Simultaneous copper(I)-catalyzed azide-alkyne cycloaddition (CuAAC) and living radical polymerization

Article abstract of DOI:10.1002/anie.200800179

(Chemical Equation Presented) All in one: CuBr/iminopyridine systems can catalyze simultaneously both copper(I)-catalyzed azide-alkyne cycloaddition (CuAAC, "click") and living radical polymerization (LRP) processes (see scheme). The relative rate of the two processes can be tailored by a judicious choice of the reaction conditions (solvent, temperature, [CuBr] ₀) leading to the development of a potentially very efficient synthetic route to well-defined functional polymers.

Full text of DOI:10.1002/anie.200800179

Communications
DOI: 10.1002/anie.200800179
Living Polymerization
Simultaneous Copper(I)-Catalyzed Azide–Alkyne Cycloaddition
(CuAAC) and Living Radical Polymerization**
Jin Geng, Josefina Lindqvist, Giuseppe Mantovani,* and David M. Haddleton*
Copper(I)-catalyzed azide–alkyne cycloaddition (CuAAC) is
a member of the family of click reactions[1] that is receivinga
steeply growing interest.[2] This has led to its use in the
synthesis of a range of materials, including dendrimers,
hydrogels, drugs, and functional polymers.[3] The mechanism
of CuAAC appears to be relatively complex and is not yet
fully understood, at least in terms of the exact nature of the
copper-containingspecies involved in the catalytic cycle.
Recent studies have also shown that in some cases more than
one catalytically active copper species may contribute to the
formation of the triazole “clicked” products.[4] In addition, the
ability of CuI to coordinate a number of species in solution,
including ligands, solvents, organic buffers, reactants and final
products, in a very dynamic equilibrium and to dispropor-
tionate to Cu0 and CuII in polar environments further
complicates the catalytic process.
The first step in the catalysis of CuAAC is the formation
of CuI acetylides I from CuI species, 1-alkynes, and a base
(Scheme 1). CuI acetylides have a tendency to form m-
coordinate bridged aggregates analogous to IIa.[5] Studies
have indicated that such species might play an important role
in CuAAC reactions.[4,5b,6] This also might be responsible for
the second-order dependence on the concentration of copper
species, as observed by Finn, Fokin and co-workers in many
cases, both in the presence[4,6b] and absence[7] of stabilizing
ligands. Interestingly, Straub and co-workers also suggested
that, with some very bulky substrates and coordinating
ligands, CuAAC may even occur without the formation of
such m-coordinate bridged CuI aggregates.[8] The second step
is the reaction of CuI acetylides with organic azides to give CuI
triazolides III, presumably via a six-membered metalla-
cycle.[2a,6c,9] Recently, the first example of a stable complex
III, a 14-valence electrons CuI triazolide bearinga N-hetero-
cyclic carbene (NHC) ligand, has been isolated, fully charac-
terized and employed as a CuAAC catalyst, providingdirect
evidence of the existence of such triazolide “click” inter-
mediates.[8] The final step is the proteolysis of III to give a 1,4-
substituted 1,2,3-triazole and recover the copper catalyst I.
Livingradical polymerization is a powerful method to
obtain macromolecules with controllable properties and
architectures. The most widely studied of these techniques,
transition-metal-mediated livingradical polymerization
(TMM LRP, often termed ATRP),[10] shares a number of
attractive features with CuAAC, includinghihg tolerance
towards unprotected functional groups and protic solvents. In
a simplified mechanism, for Cu-catalyzed LRP, a CuI complex
reacts reversibly with a suitable organic halide to give a
radical-based propagating species and a CuII complex. Percec
et al. have further suggested that in polar media the CuI
catalyst undergoes significant disproportionation to CuII and
Cu0 and the Cu0 species activates the organic halide via a
single-electron transfer mechanism (SET-LRP) to give a
radical propagating species and a CuI complex, which
subsequently disproportionates to a CuII end-cappingcom-
plex, necessary to control the polymerization process.[11] Thus
at present the mechanism of both processes seems to be not
yet fully understood and both require further investigation; it
appears that the mechanisms of CuAAC and LRP may have
little in common.
Modular processes that sequentially combine CuAAC
and LRP have proven to be a powerful route to functional
polymers with complex macromolecular architectures.[12] We
have previously shown that CuAAC and LRP can share the
same CuIBr-iminopyridine catalytic system in a sequential
one-pot reaction.[13] Sequential CuAAC/LRP one-pot pro-
cesses usingother catalytic systems have also been
reported.[14] The aim of the present study was to investigate
catalytic systems in which CuAAC and LRP could occur
simultaneously and if so, to combine them to develop a novel
synthetic tool for the synthesis of functional molecular
materials.
Scheme 1. Simplified proposed catalytic cycle for the CuAAC reac-
tion.[6b] All of the copper species are likely to exist in equilibrium with
higher-order aggregates. Complexes such as IIa, in particular, seem to
play an important role in the process.
[*] J. Geng, Dr. J. Lindqvist, Dr. G. Mantovani, Prof. D. M. Haddleton
Department of Chemistry, University of Warwick
Coventry CV4 7AL (UK)
Fax: (+44)24-7652-8267
E-mail: g.mantovani@warwick.ac.uk
d.m.haddleton@warwick.ac.uk
For this study, an alkyne-containingmonomer, propargyl
methacrylate (1), an organic azide (1-octyl azide (2), meth-
oxytriethylene glycol azide (3) or 2’-azidoethyl-a-mannopyr-
anoside (4)), a LRP initiator and CuBr/N-ethyl pyridine-
Homepage: http://www.warwick.ac.uk/go/polymers
[**] The University of Warwick is thanked for financial support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
4180
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4180 –4183

Angewandte
Chemie
Scheme 2. One-pot CuAAC/LRP process: possible pathways (benzyl 2-
bromoisobutyrate was employed as the polymerization initiator).
imine/triethylamine as catalyst were employed (Scheme 2). A
number of experimental parameters, includingsolvent, cata-
lyst concentration, temperature and nature of the azide, were
systematically varied. To enable efficient monitoringof the
kinetics of the two processes in real time, all of the reactions
were conducted in NMR tubes at the required temperature
Figure 1. CuAAC/LRP in different solvents: alkyne monomer (1) and
!
triethyleneglycol azide (3) in [D8]toluene ( ), [D7]DMF (&), and
I
*
[D6]DMSO ( ). Reaction conditions: [initiator]/[1]/[3]/[Cu Br]/[ligand]/
[Et3N]=1:50:100:1:2:5. “Click conversion” (a) refers to the conversion
of alkyne units, both in the monomer 1 and in the polymer backbone,
into 1,2,3-triazole moieties. “LRP conversion” (b) refers to the poly-
merization of both 1 and its corresponding clicked monomer ana-
logue. PDi: polydispersity index.
1
with H NMR spectra recorded at 7 min intervals.
In [D8]toluene as solvent at 608C, the processes were
found to be sequential, followingpath 1 in Scheme 2. The
CuAAC reactions usingboth azides 2 and 3 were found to
proceed virtually to completion prior to the polymerization
starting. Subsequent polymerization of the resulting “clicked”
monomer yielded a final polymer where all side-groups were
derived from the azide.
[D6]DMSO as solvent, generally regarded as a less-than-
optimal solvent for the CuAAC reaction, was also assessed.
As octyl azide (2) was not completely soluble in [D6]DMSO
(and known to form aggregates responsible for “non conven-
tional” CuAAC kinetics in polar environments[4]), triethyl-
eneglycol azide (3) was employed. Under these conditions the
two reactions were found to proceed concurrently, with the
click reaction significantly slower and the rate of the
polymerization slightly higher than with toluene as solvent.
In [D7]DMF the two reactions were again observed to
proceed simultaneously. However, the rate of the click
reaction was significantly higher than in [D6]DMSO, whereas
the rate of the polymerization was similar to when toluene
was used as the solvent (Figure 1).
was observed in DMSO. Whilst the reasons for the increased
rate of the polymerization with time in toluene still remain
unclear, in DMSO the decreasingpolymerization rate may be
an effect of the increasingnumber of 1-alkyne and triazole
moieties able to coordinate copper species in the polymer
backbone, which may result in a reduced amount of available
copper catalyst in the system.
Simultaneous CuAAC and LRP require a complex set of
reactions and equilibria to occur efficiently at the same time
(Scheme 2, path 3). Polymerization of both the un-clicked 1
and clicked monomers leads to polymers bearingboth
1-alkyne and functional 1,2,3-triazole as the pendant units,
as was also observed from 1H NMR analysis (Figure 2).
Conversely, CuAAC can occur both on the unreacted alkyne-
functional monomer 1 and on the 1-alkyne groups present in
the growing polymer chain. If the click reaction does not
proceed to completion, the final macromolecular product will
be a copolymer containinga certain amount of “un-clicked”
1-alkyne functionalities.
All polymerization reactions, in all solvents, showed a
significant deviation away from linear first-order kinetics (see
SupportingInformation), which is likely to be the result of the
complex set of equilibria existingin solution. In toluene, the
polymerization rate increased after some time; the opposite
In DMSO, the catalyst concentration was found to have a
significant effect on the rate of CuAAC (Figure 3). Virtually
complete conversion of alkyne moieties in the reaction
mixture could be achieved by simple judicious tuningof the
Angew. Chem. Int. Ed. 2008, 47, 4180 –4183
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4181

Communications
Figure 2. Partial 1H NMR spectrum of reaction mixture after 157 min
in [D6]DMSO. Reaction conditions: [initiator]/[1]/[3]/[CuIBr]/[ligand]/
[Et3N]=1:50:100:1:2:5.
Figure 3. CuAAC/LRP at different copper concentration: alkyne mono-
mer (1) with triethyleneglycol azide (3) in [D6]DMSO at different
catalyst concentrations. Reaction conditions: [initiator]/[1]/[3]/
[Et3N]=1:50:100:5. Equivalents of L2/CuIBr (L: N-ethyl pyridineimine):
concentration of [CuIBr]/[ligand]. Interestingly, in these
experiments the polymerization showed only a small rate
increase with increased catalyst concentration (Figure 3).
Conversely, in toluene, under otherwise identical exper-
imental conditions, the rate of the LRP reaction was found to
be greatly dependent on the concentration of the catalyst (see
SupportingInformation). The rate of CuAAC under the
range of catalyst concentrations employed was too high to
allow any conclusions regarding the effect of the copper
concentration on the click process under these reaction
conditions. Even though the polymerization could be signifi-
cantly accelerated by increasingthe concentration of copper
species in solution, the CuAAC was still found to be faster
than LRP for all cases. Decreasingthe temperature from 60 to
308C caused a decrease in the rate of the polymerization,
whereas the rate of the CuAAC remained extremely high.
A comparison of rates of LRP vs. click conversion shows
that in the presence of a relatively small catalyst loadingthe
rate of CuAAC in DMSO tends to be slower than (or at least
comparable to) that of polymerization, whilst in toluene, or
DMF, the click process is faster than polymerization
(Figure 4). In the absence of the iminopyridine ligand, in
[D6]DMSO at 608C, the CuAAC reaction was slightly faster
than when the chelatingligand was employed. This may be
explained by the formation of inactive species when an excess
of ligand is used, as suggested by Finn and co-workers.[6b]
Interestingly, LRP also showed to proceed under these
conditions, only at a slower rate, indicatingthat either the
formed triazoles, triethylamine or the solvent, or a combina-
!
~
0.5 (*), 1 ( ), 2 (&), and 4 ( ).
Figure 4. LRP vs. click conversion for CuAAC/LRP in different solvent
media: alkyne monomer (1) and triethyleneglycol azide (3) in
!
*
[D8]toluene ( ),[D7]DMF (&), and [D6]DMSO ( ). Reaction conditions:
[initiator]/[1]/[3]/[CuIBr]/[ligand]/[Et3N]=1:50:100:1:2:5. The solid line
represents the theoretical case of LRP conversion=click conversion.
tion of these, may act as a ligand to complex the copper
species and catalyze the polymerization. Size exclusion
chromatography (SEC) analysis of the resulting polymer,
however, showed a higher PDi (1.55) compared to when an
iminopyridine ligand was used, indicating that the polymer-
4182
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4180 –4183

Angewandte
Chemie
ization without added ligand was largely uncontrolled. To
evaluate whether a certain extent of click reaction occurred
thermally, via a non-copper catalyzed pathway, a one-pot
experiment was also conducted, at 608C, without the addition
Keywords: click chemistry · copper · cycloaddition ·
one-pot process · polymerization
.
1
[1] a) H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001,
113, 2056; Angew. Chem. Int. Ed. 2001, 40, 2004; b) H. C. Kolb,
K. B. Sharpless, Drug Discovery Today 2003, 8, 1128.
[2] a) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708; Angew. Chem. Int. Ed. 2002, 41,
2596; b) C. W. Tornoe, C. Christensen, M. Meldal, J. Org. Chem.
2002, 67, 3057.
[3] For a few examples of recent reviews on click chemistry, see
a) K. B. Sharpless, R. Manetsch, Expert Opin. Drug Discovery
2006, 1, 525; b) H. Nandivada, X. Jiang, J. Lahann, Adv. Mater.
2007, 19, 2197; c) J.-F. Lutz, Angew. Chem. 2007, 119, 1036;
Angew. Chem. Int. Ed. 2007, 46, 1018; d) C. J. Hawker, V. V.
Fokin, M. G. Finn, K. B. Sharpless, Aust. J. Chem. 2007, 60, 381.
[4] V. O. Rodionov, S. I. Presolski, S. Gardinier, Y.-H. Lim, M. G.
Finn, J. Am. Chem. Soc. 2007, 129, 12696.
[5] a) B. M. Mykhalichko, O. N. Temkin, M. G. Mysꢀkiv, Russ. Chem.
Rev. 2000, 69, 957; b) B. F. Straub, Chem. Commun. 2007, 3868.
[6] a) V. Aucagne, J. Berna, J. D. Crowley, S. M. Goldup, K. D.
Haenni, D. A. Leigh, P. J. Lusby, V. E. Ronaldson, A. M. Z.
Slawin, A. Viterisi, D. B. Walker, J. Am. Chem. Soc. 2007, 129,
11950; b) V. O. Rodionov, S. I. Presolski, D. D. Diaz, V. V. Fokin,
M. G. Finn, J. Am. Chem. Soc. 2007, 129, 12705; c) M. Ahlquist,
V. V. Fokin, Organometallics 2007, 26, 4389.
[7] V. O. Rodionov, V. V. Fokin, M. G. Finn, Angew. Chem. 2005,
117, 2250; Angew. Chem. Int. Ed. 2005, 44, 2210.
[8] C. Nolte, P. Mayer, B. F. Straub, Angew. Chem. 2007, 119, 2147;
Angew. Chem. Int. Ed. 2007, 46, 2101.
[9] F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtsev, L. Noodleman,
K. B. Sharpless, V. V. Fokin, J. Am. Chem. Soc. 2005, 127, 210.
[10] a) M. Kamigaito, T. Ando, M. Sawamoto, Chem. Rev. 2001, 101,
3689; b) K. Matyjaszewski, J. Xia, Chem. Rev. 2001, 101, 2921;
c) A. Limer, D. M. Haddleton, Prog. React. Kinet. Mech. 2004,
29, 187.
[11] V. Percec, T. Guliashvili, J. S. Ladislaw, A. Wistrand, A.
Stjerndahl, M. J. Sienkowska, M. J. Monteiro, S. Sahoo, J. Am.
Chem. Soc. 2006, 128, 14156.
[12] a) H. Gao, K. Matyjaszewski, Macromolecules 2006, 39, 4960;
b) Q. Liu, P. Zhao, Y. Chen, J. Polym. Sci. Part A. 2007, 45, 3330;
c) C. N. Urbani, C. A. Bell, D. Lonsdale, M. R. Whittaker, M. J.
Monteiro, Macromolecules 2008, 41, 76; d) J. A. Johnson, M. G.
Finn, J. T. Koberstein, N. J. Turro, Macromolecules 2007, 40,
3589; e) W. Van Camp, V. Germonpre, L. Mespouille, P. Dubois,
E. J. Goethals, F. E. Du Prez, React. Funct. Polym. 2007, 67,
1168; f) X. He, L. Liang, M. Xie, Y. Zhang, S. Lin, D. Yan,
Macromol. Chem. Phys. 2007, 208, 1797; g) H. Gao, K. Maty-
jaszewski, J. Am. Chem. Soc. 2007, 129, 6633; h) V. Ladmiral, G.
Mantovani, J. Clarkson Guy, S. Cauet, L. Irwin Jacob, M.
Haddleton David, J. Am. Chem. Soc. 2006, 128, 4823; i) B. A.
Laurent, S. M. Grayson, J. Am. Chem. Soc. 2006, 128, 4238; j) J.-
F. Lutz, H. G. Boerner, K. Weichenhan, Macromol. Rapid
Commun. 2005, 26, 514; k) J. Geng, J. Lindqvist, G. Mantovani,
G. Chen, C. T. Sayers, G. J. Clarkson, D. M. Haddleton, QSAR
Comb. Sci. 2007, 26, 1220; l) J. Geng, G. Mantovani, L. Tao, J.
Nicolas, G. Chen, R. Wallis, D. A. Mitchell, B. R. G. Johnson,
S. D. Evans, D. M. Haddleton, J. Am. Chem. Soc. 2007, 129,
15156.
of any copper species. H NMR analysis revealed that after
10 h no clicked products were present. Homopolymerizations
of 1 in the absence of any azide, conducted in toluene and
DMSO, led to polymers with broad molecular weight
distribution (PDi = 1.53–1.88), in agreement with previous
reports.[14b] This behavior may be ascribed to chain transfer to
the 1-alkyne moieties, amongst other things, and highlights
the importance of minimizingthe amount of un-clicked
alkyne groups in the growing polymer.
Synthesis of multivalent ligands presenting multiple
copies of carbohydrate-bindingepitopes for protein receptor
recognition is an area that could benefit from novel and
efficient synthetic strategies. The decreased number of
reaction steps compared to post-functionalization of an
alkyne-functional polymer with sugar azides[12h] would offer
an alternative and complementary route to such functional
glycopolymers. A one-pot CuAAC-LRP process would also
avoid the use of sugar functional methacrylate monomers,
which are known to be prone to self-polymerize.
Mannose-functional polymers were prepared by CuAAC-
LRP from propargyl methacrylate (1) and 2’-azidoethyl-a-
mannopyranoside (4). DMSO was used as solvent as it
solubilizes the mannose azide startingmaterial, 1, and the
polymer product. Online 1H NMR showed that the two
reactions occurred simultaneously, with the polymerization
slower than when triethyleneglycol azide (3) was used.
Increasingthe catalyst concentration increased the rate of
both LRP and CuAAC. The latter aspect allowed to largely
limit the amount of un-clicked alkyne groups that was present
in the final polymer when relatively low concentration of
copper catalyst was employed. SEC analysis of the resulting
glycopolymers showed PDi = 1.12–1.14, indicatingcontrolled
polymerization.
In conclusion, we report to our best knowledge the first
copper-catalyzed one-pot simultaneous CuAAC-LRP process.
ACuIBr/iminopyridine catalytic system was found to efficiently
catalyze both processes under a range of different experimental
conditions. The relative rates of CuAAC and LRP can be tuned
by appropriate changes of a number of parameters that include
the concentration of the copper species, the nature of the
solvent employed and the temperature at which reactions are
conducted. The rate of the CuAAC process appeared to be
relatively sensitive to the nature of the solvent employed, which
may be at least in part explained in terms of the different ability
of the various solvents to coordinate the copper center,
affectingthe complicated series of equilibria occurringduring
the catalytic process.
In addition, the implementation of this concept for the
synthesis of glycopolymers yielded well-defined materials in a
simplified manner, and we believe that this concept can be
used as a versatile synthetic tool in many different applica-
tions.
[13] G. Mantovani, V. Ladmiral, L. Tao, D. M. Haddleton, Chem.
Commun. 2005, 2089.
[14] a) L. Mespouille, M. Vachaudez, F. Suriano, P. Gerbaux, O.
Coulembier, P. Degee, R. Flammang, P. Dubois, Macromol.
Rapid Commun. 2007, 28, 2151; b) B. S. Sumerlin, N. V. Tsar-
evsky, G. Louche, R. Y. Lee, K. Matyjaszewski, Macromolecules
2005, 38, 7540.
Received: January 14, 2008
Published online: April 15, 2008
Angew. Chem. Int. Ed. 2008, 47, 4180 –4183
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4183