A novel way to synthesize star polymers in one pot by ATRP of N-[2-(2-bromoisobutyryloxy)ethyl]maleimide and styrene

Article abstract of DOI:10.1021/ma034542n

A one-pot approach to synthesize star polymers by atom transfer radical polymerization (ATRP) of N-[2-(2-bromoisobutyryloxy)ethyl]maleimide (BiBEMI) with a large excess of styrene (St) was described. It was based on preferential consumption of BiBEMI, as an inimer, through its copolymerization with St, to form a branched intermediate in situ as the multifunctional core, which initiated homopolymerization of the excessive St to produce a star polymer. The kinetic studies exhibited two polymerization stages corresponding to the formation of the core with a faster propagating rate and the formation of arms by homopolymerization of St, respectively. ¹H NMR spectra showed that in core formation stage random copolymer was formed. Analysis of the basic hydrolyzed products of the core by MALDI-TOF mass spectroscopy confirmed the branched structure of the core. A 6 -shaped polystyrene was also formed simultaneously, and its structure was confirmed by MALDI-TOF mass spectroscopy. Lowering the reaction temperature and using less excessive St could decrease the content of this polymer. Star polymers were characterized by ¹H NMR, hydrolysis, and intrinsic viscosity.

Full text of DOI:10.1021/ma034542n

18
Macromolecules 2004, 37, 18-26
A Novel Way To Synthesize Star Polymers in One Pot by ATRP of
N-[2-(2-Bromoisobutyryloxy)ethyl]maleimide and Styrene
Gu oh u a Den g a n d Yon gm in g Ch en *
State Key Laboratory of Polymer Physics and Chemistry, J oint Laboratory of Polymer Science and
Materials, Institute of Chemistry, The Chinese Academy of Sciences, Beijing 100080, P. R. China
Received April 27, 2003; Revised Manuscript Received October 27, 2003
ABSTRACT: A one-pot approach to synthesize star polymers by atom transfer radical polymerization
(ATRP) of N-[2-(2-bromoisobutyryloxy)ethyl]maleimide (BiBEMI) with a large excess of styrene (St) was
described. It was based on preferential consumption of BiBEMI, as an inimer, through its copolymerization
with St, to form a branched intermediate in situ as the multifunctional core, which initiated homopo-
lymerization of the excessive St to produce a star polymer. The kinetic studies exhibited two polymerization
stages corresponding to the formation of the core with a faster propagating rate and the formation of
arms by homopolymerization of St, respectively. 1H NMR spectra showed that in core formation stage
random copolymer was formed. Analysis of the basic hydrolyzed products of the core by MALDI-TOF
mass spectroscopy confirmed the branched structure of the core. A “6”-shaped polystyrene was also formed
simultaneously, and its structure was confirmed by MALDI-TOF mass spectroscopy. Lowering the
reaction temperature and using less excessive St could decrease the content of this polymer. Star polymers
1
were characterized by H NMR, hydrolysis, and intrinsic viscosity.
In tr od u ction
For “core first” approach, a multifunctional initiator
as a core was synthesized first and then followed by
polymerization of monomers to form the arms.5 Accord-
ingly, the number of arms can be controlled by the
number of functionalities, and a well-defined star
polymer with narrow molecular weight distribution can
be obtained. However, in many cases the multifunc-
tional initiators must be presynthesized, and limited
studies6b,10 employing ionic polymerizations have been
documented using this method due to the poor solubility
of multiply charged species needed to initiate ionic
polymerizations.
Star polymer is a class of the simplest branched
polymer that consists of multiple linear chains linked
to a central core.1 Because of its branched structure, star
polymer often exhibits lower solution and melt viscosi-
ties compared to those of linear analogues of the same
molar mass,2 a characteristic that is very important for
the applications, especially in fabrication processes of
polymer materials. Star polymer also contains a higher
degree of end-group functionalities, and this is quite
important in many specialized applications. Because of
these novel properties, star polymer has attracted
considerable attention in recent years.
Although a variety of star polymers have been pre-
pared by living ionic procedures,5,6b the rigorous reaction
conditions and monomer selections limit the applicabil-
ity of this strategy. In contrast, controlled radical
polymerization (CRP) has become a powerful tool
in preparation of well-defined star polymers during
the latest few years because it can overcome these
limitations.11-14 CRPs are often referred to atom trans-
fer radical polymerization (ATRP)11 or metal-catalyzed
living-radical polymerization,12 nitroxide-mediated
polymerization (NMP),13 and reversible addition-
fragmentation chain transfer (RAFT)14 polymerization.
Using the arm first approach, star polymers were
prepared by CRP,15 but many reaction parameters may
influence the structure of the resulting star polymers.
This deficiency has been overcome by high-throughput
combinatorial techniques in NMP.16 The core first
approach is the most frequently used method in the
preparation of star polymers by CRP. Various families
of initiators with defined number of functionalities were
designed to serve as the central cores, and stars of
polystyrene, poly(alkyl methacrylate), and poly(alkyl
acrylate) as well as block copolymers were prepared.
Several groups, for example, Matyjaszewski,11,17 Sawa-
moto,12,18 Hawker,13,19 Miller,20 Gnanou,21 Haddleton,22
Pan,23 and Rizzardo,24 have contributed much to this
area. No matter which approach was used in prepara-
tion of star polymers in the literature, either the core
The preparation of well-defined star polymers was
first reported by applying anionic living polymerization
of vinyl monomers.3 Cationic living polymerization was
also used successfully in preparing star polymers since
it was reported by Kennedy and co-workers.4 Several
recent reviews have summarized comprehensively on
the syntheses of star polymers using anionic living
polymerization5 and cationic living polymerizations6.
Employing ionic polymerizations, preparation of star
polymers documented in the literature can be catego-
rized into two approaches, namely the “arms first” and
the “core first” methods.
In the “arms first” approach, the living linear arms
of the star polymer were prepared first and then
followed by polymerization of the arms with a small
amount of a suitable multifunctional monomer, such as
divinylbenzene7 or ethylene glycol dimethacrylate,8 to
form a cross-linked microgel core and give the star
polymer. It could also be achieved by coupling reaction
of the reactive linear arm with a multifunctional linking
agent with a defined number of quenching function,9
whereas sometimes long reaction time was required and
fractionation was needed to remove unreacted linear
polymer to obtain pure star polymer.5
* Corresponding author: phone +0086-10-62659906; Fax +0086-
10-62559373; e-mail ymchen@iccas.ac.cn.
10.1021/ma034542n CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/13/2003

Macromolecules, Vol. 37, No. 1, 2004
ATRP of BiBEMI 19
Sch em e 1
light scattering measurements, the sample solution was
centrifuged with a velocity of 12 000 rpm and then filtered
using filters with a pore size of 0.2 µm. The refractive index
increment (dn/dc) of polySt was used as that of the star
polymer because the content of BiBEMI unit in the polymer
is not more than 4% calculated from the conversion of St.
Intrinsic viscosity was measured by an Ubbelohde-type capil-
lary viscometer in toluene at 35 °C.
or the arms should be synthesized in a separate step;
at least chemicals were added step by step. That is to
say that two steps are needed to get a star polymer. It
may bring problems in industrial preparation. There-
fore, a simple way of synthesis is highly appreciated.
This paper reports a novel one-pot approach to
prepare star polymers by ATRP. The strategy is based
on ATRP of BiBEMI and St (Scheme 1). When excess
St is used, preferentially copolymerization of BiBEMI,
as an inimer, with St may be expected through charge-
transfer complex (CTC)25 and give an intermediate,
which is supposed to be a branched structure, by self-
condensing of vinyl polymerization (SCVP).26 Once the
inimer is consumed, a multifunctional core is formed
in situ, and grafting polymerization of the excessive St
from the core as an initiator may proceed to form a star
polymer. When this paper was under submission,
another example of one-pot strategy was reported by
Ishizu and co-workers using a system of free radical
photopolymerization applying a similar principle.27 In
their system, polymerization of excessive methyl meth-
acrylate, inimer (N,N-diethyldithiocarbamyl)methyl-
styrene, and 4,4′-bismaleimidediphenylmethane in ben-
zene under UV irradiation was studied.
Syn th esis of N-(2-Hyd r oxyeth yl)m a leim id e. AEMI (9.15
g, 50 mmol) and p-toluenesulfonic acid hydrate (0.915 g, 4.8
mmol) were dissolved in a 180 mL mixture of methanol and
water (volume ratio, 5:1). The solution was allowed to reflux
for 72 h, and then methanol and water were evaporated. The
crude product was purified by recrystallization in methanol
to give the title compound as a white crystal: 5.71 g (yield,
81%); mp 71.5-72.5 °C (lit. mp 70-71 °C).28 1H NMR
(CDCl3): δ (ppm): 2.02 (s, 3H, -CO-CH3), 3.80 (t, 2H, )N-
CH2-), 4.22 (t, 2H, -CH2-O-), 6.73 (s, 2H, -CHdCH-).
Syn th esis of BiBEMI. 2-Bromoisobutyryl bromide (3.4 mL,
27.6 mmol) was added dropwise to the mixture of N-(2-
hydroxyethyl)maleimide (3.24 g, 23 mmol) and dry triethyl-
amine (5 mL, 34 mmol) in 60 mL of THF at 0 °C for 1 h. The
mixture was stirred for 4 h at 0 °C and at room temperature
for 38 h. The triethylammonium bromide was removed by
filtration, and the solvent was removed by rotary evaporation.
The crude product was purified by column chromatography
using 1:5 (volume ratio) ethyl acetate/petroleum ether as
eluent to obtain the title compound: 4.58 g (yield, 69%); mp
64.5-65.5 °C. 1H NMR (CDCl3): δ (ppm): 1.89 (s, 6H, -C(Br)-
CH3), 3.86 (t, 2H, )N-CH2-), 4.34 (t, 2H, -CH2-O-), 6.74
Exp er im en ta l Section
Ma ter ia ls. N-(2-Acetoxyethyl)maleimide (AEMI), mp 78.5-
79.5 °C (lit. mp 79 °C), and CuBr were prepared according to
the literature.28,29 Styrene was stirred over CaH2 overnight and
distilled under reduced pressure. Triethylamine was dried by
KOH. Tetrahydrofuran (THF) and dioxane were distilled over
Na prior to use. Anisole was washed with 10% NaOH solution
and dried by CaCl2 and distilled before use. 2-Bromoisobutyryl
bromide (98%, Aldrich), N,N,N′,N′,N′′-pentamethyldiethylene-
triamine (PMDETA; 99%, Aldrich), and other reagents were
commercialized chemicals and used as received.
(s, 2H, -CHdCH-). Elemental analysis: Calcd for C10H12
-
BrO4N: C, 41.40; H, 4.17; Br, 27.54; N, 4.83. Found: C, 41.82;
H, 4.22; Br, 28.26; N, 4.85.
Syn th esis of Sta r P olym er s. A typical procedure is as
follows: In a 25 mL Schlenk flask, CuBr (0.058 g, 0.4 mmol)
was purged three times with N2, and then deoxygenated
PMDETA (0.084 mL, 0.4 mmol) was added. Afterward, a
deoxygenated mixture of St (4.6 mL, 40 mmol), BiBEMI (0.232
g, 0.8 mmol), and dioxane (2.3 mL) was transferred to the flask,
and the flask was placed in an oil bath at 80 °C with stirring.
Samples with different time intervals were taken to analyze
St conversion by GC and monitor the polymerization by GPC.
The catalyst was removed by passing through a basic alumina
column prior to GPC analysis. The samples for 1H NMR
analysis were precipitated from petroleum ether (boiling range
60-90 °C) and dried under vacuum.
Hyd r olysis of th e Sta r P olym er . A typical procedure is
as follows: The star polymer (0.3 g) in THF (40 mL) and KOH
ethanol solution (1 M, 30 mL) were mixed and refluxed for 72
h. The solution was neutralized by dilute hydrochloric acid and
then evaporated to dryness. The residue was dissolved in THF,
and insoluble material was removed by filtration.
Mea su r em en t. Monomer conversion was determined by a
Beifen model SQ204 gas chromatograph (GC) equipped with
a FID detector using a 25 m capillary column; dioxane or
anisole was used as internal standards. Gel permeation
chromatography (GPC) was performed by a set of a Waters
515 HPLC pump, a Waters 2414 differential refractometer,
and three Waters Styragel columns (HT2, HT3, and HT4)
using THF as eluent at a flow rate of 1.0 mL/min at 35 °C.
Polystyrene standards were used for the calibration. 1H NMR
spectra were recorded on a Varian 300 MHz spectrometer in
CDCl3 at room temperature. Matrix-assisted laser desorption/
ionization time-of-flight (MALDI-TOF) mass spectrometry
was performed on a Bruker Biflex III spectrometer equipped
with a 337 nm nitrogen laser. Both matrix 4-hydroxy-R-
cyanocinnamic acid (CCA) and sample were dissolved in 1:1
(v/v) acetonitrile:water with 1% trifluoroacetic acid. 0.5 µL of
this mixture solution was placed on a metal sample plate and
air-dried at ambient temperature. Mass spectra were acquired
in positive reflector mode using an acceleration voltage of 19
kV. External mass calibration was performed using a standard
peptide mixture. Spectra were obtained by setting the laser
power close to the threshold of ionization, and generally 100
pulses were acquired and averaged. Static light scattering
(SLS) experiments were carried out on a ALV DLS/SLS-5022F
instrument using a He-Ne laser (λ0 ) 632.8 nm). Prior to the
Resu lts a n d Discu ssion
P olym er iza tion Kin etics. Figure 1 shows ATRP of
BiBEMI and St in dioxane at 80 °C. The plot of ln([M]0/
[M]) against polymerization time seems to be composed
of two stages. During the period of initial 30 min, the
apparent propagation rate constant (kpapp) is obviously
larger than that of the later period. The kpapp(bulk, 80
°C) ) 3.3 × 10-4 s-1 of copolymerization of St and AEMI
estimated from kinetic studies given in the literature25a

20 Deng and Chen
Macromolecules, Vol. 37, No. 1, 2004
F igu r e 1. Kinetic plot of the copolymerization of BiBEMI with
St by ATRP. [BiBEMI]0:[St]0:[CuBr]:[PMDETA] ) 1:50:0.5:0.5,
at 80 °C in dioxane; Vdioxane:VSt ) 1:2.
F igu r e 3. GPC traces of the copolymerization of BiBEMI with
St by ATRP. [BiBEMI]0:[St]0:[CuBr]:[PMDETA] ) 1:50:0.5:0.5,
at 80 °C in dioxane; Vdioxane:VSt ) 1:2.
F igu r e 2. Kinetic plot of the copolymerization of BiBEMI with
St by ATRP. [BiBEMI]0:[St]0:[CuBr]:[PMDETA] ) 1:50:0.5:0.5,
at 100 °C in anisole; Vanisole:VSt ) 1:2.
is bigger than the kpapp(bulk, 110 °C) ) 1.6 × 10-4 s-1
of homopolymerization of St reported by Matyjaszew-
ski30 under similar ATRP conditions. Also, the propaga-
tion rate constant of alternating copolymerization by
conventional radical polymerization is much faster than
that of homopolymerization of vinyl monomers, for
example, St.31 In present study, when excess St was
present, the inimer, BiBEMI, should be preferentially
consumed by copolymerization with St at a much faster
rate, and an intermediate in the initial stage was
expected to form. This stage took less than 30 min under
present conditions. From this intermediate as a core,
homopolymerization of St would proceed at a lower rate
in a second stage.
F igu r e 4. GPC traces of the copolymerization of BiBEMI with
St by ATRP. [BiBEMI]0:[St]0:[CuBr]:[PMDETA] ) 1:50:0.5:0.5,
at 100 °C in anisole; Vanisole:VSt ) 1:2.
before and after grafting arm chains. The core, the
branched oligomer copolymers, had a broad distribution,
and the molar mass differences were not big. However,
because of the arm grafting, the molar mass differences
between the stars with different number of arms became
much larger and overlapped curves became obvious. In
the low molecular weight region, there was a mono-
modal peak, Pl, with a small Mw/Mn (<1.2). Analysis of
the samples from other polymerizations with different
conditions showed that different solvents and temper-
atures had no great effect on the contour of every curve,
except the relative area of Pl to the main composition.
Samples taken with different time intervals were
analyzed by 1H NMR spectroscopy, which are shown in
Figure 5. For comparison, a linear alternating copolymer
of BiBEMI and St was synthesized by traditional free
But the two stages were not observed when the
polymerization temperature was 100 °C in anisole.
Figure 2 shows that the plot of ln([M]0/[M]) against
polymerization time is linear. The kpapp (6.47 × 10-5 s-1
)
app
is bigger than kp
(1.51 × 10-5 s-1) calculated from
1
the second stage of the plot in Figure 1. Thus, the
polymerization under 100 °C was much faster than that
under 80 °C, and the consumption of BiBEMI might be
finished in a short time under higher temperature.
Therefore, the initial stage “disappears” from the plot,
and the observed straight line may be attributed to the
homopolymerization of St from the core. The two stages
were not observed in Ishizu’s system.27
radical polymerization using AIBN as initiator. Its H
NMR spectrum (Figure 6) gave a strong signal, a,
between δ 1.7 and 2.0 ppm due to the protons of the
-C(CH3)2Br groups. The two peaks, b and c, between δ
3.5 and 4.5 ppm are attributed to the -NCH2CH2O-
groups in the linear structure. The broad peak around
δ 5.8-7.5 ppm in Figure 6 is the signal of the phenyl
protons. In Figure 5, the strong peak in the upfield (δ
0.6-1.2 ppm, region 1) is due to -OCOC(CH3)2-C-
groups, indicating -OCOC(CH3)2Br groups have been
transformed to -OCOC(CH3)2-C- groups by initiating
ATRP.32 The peak between δ 3.0 and 4.0 ppm (region 2
in Figure 5) is attributed to the -NCH2CH2O- groups.
These differences in Figures 5 and 6 showed that the
structure of the polymer formed in the initial stage was
GPC traces of kinetic samples taken from two poly-
merizations are illustrated in Figures 3 and 4. These
figures show that the peaks of main proportion in higher
molecular direction are broad at the early stage of
copolymerization but changed to overlapped multipeaks
with the increase of polymerization time. This change
may be explained by the molecular mass difference

Macromolecules, Vol. 37, No. 1, 2004
ATRP of BiBEMI 21
Ta ble 1. In cr em en t of th e Mola r Ra tio of St to BiBEMI
Un its (r S:B) in th e P olym er d u r in g th e P olym er iza tion a
c
d
reaction time
conv of St (%)b
rS:B(th)
rS:B(NMR)
5 min
10 min
30 min
1 h
2.7
3.7
7.5
7
13
17
3.5
6.5
8.5
9.8
2 h
10 h
21.5 h
23
45
75
100
11.5
22.5
37.5
50
12.8
21.6
36.7e
a
[BiBEMI]0:[St]0:[CuBr]:[PMDETA] ) 1:50:0.5:0.5, at 80 °C in
dioxane; Vdioxane:VSt ) 1:2. Determined by GC. c Calculated from
b
d
conversion of St: rS:B(th) ) 50 × conversion of St/1. Calculating
from areas of region 4 and region 2 in Figure 5: rS:B(NMR) ) (area
of region 4/5):(area of region 2/4). e (Area of region 4/5):(area of
region 1/6).
7.5 ppm (region 4 in Figure 5) is ascribed to the phenyl
protons of St units. Along with the increase of the
reaction time and conversion of St, the shape of this
peak gradually changed to a typical shape of pure
polySt, indicating the structure of the initially formed
polymer was different from the polymer formed after-
ward. The peaks between δ 4.3 and 4.7 ppm (region 3
in Figure 5) are ascribed to the protons attached to the
ω-end group, C-CH-Br, and its intensities decrease
gradually with the time. The above observations dem-
onstrate that there were polySt chains with terminal
Br produced in ATRP. The small sharp peak at δ 6.7
ppm in the polymer given at 5 min is the protons of the
R-end vinyl groups from maleimide moieties, and this
peak is overlapped by the phenyl protons in polySt.
From these spectra, the molar ratio of St to BiBEMI
units (rS:B(NMR)) in the polymers was calculated as listed
in Table 1. The theoretical values were calculated from
the conversion of St assuming that all the BiBEMI were
consumed and contained in the polymers. The observed
F igu r e 5. 1H NMR spectra of the polymers in different
reaction times from the polymerization as in Figure 1.
rS:B(NMR) agrees well with the theoretical value rS:B(th)
.
The rS:B value reaches nearly 3 at 5 min of polymeri-
zation and increases greatly with the time. This result
demonstrates that the copolymerization of two mono-
mers in the core formation stage did not give strictly
alternating structure and the core formation was earlier
than St homopolymerization. The random copolymeri-
zation in core formation stage is not difficult to under-
stand because the feed ratio of St to BiBEMI is very
large.
F or m a tion a n d Str u ctu r e of th e Cor e. The core
of the star polymer is proposed to originate from the
copolymerization of BiBEMI, the inimer, and St in a
similar way as SCVP of AB* type monomer studied in
the literature.26b,c Its random copolymerization has been
proved in the former section since a large excess of St
was used. Two kinds of structures A* and B* (shown in
Scheme 2) could act as chain propagation points. The
possible structures of the initially formed polymers may
be categorized roughly into three types, as shown in
Scheme 2. Type a is given by vinyl propagation only at
the A* points while B* is less active, and type b is
formed when only B* acts as radical addition propaga-
tion point. These two types are linear polymers in the
extreme cases. If both A* and B* act as propagation
points, type c, the branched structure can be formed.
Of course, no matter which structure of a , b, or c is the
core, star polymer may be produced since the core bears
multifunctionalities and if polySt branches are long
F igu r e 6. 1H NMR spectrum of a linear alternating copolymer
of BiBEMI and St synthesized in chloroform at 60 °C using
AIBN as initiator (Mn ) 2.2 × 105, Mw/Mn ) 2.2 by GPC). Its
alternating structure was proved by elemental analysis results.
The molar ratio of the two monomer units in the copolymer-
ization products at low conversions was always close to 1:1
when the molar feed of BiBEMI:St changed from 1:3, 1:2, 1:1,
2:1, to 3:1.
different from the linear alternating polymer, revealing
that the main structure of the initially formed polymer
was unlikely to be a linear alternating copolymer. The
proportion of -OCOC(CH3)2-C- structure is 86% cal-
culated from [(area of region 1)/6]/[(area of region 2)/4],
and that of -OCOC(CH3)2Br is deduced to be 14%. That
is to say, at the polymerization time of 5 min, 86% of
the -OCOC(CH3)2Br groups were transformed to the
-OCOC(CH3)2-C- structure, and 14% still existed in
the polymer. The broad and strong peak around δ 6.0-
1
enough. As discussed above, H NMR spectra (Figure
5) show that the polymer formed in the initial stage is

22 Deng and Chen
Macromolecules, Vol. 37, No. 1, 2004
Sch em e 2
Ta ble 2. m /z Va lu es of Ma in P ea k s in th e MALDI-TOF Ma ss Sp ectr u m (F igu r e 7) a n d Th eir Cor r esp on d in g Molecu la r
Str u ctu r es
quite different from the linear alternating copolymer of
BiBEMI and St synthesized by conventional radical
polymerization. So the first linear structure, a , may be
could be broken into small molecules with carboxyl acid
groups. The imide rings were opened simultaneously,
but the resulted amide bonds were difficult to undergo
further transformation to carboxyl acid.33 The end
bromine connected to the second carbon atoms may be
removed by elimination of bromine hydride during the
hydrolysis process.
The mass spectrum of the hydrolysis products is
shown in Figure 7. Although the mass spectrum is
complex due to the variety of hydrolysis products, the
peaks with relatively high intensity still can be assigned
to the structures in Table 2 with a deviation less than
1 Da between the observed masses and the calculated
values, and the observed isotopic distributions also
agree very well with the simulated ones.34 The mass
series A and B with a mass interval of 104 correspond-
ing to the molar mass of St can be easily determined to
be from the original polymer formed by random copo-
1
excluded. But H NMR cannot distinguish b and c since
the signal of the protons of B* was overlapped by the
protons from the main chain. From Scheme 2, one may
note that there are ester groups inherited from the
inimer in the structure. For b, every BiBEMI unit, with
some St units, is separated by one ester group. Cutting
ester group by hydrolysis may give fragments of one
BiBEMI unit with a few St units. For c, because of
branching, hydrolysis may give the pieces of at least two
BiBEMI units with a few St units.
The polymer formed in the core forming stage at a
reaction time of 5 min was thus hydrolyzed under basic
conditions. The hydrolysis product was then analyzed
by MALDI-TOF mass spectroscopy. Under the basic
hydrolysis, the linear and branched polymer, b and c,

Macromolecules, Vol. 37, No. 1, 2004
ATRP of BiBEMI 23
F igu r e 9. Pl fraction from the polymerization product.
[BiBEMI]0:[St]0:[CuBr]:[PMDETA] ) 1:50:0.5:0.5, at 80 °C in
dioxane; Vdioxane:VSt ) 1:2; Pl: Mn ) 1500, Mw/Mn ) 1.15 by
GPC.
F igu r e 7. MALDI-TOF mass spectrum of the hydrolysis
product of the polymer from the polymerization as shown in
Figure 1 with reaction time of 5 min.
F igu r e 8. Peak molecular weight of Pl against conversion of
St in the copolymerization of BiBEMI with St by ATRP.
[BiBEMI]0:[St]0:[CuBr]:[PMDETA] ) 1:50:0.5:0.5 (O, at 80 °C
in dioxane; b, at 100 °C in anisole).
lymerization of BiBEMI and St. The actual sequence of
the BiBEMI and St units in the polymer may be
different from the listed structures, but the composition
is kept the same. Thus, series B in Table 2 originates
from branch structure of the core, structure c. Series A
may originate from b or from a segment of the branched
core. The branched core is thus demonstrated although
the actual core structure may be very complicated. The
grafting polymerization from the initiating sites, A* and
B*, of the core may generate the arms of the star
polymers.
Str u ctu r e of P l. To explore the origin of Pl, plots of
peak molecular weights (Mp) of Pl (shown in Figures 3
and 4) against conversion of St are shown in Figure 8.
It is interesting to observe that both plots can be fitted
linearly rather well. This result, also from kinetic
studies in Figures 1 and 2, reveals the controlled
polymerization nature for Pl. To identify its structure,
Pl was fractionated from the polymerization product
with reaction time of 2 h under the same polymerization
conditions as listed in Figure 1. GPC curves of the
product and Pl fraction are shown in Figure 9. Pl was
first subjected to hydrolysis under basic conditions, and
no difference was noticed as revealed by GPC analysis.
It means the main chain is stable against hydrolysis.
Both P1 and its hydrolysis products were then analyzed
by MALDI-TOF mass spectroscopy, which gave results
presented in Figure 10. From the mass spectrum of Pl
(Figure 10I), a series of peaks with a mass interval of
104 between adjacent peaks, which corresponds to the
molar mass of St, are observed. The series of peaks may
F igu r e 10. MALDI-TOF mass spectra of fraction Pl (Mn
1500, Mw/Mn ) 1.15) and its hydrolysis product.
)
be attributed to structure a or b shown in Scheme 3
since they have the same molecular weights. The
calculated molar masses from the formulas agree very
well with the experimental ones, and the elimination
of HBr is also observed.21c,35 The simulated and the
observed isotopic distributions also agree very well for
each main peak. But the structure a and b cannot be
distinguished in Figure 10I. Structure a may be pro-
duced by ATRP of St with BiBEMI as an initiator while
b may be given in a similar way but its maleimide end
group copolymerized with St by tail biting of its own.
As shown in Scheme 3, hydrolysis products of a and b
are a ′ and b′, respectively, which have different molar
masses and can be distinguished by mass spectroscopy.
As shown in Figure 10II, the main series of the peaks
labeled can be clearly attributed to structure b′ in
Scheme 3, which is given by b. The attribution of other
peaks with relatively high intensity in Figure 10II could

24 Deng and Chen
Macromolecules, Vol. 37, No. 1, 2004
Sch em e 3
Ta ble 3. Ch a r a cter iza tion of th e Sta r P olym er a
be made, but no peak from structure a ′ was found as
the same case observed in the analysis of hydrolysis
products of the initially formed core in the former
section. These results show that the structure of Pl
corresponds to structure b. This “6”-shaped polymer is
supposed to be formed during the core forming stage.
In early stage of ATRP in large excess of St, BiBEMI
might initiate the polymerization of St to give a tele-
chelic polySt oligomer besides the SCVP of BiBEMI and
St. If more BiBEMI and St were added to the polymer,
the core for star would be produced. However, in the
case of the radical attacked the maleimide of its own in
the end, cyclic ring was formed and further initiated
polymerization of St which would give a “6”-shaped
polymer.
Mw,GPC
(PDI)
Mw,arm
(PDI)
g′
Mw,SLS
arm no.
12
[η]
(mL/g)b
43000 (1.35) 61170 5120 (1.12)
0.077
0.24
a
Fractionated from the product of the polymerization under the
similar conditions as shown in Figure 1 with reaction time of 21.5
b
0.71
h. g′ ) [η]/[η]L, [η]L ) 1.26 × 10-4Mw,SLS
) 0.32.
Ch a r a cter iza tion of th e Sta r P olym er a n d Con -
d ition Op tim iza tion . To prove the structure of main
proportion with high molecular weight, the whole
product was hydrolyzed under basic conditions due to
the presence of ester groups in the star core originated
from the inimer (Scheme 2). The representative GPC
traces of the polymer and its hydrolysis product are
given in Figure 11 (lines a and a′). The GPC trace
obtained after hydrolysis is monomodal, and its poly-
dispersity is low (Mw/Mn ) 1.11). It is interesting to
notice that GPC shape and molecular weight of the
hydrolysis product were almost the same as those of Pl
fraction in the polymer synthesized under such condi-
F igu r e 11. GPC traces of polymers and their hydrolysis
product: a, polymer obtained under the same conditions as
Figure 4 for 9.5 h; a′, hydrolysis product of a, Mn ) 6880, Mw/
Mn ) 1.11; b, high molecular weight fraction from a, Mn
)
46 710; b′, hydrolysis product of b, Mn ) 6850, Mw/Mn ) 1.12;
c, low molecular weight fraction from a; c′, hydrolysis product
of c, Mn ) 6720, Mw/Mn ) 1.05.
1
tions. The results, together with the H NMR studies
in first section, suggest that the main proportion is a
star polymer with many polySt in low polydispersities
covalently bonding to a core. No shoulder peaks ob-
served also demonstrates no further branch point was
formed during the homopolymerization of St once the
core was produced. This is another proof that the cores
are formed at the initial stage in a very short time.
It is known that coupling termination from two
growing chains happens easily during the synthesis of
a star polymer by ATRP from a core initiator. There are
two types of possible coupling terminations, interstar
and intrastar. One may notice that a shoulder peak in
high molar mass region appears gradually from GPC
traces in Figures 3 and 4. Because the terminations may
couple two chains into one chain by C-C bond that is
stable against hydrolysis and double the molecular
weight, the star polymer was hydrolyzed to study the
termination. Two fractions (lines b and c, Figure 11)
with different molecular weights obtained by fraction-
ation of the polymer using cyclohexane were hydrolyzed,
respectively. GPC traces of their hydrolysis products are
b′ and c′, respectively, in Figure 11. Although given from
different fractions, the hydrolysis products almost had
the same Mn and Mw/Mn. A small leading hump was
observed on the GPC curve (line b′) of the hydrolysis
product from the high molecular weight fraction, while
there was no such hump on that (line c′) of the low
molecular weight fraction. The average molecular weight
of the small leading hump was approximately twice of
the main peak value. The results demonstrated that
there was small extent of coupling termination during
the polymerization at high St conversion.
To elucidate the arm number and the compact nature
of the star polymer, one sample was fractionated from
the product of the polymerization. Its molecular weights
and intrinsic viscosity, [η], were determined as shown
in Table 3. Its intrinsic viscosity, [η], was smaller than
[η]L of linear polySt having the same molecular weight
calculated by the equation [η]L ) 1.26 × 10-4Mw,SLS
,
0.71
and the g′ value ([η]/[η]L ) 0.24) reveals its compact

Macromolecules, Vol. 37, No. 1, 2004
Ta ble 4. Effect of Rea ction Con d ition s on th e P olym er iza tion P r od u cts
ATRP of BiBEMI 25
a
a
entry
[BiBEMI]0:[St]0:[CuBr]:[PMDETA]
T (°C)
time (h)
solvent
Mn,star
Mn,Pl
Pl content (%)b
1
2
3
4
5
6
7
8
9
1:50:0.5:0.5
1:50:0.5:0.5
1:50:1:1
80
45
23
45
45
100
23
45
23
80
7
30
72
48
48
3.5
72
19
72
2.5
dioxanec
dioxanec
dioxanec
dioxaned
dioxaned
anisolec
DMEc,e
DMEc
9830
8740
7670
18720
5170
13720
7160
2190
1610
1510
2620
910
3170
1450
840
29.8
19.2
18.4
14.8
15.0
31.6
20.7
21.4
15.0
34.6
1:20:1:1
1:20:0.5:0.5
1:50:0.5:0.5
1:50:1:1
1:50:0.5:0.5
1:50:1:1
4260
bulk
bulk
14440
2250
10
1:100:0.5:0.5
12630f
2920f
a
b
Determined by GPC. Obtained from GPC curves by comparing the area of the whole curve with that of Pl. c Solvent volume:St
volume ) 1:2. Solvent volume:St volume ) 1.3:1. e 1,2-Dimethoxyethane. f Peak value.
d
nature.36 The star branched structure was further
proven. Using the weight-average molecular weight of
its hydrolysis product as the molecular weight of the
arm, the average arm number is estimated to be 12 by
comparing the absolute molecular weight Mw,SLS of the
star polymer and the molecular weight of the arm,
MALDI-TOF mass spectroscopy confirms the branched
structure of the core. The star polymer bearing many
arms with nearly the same length has been confirmed
by hydrolysis studies. The main proportion of the
reaction product is further proved to be a star polymer
by comparing its intrinsic viscosity with that of linear
polymer having the same molecular weight. The Pl
fraction with small proportion, which coexisted with the
star polymers, was found to be a “6”-shaped polySt.
Lowering the reaction temperature and using less
excessive St may get products with less Pl fraction.
Although the star polymers prepared by the present
developed way have relatively broad distributions, a
simple approach in one pot is its main advantage and
should be important in terms of practical application.
The application of this one-pot approach to new systems
is currently under investigation in order to get a better
controlled star polymer and generalize this principle to
other systems.
Mw,arm
.
Obviously, the polymer prepared by the present one-
pot approach was a mixture of star polymer and Pl,
although the content of Pl was low. To optimize the
polymerization conditions and obtain a product with less
Pl fraction, polymerizations were carrier out under
various conditions. Table 4 shows that lowering the
reaction temperature and using less excessive St can
bring the expected results. Keeping the feed ratio of
BiBEMI and St, the content of Pl fraction decreases
from 29.8% to 19.2% to 18.4% when the polymerization
temperature is lowered from 80 to 45 to 23 °C (see
entries 1, 2, and 3 in Table 4). Comparing entries 3, 7,
and 9, one can see that dioxane and 1,2-dimethoxy-
ethane, as polar solvents, may facilitate the cyclization.
The highest content of Pl fraction in entry 10 is due to
the largest feed ratio of BiBEMI to St (1:100) and high
reaction temperature (80 °C). When the feed ratio of
BiBEMI to St changed from 1:50 to 1:20 (entries 2 and
5 in Table 4), the content of Pl fraction decreases from
19.2% to 15%. More feed of the catalyst seems have no
apparent effect on the composition of the polymer
(entries 4 and 5 in Table 4). The above results reveal
that temperature, solvent, and feed ratio have influ-
ences on the ratio of P1 in the product. Such effects
should play roles in the core formation stage. The
radical in PS oligomer bearing maleimide end in early
stage has two choices. One is to initiate the polymeri-
zation of the maleimide of another oligomer to give a
potential branched core, i.e., intermolecular propaga-
tion. Another is to initiate the polymerization of the
maleimide of its own in ω-end to form a cycle, i.e.,
intramolecular interaction. The less polar solvent and
lower temperature may favor intermolecular propaga-
tion to produce the core of a star instead of intramo-
lecular interaction to give a molecular cycle.
Ack n ow led gm en t. The authors thank the National
Natural Science Foundation of China (20374052), the
Bairen Program and Directional Project of the Chinese
Academy of Sciences, and the project sponsored by SRF
for ROCS, SEM for support.
Refer en ces a n d Notes
(1) Roovers, J . In Encyclopedia of Polymer Science and Engineer-
ing, 2nd ed.; Kroschwitz, J . I., Ed.; Wiley-Intescience: New
York, 1985; Vol. 2, p 478.
(2) (a) Huber, K.; Bantle, S.; Burchard, W.; Fetters, L. J .
Macromolecules 1986, 19, 1404. (b) Mishra, M. K.; Kobayashi,
S. Star and Hyperbranched Polymers; Marcel Dekker: New
York, 1999.
(3) Morton, M.; Helminiak, T. E.; Gadkary, S. D.; Bueche, F. J .
Polym. Sci. 1962, 57, 471.
(4) Mishra, M. K.; Wang, B.; Kennedy, J . P. Polym. Bull. (Berlin)
1987, 17, 307.
(5) Hadjichristidis, N.; Pitsikalis, M.; Pispas, S.; Iatrou, H. Chem.
Rev. 2001, 101, 3747.
(6) (a) Charleux, B.; Faust, R. Adv. Polym. Sci. 1999, 142, 1. (b)
Kennedy, J . P.; J acob, S. Acc. Chem. Res. 1998, 31, 835.
(7) (a) Tsitsilianis, C.; Graff, S.; Rempp, P. Eur. Polym. J . 1991,
27, 243. (b) Bi, L.-K.; Fetters, L. J . Macromolecules 1976, 9,
732.
(8) Efstratiadis, V.; Tselikas, Y.; Hadjichristidis, N.; Li, J .;
Yunan, W.; Mays, J . W. Polym. Int. 1994, 33, 171.
(9) Hsieh, H. L.; Quirk, R. P. Anionic Polymerization. Principles
and Practical Applications; Marcel Dekker: New York, 1996.
(10) (a) Shohi, H.; Sawamoto, M.; Higashimura, T. Makromol.
Chem. 1992, 193, 2027. (b) Sawamoto, M. In Cationic
Polymerizations; Matyjaszewski, K., Ed.; Marcel Dekker:
New York, 1996; p 381. (c) Cloutet, E.; Fillaut, J .; Astruc,
D.; Gnanou, Y. Macromolecules 1998, 31, 6748. (d) Quirk, R.
P.; Tsai, Y. Macromolecules 1998, 31, 8016.
Con clu sion
The one-pot approach to prepare star polymers based
on ATRP of BiBEMI with a large excessive amount of
St is presented. An intermediate as a multifunctional
core is proven to be formed in situ at the initial stage of
the polymerization in a short time by preferential
consumption of BiBEMI; homopolymerization of the
excessive St from the multifunctional cores as initiator
can result in star polymers. Analysis of the hydrolysis
products of the polymer formed in the initial stage by
(11) Matyjaszewski, K.; Xia, J . H. Chem. Rev. 2001, 101, 2921.
(12) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001, 101,
3689.

26 Deng and Chen
Macromolecules, Vol. 37, No. 1, 2004
(13) Hawker, C. J .; Bosman, A. W.; Harth, E. Chem. Rev. 2001,
101, 3661.
S.; Bon, S. A. F.; J ackson, S. G.; Edmonds, R.; Kelly, E.;
Kukulj, D.; Waterson, C. Macromol. Symp. 2000, 157, 201.
(c) Haddleton, D. M.; Edmonds, R.; Heming, A. M.; Kelly, E.
J .; Kukulj, D. New J . Chem. 1999, 23, 477. (d) Haddleton, D.
M.; Waterson, C. Macromolecules 1999, 32, 8732.
(23) (a) Feng, X. S.; Pan, C. Y. Macromolecules 2002, 35, 4888.
(b) Pan, C. Y.; Tao, L.; Wu, D. C. J . Polym. Sci., Part A:
Polym. Chem. 2001, 39, 3062.
(14) Chiefari, J .; Chong, Y. K.; Ercole, F.; Kristina, J .; J effrey, J .;
Le, T. P.; Mayadunne, R. T.; Meijs, G. F.; Moad, C. L.; Moad,
G.; Rizzardo, E.; Thang, S. H. Macromolecules 1998, 31, 5559.
(15) (a) Xia, J . H.; Zhang, X.; Matyjaszewski, K. Macromolecules
1999, 32, 4482. (b) Baek, K. Y.; Kamigaito, M.; Sawamoto,
M. Macromolecules 2001, 34, 215. (c) Baek, K. Y.; Kamigaito,
M.; Sawamoto, M. J . Polym. Sci., Polym. Chem. 2002, 40,
1972. (d) Baek, K. Y.; Kamigaito, M.; Sawamoto, M. J . Polym.
Sci., Polym. Chem. 2002, 40, 2245. (e) Baek, K. Y.; Kamigaito,
M.; Sawamoto, M. J . Polym. Sci., Polym. Chem. 2002, 40,
633.
(16) (a) Bosman, A. W.; Heumann, A.; Klaerner, G.; Benoit, D.;
Fre´chet, J . M. J .; Hawker, C. J . J . Am. Chem. Soc. 2001, 123,
6461. (b) Bosman, A. W.; Vestberg, R.; Heumann, A.; Fre´chet,
J . M. J .; Hawker, C. J . J . Am. Chem. Soc. 2003, 125, 715.
(17) (a) Matyjaszewski, K.; Miller, P. J .; Fossum, E.; Nakagawa,
Y. Appl. Organomet. Chem. 1998, 12, 667. (b) Matyjaszewski,
K.; Miller, P. J .; Pyun, J .; Kickelbick, G.; Diamanti, S.
Macromolecules 1999, 32, 6526.
(24) Mayadunne, R. T. A.; J effery, J .; Moad, G.; Rizzardo, E.
Macromolecules 2003, 36, 1505.
(25) (a) Chen, G. Q.; Wu, Z. Q.; Wu, J . R.; Li, Z. C.; Li, F. M.
Macromolecules 2000, 33, 232. (b) Lutz, J .-F.; Kirci, B.;
Matyjaszewski, K. Macromolecules 2003, 36, 3136.
(26) (a) J iang, X. L.; Zhong, Y. L.; Yan, D. Y.; Yu, H.; Zhang, D.
Z. J . Appl. Polym. Sci. 2000, 78, 1992. (b) Matyjaszewski,
K.; Gaynor, S. G.; Kulfan, A.; Podwika, M. Macromolecules
1997, 30, 5192. (c) Matyjaszewski, K.; Gaynor, S. G.; Mu¨ller,
A. H. E. Macromolecules 1997, 30, 7034.
(27) Ishizu, K.; Park, J .; Shibuya, T.; Sogabe, A. Macromolecules
2003, 36, 2990.
(28) Yamada, M.; Takase, I.; Hayashi, K.; Hashimoto, Y.; Komiya,
Y. Yuki Gosei Kagaku Kyokaishi 1965, 23, 166.
(29) Keller, R. N.; Wycoff, H. D. Inorg. Synth. 1946, 2, 1.
(30) Matyjaszewski, K.; Patten, T. E.; Xia, J . H. J . Am. Chem.
Soc. 1997, 119, 674.
(31) Cowie, J . M. G. Alternating Copolymer; Plenum Press: New
York, 1985; p 48.
(18) (a) Ueda, J .; Kamigaito, M.; Sawamoto, M. Macromolecules
1998, 31, 6762. (b) Ueda, J .; Matsuyama, M.; Kamigaito, M.;
Sawamoto, M. Macromolecules 1998, 31, 557.
(19) Hedrick, J . L.; Trollsås, M.; Hawker, C. J .; Atthoff, B.;
Claesson, H.; Heise, A.; Miller, R. D.; Mecerreyes, D.; J e´roˆme,
R.; Dubois, P. Macromolecules 1998, 31, 8691.
(20) (a) Heise, A.; Hedrick, J . L.; Trollsås, M.; Miller, R. D.; Frank,
C. W. Macromolecules 1999, 32, 231. (b) Heise, A.; Hedrick,
J . L.; Frank, C. W.; Miller, R. D. J . Am. Chem. Soc. 1999,
121, 8647. (c) Heise, A.; Nguyen, C.; Malek, R.; Hedrick, J .
L.; Frank, C. W.; Miller, R. D. Macromolecules 2000, 33, 2346.
(21) (a) Angot, S.; Murthy, K. S.; Taton, D.; Gnanou, Y. Macro-
molecules 1998, 31, 7218. (b) Angot, S.; Murthy, K. S.; Taton,
D.; Gnanou, Y. Macromolecules 2000, 33, 7261. (c) Francis,
R.; Lepoittevin, B.; Taton, D.; Gnanou, Y. Macromolecules
2002, 35, 9001.
(32) Heise, A.; Diamanti, S.; Hedrick, J . L.; Frank, C. W.; Miller,
R. D. Macromolecules 2001, 34, 3798.
(33) (a) Mosig, J .; Gooding, C. H.; Wheeler, A. P. Ind. Eng. Chem.
Res. 1997, 36, 2163. (b) Matsubara, K.; Nakato, T.; Tomida,
M. Macromolecules 1998, 31, 1466. (c) Stephans, L. E.; Myles,
A.; Thomas, R. R. Langmuir 2000, 16, 4706.
(34) Nielen, M. W. F. Mass Spectrom. Rev. 1999, 18, 309.
(35) Nonaka, H.; Ouchi, M.; Kamigaito, M.; Sawamoto, M. Mac-
romolecules 2001, 34, 2083.
(36) Nakata, M. Makromol. Chem. 1971, 149, 99.
(22) (a) Hovestad, N. J .; van Koten, G.; Bon, S. A. F.; Haddleton,
D. M. Macromolecules 2000, 33, 4048. (b) Haddleton, D. M.;
Heming, A. M.; J arvis, A. P.; Khan, A.; Marsh, A.; Perrier,
MA034542N