Ultrafast synthesis of ultrahigh molar mass polymers by metal-catalyzed living radical polymerization of acrylates, methacrylates, and vinyl chloride mediated by SET at 25 °C

Article abstract of DOI:10.1021/ja065484z

Conventional metal-catalyzed organic radical reactions and living radical polymerizations (LRP) performed in nonpolar solvents, including atom-transfer radical polymerization (ATRP), proceed by an inner-sphere electron-transfer mechanism. One catalytic system frequently used in these polymerizations is based on Cu(I)X species and N-containing ligands. Here, it is reported that polar solvents such as H₂O, alcohols, dipolar aprotic solvents, ethylene and propylene carbonate, and ionic liquids instantaneously disproportionate Cu(I)X into Cu(0) and Cu(II)X₂ species in the presence of a diversity of N-containing ligands. This disproportionation facilitates an ultrafast LRP in which the free radicals are generated by the nascent and extremely reactive Cu(0) atomic species, while their deactivation is mediated by the nascent Cu(II)X₂ species. Both steps proceed by a low activation energy outer-sphere single-electron-transfer (SET) mechanism. The resulting SET-LRP process is activated by a catalytic amount of the electron-donor Cu(0), Cu₂Se, Cu₂Te, Cu₂S, or Cu₂O species, not by Cu(I)X. This process provides, at room temperature and below, an ultrafast synthesis of ultrahigh molecular weight polymers from functional monomers containing electron-withdrawing groups such as acrylates, methacrylates, and vinyl chloride, initiated with alkyl halides, sulfonyl halides, and N-halides.

Full text of DOI:10.1021/ja065484z

Published on Web 10/05/2006
Ultrafast Synthesis of Ultrahigh Molar Mass Polymers by
Metal-Catalyzed Living Radical Polymerization of Acrylates,
Methacrylates, and Vinyl Chloride Mediated by SET at 25 °C
Virgil Percec,* Tamaz Guliashvili, Janine S. Ladislaw, Anna Wistrand,
Anna Stjerndahl, Monika J. Sienkowska, Michael J. Monteiro, and Sangrama Sahoo
Contribution from the Roy and Diana Vagelos Laboratories, Department of Chemistry,
UniVersity of PennsylVania, Philadelphia, PennsylVania 19104-6323
Received July 29, 2006; E-mail: percec@sas.upenn.edu
Abstract: Conventional metal-catalyzed organic radical reactions and living radical polymerizations (LRP)
performed in nonpolar solvents, including atom-transfer radical polymerization (ATRP), proceed by an inner-
sphere electron-transfer mechanism. One catalytic system frequently used in these polymerizations is based
on Cu(I)X species and N-containing ligands. Here, it is reported that polar solvents such as H
dipolar aprotic solvents, ethylene and propylene carbonate, and ionic liquids instantaneously disproportionate
Cu(I)X into Cu(0) and Cu(II)X species in the presence of a diversity of N-containing ligands. This
disproportionation facilitates an ultrafast LRP in which the free radicals are generated by the nascent and
extremely reactive Cu(0) atomic species, while their deactivation is mediated by the nascent Cu(II)X species.
Both steps proceed by a low activation energy outer-sphere single-electron-transfer (SET) mechanism.
The resulting SET-LRP process is activated by a catalytic amount of the electron-donor Cu(0), Cu Se,
Cu Te, Cu S, or Cu O species, not by Cu(I)X. This process provides, at room temperature and below, an
2
O, alcohols,
2
2
2
2
2
2
ultrafast synthesis of ultrahigh molecular weight polymers from functional monomers containing electron-
withdrawing groups such as acrylates, methacrylates, and vinyl chloride, initiated with alkyl halides, sulfonyl
halides, and N-halides.
Introduction
Metal-catalyzed LRP is considered to proceed by an inner-
sphere electron-transfer mechanism in which a low oxidation
state metal complex acts as the catalyst, mediating a fast
exchange between radicals and their dormant alkyl halide
Metal-catalyzed living radical polymerization (LRP) initiated
1
2
3
with alkyl halides, sulfonyl halides, and N-halides has been
1
a-f
accomplished in organic and aqueous media
as well as in
1
c,f,g,h,2b
3
1,2a,b
species.
The equilibrium between active and dormant
ionic liquids, mostly for activated monomers such as styrene,
acrylates,1a-f,5
1a-f,2,3,7
6
species is shifted toward the dormant species by an excess of a
high oxidation state catalyst generated by a small extent of
bimolecular radical dimerization during the initial stages of the
polymerization. This concept is known as internal suppression
methacrylates,
and acrylonitrile. Complex
macromolecular architectures, such as dendrimers, have also
been synthesized by metal-catalyzed LRP.7
8
(
1) (a) Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T. Macromol-
ecules 1995, 28, 1721-1723. (b) Wang, J. S.; Matyjaszewski, K. J. Am.
Chem. Soc. 1995, 117, 5614-5615. (c) Matyjaszewski, K.; Xia, J. Chem.
ReV. 2001, 101, 2921-2990. (d) Kamigaito, M.; Ando, T.; Sawamoto, M.
Chem. ReV. 2001, 101, 3689-3746. (e) Limer, A.; Haddleton, D. M. Prog.
React. Kinet. 2004, 29, 187-241. (f) Matyjaszewski, K. Macromol. Symp.
of fast reactions or the persistent radical effect (PRE). The
inner-sphere radical process is also known as atom-transfer
radical addition (ATRA). The corresponding radical polymer-
ization was, therefore, named atom-transfer radical polymeri-
1b,c,f
1
7
3
998, 134, 105-118. (g) Taube, H. Myers, H. J. Am. Chem. Soc. 1954,
6, 2103-2111. (g) Taube, H. Angew. Chem., Int. Ed. Engl. 1984, 23,
29-339.
zation (ATRP).
Nonactivated monomers that generate stable
alkyl halide dormant species, such as vinyl acetate, vinyl
chloride, and ethylene, do not and are not expected to polymerize
(
2) (a) Percec, V.; Barboiu, B. Macromolecules 1995, 28, 7970-7972. (b)
Percec, V.; Barboiu, B.; Kim, H.-J. J. Am. Chem. Soc. 1998, 120, 305-
9
by the current generation of ATRP catalysts. Of specific interest
3
16. (c) Grigoras, C.; Percec, V. J. Polym. Sci., Part A: Polym. Chem.
2
005, 43, 319-330. (d) Percec, V.; Grigoras, C. J. Polym. Sci., Part A:
is vinyl chloride (VC) because it has proven impossible to
Polym. Chem. 2005, 43, 3920-3931.
1
0
11a,b
(
3) Percec, V.; Grigoras, C. J. Polym. Sci., Part A: Polym. Chem. 2005, 43,
polymerize by any living mechanism. Recently,
our
5
283-5299.
laboratory reported the metal-catalyzed LRP of VC initiated by
CHI3 and mediated by a competition between outer-sphere
(
4) (a) Percec, V.; Grigoras, C. J. Polym. Sci., Part A: Polym. Chem. 2005,
4
3, 5609-5619. (b) Kubisa, P. J. Prog. Polym. Sci. 2004, 29, 3-12. (b)
Kubisa, P. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 4675-4683.
5) Wayland, B. B.; Poszmik, G.; Mukerjee, S. L.; Fryd, M. J. Am. Chem.
Soc. 1994, 116, 7943-7944.
(
(8) (a) Fischer, H. Chem. ReV. 2001, 101, 3581-3610. (b) Studer, A. Chem.s
Eur. J. 2001, 7, 1159-1164.
(
(
6) Barboiu, B.; Percec, V. Macromolecules 2001, 34, 8626-8636.
(9) Queffelec, J.; Gaynor, S. G.; Matyjaszewski, K. Macromolecules 2000,
33, 8629-8639.
7) (a) Percec, V.; Barboiu, B.; Grigoras, C.; Bera, T. K. J. Am. Chem. Soc.
2
003, 125, 6503-6516. (b) Percec, V.; Grigoras, C.; Kim, H.-J. J. Polym.
(10) (a) Stockland, R. A.; Jordan, R. F. J. Am. Chem. Soc. 2000, 122, 6315-
6316. (b) Stockland, R. A.; Foley, S. R.; Jordan, R. F. J. Am. Chem. Soc.
2003, 125, 796-809. (c) Foley, S. R.; Stockland, R. A.; Shen, H.; Jordan,
R. F. J. Am. Chem. Soc. 2003, 125, 4350-4361.
Sci., Part A: Polym. Chem. 2004, 42, 505-513. (c) Percec, V.; Grigoras,
C.; Bera, T. K.; Barboiu, B.; Bissel, P. J. Polym. Sci., Part A: Polym.
Chem. 2005, 43, 4894-4906.
14156
9
J. AM. CHEM. SOC. 2006, 128, 14156-14165
10.1021/ja065484z CCC: $33.50 © 2006 American Chemical Society

Ultrafast Synthesis of Ultrahigh MW Polymers
A R T I C L E S
Scheme 1. Mechanism of SET-LRP
Table 1. EHOMO of Various Cu Catalysts Calculated Using HF/DFT
Methods
E
HOMO
(eV)
catalyst
method
-
-
-
-
-
-
-
7.00
7.34
7.62
7.67
8.26
8.84
9.71
Cu2Te
Cu2Se
Cu(0)
Cu2S
Cu2O
CuI
HF/LACVP+*//B3LYP/LACVP*
HF/6-31+G*//B3LYP/6-31G*
UHF/6-31+G*//UB3LYP/6-31G*
HF/6-31+G*//B3LYP/6-31G*
HF/6-31+G*//B3LYP/6-31G*
HF/6-31+G*//B3LYP/6-31G*
HF/6-31+G*//B3LYP/6-31G*
HF/LACVP+*//B3LYP/LACVP*
CuBr
CuCl
-
10.24
electron-acceptor alkyl halide, sulfonyl halide, or N-halide
1
1a,b
initiator.
Subsequently, the Cu(I) generated in this step
instantaneously disproportionates into Cu(II) and Cu(0) species.
The disproportionation of Cu(I) in H2O has been known for
over 100 years (eq 1) and proceeds in the presence of chelating
compounds with an equilibrium constant (Kdis, eq 2) of about
single-electron transfer (SET)11d and degenerative chain transfer
(DT) mechanisms (SET-DTLRP). This polymerization proceeds
at 25 °C in H2O and uses Cu(0) and/or “nascent” Cu(0)
generated in situ by the disproportionation of various Cu(I)
precursors as a catalyst. In this polymerization, Cu(0) species
act as electron donors, and the initiator and dormant propagating
species act as electron acceptors. The Cu(I) species generated
during the formation of radicals spontaneously disproportionate
into extremely reactive nascent Cu(II) and Cu(0) atomic species
that mediate the initiation and the reversible termination. This
disproportionation generates, by a self-regulated mechanism, in
situ, the Cu(II) species that, in the case of VC, would not be
accessible by a conventional PRE mechanism since the radical
polymerization of VC is dominated by chain transfer to
7
12
10 .
Kdis
2
Cu(I)X { } Cu(0) + Cu(II)X2
(1)
(2)
-2
6
7
Kdis ) [Cu(II)X ] [Cu(I)X] ) 10 to 10
2
Previously, we have shown that Cu(I) spontaneously dispro-
portionates in H2O in the presence of tris(2-aminoethyl)amine
11a
(
TREN) and poly(ethylene imine) (PEI). The very reactive,
nascent Cu(II) species generated from the disproportionation
reaction provide the reversible deactivation (kdeact) of the radicals
into dormant alkyl halide species. Cu(0) promotes the reactiva-
tion of the dormant species. Both processes occur by an SET
mechanism. To assess the electron-donating character of Cu(0)
and various Cu(I) species, their EHOMO values were calculated.
EHOMO correlates with the ionization potential (IP) and therefore,
estimates the electron-donor ability of various Cu derivatives.
Table 1 reports the EHOMO in electronvolts (eV) for Cu(0), CuCl,
CuBr, CuI, Cu2O, Cu2S, Cu2Se, and Cu2Te. Cu(0) is known to
1
c,11a,b
monomer
rather than bimolecular termination. By this
mechanism, the inactive Cu(I) species are spontaneously
consumed and the catalytically active Cu(0) species are continu-
ously produced. Here, we report that, under suitable conditions,
11a,b
the DT part of the SET-DTLRP
can be eliminated and the
newly elaborated LRP becomes SET-LRP. This polymerization
process takes place in H2O, protic, dipolar aprotic, and other
polar solvents that, in the presence of N-ligands (Supporting
Scheme SS1), were discovered to disproportionate Cu(I) into
Cu(0) and Cu(II). These solvent and ligand combinations also
favor an SET process.1 SET-LRP occurs under very mild
reaction conditions, at room temperature and below, uses a
catalytic rather than a stoichiometric amount of catalyst, and,
although proceeds ultrafast, generates polymers with unprec-
edently high molecular weight. SET-LRP is general and applies
to both nonactivated and activated monomers containing
electron-withdrawing functional groups, such as vinyl chloride
and other halogenated monomers, acrylates, and methacrylates.
It also applies to organic reactions and tolerates a diversity of
functional groups.
14
be an efficient single-electron donor agent. In addition, Cu2O,
Cu2S, Cu2Se, and Cu2Te were investigated because our labora-
tory has developed them as the most efficient self-regulated
catalysts for LRP initiated with sulfonyl halides and N-halides
1c
2c,d,3,7,15e
6
4a
in nonpolar
and polar solvents, in ionic liquids, and
11a
for SET-DTLRP of VC in H2O. Their mechanism of catalysis
for any of the previously reported LRP reactions, however, is
not known. CuCl, CuBr, and CuI were studied because,
depending on the structure of the initiator, they are expected to
be transient, inactive species, as outlined in the reaction
mechanism in Scheme 1. The results from Table 1 are
(
12) (a) Luther, R. Z. Physik. Chem. 1901, 36, 385. (b) Fenwick, F. J. Am.
Chem. Soc. 1926, 48, 860-870. (c) Tindall, G. W.; Bruckenstein, S. Anal.
Chem. 1968, 40, 1402-1404. (d) Ciavatta, L.; Ferri, D.; Palombari, R. J.
Inorg. Nucl. Chem. 1980, 42, 593-598. (e) Solari, E.; Latronico, M.; Blech,
P.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Inorg. Chem. 1996, 35, 4526-
Results and Discussion
Selecting Catalysts. The mechanism proposed for SET-LRP
is outlined in Scheme 1.
The initiation (activation) step (kact) is mediated by an SET
from the Cu(0) electron donor (or other donor species) to the
4528. (f) Hataway, B. J. In ComprehensiVe Coordination Chemistry;
Wikinson, G., Gillard, R. B., McCleverty, J. A., Eds.; Pergamon: Oxford,
U.K., 1987; Vol. 5, pp 533-774. (g) Shriver, D.; Atkins, P. Inorganic
Chemistry, 3rd ed.; Freeman, New York, 1999; pp 195-196.
13) (a) Kong, J. et al. J. Comput. Chem. 2000, 21, 1532-1548. (b) Gillies, M.
B.; Matyjaszewski, K.; Norrby, P.-O.; Pintauer, T.; Poli, R.; Richard, P.
Macromolecules 2003, 36, 8551-8559.
(
(
11) (a) Percec, V.; Popov, A. V.; Ramirez-Castillo, E.; Monteiro, M.; Barboiu,
B.; Weichold, O.; Asandei, A. D.; Mitchell, C. M. J. Am. Chem. Soc. 2002,
1
24, 4940-4941. (b) Percec, V.; Popov, A. V.; Ramirez-Castillo, E.;
(14) (a) Chen, Q.-Y. Isr. J. Chem. 1999, 39, 179-192. (b) Sato, K.; Nakazato,
S.; Enko, H.; Tsujita, H.; Yamamoto, T.; Omote, M.; Ando, A.; Kumadaki,
I. J. Fluorine Chem. 2003, 121, 105-107. (c) Sato, K.; Ogawa, Y.; Tamura,
M.; Harada, M.; Ohara, T.; Omote, M.; Ando, A.; Kumadaki, I. Collect.
Czech. Chem. Commun. 2002, 67, 1285-1295. (d) Brace, N. O. J. Fluorine
Chem. 2001, 108, 147-175.
Weichold, O. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 3283-3299.
c) Bunnett, J. F.; Scamehorn, R. G.; Traber, R. P. J. Org. Chem. 1976,
1, 3677-3682. (d) Jordan, R. B. Reaction Mechanisms of Inorganic and
Organometalic Systems; Oxford University Press: Oxford, U.K., 1991; pp
68 and 183.
(
4
1
J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006 14157

A R T I C L E S
Percec et al.
Deoxygenated mixtures containing solvent, ligand, and Cu-
(I)X or Cu(II)X2 were prepared, and their UV-vis spectra were
recorded after 10 min in order to allow the sedimentation of
Cu(0) at the bottom of the UV-vis cell, unless otherwise
11a
noted. The Cu(II)X2 solution was used as a standard to assess
the disproportionation of Cu(I)X. Figure 1a shows that, in the
absence of a ligand, CuBr does not disproportionate in DMSO.
However, complete and instantaneous disproportionation of
CuBr into CuBr2 and Cu(0) occurs in the presence of tris(2-
dimethylaminoethyl)amine (Me6-TREN) in DMSO (Figure 1b).
When Me6-TREN was replaced with 2,2′-bipyridine (bpy),
disproportionation of CuCl was also observed in DMSO, but it
was not as fast as that in the presence of Me6-TREN in DMSO
(Figure 1c). A survey of different combinations of ligands and
solvents is available in Supporting Figures SF1 to SF28. Me6-
TREN, TREN, PEI, bpy, N,N,N′,N′′,N′′-pentamethyldiethylene-
triamine (PMDETA), and many other N-ligands disproportionate
1
1a,12
Cu(I)X into Cu(II)X2 and Cu(0), not only in H2O
but also
in protic solvents (MeOH, EtOH, ethylene glycol, diethylene
glycol, triethylene glycol, 2-(2-ethoxyethoxy)ethanol, tetra-
ethylene glycol, glycerine, HEMA, phenols), dipolar aprotic
4
solvents (DMSO, DMF, DMAc, NMP, etc.), ionic liquids, and
ethylene and propylene carbonate, but not in CH3CN which is
a good stabilizing ligand for Cu(I)X and, therefore, does not
favor disproportionation. However, N-n-propyl-2-pyridyl-meth-
16
animine in DMSO does not disproportionate CuBr (Figure 1d).
This suggests that, from the ligands investigated (SS1), only
1
6
Haddleton’s ligand may mediate an inner-sphere “ATRP”
metal-catalyzed LRP in both nonpolar and polar solvents.
Therefore, any combination of catalyst, ligand, and solvent that
favored disproportionation of Cu(I) into Cu(0) and Cu(II) in
this study is expected to be efficient in mediating SET-LRP.
Figure 1. UV-vis spectra of (a) CuBr2 (0.01 mmol/mL, red) and CuBr
Selecting Initiators. The inner-sphere atom transfer
(
0.01 mmol/mL, green) in DMSO (both CuBr2 and CuBr are soluble in
1
f,g,h,11d
DMSO without ligand); (b) CuBr2/Me6-TREN (0.01 mmol/mL, red) and
CuBr/Me6-TREN (0.01 mmol/mL, green) in DMSO; (c) CuCl2/bpy, 1/3
mechanism
is known to show a strong dependence of
the metal-catalyzed kact on the nature of the halogen from the
(0.005 mmol/mL, red) and CuCl/bpy, 1/3 (0.005 mmol/mL, green/blue) in
structure of R-X. In two independent series of experiments
DMSO at 25 °C after 12 h (blue) and at 70 °C (green); (d) CuBr2/N-n-
propyl-2-pyridylmethanimine, 1/2 (0.01 mmol/mL solution) and CuBr/N-
n-propyl-2-pyridylmethanimine (0.005 mmol/mL, green/blue) in DMSO at
3
the ratio kact(R-Br)/kact(R-Cl) was in the range of 10 to 9 ×
4 17
1
0 . However, in the case of the outer-sphere mechanism, kact
2
5 °C after 12 h (green) and at 70 °C after 12 h (blue).
has a much smaller dependence on the nature of the halogen
from R-X. For example the ratio, kact(R-I)/kact(R-Br) ≈
1
8
kact(R-Br)/kact(R-Cl) ≈ 1 to 10. Based on these data, it is
expected that the range of initiators available for the case of
metal-mediated SET-LRP must be much broader than that for
the case of the metal-mediated inner-sphere process. Alkyl
halides, sulfonyl halides, and N-halides containing Cl, Br, and
I as the halide must be efficient in the case of SET-LRP
rewarding. According to the data from Table 1, the most efficient
electron-donor SET catalysts are Cu2Te followed by Cu2Se, Cu-
(
0), Cu2S, and Cu2O. Under the polymerization conditions these
11a
Cu derivatives disproportionate, and therefore, their role in
catalysis is determined by the ratio between their rate of
disproportionation and direct activation by SET.
(Supporting Scheme SS2). In the case of metal-catalyzed inner-
CuI, CuBr, and, specifically, CuCl are very poor electron
donors although they are catalysts of choice for inner-sphere
sphere ATRP, iodine-containing initiators undergo a competition
1a-f
ATRA and ATRP.
Since the lifetime of Cu(I)X is very short
(
15) (a) Percec, V.; Barboiu, B.; van der Sluis, M. Macromolecules 1998, 31,
053-4053. (b) Percec, V.; Asandei, A. D.; Asgarzadeh, F.; Bera, T. K.;
Barboiu, B. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 3839-3843.
c) Percec, V.; Barboiu, B.; Bera, T. K.; van der Sluis, M.; Grubbs, R. B.;
in comparison with that of Cu(0) and Cu(II)X2 species, during
reaction conditions that favor disproportionation of Cu(I) into
Cu(II) and Cu(0), the SET catalytic activity of Cu(I)X halides
can be neglected. Therefore, Cu2Te, Cu2Se, Cu(0), Cu2S, and
Cu2O (but not Cu(I)X) are expected to provide the most efficient
catalysts for the SET-LRP.
4
(
Fr e´ chet. J. M. J. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 4776-
4791. (d) van der Sluis, M.; Barboiu, B.; Pesa, N.; Percec, V. Macromol-
ecules 1998, 31, 9409-9412. (e) Percec, V.; Kim, H.-J.; Barboiu, B.
Macromolecules 1997, 30, 6702-6705. (f) Robello, D. R.; Andr e´ , A.;
McCovic, T. A.; Krana, A.; Mourey, T. H. Macromolecules 2002, 35,
9334-9344.
Selecting Ligands and Solvents. A UV-vis spectroscopy
study was performed to assess the activity of various solvents
(16) Haddleton, D. M.; Jasieczek, C. B.; Hannon, M. J.; Shooter, A. J.
Macromolecules 1997, 30, 2190-2193.
(
17) (a) Howes, K. R.; Bakac, A.; Espenson, J. H. Inorg. Chem. 1988, 27, 3147-
3151. (b) Matyjaszewski, K.; Paik, H.-j.; Zhou, P.; Diamanti, S. J.
Macromolecules 2001, 34, 5125-5131.
(in addition to H2O) and ligands for the disproportionation of
Cu(I) into Cu(II) and Cu(0). Representative examples of UV-
vis experiments are shown in Figure 1.
(
18) Marzilli, L. G.; Marzilli, P. A.; Halpern, J. J. Am. Chem. Soc. 1970, 92,
5752-5753.
14158 J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006

Ultrafast Synthesis of Ultrahigh MW Polymers
A R T I C L E S
for [MBP]0/[Cu(0)]0 ) 1/1 (mole/mole) (Figure 2a, b). The
reaction time is similar when [MBP]0/[Cu(0)]0 ) 1/0.1 (mole/
mole) (Figure 1c, d) and even with some excess of externally
added CuBr2 (in addition to the one generated by dispropor-
tionation) (Figure 2e, f). In all cases, kinetic plots show an
internal first order of reaction in monomer and growing species,
and the resulting poly(methyl acrylate) (PMA) has a low
molecular weight distribution. Additional examples with MBP
and BPN initiators are depicted in SF29, SF30, and SF31. These
figures also describe the LRP of several different acrylates. The
amount of catalyst can be decreased from 100% versus the
initiator to 10% and even to 1%. When it contains less than
1
0% catalyst, the reaction mixture is colorless.
External Orders of the Polymerization in [Cu(0)]0, [Cu-
Br2]0, and [DMSO]0. Kinetic experiments, processed according
2
b
to the method used previously, provided the external orders
of reaction of the polymerization in [Cu(0)]0, [DMSO]0, and
[CuBr2]0 (Figure 3). As expected for the mechanism illustrated
in Scheme 1, a complex, less than one, order of reaction in
[
[
Cu(0)]0 is observed. The orders of reaction in [DMSO]0 and
CuBr2]0 are about one. This demonstrates the expected catalytic
effect of DMSO in this reaction and establishes that the LRP
may be performed at a reasonably high rate in the presence of
a high concentration of CuBr2. These data provide instructions
on how to control the rate of polymerization. For example, the
amount of DMSO used in these experiments can be less than
the amount of monomer. Therefore, the reaction mixture can
be diluted with only a catalytic amount of solvent and be
performed almost in bulk.
Synthesis of Ultrahigh Molecular Weight PMA by Cu-
(
0)/Me6-TREN Catalyzed LRP Initiated with MBP. The
Figure 2. Kinetic plots for the Cu(0)-mediated living radical polymerization
of methyl acrylate (MA) at 25 °C in DMSO, initiated with methyl
simplest test of the efficiency of a living polymerization is to
assess its ability in the synthesis of polymers with ultrahigh
molecular weight, i.e., with number average molecular weight
2
)
-bromopropionate (MBP) using Me6-TREN as a ligand (conditions: MA
2 mL, DMSO ) 1 mL, [MA]0 ) 7.4 mol/L, and [MA]0/[MBP]0 ) 222).
6
(Mn) larger than 10 . Previously the highest molecular weight
1c
between inner-sphere and degenerative transfer that is expected
to be eliminated in the case of the outer-sphere SET-LRP
process.
linear polymers synthesized by the inner-sphere LRP metal-
catalyzed technique were Mn ) 300 000 for poly(n-butyl
methacrylate), Mn ) 367 000 for poly(methyl methacrylate),
Mn ) 554 000 for poly(methyl acrylate),
15e
20a
SET-LRP of Acrylates Initiated with Methyl-2-bromopro-
pionate (MBP) and 2-Bromopropionitrile (BPN) and Cata-
lyzed with Cu(0)/Me6-TREN in DMSO. Cu(0)/bpy-catalyzed
LRP of acrylates initiated with sulfonyl halides in nonpolar
5,20b
and Mn ) 823 000
2
0c
for poly(tert-butyl acrylate).
Figure 4 shows representative examples of kinetic experi-
ments in which [MA]0/[MBP]0 ) 11 000 (Figure 4a-d) and
2
1
15a,d
solvents at high temperatures was reported by our laboratory
2 200 (Figure 4e, f). PMAs with Mn ) 690 000 (Mw/Mn )
.10), 950 000 (Mw/Mn ) 1.12), and 1 420 000 (Mw/Mn ) 1.15),
and was shown to be accelerated by many additives, including
ethylene glycol.1 Mixtures of Cu(0)/Cu(II)X2 or even Cu(0)
alone were used to catalyze the LRP initiated with alkyl halides
5a
respectively, with a narrow molecular weight distribution were
obtained in 3, 6, and 10 h, respectively. The higher molecular
weight polymers obtained by SET-LRP suggests significantly
less termination and a much higher rate of polymerization at
room temperature than those in previous metal-catalyzed LRP
performed in nonpolar solvents at higher temperature.1
These are remarkable results considering that the reaction
conditions used were not optimized. Radical polymerizations
and LRPs of acrylates are known to be accompanied by
intramolecular chain transfer to polymer and, therefore, to
produce branched polyacrylates when the reaction is performed
1
9
at high temperature in bulk and in nonpolar solvents. It was
concluded that Cu(0) increases the rate of the reaction since it
reduces Cu(II)X2 to Cu(I)X, the latter being considered the
catalyst.1
9a,b,d
Cu(0) alone was shown to provide poor control
a-f,2,3
of molecular weight and molecular weight distribution since
the Cu(I)X species generated cannot react with radicals to
produce the dormant alkyl halides.1
9a,b
Figure 2 shows examples of LRP of methyl acrylate (MA)
catalyzed by MBP/Cu(0)/Me6-TREN carried out at 25 °C in
DMSO. Complete conversions are obtained in less than 50 min
2
1
at high temperatures. The analysis of the HSQC and HMBC
(
19) (a) Matyjaszewski, K.; Coca, S.; Gaynor, S. G.; Wei, M.; Woodworth, B.
E. Macromolecules 1997, 30, 7348-7350. (b) Woodworth, B. E.; Metzner,
Z.; Matyjaszewski, K. Macromolecules 1998, 31, 7999-8004. (c) Cheng,
G. L.; Hu, C. P.; Ying, S. K. J. Mater. Catal. A: Chem. 1999, 144, 357-
(20) (a) Xue, L.; Agarwal, U. S.; Lemstra, P. J. Macromolecules 2002, 35, 8650-
8652. (b) Wayland, B. B.; Basickes, L.; Mukerjee, S.; Wei, M.; Fryd, M.
Macromolecules 1997, 30, 8109-8112. (c) Percec, V.; Ramirez-Castillo,
E.; Popov, A. V.; Hinojosa-Falcon, L.; Guliashvili, T. J. Polym. Sci., Part
A: Polym. Chem. 2005, 43, 2178-2184.
3
62. (d) Johnson, R. M.; Ng, C.; Samson, C. C. M.; Fraser, C. L.
Macromolecules 2000, 33, 8618-8628.
J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006 14159

A R T I C L E S
Percec et al.
Figure 3. Determination of the external order of reaction in [Cu(0)]0, [CuBr2]0, and [DMSO]0 for the Cu(0)/Me6-TREN-catalyzed polymerization of methyl
app
acrylate (MA) in DMSO at 25 °C initiated with methyl 2-bromopropionate (MBP). [MA]0/[MBP]0 ) 200/1; (a) ln kp vs ln[Cu(0)]0, ratio [Cu(0)]0/[MA]0
app
was varied from 0.01 to 1 at an equal ratio of [Cu(0)]0/[ Me6-TREN]0 (MA ) 2 mL, DMSO ) 1 mL, [MA]0 ) 7.4 mol/L); (b) ln kp vs [DMSO]0, DMSO
app
was varied from 0.2 to 0.8 mL with 1.8 mL of MA. ([MA]0/[MBP]0/[Me6-TREN]0/[Cu(0)]0 ) 200/1/0.4/0.4); (c) ln kp vs -ln[CuBr2]0, ratio of [CuBr2]0/
[Cu(0)]0 was varied from 0.1 to 2, [Me6-TREN]0 was equal to sum of [Cu(0)]0 and [CuBr2]0. ([MA]0/[MBP]0/[Cu(0)]0 ) 200/1/0.4).
Figure 4. Kinetic plots for the Cu(0)-mediated living radical polymerization
of MA initiated with methyl 2-bromopropionate (MBP) at 25 °C in DMSO
using Me6-TREN as a ligand. Conditions: (a and b) MA ) 2.5 mL, DMSO
Figure 5. Kinetic plots for the Cu(0)-mediated living radical polymerization
of MA initiated from: CHCl3, CHBr3, CHI3 using Me6-TREN as a ligand.
Conditions: (a and b) MA ) 1 mL, DMSO ) 1 mL, [MA]0 ) 5.55 mol/L,
and [MA]0/[CHCl3]0 ) 222; (c and d) MA ) 1 mL, DMSO ) 1 mL, [MA]0
) 5.55 mol/L, and [MA]0/[CHBr3]0 ) 222; and (e and f) MA ) 2 mL,
DMSO ) 1 mL, [MA]0 ) 7.4 mol/L, and [MA]0/[CHI3]0 ) 222.
)
2.5 mL, [MA]0 ) 5.55 mol/L, and [MA]0/[MBP]0 ) 11 000; (c and d)
MA ) 5 mL, DMSO ) 2.5 mL, [MA]0 ) 7.4 mol/L, and [MA]0/[MBP]0
)
11 000; (e and f) MA ) 2 mL, DMSO ) 4 mL, [MA]0 ) 3.7 mol/L,
and [MA]0/[MBP]0 ) 22 200.
CHCl3, CHBr3, and CHI3 as Monofunctional and Bifunc-
tional Initiators for the LRP of MA Catalyzed by Cu(0),
Cu2Te, Cu2Se, Cu2S, Cu2O/Me6-TREN in DMSO. The
simplest monofunctional and difunctional initiators1 for the
synthesis of R,ω-difunctional telechetic polyacrylates with
different or identical chain ends are CHCl3, CHBr3, and CHI3.
Figure 5 shows kinetic experiments for all initiators with Cu-
(0)/Me6-TREN as the catalyst. Only CHCl3 requires a small
amount of additional CuCl2 to control the LRP. The most
important message provided by this figure is that the apparent
NMR spectra of linear poly(n-butyl acrylate) obtained at 25 °C
revealed no detectable branching in the polymer sample
1a,b
(
Supporting Figures SF57 and SF58). A linear PMA, free of
branching, was obtained in the polymerization executed at 25
C.
°
(
21) (a) Farcet, C.; Belleney, J.; Charleux, B.; Pirri, R. Macromolecules 2002,
35, 4912-4918. (b) Plessis, C.; Arzamendi, G.; Alberdi, J. M.; van Herk,
A. M.; Leiza, J. R.; Asua, J. M. Makromol. Chem., Rapid Commun. 2003,
4, 173-177.
3
2
14160 J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006

Ultrafast Synthesis of Ultrahigh MW Polymers
A R T I C L E S
The bifunctional CHI3 and the monofunctional CHCl3 and
CHBr3 initiators were also tested for the synthesis of ultrahigh
molecular weight linear PMA. The results of selected kinetic
experiments using these initiators are presented in Figure 6.
PMA with an Mn up to 1 400 000 and a narrow molecular
weight distribution is easily accessible in a reasonable reaction
time. The best results using CHCl3 as an initiator were obtained
when a small amount of CuCl2 additive was used to regulate
the polymerization. The nature of the haloform does not
app
significantly change the kinetic parameters (including kp ) of
these polymerizations under similar reaction conditions. This
observation was also valid (see above) when various haloforms
were employed for the synthesis of low molecular weight PMA
(Figure 5). The use of these easily available and inexpensive
solvents as dual mono- and bifunctional initiators for the
synthesis of ultrahigh molecular weight PMA could be attractive
from an industrial point of view and for the synthesis of AB,
ABA, and more complex block copolymers that are not
accessible with the more expensive monofunctional and bifunc-
tional initiators based on MBP.
Synthesis of Ultrahigh Molecular Weight 8- and 4-Arm
Star PMA. The Cu(0) based catalytic system was also utilized
for the synthesis of 4- and 8-arm star PMA. The bromopro-
1d
pionate based 4-armed (4BrPr derived from pentaerythritol)
and 8-armed (8BrPr derived from 4-tert-butyl calyx[8]arene)1d
Supporting Scheme SS2) initiators were used in these experi-
ments. Representative kinetic plots of LRP of MA initiated with
BrPr and 8BrPr star-shaped initiators are shown in Figures 7
(
4
and SF40 to SF42. In all experiments, the monomer conversion
reaches up to 90%. The plots of ln[MA]0/[MA]0 as a function
of time and those of experimental versus theoretical Mn increase
linearly. The Mw/Mn values remain very low throughout the
polymerization. These results indicate the living nature of these
polymerizations and the absence of star-star coupling as well
as of other side reactions. It is remarkable that both 4- and 8-arm
star PMA with an Mn up to 1 000 000 can be obtained in a
living manner at room temperature in a very short time (SF40
to SF42). The decreased catalyst concentration (Figure 7a and
Figure 6. Kinetic plots for the Cu(0)-mediated living radical polymerization
of MA initiated from CHBr3 using Me6-TREN as a ligand. Conditions:
MA ) 2 mL, DMSO ) 4 mL, [MA]0 ) 3.7 mol/L. (a and b) [MA]0/
[
CHBr3]0 ) 11 100; (c and d) [MA]0/[CHBr3]0 ) 22 200; (e and f) [MA]0/
CHCl3]0 ) 18 000).
[
rate constant of propagation (kpapp) does not depend strongly
on the structure of the haloform. Although this is a complex
rate constant, its trend resembles that observed in other SET-
b) as well as the use of a small amount of CuBr as an additive
2
18
mediated processes and differs from that seen in ATRA and
ATRP.1 Additional kinetic experiments mediated by Cu(0)
are shown in Figures SF32 and SF33, while kinetic experiments
mediated by Cu2O, Cu2S, Cu2Se, and Cu2Te are depicted in
Figures SF34, SF35, SF36, SF37, SF38, and SF39. Under these
reaction conditions, CHX3 with X ) Cl, Br provide monofunc-
tional initiators, while X ) I behaves as a bifunctional initiator
(Figure 7c and d) provide an excellent control of the polym-
,19
erization, yielding high molecular weight 4- and 8-arm PMA
with a low molecular weight distribution. Reports on the
synthesis of multiarm star polyacrylates and polymethacrylates
using the same 4BrPr and 8BrPr and other multifunctional
initiators under inner-sphere metal-catalyzed LRP conditions are
1d,7,15c,f
available.
All previous synthetic procedures required high
(
SF51 to SF58). The resulting PMAs are telechelic in the case
polymerization temperatures and long reaction times and were
performed in nonpolar solvents.
of CHCl3 (SF51) and CHBr3 (SF52, SF53), with their two chain
ends having different chemical functionalities (one derived from
initiator and one from growing species) and reactivities. When
iodoform is used as an initiator, the resultant polymer contains
two chain ends which at low conversion are different. However,
at high conversion and high [M]0/[I]0 ratio the two chain ends
exhibit identical structure and reactivity (SF55). A detailed
kinetic analysis of this process and demonstrations of its
synthetic capabilities will be reported.
LRP of MA Initiated with Haloforms and Catalyzed by
Cu(0), Cu2O, Cu2S, Cu2Se, Cu2Te/TREN in DMSO. The
simplest and the least expensive ligands that can be used in
this LRP are TREN and PEI. They were previously used in the
11a,b
LRP of VC in H2O together with the same catalysts.
Figure
8
provides several examples of kinetics that demonstrate the
very fast synthesis of different medium molecular weight (Figure
a, b) and very high molecular weight PMAs (Figure 8c, d).
8
Synthesis of Ultrahigh Molecular Weight PMA by Cu-
0)/Me6-TREN Catalyzed LRP Initiated with Haloforms. The
initiators CHBr3 and CHCl3 were used to demonstrate the
synthesis of ultrahigh molecular weight PMA (Figure 6).
This polymerization can be carried out in the presence or
absence of CuX2 (Figures 8, SF43, SF44). Catalysis by Cu2O,
Cu2S, Cu2Se, and Cu2Te/TREN in DMSO was demonstrated
by the kinetic experiments summarized in Figure SF45. The
(
J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006 14161

A R T I C L E S
Percec et al.
Figure 7. Kinetic plots for the living radical polymerization of MA initiated
with (a and b) (8BrPr)5,11,17,23,29,35,41,47-octa-tert-butyl-49,50,51,52,
Figure 9. Kinetic plots for the Cu(0)-mediated living radical polymerization
of MMA initiated with the following: (a and b) 2,2-dichloroacetophenone
5
)
1
3,54,55,56-octakis-(2-bromopropionyloxy) calyx[8]arene (conditions: MA
1 mL, DMSO ) 0.5 mL, [MA]0 ) 7.4 mol/L, and [MA]0/[8BrPr]0 )
1 100); (c and d) (4BrPr) pentaerythritol tetrakis(2-bromopropionate)
(
DCAP) (conditions: MMA ) 1 mL, DMSO ) 0.5 mL), [MMA]0 ) 6.2
mol/L, and [MMA]0/[DCAP]0 ) 200) using PMDETA as a ligand; (c and
d) phenoxybenzene-4,4′-disulfonyl chloride (PDSC) (conditions: MMA )
(
conditions: MA ) 2.5 mL, DMSO ) 2.5 mL, [MA]0 ) 5.55 mol/L, and
1
2
mL, NMP ) 0.5 mL, [MMA]0 ) 6.2 mol/L, and [MMA]0/[PDSC]0 )
00) using bpy as a ligand.
[MA]0/[4BrPr]0 ) 11 100) using Me6-TREN as a ligand.
experiments are presented in Figure 9. The first one is initiated
with DCAP and catalyzed by Cu(0)/PMDETA in DMSO (Figure
9
a, b). The second one is initiated with the bifunctional PDSC
(Supporting Scheme SS2) and catalyzed with Cu(0)/bpy in
NMP. Both experiments were performed at 25 °C. Since
sulfonyl halides are known to undergo side reactions in the
2
2a
21b
presence of DMSO and of aliphatic N-containing ligands,
the classic bpy ligand employed previously with sulfonyl halide
initiators
2
-5,7,15
was used. These two examples demonstrate a
dramatic acceleration when compared with related experiments
1
c,d,2-4,6,7,15
carried out in nonpolar solvents.
Additional kinetic
experiments are presented in Figures SF46 and SF47.
LRP of VC Initiated with CHBr3 and Catalyzed with Cu-
(0), Cu2O, Cu2S, Cu2Se, Cu2Te/TREN in DMSO. Previously,
the SET-DTLRP of VC initiated with CHI3 and catalyzed by
the same catalytic systems in H2O at 25 °C was reported.1
The resulting PVC contains two functional chain ends that were
used in further functionalization and block copolymerization
experiments. However, since CHI3 is a good chain transfer agent
and the polymerization mixture was heterogeneous, a two-stage
kinetic process leading to a limited VC conversion was observed.
CHBr3 would be a less reactive, more UV and thermally stable,
and less expensive initiator for the LRP of VC. Since reactivity
studies on MA have demonstrated little difference between the
nature of X from the structure of R-X and its kact, reaction
conditions for the LRP of VC initiated with CHBr3 were
investigated. Me6-TREN as ligand in conjunction with Cu(0)
mediates the polymerization of VC only to low conversion.
However, the replacement of Me6-TREN with TREN in DMSO
provided a rewarding result that is determined by the difference
between the ratio of the rates of activation/deactivation mediated
1a,b
Figure 8. Kinetic plots for the Cu(0)-mediated living radical polymerization
of MA initiated from CHBr3 using TREN as a ligand. Conditions: (a and
b) MA ) 2 mL, DMSO ) 1 mL, [MA]0 ) 7.4 mol/L, and [MA]0/[CHBr3]0
)
222; (c and d) MA ) 2 mL, DMSO ) 2 mL, [MA]0 ) 5.55 mol/L, and
[MA]0/[CHBr3]0 ) 11 100.
resulting PMA contains two different chain ends one derived
from initiator and one from growing species that have different
reactivities (SF52, SF54). This suggests new strategies for the
synthesis of multifunctional block copolymers and other com-
plex architectures.
LRP of MMA Initiated with 2,2-Dichloroacetophenone
(DCAP) and Phenoxybenzene-4,4′-Disulfonyl Chloride (PDSC)
Catalyzed by Cu(0)/PMDETA and bpy in Dipolar Aprotic
Solvents. Two examples of preliminary, nonoptimized kinetic
(
22) (a) Boyle, R. E. J. Org. Chem. 1966, 31, 3880-3882. (b) Gurr, P. A.;
Mills, M. F.; Qiao, G. G.; Solomon, D. H. Polymer 2005, 46, 2097-2104.
14162 J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006

Ultrafast Synthesis of Ultrahigh MW Polymers
A R T I C L E S
Figure 11. Kinetic plots for the living radical polymerization of VC initiated
with CHBr3 using TREN as a ligand (conditions: VC ) 2.2 g, DMSO )
2
mL, [VC]0 ) 8.8 mol/L, and [VC]0/[CHBr3]0 ) 350). * denotes Cu(0)
catalyst was used as a wire wrapped on the magnetic stirring bar.
Mechanistic Considerations. The experimental observations
that Cu(I)X formed in situ undergoes spontaneous dispropor-
tionation under the polymerization conditions and that most of
these polymerizations do not require externally added Cu(II)-
X2 deactivator demonstrate that the active catalyst is the nascent
Cu(0) atomic species (Scheme 1). When Cu2Te, Cu2Se, Cu2S,
and Cu2O are used, depending on the ratio between SET
initiation and disproportionation they also can act as initial
electron-donor catalysts. The rather interesting question is the
following: By what mechanism are the radicals formed under
these conditions? Several experimental observations as well as
literature data help to elaborate the most probable mechanistic
hypothesis for the activation of the dormant species. On one
app
hand, the kp during LRP of MA strongly depends on the
concentration of DMSO used in the polymerization mixture
(Figure 3b). Increasing DMSO concentration leads to faster
polymerization while maintaining good control over molecular
weight and molecular weight distribution. If the radicals were
formed through an inner-sphere homolytic atom-transfer process,
Figure 10. Kinetic plots for the Cu(0)-mediated living radical polymeri-
zation of VC initiated with CHBr3 using TREN as a ligand (conditions:
VC ) 2.2 g, DMSO ) 2 mL, [VC]0 ) 8.8 mol/L).
1c,f
like in the case of ATRA and ATRP, then the polymerization
by these two ligands. Figure 10 shows selected experiments of
preliminary experiments catalyzed with different concentrations
of Cu(0)/TREN and DMSO.
Without optimization, about 90% VC conversion can be
reached in a very short period of time. The molar ratio [CHBr3]0/
app
rate and, hence, the kp would have to exhibit very little
dependence on the polarity of the reaction medium. The rate of
a reaction in which radical intermediates are involved is not
strongly dependent on the polarity of the medium. In principle,
the opposite effect should be observed, i.e., the polymerization
rate must decrease when the monomer, catalyst, and/or initiator
concentrations decrease. This observation is an indication that
the formation of radicals occurs by a different mechanism. One
[Cu(0)]0 can be reduced from 1/1 (Figure 10 a, b) to 1/0.5
(
Figure 10 c, d) and to 1/0.1 (SF48) with higher rates than
1
1a,b
previously reported with CHI3 as the initiator.
Since the
amount of DMSO used in these polymerizations is very small,
at the end of the polymerization the reaction mixture contains
a white solid PVC plasticized with some DMSO. Therefore,
purification of the PVC requires only washing with H2O or
MeOH. The limited amount of Cu(0) powder required to mediate
this polymerization suggested that Cu(0) wire or any other solid
form of a Cu alloy would be sufficient to promote it. Indeed,
Figure 11 demonstrates this concept (compare Figure 10 a, b
with Figure 11 a, b). Higher molecular weight PVC was obtained
with higher [VC]0/[CHBr3]0 ratios in the presence of Cu(0)
2
3
option is that ionic intermediates are involved in the rate-
limiting step of the radical formation. In this case, the rate of
radical generation must dependent strongly on the polarity of
the reaction medium and this was indeed observed experimen-
tally (Figure 3b). The radicals are most likely generated through
the decomposition of radical-anion intermediates23 that are
initially formed by the SET reaction between Cu(0) species and
the halogen-containing substrates, such as the initiator, and
halogen-terminated polymeric chain ends. These considerations
are supported by literature data that show that dipolar aprotic
as well as protic solvents facilitate fast SET reactions between
electron donors (including Cu(0)) and various halogen-contain-
(
(
Figure 10 e, f and SF48), Cu2Te, Cu2Se, Cu2S, and Cu2O
SF49). All kinetic data demonstrate an SET-LRP single stage
process. Depending on the reaction conditions and conversion,
the resulting PVC can be designed as a bifunctional initiator
with two different or identical chain ends. Therefore, it can be
used for the synthesis of AB and ABA block copolymers (SF59
to SF65). This telechelic PVC can also be used for the synthesis
of new complex architectures based on PVC. Details of these
synthetic capabilities will be reported.
1
4,23
ing compounds (including haloforms and tetrahalomethanes).
1
3a
A series of quantum chemical calculations was performed
on model systems to verify the path that best describes the
(23) (a) Pause, L.; Robert, M.; Sav e´ ant, J.-M. J. Am. Chem. Soc. 2000, 122,
9
829-9835. (b) Cardinale, A.; Isse, A. A.; Gennaro, A.; Rosset, M.;
Sav e´ ant, J.-M. J. Am. Chem. Soc. 2002, 124, 13533-13539. (c) Costentin,
C.; Robert, M.; Sav e´ ant, J.-M. J. Am. Chem. Soc. 2003, 125, 10729-10739.
J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006 14163

A R T I C L E S
Percec et al.
Scheme 3. Proposed Catalytic Cycle of SET-LRP Mediated by
Scheme 2. Relative C-X Bond Dissociation Energies in Methyl
2
-Halopropionatesa
Cu(0) Species in Polar Media
to a similar ion cage composed of a radical-anion [R- - -X]•
-
+
and a CuX/L countercation. The mechanism depicted in
Scheme 3 was simplified in order to facilitate the understanding
of the main chemical events that might be involved in this
catalytic process. However, the goal of this report was not to
provide a complete mechanistic explanation of the SET-LRP
process mediated by nascent Cu(0) species but rather to develop
the SET-LRP synthetic method and to compare it with that based
on the inner-sphere process. More experimental and theoretical
studies are necessary to elucidate the hypothetic mechanism
outlined in Scheme 3. Investigations in this direction are in
progress and will be reported in due time. Nevertheless, the
SET-LRP reported here, together with its mechanistic hypoth-
esis, explains numerous literature examples in which an ac-
celeration of the “Cu(I)X”-catalyzed LRP initiated with alkyl
halides, sulfonyl halides, and N-halides was observed in
a
Homolytic bond dissociation (top diagram) and formation and decom-
position of anion-radicals formed by an SET process (bottom diagram)
calculated using (U)B3LYP/6-31+G*//(U)B3LYP/6-31+G* (for Cl and Br
case) or (U)B3LYP/LACVP+*// (U)B3LYP/LACVP+* (for I case)
methods.
24a-f
24g-j
15a,21j,k
24l-q
24p
H2O,
alcohols,
ethyleneglycol,
DMSO,
DMF,
4
a
experimental observation that the nature of the halogen does
not affect the rate of polymerization of MA when Cu(0) catalyst
is used under similar conditions. The details of the quantum
chemical calculations performed are described in the Supporting
Information, and the results are shown in Scheme 2.
and ionic liquids, including a recent example of high molar
mass polymethacrylate. This reaction will also facilitate access
to complex organic synthesis by radical reactions performed
24s
2
5
under mild conditions. Research along this line will be
reported.
It is obvious that the homolytic C-X bond dissociation
energy (BDE) in the initiators MCP (X ) Cl), MBP (X ) Br),
and MIP (X ) I) (Supporting Scheme SS2) strongly depends
on the nature of the halogen and usually increases in the
following order: Cl > Br > I (Scheme 2, top diagram). The
calculated BDEs are in good agreement with previously
calculated values obtained for the same methyl 2-halopropionate
systems. The C-X bond dissociation energy depends very
little on the nature of the halogen (X) if it occurs through the
decomposition of the radical-anion formed via a SET process
(
24) (a) Robinson, K. L.; Khan, M. A.; de Paz Banez, M. V.; Wang, X. S.;
Armes, S. P. Macromolecules 2001, 34, 3155-3158. (b) Wang, X.-S.;
Armes, S. P. Macromolecules 2000, 33, 6640-6647. (c) Wang, X.-S.;
Lascelles, S. F.; Jackson, R. A.; Armes, S. P. Chem. Commun. 1999, 1817-
1818. (d) Wang, X.-S.; Jackson, R. A.; Armes, S. P. Macromolecules 2000,
3
3, 255-257. (e) Zeng, F.; Shen, Y.; Zhu, S. S.; Pelton, R. J. Polym. Sci.,
Part A: Polym. Chem. 2000, 38, 3821-3827. (f) Bontempo, D.; Maynard,
H. D. J. Am. Chem. Soc. 2005, 127, 6508-6509. (g) Lobb, E. J.; Ma, I.;
Billingham, N. C.; Armes, S. P. J. Am. Chem. Soc. 2001, 123, 7913-
7914. (h) McDonald, S.; Rannard, S. P. Macromolecules 2001, 34, 8600-
1
3b
8
602. (i) Kimani, S. M.; Moratti, S. C. J. Polym. Sci., Part A: Polym.
Chem. 2005, 43, 1588-1598. (j) Bontempo, D.; Heredia, K. L.; Fish, B.
A.; Maynard, H. D. J. Am. Chem. Soc. 2004, 126, 15372-15373. (k) Perrier,
S.; Gemici, H.; Li, S. Chem. Commun. 2004, 604-605. (l) Percec, V.;
Guliashvili, T.; Popov, A. V. J. Polym. Sci., Part A: Polym. Chem. 2005,
43, 1948-1954. (m) Percec, V.; Guliashvili, T.; Popov, A. V.; Ramirez-
Castillo, E. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 1935-1947.
(Scheme 2, bottom diagram). The generalized catalytic cycle
of SET-LRP is depicted in Scheme 3.
The reaction starts with an SET reaction between Cu(0)
species and the halogen-containing substrate (initiator or
halogen-terminated polymeric chain end). Any polar solvent,
including DMSO, facilitates the decrease of interaction between
(
n) Percec, V.; Guliashvili, T.; Popov, A. V.; Ramirez-Castillo, E.; Hinojosa-
Falcon, L. A. Polym. Sci., Part A: Polym. Chem. 2005, 43, 1660-1669.
o) Percec, V.; Guliashvili, T.; Popov, A. V.; Ramirez-Castillo, E.; Coelho,
J. F. J.; Hinojosa-Falcon, L. A. J. Polym. Sci., Part A: Polym. Chem. 2005,
(
4
3, 1649-1659. (p) Bontempo, D.; Li, R. C.; Ly, T.; Brubaker, C. E.;
Maynard, H. D. Chem. Commun. 2005, 4702-4704. (q) Monge, S.; Darcos,
V.; Haddleton, D. M. J. Polym. Sci. Part A: Polym. Chem. 2004, 42, 6299-
6308. (r) Mao, B. W.; Gan, L. H.; Gan, Y. Y. Polymer 2006, 47, 3017-
-
•
the anion (X ) and the electrophilic radical (R ) from the
radical-anion pair.23 The radical-anion cluster and Cu/L+
3020.
countercation must be in close proximity in the so-called caged
(25) (a) Bellus, D. Pure Appl. Chem. 1985, 57, 1287-1838. (b) Pirrung, F. O.
H.; Hiemstra, H.; Speckamp, W. N.; Kaptein, B.; Schoemaker, H. E.
Synthesis 1995, 458-472. (c) Iqbal, J.; Bhatia, B.; Navyar, N. K. Chem.
ReV. 1994, 94, 519-564. (d) Kamigata, N.; Shimizu, T. ReV. Heteroat.
Chem. 1997, 17, 1-50. (e) Gossage, R. A.; van de Kuil, L. A.; van Koten,
G. Acc. Chem. Res. 1998, 31, 423-431. (f) Clark, A. J. Chem. Soc. ReV.
2002, 31, 1-11. (g) Delaude, L.; Demonceau, A.; Noels, A. F. Top.
Organomet. Chem. 2004, 11, 155-171. (h) Severin, K. Curr. Org. Chem.
ion pair, thus further facilitating decomposition of the radical-
-
anion into the electrophilic radical and the anion (X ). The
reaction between a radical and the deactivator CuX2/ligand in
polar media is a more complex process. One of the possible
-
mechanisms may involve the transfer of the halide anion X
2
006, 10, 217-224. (i) Quebatte, L.; Thommes, K.; Severin, K. J. Am.
•
from the deactivator to the propagating macroradical R , leading
Chem. Soc. 2006, 128, 7440-7441.
14164 J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006

Ultrafast Synthesis of Ultrahigh MW Polymers
A R T I C L E S
Conclusions
drawing groups. Without optimization, polymers with M in the
n
6
range of 1.5 × 10 with a narrow molecular weight distribution
The disproportionation of Cu(I)X species into Cu(0) and Cu-
are obtained in a short reaction time with no purification. This
reaction also applies to the polymerization of a diversity of
functional monomers and is expected to facilitate the rapid
(II)X2 species in alcohols, dipolar aprotic solvents, ethylene and
propylene carbonate, and ionic liquids in the presence of a
diversity of N-containing ligands is reported. This discovery
complements the well-known disproportionation of Cu(I)X in
H2O. In water, alcohols, dipolar aprotic solvents, ethylene and
propylene carbonate, and ionic liquids, Cu(0) and its alloys as
powder, wire, or any other form (Figure 11, SF50), Cu2Se, Cu2-
Te, Cu2S, and Cu2O and the nascent, extremely reactive atomic
Cu(0) and Cu(II)X2 species obtained by disproportionation all
mediate an ultrafast LRP of acrylates, methacrylates, and vinyl
chloride at room temperature and below, initiated with alkyl
halides, sulfonyl halides (Figure 9 and SF47), and N-halides
7
25
synthesis of complex macromolecular and organic compounds
under mild conditions. Since activation and deactivation steps
in SET-LRP are expected to be faster than chain transfer to
initiators and/or dormant species of the general structure R-X
with X ) Cl, Br, I, all these initiators can be used without
competition with degenerative chain transfer. Preliminary
simulations of the SET-LRP kinetic experiments demonstrated
that, under suitable reaction conditions, the extent of bimolecular
termination is below the level detectable by any available
analytical method. Recently, accelerated methods for the inner-
sphere metal-catalyzed ATRP process were also reported.26
(SF47). This polymerization proceeds by an outer-sphere SET
mechanism in which Cu(0) species act as electron donors, and
the dormant initiator and propagating R-X species act as
electron acceptors. By contrast with other metal-catalyzed LRPs,
including ATRP, in which excess Cu(II)X2 species are created
by the undesired radical dimerization, here, the Cu(II)X2 species
responsible for the reversible deactivation of the radicals are
formed by the disproportionation of Cu(I)X. The previously
reported metal-catalyzed LRP, including ATRP, generate radi-
cals by an inner-sphere process that requires a high activation
energy. The outer-sphere SET process involved in this new
polymerization has a very low activation energy. Therefore,
SET-LRP involves very fast activation and deactivation steps
and negligible bimolecular termination at room temperature.
Since activation takes place at room temperature in the presence
of excess Cu(II)X2, an extremely good control of the reaction
is accomplished with only a very small amount of catalyst. This
provides an ultrafast polymerization of both activated and
nonactivated functional monomers containing electron-with-
Acknowledgment. Financial support by the National Science
Foundation (DMR-0548559 and DMR-0520020) and the P. Roy
Vagelos Chair at Penn are gratefully acknowledged. A.W. and
A.S. acknowledge the Swedish Foundation for Strategic Re-
search (Grant A 302:132) for financial support during their study
at Penn.
Supporting Information Available: Experimental section
containing materials, techniques, synthesis, structural analysis,
additional kinetic experiments, and method used in the quantum
chemical calculations. Also complete ref 13a. This material is
available free of charge via the Internet at http://pubs.acs.org.
JA065484Z
(
26) (a) Min, K.; Gao, H.; Matyjaszewski, K. J. Am. Chem. Soc. 2005, 127,
825-3830. (b) Jakubowski, W.; Min, K.; Matyjaszewski, K. Macromol-
ecules 2006, 39, 39-45.
3
J. AM. CHEM. SOC.
9
VOL. 128, NO. 43, 2006 14165