Controlled radical polymerization of an acrylamide containing L-phenylalanine moiety via RAFT

Article abstract of DOI:10.1021/ma0509558

Homopolymers of a monosubstituted acrylamide having an amino acid moiety in the side chain, N-acryloyl-L-phenylalanine methyl ester (A-Phe-OMe), have been synthesized by reversible addition - fragmentation chain transfer (RAFT) polymerization. Two dithioe

Full text of DOI:10.1021/ma0509558

Macromolecules 2005, 38, 9055-9065
9055
Controlled Radical Polymerization of an Acrylamide Containing
L-Phenylalanine Moiety via RAFT
Hideharu Mori,* Kazuhiko Sutoh, and Takeshi Endo*
Department of Polymer Science and Engineering, Faculty of Engineering, Yamagata University,
4
-3-16, Jonan, Yonezawa 992-8510, Japan
Received May 7, 2005; Revised Manuscript Received August 9, 2005
ABSTRACT: Homopolymers of a monosubstituted acrylamide having an amino acid moiety in the side
chain, N-acryloyl-L-phenylalanine methyl ester (A-Phe-OMe), have been synthesized by reversible
addition-fragmentation chain transfer (RAFT) polymerization. Two dithioesters were used as chain
transfer agents: benzyl 1-pyrrolecarbodithioate (CTA 1) and benzyl dithiobenzoate (CTA 2). The controlled
character of the polymerization in the presence of CTA 1 in dioxane at 60 °C was confirmed by the
formation of narrow polydispersity products, the molecular weight controlled by the monomer/CTA molar
ratio, the linear relationship between the molecular weight and conversion, and the ability to increase
the molecular weight by a second addition of monomer. Poly(A-Phe-OMe) with controlled molecular weight,
low polydispersity, and enhanced isotacticity was also prepared by RAFT polymerization in the presence
of catalytic amounts of Lewis acid, Y(OTf)
C and in methanol-toluene mixture at 60 °C were also found to afford the polymers with narrow
molecular weight distributions.
3
. The RAFT polymerizations of A-Phe-OMe in methanol at 45
°
Introduction
biologically produced polymers. The desirable require-
ment is to develop synthetic methods to control the
sequence and composition of amino acids, chain chiral-
ity, conformation, amphiphilicity, and chain length. It
is also important to produce polymers with narrow chain
length distributions, which can assemble linear macro-
molecules into precisely defined nanostructures. Here,
we report the synthesis of a polyacrylamide with amino
acid moieties in the side chains, poly(N-acryloyl-L-
phenylalanine methyl ester) (poly(A-Phe-OMe)), with
narrow molecular weight distributions, controlled mo-
lecular weights, and tacticity (Scheme 1). A variety of
poly[(meth)acrylamide]s with amino acid moieties have
been developed to study their characteristic polymeri-
zation behavior, structures, and properties mainly based
on the functional amino acid moieties.10 Because of their
unique properties, these polymers not only are of
fundamental interest but also have many potential
Amino acids are the constitutional components of
peptides and proteins, which are able to produce highly
ordered hierarchical structures scaling from a few
nanometers to several microns. Their secondary struc-
tures (R-helix and â-sheet) and higher-order (tertiary
and quaternary) structures are assembled through
intra- and interchain associations by noncovalent forces,
such as hydrogen bonding, hydrophobic stacking, elec-
trostatic, and dipolar interactions. The primary struc-
tures of the biomacromolecules are constructed via
covalent linkage of the building blocks, and the molec-
ular information such as amino acid sequence, chain
chirality, and amphiphilicity encoded in the primary
structures determines their highly ordered structures.
Incorporation of amino acid residues into synthetic
polymers is of interest because these combinations may
create new nonbiological macromolecules with biomi-
metic structures and properties.
11
applications, such as polyelectrolytes, optically active
adsorbents,12 photochromic materials, controlled re-
13
In recent years, much interest has been devoted to
preparing mimics of these natural macromolecules and
artificial polypeptides for various biorelated applica-
1
4,15
16
lease systems,
biologically active materials, and
stimuli-responsive materials.1
7-19
The ease of monomer
synthesis, in which various amino acid-containing mono-
mers can be easily prepared by the reaction of (meth)-
acryloyl chloride with amino acids, is promising for
producing functional polymers for various applications.
However, the preparation of these polymers is based
almost entirely on conventional free radical polymeriza-
tions, which yield poorly defined polymers.
tions. An attractive approach is the self-organization of
block copolymers containing amino acids segments,1-3
which can form a variety of hierarchical structures,
depending upon the structure and property of each
segment, composition, chain length, and secondary
structures of poly(amino acid) segments. Recent devel-
opments in novel polymerization methods, such as ring-
opening polymerization4-6 and polycondensation, can
produce amino acid-based polymers with novel struc-
tures and architectures. Another direction is to produce
peptide-polymer hybrids by combining solid-phase
synthesis and controlled radical polymerization.8,9 How-
ever, there is still a large gap between the performance
and architectures of most synthetic polymers and
7
Recent developments in controlled radical polymer-
izations enable the synthesis of functional polymers
with controlled molar mass, narrow molecular weight
distribution, and well-defined architectures and func-
tionalities. The systems include atom transfer radical
2
0,21
polymerization,
nitroxide-mediated radical polym-
erization,22 and reversible addition-fragmentation chain
transfer (RAFT) polymerization.2
3,24
Among these con-
trolled radical polymerizations, RAFT has been suc-
cessfully applied for controlled polymerization of acryl-
amide derivatives, such as N,N-dimethylacrylamide,2
*
To whom correspondence should be addressed: e-mail
h.mori@yz.yamagata-u.ac.jp or tendo@yz.yamagata-u.ac.jp;
Ph +81-238-26-3765; Fax +81-238-26-3092.
5-27
1
0.1021/ma0509558 CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/28/2005

9056 Mori et al.
Macromolecules, Vol. 38, No. 22, 2005
Scheme 1
2
8-31
N-isopropylacrylamide,
and N-acryloylmorpho-
then the solution was degassed by three freeze-evacuate-
thaw cycles. After the ampule was flame-sealed under vacuum,
it was held at 60 °C for 24 h. The characteristic pale yellow
color kept constant during the polymerization. The reaction
was stopped by rapid cooling with liquid nitrogen. The reaction
mixture was precipitated in a large excess of diethyl ether and
isolated by filtration. The resulting product was freeze-dried
from dioxane and finally dried under vacuum at room tem-
perature to yield poly(A-Phe-OMe) as a white or pale yellow
powder. The polymer had Mn ) 7800 and M /M ) 1.23
3
2-34
line.
These results confirm the ability of RAFT to
produce well-controlled polymer chains from either
monosubstituted or disubstituted acrylamide deriva-
tives with narrow molecular weight distribution. Fur-
thermore, stereocontrol over radical polymerizations by
the addition of Lewis acid has recently been demon-
26,30,31
strated for RAFT polymerization of acrylamides
and methacrylamides.35
w
n
In this study, we investigated radical polymerization
of a monosubstituted acrylamide with an amino acid
moiety, N-acryloyl-L-phenylalanine methyl ester (A-Phe-
OMe), in the presence of a chain transfer agent (CTA)
under various conditions. First, the effects of the nature
of CTA, CTA/initiator molar ratio, and polymerization
temperature were studied to optimize the polymeriza-
tion conditions. The RAFT polymerization of A-Phe-OMe
using a suitable CTA was then evaluated by performing
polymerizations at different monomer/CTA molar ratios,
kinetic measurements, and a chain extension experi-
ment. Control of the structures of the amino acid-based
polymers was attempted in terms of the molecular
weight, polydispersity, and tacticity, which are crucial
for manipulating the properties and in further practical
applications.
according to GPC using polystyrene calibration. The structure
1
13
of the resulting polymer was confirmed by H and C NMR
measurements (Figures S1 and S2 in Supporting Information).
1
The monomer conversion was determined by H NMR
3
spectroscopy of the polymerization mixture in CDCl at room
temperature by integration of the monomer CdC-H peak at
around 5.7 ppm, compared with the sum of N-C-H peak
intensity of the polymer and the monomer at around 4.2-5.0
ppm. Conversion determined by this method was 93%. Ad-
ditionally, the polymer yield was gravimetrically determined
from the ether-insoluble polymer sample (yield ) 87%, 0.43
g). The resulting polymer was soluble in most organic solvents,
such as dichloromethane, acetone, dioxane, DMF, and DMSO,
and insoluble in diethyl ether, hexane, and water. For the
polymerization in the presence of a Lewis acid, the resulting
polymer was precipitated in water and then in diethyl ether.
The theoretical number-average molecular weight on conver-
sion is defined as follows:
Experimental Section
[
monomer]0
M (theor) )
× Mmonomer
kdt
×
Materials. L-Phenylalanine methyl ester hydrochloride
n
-
[
CTA] + 2f [I] (1 - e
)
(
Kanto Chemical, 98%) was used as received. 2,2′-Azobis-
0
0
(
isobutyronitrile) (AIBN, Kanto Chemical, 97%) was purified
conv + MCTA (1)
by recrystallization from methanol. Triethylamine (Kanto
Chemical, 98%) and pyrolle (Kanto Chemical, 99%) were
in which MCTA and Mmonomer are molecular weights of chain
transfer agent and monomer and [monomer] and [CTA] are
distilled from CaH
9%) was distilled from sodium wire, and toluene (Kanto
Chemical, 99.5%) was distilled before use. Yttrium trifluo-
romethanesulfonate (triflate; Y(OTf) , Aldrich, 98%), yitter-
bium trifluoromethanesulfonate (triflate; Yb(OTf) , Aldrich,
9.99%), and scandium trifluoromethanesulfonate (triflate; Sc-
OTf) , Aldrich, 99%) were dried under vacuum before use.
2
before use. 1,4-Dioxane (Kanto Chemical,
0
0
9
the initial concentrations of monomer and chain transfer
agent, respectively. The right-hand side of the denominator
accounts for radicals derived from initiator with an initial
3
3
0 d
concentration [I] at time t with a decomposition rate, k . The
9
initiator efficiency is represented by f. In an ideal RAFT
process, polymer directly derived from the initiators is mini-
mal, and thus the second term in the denominator becomes
negligible and eq 1 can be simplified to eq 2.
(
3
1
,1,1,3,3,3-Hexafluoro-2-propanol (HFIP, Wako Pure Chemical,
9%) and methanol (Wako Pure Chemical, 99.8%) were used
9
as received. Other materials were used without further
purification.
[
monomer]0
Synthesis of Chain Transfer Agent. Benzyl 1-pyrrole-
M (theor) )
× Mmonomer × conv + MCTA (2)
n
3
6
37
[CTA]0
carbodithioate (CTA 1) and benzyl dithiobenzoate (CTA 2)
were synthesized by literature procedures. CTA 1 was purified
by column chromatography on silica with n-hexane as the
eluent to afford a yellow oil. CTA 2 was purified by sublimation
to give red oil.
Chain Extension Using Poly(A-Phe-OMe) as Macro-
CTA. For the chain extension experiments, low molecular
weight poly(A-Phe-OMe) was prepared according to the below
procedure. A-Phe-OMe (0.497 g, 2.13 mmol), CTA 1 (9.9 mg,
0.042 mmol), AIBN (3.5 mg, 0.021 mmol), and dioxane (1.0
mL) were placed in an ampule, and then the solution was
degassed by three freeze-evacuate-thaw cycles. The polym-
erization was conducted at 60 °C for 3 h. Conversion of the
N-Acryloyl-L-phenylalanine Methyl Ester (A-Phe-
OMe). The monomer was prepared by the reaction of acryloyl
chloride with L-phenylalanine methyl ester hydrochloride
3
8
25
according to a method reported previously. [M]
D
) 102.2°,
) 141.7°, c
). The H and C NMR spectra of the monomer
3
8
27
c ) 0.1 g/dL, tetrahydrofuran (THF) (lit. [M]
D
1
13
1
)
1.0, CHCl
3
double bonds, as detected by H NMR spectroscopy, was 64%.
are shown in Figures S1 and S2, respectively (see Supporting
Information).
n
The resulting poly(A-Phe-OMe) had an M (as determined by
SEC) of 6400 and a polydispersity index of 1.29. The product
was purified by precipitation into diethyl ether and then
isolated by filtration. Finally, the resulting poly(A-Phe-OMe)
was freeze-dried from dioxane and dried under vacuum at
room temperature.
Polymerization. A representative example is as follows.
The monomer (0.50 g, 2.1 mmol), CTA 1 (9.9 mg, 0.042 mmol),
AIBN (3.5 mg, 0.021 mmol), and dioxane (1.0 mL) were placed
in a dry glass ampule equipped with a magnetic stir bar, and

Macromolecules, Vol. 38, No. 22, 2005
Controlled Radical Polymerization of an Acrylamide 9057
Scheme 2
Table 1. Reversible Addition-Fragmentation Chain
Transfer (RAFT) Polymerization of
N-Acryloyl-L-phenylalanine Methyl Ester (A-Phe-OMe) in
Dioxane for 24 ha
temp convc
Mw/Mnf
entry CTAb (°C) (%) Mnd (theory) Mn (SEC) (SEC)
e
1
2
3
4
5
60
60
90
60
90
100
93
97
45
97
89000
7800
9200
4000
9900
3.36
1.23
1.29
1.20
1.48
A representative example of chain extension experiment is
as follows. The dithiocarbamate-terminated poly(A-Phe-OMe)
CTA1
CTA1
CTA2
CTA2
11000
11600
5400
(
(
(
n w n
0.133 g, 0.018 mmol; M ) 6400, M /M ) 1.29), A-Phe-OMe
0.485 g, 2.1 mmol), AIBN (1.7 mg, 0.010 mmol), and dioxane
4.0 mL) were placed in a dry ampule, and then the solution
11600
a
was degassed by three freeze-evacuate-thaw cycles. The
ampule was subsequently immersed in an oil bath preheated
to 60 °C, and it was held for 24 h before being quenched by
rapid cooling with liquid nitrogen. Conversion of the double
[M]0/[CTA]0 ) 50, [CTA]0/[AIBN]0 ) 2, monomer concentration
)
0.5 g/mL ([M] ) 1.38 mol/L), where M ) acryloyl-L-phenylala-
nine methyl ester (A-Phe-OMe), AIBN ) 2,2′-azobis(isobutyroni-
b
trile). CTA 1 ) benzyl 1-pyrrolecarbodithioate, CTA 2 ) benzyl
c
1
1
dithiobenzoate (see Scheme 2). Calculated by H NMR in CDCl3.
bonds, as detected by H NMR spectroscopy, was 97%. The
d
The theoretical molecular weight (Mn,theory) ) (MW of M) × conv
n
resulting poly(A-Phe-OMe) had an M (as determined by SEC)
e
×
[M]0/[CTA]0 + (MW of CTA). Number-average molecular
of 23 000 and a polydispersity index of 1.38.
weight (Mn) was measured by size-exclusion chromatography
Instrumentation. H (270 and 500 MHz) and 13C NMR
1
(
(
SEC) using polystyrene standards in N,N-dimethylformamide
(
67.5 and 125 MHz) spectra were recorded with a JEOL EX-
f
DMF, 10 mM LiBr). Molecular weight distribution (Mw/Mn) was
2
70 and a Varian INOAVA-500. The dyad tacticity of poly(A-
Phe-OMe) was determined from the methylene proton peaks
of the polymer recorded in DMSO-d at 170 °C. Specific
rotations ([R] ) were measured on a JASCO DIP-1000 digital
measured by SEC in DMF (10 mM LiBr).
6
has been polymerized successfully via RAFT using
benzyl 1-pyrrolecarbodithioate (CTA 1).29 Benzyl dithio-
benzoate (CTA 2) is a RAFT agent that has been
employed for well-controlled polymerization of N,N-
D
polarimeter equipped with a sodium lamp as a light source.
Circular dichroism (CD) spectra were measured on a JASCO
J-720 spectropolarimeter. Number-average molecular weight
25
28
dimethylacrylamide and N-isopropylacrylamide. The
R group in CTA must be a good free radical leaving
group and efficient at reinitiating polymerization. CTA
(
n w n
M ) and molecular weight distribution (M /M ) were esti-
mated by size-exclusion chromatography (SEC) using a Tosoh
HPLC HLC-8220 system equipped with refractive index and
ultraviolet detectors at 40 °C. The column set was as follows:
three consecutive hydrophilic vinyl polymer-based gel columns
1
and CTA 2 have the same R group, which yields a
benzyl radical species upon fragmentation. As the Z
group influences strongly the stability of the dithioester
intermediate radical, strong stabilizing groups will favor
the formation of the intermediate radical and therefore
enhance the reactivity of the SdC bond toward radical
addition. However, the stability of the intermediate
needs to be finely tuned to favor the fragmentation,
which will release the reinitiating group (R).
[
TSK-GELs (bead size, exclusion limited molecular weight):
6
5
SuperAW5000 (7 µm, 4 × 10 ), SuperAW4000 (6 µm, 4 × 10 ),
4
SuperAW3000 (4 µm, 6 × 10 ), 15 cm each] and a guard column
[
TSK-guardcolumn Super AW-H, 3.5 cm]. The system was
operated at a flow rate of 0.6 mL/min, using N,N-dimethyl-
formamide (DMF) containing 10 mM LiBr as an eluent.
Polystyrene standards were employed for calibration. GPC
with a multiangle light scattering detector (GPC-MALS) was
also performed to determine the true molecular weights of the
resulting polymer. The measurement was conducted using a
Shodex GPC system 21 with two consecutive columns (Shodex
To find suitable systems for the synthesis of well-
defined polyacrylamides containing amino acid moieties
in the side chains, we initially investigated the influence
of the nature of CTA on the radical polymerization of
N-acryloyl-L-phenylalanine methyl ester (A-Phe-OMe).
A-Phe-OMe was polymerized using CTA 1 or CTA 2
under various conditions, and the results are sum-
marized in Table 1. When A-Phe-OMe was polymerized
using CTA 1 with AIBN as an initiator at [A-Phe-OMe]0/
7
KD-806M × 2, exclusion limited molecular weight ) 2 × 10 ,
3
0 cm each) and DAWN DSP-F (Wyatt Technology Co.)
detector equipped with He-Ne laser (632.8 nm). DMF con-
taining 10 mM LiBr was used as an eluent at a flow rate of
1.0 mL/min. Excess refractive index increment (dn/dc) was
measured using a differential refractometer DRM1021 at 25
C.
Thermogravimetric analysis (TGA) was performed on a
SEIKO SSC/5200 at a heating rate of 10 °C/min under N . For
°
[
CTA 1]0/[AIBN]0 ) 100/2/1 in dioxane (0.5 g/mL), which
corresponds to [M] ) 1.38 mol/L, [CTA] ) 0.0276 mol/
L, and [I] ) 0.0138 mol/L, almost full conversion (93%,
2
differential scanning calorimetry (DSC) measurements, a
SEIKO DSC/6200 apparatus was used (heating rate: 10 °C/
min; cooling rate: 10 °C/min). Samples were heated from 50
to 250 °C at a rate of 10 °C/min, kept for 5 min, and cooled at
a rate of 10 °C/min. The data collection was carried out on
the second heating process, and the glass transition temper-
ature (T ) was taken to be the midpointsthe temperature
g
corresponding to half of the endothermic shift. The calorimeter
was calibrated with an indium standard.
1
as determined by H NMR spectroscopy) was obtained
at 60 °C after 24 h. The characteristic pale yellow color
remained throughout the polymerization without sig-
nificant change in the viscosity. The resulting polymer
showed sharp symmetrical unimodal SEC peak (Mw/Mn
)
1.23) without shoulders and tailings. The number-
average molecular weight of the poly(A-Phe-OMe),
measured by a GPC in DMF with 10 mM LiBr, was Mn
)
7800, which is roughly comparable to the theoretical
Results and Discussion
value (Mn ) 11 000) calculated from the monomer/CTA
molar ratio and the monomer conversion using eq 2.
However, a conventional radical polymerization of A-
Phe-OMe under the similar condition in the absence of
CTA afforded a high molecular weight homopolymer
with high polydispersity (Mn ) 89 000 and Mw/Mn )
3.36, entry 1). The difference in the molecular weights
of the polymers obtained in the presence and absence
of CTA under the similar conditions supports the
effectiveness of the reaction conditions to achieve con-
Preliminary Comparison of CTA 1 and CTA 2.
One of the key points for the synthesis of well-defined
products via RAFT process is the design of the chain
transfer agent (CTA), as the choice of R and Z groups
depends on the monomer.36,37 In this study, we selected
two different chain transfer agents, namely benzyl
1
-pyrrolecarbodithioate (CTA 1) and benzyl dithioben-
zoate (CTA 2), as shown in Scheme 2. An important
monosubstituted acrylamide, N-isopropylacrylamide,

9058 Mori et al.
Macromolecules, Vol. 38, No. 22, 2005
trolled polymerization.39 Note that the significant in-
crease in the viscosity could not be observed visually
even at almost full conversion during the RAFT polym-
erization, which was due to relatively low molecular
weights of the resulting product (Mn ) 7800, Mw/Mn )
Table 2. Effect of Chain Transfer Agent/Initiator Molar
Ratio on Reversible Addition-Fragmentation Chain
Transfer (RAFT) Polymerization of
N-Acryloyl-L-phenylalanine Methyl Ester (A-Phe-OMe) in
a
the Presence of CTA 1 in Dioxane for 24 h
temp [CTA 1]0/ convb
Mnc
Mnd
Mw/Mne
(SEC)
1.23). Whereas, apparent viscosity increase was detected
entry
(°C)
[AIBN]0
(%)
(theory) (SEC)
in the case of the conventional radical polymerization
in the absence of CTA, in which high molecular weight
product (Mn ) 89 000, Mw/Mn ) 3.36) was produced.
1
2
3
4
5
6
7
8
60
60
60
60
60
90
90
90
1.5
2
90
93
91
94
92
97
92
91
10 700
11 100
10 800
11 200
11 000
11 500
11 000
10 800
17000
7800
9400
9500
8800
9200
9200
8500
1.35
1.23
1.25
1.27
1.23
1.29
1.23
1.24
3
The polymerization of A-Phe-OMe was also conducted
using benzyl dithiobenzoate (CTA 2) to compare the
effect of different Z groups, phenyl and pyrrole, on the
polymerization behavior. As shown in Table 1 (entry 4),
the polymerization with AIBN in the presence of CTA
5
10
2
5
10
a
[
M]0/[CTA 1]0 ) 50, monomer concentration ) 0.5 g/mL ([M]
2
at 60 °C produced a polymer with a relatively narrow
)
1.38 mol/L), where M ) acryloyl-L-phenylalanine methyl ester
molecular weight distribution (Mw/Mn ) 1.20), while
achieving only 45% conversion even after 24 h. This is
an indication that the polymerization with CTA 2 is
much slower than that with CTA 1 at 60 °C. At 90 °C,
the polymerization with CTA 2 led to increase of the
reaction rate but afforded the polymer with broad
molecular weight distribution (Mw/Mn ) 1.48). The
retardation of the polymerization rate, depending on the
structure of CTA, has been intensively discussed in the
(
A-Phe-OMe), AIBN ) 2,2′-azobis(isobutyronitrile), and CTA 1 )
b
1
benzyl 1-pyrrolecarbodithioate (see Scheme 2). Calculated by H
NMR in CDCl . c The theoretical molecular weight (Mn,theory) )
3
d
(MW of M) × conv × [M]0/[CTA]0 + (MW of CTA). Number-
average molecular weight (Mn) was measured by size-exclusion
chromatography (SEC) using polystyrene standards in N,N-
e
dimethylformamide (DMF, 10 mM LiBr). Molecular weight dis-
tribution (Mw/Mn) was measured by SEC in DMF (10 mM LiBr).
[
CTA 1]0:[AIBN]0 ratios between 1.5 and 10, keeping the
monomer-to-chain transfer agent at a constant value of
A-Phe-OMe]0/[CTA 1]0 ) 50/1. As shown in Table 2, the
3
9-42
literature.
The rate retardation effect is attributed
to the main equilibrium after all initial CTA have been
transformed into macro-CTA (so-called macro-RAFT
agent) and may therefore be ascribed to different
stabilities of the intermediate radicals in the main
equilibrium. It has frequently been observed that
dithiobenzoates show pronounced retardation phenom-
ena due to the significant stabilization effect of phenyl
group. The transfer constant is another important
parameter in the RAFT process. The transfer constants
of a series of benzyl thiocarbonylthio compounds of
general structures, Z-C(dS)-CH2Ph, in styrene po-
lymerization have been reported, and it was demon-
strated that the transfer constant with CTA 2 (Z ) Ph,
Ctr ) 26) is apparently larger than that with CTA 1 (Z
[
molecular weight obtained at [CTA]0/[AIBN]0 ) 2/1 is
slightly lower than that at [CTA]0/[AIBN]0 ) 10/1, and
the molecular weight distribution remains narrow (Mw/
Mn ) 1.23-1.27). However, low initiator concentration
(
[CTA 1]/[I] ) 1.5, entry 1 in Table 2) led to remarkable
increase in the molecular weights with broader poly-
dispersity, suggesting the loss of the controlled charac-
ter. In RAFT polymerization, the total number of chains
is determined by the number of CTAs that have suc-
cessfully fragmented and reinitiated polymerization plus
the number of initiator-derived chains. In other words,
the molecular weights of resulting polymers should be
decreased with decreasing [CTA]0/[I]0 ratio under the
same monomer-to-chain transfer agent ratio and mono-
mer conversion. In an ideal RAFT process, however, the
number of polymer chains directly derived from the
initiator molecules should be minimal, and the initial
CTA concentration is large enough compared to the
3
9
)
pyrrole, Ctr ) 9). Note that the chain transfer
constant (Ctr) is given by the ratio of the rate constant
for chain transfer to that for propagation (ktr/kp) in the
case of conventional radical polymerization, while ktr is
composed of two factors: the rate constant for addition
to the thiocarbonyl group and the partitioning of the
intermediate radical between starting materials and
products in the RAFT process. Since the leaving group
43,44
number of the initiator-derived chains.
Further, the
reactivity of the CTAs is usually substantially higher
•
than that of monomer, favoring initiation by R frag-
(
benzyl radical) of CTA 2 is the same to that of CTA 1,
ments. The consumption of CTA and reversible frag-•
the retardation with CTA 2 should be related to the
slower fragmentation rate from the phenyl-substituted
intermediate radical than that from pyrrole one and/or
fast addition rate of expelled radical to the thiocarbonyl
group. The high transfer constant for CTA 2 is reflecting
the stability of the intermediate radicals because the
value directly leads to increased addition rates. This
process, however, may mainly apply for the initial stage
mentation of intermediate to produce reinitiating R
fragment are often referred to as the “preequilibrium”.
The narrow molecular weight distribution is apparently
due to rapid establishment of the preequilibrium, ef-
•
ficient reinitiation from the R fragment, and attainment
of the so-called “main equilibrium” in which the popula-
tion of dormant chains and/or intermediate radicals is
much higher than the total number of propagating
chains.45 In our system, the molecular weight obtained
(preequilibrium), which is related closely to the induc-
tion period discussed in the next section. From these
preliminary results, we selected CTA 1 for our further
investigations toward the precise synthesis of poly(A-
Phe-OMe)s having low polydispersity, controlled mo-
lecular weights, and tacticity. In the next stage, the
effects of the polymerization temperature and the ratio
of CTA 1 to initiator were investigated in terms of the
molecular weights and the polydispersity of the result-
ing poly(A-Phe-OMe). The polymerization of A-Phe-OMe
was conducted in dioxane at 60 °C for 24 h at different
at [CTA] /[AIBN]0 ) 10/1 is comparable to that at
0
[CTA] /[AIBN] ) 2/1, with keeping low polydispersity.
0
0
A similar tendency was also reported in another RAFT
system,44 in which no deterioration in the control of the
polymerization was observed as [CTA] /[I] decreased
0
0
from 8 to 1.5.
The polymerization at 90 °C under the same condi-
tions ([A-Phe-OMe]0/[CTA 1]0/[AIBN]0 ) 100/2/1, 24 h)
afforded the polymer with low polydispersity (Mw/Mn )
1.29) and reasonable molecular weights. Note that the

Macromolecules, Vol. 38, No. 22, 2005
Controlled Radical Polymerization of an Acrylamide 9059
substantially the same in all cases. Nevertheless, the
result suggests that the molecular weights of the poly-
(
A-Phe-OMe)s can be easily adjusted by the monomer-
to-CTA ratio. In all cases, the SEC traces are unimodal
with no evidence of high molecular weight species, as
can be seen in Figure 1b. The phenomenon, namely the
linearity of the Mn vs [A-Phe-OMe]0:[CTA 1]0 ratio with
keeping low polydispersity, was also observed for the
polymerization under nitrogen at 90 °C (Figure S3, see
Supporting Information). These results suggest that
uncontrolled polymerization and/or termination by cou-
pling of propagating polymer radicals can be negligible
under the condition used in this study. A possibility to
form a three-arm star chain reported in the RAFT
polymerization of styrene by Kwak et al.41 can be also
excluded, as judged from the SEC charts of the poly(A-
Phe-OMe)s obtained under various conditions.
In all cases, the experimental molecular weights were
slightly lower than calculated ones, as shown in Tables
1
and 2. These discrepancies may result from the
difference in hydrodynamic volume between poly(A-Phe-
OMe) and the linear polystyrene standards used for
GPC calibration. Another possible explanation is that
the number of living polymer chains is larger than that
of CTA due to an initiator-derived chain. To clarify the
point, GPC with a multiangle light scattering detector
(
GPC-MALS) was applied for the determination of the
absolute molecular weights of representative samples
Figure S4, see Supporting Information). The polymers
(
obtained at [A-Phe-OMe]0:[CTA]0 ratios ) 75 and 100
had Mw ) 23 400 and Mw/Mn ) 1.20 (Mn,GPC-MALS )
Figure 1. (a) Dependence of number-average molecular
1
9 400, as determined by GPC-MALS), compared to
0 0
weight and molecular weight distribution on [M] :[CTA 1]
ratio for the polymerization of N-acryloyl-L-phenylalanine
methyl ester (A-Phe-OMe) with 2,2′-azobis(isobutyronitrile)
Mn,calcd ) 16 000 and Mn,GPC ) 13 000 (Mw/Mn ) 1.26),
and Mw ) 30 000 and Mw/Mn ) 1.19 (Mn,GPC-MALS )
(
AIBN) in the presence of benzyl 1-pyrrolecarbodithioate (CTA
2
5 200, as determined by GPC-MALS), compared to
1
, see Scheme 2) in dioxane at 60 °C. Monomer concentration
Mn,calcd ) 20 000 and Mn,GPC ) 17 000 (Mw/Mn ) 1.25),
respectively. There was no significant difference be-
tween Mn,GPC-MALS and Mn,calcd, suggesting that the
number of living polymer chains is comparable to that
of CTA. In the SEC experiments of the two samples,
the RI detector and the UV detector monitoring the
absorption due to the phenyl groups in the polymer at
)
0 0
0.5 g/1.0 mL ([M] ) 1.38 mol/L). [CTA 1] /[AIBN] ) 2/1.
Monomer conversion ) 93-95%. (b) Size-exclusion chroma-
tography (SEC) traces of the corresponding poly(A-Phe-OMe)s
0 0
obtained at different [M] :[CTA 1] .
half-life time of AIBN at 90 °C is only 17 min; that is,
after 170 min ) 2.9 h of reaction time the initiator has
decomposed quantitatively. No significant influence on
the conversion, molecular weight, and polydispersity
was observed, as the concentration ratio of CTA to
initiator increased up to [CTA 1]0/[AIBN]0 ) 10/1
2
80 nm gave similar curves within the same elution
time range (Figure S5, see Supporting Information).
The polymerizations of A-Phe-OMe in the presence
of CTA 1 in dioxane at 60 and 90 °C were investigated
at a constant monomer/chain transfer agent/initiator
molar ratio, [A-Phe-OMe]0/[CTA 1]0/[AIBN]0 ) 200/2/1.
The time-conversion and the pseudo-first-order kinetic
plots are shown in Figure 2. The polymerization at 90
°C was very fast, in which more than 90% conversion
was reached within 0.5 h. Whereas, less than 10%
conversion could be reached in the same time period
when the polymerization was conducted at 60 °C.
Indeed, an induction period of less than 2 h is seen in
the pseudo-first-order kinetic plot at 60 °C. The induc-
tion period roughly estimated simply by extrapolating
the linear part of each curve to the time axis is about
50 min at 60 °C, while it decreased to less than 10 min
at 90 °C. An induction period is often observed in RAFT
(entries 6-8, Table 2), suggesting that the difference
in the initiator concentration had minor effect on the
total number of the polymer chains. Nevertheless, it is
hard to evaluate the number of the initiator-derived
chain existed in this system. Further investigations,
such as MALDI-TOF MS investigation, may be required
to clarify this point, which will be reported elsewhere.
Polymerization in the Presence of CTA 1. The
polymerization of A-Phe-OMe was investigated with
AIBN in the presence of CTA 1 in dioxane at 60 °C for
2
4 h at different [A-Phe-OMe]0:[CTA 1]0 ratios between
5 and 100, while the AIBN:CTA 1 molar ratio was held
2
constant at 1:2. Under the conditions, the conversions
1
were quantitative (>90%, as determined by H NMR)
in all cases. Figure 1a shows the relation of the
molecular weight and polydispersity with the [M]0:[CTA
polymerization of various monosubstituted and disub-
stituted acrylamides.2
7,29,31,32
The reasons for the induc-
1
]0 ratio. A linear increase of the number-average
molecular weight with the ratio is observed clearly, and
the molecular weight distributions remain narrow (Mw/
Mn ) 1.21-1.26), indicating a feasibility to control the
molecular weights. Note that the straight line could be
obtained only when the monomer conversions are
tion periods with some CTAs are not clearly understood,
but a number of possible explanations have been
suggested,4
6-50
including slow fragmentation of the
initiating leaving group radical, slow reinitiation by the
expelled radical, increased stability of the intermediate

9060 Mori et al.
Macromolecules, Vol. 38, No. 22, 2005
Figure 3. (a) Evolution of size-exclusion chromatography
(
SEC) traces with conversion and (b) number-average molec-
Figure 2. Conversion-time (squares) and the first-order
kinetic plots (circles) for the polymerization of N-acryloyl-L-
phenylalanine methyl ester (A-Phe-OMe) with 2,2′-azobis-
ular weight (circles) and polydispersity (squares) as a function
of conversion at 60 °C. See Figure 2 for detailed polymerization
conditions.
(
isobutyronitrile) (AIBN) in the presence of benzyl 1-pyrrole-
carbodithioate (CTA 1, see Scheme 2) in dioxane at 60 °C (a)
and 90 °C (b). Monomer concentration ) 0.5 g/1.0 mL ([M] )
0 0 0
1.38 mol/L). [A-Phe-OMe] /[CTA 1] /[AIBN] ) 200/2/1.
radical (with and without intermediate radical termina-
tion), tendency of the expelled radical to add to the CTA
rather than to monomer, and impurities in the CTA.
The SEC traces of poly(A-Phe-OMe) obtained at 60 °C
at different polymerization times clearly illustrate the
increase in molar mass with time (Figure 3a). Sym-
metrical unimodal SEC peaks without shoulders and
tailings are observed for the polymers obtained even at
higher conversion (>90%). The polydispersity indices
(
Mw/Mn) for all samples ranged between 1.15 and 1.36,
Figure 4. Size-exclusion chromatography (SEC) traces of the
regardless of the polymerization temperature. Figure
parent poly(A-Phe-OMe) macro-CTA (dotted trace, M
n
) 6400,
3
b shows the evolution of Mn and Mw/Mn with conver-
M
M
w
/M
n
) 1.29) and the chain extended polymer (solid trace,
/M ) 1.38) obtained after the polymerization
sion during the polymerization of A-Phe-OMe. The
n
) 23 000, M
w
n
linearity of the Mn vs conversion plot was observed at
with A-Phe-OMe.
6
0 °C, indicating that the polymerization proceeded in
a controlled fashion without nondegenerative chain
transfer. The linear relationship between the Mn and
the conversion with maintaining low polydispersity was
also observed at 90 °C (Figure S6, see Supporting
Information). These results suggest that the polymer-
ization of A-Phe-OMe mediated by CTA 1 shows a good
control at both temperatures, 60 and 90 °C.
[AIBN] ) 200/2/1, almost full conversion was reached
0
after 24 h. SEC traces of the starting poly(A-Phe-OMe)
macro-CTA and the second-growth polymer are shown
in Figure 4. A shift of the SEC trace toward a higher
molecular weight region clearly shows that the exten-
sion was successful. Tiny tailing can be seen, suggesting
that a small amount of residual low molecular weight
dead chains remains in the final product. Nevertheless,
the polydispersity remained below 1.4, even at almost
full conversion. These results suggest that most of the
chain ends of poly(A-Phe-OMe) are functionalized with
the dithiocarbamate end groups, which can be used as
a macro-CTA for further chain extension reactions.
Experiments aiming to synthesize well-defined block
copolymers containing poly(A-Phe-OMe) segment by
To get further insights into to the controlled/living
character of the polymerization, the poly(A-Phe-OMe)
obtained from the polymerization in the presence of CTA
1
was isolated and then used as a macro-CTA for further
reaction. The chain extension was conducted using the
poly(A-Phe-OMe) (Mn ) 6400, Mw/Mn ) 1.29) as the
macro-CTA, A-Phe-OMe as the monomer, and AIBN as
the initiator in dioxane at 60 °C. When the polymeri-
zation was conducted at [A-Phe-OMe]0/[macro-CTA]0/

Macromolecules, Vol. 38, No. 22, 2005
Controlled Radical Polymerization of an Acrylamide 9061
Table 3. Reversible Addition-Fragmentation Chain Transfer (RAFT) Polymerization of N-Acryloyl-L-phenylalanine
Methyl Ester (A-Phe-OMe) in the Presence of Lewis Acid in Dioxane at 60 °C for 6 ha
b
c
Mnd (SEC)
e
tacticity m/rf
entry
Lewis acid
[Lewis acid]/[M]
conv (%)
Mn (theory)
Mw/Mn (SEC)
1
2
3
4
5
93
93
98
98
47
23
95
99
84
21
22 000
22 000
23 500
23 500
11 000
5 600
23 000
23 500
20 000
5 100
17 000
14 000
18 500
18 000
16 900
1.25
1.35
1.53
1.55
1.80
49/51
62/38
67/33
68/32
69/31
Y(OTf)3
Y(OTf)3
Y(OTf)3
Y(OTf)3
Y(OTf)3
Yb(OTf)3
Yb(OTf)3
Sc(OTf)3
Sc(OTf)3
0.05
0.1
0.2
0.5
0.2
0.05
0.2
0.05
0.2
g
6
7
8
9
21 000
18 000
17 000
1.63
1.59
1.55
65/35
66/34
61/39
1
0
a
[
CTA 1]0/[AIBN]0 ) 2, [M]0/[CTA 1]0 ) 100, monomer concentration ) 0.5 g/mL ([M] ) 1.38 mol/L), where M ) acryloyl-L-phenylalanine
b
methyl ester (A-Phe-OMe), AIBN ) 2,2′-azobis(isobutyronitrile), and CTA 1 ) benzyl 1-pyrrolecarbodithioate (see Scheme 2). Calculated
1
c
d
by H NMR in CDCl3. The theoretical molecular weight (Mn,theory) ) (MW of M) × conv × [M]0/[CTA]0 + (MW of CTA). Number-
average molecular weight (Mn) was measured by size-exclusion chromatography (SEC) using polystyrene standards in N,N-dimethylfor-
e
f
1
mamide (DMF, 10 mM LiBr). Molecular weight distribution (Mw/Mn) was measured by SEC in DMF (10 mM LiBr). Determined by H
g
NMR in dimethyl-d6 sulfoxide (DMSO-d6) at 170 °C. Polymerization was conducted at 45 °C for 24 h.
using this method are now in progress, which will be
reported elsewhere.
RAFT Polymerization in the Presence of Lewis
Acids. Recently, Lewis acids, such as rare earth metal
trifluoromethanesulfonates (triflates), Yb(OTf)3, Y(OTf)3,
and Sc(OTf)3, have been successfully applied to attain
stereocontrol over conventional radical polymerization
5
1
52,53
of acrylamides, methacrylamides,
R-alkoxymethyl
54
55
acrylates, and methacrylates. The effect of Lewis acid
on the stereocontrol was also reported on conventional
radical polymerization of an optically active methacry-
lamide containing amino acid moiety, N-[(R)-R-meth-
5
6
oxycarbonylbenzyl]methacrylamide. The mechanism
controlling tacticity is believed to involve coordination
of the Lewis acid with the last two segments of a
growing polymer chain, which forces them into the meso
configuration during the monomer addition and leads
to isotactic polymers. A suitable choice of the nature of
Lewis acid as well as Lewis acid/monomer ratio is
crucial to attain tacticity control without harming the
narrow polydispersity and controlled molecular weights
1
6
Figure 5. H NMR (500 MHz, DMSO-d , 170 °C) spectrum
of poly(A-Phe-OMe) prepared in the absence of Lewis acid.
1
H NMR in DMSO-d6 at 170 °C. Figure 5 shows the
1
representative H NMR spectrum of poly(A-Phe-OMe)
obtained by the RAFT polymerization in the absence of
Lewis acid. The characteristic peaks at 7.4-6.8 (aro-
matic protons and amide methine proton), 4.8-4.5
26,30,31,35
of resulting polymers.
For example, Okamoto et
(
methine proton attached to chiral carbon atom), 3.4-
al. reported that a ratio of Lewis acid/monomer equal
3
.6 (methyl protons), 3.2-2.8 (methylene protons), 2.6-
to 0.1 is sufficient to achieve successful stereocontrol of
the polymerization of various acrylamide derivatives.51
2.0 (backbone methine proton), and 1.9-1.2 ppm (back-
bone methylene protons) are clearly seen. As reported
However, Matyjaszewski et al. demonstrated that a
RAFT polymerization of N,N-dimethylacrylamide ex-
hibited uncontrolled molecular weight and a relatively
broad molecular weight distribution at a ratio of Lewis
acid/monomer equal to 0.1, and the RAFT polymeriza-
26
in poly(N,N-dimethylacrylamide) and poly(N-isopro-
pylacrylamide),3
0,31
the proportion of meso and racemic
dyads and the degree of isotacticity in the polymer can
be estimated from the backbone methylene region of the
1
H NMR spectra. In the signal from the meso dyads,
tion was therefore conducted in the presence of a lower6
amount of Lewis acid (Lewis acid/monomer ) 0.05).2
the two methylene protons are not equivalent and lead
to two broad peaks of almost equal intensity, as can be
seen in Figure 6. For poly(A-Phe-OMe) prepared in this
study, these meso dyads peaks are observed at around
The stereocontrol over radical polymerization strongly
depends on the monomer structure, involving the nature
of the R-substituted group and of the coordination
functions, such as ester or amide group. In the cases of
free radical polymerization without Lewis acid, the
effect of R-substituted group is important. For acrylate
and acrylamide monomers, the probability of meso and
1
.75 and 1.35 ppm. However, the methylene protons of
racemo dyads are equivalent and the racemo peak is
seen at 1.55 ppm.
In the absence of Lewis acid, the polymerization of
A-Phe-OMe with CTA 1 at 60 °C yielded the polymer
with a similar proportion of meso and racemic dyads
(m ) 49%, entry 1 in Table 3, Figure 6a), namely an
atactic polymer. The polymerization in the presence of
Y(OTf)3 (0.05 equiv of Y(OTf)3 as compared to monomer)
produced the polymer with higher isotacticity (m ) 62%,
entry 2 in Table 3, Figure 6b). The polymer yield was
almost quantitative after 6 h, and the control over
molecular weight and molecular weight distribution
remained as good as in the absence of Lewis acid. Thus,
3
5
racemo addition is nearly equal, resulting in atactic
polymers. In the presence of Lewis acid, the coordination
functions are predominant parameter for the stereo-
control. We, therefore, examined the effect of Lewis acid
on RAFT polymerization of A-Phe-OMe in dioxane using
AIBN and CTA 1. The monomer/chain transfer agent/
initiator molar ratio was maintained at [A-Phe-OMe]0/
[
CTA 1]0/[AIBN]0 ) 200/2/1. The results are summarized
in Table 3. The microstructure of poly(A-Phe-OMe)s
obtained under different conditions was investigated by

9062 Mori et al.
Macromolecules, Vol. 38, No. 22, 2005
Figure 7. Circular dichroism (CD) spectra (c ) 0.1 g/dL, THF)
of A-Phe-OMe and poly(A-Phe-OMe)s prepared by RAFT
polymerization in the absence and presence of Lewis acid.
1
Figure 6. H NMR (500 MHz, DMSO-d
6
, 170 °C) spectra of
poly(A-Phe-OMe)s prepared in the absence (a) and presence
b-d) of Lewis acids.
(
13
ducted preliminary experiments by C NMR analyses
the combination of RAFT and the Lewis acid complex-
ation technique allows the synthesis of well-defined
poly(A-Phe-OMe) with enhanced isotacticity. Increasing
the concentration of Y(OTf)3 to 0.1 and 0.2 equiv led to
further increase in the tacticity (m ) 67-68%), whereas
the molecular weight distributions of the resulting
polymers were relatively broad (Mw/Mn ) 1.53-1.55).
Further increase of the Y(OTf)3 concentration (Lewis
acid/monomer ) 0.5) led to the retardation of the
polymerization, in which the monomer conversion was
ca. 47% even after 6 h. In this case, the color of the
reaction mixture turned form a typical pale yellow into
brown during the polymerization, resulting in the
polymer (m ) 69%) with broad polydispersity (Mw/Mn
13
in CDCl3 at room temperature. In the C NMR mea-
surements of the monomer (A-Phe-OMe), apparent
chemical shifts were observed for N-CdO and CdO
carbons with the addition of Y(OTf)3 (Table S1, see
Supporting Information), suggesting that the Lewis acid
interacts with ester and amide groups of the monomer.
The changes in the chemical shifts of the CTA with the
addition of Y(OTf)3 were found to be smaller than those
of the monomer (Table S2, see Supporting Information)
under the same conditions. Nevertheless, further pa-
rameters, such as interaction of the growing polymer
chains with Lewis acid, concentration and molar ratio
of the reagents, and effects of the temperature and
solvent, have to take into account for the detailed
investigations of the complexations in real polymeriza-
tion system, which will be reported in our forthcoming
paper.
)
1.80).
Similar observations were also reported in the litera-
ture, in which RAFT polymerization of acryamides in
the presence of Lewis acid led to the higher polydisper-
The effects of three triflates were compared at Lewis
2
6,31
sity.
Several assumptions have been reported to
acid/monomer ) 0.05. Compared with Y(OTf) , Yb(OTf)
3
3
explain the observed higher polydispersity in the pres-
ence of Lewis acid, which involve the increase of the
polymerization rate owing to the complexation of the
Lewis acid with the monomer and propagating radical
and the decrease of the exchange rate within the
dormant intermediate radical complexed with Lewis
acid and the active propagating radical. Both behaviors
should be responsible for the rate enhancement of RAFT
polymerization. In our case, actually, the presence of
Lewis acid (Lewis acid/monomer ) 0.1 or 0.2) in the
RAFT polymerization of A-Phe-OMe in dioxane led to
the increase in the monomer conversion, as shown in
Table 3. Drastic decrease in the polymerization rate was
also observed in the RAFT polymerization of A-Phe-OMe
with higher Y(OTf)3 concentration (Lewis acid/monomer
and Sc(OTf) exhibited a similar isotacticity enhancing
3
effect (m ) 65 and 61%) but led higher polydispersity
(M /M ) 1.63 and 1.55). In these runs, the conversions
w
n
1
were nearly quantitative (>84%, as determined by H
NMR). When the polymerization was conducted with
Yb(OTf) , the Lewis acid/monomer ratio had no signifi-
3
cant influence on the tacticity, conversion, molecular
weight, and polydispersity. However, the significant
retardation was observed at Lewis acid/monomer ) 0.2
in the case of the polymerization with Sc(OTf) . These
3
results suggest that moderate isotacticity enhancing
effect (m ) 61-69%) can be achieved by all Lewis acids
used in this study, but the nature of Lewis acid has
remarkable influence on the polymerization rate and
polydispersity of the resulting poly(A-Phe-OMe)s. Figure
7 illustrates CD spectra of A-Phe-OMe and poly(A-Phe-
OMe)s measured in THF solutions (c ) 0.1 g/dL). The
monomer showed positive Cotton effects at 205 and 220
nm, and the molecular ellipticity at the maximum is
)
0.5), suggesting the existence of some unknown side
reactions assisted by Lewis acid. To achieve a better
control over the stereoregurality, we attempted to slow
the growth of the polymer chains by decreasing the
temperature to 45 °C. However, the conversion was low
2
43 186 deg‚cm /dmol. All polymers possessed positive
(
<30%) even at moderate Y(OTf)3 concentration (Lewis
Cotton effects at 218-219 nm, and the values of the
ellipticity were lower than that of the monomer. Among
the polymers, poly(A-Phe-OMe) obtained by RAFT po-
lymerization in the absence of Lewis acid ([M]D25 ) 37.0,
acid/monomer ) 0.2) after 24 h (entry 6 in Table 3).
To obtain information on the complexation of the
Lewis acid with the monomer as well as a possible
interaction between the Lewis acid and CTA, we con-
2
m ) 49%) exhibited a higher ellipticity (36 944 deg‚cm /

Macromolecules, Vol. 38, No. 22, 2005
Controlled Radical Polymerization of an Acrylamide 9063
Table 4. Reversible Addition-Fragmentation Chain Transfer (RAFT) Polymerization of N-Acryloyl-L-phenylalanine
Methyl Ester (A-Phe-OMe) in Alcoholsa
temp
(°C)
time
(h)
Mw/Mne
(SEC)
tacticity
b
c
Mnd (SEC)
f
entry
solvent
[Y(OTf)3]/[M]
conv (%)
Mn (theory)
m/r
1
2
3
4
5
6
7
MeOH
60
60
60
45
60
45
60
24
6
96
41
39
81
96
38
52
23000
9500
15000
5700
1.28
1.35
2.34
1.20
1.28
1.85
1.71
46/54
54/46
62/38
47/53
48/52
46/54
48/52
MeOH
0.05
0.2
MeOH
MeOH
6
9300
4100
24
24
24
24
19000
23000
9100
19000
28000
6100
MeOH/toluene (1/1)
g
HFIP
HFIP
g
12000
7200
a
[
CTA1]0/[AIBN]0 ) 2, [M]0/[CTA1]0 ) 100, monomer concentration ) 0.5 g/mL ([M] ) 1.38 mol/L), where M ) acryloyl-L-phenylalanine
b
methyl ester (A-Phe-OMe), AIBN ) 2,2′-azobis(isobutyronitrile), and CTA 1 ) benzyl 1-pyrrolecarbodithioate (see Scheme 2). Calculated
1
c
d
by H NMR in CDCl3. The theoretical molecular weight (Mn,theory) ) (MW of M) × conv × [M]0/[CTA]0 + (MW of CTA). Number-
average molecular weight (Mn) was measured by size-exclusion chromatography (SEC) using polystyrene standards in N,N-dimethylfor-
e
f
1
mamide (DMF, 10 mM LiBr). Molecular weight distribution (Mw/Mn) was measured by SEC in DMF (10 mM LiBr). Determined by H
g
NMR in dimethyl-d6 sulfoxide (DMSO-d6) at 170 °C. 1,1,1,3,3,3-Hexafluoro-2-propanol.
dmol). The value decreases slightly with increasing the
2
m content in the polymers (35 542 deg‚cm /dmol at m
2
)
62% and 32 507 deg‚cm /dmol at m ) 68%). The
tendency may be attributed to small conformation
difference based on the tacticity. The polymers obtained
by the RAFT polymerization in the absence and pres-
ence of Y(OTf)3 (Lewis acid/monomer ) 0.05) showed
2
5
25
the specific rotations of [R] D ) 37.0 and [R] D ) 13.0,
respectively. The significant decrease of the specific
rotations was observed in the transformation of the
2
5
monomer ([R] D ) 102.2) to the polymers. The thermal
properties of the resulting polymers were also evaluated
by differential scanning calorimetry (DSC) and ther-
mogravimetric analysis (TGA) measurements. The glass
transition temperature (Tg) of poly(A-Phe-OMe) pre-
pared in the presence of Lewis acid (m content ) 62%,
sample; entry 2 in Table 3) was found to be 87.2 °C,
which is slightly higher than that in the absence of
Lewis acid (Tg ) 82.1 °C, sample; entry 1 in Table 3).
The temperatures for 10% weight loss of poly(A-Phe-
OMe)s under nitrogen atmosphere were in the range
Figure 8. Size-exclusion chromatography (SEC) traces of
poly(A-Phe-OMe)s obtained by RAFT polymerization in dif-
ferent solvents.
polymer shows SEC peak with a shoulder at high
molecular weight region, as can be seen in Figure 8. This
is frequently observed for RAFT polymer obtained at
high monomer conversion, which is most probably
attributed to species arising from bimolecular termina-
tion reactions of the growing polymer chains. In con-
trast, the polymerization in methanol at lower temper-
ature (45 °C) was relatively successful in terms of the
suppression of the termination reactions and afforded
the polymer with low polydispersity (Mw/Mn ) 1.20) with
reduced shoulder peak, as shown in Figure 8. The SEC
traces with UV detector also showed the same tendency
(Figure S7, see Supporting Information). This behavior
may be due to that decreasing the temperature leads
to a decrease in the radical reactivity, resulting in the
increased selectivity of the various radical reactions.
Another possible explanation is that lower polymeriza-
tion temperature causes a remarkable decrease of the
fragmentation rate constant, leading to the decreased
number of the active propagating radical. In the case
of the polymerization in methanol at 45 °C, the mono-
mer conversion was 81% after 24 h, and the tacticity of
the resulting poly(A-Phe-OMe) was m ) 47%. In the
RAFT polymerization of A-Phe-OMe in methanol at 60
°C, the Lewis acid, Y(OTf)3, exhibited a less isotacticity
enhancing effect (m ) 54% at Y(OTf)3/monomer ) 0.05),
as compared in dioxane (m ) 62%). In both cases, the
Lewis acid enhanced isotacticity moderately, but higher
Lewis acid concentration resulted in the polymers
having higher polydispersity. Interestingly, the signifi-
cant retardation of the polymerization rate was observed
3
38-340 °C.
RAFT Polymerization in Alcohols. In the next
stage, we investigated the effect of polar solvent, in
particular alcohols, on the control over stereoregularity
and molecular weights of poly(A-Phe-OMe)s obtained by
RAFT polymerization. Alcohols, such as methanol and
butanol, are effective solvents for the stereocontrol of
26,30,31,35
poly(meth)acrylamides with Lewis acid.
Higher
levels of the stereocontrol in radical polymerizations
required not only a complexing Lewis acid but also the
presence of relatively polar solvents. Recently, Okamoto
et al. demonstrated another effective approach to achieve
control of stereoregularity by radical polymerization.
They reported that fluoro alcohols, such as 1,1,1,3,3,3-
hexafluoro-2-propanol (HFIP), were effective solvents for
producing highly syndiotactic poly(methyl methacrylate)
5
7,58
by conventional radical polymerization
and atom
5
9
transfer radical polymerization. The control of the
tacticity is considered to be due to the hydrogen-bond
interaction between the alcohol and monomers or grow-
ing species. We therefore attempted to extend this
methodology to the synthesis of well-defined poly(A-Phe-
OMe)s by RAFT polymerization. The polymerization
was carried out in methanol, methanol/toluene, and
HFIP at [A-Phe-OMe]0/[CTA 1]0/[AIBN]0 ) 200/2/1, and
the results are shown in Table 4. When A-Phe-OMe was
polymerized using CTA 1 with AIBN in methanol,
1
almost full conversion (96%, as determined by H NMR)
was obtained at 60 °C after 24 h. Although the poly-
dispersity remains low (Mw/Mn ) 1.28), the resulting

9064 Mori et al.
Macromolecules, Vol. 38, No. 22, 2005
by the addition of the Lewis acid in methanol, which is
the opposite tendency observed in the case of polymer-
ization in dioxane. Higher polarity of methanol com-
pared to that of dioxane may disturb the interaction of
Lewis acid with the monomer, resulting in the depres-
sion of the isotacticity enhancing effect and polymeri-
zation rate. Narrow polydispersity product (Mw/Mn )
acid moieties in the side chains and controlled archi-
tectures, such as graft, star, and block copolymers. Since
specific intra- and intermolecular interactions via hy-
drogen bonding may be manipulated by the nature of
the amino acid moiety, the self-organization of the well-
defined polymers can provide a viable route to the
production of tailored amino acid-based materials with
unique properties for various applications, such as
controlled release, biochemical sensing, biocompatible
materials, and optical resolution.
1
.28) was obtained by the RAFT polymerization in
methanol-toluene mixture at 60 °C. Under the similar
experimental conditions, the polymerizations in HFIP,
which can be regarded as the highly polar solvent, led
to insufficient monomer conversions and broad molec-
ular weight distributions (Mw/Mn > 1.7). Different from
the cases in methanol, the polymerizations in HFIP at
low temperature (45 °C) was relatively unsuccessful and
produced the polymer with broader polydispersity and
lower conversion.
Acknowledgment. This work has been partially
supported by a Grant-in-Aid for Scientific Research from
the Ministry of Education, Culture, Sports, Science, and
Technology, Japan (17550112). The authors acknowl-
edge Associate Professor Seigou Kawaguchi for GPC-
MALS measurement.
From the polymerization results mentioned above, it
was demonstrated that the synthesis of the well-defined
poly(A-Phe-OMe)s with predetermined molecular weight
and narrow molecular weight distribution was possible
by RAFT polymerization using CTA 1, even in the
presence of Lewis acid, in dioxane or alcohols. This
indicates the feasibility to achieve simultaneous control
of the molecular weight and stereochemistry, but fur-
ther improvement is needed for the preparation of the
amino acid-based polymer with higher stereoregularity.
It seems reasonable to expect that the bulky phenyla-
lanine side chain in A-Phe-OMe has negative influence
on the stereocontrol because the coordination of the
Lewis acid with monomers or growing polymer chains
is crucial to attain the stereocontrol by Lewis acid, and
the hydrogen-bond interaction is key factor to produce
syndiotactic polymers in fluoro alcohols. The bulky
phenylalanine side chain in A-Phe-OMe may disturb the
coordination and the hydrogen-bond interaction, result-
ing in insufficient control of the tacticity. In other words,
RAFT polymerization of a monosubstituted acrylamide
having a smaller amino acid moiety, such as alanine
and glycine, in the side chain may help to achieve
simultaneous control of the molecular weight and ster-
eochemistry. Further studies for such directions are now
in progress, which will be reported separately.
1
Supporting Information Available: Figures showing H
1
3
and C NMR spectra of the monomer and polymer, experi-
mental conditions and results of the polymerizations at dif-
ferent [M] :[CTA 1] ratios under nitrogen at 90 °C, and kinetic
0 0
results of the polymerization with CTA 1 at 90 °C, comparison
of UV (280 nm) and RI detector responses of the SEC traces,
comparison of light scattering (90°) and RI detector responses
of the SEC traces, and changes in the C NMR chemical shifts
of the monomer (A-Phe-OMe) and CTA 1 with the addition of
3
Y(OTf) . This material is available free of charge via the
13
Internet at http://pubs.acs.org.
References and Notes
(1) Klok, H.-A.; Langenwalter, J. F.; Lecommandoux, S. Macro-
molecules 2000, 33, 7819-7826.
(
2) Schlaad, H.; Kukula, H.; Smarsly, B.; Antonietti, M.; Pakula,
T. Polymer 2002, 43, 5321-5328.
(
3) Floudas, G.; Papadopoulos, P.; Klok, H.-A.; Vandermeulen,
G. W. M.; Rodriguez-Hernandez, J. Macromolecues 2003, 36,
3673-3683.
(
(
4) Deming, T. J. Nature (London) 1997, 390, 386-389.
5) Deming, T. J. J. Polym. Sci., Part A: Polym. Chem. 2000,
38, 3011-3018.
(
(
(
(
6) Nagai, A.; Sato, D.; Ishikawa, J.; Ochiai, B.; Kudo, H.; Endo,
T. Macromolecules 2004, 37, 2332-2334.
7) Naka, K.; Nemoto, T.; Chujo, Y. J. Polym. Sci., Part A: Polym.
Chem. 2003, 41, 1504-1510.
8) Mei, Y.; Beers, K. L.; Byrd, H. C. M.; VanderHart, D. L.;
Washburn, N. R. J. Am. Chem. Soc. 2004, 126, 3472-3476.
9) Becker, M. L.; Liu, J. Q.; Wooley, K. L. Chem. Commun.
(
Cambridge) 2003, 180-181.
Conclusion
(
(
(
10) Sanda, F.; Endo, T. Macromol. Chem. Phys. 1999, 200, 2651-
2661.
We have demonstrated the first successful controlled
radical polymerization of a monosubstituted acrylamide
having an amino acid moiety, N-acryloyl-L-phenyl-
alanine methyl ester (A-Phe-OMe), via the RAFT pro-
cess. Benzyl 1-pyrrolecarbodithioate (CTA 1) is efficient
as the chain transfer agent for the preparation of near-
monodisperse poly(A-Phe-OMe)s with controlled molec-
ular weights. Good control of the polymerization in
dioxane at 60 °C was confirmed by the linear relation-
ship between the molecular weight and the monomer/
CTA molar ratio, the linear increase in the molecular
weight with the conversion, and successful chain exten-
sion. The poly(A-Phe-OMe)s with low polydispersity
could be obtained also under various conditions (solvent
11) Methenitis, C.; Morcellet, J.; Pneumatikakis, G.; Morcellet,
M. Macromolecues 1994, 27, 1455-1460.
12) Blaschke, G. Angew. Chem., Int. Ed. Engl. 1980, 19, 13-23.
(13) Angiolini, L.; Caretti, D.; Carlini, C.; Salatelli, E. Macromol.
Chem. Phys. 1995, 196, 2737-2750.
(
14) Barbucci, R.; Casolaro, M.; Magnani, A. J. Controlled Release
1991, 17, 79-88.
(
15) Casolaro, M.; Barbucci, R. Int. J. Artif. Organs 1991, 14, 732-
738.
(16) Bentolila, A.; Vlodavsky, I.; Ishai-Michaeli, R.; Kovalchuk,
O.; Haloun, C.; Domb, A. J. J. Med. Chem. 2000, 43, 2591-
2600.
(
17) Bignotti, F.; Penco, M.; Sartore, L.; Peroni, I.; Mendichi, R.;
Casolaro, M.; D’Amore, A. Polymer 2000, 41, 8247-8256.
(18) Casolaro, M. Macromolecues 1995, 28, 2351-2358.
(19) Yoshida, M.; Safranj, A.; Omichi, H.; Katakai, R. Macromol-
ecules 1996, 29, 2321-2323.
)
dioxane, methanol, and methanol/toluene; tempera-
(
(
20) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921-2990.
21) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001, 101,
3689-3745.
ture ) 45, 60, and 90 °C). This work presents the
feasibility that the combination of RAFT and Lewis acid
complexation allows the synthesis of well-defined amino
acid-based polymers with predetermined molecular
weight, narrow molecular weight distribution, and
improved tacticity. This procedure may extend to the
synthesis of well-defined polymers having various amino
(22) Hawker, C. J.; Bosman, A. W.; Harth, E. Chem. Rev. 2001,
101, 3661-3688.
(
23) Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina, J.; Jeffery, J.;
Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, C. L.;
Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecues 1998,
31, 5559-5562.

Macromolecules, Vol. 38, No. 22, 2005
Controlled Radical Polymerization of an Acrylamide 9065
(
24) Rizzardo, E.; Chiefari, J.; Mayadunne, R.; Moad, G.; Thang,
S. Macromol. Symp. 2001, 174 (Polymerization Processes and
Polymer Materials I), 209-212.
(42) Barner-Kowollik, C.; Quinn, J. F.; Morsley, D. R.; Davis, T.
P. J. Polym. Sci., Part A: Polym. Chem. 2001, 39, 1353-
1365.
(
(
(
(
25) Chong, Y. K.; Le, T. P. T.; Moad, G.; Rizzardo, E.; Thang, S.
(
43) Thomas, D. B.; Convertine, A. J.; Myrick, L. J.; Scales, C.
W.; Smith, A. E.; Lowe, A. B.; Vasilieva, Y. A.; Ayres, N.;
McCormick, C. L. Macromolecules 2004, 37, 8941-8950.
H. Macromolecules 1999, 32, 2071-2074.
26) Lutz, J.-F.; Neugebauer, D.; Matyjaszewski, K. J. Am. Chem.
Soc. 2003, 125, 6986-6993.
(
(
(
44) Vasilieva, Y. A.; Thomas, D. B.; Scales, C. W.; McCormick,
27) Donovan, S. M.; Lowe, B. A.; Sumerlin, S. B.; McCormick, L.
C. Macromolecues 2002, 35, 4123-4132.
C. L. Macromolecules 2004, 37, 2728-2737.
45) McCormick, C. L.; Lowe, A. B. Acc. Chem. Res. 2004, 37, 312-
28) Ganachaud, F.; Monteiro, M. J.; Gilbert, R. G.; Dourges, M.-
A.; Thang, S. H.; Rizzardo, E. Macromolecules 2000, 33,
325.
46) Vana, P.; Davis, T. P.; Barner-Kowollik, C. Macromol. Theory
6738-6745.
Simul. 2002, 11, 823-835.
(29) Schilli, C.; Lanzend o¨ rfer, M. G.; M u¨ ller, A. H. E. Macromol-
ecules 2002, 35, 6819-6827.
(30) Ray, B.; Isobe, Y.; Morioka, K.; Habaue, S.; Okamoto, Y.;
Kamigaito, M.; Sawamoto, M. Macromolecules 2003, 36, 543-
(47) Perrier, S.; Barner-Kowollik, C.; Quinn, J. F.; Vana, P.; Davis,
T. P. Macromolecules 2002, 35, 8300-8306.
(48) Plummer, R.; Goh, Y.-K.; Whittaker, A. K.; Monteiro, M. J.
545.
Macromolecules 2005, 38, 5352-5355.
(31) Ray, B.; Isobe, Y.; Matsumoto, K.; Habaue, S.; Okamoto, Y.;
(
(
49) Coote, M. L. Macromolecules 2004, 37, 5023-5031.
Kamigaito, M.; Sawamoto, M. Macromolecues 2004, 37,
50) McLeary, J. B.; Calitz, F. M.; McKenzie, J. M.; Tonge, M. P.;
1702-1710.
Sanderson, R. D.; Klumperman, B. Macromolecules 2005, 38,
(
(
(
(
(
32) Favier, A.; Charreyre, M. T.; Chaumont, P.; Pichot, C.
Macromolecules 2002, 35, 8271-8280.
3151-3161.
(
51) Isobe, Y.; Fujioka, D.; Habaue, S.; Okamoto, Y. J. Am. Chem.
33) D’Agosto, F.; Hughes, R.; Charreyre, M.; Pichot, C.; Gilbert,
R. G. Macromolecules 2003, 36, 621-629.
Soc. 2001, 123, 7180-7181.
34) Favier, A.; Ladaviere, C.; Charreyre, M. T.; Pichot, C.
Macromolecules 2004, 37, 2026-2034.
(52) Isobe, Y.; Suito, Y.; Habaue, S.; Okamoto, Y. J. Polym. Sci.,
Part A: Polym. Chem. 2003, 41, 1027-1033.
35) Lutz, J. F.; Jakubowski, W.; Matyjaszewski, K. Macromol.
Rapid Commun. 2004, 25, 486-492.
(53) Suito, Y.; Isobe, Y.; Habaue, S.; Okamoto, Y. J. Polym. Sci.,
Part A: Polym. Chem. 2002, 40, 2496-2500.
36) Chiefari, J.; Mayadunne, R. T. A.; Moad, C. L.; Moad, G.;
Rizzardo, E.; Postma, A.; Skidmore, M. A.; Thang, S. H.
Macromolecules 2003, 36, 2273-2283.
(
(
(
54) Baraki, H.; Habaue, S.; Okamoto, Y. Macromolecules 2001,
4, 4724-4729.
3
55) Isobe, Y.; Nakano, T.; Okamoto, Y. J. Polym. Sci., Part A:
(37) Chong, Y. K.; Krstina, J.; Le, T. P. T.; Moad, G.; Postma, A.;
Polym. Chem. 2001, 39, 1463-1471.
Rizzardo, E.; Thang, S. H. Macromolecules 2003, 36, 2256-
56) Morioka, K.; Suito, Y.; Isobe, Y.; Habaue, S.; Okamoto, Y. J.
Polym. Sci., Part A: Polym. Chem. 2003, 41, 3354-3360.
2272.
(38) Sanda, F.; Abe, T.; Endo, T. J. Polym. Sci., Part A: Polym.
Chem. 1997, 35, 2619-2629.
(39) Moad, G.; Chiefari, J.; Chong, Y. K.; Krstina, J.; Mayadunne,
R. T. A.; Postma, A.; Rizzardo, E.; Thang, S. H. Polym. Int.
(57) Isobe, Y.; Yamada, K.; Nakano, T.; Okamoto, Y. Macromol-
ecules 1999, 32, 5979-5981.
(58) Isobe, Y.; Yamada, K.; Nakano, T.; Okamoto, Y. J. Polym.
Sci., Part A: Polym. Chem. 2000, 38, 4693-4703.
2000, 49, 993-1001.
(
(
40) Monteiro, M. J.; de Brouwer, H. Macromolecules 2001, 34,
(
59) Miura, Y.; Satoh, T.; Narumi, A.; Nishizawa, O.; Okamoto,
349-352.
Y.; Kakuchi, T. Macromolecules 2005, 38, 1041-1043.
41) Kwak, Y.; Goto, A.; Tsujii, Y.; Murata, Y.; Komatsu, K.;
Fukuda, T. Macromolecules 2002, 35, 3026-3029.
MA0509558