European Polymer Journal 48 (2012) 1646–1659

Contents lists available at SciVerse ScienceDirect

European Polymer Journal

journal homepage: www.elsevier.com/locate/europolj

Feature Article

Grafting of cellulose by ring-opening polymerisation – A review

Anna Carlmark, Emma Larsson, Eva Malmström ⇑
KTH Royal Institute of Technology, School of Chemical Science and Engineering, Fibre and Polymer Technology, Teknikringen 56, SE-100 44 Stockholm, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: In this review, homogeneous and heterogeneous grafting from cellulose and cellulose
Received 4 May 2012 derivatives by ring-opening polymerisation (ROP) are reported. Cellulose is biorenewable
Received in revised form 11 June 2012 and biodegradable as well as a stiff material with a relatively low specific weight, foreseen
Accepted 19 June 2012
to be an excellent replacement for synthetic materials. By utilising ROP of monomers such
Available online 2 July 2012
as e-caprolactone or L-lactide from cellulose, composite materials with new and/or
improved properties can be obtained. Grafting of solid cellulose substrates, such as cotton,
Keywords:
microfibrillated cellulose (MFC) or cellulose nanocrystals, renders cellulose that can easily
Ring-opening polymerization (ROP)
Cellulose
be dispersed into polymer matrices and may be used as reinforcing elements to improve
Cellulose derivatives mechanical and/or barrier properties of biocomposites. A surface grafted polymer can also
Surface grafting tailor the interfacial properties between a matrix and the fibrillar structure of cellulose.
Biocomposites When derivatives of cellulose are grafted with polymers in homogenous media, amphi-
philic materials with interesting properties can be achieved, anticipated to be utilised for
applications such as encapsulation and release.
Ó 2012 Elsevier Ltd. Open access under CC BY-NC-ND license.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1647
2. Cellulosic substrates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1647
2.1. Nanocelluloses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1647
2.2. Cellulose derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1648
2.3. Cellulose modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1648
3. Biocomposites. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1648
4. Ring-opening polymerisation (ROP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1649
5. Ring opening polymerisation from cellulose fibres – heterogeneous grafting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1649
5.1. ‘‘Grafting from’’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1649
5.2. ‘‘Grafting to’’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1653
6. Ring opening polymerisation from cellulose derivatives – homogeneous grafting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1653
6.1. ‘‘Grafting from’’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1653
6.2. ‘‘Grafting from’’ by a combination of ROP and atom transfer radical polymerisation (ATRP) . . . . . . . . . . . . . . . . . . . . . 1656
7. Enzymatic grafting of cellulose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1657
8. Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1657
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1657
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1657

⇑ Corresponding author. Tel.: +46 8 7907225; fax: +46 8 7908283.

E-mail address: mavem@kth.se (E. Malmström).

0014-3057 Ó 2012 Elsevier Ltd. Open access under CC BY-NC-ND license.

http://dx.doi.org/10.1016/j.eurpolymj.2012.06.013
 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
1. Introduction
Cellulose, being the most abundant naturally occurring
polymer on earth, is a highly interesting material due to its
renewability, low price, high availability, and good
mechanical properties [1–4]. Cellulose is composed of
repeating units of glucose joined together by hydrogen
bonds to form sheets. The sheets in turn form microfibrils,
which aggregate. These aggregates form layers with differ-
ent orientation in the cell wall, see Fig. 1 [1–3,5]. The high
molecular weight and numerous hydroxyl groups bring
about strong inter- and intra-molecular interactions due
to hydrogen bonding, so strong that cellulose is practically
insoluble in water, although highly hydrophilic. Further-
more, these hydrogen bonds give cellulose fibres excellent
mechanical properties, [1,2] similar to those of synthetic
fibres.
The interest in cellulose has reawakened due to the con-
cern for the environmental impact of synthetic polymers
and fibres, and as the price of the raw material for produc-
tion of synthetic polymers is rapidly increasing. Since cel-
lulose is both biorenewable and biodegradable, several
studies have focused on the modification of cellulose, to
investigate it as a replacement material for synthetic poly-
mers [1,6,7]. The hydroxyl groups can act as chemical han-
dles and cellulose can either be modified with small
organic molecules [2,4,8] or by grafting of polymer from
the cellulose backbone. Grafting of polymer chains can be
performed by several different methods, either by ‘‘grafting
from’’, ‘‘grafting to’’ or ‘‘grafting through’’ [1,4,8]. There are
several reports in the literature of the grafting of cellulose
by free radical polymerisation, coupling chemistries, con-
trolled radical polymerisation etc. [1,7,9]. The focus of this
Fig. 1. The hierarchical structure of cellulose from wood. Reproduced from Ref. [5].
1647
review is, however, the grafting of polymers from cellulose
and cellulose derivatives utilising Ring-Opening Polymeri-
sation (ROP). The review is divided into heterogeneous (i.e.
from solid cellulose fibres and fibrils), homogeneous (i.e.
from soluble cellulose derivatives and dissolved cellulose)
and enzymatic grafting, as well as in ‘‘grafting to’’ and
grafting from’’, referring to the way the polymer is an-
chored to the substrate. Furthermore, for every method,
each substrate of cellulose is described separately. A short
introduction to the different cellulosic substrates that have
been utilised for ROP is given below.
2. Cellulosic substrates
There are several natural sources of cellulose; plants
(major sources are wood and cotton), some animals, bacte-
ria, and fungi [3,6,10]. Depending on the source, molecular
properties such as molecular weight and crystallinity will
differ. Wood pulp is the most important of all cellulosic
raw materials, which is used in the paper industry. Only
around 2% of the wood pulp is modified to be used for pro-
duction of regenerated fibres and chemically modified fi-
bres. The molecular weight of cellulose in wood pulp
varies with the source of the pulp, but it typically has a de-
gree of polymerisation (DP) in the region of 300–1700 [2].
2.1. Nanocelluloses
Recently, much research has focused on the investiga-
tion and disintegration of cellulose into its nanostructural
components. Microfibrillated cellulose (MFC), also called
nanofibrillated cellulose (NFC), is cellulose that has been
1648 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
mechanically processed in a high pressure homogenizer,
and eventually turned into small aggregates of cellulose fi-
brils. It is common to pre-treat the cellulose chemically,
physically or enzymatically prior to the mechanical pro-
cessing, to reduce the energy needed for an otherwise ex-
tremely energy consuming production. Typically, MFC
has diameters ranging from 5–60 nm and lengths of sev-
eral micrometres. MFC is of significant interest, as it has
shown to have high strength and stiffness and rather low
weight [11].
Nanocrystalline cellulose (NCC), also called cellulose
nanowhiskers (CNW), is cellulose that is nearly completely
crystalline. The structure is achieved after acidic hydroly-
sis, commonly followed by ultrasonication, to remove
amorphous regions in the cellulose. The NCC has a rod like
shape, 5–70 nm width and a length varying from 100 nm
to a few micrometres. It has a high strength and is there-
fore considered to be of potential use as a filler material
in composites [11].
Bacterial nanocellulose (BNC) has glucose chains with
DPs that varies from 2000–8000 and diameters in the
range of 20–100 nm. This makes the cellulose chains of
BNC much longer than those of cellulose from plant
sources. BNC also has a higher purity than cellulose from
plants, as it contains no traces of hemicelluloses or lignin.
It has a high degree of crystallinity and high water content,
at least 90 wt.%. The material shows great mechanical
properties due to its crystalline structure [2,11].
2.2. Cellulose derivatives
By reacting some, or all, of the hydroxyl groups on the
repeat unit in cellulose, cellulose derivatives are obtained.
Usually some degradation of the cellulose backbone can be
expected but also tolerated during modification. The most
commonly produced chemically modified cellulose deriva-
tives are cellulose esters and ethers. Cellulose esters are
made by substitution of the hydroxyl groups on cellulose
with organic or inorganic acids, and depending on degree
of substitution, type of substituent, and the distribution
of the substituents along the cellulose chain, a large variety
of properties can be achieved. Some common cellulose es-
ters are cellulose acetate (CA) and cellulose acetate propi-
onate (CAP) [2], Fig. 2.
Production of cellulose ethers is performed in alkaline
solutions and one of the most common processes is oxygen
alkylation with alkyl halides. Cellulose ethers typically
have high chemical stability and a low toxicity. The solu-
bility of cellulose ethers in water can, to some extent, be
controlled by the degree and type of substitution as well
Fig. 2. Schematic structure of the repeating unit of some cellulose derivatives.
as the distribution of ether groups along the chain.
Common cellulose ethers are alkyl ethers, like ethyl cellu-
lose (EC), carboxymethyl cellulose (CMC), hydroxyethyl
cellulose (HEC), and hydroxypropyl cellulose (HPC) [2],
Fig. 2.
2.3. Cellulose modifications
When grafting is performed from native cellulose, like
cotton or wood fibres, the reaction is typically performed
in a heterogeneous mixture, since these fibres are insoluble
in common organic solvents, as well as in water. However,
rather recently ionic liquids have been used as solvents
and it has been shown possible to graft from unmodified
cellulose in homogeneous mixtures as well [12,13]. Heter-
ogeneous grafting is also performed from the nanostruc-
tural elements of cellulose, like MFC and NCC.
With cellulose derivatives as substrates, polymerisation
reactions can be performed in homogenous reaction mix-
tures since cellulose derivatives are soluble in various sol-
vents. Furthermore, since some of the hydroxyl groups
have been reacted into other functional groups the chemi-
cal and mechanical properties are altered as compared to
native cellulose.
3. Biocomposites
A composite is composed of two (or more) different
constituents, where one is generally a matrix and the other
some kind of fibre, typically added to reinforce the matrix
material. Biocomposites are a rapidly growing group of
materials where the fibre originates from nature while
the matrix may or may not be based on a renewable mate-
rial [6]. Polyethylene (PE) and polypropylene (PP) are
examples of durable matrix polymers while poly(e-capro-
lactone) (PCL) and poly-L-lactide (PLLA) are examples of
degradable matrices. Typically, the strengthening fibre-
component can either be wood fibres or emanate from
other species in the plant kingdom such as wheat, hemp,
jute, sisal, flax, or kenaf. Depending on source and proce-
dure used for fibre isolation, a wide range rather different
biofibres can be obtained [4,6].
Biofibres often have a lower specific weight than inor-
ganic alternatives which justifies the replacement of inor-
ganic fibres by natural fibres given that mechanical
properties remain comparable [4,14]. Furthermore, the
development of biocomposites is attractive from a sustain-
able perspective; natural fibres are carbon dioxide neutral
and overall environmentally friendly to produce and use
[14].
 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
Biofibres also bring about challenges since they have
very polar surfaces, carrying mainly hydroxyl groups, ren-
dering them hydrophilic and thereby susceptible to water
adsorption. The matrix polymer is often hydrophobic by
nature and the adhesion between a natural fibre and a
polymer matrix often needs to be carefully controlled in
order to reduce the risk of failure. The water adsorption,
depending on the relative humidity in the surrounding,
gives rise to poor dimensional stability as a consequence
of changing water content in the fibres. Suitable surface
functionalization is often required to provide the natural
fibres with appropriate surface properties for the intended
application [4,6,14]. Another challenge is that the resis-
tance to UV-irradiation is quite poor and therefore UV-
protection has to be employed [4].
By grafting biofibres with polymer chains, the fibres be-
come more compatible with polymer matrices, and several
of the drawbacks described above can be addressed, en-
abling them to be utilised in biocomposites. Hence, much
of the literature describing the ROP from cellulose deals
with this particular aspect where the modified fibres have
been incorporated into polymer matrices.
4. Ring-opening polymerisation (ROP)
ROP is a well-established technique to polymerize cyc-
lic monomers such as lactones and lactides. An alcohol (or
hydroxyl group) is generally used as the initiator for ROP
which makes it especially interesting to utilise ROP of cyc-
lic monomers for the polymer modification of cellulose or
cellulose derivatives [15].
ROP operates through different mechanisms depending
on which monomer, initiator and catalytic system that are
utilised. Tin(II) 2-ethylhexanoate (Sn(Oct)2) is a commonly
used catalyst for the polymerisation for monomers such as
e-caprolatone (e-CL), lactide (LA) and p-dioxanone. Several
different mechanisms have been hypothesised for this sys-
tem, but the most commonly accepted mechanism for the
initiation is that Sn(Oct)2 is converted into tin alkoxide, the
actual initiator, by reaction with alcohols or other protic
Scheme 1. Schematic drawing of the grafting of cellulose with PCL.
1649
compounds/impurities (1 and 2), i.e. the ‘‘coordination-
insertion’’ mechanism. Effectively, by tuning the alcohol-
to-monomer ratio, the molecular weight of the final
polymer can be controlled [15–17].
SnðOctÞ2 þ R-OH ! Oct  Sn  OR þ OctH ð1Þ
Oct  Sn  OR þ R  OH ! SnðORÞ2 þ OctH ð2Þ
5. Ring opening polymerisation from cellulose fibres –
heterogeneous grafting
5.1. ‘‘Grafting from’’
Since ROP can be initiated by hydroxyl groups, cellulose
in its native form is a multifunctional initiator for this reac-
tion. This is one great advantage with performing ROP from
cellulose, since no chemical treatment of the cellulose is
necessary prior to the grafting reaction, as with other poly-
merisation methods. This was exploited by Hafrén and
Córdova in 2005 when they published the first report of
surface-initiated ROP (SI-ROP) from a solid cellulose sub-
strate [18]. In their initial study, e-CL was grafted from fil-
ter paper and cotton, using organic- or amino acids as
catalysts, Scheme 1. The successful grafting was confirmed
by FTIR spectroscopy by the appearance of a peak at
1730 cm1, assigned to the carbonyl group in PCL. FTIR
spectroscopy is, due to its ease of operation and non-
destructive character, frequently used to characterise sur-
faces after grafting reactions. Hafrén and Cordova also used
contact angle measurements to investigate the hydropho-
bicity after grafting, as observed by high contact angles
against water. The weight gain of the samples was deter-
mined by gravimetric analysis to be 11%. In their compar-
ison, Hafrén and Cordova found that tartaric acid was the
most efficient catalyst for the grafting.
The same group also reported a similar approach where
SI-ROP of e-CL catalysed by tartaric acid was used to graft
polymer from newspaper based on thermo-mechanical
pulp (TMP) [19]. The modified surfaces were compared
with unmodified surfaces using SEM and, according to
1650 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
Scheme 2. Modification of cellulose with bis-MPA. Reprdouced from Ref. [5].
the authors; the fibre structure of the papers appeared
slightly more open after the modification. From the SEM
images it was also determined that the PCL was evenly dis-
tributed across the samples. Hafrén and Córdova also poly-
merized L-LA from filter paper using tartaric acid as
catalyst [20]. It was observed that it was possible to graft
oligomers of L-LA from the cellulose surface even without
the presence of a catalyst, but with the addition of tartaric
acid the reaction was faster and higher DṔs were reached.
Hult et al. performed SI-ROP of e-CL and L-LA from cellu-
lose fibres with the aim to increase the amount of grafted
polymer on fibres [21]. This was approached by increasing
the number of available hydroxyl groups on the surface of
filter paper by reacting the native hydroxyl groups with
the anhydride of acetonide-protected bis-MPA, Scheme 2.
After deprotection of the acetonide group the polymerisa-
tions were conducted, catalysed by Sn(Oct)2. Also, sacrificial
initiator was added to form a free, unbound, polymer. The
free polymer formed was analysed with 1H-NMR and SEC
to determine molecular weight and polydispersity, and
was used to estimate the chain length of the grafted poly-
mer, assuming that the kinetics of ROP is similar to that of
SI-ROP. To further vary the amount of available hydroxyl
groups on the substrate, a second method was attempted
in which xyloglucan-bis(methylol)-2-methylpropanamide
(XG-bis-MPA) was physiosorbed to the substrate. Xyloglu-
can is a hemicellulose known for its high affinity to cellulose
and it has previously been explored by Teeri et al. for the
modification of cellulose [22]. Polymerisations of e-CL and
L-LA by ROP from both unmodified and modified filter paper
Scheme 3. Grafting of hyperbranched polyether from a cellulose fibre surface. Reproduced with permission from Springer from Ref. [23].
were successful and high conversions were obtained in all
reactions. FTIR indicated that the cellulose modified with
bis-MPA had the largest amount of grafted PCL and PLLA.
To estimate a change in hydrophobicity of the samples after
grafting, contact angle measurement was performed. The
cellulose surface that had been modified with bis-MPA prior
to grafting showed the highest contact angle of the cellulose
surfaces grafted with PCL with a value of 111°.
To investigate how well the cellulose substrates were
covered with polymer grafts, an enzymatic degradation
study was performed utilising a wild-type cellulase. The
study showed that unmodified paper, paper modified with
bis-MPA and polymer grafted XG-bis-MPA papers were
disintegrated into fibre dispersions within 24 h. Polymer
grafted unmodified paper and polymer grafted bis-MPA
modified paper showed stronger resistance to degrading
than unmodified paper, this resistance increasing with
increasing length of the grafts [21].
Pan et al. grafted filter paper with hyperbranched
poly(3-methyl-3-oxetanemethanol) (HBPO) by SI-ROP of
3-methyl-3-oxetanemethanol from the hydroxyl groups
of cellulose, Scheme 3 [23]. The grafting of this hyper-
branched polymer yielded an increased amount of hydro-
xyl groups on the surface of the fibres, as determined by
XPS. The HBPO content of different samples varied be-
tween 44–99%, however, the results calculated from XPS
may overestimate the HBPO content, due to an underesti-
mation of the O/C ratio in cellulose.
Hult et al. were the first to report ring-opening poly-
merisation from microfibrillated cellulose (MFC) [24]. The
 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
Scheme 4. ROP of e-caprolactone from microfibrillated cellulose. Reproduced with permission of Elsevier from Ref. [24].
Fig. 3. FTIR-spectra of unmodified MFC and MFC-PCL, target DP 300, 600,
and 1200. Reproduced with permission of Elsevier from Ref. [24].
Fig. 4. Unmodified MFC (left) and PCL-grafted MFC (right) dispersed in
THF. Reproduced with permission of Elsevier from Ref. [24].
grafting was performed by SI-ROP of e-CL in the presence
of benzyl alcohol as a free initiator, Scheme 4, which cre-
ated grafts on the MFC as well as an unbound, free
polymer.
The MFC was grafted with polymer chains of different
lengths, and the grafted samples were characterised with
FTIR in which an increase of the intensity of the carbonyl
1651
peak can be seen in the samples where higher ratios of free
initiator: monomer have been utilised, Fig. 3.
The grafted MFC was also analysed by Thermal Gravi-
metric Analysis (TGA) and Differential Scanning Calorime-
try (DSC). The thermal properties of the modified cellulose
were compared to properties of linear PCL, with the pur-
pose of understanding how the interaction between the
grafts on MFC and the polymer matrix would appear in a
biocomposite material. This was conducted by analysing
and comparing the crystallisation behaviour of the sur-
face-confined (grafted) and the unbound PCL. The TGA
measurements showed that the amount of PCL increased
for samples containing longer grafts, which also was sup-
ported by the results from the FTIR measurements. From
the DSC results it was possible to determine that the melt-
ing temperature increased with increased chain length of
grafted PCL and that the melting temperature of PCL
grafted from cellulose was lower than the melting temper-
ature of free PCL, probably due to morphological changes
in the PCL structure in the interface region. From the ther-
mograms from the DSC analysis, the degree of crystallisa-
tion was determined and for the PCL grafted on cellulose
it varied from 10–30 wt.%, depending on cooling rate and
polymer chain length [24].
To evaluate the dispersibility of grafted MFC samples in
nonpolar organic solvents, tests were performed and the
results showed that the dispersibility and the stability of
the dispersions increased with increased length of the
grafted PCL [24]. Unmodified MFC was, due to its hydro-
philic nature, not possible to disperse in the nonpolar sol-
vent, Fig. 4.
PCL-grafted MFC has been utilised in composites
[25,26]. Hult and Malmström et al. hot-pressed PCL-
grafted MFC together with pure PCL-films to produce bi-
layer laminates [25], Fig. 5. The aim of the study was to
investigate if the chain length of the PCL grafts would af-
fect the adhesive properties in the interfacial region be-
tween the modified fibre and the polymer matrix. For
this study, MFC-films grafted with PCL of several different
lengths were compared.
The toughness of the interface of the resulting bilayer
laminates was determined by a peel test using Dynamic
Mechanical Analysis (DMA). From the results, it could be
seen that the interfacial toughness increased with increas-
ing graft length of the PCL, and that the peeling energy was
1652 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
Fig. 5. Production of the bilayer laminates by hot pressing. Reproduced with permission of Elsevier from Ref. [25].
Fig. 6. Peel test of bilayer composites, target DP of 75, 150, 300 and 600 of
the PCL grafts. Reproduced with permission of Elsevier from Ref. [25].
so high suggested that plastic deformation in the polymer
matrix occured Fig. 6. This study indicated that physical
entanglements are of importance when designing biocom-
posites [25].
Malmström and Lindström et al. continued the work by
producing biocomposites of PCL grafted MFC dispersed in
a PCL matrix [26]. MFC polymer grafted samples were
prepared with three different graft lengths. Bionanocom-
posites were prepared by blending different amounts (0–
10 wt.%) of PCL-grafted MFC with pure PCL, followed by
hot-pressing. Tensile testing was used to evaluate the
Fig. 7. Cellulose nanowhiskers (left) and PCL grafted cellulose nanowhiskers (right). Reproduced with permission of The Royal Society of Chemistry from
Ref. [28].
mechanical properties of the resulting materials. The bio-
composites reinforced with 3 wt.% of polymer grafted
MFC showed an enhanced ductility compared to when
3 wt.% unmodified MFC was incorporated. At a MFC load-
ing of 10 wt.% in the matrix, increased graft length resulted
in higher strength. The improvement could be due to im-
proved adhesion at the interface, better dispersion of the
MFC and/or decreased water uptake. This result also indi-
cated that a network of MFC, with some ability to carry
strength, was formed in the material.
Several studies have been conducted performing SI-ROP
of e-CL and/or L-LA from cellulose nanocrystals (cellulose
nanowhiskers, CNW) [27–30]. As with the filter paper
and MFC, SI-ROP was performed from the surface of the
nanocrystals. The grafted CNWs showed improved disper-
sibility in organic solvents. Dufresne et al. studied the na-
tive CNW’s and PCL grafted CNW’s by TEM (PCL-g-CNW),
Fig. 7, where a clear difference can be seen before and after
the grafting [27]. The PCL-g-CNW’s were subsequently
incorporated into a PCL or PLLA matrix, respectively, either
via solvent casting or by melt blending. From DMA mea-
surements it could be seen that the storage modulus in
the materials increased with increasing content of PCL-g-
CNW’s. Furthermore, the addition of unmodified CNW’s
to PCL gave no change in the storage modulus, while
PCL-g-CNW’s resulted in an increased storage modulus
with increasing amounts of modified nanocrystals. From
rheological studies it was possible to determine that the
addition of unmodified CNW’s to the matrix gave no
change in visco-elastic properties, while the addition of
relatively large amounts of PCL-g-CNW’s resulted in a solid
 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
like behaviour, indicating that a network structure was
formed in the material.
In the case of PLA-grafted CNWs incorporated in a PLLA
matrix, DMA showed that the composite stiffness, above Tg,
was higher than for unmodified PLA and that the stiffness
increased with increasing nanofiller content, this most
likely due to the formation of a network. Below Tg no rein-
forcing effect could be seen due to the plasticizing effect of
the filler [27]. Huang et al. grafted CNW’s with PLA by
microwave assisted ROP [28]. In a second step the modified
CNW’s were dispersed in a PCL matrix and the mechanical
properties of the produced material were investigated and
compared to those of materials produced with starch
nanocrystals. Composites produced using the modified
CNW’s in a PLA matrix showed that both strength and
the elongation of the material could be increased at the
same time, as compared to pure PLA. It was also seen that
material produced from CNW’s gave a material with better
mechanical properties than a material produced from PCL
grafted starch nanocrystals.
From these studies it can be concluded that PCL-g-
CNW’s can be used for the manufacturing of bio-nanocom-
posites to produce more ductile materials with a higher
elongation at break.
In a recent study, Thielemans and Labet showed that it
was important to carefully wash the CNW’s prior to graft-
ing, in order to increase the reproducibility of the results
[29]. The authors recommended that 24 h soxhlet extrac-
tion with ethanol should be performed on all nanocrystals
prior to any kind of surface modification since impurities,
such as xylobiose, 1,6-anhydroglucose, vanillic acid and
3,4,5-trimethoxyphenol, affect the surface grafting and
could even initiate polymerisations.
Carlmark et al. grafted a cellulose model surface by ROP
of PCL in situ in a Quartz Crystal Microbalance with Dissipa-
tion (QCM-D) [31]. The study was performed to monitor the
polymerization in real-time. An organic catalyst, 1,5,7-
triazabicyclo[4.4.0]dec-5-ene (TBD), was used for the reaction
in two different concentrations (0.5 and 1.0 mol%) and it
was performed in bulk at room temperature. Measurements
showed that doubling the amount of catalyst increased the
amount of grafted polymer from the surface, although this
increase was less than twice the amount. The polymeriza-
tion was performed in cycles, with rinsing in between,
meaning that the attached chains had active end-groups
since the surface polymerization could be reinitiated.
In a more fundamental study, Rutland and Malmström
et al. grafted PCL from cellulose microspheres and utilised
them for AFM force measurements in a colloid probe
experiment [30]. The AFM measurements were performed
between a cellulose sphere mounted on the cantilever and
a grafted or ungrafted cellulose sphere on a flat surface to
investigate the macromolecular contact level. From the
results it could be seen that there was a stronger adhesion
between two grafted spheres than between a grafted and
an ungrafted sphere which was explained as being due
to entanglements. It was also shown that the adhesion
between two cellulose grafted spheres increased when
the temperature was raised above the melting temperature
of PCL since this allowed for co-crystallisation between
chains in the bulk and surface confined chains.
1653
5.2. ‘‘Grafting to’’
The ‘‘grafting to’’ methodology implies that preformed
polymer chains are coupled to a solid substrate carrying
suitable reactive sites. This can be performed using phys-
iosorption (i.e. by electrostatic forces) or by covalent
attachment. For the latter to be successful, it is necessary
to utilise viable and efficient coupling chemistry for the
reaction or the grafting density will be low. There are only
a few reports where cellulose has been grafted via ‘‘graft-
ing to’’ of polymers synthesized via ROP. Tsubokawa et al.
have grafted poly(2-methyl-2-oxazoline) (polyMeOZO), to
the surface of amino functional cellulose powder [32]. The
polymer was grafted to the substrate by a reaction be-
tween a living cationic chain end group of the polymer
and amino groups on the surface of the cellulose powder.
It was shown that the number of grafted polymer chains
increased with increasing number of amino groups on
the cellulose surface, even though the percentage of
amino groups used for grafting was decreased. It was also
observed that the number of grafts on the surface
followed the expected trend, and decreased with increas-
ing molecular weight of the polymer chains due to in-
creased steric hindrance. By grafting of polyMeOZOto to
the surface, the wettability of the surface was increased.
Hereby, they concluded that the wettability of cellulose
powder surfaces could be controlled by grafting polymer
to the surface.
In a second report using ‘‘grafting to’’, modification of
native cellulose by thiol-ene, ‘‘click’’ chemistry in combina-
tion with organo-catalysis has been reported by Córdova
and Hafrén et al. [33]. The thiol-ene chemistry is regarded
as an interesting environmentally friendly route as it does
not require metal catalysts or solvents for the modification
to be performed. Initially, the cellulose surface was ene-
modified, by reaction of 9-decenoic acid in the presence
of tartaric acid, and subsequently further reacted with sev-
eral different molecules with thiol end-groups, including
PCL, Scheme 5. Due to the low reactivity of the solid cellu-
lose, characterization was difficult, but successful grafting
of PCL chains to the surface could be verified by FTIR
spectroscopy.
6. Ring opening polymerisation from cellulose
derivatives – homogeneous grafting
6.1. ‘‘Grafting from’’
There are several reports in the literature describing
‘‘grafting from’’ soluble celluloses by ROP. Such materials
have gained significant interest in recent years, since
the grafting reaction often results in amphiphilic block-
copolymers with a comb-like architecture, giving rise to
interesting three-dimensional structures. The resulting
materials are often biodegradable and may be foreseen
to function as nanocarriers and find use in biomedical
applications, such as drug-delivery.
The first report of ROP from soluble cellulose originates
from 1986 and was published by Kahovec et al. [34]. In the
study, tosylated cellulose was used as the substrate for the
1654 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659

Scheme 5. Grafting-to via thiol-ene ‘‘click’’ chemistry of PCL onto cellulose.

Scheme 6. Schematic picture over the ring-opening polymerisation of methyloxazoline from tosylated cellulose.

grafting of 2-methyloxazoline, Scheme 6. The purpose of
the study was to investigate if the grafted polymer could
show catalytic properties; however, it only showed low
activity and the grafted cellulose was not characterised.
Similarly, in 2000, Binder et al. studied grafted 2-
methyloxazoline from tosylated cellulose and herein the
resulting cellulose-graft-poly(N-acetylentyleneimine) was
analysed by FTIR and 13C CP MAS NMR spectroscopy, as
well as elemental analysis and [35]. From the study it
was possible to conclude that the amount of polymer and
the graft density increased with the number of tosyl groups
on cellulose, and that essentially all tosyl groups on the
cellulose acted as initiators for ROP of 2-methyloxazoline.
Several groups have utilised cellulose diacetate (CDA)
for ROP where the remaining hydroxyl groups initiate the
grafting [36–42]. Nishio and Teramoto grafted CDA with
PLA through ROP of L-LA, both in solution and in bulk
[36]. The results showed that longer side chains are formed
when the polymerisation was performed in bulk as com-
pared to in solution. It was also found that the polymer
chain length could be controlled by the amount of mono-
mer added to the reaction mixture [36]. The development
of supramolecular structures of the formed materials was
investigated, and it was found that the materials showed
spherulitic growth (Fig. 8). The growth of crystals was
much slower for the grafted cellulose diacetate (CDA-g-
PLLA) than for pure PLLA. Enthalpy relaxation times, as
determined by DSC, of the materials were found to be quite
long compared to those of pure PLLA [37].
In an attempt to find a general relationship between the
compositions of polymer grafted cellulose acetates (CA’s)
and their thermal properties, Nishio et al. grafted L-LA, e-
CL, d-valerolactone (VL), and 3-hydroxy butyrate (HB) from
CA using ROP [38]. Both graft lengths and grafting densities
were varied during the polymerisations, and it was deter-
mined that an expression, based on a theory developed
by Reimschuessel [43] for comb-like polymer based on
side chains graft of alkyl chains, sometimes can be used
to determine the Tg. The results also showed that if the
grafting density of the chains was decreased, an increased
graft length was needed for crystallisation to occur. In
1999 Shirashi et al. co-grafted e-CL and L-LA by ROP from
CA, with the aim of introducing thermoplastic properties
into cellulose [39]. The grafting reactions were performed
both in a flask reactor and through reactive processing
using a twin-screw extruder. It was shown that short reac-
tion times (10–30 min) were sufficient for successful mod-
ification and 13C NMR spectroscopy showed that the
structure of the co-polymer contained short blocks of
P(L-LA) together with relatively large amounts of randomly
 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659 1655

Fig. 8. Polarized optical micrographs of typical spherulites observed for CDA-g-PLLAs and pure PLLAs: (a) CDA-g-PLLA (MS = 22) at Tic = 107 °C and
t = 4000 min; (b) CDA-g-PLLA (MS = 58) at Tic = 119 °C and t = 365 min; (c) CDA-g-PLLA (MS = 77) at Tic = 110 °C and t = 300 min and (d) pure PLLA-L at
Tic = 130 °C and t = 30 min. Tic denotes the isothermal crystallisation temperature and t the elapsed time and MS the molar substitution. Reproduced with
permission from the American Chemical Society, copyright 2004, from Ref. [37].

Scheme 7. ROP of e-CL from HEC and subsequently ROP of L-LA from HEC-graft-PCL. Reproduced with permission from the American Chemical Society,
copyright 2007, from Ref. [50].

polymerized sequences. To investigate the properties of as rather low values for the tensile strength and Younǵs
the grafted cellulose, flow testing and tensile testing modulus indicating that the modified cellulose had elasto-
were used, and the results show both a lowering in flow meric properties. Cheng et al. studied the structure and
temperature and an enhanced elongation at break as well thermal properties of CDA grafted with PLA and showed
1656 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
that the melting temperature was decreased by polymer
grafting, which would make an improvement in the ther-
mal processing of CDA [40].
Other cellulose derivatives such as HEC, HPC and EC
have also gained large interest as backbone for the ROP
grafting [12,13,44–53]. In 1999 Wang et al. showed that
LiCl could be used as a catalyst for the grafting of L-LA by
ROP from HEC and HPC [44]. The advantage of using LiCl
over the most common catalyst, SnOct2, is that it has low
toxicity and a low price.
In 2003, Burt and Shi published a report on ROP of e-CL
from HPC, the reactions were performed in bulk without
the use of a catalyst [45]. Films made from the grafted cel-
lulose were heterogeneous, due to the formation of micro-
crystalline regions in the PCL, as detected by optical
microscopy. Tan et al. and Yang et al. modified some of
the hydroxyl groups on HPC and HEC, respectively, with
trimethylsilyl (TMS) protecting groups prior to the ROP
grafting, which made it possible to dissolve the cellulose
derivatives in organic solvents [46,47]. This protocol was
also utilised to govern the grafting density from the HPC
and HEC, as the TMS-group do not initiate the polymerisa-
tion. Wang et al. grafted p-dioxanone (PDO) from EC by
ROP, the polymerisation was performed in bulk using
Sn(Oct)2 as catalyst [52]. It was found that copolymers
with longer poly(p-dioxanone) (PPDO) grafts had better
thermal stability than those having shorter grafts. PPDO-
graft-EC degraded faster than linear PPDO, due to a re-
duced crystallinity.
Yuan et al. grafted a block-copolymer of e-CL and L-LA
from EC using ROP, Scheme 7 [50]. Grafts were produced
with varying lengths of the PCL- and PLLA blocks to inves-
tigate if varying the sizes and proportions of the blocks
could give control over the thermal properties and the bio-
degradation time. The results showed that grafted EC con-
taining longer blocks of PCL had higher melting
temperatures than those with shorter PCL blocks. The re-
sults also showed that the introduction of a PLLA block in-
creased the degradation time of the material. Block-
copolymers from EC have also been produced by Yuan
et al. who grafted e-CL from EC by ROP and in a second step
added a block of poly(ethylene oxide) by a DCC-coupling
reaction [51]. a-Cyclodextrine was then trapped onto the
double grafts of the resulting comb-polymer, forming
deformable supramolecular objects such as micelles, cylin-
ders, vesicles and sheets, as shown by AFM.
Homogeneous grafting from unmodified, dissolved, cel-
lulose by ROP was first reported in 2008 by Wang and
coworkers [12]. An efficient solvent is required to molecu-
larly dissolve the cellulose fibres and the polymerisation
was performed in the ionic liquid 1-N-butyl-3-methylimi-
dazolium chloride ([Bmim]Cl). The monomer used for the
ROP was p-dioxanone, chosen for its ability to biodegrade.
It was determined that the resulting material exhibited a
glass transition, which cannot be detected for pure
cellulose. The material showed no ability to crystallise
which, according to the authors, indicated that the grafted
polymer prohibited hydrogen bonding in and between
cellulose chains. Similarly, in 2009 Wu and Zhang et al.
grafted L-LA from unmodified cellulose in ionic liquid
(1-allyl-3-methylimidazolium chloride (AmimCl)) [13]. It
was determined that the material exhibited thermoplastic
properties at higher molar substitution values, and that it
exhibited a lower thermal stability than both pure cellu-
lose and PLLA. WAXD and DSC showed that the material
was amorphous indicating, according to the authors, that
no hydrogen bonding was present in the material. Disks
made from the grafted fibres showed a high homogeneity
and were transparent when the moulding temperature
was relatively low; at higher temperatures the disks be-
came slightly yellow due to degradation of the material.
6.2. ‘‘Grafting from’’ by a combination of ROP and atom
transfer radical polymerisation (ATRP)
There are a few reports in the literature where cellulose
derivatives have been grafted utilising a combination of
ROP and other polymerisation methods, mainly Atom
Transfer Radical Polymerisation (ATRP). For example, Vlček
et al. grafted two different polymers from CA by combining
ROP with ATRP, creating so-called dual-grafts [41]. Prior to
the polymerisations, some of the hydroxyl groups on the
cellulose were modified with initiator groups for ATRP.
ROP of e-CL was then initiated from the remaining hydroxyl
groups on the cellulose and in a later step ATRP of styrene,
butyl acrylate or methyl methacrylate was performed. AFM
analysis showed that the copolymers formed worm-like
structures. In a similar approach, Yuan and Wei et al. used
a combination of ATRP and ROP to graft PDMAEMA and
PCL from EC [53]. The polymer grafted cellulose showed po-
tential as being used as nanocarrier for controlled drug re-
lease. Oishi et al. were able to graft cellulose chloroacetate
with several different polymers by grafting two different
polymers simultaneously using ATRP (for 3-ethyl-methac-
ryloyloxy-methyloxetane and methyl methacrylate) and
ROP (for e-CL and L-LA) [42]. The ATRP was initiated by
the chloroacetate modified groups on the cellulose while
unmodified hydroxyl groups initiated the ROP. The glass
transition temperatures, and the temperature of maximum
degradation rate of the grafted cellulose varied depending
on which monomers that were used for grafting.
Combinations of ATRP and ROP have also been used to
create block-copolymer grafts from cellulose derivatives.
Berthier et al. grafted HPC with the purpose of producing
a system for fragrance delivery in a solution containing
high levels of ethanol [48]. The grafting was performed
by ROP of L-LA from the HPC, followed by ATRP of tert-bu-
tyl acrylate (t-BA) from the PLLA chain-end. In a third step
the poly(tert-butyl acrylate) (PtBA) was hydrolysed into a
hydrophilic poly(acrylic acid) (PAA) block, and the final
material showed good properties for potential use in the
desired area. Similarly, Malmström et al. used HPC for
the ROP of e-CL followed by end group modification of
the PCL into ATRP initiators, and in a third step, t-BA was
grafted from the PCL by ATRP [49]. To give the resulting
copolymer amphiphilic properties, for potential applica-
tion for formation of micelles for drug delivery, the PtBA
block was deprotected to give PAA by acidolysis and
was subsequently cross-linked. The cross-linked polymer
(HPC-g-(PCL-b-PAA-x)) was spin-casted onto mica and
analysed by AFM (Fig. 9), it showed some interesting nano-
sized features.
 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
Fig. 9. Schematic pictures of the comb-like polymers synthesized with an HPC backbone (left) and an AFM micrograph of the cross-linked polymer spin-
casted onto mica (right). Reproduced with permission from the American Chemical Society, copyright 2008 from Ref. [49].
7. Enzymatic grafting of cellulose
Two studies in which enzymes are used to catalyse ROP
from cellulose substrates have been published [54,55]. In
1999, Wang et al. grafted HEC films with PCL by the use
of a lipase [54]. The grafted films were investigated by FTIR
spectroscopy and TGA measurements which confirmed a
successful grafting. The degree of substitution was esti-
mated by results from 1H NMR and found to be between
0.10 and 0.32 per anhydroglucose unit. In 2004 Martinelle
et al. used a lipase for the ROP of e-CL in close proximity to
cellulose fibres in a filter paper [55]. In a first step, the en-
zyme was immobilised on the filter paper used as sub-
strate, and in a second step the polymerisation was
performed. This did not create covalently bonded grafts,
but the PCL still coated the cellulose surface, as determined
by FTIR spectroscopy and contact angle measurements.
8. Outlook
Grafted cellulosic fibres and cellulose derivatives are
anticipated to find a wide range of applications. Biocom-
posites are mainly being explored in two areas; as various
construction materials (load-bearing or not) and in a range
of biomedical applications. Today, cellulose fibre rein-
forced composites are already finding use in many fields,
ranging from construction industry to the automotive
industry. By grafting polyesters from the surface of cellu-
loses, and incorporating them into a polyester matrix, com-
posites that are completely degradable, yet have good
mechanical properties, are created. Most likely, in the next
years to come, biofibres and the corresponding biocompos-
ites will find many new applications, given that their sur-
face properties can be appropriately controlled and
tailored. One such foreseen application area is biomedical
materials since cellulose is biocompatible but does not de-
grade in the body. For example, Gatenholm et al. have pro-
duced artificial blood vessels of bacterial cellulose (BC)
[56]. BC has also proven to function as excellent scaffolds
for tissue engineering, as it can be grown in many different
forms and shapes. The market for nanocomposites is also
believed to grow in the near future due to the interesting
properties of the MFC fibres, which give a low weight
1657
material with high strength and stiffness [3,10,11]. NCC
is of great interest as filler material as it is easier to dis-
perse in a polymer matrix and as it does not adsorb as
much moisture as traditional fibres [11].
In terms of polymer grafted cellulose derivatives, such
as HPC, HEC, EC etc., one future application is believed to
be as drug carrier for controlled release, this due to that
the materials often have good biocompatibility, often are
biodegradable and that the properties of the fibres are rel-
atively easy to modify. By grafting the cellulose with vary-
ing amounts and combinations of polymers, e.g. PCL and
PLLA, the biodegradability rate can be controlled and step-
wise degradation can be achieved [50], this could increase
the ability for the material to be used for controlled re-
lease. The grafted fibres commonly self-assemble into mi-
celles in aqueous solutions which may lead to future
applications as solubilizing agents for hydrophobic drugs.
As grafted cellulose derivatives often have good mechani-
cal properties in combination with good biocompatibility
they are utilised as material for surgical repair, an area that
potentially could expand further. The ability of polymer
grafted cellulose derivatives to be used to encapsulate vol-
atile molecules for slow release in ethanol has been inves-
tigated by Berthier et al., [48] where a possible application
could be in perfumes.
Acknowledgements
The Swedish Research Council (VR), the Swedish Re-
search Council FORMAS, Biofibre Materials Centre (BiMaC
Innovation), the Swedish Centre for Biomimetic Fibre Engi-
neering (Biomime) and the Wallenberg Wood Science Cen-
tre (WWSC), are gratefully acknowledged for financial
support over the years.
References
[1] Roy D, Semsarilar M, Guthrie JT, Perrier S. Cellulose modification by
polymer grafting: a review. Chem Soc Rev 2009;38(7):2046–64.
[2] Klemm D, Heublein B, Fink H-P, Bohn A. Cellulose: fascinating
biopolymer and sustainable raw material. Angew Chem Int Ed
2005;44(22):3358–93.
[3] Eichhorn SJ, Dufresne A, Aranguren M, Marcovich NE, Capadona JR,
Rowan SJ, et al. Review: current international research into cellulose
nanofibres and nanocomposites. J Mater Sci 2010;45(1):1–33.
1658 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
[4] Saheb DN, Jog JP. Natural fiber polymer composites: a review. Adv
Polym Technol 1999;18(4):351–63.
[5] Lönnberg H. Ring-Opening Polymerization from Cellulose for
Biocomposite Applications [Doctoral Thesis]. Stockholm: KTH Royal
Institute of Technolology; 2009.
[6] Eichhorn SJ, Baillie CA, Zafeiropoulos N, Mwaikambo LY, Ansell MP,
Dufresne A, et al. Review: current international research into
cellulosic fibres and composites. J Mater Sci 2001;36(9):2107–31.
[7] Tizzotti M, Charlot A, Fleury E, Stenzel M, Bernard J. Modification of
polysaccharides through controlled/living radical polymerization
grafting-towards the generation of high performance hybrids.
Macromol Rapid Commun 2010;31(20):1751–72.
[8] George J, Sreekala MS, Thomas S. A review on interface modification
and characterization of natural fiber reinforced plastic composites.
Polym Eng Sci 2001;41(9):1471–85.
[9] Malmström E, Carlmark A. Controlled grafting of cellulose fibres – an
outlook beyond paper and cardboard. Polym Chem 2012;3(3):
727–33.
[10] Siro I, Plackett D. Microfibrillated cellulose and new nanocomposite
materials: a review. Cellulose 2010;17(3):459–94.
[11] Klemm D, Kramer F, Moritz S, Lindström T, Ankerfors M, Gray D,
et al. Nanocelluloses: a new family of nature-based materials.
Angew Chem Int Ed 2011;50(24):5438–66.
[12] Zhu J, Wang W-T, Wang X-L, Li B, Wang Y-Z. Green synthesis of a
novel biodegradable copolymer base on cellulose and poly(p-
dioxanone) in ionic liquid. Carbohydrate Polym 2009;76(1):139–44.
[13] Yan C, Zhang J, Lv Y, Yu J, Wu J, Zhang J, et al. Thermoplastic
cellulose-graft-poly (L-lactide) copolymers homogeneously
synthesized in an ionic liquid with 4-dimethylaminopyridine
catalyst. Biomacromolecules 2009;10(8):2013–8.
[14] Bledzki AK, Gassan J. Composites reinforced with cellulose based
fibres. Prog Polym Sci 1999;24:221–74.
[15] Jerome C, Lecomte P. Recent advances in the synthesis of aliphatic
polyesters by ring-opening polymerization. Adv Drug Deliv Rev
2008;60(9):1056–76.
[16] Dubois P, Coulembier O, Raquez J.-M, editors. Handbook of Ring-
Opening Polymerization, 2009.
[17] Dechy-Cabaret O, Martin-Vaca B, Bourissou D. Controlled ring-
opening polymerization of lactide and glycolide. Chem Rev
(Washington, DC, United States) 2004;104(12):6147–76.
[18] Hafren J, Cordova A. Direct organocatalytic polymerization from
cellulose fibers. Macromol Rapid Commun 2005;26(2):82–6.
[19] Cordova A, Hafren J. Direct organic acid-catalyzed polyester
derivatization of lignocellulosic material. Nord Pulp Pap Res J l
2005;20(4):477–80.
[20] Hafren J, Cordova A. Direct Bronsted acid-catalyzed derivatization of
cellulose with poly(L-lactic acid) and D-mandelic acid. Nord Pulp Pap
Res 2007;22(2):184–7.
[21] Lönnberg H, Zhou Q, Brumer III H, Teeri T, Malmström E, Hult A.
Grafting of cellulose fibers with poly (Îl-caprolactone) and poly (L-
lactic acid) via ring-opening polymerization. Biomacromolecules
2006;7(7):2178–85.
[22] Zhou Q, Greffe L, Baumann MJ, Malmström E, Teeri TT, Brumer III H.
Use of xyloglucan as a molecular anchor for the elaboration of
polymers from cellulose surfaces: a general route for the design of
biocomposites. Macromolecules 2005;38(9):3547–9.
[23] Yang Q, Pan X, Huang F, Li K. Synthesis and characterization of
cellulose fibers grafted with hyperbranched poly(3-methyl-3-
oxetanemethanol). Cellulose (Dordrecht, Netherlands)
2011;18(6):1611–21.
[24] Lönnberg H, Fogelström L, Berglund L, Malmström E, Hult A. Surface
grafting of microfibrillated cellulose with poly (Îl-caprolactone) –
Synthesis and characterization. Eur Polym J 2008;44(9):2991–7.
[25] Lönnberg H, Fogelström L, Zhou Q, Hult A, Berglund L, Malmström E.
Investigation of the graft length impact on the interfacial toughness
in a cellulose/poly(epsilon-caprolactone) bilayer laminate. Comp Sci
Technol 2011;71(1):9–12.
[26] Lönnberg H, Larsson K, Lindström T, Hult A, Malmström E. Synthesis
of polycaprolactone-grafted microfibrillated cellulose for use in
novel bionanocomposites-influence of the graft length on the
mechanical properties. Acs Appl Mater Interface 2011;3(5):1426–33.
[27] Goffin A-L, Raquez J-M, Duquesne E, Siqueira G, Habibi Y, Dufresne A,
et al. From interfacial ring-opening polymerization to melt
processing of cellulose nanowhisker-filled polylactide-based
nanocomposites. Biomacromolecules 2011;12(7):2456–65.
[28] Lin N, Chen G, Huang J, Dufresne A, Chang PR. Effects of polymer-
grafted natural nanocrystals on the structure and mechanical
properties of poly(lactic acid): a case of cellulose whisker-graft-
polycaprolactone. J Appl Polym Sci 2009;113(5):3417–25.
[29] Labet M, Thielemans W. Improving the reproducibility of chemical
reactions on the surface of cellulose nanocrystals: ROP of e-
caprolactone as a case study. Cellulose (Dordrecht, Netherlands)
2011;18(3):607–17.
[30] Nordgren N, Lönnberg H, Hult A, Malmström E, Rutland MW. ACS
Appl Mater Interfaces 2009;1(10):2098–103.
[31] Carlsson L, Utsel S, Wågberg L, Malmström E, Carlmark A. Surface-
initiated ring-opening polymerization from cellulose model surfaces
monitored by a quartz crystal microbalance. Soft Matter
2012;8(2):512–7.
[32] Tsubokawa N, Iida T, Takayama T. Modification of cellulose powder
surface by grafting of polymers with controlled molecular weight
and narrow molecular weight distribution. J Appl Polym Sci
2000;75(4):515–22.
[33] Zhao GL, Hafren J, Deiana L, Cordova A. Heterogeneous ‘‘organoclick’’
derivatization of polysaccharides: photochemical thiol-ene click
modification of solid cellulose. Macromol Rapid Commun
2010;31(8):740–4.
[34] Kahovec J, Jelinkova M, Janout V. Polymer-supported oligo(N-
acetyliminoethylenes). New phase-transfer catalysts. Polym Bull
(Berlin, Germany) 1986;15(6):485–90.
[35] Mais U, Binder WH, Knaus S, Gruber H. Synthesis and 13C CP MAS
NMR spectroscopy of cellulose-graft-poly(N-acetylethylenimine).
Macromol Chem Phys 2000;201(16):2115–22.
[36] Teramoto Y, Nishio Y. Cellulose diacetate-graft-poly (lactic acid)s:
synthesis of wide-ranging compositions and their thermal and
mechanical properties. Polymer 2003;44(9):2701–9.
[37] Teramoto Y, Nishio Y. Biodegradable cellulose diacetate-graft-
poly(L-lactide)s: thermal treatment effect on the development
of supramolecular structures. Biomacromolecules 2004;5(2):
397–406.
[38] Teramoto Y, Ama S, Higeshiro T, Nishio Y. Cellulose acetate-graft-
poly (hydroxyalkanoate)s: synthesis and dependence of the thermal
properties on copolymer composition. Macromol Chem Phys
2004;205(14):1904–15.
[39] Yoshioka M, Hagiwara N, Shiraishi N. Thermo-plasticization of
cellulose acetates by grafting of cyclic esters. Cellulose (Dordrecht,
Netherlands) 1999;6(3):193–212.
[40] Deng W, Zhuang X, Guan K, Cheng B. Structure and thermal
properties of cellulose diacetate-graft-poly(lactic acid) copolymer.
Adv Mater Res (Durnten-Zurich, Switzerland). 2011;221(Advanced
Polymer Science and Engineering):85–89.
[41] Vlcek P, Janata M, Latalova P, Dybal J, Spirkova M, Toman L.
Bottlebrush-shaped copolymers with cellulose diacetate backbone
by a combination of ring opening polymerization and ATRP. J Polym
Sci Part A: Polym Chem 2007;46(2):564–73.
[42] Chang F, Yamabuki K, Onimura K, Oishi T. Modification of cellulose
by using atom transfer radical polymerization and ring-opening
polymerization. Polym J (Tokyo, Japan) 2008;40(12):1170–9.
[43] Reimschuessel HK. Glass transition temperature of comblike
polymers: effects of side chain length and backbone chain
structure. J Polym Sci Polym Chem Edition 1979;17(8):2447–57.
[44] Xie W, Chen D, Fan X, Li J, Wang PG, Cheng HN, et al. Lithium
chloride as catalyst for the ring-opening polymerization of lactide in
the presence of hydroxyl-containing compounds. J Polym Sci Part A:
Polym Chem 1999;37(17):3486–91.
[45] Shi R, Burt HM. Synthesis and characterization of amphiphilic
hydroxypropylcellulose-graft-poly(e-caprolactone). J Appl Polym
Sci 2003;89(3):718–27.
[46] Wang C, Dong Y, Tan H. Biodegradable brushlike graft polymers. I.
Polymerization of e-caprolactone onto water-soluble
hydroxypropyl cellulose as the backbone by the protection of the
trimethylsilyl group. J Polym Sci Part A: Polym Chem 2002;41(2):
273–80.
[47] Jiang C, Wang X, Sun P, Yang C. Synthesis and solution behavior of
poly(e-caprolactone) grafted hydroxyethyl cellulose copolymers. Int
J Biol Macromol 2011;48(1):210–4.
[48] Berthier DL, Herrmann A, Ouali L. Synthesis of hydroxypropyl
cellulose derivatives modified with amphiphilic diblock copolymer
side-chains for the slow release of volatile molecules. Polym Chem
2011;2(9):2093–101.
[49] Östmark E, Nyström D, Malmström E. Unimolecular nanocontainers
prepared by ROP and subsequent ATRP from hydroxypropyl-
cellulose. Macromolecules (Washington, DC, United States).
2008;41(12):4405–4415.
[50] Yuan W, Yuan J, Zhang F, Xie X. Syntheses, characterization, and
in vitro degradation of ethyl cellulose-graft-poly (e-caprolactone)-
block-poly (L-lactide) copolymers by sequential ring-opening
polymerization. Biomacromolecules 2007;8(4):1101–8.
 A. Carlmark et al. / European Polymer Journal 48 (2012) 1646–1659
[51] Yan Q, Yuan J, Kang Y, Yin Y. Dynamic supramacromolecular self-
assembly: deformable polymer fabricated nanostructures through a
host-controlled approach. Polym Chem 2010;1(4):423–5.
[52] Zhu J, Dong X-T, Wang X-L, Wang Y-Z. Preparation and properties of
a novel biodegradable ethyl cellulose grafting copolymer with poly
(p-dioxanone) side-chains. Carbohydrate Polym 2010;80(2):350–9.
[53] Yan Q, Yuan J, Zhang F, Sui X, Xie X, Yin Y, et al. Cellulose-based
dual graft molecular brushes as potential drug nanocarriers:
stimulus-responsive micelles, self-assembled phase transition
behavior, and tunable crystalline morphologies. Biomacromolecules
2009;10(8):2033–42.
[54] Li J, Xie W, Cheng HN, Nickol RG, Wang PG. Polycaprolactone-
modified hydroxyethyl cellulose films prepared by lipase-catalyzed
ring-opening polymerization. Macromolecules 1999;32(8):2789–92.
[55] Gustavsson Malin T, Persson Per V, Iversen T, Hult K, Martinelle M.
Polyester coating of cellulose fiber surfaces catalyzed by a cellulose-
binding module-Candida antarctica lipase B fusion protein.
Biomacromolecules 2004;5(1):106–12.
[56] Petersen N, Gatenholm P. Bacterial cellulose-based materials and
medical devices: current state and perspectives. Appl Microbiol
Biotechnol 2011;91(5):1277–86.
Eva Malmström received her PhD in polymer
technology in 1996 from KTH Royal Institute
of Technology, Sweden. After a postdoctoral
stay at IBM Almaden Research Center, US, she
re-joined KTH as an assistant professor in
1997. In 2005 she became full professor and
in 2009 she was appointed deputy president
for KTH. She has published more than 100
papers and is the co-inventor of two patents
and one of the founders of Polymer Factory
Sweden AB. Professor Malmström’s main
research interests are the interplay macro-
molecular architecture and properties, renewable materials and con-
trolled polymerisations.
1659
Dr. Anna Carlmark received her PhD in
polymer technology in 2004 from Fibre and
Polymer Technology at KTH Royal Institute of
Technology in Sweden, under the supervision
of Prof. Eva Malmström. Between 2004 and
2007 she was employed as a researcher in
several industries in Sweden (GE Healthcare
AB, Gyros AB and SweTree Technologies AB)
before she re-joined the group of Prof.
Malmström in 2007 as an assistant professor
in the Division of Coating Technology at KTH.
Her research focus is within the fields of
controlled radical polymerisation, the synthesis of complex macromo-
lecular architectures, functional surfaces and (bio)fibre modifications.
Emma Larsson received her master’s degree
in chemical science and engineering in 2010
from KTH Royal Institute of Technology in
Sweden. She is currently working to earn her
PhD under the supervision of Prof. Eva
Malmström and Assistant Professor Anna
Carlmark in the Division of Coating Technol-
ogy and within Wallenberg Wood Science
Center (WWSC). Her research is in the field of
wood fibre modification with tailored poly-
mers.