Binary copolymerization with catalytic chain transfer. A method for synthesizing macromonomers based on monosubstituted monomers

Article abstract of DOI:10.1021/ma0501949

With appropriate choice of conditions, copolymerization of monosubstituted monomers [CH₂=CHX: e.g., styrene, butyl acrylate (BA)] in the presence of small amounts of an α-methylvinyl monomer [CH ₂=C(CH₃)Y: e.g., α-methylstyrene (AMS), methyl methacrylate (MMA), methacrylonitrile (MAN)] and a cobaloxime as chain transfer catalyst provides a route to macromonomers that are composed largely of the monosubstituted monomer and yet have a chain end derived from the α-methylvinyl monomer (-CH₂-C(=CH₂)Y). The various factors (temperature, concentrations, type of cobaloxime, types of monomer) that influence molecular weight and end group purity and the importance of the various side reactions that may complicate the process are described. Macromonomer purity is enhanced by increasing the concentration of the α-methylvinyl monomer and reducing the cobaloxime concentration. It also depends on the structure of the cobaloxime, increasing in the series where the ligands are derived from dimethyl glyoxime < diethyl glyoxime ～ diphenyl glyoxime, and the α-methylvinyl monomer, increasing in the series where Y is CO₂R < CN < Ph. For styrene-AMS copolymerization, macromonomer purity (the fraction of AMS-derived ends) was enhanced by increasing the reaction temperature. For BA-AMS copolymerization, macromonomer purity was enhanced by decreasing the reaction temperature. However, it is necessary to use a high reaction temperature to limit the extent of macromonomer copolymerization that occurs as a side reaction at high monomer conversion. High purity, AMS terminal BA macromonomers can be prepared by BA-AMS copolymerization at 125°C with as little as 2 mol % AMS. We also show how the overall composition, molecular weight, and end group functionality of copolymers formed in the presence of a chain transfer agent can be predicted using classical statistics and point out some problems with previous treatments. Analytical expressions which describe zero conversion binary copolymerization in the presence of a transfer agent are derived and successfully applied to the above-mentioned system. The effective transfer constants of cobaloxime transfer catalysts are reported. Those observed in copolymerizations of vinyl and α-methylvinyl monomers are reduced with respect to values observed in homopolymerization of α-methylvinyl monomers because of reversible consumption of the cobalt complex as an adduct to the vinyl monomer. Transfer constants of macromonomers with AMS end groups in styrene polymerization at 120°C are in the range 0.11-0.15. At lower temperatures (80°C), macromonomer copolymerization dominates over chain transfer and the effective transfer constant is ～0.

Full text of DOI:10.1021/ma0501949

Macromolecules 2005, 38, 9037-9054
9037
Binary Copolymerization with Catalytic Chain Transfer. A Method for
Synthesizing Macromonomers Based on Monosubstituted Monomers
John Chiefari, Justine Jeffery, Julia Krstina, Catherine L. Moad, Graeme Moad,*
Almar Postma, Ezio Rizzardo, and San H. Thang
CSIRO Molecular Science, Bag 10, Clayton South, Victoria 3169, Australia
Received January 30, 2005; Revised Manuscript Received August 4, 2005
ABSTRACT: With appropriate choice of conditions, copolymerization of monosubstituted monomers [CH
2
d
CHX: e.g., styrene, butyl acrylate (BA)] in the presence of small amounts of an R-methylvinyl monomer
[
2 3
CH dC(CH )Y: e.g., R-methylstyrene (AMS), methyl methacrylate (MMA), methacrylonitrile (MAN)]
and a cobaloxime as chain transfer catalyst provides a route to macromonomers that are composed largely
of the monosubstituted monomer and yet have a chain end derived from the R-methylvinyl monomer
(
2 2
-CH -C(dCH )Y). The various factors (temperature, concentrations, type of cobaloxime, types of
monomer) that influence molecular weight and end group purity and the importance of the various side
reactions that may complicate the process are described. Macromonomer purity is enhanced by increasing
the concentration of the R-methylvinyl monomer and reducing the cobaloxime concentration. It also
depends on the structure of the cobaloxime, increasing in the series where the ligands are derived from
dimethyl glyoxime < diethyl glyoxime ∼ diphenyl glyoxime, and the R-methylvinyl monomer, increasing
2
in the series where Y is CO R < CN < Ph. For styrene-AMS copolymerization, macromonomer purity
(
the fraction of AMS-derived ends) was enhanced by increasing the reaction temperature. For BA-AMS
copolymerization, macromonomer purity was enhanced by decreasing the reaction temperature. However,
it is necessary to use a high reaction temperature to limit the extent of macromonomer copolymerization
that occurs as a side reaction at high monomer conversion. High purity, AMS terminal BA macromonomers
can be prepared by BA-AMS copolymerization at 125 °C with as little as 2 mol % AMS. We also show
how the overall composition, molecular weight, and end group functionality of copolymers formed in the
presence of a chain transfer agent can be predicted using classical statistics and point out some problems
with previous treatments. Analytical expressions which describe zero conversion binary copolymerization
in the presence of a transfer agent are derived and successfully applied to the above-mentioned system.
The effective transfer constants of cobaloxime transfer catalysts are reported. Those observed in
copolymerizations of vinyl and R-methylvinyl monomers are reduced with respect to values observed in
homopolymerization of R-methylvinyl monomers because of reversible consumption of the cobalt complex
as an adduct to the vinyl monomer. Transfer constants of macromonomers with AMS end groups in styrene
polymerization at 120 °C are in the range 0.11-0.15. At lower temperatures (80 °C), macromonomer
copolymerization dominates over chain transfer and the effective transfer constant is ∼0.
Introduction
Macromonomers with end group structure 2 have
Scheme 1
1
-7
utility as chain transfer agents
and as precursors to
8-10
1,2,10-18
19-21
block,
graft,
and end-functional polymers.
Polymerizations of methacrylate esters [e.g. methyl
methacrylate (MMA); Y ) CO2Me)] and some other
monomers with an R-methyl group in the presence of
cobaloximes as chain transfer catalysts can provide such
macromonomers in high yield. The generally accepted
mechanism for chain transfer is shown in Scheme 1 and
the overall process may be viewed as a catalyzed chain
22
transfer to monomer. The cobaloxime may then either
2
2
be called a “chain transfer catalyst” or a “catalytic
2
3
chain transfer agent”. The process of catalytic chain
transfer has been the subject of a number of recent
mentioned applications. A side reaction, involving re-
versible coupling of the propagating radical (3) with the
cobalt(II) catalyst, to give a relatively stable alkylcobalt-
(III) species (4), further complicates the polymerizations
(Scheme 2). Probably as a consequence of this side
reaction, inhibition, retardation and/or loss of catalyst
activity are often reported.2
A mechanism for the transfer step in copolymerization
of a monosubstituted monomer (S) with an R-methylvi-
nyl monomer (A) in the presence of a chain transfer
catalyst is shown in Scheme 3. Provided there is
specificity for reaction between the chain transfer
catalyst and the propagating radical (1) which has a
3
,24-26
reviews.
Polymerizations of monosubstituted monomers [e.g.
styrene, butyl acrylate (BA), methyl acrylate (MA)] in
the presence of cobaloximes as chain transfer catalysts
provides polymer chains possessing an internal (1,2-
disubstituted) double bond as an end group structure
i.e., 5, Scheme 2). These double bonds have a low
reactivity in polymerization and consequently the poly-
mers have little utility with respect to the above-
2,27,28
(
*
Author for correspondence. E-mail graeme.moad@csiro.au.
1
0.1021/ma0501949 CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/04/2005

9038 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
Scheme 2
and an R-methylvinyl monomer [e.g. MMA, Y ) CO2-
CH3; R-methylstyrene (AMS), Y ) Ph; methacrylonitrile
(
MAN), Y ) CN] can yield predominantly macromono-
mers with end group structure 2. In such copolymeriza-
tions, the relative yield of end groups 2 and 5 depends
on a number of factors including the nature and
concentration of the R-methylvinyl monomer and cobalt
complex and the reaction temperature.30 Other comono-
mers with appropriate reactivity and transfer constants
may also be used in the process to yield end-functional
polymers.37 With higher reaction temperatures (>100
°
C), the inhibition or retardation, often seen in poly-
merizations of monosubstituted monomers, can be sub-
stantially reduced or eliminated.2
2,27,28
38
The copolymerizations of styrene with AMS, BA
3
9
with AMS, styrene with MMA or other methacry-
lates,4
0-42
and BA or MA with MMA
43-45
in the presence
of cobaloxime chain transfer catalysts have been the
subject of a number of other recent studies. A kinetic
Scheme 3
model3
8,40,46
was proposed to predict, among other
things, the end group composition of copolymers formed
in these copolymerizations. The model suggests that for
the case of styrene-AMS copolymerization, the product
should have overwhelmingly structure 2. Although this
may be consistent in general terms with our3 and other
experimental findings on end group composition, the
proposed kinetic model significantly overestimates ex-
perimentally determined end group purities (see below).
Recently, Pierik and van Herk have also examined the
0
kinetics of copolymerization of MA with MMA and BA
with MMA.4
3,44
However, the quantification of the end
group functionality of the polymers produced was not
directly addressed in their studies.
terminal R-methylvinyl monomer unit, it should be
possible to control the process to provide macromono-
mers with end group structure 2 that are composed
largely of monosubstituted monomers and thus avoid
Macromonomers based on monosubstituted mono-
mers (styrene, BA) have previously been prepared using
addition-fragmentation chain transfer agents such as
R-methylstyrene dimer (AMS dimer, 6), allyl sulfides
2
9,30
some of the above-mentioned complications.
(e.g. 7, 8) and allyl bromides (vide infra). Very recently,
Copolymerizations of monosubstituted monomers (in
particular, acrylates and styrene, usually as a minor
component) and methacrylate monomers are mentioned
in the early patents on the use of catalytic chain transfer
for molecular weight control.31-33 However, these pat-
ents contain little or no information on the precise end
group structure of the copolymers formed nor do they
indicate how it might be controlled. Gruel and Har-
we found that a narrow dispersity BA macromonomer
can also be prepared by thermolysis of poly(butyl
acrylate) prepared by RAFT polymerization using a
47
trithiocarbonate RAFT agent. These methods make
use of a reagent that must be stoichiometric with the
moles of polymer formed.
In this paper, we provide further details of the
copolymerization of monosubstituted monomers with
R-methylvinyl monomers in the presence of chain
transfer catalysts taking into account the various factors
that may influence macromonomer purity. We also
propose analytical expressions that may be used to
predict copolymer composition, molecular weight and
end group functionality for copolymerization in the
presence of a chain transfer agent.
3
4
wood appear to have reported the first systematic
study of such a copolymerization where attention was
paid to the nature of the chain ends formed. They
copolymerized styrene and MMA and reported that the
fraction of chain ends derived from the two monomers
directly reflected the overall copolymer composition.
3
0
We reported in a patent and in several preliminary
2
9,35,36
communications
that with appropriate choice of
reaction conditions polymerization of a monosubstituted
monomer (e.g. BA, styrene, vinyl benzoate) as the major
component in the presence of a chain transfer catalyst
Experimental Section
1H NMR spectra were obtained with a Bruker DRX 500
spectrometer on samples dissolved in deuteriochloroform

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9039
AMS dimer (6) was supplied by NOF (Japan). The allyl
sulfides, R-(tert-butylthiomethyl)styrene (7) and ethyl 2-(tert-
butylthiomethyl)acrylate (8) were synthesized according to the
5
0
literature procedures.
n
Figure 1. COSY spectrum of polystyrene (14, Mh 1630)
showing connectivities between the end group and adjacent
hydrogens. The spectrum was obtained at 298 K in CDCl
3
using a Bruker DRX 500 spectrometer at (see text for further
experimental details).
(
3
CDCl ). Chemical shifts are reported in ppm from TMS
rounded to the nearest 0.05 ppm and coupling constants are
quoted to the nearest 0.5 Hz. The COSY experiment (Figure
1
) used a standard Bruker library sequence (cosygpqf45) with
the following parameters: 4096 FID data points, 6510 Hz
sweep width, 0.31 s acquisition time, 1.0 s relaxation delay,
and 512 experiments, multiplied by an unshifted sine function
in both dimensions and Fourier transformed over 2048 × 1024
points. Number-average degrees of polymerization (DP),
MeCoIII(DEG)BF
mato)(2-)-O,O′]]tetrafluorodiborato(2-)-N,N′,N′′,N′′′](methyl)-
2 2
) . The [bis[µ-[(2,3-hexanedione dioxi-
III
(aquo)cobalt [MeCo (DEG-BF
2 2
) ] was prepared following a
i
III
procedure similar to that used previously for PrCo (DMG-
5
1
number-average molecular weights (Mh
ratio of the weight-average and the number-average molecular
weightss Mh / Mh ) were determined by gel permeation chroma-
n
) and dispersities (the
2 2
BF ) .
A solution of 3,4-hexandione (5.3 g, 0.046 mol), hydroxy-
lamine hydrochloride (5.3 g, 0.082 mol) in ethanol (53 mL),
and pyridine (5 mL) was heated under reflux for 1 h and
allowed to cool to ambient temperature when ice water (50
mL) was added. The gray crystals of 3,4-hexanedionedioxime
that separated on standing at 0 °C were collected and
recrystallized from ethanol (6.4 g, 95%). 3,4-Hexanedione
dioxime (diethyl glyoxime, DEG) had mp 183-184 °C.
Sodium hydroxide (1.25 g, 0.031 mol in 50% aqueous
w
n
tography (GPC) performed on a Waters Associates liquid
chromatograph equipped with differential refractometer and
a set of Waters UltraStyragel columns (300 mm × 7.5 mm,
6
5
4
3
5
µm particle size, and 10 , 10 , 10 , 10 , 500 and 100 Å pore
size). The column set gave an approximately linear calibration
7
-1
for the molecular weight range 200 to 1 × 10 g mol that
was fitted to a third-order polynomial. Tetrahydrofuran (flow
rate of 1.0 mL/min) was used as eluent at 22 ( 1 °C. The
columns were calibrated with narrow polydispersity poly-
styrene standards. High performance liquid chromatography
2 2
solution) was added to a stirred solution of CoCl ‚6H O (3.2 g,
0.013 mol) and 3,4-hexanedione dioxime (4.6 g, 0.032 mol) in
degassed methanol (70 mL) at room temperature. After 10 min,
pyridine (1.02 mL, 0.013 mol) was added slowly. The reaction
mixture was then cooled to -20 °C and stirred for 20 min
under nitrogen. Then sodium hydroxide (0.81 g of 50% aqueous
solution, 0.021 mol) and sodium borohydride (0.56 g, 0.015 mol)
were added slowly. Iodomethane (1.09 mL, 2.48 g, 0.017 mol)
was then added dropwise over 20 min, and the reaction
mixture was allowed to warm to room temperature. The
solution was concentrated under vacuum to ca. half its original
(
HPLC) was performed with a Hewlett-Packard 1090 HPLC
system equipped with a diode array detector and a Beck-
man Ultrasphere ODS column (250 mm × 10 mm) with 60:40
water:methanol (flow rate of 3 mL/min) as eluent at 30 °C.
Conversions were determined gravimetrically by determining
the mass of the residue after exhaustive evaporation of
monomer.
Materials. Monomers (styrene, BA, AMS, MMA, MAN)
were obtained from Aldrich and were purified by filtration
through alumina (to remove inhibitors) and flash distilled
under vacuum immediately prior to use.
volume and 40 mL of cold water was added. The precipitated
III
2
Me(py)Co (DEG) was isolated by filtration, washed with an
aqueous solution of pyridine (5% v/v) and dried under vacuum
over P to provide an orange powder (4.9 g, 82%) that was
used directly as follows.
2 5
O
Reagent chemicals, 3,4-hexanedione (Aldrich, 95%), cobalt-
(
(
II) chloride hexahydrate (Aldrich, 98%), sodium borohydride
Aldrich, 99%), and boron trifluoride etherate (Aldrich), were
1
H NMR (CDCl
CH CH ), 2.6 (m, 8H, CH
Me(py)Co (DEG)
3
): δ 0.8 (s, 3H, CoCH
CH ), 8-9.2 (m, pyridine).
(3.6 g, 0.0082 mol) was added over 20
3
), 0.11-0.95 (m, 12H,
2
3
2
3
used without purification.
III
2
Initiators. azobis(isobutyronitrile) (AIBN) (Fluka) was
purified by crystallization from chloroform/methanol; 1,1′-
azobis(1-cyclohexanecarbonitrile) (ACHN) (DuPont VAZO-88)
and 2,2′-azobis(2,4-dimethylpentane) (WAKO VR110) were
used without purification.
min to a solution of boron trifluoride etherate (9.2 g, 0.064 mol)
in diethyl ether (5 mL) at -20 °C and the reaction mixture
was allowed to warm to room temperature and the suspended
III
2 2
Me(py)Co (DEG-BF ) was isolated by filtration, washed with
ether and dried under vacuum to provide 3.2 g of orange-brown
powder. A further 1.1 g was isolated after the filtrate was
allowed to stand. Total yield was 4.3 g (98%).
II
48
and iPrCoIII(DMG-
The syntheses of Co (DPG-BF
)
2 2
4
9
BF
2
)
2
are reported elsewhere.

9040 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
Table 1. Properties of Polymers Formed in Feed Polymerization of Styrene with PrCoIII(DMG-BF2)2 in n-Butyl Acetate
i
at 125 °C
a
b
Mh ne
time
min)
[Co(III)] initial
[Co(III)] feed
[ACHN]
[M]
4
4
c
d
(g mol-1)
Mh w/ Mh nf
(
(M × 10 )
(M × 10 )
feed (M)
initial
convn (%)
[2] (%)
3
6
1
0
1.10
17.7
0.065
3.10
1050
1150
2.18
2.21
2.18
1.69
2.06
2.30
2.03
3.41
3.52
3.26
1.70
1.76
1.68
1.70
1.81
1.70
h
30
52
0
20
ppt
1100
g
1630
0
0
0
6
1
0
0.562
0.296
0
8.46
4.77
0
0.070
0.068
0.066
3.09
3.07
2.71
2010
34
56
20
1720
g
ppt
0
20
1940
6
1
ppt
3270
35
53
2710
g
2750
6
1
1
2
3
3
0
32 200
33 800
38 100
39 500
37 400
39 400
28
46
58
64
75
83
20
80
40
00
60
a
i
III
b
i
III
Initial concentration of PrCo (DMG-BF2)2 in reaction mixture. Concentration of PrCo (DMG-BF2)2 in initiator/complex feed.
c
d
For details of conditions see Experimental Section. Concentration of 1,1′-azobis(1-cyclohexanecarbonitrile) in feed. Initial concentration
e
of styrene in n-butyl acetate. Additional monomer added as feed; see Experimental Section. Number-average molecular weight determined
by GPC. Molecular weight dispersity. g Precipitated (120 min) sample. Not determined.
f
h
Table 2. Properties of Copolymers Formed in Feed Copolymerization of Styrene and AMS (∼10:1 w/w) with
i
III
PrCo (DMG-BF2)2 in n-Butyl Acetate at 125 °C
a
b
Mh ne
time
min)
[Co(III)]initial
[Co(III)]feed
[ACHN]
[M]
4
4
c
d
(g mol-1)
Mh w/ Mh n f
g
(
(M × 10 )
(M × 10 )
feed (M)
initial
convn (%)
[2] (%)
3
6
1
0
0.995
17.7
0.065
3.28
730
740
2.38
2.25
2.06
1.43
2.17
2.21
1.80
2.11
2.11
1.89
2.56
2.55
2.28
1.72
1.87
1.80
8
19
34
0
20
ppt
690
h
1270
32
56
65
6
1
0
0.537
0.295
0
8.84
4.77
0
0.064
0.071
0.066
3.28
3.26
2.90
1170
24
40
20
1040
h
ppt
0
20
1470
6
1
1370
22
39
1270
h
ppt
1660
2
4
6
1
2
3
0
0
0
20
40
60
19 700
14 900
17 100
24 400
27 400
29 400
i
i
17
30
45
68
a
i
III
b
i
III
Initial concentration of PrCo (DMG-BF2)2 in reaction mixture. Concentration of PrCo (DMG-BF2)2 in feed. For details of
c
d
conditions, see Experimental Section. Concentration of 1,1′-azobis(1-cyclohexanecarbonitrile) in feed. Initial concentration of total
e
monomers in n-butyl acetate. Additional monomer added as feed; see Experimental Section. Number-average molecular weight determined
f
g
1
by GPC. Molecular weight dispersity. Fraction of macromonomer chain ends (2) from H NMR analysis () 100.[2]/([2] + [5])).
h
Precipitated (120 min) sample. i Not determined.
1H NMR (CDCl
): δ 0.8 (s, 3H, CoCH
), 1.0-1.2 (m, 12H,
), 8-9.2 (m, pyridine).
(3.2 g, 0.0060 mol) was added to
22, and 69 h. The NMR spectra broadened substantially
indicating formation of a paramagnetic Co(II) complex.
Polymerizations. Experiments Using a Monomer Feed.
The following procedure is typical. Details of other experiments
are provided in Tables 1-3.
3
3
CH
2
CH
3
2 3
), 2.65 (m, 8H, CH CH
III
Me(py)Co (DEG-BF
2
)
2
degassed water under nitrogen at 30 °C. The solution was held
III
2 2
at 30 °C for 40 min. The Me(aquo)Co (DEG-BF ) was
isolated by filtration, washed with water, and dried to constant
weight under vacuum. The bright orange powder (2.8 g, 99%)
The solvent, n-butyl acetate (20 g), in a 5 neck 250 mL
reactor, equipped with condenser and a mechanical stirrer was
purged with nitrogen. Styrene (10 g) and AMS (1 g) were added
and the solution was purged with nitrogen for a further 10
min. The reactor was then heated to reflux (ca. 125 °C) and
1
was used without further purification. H NMR and reverse
phase HPLC analysis indicated >95% purity.
1
H NMR (CDCl
CH ), 2.67 (q, 8H, CH
acetone-d
3
): δ 0.8 (s, 3H, CoCH
3
), 1.0-1.1 (m, 12H,
1
CH
2
3
2
CH
3
), 4.8 (br s, 2H, H
), 1.2 (t, 12H, CH
, J ) 7.5 Hz), 4.9 (br s, 2H,
2
O). H NMR
i
III
2 2
PrCo (DMG-BF ) (1.4 mg in 5 mL of degassed butyl acetate)
(
)
H
6
): δ 1.0 (br s, 3H, CoCH
3
2
CH
3
J
added. The monomer (styrene 13.6 g and AMS 1.6 g at 0.139
7.5 Hz), 2.85 (q, 8H, CH
O).
CH
2 3
mL/min) and initiator/transfer agent (1,1′-azobis(1-cyclohex-
i
III
2
2 2
anecarbonitrile), 0.93 g, PrCo (DMG-BF ) , 4.6 mg in de-
A sample of MeCoIII(DEG-BF
) (15 mg) in acetone-d (1
2 2 6
gassed n-butyl acetate, 6.7 g at 0.063 mL/min) feeds were
commenced immediately and completed over 120 min. The
reaction mixture was then cooled. The reactor was sampled
at regular intervals to monitor intermediate molecular weights
mL) was degassed by three freeze-thaw-evacuate cycles,
-
2
sealed under vacuum (10 mmHg), and heated at 80° for 69
h. NMR spectra were recorded after 1, 22, and 69 h. The NMR
1
spectra were unchanged from the original spectrum.
(GPC) and conversions ( H NMR). Precautions to exclude
III
A sample of MeCo DEG-BF
2
(16 mg) and AIBN (2.4 mg)
ingress of air were taken during all process and sampling
steps. End groups were also determined on samples that were
precipitated into a 20-fold excess of methanol. In this process
some fractionation occurred on precipitation with loss of lower
in acetone-d (1 mL) was degassed by three freeze-thaw-
6
-
2
evacuate cycles, sealed under vacuum (10 mmHg), and
heated at 80 °C for 69 h. NMR spectra were recorded after 1,

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9041
Table 3. Properties of Copolymers Formed in Feed Copolymerization of Styrene and AMS (∼5:1) with
i
III
PrCo (DMG-BF2)2 in n-Butyl Acetate at 125 °C
b
Mh ne
time
min)
[Co(III)]
[Co(III)] feed
[ACHN]
[M]
4
a
4
c
d
(g mol-1)
Mh w/ Mh n f
g
(
(M × 10 )
(M × 10 )
feed (M)
initial
convn (%)
[2] (%)
6
1
0
0.965
0.498
0.270
23.8
9.10
4.68
0.066
0.067
0.066
3.45
3.44
3.42
380
410
1310
810
780
1180
1760
1640
2140
2660
3300
4650
5160
16 700
19 500
19 600
2.45
2.10
1.83
2.06
1.96
1.53
1.98
1.93
1.60
1.87
1.63
1.81
1.64
1.94
1.79
1.83
i
i
20
h
ppt
6
1
0
10
22
20
h
ppt
0
20
68
6
1
ppt
20
35
h
>95
>70
>70
1
20
0.173
0.090
0
3.43
3.43
3.99
20
12
h
ppt
6
0
h
ppt
6
1
0
0
0.066
i
i
20
h
ppt
a
i
III
b
i
III
Initial concentration of PrCo (DMG-BF2)2 in reaction mixture. Concentration of PrCo (DMG-BF2)2 in feed. For details of
c
d
conditions, see Experimental Section. Concentration of 1,1′-azobis(1-cyclohexanecarbonitrile) in feed. Initial concentration of total
e
monomers in n-butyl acetate. Additional monomer added as feed; see Experimental Section. Number-average molecular weight determined
f
g
1
by GPC. Molecular weight dispersity.i Percentage of macromonomer chain ends (2) from H NMR analysis () 100.[2]/([2] + [5])).
h
Precipitated (60 or 120 min) sample. Not determined.
Table 4. Properties of Copolymers Formed in Batch Copolymerization of Styrene and AMS (∼10:1) with
i
III
PrCo (DMG-BF2)2 in n-Butyl Acetate
c
-5
Mh n (g mol-1)
d
Mh w/ Mh ne
f
temp (°C)
25a
[styrene] (M)
2.53
[AMS] (M)
[Co]/[M] × 10
convn (%)
[2] (%)
1
0.257
0.251
0.246
0.236
0.242
0.237
0.232
0.223
0
64 500
1410
750
1.7
1.8
1.8
1.6
2.0
1.6
1.5
1.3
2
2
2
.47
.42
.32
1.89
3.77
7.54
0
1.84
3.70
7.39
54
39
36
70
56
52
450
8
0b
2.38
32 600
1460
1090
660
2
2
2
.33
.28
.19
45
33
22
70
50
33
a
,2′-Azobis(2,4-dimethylpentane) initiator. 1,1′-Azobis(4-cyclohexanecarbonitrile) initiator. Ratio of [ PrCoIII(DMG-BF2)2] to total
b
c
i
2
d
e
f
monomer concentration. Number-average molecular weight determined by GPC. Molecular weight dispersity. Percentage macromono-
mer chain ends (2) from H NMR analysis ()100.[2]/([2] + [5])).
1
Table 5. Properties of Copolymers Formed in Batch Copolymerization of Styrene and AMS (10:1) with Various Cobalt
Complexes in n-Butyl Acetate at 125 °C and 2,2′-Azobis(2,4-dimethylpentane) Initiator (0.0003 M)29
d
-5
e
-1)
Mh w/ Mh nf
g
complex
[styrene] (M)
[AMS] (M)
[Co]/[M] × 10
Mh n (g mol
convn (%)
[2] (%)
control
DMGa
control
2.48
2.48
2.48
2.48
2.46
2.46
0.248
0.248
0.248
0.248
0.246
0.246
0
58 300
1060
72 300
1390
71 900
1450
1.80
1.62
1.73
1.70
1.70
1.74
13
19
15
19
12
23
6.1
0
6.1
0
6.1
0.71
0.85
0.91
b
DEG
control
DPGc
a
iPrCoIII(DMG-BF2)2. b MeCoIII(DEG-BF2)2. c Co (DPG-BF2)2. Ratio of cobalt complex concentration to total monomer concentration.
II
d
e
f
f
Number-average molecular weight determined by GPC. Molecular weight dispersity. Mole fraction of macromonomer chain ends (2)
from H NMR analysis () [2]/([2] + [5])).
1
1
molecular weight material. The results of this experiment
that was analyzed by H NMR and GPC. No precipitation or
(
molecular weights, end group purities) are provided as the
fractionation was performed. Details of concentrations and
first set of entries in Table 2.
reaction conditions used are provided in Tables 4-9. The
The rate of monomer feed was chosen such that the
concentration of polymer did not exceed 30 wt % of monomer
to minimize further reaction of the formed macromonomer. The
molecular weights given in the tables for samples with Mh
n
<600 are an overestimate because of the interference from
peaks in the chromatogram due to low molecular weight
impurities (e.g. solvent, monomer, cobalt complex, etc.).
The typical procedure was as follows:
i
III
initial concentration of PrCo (DMG-BF
2 2
) and the feed rate
were chosen so as to maintain the molecular weight by keeping
i
III
the ratio [ PrCo (DMG-BF
2
)
2
]:[total monomers] approxi-
A mixture of styrene (1.3 g, 12.5 mmol), R-methylstyrene
mately constant. The results of these experiments are pre-
sented in Tables 1-3.
(0.15 g, 1.27 mmol) (molar monomer ratio 10/1), n-butyl acetate
-
5
(3 g), 2,2′-azobis(2,4-dimethylpentane) (8.9 × 10 g), and the
i
III
Batch Polymerization Procedure. Stock solutions of the
monomers, initiator, and the cobalt complex in n-butyl acetate
were prepared and appropriate aliquots transferred to ampules
which were degassed through four freeze-evacuate-thaw
cycles. The ampules were sealed under vacuum and heated in
a thermostated bath for the requisite times. The ampules were
then rapidly cooled, opened and the solvent and excess
monomers were removed under vacuum to provide a residue
2 2
appropriate amount of PrCo (DMG-BF ) (to provide the
concentration shown in Table 4) was placed in each of four
ampules which were degassed by four freeze-evacuate-thaw
cycles. The ampules were then sealed and the polymerization
mixtures heated at 125 °C for 2 h. The ampules were then
cooled rapidly and opened. A sample of each polymerization
mixture (∼10 µL) was transferred to a NMR tube, diluted with
1
CDCl
3
(∼0.5 mL) and analyzed by H NMR to determine the

9042 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
Table 6. Properties of Copolymers Formed in Batch Copolymerization of BA and AMS Copolymerization with
i
III
PrCo (DMG-BF2)2 in n-Butyl Acetate
temp (°C) (BA:AMS)a
[BA] (M)
2.62
[AMS] (M)
[Co]/[M] × 10
d
-5
Mh n (g mol-1)
e
Mh w/ Mh nf
convn (%)
AMSg
[2]h
b
1
25 (∼50:1)
0.044
0.043
0.042
0.041
0.038
0.033
0.089
0.087
0.085
0.081
0.075
0.047
0.046
0.045
0.043
0.040
0.035
0.093
0.090
0.088
0.085
0.078
0.187
0.183
0.179
0.172
0.365
0.355
0.350
0.337
0
18 900
3460
1950
1620
1200
1070
18 100
2230
1400
1200
1070
56 100
1880
1600
1480
1210
1170
41 400
1150
1060
960
1.85
2.10
1.92
1.81
1.72
1.65
1.77
2.10
1.68
1.73
1.58
1.76
1.96
1.82
1.80
1.76
1.64
1.69
1.48
1.42
1.39
1.37
1.64
1.33
1.31
1.27
1.75
1.20
1.20
1.20
6
43
32
30
21
21
13
20
13
13
12
14
11
11
13
10
10
9
0.05
0.05
0.06
0.06
0.07
0.06
0.17
0.2
0.13
0.13
0.12
0.12
0.10
0.10
0.09
0.09
0.09
0.17
0.2
0.18
0.17
0.15
0.38
0.36
0.35
0.36
0.39
0.41
0.38
0.43
2
2
2
2
1
.56
.51
.41
.23
.93
1.91
3.81
7.61
15.22
30.41
0
1.88
3.75
7.49
14.97
0
1.73
3.44
6.88
13.77
27.55
0
1.81
3.59
7.20
14.41
0
1
0.87
0.77
0.73
i
b
1
25 (∼25:1)
2.62
2
2
2
2
.56
.51
.41
.22
1
1
0.93
0.85
c
8
0 (∼50:1)
2.91
2
2
2
2
2
.83
.78
.67
.47
.15
1
1
0.91
0.85
0.8
c
8
0 (∼25:1)
2.72
2
2
2
2
.67
.61
.50
.29
7
1
1
0.96
0.91
6
5
940
6
c
8
0 (∼10:1)
2.30
28 200
640
4
2
2
2
.25
.20
.11
2.03
4.06
8.12
0
1.94
3.86
7.71
5
1
1
1
610
550
5
3
c
8
0 (∼5:1)
2.24
23 500
500
3
2
2
2
.18
.15
.07
4
1
1
1
490
480
4
3
a
b
-4
g). c Initiator was
Nominal ratio of monomers in parentheses. Initiator was 2,2′-azobis(2,4-dimethylpentane) (3.74 × 10
-4
d
e
azobis(isobutyronitrile) (3.74 × 10 g). Ratio of cobalt complex concentration to total monomer concentration. Number-average molecular
f
g
h
weight determined by GPC. Molecular weight dispersity. Mole fraction of AMS in copolymer. Mole fraction of macromonomer chain
1
1
ends (2) from H NMR analysis ()[2]/([2] + [5])). Value of “1” indicates that the end groups (5) were not detectable by H NMR (i.e., >97%
i
(
2)). Not determined.
Table 7. Properties of Copolymers Formed in Batch Copolymerization of BA and MMA (∼10:1) with PrCoIII(DMG-BF2)2
i
in n-Butyl Acetate
c
-5
Mh nd (g mol-1
Mh w/ Mh n e
MMAf
[2]g
temp °C
0a
[BA] (M)
1.73
[MMA] (M)
[Co]/[M] × 10
)
convn (%)
8
0.168
0.164
0.161
0.154
0.143
0.124
0.168
0.164
0.161
0.154
0.143
0.124
0
75 500
12 120
2600
1500
1200
920
10 400
3510
1990
1220
1030
830
2.08
1.82
1.66
1.50
1.43
1.31
2.56
1.74
1.7
54
53
21
9
10
8
76
45
32
14
15
9
0.14
0.14
0.17
0.18
0.17
0.14
0.11
0.14
0.15
0.17
0.15
0.14
1
1
1
1
1
.69
.65
.59
.47
.28
2.67
5.34
11.0
21.0
43.0
0
2.67
5.34
11.0
21.0
43.0
1
0.94
0.92
0.86
0.75
1
25b
1.73
1.69
1.65
1.59
1.47
1.27
0.93
0.92
0.91
0.87
0.79
1.6
1.73
1.51
a
Initiator was 2,2′-azobis(2,4-dimethylpentane) (3.74 × 10 g). b Initiator was azobis(isobutyronitrile) (3.74 × 10 g). c Ratio of cobalt
-4
-4
d
e
complex concentration to total monomer concentration. Number-average molecular weight determined by GPC. Molecular weight
dispersity. Mole fraction of MMA in copolymer. g Fraction of macromonomer chain ends (2) from H NMR analysis ()[2]/([2] + [5])).
f
1
1
Value of “1” indicates that the end groups (5) were not detectable by H NMR (i.e., >97% (2)).
3
5
monomer conversion. The unconverted monomer and solvent
chain transfer have been published elsewhere. The unsatur-
-
1
1
removed under vacuum (∼10 mmHg) at <40 °C to provide
ated end groups give rise to signals in the H NMR spectrum
1
a residue that was analyzed by H NMR and GPC. The results
3
(CDCl ) as follows: AMS-derived terminal methylene double
of these experiments are shown in Table 4.
2
bond (13), δ 4.8, 1H and δ 5.2, 1H, -C(Ph)dCH ; 1,3-
Structures 9 (AMS-co-S), 10 (BA-co-AMS), 11 (BA-co-MAN),
and 12 (BA-co-MMA) are used to refer to the products of
copolymerizations. The polymer chain will in each case be a
copolymer chain. The end groups shown are the predominant
end groups. The structures a and c are formed by H-transfer
to the monosubstituted monomer (BA or S). Structures b and
d are formed by H transfer to the R-methylvinyl monomer
diphenylpropen-3-yl end group (14), δ 6.1-6.3, 2H, -CH(Ph)-
CHdCH-Ph and δ 3.1, 1H, -CH(Ph)-CHdCH-Ph. The ratio
of the signals at δ 3.1 and δ 4.8 gave the best estimate of
3
5
terminal double bond content. In a previous report it was
suggested the signals at δ 6.1-6.3 were due to only one
hydrogen of 14 with the other appearing within the aromatic
envelope. It has now been established by 2D NMR (COSY) on
(AMS, MMA). Structures a and b have a terminal unit derived
n
a sample of 14 (precipitated sample with Mh 1630, refer to
from the R-methylvinyl monomer. Structures c and d have a
terminal unit derived from the monosubstituted monomer.
Characterization of Copolymers. Poly(AMS-co-S) (9).
Table 1, first-mentioned experiment) that both olefinic reso-
nances appear at δ 6.1-6.2 (Figure 1). Integration supports
this assignment. The signals at δ 6.1-6.3 (-CH(Ph)-CHd
1
The H NMR spectra of S/AMS copolymers made by catalytic
2
CH-Ph), δ 3.1 (-CH(Ph)-CHdCH-Ph), and δ 0.9-1.1 (-CH -

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9043
Table 8. Properties of Copolymers Formed in Batch Copolymerization of BA and MAN (10:1) with PrCoIII(DMG-BF2)2 in
i
n-Butyl Acetate at 125 °Ca
b
-5
Mh n (g mol-1)
c
Mh w/ Mh nd
[2]e
[
BA] (M)
[MAN] (M)
[Co]/[M] × 10
convn (%)
1.84
1.80
1.76
1.69
1.56
1.36
0.179
0.175
0.171
0.164
0.152
0.132
0
9300
1300
1040
960
800
670
1.90
1.54
1.62
1.49
1.47
1.50
8
8
8
8
7
6
2.50
4.99
10.0
20.0
40.0
0.81
0.76
0.80
0.87
0.86
a
Initiator was 2,2′-azobis(2,4-dimethylpentane) (3.74 × 10 g). b Ratio of cobalt complex concentration to total monomer concentration.
-4
c
d
e
Number-average molecular weight determined by GPC. Molecular weight dispersity. Fraction of macromonomer chain ends (2) from
H NMR analysis () [2]/([2] + [5])). Value of “1” indicates that the end groups (5) were not detectable by H NMR (i.e., >97% (2)).
1
1
Table 9. Properties of Copolymers Formed in Batch Copolymerization of BA and AMS Copolymerization (∼50:1) with
a,b
Various Complexes in n-Butyl Acetate at 125 °C
e
-5
f
-1)
Mh w/ Mh ng
AMSh
[2]i
complex
DPGc
[BA] (M)
2.31
[AMS] (M)
[Co]/[M] × 10
Mh n (g mol
convn (%)
0.046
0.045
0.045
0.043
0.041
0.036
0.046
0.046
0.045
0.044
0.042
0
104 700
5400
1960
1820
1420
1260
49 300
2370
1650
1160
1130
2.07
2.10
1.73
1.71
1.64
1.58
1.74
1.85
1.70
1.66
1.57
17
3
0.09
0.13
0.14
0.13
0.12
0.11
0.12
0.13
0.12
0.12
0.12
2
2
2
2
1
.27
.23
.16
.03
.81
2.106
4.225
8.44
16.86
33.75
0
2.07
4.13
8.26
16.58
1
1
1
1
1
1
1
1
1
1
1
1
1
1
DEGd
2.31
11
11
10
5
2
2
2
2
.28
.26
.21
.11
4
a
Initiator was 2,2′-azobis(2,4-dimethylpentane) (3.74 × 10 g). b Results for similar polymerizations with PrCoIII(DMG-BF2)2 are
-4
i
c
II
d
III
e
presented in Table 6. Co (DPG-BF2)2. MeCo (DEG-BF2)2. Ratio of cobalt complex concentration to total monomer concentration.
f
g
h
i
Number-average molecular weight determined by GPC. Molecular weight dispersity. Mole fraction of AMS in copolymer. Mole fraction
1
of macromonomer chain ends (2) from H NMR analysis () [2]/([2] + [5])). Value of “1” indicates that the end groups (5) were not detectable
1
e
by H NMR (i.e., >97% (2)). Fraction of AMS in copolymer.
derived chain ends [2] ()[2]/([2] + [5]) shown in the tables
are based on the ratio of the areas of the signal at δ 5.8 and
those assigned to the -CH
romonomer (10-12a,b).
2
-C(Y)dCHH end of the mac-
A more comprehensive study of the NMR of 9 has recently
3
9
been published by Chui et al. The comparatively high levels
3
9
of AMS used by Chui et al. mean that unsaturated chain
ends derived from BA were not observed.
Poly(AMS-co-BA) Macromonomer (10a,b). 1H NMR
(
acetone-d
CH -C(Ph)d; 3.95, OCH
.25, ArH.
Poly(MAN-co-BA) Macromonomer (11a,b). 1H NMR
CDCl ): δ 0.95, butyl CH ; 1.0-2.3, butyl, backbone H; 2.6,
-C(CN)d; 4.0, OCH ; 5.7, dCHH; 5.85, dCHH.
Poly(MMA-co-BA) Macromonomer (12a,b). 1H NMR
CDCl ): δ 0.9, butyl CH ; 1.0-2.3, butyl, backbone H; 2.55,
-C(CO CH )d; 3.6, OCH ; 4.0, OCH ; 5.5, dCHH; 6.15,
6
): δ 0.9, butyl CH
3
; 1.0-2.3 butyl, backbone H; 2.55,
-
2
2
; 5.0, dCHH; 5.2, dCHH; 7.15-
7
(
-
3
3
CH
2
2
(
-
3
3
CH
2
2
3
3
2
CH(Ph)CH
3
) appear in a 2:1:3 ratio. This COSY spectrum also
dCHH.
enables us to assign a signal at δ 2.3 on the fringe of the
aliphatic envelope to -CH
The use of the signals at δ 6.1 for integration is complicated
by its proximity to the broad aromatic resonance at δ 7.6-
2
CH(Ph)CH
3
.
Macromonomer Chain Transfer Agents. Macromono-
mer 16. Styrene (40 mL), AIBN (0.110 g) and allyl sulfide 7
(5 g) were dissolved in benzene (40 mL). The mixture was
degassed using four freeze-pump-thaw cycles, sealed under
vacuum and heated at 60 °C for 64 h. The ampules was then
cooled, opened, the solvent was removed on a rotary evaporator
7
.2, The signal at δ 3.1 may be low if the internal double bond
product is a mixture of 14 and 15. However, the yield of 15 is
anticipated to be small.
and the residue precipitated into methanol (Mh
.94). The precipitate was redissolved in ethyl acetate and
precipitated into methanol for a total of five precipitations to
n w n
1710, Mh / Mh
1
yield 12.3 g (34% conversion) of macromonomer 16 with Mh
n
2
380 and Mh
w
/Mh
n
1.53. NMR analysis demonstrated no residual
7
.
Macromonomer 17. A solution of styrene (10 g) and allyl
sulfide 8 (1.63 g) in butyl acetate (30 g) was degassed under
nitrogen in a multinecked 250 mL reactor equipped with a
mechanical stirrer. The mixture was heated to reflux (125 °C)
under nitrogen and the two feeds (feed 1, allyl sulfide 8 (6.67
g) in styrene (40 g) at 0.21 mL/min; feed 2, the initiator ACHN
(283 mg) in butyl acetate (20 g)) at 0.063 mL/min) added by
1
BA Copolymers (10-12). Examples of H NMR spectra of
3
9
BA/AMS copolymers have been published elsewhere. The
internal double bond of the unsaturated butyl acrylate-derived
chain end (10-12c,d) gives rise to a relatively sharp “doublet”
2
Bu) and a broader multiplet at δ 7.0
-CHdCH-CO Bu). Fractions of R-methylvinyl monomer-
at δ 5.8 (-CHdCH-CO
(
2
n
syringe pump. After 240 min, the macromonomer 17 ( Mh 1890,

9044 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
II
Table 10. Transfer Constants (CT) of Co (DMG-BF2)2 in
Table 11. Reactivity Ratios (rA, rS) and
Polymerizations of Selected Monomers
Propagation-depropagation Equilibrium Constants (K)
for Selected Monomer Pairs
monomer
BAa
CT
solvent
T (°C)
ref
monomer A monomer S
rA
rS
K
T (°C)
ref
62
∼45 BuAc/MMA
60
80
125
60
60
40
50
60
70
125
80
125
60
60
60
60
60
60
70
80
90
40
50
60
70
60
60
40
50
80
125
80
125
this work
∼
∼
65
90
MMA
BA
2.24
1.789
3.26
2.24
1.62
2.68
2.28
0.67
0.31
0.149
0.022
0.557
0.502
0.32
0.41
0.42
0.14
0.15
0.3
0.414
0.298
0.43
50
91
92
93
93
b
b
BA
∼650
54
55
38
50
80
styrenea
styrenea
1500 bulk
1027 bulk
0.38
0.37
120
5
6
2
33
60
65
MAN
AMS
BA
BA
0.33
0.55
c
0.096
0.20
b
c
styrenea
2700 BuAc
this work
∼
2000 BuAc/AMS
1500 BuAc/AMS
0.153
0.166
0.148
80
120
80
120
60
68
68
e, 69
e, 69
94
∼
b
styrene
3-7000 bulk
∼8-9000 bulk
24 300 33%(v/v) in toluene
36 140 bulk
53,54
53,54
55
6.1
styrenec
MMA
MMA
MMA
MMA
0.171 22.0
0.39
MAN
styrene
55
78
0.37
90
17 900 toluene
33 500 bulk
0.38
120
-20
60
74,76
AMS
AMS
styrene
styrene
10.1
95
67
a
3
3
3
2 800
5 200
2 500
1.0
7.1
17.2
28.5
67
1.09
1.13
1.2
90
0.4
110
150
20
MMA
33 000 bulk
74
0.8
3
3
2
9 000
4 000
7 000
AMS
MMA
0.3
0.5
1.7
96
0.4
0.55
0.55
0.6
5.1
50
0.35
0.2
7.1
60
MMA
BMA
AMS
39 800 bulk, toluene
28 000 bulk, toluene
815 000 bulk
93 000
∼500 000 BuAc/styrene
300 000
∼70 000 BuAc/BA
100 000
90
90
38
12.9
22.9
80
0.05
0.65
100
a
Values calculated using the relationship log rA ) 2.48 - 1094/
8
6
7
b
70,72
T, log rS ) 0.378 - 125/T.
Estimated from Q - e values.
AMS
AMS
this work
this work
c
71 d
Estimated using the patterns of reactivity scheme.
K is the
∼
rate constant for AMS depropagation relative to cross propagation,
6
9
in bulk BA-AMS copolymerization obtained using the Kruger
∼
7
3
model.
a
Apparent transfer constant with monosubstituted monomers
should be considered as a lower limit due to reversible consumption
of the cobalt complex under the reaction conditions (see text).
Value dependent on initiator concentration. Estimated value
of transfer constant after allowing for reversible consumption of
the cobalt complex.
Results and Discussion
b
c
A series of binary copolymerizations of a monosub-
stituted monomer (styrene or BA) with an R-methylvinyl
monomer (MMA, MAN, or AMS) in the presence of a
chain transfer catalyst were performed. Alkylcobalt(III)
Mh
w
/Mh
n
1.58) was isolated by two precipitations into acidified
methanol. The conversion based on isolated macromonomer
was 50%.
i
III
complexes (e.g., Pr-Co (DMG-BF2)2) were used in
most of the polymerizations reported in this paper.
i
These complexes undergo unimolecular homolysis ( Pr-
Macromonomer Transfer Constants. Aliquots of a stock
solution of the initiator (AIBN at 60 or 80 °C; tert-butyl
peroxybenzoate at 100 or 120 °C) in monomer were transferred
to ampules containing the appropriate amount of macromono-
mer (9, 16, or 17) or R-methylstyrene dimer (6) in monomer
which were then degassed through three freeze-evacuate-
thaw cycles. The ampules were sealed under vacuum and
heated in a thermostated bath for the requisite times. The
ampules were then cooled, opened and the solvent and excess
monomers were removed under vacuum to provide a residue
that was analyzed by GPC. No precipitation or fractionation
was performed. Reaction conditions were chosen to provide
conversions of <10%. Reaction conditions and molecular
weight data are summarized in Table 12. Values of transfer
constants evaluated by analysis of the chain length distribu-
tions according to the log CLD method are provided in Table
III
Co (DMG-BF2)2sScheme 5) or radical induced decom-
III
position (Me-Co (DEG-BF2)2sScheme 6) to produce
the corresponding cobalt(II) complex as the active
9
transfer agent under polymerization conditions. The
cobalt(III) complexes are preferred over the analogous
cobalt(II) complexes as they are substantially less air
sensitive particularly in solution and this simplifies the
protocol for setting up the reactions. Generation of the
active cobalt(II) species should not be rate limiting at
temperatures >60 °C. In general, conditions were
chosen so as to give low molecular weights ( Mh n in range
5
00-5000). The polymers formed were analyzed by GPC
1
for molecular weight and by H NMR for end group
composition. These results are summarized in Tables
1-9.
13. The typical procedure was as follows:
Aliquots (5 mL) of a solution on tert-butyl perbenzoate (11.4
Good molecular weight control was observed in copo-
lymerizations of styrene with AMS (see Figures 2-4 and
Tables 1-4). A first series of experiments (Tables 1-3)
was conducted using a feed addition protocol (See
Experimental Section) under which monomers and
transfer agent were fed to a stirred reactor at a rate
such that the macromonomer molecular weight should
remain constant by keeping the ratio [ PrCo (DMG-
BF2)2]:[total monomers] approximately constant. This
strategy also maintained polymer concentration at <30
wt % vs monomer thus minimizing the likelihood that
mg in styrene 25 mL, 3.4 M) were added to each of four
ampules containing macromonomer 17 (0, 24.5 mg, 50.9 mg,
7
4.9 mg respectively). The ampules were degassed through
three freeze-evacuate-thaw cycles. The ampules were then
sealed under vacuum and heated in a thermostated bath at
1
20.0(0.1 °C for 30 min. The ampules were then cooled
rapidly, opened and the remaining monomer removed under
vacuum at 30 °C. The monomer conversions were <10%. The
molecular weights determined by GPC analysis are reported
in Table 12. The transfer constant evaluated from these data
is given in Table 13.
i
III

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9045
Table 12. Molecular Weights Obtained in Styrene
Polymerizations in the Presence of Macromonomera
Chain Transfer Agents (9, 16, 17) and AMS Dimer (6)
Table 13. Transfer Constants (CT) Measured for
Macromonomer Transfer Agents (9, 16, 17) and AMS
Dimer (6) Determined by log CLD Method
transfer
agent
temp
(°C)
[CTA]
Mh nk
temp CT (this
2
(g mol-1)
CTa,89 CTa,79
(M × 10 )
Mh w/ Mh n
transfer agent
monomer (°C)
work)
9a
9b
100
0
265 000
222 000
183 000
191 000
224 000
209 000
146 000
97 000
64 600
70 900
69 400
68 400
180 000
169 000
147 000
133 000
67 400
70 800
68 600
66 800
125 000
115 000
97 100
88 600
567 700
315 000
178 100
156 400
93 900
38 700
25 600
19 100
387 600
43 800
17 700
14 400
140 100
24 400
15 500
1180
1.76
1.72
1.74
1.68
2.21
1.74
2.00
1.86
1.65
1.50
1.45
1.42
2.25
1.64
1.59
1.60
1.59
1.47
1.44
1.43
1.83
1.70
1.63
1.57
1.90
1.86
1.86
1.80
1.63
1.86
1.70
1.70
1.81
2.03
2.18
1.89
1.90
2.21
2.11
1.93
AMS dimer (6)
0.16
1
1
1
00
00
00
2.197
4.376
6.513
0
sty
60
80
0.24c
0.27
0.25
0.23
100
110
120
130
120
110
100
120
80
120
80
120
120
0.26
120
0.28 0.20b
0.36
120
120
120
80
2.085
4.333
6.538
0
1.09
2.15
3.14
0
1.016
2.098
3.164
0
1.31
2.64
4.01
0
1.357
2.707
3.984
0
1.340
2.660
4.048
0
9.689
17.73
25.09
0
8.335
17.18
24.71
0
0.28
0.09
0.48
BMA
MMA
sty
1
1
1
1
1
6c
0.13
8
8
8
0
0
0
macromonomer 9c
macromonomer 16d
macromonomer 17e
0.03
0.16
∼0
sty
sty
6d
120
sty
0.11
∼0
120
120
120
80
sty
sty
0.15
0.14
BMA
7e
a
Determined by Mayo method. b A mixed MMA-AMS dimer
8
8
8
0
0
0
was reported to have a similar transfer constant (0.19 with
79 c
styrene, 0.10 with MMA).
Styrene macromonomer 9, Mh n 1180,
7f
prepared by copolymerization of AMS with Sty and 50 ppm of
120
Co(III). Value takes into account the end group purity ()[2]/{[2]
1
1
1
20
20
20
d
+
[5]} × 100) of 68% estimated by NMR (see Table 3). Styrene
macromonomer 16, Mh n 2380, prepared by addition-fragmentation
e
7/BMAg
chain transfer with corresponding allyl sulfide 7. Styrene mac-
120
romonomer 17, Mh n 1880, prepared by addition-fragmentation
chain transfer with corresponding allyl sulfide 8.
120
120
120
80
AMS dimer (6)h
Scheme 4
8
8
8
0
0
0
AMS dimer (6)i
AMS dimer (6)j
100
1
1
1
00
00
00
120
1
1
1
20
20
20
9.520
17.01
25.89
a
Initiators used tert-butyl perbenzoate (tBPB) and azobis-
Scheme 5a
isobutryronitrile (AIBN). The reaction time and initiator concen-
b
-3
-3
trations provided in footnotes b-k. 147 min, tBPB 3.43 × 10
M. c 35 min, tBPB 2.951 × 10
-3
M. 80 min, AIBN 6.18 × 10
d
M. 32 min, tBPB 3.40 × 10 M. f 80 min, AIBN 6.28 × 10 M.
e
-3
-3
g
-3
h
BMA polymerization, 30 min, tBPB 3.40 × 10 M. 31 min,
tBPB 6.08 × 10 M. i 90 min, tBPB 6.22 × 10 M. 180 min,
-
4
-3
j
-
3
k
-3
l
AIBN 2.95 × 10 M. 46 min, tBPB 3.07 × 10 M. Values
provided to 3 significant figures.
the macromonomers 2 would undergo further reaction
by copolymerization or chain transfer (see later discus-
sion). The polymerizations were carried out at 125 °C
a
Z is solvent.
(
n-butyl acetate at reflux) with continuous addition of
a rapidly decomposing initiator [ACHN; t1/2 at 125 °C
5 min] to maintain the polymerization rate. Thermal
The product of styrene-AMS copolymerization con-
tained a mixture of end group structures 2 and 5 (Tables
-4). The absolute number of end groups determined
∼
initiation also occurs under these conditions. However,
the relative importance of the two initiation processes
was not quantified.
1
by NMR (see Experimental Section) was in accord with
that expected on the basis of the GPC determined
molecular weight. By selecting the ratio of AMS:styrene
and the cobalt concentration, conditions could be chosen
such that product 2 dominates (favored by high AMS
concentration and low cobalt concentration (see Tables
A second set of experiments was conducted in sealed
ampules at 80 or 125 °C and only taken to low conver-
sion of monomer (Table 4). Slightly higher molecular
weights and slightly lower end group purities were
observed in the sealed tube experiments. Given the
difference in reaction conditions and that end group
analysis on the product from the feed experiments was
carried out on precipitated samples, the agreement
between the experiments is considered within experi-
mental error.
1
1-4) and 5 is not detectable by H NMR. End group
compositions observed in the sealed tube experiments
conducted at 125 and 80 °C are shown in Figure 3 and
Figure 4, respectively. Higher macromonomer purity
(greater amount of 2) was observed for the higher
reaction temperature. The temperature effect may in

9046 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
Scheme 6a
a
Z is solvent; the byproduct (?) was not determined.
Figure 4. Comparison of observed and predicted properties
of styrene-AMS formed in copolymerization ([styrene]:[AMS]
i
III
∼
2 2
5:1) in the presence of Pr-Co (DMG-BF ) at 125 °C: (a)
fraction of AMS-derived chain ends [2]/([2] + [5]) (experimen-
tal, b; predicted, - -), (b) predicted fraction of AMS in copolymer
(
n
- -), and (c) number-average molecular weight ( Mh experi-
n
mental, 4; Mh predicted, -). Experimental conditions are
summarized in Table 4. Parameters used in the fit are given
in Table 14.
phenomena have been associated with the reversible
2
8,52-54
formation of a polystyryl-cobalt(III) species.
At
1
25 °C, there may be less formation of such species or
the reaction may be more readily reversible. The
estimated transfer constant to styrene (ca. 2700, from
Figure 2. “Mayo” plot (reciprocal degree of polymerization
1/ Xh ) vs [catalyst]/([total monomers]) for styrene-AMS feed
copolymerization at 125 °C in the presence of Pr-Co (DMG-
BF
with [styrene]:[AMS] (a) ∼100:0 (b, - -), (b) ∼90:10 (0,
Mayo plot shown in Figure 2) appears approximately
(
n
8,55
i
III
2-fold higher than that reported3
at lower tempera-
tures with the corresponding cobalt(II) complex (see
Table 10). However it is still significantly lower than
the estimated value of transfer constant after allowing
for reversible consumption of the cobalt complex.5
The following expression, eq 1a, which defines the
meaning of the average transfer constant Ch tr as may
be derived from a Mayo plot of 1/ Xh n vs total monomer
concentration.
2
)
2
-
-), (b) ∼80:20 (O, -). Experimental conditions are sum-
marized in Tables 1-3.
3,54
1
1
[T]
[A] + [S]
)
+ Ch tr
(1a)
Xh n Xh n0
where Ch tr ) kh tr/ kh p. A, S, and T are defined in Scheme 6.
By making use of a long chain approximation the
following expressions (eq 1b,c), relating Ch tr to values of
the reactivity ratios and the individual transfer con-
stants for a particular monomer ratio, can be de-
rived.5
6,57
Figure 3. Comparison of observed and predicted properties
of styrene-AMS formed in copolymerization ([styrene]:[AMS]
[A] + [S]
[A] + [S]
i
III
∼
5:1) in the presence of Pr-Co (DMG-BF
)
2 2
at 80 °C: (a)
Ch tr ) ACa
+ SCs
(1b)
(1c)
[
A] + r [S]
rs[A] + [S]
fraction of AMS-derived chain ends [2]/([2] + [5]) (experimen-
tal, b; predicted, - -), (b) predicted fraction of AMS in copolymer
a
AC r + SC r
(
- -), and (c) number-average molecular weight (Mh
n
experi-
a
a
s
s
Ch tr
)
n
mental, 4; Mh predicted, -). Experimental conditions are
2
2
A r + 2AS + S r
summarized in Table 4. Parameters used in the fit are given
in Table 14.
a
s
where A is the mole fraction of monomer A. It can be
seen that Ch tr is a complex function of the reactivity
ratios and individual transfer constants.
large part be associated with the temperature depen-
dence of the reactivity ratios (see below).
II
Good molecular weight control was also observed in
homopolymerization of styrene at 125 °C (Table 1). No
dependence of conversion on cobalt concentration was
seen and no broadening of molecular weight distribution
was observed for higher conversions (See Table 1). Many
previous studies on catalytic chain transfer in styrene
polymerization at lower reaction temperatures have
indicated that inhibition periods, catalyst poisoning and/
or retardation complicate the polymerization.27,28 These
The transfer constant for Co (DMG-BF2)2 toward the
AMS propagating radical were estimated from the
slopes of the Mayo plots shown in Figure 2 for the
styrene-AMS copolymerizations at 125 °C by solving
eq 1a. In this analysis, 1/ Xh n for the copolymers was
II
approximated as 104/ Mh n. The transfer constant of Co -
(DMG-BF2)2 in styrene polymerization was taken to be
2700 (see above). The values of the reactivity ratios for
styrene-AMS copolymerization used are shown in Table

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9047
Figure 6. “Mayo” plot (reciprocal degree of polymerization
Xh ) vs [catalyst]/([total monomers]) for BA-MMA copolym-
erization in the presence of Pr-Co (DMG-BF ) for [BA]:
Figure 5. “Mayo” plot (reciprocal degree of polymerization
(
n
( Xh ) vs [catalyst]/([total monomers]) for BA-AMS copolymer-
n
i
III
i III
ization in the presence of Pr-Co (DMG-BF
2
)
2
at 80 °C with
BA]:[AMS]: (a) ∼5:1 (2, - -), (b) ∼10:1 (O, -), (b) ∼25:1 (0,
-), and (c) ∼50:1 (4, - -). Experimental conditions are
2 2
[
MMA] ∼10:1 at (a) 60 °C (O, -), (b) 80 °C (2, - -), and (c)
[
-
125 °C (0, - -). Experimental conditions are summarized in
Table 5.
summarized in Table 7.
1
1. This analysis suggests a transfer constant for CoII-
DMG-BF2)2 toward the AMS propagating radical in
styrene-AMS copolymerization at 125 °C of ∼80000 (
000. This transfer constant, while very high, is sub-
stantially lower than that reported to apply at lower
(
5
3
8
temperatures (see Table 10). A similar determination
of the transfer constant at 80 °C was not possible
because of significant curvature in the Mayo plot. The
reasons for curvature are discussed further below.
Molecular weight control and macromonomer purity
obtained in copolymerizations of BA in the presence of
i
III
PrCo (DMG-BF2)2 were strongly dependent on the
comonomersAMS (Table 6), MMA (Table 7), or MAN
Table 8). Macromonomer purity is significantly higher
(
Figure 7. Fraction of AMS-derived chain ends 100.[2]/([2] +
[5]) observed in low conversion BA-AMS copolymerization in
with AMS as comonomer. In copolymerizations of BA
with AMS, macromonomer purity is higher for lower
reaction temperatures. The overall copolymer composi-
tion is also strongly temperature dependent which is
attributable to the temperature dependence of the
reactivity ratios. For BA-MMA copolymerization the
temperature sensitivity of copolymer composition and
end group composition was within experimental error
i
III
the presence of Pr-Co (DMG-BF
)
2 2
at 80 °C with [BA]:
[
(
AMS]: (a) 5:1 (2, - - -), (b) 10:1 (0, -), (b) 25:1 (O, - -), and
c) 50:1 (4, - -). Experimental conditions are summarized in
Table 7.
ratio of the two monomers ([S]:[Α]) in the feed. This
effect is discussed further below. End group purity is,
nonetheless, affected by cobalt concentration. Higher
cobalt concentrations generally lead to a lower fraction
of the desired macromonomer chain ends 2ssee for
example Figure 7.
(Table 7). AMS appears to be the most effective R-me-
thylvinyl comonomer of those used since it provides
relatively high macromonomer purities for very low
AMS concentrations (2 mol % relative to BA).
The effect of cobalt complex type on end group purity
was also briefly explored in styrene-AMS copolymeri-
zation (Table 5)29 and BA-AMS copolymerization (Tables
6 and 9). A higher fraction of the desired macromonomer
Analysis of the BA copolymerization data by way of
a “Mayo analysis” revealed a marked deviation from a
linear trend of 1/ Xh n with Mh n for high cobalt concentra-
tions. The effect is most marked for BA-AMS copolym-
erization (Figure 5) though similar trends are seen in
BA-MMA (Figure 6) and BA-MAN copolymerizations.
There appears to be a “saturation effect” such that above
a certain cobalt concentration, the effect on Mh n of further
increasing cobalt concentration is dramatically reduced.
This effect is attributed to the two propagating species
III
chain ends 2 (i.e. higher [2]:[5]) was found when MeCo -
II
(DEG-BF ) or Co (DPG-BF ) was used. In the case
2
2
2 2
of styrene-AMS (∼10:1) copolymerization with the
amount of the undesired product [5] can be more than
halved (Table 5). In the case of BA-AMS copolymeri-
zation with the latter two complexes, macromonomer
purities are improved to the extent that only chain ends
2 were detectable for BA-AMS ∼ 50:1 (Table 9).
BA-AMS copolymerization was substantially re-
(
that ending in BA and that ending in R-methylvinyl
monomer) having markedly different reactivities toward
the chain transfer catalyst. Transfer constants in poly-
merizations of R-methylvinyl monomers are much higher
than those for monosubstituted monomers (Table 10),
and this is confirmed in the present work.
II
tarded with Co (DPG-BF ) at 125 °C even with very
2 2
low concentrations of the complex (Table 9). No similar
III
retardation was observed with MeCo (DEG-BF ) or
2 2
i
III
PrCo (DMG-BF ) . In BA-MMA and BA-MAN co-
2
2
i
III
The molecular weights in BA/AMS copolymerizations
are dramatically influenced by the level of the R-meth-
ylvinyl monomer (see Figure 7). For high catalyst
concentrations, it appears that the molecular weight is
not controlled by the catalyst level but rather by the
polymerization with PrCo (DMG-BF2)2 at >80 °C,
conversions are largely independent of the cobalt con-
centration.
Molecular weight reductions obtained with all three
complexes were similar. This suggests that the im-

9048 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
Scheme 7
affect the copolymer composition and can be safely
neglected.
On reaction with T, one chain is terminated (and an
AT dyad is formed) and a new chain is started (to
produce a TA dyad). The fraction of transfer agent-
derived chain ends with a penultimate unit of monomer
A (FA) can be expressed as follows (eq 2):
P(AT)
P(A)P(A|T)
FA
)
)
(2)
P(AT) + P(ST) P(A)P(A|T) + P(S)P(S|T)
where P(A) is the probability of finding monomer A in
the chain and P(A|T) is the probability that a chain
ending in A will react with transfer agent. P(S) and
P(S|T) are defined analogously.
P(A|T) can be expressed as the rate of formation of
end groups (AT) divided by the sum of the rates of all
•
reactions of the propagating species A (eq 3).
•
k [A ][T]
at
P(A|T) )
(3)
•
•
•
k [A ][S] + k [A ][A] + k [A ][T]
at
aa
at
proved specificity referred to above is due to a lower
transfer constant to the monosubstituted monomer with
III
II
•
58
MeCo (DEG-BF2)2 or Co (DPG-BF2)2 rather than a
higher transfer constant to AMS. However, further work
is required to more rigorously establish values for the
transfer constants. We have previously observed that
Dividing through by kaa[A ] gives eq 4.
C [T]
a
P(A|T) ) [S]/r
(4)
i
III
+ [A] + C [T]
for MMA polymerization in solution PrCo (DMG-
as
a
III
BF2)2 and MeCo (DEG-BF2)2 have similar transfer
II
9
constants while that for Co (DPG-BF2)2 is lower.
Equations 5 and 6 are derived in the same manner.
Other work also suggests that the transfer constant in
II
MMA polymerization with Co (DPG-BF2)2 is lower
[
A]
II
than that with Co (DMG-BF2)2.
P(A|A) ) [S]/r
P(A|S) ) [
(5)
(6)
+ [A] + C [T]
as
a
Kinetic Model for Binary Copolymerization in
the Presence of a Transfer Agent. In earlier work,
we have shown that kinetic simulation using Monte
[S]/ras
S]/ras + [A] + C [T]
58
59
a
Carlo methods or numerical integration can be used
to predict the molecular weight, composition, and end
group functionality of multicomponent copolymers formed
by free radical polymerization in the presence of chain
transfer agents. These methods also enabled calculation
of the end group composition and molecular weight
distribution based on knowledge of the reactivity ratios,
and transfer constants and relative initiation rates.
Given a long kinetic chain length, no absolute rate data
where rx are the reactivity ratios (rx ) kxx/kxy) and Cx
are the transfer constants (Cx ) kxt/kxx).
Similar expressions for P(S|T), P(S|A) and P(S|S) can
be derived.
•
Note that the transfer agent-derived species (T ) does
not react the with transfer agent (or does so in a
degenerate process). Thus, P(T|T) ) 0 and
(
e.g. propagation rate constants or radical concentra-
tions) are required to evaluate these parameters.
•
k [T ][A]
ta
Even though analytical expressions have been pro-
posed previously to predict the end group composition
of binary copolymers formed in the presence of transfer
P(T|A) )
(7)
(8)
•
•
k [T ][S] + k [T ][A]
ts
ta
[A]
3
8,40,60
agents,
when applied to the systems described
)
herein, the predictions appear inconsistent with the
results of kinetic simulation
[S]/R + [A]
t
58,59
and, more importantly,
with the experimental findings reported herein. The
equations substantially overestimate the fraction of end
groups formed from the monomer with the lower
propagation rate constant. The problem with the previ-
Similarly,
[
S]
P(T|S) )
(9)
3
8,40,60
[A]/R + [S]
ous treatment
lies with the use of a simplified
t
steady-state assumption and a long chain assumption
both of which are inappropriate in the present context.
where Rt reflects the specificity of the transfer agent-
•
The reactions that comprise terminal model copolym-
erization of two monomers (designated A and S) in the
presence of transfer agent (T) are summarized sche-
matically in (Scheme 7). If the kinetic chain length is
long, initiation (by initiator-derived radicals) and ter-
mination reactions (by radical-radical reaction) do not
derived radical (T ) for the two monomers (Rt ) kta/kts).
•
Since T does not react with T an expression for P(A):
P(S):P(T) can be derived that is analogous to that
derived by Alfrey and Goldfinger61 to describe ternary
copolymerization where one monomer (in our case, the
transfer agent, T) cannot homopolymerize. Thus

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9049
Table 14. Parameters Used to Model Copolymerizations
P(A) ∝ pa )
i
III
in Presence of Pr-Co (DMG-BF2)2
R [A] [S] R [T]
[S]
ra
[T]
t
t
[
A]
+
+
[A] +
[S] +
+
+
(10)
(11)
temp monomer A monomer S
ra
rs
Ca
Cs
R
{ r
}{
1/Ca}
s
rs
1/Cs
6
8
0
0
MMA
MMA
MMA
AMS
AMS
AMS
AMS
BA
BA
BA
BA
BA
S
2.24 0.414
2.24 0.414
2.24 0.414
0.5 0.13
0.5 0.18
0.24 1.06 500 000 2000
0.54 1.16 300 000 1500
1000
1200
1700
80 000 150
80 000 90
45
65
90
1
1
1
1
1
1
1
P(S) ∝ ps )
1
25
R [A] [S]
[T]
1/Ca
[A]
rs
[T]
1/Cs
t
80
125
[
S]
+
+
{ r
}{
}
a
ra
8
0
1
25
S
[
A]
[S]
[T]
P(T) ∝ pt ) [T]
+
+
{r [A] + [S]}
t
{
}
rs/Ca ra/Cs 1/C C
a
s
In principle, one might use the kinetic model to
(12)
estimate rate constants from the experimental data
using a nonlinear least-squares fitting technique. Re-
activity ratios for many systems are well established
or can be estimated (Table 11). Transfer constants for
cobalt complexes in AMS, styrene, and MMA polymer-
izations have been reported (Table 10). The main
unknowns are the reinitiation specificity, the transfer
constants to other monomers, and the equilibrium
constant for association of the propagating species with
the cobalt complex. In practice, this approach has not
been successful when applied to obtain all unknown
parameters simultaneously. Consequently, a trial and
error approach has been used in estimating and estab-
lishing sensitivity to the various kinetic parameters.
Reactivity Ratios and Transfer Constants. For
the monomer ratios used in our experiments (low AMS),
it is not appropriate to evaluate reactivity ratios by
analysis of the composition data directly. For example,
in BA-AMS copolymerization with low AMS, the overall
copolymer composition is very sensitive to rBA but
relatively insensitive to rAMS. The end group composition
and the molecular weight are, however, sensitive to
value of rAMS.
and
pa
P(A) )
(13)
pa + ps + pt
Therefore, given values for the reactivity ratios and
transfer constants, it is possible to apply eq 2 to
calculate the fraction of chain transfer agent-derived
end groups.
The degree of polymerization is approximately equal
to the reciprocal of P(T).
1
Xh ≈
(14)
n
P(T)
Since a one unit chain would be monomer, the applica-
tion of eq 2 or 14 to polymerization in systems with
catalytic chain requires the contribution of 1 unit chain
to be deducted. Since
F ) F (1)P(1) + F (>1)P(>1)
(15)
A
A
A
where FA(n) is the fraction of n unit chains terminated
with monomer A and P(n) is the probability of forming
a chain of length n.
Reactivity ratios for copolymerizations of BA and
styrene with various R-methylvinyl monomers are sum-
marized in Table 11. The literature indicates that the
62
Noting that P(>1) ) 1 - P(1), then
composition (but not the kinetics) of MMA-BA and
MMA-styrene6
3-66
copolymers formed by radical po-
F - F (1)P(1)
lymerization are adequately described by terminal
model reactivity ratios. Reported reactivity ratios for
A
A
F (>1) )
(16)
A
1
- P(1)
AMS-styrene67 and AMS-BA
68,69
show significant
temperature dependence which is attributed to the
reversibility of propagation. Reactivity ratios for copo-
lymerizations of BA with MAN have not been reported.
where
F (1) ) P(T|A) P(A|T)/P(1)
P(1) ) P(T|A) P(A|T) + P(T|S) P(S|T)
(17)
(18)
A
70
Estimates based on the use of the Q-e scheme and
the patterns of reactivity scheme71 are shown in Table
1
1.
Similarly, since the degree of polymerization, Xh n ) Xh n
>1)P(>1) + P(1)
For AMS-styrene copolymerization, reactivity ratios
(
based on the Arrhenius parameters reported by
Fischer67 have been used in our modeling (Table 14).
These values allow prediction of the observed copolymer
composition and values of the transfer constants can be
chosen which describe of the end group composition and
the molecular weight.
Xh - P(1)
n
Xh (> 1) )
(19)
n
1
- P(1)
The expressions simplify considerably if the chains
formed are long since in that case P(1)f0 and the above-
mentioned corrections are unnecessary and the expres-
sions for P(A) and P(S) can be approximated by those
for binary copolymerization in the absence of a chain
transfer agent.
When our analysis was initiated, the only reported
reactivity ratio data for the AMS-BA system was that
of McManus et al.68 Their work indicated a very low
value for rAMS particularly at high temperatures (Table
11). If rAMS is low, then higher values of the transfer
constants can be chosen to predict the observed chain
end composition and molecular weight. However, these
reactivity ratios were apparent reactivity ratios based
on the use of a terminal model which did not allow for
depropagation and are therefore perhaps not appropri-
ate for use in the present circumstances. In our experi-
ments, the extent of depropagation is reduced because
Analytical expressions might also be derived to en-
compass more complex situations such as the effects of
penultimate units on reactivity ratios and various side
reactions as discussed below. However, the complexity
of the resultant equations would severely limit their
utility and the use of numerical simulation4
3,59
is
recommended in this case.

9050 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
the propagating radical with a terminal AMS unit can
be trapped by chain transfer. Application of the empiri-
Scheme 8
7
0,72
71
cal Q-e
or patterns schemes suggested a rBA
6
8
similar to that reported but a markedly higher value
for rAMS (Table 11). Notwithstanding the abovemen-
tioned limitations of our data with respect to reactivity
ratio estimation, fitting our experimental data for 80
°
C also suggested higher values of rAMS closer to those
predicted by the Q-e scheme. Very recently the Penlidis
6
9
group has reported reactivity ratios and their tem-
perature dependence for AMS-BA copolymerization
7
3
that were estimated using the Kruger model (Table
has no marked effect on the composition, it may affect
the rate of copolymerization.
7
3
1
1). The Kruger model takes into account all possible
depropagation reactions. These values are entirely
consistent with our experimental observations and have
been used in the present work.
Side Reactions. Depropagation. The kinetic model
described above does not allow for depropagation, and
thus some caution should be exercised in their direct
application to copolymerizations of AMS, particularly
at high temperatures. AMS has a very low ceiling
temperature and depropagation should not be ignored.
Reported propagation-depropagation equilibrium con-
stants (K) for AMS in copolymerizations with styrene
and BA are shown in Table 11.80 Predictions of end
group composition may still be valid if only homodepro-
pagation occurs since, in that case, loss of an AMS end
will generate another AMS end. Furthermore, the
probability of forming an AMS-AMS dyad under the
conditions used (low AMS concentration) studied is very
low. However, if cross depropagation occurs these
expressions will tend to overestimate end group purity.
Fischer67 has indicated that cross depropagation in
styrene-AMS copolymerization may become significant
at temperatures >110 °C. No data are available for BA-
AMS copolymerization, though it is anticipated on the
basis of relative bond strengths that the cross depro-
pagation should be less favorable in this system than
for styrene-AMS copolymerization.
II
Literature transfer constants for Co (DMG-BF2)2
observed in polymerizations of various (meth)acrylate
esters, styrene and AMS are collected in Table 10. It
has been suggested that chain transfer to the meth-
acrylate esters may be a diffusion-controlled process.74
In the case of styrene and monosubstituted monomers,
chain transfer constants are lower, not only because the
reaction of the propagating species with cobalt is slower
but also because the complex reacts reversibly with
propagating radicals by the process of Scheme 5 (where
•
R is a propagating radical) to form the corresponding
alkylcobalt(III) complex, thus rendering some fraction
2
8,53,54,75
of the cobalt complex in a dormant state.
Some
data also suggest that the transfer constant to MMA
may be chain length dependent for chain lengths <7
7
6
units.
While there have been a few studies of hydrogen atom
transfer to monomers from cobalt hydrides,77,78 there
are as yet no reliable data for relative rates of initiation
(
Rt) by this species for the present systems. On the basis
Transfer to AMS. Reported rate constants for trans-
fer to monomer are more than 2 orders of magnitude
of an analysis of the dimeric products formed in styrene-
II
MMA copolymerization in the presence of Co (DMG)2,
-3
-1 -1
higher for AMS (ca. 4 × 10
M
s ) than for other
3
4
Gruel and Harwood suggested that there is a prefer-
ence for hydrogen atom transfer to styrene over MMA.
Analysis of the dimeric products formed in copolymer-
-5
-1 -1
81
monomers (ca. 10
M
s ) used in this study, and
it has recently been reported that copolymerization of
AMS with, for example, styrene may in itself provide a
mechanism of molecular weight control and a method
79
izations of AMS and MMA (or ethyl methacrylate) and
of AMS with MAN in the presence of Bz(py)Co (DMG)2
III
82
for making functional polymers. However, the con-
suggests that MMA chain ends are formed in preference
to AMS or MAN. Some caution is required in interpret-
ing these results. In both sets of experiments very high
cobalt concentrations were used (0.0005-0.001 M) such
that the observed product was mainly dimer. It is likely
that the major (nonobserved) product was monomer,
since, under these conditions, a large fraction of the
initially formed monomeric radicals are also likely to
be trapped by the transfer agent. The very much higher
reactivity of propagating species with a terminal MMA
vs styrene or AMS vs MMA means that a large fraction
of those species are trapped. Thus, the dimer distribu-
tion will not reflect the relative rate of reaction of the
cobalt hydride with the monomers.
centration of AMS required to give low molecular weight
polymers is very high. Given the relatively low concen-
trations of AMS used in the present experiments
(
always <10 wt %), transfer to monomer should be of
little significance.
Backbiting)Fragmentation. Another side reaction
that should be considered in polymerization of mono-
substituted monomers is the formation of a macromono-
29,43,83,84
merchainendsbybackbiting/â-scission(Scheme8).
It is believed that this side reaction is the main factor
limiting molecular weight in high temperature polym-
erization or copolymerization of acrylate esters. The
reaction may also occur during styrene polymerization
29,83
but it is substantially less significant.
Backbiting/
From our modeling, it is clear that the end group
composition and molecular weight of the product (after
correction for the formation of 1 unit chains) is insensi-
tive to the value of Rt. The reason is that, under our
conditions, irrespective of the value of Rt, essentially
all initiation is by way of monomer S. Initiating species
formed by transfer of hydrogen to monomer A do not
propagate but are immediately trapped as monomer.
The value of Rt has therefore been arbitrarily assigned
as 1.0. We anticipate that even though the value of Rt
â-scission produces a polymer with macromonomer
chain end (similar to 2) but derived from the monosub-
stituted monomer as shown in Scheme 8.
For cases where the substituents Y (on the R-meth-
ylvinyl monomer) and X (on the monosubstituted mono-
mer) are different, the formation of macromonomer
chain ends by backbiting â-scission will be evident in
1
1
the H NMR spectrum. Thus, the H NMR spectra of
high conversion copolymers of BA and AMS produced
in the presence of a chain transfer catalyst shows

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9051
Scheme 9
evidence of both -CH2-C(Ph)dCH2 and -CH2-C(CO2-
Bu)dCH2 chain ends. For cases where Y and X are the
same, the fraction of chains formed by backbiting
â-scission should be independent of the concentration
the cobaloxime chain transfer catalyst. Thus, an upper
limit for the fraction of chain ends formed by this
mechanism is the ratio of molecular weights seen in the
absence and presence of the chain transfer catalyst. For
the experiments reported in this paper where conver-
sions are low, the conditions are such that the formation
of chain ends by backbiting-fragmentation is negligible.
Macromonomer Copolymerization. The initially
formed macromonomers are likely to be reactive under
the polymerization conditions, particularly at high
monomer conversions. The radical formed by addition
to the macromonomer may undergo â-scission (to return
to starting materials or to give chain transfer) or it may
propagate leading to branch formation (Scheme 9). This
latter pathway may be desirable in some circumstances
since it provides a route to hyperbranched polymers
tion at 60 °C when polymerized with the respective
monomers.
The polymerizations reported herein were carried out
in solution, were generally carried out at high reaction
temperatures (120 °C) and were only taken to low
conversion such that the molar concentration of mac-
romonomer is very low. The measured transfer con-
stants of styrene macromonomers with terminal AMS
chain ends in polymerizations of styrene are 0.11-0.15
at 120 °C as compared to (∼0) at 80 °C (Table 12). These
results are consistent with cross propagation being
favored over fragmentation at the lower temperature
(80 °C) and with fragmentation being a preferred
pathway at the higher temperature (120 °C) such that
graft copolymer formation becomes less important.
Chain transfer to formed macromonomer by addition
fragmentation (RAFT) does not give rise to byproduct
formation. It also provides a mechanism for chain
equilibration process as shown in Scheme 9 and may
lead to lower polydispersities.9
Predictions of Copolymer Molecular Weight,
End Groups, and Composition. Predictions of mo-
lecular weight and end group composition using our
kinetic model for AMS-styrene copolymerization are
shown in Figure 3 (125 °C) and Figure 4 (80 °C). The
kinetic parameters used are given in Table 14. It can
be seen that we are able to predict all of the generic
trends observed experimentally. It would seem that the
lower macromonomer purity observed for the lower
reaction temperatures is largely attributable to less
favorable reactivity ratios.
85
based on acrylate monomers. Macromonomer products
with end group 2, rather than being shown as (9a-12a
or 9c-12c), should be represented by structure 18. The
nonmacromonomer products with end group 5 (9b-12b
or 9d-12d) can be similarly described. The fraction of
the repeat unit CH2-C(CH3)Z in 18 (i.e. the extent of
branching) should be reduced by increasing tempera-
ture, by increasing dilution and is favored by low
8
,9
monomer conversions.
Predictions of molecular weight and end group com-
position and overall copolymer composition for BA
copolymerizations are shown in Figure 8 (BA-AMS 80
°
C), Figure 9 (BA-AMS 125 °C), Figure 10 (BA-MMA
6
0 °C) and Figure 11 (BA-MMA 125 °C), respectively.
Again, all of the experimentally observed trends are
predicted by the kinetic model.
The estimated transfer constant of cobalt complex to
the BA propagating species is small (∼45-90 in MMA
copolymerization ∼150-200 in AMS copolymerization).
The apparent transfer constant to the MMA propagating
species in these copolymerizations is in the range 1000-
1700. This is up to 30-fold lower than observed in MMA
homopolymerization (see Table 10). The apparent trans-
fer constant to the MMA propagating species is also
much lower than previously reported. A probable ex-
Graft formation will also complicate product analysis
of high conversion polymerizations. A recent study on
II
bulk BA-MMA copolymerization with Co (DMG-BF2)2
at 60 °C indicates that graft formation by copolymeri-
zation of macromonomer is significant at high conver-
4
3
86
sions. Yamada et al. have reported that MA and
styrene macromonomers give exclusively copolymeriza-

9052 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
Figure 10. Comparison of observed and predicted properties
of BA-MMA formed in copolymerization with [BA]:[MMA]
Figure 8. Comparison of observed and predicted properties
of BA-AMS formed in copolymerization with [BA]:[AMS]
i
III
∼
2 2
10:1 in the presence of Pr-Co (DMG-BF ) at 60 °C: (a)
∼
50:1 (closed symbols) or ∼25:1 (open symbols) in the presence
i III
2 2
fraction of MMA-derived chain ends [2]/([2] + [5]) (experimen-
tal, b; predicted, - -), (b) predicted fraction of AMS in copolymer
of Pr-Co (DMG-BF
chain ends [2]/([2] + [5]) (experimental, b; predicted, - -),
b) predicted fraction of AMS in copolymer (experimental, 9;
predicted, - -), and (c) number-average molecular weight (Mh
) at 80 °C: (a) fraction of AMS-derived
(
experimental, 9; predicted, - -), and (c) number-average
(
n n
molecular weight ( Mh experimental, 2; Mh predicted, -).
n
Experimental conditions are summarized in Table 7. Param-
eters used in the fit are given in Table 14.
n
experimental, 2; Mh predicted, -). Experimental conditions
are summarized in Table 6. Parameters used in the fit are
given in Table 14.
Figure 11. Comparison of observed and predicted properties
of BA-MMA formed in copolymerization with [BA]:[MMA]
Figure 9. Comparison of observed and predicted properties
of BA-AMS formed in copolymerization with [BA]:[AMS]
i
III
∼
2 2
10:1 in the presence of Pr-Co (DMG-BF ) at 125 °C: (a)
fraction of MMA-derived chain ends [2]/([2] + [5]) (experimen-
tal, b; predicted, - -), (b) predicted fraction of AMS in copolymer
∼
50:1 (closed symbols) or ∼25:1 (open symbols) in the presence
i III
2 2
of Pr-Co (DMG-BF
chain ends [2]/([2] + [5]) (experimental, b; predicted, - -),
b) predicted fraction of AMS in copolymer (experimental, 9;
predicted, - -), and (c) number-average molecular weight (Mh
) at 125 °C: (a) fraction of AMS-derived
(
experimental 9, predicted - -), and (c) number-average
n n
molecular weight ( Mh experimental, 2; Mh predicted, -).
(
Experimental conditions are summarized in Table 7. Param-
eters used in the fit are given in Table 14.
n
n
experimental, 2; Mh predicted, -). Experimental conditions
are summarized in Table 6. Parameters used in fit are given
in Table 14.
conditions of our experiments which provide relatively
low molecular weight polymers. The onset of the “satu-
ration effect” occurs when the degree of polymerization
approaches the number-average sequence length for the
monosubstituted monomer (S) (eq 20). The dependence
of the fraction macromonomer chain ends [2]:[5] on
cobalt concentration can also be explained in these
terms.
planation is that much of the cobalt complex in the
predominantly BA reaction medium is in an inactive
form as the alkylcobalt(III) complex formed by reversible
coupling between the BA propagating species and the
active cobalt(II) complex (see above).
The predicted dependence of the molecular weight on
cobalt concentration is similar to that observed experi-
mentally. The “saturation effect” referred to above can
be related to two factors:
number-average sequence length ∼
1
(
20)
(
a) The first is the fact that the transfer constants Cs
P(S|A) + P(S|T)
involving the propagating species 3 (with a terminal
unit derived from the monosubstituted monomer) are
much lower than Ca involving the propagating species
Transfer Constants of Macromonomers. The ap-
parent transfer constant for a styrene macromonomer
1.
(9) prepared by AMS-styrene copolymerization was
(
b) The second is due to the monomer concentrations
evaluated to both to provide another estimate of the
macromonomer purity and to gauge the importance of
macromonomer copolymerization as a side reaction.
The measured transfer constants for the styrene
macromonomer (9) can be compared with those for a
macromonomer with a similar end group prepared with
and reactivity ratios which dictate the likelihood of there
being a R-methyl vinyl monomer at the chain end.
The failure of the traditional “Mayo analysis” when
applied to BA copolymerizations occurs in part because
the long chain approximation is not valid under the

Macromolecules, Vol. 38, No. 22, 2005
Copolymerization with Catalytic Chain Transfer 9053
an allyl sulfide transfer agent (16) (refer to Scheme 4),50
a macromonomer with a methacrylate chain end (17)
and those for AMS dimer 6 in Table 13. The transfer
constants were determined using the log CLD method
of the two monomers ([S]:[Α]) in the feed. This has
important practical consequences, particularly if the
primary aim is molecular weight control and end group
purity is less relevant. It means that the outcome of
polymerization is less affected by such factors as deg-
radation of the catalyst with reaction time. It also means
less care is required in selecting the amount/quality of
catalyst. This is an important consideration given the
air sensitivity of certain catalysts and the very small
quantities involved.
4
,87
79,88,89
as described in earlier work.
Literature values
for AMS dimer 6 are also shown in Table 13. The
transfer constants of the styrene macromonomers 9 and
1
6 are the same within experimental error after allow-
ing for the macromonomer purity (consistent with the
∼
68% of chains with end groups 2 determined by NMR
analysis).
It is now possible to summarize the features impor-
tant in the R-methylvinyl monomer that provide for high
macromonomer purity as follows:
The apparent transfer constant for the styrene mac-
romonomers in styrene polymerization increases sig-
nificantly with increasing temperature; from ∼0 at 80
(a) A low reactivity ratio r should exist, such that
s
°C to 0.11-0.16 at 120 °Cssee Table 13. For the
the propagating radical 3 prefers to undergo cross
polymerizations of styrene in the presence of the styrene
macromonomer (16 or 17) at 80 °C there was no
significant variation in molecular weight of the poly-
styrene formed with macromonomer concentration.
However, some narrowing of the molecular weight
distribution may be a consequence of the extent of chain
branching incorporated by macromonomer copolymer-
ization.
propagation (add to the R-methyl vinyl monomer).
(b) The transfer constant Ca should be high and
preferably much greater than the transfer constant Cs
associated with the monosubstituted monomer.
(c) The reactivity ratio ra is less crucial. It may be
considered desirable that the radical derived from
monomer A does not undergo propagation since this
maximizes the likelihood that the macromonomer con-
tains the R-methylvinyl monomer only at the chain end.
When the R-methylvinyl monomer contains appropri-
ate functionality the chemistry described provides a
route to end-functional macromonomers. Thus, hydroxy
end functional polymers have been prepared using
hydroxyethyl methacrylate as comonomer and isocy-
anato-functional polymers have been prepared using
R,R-dimethyl-m-isopropenylbenzyl isocyanate as comono-
mer.2
In contrast, we find the transfer constant for AMS
dimer (6) shows no marked temperature dependence in
styrene polymerization over the range 60-120 °C. It has
a significantly lower transfer constant in BMA polym-
erization than in styrene polymerization at 120 °C.
The different temperature dependence seen for AMS
dimer (6) and the styrene macromonomers is related to
the propensity of the styrene macromonomer (9) and
9,30
(16) to undergo copolymerization with styrene at lower
temperatures (refer to Scheme 9 for the mechanism).
Chain transfer by addition-fragmentation predomi-
nates at 120 °C. AMS dimer (6) does not undergo
copolymerization with styrene in the temperature range
Note Added after ASAP Publication. This article
was released ASAP on October 4, 2005. References 4,
5
7, 80, 83, and 94 have been revised. The correct version
was posted on October 12, 2005.
6
0-120 °C, with fragmentation being favored over
propagation over the entire temperature range. The
difference in behavior is attributed to the cumyl radical
being a substantially better free radical leaving group
than the polystyryl propagating radical.
Acknowledgment. We are grateful to Drs. C. T.
Berge, M. Fryd, A. A. Gridnev, S. D. Itell, and R. A.
Matheson of DuPont Performance Coatings for their
support of this work and for a valuable discussion.
The transfer constants for the macromonomers (9, 16,
or 17) at 120 °C in styrene polymerization are similar
to those observed in BMA polymerization (Table 12) and
only slightly lower than those reported for methacrylate-
based macromonomers in polymerizations of methacry-
References and Notes
(1) Cacioli, P.; Hawthorne, D. G.; Laslett, R. L.; Rizzardo, E.;
Solomon, D. H. J. Macromol. Sci., Chem. 1986, A23, 839-
8
52.
4,5
late esters. The result suggests a possible application
of these macromonomers in block copolymers synthesis
at higher temperatures (>120 °C).
(
2) Rizzardo, E.; Meijs, G. F.; Thang, S. H. Macromol. Symp.
1
995, 98, 101-123.
(3) Gridnev, A. A.; Ittel, S. D. Chem. Rev. 2001, 101, 3611-3659.
(4) Moad, C. L.; Moad, G.; Rizzardo, E.; Thang, S. H. Macromol-
ecules 1996, 29, 7717-7726.
The results also indicate that macromonomer copo-
lymerization should occur and may be a dominant
reaction during synthesis of macromonomers (9) in
polymerizations carried to significant monomer conver-
sion with a reaction temperature at or below 100 °C. A
similar situation is likely to apply in the case of acrylate
based macromonomers.
(
5) Hutson, L.; Krstina, J.; Moad, C. L.; Moad, G.; Morrow, G.
R.; Postma, A.; Rizzardo, E.; Thang, S. H. Macromolecules
2004, 37, 4441-4452.
(6) Miyake, K.; Zetterlund, P. B.; Yamada, B. Macromol. Rapid
Commun. 2004, 25, 1905-1911.
(
(
(
7) Sato, E.; Zetterlund, P. B.; Yamada, B. Macromolecules 2004,
37, 2363-2370.
8) Krstina, J.; Moad, G.; Rizzardo, E.; Winzor, C. L.; Berge, C.
Conclusions
T.; Fryd, M. Macromolecules 1995, 28, 5381-5385.
9) Krstina, J.; Moad, C. L.; Moad, G.; Rizzardo, E.; Berge, C.
T.; Fryd, M. Macromol. Symp. 1996, 111, 13-23.
Copolymerization of monosubstituted monomer in the
presence of R-methylvinyl monomers and a cobalt chain
transfer catalyst provides a useful route to macromono-
mers that are composed largely of the monosubstituted
monomer.
The molecular weight is dramatically influenced by
the level of the R-methylvinyl monomer. For higher
catalyst concentrations, the molecular weight is not
controlled by the catalyst level but rather by the ratio
(
10) Huybrechts, J.; Bruylants, P.; Kirshenbaum, K.; Vrana, J.;
Snuparek, J. Prog. Org. Coatings 2002, 45, 173-183.
(11) Rizzardo, E.; Harrison, D.; Laslett, R. L.; Meijs, G. F.; Morton,
T. C.; Thang, S. H. Prog. Pacific Polym. Sci. 1991, 2, 77-88.
(
12) Abbey, K. J.; Trumbo, D. L.; Carlson, G. M.; Masola, M. J.;
Zander, R. A. J. Polym. Sci., Part A: Polym. Chem. 1993,
31, 3417-3424.
(
13) Trumbo, D. L.; Abbey, K. J. J. Polym. Sci., Part C: Polym.
Lett. 1987, 25, 229-232.

9054 Chiefari et al.
Macromolecules, Vol. 38, No. 22, 2005
(
(
(
(
(
(
(
(
(
14) Rajatapiti, P.; Dimonie, V. L.; El-Aasser, M. J. Appl. Polym.
Sci. 1996, 61, 891-900.
(54) Pierik, S. C. J.; Vollmerhaus, R.; van Herk, A. M.; German,
A. L. Macromol. Symp. 2002, 182, 43-52.
15) Muratore, L. M.; Davis, T. P. J. Polym. Sci., Part A: Polym.
Chem. 2000, 38, 810-817.
(55) Suddaby, K. G.; Maloney, D. R.; Haddleton, D. M. Macro-
molecules 1997, 30, 702-713.
16) Muratore, L. M.; Steinhoff, K.; Davis, T. P. J. Mater. Chem.
(56) Alfrey, T., Jr.; Hardy, V. J. Polym. Sci. 1948, 3, 500-502.
(57) Smith, W. V. J. Am. Chem. Soc. 1946, 68, 2069-2071.
(58) Galbraith, M. N.; Moad, G.; Solomon, D. H.; Spurling, T. H.
Macromolecules 1987, 20, 675-679.
(59) Spurling, T. H.; Deady, M.; Krstina, J.; Moad, G. Makromol.
Chem., Macromol. Symp. 1991, 51, 127-146.
(60) Heuts, J. P. A.; Coote, M.; Davis, T. P.; Johnston, L. P. M.
ACS Symp. Ser. 1998, 685, 120-144.
1999, 9, 1687-1691.
17) Norman, J.; Moratti, S. C.; Slark, A. T.; Irvine, D. J.; Jackson,
A. T. Macromolecules 2002, 35, 8954-8961.
18) Bon, S. A. F.; Morsley, S. R.; Waterson, C.; Haddleton, D. M.
Macromolecules 2000, 33, 5819-5824.
19) Haddleton, D. M.; Topping, C.; Hastings, J. J. Macromol.
Chem. Phys. 1996, 197, 3027-3042.
20) Haddleton, D. M.; Topping, C.; Kukulj, D.; Irvine, D. Polymer
(61) Alfrey, T.; Goldfinger, G. J. Chem. Phys. 1946, 14, 115-116.
(62) Hutchinson, R. A.; McMinn, J. H.; Paquet, D. A.; Beuermann,
S.; Jackson, C. Ind. Eng. Chem. Res. 1997, 36, 1103-1113.
(63) Coote, M. L.; Johnston, L. P. M.; Davis, T. P. Macromolecules
1997, 30, 8191-8204.
1998, 39, 3119-3128.
21) Monteiro, M. J.; Bussels, R.; Wilkinson, T. S. J. Polym. Sci.,
Part A: Polym. Chem. 2001, 39, 2813-2820.
22) Enikolopyan, N. S.; Smirnov, B. R.; Ponomarev, G. V.;
Belgovskii, I. M. J. Polym. Sci., Polym. Chem. Ed. 1981, 19,
(64) Coote, M. L.; Zammit, M. D.; Davis, T. P.; Willet, G. D.
Macromolecules 1997, 30, 8182-8190.
879-889.
(
(
(
(
(
23) Burczyk, A. F.; O’Driscoll, K. F.; Rempel, G. L. J. Polym. Sci.,
(65) Maxwell, I. A.; Aerdts, A. M.; German, A. L. Macromolecules
1993, 26, 1956-1964.
Polym. Chem. Ed. 1984, 22, 3255-3262.
24) Davis, T. P.; Kukulj, D.; Haddleton, D. M.; Maloney, D. R.
Trends Polym. Sci. 1995, 3, 365-373.
25) Gridnev, A. J. Polym. Sci., Part A: Polym. Chem. 2000, 38,
(66) Fukuda, T.; Ma, Y.-D.; Inagaki, H. Macromolecules 1985, 18,
17-26.
(67) Fischer, J. P. Makromol. Chem. 1972, 155, 211-225.
(68) McManus, N. T.; Lona, L. M. F.; Penlidis, A. Polym. React.
Eng. 2002, 10, 285-309.
1753-1766.
26) Heuts, J. P. A.; Roberts, G. E.; Biasutti, J. D. Aust. J. Chem.
2002, 55, 381-398.
(69) Wang, T. J.; Leamen, M. J.; McManus, N. T.; Penlidis, A. J.
Macromol. Sci., Chem. 2004, A41, 1205-1220.
(70) Greenley, R. Z. In Polymer Handbook, 4th ed.; Immergut, E.
H., Ed.; Wiley: New York, 1999; pp II/309-319.
(71) Jenkins, A. D.; Jenkins, J. In Polymer Handbook, 4th ed.;
Grulke, E. A., Ed.; John Wiley and Sons: New York, 1999;
pp II/321-328.
27) Smirnov, B. R.; Plotnikov, V. D.; Ozerkovskii, B. V.; Rosh-
chupkin, V. P.; Enikolopyan, N. S. Polym. Sci. USSR 1981,
23, 2807-2816.
(
(
(
28) Heuts, J. P. A.; Forster, D. J.; Davis, T. P.; Yamada, B.;
Yamazoe, H.; Azukizawa, M. Macromolecules 1999, 32,
2511-2519.
29) Chiefari, J.; Jeffery, J.; Mayadunne, R. T. A.; Moad, G.;
(72) Laurier, G. C.; O’Driscoll, K. F.; Reilly, P. M. J. Polym. Sci.,
Polym. Symp. 1985, 72, 17-26.
Rizzardo, E.; Thang, S. H. ACS Symp. Ser. 2000, 768, 297-
312.
(73) Kruger, H.; Bauer, J.; Rubner, J. Makromol. Chem. 1987, 188,
2163-2175.
30) Moad, G.; Rizzardo, E.; Moad, C. L.; Ittel, S. D.; Wilczek, L.;
Gridnev, A. A. Int. Patent Appl. WO 9731030, 1997.
31) Lin, J.-C.; Abbey, K. J. US4680354, 1987.
(74) Heuts, J. P. A.; Forster, D. J.; Davis, T. P. Macromolecules
1999, 32, 3907-3912.
(
(
(
(
32) Hawthorne, D. G. US5324879, 1994.
(75) Roberts, G. E.; Heuts, J. P. A.; Davis, T. P. Macromolecules
2000, 33, 7765-7768.
33) Janowicz, A. H.; Melby, L. R. US4680352, 1987.
34) Greuel, M. F.; Harwood, H. J. Polym. Prepr. (Am. Chem. Soc.,
Div. Polym. Chem) 1991, 32, 545-546.
(76) Sanayei, R. A.; O’Driscoll, K. F. J. Macromol. Sci. Chem.
1989, A26, 1137-1149.
(
35) Chiefari, J.; Rizzardo, E. In Handbook of Radical Polymer-
ization; Matyjaszewski, K., Ed.; John Wiley & Sons: Hobo-
ken, NJ, 2002; pp 263-300.
(77) Gridnev, A. A.; Ittel, S. D.; Wayland, B. B.; Fryd, M.
Macromolecules 1996, 15, 5116-5126.
(78) Suddaby, K. G.; Hunt, K. H.; Haddleton, D. M. Macromol-
ecules 1996, 29, 8642-8649.
(36) Moad, G.; Chiefari, J.; Moad, C. L.; Postma, A.; Mayadunne,
R. T. A.; Rizzardo, E.; Thang, S. H. Macromol. Symp. 2002,
(79) Yamada, B.; Tagashira, S.; Aoki, S. J. Polym. Sci., Part A:
Polym. Chem. 1994, 32, 2745-2754.
182, 65-80.
(
(
(
37) Gridnev, A. A.; Simonsick, W. J.; Ittel, S. D. J. Polym. Sci.,
Part A: Polym. Chem. 2000, 38, 1911-1918.
(80) McCormick, H. W. J. Polym. Sci. 1957, 25, 488-490.
(81) Kukulj, D.; Davis, T. P.; Gilbert, R. G. Macromolecules 1998,
31, 994-999.
38) Kukulj, D.; Heuts, J. P. A.; Davis, T. P. Macromolecules 1998,
31, 6034-6041.
(82) Barner-Kowollik, C.; Davis, T. P. Macromol. Theory Simul.
2001, 10, 255-261.
39) Chiu, T. Y. J.; Heuts, J. P. A.; Davis, T. P.; Stenzel, M. H.;
Barner-Kowollik, C. Macromol. Chem. Phys. 2004, 205, 752-
(83) Chiefari, J.; Jeffery, J.; Moad, G.; Mayadunne, R. T. A.;
Rizzardo, E.; Thang, S. H. Macromolecules 1999, 32, 7700-
7702.
761.
(
(
(
(
(
(
(
(
40) Heuts, J. P. A.; Kukulj, D.; Forster, D. J.; Davis, T. P.
Macromolecules 1998, 31, 2894-2905.
(
84) Peck, A. N. F.; Hutchinson, R. A. Macromolecules 2004, 37,
41) Heuts, J. P. A.; Muratore, L. M.; Davis, T. P. Macromol.
Chem. Phys. 2000, 201, 2780-2788.
5944-5951.
(85) Wilczek, L.; McCord, E. F. US 6 100 350, 2000.
(86) Yamada, B.; Oku, F.; Harada, T. J. Polym. Sci., Polym. Chem.
42) Beuermann, S.; Buback, M.; Jurgens, M. Ind. Eng. Chem.
Res. 2003, 42, 6338-6342.
2003, 41, 645-654.
43) Pierik, S. C. J.; van Herk, A. M. Macromol. Chem. Phys. 2003,
(
87) Moad, G.; Moad, C. L. Macromolecules 1996, 29, 7727-7733.
88) Watanabe, Y.; Ishigaki, H.; Okada, H.; Suyama, S. Chem.
Lett. 1993, 1089-1092.
204, 1406-1418.
(
44) Pierik, S. C. J.; Smeets, B.; van Herk, A. M. Macromolecules
003, 36, 9271-9274.
45) Pierik, B.; Masclee, D.; van Herk, A. Macromol. Symp. 2001,
65, 19-27.
46) Heuts, J. P. A.; Morrison, D. A.; Davis, T. P. ACS Symp. Ser.
2
(
(
(
(
(
(
(
(
89) Yamada, B.; Inoue, K. Kobunshi Ronbunshu 2004, 61, 275-
281.
1
90) Pierik, S. C. J.; Vollmerhaus, R.; van Herk, A. M. Macromol.
Chem. Phys. 2003, 204, 1090-1101.
2000, 768, 313-331.
91) Madruga, E. L.; Fernandez-Garcia, M. Macromol. Chem.
Phys. 1996, 197, 3743-3755.
47) Postma, A.; Davis, T. P.; Moad, G.; O’Shea, M. Macromol-
ecules 2005, 38, 5371-5374.
92) Buback, M.; Feldermann, A.; Barner-Kowollik, C.; Lacik, I.
Macromolecules 2001, 34, 5439-5448.
(48) Schrauzer, G. N. Inorg. Synth. 1968, 61-70.
49) Schrauzer, G. N.; Windgassen, R. J. J. Am. Chem. Soc. 1966,
(
93) McManus, N. T.; Dube, M. A.; Penlidis, A. Polym. React. Eng.
88, 3738-3743.
1999, 7, 131-145.
(50) Meijs, G. F.; Rizzardo, E.; Thang, S. H. Macromolecules 1988,
94) Rudin, A.; Yule, R. G. J. Polym. Sci., Part A-1 1971, 9, 3009-
21, 3122-3124.
3025.
(
51) Moad, G.; Moad, C. L.; Krstina, J.; Rizzardo, E.; Thang, S.
H.; Fryd, M. US 5 756 605, 1998.
95) Ivin, K. J.; Spensley, R. H. J. Macromol. Sci., Chem. 1967,
A1, 653-668.
96) Wittmer, P. Makromol. Chem. 1967, 103, 188-213.
(
52) Gridnev, A. A. Polym. Sci. USSR 1989, 31, 2369-2376.
53) Roberts, G. E.; Davis, T. P.; Heuts, J. P. A. J. Polym. Sci.,
Part A: Polym. Chem. 2003, 41, 752-765.
(
MA0501949