A dual catalyst system for atom transfer radical polymerization based on a halogen-free neutral Cu(I) complex

Article abstract of DOI:10.1021/ma034979v

A neutral phenoxy-triamine copper(I) complex was synthesized, and its performance as a catalyst for the ATRP of n-butyl acrylate (BA) was investigated. This halogen-free neutral cuprous complex catalyzed BA polymerization, but while the polymerization rat

Full text of DOI:10.1021/ma034979v

7432
Macromolecules 2003, 36, 7432-7438
A Dual Catalyst System for Atom Transfer Radical Polymerization
Based on a Halogen-Free Neutral Cu(I) Complex
Yosh ih isa In ou e a n d Kr zysztof Ma tyja szew sk i*
Center for Macromolecular Engineering, Department of Chemistry, Carnegie Mellon University,
4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213
Received J uly 11, 2003; Revised Manuscript Received August 12, 2003
ABSTRACT: A neutral phenoxy-triamine copper(I) complex was synthesized, and its performance as a
catalyst for the ATRP of n-butyl acrylate (BA) was investigated. This halogen-free neutral cuprous complex
catalyzed BA polymerization, but while the polymerization rate was high there was poor control over
polymerization. The level of control over the polymerization process was improved dramatically when a
small amount of a suitable copper(II) complex was added as a deactivator. Several copper(II) complexes
including CuBr2, dNbpy/CuBr2, PMDETA/CuBr2, and Me6TREN/CuBr2 were investigated as deactivators.
Among these deactivators, Me6TREN/CuBr2 was the most efficient at improving the level of control over
the polymerization. When 3 mol % of Me6TREN/CuBr2 (versus the phenoxy-triamine copper complex)
was added, the molecular weight increased linearly with conversion and agreed with theoretical values,
and the polymer displayed a narrow molecular weight distribution (Mw/Mn ) 1.17). The mechanism of
the dual catalyst system might be similar to that proposed earlier for another binary catalyst system,
the immobilized/soluble hybrid catalyst system. The high deactivation rate constant and high reducing
power of Me6TREN complex might be the key properties that allow successful control over the
polymerization in this dual catalyst system.
In tr od u ction
ligands possess a multidentate structure and neutral
charge; therefore, counteranions must be present to
neutralize the positive charge of the copper in complexes
formed with these ligands. Halides such as chloride and
bromide are the most common counterions for copper-
mediated ATRP, though hexafluorophosphate (PF6),24
triflate (OTf),25 tetraphenylborate (BPh4),26 carboxylate
(e.g., OAc),27 thiocyanate (SCN),28,29 and cuprous di-
halide (e.g., [CuBr2]-) have also been used as counter-
anions. Sometime (pseudo)halogens may bind directly
to copper in certain neutral tridenate complexes, e.g.,
complexes formed with PMDETA.30 Ligands/counterions
may participate in exchange reactions with other com-
ponents of the polymerization system. In some cases
these exchange reactions are beneficial, e.g., a “halogen
exchange” technique which significantly improves con-
trol and efficiency of block copolymer formation,31 but
sometimes, halogens may be displaced by other ligands,
e.g., water, leading to reduced control over the poly-
merization. Thus, appropriate selection of the counter-
anion is critical for proper control in ATRP.
While many efficient copper-mediated catalyst sys-
tems have been identified, there are several reasons to
continue the considerable effort that has been directed
at the development of new catalysts for ATRP. These
include development of highly active catalysts, less
expensive catalysts, catalysts suitable for use in inor-
ganic media including water, and catalysts that can
polymerize monomers that are presently difficult to
polymerize using existing ATRP catalysts.
Controlled/living radical polymerization (CRP)1-4 has
been one of the most noteworthy research fields in
polymer chemistry in the past decade, attracting in-
creased academic and industrial interest. CRP provides
access to materials with complex polymer architectures,
some of which have never been synthesized before via
conventional radical polymerization. Among the several
CRP techniques examined, atom transfer radical poly-
merization (ATRP)5-12 is one of the most promising
synthetic methods that can be applied to the copoly-
merization of a variety of monomers and provides well-
defined polymers with controlled topology and function-
ality. Several middle and late transition metal com-
plexes have been investigated as ATRP catalysts;
however, copper complexes are among most successful
for ATRP due to their availability, versatility, and low
cost.5,13,14
In ATRP, the activity of the catalyst, and the degree
of control over the polymerization process, strongly
correlate with the redox potentials and halogenophilicity
of the metal center.15 These primary properties of the
catalyst are mostly determined by the ligands and
counteranions present in the complexes. Therefore, the
selections of ligands and counterions are key in con-
ducting a successful ATRP reaction. In copper-mediated
ATRP, highly active catalysts capable of conducting
well-controlled polymerizations have been prepared
using a broad spectrum of nitrogen-containing ligands,
including pyridine derivatives (e.g., 4,4′-(di-5-nonyl)-2,2′-
bipyridine (dNbpy)16), linear, branched, or cyclic ali-
phatic polyamines (e.g., N,N,N′,N′′,N′′-pentamethyl-
diethylenetriamine (PMDETA)17 and tris(2-(dimethyl-
amino)ethyl)amine (Me6TREN)18),19 imine derivatives,20
picolylamine derivatives,21 phenanthroline derivatives,22
and terpyridine derivatives.23 All of these suitable
There have been only a few attempts to use negatively
charged ligands, which should provide neutral Cu(I)
complexes as ligands for ATRP. Such complexes should
have higher stability due to a strong bond between the
ligand and the metal center. This strong bond should
also affect the ATRP equilibrium and could enhance the
polymerization rate. Furthermore, charged ligands pro-
vide a counterion-free copper complex. Herein, we report
the catalytic performance of a new neutral copper(I)
* Corresponding author. E-mail: km3b@andrew.cmu.edu.
10.1021/ma034979v CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/13/2003

Macromolecules, Vol. 36, No. 20, 2003
Dual Catalyst System for ATRP 7433
Sch em e 1. Syn th esis of th e P h en oxy-Tr ia m in e
hydride prior to use. All other reagents and solvents were used
as received.
Liga n d a n d th e Cop p er (I) Com p lex
Syn t h esis of 2,4-Dim et h yl-6-b is(2-(d iet h yla m in o)-
eth yl)a m in om eth ylp h en ol.36 Paraformaldehyde (95% pu-
rity, 0.51 g, 16.1 mmol) and N,N,N′,N′-tetraethyldiethylene-
triamine (90% purity, 3.87 g, 16.2 mmol) were added to a 50
mL round-bottom flask and stirred at 80 °C for 1 h. A solution
of 2,4-dimethylphenol (2.0 g, 16.2 mmol) in methanol (10 mL)
was then added to the resulting pale yellow oil, and this
solution was stirred for 24 h under gentle reflux. After the
reaction was complete, the methanol was removed through
evaporation and the residual oil was purified by column
chromatography with silica gel (CH2Cl2/methanol ) 9/1 vol).
After evaporation of the solvent, the desired compound was
obtained as an yellow oil in 70% yield. 1H NMR (300 MHz,
CDCl3): δ 0.96 (t, 12 H), 2.16 (s, 3 H), 2.17 (s, 3 H), 2.47 (q, 8
H), 2.60 (m, 8 H), 3.67 (s, 2 H), 6.59 (s, 1 H), 6.81 (s, 1 H).
Anal. Calcd for C21H39N3O: C, 72.16; H, 11.24; N, 12.02.
Found: C, 70.92; H, 11.43; N, 11.97.
Syn t h esis of 2,4-Dim et h yl-6-b is(2-(d iet h yla m in o)-
eth yl)a m in om eth ylp h en oxycop p er (I). Synthesis of the
copper complex was performed using standard Schlenk
techniques. 2,4-Dimethyl-6-bis(2-(diethylamino)ethyl)amino-
methylphenol (1.06 g, 3.0 mmol) was added to a 50 mL Schlenk
flask degassed under vacuum and placed under nitrogen. This
phenol compound was dissolved in acetonitrile (15 mL), and
then a solution of 2.5 N n-butyllithium in hexane (1.2 mL, 3.0
mmol) was added at 0 °C. After the addition of n-butyllithium,
the reaction mixture was allowed to warm to room tempera-
ture and stirred for 2 h.
complex bearing a phenoxy-triamine ligand (Scheme
1). We designed this complex as a potential catalyst
since it has both strong copper-oxygen bond and
branched triamine structure. One could anticipate high
complex stabilities due to the covalent nature of the
copper-oxygen bond and high catalytic activity due to
the branched triamine structure of the ligand that is
similar to the most active ATRP catalyst, Me6TREN/
CuBr.
This phenoxy-triamine copper complex does display
high activity in the polymerization of n-butyl acrylate
but, unfortunately, also provides poor control. However,
control over the polymerization was dramatically im-
proved by the addition of a small amount of a Cu(II)
deactivator, especially Me6TREN/CuBr2. This dual cata-
lyst system reminds us of a similar binary catalyst
system discussed earlier, the immobilized/soluble hybrid
catalyst system.32,33 At that time we reported that high
deactivation rate constant and high reducing power of
Me6TREN/CuBr2 enhanced the level of control obtained
with the immobilized/soluble hybrid catalyst system in
a range of ATRP polymerization reactions. A similar
mechanism might be expected to operate in the new
dual catalyst system formed with the new neutral
copper(I) complex bearing a phenoxy-triamine ligand
as primary activator. The plausible mechanism for this
dual catalyst system will be discussed.
The resulting yellow solution was added to a suspension of
CuCl (0.30 g, 3.0 mmol) in acetonitrile (10 mL) at 0 °C. A white
solid was formed instantly, and the solution became light
brown in color. The reaction mixture was allowed to warm to
room temperature and stirred for 3 h. After filtration, the
filtrate was concentrated to ca. 15 mL and stored at -20 °C
overnight, yielding a white powdery precipitate. This white
powder was collected through decantation and recrystallized
from acetonitrile to give the desired copper complex as colorless
1
crystals in 56% yield. H NMR (300 MHz, C6D6): δ 0.90 (t, 12
H), 2.1-2.9 (m, 16H), 2.38 (s, 3 H), 2.44 (s, 3 H), 3.82 (s, 2H),
6.86 (s, 1H), 7.14 (s, 1H). Anal. Calcd for C21H38CuN3O: C,
61.21; H, 9.29; N, 10.20. Found: C, 60.80; H, 9.34; N, 10.13.
Cu Br 2 Dea ct iva t or St ock Solu t ion . CuBr2 deactivator
stock solution was prepared just prior to use. CuBr2 (5.32 mg,
2.38 × 10-5 mol) was added to a 25 mL Schlenk flask, and
then the flask was evacuated and backfilled with nitrogen
three times. Acetone (10 mL) degassed by bubbling with
nitrogen for 30 min and, optionally, Me6TREN (6.28 µL, 2.38
× 10-5 mol), PMDETA (4.97 µL, 2.38 × 10-5 mol), or dNbpy
(19.5 mg, 4.77 × 10-5 mol) were added to the flask, and the
resulting solution was stirred for 10 min at room temperature.
These solutions, containing Cu(II) (2.38 × 10-6 mol/mL) were
used as stock CuBr2 deactivator solutions.
Exp er im en ta l Section
1
Ch a r a cter iza tion . H NMR spectra of the ligand and the
copper complex were collected using Bruker 300 MHz 1H NMR
with CDCl3 and benzene-d6 as solvents, respectively. Elemen-
tal analysis was performed by Midwest Microlab, LLC. Con-
version of monomer was determined by gas chromatography
using a Shimadzu GC 14-A gas chromatograph equipped with
a FID detector and J &W Scientific 30 m DB WAX Megabore
column. The molecular weight and molecular weight distribu-
tion of the polymers were determined by GPC using a PSS
column (styrogel 105, 103, 102 Å) with RI detectors. GPC was
performed using THF as an eluent at the flow rate of 1 mL/
min. Linear polystyrene standards were used for calibration
of poly(n-butyl acrylate). Theoretical molecular weights were
calculated by following eq 1.
ATRP of n -Bu tyl Acr yla te (BA). A typical ATRP poly-
merization procedure was performed as follows. The phenoxy-
triamine copper complex was placed in a 25 mL Schlenk flask,
and then the flask was evacuated and backfilled with nitrogen
three times. Toluene (1.4 mL), anisole (0.1 mL), and BA (1.5
mL, 1.05 × 10-2 mol) were successively added to the flask after
degassing by bubbling with nitrogen for 30 min. The desired
amount of CuBr2 deactivator stock solution was added. Finally
the initiator, ethyl 2-bromopropionate, was added. The result-
ing mixture was warmed to 70 °C to start the polymerization.
Samples were taken periodically via a syringe to follow the
kinetics of the polymerization process. The samples were
diluted with tetrahydrofuran followed by filtration through a
Gelman Acrodisc 0.2 µm PTFE filter prior to the analysis by
gas chromatography (GC) and gel permeation chromatography
(GPC).
Mn,th ) ([monomer]0/[initiator]0) × conversion ×
MW(monomer) + MW(initiator) (1)
Ma ter ia ls. Copper(I) bromide,34 copper(I) chloride,34 tris-
(2-(dimethylamino)ethyl)amine (Me6TREN),35 and 4,4′-(di-5-
nonyl)-2,2′-bipyridine (dNbpy)34 were purified and prepared
as detailed in previous reports. N,N,N′,N′′,N′′-Pentamethyldi-
ethylenetriamine (PMDETA) was purified by distillation under
vacuum. n-Butyl acrylate (BA) was passed through a column
filled with neutral alumina to remove stabilizer, dried over
calcium hydride, and distilled under reduced pressure before
use. Acetonitrile was distilled under nitrogen over calcium
Resu lts a n d Discu ssion
Syn th esis of Liga n d a n d Cop p er Com p lex. The
synthetic scheme for the preparation of the phenoxy-

7434 Inoue and Matyjaszewski
Macromolecules, Vol. 36, No. 20, 2003
F igu r e 2. Kinetics plots for the polymerization of BA using
phenoxy-triamine Cu(I) complex. See Table 1 for polymeri-
zation conditions.
1
F igu r e 3. Evolution of Mn and Mw/Mn vs conversion for the
polymerization of BA using phenoxy-triamine Cu(I) complex.
See Table 1 for polymerization conditions.
F igu r e 1. 300 MHz H NMR spectra of phenol-triamine in
CDCl3 (a) and phenoxy-triamine copper(I) complex in C6D6
(b).
triamine ligand and the resulting copper complex is
summarized in Scheme 1. To avoid any reaction at the
para position of phenol, which would result in the
formation of a multisubstitution reaction product, 2,4-
dimethylphenol was used as the starting material. 2,4-
Dimethyl-6-bis(2-(diethylamino)ethyl)aminomethyl-
phenol was synthesized through the reaction of
N,N,N′,N′-tetraethyldiethylenetriamine, formaldehyde,
and 2,4-dimethylphenol in methanol. After purification
by passage through a silica column, the desired phenol-
triamine compound was obtained as a yellow oil in 70%
yield. Subsequently, the compound was reacted with
n-butyllithium followed by CuCl in acetonitrile to give
the desired neutral copper(I) complex. After recrystal-
lization from acetonitrile, the complex was obtained as
colorless crystals in 56% yield.
Figure 1 shows 1H NMR spectra of the phenol-
triamine compound and the phenoxy-triamine copper
complex. Signals of aromatic protons and aromatic
methyl protons (protons a, b, c and d) are shifted to a
lower magnetic field, and the methylene protons (proton
f, g, and h) show more complex signals due to the
coordination to the copper metal center. Along with
elemental analysis, these results support the proposed
structure for the complex shown in Scheme 1. This
neutral copper complex has good solubility in both polar
and nonpolar organic solvents. The crystalline form of
this copper complex was stable, at least for several
hours, under ambient atmosphere; however, it decom-
posed easily in solution, forming a brown solution,
probably due to oxidation of copper(I). An X-ray crystal-
lography study of the structure of this copper complex
is underway.
BA P olym er iza tion in th e Absen ce a n d P r esen ce
of Cu Br 2 Dea ctiva tor s. The phenoxy-triamine copper
complex was used for the polymerization of BA in 50%
toluene at 70 °C. The ratio of [BA]0/[I]0/[Cu(I) complex]0
was set at 300/1/1 for all experiments. Under these
conditions, the phenoxy-triamine catalyst promoted a
relatively fast polymerization reaction. As shown in
Figure 2, semilogarithmic kinetic plots showed linearity
vs polymerization time and monomer conversion reached
84% in 4 h. However, this phenoxy-triamine catalyst
did not control the polymerization reaction very well.
As shown in Figure 3, the molecular weight jumped to
74 000 at 19% monomer conversion, approximately 10
times higher than the value expected for quantitative
initiation. Although molecular weight was decreasing
with increasing conversion, it was still considerably
higher than theoretical molecular weight throughout the
polymerization. The molecular weight distribution was
broad with PDI values around 2, also indicating poor
control. These facts indicate that the phenoxy-triamine
catalyst could promote relatively fast activation of the
dormant species but slow deactivation of growing radi-
cals. Therefore, this catalyst system leads to inefficient
initiation, higher molecular weight polymers, and broad
molecular weight distribution.
To increase the rate of deactivation and improve
control over the polymerizations, we decided to add a
small amount of an additional Cu(II) deactivator com-
plexed with various ligands. Fast activation by phe-
noxy-triamine catalyst and fast deactivation by an
additional deactivator was anticipated. Thus, 3 mol %
(versus phenoxy-triamine complex) of CuBr2, dNbpy/
CuBr2, PMDETA/CuBr2, or Me6TREN/CuBr2 was added

Macromolecules, Vol. 36, No. 20, 2003
Dual Catalyst System for ATRP 7435
Ta ble 1. Con d ition s a n d Resu lts for ATRP of BA Usin g P h en oxy-Tr ia m in e Cu (I) Com p lex in th e Absen ce a n d P r esen ce
of Cu (II) Dea ctiva tor a
expt
Cu(II) complex
[Cu(I)]0/[Cu(II)]0
t (h)
conv (%)
Mn (×10-4
)
Mn,theo (×10-4
)
Mw (×10-4
)
Mw/Mn
1
2
3
4
5
4
5
8
9
84.1
66.5
72.5
65.6
67.4
5.20
2.11
2.89
2.62
2.26
3.25
2.58
2.81
2.54
2.61
10.3
2.47
5.00
5.96
3.88
1.99
1.17
1.73
2.28
1.71
Me6TREN/CuBr2
PMDETA/CuBr2
dNbpy/CuBr2
CuBr2
1/0.03
1/0.03
1/0.03
1/0.03
8.5
a
Polymerization conditions: initiator ) ethyl 2-bromopropionate, temperature ) 70 °C, Cu(II) complexes were added as a stock solutions
in acetone. Concentration of Cu(II) in stock solution was 2.38 mM (CuBr2/Me6TREN ) CuBr2/PMDETA ) 1/1 mol/mol, CuBr2/dNbpy )
1/2 mol/mol in stock solution); for expt 1, [BA]0 ) 3.49 M, [BA]0/[I]0/[Cu(I) complex]0 ) 300/1/1, BA/toluene ) 1/1 v/v; for expt 2-5, [BA]0
) 3.05 M, [BA]0/[I]0/[Cu(I) complex]0/[Cu(II) complex]0 ) 300/1/1/0.03, BA/toluene/acetone ) 1/1/0.29 v/v/v.
F igu r e 5. Evolution of Mn and Mw/Mn vs conversion for the
polymerization of BA using phenoxy-triamine Cu(I) complex
in the presence of 3 mol % of Me6TREN/CuBr2. See Table 1
for polymerization conditions.
F igu r e 4. Kinetics plots for the polymerization of BA using
phenoxy-triamine Cu(I) complex in the presence of vari-
able Cu(II) deactivators. See Table 1 for polymerization
conditions.
to the polymerization system as a deactivator. The
results of these polymerizations using the various dual
catalyst systems are summarized in Table 1, along with
the result of ATRP using the phenoxy-triamine com-
plex alone. Polymerization rates decreased in the pres-
ence of deactivators due to higher concentration of
Cu(II) species. The overall polymerization rate depended
on the structure of deactivator, and the order of de-
creasing rate was Me6TREN/CuBr2 > PMDETA/CuBr2
> dNbpy/CuBr2 ∼ CuBr2 (Figure 4). The order of
polymerization rates correlates with the order of acti-
vation rate constant and equilibrium constants for
Me6TREN, PMDETA, and dNbpy.37 As mentioned later,
some amount of Cu(I) complex can be generated from
the added deactivator through halogen migration reac-
tion, and it might have some contribution to the activa-
tion process. Thus, the smaller rate reduction in the
presence of the Me6TREN and PMDETA complex, when
compared to dNbpy and CuBr2 alone, could be due to
the contribution of Me6TREN/CuBr and PMDETA/CuBr
to the activation process.
F igu r e 6. Evolution of Mn and Mw/Mn vs conversion for the
polymerization of BA using phenoxy-triamine Cu(I) complex
in the presence of 3 mol % of PMDETA/CuBr2. See Table 1 for
polymerization conditions.
weight from predicted value was observed at high
conversion, plausibly due to the contribution of chain
transfer reactions with an excess ligand (Figure 1b).38,39
Thus, a small amount of the proper deactivator
enhanced the deactivation step and promoted a well-
controlled polymerization, while the polymerization rate
was slightly reduced. These results remind of a similar
binary catalyst system, the immobilized/soluble hybrid
catalyst system.32,33,40,41
BA P olym er iza tion in th e P r esen ce of Me6TREN/
Cu Br 2. To determine the minimum amount of the
additional deactivator required for preparation of an
efficient dual catalyst system, the polymerization was
conducted in the presence of 0.5, 1, and 2 mol % of
Me6TREN/CuBr2. The results are summarized in Table
2. As shown in Figure 9, polymerization rates correlate
with the amount of deactivator; with less deactivator
leading to higher polymerization rates. Regarding the
level of control over the polymerization process, the
In the presence of CuBr2, dNbpy/CuBr2, or PMDETA/
CuBr2, the level of control over the polymerization was
improved, but it was still insufficient (Figures 6-8).
Molecular weights increased with conversion; how-
ever, some deviations from predicted values were ob-
served, and molecular weight distributions were still
quite broad (Mw/Mn > 1.6). However, ATRP in the
presence of Me6TREN/CuBr2 was surprisingly well-
controlled.
Molecular weights increased linearly with conversion
and agreed with predicted values. PDI values decreased
during the polymerization and eventually reached Mw/
Mn ) 1.17 (Figure 5). Some deviation of molecular

7436 Inoue and Matyjaszewski
Macromolecules, Vol. 36, No. 20, 2003
Ta ble 2. Con d ition s a n d Resu lts for ATRP of BA Usin g P h en oxy-Tr ia m in e Cu (I) Com p lex in th e P r esen ce of Va r ia ble
a
Am ou n t of Me6TREN/Cu Br 2
expt
catalyst
[I]0/[Cu(I)]0/[Cu(II)]0
t (h)
conv (%)
Mn (×10-4
)
Mn,theo (×10-4
)
Mw (×10-4
)
Mw/Mn
1
2
3
4
5
6
phenoxy-triamine
phenoxy-triamine
phenoxy-triamine
phenoxy-triamine
phenoxy-triamine
Me6TREN/CuBr
1/1/0
4
4
5
5
5
6
84.1
84.4
80.3
72.9
66.5
32.6
5.20
2.60
2.65
2.30
2.11
1.39
3.25
3.26
3.11
2.82
2.58
1.27
10.3
4.23
3.69
2.86
2.47
2.39
1.99
1.63
1.39
1.24
1.17
1.72
1/1/0.005
1/1/0.01
1/1/0.02
1/1/0.03
1/0.01/0
a
Polymerization conditions: initiator ) ethyl 2-bromopropionate, temperature ) 70 °C, Me6TREN/CuBr2 was added as a stock solutions
in acetone. Concentration of Me6TREN/CuBr2 in stock solution was 2.38 mM (CuBr2/Me6TREN ) 1/1 mol/mol in stock solution); for expt
1, [BA]0 ) 3.49 M, [BA]0/[I]0/[Cu(I) complex]0 ) 300/1/1, BA/toluene ) 1/1 v/v; for expt 2, [BA]0 ) 3.40 M, [BA]0/[I]0/[Cu(I) complex]0/
[Me6TREN/CuBr2]0 ) 300/1/1/0.005, BA/toluene/acetone ) 1/1/0.05 v/v/v; for expt 3, [BA]0 ) 3.32 M, [BA]0/[I]0/[Cu(I) complex]0/[Me6TREN/
CuBr2]0 ) 300/1/1/0.01, BA/toluene/acetone ) 1/1/0.10 v/v/v; for expt 4, [BA]0 ) 3.18 M, [BA]0/[I]0/[Cu(I) complex]0/[Me6TREN/CuBr2]0 )
300/1/1/0.02, BA/toluene/acetone ) 1/1/0.19 v/v/v; for expt 5, [BA]0 ) 3.05 M, [BA]0/[I]0/[Cu(I) complex]0/[Me6TREN/CuBr2]0 ) 300/1/1/
0.03, BA/toluene/acetone ) 1/1/0.29 v/v/v; for expt 6, [BA]0 ) 3.49 M, [BA]0/[I]0/[CuBr]0/[Me6TREN]0 ) 300/1/1/0.01/0.01, BA/toluene )
1/1 v/v.
F igu r e 7. Evolution of Mn and Mw/Mn vs conversion for the
F igu r e 9. Kinetics plots for the polymerization of BA using
polymerization of BA using phenoxy-triamine Cu(I) complex
phenoxy-triamine Cu(I) complex in the presence of variable
amounts of Me6TREN/CuBr2. See Table 2 for polymerization
conditions.
in the presence of 3 mol % of dNbpy/CuBr2. See Table 1 for
polymerization conditions.
F igu r e 8. Evolution of Mn and Mw/Mn vs conversion for the
polymerization of BA using phenoxy-triamine Cu(I) complex
in the presence of 3 mol % of CuBr2. See Table 1 for
polymerization conditions.
F igu r e 10. Evolution of Mn and Mw/Mn vs conversion for the
polymerization of BA using phenoxy-triamine Cu(I) complex
in the presence of variable amounts of Me6TREN/CuBr2. See
Table 2 for polymerization conditions.
polymerization was still controlled even in the presence
of 1 or 2 mol % deactivator, though the degree of control
was reduced gradually with decreasing amount of
deactivator. For instance, in the presence of 1 mol %
Me6TREN/CuBr2, some deviation of molecular weight
at low conversion and relatively high PDI values (Mw/
Mn ∼ 1.4) were observed. When the amount of deactiva-
tor was decreased to 0.5 mol %, the level of control
attained in the polymerization was poor. Molecular
weight stayed at constant values, around 25 000, and
PDI values were higher than 1.6 throughout the poly-
merization. Therefore, at least 1 mol % of Me6TREN/
CuBr2 deactivator is necessary to control the polymer-
ization. This amount corresponds to ∼30 ppm vs BA
monomer (molar basis). It is surprising that such a
small amount of additional Cu(II) species provides an
efficient control over polymerization.
As indicated above and described later, some
Me6TREN/CuBr should exist in this polymerization
system due to the halogen migration reaction be-
tween the phenoxy-triamine copper(I) complex and
Me6TREN/CuBr2. Me6TREN/CuBr is the most active
ATRP catalyst for BA polymerization and promotes
well-controlled polymerization using a significantly
lower amount of catalyst.35 Therefore, we conducted
polymerization of BA in the presence of only Me6TREN/
CuBr with [BA]0/[I]0/[Me6TREN/CuBr]0 ) 300/1/0.01 in

Macromolecules, Vol. 36, No. 20, 2003
Dual Catalyst System for ATRP 7437
Sch em e 2. Ha logen Migr a tion Rea ction betw een th e
P h en oxy-Tr ia m in e Cop p er Com p lex a n d th e
Ad d ition a l Dea ctiva tor s
Sch em e 3. P r op osed Mech a n ism for th e Du a l
Ca ta lyst System
Ta ble 3. Equ ilibr iu m Con sta n t of ATRP (Keq) a n d
Dea ctiva tion Ra te Con sta n t (kd ) for a Va r iety of Ca ta lyst
System s
a
b
catalyst system
Keq
kd (L mol-1 s-1
)
Me6TREN/CuI/II
PMDETA/CuI/II
dNbpy/CuI/II
10-6
10-8
10-9
1.4 × 107
6.1 × 106
2.5 × 107
while they have quite different Keq. The Me6TREN/Cu
system has the highest Keq, and this indicates that the
Me6TREN complex prefers to be in the higher oxidation
state as compared to PMDETA and dNbpy complexes.
This was confirmed earlier by the measurement of
halogen migration equilibrium constants between
Me6TREN and dNbpy complexes and between PMDETA
and dNbpy complexes.33 This high reducing power forces
the Me6TREN complex to stay in the higher oxidation
state in the dual catalyst system, and the resulting
Me6TREN/CuBr2 deactivates the growing radical ef-
ficiently. The resulting Me6TREN/CuBr should then
reduce the phenoxy-triamine Cu(II) complex to its
lower oxidation state and return again to the Cu(II)
state. Higher concentration of a deactivator having high
kd should decrease the polydispersity index as the latter
is inversely proportional to the deactivation rate con-
stant and deactivator concentration (eq 2). This pro-
posed “shuttling” mechanism is again similar to the
mechanism for the immobilized/soluble hybrid catalyst
system (Scheme 3).32,33
a
Equilibrium constant of ATRP for methyl acrylate measured
at 90 °C for Me6TREN and PMDETA and at 100 °C for dNbpy.
b Deactivation rate constant measured at 75 °C in acetonitrile with
1-phenylethyl radical.
order to confirm the contribution of Me6TREN/CuBr-
mediated polymerization in the dual catalyst system.
The result is shown in experiment 6 in Table 2. Slower
rate of polymerization and limited control over poly-
merization were observed when compared to the poly-
merization catalyzed by the dual catalyst system using
the same amount of Me6TREN/CuBr2. Although molec-
ular weights were properly controlled (fast initiation),
polydispersities were relatively high (too low deactivator
concentration). Therefore, high activity and well-
controlled polymerization behavior are achieved by
mutual contribution of both the phenoxy-triamine
catalyst and Me6TREN/CuBr2.
P r op osed Mech a n ism for t h e Du a l Ca t a lyst
System . When polymerization was conducted using the
phenoxy-triamine catalyst alone, relatively fast poly-
merization was demonstrated; however, poor control of
polymerization was attained, i.e., low initiation effi-
ciency resulting in high molecular weight and high poly-
dispersity. This suggests that although the phenoxy-
triamine Cu(I) complex could activate the dormant
species sufficiently fast, the corresponding Cu(II) com-
plex deactivates too slow. Thus, the rate of deactivation
(Rd) should be increased to improve control over the
polymerization process. The addition of a small amount
of a strong Cu(II) deactivator can increase overall Rd in
ATRP, as demonstrated for the immobilized/soluble
hybrid catalyst system.
Mw/Mn ) 1 + (kp[I]0/kd[deactivator])(2/p - 1) (2)
Addition of CuBr2 without ligand was also inves-
tigated. CuBr2 should increase the concentration of
phenoxy-triamine Cu(II) complex through halogen
migration reaction. However, control over polymeriza-
tion did not improve significantly. This again indicates
that the phenoxy-triamine Cu(II) complex is not a good
deactivator, and it is necessary to increase both de-
activator concentration and overall kd in order to
improve the level of control over polymerization.
The Me6TREN complex is perhaps the best deactiva-
tor for this dual catalyst system because of its high
reducing power and high deactivation rate constant.
Me6TREN/CuBr2, PMDETA/CuBr2, and dNbpy/CuBr2
were investigated as deactivators for this dual catalyst
system. When these Cu(II) species are added to a
solution the Cu(I) complex, halogen migration reaction
between the phenoxy-triamine complex and these
additional deactivators could occur to yield a mixture
of phenoxy-triamine CuI/II complexes and additional
CuI/II ligand species (Scheme 2). There are two impor-
tant factors to consider in the desire to increase overall
Rd by an addition of Cu(II) deactivator: one is a high
deactivation rate constant of the additional deactivator
but another, perhaps even more important, is the
presence of a sufficient concentration of this deactivator.
The latter depends on the relative values of the equi-
librium constants. Values for the deactivation rate
constants (kd)37 and equilibrium constants (Keq ) ka/kd)35
for the Me6TREN/Cu, PMDETA/Cu, and dNbpy/Cu
systems are listed in Table 3 to allow a comparison of
these properties. All these systems possess similar and
relatively high deactivation rate constants (kd ∼ 107),
Con clu sion s
A phenoxy-triamine copper complex promotes rela-
tively fast polymerization of BA; however, this complex
does not control the polymerization well. Analysis of the
kinetics of the polymerization indicates that this is due
to a slow deactivation step, indicating the phenoxy-
triamine Cu(II) complex is a poor deactivator. When a
small amount of an efficient deactivator was added, the
level of control over the polymerization process was
improved. Especially, the Me6TREN/CuBr2 system im-
proved the control dramatically, and the addition of only
∼30 ppm of the deactivator vs BA monomer was
effective in improving the control to the desired degree.
Both the high deactivation rate constant and the high
reducing power of the Me6TREN complex are respon-
sible for the success of Me6TREN/CuBr2 in the prepara-
tion of this dual catalyst system. The Me6TREN/Cu
complex might act as “shuttling” agent and in addi-
tion to assisting the deactivation process may also

7438 Inoue and Matyjaszewski
Macromolecules, Vol. 36, No. 20, 2003
Chemical Society: Washington, DC, 2003; p 130.
(16) Patten, T. E.; Xia, J .; Abernathy, T.; Matyjaszewski, K.
Science 1996, 272, 866.
(17) Xia, J .; Matyjaszewski, K. Macromolecules 1997, 30, 7697.
(18) Xia, J .; Gaynor, S. G.; Matyjaszewski, K. Macromolecules
1998, 31, 5958.
(19) Xia, J .; Zhang, X.; Matyjaszewski, K. In Transition Metal
Catalysis in Macromolecular Design; ACS Symposium Series
Vol. 760; Boffa, L. S., Novak, B. M., Eds.; American Chemical
Society: Washington, DC, 2000; p 207.
reduce the phenoxy-triamine Cu(II) complex. This
proposed mechanism is similar to that proposed for the
immobilized/soluble hybrid catalyst system where
Me6TREN/CuBr2 overcomes slow deactivation due to
diffusion limitation in the immobilized/soluble hybrid
catalyst system. This study demonstrates that addition
of small amounts of good deactivators can overcome the
slow deactivation step due to low deactivation rate
constant of the major catalyst. This result indicates the
some catalyst systems inherently providing poorly
controlled ATRP could be improved by the addition of a
small amount of Cu(II) deactivator even in a homoge-
neous system.
(20) Haddleton, D. M.; J asieczek, C. B.; Hannon, M. J .; Shooter,
A. J . Macromolecules 1997, 30, 2190.
(21) Xia, J .; Matyjaszewski, K. Macromolecules 1999, 32, 2434.
(22) Destarac, M.; Bessiere, J . M.; Boutevin, B. Macromol. Rapid
Commun. 1997, 18, 967.
(23) Kickelbick, G.; Matyjaszewski, K. Macromol. Rapid Commun.
1999, 20, 341.
Ack n ow led gm en t. We greatly appreciate the fi-
nancial support from Mitsui Chemicals, the National
Science Foundation (DMR-0090409), and CRP Consor-
tium at Carnegie Mellon University.
(24) Davis, K. A.; Paik, H.-j.; Matyjaszewski, K. Macromolecules
1999, 32, 1767.
(25) Woodworth, B. E.; Metzner, Z.; Matyjaszewski, K. Macro-
molecules 1998, 31, 7999.
(26) J ohnson, R. M.; Ng, C.; Samson, C. C. M.; Fraser, C. L.
Macromolecules 2000, 33, 8618.
Refer en ces a n d Notes
(27) Matyjaszewski, K.; Wei, M.; Xia, J .; Gaynor, S. G. Macromol.
Chem. Phys. 1998, 199, 2289.
(28) Singha, N. K.; Klumperman, B. Macromol. Rapid Commun.
2000, 21, 1116.
(1) Matyjaszewski, K., Ed.; Controlled Radical Polymerization;
ACS Symposium Series Vol. 685; American Chemical Soci-
ety: Washington, DC, 1998.
(2) Matyjaszewski, K., Ed.; Controlled/ Living Radical Poly-
merization. Progress in ATRP, NMP, and RAFT; ACS Sym-
posium Series Vol. 768; American Chemical Society: Wash-
ington, DC, 2000.
(3) Matyjaszewski, K., Ed.; Advances in Controlled/ Living Radi-
cal Polymerization; ACS Symposium Series Vol. 854; Ameri-
can Chemical Society: Washington, DC, 2003.
(4) Matyjaszewski, K., Davis, T. P., Eds.; Handbook of Radical
Polymerization; Wiley-Interscience: Hoboken, 2002.
(5) Matyjaszewski, K.; Xia, J . Chem. Rev. 2001, 101, 2921.
(6) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001, 101,
3689.
(29) Davis, K.; O’Malley, J .; Paik, H.-J .; Matyjaszewski, K. Polym.
Prepr. 1997, 38, 687.
(30) Pintauer, T.; Reinoehl, U.; Feth, M.; Bertagnolli, H.; Maty-
jaszewski, K. Eur. J . Inorg. Chem. 2003, 2082.
(31) Matyjaszewski, K.; Shipp, D. A.; Wang, J .-L.; Grimaud, T.;
Patten, T. E. Macromolecules 1998, 31, 6836.
(32) Hong, S. C.; Paik, H.-J .; Matyjaszewski, K. Macromolecules
2001, 34, 5099.
(33) Hong, S. C.; Matyjaszewski, K. Macromolecules 2002, 35,
7592.
(34) Matyjaszewski, K.; Patten, T. E.; Xia, J . J . Am. Chem. Soc.
1997, 119, 674.
(35) Queffelec, J .; Gaynor, S. G.; Matyjaszewski, K. Macromol-
ecules 2000, 33, 8629.
(36) Lubben, M.; Hage, R.; Meetsma, A.; Byma, K.; Feringa, B.
L. Inorg. Chem. 1995, 34, 2217.
(7) Patten, T. E.; Matyjaszewski, K. Adv. Mater. 1998, 10, 901.
(8) Patten, T. E.; Matyjaszewski, K. Acc. Chem. Res. 1999, 32,
895.
(9) Davis, K. A.; Matyjaszewski, K. Adv. Polym. Sci. 2002, 159,
2.
(37) Matyjaszewski, K.; Paik, H.-j.; Zhou, P.; Diamanti, S. J .
(10) Coessens, V.; Pintauer, T.; Matyjaszewski, K. Prog. Polym.
Sci. 2001, 26, 337.
(11) Pyun, J .; Matyjaszewski, K. Chem. Mater. 2001, 13, 3436.
(12) Matyjaszewski, K. Chem.sEur. J . 1999, 5, 3095.
(13) Wang, J .-S.; Matyjaszewski, K. J . Am. Chem. Soc. 1995, 117,
5614.
(14) Wang, J .-S.; Matyjaszewski, K. Macromolecules 1995, 28,
7901.
(15) Pintauer, T.; McKenzie, B.; Matyjaszewski, K. In Advances
in Controlled/ Living Radical Polymerization; ACS Sympo-
sium Series Vol. 854; Matyjaszewski, K., Ed.; American
Macromolecules 2001, 34, 5125.
(38) Bednarek, M.; Biedron, T.; Kubisa, P. Macromol. Chem. Phys.
2000, 201, 58.
(39) Schoen, F.; Hartenstein, M.; Mueller, A. H. E. Macromolecules
2001, 34, 5394.
(40) Hong, S. C.; Neugebauer, D.; Inoue, Y.; Lutz, J .-F.; Matyjas-
zewski, K. Macromolecules 2003, 36, 27.
(41) Hong, S. C.; Lutz, J .-F.; Inoue, Y.; Strissel, C.; Nuyken, O.;
Matyjaszewski, K. Macromolecules 2003, 36, 1075.
MA034979V