Room temperature atom transfer radical polymerization of glycidyl methacrylate mediated by copper(I)/N-alkyl-2-pyridylmethanimine complexes

Article abstract of DOI:10.1021/ma0359032

The homogeneous controlled/"living" free radical polymerization of glycidyl methacrylate (GMA) by atom transfer radical polymerization (ATRP) using Cu(I)X/N-alkyl-2-pyridylmethanimine complexes with various initiators R-X (X = Cl, Br) and solvents was investigated. Most of these systems display characteristics of a living radical polymerization as indicated by (a) linear first-order kinetic plots of ln[M]₀/[M] vs time, (b) an increase in the number-average molecular weight (M_n) vs conversion, and (c) relatively narrow polydispersities indicating a constant number of propagating species throughout the polymerization with negligible contribution of termination or transfer reactions. The dependence of the rate of polymerization on the concentrations of initiator, ligand, and temperature is presented. We observed comparable rates of polymerization linear increase of molecular weight with conversion and low polydispersities in polar solvents. No polymerization was observed in nonpolar solvents such as toluene and xylene at room temperature. The order of controlled polymerization with different initiator system is CuBr/BPN > CuCl/BPN > CuBr/ClPN, and the polymerization did not proceed with CuCl/ClPN initiator system at room temperature. The high functionality of bromine end groups present in the polymer chains was confirmed by ESI MS analysis. The thermal stability of PGMA prepared by the CuBr/PPMI/BPN initiation system is higher than by the other three systems, indicating the high regioselectivity and the virtual absence of termination reactions in the former case. The ligand alkyl chain length from R = propyl to octyl did not affect the rate of polymerization. The molecular weight (M_n) increases linearly with conversion, and these polymers showed narrow polydispersities.

Full text of DOI:10.1021/ma0359032

3
614
Macromolecules 2004, 37, 3614-3622
Room Temperature Atom Transfer Radical Polymerization of Glycidyl
Methacrylate Mediated by Copper(I)/N-Alkyl-2-pyridylmethanimine
Complexes†
R. Kr ish n a n a n d K. S. V. Sr in iva sa n *
Polymer Division, Central Leather Research Institute, Adyar, Chennai 600 020, India
Received December 15, 2003; Revised Manuscript Received March 12, 2004
ABSTRACT: The homogeneous controlled/“living” free radical polymerization of glycidyl methacrylate
(GMA) by atom transfer radical polymerization (ATRP) using Cu(I)X/N-alkyl-2-pyridylmethanimine
complexes with various initiators R-X (X ) Cl, Br) and solvents was investigated. Most of these systems
display characteristics of a living radical polymerization as indicated by (a) linear first-order kinetic plots
of ln[M] /[M] vs time, (b) an increase in the number-average molecular weight (M ) vs conversion, and (c)
0 n
relatively narrow polydispersities indicating a constant number of propagating species throughout the
polymerization with negligible contribution of termination or transfer reactions. The dependence of the
rate of polymerization on the concentrations of initiator, ligand, and temperature is presented. We observed
comparable rates of polymerization linear increase of molecular weight with conversion and low
polydispersities in polar solvents. No polymerization was observed in nonpolar solvents such as toluene
and xylene at room temperature. The order of controlled polymerization with different initiator system
is CuBr/BPN > CuCl/BPN > CuBr/ClPN, and the polymerization did not proceed with CuCl/ClPN initiator
system at room temperature. The high functionality of bromine end groups present in the polymer chains
was confirmed by ESI MS analysis. The thermal stability of PGMA prepared by the CuBr/PPMI/BPN
initiation system is higher than by the other three systems, indicating the high regioselectivity and the
virtual absence of termination reactions in the former case. The ligand alkyl chain length from R ) propyl
to octyl did not affect the rate of polymerization. The molecular weight (M
n
) increases linearly with
conversion, and these polymers showed narrow polydispersities.
In tr od u ction
Living” polymerization of vinyl monomers is a route
Transition-metal-mediated atom transfer radical po-
lymerization is a versatile technique for the controlled
polymerization of wide variety of functional monomers
and has been used successfully to prepare well-defined
“
to synthesize a wide range of well-defined polymers. It
allows the synthesis of polymers with well-defined
compositions, architectures, functionalities, chain topol-
ogy of macromolecules with narrow molecular weight
distribution, targeted number-average molecular weights,
specific end groups, etc. Traditionally, this has been
9
polymers such as styrene, substituted styrenes, (meth)-
1
0
11
acrylates, and acrylonitrile. Atom transfer radical
9
-11
6
12
13
polymerization systems using Cu,
Ru, Rh, Ni,
1
4
15
Pd, and iron based transition metals in conjunction
with suitable ligands such as substituted and unsub-
stituted bipyridines, phosphorus-containing ligands, and
multidentate amines have been used as catalysts. This
living radical polymerization technique tolerates a wide
range of functional groups in the monomer, solvent, or
initiator. Although ATRP is usually performed in the
bulk or in an organic solvent, the efficient polymeriza-
tion of various hydrophilic monomers in protic media
1
-3
accomplished by living ionic polymerization techniques.
However, such polymerization techniques require high-
purity monomers, solvents, and reagents, coupled with
low temperatures, which thereby makes ionic polymer-
izations relatively difficult processes. Another major
limitation of ionic polymerization is the difficulty in
polymerizing monomers containing polar and/or func-
tional groups, e.g., 2-hydroxyethyl methacrylate, (meth)-
acrylic acid, and glycidyl methacrylate, as they can
complicate reaction pathways. Functional monomers
can be readily polymerized via free-radical routes. The
development of controlled/“living” radical polymerization
processes has been the focus of much activity in recent
years. In the past decade, the development of controlled/
1
6
1
7-21
has been demonstrated at ambient temperature
and
2
2
also for hydrophobic monomers by using a more
powerful catalyst system such as CuBr/Me6-TREN and
CuBr/Me4Cyclam under mild conditions or at room
temperature.
Haddleton and co-workers developed a catalyst based
“
living” radical polymerization (CRP) techniques allows
on Cu(I)X and N-alkyl-2-pyridylmethanimine Schiff
the synthesis of polymers using less vigorous conditions.
Many controlled polymerization techniques have been
proposed, including nitroxide-mediated living radical
23
base ligands for the living-radical polymerization, and
the same catalyst was used in an aqueous media
20
without hydrolysis of the ligand. The Schiff based
4
polymerization (NMRP), atom transfer radical polym-
ligands are effective for the polymerization of meth-
erization (ATRP),5 reversible addition and fragmenta-
,6
24,25
acrylates
in toluene or xylene solutions in conjunc-
7
tion chain transfer (RAFT), and group transfer polym-
erization (GTP).
tion with copper(I) halides and suitable alkyl halide
initiators over a range of temperatures (even as low as
8
2
6
-
15 °C). This paper deals with the controlled/“living”
†
Part of this work has been presented at the 224th ACS
radical polymerization of glycidyl methacrylate (GMA)
by homogeneous ATRP by using Cu(I)X complexed with
various N-alkyl-2-pyridylmethanimine ligand as cata-
National Meeting, Boston, Aug 18-22, 2002.
*
Corresponding author: Ph +91-044-24422059; Fax +91-044-
2
4911589; e-mail srinivasan1@hotmail.com.
1
0.1021/ma0359032 CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/24/2004

Macromolecules, Vol. 37, No. 10, 2004
ATRP of Glycidyl Methacrylate 3615
Exp er im en ta l Section
Ma ter ia ls. Glycidyl methacrylate (GMA, Aldrich, 97%) was
purified by vacuum distillation under reduced pressure and
stored in a refrigerator under a nitrogen atmosphere. n-
Propylamine (S.D. Fine Chem., India, 98%), n-butylamine
(Aldrich, 99.5%), n-hexylamine (Aldrich, 99%), n-octylamine
(Aldrich, 99%), pyridine-2-carboxaldehyde (Lancaster, 99%),
and cuprous chloride (CuCl, Aldrich, 99+%) were all used as
received. Initiators methyl 2-bromopropionate (M 2-BP, Ald-
rich, 98%), 2-bromopropionitrile (BPN, Aldrich, 97%), 1-phen-
ylethyl bromide (1-PEBr, Aldrich, 97%), and 2-chloropropioni-
trile (ClPN, Aldrich, 95%) were used as received. Cuprous
bromide (CuBr, Fluka, 98%) was purified by washing with
glacial acetic acid, followed by absolute ethanol and diethyl
ether, and then dried under vacuum at room temperature. The
initiator AIBN was recrystallized from ether and was dried
at room temperature under vacuum. Diphenyl ether (DPE) was
dried over molecular sieves and stored in a nitrogen atmo-
sphere. All other chemicals and solvents were purchased from
commercial sources and used after standard purification
procedures.
F igu r e 1. Effect of initiator concentration on kinetics of
homogeneous ATRP of GMA in bulk polymerization at ambient
temperature. [GMA]
and 0.029 M (1). [CuBr]
0
) 7.3 M; [BPN]
0
) 0.097 (9), 0.048 (2),
Gen er a l P r oced u r e for th e ATRP of GMA. The ATR
polymerization of glycidyl methacrylate was carried out in bulk
and in solution. In a typical bulk polymerization, a dry glass
tube was charged with CuBr (21.0 mg, 0.146 mmol) and N-(n-
propyl)-2-pyridylmethanimine (PPMI) (43.2 mg, 0.292 mmol),
GMA (2.0 mL, 0.014 mol), and a magnetic stir bar. The tube
was fitted with a rubber septum and degassed by three freeze-
pump-thaw cycles. The solution turned dark brown and
homogeneous. Initiator 2-bromopropionitrile (12.6 µL, 0.146
mmol) was added via a degassed syringe, and the solution
became progressively more viscous, indicating the onset of
polymerization. After various time intervals, the polymer was
dissolved in tetrahydrofuran and then filtered through a silica
gel column to remove the copper catalyst. The polymer solution
was then precipitated by using excess of petroleum ether and
0
) [PPMI] /2 ) 0.058 M.
0
Sch em e 1
dried in a vacuum (time 90 min, 88% conversion, Mn,th
2 500, Mn,GPC ) 12 150, M /M ) 1.21).
Ch a r a cter iza tion . Conversion was determined by gravim-
etry. Molecular weights and molecular weight distributions
)
1
w
n
lyst and various initiators in bulk and various solvents
at an ambient temperature (30 °C).27
F igu r e 2. Plot of M
GMA at ambient temperature. [GMA]
solid lines indicate Mn,th values corresponding to the [M]/[I]
n
and M
w
/M
n
as a function of monomer conversion on differing ratios of monomer:initiator for bulk ATRP of
) 7.3 M; [BPN] ) 0.097 (9), 0.048 (2), and 0.029 M (1). [CuBr] ) [PPMI] /2 ) 0.058 M
ratios).
0
0
0
0
(
0

3
616 Krishnan and Srinivasan
Macromolecules, Vol. 37, No. 10, 2004
F igu r e 5. Dependence of molecular weight (filled symbols)
and molecular weight distributions (open symbols) as a func-
tion of monomer conversion for ATRP of GMA at an ambient
F igu r e 3. Dependence of the apparent rate constant of
propagation (k ) on the initiator concentration for homoge-
app
temperature in different solvent systems. [GMA]
BPN] ) [CuBr] ) [PPMI] /2 ) 0.036 M (monomer:solvent
50% v/v).
0
) 3.65 M;
neous ATRP of GMA in bulk at ambient temperature. [GMA]
7.3 M; [CuBr] ) [PPMI] /2 ) 0.058 M.
0
[
0
0
0
)
0
0
)
F igu r e 6. Semilogarithmic kinetic plot for the bulk polym-
erization of GMA at ambient temperature using three different
F igu r e 4. Semilogarithmic kinetic plot for ATR polymeriza-
tion of GMA in different solvents at an ambient temperature.
initiation systems. [GMA]
[PPMI]/2 ) 0.073 M.
0
) 7.3 M; [initiator]
0
) [CuBr] )
0
[
(
GMA]
0
) 3.65 M; [BPN]
0
) [CuBr]
0
) [PPMI]
0
/2 ) 0.036 M
monomer:solvent ) 50% v/v).
and a wide range of functionalities could be introduced
with subsequent reactions. The homopolymerization and
copolymerization of GMA are reported by controlled
were measured by gel permeation chromatography (GPC). The
GPC setup consisted of a Waters 515 LC pump and a Waters
4
10 differential refractive index detector, with four Ultrastyra-
2
7,28
GTP,8
3
4
5
polymerization techniques, such as ATRP,
gel columns (guard, 10 , 10 , 10 Å). Calibration was based on
linear polystyrene standards. The GPC eluent was HPLC
grade THF at a flow rate of 1.0 mL/min. The instrument was
2
9
and NMRP.
The kinetic plot of ln[M]0/[M] vs time for homogeneous
bulk ATRP of GMA catalyzed by CuBr/PPMI initiated
by 2-bromopropionitrile at three different initiator
concentrations is shown in Figure 1. The straight lines
passing through the origin show that the polymerization
proceeded with an approximately constant number of
active species present throughout the polymerization,
and therefore the contribution of termination reactions
could be negligible. As expected, the apparent rate
1
set at 40 °C before starting the analysis. H NMR spectra were
recorded in CDCl
eter operating at 300 MHz. The Fourier transform infrared
FT-IR) spectra of the ligands were recorded with a Nicolet
Impact 400 spectrometer on solution cast onto KBr disks.
3
solvent using a Bruker MSP 300 spectrom-
(
1
13
(
Representative H, C NMR, and FT-IR spectra of polymer
and ligands are given in the Supporting Information.) Elec-
trospray ionization mass spectrometry of the polymer was
analyzed by a Hewlett-Packard 1100 MSD mass spectrometer.
Thermogravimetric analysis was carried out on a Seiko model
SSC 5200H system attached to a thermogravimetry/dynamic
thermal analysis module at a heating rate of 20 °C/min in a
nitrogen atmosphere (flow rate 100 mL/min).
app
constant of polymerization (k ) decreased with de-
creasing initiator concentration, as determined from the
kinetic slopes. The molecular weight evolution, M , and
n
molecular weight distribution, Mw/Mn, as a function of
monomer conversion are shown in Figure 2. The mo-
lecular weights determined by GPC against linear
polystyrene standards generally agreed with the theo-
retical lines calculated from the degree of polymerization
Resu lts a n d Discu ssion
Poly(glycidyl methacrylate) (Scheme 1) is of great
interest since the pendant oxirane ring could be opened

Macromolecules, Vol. 37, No. 10, 2004
ATRP of Glycidyl Methacrylate 3617
F igu r e 7. Dependence of number-average molecular weight
and polydispersities on monomer conversion in the bulk
polymerization of GMA at ambient temperature using three
F igu r e 9. Dependence of molecular weights (open symbols)
and polydispersities (filled symbols) on monomer conversion
for bulk ATRP of GMA at an ambient temperature using Cu-
(I)X/XPN initiation systems, where X ) Cl or Br (solid line
different initiation systems. [GMA]
CuBr] ) [PPMI]/2 ) 0.073 M.
0
) 7.3 M; [initiator] )
0
[
0
represents Mn,th values). [GMA]
[PPMI] /2 ) 0.073 M.
0
) 7.3 M; [XPN]
0
) [CuBr]
0
)
0
F igu r e 10. First-order kinetic plots for the polymerization
of GMA with various N-(n-alkyl)-2-pyridylmethanimine ligands.
F igu r e 8. Semilogarithmic kinetic plots for the bulk ATRP
of GMA at an ambient temperature using Cu(I)X/XPN initia-
[GMA]
0
) 7.3 M; [BPN]
0
) [CuBr]
0
) [ligand] /2 ) 0.097 M.
0
tion systems, where X ) Cl or Br. [GMA]
CuBr] ) [PPMI] /2 ) 0.073 M.
0
) 7.3 M; [XPN] )
0
slower rate of polymerization than the bulk polymeri-
zation. The polymerization of GMA in diphenyl ether
[
0
0
(
DPE) medium proceeds at a much faster rate compared
to the other polar solvents such as methyl ethyl ketone
MEK), methanol, and anisole. The polymerizations did
as predicted by the initial ratio of monomer:initiator.
The experimental molecular weight (Mn,GPC) values were
very close to theoretical lines (Mn,th), indicating that
BPN was an efficient initiator for ATR polymerization
of GMA. The molecular weights increased linearly with
monomer conversion, and low polydispersities Mw/Mn
(
not occur in nonpolar solvents such as toluene and
xylene even after 24 h. The plots of Mn and Mw/Mn as a
function of monomer conversion for different polymer-
ization systems are shown in Figure 5. In the case of
bulk and diphenyl ether solution polymerization sys-
tems experimental molecular weights (Mn,GPC) follow the
theoretical line and narrow polydispersities were ob-
served (Mw/Mn < 1.3), whereas in other polar solvents
such as MEK, anisole, and methanol, molecular weights
are higher than the predicted theoretical values. This
indicates inefficient initiation in the polar solvents
presumably caused by a slower deactivation, resulting
in slightly higher polydispersities Mw/Mn < 1.5.
<
1.35 were observed for all polymerizations. Polydis-
persities slightly increased with increasing [M]/[I]0 ratio
(
Figure 2).
Plotting ln kapp vs ln[initiator] elucidated the depen-
dence of the rate of polymerization on the initiator
concentration. For this system, the slope of the line
indicated an apparent 0.8 order with respect to initiator
and (Figure 3) differs slightly from the first-order
dependence of the rate on the initiator concentration.
Different solvents were investigated for the ATRP of
glycidyl methacrylate using the CuBr/PPMI catalyst
system initiated by BPN at room temperature (30 °C)
with a monomer concentration of 50% (v/v) for all
reactions. The first-order kinetic plots of the polymer-
izations are shown in Figure 4. All solvents showed a
Three different initiators were examined for bulk
ATRP of GMA such as 2-bromopropionitrile, methyl
2-bromopropionate, and 1-phenylethyl bromide along
with CuBr/PPMI catalyst system at room temperature,
and the obtained results are shown in Figures 6 and 7.
The polymerization of GMA proceeds with first-order

3
618 Krishnan and Srinivasan
Macromolecules, Vol. 37, No. 10, 2004
Ta ble 1. Effect of Alk yl Ch a in Len gth in
N-Alk yl-2-p yr id ylm eth a n im in e Liga n d s on th e Atom
Tr a n sfer Ra d ica l P olym er iza tion of GMA in Bu lk a t RTa
time/
min
ligand
PPMI
Mn,GPC
Mw/Mn
% conv kapp × 10-4 (s-1
)
15
1580
3130
4150
4920
6980
1870
3730
4910
5980
7650
1855
3490
4680
5590
7120
1780
3650
4680
5885
7350
1.19
1.17
1.20
1.18
1.20
1.23
1.25
1.20
1.24
1.22
1.18
1.17
1.21
1.19
1.19
1.16
1.15
1.10
1.12
1.13
19.0
35.0
46.2
57.3
73.0
20.6
37.0
47.8
59.5
75.5
20.0
35.6
45.7
57.0
72.5
19.0
35.0
45.7
56.5
72.3
2.415
2.568
2.380
2.368
3
4
6
9
0
5
0
0
BPMI
HPMI
OPMI
15
3
4
6
9
0
5
0
0
15
3
4
6
9
0
5
0
0
15
3
4
6
9
0
5
0
0
F igu r e 11. Plot of kapp vs equivalents of PPMI used for bulk
ATRP of GMA using CuBr/PPMI as the catalyst at room
temperature. [GMA]
0
) 7.3 M; [BPN]
0
) [CuBr] ) 0.048 M.
0
a
[GMA]0 ) 7.3 M; [BPN]0 ) [CuBr]0 ) [ligand]0/2 ) 0.097 M.
kinetics with respect to monomer concentration for BPN
and M 2-BP initiation systems, whereas 1-PEBr showed
significant curvature in the first-order kinetic plot,
which indicates that termination is detectable as shown
in Figure 6. The number-average molecular weight (Mn)
increases linearly with monomer conversion, which
indicates constant concentration of growing chains
present throughout the reaction. In BPN initiator, the
Mn,GPC values follow the theoretical line, whereas in the
M 2-BP system a higher value than theoretical line
indicates inefficient initiation caused by slow deactiva-
tions. In the 1-PEBr/CuBr initiation system, the rate
of initiation is much slower than propagation, resulting
an inefficient initiation. Consequently, deviation of
molecular weights from the theoretical line and higher
polydispersities were also observed.
2
-Halopropionitriles (XPN, X ) Br or Cl) are good
app
F igu r e 12. Plot of ln k
7.3 M; [BPN]
p
vs 1/T for ATRP of GMA. [GMA]
) [ligand] /2 ) 0.097 mM.
0
initiators for copper-based ATR polymerization of meth-
acrylates and acrylonitrile. Comparison between the
initiation systems CuBr/BPN and CuCl/ClPN shows
that CuBr/BPN gave better control of the polymeriza-
tion, Mn,GPC values are very close to the theoretical line,
and low polydispersities were observed throughout the
polymerization, which indicates an efficient activation/
deactivation process. On the other hand, no polymeri-
zation was observed in the CuCl/ClPN system at room
temperature. The results of mixed halide/initiator sys-
tem for ATRP of GMA are shown in Figures 8 and 9. In
the mixed halide initiation systems, CuCl/BPN gave
better control of polymerization and the Mn,GPC values
followed the theoretical line, whereas the CuBr/ClPN
system showed a significant curvature in a first-order
kinetic plot; poor control of molecular weight and
molecular weight distribution was observed (Figure 9).
The results imply that the mixed halide system CuCl/
BPN should give faster initiation and slower propaga-
tion and therefore better control of molecular weight and
lower polydispersities, whereas the CuBr/ClPN initia-
tion system gave essentially uncontrolled polymeriza-
tion, which was attributed to the strong C-Cl bond and
inefficient initiation.30 In the mixed halide system the
presence of a chlorine originating from either the
initiator or the catalyst becomes the ω´ -terminus of the
)
0
) [CuBr]
0
0
polymer, leading to the lowering of the rate of polym-
erization compared to the CuBr/BPN initiation system.
The alkyl chain length in the Schiff base ligands can
be easily varied by appropriate choice of primary amine
reacting with pyridine-2-carboxaldehyde. Figure 10
shows the first-order kinetic plots of bulk polymerization
of GMA with four different N-alkyl-2-pyridylmetha-
nimine ligands ranging from n-propyl to n-octyl, and
very similar rates of polymerization were observed for
all the reactions. The nature of the alkyl group dramati-
cally affects the solubility of the copper complexes in a
range of solvents, especially at ambient temperature.
Normally, when the length of the alkyl group is in-
creased, the system becomes more soluble in nonpolar
solvents. But in more polar monomers such as MMA
and GMA, the shorter alkyl chain ligands are more
soluble, and when the concentration of the monomer in
the reactions is increased, the solubility of the copper
complexes is also increased. In all polymerizations, the
Mn increases linearly with monomer conversion, and the
polydispersity index remains relatively narrow; the
same is shown in Table 1. Mn,GPC values are slightly
lower than the predicted values. These results suggest

Macromolecules, Vol. 37, No. 10, 2004
ATRP of Glycidyl Methacrylate 3619
F igu r e 13. (a) ESI MS spectrum of PGMA prepared by CuBr/BPN/PPMI initiation system. (b) Expanded ESI MS spectrum.
that the variation of alkyl chain length from n-propyl
to n-octyl has no appreciable changes in the rate of
polymerization.
ing significantly as ligand concentration is raised to this
value, but beyond this concentration the rate of polym-
erization is slightly decreased. These reactions are
usually carried out with 2 times molar excess of Schiff
base ligand with respect to Cu(I)X. Because the Cu(I)
prefers a tetrahedral geometry, which can be achieved
with two bidendate ligands. However, further increasing
the ligand concentration, coordination of the ligand to
the metal center may hinder the activity of the catalyst,
The dependence of the rate of polymerization in the
homogeneous ATRP of GMA with catalyst concentration
was also investigated. Figure 11 shows the dependence
of the apparent rate constant of polymerization as a
function of [ligand]:[CuBr] ratio. The optimum ratio is
found to be 2:1 with the rate of polymerization increas-

3
620 Krishnan and Srinivasan
Macromolecules, Vol. 37, No. 10, 2004
F igu r e 14. Comparison of (a) TG and (b) DTG curves for PGMA prepared by five different initiation systems. Reaction
conditions: [GMA] ) 7.3 M; [XPN] ) [CuBr] ) [PPMI] /2 ) 0.073 M.
0
0
0
0
resulting in a slight decrease in the rate of polymeri-
zation.
catalyst system to initiate the polymerization of the
second monomer to form a block, graft, star polymers,
etc., depending on the position and number of initiation
sites. The existence of chain end functionality was
confirmed using ESI MS (Figure 13a,b). A series cor-
responding to the bromo-terminated PGMA with a
sodium cation [m ) 55 + n(142.16) + 79/81 + 23; for n
) 7, m ) 1152/1154] and the major series match the
molecular weight of the polymer, which is doubly
charged with two sodium cations [m ) 55 + n(142.16)
+ 79/81 + 23 + 23]/2; for n ) 13, m ) 1014/1016]. The
entire range of ESI mass spectrum is given in Figure
13a. The doubly charged species is more intense than
the singly charged species. The obtained ESI mass
spectrum indicated that there was no significant loss
of end group. The results suggest that high functionality
The temperature dependence on the reaction rates
was determined by the Arrehenius parameters for the
apparent first-order rate constants (ln kapp). The Arrhe-
nius plot for the CuBr/PPMI/BPN-catalyzed polymeri-
zation of GMA is shown in Figure 12. The apparent
activation energy (Ea) calculated for the CuBr/PPMI/
BPN initiation system is 86.35 kJ /mol. Such an appar-
ent activation energy is higher than that reported for
the homogeneous dNbpy/CuBr ATRP system (50 kJ /
3
1
mol)
En d-Gr ou p An alysis of P GMA-Br . End-group analy-
sis of ATRP polymers is important because polymer
chains with halogen end groups act as a macroinitiator.
It can be reactivated in the presence of the ATRP

Macromolecules, Vol. 37, No. 10, 2004
ATRP of Glycidyl Methacrylate 3621
of the bromine end groups is present in PGMA. The
expanded ESI MS region is shown in Figure 13b. This
is also a supporting evidence for the observed high
thermal stability of PGMA-Br prepared by CuBr/BPN,
suggesting virtually the absence of termination reac-
tions.
gation, and therefore better control of molecular weight
and polydispersities. The use of the CuBr/ClPN initia-
tion system essentially proceeds through an uncon-
trolled polymerization, which is likely due to the strong
C-Cl bond and inefficient initiation. The variation of
alkyl chain length from n-propyl to n-octyl does not seem
to have an effect on the rate of polymerization and at
least 2 mol equiv ligand to copper salt being required
for optimum rate of polymerization. Increase in ligand
concentration slightly decreases the rate of polymeri-
zation. ESI MS suggests that there was no significant
loss of bromine end groups present in the polymer chain,
and doubly charged species were more intense than
singly charged species. The TG spectra of PGMA
indicate that the CuBr/PPMI/BPN system show an
absence of head-to-head, vinylidene end groups, and
high thermal stability compared to the mixed halide
initiation system and free radical polymerization.
Th er m a l Sta bility of P GMA. The thermal stability
of polymer is in direct relationship with the polymer
structure, and therefore one could expect desirable
information on the regioselectivity and the extent of
termination reactions during the polymerization pro-
1
3,32
cess.
Figure 14a shows the TG spectra of PGMA
synthesized via five different initiating systems: (1) by
conventional free-radical polymerization (FRP) using
3
3
AIBN initiator (Mn ) 91 850, Mw/Mn ) 1.49); (2) by
ATRP (a) CuBr/BPN (Mn ) 6800, Mw/Mn ) 1.30), (b)
CuBr/ClPN (Mn ) 15 750, Mw/Mn ) 1.76), (c) CuCl/ClPN
Mn ) 19 560, Mw/Mn ) 2.14),34 and (d) CuCl/BPN (Mn
(
)
8100, Mw/Mn ) 1.35). The DTG spectra of PGMA
Ack n ow led gm en t. Mr. R. Krishnan thanks the
Council of Scientific and Industrial Research (CSIR,
New Delhi), India for an award of Senior Research
Fellowship.
synthesized via five different initiating systems is shown
in Figure 14b. PGMA produced by CuCl/BPN displays
a three-step degradation corresponding to (a) head-to-
head linkage (around 80-190 °C), (b) chain-end initia-
tion from the vinylidine ends (around 190-237 °C), and
Su p p or tin g In for m a tion Ava ila ble: The preparation
(
2
c) a random scission with in the polymer chain (around
37-377 °C) (Figure 14b). The first two steps (a, b) are
1
13
and characterization of ligands (FT-IR, H and C NMR),
1
polymers ( H NMR, TG/DTA), and kinetic results. This mate-
more pronounced than in PGMA prepared from CuBr/
BPN, indicating the termination reactions resulting in
head-to-head linkages and vinylidene end groups. Ther-
mal degradation of PGMA synthesized by ATRP with
the CuBr/PPMI/BPN initiation system shows a single
step degradation around 400 °C, originating only from
random scission of polymer chain. This polymer is about
rial is available free of charge via the Internet at http://
pubs.acs.org.
Refer en ces a n d Notes
(
1) Szwarc, M. Nature (London) 1956, 178, 1168.
(2) Szwarc, M.; Levy, M.; Milkovich, R. J . Am. Chem. Soc. 1956,
7
8, 2657.
8
0 °C more stable than the PGMA produced by the FRP
(
3) Webster, O. W. Science 1991, 251, 887.
method. This result further indicates the absence of
abnormal linkages, such as head-to-head, and vi-
nylidene end groups. Therefore, this confirms the high
regioselectivity and complete absence of termination
reactions when using the CuBr/BPN system, whereas
the other three combinations such as CuCl/ClPN, CuCl/
BPN, and CuBr/ClPN show three-step degradations like
FRP, indicating head-to-head and vinylidene linkages
present in the polymer chains. This suggests that the
GMA polymerization by ATRP other than CuBr/BPN
systems shows the extent of termination reactions
during the polymerization.
(4) (a) Georges, M. K.; Veregin, R. P. N.; Kazmaier, P. M.; Hamer,
G. K. Macromolecules 1993, 26, 2987. (b) Hawker, C. J .
Trends Polym. Sci. 1994, 4, 183. (c) Hawker, C. J .; Bosmon,
A. W.; Harth, E. Chem. Rev. 2001, 101, 3661.
(
5) Wang, J .-S.; Matyjaszewski, K. J . Am. Chem. Soc. 1995, 117,
5614.
(6) Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T.
Macromolecules 1995, 28, 1721.
(
7) Chiefari, J .; Chong, Y. K.; Ercole, F.; Krstina, J .; J effery, J .;
Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G. G.; Moad, C. L.;
Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecules 1998,
31, 5559.
(
8) Webster, O. W.; Hertler, W. R.; Sogah, D. Y.; Farnham, W.
B.; Rajan Babu, T. V. J . Am. Chem. Soc. 1983, 105, 5706.
9) (a) Wang, J .-S.; Matyjaszewski, K. Macromolecules 1995, 28,
(
7
572. (b) Qiu, J .; Matyjaszewski, K. Macromolecules 1997,
Con clu sion s
30, 5643.
(
10) (a) Wang, J .-S.; Matyjaszewski, K. Macromolecules 1995, 28,
A range of Schiff base ligands, initiators, and solvents
in conjunction with Cu(I)X have been shown to be
effective for the homogeneous controlled atom transfer
radical polymerization of glycidyl methacrylate at room
temperature. Kinetic results showed an apparent 0.8
order dependence of the rate of polymerization on the
concentration of initiator, and there is slight deviation
from the first-order kinetics. Bulk and diphenyl ether
solution polymerization show a better control of polym-
erization with quite narrow polydispersities. No polym-
erization in nonpolar solvents, deviation from theoreti-
cal molecular weight, and slightly higher polydispersities
were observed in polar solvents at room temperature.
BPN and M 2-BP show a better control of polymeriza-
tion with faster initiation and slower propagation, and
7901. (b) Grimaud, T.; Matyjaszewski, K. Macromolecules
1
997, 30, 2216.
(
11) Matyjaszewski, K.; J o, S. M.; Paik, H.-J .; Gaynor, S. G.
Macromolecules 1997, 30, 6398.
12) Percec, V.; Barboiu, B.; Neumann, A.; Ronda, J . C.; Zhao, M.
Macromolecules 1996, 29, 3665.
(
(13) Granel, C.; Dubois, P.; J erome, R.; Teyssie, P. Macromolecules
1
996, 29, 8576.
(
14) Lecomte, P.; Drapier, I.; Dubois, P.; Teyssie, P.; J erome, R.
Macromolecules 1997, 30, 7631.
15) (a) Louie, J .; Grubbs, R. H. Chem. Commun. 2000, 1479. (b)
Ando, T.; Kamigaito, M.; Sawamoto, M. Macromolecules 1997,
(
3
0, 4507.
16) (a) Matyjaszewski, K.; Xia, J . Chem. Rev. 2001, 101, 2921.
b) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001,
01, 3686 and references therein.
(
(
1
(17) Wang, X.-S.; Luo, N.; Ying, S. K. Polymer 1999, 40, 4157.
(18) (a) Teodorescu, M.; Matyjaszewski, K. Macromolecules 1999,
3
2, 4826. (b) Teodorescu, M.; Matyjaszewski, K. Macromol.
1
-PEBr shows a curvature in first-order kinetic plot and
Rapid Commun. 2000, 21, 190. (c) Rademacher, J . T.; Baum,
M.; Pallack, M. E.; Brittain, W. J .; Simonsick, W. J . Macro-
molecules 2000, 33, 284. (d) Zhang, H.; Xia, J .; Matyjaszewski,
K. Macromolecules 1998, 31, 5167.
uncontrolled molecular weight was observed, indicating
that termination was taking place. The mixed halide
system CuCl/BPN gave faster initiation, slower propa-

3
622 Krishnan and Srinivasan
Macromolecules, Vol. 37, No. 10, 2004
(
(
(
19) Wang, X. S.; J ackson, R. A.; Armes, S. P. Macromolecules
(28) Dayananda, K.; Dhamodharan, R. J . Polym. Sci., Part A:
Polym. Chem. 2004, 42, 902.
(29) Grubbs, R. B.; Dean, J . M.; Broz, M. E.; Bates, F. S.
Macromolecules 2000, 33, 9522.
(30) Matyjaszewski, K.; Shipp, D. A.; Wang, J . L.; Grimaud, T.;
Patten, T. E. Macromolecules 1998, 31, 6836.
(31) Matyjaszwski, K.; Xia, J .; Patten, T. E.; Abernathy, T. J . Am.
Chem. Soc. 1997, 119, 674.
(32) (a) Norman, J .; Moratti, S. C.; Slark, A. T.; Irvine, D. J .;
J ackson, A. J . Macromolecules 2002, 35, 8954. (b) Wunder-
lich, W.; Bentaremo, N.; Klapper, M.; Mullen, K. Macromol.
Rapid Commun. 1996, 17, 433.
(33) Free radical polymerization was carried out in toluene at 60
°C. Reaction conditions: [GMA]0 ) 7.3 M; [AIBN] ) 0.036
M; monomer:solvent ) 50% v/v.
(34) Reaction conditions: [GMA]0 ) 7.3 M; [ClPN]0 ) [CuCl]0 )
[PPMI]0/2 ) 0.073 M at 90 °C.
2
000, 33, 255.
20) Wang, X. S.; Malet, F. L. G.; Armes, S. P.; Haddleton, D. M.;
Perrier, S. Macromolecules 2001, 34, 162.
21) (a) Zeng, F.; Shen, Y.; Zhu, S.; Pelton, R. J . Polym. Sci., Part
A: Polym. Chem. 2000, 38, 3821. (b) Zeng, F.; Shen, Y.; Zhu,
S.; Pelton, R. Macromolecules 2000, 33, 1628.
(
(
(
(
(
(
22) Kim, J .-B.; Bruening, M. L.; Baker, G. L. J . Am. Chem. Soc.
2
000, 123, 7616.
23) Haddleton, D. M.; J asieczek, C. B.; Hannon, M. J .; Shooter,
A. J . Macromolecules 1997, 30, 2190.
24) Haddleton, M.; Crossman, M. C.; Hunt, K. H.; Topping, C.;
Waterson, C.; Suddaby, K. G. Macromolecules 1997, 30, 3992.
25) Haddleton, M.; Clark, A. J .; Duncalf, D. J .; Heming, A. H.;
Kukulj, D.; Shotter, A. J . J . Mater. Chem. 1998, 1525.
26) Haddleton, M.; Kukulj, D.; Duncalf, D. J .; Heming, A. H.;
Shotter, A. J . Macromolecules 1998, 31, 5201.
27) Krishnan, R.; Srinivasan, K. S. V. Macromolecules 2003, 36,
1
769.
MA0359032