Macroinitiator halide effects in galactoglucomannan-mediated single electron transfer-living radical polymerization

Article abstract of DOI:10.1002/pola.24855

Chloro (Cl)- and bromo (Br)-functionalized macroinitiators were successfully prepared from the softwood hemicellulose O-acetylated galactoglucomannan (AcGGM) and then explored and evaluated with respect to their ability and efficiency of initiating single electron transfer-living radical polymerization (SET-LRP). Both halogenated species effectively initiate SET-LRP of an acrylate and a methacrylate monomer, respectively, yielding brushlike AcGGM graft copolymers, where the molecular weights are accurately controlled via the monomer:macroinitiator ratio and polymerization time over a broad range: from oligomeric to ultrahigh. The nature of the halogen does not influence the kinetics of polymerization strongly, however, for acrylate graft polymerization, AcGGM-Cl gives a somewhat higher rate constant of propagation, while methacrylate grafting proceeds slightly faster when the initiating species is AcGGM-Br. For both monomers, the macroinitiator efficiency is superior in the case of AcGGM-Br.

Full text of DOI:10.1002/pola.24855

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Macroinitiator Halide Effects in Galactoglucomannan-Mediated Single
Electron Transfer-Living Radical Polymerization
Ulrica Edlund, Ann-Christine Albertsson
Fibre and Polymer Technology, School of Chemical Science and Engineering, Royal Institute of Technology (KTH), 100 44
Stockholm, Sweden
Correspondence to: A.-C. Albertsson (E-mail: aila@polymer.kth.se)
Received 23 June 2011; accepted 24 June 2011; published online 18 July 2011
DOI: 10.1002/pola.24855
ABSTRACT: Chloro (Cl)- and bromo (Br)-functionalized macroini-
tiators were successfully prepared from the softwood hemicel-
lulose O-acetylated galactoglucomannan (AcGGM) and then
explored and evaluated with respect to their ability and effi-
ciency of initiating single electron transfer-living radical poly-
merization (SET-LRP). Both halogenated species effectively
initiate SET-LRP of an acrylate and a methacrylate monomer,
respectively, yielding brushlike AcGGM graft copolymers,
where the molecular weights are accurately controlled via the
monomer:macroinitiator ratio and polymerization time over a
broad range: from oligomeric to ultrahigh. The nature of the
halogen does not influence the kinetics of polymerization
strongly, however, for acrylate graft polymerization, AcGGM-Cl
gives a somewhat higher rate constant of propagation, while
methacrylate grafting proceeds slightly faster when the initiat-
ing species is AcGGM-Br. For both monomers, the macroinitia-
C
V
tor efficiency is superior in the case of AcGGM-Br.
2011
Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 49:
4139–4145, 2011
KEYWORDS: graft copolymers; hemicellulose; living polymeriza-
tion; macroinitiator; polysaccharides; single electron transfer-
living radical polymerization (SET-LRP); single electron transfer
(SET)
INTRODUCTION The utilization of lignocellulosic feedstock
through the conversion of its components into chemicals,
materials, and energy is the core idea of the wood biorefi-
nery concept. The recovery and purification of noncellulosic
poly and oligosaccharides from such processes is possible
via several pathways and opens up new opportunities for
green material design. In recent years, hemicelluloses have
attracted more and more attention as cheap and renewable
resources, and we have elaborated pathways for various
chemical modifications and the development of functional
materials from the softwood hemicellulose O-acetylated gal-
actoglucomannan (AcGGM).1–5 In a recent proof-of-concept
study,6 this hemicellulose was successfully converted to a
macroinitiator via an imidazole-driven activation of the sugar
backbone pendant hydroxyl groups. The macroinitiator medi-
ated the single electron transfer-living radical polymerization
(SET-LRP) of a model vinyl monomer affording a hybrid graft
glucopolymer with a brush-like architecture.
temporarily terminated macroradical, allowing for propaga-
tion until the catalyst/ligand complex generated in the pro-
cess deactivates the propagating chain end back to a dor-
mant state. The disproportionation of the catalyst/ligand
complex in coordinating solvents such as water, alcohols, and
other polar solvents is rapid and the driving force for a
highly efficient polymerization.9,11 Compared with estab-
lished living radical polymerization (LRP) techniques such as
reversible addition-fragmentation-transfer12 or atom transfer
radical polymerization (ATRP),13 SET-LRP is robust toward
oxygen,14 readily conducted in water containing media,11
and offer facile removal of the metal catalyst.15 Simple
monofunctional and difunctional initiators, such as the halo-
forms CHCl3 and CHBr3, which are commonly used in LRP
approaches, typically in ATRP but more recently also in SET-
LRP. Also, alkyl halides and sulfonyl halides are viable and
often used initiators. It has been shown that the nature of
the halogen exerts only a minor effect on the polymerization,
including the apparent rate constant of propagation (kappp) in
SET-LRP,7,9 whereas in other metal-catalyzed LRP, typically
ATRP, the kact of the brominated initiator has been reported
to be in the order of 103–104 greater than kact of the corre-
sponding chlorinated species.16 The outer sphere mechanism
proposed for SET-LRP could explain this robustness of the
SET-LRP process. Turning from small to macromolecular ini-
SET-LRP is a robust and versatile method for the living and
controlled radical polymerization offering a fast pathway to
well-defined vinyl polymers with a high level of control of
the product structures, architectures, end groups, and molec-
ular weights.7–10 In SET-LRP, the halide initiator induces
reversibly terminated radical polymerization. The catalyst
[typically a Cu(0) species] transfers a single electron to the
tiators presents new challenges but also
a range of
C
V
2011 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4139–4145
4139

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 1 Synthetic pathway for the preparation of AcGGM-based macroinitiators by 2-chloropropionic acid and a-bromo isobu-
tyric acid. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
opportunities. Various copolymer architectures are possible
and far greater molecular weights are possible, but the
higher initial molecular weight will affect the solubility and
the reaction mixture viscosity posing potential limitations at
higher conversions, successively decreasing the initiator effi-
ciency over time, decelerating propagation and causing spon-
taneous precipitation before full conversion is reached. A
multisite macroinitiator enable grafted and branched archi-
tectures with functional chain ends that allow for possible
further coupling; however, small variations in site reactivity
and accessibility might influence the kinetics and final struc-
ture and their viability for initiating polymerization in a liv-
ing manner cannot be directly estimated from their low mo-
lecular weight analogs.
tyl pendant groups. In the reaction with 2-chloropropionic
acid, a methyl and a methine group is introduced, together
with the ACl as a pendant moiety to the polysaccharide
backbone, and these protons give rise to a set of nonresolved
1
peaks ranging from 1.9 to 1.2 ppm in the H NMR spectrum
of AcGGM-Cl. The overlaid nature of these peaks is because
of the substitution that may occur on several different
carbons in the polysaccharide repeating unit. Substitution on
C-6 is most likely due to steric effects and accordingly, one
peak—at 1.69 ppm—is more pronounced than the others.
For AcGGM-Br, AC(CH3)2-Br moieties are introduced as
pendant groups to the polysaccharide backbone giving rise
to several methyl hydrogen peaks is the 2–1.1 ppm range.
The aim of this work was to prepare chloro (Cl)- and bromo
(Br)-substituted hemicellulose derived multisite macroinitia-
tors for the grafting-from LRP and to provide a comparative
assessment of the effect of the nature of the halogen in the
SET-LRP of an acrylate and a methacrylate initiated from
such glucomacroinitiators.
RESULTS AND DISCUSSION
Cl- and Br-functionalized macroinitiators were successfully pre-
pared from the softwood hemicellulose AcGGM. These func-
tionalized polysaccharides were then explored and evaluated
with respect to their ability and efficiency of initiating the SET-
LRP of two monomers, methyl acrylate (MA) and methyl meth-
acrylate (MMA).
Both Cl- and Br-substituted hemicellulose-derived macroini-
tiators were prepared from the same type of hemicellulose
backbone, a highly purified AcGGM. Bromination was per-
formed using a-bromo isobutyric acid (aBrIBA) according to
Scheme 1, whereas chlorination was afforded by 2-chloro-
propionic acid, yielding AcGGM-Br and AcGGM-Cl, respec-
tively. The functionalization of AcGGM was structurally veri-
1
fied by H NMR as shown in Figure 1. Neat AcGGM shows a
characteristic cluster of overlaid peaks at 5.5–3.2 ppm stem-
ming from the repeating unit anomeric hydrogens. The sharp
peak at 2.16 ppm is assigned to the three protons in the ace-
FIGURE 1 1H NMR in D2O of native AcGGM (bottom), AcGGM-
Br macroinitiator (middle), and AcGGM-Cl macroinitiator (top).
4140
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4139–4145

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 2 Synthetic pathway for the SET-LRP of MA and MMA using AcGGM-based macroinitiators. X is either Cl or Br, R1 and
R2 denotes H or CH3.
Again, some anomeric positions are preferred in the substi-
tution reaction and accordingly, peaks at 1.96 and 1.80 ppm
are the most pronounced.
The halogenated AcGGMs were subsequently used in the
SET-LRP of MA and MMA to elaborate their potential as mac-
roinitiators. The resulting products are graft copolymers as
schematically illustrated in Scheme 2. A model acrylate, MA,
was polymerized by SET-LRP initiated by chlorine- and bro-
mine-functionalized AcGGM, respectively, under identical
reaction conditions. Kinetic plots are shown in Figures 2 and
3, respectively. The time dependence of ln [M0]/[M] as well
as the relationship between number-average molecular
weight and conversion is close to linear in both cases sug-
gesting a first-order rate of propagation and a living poly-
merization. The molecular weight distributions are not
broadening over the course of polymerization but remains
steadily around the values associated with the polysaccha-
ride macroinitiators before grafting: 3.5 for AcGGM-Cl and
4.6 for AcGGM-Br, respectively, further sustaining this conclu-
sion. Monomodal SEC traces are observed throughout the
polymerizations. The initiator and monomer concentrations
were purposely kept low to avoid very high viscosities with
increasing conversions, yet the formation of a high viscosity
reaction mixture was observed at high conversions, making
it very difficult to reach full conversions. AcGGM-Br-initiated
MA grafting proceeded at an apparent rate constant of prop-
agation (kapp) of 0.0031 minꢀ1, while AcGGM-Cl-initiated MA
grafting under identical conditions differed only to a minor
extent with a kapp of 0.0035 minꢀ1. Compared with the SET-
LRP of MA with a low molecular weight initiator analog,
The degree of acetylation (DSAc) in this particular hemicellu-
lose sample is 0.3 as determined by size exclusion chroma-
tography (SEC) calibrated with matrix assisted laser induced
desorption and ionization - time of flight - mass spectrome-
try (MALDI-TOF-MS).17,18 Hence, the acetylation peak can be
used as an internal standard in the calculation of the
degree of substitution (DS) of Br- and Cl-substituents
introduced to the hemicellulose backbone. Because of
the more pronounced bulkiness of the aBrIBA compared with
2-chloropropionic acid, the DS of the Br substituent (DSBr) of
functionalized AcGGM became somewhat lower than the
corresponding DSCl, although the reaction time was doubled
in the former case. DSBr was 0.2, while DSCl was 0.3. The neat
AcGGM had a Mw of 7500 g molꢀ1 and the chlorinated
AcGGM (AcGGM-Cl) had a Mw of 6630 g molꢀ1, so there is a
slight ꢁhydrolysis occurring at reaction times around 30 min
at 50 C. Brominated AcGGM (AcGGM-Br), on the other hand,
had a Mw of 3450 g molꢀ1 indicating that the longer reaction
time, 1 h, caused a more significant hydrolysis of the AcGGM
polysaccharide backbone. This difference in initiator molecu-
lar weight and degree of halogen substitution will affect the
molecular weight of resulting graft copolymers that is possi-
ble to achieve during LRP at a fixed monomer:initiator ratio.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4139–4145
4141

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 2 Kinetic plots for SET-LRP of MA initiated by AcGGM-Br. Each data point represents an individual experiment. Reaction
conditions: [MA]0/[Cl-AcGGM]0/[Me6-TREN]0 ¼ 200/1/0.2, 6.25 cm Cu-wire, DMSO ¼ 3 mL, [MA]0 ¼ 0.83 mol Lꢀ1, 25 ꢁC.
methyl 2-bromopropionate, under otherwise very similar con-
MMA grafting-from AcGGM macroinitiators was typically
more sluggish than MA grafting under the same conditions.
Linear dependencies of ln [M0]/[M] versus time and num-
ber-average molecular weight versus conversion still indicate
a living behavior. Kinetic plots for MMA polymerization from
AcGGM-Br and AcGGM-Cl are shown in Figures 4 and 5,
respectively. Compared with AcGGM-Br-initiated MA grafting
(kapp ¼ 0.0031 minꢀ1) under identical conditions, kapp was
significantly lower in the case of MMA grafting: 0.0011
minꢀ1. In the case of AcGGM-Cl-initiated MMA grafting, the
7
ꢁ
ditions [dimethyl sulfoxide (DMSO), 25 C, Cu(0), [M]0/[I]0/
[L]0 ¼ 2220/10/1], the rate constants are a factor 16 lower
for the AcGGM-macroinitiated SET-LRP of MA illustrating the
change in initiator site accessibility when increasing the molar
mass significantly. The ratio kapp (AcGGM-Cl)/kapp (AcGGM-Br)
is 1.13, which is sufficiently close to 1, to conclude that the na-
ture of the halogen does not exert a strong influence on the
propagation rate of MA when these AcGGM-derived multisite
macroinitiators are used, in line with the outer sphere mecha-
nism proposed for SET-LRP. However, it should be noted that
the kapp (AcGGM-Cl) is slightly higher than kapp (AcGGM-Br)
for the MA polymerization. The same trend was observed in
the SET-LRP of MA initiated by haloforms, where kapp (CHCl3)/
kapp (CHBr3) was 1.25.7 In radical polymerization, the acryl-
ates in general are expected to be more efficiently initiated by
Br-initiators than Cl-initiators, whereas methacrylates, on the
other hand, work better with Cl-initiators.10 This is not sus-
tained by the results from AcGGM-initiated SET-LRP of MA,
but the difference between the halide types is again not exten-
sive as proposed for SET-LRP.7
rate was even slower with a calculated kapp ¼ 0.0006 minꢀ1
.
The slower propagation of MMA compared with MA is also
seen in conventional radical polymerization and other living
approaches and is explained by the lower intrinsic kp of
methacrylates. The ratio kapp (AcGGM-Cl)/kapp (AcGGM-Br) is
in this case 0.56 which, in analogy with the MA polymeriza-
tion described above, is sufficiently close to 1 to conclude
that the nature of the halogen does not exert a strong influ-
ence on the propagation rate. In contrast to the MA polymer-
ization, however, the kapp (AcGGM-Cl) is slightly lower than
kapp (AcGGM-Br) for the MMA polymerization. As outlined
FIGURE 3 Kinetic plots for SET-LRP of MA initiated by AcGGM-Cl. Each data point represents an individual experiment. Reaction
conditions: [MA]0/[Cl-AcGGM]0/[Me6-TREN]0 ¼ 2000/10/2, 6.25 cm Cu-wire, DMSO ¼ 3 mL, [MA]0 ¼ 0.83 mol Lꢀ1, 25 ꢁC.
4142
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4139–4145

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 4 Kinetic plots for SET-LRP of MMA initiated by AcGGM-Br. Each data point represents an individual experiment. Reaction
conditions: [MMA]0/[Cl-AcGGM]0/[Me6-TREN]0 ¼ 2000/10/2, 6.25 cmCu-wire, DMSO ¼ 3 mL, [MMA]0 ¼ 0.83 mol Lꢀ1, 25 ꢁC.
previously, Cl-initiators are supposed to be better suited for
the methacrylates than Br-initiators and vice versa for acryl-
ates.10 As concluded for the AcGGM-initiated SET-LRP of MA
in the previous paragraph, this relationship is not sustained
by the results presented here. The halide affinities seems to
be somewhat offset in the AcGGM-initiated SET-LRP, but still,
the halide effects are small compared with those reported
for other metal-catalyzed LRP.1
being only 50% of that of AcGGM-Cl. This means that much
higher molecular weights are theoretically possible in the lat-
ter case but, at the same time, there is an increasing effect
of steric hindrance in a random coil conformation of the lon-
ger chains potentially making initiator sites less available.
Also, the viscosity buildup at higher conversions is more
extensive hindering efficient diffusion and hence propaga-
tion. Clearly, the molar mass of the macroinitiator is a pa-
rameter significantly affecting the ability to effectively medi-
ate SET-LRP and also a moderate increase in molar mass of
the polysaccharide limit the initiator efficiency. Comparing
Ieff values for the two monomers, respectively, sustain the
small differences observed in halide suitability toward Cl-
and Br-initiators. For the methacrylate, AcGGM-Br gave the
higher kapp and also this initiator showed a higher Ieff for
MMA (Ieff ¼ 84%) compared with MA (Ieff ¼ 74%) polymer-
ization. For the acrylate, in contrast, AcGGM-Cl gave the
higher kapp and, furthermore, the Ieff values for the Cl-initia-
In terms of product number-average molecular weight, there
is a marked difference between polymerization initiated by
the two AcGGM-based macroinitiators and a deeper perspec-
tive on their ability to function as macroinitiators is achieved
by estimating the initiator efficiencies. When considering the
products’ number-average molecular weight as a function of
the theoretical molecular weight as calculated according to
the living polymerization theoretical dependence, a clear dif-
ference between the macroinitiators becomes apparent (Fig.
6). Initiator efficiencies, Ieff, of AcGGM-Cl range from 32 to
40%, whereas AcGGM-Br shows efficiencies in the range of
74–84%. The increase in Ieff for AcGGM-Br by a factor 2
coincides with the molecular weight of this macroinitiator
tor was higher for MA (Ieff ¼ 40%) than for MMA (Ieff
¼
32%) polymerization. Hence, the Ieff values sustain the prop-
agation rate trends, both trends contradicting the theoretic
prediction that Cl-initiators are supposed to be better suited
FIGURE 5 Kinetic plots for SET-LRP of MMA initiated by AcGGM-Cl. Each data point represents an individual experiment. Reaction
conditions: [MMA]0/[Cl-AcGGM]0/[Me6-TREN]0 ¼ 2000/10/2, 6.25 cm Cu-wire, DMSO ¼ 3 mL, [MMA]0 ¼ 0.83 mol Lꢀ1, 25 ꢁC.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4139–4145
4143

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
mide 99.8% (Fisher), were used as received. DMSO ꢂ 99.5%
(Fluka), MMA > 99% (Merck), and MA 99% (Merck) were
distilled at reduced pressure before use.
O-Acetyl-galactoglucomannan (AcGGM) was recovered from
the process water of thermomechanical pulping of spruce
(Picea abies). The hemicellulosic fraction was concentrated
by ultrafiltration, further purified by diafiltration, and finally
lyophilized to yield off-white powder. The AcGGM had a
weight average molecular weight of about 7500 g molꢀ1
,
and a polydispersity index (PDI) of ꢃ1.3. The DSAc of 0.30 as
determined by SEC calibrated with MALDI-TOF-MS.17,18
Me6-TREN Synthesis
Me6-TREN was synthesized from 25.5 mL of 98% formic
acid, 25 mL of 37% formaldehyde, and 6.58 g of tris(2-ami-
noethyl)amine (0.045 mol, 1 equiv) in 25 mL of H2O accord-
ing to a literature procedure.19
Yield: 10%. 1H NMR (DMSO-d6): d ¼ 2.50 ppm (tr, 6 H,
CH2); 2.26 ppm (tr, 6 H, CH2); 2.12 ppm (s, 18 H, N-(CH3)2);
13C-NMR (DMSO-d6): d ¼ 57.43, 52.52, 45.64 ppm.
FIGURE 6 Molecular weights as measured by SEC versus theo-
retical molecular weights for macroinitiated SET-LRP: (x) MMA
initiated by AcGGM-Br, (n) MA initiated by AcGGM-Br, (^) MA
initiated by AcGGM-Cl, and (~) MMA initiated by AcGGM-Cl.
Each data point represents an individual experiment. Reaction
conditions: [M]/[I]0/[Me6-TREN]0 ¼ 2000/10/2, 6.25 cm Cu-wire,
DMSO ¼ 3 mL, 25 ꢁC.
AcGGM-Cl Macroinitiator Synthesis
AcGGM was converted to a macroinitiator by esterification of
the sugar hydroxyl groups. In the first step, 2.130 g of
CPA (23.25 mmol) was activated through coupling to CDI
(3.768 g/23.25 mmol) in 20 mL of DMSO at room tempera-
ture. After 1 h, the temperature was raised to 50 ꢁC and
1.50 g of AcGGM (8.65) dissolved in 100 mL of DMSO was
added to the mixture and allowed to react with the imida-
zoyl activated CPA for 30 min. The product was precipitated
in ethyl acetate and washed repeatedly with ethyl acetate.
The crude precipitate was collected by centrifugation and
purified with Soxhlet extraction for 2 days. The product was
finally dissolved in water, lyophilized, and characterized with
1H NMR and SEC.
for the methacrylates than Br-initiators and vice versa for
acrylates.10
To further elaborate the efficiency of Cl- and Br-AcGGM mac-
roinitiators to mediate a polymerization in a living manner,
the preparation of very high molecular weight polymers was
attempted. Both types of AcGGM macroinitiators were, thus,
used in the SET-LRP of MMA and MA with a [M]0/[I] ratio of
6000/1, catalyzed by Cu(0) and in DMSO at 25 ꢁC as for the
other experiments. Even though conducted at much lower con-
centrations to avoid a very high viscosity buildup, the reaction
mixtures became close to solid, and higher conversions were
not achieved. After 48 h, conversions up to 40% were reached.
Still, very high and in a couple of experiments, ultrahigh, mo-
lecular weights were achieved. Similar Ieff values as calculated
AcGGM-Br Macroinitiator Synthesis
AcGGM was brominated analogous to the above described
chlorination route using aBrIBA. aBrIBA (10.02 g/60 mmol)
was activated with equimolar amounts of CDI (9.73 g) in
80 mL DMSO at room temperature. After 1 h, the tempera-
ture was raised to 50 ꢁC and 2.60 g of AcGGM dissolved in
100 mL of DMSO was added together with an additional
4.08 g of CDI. The reaction was allowed to proceed for 60
min and quenched by precipitation in 2-propanol. The crude
product was collected by centrifugation and purified with
Soxhlet extraction with 2-propanol for 2 days. The product
was finally dissolved in water, lyophilized, and characterized
for the [M]0/[I] ¼ 200/1 experiments are observed here, Ieff
31% for AcGGM-Cl and Ieff ¼ 60% for AcGGM-Br. AcGGM
grafted with poly(methyl methacrylate) (PMMA) where Mn
¼
¼
540,000 and Mw ¼ 1,141,000 (Mw/Mn ¼ 2.1) were obtained
after 48 h.
1
with H NMR and SEC.
EXPERIMENTAL
Polymerizations
Materials
All polymerizations were conducted in DMSO using Me6-
TREN as the ligand, and MA or MMA as the monomer. In a
typical polymerization, the monomer, macroinitiator (Br-
AcGGM or Cl-AcGGM), and ligand were dissolved in DMSO in
a 2000:10:2 molar ratio. For 3 mL of solvent, 16.35 mg of
Br-AcGGM or 8.4 mg of Cl-AcGGM were used. The monomer,
macroinitiator, and ligand solution was in each polymeriza-
tion transferred to a dry Schlenk-tube and degassed with six
N,N0-Carbonyldiimidazole (CDI) 97% (Aldrich), 2-chloropro-
pionic acid (CPA) 98% (Sigma), aBrIBA 98% (Aldrich),
tris(2-aminoethyl)amine 96% (Aldrich), formic acid ꢂ 96%
(Aldrich), formaldehydeaq 37% (AlfaAesar), dichloromethane
99.8% (Fisher), sodium chloride 99.5% (Merck), 2-propanol
ꢂ 99% (LabScan), methanol ꢂ 99.8% (LabScan), copper
wire (Fisher), acetone (Fischer, purum), and dimethylaceta-
4144
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4139–4145

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
consecutive freeze-pump-thaw cycles, flushing with nitrogen
after thawing. Copper wire (6.25 cm) with a diameter of
ꢃ0.812 mm corresponding to American wire gauge 20, was
used as a catalyst. The wire was wrapped around the stir-
ring bar and kept over the reaction mixture throughout the
degassing. Then, the tube was immersed in a thermostated
oil bath at 25 ꢁC, and the polymerization was started by
dropping the stirring bar with the copper catalyst into the
monomers, the macroinitiator efficiency of AcGGM-Br is
twice as high when compared with AcGGM-Cl, influenced by
the higher initial chain length of the latter and the associated
viscosity effects at high molecular weights of the hybrid graft
copolymer products. The molar mass of the macroinitiator is
a parameter significantly affecting the ability to effectively
mediate SET-LRP, even a moderate increase in molar mass of
the polysaccharide limit the initiator efficiency. Both Cl- and
Br-functionalized AcGGM successfully mediate the formation
of very high or ultrahigh molecular weight graft copolymers.
1
solution. The polymerization was monitored via H NMR and
SEC. Each data point is derived from an individual experi-
ment. When reaching the individual stop time for each
experiment, an aliquot of the reaction mixture was first
sampled for NMR analysis under a flow of N2 (g) using a sy-
ringe. The polymerizations were then terminated by precipi-
tation of the reaction mixture in cold methanol. The product
graft copolymers were then purified by redissolution in
DMSO and reprecipitation in methanol followed by centrifu-
gation and vacuum drying. For polymerizations aiming at
ultrahigh molecular weights, monomer:macroinitiator:ligand
ratios of 60,000:10:2 were used. The amounts were reduced
to one-fourth of the amounts used for all other polymeriza-
tions as described. Sampling, product recovery, and purifica-
tion were done according to the procedure described above.
Weight average molecular weights exceeding
1 million
g molꢀ1 were achieved from AcGGM-Cl-initiated SET-LRP.
REFERENCES AND NOTES
1 Lindblad, M. S.; Ranucci, E.; Albertsson, A. C. Macromol
Rapid Commun 2001, 22, 962–967.
2 Hartman, J.; Albertsson, A. C.; Sjoberg, J. Biomacromole-
cules 2006, 7, 1983–1989.
3 Edlund, U.; Albertsson, A. C. J Bioact Compat Polym 2008,
23, 171–186.
4 Voepel, J.; Edlund, U.; Albertsson, A. C. J Polym Sci Part A:
Polym Chem 2009, 47, 3595–3606.
Characterization
5 Edlund, U.; Ryberg, Y. Z.; Albertsson, A. C. Biomacromole-
cules 2010, 11, 2532–2538.
Molecular weights were determined with a Shimadzu SEC
system equipped with four PLgel 20 lm Mixed-A columns
and using N,N-dimethylacetamide (DMAc) as the eluent with
a flow rate of 0.5 mL minꢀ1 at 80 ꢁC and an injection vol-
ume of 200 lL. Pullulan standards with narrow molecular
weight distributions were used for calibration. All samples
were dissolved in DMAc containing LiCl (0.5%, w/w) and fil-
tered before analysis.
6 Voepel, J.; Edlund, U.; Albertsson, A. C.; Percec, V. Biomacro-
molecules 2011, 12, 253–259.
7 Percec, V.; Guliashvili, T.; Ladislaw, J. S.; Wistrand, A.;
Stjerndahl, A.; Sienkowska, M. J.; Monteiro, M. J.; Sahoo, S. J
Am Chem Soc 2006, 128, 14156–14165.
8 Lligadas, G.; Percec, V. J Polym Sci Part A: Polym Chem
2008, 46, 2745–2754.
1H NMR spectra were recorded at 400 MHz on a Bruker
DMX-400 nuclear magnetic resonance spectrometer with
Bruker software. DMSO-d6 and D2O (Larodan Fine Chemicals
AB) were used as solvents, and spectra were recorded in
sample tubes with an outer diameter of 5 mm.
9 Lligadas, G.; Percec, V. J Polym Sci Part A: Polym Chem
2008, 46, 4917–4926.
10 Percec, V.; Rosen, B. M. Chem Rev 2009, 109, 5069–5119.
11 Percec, V.; Nguyen, N. H.; Rosen, B. M.; Jiang, X.; Fleisch-
mann, S. J Polym Sci Part A: Polym Chem 2009, 47, 5577–5590.
12 Moad, G.; Rizzardo, E.; Thang, S. H. Aust J Chem 2005, 58,
379–410.
CONCLUSIONS
13 Matyjaszewski, K.; Xia, J. H. Chem Rev 2001, 101,
2921–2990.
The hemicellulose AcGGM was successfully converted to Cl-
and a Br-functionalized macroinitiators using an imidazole-
mediated esterification of the pendant hydroxyl groups on
the polysaccharide backbone. Both halogenated species effec-
tively initiated SET-LRP of methyl acrylate (MA) and methyl
methacrylate (MMA) in DMSO solution yielding grafted
AcGGM copolymers. The Cl- and the Br-functionalized macro-
initiators show small differences in rates of SET-LRP. For ac-
rylate graft polymerization, AcGGM-Cl gives the highest appa-
14 Percec, V.; Jiang, X. A.; Rosen, B. M. J Polym Sci Part A:
Polym Chem 2010, 48, 2716–2721.
15 Percec, V.; Nguyen, N. H.; Rosen, B. M.; Lligadas, G. Macro-
molecules 2009, 42, 2379–2386.
16 Matyjaszewski, K.; Paik, H. J.; Zhou, P.; Diamanti, S. J. Mac-
romolecules 2001, 34, 5125–5131.
17 Jacobs, A.; Lundqvist, J.; Stalbrand, H.; Tjerneld, F.; Dahl-
man, O. Carbohydr Res 2002, 337, 711–717.
rent rate constant (kapp
methacrylate grafting proceeds at a higher kapp (0.0011
¼
0.0035 minꢀ1), while
18 Dahlman, O.; Jacobs, A.; Liljenberg, A.; Olsson, A. I. J Chro-
matogr A 2000, 891, 157–174.
minꢀ1), when the initiating species is AcGGM-Br. For both
19 Ciampolini, M.; Nardi, N. Inorg Chem 1966, 5, 41–44.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4139–4145
4145