This document summarizes a study investigating the initialization behavior of reversible addition-fragmentation chain transfer (RAFT)-mediated styrene-maleic anhydride copolymerizations using in situ 1H NMR spectroscopy. The results indicate specificity of addition of the RAFT agent leaving groups for either styrene or maleic anhydride. Analysis of the NMR spectra also showed that monomers are added individually, favoring the penultimate unit model of polymer propagation over other proposed mechanisms. Stereoselectivity was observed during monomer addition to the RAFT agent.

1. RESEARCH FRONT
CSIRO PUBLISHING
Full Paper
Aust. J. Chem. 2006, 59, 742–748 www.publish.csiro.au/journals/ajc
Investigation into the Initialization Behaviour of RAFT-Mediated
Styrene–Maleic Anhydride Copolymerizations
Eric T. A. van den Dungen,A Jacques Rinquest,A Nadine O. Pretorius,A Jean M. McKenzie,B
James B. McLeary,A Ron D. Sanderson,A and Bert KlumpermanA,C,D
A UNESCO Centre for Macromolecules and Materials, Department of Chemistry and Polymer Science,
University of Stellenbosch, Matieland 7602, South Africa.
B Central Analytical Facility, University of Stellenbosch, Matieland 7602, South Africa.
C Laboratory of Polymer Chemistry, Eindhoven University of Technology,
5600 MB Eindhoven, The Netherlands.
D Corresponding author. Email: l.klumperman@tue.nl
The living radical alternating copolymerization of styrene and maleic anhydride mediated by the reversible addition–
fragmentation chain transfer (RAFT) polymerization process has been studied at short chain lengths using two
different dithiobenzoate RAFT agents.The results indicate specificity of addition of the RAFT-agent leaving groups
for either styrene or maleic anhydride. The addition rate of the monomers and the fact that monomers are added
individually favour the penultimate unit model of polymer propagation.
Manuscript received: 7 July 2006.
Final version: 14 September 2006.
Introduction
Radical polymerization is a convenient commercially appli-
cable method for the synthesis of copolymers. The rapid
chain growth and short lifetimes of chains in conventional
free radical polymerization are, however, disadvantages that
lead to chains with broad chemical composition distribu-
tions within a single reaction. To ensure that all chains have
similar properties it is necessary to increase the lifetime of
chains and ensure that they are able to propagate for the full
reaction time. To achieve this, living radical polymerization
techniques have been developed. Among the most widely
used living radical polymerization techniques are stable free
radical polymerization (SFRP),[1] atom transfer radical poly-
merization (ATRP),[2] and reversible addition–fragmentation
chain transfer (RAFT) polymerization.[3–6]
Recently, a significant amount of attention has been paid
to the synthesis of alternating styrene (Sty)–maleic anhy-
dride (MAh) copolymers by living radical polymerization
techniques.[7–11] The copolymerization of Sty and MAh by
conventional radical polymerization provides an alternating
copolymer under correctly selected conditions.[12,13] How-
ever, the molecular weight and molecular-weight distribution
are not well defined, and the chemical-composition distri-
bution of the copolymer obtained is complex unless the
monomer composition is controlled. In conventional radical
polymerization the chains formed during the early stage of the
polymerization have a different composition to those at a later
stage of the polymerization as chains are initiated throughout
the polymerization. It is advantageous to form copolymers
by living radical polymerization, as all chains start growing
at approximately the same time and thus all chains will have
a similar composition.
Several research groups have reported the living radical
copolymerization of Sty and MAh. Park et al. and Benoit et al.
performed Sty–MAh copolymerizations using SFRP. Park
etal.usedbenzylperoxideand2,2,6,6-tetramethylpiperidine-
1-oxyl (TEMPO)[14] and synthesized random copolymers
of Sty and MAh using very low amounts of MAh (up
to 17.5 mol-%). The polymerization only showed living
behaviour when carried out at 120◦C. At different tem-
peratures the copolymerizations deviated from the lin-
ear relationship between molecular weight and conversion,
which indicates that the copolymerization deviated from the
expected living/controlled behaviour.[3,4] Benoit et al. suc-
cessfully synthesized copolymers that consist of poly[(Sty-
ran-MAh)-block-Sty] in the presence of a mixture of an
α-hydrido alkoxyamine and a nitroxide.[7] Several research
groups have reported Sty–MAh copolymerizations mediated
by the RAFT technique. De Brouwer et al. copolymerized
Sty and MAh by RAFT to result in a copolymer with pre-
determined molecular weight and narrow molecular weight
distribution.[8] You et al. used the RAFT technique for the
copolymerization of Sty and MAh at 22◦C without using
an initiator, to result in an alternating copolymer. This was
© CSIRO 2006 10.1071/CH06238 0004-9425/06/100742

2. RESEARCH FRONT
Initialization Behaviour of Styrene–Maleic Anhydride Copolymerizations 743
Table 1. Composition of reaction mixtures for both copolymerizations
Exp Sty MAh RAFT agent AIBN 1,4-Dioxane [D6]Benzene Mn,th Ratio
[g] [mmol] [g] [mmol] [g] [mmol] [g] [mmol] [g] [mmol] [g] [mmol] [g mol−1] RAFT/AIBN
1A 0.101 0.97 0.090 0.92 0.053 0.24 0.008 0.05 0.123 1.39 0.302 3.59 1044 4.93
2B 0.111 1.07 0.101 1.03 0.079 0.29 0.009 0.06 0.132 1.49 0.276 3.28 1020 5.03
A RAFT agent: CIPDB, T 70◦C.
B RAFT agent: CDB, T 60◦C.
confirmed by 13C and DEPT NMR experiments.[9] Zhu et al.
and Harrisson et al. reported on the copolymerization of
poly(Sty-alt-MAh) and poly[(Sty-alt-MAh)-block-polySty],
respectively, by RAFT-mediated polymerization, and the sub-
sequent self-assembly of the copolymer in water to form
uniform nanoparticles.[15,16] Du et al. investigated the nature
of the propagating radical attached to the RAFT agent
by electron spin resonance (ESR) spectroscopy for a ben-
zyldithiobenzoate-mediatedSty–MAhcopolymerization.[12]
They found that the intermediate radical formed during the
copolymerization is derived from active MAh-ended propa-
gating radicals. From the nature of the intermediate radical
they concluded that the mechanism of propagation is by
the charge-transfer complex (CTC) model, because there
should be only one kind of propagating radical present in
this case. However, other models cannot be excluded as the
active propagating radical could not directly be observed
by the ESR experiments. Chernikova et al. investigated the
kinetics of RAFT-mediated Sty–MAh copolymerizations and
concluded that an increase in the amount of MAh in the
comonomer feed leads to an increase in the polymerization
rate and a decrease in the control over the reaction.[10] This
was attributed to an increase in the average rate of polymer-
ization (kp) for higher amounts of MAh in the comonomer
feed (which was also found for conventional radical copoly-
merization of Sty and MAh),[18] which leads to a lower chain
transfer constant. Davies et al. carried out copolymerizations
of MAh with a variety of substituted styrenes, all of which
resulted in controlled polymerizations, except for α-methyl
styrene.[11]
The mechanism of the alternating copolymerization of
Sty–MAh has been a subject of debate in the literature and
several models have been suggested.[17–20] Studies of Sty–
MAh copolymerizations reported that the temperature must
be kept below 80◦C in order for the copolymerization to
proceed in an alternating fashion, because at higher tem-
peratures CTCs cannot be formed and the copolymerization
would rather proceed in a random manner.[9] A significant
amount of evidence has, however, been published, which
shows that the experimental results obtained from Sty–MAh
copolymerizations are best described by the penultimate
unit (PU) model.[17,19] The study presented here reports the
initialization behaviour of RAFT-mediated Sty–MAh copoly-
merizations using 2-cyanoprop-2-yl dithiobenzoate (CIPDB)
and cumyl dithiobenzoate (CDB) as RAFT agents at 60 and
70◦C, investigated by in situ 1H NMR spectroscopy. Further-
more, the alternating behaviour for the comonomer pair using
an equimolar amount of both monomers is investigated.
Experimental
Chemicals
Sty (Plascon Research Centre, University of Stellenbosch, estimated
purity ∼99% by 1H NMR) was washed three times with a 0.3 M KOH
solution to remove the inhibitor, and the Sty was subsequently distilled
under reduced pressure. 2,2-Azobis(isobutyronitrile) (AIBN, Riedel-
de Haën) was recrystallized from ethanol and dried under vacuum.
MAh (Acros organics 99%), 1,4-dioxane (Saarchem uniLAB 99%), and
[D6]benzene (Acros Organics) were used as received.
Synthesis of RAFT Agents
CIPDB[3] and CDB[3] were synthesized according to procedures readily
available in the literature.The purity of the RAFT agents, as determined
by 1H NMR spectroscopy, appeared to be 96% for CIPDB and 99%
for CDB.
Copolymerizations
Sty, MAh, 1,4-dioxane, RAFT agent (CIPDB or CDB), AIBN, and
[D6]benzene were accurately weighed into a glass vial and homoge-
nized. The solution was degassed in the NMR tube by purging with
nitrogen. The experiment was conducted in a closed NMR tube and the
cavity of the magnet was kept under an N2 flux that maintained the
inert atmosphere. Table 1 shows the amounts weighed of the various
compounds for the copolymerizations carried out. The copolymeriza-
tions were performed at either 60 or 70◦C and followed online by in-situ
1H NMR spectroscopy.A spectrum was collected every minute for about
three hours.
1H NMR Analysis
In situ 1H NMR experiments were carried out on a 600 MHz Varian
Unity Inova spectrometer. A 5 mm inverse detection PFG probe was
used for the experiments and the probe temperature was calibrated using
an ethylene glycol sample in the manner suggested by the manufacturer
using the method of Van Geet.[21] 1H NMR spectra were acquired with
a 3 µs (40◦) pulse width and a 4 s acquisition time. For the 1H NMR
kinetic experiments, samples were inserted into the magnet at 25◦C and
the magnet was fully shimmed on the sample.A spectrum was collected
at 25◦C to serve as a reference. The sample was then removed from
the magnet and the cavity of the magnet was raised to the required
temperature (60 or 70◦C). Once the magnet cavity had stabilized at
the required temperature, the sample was re-inserted (time zero) and
allowed to equilibrate. Additional shimming was then carried out to
fully optimize the system and the first spectra were recorded ∼5 min
after the sample was inserted into the magnet. Integration of the spectra
was carried out using ACD laboratories 7.0 1-D 1H NMR processor
software.
Results and Discussion
The 1H NMR spectra obtained from the Sty–MAh copoly-
merizations were analyzed in a similar way to that of earlier
published work.[22–24] The first monomer adduct of the
reaction of CIPDB and Sty was identified by means of a
detailed NMR study (see Table 2). During the analysis of

3. RESEARCH FRONT
744 E. T. A. van den Dungen et al.
Table 2. Peak assignment and numbering of the atoms for the first
monomer adduct of a CIPDB-mediated Sty homopolymerization
S
S
NC
*
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Position δC δH (J [Hz])
1 123.55
2 31.49
3/4 26.62, 27.44 1.18 s, 1.36 s
5 44.46 2.32 dd (9.9, 14.2)
2.39 dd (4.9, 14.2)
6 51.99 5.34 dd (4.9, 9.9)
7 138.29
8/12 128.06
9/11 128.38 7.30 m
10 128.23
13 225.10
14 144.28
15/19 126.65 7.91 dd (1.2, 8.5)
16/18 128.69 7.43 m
17 132.29
the NMR spectra of the copolymerizations, some interesting
observations were made regarding the stereochemistry of the
products formed.This is discussed briefly here. During a radi-
cal addition reaction, the carbon bearing the unpaired electron
in the radical intermediate is sp2 hybridized and the three
substituent groups on the carbon lie in a plane. The added
monomer has equal access to both sides of the plane, which
results in equal formation of both R and S enantiomers. To
understand the copolymerization of Sty and MAh it should
be noted that Sty may homopolymerize while MAh does not.
For this reason it is relevant to examine the behaviour of
Sty homopolymerization. For the second monomer addition
of Sty to a Sty unimer, a chiral centre already exists in the
propagating radical, and the addition of a second monomer
results in the formation of a second chiral centre (after addi-
tion to the RAFT agent) to result in diastereomers. Unequal
formation of R and S configurations occurs at the second
centre, because of greater access on one side of the radical
intermediate as a result of steric constraints when adding to
the RAFT agent. Four stereocombinations are now possible,
being RR, SS, RS, or SR. The RR and SS configurations are
one enantiomer pair (R∗R∗), while the RS and SR configu-
rations constitute another (R∗S∗). It should be noted at this
point that NMR spectroscopy only allows one to distinguish
between diastereomers but not between enantiomers (unless
an auxillary reagent is used, which was not feasible in this
work because of the complexity of the systems). Thus the
R∗R∗ stereoisomers appear as one signal and the R∗S∗ pair
also appear as a single resonance though at a different chem-
ical shift from the R∗R∗ pair. In the 1H NMR spectrum of the
second Sty–Sty monomer adduct the peaks exhibit a ratio of
45:55, which indicates that stereoselectivity occurs during the
addition reaction to the RAFT agent. Bulk polystyrene has
S
S
Ph
H H
O
O
O
H
H
H3C
CH3
Ph
H Ph
Fig. 1. R∗R∗S∗ stereoisomer of the second monomer adduct C-MAh-
Sty-D showing selected nOe interactions that were detected by one-
dimensional NOE spectra.
a meso-diad fraction P(m) = 0.46, i.e., an excess of syndio-
tacticity exists over isotacticity. Syndiotacticity implies that
the relative configuration of the chiral centres alternate and
that racemic dyads dominate. From one-dimensional NOE
spectra it was possible to determine that R∗R∗ configurations
were preferred over R∗S∗, which implies that the stereochem-
istry was similar to bulk polystyrene as R∗R∗ corresponds to
a racemic dyad in the second monomer addition.
For the determination of the position of the methyl protons
of the RAFT agent for the first monomer adduct containing a
MAh unit, eitherA-MAh-D or C-MAh-D (A and C represent
the leaving groups 2-cyanoprop-2-yl and cumyl respectively,
and D for the dithiobenzoate moiety), two in-situ 1H NMR
experiments were conducted with MAh, CDB or CIPDB,
AIBN, and 1,4-dioxane in [D6]benzene. From these experi-
mentsthestructure,peaks,andcouplingsofthefirstmonomer
adducts were determined. It was observed that the first
monomer adduct that contained a MAh unit (A-MAh-D and
C-MAh-D) contained only two peaks for the methyl groups
instead of four, as expected because of the two chiral carbon
centres (since the methyl groups behave diastereotopically).
With the addition of Sty to form the second monomer adduct,
four methyl peaks were observed. The stereochemistry of the
second monomer adduct was more closely investigated for
C-MAh-Sty-D. From one-dimensional NOE spectra it was
found that the CH protons of the MAh unit are anti for both
the diastereomers that can be observed in the 1H NMR spec-
trum. This also explains the observations from the 1H NMR
spectra of the first monomer adductsA-MAh-D and C-MAh-
D. The chiral centre formed with the addition of Sty can be
either R or S and it was found that the R∗R∗S∗ diastereomer is
marginally preferred to the S∗R∗S∗ diastereomer with a ratio
of ∼58:42.The assignment of the peaks in the 1H NMR spec-
trum, because of the two diastereomers present, was again
achieved using selective one-dimensional NOE experiments.
Fig. 1 shows selected NOE correlations observed for the
R∗R∗S∗ diastereomer, which made the assignment possible.
The peaks used for the quantification of species in this
study were predominantly the methyl peaks of the leav-
ing group of the RAFT agent for the initial RAFT agent
(2-cyanoprop-2-yl or cumyl, AD/CD), the first monomer
adduct (AMD/CMD, where M represents one monomer
unit), and the second monomer adduct (AM2D/CM2D, where
M2 represents two monomer units). The main focus of the
study was to investigate the presence of an initialization

4. RESEARCH FRONT
Initialization Behaviour of Styrene–Maleic Anhydride Copolymerizations 745
period[22–24] for these copolymerizations together with the
preference of the leaving group of the RAFT agent to add to
one of the monomers (Sty or MAh) during the initialization
period. The peaks that belong to the monomers were also
1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7
15 min
30 min
45 min
60 min
CIPDB
⫺C(CH3)2CN
AIBN CIP-MAh-DB
⫺C(CH3)2CN
CIP-Sty-DB
⫺C(CH3)2CN
CIP-Sty-MAh-DB
⫺C(CH3)2CN
d [ppm]
5 min
Fig. 2. 1H NMR spectra of a CIPDB-mediated Sty–MAh copolymerization at 70◦C, enlarged at the methyl
region of the leaving group of the RAFT agent (δH 0.7–1.6). The spectra have been taken after 5, 15, 30,
45, and 60 min and shown are the leaving group (CIP), –C(CH3)2CN, the MAh-derived first monoadduct,
CIP-MAh-DB, the Sty-derived first monoadduct, CIP-Sty-DB, and the second monomer adduct derived from
one Sty unit and one MAh unit, CIP-Sty-MAh-DB.
1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8
d [ppm]
5 min
CDB
⫺C(CH3)2Ph
C-MAh-DB
⫺C(CH3)2Ph
C-MAh-DB
⫺C(CH3)2Ph
10 min
20 min
30 min
40 min
AIBN C-MAh-Sty-DB
⫺C(CH3)2Ph
Fig. 3. 1H NMR spectra of a CDB-mediated Sty–MAh copolymerization at 60◦C, enlarged at the
methyl region of the leaving group of the RAFT agent (δH 0.8–1.9). The spectra have been taken after
5, 10, 20, 30, and 40 min and shown are the leaving group (C), –C(CH3)2Ph, the MAh-derived first
monoadduct, C-MAh-DB, and the second monomer adduct derived from one MAh unit and one Sty unit,
C-MAh-Sty-DB.
analyzed in order to calculate the conversion and examine
the alternating behaviour for the copolymerizations. In Figs 2
and 3 spectra obtained from in situ 1H NMR experiments of
CIPDB and CDB-mediated Sty–MAh copolymerizations at

5. RESEARCH FRONT
746 E. T. A. van den Dungen et al.
70◦C and 60◦C, respectively, are shown.The expansion of the
spectra in the methyl region of the leaving group shows that
all relevant species can clearly be observed, i.e., the peaks of
the leaving group of the RAFT agent that belongs to the initial
RAFT agent (AD or CD), the first (A-Sty-D or C-MAh-D),
and the second (A-Sty-MAh-D or C-MAh-Sty-D) monomer
adducts.
Fig. 4 shows the consumption of the initial RAFT agent
and the formation of the first, second, and third monomer
adducts for a CIPDB-mediated Sty–MAh copolymerization
at 70◦C, monitored by means of in situ 1H NMR spec-
troscopy. The initialization period can clearly be observed
for the copolymerization of Sty and MAh, which means that
the RAFT agent is converted into a macroRAFT agent with
high selectivity, and that only one monomer unit is attached
to this RAFT agent during the initialization period. As Fig. 4
indicates,themajorityoftheRAFTagentisconvertedintothe
Sty-derived monoadduct, and only a very small amount of the
RAFT agent is converted into the MAh-derived monoadduct.
Therefore, it can be deduced that for this copolymerization
the addition rate of the leaving group to Sty is higher than
the addition rate of the leaving group to MAh. The second
monomer adduct, which is being formed at very low concen-
trations from early on in the polymerization, consists entirely
of one Sty and one MAh unit, as evidenced from 1H NMR
analysis and conversion determination of both monomers.
Only after the RAFT agent has been initialized does the con-
centration of the second monomer adduct start to increase
significantly. Furthermore, the third monomer adduct could
be resolved from the spectra, which consisted of an A-Sty-
MAh-Sty-D sequence, as confirmed by one-dimensional
TOCSY experiments. From this single experiment several
conclusions can be drawn. First, Du et al. previously inves-
tigated the nature of the intermediate radical by ESR studies
and concluded that it is derived from active MAh-terminal
propagating radicals.[12] From this they concluded that the
mechanism of propagation is by the CTC model, although
they were not able to detect the propagating radical itself.
Although in their experiments they used benzyl dithioben-
zoate and for our experiment we used CIPDB, the effect of
the leaving group should not change the nature of the prop-
agation reaction from the CTC model to the PU model. In
the case of the CTC model operating, two monomer units
add during one addition–fragmentation cycle, as the electron
donor (Sty) and the electron acceptor (MAh) form a CTC.
From the data obtained from the in situ 1H NMR copolymer-
ization, it is obvious that only one monomer unit is added to
the growing chain during each addition–fragmentation cycle,
which favors the PU model. Du et al. based their findings on
the appearance of the intermediate radical by ESR experi-
ments, which showed that the MAh unit was attached to the
intermediate radical and suggested that the propagating rad-
ical has a MAh terminal unit. Although only this type of
intermediate radical could be observed, it is evident from the
current experiment that chains with a Sty terminal unit can
also be attached to the RAFT-agent moiety and thus will form
a portion of the intermediate radical population in the reac-
tion. The reason that no intermediate radicals formed by the
addition of propagating chains with a Sty terminal unit could
be observed by ESR studies can be attributed to the difference
intherateoffragmentationbetweenaMAhleavinggroupand
a Sty leaving group. The ESR spectroscopy experiment only
allows easy observation of the intermediate radical species
of highest concentration. Therefore, during the course of the
polymerization, if the rate of fragmentation is substantially
higher for a chain with a Sty terminal unit, only intermediate
radicals formed by two chains with a MAh terminal unit will
be observed by ESR spectroscopy. The intermediate radicals
formed from a combination of chains that include a styrene
terminalunitwouldhaveamuchlowerlifetimeand,therefore,
a lower steady-state concentration.
Fig. 5 shows the monomer conversion for the CIPDB-
mediated copolymerization of Sty and MAh (Table 1,
experiment 1). During the initialization period, Sty monomer
0 20 40 60 80 100 120 140 160 180
0
20
40
60
80
100
Percentage
dithiobenzoates
[%]
Time [min]
AD
A-Sty-D
A-MAh-D
A-Sty-MAh-D
A-Sty-MAh-Sty-D
Fig. 4. Conversion of the initial RAFT agent and the formation of
the first, second, and third monomer adducts for a CIPDB-mediated
Sty–MAh copolymerization at 70◦C. AD represents the initial RAFT
agent CIPDB, andA-Sty-D andA-MAh-D are the first monomer adducts
in the case of a styrene monomer unit and a maleic anhydride monomer
unit addition, respectively, to the initial RAFT agent. A-Sty-MAh-D
and A-Sty-MAh-Sty-D are the second and third monomer unit adducts
formed from a styrene-derived first monomer adduct.
0 20 40 60 80 100 120 140 160 180
0
5
10
15
20
25
30
0
1
2
3
4
5
6
7
8
9
10
Conversion
[%]
Time [min]
Conv Sty
Conv MAh
Ratio Sty/MAh
Ratio
Sty/MAh
Fig. 5. Conversion versus time plot for a CIPDB-mediated Sty–MAh
copolymerization at 70◦C. The ratio of Sty to MAh consumption is
plotted versus time.

6. RESEARCH FRONT
Initialization Behaviour of Styrene–Maleic Anhydride Copolymerizations 747
is preferentially consumed and only a very small amount of
MAh is consumed. This is in agreement with Fig. 4, which
shows that the majority of the first monomer adduct consists
of a Sty-derived monoadduct. Once the copolymerization has
passed the initialization period, MAh is consumed much more
rapidly, which confirms that a MAh unit is being added to
the Sty-derived monoadduct. This is also in agreement with
copolymer reactivity ratios given in the literature.[25] When
Figs 4 and 5 are examined in more detail, it can be observed
thatStyisbeingconsumedmuchslowerandMAhmuchfaster
up to the point that the formation of the second monomer
adduct (which consists of one Sty and one MAh unit) reaches
its maximum. After this point the third monomer adduct is
formed. After the second monomer adduct has been formed,
the slopes of both the Sty and the MAh conversion versus
time become fairly similar, which means that both monomers
are being consumed at approximately the same rate. This
implies that the Sty–MAh copolymerization proceeds in an
alternating fashion that leads to a well-defined structure. The
ratio of the Sty and MAh consumption is also shown in Fig. 5,
which decreases from a maximum during the initialization
period to a value close to unity after the initialization period,
and indicates that an alternating copolymer is being formed.
A second Sty–MAh copolymerization was carried out
at 70◦C and observed by in situ 1H NMR spectroscopy,
using CDB instead of CIPDB as the RAFT agent. Again the
focus was on the initialization period of this copolymeriza-
tion and the alternating behaviour of the monomers in the
copolymerization. It appeared that the rate of addition of the
leaving group radical (cumyl radical) to MAh was so fast
that by the time the first 1H NMR spectrum was obtained, the
copolymerization had already proceeded past the initializa-
tion period. The two peaks belonging to the first monomer
adduct C-MAh-D could be observed to decrease, which indi-
cates that the second monomer adduct C-MAh-Sty-D was
already being formed. As already mentioned, the increase in
the mean rate of polymerization with an increase of the MAh
content for a Sty–MAh copolymerization has been observed
previously and, as the electronic structure of the cumyl radi-
cal is similar to that of Sty and it acts as an electron donor to
MAh, the addition rate is expected to be at least similar to that
of cross propagation in a Sty–MAh polymerization.[10,18] To
achieve greater resolution of the formation of oligomers the
same experiment was repeated at 60◦C.
Fig. 6 shows the consumption of CDB and the formation of
the first (C-MAh-D) and second (C-MAh-Sty-D) monomer
adducts for the CDB-mediated Sty–MAh copolymerization
at 60◦C, as observed by in situ 1H NMR spectroscopy. An
initialization period was observed for the CDB-mediated
Sty–MAh copolymerization, which took ∼40 min to com-
plete. The first monomer adduct consists almost entirely of
single MAh adducts, forming C-MAh-D, and no peaks were
observed for a Sty addition to the initial RAFT agent. From
Fig. 6 it can be seen that the addition rate of the cumyl leav-
ing group to a MAh unit is much faster than to a Sty unit.
Close to the end of the initialization period, the formation of
the second monomer adduct starts to increase in frequency,
whichconsistsentirelyofaStyunitthatformsC-MAh-Sty-D.
This can also be observed from Fig. 7, which shows the
conversion–time plot for a CDB-mediated Sty–MAh copoly-
merization.The preference of CDB to add a MAh unit can be
attributed to the leaving group. As the leaving group rad-
ical is nucleophilic, it acts as an electron donor to MAh.
The 2-cyanoprop-2-yl radical fragmented from CIPDB, on
the other hand, is electrophilic and therefore adds to Sty,
which acts as an electron donor. Initially, the conversion
of Sty remains very low and only once the majority of the
RAFTagenthasbeenconvertedintothefirstmonomeradduct
C-MAh-D does the conversion of Sty start to increase to form
the second monomer adduct C-MAh-Sty-D. The ratio of the
Sty to MAh consumption shows a minimum just before the
end of the initialization period, which indicates that the first
monomer adduct consists mainly of C-MAh-D and only after
initialization is the Sty consumed more rapidly. During the
addition of the second monomer unit the ratio of the Sty to
MAh consumption increases to form the second monomer
0 20 40 60 80 100 120 140 160 180
0
20
40
60
80
100
Percentage
dithiobenzoates
[%]
Time [min]
CD
C-MAh-D
C-MAh-Sty-D
Fig. 6. Conversion of the initial RAFT agent and the formation of
the first and second monomer adducts for a CDB-mediated Sty–MAh
copolymerization at 60◦C. CD represents the initial RAFT agent CDB,
and C-MAh-D and C-MAh-Sty-D are the first and second monomer
adducts, respectively.
0 20 40 60 80 100 120 140 160 180
0
5
10
15
20
25
30
35
40
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Conversion
[%]
Time [min]
Conv Sty
Conv MAh
Ratio Sty/MAh
Ratio
Sty/MAh
Fig. 7. Conversion versus time plot for a CDB-mediated Sty–MAh
copolymerization at 60◦C. The ratio of Sty to MAh consumption is
plotted versus time.

7. RESEARCH FRONT
748 E. T. A. van den Dungen et al.
adduct C-MAh-Sty-D and tends towards equality. This is
in agreement with alternating behaviour for the Sty–MAh
copolymerization.
Conclusions
The current experiments show that an initialization period
occurs for RAFT-mediated copolymerizations of Sty and
MAh during which the initial RAFT agent, CIPDB or CDB,
is converted into a first monomer adduct. This first monomer
adduct consists of the initial RAFT agent, which selectively
adds a preferred monomer unit, either Sty or MAh. CIPDB
predominantly adds one Sty unit during the initialization
period whereas CDB mainly adds one MAh unit. The prefer-
ence of the cumyl leaving group to add to MAh is determined
by its electronic structure, which acts as an electron donor
for MAh. During the early stages of the copolymerization, it
is observed that for each addition–fragmentation cycle only
one monomer unit is added to the growing chain.This implies
that the Sty–MAh copolymerizations correspond to the PU
model not the CTC model, which requires that the electron
donor and acceptor add as a couple to the growing chain.
Moreover, as both monomers are being consumed at a similar
rate shortly after initialization, the Sty–MAh RAFT-mediated
copolymerization proceeds in an alternating fashion. This
study shows that the choice of RAFT-agent leaving group
is critical for alternating copolymers prepared via living rad-
ical polymerization. Correct choice of leaving group leads
to faster initialization rates and determines which monomer
will be found at the terminus of the chain.[26]
Acknowledgments
We thank the NMR laboratories of the University of Stellen-
bosch for the generous instrument time, the Dutch Polymer
Institute for the financial support of the project, and the
National Research Foundation of South Africa.
References
[1] G. Moad, E. Rizzardo, D. H. Solomon, Macromolecules 1982, 15,
909. doi:10.1021/MA00231A042
[2] J.-S. Wang, K. Matyjaszewski, J. Am. Chem. Soc. 1995, 117,
5614. doi:10.1021/JA00125A035
[3] T. P. Le, G. Moad, E. Rizzardo, S. H. Thang, PCT Int. Appl.
WO98/01478 1998.
[4] J. Chiefari, Y. K. B. Chong, F. Ercole, J. Krstina, J. Jeffery,
T. P. T. Le, R. T. A. Mayadunne, G. F. Meijs, C. L. Moad,
G. Moad, E. Rizzardo, S. H. Thang, Macromolecules 1998, 31,
5559. doi:10.1021/MA9804951
[5] G. Moad, E. Rizzardo, S. H. Thang, Aust. J. Chem. 2005, 58, 379.
doi:10.1071/CH05072
[6] C. Barner-Kowollik, M. Buback, B. Charleux, M. L. Coote,
M. Drache, T. Fukuda, A. Goto, B. Klumperman, A. B. Lowe,
J. B. McLeary, G. Moad, M. J. Monteiro, R. D. Sanderson,
M. P. Tonge, P. Vana, J. Polym. Sci., Part A: Polym. Chem. 2006,
44, 5809. doi:10.1002/POLA.21589
[7] D. Benoit, C. J. Hawker, E. E. Huang, Z. Lin, T. P. Russell,
Macromolecules 2000, 33, 1505. doi:10.1021/MA991721P
[8] H. de Brouwer, M. A. J. Schellekens, B. Klumperman,
M. J. Monteiro, A. L. German, J. Polym. Sci., Part A: Polym.
Chem. 2000, 38, 3596. doi:10.1002/1099-0518(20001001)38:
19<3596::AID-POLA150>3.0.CO;2-F
[9] Y.-Z. You, C.-Y. Hong, C.-Y. Pan, Eur. Polym. J. 2002, 38, 1289.
doi:10.1016/S0014-3057(01)00309-3
[10] E. Chernikova, P. Terpugova, C. Bui, B. Charleux, Polymer
2003, 44, 4101. doi:10.1016/S0032-3861(03)00397-5
[11] M. C. Davies, J. V
. Dawkins, D. J. Hourston, Polymer 2005, 46,
1739. doi:10.1016/J.POLYMER.2004.12.037
[12] F.-S. Du, M.-Q. Zhu, H.-Q. Guo, Z.-C. Li, F.-M. Li,
M. Kamachi, A. Kajiwara, Macromolecules 2002, 35, 6739.
doi:10.1021/MA0202179
[13] J. Brandrup, E. H. Immergut, E. A. Grulke, Polymer Handbook
1999 (John Wiley: New York, NY).
[14] E.-S. Park, M.-N. Kim, I.-M. Lee, H. S. Lee, J.-S. Yoon,
J. Polym. Sci., PartA: Polym. Chem. 2000, 38, 2239. doi:10.1002/
(SICI)1099-0518(20000615)38:12<2239::AID-POLA120>3.0.
CO;2-U
[15] M.-Q. Zhu, L.-H. Wei, M. Li, L. Jiang, F.-S. Du, Z.-C. Li,
F.-M. Li, Chem. Commun. 2001, 365. doi:10.1039/B009815I
[16] S. Harrisson, K. L. Wooley, Chem. Commun. 2005, 3259.
doi:10.1039/B504313A
[17] R.A. Sanayei, K. F. O’Driscoll, B. Klumperman, Macromolecules
1994, 27, 5577. doi:10.1021/MA00098A010
[18] H. K. Hall, Jr, A. Buyle Padias, J. Polym. Sci., Part A: Polym.
Chem. 2001, 39, 2069. doi:10.1002/POLA.1183
[19] D. J. T. Hill, J. H. O’Donnell, P. W. O’Sullivan, Macromolecules
1985, 18, 9. doi:10.1021/MA00143A002
[20] N.-T. H. Ha, K. Fujimori, Acta Polym. 1998, 49, 404.
doi:10.1002/(SICI)1521-4044(199808)49:8<404::AID-APOL
404>3.0.CO;2-5
[21] A. L. Van Geet, Anal. Chem. 1968, 40, 2227. doi:10.1021/
AC50158A064
[22] J. B. McLeary, F. M. Calitz, J. M. McKenzie, M. P. Tonge,
R. D. Sanderson, B. Klumperman, Macromolecules 2004, 37,
2383. doi:10.1021/MA035478C
[23] J. B. McLeary, F. M. Calitz, J. M. McKenzie, M. P. Tonge,
R. D. Sanderson, B. Klumperman, Macromolecules 2005, 38,
3151. doi:10.1021/MA047696R
[24] J. B. McLeary, J. McKenzie, M. Tonge, R. Sanderson,
B. Klumperman, Chem. Commun 2004, 1950. doi:10.1039/
B404857A
[25] B. Klumperman, G. Vonk, Eur. Polym. J. 1994, 30, 955.
doi:10.1016/0014-3057(94)90030-2
[26] B. Klumperman, J. B. McLeary, E. T. A. van den Dungen,
W.-J. Soer, J. S. Bozovic, in Progress in Controlled/Living Radical
Polymerization, ACS Symp. Ser. 944 (Ed. K. Matyjaszewski)
2006, p. 501 (ACS: Washington, DC).