisgen cycloaddition of
azide and alkyne (CuAAC) is an efficient method for pre-
paring a covalent linkage, a 1,2,3-triazole, between two mo-
lecular units.[1,2] Contrary to the thermal cycloaddition, the
copper(I)-catalyzed variant affords rapid formation of regio-
specific 1,4-disubstituted 1,2,3-triazoles. With its high yields,
wide substrate scope, simple reaction and purification condi-
tions, CuAAC is considered a “click reaction” as defined by
Kolb, Finn, and Sharpless a decade ago.[3,4] Due to a sub-
stantially lowered energy of activation compared to the ther-
mal reaction, the copper(I)-catalyzed reaction proceeds
under very mild conditions, typically without the need for
Our group reported the use of copper(II) acetate (Cu-
AHCTUNGRTEGNUN(N OAc)2) to promote CuAAC reactions in the absence of an
added reducing agent, for example, sodium ascorbate.[6] We
hypothesized that the catalytic copper(I) species is generat-
ed upon reduction of copper(II) by either an oxidizable sol-
vent such as methanol[7] or the terminal alkyne substrate by
means of the oxidative homocoupling reaction.[8] A subset
of the azide substrates rapidly convert to triazole products
under the CuACTHUNRGTNEUNG(OAc)2-mediated conditions. Those azides,
which are termed “chelating azides,” contain auxiliary li-
gands capable of assisting the azido group in binding the
copper centre.[9] To expand the substrate scope of the Cu-
AHCTUNGRTEG(NUNN OAc)2-mediated procedure to include nonchelating azides,
we investigated conditions that employ an accelerating
ligand.
[a] H. A. Michaels, Dr. L. Zhu
Department of Chemistry and Biochemistry
The Florida State University
Polytriazoles were shown to accelerate CuAAC reactions.
Tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA) is
a superior accelerating ligand.[10] The acceleratory effect of
TBTA and other polytriazole ligands has been attributed
to[11,12] 1) its basicity, which helps deprotonate the alkyne to
facilitate the formation of copper(I) acetylide; 2) its multi-
Tallahassee, FL 32306-4390 (USA)
E-mail: lzhu@chem.fsu.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100426.
Chem. Asian J. 2011, 6, 2825 – 2834
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2825

FULL PAPERS
denticity by which the ligand encapsulates and thus protects
copper(I) from oxidation by molecular oxygen[13] and pre-
vents the formation of inactive polynuclear copper(I) aggre-
gates; and 3) its coordination lability that allows fast ligand
exchange, which is beneficial to catalytic turnover. In the
past few years, several other polydentate aza-ligand frame-
works have also been found to aid CuAAC reactions.[14–19]
we investigated the acceleratory effects of other polytria-
zoles in Cu(OAc)2-mediated AAC reactions. The kinetic
studies were also carried out to provide some insights on the
mechanistic involvement of TBTA (as an example of poly-
AHCTUNGTRENNUNG
triazole ligands) under the CuACHTNUGRTENUNG(OAc)2-mediated conditions.
We recently found that TBTA also accelerates Cu
U
Results and Discussion
Our previously reported CuACHTNUTRGNENG(U OAc)2-mediated AAC condi-
N
tions[6] work exceptionally well to produce various polytria-
zole ligands, including TBTA (Scheme 1). Tripropargyla-
Scheme 1. Polytriazole ligands synthesized by utilizing the CuACHTUNRGTENN(UG OAc)2-mediated CuAAC reactions. a) CuCAHTUNGTERN(NUNG OAc)2 (5 mol%), CH3OH, RT.
2826
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2825 – 2834

Azide–Alkyne Cycloaddition
mine [Eq. (1) in Scheme 1] and tris(2-azidoethyl)amine
[Eq. (2)] serve as the tertiary amino cores and react effi-
ciently with corresponding azide and alkyne substrates, re-
spectively, in CH3OH with the addition of CuACTHUNTRGNE(NUG OAc)2. In this
manner we prepared TBTA (1) and its derivatives (2–4).
TBTA (1)[10,20] and its analogues TTMA (2)[21,22] or THPTA
(3)[23] accelerate CuAAC reactions under various conditions.
The tris(2-azidoethyl)amino core [see Eq. (2), Scheme 1] af-
fords tris(triazolyl)-based ligands 5–8 with a different coor-
dination configuration. One of the initial motivations of this
work was to compare the reactivities of these two configura-
tions in ligand-assisted CuAAC reactions.
The reactions [e.g., Eqs. (1) and (2) in Scheme 1] to pro-
duce polytriazole ligands proceed cleanly and are often ac-
companied by the release of a significant amount of heat
following an induction period (see a video of the formation
of tris(2-{4-[(dimethylamino)methyl]-1H-1,2,3-traizol-1-yl}e-
thyl)amine (6, DTEA) in the Supporting Information). Pu-
rification in most cases only requires an extractive isolation
using an aqueous solution of ethylenediaminetetraacetic
acid (EDTA), or a treatment with Chelex to remove copper-
containing species followed by trituration with either hex-
anes or diethyl ether. The polytriazole ligands were then
used as additives in the CuACTHNUTRGNE(UNG OAc)2-mediated AAC reactions.
The reaction progress of 1-azidooctane and tripropargyla-
mine to afford 4 (E in Scheme 2) was monitored by
1H NMR spectroscopy.[24] In this experiment, A and B are
defined as the tripropargylamine and 1-azidooctane sub-
strates, respectively, whereas C, D, and E are mono-, bis-,
and tris-triazole product molecules. The reaction proceeds
in a stepwise fashion (Figure 1). First, mono-triazole C
Figure 1. 1H NMR (500 MHz, CD3CN) spectra of the reaction in
Scheme 2 collected over time (from bottom to top). Components A, C,
D, and E are assigned as indicated by the arrows.
ꢀ
forms and its sharp triazolyl C5 H signal (d=7.67 ppm) can
be monitored in the aromatic region over the entire course
amine is converted to tris-triazolyl ligand in a stepwise
manner.
ꢀ
of the reaction. Next, the triazolyl C5 H of bis-triazole D
can be seen slightly downfield from that of C. The increased
The time courses following the changes of A, C, D, and E
are shown in Figure 2. An in-
duction period is observed fol-
lowed by a triazole-producing
reaction phase. Mono-triazole
C, bis-triazole D, and tris-tria-
zole E form successively. The
attempt to fit the kinetic traces
using the model in Scheme 3, in
denticity shall lead to a higher affinity of bis-triazole D to
1
copper
(I/II) than mono-triazole C. The broadened H NMR
spectroscopic signal of C5 H in D can therefore be attribut-
ed to the formation of copper(I/II) complex of D that equili-
brates on the H NMR spectroscopic timescale. Finally, tris-
triazole E emerges with an even further downfield triazolyl
ꢀ
AHCTUNGTRENNUNG
1
ꢀ
C5 H signal that is nearly level with the baseline of the
1H NMR spectrum. The successive broadening and down-
which kC, kD, and kE are defined
Scheme 3. Model used to simu-
ꢀ
field shift of the triazolyl C5 H signal is indicative of the
as the apparent second-order
rate constants for each step,
late the time courses shown in
Figures 2 and 5.
progressively tighter binding to copper
N
Scheme 2. Stepwise triazole formation en route to ligand 4 (E). R=n-octyl; a) Cu
kD, and kE are apparent second-order rate constants in each step.
ACHTUGNTRNE(UNNG OAc)2 (5 mol%), 1-azidooctane (B), CD3CN, RT. The values of kC,
Chem. Asian J. 2011, 6, 2825 – 2834
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2827

H. A. Michaels and L. Zhu
FULL PAPERS
was not successful. We resorted to simulate various kinetic
scenarios using the software Pro-Kineticist II (Applied Pho-
tophysics). The simulated kinetic traces in Figure 3A resem-
ble the experimental traces of the reaction phases in
Figure 2, in which kC, kD, and kE are 0.6, 0.6, 0.3mꢀ1 minꢀ1,
respectively. The slowest step as simulated is the conversion
of bis-triazole D to tris-triazole E (kE). Therefore, under the
CuACHTNUTRGENNG(U OAc)2-mediated conditions in this particular case, no au-
toinductive kinetics were observed, as might have been sus-
pected due to the apparent acceleratory effect of the poly-
triazoles.[15] The product inhibitory effect as manifested by
the lowest rate of the tris-triazole E formation allows the ac-
cumulation of mono- and bis-triazole intermediates C and
D, which were observed experimentally. We simulated situa-
tions in which the last step (kE) is the fastest as in an autoin-
ductive process (e.g., see Figure 3C). Very little intermediate
(C and D) accumulation would occur in these cases.
Similarly, the reaction between tris(2-azidoethyl)amine A’
and phenylacetylene B’ was monitored over time
(Scheme 4). The reaction proceeds in a stepwise fashion.
Figure 2. The concentration changes over time of tripropargylamine A
~
^
(square), mono-, bis-, and tris-triazoles C ( ), D ( ), and E (ꢁ), respec-
tively, in the reaction in Scheme 2.
1
The H NMR spectroscopy signals of bis- and tris-triazoles
D’ and E’ (Figure 4) are significantly sharper than those of
their counterparts D and E in Scheme 2, thus suggesting
that copper(II) binds to D’ and E’ less tightly than to D and
E. Compared to the reaction to afford 4 (E), a much shorter
induction period en route to 5 (E’) was observed (Figure 5).
Furthermore, the AAC reaction appears to be more effi-
cient. The shorter induction period may be attributed to two
factors: 1) the inducting oxidative homocoupling reaction is
faster because tris(2-azidoethyl)amine is more basic than tri-
propargylamine and/or phenylacetylene is less sterically hin-
dered than tripropargylamine, and 2) due to the high reac-
tivity that arises from the chelation assistance of tris(azidoe-
thyl)amine,[9] the current reaction requires less copper(I),
hence a shorter induction period for its generation, to pro-
ceed.
Simulated kinetic traces using the model in Scheme 3
reveal that the experimental traces match well with the sce-
nario of product inhibitory effect (Figure 3B). A product in-
hibition observed in the syntheses of ligands with both coor-
dination configurations (Schemes 2 and 4) suggests that al-
though the ligands may carry beneficiary effects in CuAAC
reactions as outlined in the Introduction, certain portions of
Figure 3. Simulated kinetic traces based on the model in Scheme 3 using
different sets of apparent second-order rate constants kC, kD, and kE
(listed in the top portions of the figures).
copper
gands.
ACHTUNGTREN(NGNU I/II) may form unproductive complexes with the li-
Scheme 4. Stepwise triazole formation of ligand 5 (E’). R’=phenyl; (a) CuACTHNUTRGNE(UNG OAc)2 (5 mol%), phenylacetylene (B’), CD3CN, rt. kC, kD, and kE are appar-
ent second-order rate constants in each step.
2828
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2825 – 2834

Azide–Alkyne Cycloaddition
Both aromatic (Table 1, entries 1–5) and aliphatic (en-
tries 6–10) alkyne substrates are included. 2-Picolylazide
represents chelating azides with a copperACTHNUTRGENUG(N I/II)-binding auxil-
iary, whereas benzylazide lacks the chelating ability. 1-Azi-
dooctane was chosen to represent sterically unhindered ali-
phatic azides. p-Azidophenylmethylalcohol represents aro-
matic azides.
In Table 1, the entries where TBTA does not exert a sig-
nificant acceleratory effect include reactions that involve 2-
picolylazide, N,N-diethylpropargylamine, and 3,3-dimethyl-
butyne. 2-Picolylazide reacts rapidly through the chelation-
assisted pathway. The strongly basic N,N-diethylpropargyla-
mine (entry 7) also rapidly converts in the absence of
TBTA, possibly due to the formation of a bidentate triazole
product that slows down the detrimental aggregation pro-
cess of copper(I) catalytic species. The tert-butyl group in
3,3-dimethylbutyne creates a severe steric effect that is diffi-
cult to overcome under the tested conditions. TBTA aids
the CuACTHUNTGRNE(UNG OAc)2-mediated AAC reactions of other azide/
alkyne combinations, in particular reactions that involve ar-
omatic azides. In the absence of TBTA, p-azidobenzylalco-
hol is almost inactive under CuACHTNUGRTENUNG(OAc)2-mediated conditions.
Figure 4. 1H NMR (500 MHz, CD3CN) spectra of the reaction in
Scheme 4 collected over time (from bottom to top). Components A’, C’,
D’, and E’ are assigned as indicated by the arrows.
Effect of Other Polytriazole Ligands in Cu
AAC Reactions
ACHTUNGRTEN(NUNG OAc)2-Mediated
Several reactions from the substrate screening were chosen
to assess the acceleratory effects of other polytriazole li-
gands and triethylamine (TEA) on the CuACTHNURTGNENG(U OAc)2-mediated
AAC reactions (Table 2). If the basic nature of TBTA is the
primary contributor to its acceleratory ability, TEA would
also perform well. However, in most of the reactions, TEA
performs only marginally better than the ligand-free reac-
tions.
Consistently, ligand 6 (DTEA) outperforms TBTA in the
AAC reactions in Table 2, in particular in reactions that in-
volve an aromatic azide (Table 2, entry 3). The acceleratory
effect of DTEA is tentatively attributed to its high basicity
(4 tertiary amino groups) and its ability to bind copperACHTUNGTRENNUNG(I/II)
in a five-membered ring chelation similar to the coordina-
tion mode of TBTA.
The performance of DTEA was then evaluated in various
protic and aprotic solvents by using the reaction between
aliphatic azide 1-azidooctane and phenylacetylene. Com-
plete conversions within 1 h were recorded in CH3OH,
iPrOH, tBuOH, as well as in aprotic toluene, acetonitrile,
and THF (Table 3). In dichloromethane and HEPES buffer
(pH 7.0), the reaction shows marginal conversion to product
within 1 h but complete conversion within 24 h. Hexanes
proved to be a poor solvent even over a 24 h reaction time.
We attribute the decreased reactivity in HEPES, dichloro-
methane, and hexanes to substrate/solvent incompatibility
rather than the impaired ligand efficiency, based on the fact
that the ligand is soluble in all three solvents. The satisfacto-
ry solubility of ligand 6 in nonpolar solvents such as toluene
and polar solvents such as water makes it advantageous for
Figure 5. The concentration changes over time of tris(2-azidoethyl)amine
A’, mono-, bis-, and tris-triazoles C’, D’, and E’, respectively in the reac-
tion in Scheme 4.
Substrate Scope of TBTA-Assisted, Cu
AAC Reactions
ACHTUNGRTEN(NNUG OAc)2-Accelerated
As previously shown, chelating azides display an accelerato-
ry effect in Cu(OAc)2-mediated AAC reactions. To deter-
ACHTUNGTRENNUNG
mine whether a ligand would aid reactions that involved
nonchelating azides, we screened several azides with differ-
ent alkyne partners in the presence and absence of TBTA in
tBuOH (Table 1).
the CuACTHUNTGRNE(UNG OAc)2-mediated AAC reactions in which solvent
Chem. Asian J. 2011, 6, 2825 – 2834
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2829

H. A. Michaels and L. Zhu
FULL PAPERS
Table 1. Substrate screening of the TBTA-assisted Cu
min (conversion%) are listed.[b]
(OAc)2-accelerated AAC reaction.[a] Reaction times in
ACHTUNGTRENNUNG
ated AAC reactions that in-
volve the nonchelating benzyla-
zide in CD3CN. Phenylacety-
lene and 1-ethynylcyclohexanol
are selected as the alkyne sub-
strates in two sets of measure-
ments. The kinetic orders of in-
dividual components are listed
in Table 4, along with results of
Finn and co-workers for com-
parison.
The time course of a typical
1H NMR spectroscopic experi-
ment is separated into induc-
tion and reaction phases (e.g.,
see Figure 6). In the induction
phase, no triazole product
forms, while presumably the
copper(I) catalytic species is
generated. Triazole product
emerges during the reaction
phase. The initial rate of the re-
action is considered to be the
slope of the reaction phase in
its steady state.
In the reaction between phe-
nylacetylene and benzylazide,
the kinetic order in phenylace-
tylene [x in Eq. (3)] is 1.3
(Figure 6), which is consistent
with kinetically significant de-
protonation of alkyne to form a
copper acetylide intermedi-
ate.[26] At high alkyne concen-
trations, saturation kinetics ap-
pears to emerge, which indi-
Entry
A. 35 (100%)
B. 90 (100%)
A. <5 (100%)
B. 60 (100%)
A. 90 (100%)
B. 300 (0%)
A. 60 (100%)
B. 300 (40%)
1
2
3
4
A. <10 (100%) A. <10 (100%) A. 300 (100%)
B. 30 (100%)
A. 300 (100%)
B. 300 (0%)
B. 240 (100%)
B. 300 (0%)
A. <5 (100%)
B. 300 (100%)
A. <5 (100%)
B. <5 (100%)
A. 60 (100%)
B. 300 (0%)
A. 60 (100%)
B. 300 (0%)
A. 10 (100%)
B. 20 (100%)
A. 60 (100%)
B. 165 (100%)
A. 300 (71%)
B. 300 (27%)
A. 30 (100%)
B. 60 (100%)
A. <5 (100%)
B. 10 (100%)
A. <5 (100%)
B. <5 (100%)
A. 30 (100%)
B. 300 (100%)
A. 30 (98%)
B. 300 (100%)
5
6
7
8
9
A. 90 (100%)
B. 300 (84%)
A. <5 (100%)
B. <5 (100%)
A. 15 (100%)
B. 30 (100%)
A. <5 (100%)
B. <5 (100%)
A. 300 (91%)
B. 300 (0%)
A. 15 (100%)
B. 10 (100%)
A. 300 (97%)
B. 300 (0%)
A. <5 (100%)
B. <5 (100%)
A. 10 (100%)
B. 45 (100%)
A. 15 (100%)
B. 60 (100%)
A. <5 (100%)
B. <5 (100%)
A. 300 (93%)
B. 15 (100%)
A. 15 (100%)
B. 300 (0%)
A. 120 (100%)
B. 300 (0%)
A. 15 (100%)
B. 300 (100%)
A. 120 (100%)
B. 300 (0%)
A. 300 (24%)
B. 300 (18%)
A. 300 (75%)
B. 300 (0%)
A. 300 (0%)
B. 300 (0%)
A. 300 (0%)
B. 300 (0%)
10
[a] Reaction conditions: azide (0.2 mmol), alkyne (0.22 mmol), tBuOH (0.5 mL), CuACTHNUTRGNE(UNG OAc)2 (25 mL, 0.4m in
H2O), TBTA (5 mol% of azide). [b] Reaction times were determined by monitoring the disappearance of
azide through TLC. The conversion values in the parentheses were determined based on 1H NMR spectra of
the crude products.
choice is highly dependent on the nature of the starting ma-
terials.
cates that the rate-determining step, or another kinetically
significant step, follows the formation of copper acetylide.
The induction period is progressively longer as [alkyne] de-
creases, thereby suggesting that the alkyne-requiring homo-
coupling reaction proceeds in the induction period.[27]
The kinetic order in benzylazide [y in Eq. (3)] is ꢀ0.3
(Figure 7), which indicates an inhibitory effect. Interestingly,
the inhibitory effect was also observed in the induction
period, which is prolonged with increasing [benzylazide].
This observation suggests that benzylazide is able to com-
Mechanistic Investigations of the TBTA-Assisted Cu
Accelerated AAC Reaction
ACHTUNGERTN(NUNG OAc)2-
Finn and co-workers conducted mechanistic studies on
ligand-assisted CuAAC reactions.[11,18] The reaction between
phenylacetylene and benzylazide in water shows a first-
order dependence on the copper(I)/TBTA 1:1 complex, a
zero-order dependence on azide, and a slight inhibitory
effect from alkyne [Eq. (3), x= ꢀ0.28, y=0, z=1.01]:
pete with phenylacetylene for copperACTHNUTRGNEUNG(I/II) centers, which ul-
timately slows down the copper acetylide formation in both
induction and reaction phases.
The putative 1:1 TBTA/CuACTHNUTRGNE(UNG OAc)2 complex has a fraction-
d½triazoleꢁ
x
y
z
ð3Þ
rate ¼
¼ ½alkyneꢁ ½azideꢁ ½Cu=TBTAꢁ
al kinetic order of 0.4 (Figure 8). As we speculated upon ob-
serving product inhibition in the preparation of the tris-tria-
zole ligands, TBTA may form both active and inactive com-
dt
1
We recently developed an H NMR spectroscopic assay to
follow the progress of a homogenous Cu
(OAc)2-mediated
plexes with Cu
contributions from both TBTA and copperACTHUNGTRENNUNG
lack of a knowledge base of the coordination chemistry of
TBTA and tris(triazolyl)-based ligands in general,[13] we do
ACHTUNGTRENNUNG
AAC reaction that involved chelating azides in CD3CN.[25]
1
Herein, the H NMR spectroscopic assay is applied to inves-
tigate the kinetic involvement of TBTA in CuACTHNURGTNEUNG(OAc)2-medi-
2830
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2825 – 2834

Azide–Alkyne Cycloaddition
Table 2. Ligand effect in CuACHTNUGRTNEUNG
(OAc)2-accelerated AAC reactions.[a]
Entry
Azide
Alkyne
Ligand
t [min][b]
Conv. [%]
none
TEA
60
20
5
25
20
2.5
7
100
100
100
94
1 (TBTA)
2 (TTMA)
3 (THPTA)
6 (DTEA)
7
1
100
100
100
100
100
100
100
100
100
100
100
100
0
8
25
60
60
20
15
30
15
20
45
60
60
20
30
60
<5
20
40
60
60
60
60
60
60
60
60
none
TEA
1 (TBTA)
2 (TTMA)
3 (THPTA)
6 (DTEA)
7
2
3
4
8
none
TEA
0
1 (TBTA)
2 (TTMA)
3 (THPTA)
6 (DTEA)
7
100
100
100
100
100
100
0
34
74
87
0
8
none
TEA
1 (TBTA)
2 (TTMA)
3 (THPTA)
6 (DTEA)
7
100
48
35
8
[a] Reaction conditions: azide (0.2–0.22 mmol), alkyne (1.0–1.05 equiv), tBuOH (0.5 mL), CuACTHNUTRGNE(UNG OAc)2 (25 mL, 0.4m in H2O), ligand (5 mol%). [b] Reaction
1
times were determined by monitoring the disappearance of the azide through TLC. The conversion values were determined based on the H NMR spec-
tra of the crude products.
Table 3. Solvent effect on 6 (DTEA)-assisted CuACTHNUTRGNEUNG
(OAc)2-mediated AAC reaction.[a]
Entry
Solvent
Reaction time [h][b]
Conversion [%]
1
2
3
4
5
6
7
8
9
toluene
CH2Cl2
hexanes
acetonitrile
methanol
t-butanol
HEPES (pH 7.0)
tetrahydrofuran
isopropanol
1
24
24
1
1
1
24
1
1
100
100
<20
100
100
100
100
100
100
[a] Reaction conditions: 1-azidooctane (0.2–0.22 mmol), phenylacetylene (1.0–1.05 equiv), solvent (0.5 mL), CuACHTUNRGTNEUNG(OAc)2 (25 mL, 0.4m), ligand 6 (5 mol%)
[b] Reaction times were determined by monitoring the disappearance of the azide through TLC. The conversion values were determined based on the
1H NMR spectra of the crude products.
Chem. Asian J. 2011, 6, 2825 – 2834
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2831

H. A. Michaels and L. Zhu
FULL PAPERS
Table 4. Kinetic orders in various components of TBTA-assisted, CuACHTUNTRGNEUNG(OAc)2-mediated AAC reactions (entries 1–6). The kinetic orders determined by
Finn et al. using CuSO4/sodium ascorbate combination are shown in entries 7–9.[11]
Entry
Alkyne
Solvent
G
E
[Cu] [mm]
Varied component
alkyne
Kinetic order
1.3
1
CD3CN
0.8
2
3
–
–
CD3CN
CD3CN
16
20
8–18
20
0.8
0.2–1
azide
Cu/TBTA
ꢀ0.3
0.4
4
CD3CN
10–20
20
1
alkyne
1.5
5
6
–
–
CD3CN
CD3CN
20
20
10–20
20
1
azide
Cu/TBTA
ꢂ0.5
0.2–1
1.3
7
H2O
10–50
1
0.05
alkyne
ꢀ0.28
8
9
–
–
H2O
H2O
1
1
10–50
1
0.05
0.02–0.25
azide
Cu/TBTA
0
1.01
Figure 6. Dependence of reaction time course on [phenylacetylene] (8–
20 mm). Conditions: [benzylazide]=16 mm, [TBTA]=[Cu(OAc)2]=
0.8 mm in CD3CN, 300 K.
Figure 8. Dependence of reaction time course on [CuACTHUNRGTNEUNG(OAc)2/TBTA]
(0.2–1 mm). Conditions: [phenylacetylene]=20 mm, [benzylazide]=
20 mm in CD3CN, 300 K.
AHCTUNGTRENNUNG
not attempt to offer a structural hypothesis on the remark-
able acceleratory effect of TBTA on Cu
AAC reactions. On the other hand, based on the kinetic
orders of copper(I/II)/TBTA complexes determined by Finn
ACHTUNGRTEN(NUNG OAc)2-mediated
AHCTUNGTRENNUNG
et al. and us, we incline to argue that TBTA-assisted
CuAAC reactions may take a different pathway than the
ligand-free version of the reaction, which has been shown to
be second-order in copper.[25,28,29]
The kinetic orders of the reaction that involves 1-ethynyl-
cyclohexanol reveal interesting substrate-dependent kinetic
behaviors. Compared to phenylacetylene, 1-ethynylcyclohex-
anol carries a similar kinetic order of 1.5 (Figure 9). Howev-
er, the induction periods are much shorter. This observation
suggests that 1-ethynylcyclohexanol is a much better sub-
strate than phenylacetylene in alkyne oxidative homocou-
pling reactions, probably due to its propensity to interact
with copper(II).[30] For example, in the following hypothe-
sized structure, the hydroxyl-directed p-complex formation
aids the deprotonation of the ethynyl group and the subse-
Figure 7. Dependence of reaction time course on [benzylazide] (8–
18 mm). Conditions: [phenylacetylene]=16 mm, [TBTA]=[Cu
0.8 mm in CD3CN, 300 K.
ACHTUNGRTEN(NUNG OAc)2]=
2832
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2825 – 2834

Azide–Alkyne Cycloaddition
come the inhibitory effect of benzylazide observed in the
previous case that involves phenylacetylene. Consequently,
the participation of benzylazide becomes kinetically signifi-
cant, as a roughly 0.5 positive kinetic order was observed.
CuACTHNUTRGEN(UGN OAc)2/TBTA complex shows a kinetic order of 1.3
(Figure 11), which is drastically larger than the previous
case that involves phenylacetylene. Due to the high rate of
ligand-free, CuACTHNUTRGNEUNG(OAc)2-mediated AAC reaction that involves
1-ethynylcyclohexanol (Table 2, entry 2), the second order in
copper of the ligand-free pathway, which is the background
reaction, may be factored into the ligand-assisted kinetic
order.[31]
Conclusion
Tris(triazolyl)-based ligands greatly facilitate CuACHTUNGTRENNUNG(OAc)2-
mediated AAC reactions that involve nonchelating azides.
Based on this study, the known ligand TBTA, which has
been applied extensively in assisting typical copper(I)-AAC
reactions, and a new ligand DTEA are most effective in var-
Figure 9. Dependence of reaction time course on [1-ethynylcyclohexanol]
(10–20 mm). Conditions: [benzylazide]=20 mm, [TBTA]=[Cu
1 mm in CD3CN, 300 K.
ACHTUNGRTEN(NUNG OAc)2]=
ious Cu
DTEA can be conveniently synthesized by using the Cu-
ACHTUNGRTEN(NNUG OAc)2-mediated procedure. The three triazolyl groups form
sequentially, with a product inhibition effect for the last tria-
zole formation step. The kinetic-order determination reveals
interesting substrate dependence of the reaction mechanism.
An inhibitory effect of benzylazide can be overcome when
ACHTUNGRTENN(UNG OAc)2-mediated AAC reactions. Both TBTA and
quent formation of the dimeric
copper(II) acetylide,[30] an inter-
mediate in alkyne homocou-
pling reactions (Scheme 5).[8]
The strong 1-ethynylcyclo-
hexanol/copper
is further supported by the lack
intermediate in the oxidative of azide dependence on the in-
(I/II) interaction
an alkyne with a strong copper
1-ethynylcyclohexanol. In addition to suggesting that com-
petitive interactions with copper(I/II) by alkyne and azide
lead to distinct mechanistic pathways of CuAAC reactions,
this work provides evidence for the mechanistic synergy be-
tween CuAAC and alkyne oxidative homocoupling reac-
ACHTUNGTNER(NUNG I/II) affinity is used, such as
Scheme 5. Postulated dinuclear
ACHTUNGTRENNUNG
homocoupling of 1-ethynylcy-
duction period (Figure 10),
clohexanol based on the report
thereby suggesting that 1-ethy-
by Fedenok and Shvarts-
nylcyclohexanol is able to over-
berg.[30]
Figure 10. Dependence of reaction time course on [benzylazide] (10–
Figure 11. Dependence of reaction time course on [CuACTHUNGTERNNU(G OAc)2/TBTA]
20 mm). Conditions: [1-ethynylcyclohexanol]=20 mm, [TBTA]=[Cu-
ACHTUNGTRENNUNG(OAc)2]=1 mm in CD3CN, 300 K.
(0.2–1 mm). Conditions: [1-ethynylcyclohexanol]=20 mm, [benzylazide]=
20 mm in CD3CN, 300 K.
Chem. Asian J. 2011, 6, 2825 – 2834
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2833

H. A. Michaels and L. Zhu
FULL PAPERS
[7] N. Kirai, Y. Yamamoto, Eur. J. Org. Chem. 2009, 1864.
[8] P. Siemsen, R. C. Livingston, F. Diederich, Angew. Chem. 2000, 112,
2740; Angew. Chem. Int. Ed. 2000, 39, 2632.
[9] G.-C. Kuang, H. A. Michaels, J. T. Simmons, R. J. Clark, L. Zhu, J.
Org. Chem. 2010, 75, 6540.
tions. The structural details of the involvement of TBTA
and other polytriazole ligands in CuAAC reactions, howev-
er, are awaiting advances in the coordination chemistry of
these ligands.
[10] T. R. Chan, R. Hilgraf, K. B. Sharpless, V. V. Fokin, Org. Lett. 2004,
6, 2853.
[11] V. O. Rodionov, S. I. Presolski, D. D. Dꢂaz, V. V. Fokin, M. G. Finn,
J. Am. Chem. Soc. 2007, 129, 12705.
Experimental Section
[12] X. Chen, G. N. Khairallah, R. A. J. OꢃHair, S. J. Williams, Tetrahe-
dron Lett. 2011, 52, 2750.
[13] X-ray single-crystal structures of TBTA complexes with CuCl2 and
General Procedure of CuACTHNUTRGNE(NUG OAc)2-Mediated AAC Reactions
All azides and alkynes were used “as is” from the supplier or prepared
from previously reported procedures.[32,9] Azide (0.80m) and alkyne
(0.82m) stock solutions were prepared in tBuOH (>99.5%). An aqueous
[CuACHTUNGTRNEUN(G CH3CN)4]BF4 were reported in: P. S. Donnelly, S. D. Zanatta,
S. C. Zammit, J. M. White, S. J. Williams, Chem. Commun. 2008,
2459.
stock solution of CuACHTUNGTRENNUNG(OAc)2 (0.4m) was also prepared and used within
2 w. For reactions that involved an assisting ligand, 0.01 mmol (5 mol%)
of the ligand was weighed into a reaction vial first. Azide (250 mL) and
alkyne (250 mL) stock solutions were then added to the reaction vial
along with an appropriately sized stir bar. No effort was made to exclude
[14] W. G. Lewis, F. G. Magallon, V. V. Fokin, M. G. Finn, J. Am. Chem.
Soc. 2004, 126, 9152.
[15] V. O. Rodionov, S. I. Presolski, S. Gardinier, Y.-H. Lim, M. G. Finn,
J. Am. Chem. Soc. 2007, 129, 12696.
[16] N. Candelon, D. Lastecoueres, A. K. Diallo, J. R. Aranzaes, D.
Astruc, J. Vincent, Chem. Commun. 2008, 741.
[17] S. ꢄzÅubukÅu, E. Ozkal, C. Jimeno, M. A. Pericꢅs, Org. Lett. 2009,
11, 4680.
[18] S. I. Presolski, V. Hong, S.-H. Cho, M. G. Finn, J. Am. Chem. Soc.
2010, 132, 14570.
[19] See more examples in: S. Diꢆz-Gonzꢇlez, Catal. Sci. Technol. 2011,
1, 166.
molecular oxygen. Upon injection of CuACHTNUTRGNEUNG(OAc)2 stock solution (25 mL),
the reactions were monitored with TLC. TLC was taken every 2.5 min
for the first 15 min of reaction; between 15 min to 1 h, TLC was taken
every 5 min. The frequency was reduced to every 30 min when the reac-
tion ran over 1 h. The reaction mixture was diluted by ethyl acetate
(2 mL) and was filtered through a pipette-sized column of alumina to
produce the crude product. Chromatographic purification on an alumina
column was carried out to afford the pure product, if needed.
[20] A. J. Link, D. A. Tirrell, J. Am. Chem. Soc. 2003, 125, 11164.
[21] Z. Zhou, C. J. Fahrni, J. Am. Chem. Soc. 2004, 126, 8862.
[22] N. K. Devaraj, P. H. Dinolfo, C. E. D. Chidsey, J. P. Collman, J. Am.
Chem. Soc. 2006, 128, 1794.
[23] V. Hong, S. I. Presolski, C. Ma, M. G. Finn, Angew. Chem. 2009, 121,
10063; Angew. Chem. Int. Ed. 2009, 48, 9879.
1H NMR Spectroscopic Kinetic Experiments
Stock solutions (2 mL each) of phenylacetylene (50 mm), benzylazide
(50 mm), and the 1:1 complex of Cu
fresh in CD3CN for each set of data obtained. Aliquots of azide, alkyne,
and CD3CN were first combined in a small vial. The Cu(OAc)2/TBTA
ACHTUNGERTN(NUNG OAc)2/TBTA (5 mm) were prepared
AHCTUNGTRENNUNG
catalyst was added last and the mixture was immediately transferred to
an NMR spectroscopy tube and injected into the instrument. The total
volume was 550 mL. Spectra were recorded at 300 K every 5 min. Data
was collected on the same day for each experiment.
[24] The reaction of benzylazide and tripropargylamine, which forms
TBTA, was also monitored but the signals were not as well resolved.
[25] G.-C. Kuang, P. M. Guha, W. S. Brotherton, J. T. Simmons, L. A.
Stankee, B. T. Nguyen, R. J. Clark, L. Zhu, J. Am. Chem. Soc. 2011,
DOI: 10.1021/ja203733q.
[26] F. Bohlmann, H. Schçnowsky, E. Inhoffen, G. Grau, Chem. Ber.
1964, 97, 794.
[27] TBTA is shown to assist the oxidative homocoupling of phenylacety-
lene. See Table S2 in the Supporting Information.
[28] V. O. Rodionov, V. V. Fokin, M. G. Finn, Angew. Chem. 2005, 117,
2250; Angew. Chem. Int. Ed. 2005, 44, 2210.
[29] J. E. Hein, V. V. Fokin, Chem. Soc. Rev. 2010, 39, 1302.
[30] L. G. Fedenok, M. S. Shvartsberg, Tetrahedron Lett. 2011, 52, 3776.
Acknowledgements
This work was generously supported by the National Science Foundation
(CHE-0809201). We are indebted to Dr. Mioara Larion and Professor
Brian Miller for guidance in using Pro Kineticist II software.
[31] The apparent kinetic order of Cu
ACHTUNGERTN(NUNG OAc)2 includes the contributions
of copper(I/II) to both induction and reaction phases. It should be
AHCTUNGTRENNUNG
[1] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew.
Chem. 2002, 114, 2708; Angew. Chem. Int. Ed. 2002, 41, 2596.
[2] C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002, 67,
3057.
[3] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056; Angew. Chem. Int. Ed. 2001, 40, 2004.
[4] M. G. Finn, V. V. Fokin, Chem. Soc. Rev. 2010, 39, 1231.
[5] H. C. Kolb, K. B. Sharpless, Drug Discovery Today 2003, 8, 1128.
[6] W. S. Brotherton, H. A. Michaels, J. T. Simmons, R. J. Clark, N. S.
Dalal, L. Zhu, Org.Lett. 2009, 11, 4954.
noted that copper(I) may still be produced in the reaction phase, es-
pecially in relatively slow reactions. Therefore, we are at this point
not in a position to propose a structural model to account for the ki-
netic orders of Cu
ported herein.
ACHTUNGRTENUN(NG OAc)2 in the ligand-assisted AAC reactions re-
[32] S. Huang, R. J. Clark, L. Zhu, Org. Lett. 2007, 9, 4999.
Received: May 2, 2011
Published online: August 29, 2011
2834
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2825 – 2834