Enhanced Spin-capturing polymerization and radical coupling mediated by cyclic nitrones

Article abstract of DOI:10.1071/CH12182

A series of cyclic nitrones have been tested for their spin-trapping activity in the enhanced spin-capturing polymerization of styrene and in nitrone-mediated radical coupling reactions. rac-2-Isopropyl-2,3-dimethyl-1-oxy- 2,3-dihydro-imidazol-4-one was found to be the most efficient nitrone. The specific polystyrene macroradical addition rate to this nitrone was determined to be 8.010³Lmol^-1s^-1, which is by a factor of 10 higher than for previously studied compounds. Via enhanced spin-capturing polymerization, polymers in the range of oligomers to 30000gmol^-1 were obtained. A strong dependence of molecular weight on monomer conversion was observed, which can be explained by the high trapping rate. In nitrone-mediated radical coupling, almost ideal coupling of bromine-functional polymers was obtained and the successful introduction of the residual alkoxyamine functionality confirmed.

Full text of DOI:10.1071/CH12182

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem.
Full Paper
http://dx.doi.org/10.1071/CH12182
Enhanced Spin-capturing Polymerization and Radical
Coupling Mediated by Cyclic Nitrones
A
A
B
Kayte Ranieri, Matthias Conradi, Pierre-Yves Chavant, Veronique
B
C
,
D
,
A D
Blandin, Christopher Barner-Kowollik,
and Thomas Junkers
A
Polymer Reaction Design Group, Institute for Materials Research, Universiteit Hasselt,
Agoralaan, Gebouw D, BE-3590 Diepenbeek, Belgium.
B
D e´ partement de Chimie Mol e´ culaire, UMR-5250, ICMG FR-2607, CNRS,
Universit e´ Joseph Fourier, BP-53, 38041 Grenoble Cedex 9, France.
C
Preparative Macromolecular Chemistry, Institut f u¨ r Technische Chemie und
Polymerchemie, Karlsruhe Institute of Technology (KIT), Engesserstrasse 18,
76128 Karlsruhe, Germany.
D
Corresponding authors. Email: christopher.barner-kowollik@kit.edu;
thomas.junkers@uhasselt.be
A series of cyclic nitrones have been tested for their spin-trapping activity in the enhanced spin-capturing polymerization
of styrene and in nitrone-mediated radical coupling reactions. rac-2-Isopropyl-2,3-dimethyl-1-oxy-2,3-dihydro-imidazol-
4
-one was found to be the most efficient nitrone. The specific polystyrene macroradical addition rate to this nitrone was
3
ꢁ1 ꢁ1
determined to be 8.0 ꢀ 10 L mol
s , which is by a factor of 10 higher than for previously studied compounds.
ꢁ1
Via enhanced spin-capturing polymerization, polymers in the range of oligomers to 30000 g mol were obtained.
A strong dependence of molecular weight on monomer conversion was observed, which can be explained by the high
trapping rate. In nitrone-mediated radical coupling, almost ideal coupling of bromine-functional polymers was obtained
and the successful introduction of the residual alkoxyamine functionality confirmed.
Manuscript received: 5 April 2012.
Manuscript accepted: 1 May 2012.
Published online: 16 July 2012.
Introduction
[3,12,13]
traps as well).
The so-formed macronitroxide is then
Nitrone radical spin traps – mostly known from the area of
electron-spin resonance – are efficient mediators in radical
available to terminate the growth of other growing chains
(see Scheme 1). Thus, each spin-trapped molecule ends the
growth cycle of two radicals. If a sufficiently high spin-trapping
rate is achieved, conventional termination of chains is effectively
suppressed owing to the persistent radical effect, and all
produced polymers feature an alkoxyamine functionality
(precisely) in the middle of the chain. ESCP – as it is a strictly
non-living process – thus yields polymers with a dispersity close
to the ideal case of a termination by a combination-dominated
system. Narrow-molecular-weight analogues are obtained when
the radical source is changed, for example by using ATRP
polymers that release macroradicals of defined size and
[1]
reactions. When added directly to a free-radical polymeriza-
tion, control over molecular weight is achieved and polymers
bearing an alkoxyamine functionality in the backbone in a
midchain position are obtained (see Scheme 1). This process,
which was termed enhanced spin-capturing polymerization
(ESCP), was shown to operate with many different
monomers.
[
2–4]
ESCP is somewhat related to in situ nitroxide-
[
5,6]
mediated polymerization ; however, it should not be con-
fused with a living process such as in situ nitroxide mediated
polymerization (NMP) itself, classical NMP,
addition–fragmentation transfer (RAFT) polymerization or
atom-transfer radical polymerization, ATRP. Alternatively,
nitrones can also be employed as mediators in nitrone-mediated
radical coupling (NMRC) reactions
[
7]
reversible
[
8]
dispersity on reaction with copper catalysts, as in the so-called
[14]
[
9]
atom-transfer radical coupling method.
In this process,
NMRC, essentially the same spin-trapping reaction occurs and
the principle of product formation is identical, which is impor-
tant in the sense that kinetic information gathered on one
mechanism also holds for the other.
[
10,11]
to couple preformed
polymers with defined molecular weight distributions. Both
processes allow the efficient synthesis of midchain-functiona-
lized macromolecules and the only difference in the products is
evident in their disparate dispersities: ESCP generally yields
materials with polydispersity indexes (PDI) close to 1.5,
whereas NMRC is able to produce materials with dispersities
close to unity.
The midchain-functional alkoxyamine polymers can be
employed in post-polymerization reactions in two ways: one is
reactivation of the chains in a nitroxide-mediated polymeriza-
[4]
tion, thus using the ESCP or NMRC product effectively as a
macroinitiator; the other is to employ the nitrone or alkoxyamine
respectively as a functional carrier that introduces a certain
functional group at a precisely controlled position in the chain.
In ESCP, growing macroradicals are trapped by a spin trap
usually irreversibly by nitrones, but other compounds can act as
(
Journal compilation Ó CSIRO 2012
www.publish.csiro.au/journals/ajc

B
K. Ranieri et al.
ꢀ
O
ꢁ
N
N
Nitrone
ATRP polymer
O
Br
NMRC
Cu(0)
ꢀ
CuBr
•
O
N
•
ESCP
N
O
Macroradical
Macronitroxide
•
ꢁ
O
N
O
N
ꢂ Initiator
ꢂ Repeating unit
Mid-chain functionalized polymer
Scheme 1. General reaction mechanism for nitrone-mediated radical coupling (NMRC) and enhanced spin-capturing
polymerization (ESCP).
[
18–20]
glycine synthons.
The activity of the compounds is
assessed on the basis of their performance in ESCP of styrene
as well as in NMRC reactions. As will be described below,
especially compound 2 exhibits high spin-trapping rates and is
thus the main focus of the present kinetic studies.
ꢀ
ꢀ
ꢁ
ꢀ
O
ꢁ
O
O
ꢁ
N
N
N
N
N
N
O
O
O
1
2
3
Experimental Section
Materials
0
0
0
Methyl2-bromopropionate (MBrP, Aldrich, 98%), N,N,N ,N ,N -
pentamethyldiethylenetriamine (PMDETA, Acros, 99 %),
Scheme 2. Structures of nitrones under investigation.
0
copper powder ,63mm (Cu , Merck, 99.7 %), toluene (VWR,
9
9 %), tributyl tin hydride (TBTH, Aldrich, 97 %), tetrahydro-
furan (THF, VWR, 99 %), dimethyl sulfoxide (DMSO, VWR,
9 %) and dimethyl formamide (DMF, VWR, 99 %) were used
In previous studies, we used alkyne-functional phenyl nitrones
based on the commercially available N-tert-butyl phenylnitrone,
PBN, for the modular design of, for example, star and H-shaped
9
as received. Styrene (Sty, Aldrich, 99 %) monomer was
de-inhibited by percolating over a column of basic alumina.
Copper(I) bromide (CuBr, Acros, 98 %) was washed with glacial
acetic acid at 808C overnight to remove any soluble oxidized
species before being filtered, rinsed with ethanol and dried. The
[
15–17]
polymers.
Generally, PBN and PBN-type nitrones are
good and versatile spin traps that are associated with reasonable
trapping rates. For most styrene polymerization, the addition
rate coefficient for the addition of macroradicals (kad,macro) to
PBN was determined to be similar to the monomer propagation
0
initiator 2,2 -azobisisobutynitrile (AIBN, Aldrich, 98 %) was
recrystallized twice from methanol.
rate coefficient (k ), leading to so-called spin-trapping constants
p
[
3]
CSC close to unity. As can be readily inferred from the kinetic
equations, more effective spin traps may impose some
difficulties in ESCP, but may be highly favourable in the NMRC
process, because a higher kad,macro would result in lower
amounts of nitrone required for effective mediation of the
coupling reactions. It is thus a constant effort to explore
alternative types of nitrones that may be more effective spin
traps for macroradicals.
Nitrones 2 (rac-2-isopropyl-2,3-dimethyl-1-oxy-2,3-dihydro-
[18,19]
imidazol-4-one)
oxy-5-phenyl-2,3-dihydro-imidazol-4-one)
according to published methods.
and 3 (rac-2-isopropyl-2,3-dimethyl-1-
[20]
were prepared
Preparation of 1-Methyl-4-oxy-1,4-diaza-spiro[4.5]dec-
-en-2-one 1
3
In the present contribution, we thus focus on a series of
,2-dialkyl-3-methyl-1-oxy-2,3-dihydro-imidazol-4-ones (see
In a 100-mL Erlenmeyer flask under magnetic stirring,
methylamine (8 M in ethanol, 32 mL, 0.25 mol) was added to
glycine ethyl ester hydrochloride (7.0 g, 50 mmol). After 16 h at
2
Scheme 2), which were originally designed for use as chiral

Polymerization and Radical Coupling Mediated by Nitrones
C
ambient temperature, the crude product was taken up in ethanol
(100 mL) and the solvent and excess methylamine were
evaporated. This operation was repeated twice to give the
amino-amide hydrochloride (7.8 g, quantitative) as an
amorphous solid that was introduced into a 100-mL flask fitted
with a reflux condenser. Ethanol (50 mL), cyclohexanone
Synthesis of polystyrene (PS) by ATRP
To a 100-mL round-bottom flask were added CuBr (143 mg,
mmol), Sty (50 mL, 435 mmol) and MBrP (112 mL, 0.8 mmol)
and the solution was degassed by purging with nitrogen for
0 min in an ice bath. Degassed PMDETA (418 mL, 2 mmol)
was transferred into the flask via a degassed syringe. The
reaction mixture was heated at 808C in a thermostatted oil bath
for 90 min and the polymerization was stopped by immersing
the sealed flask into an ice bath before opening to air. Fresh THF
was added to the flask and the polymer was purified by passing
over a column of silica gel in order to remove the copper com-
plexes. The copper-free polymer solution was subsequently
1
3
(
10.4 mL, 100 mmol), triethylamine (7.0 mL, 50 mmol) and
˚
activated 4-A molecular sieves (beads, 6 g) were added and the
reaction mixture was heated for 16 h at reflux without stirring.
The sieves were filtered off and the solvent was evaporated
under reduced pressure. The crude slurry was taken up in ethyl
acetate (250 mL) and washed with dilute HCl (50 mL), then
water (2 ꢀ 50 mL), and dried over anhydrous sodium sulphate.
Evaporation of the solvent yielded the imidazolidinone
concentrated and precipitated in a cold methanol/water mixture
ꢁ1
(
(
4 : 1) yielding 1.3 g of polymer with Mn,SEC ¼ 1670 g mol
1
8
-methyl-1,4-diaza-spiro[4.5]decan-2-one (7.40 g, 44 mmol,
8 %) as a brown oil that crystallized on standing, mp 1018C.
number-average molecular weight determined by size-
ꢁ1
exclusion chromatography), MnNMR ¼ 1730 g mol (number-
average molecular weight determined by nuclear magnetic
resonance spectroscopy) and PDI ¼ 1.10.
ꢁ
1
nmax (KBr)/cm 3299, 2931 2852, 1679. d (300 MHz, CDCl )
H
3
3
CDCl ) 173.1 (C), 79.4 (C), 48.1 (CH ), 33.9 (CH ), 24.9 (CH ),
.45 (2H, s), 2.78 (3H, s), 1.73–1.53 (10H, m). d (100 MHz,
C
3
2
2
2
þ
2
2.4 (CH ). High-resolution (HR)MS (electrospray ionization )
2
Typical Procedure for the Radical–Radical Conjugation
Reactions with 2
þ
m/z 191.11522 [M þ Na] ; C H N ONa requires 191.11548.
9
16
2
In a 250-mL flask were introduced methanol (70 mL) and
methyltrioxorhenium (MTO, 12 mg, 46 mmol), followed by urea
hydroperoxide (UHP, 5.22 g, 55 mmol). A solution of diaza-
spiro[4.5]decan-2-one (7.40 g, 44 mmoles) in methanol (30 mL)
was then added in one portion at 258C. Fresh fractions of MTO
0
ATRP-made PS (0.1 mmol), 2 (0.5 mmol), Cu (0.1 mmol) and
I
Cu Br (0.01 mmol) were dissolved in DMSO (500 mL) in a 5-mL
ground-glass test tube (to make up a 0.14 M solution of ATRP
PS) and purged with nitrogen for 10 min in an ice bath. Degassed
PMDETA (0.1 mmol) was transferred into the flask via a
degassed syringe. The reaction mixture was heated at 608C in a
thermostatted oil bath for 20 h. The resulting conjugated PS was
purified by passing over a column of silica gel to remove copper
complexes. The copper-free polymer solution was subsequently
concentrated and precipitated in a cold methanol/water (4 : 1)
(12 mg) were added when the yellow coloration faded, approxi-
mately every 30 min. The reaction was monitored by TLC and
conversion of the starting material was complete within 3 h.
MnO (16.07 g, 185 mmol) was subsequently added portion-
2
wise, both to destroy the excess hydroperoxide and to complete
the oxidation of the intermediate hydroxylamine. The reaction
mixture was stirred for 2 h at room temperature and concentrated
under reduced pressure. The residue was taken up in dichlor-
omethane (50 mL) and anhydrous Na SO was added. The
1
mixture before being analysed by H NMR and THF-SEC.
Typical Procedure for Polymer Quenching
2
4
In a 50-mL round-bottom flask, coupled PS (10 mmol) was
dissolved in DMF (538 mL) before adding TBTH (538 mL,
solids were removed by filtration over Celite and rinsed with
ethyl acetate. The filtrate was evaporated to dryness and the
crude product was purified by filtration over silica gel (ethyl
acetate/cyclohexane 80 : 20) to yield 1 as a white solid (5.55 g,
2
mmol) such that the ratio of polymer to quencher was 1 : 200.
The solution was heated at 1258C overnight. The solvent was
removed by evaporation overnight before the quenched polymer
was analyzed by THF-SEC.
ꢁ1
3
2
3
1
2
1 mmol, 66 %), mp 95.7–95.88C. nmax (KBr)/cm
2924,
854, 1701, 1550, 1260. dH (300 MHz, CDCl ) 7.03 (1H, s),
3
.10 (3H, s), 2.24–1.46 (10H, m). dC (100 MHz, CDCl3)
62.2 (C), 124.1 (CH), 89.8 (C), 33.8 (CH ), 26.2 (CH ),
Characterization
2
3
3.6 (CH ), 21.2 (CH ). MS (desorption/chemical ionization)
2
Analysis of the molecular weight distribution of the polymer
samples was performed on samples diluted with THF, with a
TOSOH EcoSEC HLC-8320GPC, comprising an autosampler,
a PSS SDV 5 microguard (5 cm), followed by three PSS SDV 5
linear XL (30-cm) columns thermostatted at 408C (column
2
þ
þ
m/z 167.5 (36), 183.3 (100, [M þ H] ), 200.0 (27, [M þ NH ] ).
4
Found: C 59.53, H 7.89, N 15.51. C H N O requires C 59.32,
9
14 2 2
H 7.74, N 15.37 %.
Synthesis of ESCP Polymers
2
6
ꢁ1
molecular weight range: 1 ꢀ 10 –1 ꢀ 10 g mol ). THF was
ꢁ
1
In a 25-mL round bottom flask, AIBN (0.17 g, 1 mmol) and
(0.34 g, 2 mmol) were dissolved in styrene (10 mL, 87 mmol).
used as the eluent with a 1.0 mL min flow rate and toluene as
flow marker. Calibration was performed with polystyrene
standards from PSS Laboratories in the range of 470–
7.5 ꢀ 10 g mol . H NMR spectra were recorded with two
NMR spectrometers (300 and 400 MHz, 64 scans, 12-s relaxa-
tion) from Oxford Instruments Ltd using a Varian probe (9 mm–
four-nucleus AutoSWPFG) in CDCl3.
2
The solution was then divided into 10 portions and placed into
glass vials, which were subsequently sealed with rubber septa.
The solutions were degassed with nitrogen for ,2 min. The vials
were heated in an 808C thermostatted oil bath. After
predetermined reaction times, the polymerization of individual
vials was stopped by immersing the sealed vials into an ice bath
before opening to the air. The contents of the vials were
transferred into aluminium pans and the conversion was
determined gravimetrically by weighing the samples before and
after evaporation of the residual monomer. All samples were
subjected to molecular weight characterization without further
treatment.
6
ꢁ1
1
Results and Discussion
The addition rate coefficient of macroradicals kad,macro to the
spin trap is a crucial parameter for both ESCP and NMRC
reactions. Similarly to the chain-transfer constant in classical
chain-transfer polymerization, a spin-capturing constant CSC

D
K. Ranieri et al.
[2] ꢂ 0.01 mol Lꢀ
1
can be defined, which is the ratio of kad,macro over the specific
propagation rate k of the monomer in use. For ESCP, this
1
3
6
9
5 minutes
0 minutes
0 minutes
0 minutes
p
constant is a direct measure of quality of the spin-trapping
activity because macroradical addition to the nitrone is in direct
competition with the propagation reaction. A higher CSC can
thus be directly translated into lower concentrations that are
required to achieve good control over molecular weight. For
NMRC, the pure addition rate is of higher interest, because there
the addition is in competition with the conventional termination
reaction.
120 minutes
180 minutes
The CSC constant can be directly determined from the change
in molecular weight with varying nitrone concentration. Analo-
[
2,3]
gous to the Mayo equation, a simple relation can be derived
:
cnitrone
ꢁ
1
ꢁ1
DP ¼ DP þ 0:5CSC
ð1Þ
n
1
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
cm
log M [g molꢀ1]
where DP is the degree of polymerization, infinity marks the
n
chain length of polymer obtained from uncontrolled polymeri-
zation under otherwise identical reaction conditions, cnitrone is
the nitrone concentration and c is the monomer concentration.
m
Fig. 1. Molecular weight distribution of polymer generated via enhanced
ꢁ1
spin-capturing polymerization from styrene with a [2] ¼ 0.01 mol L at
08C for variable reaction times.
8
Thus, from plotting the inverse of the degree of polymerization
of polymer obtained from ESCP v. the nitrone concentration
employed, the spin-trapping constant (and thereby kad,macro if the
propagation rate coefficient is known) is directly accessible.
ESCP serves thus not only as a polymerization method, but also
as a kinetic experiment to derive the spin-trap specific kinetic
parameters.
profound change in molecular weight with increasing polymer-
ization time and monomer conversion was observed at the same
time. What may seem to be surprising at first glance – usually
more or less constant molecular weight is observed over the
course of polymerization – can be readily explained on the basis
of Eqn 1. If CSC is significantly above unity, then the nitrone is
All nitrones shown in Scheme 2 were assessed for their
performance in styrene polymerizations. Styrene is a suitable
monomer for probing the kinetics because practically all
relevant parameters for this system are known. In principle,
good performance during polymerization can a priori be
expected for 1 and 2. Both nitrones are structurally relatively
similar and only differ in the steric shielding caused from the
bulky alkyl residues. Compound 3 carries an additional phenyl
substituent on the C¼N double bond. Such fully substituted
nitrones are relatively less accessible and have therefore been
less studied; they could differ from H-substituted compounds
both electronically and sterically. Interestingly, when the
compounds were screened, very different results were obtained
from the individual compounds. 1 resulted in polymers of
consumedata faster ratethanthe monomerandthe ratio cnitrone/c
m
decreases with progressing polymerization. The result is then a
profound increase of molecular weight with conversion.
Fig. 1 depicts the molecular weight distributions of polysty-
ꢁ1
rene prepared via ESCP mediated by 0.01 mol L 2 at 808C
after different reaction times. As can be seen, initially, quite low
molecular weights are obtained, but a strong shift of the
molecular weight distributions occurs over the course of the
reaction. Whereas after 30-min reaction time polymers with an
average molecular weight of few thousands are obtained, a more
than 10-fold increase is observed after 3 h. This increase in
molecular weight should not be confused with ‘livingness’ of the
reaction. All chains are irreversibly terminated within a short
time span and the increase in molecular weight is a mere
consequence of the polymer with higher masses overshadowing
the material formed earlier in polymerization. Regardless, Fig. 2
ꢁ
1
molecular weights in the range of 2000 to 30000 g mol when
ꢁ1
nitrone concentrations between 0.4 and 2.5 ꢀ 10ꢁ mol L
2
were employed. Although a respectable variation in polymer
size was achieved, no improvement over the N-alkyl phenyl
nitrones employed in previous studies could be seen. 3 resulted
in even less effective chain-length control. Some variation in
molecular weight with nitrone concentration was observed, yet
significantly below the activity of 1. 2, however, exhibited very
good control over styrene polymerization. Even small concen-
depicts the change of the number-average molecular weight M
with time (Fig. 2a) and conversion (Fig. 2b).
n
The steady increase in molecular weight with proceeding
polymerization is nicely shown in Fig. 2a. An almost linear
increase is observed, with molecular weights ranging from as
ꢁ1
ꢁ1
low as 1000 g mol to close to 30000 g mol at later stages.
When plotted against conversion, one can, however, observe
that the increase is not completely linear and a curved plot is
obtained, in line with the above explanation of changing
concentration ratios (the largest effect is seen at early stages
of the polymerizations, where comparatively more nitrone is
consumed than monomer). It should be noted that the data show
some scatter, which is a result of the relatively broad distribu-
tions that are obtained. In addition, in Fig. 1 it is already evident
that some distributions are somewhat broader than others. The
cause of such an effect, which we have also observed in methyl
trations such as 2.5 ꢀ 10ꢁ mol L resulted in polymers with a
2
ꢁ1
ꢁ
1
molecular weight significantly below 5000 g mol . Although
the lowered activity of 3 may be easily explained on the basis of
steric hindrance, the difference between 1 and 2 comes as a
significant surprise. In principle, both the isopropyl and methyl
substituent are comparatively bulky when compared with the
cyclohexyl group. It may be hypothesized that the isopropyl
group leads in principle to a more pronounced shielding from
one side and thus radicals will preferentially attack from the side
of the methyl residue. Regardless, considering all factors,
[
21]
2
appears to be nevertheless a better macroradical trap than 1.
To date, 2 is thus the most effective spin trap for polymeri-
zation control that was tested. On the downside, however, a
methacrylate polymerization mediated by PBN,
is not
completely known and seems to be not fully reproducible (it is
important to add that all data in the present study were obtained

Polymerization and Radical Coupling Mediated by Nitrones
E
(
a)
0.12
4
3
3
2
2
1
1
0000
5000
0000
5000
0000
5000
0000
0.10
0.08
0.06
0.04
0.02
0
5
000
0
0
20
40
60
80 100 120 140 160 180 200
0
0.02
0.04
0.06
0.08
0.10
0.12
0.14
t [min]
[mol Lꢀ1]
2
(
b) 40000
n
Fig. 3. Inverse degree of polymerization (DP ) plotted as a function of [2]
in the enhanced spin-capturing polymerization of styrene at 808C and 30 min
ꢁ
3
3
2
2
1
1
5000
0000
5000
0000
5000
0000
1
with 0.02 mol L AIBN. Polymer samples of low monomer conversions
1–10 %) were used to determine the DP
(
n
.
The two different regions become explainable on the basis of
conversion. The higher the nitrone concentration, the more the
polymerization speed is reduced. Consequently, longer time is
required to reach significant conversions. Thus, even though the
reaction time was already successively reduced when decreas-
ing the nitrone content, differences are seen, and whereas the
data points taken at larger nitrone content represent monomer
conversions of 2–3 %, some data points at low nitrone content go
up to 8–9 % conversion. Thus, taking the increase in molecular
5000
0
0
0.1
0.2
0.3
0.4
0.5
ꢁ1
Conversion, X
weight into account (see Fig. 2b), decreased DPn are to be
expected, resulting in a increased slope of the pseudo-Mayo
plot. For this reason, only the second regime was fitted and used
for the determination of CSC (see dotted line for best fit of the
data) because the results from these concentrations are deemed
to be more reliable. From the slope in conjunction with the initial
monomer concentration, a CSC of 12.2 was deduced, which is –
as expected from the qualitative interpretation above – a
relatively high value. Compared with PBN, nitrone 2 thus
Fig. 2. (a) Time dependence of the number-average molecular weight (M
n
)
for the enhanced spin-capturing polymerization of styrene at 808C and a
ꢁ
1
concentration of 2 of 0.01 mol L ; (b) dependence of number-average
molecular weight on styrene conversion for the same data.
from individual polymerization samples rather than series of
samples taken from a single polymerization), but is limited to
earlier stages of the polymerization.
3
ꢁ1 ꢁ1
ꢁ1 ꢁ1 [22]
shows with 8.0 ꢀ 10 L mol
s
(with k
¼ 660L mol
s )
p
A change in molecular weight during polymerization poses a
problem when Eqn 1 is supposed to be used for CSC determina-
tion. As in most ESCP polymerizations, constant molecular
weight is observed, samples can be taken at any conversion and
seen as representative for the specific reaction conditions. In the
case of 2, this is not possible. The exact concentration of the
nitrone is only known at the start of polymerization and thus
samples need to be taken at negligible monomer consumption.
As this is not easily achieved, exact determination of conver-
sions is required in order to be able to judge the quality of the
kinetic parameters deduced.
Fig. 3 presents the inverse of the degree of polymerization
obtained at different nitrone concentrations. Two linear regions
can be identified: a steep slope for concentrations below
0.02 mol L and a less pronounced, but also steady increase
in the region 0.02–0.12 mol L . Samples were also taken for
a kad,macro higher by a factor of 10, which is an exceptional
spin-trapping performance.
Even though the benefits of the pseudo-living behaviour of
the above polymerizations still needs to be fully explored, one of
the advantages of ESCP is, however, that constant-molecular-
weight material is formed throughout the reaction. Thus, in
many cases, a CSC of approximately unity is favourable and
nitrone 2 may not be the best option to be used directly in ESCP.
In NMRC, where no competition with the propagation reaction
exist, a higher kad,macro is a direct benefit because it allows – in
principle – high-quality coupling results even when a reduced
nitrone concentration is employed. Thus, we tested 2 for use in
NMRC reactions. ATRP-made polystyrene with an
ꢁ
1
average molecular weight of 1670 g mol was reacted with
ꢁ
1
0
the nitrone in presence of Cu /CuBr, the ligand PMDETA and
ꢁ1
the solvent DMSO. After an extended reaction time of 20 h,
successful coupling of polymer chains was observed.
ꢁ1
nitrone concentrations of up to 0.4 mol L . As these conditions,
however, only allow for production of oligomers (at low
conversions), they were not included in the graph as M values
determined from SEC are obscured by the limitations in calibra-
tion accuracy in this molecular weight regime as well as by
overlap of polymer distributions with the residual control agent.
Fig. 4 shows the respective weight distributions and the
corresponding molecular weights are given in Table 1. An
almost perfect coupling is achieved with precise doubling in
molecular weight and a slight reduction in PDI for the coupled
product. In order to prove that the coupling was indeed promoted
n

F
K. Ranieri et al.
the average molecular weight of polystyrene increases signifi-
cantly during polymerization. With regards to NMRC, almost
ideal coupling of ATRP-made chains can be observed under
PS-Br polymer
Nitrone coupled
Quenched polymer
[
23]
single-electron-transfer conditions
and the presence of
alkoxyamine linkages in the coupled product was confirmed via
a quenching experiment.
Based on nitrone 2, further structures will be tested. The full
potential of ESCP and NMRC are only accessed when distinct
functionalities are introduced into the chains; thus the synthesis
of functionalized derivatives is a priority. In addition, further
structural variations should be carried out. In principle, the
exchange of the isopropyl residue for a methyl group should
allow less steric hindrance and thus even higher macroradical
addition rates. Further, the formed alkoxyamines will be
tested for their reactivity in NMP as these specific compounds
have not yet been used for such purpose. The quenching
experiments performed demonstrate that the combination–
disproportionation equilibrium falls within a suitable kinetic
window, and thus controlled polymerizations to build block
copolymer structures should be feasible.
2
.5
2.7
2.9
3.1
3.3
3.5
3.7
3.9
log M [g molꢀ1
]
Fig. 4. Molecular weight distribution of polystyrene (PS) before nitrone
coupling, after coupling, and quenching of the alkoxyamine bond.
Table 1. Result of coupling reactions of bromine functional poly-
styrene (PS-Br) polymer using nitrone-mediated radical coupling and
quenching of the alkoxyamine bond
Acknowledgements
C. B.-K. acknowledges continued support from the Karlsruhe Institute of
Technology in the context of the Excellence Initiative for leading German
universities. K. R. and T. J. are grateful for funding from the Fonds
Wetenschappelijk Onderzoek (FWO) via project G.0491.11N and the
Odysseus scheme.
M
n
p
, number-average molecular weight; M , peak molecular weight; PDI,
polydispersity index
ꢁ1
ꢁ1
[g mol ]
M
n
[g mol
]
PDI
M
p
PS-Br precursor polymer
Coupled
1700
3500
1700
1.10
1.08
1.10
2000
3800
1900
Quenched
References
[
[
[
1] E. H. H. Wong, T. Junkers, C. Barner-Kowollik, Polym. Chem. 2011,
2, 1008. doi:10.1039/C0PY00377H
2] E. H. H. Wong, T. Junkers, C. Barner-Kowollik, J. Polym. Sci. A
Polym. Chem. 2008, 46, 7273. doi:10.1002/POLA.23025
3] E. H. H. Wong, M. H. Stenzel, T. Junkers, C. Barner-Kowollik,
via the nitrone, a quenching experiment was also performed in
which the alkoxyamine bond was selectively quenched at high
temperatures. Here also, an almost ideal result was observed and
the starting distribution was re-obtained after the quenching.
If the nitrone had not been effective, then conventional termina-
tion events should also have taken place, resulting in stable C–C
bonds that cannot be cleaved. Thus, from the reversion of the
coupling reaction, it can be concluded that all chains in the
coupled product did indeed feature the alkoxyamine moiety. As
a side effect, quenching also demonstrates in this respect that the
alkoxyamine bond can indeed be cleaved at higher tempera-
tures, which makes future chain insertions via nitroxide-
mediated polymerization feasible.
J. Polym. Sci. A Polym. Chem. 2009, 47, 1098. doi:10.1002/POLA.
2
4] T. Junkers, E. H. H. Wong, M. H. Stenzel, C. Barner-Kowollik,
3221
[
Macromolecules 2009, 42, 5027. doi:10.1021/MA900356P
[5] V. Sciannamea, R. Jerome, C. Detrembleur, Chem. Rev. 2008, 108,
1104. doi:10.1021/CR0680540
[6] D. F. Grishin, L. L. Semenycheva, E. V. Kolyakina, Russ. J. Appl.
Chem. 2001, 74, 494. doi:10.1023/A:1012766216412
[
7] C. J. Hawker, A. W. Bosman, E. Harth, Chem. Rev. 2001, 101, 3661.
doi:10.1021/CR990119U
8] (a) J. Chiefari, Y. K. Chong, F. Ercole, J. Krstina, J. Jeffery, T. P. T. Le,
R. T. A. Mayadunne, G. Meijs, C. L. Moad, G. Moad, E. Rizzardo,
S. H. Thang, Macromolecules 1998, 31, 5559. doi:10.1021/
MA9804951
[
Conclusions
(
b) G. Moad, E. Rizzardo, S. H. Thang, Aust. J. Chem. 2005, 58, 379.
A series of cyclic nitrones were assessed for their performance
in ESCP and NMRC. 2-Isopropyl-2,3-dimethyl-1-oxy-
2,3-dihydro-imidazol-4-one 2 exhibited superior performance
and is associated with a very high addition rate of polystyrene
doi:10.1071/CH05072
[9] K. Matyjaszewski, J. Xia, Chem. Rev. 2001, 101, 2921. doi:10.1021/
CR940534G
[
10] E. H. H. Wong, C. Boyer, M. H. Stenzel, C. Barner-Kowollik,
T. Junkers, Chem. Commun. 2010, 1959. doi:10.1039/B925390D
11] A. Debuigne, M. Hurtgen, C. Detrembleur, C. J e´ r oˆ me, C. Barner-
Kowollik, T. Junkers, Prog. Polym. Sci. 2012, 37, 1004. doi:10.1016/
J.PROGPOLYMSCI.2012.01.003
12] T. Junkers, M. H. Stenzel, T. P. Davis, C. Barner-Kowollik, Macromol.
Rapid Commun. 2007, 28, 746. doi:10.1002/MARC.200600885
13] C. Zhang, Q. Wang, J. Polym. Sci. A Polym. Chem. 2011, 49, 612.
doi:10.1002/POLA.24469
3
ꢁ1 ꢁ1
macroradicals to the nitrone of 8.0 ꢀ 10 L mol
s at 808C,
[
which is approximately a factor of 10 higher than for the nitrone
PBN, which is extensively employed to date. It is known from
previous studies that the spin-capturing constant (in the present
case 12.2) is largely applicable also to other monomer families
and thus outstanding performance in acrylate polymerization
can also be expected. Owing to the high spin-capturing activity,
pseudo-living polymerization characteristics are obtained and
[
[
[14] C. Yoshikawa, A. Goto, T. Fukuda, e-Polymers 2002, 13.

Polymerization and Radical Coupling Mediated by Nitrones
G
[
15] E. H. H. Wong, M. H. Stenzel, T. Junkers, C. Barner-Kowollik,
Macromolecules 2010, 43, 3785. doi:10.1021/MA100263K
[20] M. Thiverny, D. Farran, C. Philouze, V. Blandin, P. Y. Chavant,
Tetrahedron Asymmetry 2011, 22, 1274. doi:10.1016/J.TETASY.
2011.07.022
[
16] C. Detrembleur, A. Debuigne, O. Altintas, M. Conradi, E. H. H. Wong,
C. J e´ r oˆ me, C. Barner-Kowollik, T. Junkers, Polym. Chem. 2012, 3,
[21] L. Zang, E. H. H. Wong, C. Barner-Kowollik, T. Junkers, Polymer
2010, 51, 3821. doi:10.1016/J.POLYMER.2010.06.040
[22] M. Buback, R. G. Gilbert, R. A. Hutchinson, B. Klumpermann,
F.-D. Kuchta, B. Manders, K. F. O’Driscoll, G. T. Russell, J. Schweer,
Macromol. Chem. Phys. 1995, 196, 3267. doi:10.1002/MACP.1995.
021961016
135. doi:10.1039/C1PY00297J
[
[
[
17] E. H. H. Wong, O. Altintas, M. H. Stenzel, C. Barner-Kowollik,
T. Junkers, Chem. Commun. 2011, 5491. doi:10.1039/C1CC10322A
18] M. Thiverny, C. Philouze, P. Y. Chavant, V. Blandin, Org. Biomol.
Chem. 2010, 8, 864. doi:10.1039/B918612C
19] M. Thiverny, E. Demory, B. Baptiste, C. Philouze, P. Y. Chavant,
V. Blandin, Tetrahedron Asymmetry 2011, 22, 1266. doi:10.1016/
J.TETASY.2011.07.002
[23] V. Percec, T. Guliashvili, J. S. Ladislaw, A. Wistrand, A. Stjerndahl,
M. J. Sienkowska, M. J. Monteiro, S. Sahoo, J. Am. Chem. Soc. 2006,
128, 14156. doi:10.1021/JA065484Z