Recyclable catalytic dendrimer nanoreactor for part-per-million Cu I catalysis of "click" chemistry in water

Article abstract of DOI:10.1021/ja5061388

Upon catalyst and substrate encapsulation, an amphiphilic dendrimer containing 27 triethylene glycol termini and 9 intradendritic triazole rings serves as a catalytic nanoreactor by considerably accelerating the Cu ^I-catalyzed alkyne-azide cycl

Full text of DOI:10.1021/ja5061388

Article
pubs.acs.org/JACS
Recyclable Catalytic Dendrimer Nanoreactor for Part-Per-Million CuI
Catalysis of “Click” Chemistry in Water
̈
Christophe Deraedt, Noel Pinaud, and Didier Astruc*
́
ISM, Univ. Bordeaux, 351 Cours de la Liberation, 33405 Talence, France
*
S Supporting Information
ABSTRACT: Upon catalyst and substrate encapsulation, an amphiphilic
dendrimer containing 27 triethylene glycol termini and 9 intradendritic
triazole rings serves as a catalytic nanoreactor by considerably accelerating
I
the Cu -catalyzed alkyne−azide cycloaddition (CuAAC) “click” reactions of
various substrates in water using the catalyst Cu(hexabenzyltren)Br (tren =
triaminoethylamine). Moreover this recyclable nanoreactor with intra-
dendritic triazole rings strongly also activates the simple Sharpless−Fokin
catalyst CuSO + sodium ascorbate in water under ambient conditions
4
leading to exceptional TONs up to 510 000. This fully recyclable catalytic
nanoreactor allows to considerably decrease the amount of this cheap
copper catalyst down to industrially tolerable residues, and some biomedical and cosmetic applications are exemplified.
INTRODUCTION
the order of 1%) compared to the original, simple, and practical
catalyst CuSO4 + sodium ascorbate that is still the most
commonly utilized catalyst but in much larger quantities that
■
Since Breslow’s concept of supramolecular nanoreactors with
1
cyclodextrins, nanoreactors are becoming increasingly inves-
25−27
2
are often even stoichiometric or superior to stoichiometry.
tigated in catalysis as illustrated by Fujita’s capsule M L ,
6
4
3
4
5
The large quantity of metal catalyst used in the CuAAC “click”
reaction and its difficult complete removal presently remain the
main problems, inhibiting the utilization of such “click” chemistry
in electronics and biomedicine.
Rebek’s soft ball, cucurbit[6]uril, and biresorcinarenes,
6
respectively, studied by Mock and Warmuth, porphyrin used
by Reek and Van Leuuwen, Sanders’ porphyrine macrocycle
7
8
,9
and Bergman and Raymond’s capsule. Surfactants and ionic
1
0
11
liquids, copolymers that form micelles and polymersomes,
Now the use of the dendrimer 1 terminated by 27 triethylene
1
2−29
39
and dendrimers
are useful for the encapsulation or stabili-
glycol (TEG) termini (Scheme 1a) is proposed as an amphi-
1
2−18
zation of catalytically active nanoparticles.
Newkome has proposed dendritic molecular micelles,
Crooks has used dendrimers as nanofilters for nanoparticle
In particular,
philic micellar nanoreactor, stabilizer, and activator of CuAAC
reactions in water leading to a considerable decrease of the
1
9−21
and
I
required Cu catalyst for these reactions. The water-soluble
1
2−14
catalysis.
Along this line, dendrimers have the potential
dendrimer 1 is also fully recyclable and reusable many times
without consumption or decomposition.
Two approaches are shown here to be extremely efficient for
this purpose. First, the reactions of various substrates are
2
2−24,29
to serve as molecular containers,
internal triazole rings on their tethers have been shown to
and dendrimers with
facilitate the catalysis by Pd nanoparticles of C−C cross-coupling
18,23−27,30
reactions.
catalyzed in water using the complex [hexabenzyltren-Cu]Br,
It was then of interest to examine whether these dendritic
nanoreactors are of general application in catalysis and, in
particular, if they can also facilitate or activate catalysis by mole-
cular catalysts. Therefore, we have now addressed their utility in
37
2
,
with the dendrimer 1 taking advantage of its molecular-
micelle effect. In this strategy, it is possible to catalyze click
I
reactions with down to only 0.1% of the Cu catalyst 2 and to
localize this hydrophobic solid catalyst 2 inside the dendrimer
I
the Cu -catalyzed Huisgen-type azide−alkyne 1,3-cycloaddition
1
in D O by 600 MHz H NMR, bringing an enlightening sup-
2
(
CuAAC), the most commun “click” reaction allowing to link
port for the role of 1 as a nanoreactor. Second, now utilizing
together two organic, bioorganic, or other functional molecular
3
1,32
I
both advantages of micellar dendritic encapsulation and catalytic
fragments.
The catalytically active Cu species is usually
I
activation of “naked” Cu by intradendritic triazole ligands,
generated from CuSO and sodium ascorbate in excess, a very
4
32
the recyclable dendrimer 1 in very small quantities allows the
convenient system that works well in aqueous solvents. Various
I
I
use of only part-per-million Cu catalyst (CuSO + sodium
4
genuine Cu catalysts that do not require a sacrificial reductant
I
ascorbate) at 30 °C in only water, an exceptional efficacy of this
convenient, simple, and commercial catalyst.
are also known, the advantage of liganded Cu , particularly
with nitrogen ligands, being the rate acceleration, for instance
3
3,34
with the efficient polytriazoles
and tris(2-aminoethyl)amine
The nitrogen ligands allow the use of
Cu catalysts in amounts that are much reduced (most often of
35−38
derivatives (tren).
Received: June 18, 2014
Published: August 5, 2014
I
©
2014 American Chemical Society
12092
dx.doi.org/10.1021/ja5061388 | J. Am. Chem. Soc. 2014, 136, 12092−12098

Journal of the American Chemical Society
Article
Scheme 1. CuAAC Reaction between Benzyl Azide and
Phenyl Acetylene Catalyzed by the Cu Complex 2 in the
Presence of Small Amounts of the Dendrimer 1 (a, b).
of entry 2 was repeated 10 times, with the same recycled den-
I
drimer 1 without change in reaction yield). The next-generation
39
dendrimer 3 containing a hydrophobic core and 81 TEG
termini (see structure in the Supporting Information Scheme S1)
was also used in the test CuAAC reaction between benzyl azide
and phenyl acetylene under the conditions indicated in Table 1
in order to check its behavior as a nanoreactor. The reaction
performed in the presence of 3 was nearly quantitative (90%)
after 3 h as in the presence of 1 (91%, entry 2), indicating that
the micellar effect of 1 and 3 in the catalysis of CuAAC reactions
was similar. Because the synthesis of 3 is longer than that of 1,
the following studies of the “click” reactions were only conducted
with the dendrimer 1.
The CuAAC reaction has been conducted using the nano-
reactor 1 in water for seven other substrates in order to check
the applicability and generality of the method for this reaction
(
eq 1 and Table 2)
Table 2. CuAAC Reactions between Various Azides and
Alkynes Using 0.1 Mol % of Catalyst 2 in the Presence of
Catalytic Amounts of Dendrimer 1
RESULTS AND DISCUSSION
a
■
Activation of Click Catalysis by Micellar Dendrimer
Effect: the Dendrimer 1 As a Nanoreactor. The CuAAC
reaction is first conducted in water during 3 h between the
water-insoluble substrates benzyl azide and phenyl acetylene
with various amounts of catalyst 2 in the presence of 1% mol of 1
(
Scheme 1a,b and Table 1).
Table 1. CuAAC Reactions between Benzyl Azide and
a
Phenyl Acetylene
b
entry
catalyst 2 (mol %)
dendrimer 1 (mol %)
yield (%)
2
1
2
3
4
0.1
0.1
0.2
0.5
0
1
1
1
c
91
92
98
a
From Scheme 1b. All the reactions were carried out with 0.1 mmol
of azide, 0.105 mmol of alkyne, 2 mL of water at 25 °C during 3 h.
b
c
Isolated yield. This reaction was repeated 10 times with the same
recycled dendrimer 1, and after reusing 1 ten times, the yield remained
1%.
9
a
See eq 1. All the reactions are carried out with 0.1 mmol of azide,
0
.105 mmol of alkyne, 1% mol of dendrimer 1, 0.1% mol of 2, and 2 mL
According to Table 1, the “click” reaction with 1% of 1 works
b
1
of water at 25 °C, during 3 h. Isolated yield/ H NMR conversion.
well when the amount of catalyst 2 is in the range 0.1−0.5%
mol (yields: 92−98%). When the same reaction is conducted
without 1, the reaction does not work under these conditions,
the yield being only 2% (entry 1).
The method that is proposed here is efficient for the “click”
reaction of various substrates, leading to yields of around 90%
or more for classic nonactivated substrates. Finally the catalyst
2 was used during the “autocatalytic” synthesis of the nano-
reactor 1 itself. Remarkably, with only 8% mol of 2 per branch,
the “click” reaction between the nona-core azide and the tris-
TEGylated alkynyl dendron (eq 2) is completed in 570 min at
30 °C in the presence of only 1 mol % 1 per branch, leading
to 1 in 81% isolated yield. If 1 is absent at the beginning of
the reaction, the yield under these conditions is only 39%
(in 1200 min), clearly showing the strong autocatalytic effect
of very small amounts of 1 on its own formation. The kinetic
study of the reaction (see Supporting Information, Figures S29
The water/organic compatible TEG termini render the
dendrimer 1 water-soluble, and the hydrophobic core allows
solubilization in water of the hydrophobic catalyst 2 and the
substrates. The results from Table 1 confirm that the hydro-
phobic substrates and catalyst 2 meet more easily in the hydro-
phobic core of 1 than outside 1 in water. Scheme 1a illustrates
the schematic principle of the reaction. The insolubility of 1
in diethyl ether allows the complete extraction of the organic
products from the reaction medium upon keeping 1 in the
water phase. In this way, 1 was recycled more than ten times
without change of structure during the catalysis (the experiment
1
2093
dx.doi.org/10.1021/ja5061388 | J. Am. Chem. Soc. 2014, 136, 12092−12098

Journal of the American Chemical Society
Article
experiment suggests that, when 1 is concentrated, assemblies
of dendrimers are formed with interdendritic interlocked TEG
termini, whereas in diluted conditions, these assemblies are split.
This observation is also confirmed by DOSY NMR experiments
that show different diffusion coefficients when 1 is concentrated
(
2.8 × 10− M) or diluted (8.7 × 10 M). The hydrodynamic
3
−5
diameter of concentrated 1 calculated from the diffusion co-
efficient using the Stokes−Einstein law is 13.3 (±0.2) nm, and
1
0.2 (±0.2) nm when 1 is diluted (the calculated maximum
and S30) shows that after 90 min, the conversion of the starting
material is already 45.8% in the presence of 1 against only 2%
without 1 at the beginning of the reaction. In the absence of 1
at the beginning of the reaction, its synthesis using catalyst 2
requires 2 days to reach completion; thus, the presence of 1 at
the beginning of the reaction clearly accelerates its synthesis.
diameter of 1 at full extension is approximately 6 nm).
1
The H NMR spectrum of 1 + 2 in D O presents, compared
2
to that of 1 alone, the appearance of new signals at 7.47−
7
.49 ppm (Figure 1b) corresponding to the protons of the
phenyl groups of 2, showing the solubilization in water of
the hydrophobic complex 2. It is also remarkable that in the
presence of 2 the signals of the protons of the hydrophobic
groups of the tethers of 1 (red and green regions of Scheme 1a)
are slightly shifted (0.03 ppm in the red region, see Figure S2/
Table S1 of the Supporting Information), especially the triazole
proton (around 0.1 ppm shift, see Figure 1b). This latter shift
possibly results from interaction of this triazole with the Cu
atom of 2 subsequent to reversible decoordination of a nitrogen
ligand of 2. This would eventually add to the driving force
provided by the hydrophobic shelter for the encapsulation of 2
by 1. NOESY NMR shows a weak interaction between the new
1
H NMR Characterization of the Solubilization of the
Hydrophobic Catalyst 2 in Water and Encapsulation in
the Dendrimer 1. In order to provide further insight into the
1
catalytic activation by the dendrimer 1, H NMR and DOSY
NMR studies of 1 have been conducted.
1
The H NMR study of 1 in D O shows a minute shift of
2
the triazole proton (from 7.79 to 7.76 ppm) upon diluting 1
−3
−5
3
2 times (from 2.8 × 10 to 8.7 × 10 M, Figure 1a). This
peaks of 2 and the CH substituents of the Si groups of 1,
3
which also confirms the presence of the catalyst 2 very close to
the hydrophobic interior of 1 (see Figure S4 of the Supporting
Information). The bulk of the core and the barrier of the
methyl substituents of the Si atoms prevent significantly
deeper interaction of 2 beyond the SiMe groups near the
2
dendrimer core.
Upon diluting the solution 1 + 2 from [1] = 2.8 × 10−
3
M
to [1] = 3.2 × 10− M (condition of the CuAAC reactions)
the ratio of water-soluble 2 per mol 1 increases from 1/15 to 1/5
4
(Supporting Information, Figure S3), which is in agreement with
the stoichiometry used during the catalytic CuAAC reaction with
1
% dendrimer 1 and 0.1−0.2% catalyst 2 (Table 1, entries 1−3
and Table 2, entries 5−12).
Intradendritic Triazole Ligands of 1 As Additional
I
Activators of Cu Catalysis. The second strategy uses, in
addition to and in synergy with the micellar effect of the
amphiphilic dendrimer illustrated above, the intradendritic
I
triazole activation of the Cu catalysis. In this case, it is not
I
necessary to synthesize a Cu catalyst such as 2 with a nitrogen
ligand because this role is played by the nine intradendritic
triazole ligands of 1. In precedent work on catalytically efficient
dendrimer-encapsulated Pd nanoparticles, the role of the
18,39
triazole ligands was the stabilization of the nanoparticles.
On the other hand, here, the intradendritic triazole ligands
I
activate Cu by increasing the electronic density for improved
catalytic efficiency. Thus, the classic CuSO ·5H O source can
4
2
advantageously be used as precatalyst of the CuAAC reaction,
II
I
and sodium ascorbate (NaAsc) as reducing agent of Cu to Cu ,
31
that is, the initially reported conditions. As 1 contains nine
II
triazolyl rings, 9 equiv. Cu per dendrimer are used. After
Figure 1. 1H NMR characterization of the solubilization of the
II
adding CuSO ·5H O to an aqueous solution of 1, Cu
4
2
hydrophobic catalyst 2 in water and its encapsulation in the dendrimer
II
coordinates the intradendritic triazolyl ligands, then Cu is
reduced in situ to Cu for intradendritic catalysis (Figure 2).
The coordination of Cu to the triazole rings of 1 has been
checked by H NMR spectroscopy. After adding CuSO ·5H O
1
1
[
(
. (a) 600 MHz H NMR in D O of 1 from concentrated solution
2
I
‑
3
‑5
1] = 2.8 × 10 M (left) to diluted solution [1] = 8.7 × 10 M (right).
b) 600 MHz H NMR of 1 + 2 in D O. New peaks appear around
2
II
1
1
7.5 ppm and shifts are observed, inter alia in the portion of the spectrum
4
2
represented here, for two types of protons (triazole at 7.85 ppm and
aromatic rings of the TEG dendron 6.62 ppm).
to a deuterium oxide solution of 1 (1 equiv per triazole), the
NMR signal of the triazole proton of 1 at 7.90 ppm vanishes to
1
2094
dx.doi.org/10.1021/ja5061388 | J. Am. Chem. Soc. 2014, 136, 12092−12098

Journal of the American Chemical Society
Article
Figure 2. (a) Dendrimer 1 as nanoreactor/ligand for CuAAC catalysis. (b) Comparison of the NMR signals of the triazole proton of 1 alone
II
I
(
1
7.90 ppm), with Cu (very broad due to the paramagnetism) and Cu (shift to 8.08 ppm) showing the coordination of the intradendritic triazoles of
to the copper ions.
II
give a very broad signal due to the paramagnetic Cu species.
The CuAAC reaction is also carried out in water at 30 °C
II
I
When NaAsc is added to reduce Cu to Cu , the NMR signal
of the triazole proton of 1 reappears but is shifted (8.08 ppm
instead of 7.90 ppm when 1 is alone) showing the coordination
between various alkynes and azides (aromatic and aliphatic)
I
with parts-per-million amounts of Cu (eq 3 and Table 4)
I
of all the triazole rings to Cu (Figure 2b and Figure S8 of the
40
Supporting Information).
The efficiency of the CuAAC catalyst involving this system
has been tested again for the classic reaction between benzyl
azide and phenyl acetylene with various amounts of catalyst,
down to 1 ppm of copper. When the quantities of 1 and
In order to check if the dendritic effect is positive and to
distinguish between the micellar dendrimer effect and the CuI
activation by coordination of a nitrogen atom of the “clicked”I
triazole in water, the results obtained in the presence of the Cu
complex of dendrimer 1 were compared on one hand with those
CuSO ·5H O are 1%, the yield is 100% in 2.5 h of reaction
4
2
against only 15% when the reaction is conducted without 1.
I
The reaction is quantitative with only 4 ppm of Cu in water at
3
1
0 °C during 24 h (entry 18) and reaches 50% of yield with
I
I
ppm of Cu (Table 3).
obtained with the Cu complex of the nondendritic molecule 4
(
Scheme 2a), and on the other hand with those obtained with
I
41
Table 3. CuAAC Reactions between Benzyl Azide and
Phenyl Acetylene Using Various Amounts of 1−Cu
Cu in the presence of the water-soluble dendrimer 5 that does
not contain triazole ligands (Scheme 2b).
I
a
I
With Cu stabilized by the nondendritic triazole ligand of 4,
I
TOF (h−1
)
entry
Cu ppm
time (h) yield (%)
TON
99
I
the same conditions as in entry 15 (200 ppm of Cu ) are
b
13
14
15
16
17
18
19
20
10000 (1 mol %)
2.5
19
99
99
99
99
99
99
50
0
39.6
26.0
followed, and the yield is also quantitative. When the amount of
catalyst is reduced to 4 ppm as in entry 18, however, no reaction
is observed upon several attempts, which emphasizes the key
molecular micelle role of the dendritic nanostructure. With
dendrimer 5 that does not contain triazole rings (Scheme 2b),
Cu is introduced under the same conditions as with 1 and as
in entry 17 (20 ppm of Cu ). After 24 h, only 27% of isolated
yield is obtained, whereas with 1−Cu 99% is obtained in 19 h.
This shows the crucial activation role of the triazole ring. When
the reaction is performed without any dendrimer (using only
b
2000
200
40
20
4
495
4 950
19
19
19
24
24
24
260
24 700
49 500
247 000
510 000
0
1 300
2 600
10 300
21 200
0
I
1
I
0
I
a
All the reactions were carried out with 1 mmol of azide, 1.05 mmol of
b
alkyne in 1 mL of H O. The reaction was carried out with 5 mL of water.
2
I
the water solution of Cu ), the isolated yield is 9%. These
Such an extremely low quantity of copper has never been
successfully used in only water at 30 °C for CuAAC reactions
before the present study. At such extremely low catalyst amounts,
the noncatalyzed Huisgens reaction is in competition with the
CuAAC reaction at high temperature using the standard
experiments show that both the micellar effect of the dendrimer
1 and the activation by coordination of Cu by the intradendritic
I
triazole ligands have a very positive effect on the catalytic
efficiency and that these two effects are cumulative, resulting in
the considerable benefit of the dendritic nanoreactor 1 on the
“click” reactions with the simple Sharpless-Fokin catalyst
CuSO ·5H O + sodium ascorbate in water.
3
7
substrates. At 30 °C, however, neither the Huisgen reaction
I
in the absence of Cu catalyst nor the CuAAC reaction (entry 20)
4
2
proceeds. In the case of very small amounts of copper, an
excess of sodium ascorbate vs the low catalyst amount is added
to the reaction in order to avoid oxidation of Cu that is air
These results advantageously compare with relatively recent
ones (2008−2014) from the literature in Table 5, a large
number of active catalysts for the CuAAC reactions being
known.
I
sensitive.
1
2095
dx.doi.org/10.1021/ja5061388 | J. Am. Chem. Soc. 2014, 136, 12092−12098

Journal of the American Chemical Society
Article
Table 4. CuAAC Reactions between Various Azides and
Alkynes with 1−Cu (See eq 3)
Table 5. Examples of Recent Efficient CuAAC Reactions in
the Presence of Various Catalysts and Reaction Conditions
I
a
a
Cu
(% mol)
time
(h)
temperature
catalystref
solvent
(°C)
4
2
Cu/Fe
Cu-NHC/Phen
5 wt %
1
8
CH Cl2
30
20
2
4
3
18
t-BuOH/
H O 2/1
2
44
Cu/AlO(OH)
Cu O/benzoic acid
3
6
n-hexane
25
20
50
45
1
0.13 H O
2
2
poly(imidazole-acrylamide)/
CuSO /NaAsc
0.25
1.5 t-BuOH/
4
6
H O 1/3
2
4
poly(imidazole-acrylamide)/
0.00045
31
t-BuOH/
50
46
CuSO /NaAsc
H O 1/3
4
2
47
bis-NHC-dicopper complex
TBTA-CuI
0.5
0.75 CH Cl2
20
20
2
33
0.25−1
24
t-BuOH/
H O 2/1
2
48
nano-FGT-Cu
Fe O −NHC-Cu
2.4
0.25
5
0.16 H O
Mw 120
2
I
49
18
H O
2
20
20
20
2
3
50
β-CD/CuSO ,NaAsc
0.25 H O
2
4
[
Cu (l4-H){S P(OEt) } ]
0.4
4
CH CN
3
8
2
2
6
51
(
PF6)
52
CuSO /N H −H O
ammonium NHC-Cu
0.0025
5
4
3
No
20
20
20
4
2
4
2
I
53
H O
2
Fe O /SiO Tris(triazolyl)
0.5
20
H O
2
2
Cu
3
2
I
54
PS/SiO Tris(triazolyl)
0.5
4
H O
20
2
2
I
55
methane−Cu
a
I
I
56
All the reactions are carried out with 1−Cu at 30 °C during 24 h in
SBA-15-imine/Cu
0.1
0.5
2
6
H
O
2
20
20
20
22
b
57
1
mL of H O. Pent-1-yne is hydrophobic and has a low boiling point,
Cu(PPh )NO3
0.66 H O
2
3
2
58
so that it is difficult to conduct experiments with a lower amount of
[
(NHC )Cu]PF
1.5 H O
Cy
6
2
c
d
catalyst. Isolated yield. Reaction performed at 35 °C.
37
Catalyst 2
0.1
24
PhCH3
a
All the above results correspond to the CuAAC reaction between
benzyl azide and phenylacetylene. NHC: N-heterocyclic carbene, TBTA:
tris(benzyltriazolylmethyl)amine, FGT: ferrite-glutathione, β-CD: β-
One of the best results has been recently reported by Yamada
et al., who have described the synthesis of an amphiphilic
self-assembled polymeric copper catalyst. This catalyst is active
in the CuAAC reaction between benzyl azide and phenyl
acetylene with 4.5 ppm of copper at 50 °C during 32 h in the
cyclodextrin, SBA-15: mesoporous silica, NHC : NHC ligand with
Cy
cyclohexyl substituents on the nitrogen atoms. Yields are between 90 to
1
00% except with Cu-NHPhen (78%) and CuSO /N H −Cu (82%).
4
2
4
45
mixed solvent t-BuOH/H O (1/3). The presence of t-BuOH
2
57
as cosolvent and 50 °C are necessary for a complete reaction.
Another remarkable example was also recently reported by Shin
et al. using β-cyclodextrin as nanoreactor for “click” reactions
[(NHC )Cu]PF6 are also active in water but 0.5% mol and
Cy
2% mol respectively are necessary for a complete reaction
between benzyl azide and phenyl acetylene. Generally, the
CuAAC “click” reaction is possible in water when the amount
of liganded copper exceeds 0.1% mol (see Table 5); therefore,
the activity of the present system is incomparable.
in water in the presence of CuSO and sodium ascorbate at rt,
4
but 5% mol of copper was needed in that case for a quantitative
49
56
reaction in only water. The catalysts Cu(PPh )NO3 and
3
II
Scheme 2. (a) Complexation of the reference compound 4 by Cu followed by reduction to water-soluble, catalytically active
I
Cu -triazole for CuAAC reactions. (b) Structure of dendrimer 5 that does not contain triazolyl groups.
1
2096
dx.doi.org/10.1021/ja5061388 | J. Am. Chem. Soc. 2014, 136, 12092−12098

Journal of the American Chemical Society
Article
I
Scheme 3. CuAAC Reactions with 1−Cu As Catalyst for Various Applications
To evaluate the scope and the applicability of the present
with the water-soluble dendrimer 5 that has a hydrophobic core
but does not contain triazole ligands show that both the micellar
and intradendritic triazole coordination share key roles in the
considerable catalyst activation in water. This opens the route to
future biomedical and nanomaterials applications of the common
CuAAC reaction with the cheap Sharpless−Fokin catalyst
CuSO ·5H O + sodium ascorbate, as already exemplified here
I
system, the CuAAC reaction with 1−Cu was tested on hydro-
phobic biomolecules with medicinal, targeting, and labeling
interests (Scheme 3). As a candidate, the 1-ethynylcyclohexanol
6
was chosen because of its simple structure and because it is an
active metabolite of the old central nervous system depressant
drug ethinamate. The 7-(propargyloxy)coumarin 7 belonging
to the coumarin family was also tested in the CuAAC “click”
4
2
in a few examples of biomedical or cosmetic interest.
I
reaction with 1−Cu . Coumarins are often used for their anti-
oedematous properties, and their flavor properties have rendered
this family famous in the perfume industry. Moreover coumarins
are known as fluorescence dyes. The 3-(D-biotinylamido)-1-
propyne 8 is an alkyne derivative of biotin that is known for
instance for its role as a vitamin and coenzyme in the synthesis of
fatty acids. Biotin is also known for its extremely high affinity
ASSOCIATED CONTENT
Supporting Information
■
*
S
I
General data, syntheses of the catalysts 2 and 1−Cu ,
procedure and data for the encapsulation of catalyst 2 in 1,
I
1
complexation of Cu to the triazoles of 1, syntheses and H
NMR spectra and data of the “click” reaction products,
spectroscopic characterization of 3−5, kinetic study of the
autocatalytic synthesis of 1, and characterization of 1 after
reuse. This material is available free of charge via the Internet at
http://pubs.acs.org.
I
with avidin. As exposed in Scheme 3, the catalyst 1−Cu is very
active for the CuAAC “click” reaction even for these three
I
biological molecules with only 100−200 ppm of Cu .
CONCLUDING REMARKS
■
AUTHOR INFORMATION
Corresponding Author
d.astruc@ism.u-bordeaux1.fr
The water-soluble dendrimer 1 acts as a molecular micelle
nanoreactor in catalytic quantities (1% vs substrates or less)
and is recycled and reused many times without any loss or
decomposition. Under these conditions (water, 25 °C, 3 h), it
considerably facilitates catalysis by [Cu(hexabenzyltren)]Br
■
Notes
The authors declare no competing financial interest.
(
0.1% vs substrate) of the CuAAC reactions that are nearly
quantitative in the presence of this dendritic nanoreactor and
do not work in its absence. Along this line, the “autocatalyzed”
dendritic nanoreactor synthesis of 1 is remarkable.
ACKNOWLEDGMENTS
■
Helpful discussions with Dr Jaime Ruiz (Univ. Bordeaux) and
1
financial support from the Universite
́
de Bordeaux, the CNRS,
e de l′Enseignement Superieur et de la Recherche
Ph.D. grant to C.D.) and L′Oreal are gratefully acknowledged.
This catalytic nanoreactor effect is confirmed by H NMR
the Minister
̀
́
evidence of the solubilization of the hydrophobic catalyst 2
in water in the presence of the dendrimer 1, and the catalyst−
dendrimer interaction is shown by the selective NMR shifts
of intradendritic protons in the hydrophobic region, providing a
proof for the solubilization role of 1.
(
́
REFERENCES
■
(
1) Breslow, R.; Overman, L. E. J. Am. Chem. Soc. 1970, 92, 1075−
077.
2) Yoshizawa, M.; Tamura, M.; Fujita, M. Science 2006, 312, 251−
1
In addition to and in synergy with this effect, the presence of
intradendritic triazole ligands allows catalyzing CuAAC reactions
at 30 °C with down to 4 ppm of commercial CuSO ·5H O and
(
2
54.
4
2
(
(
(
3) Kang, J.; Rebek, J. Nature 1997, 385, 50−52.
sodium ascorbate for quantitative yields and 1 ppm with 50%
yield. The reaction with hydrophobic biomolecules has also been
performed in only water at 30 °C, leading to quantitative yields.
Comparisons of 1 with the nondendritic triazole ligand 4 and
4) Mock, W. L. Top. Curr. Chem. 1995, 175, 1−24.
5) Warmuth, R. Angew. Chem., Int. Ed. 1997, 36, 1347−1350.
(6) Slagt, V. F.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Angew.
Chem., Int. Ed. 2003, 42, 5619−5623.
1
2097
dx.doi.org/10.1021/ja5061388 | J. Am. Chem. Soc. 2014, 136, 12092−12098

Journal of the American Chemical Society
Article
(
7) Walter, C. J.; Anderson, C. J.; Sanders, J. K. M. J. Chem. Soc.,
Chem. Commun. 1993, 458−460.
8) Fiedler, D.; Leung, D. H.; Bergman, R. G.; Raymond, K. N. Acc.
Chem. Res. 2005, 38, 351−360.
9) Wang, Z. J.; Clary, K. N.; Bergman, R. G.; Raymond, K. N.;
Toste, F. D. Nat. Chem. 2013, 5, 100−103.
10) Yu, X.; Yue, K.; Hsieh, I.-F.; Li, Y.; Dong, X.-H.; Liu, C.; Xin, Y.;
(41) Boisselier, E.; Diallo, A. K.; Salmon, L.; Ornelas, C.; Ruiz, J.;
Astruc, D. J. Am. Chem. Soc. 2010, 132, 2729−2742.
(
́ ́ ́
(42) Kovacs, S.; Zih-Perenyi, K.; Revesz, A.; Novak, S. Synthesis 2012,
44, 3722−3730.
(
(43) Teyssot, M.-L.; Chevry, A.; Traïkia, M.; El-Ghozzi, M.;
Avignant, D.; Gautier, A. Chem.Eur. J. 2009, 15, 6322−6326.
(44) Park, I. S.; Kwon, M. S.; Kim, Y.; Lee, J. S.; Park, J. Org. Lett.
2008, 10, 497−500.
(
Wang, H.-F.; Shi, A.-C.; Newkome, G. R.; Ho, R.-M.; Chen, E.-Q.;
Zhang, W.-B.; Cheng, S. Z. D. Proc. Nat. Acad. Sci. U. S. A. 2013, 110,
(45) Shao, C.; Zhu, R.; Luo, S.; Zhang, Q.; Wang, X.; Hu, Y.
Tetrahedron Lett. 2011, 52, 3782−3785.
1
(
0078−10083.
11) Discher, D. E.; Eisenberg, A. Science 2002, 297, 967−973.
(46) Yamada, Y. M. A.; Sarkar, S. M.; Uozumi, Y. J. Am. Chem. Soc.
2012, 134, 9285−9290.
(
12) Crooks, R. M.; Zhao, M.; Sun, L.; Chechik, V.; Yeung, L. K. Acc.
Chem. Res. 2001, 34, 181−190.
13) Scott, R. W. J.; Wilson, O. M.; Crooks, R. M. Phys. Chem. B
005, 109, 692−704.
14) Myers, V. S.; Weier, M. G.; Carino, E. V.; Yancey, D. F.; Pande,
S.; Crooks, R. M. Chem. Sci. 2011, 2, 1632−1646.
15) Bernechea, M.; de Jesus, E.; Lopez-Mardomingo, C. Inorg.
Chem. 2009, 48, 4491−4496.
16) Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev. 2010, 110,
(47) Berg, R.; Straub, J.; Schreiner, E.; Mader, S.; Rominger, F.;
Straub, B. F. Adv. Synth. Catal. 2012, 354, 3445−3450.
(48) Baig, R. B. N.; Varma, R. S. Green Chem. 2012, 14, 625−632.
(
2
(
(49) Collinson, J.-M.; Wilton-Ely, J. D. E. T.; Díez-Gonzal
Chem. Commun. 2013, 49, 11358−11360.
50) Shin, J.-A.; Lim, Y.-G.; Lee, K.-L. J. Org. Chem. 2012, 77, 4117−
122.
51) Lee, B.-H.; Wu, C.-C.; Fang, X.; Liu, C.-W.; Zhu, J.-L. Catal.
Lett. 2013, 143, 572−577.
52) Pathigoolla, A.; Pola, R. P.; Sureshan, K. M. Appl. Catal., A 2013,
53, 151−158.
53) Wang, W.; Wu, J.; Xia, C.; Li, F. Green Chem. 2011, 13, 3440−
445.
54) Wang, D.; Etienne, L.; Echeverria, M.; Moya, S.; Astruc, D.
Chem.Eur. J. 2014, 20, 4047−4054.
55) Ozkal, E.; Llanes, P.; Bravo, F.; Ferrali, A.; Pericas
Synth. Catal. 2014, 356, 857−869.
56) Roy, S.; Chatterjee, T.; Pramanik, M.; Singha Roy, A.; Bhaumik,
A.; Islam, SK. M. J. Mol. Catal. A: Chem. 2014, 386, 78−85.
57) Wang, D.; Li, N.; Zhao, M.; Shi, W.; Ma, C.; Chen, B. Green
Chem. 2010, 12, 2120−2123.
58) Diez-Gonzalez, S.; Nolan, S. P. Angew. Chem., Int. Ed. 2008, 47,
881−8884.
́
ez, S.
(
4
(
(
́
(
1
(
857−1959.
(
4
(
3
17) Astruc, D. Nat. Chem. 2012, 4, 255−267.
18) Deraedt, C.; Astruc, D. Acc. Chem. Res. 2014, 46, 494−503.
(
(
19) Newkome, G. R.; Yao, Z.; Baker, G. R.; Gupta, V. K. J. Org.
Chem. 1985, 50, 2003−2004.
20) Newkome, G. R.; Moorefield, C. N.; Baker, G. R.; Saunders, M.
J.; Grossman, S. H. Angew. Chem., Int. Ed. 1991, 30, 1178−1180.
21) Newkome, G. R.; Shreiner, C. Chem. Rev. 2010, 110, 6338−
442.
22) Jansen, J. F. G. A.; de Brabander-van den Berg, E. M. M.; Meijer,
E. W. Science 1999, 266, 1226−1229.
23) Zimmerman, S. C.; Wendland, M. S.; Rakow, N. A.; Zharov, I.;
Suslick, K. S. Nature 2002, 418, 399−403.
24) Ornelas, C.; Ruiz, J.; Belin, C.; Astruc, D. J. Am. Chem. Soc.
009, 131, 590−601.
25) Ornelas, C.; Ruiz, J.; Cloutet, E.; Alves, S.; Astruc, D. Angew.
Chem., Int. Ed. 2007, 46, 872−877.
26) Diallo, A. K.; Ornelas, C.; Salmon, L.; Ruiz, J.; Astruc, D. Angew.
Chem., Int. Ed. Engl. 2007, 46, 8644−8648.
27) Astruc, D.; Liang, L.; Rapakousiou, A.; Ruiz, J. Acc. Chem. Res.
012, 45, 630−640.
28) Helms, B.; Frec
148.
29) Frec
713−3725.
30) Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel, A.;
Voit, B.; Pyun, J.; Frechet, J. M. J.; Sharpless, K. B.; Fokin, V. V. Angew.
Chem., Int. Ed. 2004, 43, 3928−3932.
31) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596−2599.
32) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002,
7, 3057−3064.
33) Chan, T. R.; Hilgraf, R.; Sharpless, K. B.; Fokin, V. V. Org. Lett.
004, 6, 2853−2855.
34) Hong, V.; Presolski, S. I.; Ma, C.; Finn, M. G. Angew. Chem., Int.
Ed. 2009, 48, 9879−9883.
35) Golas, P. L.; Tsarevsky, N. V.; Sumerlin, B. S.; Matyjaszewski, K.
Macromolecules 2006, 39, 6451−6457.
36) Golas, P. L.; Tsarevsky, N. V.; Matyjaszewski, K. Macromol.
Rapid Commun. 2008, 29, 1167−1171.
37) Liang, L.; Ruiz, J.; Astruc, D. Adv. Syn. Catal. 2011, 353, 3434−
450.
38) Candelon, N.; Lastec
Vincent, J.-M. Chem. Commun. 2008, 741−743.
39) Deraedt, C.; Salmon, L.; Etienne, L.; Ruiz, J.; Astruc, D. Chem.
Commun. 2013, 49, 8169−8171.
40) For the complexation site of “clicked” 1,2,3-triazole ligands with
various transition metals, see Badeche, S.; Daran, J.-C.; Ruiz, J.; Astruc,
D. Inorg. Chem. 2008, 47, 4903−4908.
(
(
(
̀
, M. A. Adv.
(
6
(
(
(
(
(
8
(
2
(
(
(
2
(
1
(
3
(
́
het, J. M. J. Adv. Synth. Catal. 2006, 348, 1125−
́
het, J. M. J. J. Polym. Sci., Part A: Polym. Chem. 2003, 41,
́
(
(
6
(
2
(
(
(
(
3
(
́ ̀
ouere, D.; Diallo, A. K.; Ruiz, J.; Astruc, D.;
(
(
̀
1
2098
dx.doi.org/10.1021/ja5061388 | J. Am. Chem. Soc. 2014, 136, 12092−12098