polymer

Received25th September2008

First published as an Advance Article on the web1st May2009

DOI:10.1039/b808639g

This critical review is concerned with the recent advances in graft polymerisation techniques involving cellulose and its derivatives.It summarises some of the features of cellulose structure and cellulose reactivity.Also described are the various techniques for grafting synthetic polymers from the cellulosic substrate.In addition to the traditional grafting techniques,we highlight the recent developments in polymer synthesis that allow increased control over the grafting process and permit the production of functional celluloses that possess improved physical properties and chemical properties(1references).

Introduction

Cellulose is the most abundant organic raw material andfinds applications in areas as diverse as composite materials, textiles,drug delivery systems and personal care products. Since it wasfirst characterised in1838,1,2this inexpensive, biodegradable and renewable resource has received a great deal of attention for its physical properties and chemical reactivity.Many properties of cellulose,both physical and chemical are significantly different from those of synthetic polymers.Despite all its well established and interesting properties,cellulose lacks some of the versatile properties of synthetic polymers.This point explains the ever continuing interest in the understanding of the nature of cellulose, cellulose derivatives,cellulose composites,cellulose copoly-mers and blends that contain cellulose.

In recent years,environmental awareness has driven research into the modification of biofibres such as cellulose to increase their functionality and the scope of their use. Modification by graft polymerisation provides a means of altering the physical and chemical properties of cellulose and increasing its functionality.With the recent progresses in polymer synthesis,new routes are now available for the production of functional and sustainable cellulose–based materials.In this critical review,the structure of cellulose and its reactivity,together with highlights of the recent advances in techniques for cellulose grafting are considered.

1.Cellulose as a substrate

1.1Structural aspects of cellulose

The chemical and physical properties of the cellulose bio-polymer are largely dependent on its specific structure.The

a The University of Leeds,Leeds UK LS29JT.

E-mail:ccdjtg@leeds.ac.uk

b Department of Chemistry,Southern Methodist University,Dallas, Texas75275-0314,USA

c Key Centre for Polymers&Colloids,School of Chemistry,

the University of Sydney,Sydney,NSW2006,Australia.

E-mail:s.perrier@chem.usyd.edu.au;Fax:+61(0)293513329; Tel:+61(0)29351

3366 Debashish Roy Debashish Roy received his PhD in Polymer Chemistry from the University of Leeds, UK in2007.His studies with A/Professor Se´bastien Perrier focused on the design and syntheses of natural–synthetic hybrid materials from cellu-lose using a RAFT polymeri-zation technique.His research interests relate to the surface modification of inorganic and organic substrates,complex macromolecular architectures and stimuli responsive poly-mers.He is currently working

in the group of Professor Brent S.Sumerlin at Southern Methodist University,USA as a postdoctoral fellow.His current research project involves investigation of the syntheses and applications of advanced macromolecular architectures based on boronic acid-containing

polymers.

Mona Semsarilar

Mona Semsarilar obtained her

BSc in Pure Chemistry from

Azad University in Iran,

graduating in2003.She then

undertook an MSc in Phy-

sical Chemistry and Poly-

mer Science at the Eastern

Mediterranean University in

Northern Cyprus,graduating

in2006.She is now completing

her PhD in Polymer Chemistry

under the supervision of

A/Professor Se´bastien Perrier

at the University of Sydney.

Her work entails the use of

living radical polymerization to fabricate and study novel natural–synthetic polymer hybrid materials.

2046|Chem.Soc.Rev.,2009,38,2046–20This journal is c The Royal Society of Chemistry2009 CRITICAL REVIEW www.rsc.org/csr|Chemical Society Reviews

polymeric structure of cellulose was first demonstrated by Staudinger in 1920.He used the acetylation and de-acetylation reactions of cellulose and recognised that the structure was not simply based on an aggregation of relatively small,ring-like anhydroglucose oligomers held together by secondary bonds.Rather,the anhydroglucose units (AGUs)were found to be linked to each other by covalent bonds to form a large macromolecular chain.1,2

Understanding the structure of cellulose is a pre-requisite to controlling its modification.Fig.1shows the molecular structure of cellulose generated from repeating b -D -anhydro-glucopyranose units that are joined together covalently through acetal functions between the equatorial group of the C4carbon atom and the C1carbon atom (b -1,4-glycosidic bonds).1The chemical stability of the cellulose molecule is determined by the sensitivity to hydrolytic attack of the b -1,4-glycosidic linkages between the glucose repeating units.3

The molecular structure of cellulose suggests that it is a linear-chain polymer with a large number of hydroxyl groups (three OH groups per AGU unit).The linearity of structure arises from the b -glucose link at C1–O–C4to yield cellobiose units.This linear structure can be extended to molecules containing 1000–1500b -glucose units.3The degree of linearity enables the molecules to approach together.Thus,cellulose has a high cohesive energy that is greatly enhanced by the fact that the hydroxyl groups are capable of forming extensive hydrogen bond networks between the chains and within the chains.These aspects are responsible for the stiffand straight

chain nature of the cellulose molecules as well as the cause of the considerable tendency of the chains to organise in parallel arrangements into crystallities and crystallite strands,the basic elements of the supramolecular structure of the cellulose fibrils and the cellulose fibres.2,3Thus,the physical properties and the chemical reactivity of fibrous cellulose are not only influenced by the chemical constitution of the cellulose molecules but are also determined by the supramolecular structure of cellulose,i.e.the overall arrangement of chain molecules in a fibre.3

Cellulose can exist in at least five allomorphic forms.4Cellulose I is the form found in nature.In the 1980s,it was discovered that native cellulose,from almost every source,is also present in two different crystalline cellulose I modifica-tions (I a and I b ),which can be found alongside each other in statistically variable proportions.5In the cellulose I crystal lattice structure,the hydroxyl groups on the carbon atoms,C3,of the glucopyranose rings are involved in intramolecular hydrogen bonds with the pyranose ring oxygen atom,O50,of neighbouring glucose units in the same molecular chain.A possible second intramolecular hydrogen bonding exists between the hydroxyl groups on the carbon atoms C6and C20of neighbouring glucose base units.Intermolecular hydro-gen bonding also exists between the hydroxyl groups on the carbon atoms C6and C300of cellulose molecules that are adjacently located on the same plane.Both types of intra-molecular hydrogen bonding and intermolecular hydrogen bonding are illustrated in Fig.2and 3.1

In a recent study,Nishiyama et al.6,7reported the crystal structure and hydrogen bonding system of cellulose I a and cellulose I b ,using atomic-resolution synchrotron X-ray and neutron diffraction.In cellulose I a ,alternating glucose units in each chain differ slightly in conformation,but all of the chains are the same.However,cellulose I b has chains of two kinds arranged in alternating sheets.The authors pointed out that C–H ÁÁÁO hydrogen bonds play a prominent part in the cohesion of sheets of cellulose chains,together in a

stack.

James T.Guthrie

Professor J.T.Guthrie is Professor of Polymer Science at the University of Leeds via progression from 1972–to date.His research interests include interactions in poly-mer composite systems;membranes and membrane transport;polymer materials in cosmetics;co-polymer com-posites and blends;polymers in plastics,coatings and inks;high throughput methods for polymer solution and disper-sion delivery;polymers in bio-sensors,biomedical products

and reagent delivery;solid state,solution and melt properties of polymeric systems;polymer characterisation methods;migra-tion behaviour in polymeric compositions,development and removal of transients;and additives in the optimisation of the performance properties of polymer composites and

blends.

Se

bastien Perrier A/Professor Se ´bastien Perrier

graduated from the Ecole National Supe `rieure de Chimie de Montpellier,France,in 1998.He undertook his PhD at the University of Warwick,England,in polymer chemistry and spent one year as a postdoctoral fellow at the University of New South Wales,Australia.In 2002,he was appointed a lecturer at the University of Leeds,and promoted to senior lecturer in 2005.In 2007,he moved to the University of Sydney,and was

appointed director of the Key Centre for Polymers and Colloids.He leads a team working at the interface of organic chemistry,polymer synthesis,and materials science.

This journal is

c

The Royal Society of Chemistry 2009Chem.Soc.Rev.,2009,38,2046–20|2047

Cellulose may occur in other crystal structures denoted cellulose II,III and IV.Among these crystal structures, denoted cellulose II is the most stable structure of technical relevance.This structure can be formed from cellulose I by treatment with an aqueous solution of sodium hydroxide or when native cellulose is regenerated from solutions of semi-stable derivatives.1,2

According to modern theories of cellulosefibre structure, the long cellulose molecules can individually form part of two or more crystalline(ordered)regions.8Thus,the ordered regions are held together by cellulose molecules that run from one crystalline region to another.The cellulose molecules are arranged in a random fashion between the crystalline regions, forming regions of amorphous cellulose.The relative amounts of crystalline cellulose and amorphous cellulose have a considerable influence on the properties of the cellulosefibre.8 It was suggested by Howsmon and Sisson9that there is a range of degrees of order in the packing of chain molecules in afibre. Chemical reagents cannot penetrate the crystalline regions unless they can simultaneously disrupt them.10The penetra-tion of reagents is thereby restricted to those disordered regions that are accessible to reagents.These regions range from low degrees of disorder to the completely disordered (amorphous)regions.Adsorption/absorption of the reagents may also be possible on/through the surface of the crystalline regions.However,this would be dependent on the packing of the chain molecules.Thefibre accessibility is,thus,dependent on the crystalline–amorphous structure.Transport of chemical reagents to the reaction sites can occur within the less-ordered (amorphous)regions.Swelling of thefibre structure may help to increase the degree of thefibre accessibility.3

The morphological build-up of the native cellulosefibre also has an important influence on its physical and chemical properties.The accessibility of reagents to the functional groups of cellulose may be restricted due to the complex physical structure.11The morphological structure of cellulose (cotton)fibre consists of three major parts.These are the primary wall,the secondary wall and the lumen.11The reaction behaviour of cellulose is decisively determined by the different cell wall layers,particularly the primary wall in which thefibrils are laid down in a criss-cross fashioned helical manner,and the outer layer of the secondary wall that has a strong effect on the swelling behaviour and,thus,physical properties and the chemical properties of cellulose.3

The primary wall consists of a network of cellulosefibrils that are covered with an outer layer(cuticle)of pectin,protein, mineral matter and wax.The waxy layer renders thefibre impermeable to water and to aqueous solutions.10The lumen is the channel in the centre of thefibre from which the layers of the cellulose were laid down in the secondary wall while the fibre was growing.The secondary wall is almost pure cellulose and represents about90%of the totalfibre weight.It consists almost entirely of cellulosicfibrils that are packed together in a near-parallel arrangement.The layers of thefibrils lie in spiral formation along thefibre.The direction of the spirals can change(from S-to Z-twist and vice versa)many times in the same layer.Thefibrils are made of the bundles of smaller microfibrils that may be0.02–0.03m m in thickness and approximately10m m in length.Furthermore,the microfibrils consist of smallerfibrils that are about0.004Â0.006m m in cross-section.The assembly offibrils in the cottonfibre is such that it causes chemical reagents to be accessible to thefibrillar surfaces only,by way of a system of voids and channels.Some of the surfaces of the elementaryfibrils are very close to one another,causing them to be completely inaccessible to all chemical reagents without a prior swelling operation.These structural features indicate that the accessibility of the chain segments to various reagents varies across thefibre.10Other physical forms of cellulose such as regenerated cellulose,e.g. viscose(stillfibrillar but different from native cellulose)or micron sized cellulose, e.g.whiskers and microcrystalline cellulose,follow a similar analogy although their structures are not as complicated as that of the native cellulose infibrillar form.1,3There are other physical forms of cellulosics,namely cellulose derivatives and dissolved cellulose in which the structure is made from polyglucan chains.3,9

1.2Cellulose reactivity

According to its molecular structure,cellulose is an active chemical due to the presence of the three hydroxyl groups in each glucose residue.In most instances the hydroxyl groups at the2and3positions behave as secondary alcohols while the hydroxyl group in the6position acts as a primary alcohol (Fig.4).

These hydroxyl groups are mainly responsible for the reac-tions of cellulose.If all three of the hydroxyl groups of anhydroglucose unit are substituted due to reaction,the situation that exists is usually expressed such that the degree of substitution(DS)is3.0.Therefore,the DS indicates the average number of OH groups of the anhydroglucose unit of the cellulose molecule that have been substituted.12However, the reaction of cellulose should not simply be regarded as being that of a trihydric alcohol that is similar in its

chemistry 2048|Chem.Soc.Rev.,2009,38,2046–20This journal is c The Royal Society of Chemistry2009

to sugars having three hydroxyl groups.This is because cellulose is a fibre-forming and a high molar mass substance.In addition,the reactions of cellulose nearly always take place under heterogeneous conditions (this is only the case if the soluble forms of cellulose are not being considered).The heterogeneity arises from the fact that the solid cellulose is usually suspended in a liquid reaction medium.Moreover,cellulose itself is heterogeneous in nature,as different parts of its constituent fibrils display very different accessibilities to the same reagent.11In a way,one may consider that the majority of cellulose derivatives are cellulosic,‘‘very random block’’,copolymers.This applies in circumstances when the DS value is neither zero nor 3.In such derivative ‘‘block’’copolymers,anhydroglucose units with different degrees of substitution will be present in the polymer chain,along with anhydroglucose units that carry no substituent groups and anhydroglucose units that are fully substituted.

The reactivity of these three hydroxyl groups under heterogeneous conditions can be affected by their inherent chemical reactivity,by steric effects that are produced by the reacting agent,and by steric effects that are derived from the supramolecular structure of cellulose.12From esterification studies it has been found that the C6OH-group can react ten times faster than the other groups.The C2OH-group has been observed to react twice as fast as the C3OH-group in etherification reactions.In comparison with the other two secondary alcoholic groups,the primary alcoholic group at C6has an axis of free rotation around the C5to C6bond.It has also been observed by infrared spectroscopy that rotational isomers exist in alkali-swollen cellulose.The reactivity of the primary alcoholic groups seems to be related to this isomerisation.13However,in general,the relative reactivity of the hydroxyl groups can be expressed as OH–C6c OH–C24OH–C3.13

Thus,the overall reactivity and processability of cellulose depend on the constitution of the cellulose molecule,the fibre morphology and its regularity,the degree of crystalline order,the interfibrillary bonds and interfibrillar interstices,the voids and the capillaries.Activation treatments can enhance the accessibility and the reactivity of cellulose for subsequent reactions by (i)opening surface cannulae,internal pores and cavities,and interfibrillar interstices,(ii)disrupting fibrillar aggregations,in order to make available additional accessible surface,(iii)disturbing the crystalline order and (iv)altering the crystal modification,thus changing the hydrogen bonding scheme and the relative availability of the reactive hydroxyls.1,3As a result,a higher degree of substitution with a more homogeneous substituent distribution,and thus a higher yield can be achieved,resulting in a product with better properties.Accessibility treatments of cellulose can be performed in various ways such as swelling,solvent exchange,inclusion,

degradation or mechanical grinding.1,3Among these,swelling

is the most frequently used activation method for cellulose modification.Cellulose undergoes swelling in solutions of acids,bases,and salts as well as in some organic solvents.Swelling agents generally penetrate the highly ordered regions,and split bonds between chains and fibrils of cellulose.Thus,cellulose can behave as an acid,as a base and as an amphoteric species,depending on the prevailing medium in which it is present.This logic helps in explaining the increased cellulose grafting efficiency that can occur under acidic conditions.An aqueous sodium hydroxide solution is the most frequently used swelling agent for cellulose to improve its chemical and physical properties.1Its effect on the structure,morphology,accessibility and reactivity of cellulose is now well established.1,14,15After alkali treatment,the structure of native cellulose fibres remains fibrillar but the degree of disorder increases.Moreover,conversion of the crystallites in the native fibre,on change from the cellulose I form to the cellulose II form can be retained after the sodium hydroxide has been washed out and the sample dried.14,16For systems in which organic solvents are present,water can be used as the swelling medium,as long as the organic solvent has compatibility with water,e.g.,the THF–water system.

Treatment involving the use of alkaline media largely destroys the microfibrillar integrity of the cellulose.Thus,this process has an effect on the supramolecular structure of cellulose and enhances interfibrillar swelling,weakening the intermolecular hydrogen bonds between cellulose microfibrils and loosening the packing.As a result,the accessibility of chemical reagents to the molecules is improved.1.3Chemical modification of cellulose

Cellulose is itself a unique polymeric product and possesses several attributes such as a fine cross section,the ability to absorb moisture,high strength and durability,high thermal stability,good biocompatibility,relatively low cost and low density yet good mechanical properties.However,cellulose has some inherent drawbacks.These include poor solubility in common solvents,poor crease resistance,poor dimensional stability,lack of thermoplasticity,high hydrophilicity (not desirable for several composite applications)and lack of antimicrobial properties.To overcome such drawbacks,the controlled physical and/or chemical modification of the cellulose structure is necessary.13

The constitution and chemical composition of natural cellulose cannot be modified in the same ways as those that can be applied to a synthetic polymer because these features of cellulose are determined by biosynthesis.Introducing functional groups into cellulose molecules through chemical modification is one of the key ways of alleviating this problem.These functional groups can confer new properties to the cellulose without destroying its many desirable intrinsic properties.

The industrial history of the chemical modification of cellulose to impart new properties can be tracked back to 1870with the production of the first thermoplastic polymeric material ‘‘celluloid’’(cellulose nitrate plasticized with camphor)which was manufactured by the Hyatt

Manufacturing

Fig.4The numbering system for carbon atoms in anhydroglucose unit of cellulose.

This journal is

c

The Royal Society of Chemistry 2009

Chem.Soc.Rev.,2009,38,2046–20|2049

The use of etherification reactions provides another important way of modifying the structure of cellulose chemically.Some commercially important cellulose ethers are methyl cellulose,carboxymethyl cellulose and the hydroxyalkyl celluloses.These cellulose esters and ethers have potential applications in coatings,laminates,opticalfilms, sorption media,pharmaceuticals,foodstuffs and cosmetics.2 The reaction of cellulose with bi-or poly-functional compounds to form cross-linked or resinification products in the cellulose matrix,provides another possible way of modifying the structure of cellulose.17These methods can bring stability to the structure of cellulose and can impart crease-resistance

(or‘‘durable press’’properties)to cellulose.18

The synthesis of cellulose graft copolymers marked thefirst departure from the traditional means of modifying cellulose, until then undertaken via the chemical reactions described above.In1943,Ushakov attempted to copolymerise some allyl esters and vinyl esters of cellulose with the esters of maleic acid.He obtained some insoluble products that were probably thefirst graft copolymers ever reported.1Since this invention, extensive studies have been carried out on the synthesis, properties and applications of cellulose graft copolymers.19–24 Research studies on the cellulose graft copolymers through to 1986have been reviewed by Hebeish and Guthrie,13and by Samal et al.21

2.Methods of cellulose graft copolymerisation

A graft copolymer generally consists of a long sequence of one monomer,referred to as the backbone polymer(main chain) with one or more branches(grafts)of long sequences of a different monomer.25

Among the methods of modification of polymers,graft copolymerisation offers an attractive and versatile means of imparting a variety of functional groups to a polymer.26In fact,polymeric materials with valuable properties can be achieved via graft copolymerisation by changing parameters such as the polymer types,the degree of polymerisation and the polydispersities of the main chain and the side chains,the graft density(average spacing in between the side chains)and the distribution of the grafts(graft uniformity).Graft copoly-merisation permits one to combine the best properties of two or more polymers in one physical unit.According to the end use or specific needs,tailor-made graft copolymers can be synthesised using specific polymerisation methods.27

The synthesis of cellulose graft copolymers is one of the key ways of modifying the physical properties and chemical prop-erties of cellulose.13This is usually achieved by modifying the cellulose molecules through the creation of branches(grafts)of synthetic polymers that impart specific properties onto the cellulose substrate,without destroying its intrinsic properties.13 Indeed,depending on the polymer that is grafted onto the cellulose,it is possible to attain properties such as dimensional-stability,resistance to abrasion and wear,wrinkle recovery,oil and water repellence,elasticity,sorbancy,ion exchange capabilities,temperature responsiveness,thermal resistance and resistance to microbiological attack.28–33 Fig.5illustrates a schematic structure of a cellulose graft copolymer.

The graft copolymerisation of many monomers onto cellulose and onto cellulose derivatives has been carried out by different methods that can be generally classified into three major groups:(i)free radical polymerisation,(ii) ionic and ring opening polymerisation and(iii)living radical polymerization.

The methods of cellulose graft copolymerisation are based on one or more of three approaches,‘‘grafting-to’’the cellulose,‘‘grafting-from’’the cellulose,and‘‘grafting-through’’the cellulose.

The‘‘grafting-to’’approach.In the‘‘grafting-to’’approach, an end functional pre-formed polymer with its reactive end-group is coupled with the functional groups that are located on the cellulose backbone.The‘‘grafting-to’’approach is expressed schematically in Fig.6.25

The‘‘grafting-from’’approach.In the‘‘grafting-from’’approach,the growth of polymer chains occurs from initiating sites on the cellulose backbone.The‘‘grafting-from’’approach is expressed schematically in Fig.7.25

The‘‘grafting-through’’approach.In the‘‘grafting-through’’approach,a macromonomer,usually a vinyl macromonomer of cellulose,is copolymerised with a low molecular weight co-monomer as shown in Fig.8.25

Among these three approaches of grafting,the‘‘grafting-from’’approach is the most commonly used procedure. One of the major advantages of this approach is that a high graft density can be achieved due to the easy access of the reactive groups to the chain ends of the growing polymers.The‘‘grafting-to’’approach is inherently limited by the crowding of chains at the surface.This hinders the diffusion of chain ends to the surface for further attach-ment and consequently limits thefinal graft density.34 Although the‘‘grafting-through’’approach is relatively con-venient,it still requires the synthesis of cellulose-derived

macromonomers.

Fig.5A schematic representation of cellulose graft copolymer.

2050|Chem.Soc.Rev.,2009,38,2046–20This journal is c The Royal Society of Chemistry2009

2.1Conventional free radical graft copolymerisation

Radical polymerisation has received the greatest amount of attention among all of the polymerisation methods.About 60%of all available polymers are still obtained by this method.35This paramount position of free radical

polymerisation is due to its many attractive characteristics,including:34,36,37

(i)its applicability to the polymerisation of a wide range of monomers such as (meth)acrylates,(meth)acrylamides,styrene,butadiene,vinyl acetate and the water-soluble mono-mers such as acrylic acid,the hydroxyacrylates and N -vinyl pyrrolidone.

(ii)its ability to provide an unlimited number of copolymers.

(iii)its tolerance to a wide range of functional groups (e.g.OH,NR 2,COOH,CONR 2)and reaction conditions (bulk,solution,emulsion,mini-emulsion,suspension).

(iv)its tolerance to water or other impurities in contrast to the great sensitivities of ionic polymerisation and of coordination polymerisation.

(v)the convenient,mild reaction conditions and wide temperature range under which it can be conducted.

(vi)it is simple to implement and inexpensive compared with competitive technologies.

Radical polymerisation 34is a chain reaction process,consisting of mainly three steps,initiation,propagation and termination.The chains are initiated by initiator derived radicals and add to monomers.The subsequent addition of monomers to the propagating radical is the basis of the step of chain propagation.Chain termination takes place when the propagating radicals react by combination,dispropor-tionation and transfer.

Vinyl graft polymerisation and acrylic graft copolymerisation to a pre-existing polymeric backbone via a free radical mechanism generally involves the ‘‘grafting-from’’

approach.38

The reaction starts with the generation of free radical sites on the main backbone.The vinyl or acrylic monomer then reacts with the free radical sites to propagate a new polymer chain that can be covalently anchored to the backbone.The presence of new polymeric chains gives new properties to the backbone polymer.

In the free radical graft copolymerisation of cellulose,free radicals on the cellulosic backbone can be formed by chemical initiators,15by irradiation with ultraviolet light,39by irradia-tion with gamma rays,40and by exposure to plasma ion beams.41The free radicals thus formed can add to monomers to form a covalent bond between the monomer and the cellulose.A free radical site is also formed on the newly formed branch.Many monomers may add subsequently to the free radical site of the branch.The propagation of the branch continues until termination occurs either by combination of two growing cellulose chains or by a disproportionation mechanism,where a hydrogen atom is abstracted by another growing polymeric chain.Termination may also occur by chain transfer to monomer,to initiator,to dead polymer,to additives and to impurities.The general mechanism of the free radical grafting of vinyl and acrylic monomers onto cellulose is illustrated in Scheme 1.13

2.1.1Chain transfer.Chain transfer reactions have been extensively used in cellulose grafting.42,43In this method,the termination of a polymerisation is carried out in the presence of cellulose.The polymerisation is then terminated

via

Fig.7A schematic representation of the ‘‘grafting-from’’

approach.

Fig.6A schematic representation of the ‘‘grafting-to’’approach.

initiators,such as azobisisobutyronitrile(AIBN),hydrogen peroxide and benzoyl peroxide can be used.1,44

(a)Introducing thiol groups or xanthate ester groups.Since the graft copolymerisation that is mediated by purely radical transfer is not amenable to obtaining high graft yields, compounds with greater chain transfer activity,such as thiol groups,can be introduced into the cellulose molecules before grafting,for example by reaction with ethylene sulfide (see Scheme3).Excellent yields of graft copolymer can be achieved.In this process there is a reduction of at least50%in the amount of homopolymer that would otherwise be produced.45

Prior to grafting,the cellulose can also be modified to provide a few xanthate ester groups.46The high radical transfer activity of the xanthate ester group results in high polymer add-ons with reduced homopolymer formation.

Different vinyl and acrylic monomers have been grafted onto cellulose via the xanthate method.The graft poly-merisation of acrylonitrile,mediated by the cellulose xanthogenate–hydrogen peroxide redox system,is illustrated in Scheme4.

In a recent study,Zahran et al.47reported the use of a cellulose thiocarbonate–azobisisobutyronitrile initiation system to induce graft copolymerisation of methyl acrylate and other acrylic monomers onto the cellulosic backbone.

(b)Potassium persulfate initiation.Initiation via the use of potassium persulfate involves the generation of the initiating species on the swollen cellulose substrate backbone.22This is usually carried out by saturating the cellulose substrate with potassium persulfate(KPS).48–51The mechanism of persulfate initiation is shown in Scheme5.

Ghosh and Das52modified cotton cellulose by grafting acrylic acid in the presence of potassium persulfate as free radical initiator.

Suo et al.53applied this initiation technique for the graft copolymerisation of acrylic acid and acrylamide monomers onto carboxymethyl cellulose,to make cellulose-based super absorbent polymers.Similarly,Aliouche et al.synthesized cellulose substrates with improved absorption and retention via the graft copolymerisation of acrylic acid and acryronitrile using KPS as the initiator.54The graft copolymerisation

ofitaconic acid onto cellulosefibres using potassium persulfate has also been reported.55

The persulfate initiation system has been used in conjunction with xanthate systems.In a recent study,El-Hady and Ibrahim56grafted acrylamide onto a xanthate mixture of carboxymethyl cellulose with a sodium bisulfate–ammonium persulfate redox initiator system in water.Further studies have been published on methyl methacrylate(MMA)grafting via the peroxydiphosphate-metal ion-cellulose thiocarbonate system.57

(c)Fenton’s reagent initiation(Fe2+–H2O2system).The use of Fenton’s reagent to initiate graft polymerisation with methyl methacrylate onto cellulose has been described by a several research groups.58–61Fenton’s reagent involves a redox reaction between the Fe(II)ion and hydrogen peroxide to produce hydroxyl radicals.38In the grafting of cellulose,the Fe(II)ion is adsorbed on cellulose substrates in thefirst step.Subsequent contact of the substrate with an aqueous solution of hydrogen peroxide results in the formation of hydroxyl radicals.These radicals abstract a hydrogen atom from the cellulose to form cellulosic radicals that may initiate graft polymerisation in the presence of a vinyl or acrylic monomer.It is also possible for the hydroxyl radical to react with the monomer,leading to a homopolymer.22Since most of the decomposition reaction of hydrogen peroxide is expected to occur in the vicinity of cellulose on which the Fe(III)ion is adsorbed,the possibilities of homopolymer formation are lowered.Cellulose graft copolymerisation via Fenton’s reagent initiation is shown in Scheme6.

2.1.2Direct radical formation on the cellulose backbone via chemical activation

(a)Direct oxidation via Ce(IV)ions62.Among the various chemical initiation methods,the formation of free radicals on the cellulose molecules by direct oxidation with Ce(IV)ions has gained considerable importance,due to its ease of application and its high grafting efficiency compared with other known redox systems such as the Fe(II)–hydrogen peroxide system.30,44,63–67In this technique,the Ce(IV)ions produce free radicals on the cellulose backbone in the presence of acid (e.g.HNO3).When cellulose is oxidised by a salt such as a cerium ammonium nitrate,free radicals are formed on the cellulose by a single electron transfer process.13These radicals are capable of initiating the graft polymerisation of vinyl monomers.The actual initiation reaction is quite complex,22 and is thought to involve the oxidation of cellulose to form an intermediate reversible complex(which may be a chelated species) between the Ce(IV)ion and the cellulose.This is followed by the disproportionation of the complex to form a radical on the cellulose molecule.30The principles of the reaction are illustrated in Scheme7.22The rates of polymerisation obtained with this redox system are very high,even at low temperatures.65

It has been suggested that such grafting on cellulose occurs predominantly at the reducing hemiacetal end-group and at the C2-C3glycol in the anhydroglucose units rather than at the C6hydroxyl groups.68

Grafting of vinyl monomers to cellulose by the Ce(IV)ion initiation method,through the1980s has been reviewed by McDowall et al.30Recently,the Ce(IV)ion method has been studied in considerable detail by Gupta and Khandekar for cellulose graft copolymerisation.They applied this method to synthesise a thermoresponsive cellulose copolymer via the graft copolymerisation of N-isopropylacrylamide.29The graft copolymerisation of ethyl acrylate onto cellulose has also been investigated.69The graft copolymerisation of acrylamide and ethyl acrylate,acrylamide and methyl methacrylate, acrylonitrile and methyl methacrylate,acrylonitrile and ethyl methacrylate onto cellulose from their binary mixtures using the Ce(IV)ion method has been studied in detail.63,70–75In a recent publication,Chen and Hsieh76used the Ce(IV)ion technique to graft acrylic acid onto the surface of ultrafine cellulosefibres(diameter200–400nm),to investigate

the Scheme7The simplified mechanism of Ce(IV)ion initiated cellulose graft copolymerization.

absorption behaviour and activity of bound lipase enzymes.A Ce(IV )–cellulose thiocarbonate redox system was found to be effective in grafting methacrylic acid and monomers such as acrylic acid,acrylamide,acrylonitrile,butyl acrylate,methyl methacrylate,ethyl methacrylate and glycidal methacrylate.Kondo et al.77observed the acceleration effect of the Ce(IV )-initiated grafting of methyl methacrylate or acrylo-nitrile onto cellulose film in the presence of ultrasound.Higher percentages of grafting and grafting efficiency (490%)were observed when the cellulose substrates were agitated by ultra-sound.It was proposed that ultrasound activation promoted the diffusion of the Ce(IV )ions,as well as the dispersion and absorption of monomers onto the cellulose substrate.

The free radicals produced from Ce(IV )ions have relatively low efficiency in grafting initiation in aqueous media due to the hydrolysis of the Ce(IV )ions.63In principle,no homopolymer should be formed,as the radicals are almost exclusively formed on the cellulose backbone.13However,small amounts of homopolymer are always observed,due to direct reactions of the Ce(IV )ions with the monomers.22,30,78

Other oxidising agents have also been studied for use in cellulose grafting.These include Mn(III ),79Mn(IV ),33,80–82Co(III ),83,84V(V ),85,86IO 4Àions,87–and potassium ditellurato-argentate(III ).90

(b)Cellulosic initiators.If the cellulose material can be chemically converted into an initiator such as a peroxide (cellulose-hydroperoxide),it can then decompose into radicals and initiate either graft copolymerisation,or homopolymerisation,as shown in Scheme 8.Homopolymer formation can be largely reduced by the use of a reducing agent.The peroxides can be formed by ozonization or by irradiation of the substrate in air or in a hydrogen peroxide solution.

Kubota and Kuwabara 91reported the preparation of a percarboxymethyl cellulose initiator by treating cellulose with hydrogen peroxide in sulfuric acid.Photochemical decomposi-tion of the peracid group in the presence of acrylic acid,acrylamide and N -isopropyl acrylamide initiated free radical graft copolymerisation with high percentages of grafting.Esters or carbonates of N -hydroxypyridine-2-thione (Barton esters)have been immobilised onto carboxymethyl cellulose or onto hydroxypropyl cellulose.Irradiation of the cellulose-immobilised Barton esters in the presence of a monomer,initiated the free radical graft copolymerisation of styrene,acrylamide and N -isopropyl acrylamide (NIPAAm).92The graft copolymerisation of styrene with the hydroxypropyl cellulose–Barton ester derivative is shown in Scheme 9.

Karlsson and Gatenholm 93used an ozone treatment to form hydroperoxides on the cellulose substrate before the graft

polymerisation of 2-hydroxyethyl methacrylate (HEMA).Hydroperoxide radical formation was increased by treating the pre-swelled fibres with a humified ozone–oxygen mixture.The addition of a bifunctional crosslinking monomer,diethyleneglycol dimethylacrylate (DEGDMA)increased the graft ratio by up to 300%.

In addition to peroxides,diazonium salts have also been introduced into the cellulose molecule as an initiating group.94This method involves the introduction of an aromatic amine onto the cellulose backbone,followed by its conversion to its corresponding diazonium salt.Upon heating,the diazonium salt decomposes to produce free radicals,leading to graft copolymerisation,when a suitable monomer is present.The reactions involved in this method are illustrated in Scheme 10.(c)Initiation by radical attack on an unsaturated group.If a cellulose chain contains unsaturated groups,it can undergo graft copolymerisation.Unsaturated groups may be introduced onto cellulose by reacting it with compounds having two functional groups,one of which is a double bond.13Thus,the partial esterification of cellulose with methacryloyl anhydride allows the grafting of cellulose with methacrylate based monomers.

A range of vinyl or allyl cellulose derivatives has been studied.22,95The problem associated with this method is the potential crosslinking reactions that may occur if there are several unsaturated groups on each cellulose chain.Therefore,the degree of substitution and the monomer reactivity ratios must be considered.

Toledano-Thompson et al.48described a two step method for grafting acrylic acid (AA)onto cellulosic microfibres.The cellulose was initially treated with an epoxide containing a terminal double bond.Thus,the cellulose was functionalized with a terminal double bond.The grafting of the AA onto the cellulosic fibre was then carried out using a free radical initiator such as potassium persulfate.The resultant copolymer exhibited excellent water sorption capacity.

2.1.3Radiation-induced graft copolymerisation.The initia-tion of cellulose grafting using radiation with ultraviolet light 39,96–100and especially high-energy radiation (g -rays

from

Scheme 8The general mechanism of graft polymerisation via

decomposition of a cellulosic hydroperoxide

initiator.

radioactive isotopes or electron beams)40,101–103has received considerable attention over several decades.The irradiation of cellulose leads to the formation of radicals that can initiate graft polymerisation in the presence of vinyl and acrylic monomers (see Scheme 11).1

(a)g -Radiation grafting.There are two major approaches to the radiation grafting of cellulose,the pre-irradiation method and the mutual irradiation method.

In the pre-irradiation method,the cellulose substrate is irradiated first,followed by the introduction of monomers and swelling agents.40The main active species that induce the grafting reaction are trapped radicals that are located at the interphase between the crystalline regions and the amorphous regions of the irradiated samples.In a recent study,hydrogels were synthesised by the pre-irradiation grafting of several monomers such as acrylamide (AAm),acrylic acid (AA),2-hydroxypropyl acrylate (HPA),2-hydroxypropyl methacry-late (HPMA)and N ,N 0-methylene bisacrylamide (BAAm)40using 60Co g -rays as a radiation source.Temperature sensitive poly(N -isopropylacrylamide)(PNIPAAm)101,104and pH sensitive poly[2-(dimethylamino)ethyl methacrylate]102(PDMAEMA)were also synthesised via the pre-irradiation method using 60Co g -rays.The pre-irradiation method leads to considerably less homopolymer than is provided by the mutual irradiation method,105although the degradation of the

cellulose backbone is usually greater,particularly when the grafting is carried out in the presence of air or oxygen.22,40In the mutual irradiation method,the cellulose substrate is irradiated directly in the presence of the monomers or monomer–solvent mixtures.This approach usually results in a consider-able amount of homopolymer being formed,due to the direct radiolysis of the monomer.22In the mutual irradiation-induced graft copolymerisation of acrylonitrile onto cellulosic filter paper,Badawy et al.106reported that the amount of homopolymer could be reduced by the introduction of a small amount of styrene monomer.However,the mutual irradiation method leads to less degradation of cellulose fibre,due to protective action of the vinyl and acrylic monomers during the actual irradiation.22,107In a recent publication,Kumar et al.103reported the preparation of an anion-exchange matrix,poly(vinylbenzyltrimethylammonium chloride)-g -cotton cellu-lose,using the mutual,irradiation induced grafting technique with a 60Co g -radiation source.The resultant material exhibited excellent protein adsorption properties.

The grafting of a monomer to cellulose using high energy radiation has some clear advantages,including the fact that the processes can be speedy and relatively simple to operate.This is partly because there is no ‘‘synthesis’’step to modify the cellulose substrate prior to polymerisation.38However,radical formation occurs mainly along the path of the incident radiation beam,and therefore the radical generation process is unselective.38In addition,the radiation degrades cellulose in a disproportionation reaction via the splitting of glycosidic linkages,108resulting in a loss of mechanical strength of the cellulose fibres.109

(b)Ultraviolet-radiation induced grafting.The grafting of various vinyl and acrylic monomers to cellulose using low energy ultraviolet radiation causes less degradation of the backbone polymer and gives an option for better control over the grafting reaction than can be achieved by the high energy g -radiation method.39,96This process is usually carried out in the presence of photoinitiators,which accelerate the process by forming additional radicals when exposed to UV.110These radicals can enhance grafting by creating sites in the backbone polymer via abstraction reactions,as shown in Scheme 12.110The use of UV radiation for the grafting of glycidyl methacrylate,2-hydroxyethyl methacrylate,styrene and acrylo-nitrile onto cotton cellulose in the presence of various photo-initiators such as uranyl nitrate,Ce(IV )ammonium nitrate and benzoin ethyl ether has been studied.38,49,96The grafting level was increased by pre-swelling the substrate,and by introducing divinylbenzene in the reaction medium.111,112Poly(methyl methacrylate)(PMMA)was also grafted to cellulose using UV radiation,in the presence of a charge transfer (CT)monomer complex as an additive.110The UV radiation method has also been used in conjunction with an oxidation process.The oxidation of cellulose with sodium metaperiodate,followed by decomposition of the

oxidised

Scheme 11The simplified mechanism of radiation-induced cellulose graft

polymerization.

(c)Plasma grafting.Plasma treatment has also been used to graft polymers to cellulose.The successful plasma graft copolymerisation of acrylamide116and2-hydroxyethyl meth-acrylate(HEMA)117onto cellulosefibre has been reported. In conclusion,free radical methods have been applied to the graft copolymerisation of a range of vinyl and acrylic mono-mers,using various types of cellulosic substrates,under relatively mild reaction conditions.However,these methods have some drawbacks.These drawbacks will vary in type and extent depending on the grafting method that is used. Examples of the drawbacks include:(i)chain scission of the cellulose backbone during the formation of free radical grafting sites;(ii)relatively few and very long(high molecular weight) grafted side chains;(iii)little control over the molecular weights and molecular weight distributions(polydispersities) of the grafted side chains;(iv)potentially large amounts of homopolymer formation and the difficulties associated with their removal.In addition the preparation of block copolymers is almost impossible due to the unreactive end-groups of the grafted polymeric chains.20,118–120

2.2Ionic and ring opening graft polymerisation of cellulose 2.2.1Ionic graft polymerisation.The stringent reaction conditions required by ionic polymerisations have meant that relatively few studies based on this approach to grafting have been undertaken.121–126

(a)Cationic graft polymerisation.The cationically initiated grafting of isobutylene and of a-methyl styrene onto a cellulosic substrate was studied by Rausing and Sunner.127 The initiator(a Lewis salt)was formed through the chemical reaction of a Lewis acid(boron trifluoride)with a Lewis base (cellulose hydroxyl groups).Boron trifluoride was adsorbed on the cellulose surface.The resulting complex catalyst systems initiated the graft copolymerisation by reacting,the cellulose reactive sites with isobutylene and a-methyl styrene. The resulting cellulosic substrate exhibited excellent water resistance properties.Scheme13summarises the cationic graft polymerisation of cellulose with isobutylene.13

Successful cationic graft polymerisation of2-methyl-2-oxazoline onto some cellulose derivatives has been achieved by Ikeda et al.126and by Cheradame et al.128The monomer cardanol,was grafted onto cellulosefilter paper using a boron trifluoride–diethyl ether complex(BF3ÁEt2O)as the cationic initiator.129An approximate15–25%grafting yield was obtained without significant gelation of the monomer.The resultant grafted Whatman No.1filter paper showed extensive water repellency.

(b)Anionic graft polymerisation.Graft polymers of acrylo-nitrile,methacrylonitrile and methyl methacrylate onto cellulose have been prepared by anionic graft polymerisation.130 The alkali metal alkoxides of the cellulose backbone were used as the initiator.The graft polymerisations were carried out in liquid ammonia and in other inert solvents,at low temperature.Similar types of experiment with sodium cellulosates and acrylonitrile were conducted by Schwenker and Pacsu.131This system was also studied in detail by Avny and Rebenfeld,122using Rayon as the substrate and tetra-hydrofuran as the solvent.In all cases,considerable homopolymerisation occurred via chain transfer reactions to the monomer and/or to the solvent.The isolated side chains of grafted cellulose were short(molecular weight around 15000g/mol),indicative of a high degree of substitution (DS).The mechanism of the anionic graft polymerisation of acrylonitrile on sodium cellulosate is illustrated in Scheme14.131When performing the anionic graft polymerisation of propylene sulfide onto cellulose,Cohen et al.123,124observed that homopolymerisation reactions occurred extensively at high monomer conversion.The molecular weights of the grafted chains increased with reaction time and with monomer conversion,suggesting that the grafted polymeric chains were living.

Mansour et al.132studied the anionic initiation process for the graft polymerisation of several monomers such as acrylo-nitrile,methyl methacrylate,ethyl methacrylate and allyl chloride onto cotton cellulose.In their study,cotton linters were transformed into cellulose xanthate before grafting.They found that the xanthate groups had a positive influence on the ionic grafting.However,considerable homopolymer was formed if a nitrile group or a chloride group was present in the monomers.Narayan and Shay125,133described the preparation of cellulose-g-poly(styrene)using the anionic graft polymerisation of styrene on tosylated cellulose acetate using BuLi–1,1-diphenylethylene atÀ781C.The resultant cellulose acetate tosylate–styrene graft copolymer was then subjected to mild hydrolysis with aqueous ammonia to yield a cellulose–styrene graft copolymer.It was found that on average,27 styrene monomer units were attached per glucose units.

The grafting of polymeric chains from cellulose via ionic polymerisation is still challenging,as the experimental condi-tions required to perform ionic graft polymerisation are very demanding(e.g.low temperature,highly pure reagents,inert atmosphere,anhydrous condition,etc).84Also the direct use of carbanions for grafting requires the thorough protection of the cellulose hydroxyl groups,to avoid side reactions.

2.2.2Ring opening polymerisation from cellulosefibres. There are very few reports on cellulose grafting via ring opening polymerisation(ROP).Hafren and Cordova134 reported thefirst direct organic acid-catalysed,ring

opening

polymerisation of cyclic monomers such as e-caprolactone (e-CL)with solid cotton and paper cellulose as initiators (Scheme15).Tartaric acid was used as the most efficient catalyst for the production of PCL-grafted cellulose.A11% weight gain was observed after the grafting of e-CL fromfilter paper,although the percentage yield of non-grafted PCL was490%.

In a recent publication,Loennberg et al.135reported the grafting of e-caprolactone and L-lactic acid onto cellulose fibres via ring opening polymerisation.In order to introduce more available hydroxyl groups at the cellulose surface,the filter paper was modified with xyloglucan-bis(methylol)-2-methylpropanamide and2,20-bis(methylol)propionic acid (bis-MPA)for the ROP of e-caprolactone and L-lactic acid, respectively.The graft polymerisation was controlled by the ratio of the amount of added free initiator to the amount of monomer.The results suggested that cellulose samples, modified with bis-MPA,showed an improved grafting efficiency.The resultingfibres showed good resistance against enzymatic degradation.

2.3Grafting of well defined polymers onto cellulose and onto cellulose derivatives

The grafting of well-defined polymeric chains onto cellulose has received only scarce attention.119,120,136In most studies, the polymer is pre-made by anionic polymerisation or by cationic polymerisation,followed by coupling of the polymer to the cellulose backbone.

Mansson and Westfelt119described a method for the grafting of well-defined,low molecular weight

poly(styrene),(2500,12100and17100g/mol,PDI=1.1)onto cellulose acetate.The grafting method involved the esterification of the free hydroxyl groups on the cellulose acetate(acetyl DS2.5) with a carboxylic acid–terminated poly(styrene)species that was prepared anionically.The poly(styrene)content in the copolymer varied between10and80%.A grafting yield of up to83%was obtained,using strictly anhydrous conditions, with a high concentration of esterification catalyst such as 4-(dimethylamino)-pyridine,for long reaction periods(up to 3days).The grafting reaction is shown in Scheme16. Biermann et al.120described an improved method of grafting monodisperse poly(styrene)onto mesylated cellulose acetate by reaction with a carboxylate–terminated poly(styrene). This anion displaces the mesylate groups from a mesylated cellulose acetate backbone via an SN2nucleophilic reaction and forms an ester linkage with the cellulose backbone in a homogeneous phase reaction.The reaction scheme for the grafting process is shown in Scheme17.

In this method,a grafting yield of up to68%was obtained when the polymerisation was carried out in dimethyl-formamide(DMF)at751C for20h.An average grafting density of one poly(styrene)chain per17anhydroglucose units was obtained.It was observed that the grafting yield was limited by the efficiency of the carboxylation reaction. Tsubokawa et al.136investigated the grafting of poly(2-methyl-2-oxazoline)and poly(isobutyl vinyl ether). These polymers possessed controlled molecular weights and narrow molecular weight distributions.The grafting involved the use of cellulose powder containing amino groups.The poly(2-methyl-2-oxazoline)and poly(isobutyl vinyl ether)were synthesised by cationic ring opening polymerisation and by living cationic polymerisation,respectively.These living polymer cations were then successfully grafted onto cellulose via the termination of the propagating chain with the amino groups that were present on the cellulose surface.

In all of the above methods,the mole number of polymer chains grafted onto the cellulose or the cellulose derivative, decreased with increase in the molecular weight of living polymer anion or cation,due to steric hindrance of neighbouring grafted polymers.These procedures,therefore,yield poor graft densities.The polymerisation procedures are tedious because they require demanding experimental conditions,such as an inert atmosphere and anhydrous conditions.92,118

2.4Cellulose grafting via living radical polymerisation methods

The recent advances in the theme of living radical polymerisation have been applied to cellulose grafting.The‘‘grafting from’’approach,i.e.,the synthesis of graft copolymers directly from the surface-immobilised initiators,provides a unique opportunity to tailor the surface properties of graft polymers by controlling the graft length,the architecture and the composition.The formation of large amounts of non-grafted polymer can also be avoided.

The term‘‘living polymerisation’’wasfirst defined by Szwarc137as a chain growth process,without chain breaking reactions such as chain transfer or irreversible termination. Living polymerisation mechanisms provide polymers of controlled composition,architecture and molecular weight distribution.Thus,end-functional polymers with narrow polydispersity can be achieved easily.Further,high purity block copolymers,graft,stars and other polymers with complex architectures can be obtained.The combination of a living mechanism with the versatile radical process gives more freedom in choosing a wide range of monomers and reaction conditions.34The concept of living radical polymer-isation(LRP)was originally introduced by Otsu et al.138,139in the early1980s,through the use of‘‘iniferters’’(ini tiator-trans fer agent-ter minator).The most efficient iniferters are based on organosulfur compounds such as dithiuram sulfides. However,‘‘iniferter’’systems have several drawbacks.These include the provision of a high polydispersity,poor control over the molecular weight,low initiation efficiency,inefficient thermal activation,and the formation of substantial amounts of homopolymer in conjunction with block copolymer formation.140,141

From the middle of1980s,aspects of living radical polymerisation expanded dramatically.To date,the living radical polymerisation methods that have received the greatest attention from industry and academia are nitroxide-mediated polymerisations(NMP),142–145atom transfer radical polymer-isations(ATRP)146–150and reversible addition–fragmentation chain transfer polymerisation(RAFT).36,37,140,151–154The common feature of these LRP methods is the use of reagents that convert chain propagating radicals into a‘‘dormant’’form that is in equilibrium with the‘‘active form’’,as shown in Scheme18.36

Under the appropriate reaction conditions,the concentra-tion of the‘‘active’’propagating radicals is kept very low to reduce the rate of their termination by self reaction.In addition,all of the propagating radicals have an equal chance of growth,ensuring the formation of polymers with low polydispersities.After the completion of the

polymerisationreaction,the majority of polymer chains may exist in the dormant form.The end group can then be reactivated in the presence of a second monomer,providing a route to the design of the macromolecular architecture.

2.4.1Nitroxide-mediated polymerisation(NMP).Nitroxide-mediated polymerisation(NMP)belongs to a large family of stable free radical polymerisation(SFRP)processes and is based on the use of a stable nitroxide radical.25TEMPO (2,2,6,6-tetramethyl-1-piperidinyloxy)is one the more commonly used nitroxides in this respect.In this method, the propagating species(P n )reacts with a stable radical (X )to form dormant species(P n–X).Thus,deactivation of propagating radicals occurs.The resulting dormant species can then reversibly cleave to reform the free radicals.Once P n forms,it can propagate by reacting with a monomer(M)or it can terminate with other growing radicals.NMP only requires the addition of an appropriate alkoxyamine to the polymer-isation system.Polymerization is usually undertaken at high temperatures(41201C for TEMPO).155This method is widely applicable to styrene and acrylate monomers.Its use for the polymerization of methacrylates and methacrylamides

requires the presence of specially designed nitroxides.156 A simplified mechanism of NMP is shown in Scheme19. Nitroxide mediated polymerisation(NMP)was thefirst living radical polymerisation method to be used in cellulose grafting.Daly et al.92reported thefirst use of nitroxide mediated,controlled radical grafting from cellulose and cellulose derivatives.The controlled radical grafting from hydroxypropyl cellulose(HPC)was performed using TEMPO monoadducts,formed from the HPC–Barton carbonate derivative(carbonates of N-hydroxypyridine-2-thione)1 (Scheme20).The photolysis of1in the presence of styrene and TEMPO provided adduct2.Heating the macro initiator at1301C provided styrene–HPC graft copolymers3.An increase in grafted polymeric chain length with increasing polymerisation time was observed.The polydispersity of the poly(styrene)grafts ranged from 1.3–1.5.However,the grafting was limited to styrene monomer only and required the use of high temperatures.

2.4.2Atom transfer radical polymerisation(ATRP).In the ATRP approach,a suitable alkyl halide undergoes a reversible redox process that is catalysed by a transition metal complex, (activator,M t n-Y/ligand,where Y may be another ligand or counter ion)to form an active radical and a metal halide complex by the oxidative addition of the halide(X).141The active radicals can react with monomer and propagate further with additional monomers or can abstract the halide atom from the oxidised metal complex(deactivator,X–M t n+1–Y/ ligand)forming a dormant alkyl halide species and the acti-vator.The alkyl halide species is then reactivated by the metal complex activator and propagates further.ATRP can be used on the vast majority of vinyl monomers and acrylic monomers over a wide temperature range.157The reaction has some tolerance to oxygen and other inhibitors.155 However,the reaction requires unconventional initiation systems that may have poor compatibility with polymerisation media.37In addition,the residues of the transition metal ATRP catalysts colour the polymer products and may also increase their toxicity.158The catalyst has to be removed from the polymer or recycled after synthesis.This step is still a major challenge for ATRP.158A general mechanism for ATRP is shown in Scheme21.

Among all of the living radical polymerisation techniques, ATRP is the most used with respect to the modification of the surface properties of cellulose and cellulose derivatives. Haddleton and Watersonfirst reported the use of molecules, bearing a hydroxyl group,that can be transformed into ATRP initiators.159Carlmark and Malmstrom160reported thefirst application of ATRP for the grafting of monomers

oncellulosefibres at an ambient temperature(Scheme22).The hydroxyl groups of a cellulosefilter paper were modified with 2-bromoisobutyryl bromide to form initiators at the cellulose surface.Methyl acrylate(MA)was grafted from the surface, in the presence of a sacrificial initiator.The resulting PMA-grafted paper showed excellent hydrophobicity.The same research group also reported the formation of block-copolymer grafts from cellulosefibres using ATRP.118In this study,a hydrophilic second layer of2-hydroxyethyl methacrylate(HEMA)was grafted onto a hydrophobic PMA layer to demonstrate that the hydrophilic/hydrophobic behaviour of a cellulose surface can be tailored using living radical polymerisations.A free initiator was introduced in all these systems(the so-called‘‘sacrificial initiator’’)to aid in the evaluation of the molecular weight of the grafted chains. Indeed,the direct measurement of the molecular weights of the grafted chains proved to be difficult.This difficulty was attributed to the very small amount of graft that formed from the cellulose surface,making the amount of cleaved polymer difficult to collect for analysis.

In a recent study,Nystrom et al.161reported the preparation of a superhydrophobic cellulose surface via the ATRP method. In this study,poly(glycidal methacrylate)(PGMA)was grafted from the ATRP initiator-modified cellulose surface. The pendant epoxide groups in PGMA were then hydrolysed under acidic conditions.The resultant hydroxyl groups of each graft chain end were in turn converted to ATRP initiators to polymerise GMA.The obtained polymer with a‘‘graft-on-graft’’architecture was post-functionalised with penta-decafluorooctanoyl chloride to provide afluorinated cellulose surface that possessed a very high contact angle,(41701). ATRP has also been used to modify a variety of cellulosic substrates,cellulose powder,162cellulose acetate or diacetate (CDA),163–165ethyl cellulose,166,167hydroxyethyl cellulose,168 hydroxypropyl cellulose,169and jutefibre.170Of particular interest is the comparative study undertaken on the surface modification of various solid polysaccharidic substrates such asfilter paper,microcrystalline cellulose,regenerated cellulose (e.g.Lyocellfibres and dialysis tubing)and chitosanfilm via the ATRP grafting of methyl acrylate(MA)and styrene.171 A considerable difference in the amount of grafted polymer was observed,depending on the substrate.It was found that greater amount of polymers could be grafted from native cellulose substrates than from regenerated cellulose substrates. Although most of the studies have concentrated on the modification of cellulose in its solid state,recent work has shown that ATRP polymerisation can also be applied to soluble polysaccharides,in homogeneous reaction conditions. In a recent publication,Rannard et al.172reported thefirst atom transfer radical graft copolymerisation at ambient temperature in water from a soluble polysaccharide such as locust Bean Gum.The hydroxyl groups of the polysaccharide were modified with an acid imadazolide to form ATRP initiators for the polymerisation of a series of water-soluble methacrylates and styrenic monomers.High molecular weight graft copolymers with targeted graft lengths were achieved.Yan and Ishihara grafted2-methacryloyloxyethyl phosphorylcholine(MPC)onto cellulose also in a homo-geneous medium.Cellulose cotton was dissolved in an N,N-dimethyl acetamide–lithium chloride[(DMAc)–LiCl] system,then converted to the macroinitiator by reaction with 2-bromoisobutyloyl bromide(BiBBR)in the presence of pyridine(DS=0.72).The resulting macroinitiator was used to grow MPC monomers from the cellulose backbone,in DMSO. Analyses of thefinal product indicated that the obtained copolymer had a controlled,well defined architecture.173

2.4.3Reversible addition–fragmentation chain transfer (RAFT)polymerisation.Reversible addition–fragmentation chain transfer(RAFT)polymerisation is one the most versatile and convenient methods of living radical polymerisation.36,140,151,174Of all the living radical polymerisation techniques,RAFT can be applied to the widest range of radically polymerisable monomers,using reaction conditions that are similar to those of free radical polymerisation.Indeed,the RAFT process is similar to a conventional free radical polymerisation in presence of a chain transfer agent(Scheme23),where the chain transfer agent is a thiocarbonyl thio compound(so called‘‘RAFT agent’’). The use of a RAFT agent leads to polymers of narrow polydispersities and of predetermined chain length.36The thiocarbonyl thio group present at the chain end of the polymer confers colour to thefinal product.156However,this moiety can easily be removed from the polymers,175,176via reduction,thermolysis,aminolysis,exposure to ultra-violet radiation and treatment with peroxides or sodium hypochlorite.36,177,178In principle,the versatility of the RAFT process would seem to make it the candidate of choice for the grafting of vinyl monomers and acrylic monomers onto cellulose. There is a very limited number of papers concerning the modification of the solid cellulose surface via

RAFT-mediated Scheme21The mechanism for ATRP.141graft polymerisation.Stenzel et al.179,180were thefirst to report the preparation of a trithiocarbonate chain transfer agent based on the soluble cellulose derivative,hydroxy-isopropyl cellulose(HPC),by attaching the RAFT agent to cellulose through its stabilising‘‘Z’’-group(Z-group approach),to obtain poly(styrene)comb polymers.Since the chain transfer agent was attached via its Z-group,the molecular weight of the poly(styrene)branch showed a pronounced deviation from the theoretical molecular weight due to reduced accessibility of the RAFT group at high polymerisation conversions(Scheme24).

The modification of cellulose under heterogeneous conditions via RAFT polymerisation wasfirst carried out by Perrier et al.181A RAFT agent was introduced ontofilter paper via its leaving/reinitiating‘‘R’’-group(R-group approach),and was used to mediate the polymerisation of styrene,methyl acrylate and methyl methacrylate.The technique was further developed and used to polymerise styrene182–185and dimethylamino ethyl methacrylate.184–188 Direct analyses of the grafted chains,after hydrolysis of the cellulose substrate,revealed that excellent control over molecular weight and molecular weight distribution was obtained.The introduction of a free RAFT agent in to the system was shown to improve the control over the reaction.186,187The styrene-modified cellulose showed increased hydrophobic properties,180while the grafting of poly(DMAEMA)chains onto cellulose substrate conferred to it antimicrobial properties.188Scheme25summarises the structure of the RAFT agent used in these studies.

In all these studies,the chain transfer agent was initially attached to the cellulose substrate and radicals were introduced via thermal initiation.Barner and co-workers have proposed an alternative way of growing polymeric chains from cellulose,using g-radiation to initiate the free radical poly-merisation of styrene from cellulose,and introducing a free RAFT agent to control the polymerisation.They obtained excellent molecular weight control and polydispersity.1 3.Conclusions

The modification of cellulose to increase its functionality has attracted the attention of researchers since the late19th century.The graft polymerisation methods developed to date have suffered from a lack of control,which has prevented the design of well defined materials.In recent years,the polymer functionalisation of cellulosic materials has greatly benefited from advances in polymerization techniques.Living radical polymerization techniques are among the more recent processes that have enabled the introduction of well-defined polymers,with precisely controlled structure,molecular weight,polydispersity,and functionalities onto cellulose surface.Tools to produce new cellulosic materials,that have potential uses in a wider range of applications than ever before,are now available.

Acknowledgements

The authors thank the University of Leeds(DR)and the University of Sydney(MS)for the provision of scholarships. References

1H.A.Krassig,Cellulose-Structure,Accessibility and Reactivity, Gordon and Breach Science Publisher,Yverdon,1993.

2D.Klemm,B.Heublein,H.P.Fink and A.Bohn,Angew.Chem., Int.Ed.,2005,44,3358–3393.

3H. A.Krassig,in Cellulose and Its Derivatives:Chemistry, Biochemistry and Applications,ed.J.F.Kennedy,G.O.Phillips,

D.J.Wedlock and P. A.Williams,Ellis Horwood Limited,

Chichester,1985,pp.3–25.

4T.P.Nevell,in The Dyeing of Cellulosic Fibres,ed.C.Preston, Dyer’s Company Publications Trust,Bradford,1986,pp.1–54. 5R.H.Atalla and D.L.Vanderhart,Science,1984,223,283–285. 6Y.Nishiyama,P.Langan and H.Chanzy,J.Am.Chem.Soc., 2002,124,9074–9082.

7Y.Nishiyama,J.Sugiyama,H.Chanzy and P.Langan,J.Am.

Chem.Soc.,2003,125,14300–14306.

8G.J.Cook,Handbook of Textile Fibres,5th edn,Merrow Publishing Co.Ltd,Durham,1993,vol.1Natural Fibres.

9J.A.Howsmon and W.A.Sisson,in Cellulose and Cellulose Derivatives,ed. E.Ott,H.M.Spurlin and M.W.Grafflin, Wiley-Interscience,New York,Second edn,1954,vol.1,p.251. 10T.P.Nevell,in Cellulosic Dyeing,ed.J.Shore,Society of Dyers and Colourists,Bradford,1995,pp.1–73.

11Cellulose Chemistry and Its Applications,ed.T.P.Nevell and S.H.Zeronian,Ellis Horwood Limited,Chichester,1985.

12P.J.Wakelyn,in Handbook of Fiber Chemistry,ed.M.Lewin and

E.M.Pearce,Marcel Dekker,New York,1998,pp.2–654. 13A.Hebeish and J.T.Guthrie,The Chemistry and Technology of Cellulosic Copolymers,Springer-Verlag,Berlin,1981.

14S.H.Zeronian,in Cellulose Chemistry and Its Applications, ed.T.P.Nevell and S.H.Zeronian,Ellis Horwood Limited, Chichester,1985,pp.159-180.

15J.Kunze and H.P.Fink,Macromol.Symp.,2005,223,175–187. 16E.Dinand,M.Vignon,H.Chanzy and L.Heux,Cellulose,2002, 9,7–18.

17S.L.Vail,in Cellulose Chemistry and Its Applications, ed.T.P.Nevell and S.H.Zeronian,1985,pp.384–422.

18M.P.Stevens,Polymer Chemistry,Oxford University Press, New York,third edn,1999.

19K.Ward,Jr,Chemical Modification of Papermaking Fibers, Marcel Dekker Inc.,New York,1973.

20V.Stannett,ACS Symp.Ser.,1982,187,3–20.

21R.K.Samal,P.K.Sahoo,H.S.Samantaray and S.P.Bhattacharjee,J.Macromol.Sci.R.M.C.,1986,26,81–141. 22V.T.Stannett and H.B.Hopfenberg,in Cellulose and Cellulose Derivatives,ed.N.M.Bikales and L.Segal,John Wiley and Sons, New York,1971,vol.5,pp.907–936.

23E.H.Immergut,in Encyclopedia of Polymer Science and Technology, ed.N.M.Bikales,Interscience,New York,1965,vol.3,pp.242–284. 24H.A.Krassig and V.Stannett,Adv.Polym.Sci.,1965,

111.25G.Odian,Principles of Polymerization,John Wiley and Sons, Hoboken,fourth edn,2004.

26A.Bhattacharya and B.N.Misra,Prog.Polym.Sci.,2004,29, 767–814.

27H.G.Borner and K.Matyjaszewski,Macromol.Symp.,2002, 177,1–15.

28G.S.Chauhan,S.Mahajan and L.K.Guleria,Desalination, 2000,130,85–88.

29K.C.Gupta and K.Khandekar,Biomacromolecules,2003,4, 758–765.

30D.J.McDowall,B.S.Gupta and V.T.Stannett,Prog.Polym.

Sci.,1984,10,1–50.

31A.Waly,F.A.Abdel-Mohdy,A.S.Aly and A.Hebeish,J.Appl.

Polym.Sci.,1998,68,2151–2157.

32K.M.El-Salmawi,M.M.Abu Zaid,S.M.Ibraheim,A.M.

El-Naggar and A.H.Zahran,J.Appl.Polym.Sci.,2001,82, 136–142.

33A.Hebeish and Z.H.El-Hilw,J.Appl.Polym.Sci.,1998,67, 739–745.

34G.Moad and D.H.Solomon,The Chemistry of Radical Polymerization,Elsevier Ltd,Oxford,second edn,2006.

35A.Sebenik,Prog.Polym.Sci.,1998,23,875–917.

36E.Rizzardo,J.Chiefari,B.Y.K.Chong,F.Ercole,J.Krstina, J.Jeffery,T.P.T.Le,R.T. A.Mayadunne,G. F.Meijs,

C.L.Moad,G.Moad and S.H.Thang,Macromol.Symp.,

1999,143,291–307.

37G.Moad,E.Rizzardo and S.H.Thang,Aust.J.Chem.,2005,58, 379–410.

38D.W.Jenkins and S.M.Hudson,Chem.Rev.,2001,101, 3245–3273.

39S.R.Shukla,G.V.G.Rao and A.R.Athalye,J.Appl.Polym.

Sci.,1992,45,1341–1354.

40E.Takacs,L.Wojnarovits,J.Borsa,J.Papp,P.Hargittai and L.Korecz,Nucl.Instrum.Meth.B.,2005,236,259–265.

41L.Andreozzi,V.Castelvetro,G.Ciardelli,L.Corsi,M.Faetti,

E.Fatarella and

F.Zulli,J.Colloid Interface Sci.,2005,2,

455–465.

42A.Hebeish,M.I.Khalil and M.H.Elrafie,Angew.Makromol.

Chem.,1974,37,149–160.

43G.N.Richards,J.Appl.Polym.Sci.,1961,5,558–62.

44M.Pulat and C.Isakoca,J.Appl.Polym.Sci.,2006,100,2343–2347. 45D.Chaudhuri and J.J.Hermans,J.Polym.Sci.,1961,51,373. 46R.W.Faessinger and J.S.Conte,US patent,3330787,1963. 47M.K.Zahran,M.Morsy and R.I.Mahmoud,J.Appl.Polym.

Sci.,2004,91,1261–1274.

48T.Toledano-Thompson,M.I.Loria-Bastarrachea and M.J.Aguilar-Vega,Carbohydr.Polym.,2005,62,67–73.

49S.R.Shukla and A.R.Athalye,J.Appl.Polym.Sci.,1994,54, 279–288.

50G.S.Chauhan,H.Lal,R.Sharma,L.Guleria and

B.D.Sarwade,J.Appl.Polym.Sci.,2002,83,2000–2007.

51G.S.Chauhan,S.Dhiman,R.K.Sharma and L.K.Guleria, J.Polym.Mater.,2001,18,267–278.

52P.Ghosh and D.Das,Eur.Polym.J.,2000,36,2505–2511.

53A.L.Suo,J.M.Qian,Y.Yao and W.G.Zhang,J.Appl.Polym.

Sci.,2007,103,1382–1388.

54D.Aliouche,B.Sid and H.Ait-Amar,Ann.Chim.Sci.Mater., 2006,31,527–540.

55M.W.Sabaa and S.M.Mokhtar,Polym.Test.,2002,21, 337–343.

56B.A.El-Hady and M.M.Ibrahim,J.Appl.Polym.Sci.,2004,93, 271–278.

57M.K.Zahran and R.I.Mahmoud,J.Appl.Polym.Sci.,2003,87, 1879–18.

58Y.Ogiwara and H.Kubota,J.Appl.Polym.Sci.,1969,13,1613. 59Y.Ogiwara and H.Kubota,J.Polym.Sci.A1,1970,8,1069–76. 60B.N.Misra,R.Dogra,I.Kaur and J.K.Jassal,J.Polym.Sci.

Pol.Chem.,1979,17,1861–1863.

61Y.Ogiwara and H.Kubota,J.Appl.Polym.Sci.,1968,12, 2575–84.

62G.Mino and S.Kaizerman,J.Polym.Sci.,1958,31,242–243. 63K.C.Gupta and K.Khandekar,Polym.Int.,2006,55,139–150. M.K.Zahran,J.Polym.Res.,2006,13,65–71.

65S.Kaizerman,G.Mino and L.F.Meinhold,Text.Res.J.,1962, 32,136–140.

66A.Kantouch,A.Hebeish and M.H.El-Rafie,J.Appl.Polym.

Sci.,1971,15,1007–1019.

67G.Gurdag,G.Guclu and S.Ozgumus,J.Appl.Polym.Sci.,2001, 80,2267–2272.

68T.Doba,C.Rodehed and B.Ranby,Macromolecules,1984,17, 2512–2519.

69K.C.Gupta,S.Sahoo and K.Khandekar,Biomacromolecules, 2002,3,1087–1094.

70K.C.Gupta and K.Khandekar,J.Appl.Polym.Sci.,2002,86, 2631–22.

71K.C.Gupta and K.Khandekar,J.Macromol.Sci.,Pure Appl.

Chem.,2003,A40,155–179.

72K.C.Gupta and K.Khandekar,J.Appl.Polym.Sci.,2006,101, 2546–2558.

73K.C.Gupta and S.Sahoo,J.Macromol.Sci.,Pure Appl.Chem., 2000,A37,447–468.

74K.C.Gupta and S.Sahoo,J.Appl.Polym.Sci.,2000,79,767–778. 75K.C.Gupta and S.Sahoo,Biomacromolecules,2001,2,239–247. 76H.Chen and Y.L.Hsieh,Biotechnol.Bioeng.,2005,90,405–413. 77T.Kondo,H.Kubota and R.Katakai,J.Appl.Polym.Sci.,1999, 71,251–258.

78M.A.Shah,M.A.Majid and M.A.Bari,Chittagong University Studies,Part II:Science,1991,15,31–7.

79B.Ranby and D.Zuchowska,Polym.J.(Tokyo,Japan),1987, 19,623–30.

80I.C.Eromosele,J.Appl.Polym.Sci.,1994,51,1817–21.

81A.Higazy,Bulletin of the National Research Centre(Egypt), 1996,21,385–398.

82L.M.Zhang,J.Macromol.Sci.,Pure Appl.Chem.,1999,36, 1141–1152.

83K. C.Gupta and S.Sahoo,J.Appl.Polym.Sci.,2000,76, 906–912.

84K.C.Gupta and S.Sahoo,Cellulose,2001,8,233–242.

85S.Lenka,P.L.Nayak and A.K.Tripathy,J.Appl.Polym.Sci., 1982,27,2725–2728.

86S.Lenka,P.L.Nayak and M.K.Mishra,J.Appl.Polym.Sci., 1980,25,1323–33.

87J.T.Guthrie and P.D.Tune,J.Polym.Sci.,Part A1,1991,29, 1301–12.

88R.Narayan and G.T.Tsao,J.Polym.Sci.Pol.Lett.,1983,21, 437–9.

A.Takahashi and S.Takahashi,Sen’I Gakkaishi,1974,30, T35–T40.

90Y.H.Liu,L.Y.Yang,Z.Q.Shi and J.B.Li,Polym.Int.,2004, 53,1561–1566.

91H.Kubota and S.Kuwabara,J.Appl.Polym.Sci.,1997,, 2259–2263.

92W.H.Daly,T.S.Evenson,S.T.Iacono and R.W.Jones, Macromol.Symp.,2001,174,155–163.

93J.O.Karlsson and P.Gatenholm,Polymer,1997,38,4727–4731. 94G.N.Richards,J.Appl.Polym.Sci.,1961,553.

95H.S.Koenig and C.W.Roberts,J.Appl.Polym.Sci.,1974,18, 651–666.

96S.R.Shukla and A.R.Athalye,J.Appl.Polym.Sci.,1992,44, 435–442.

97F.Khan,Polym.Int.,2004,53,1581–1584.

98F.Khan,Biomacromolecules,2004,5,1078–1088.

99A.S.Bashar,M.A.Khan and K.M.I.Ali,J.Appl.Polym.Sci., 1997,63,1703–1711.

100M.K.Uddin,M.A.Khan and K.M.I.Ali,Radiat.Phys.Chem., 1996,48,511–517.

101J.Q.Liu,M.L.Zhai and H.F.Ha,Radiat.Phys.Chem.,1999, 55,55–59.

102J.Lu,M.Yi,J.Li and H.Ha,J.Appl.Polym.Sci.,2001,81, 3578–3581.

103V.Kumar,Y.K.Bhardwaj,S.N.Jamdar,N.K.Goel and S.Sabharwal,J.Appl.Polym.Sci.,2006,102,5512–5521.

104J.Lu,J.Li,M.Yi and H.F.Ha,Radiat.Phys.Chem.,2001,60, 625–628.

105K.D.N.Lawrence and D.Verdin,J.Appl.Polym.Sci.,1973,17, 2653–2666.

106S.M.Badawy,A.M.Dessouki and H.M.N.El-Din,Radiat.

Phys.Chem.,2001,61,143–148.

107V.Kumar,Y.K.Bhardwaj,K.P.Rawat and S.Sabharwal, Radiat.Phys.Chem.,2005,73,175–182.

108J.D.Wellons and V.Stannett,J.Polym.Sci.Part A,1965,3,847. 109I.Sakurada,in CHEMTECH,1972,p.376–381.

110J.L.Garnett,L.-T.Ng and V.Viengkhou,Radiat.Phys.Chem., 1999,56,387–403.

111S.R.Shukla,G.V.G.Rao and A.R.Athalye,J.Appl.Polym.

Sci.,1993,49,1423–30.

112S.R.Shukla and A.R.Athalye,Polymer,1992,33,3729–33. 113S.Margutti,S.Vicini,N.Proietti, D.Capitani,G.Conio,

E.Pedemonte and A.L.Segre,Polymer,2002,43,6183–6194. 114H.Kubota and N.Shiobara,React.Funct.Polym.,1998,37, 219–224.

115G.John,C.K.S.Pillai and A.Ajayaghosh,Polym.Bull.(Berlin Germany),1993,30,415–20.

116L.C.V.Wielen and A.J.Ragauskas,Eur.Polym.J.,2004,40, 477–482.

117Zubaidi and T.Hirotsu,J.Appl.Polym.Sci.,1996,61,1579–1584. 118A.Carlmark and E.E.Malmstrom,Biomacromolecules,2003,4, 1740–1745.

119P.Mansson and L.Westfelt,J.Polym.Sci.Pol.Chem.,1981,19, 1509–1515.

120C.J.Biermann,J.B.Chung and R.Narayan,Macromolecules, 1987,20,954–957.

121Y.Avny,B.Yomtov and A.Zilkha,J.Appl.Polym.Sci.,1965,9, 3737.

122Y.Avny and L.Rebenfeld,Text.Res.J.,1968,38,684.

123E.Cohen,Y.Avny and A.Zilkha,J.Polym.Sci.A1,1971,9,1469. 124E.Cohen,Y.Avny and A.Zilkha,J.Polym.Sci.A1,1971,9,1481. 125R.Narayan and M.Shay,Polym.Sci.Tech.(Plenum),1986,33, 137–146.

126I.Ikeda,Y.Kurushima,H.Takashima and K.Suzuki,Polym.J.

(Tokyo,Japan),1988,20,243–250.

127G.Rausing and S.Sunner,Tappi,1962,203A.

128H.Cheradame,A.U.Tadjang and A.Gandini,Makromolekul.

Chem-Rapid,1988,9,255–256.

129G.John and C.K.S.Pillai,Polym.Bul.(Berlin,Germany),19, 22,–94.

130B.A.Feit,A.Bar-Nun,M.Lahav and A.Zilkha,J.Appl.Polym.

Sci.,19,8,1869–1888.

131R.F.Schwenker and E.Pacsu,Tappi,1963,665.

132O.Y.Mansour and A.H.Basta,Nor.Pulp.Pap.Res.J.,1991,6, 184–90.

133R.Narayan and M.Shay,Polymer Preprints(American Chemical Society,Division of Polymer Chemistry),1986,27,204–5.

134J.Hafren and A.Cordova,Macromo.Rapid Comm.,2005,26, 82–86.

135H.Loennberg,Q.Zhou,H.Brumer,III,T.Teeri,E.Malmstroem and A.Hult,Biomacromolecules,2006,7,2178–2185.

136N.Tsubokawa,T.Iida and T.Takayama,J.Appl.Polym.Sci., 2000,75,515–522.

137M.Szwarc,Nature,1956,178,1168–1169.

138T.Otsu and M.Yoshida,Makromolekul.Chem-Rapid,1982,3, 127–132.

139T.Otsu,M.Yoshida and T.Tazaki,Makromolekul.Chem-Rapid, 1982,3,133–140.

140S.Perrier and P.Takolpuckdee,J.Polym.Sci.Pol.Chem.,2005, 43,5347–5393.

141K.Matyjaszewski,in Handbook of Radical Polymerization, ed.K.Matyjaszewski and T.P.Davis,John Wiley and Sons, Hoboken,2002,pp.361–406.

142D.H.Solomon,E.Rizzardo and P.Cacioli,US patent4581429, 1986.

143C.J.Hawker,A.W.Bosman and E.Harth,Chem.Rev.,2001, 101,3661–3688.

144C.J.Hawker,in Handbook of Radical Polymerization, ed.K.Matyjaszewski and T.P.Davis,John Wiley and Sons, Hoboken,2002,pp.463–521.

145D.H.Solomon,J.Polym.Sci.Pol.Chem.,2005,43,5748–57. 146M.Kato,M.Kamigaito,M.Sawamoto and T.Higashimura, Macromolecules,1995,28,1721–1723.

147J.S.Wang and K.Matyjaszewski,J.Am.Chem.Soc.,1995,117, 5614–5615.

148K.Matyjaszewski and J.H.Xia,Chem.Rev.,2001,101, 2921–2990.

149J.S.Wang and K.Matyjaszewski,Macromolecules,1995,28, 7901–7910.150V.Percec and B.Barboiu,Macromolecules,1995,28,7970–7972. 151E.Rizzardo,J.Chiefari,R.T.A.Mayadunne,G.Moad and S.H.Thang,in ACS Symp.Ser.,Controlled/Living Radical Polymerization:Progress in ATRP,NMP and RAFT, ed.K.Matyjaszewski,American Chemical Society,Washington, DC,2000,vol.768,pp.278–296.

152A.Favier and M.-T.Charreyre,Macromol.Rapid Commun., 2006,27,653–692.

153G.Moad,E.Rizzardo and S.H.Thang,Aust.J.Chem.,2006,59, 669–692.

154J.Chiefari,Y.K.Chong, F.Ercole,J.Krstina,J.Jeffery, T.P.T.Le,R.T.A.Mayadunne,G.F.Meijs,C.L.Moad,

G.Moad,E.Rizzardo and S.H.Thang,Macromolecules,1998,

31,5559–5562.

155K.Matyjaszewski,in ACS Symp.Ser.,Controlled/Living Radical Polymerization:Progress in ATRP,NMP and RAFT, ed.K.Matyjaszewski,American Chemical Society,Washington, DC,2000,vol.768,pp.2–25.

156F.Chauvin,J.-L.Couturier,P.-E.Dufils,P.Gerard,D.Gigmes, O.Guerret,Y.Guillaneuf,S.R.A.Marque,D.Bertin and P.Tordo,in ACS Symp.Ser.,Controlled/Living Radical Polymerisation-From Synthesis to Materials,ed.K.Matyjaszewski, American Chemical Society,Washington,DC,2006,vol.944, pp.326–341.

157K.Matyjaszewski,in ACS Symp.Ser.Controlled Radical Polymerization,ed.K.Matyjaszewski,American Chemical Society,Washington DC,1998,vol.685,pp.2–30.

158S.Ding,M.Radosz and Y.Shen,in ACS Symp.Ser.Controlled/ Living Radical Polymerisation-From Synthesis to Materials,ed.

K.Matyjaszewski,American Chemical Society,Washington,DC, 2006,vol.944,pp.71–84.

159D.M.Haddleton and C.Waterson,Macromolecules,1999,32, 8732–8739.

160A.Carlmark and E.Malmstrom,J.Am.Chem.Soc.,2002,124, 900–901.

161D.Nystrom,J.Lindqvist, E.Ostmark, A.Hult and

E.Malmstrom,Chem.Commun.,2006,3594–3596.

162M.Coskun and M.M.Temuz,Polym.Int.,2005,54,342–347. 163I.Ikeda,T.Higuchi and Y.Maeda,Sen-I Gakkaishi,2002,58, 308–313.

1P.Vlcek,M.Janata,P.Latalova,J.Kriz, E.Cadova and L.Toman,Polymer,2006,47,2587–2595.

165D.W.Shen and H.Yong,Polymer,2004,45,7091–7097.

166D.Shen,H.Yu and Y.Huang,Cellulose,2006,13,235–244. 167D.W.Shen,H.Yu and Y.Huang,J.Polym.Sci.Pol.Chem., 2005,43,4099–4108.

168R.Yang,Y.Wang and D.Zhou,Electrophoresis,2007,28, 3223–3231.

169E.Ostmark,S.Harrisson,K.Wooley and E.Malmstrom, Biomacromolecules,2007,8,1138–1148.

170D.Plackett,K.Jankova,H.Egsgaard and S.Hvilsted, Biomacromolecules,2005,6,2474–2484.

171J.Lindqvist and E.Malmstrom,J.Appl.Polym.Sci.,2006,100, 4155–4162.

172S.P.Rannard,S.H.Rogers and R.Hunter,Chem.Commun., 2007,362–3.

173L.Yan and K.Ishihara,J.Polym.Sci.Pol.Chem.,2008,46, 3306–3313.

174J.Chiefari and E.Rizzardo,in Handbook of Radical Polymeriza-tion,ed.K.Matyjaszewski and T.P.Davis,John Wiley and Sons, Hoboken,2002,pp.629–690.

175S.Perrier,P.Takolpuckdee,S.Brown,T.M.Legge,D.Roy, M.R.Wood,S.P.Rannard and D.J.Duncalf,in ACS Symp.

Ser.,Controlled/Living Radical Polymerisation-From Synthesis to Materials,ed.K.Matyjaszewski,American Chemical Society, Washington,DC,2006,vol.944,pp.438–454.

176S.Perrier,P.Takolpuckdee and C.A.Mars,Macromolecules, 2005,38,2033–2036.

177V.Lima,X.Jiang,J.Brokken-Zijp,P.J.Schoenmakers,

B.Klumperman and R.Van Der Linde,J.Polym.Sci.Pol.

Chem.,2005,43,959–973.

178A.Postma,T.P.Davis,G.Moad and M.S.O’Shea, Macromolecules,2005,38,5371–5374.

179M.H.Stenzel,T.P.Davis and A.G.Fane,J.Mater.Chem., 2003,13,2090–2097.

This journal is c The Royal Society of Chemistry2009Chem.Soc.Rev.,2009,38,2046–20|2063180M.Hernandez-Guerrero,T.P.Davis,C.Barner-Kowollik and M.H.Stenzel,Eur.Polym.J.,2005,41,22–2277.

181S.Perrier,P.Takolpuckdee,J.Westwood and D.M.Lewis, Macromolecules,2004,37,2709–2717.

182D.Roy,Aust.J.Chem.,2006,59,229–229.

183D.Roy,J.T.Guthrie and S.Perrier,Macromolecules,2005,38, 10363–10372.

184D.Roy,J.T.Guthrie and S.Perrier,Polymer Preprints (American Chemical Society,Division of Polymer Chemistry), 2005,46,324–325.185D.Roy,P.Takolpuckdee,Y.Zhao and S.Perrier,PMSE Preprints,2006,94,228–229.

186D.Roy,J.T.Guthrie and S.Perrier,Aust.J.Chem.,2006,59, 737–741.

187D.Roy,J.T.Guthrie and S.Perrier,Soft Matter,2008,4, 145–155.

188D.Roy,J.S.Knapp,J.T.Guthrie and S.Perrier,Biomacro-molecules,2008,9(1),91–99.

1M.Barsbay,O.Guven,M.H.Stenzel,T.P.Davis,C.Barner-Kowollik and L.Barner,Macromolecules,2007,40,7140–7147.

20|Chem.Soc.Rev.,2009,38,2046–20This journal is c The Royal Society of Chemistry2009