RAFT-Mediated emulsion polymerization of styrene with low reactive xanthate agents: Microemulsion-like behavior

Article abstract of DOI:10.1021/ma101237u

Xanthates ([1-(O-ethylxanthyl)ethyl]benzene (CTA1) and [1-(O- trifluoroethylxanthyl)- ethyl]benzene (CTA2)) have the capacity to control the molecular weight distribution in emulsion polymerizations to produce very small nanoparticles below 20 nm. We form

Full text of DOI:10.1021/ma101237u

Macromolecules 2010, 43, 7565–7576 7565
DOI: 10.1021/ma101237u
RAFT-Mediated Emulsion Polymerization of Styrene with Low Reactive
Xanthate Agents: Microemulsion-like Behavior
Mark P. F. Pepels,† Clovia I. Holdsworth,‡ Sagrario Pascual,§ and Michael J. Monteiro*,†
†Australian Institute for Bioengineering, Nanotechnology, The University of Queensland, Brisbane QLD 4072,
Australia, ‡Centre for Organic Electronics, School of Environmental and Life Sciences, University of
Newcastle, Callaghan, NSW 2308, Australia, and §UCO2M-UMR CNRS6011, LCOM-Chimie des
Polymeꢀres, Universiteꢁ du Maine, Avenue Olivier Messiaen, 72085 Le Mans Cedex 9, France
Received June 2, 2010; Revised Manuscript Received July 11, 2010
ABSTRACT: Xanthates ([1-(O-ethylxanthyl)ethyl]benzene (CTA1) and [1-(O-trifluoroethylxanthyl)-
ethyl]benzene (CTA2)) have the capacity to control the molecular weight distribution in emulsion poly-
merizations to produce very small nanoparticles below 20 nm. We form stable translucent polystyrene latexes
using surfactant (sodium dodecyl sulfate, SDS) and a small amount of pentanol as cosurfactant. The high
CTA concentration results in a greater retardation in rate until consumption of all the RAFT agent. With an
increase in CTA1 the particle size decreases from 38 to 8 nm and the particle number concentration Nc
increases from 2 ꢀ 1018 to 2 ꢀ 1020 particles/L. Although an increase in Nc should in principle lead to a faster
rate of polymerization, we observe a greater retardation in rate with increasing CTA. The higher Ctr,RAFT of
CTA2 results in a greater initial retardation until consumption of all the RAFT agent and particle diameters
lower than 5 nm and at high concentrations of CTA2 diameters that are not measurable. Kinetic simulations
solving the Smith-Ewart equations explain the anomaly between R• (formed from the fragmentation of the R
group from the RAFT agent) acting to nucleate micelles and terminate radicals within particles. The small
and mobile R• radicals can exit particles, re-enter micelles or other particles, re-exit until they either nucleate
micelles, or terminate with propagating polymeric chains. This process of exit and re-entry is similar to limit 3
in a conventional emulsion polymerization. The higher micelle nucleation rate through initiation within
micelles by R• radicals results in smaller and a greater number of particles. Exit is the dominant mechanism for
greater nucleation and retardation.
Introduction
solids (∼50wt %) in a low-viscosity environment, which is easy to
process. On the basis of all these significant advantages of
The versatile “living” radical polymerization (LRP) techniques
produce polymers with controlled molecular weight distributions
(MWDs) and architectures. These techniques comprise (i) rever-
sible addition-fragmentation chain transfer (RAFT),1 (ii) atom
transfer radical polymerization (ATRP),2 (iii) nitroxide-mediated
polymerization (NMP),3 and (iv) metal-catalyzed radical poly-
merization (e.g., single electron transfer LRP),4,5 all of which
produce new, interesting, and responsive (smart) polymeric
structures and materials. All techniques have the capability to
control the polymerization of a vast number of functional
monomers, in a wide range of acceptable solvents, and over a
wide range of temperatures.
Heterogeneous (or emulsion) polymerizations have become
the most attractive industrial method for making a wide varietyof
polymer products.6,7 The major advantages for conducting
polymerizations in dispersed media over bulk or solution include
(i) the use of environmentally friendly medium, usually water,
(ii) the use of a broad range of monomers under a wide range of
experimental conditions, and (iii) the excellent heat transfer that
provides better control of the reaction temperature and thus
better control of the final polymer latex size and molecular weight
distributions. Polymerizations in confined or compartmentalized
particles on the nanoscale result in a significant increase in the
rate of polymerization, rapidly reaching high conversions with
low monomer residuals. The resultant polymer forms in high
heterogeneous polymerization over solution or bulk, it is not
surprising that being able to carry out LRP via an emulsion
process to create polymers with well-defined molecular weight
distributions (MWDs), controlled latex particle size distributions
(PSD), and novel particle morphologies will invariably open a
new class of polymer materials for use in the coatings industry
and as specialty polymers.
Implementation of RAFT in heterogeneous polymerizations is
not as easy as originally believed.8 Previous workinvolving RAFT
ab initio emulsion polymerization with a highly reactive chain
transfer agent (CTA) reveals problems of retardation and colloi-
dal instability.8,9 In addition, this system fails to produce well-
defined molecular weight distributions and fails to control the
particle size distributions (usually broad). The main reason for
this failure is due to the slow transportation10,11 of the highly
reactive CTA from the droplets to the growing particles. Problems
of transportation are overcome by conducting emulsion polymer-
izations with multistep processes,12 using surfactant-like RAFT
agents,13 or conducting the polymerizations in miniemulsions14 or
seeded10 emulsions. Success of these methods relies on the capture
of the RAFT agent inside the growing particles. Recent work by
our group shows that a thermoresponsive nanoreactor success-
fully produces well-defined polymer but, more importantly,
produces narrow and predictable PSDs.15
Microemulsions present an alternative method to conduct
RAFT-mediated polymerizations. They are transparent emul-
sions with thermodynamic stability and allow the synthesis of
*To whom correspondence should be addressed. E-mail: m.monteiro@
uq.edu.au.
r 2010 American Chemical Society
Published on Web 08/17/2010
pubs.acs.org/Macromolecules

7566 Macromolecules, Vol. 43, No. 18, 2010
Scheme 1. Kinetic Scheme from the RAFT to the Simplified Degenerative Chain Transfer Process
Pepels et al.
very small particles with diameters between 10 and 50 nm.16 The
process involves a large amount of surfactant, generally well
above its critical micellar concentration (cmc), to stabilize tiny
monomer-swollen micelles in a continuous aqueous medium. In
many ways, this system is analogous to miniemulsions. Surpris-
ingly, there have been only a few publications on the RAFT-
mediated polymerizations in direct17,18and inverse19 microemul-
sion polymerizations. Recently, Sogabe and McCormick19 have
published the first study on RAFT-mediated inverse microemul-
sion for the homopolymerization of N,N-dimethylacrylamide
and the block copolymerization with N,N-diethylacrylamide.
The very high surfactant loadings used in microemulsions
limits their use in many applications. Pan and co-workers20,21
significantly decreased the amount of surfactant through a
modified microemulsion process they termed “differential micro-
emulsion polymerization”. Their process, for example, relied on
the slow feed of a mixture of styrene and a small amount of
pentanol into the reaction vessel containing sodium dodecyl
sulfate (SDS), water, and ammonium persulfate (APS) to main-
tain a translucent reaction through out the polymerization. The
particle size ranged between 35 and 25 nm, controlled by the
amount of SDS;the greater the SDS, the lower the particle size.
In this work, we used identical conditions to He and Pan21 but in
a one-pot reaction to conduct RAFT-mediated polymerizations
of styrene with low reactive xanthates22 ([1-(O-ethylxanthyl)ethyl]-
benzene (CTA1) and [1-(O-trifluoroethylxanthyl)ethyl]benzene
(CTA2)) under ab initio emulsion conditions (Scheme 1). The
initial opaque reaction mixture contained monomer droplets, but
after the loss of monomer droplets in the system, the reaction
polymerization mixture became translucent with a slight blue hue.
These RAFT agents were chosen because they have (i) the same
leaving R group and allowed us to study the effect of exit and its
dependence upon the chain transfer constant, Ctr,RAFT, and
(ii) have been shown to be well-behaved in all dispersion polymer-
izations in which the Mns and PDIs found in these systems are close
to that found in solution or bulk experiments.23-25 Pentanol has pre-
viously been used in combination with SDS to reduce the interfacial
tension to spontaneously form a microemulsion.26 Therefore, the
addition of pentanol in our ab initio system allows us to attain a high
weight percentage of styrene (R = 13.4%) and low weight percen-
tage of surfactant (γ=3.3%), where R = monomer/(monomer þ
water) ꢀ 100 and γ= surfactant/(surfactant þ monomer þ water) ꢀ
100. We find that both CTAs result in retardation in the rate of
polymerization, and the MWDs, in particular the number-average
molecular weight, Mn, and polydispersity index, PDI, were close to
theoretical predicition for bulk polymerizations.5,27 The particle size
remains relatively constant over the reaction consistent with findings
for most microemulsion polymerizations. In addition, an increase in
the RAFT agent concentration leads to a decrease in the particle
diameter from 38 nm without CTA down to 8 nm with CTA1. This
work provides a methodology to prepare small nanoparticles with
controlled MWDs because of the high exit rate of the R group from
the RAFT agent.
Experimental Section
Materials. All ab initio polymerizations use Milli-Q water (18.2
MΩ cm-1) generated from a Millipore Milli-Q-Academic Water
Purification System. Styrene (STY, Sigma-Aldrich, 99%) was passed
through a column of basic alumina (standard grade, Sigma-Aldrich).
Sodium dodecyl sulfate (SDS, ICN, Ultra pure), ammonium persul-
fate (APS, AR grade, APS Chemicals, 98%), 1-pentanol (Sigma-
Aldrich, 99%), tetrahydrofuran (THF, HPLC grade, LABSCAN,
99.8%), 2,2,2-trifluoroethanol (Sigma-Aldrich, 99%), carbondisulfide
(CS2, Sigma-Aldrich, 99.9%), N,N-dimethylformamide (DMF, Lab
Scan, 99.8%), sodium hydroxide (60% in mineral spirit, Aldrich),
(1-bromoethyl)benzene (Aldrich, 97%), magnesium sulfate anhy-
drous (MgSO4, Amresco), diethyl ether (ACS grade, Merck), petro-
leum spirit (bp 40-60 °C, AR grade, Ajax, 90%), and ethyl acetate
(AR grade) were used as received.
Synthesis of Chain Transfer Agents. [1-(O-Ethylxanthyl)ethyl]-
benzene (CTA1). CTA1 was synthesized according to the literature
procedure.28
[1-(O-Trifluoroethylxanthyl)ethyl]benzene (CTA2)29. To a
stirred solution of trifluoroethanol (1.4 mL, 19.5 mmol) in

Article
Macromolecules, Vol. 43, No. 18, 2010 7567
DMF (14 mL) at 0 °C under argon, a mixture of CS2 (2.6 mL,
43.1 mmol) and sodium hydroxide in mineral oil (0.967 g, 24.2
mmol) was gradually added over a period of 10 min. Upon
complete addition, the reaction mixture was stirred for 45 min at
0 °C. (1-Bromoethyl)benzene (2.5 mL, 17.7 mmol) was added
dropwise, and the mixture stirred overnight at 25 °C. The
mixture was diluted with 50 mL of diethyl ether and washed 3
times with Milli-Q water and 3 times with brine. The organic
phase was then dried over anhydrous magnesium sulfate and
filtered, and the solvents were removed under vacuum. Purifica-
tion by column chromatography (first petroleum spirit, second
95:5 petroleum spirit/ethyl acetate) gave the desired product.
After drying under vacuum 4.22 g of product was obtained
Scheme 2. Main Kinetic Events for RAFT-Mediated ab Initio Poly-
merization, Including Entry and Exit of R•
1
(85% yield). H NMR (CDCl3, 300 MHz): δ = 1.75 (3H, d,
CH3-CH(Ph)-S), 4.88 (3H, m, F3C-CH2-O and S-CH-
(Ph)-CH3), 7.30 (5H, m, aromatic protons). Element analysis
calcd (%) for C11H11F3OS2: C 47.13%, H 3.96%, S 22.88%.
Found: C 47.17%, H 4.2%, S 23.54%.
Techniques. Size Exclusion Chromatography (SEC). All
polymer samples were dried in a vacuum oven for 2 days at
25 °C, dissolved in THF to a concentration of 1 mg/mL, and
then filtered over a 0.45 μm PTFE syringe filter. Analysis of the
molecular weight distributions of the polymers was accom-
plished by using a Waters 2695 separations module, fitted with
a Waters 410 refractive index detector held at a constant
temperature of 35 °C, a Waters 996 photodiode array detector,
two Ultrastyragel linear columns (7.8 ꢀ 300 mm), and one
Styragel linear column kept in series. These columns were held
at a constant temperature of 40 °C for all analyses. The columns
used separate polymers in the molecular weight range of 500-
6 ꢀ 106 g/mol with high resolution. THF was the eluent used at a
flow rate of 1.0 mL/min. Calibration was carried out using
narrow molecular weight PSTY standards (PDI e 1.1) ranging
from 500 to 2 ꢀ 106 g/mol. Data acquisition was performed
using Empower software, and molecular weights were calcu-
lated relative to polystyrene standards.
polymerization. We are not trying to simulate an ab initio
polymerization with its complex nucleation and partitioning
processes but use the simulations to examine the effect of exit
of the R leaving group from the RAFT agent on the rate of
polymerization. In this work, we use two RAFT agents that
show little or no retardation in solution or bulk experiments
with styrene,22,31 and therefore we can simplify the RAFT
mechanism from including all the intermediate radicals to
degenerative chain transfer (Scheme 1). This is an accurate
method to simulate RAFT-mediated polymerizations in solu-
tion or bulk providing that retardation in rate is not
observed.32,33 Our model does not account for the complex
nucleation process but assumes that the initial micelle size is
close to that of the final particle size. This assumption is valid
after about 5% conversion as the particle size remains rela-
tively constant to full conversion in our styrene emulsion
experiments (see data in the Results and Discussion section).
This is also valid for microemulsion polymerizations.34
Scheme 2 gives the main kinetic events for a RAFT-
mediated emulsion where the system is compartmentalized,
and there are no monomer droplets present. First, a thermal
initiator decomposes in the aqueous phase to form radicals,
I•, that then adds to monomer. These radicals continue to
add to monomer and on their way to becoming surface-
active radicals survive termination events with all other
radicals in the aqueous phase ([T•]aq). When they reach a
critical length (of z monomer units or a “z-mer”), they
become surface active and enter a particle or micelle. The
pseudo entry rate coefficient, Finitiator, for radicals derived
from initiator is given by the following equation6,35
1H Nuclear Magnetic Resonance (NMR). All NMR spectra
were recorded on a Bruker DRX 500 MHz spectrometer using
an external lock (D2O, CDCl3).
Dynamic Light Scattering (DLS). Dynamic light scattering
measurements were performed using a Malvern Zetasizer
3000HS. The sample refractive index (RI) was set at 1.59 for
polystyrene. The dispersant viscosity was set to 0.89 N s/m2.
3
The number-average particle diameter, Z-average, and particle
PDI were measured for each sample.
Ab Initio RAFT-Mediated Polymerization of Styrene. A typi-
cal ab initio RAFT-mediated polymerization of styrene was as
follows. A mixture of styrene (3.5 mL, 3.5 ꢀ 10-2 mol), CTA1
(86.43 mg, 3.82 ꢀ 10-4 mol), and 1-pentanol (50 μL) was added
to a 50 mL round-bottom Schlenk flask containing Milli-Q
water (21 mL) and SDS (0.827 g, 2.9 ꢀ 10-3 mol). APS (0.0174 g,
7.64 ꢀ 10-5 mol) was added, and the mixture was purged with
argon for 20 min. For this polymerization weight percentages
are R = (oil/(oil þ water)) ꢀ 100 = 13.2% and γ = (surfac-
tant/(surfactant þ oil þ water)) ꢀ 100 = 3.3%. To start the
polymerization, the mixture was heated to 70 °C (t = 0) and kept
under a positive argon pressure during the reaction. Samples
were taken periodically with a syringe to monitor conversion
(gravimetry), particle size (DLS), and molecular weight distri-
bution (SEC). In experiments where CTA is added, the reaction
mixture during polymerization turns from opaque to translu-
cent once monomer droplets have been consumed.
Theoretical Discussion
Kinetic Modeling of RAFT-Mediated Dispersion Polymer-
ization. The kinetic simulations solve the Smith-Ewart
equations30 for RAFT-mediated dispersion polymerizations.
The specific purpose of these simulations is to examine the
effect of exit through reactions with CTA on the rate of
2kd½IꢁNA
rinitiator
¼
fefficiency
ð1Þ
Nc
where kd is the rate coefficient for initiator decomposition, [I]
is initiator concentration, NA is Avogadro’s number, Nc is

7568 Macromolecules, Vol. 43, No. 18, 2010
Pepels et al.
Table 1. Parameters Used in the Kinetic Simulations for the RAFT-Mediated Microemulsion Polymerization of Styrene at 70 °C
parameter
value
3 mol L-1
description
literature reference
18, 41
Cp
Dp
Nc
kp
kp1
kt, kt1
kd
monomer concentration within the particle
diameter of the particle
number of particles per L
20 nm
1 ꢀ 1020 particles L-1
477 L mol-1 s-1
4770 L mol-1 s-1
1.72 ꢀ 108 L mol-1 s-1
2.2 ꢀ 10-6 s-1
1 ꢀ 10-7 dm2 s-1
1279
propagation rate coefficient
propagation rate coefficient for R• þ M
45
termination rate coefficient
initiator decomposition rate coefficient
diffusion coefficient of R•
33
46
47
38
DR
q
partition coefficient of R•
the number of particles per liter, and fefficiency is the efficiency
to form a z-mer radical from initiator-derived radicals, I•.
In styrene emulsions where the size of the micelles and
particles are small, instantaneous termination occurs when a
z-mer enters a particle already containing a growing radical.
This “zero-one” condition permits only one or zero radicals to
reside in a particle at any one time. In conventional emulsions,
the “zero-one” condition occurs when the probability of
termination of a z-mer is high compared to propagation. To
satisfy this requirement, the rate coefficient for propagation, kp,
should be low (e.g., for styrene) and the particle size be small.
For reactive monomers with a high kp (e.g., butyl acrylate) there
is a high probability that one or more radicals can coexist per
particle even in very small particles.36 In systems with high
surfactant concentrations (i.e., with high number of micelles) as
in microemulsions, they are considered for most monomer
systems as zero-one since there are ∼1000 more micelles than
particles, and the probability of a z-mer entering a particle
containing a radical is negligible. However, in many systems
termination by a z-mer may be non-negligible.37
Small radicals formed within the particles through chain
transfer to monomer or CTA have a high probability of
exiting the small micelles or particles.8,25,38,39 The rate
coefficient for exit depends on the diffusion coefficient
(DR) of the radical, its partition coefficient (q) between the
oil and water phases, and the radius of the swollen particle
(rs) and is given as follows:40
3DR
kexit
¼
ð2Þ
qrs
Figure 1. (A) Kinetic simulations using parameters in Table 1 with “no
exit” using [I] = 3.12 ꢀ 10-3 mol L-1. Curves (a) to (e) represent 0, 1, 5,
10, and 20 [CTA]:[I], respectively. (B) Loss of CTA with conversion
according to theory.27,48
These exited R• radicals can either terminate in the aqueous
phase or re-enter a micelle or particle with rate coefficient:
kre-entry ¼ 4πDRNArs
ð3Þ
concentration in their derivation of the Smith-Ewart equa-
tion), we derive an equation for RAFT. The equation used to
describe the number fraction of particles Nji, inwhichthere are i
active polymeric radicals P• and j active R• radicals, is given as
follows:
and therefore
rre-entry ¼ kre-entry½R•ꢁaq
ð4Þ
The re-entry of R• into a micelle of ∼3 nm would occur quite
rapidly due to the great number of micelles but would also rapidly
re-escape due to their very small size (limit 3 in emulsion
polymerization6). This situation will change if they enter the
larger polymer particles (∼20-40 nm) as exit decreases by an
order of magnitude (cf. eq 2). The R• radicals have a much higher
probability to propagate due to the higher monomer concentra-
tion in the particles compared to micelles (cf. the Morton
equation41) or undergo termination if the particles contain a
polymeric or other radical species. Therefore, exited radicals play
an important role in emulsion systems where the size of the
particles are small and have found to retard the rate of poly-
merization in seeded and ab initio emulsion polymerizations with
similar RAFT agents used in this work.23,25,38
dNij
¼ rinitiatorfð1 - δiÞNj - Njg þ kexitfðj þ 1ÞNj þ 1
i - 1
i
i
dt
- jNjg þ ktr;RAFT½CTAꢁfð1 - δjÞði þ 1ÞNj - 1 - iNijg
i
i þ 1
kt
þ
fði þ 1Þði þ 2ÞNijþ 2 - iði - 1ÞNijg
NAvp
kt1
kt1
þ
fði þ 1Þði þ 2ÞNijþþ11 - ijNijg þ
NAvp
NAvp
fðj þ 1Þðj þ 2ÞNij þ 2 - jðj - 1ÞNjg þ kp1Cpfð1 - δiÞ
i
ðj þ 1ÞNij-þ11 - jNjg þ rre-entryfð1 - δjÞNj - 1 - Nijg
Similar to the derivation by Butte et al.42 for the Smith-
Ewart equations in two dimensions for NMP43 and Luo et al.44
for RAFT (but where j represents the intermediate radical
i
i
þ ktr;RAFT½MacroCTAꢁfð1 - δiÞðj þ 1ÞNj þ 1 - jNijg ð5Þ
i - 1

Article
Macromolecules, Vol. 43, No. 18, 2010 7569
Figure 2. Kinetic simulations usingparametersin Table 1 with“exit” using[I] = 3.12 ꢀ 10-3 mol L-1, inwhichcurves(a) to(e) represent0, 1, 5, 10, and
20[CTA]:[I], respectively. Ctr,RAFT = 0.7. (A) conversionvstime; (B) numberofradicals per particles (n) vs conversion;(C) numberfraction ofparticles
converted to particles by z-mers or exited R• radicals vs conversion; (D) [R•]aq vs conversion and (E) [dead polymer] within the particles vs conversion.
Curve a in (B, E) have their axis on the right-hand side.
dN01
where ktr,RAFT is the transfer rate coefficient to the RAFT
¼ rre-entryN00;micelles
ð7Þ
agent (CTA), vp is the volume of a swollen particle, Cp is the
monomer concentration within the particles, kt is the termina-
tion rate coefficient between two polymeric chains, kt1 is the
termination rate coefficient dominated by short chains, and
MacroCTA consists of a polymer chain with a RAFT end-
group. The Kronecker delta δi or δj in eq 5 represents the
condition if i or j equals 0, then the Kronecker delta value
equals 1. We set that the initial particle number to remain
constant but include in the model the number fraction of
particles that are converted to particles by z-mers or exited
R• radicals, N 00, and those not yet stung or still existing as
micelles, N00,micelles. The initial conditions are N00,micelles = 1
and N 00 = 0, and therefore additional terms (eqs 6-8) must be
included in the full differential equation in eq 5 for N10.
dt
and
dN00;micelles
¼ - ðrinitiator þ rre-entryÞN00;micelles
ð8Þ
ð9Þ
dt
The average number of P• (n) per particle is given by
X X
n ¼
iNij
i
j
The differential equations for all other species are given as
follows:
dN10
dCp
¼ rinitiatorN00;micelles
ð6Þ
¼ - kp1Cp½R•ꢁ - kpCp½P•ꢁ
ð10Þ
dt
dt

7570 Macromolecules, Vol. 43, No. 18, 2010
Pepels et al.
Figure 3. Kinetic simulations usingparametersin Table 1 with“exit” using[I] = 3.12 ꢀ 10-3 mol L-1, inwhichcurves(a) to(e) represent0, 1, 5, 10, and
20 [CTA]:[I], respectively. Ctr,RAFT = 3.5. (A) Conversion vs time; (B) number of radicals per particles (n) vs conversion; (C) number fraction of
particles converted to particles by z-mers or exited R• radicals vs conversion; (D) [R•]aq vs conversion and (E) [dead polymer] within the particles vs
conversion.
Kinetic Simulations with “No Exit”. The first set of kinetic
d½Iꢁ
¼ - kd½Iꢁ
ð11Þ
simulations examines the effect of “no exit” by setting kexit to
zero and varying the concentration of CTA1 (i.e., a CTA
with Ctr,RAFT = 0.7; a value typical for this type of
xanthate22). The model uses the same concentration of
CTA in all particles and neglects partitioning between par-
ticles. Table 1 gives all other parameters. Simulations with-
out CTA (Figure 1A, curve a) result in nearly identical
conversion-time profiles to those for all CTA1 concentra-
tions (Figure 1A, curves b-e). Simulations with a more
reactive CTA (i.e., CTA2 with a Ctr,RAFT = 3.522) also show
near-identical profiles to that without CTA regardless of the
concentration of CTA2 (data not given). Without exit these
simulations result in negligible retardation in rate.
dt
d½CTAꢁ
dt
¼ - ktr;RAFT½CTAꢁ½P•ꢁ þ ktr;RAFT½MacroCTAꢁ½R•ꢁ
ð12Þ
d½R•ꢁaq
¼ kexit½R•ꢁNc=NA - kentry½R•ꢁaqNc=NA
- kt1½R•ꢁaq½T•ꢁaq
dt
ð13Þ
The differential equations (5)-(13) are solved iteratively using
a high-order Runge-Kutta method.5,27 Consistency checks
are made to ensure that the total number of particles and
micelles remained constant with time. We use eq 14 below to
calculate [T•]aq.
The theoretical consumption of CTA in solution or bulk27
over conversion is slightly different between the two CTAs
(Figure 1B). Consumption of CTA2 is faster than CTA1,
with full consumption after 70% conversion. Extending this

Article
Macromolecules, Vol. 43, No. 18, 2010 7571
Figure 4. Kinetic data for the RAFT-mediated emulsion polymerization of styrene at 70 °C varying CTA1 ([1-(O-ethylxanthyl)ethyl]benzene).
(A) conversion vs time, (B) number-average molecular weight (Mn) vs conversion, (C) polydispersity index (PDI) vs conversion, and (D) particle
diameter vs conversion. [I] = 3.12 ꢀ 10-3 mol L-1. Dashed lines are from theory.49
into ab initio suggests there will be high CTA concentration
levels even at conversions as high as 50%. Should the CTA
influence the rate, we would expect this influence to be
effective over much of the conversion range.
termination. The aqueous phase concentration of R•
([R•]aq) increases from 1 ꢀ 10-14 M (at 20% conversion) at
the lowest CTA1 concentration to 1.5 ꢀ 10-13 M (at 20%
conversion) at the highest CTA1 concentration (Figure 2D).
These concentrations, however, are still orders of magnitude
lower than aqueous phase radical produced from initiator
([T•]aq ∼ 8 ꢀ 10-9 M). Applying the steady-state assumption,
[T•]aq is calculated as follows:
Kinetic Simulations with “Exit”. Simulations of CTA1
with Ctr,RAFT = 0.7. The second set of simulations examines
the effect of exit by varying the concentration of CTA1 and
by assuming that the partition coefficient of R• radicals is
identical to styrene due to their similarity in chemical struc-
ture. These simulations generate conversion-time profiles
with a significant retardation in rate with increasing CTA1
concentration (Figure 2A). The simulations also show that
the average number of radicals per particles (n) decrease with
increasing CTA1 (Figure 2B), suggesting that these exited R•
radicals act to reduce the number of active growing radicals
within the particles. In addition, the extremely low values of
n (i.e., well below 0.5) should result in particles that obey the
zero-one condition for styrene.
The small size of both particles and micelles allows rapid
exit (or escape) of R• radicals due their high exit rate
coefficient, resulting in these radicals being able to explore
significantly more micelles and particle than for emulsions in
the absence of CTA. In the absence of CTA, z-mers enter
only a small fraction of micelles (Figure 2C). Even at the
lowest CTA1 concentration, exited radicals enter all the
micelles at a conversion less than 20%, and this conversion
decreases with increasing CTA. However, although this does
not necessarily mean nucleation of micelles to form a grow-
ing particle, there is a high probability for greater nucleation.
This mechanism will also result in an increase in the water
phase of both the radical concentration and the rate of
ꢀ
ꢁ
0:5
kd½Iꢁ
½T•ꢁaq
¼
ð14Þ
kt1
Such a substantially lower [R•]aq compared to [T•]aq and the
decrease of n points toward termination within the particles
as the cause of retardation and not a decrease in the pseudo
entry rate coefficient due to termination in the aqueous
phase. The concentration of dead polymer within the parti-
cles in the absence of CTA (Figure 2E, curve a) is very low
(1.5 ꢀ 10-9 M) and much lower than the cumulative con-
centration of initiator-derived radicals (∼3 ꢀ 10-6 M) after
75 min. The reason for such a low amount of termination
results from the significantly higher probability of z-mers
to enter and nucleate micelles rather than enter and termi-
nate a growing particle, in agreement with classic micro-
emulsion polymerizations.49 The addition of CTA1 in the
simulations gives comparable concentrations of dead poly-
mer (curves b-e) to that from initiator-derived radicals.
Increasing the CTA1 concentration results in a higher
concentration of exited R• that preferentially acts to termi-
nate polymerization.

7572 Macromolecules, Vol. 43, No. 18, 2010
Pepels et al.
Figure 5. Kinetic data for the RAFT-mediated emulsion polymerization of styrene at 70 °C varying initiator concentration in the presence of a
constant CTA1 ([1-(O-ethylxanthyl)ethyl]benzene) concentration. (A) conversion vs time, (B) number-average molecular weight (Mn) vs conversion,
(C) polydispersity index (PDI) vs conversion, and (D) particle diameter vs conversion. [CTA1] = 1.56 ꢀ 10-2 mol L-1. Dashed lines are from theory.49
Simulations of CTA2 with Ctr,RAFT = 3.5. In the final set of
simulations, we examine the effects of exit using the more
reactive CTA2. A comparison between the conversion-time
profiles for the two CTAs (Figures 2A and 3A) shows at the
early stages of polymerization the more reactive CTA2
generates greater retardation until a conversion of ∼70%,
which is the crossover point with CTA1. This crossover
conversion represents the full consumption of CTA2 through
formation to its MacroCTA (see Figure 1B), after which
conversion we observe no retardation but acceleration in
rate. The effect of n with increasing [CTA]/[I] ratio
(Figure 3B) clearly shows a very different trend to the less
reactive CTA, with a slight increase in n with conversion
followed by near-exponential increase after 70% conversion.
A higher Ctr,RAFT value results in an initially greater
concentration of R• in the aqueous phase through exit
(Figure 3D) and consequently a much greater entry and
nucleation of micelles (Figure 3C). This high frequency of re-
entry results in greater concentrations of dead polymer,
which rapidly increase to ∼70% monomer conversion and
then plateau (Figure 3E).
The simulations show that exit and termination through re-
entry into a growing particle causes retardation in the rate of
polymerization. These findings are in agreement with simula-
tions in seeded RAFT-mediated emulsions polymerizations.50
A higher Ctr,RAFT value results in a greater initial retardation
until consumption of all CTA after which time we observe an
acceleration in rate. Through this process of exit and re-entry
until termination with a growing particle, R• can be consid-
ered as a relatively “long-lived” radical compared to its life-
time in either solution or bulk. Our simulations clearly show
that one does not need to invoke the slow fragmentation51 or
intermediate termination52 models53 to account for retarda-
tion in a dispersion (heterogeneous) system.
Results and Discussion
Even though our RAFT-mediated polymerizations are ab
initio emulsion polymerizations, the system resembles that of
microemulsion polymerizations but in our case with a signifi-
cantly much lower surfactant concentration. The reasons for this
statement are as follows. There is a small and constant particle
size over the full conversion range, similar to microemulsion
polymerizations. The polymerization mixture turns translucent
once monomer droplets have been consumed. The monomer
droplets, like the unstung monomer swollen micelles in a micro-
emulsion, will provide a source of monomer to growing particles.
Therefore, the kinetic simulations described above should pro-
vide insight into the kinetic processes in our ab initio system. The
C
tr,RAFT values for CTA1 and CTA2 were previously determined
in solution experiments to be 0.7 and 3.5, respectively.22
Emulsions Using CTA1 (Ctr,RAFT = 0.7). The first series
of ab initio emulsion polymerizations of styrene examines
the effect of CTA1 concentration on the rate of polymeriza-
tions, molecular weight distribution, and particle size distr-
ibution. The emulsion in the absence of CTA1 showed that the
rate of polymerization was fast, reaching full conversion after
only 30 min (Figure 4A, curve a). Increasing the concentra-
tion of CTA1 results in a greater decrease in rate (Figure 4A),
which is in agreement with the trend found from simulations
above. The number-average molecular weight (Mn) decreases
with increasing concentration of CTA1 and is close to theore-
tical values49 calculated from bulk (Figure 4B). We observe
similar agreement between theory and experiment in ab initio

Article
Macromolecules, Vol. 43, No. 18, 2010 7573
Figure 6. Kinetic data for the RAFT-mediated emulsion polymerization of styrene at 70 °C varying CTA2 ([1-(O-trifluoroethylxanthyl)-
ethyl]benzene). (A) conversion vs time, (B) number-average molecular weight (Mn) vs conversion, (C) polydispersity index (PDI) vs conversion,
and (D) particle diameter vs conversion. [I] = 3.12 ꢀ 10-3 mol L-1. Dashed lines are from theory.49
styrene emulsion polymerizations using CTA1 (see Supporting
Information for SEC chromatograms).23 The results suggest a
similar rate of CTA1 consumption to that of monomer in line
with theory (Figure 1B). It also suggests that CTA1 can
partition effectively from non-nucleated micelles or droplets
to growing particles. The polydispersity index (PDI) is close to 2
and remains relatively constant over conversion in agreement
with theory (Figure 4C).27 This implies that although the
MWD is similar to a bulk polymerization the rate in our
microemulsion-like system is substantially more rapid. The
reason for this rapid rate is due to compartmentalization of
growing radicals within the small nanoreactors.54 The effect of
increasing CTA1 produces polymer latex particles with a
decreasing particle size (Figure 4D), decreasing from 38 nm
without CTA to 8 nm at the highest CTA1. The particle size
remains relatively constant over the conversion range regard-
less of the concentration of CTA1. The number concentration
of particles, Nc, increase from 2 ꢀ 1018 particles/L in the
absence of CTA to 2 ꢀ 1020 particles/L in the presence of 20:1
([RAFT]:[I]). This result lends support to the high nucleation
rate of R• through exit and re-entry.
The next set of emulsion polymerizations examines the
effect of initiator concentration at a constant concentration
of CTA1. The rate of polymerization increases with increas-
ing initiator concentration (Figure 5A). There is no effect of
initiator concentration on either the Mn (Figure 5B) or the
PDI (Figure 5C) (see Supporting Information for SEC
chromatograms). More surprising is that the particle size
and particle size evolution with conversion is nearly identical
irrespective of the initiator concentration (Figure 5D). These
results provide some insight into the dominant mechanistic
role of the R• radicals. The rate increase with initiator
concentration is commensurate with the increase in the
pseudo entry rate coefficient, Finitiator (see eq 1). On the basis
of the near-identical particle sizes with increasing initiator
concentration, the exited radical R• controls and dominates
the nucleation process. In addition, these R• radicals act as
terminators to retard the rate of polymerization. Even with a
4 times increase in initiator concentration, the rate of po-
lymerization is still very much slower than without CTA.
Simulation (not shown) with a 4 times initiator concentra-
tion at the same CTA concentration shows identical nuclea-
tion rates of micelles. The invariance of the MWD on
initiator concentration also shows an independence of ini-
tiator concentration on termination, most probably due to
the very low amount of initiator that has decomposed during
the reaction.
Emulsion Polymerizations using CTA2 (Ctr,RAFT = 3.5).
We next examine the effect of the more reactive CTA2 on the
emulsion polymerizations of styrene. The series of polymer-
izations in Figure 6A show that with an increase in CTA2
concentration there is a decrease in the rate of polymeriza-
tion. Comparison between CTA1 and CTA2 rates show that
CTA2 generates a greater initial retardation followed by rate
acceleration. This is similar to the findings from simulations
(Figure 3A). The Mn decreases with increasing CTA2 and is
close to theory (Figure 6B). The PDI (Figure 6C) is similar
for all CTA2 concentrations remaining relatively constant over
conversion and is not as close to theory27 as that observed for
CTA1 (see Supporting Information for SEC chromatograms).
The effect of CTA2 concentration on the particle size is
dramatic (Figure 6D). Even at low concentrations of CTA2,

7574 Macromolecules, Vol. 43, No. 18, 2010
Pepels et al.
Figure 7. Kinetic data for the RAFT-mediated emulsion polymerization of styrene at 70 °C varying initiator concentration in the presence of a
constant CTA2 ([1-(O-trifluoroethylxanthyl)ethyl]benzene) concentration. (A) conversion vs time, (B) number-average molecular weight (Mn) vs
conversion, and (C) polydispersity index (PDI) vs conversion. [CTA2] = 1.56 ꢀ 10-2 mol L-1. Dashed lines are from theory.49
the particle size decreases from 38 nm without CTA2 to less
than 10 nm (Figure 6D, curve b, [RAFT]:[I] = 1:1). An increase
in the [CTA2] above a ratio of 2:1 to initiator results in particles
that are so small that they could not be measured by conven-
tional sizing techniques (i.e., dynamic light scattering, DLS).
These results are in agreement with simulations in which
nucleation of micelles increases with increasing CTA2 concen-
tration (Figure 3C), producing a high particle concentration
and very low particle diameter.
A variation in the initiator concentration at a constant
CTA2 concentration results in a steady increase in the rate of
polymerization (Figure 7A). The Mn and PDI are similar
regardless of initiator concentration (Figure 7B,C), and the
particle size for all polymerizations is too low to measure by
DLS (see Supporting Information for SEC chromatograms).
Again, these results support that exited R• radicals play the
dominant role in both nucleating micelles and acting as
terminators to retard the rate of polymerization. Although
a significantly higher amount of initiator is used, the rate is
still lower than that found without CTA.
faster rate of polymerization, we observe a greater retardation in
rate with increasing CTA. The higher the Ctr,RAFT, the greater the
initial retardation until consumption of all the RAFT agent.
Kinetic simulations solving the Smith-Ewart equations provide
mechanistic insight into explain the anomaly between R• acting to
nucleate micelles and terminate radicals within particles. Small and
mobile R• radicals form from the fragmentation of RAFT agent,
and they can exit particles, re-enter micelles or other particles, re-
exit until they either nucleate micelles, or terminate with propagat-
ing polymeric chains. This process of exit and re-entry is similar to
limit 3 in a conventional emulsion polymerization. The higher
micelle nucleation rate through initiation within micelles by R•
radicals results in smaller and a greater number of particles. Exit is
the dominant mechanism for higher nucleation and retardation.
And one does not need to invoke either the “slow fragmentation”
or “intermediate radical termination” models to explain the effect
of CTA concentration on the retardation of rate.
Acknowledgment. M.J.M. acknowledges financial support
from the ARC Discovery grant and receipt of an Australian
Research Council Future Fellowship. We also thank A/Prof. Per
Zetterlund for useful comments.
Conclusion
In summary, we show the successful RAFT-mediated emulsion
polymerizations of styrene using two RAFT agents with different
chain transfer constants to produce small and stable polystyrene
nanoparticles. The MWD of the resulting polymer is well-
controlled and could be predicted by theory. With an increase in
CTA1 the particle size decreases from 38 to 8 nm, and the particle
number concentration Nc increases from 2 ꢀ 1018 to 2 ꢀ 1020
particles/L. Although an increase in Nc should in principle lead to a
Supporting Information Available: SEC traces for the
RAFT-mediated emulsion polymerizations are given. This materi-
al is available free of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) (a) Moad, G.; Chong, Y. K.; Postma, A.; Rizzardo, E.; Thang, S. H.
Polymer 2005, 46 (19), 8458–8468. (b) Barner, L.; Barner-Kowollik, C.;

Article
Davis, T. P.; Stenzel, M. H. Aust. J. Chem. 2004, 57 (1), 19–24. (c)
Macromolecules, Vol. 43, No. 18, 2010 7575
(16) Dunn, A. S. Polymerization in Micelles and Microemulsions. In
Comprehensive Polymer Science; Allen, G., Bevington, J. C., Eds.;
1989; Vol. 4, pp 219-224.
(17) (a) Liu, S.; Hermanson, K. D.; Kaler, E. W. Macromolecules 2006,
39 (13), 4345–4350. (b) Hermanson, K. D.; Kaler, E. W. Macromole-
cules 2003, 36 (6), 1836–1842. (c) Hermanson, K. D.; Liu, S.; Kaler,
E. W. J. Polym. Sci., Part A: Polym. Chem. 2006, 44 (20), 6055–
6070. (d) Pascual, S.; Monteiro, M. J. Eur. Polym. J. 2009, 45 (9),
2513–2519. (e) Pascual, S.; Urbani, C. N.; Monteiro, M. J. Macromol.
React. Eng. 2010, 4 (3), 257–263.
(18) O’Donnell, J.; Kaler, E. W. Macromolecules 2008, 41, 6094–6099.
(19) Sogabe, A.; McCormick, C. L. Macromolecules 2009, 42, 5043–
5052.
(20) He, G.; Pan, Q.; Remple, G. L. Macromol. Rapid Commun. 2003,
24, 585–588.
(21) He, G.; Pan, Q. Macromol. Rapid Commun. 2004, 25, 1545–1548.
(22) Adamy, M.; van Herk, A. M.; Destarac, M.; Monteiro, M. J.
Macromolecules 2003, 36 (7), 2293–2301.
(23) Monteiro, M. J.; de Barbeyrac, J. Macromolecules 2001, 34 (13),
4416–4423.
(24) Monteiro, M. J.; Sjoberg, M.; Van der Vlist, J.; Gottgens, C. M.
J. Polym. Sci., Part A: Polym. Chem. 2000, 38 (23), 4206–4217.
(25) Monteiro, M. J.; Adamy, M. M.; Leeuwen, B. J.; van Herk, A. M.;
Destarac, M. Macromolecules 2005, 38 (5), 1538–1541.
(26) (a) Chern, C.-S.; Wu, L.-J. J. Polym. Sci., Part A: Polym. Chem.
2001, 39, 898–912. (b) Gerbacia, W.; Rosano, H. L. J. Colloid Inter-
face Sci. 1973, 44 (2), 242–248.
(27) Monteiro, M. J. J. Polym. Sci., Part A: Polym. Chem. 2005, 43 (15),
3189–3204.
(28) (a) Charmot, D.; Corpart, P.; Michelet, D.; Zard, S. Z.; Biadatti, T.
WO 9858974, 1998. (b) Bell, C. A.; Smith, S. V.; Whittaker, M. R.;
Whittaker, A. K.; Gahan, L. R.; Monteiro, M. J. Adv. Mater. 2006, 18,
582–586.
(29) Destarac, M.; Charmot, D.; Zard, S.; Franck, X. WO 0075207,
2000.
(30) Smith, W. V.; Ewart, R. H. J. Chem. Phys. 1948, 16, 592–599.
(31) Batt-Coutrot, D.; Robin, J.-J.; Bzducha, W.; Destarac, M. Macro-
mol. Chem. Phys. 2005, 206, 1709–1717.
Bernard, J.; Favier, A.; Davis, T. P.; Barner-Kowollik, C.; Stenzel, M. H.
Polymer 2006, 47 (4), 1073–1080. (d) Whittaker, M. R.; Monteiro, M. J.
Langmuir 2006, 22 (23), 9746–9752. (e) Mayadunne, R. T. A.; Jeffery,
J.; Moad, G.; Rizzardo, E. Macromolecules 2003, 36 (5), 1505–1513. (f)
Wang, W.-p.; You, Y.-z.; Hong, C.-Y.; Xu, J.; Pan, C.-Y. Polymer 2005,
46 (22), 9489–9494. (g) Monteiro, M. J.; Bussels, R.; Beuermann, S.;
Buback, M. Aust. J. Chem. 2002, 55, 433–437.
(2) (a) Matyjaszewski, K. Mol. Cryst. Liq. Cryst. 2004, 415, 23–34. (b)
Matyjaszewski, K. Curr. Org. Chem. 2002, 6 (2), 67–82. (c) Whittaker,
M. R.; Urbani, C. N.; Monteiro, M. J. J. Am. Chem. Soc. 2006, 128
(35), 11360–11361. (d) Huang, J.; Jia, S.; Siegwart, D. J.; Kowalewski,
T.; Matyjaszewski, K. Macromol. Chem. Phys. 2006, 207 (9), 801–
811. (e) Li, M.; Jahed, N. M.; Min, K.; Matyjaszewski, K. Macro-
molecules 2004, 37 (7), 2434–2441. (f ) An, S. G.; Li, G. H.; Cho, C. G.
Polymer 2006, 47 (11), 4154–4162.
(3) (a) Hawker, C. J.; Bosman, A. W.; Harth, E. Chem. Rev. 2001, 101
(12), 3661–3688. (b) Studer, A.; Schulte, T. Chem. Rec. 2005, 5 (1),
27–35. (c) Hill, N. L.; Braslau, R. Abstracts of Papers, 232nd ACS
National Meeting, San Francisco, CA, Sept 10-14, 2006; POLY-040.
(d) Chauvin, F.; Dufils, P.-E.; Gigmes, D.; Guillaneuf, Y.; Marque,
S. R. A.; Tordo, P.; Bertin, D. Macromolecules 2006, 39 (16), 5238–
5250. (e) Lagrille, O.; Cameron, N. R.; Lovell, P. A.; Blanchard, R.;
Goeta, A. E.; Koch, R. J. Polym. Sci., Part A: Polym. Chem. 2006, 44
(6), 1926–1940.
(4) (a) Percec, V.; Guliashvili, T.; Ladislaw, J. S.; Wistrand, A.;
Stjerndahl, A.; Sienkowska, M. J.; Monteiro, M. J.; Sahoo, S.
J. Am. Chem. Soc. 2006, 128 (43), 14156–14165. (b) Percec, V.;
Popov, A. V.; Ramirez-Castillo, E.; Monteiro, M.; Barboiu, B.;
Weichold, O.; Asandei, A. D.; Mitchell, C. M. J. Am. Chem. Soc.
2002, 124 (18), 4940–4941. (c) Percec, V.; Popov, A. V.; Ramirez-
Castillo, E.; Hinojosa-Falcon, L. A. J. Polym. Sci., Part A: Polym.
Chem. 2005, 43 (11), 2276–2280. (d) Percec, V.; Ramirez-Castillo, E.;
Hinojosa-Falcon, L. A.; Popov, A. V. J. Polym. Sci., Part A: Polym.
Chem. 2005, 43 (10), 2185–2187. (e) Percec, V.; Ramirez-Castillo, E.;
Popov, A. V.; Hinojosa-Falcon, L. A.; Guliashvili, T. J. Polym. Sci.,
Part A: Polym. Chem. 2005, 43 (10), 2178–2184.
(5) Monteiro, M.; Guliashvili, T.; Percec, V. J. Polym. Sci., Part A:
Polym. Chem. 2007, 45, 1835–1847.
(6) Gilbert, R. G. Emulsion Polymerization: A Mechanistic Approach;
Academic Press: London, 1995.
(32) (a) Monteiro, M. J. J. Polym. Sci., Part A: Polym. Chem. 2005, 43
(22), 5643–5651. (b) Johnston-Hall, G.; Monteiro, M. J. Macromole-
cules 2007, 40 (20), 7171–7179.
(7) (a) Fitch, R. M. Polymer Colloids; Academic Press: New York, 1997.
(b) Lovell, P. A.; E.-A., M. S. Emulsion Polymerization and Emulsion
Polymers; John Wiley & Sons Ltd.: Chichester, England, 1997.
(8) Monteiro, M. J.; Hodgson, M.; De Brouwer, H. J. Polym. Sci., Part
A: Polym. Chem. 2000, 38 (21), 3864–3874.
(33) Johnston-Hall, G.; Monteiro, M. J. J. Polym. Sci., Part A: Polym.
Chem. 2008, 46, 3155–3173.
(34) Full, A. P.; Kaler, E. W.; Arellano, J.; Puig, J. E. Macromolecules
1996, 29, 2764–2775.
(35) Maxwell, I. A.; Morrison, B. R.; Napper, D. H.; Gilbert, R. G.
(9) (a) McLeary, J. B.; Klumperman, B. Soft Matter 2006, 2 (1), 45–53.
(b) Urbani, C. N.; Monteiro, M. J. RAFT-mediated polymerization in
heterogeneous systems. In Handbook of RAFT Polymerization;
Barner-Kowollik, C., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA:
Berlin, 2008; pp 285-314. (c) Monteiro, M. J.; Charleux, B. Living
radical polymerisation in emulsion and miniemulsion. In Chemistry
and Technology of Emulsion Polymerisation; Herk, A. v., Ed.; Black-
well Publishing Ltd.: New York, 2005; pp 111-139. (d) Zetterlund,
P. B.; Kagawa, Y.; Okubo, M. Chem. Rev. 2008, 108, 3747–3794. (e)
Cunningham, M. F. Prog. Polym. Sci. 2008, 33, 365–398.
Macromolecules 1991, 24, 1629–1640.
(36) Maeder, S.; Gilbert, R. G. Macromolecules 1998, 31 (14), 4410–
4418.
(37) de Vries, R.; Co, C. C.; Kaler, E. W. Macromolecules 2001, 34,
3233–3244.
(38) Smulders, W.; Gilbert, R. G.; Monteiro, M. J. Macromolecules
2003, 36 (12), 4309–4318.
(39) Smulders, W.; Monteiro, M. J. Macromolecules 2004, 37, 4474–
4483.
(40) Ugelstad, J.; Hanson, F. K. Rubber Chem. Technol. 1976, 49, 536–
546.
(41) Morton, M.; Kaizerman, S.; Altier, M. W. J. Colloid Sci. 1954, 9,
300–312.
(42) Butte, A.; Storti, G.; Morbidelli, M. DECHEMA Monogr. 1998,
134, 497–507.
(43) Zetterlund, P. B.; Okubo, M. Macromolecules 2006, 39, 8959–8967.
(44) Luo, Y.; Wang, R.; Yang, L.; Yu, B.; Li, B.; Zhu, S. Macromole-
cules 2006, 39, 1328–1337.
(45) Manders, B. G.; Chambard, G.; Kingma, W. J.; Klumperman, B.;
van Herk, A. M.; German, A. L. J. Polym. Sci., Part A: Polym.
Chem. 1996, 34, 2473–2479.
(46) Behrman, E. J.; Edwards, J. O. Rev. Inorg. Chem. 1980, 2, 179–182.
(47) Griffiths, M. C.; Straunch, J.; Monteiro, M. J.; Gilbert, R. G.
Macromolecules 1998, 31 (22), 7835–7844.
(48) Mueller, A. H. E.; Zhuang, R.; Yan, D.; Litvinenko, G. Macro-
molecules 1995, 28 (12), 4326–33.
(49) (a) Guo, J. S.; El-Aasser, M. S.; Vanderhoff, J. M. J. Polym. Sci.,
Polym. Chem. Ed. 1989, 27, 691–710. (b) Puig, J. E.; Perez-Luna,
V. H.; Perez-Gonzales, M.; Macias, E. R.; Rodriguez, B. E.; Kaler, E. W.
Colloid Polym. Sci. 1993, 271, 114–123.
(10) Prescott, S. W.; Ballard, M. J.; Rizzardo, E.; Gilbert, R. G.
Macromolecules 2002, 35 (14), 5417–5425.
(11) Prescott, S. W.; Ballard, M. J.; Rizzardo, E.; Gilbert, R. G.
Macromolecules 2005, 38 (11), 4901–4912.
(12) (a) Ferguson, C. J.; Hughes, R. J.; Nguyen, D.; Pham, B. T. T.;
Gilbert, R. G.; Serelis, A. K.; Such, C. H.; Hawkett, B. S. Macro-
molecules 2005, 38 (6), 2191–2204. (b) Ferguson, C. J.; Hughes, R. J.;
Pham, B. T. T.; Hawkett, B. S.; Gilbert, R. G.; Serelis, A. K.; Such, C. H.
Macromolecules 2002, 35 (25), 9243–9245.
(13) (a) Rieger, J.; Osterwinter, G.; Bui, C.; Stoffelbach, F.; Charleux,
B. Macromolecules 2009, 42, 5518–5525. (b) Rieger, J.; Stoffelbach,
F.; Bui, C.; Alaimo, D.; Jerome, C.; Charleux, B. Macromolecules 2008,
41 (12), 4065–4068. (c) Rieger, J.; Stoffelbach, F.; Bui, C.; Alaimo, D.;
Jerome, C.; Charleux, B. Macromolecules 2008, 41, 4065–4068.
(14) (a) de Brouwer, H.; Tsavalas, J. G.; Schork, F. J.; Monteiro, M. J.
Macromolecules 2000, 33 (25), 9239–9246. (b) Lansalot, M.; Davis,
T. P.; Heuts, J. P. A. Macromolecules 2002, 35 (20), 7582–7591.
(c) Tsavalas, J. G.; Schork, F. J.; de Brouwer, H.; Monteiro, M. J.
Macromolecules 2001, 34 (12), 3938–3946.
(15) (a) Urbani, C. N.; Monteiro, M. J. Macromolecules 2009, 42 (12),
3884–3886. (b) Urbani, C. N.; Monteiro, M. J. Aust. J. Chem. 2009, 62
(11), 1528–1532.
(50) Peklak, A. D.; Butte, A. J. Polym. Sci., Part A: Polym. Chem. 2006,
44 (20), 6114–6135.

7576 Macromolecules, Vol. 43, No. 18, 2010
Pepels et al.
(51) Barner-Kowollik, C.; Quinn, J. F.; Morsely, D. R.; Davis,
T. P. J. Polym. Sci., Part A: Polym. Chem. 2001, 39, 1353–
(53) Barner-Kowollik, C.; Buback, M.; Charleux, B.; Coote, M. L.; Drache,
M.; Fukuda, T.; Goto, A.; Klumperman, B.; Lowe, A. B.; Mcleary,
J. B.; Moad, G.; Monteiro, M. J.; Sanderson, R. D.; Tonge, M. P.;
Vana, P. J. Polym. Sci., Part A: Polym. Chem. 2006, 44 (20), 5809–5831.
(54) Monteiro, M. J. Macromolecules 2010, 43 (3), 1159–1168.
1357.
(52) Monteiro, M. J.; de Brouwer, H. Macromolecules 2001, 34,
349–352.