Magnetic Responsive Polymer Composite Materials

This journal is c The Royal Society of Chemistry 2013 Chem. Soc. Rev.

Cite this: DOI: 10.1039/c3cs60058k

Magnetic responsive polymer composite materials

Julie Thevenot,ab Hugo Oliveira,ab Olivier Sandre*ab andSebastien Lecommandoux*ab

Magnetic responsive materials are the topic of intense research due to their potential breakthrough

applications in the biomedical, coatings, microfluidics and microelectronics fields. By merging magnetic

and polymer materials one can obtain composites with exceptional magnetic responsive features.

Magnetic actuation provides unique capabilities as it can be spatially and temporally controlled, and

can additionally be operated externally to the system, providing a non-invasive approach to remote

control. We identified three classes of magnetic responsive composite materials, according to their

activation mode and intended applications, which can be defined by the following aspects. (A) Their

ability to be deformed (stretching, bending, rotation) upon exposure to a magnetic field. (B) The

possibility of remotely dragging them to a targeted area, called magnetic guidance, which is particularly

interesting for biomedical applications, including cell and biomolecule guidance and separation. (C) The

opportunity to use magnetic induction for thermoresponsive polymer materials actuation, which has

shown promising results for controlled drug release and shape memory devices. For each category,

essential design parameters that allow fine-tuning of the properties of these magnetic responsive

composites are presented using key examples.

Key learning pointsIn this tutorial review, a critical overview is proposed to the reader, providing fundamental knowledge about:(1) How to obtain ecient magnetic-responsive polymer-based materials and the critical issues met while designing them(2) The main application fields for magnetic responsive composite polymer materials(3) The type of response that can be elicited in a polymer material through a magnetic stimulus(4) The physical mechanisms involved in the transduction of the magnetic trigger into the desired response(5) The usual characteristics (homogeneous/gradient, permanent/alternating, amplitude) of the magnetic fields employed for each type of application

1. Introduction

Stimuli-responsive materials have drawn a lot of interest over thelast few decades due to their biomimetic behaviour and theirpotential use in smart or intelligent devices. Indeed, in any livingsystem, adaptation and the ability to respond to environmentalchanges from the molecular to the macromolecular level arefascinating and crucial for maintaining and regulating normalfunctions. Stimuli-responsive or smart materials are therefore oftendefined as materials that can respond to pH, light, tempera-ture, ionic strength, electric or magnetic field variations themost commonly used triggers by changing their own properties:changes in size, surface area, solubility, permeability, shape,mechanical, and optical properties, among others.

In this research area, polymer-based materials are the mostdeveloped and studied, owing to their high versatility andability to dramatically alter their intrinsic properties as afunction of small environmental changes.1 Smart polymersare increasingly playing an important role in a wide range ofapplications, especially in the biomedical field (drug delivery,tissue engineering, biosensors, active diagnosis), coatings(smart textiles and fibres) and microelectronics (actuators,electromechanics). Indeed, the huge development of synthesisand successes in polymer chemistry allow a proper design ofwell-defined macromolecules that can incorporate stimuli-responsive building blocks for all the above-mentioned triggers.However, intrinsically magnetic responsive polymers are scarceand generally present poor ecacy. Liquid-crystalline polymersand elastomers have been designed for application as artificialmuscles, based on the seminal ideas proposed by P.-G. de Genneset al.2 Combined eects of magnetic orientation, and temperatureinduced phase transition (from nematic to isotropic phase) are

a Universite de Bordeaux, LCPO, UMR 5629, F-33600 Pessac, France.

E-mail: [email protected], [email protected]; Fax: +33 5 40 00 84 87b CNRS, LCPO, UMR 5629, F-33600 Pessac, France

Received 10th February 2013

DOI: 10.1039/c3cs60058k

www.rsc.org/csr

Chem Soc Rev

TUTORIAL REVIEW

Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article OnlineView Journal

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2013

at the origin of volume changes and deformations. In spite oftheir interesting mechanical and thermal properties, liquidcrystal polymers require intense magnetic fields for theiralignment (H B 103 kA m1) and present low switching rates.Indeed, their response time is limited by their high viscosity,especially in the bulk, and high temperatures are often requiredto reach the appropriate phase transition.3 Contrary to systemslimited by diusion processes (of either heat or mass), thosedirectly operated by exposure to a field (be it magnetic orelectric) might have shorter response times, possibly as fastas natural skeletal muscles.4

To produce highly ecient magnetic-responsive materials,the doping of polymer materials with magnetic nanoparticles(MNPs), made of inorganic matter (most often superpara-magnetic iron oxide Fe3O4 or g-Fe2O3, or soft metallic iron,but also hard magnetic materials e.g. Co, Ni, FeN, FePt,FePd. . .), appeared to be the more appealing and ecientsolution. Indeed, the magnetic moment of these smallmagnets, much larger than those of molecular magnets, allowsthem to respond to weak stimuli (static or alternating magneticfield) with a significant eect (e.g. movement, heat generation,magnetic or optical signal). The resulting composites that can

Julie Thevenot

Julie Thevenot is a post-doctoral fellow in SebastienLecommandouxs group inLaboratoire de Chimie desPolyme`res Organiques (LCPO).She obtained her PhD fromUniversity of Lyon, France, in2007, working on polymer/lipidnanoconstructs for nucleic acidand protein delivery. Herresearch interests include polymerchemistry, polymer-based drugdelivery systems, physicalchemistry of colloidal suspen-

sions and self-assembly. At LCPO, she contributed to theEuropean project Nanother that focused on the development ofnovel hybrid (polymer/magnetic inorganic material) nanoparticlesystems for cancer therapy and diagnosis.

Hugo Oliveira

Hugo Oliveira is a post-doctoral fellow in SebastienLecommandouxs group, Labo-ratoire de Chimie des Polyme`resOrganiques (LCPO-CNRS). Herehe participated in the Europeanproject Nanother that focused onthe development of novelnanoparticle systems that permitcontrolled drug release andimaging (MRI), while targetingspecific cancer tissues. Heobtained his PhD in BiomedicalEngineering in 2010 from the

University of Porto and the Instituto de Engenharia Biomedica(INEB), Portugal. In 2012 he received the Young Scientist award atthe 9th World Biomaterials Congress in Chengdu, China.

Olivier Sandre

Olivier Sandre has been a tenuredCNRS researcher since 2001. Hejoined the Laboratory of OrganicPolymers Chemistry in 2010 aftercollaborating with LCPO in 2004.He works on polymeric systemsdoped with magnetic nano-particles, especially magneticpolymersomes for theranostics(MRI combined with anti-cancertherapy). He was nominatedadjunct professor of WaterlooUniversity in Canada andreceived the 2012 Young

Researcher award of the Physical Chemistry division of FrenchChemical Society (SCF) and French Physics Society (SFP) for hisresearch on self-assembled magnetic polymer composite materials.

Sebastien Lecommandoux

Sebastien Lecommandoux is FullProfessor at the University ofBordeaux (ENSCBP-IPB) and isleading the group PolymerNanotechnology and Life Sciencesat the Laboratoire de Chimie desPolyme`res Organiques (LCPO-CNRS). His current researchinterests include polypeptide andpolysaccharide based block co-polymers self-assembly, biomimeticapproaches toward design ofsynthetic viruses and cells aswell as drug-delivery. He is the

deputy director of the LCPO and the director of the research at theIPB-ENSCBP. He is also currently chairing the ESF ResearchNetwork Programme on Precision Polymer materials P2M.Sebastien Lecommandoux is the recipient of the CNRS bronzemedal award (2004) and is an honorarium junior member of theInstitut Universitaire de France (promotion IUF 2007). He isAssociate Editor for Biomacromolecules (ACS) and in theEditorial Advisory Board of several international journals,including Polymer Chemistry and Biomaterials Science (RSC).

Tutorial Review Chem Soc Rev

Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


be named magnetic responsive polymer composites (MRPCs)are the topic of this review. The focus is thus dierent from the2009 review by C. S. Brazel5 on magnetothermally-responsivenanomaterials dedicated mostly to thermo-sensitive polymersassociated to MNPs, a class of MRPCs that will also be dis-cussed here. Our approach is also dierent from Dai andNelsons 2010 tutorial review on magnetic-responsive polymercomposites,6 and from the review by Medeiros et al. on stimuli-responsive magnetic particles for biomedical applications, inwhich the sensitivity to a magnetic field was not always coupledto the other stimuli (pH, light, temperature, electric field, ionicstrength, etc.).7 In addition to the mention of the latest workson the subject, we have opted for a totally dierent approachindeed, based on applications rather than preparation methods.Under our focus, the activation mode of MRPCs always consistsin the application of a magnetic field, either static (H dc) oralternating (H ac). In our view of MRPCs, magnetism mustdetermine their core responses rather than being an additionalfunctionality.

Magnetic nanoparticles, when combined with a polymermaterial having its own properties, can produce a large varietyof MRPCs. Hence, the main core of this review will be organizedinto three parts (Fig. 1), depending on the merged propertiesresulting from the composites. When magnetic nanoparticlesare incorporated into elastomeric polymer scaolds, controlleddeformations can be obtained such as stretching or contractionof a cylinder, bending of an elongated sample, deflection of amembrane, chaining of microparticles, rotation of anisotropicobjects, or rupture of a capsule (Section 2, Fig. 1AC). Compo-sites made from biocompatible polymer matrices have beenrecently studied for magnetic guidance and separation, such asmagnetic cell targeting and manipulation, and for sorting andseparation applications: to this aim, a magnetic field gradientis applied at a selected location in a flow circulation such as inthe blood circulation or a fluidic device (Section 3, Fig. 1DF).Combined with thermosensitive polymers, the activationof magnetic nanoparticles, when exposed to an alternatingmagnetic field, can create localized heating that was successfully

Fig. 1 Schemes depicting the dierent types of magnetic responsive materials obtained from the doping of various polymers with magnetic particles and illustrationof their response when exposed to a static magnetic field (H dc) or to an alternating magnetic field (H ac). From left to right: composites made from elastomericpolymers can be deformed in homogeneous fields or gradients in a controlled fashion; MRPC particles made of polymers designed for biomedical applications can beused for magnetic guidance for drug delivery or separation purpose; MRPC from thermoresponsive polymers can be activated by magnetic induction using alternatingfields.

Chem Soc Rev Tutorial Review

Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


used for both controlled drug release and shape memoryactuators (Section 4, Fig. 1G and H).

2. Deformation of soft materials in magneticfields

This section deals with magnetic composite polymer materialsthat can be referred to as soft matter for their mechanicalresponse (low elastic modulus) and soft magnet for theirmagnetic behaviour (with no remanent magnetization in theabsence of an applied field). The first reported materials of thisclass were obtained by association of a ferrofluid and a hydro-gel, and were thus named ferrogels.

2.1. Ferrogels: macroscopic deformation in field gradients

In their pioneering work, M. Zrinyi, L. Barsi and A. Buki loadedan aqueous dispersion of magnetite (Fe3O4) nanoparticles with a diameter of around 10 nm within a glutaraldehydecross-linked poly(vinyl alcohol) (PVA) hydrogel. Shaped likecylinders, these ferrogels exhibited stretching, contraction orbending deformations when exposed to an inhomogeneousstatic magnetic field (Fig. 2).8

Since then, many other hydrophilic polymers were proposedto host magnetic nanoparticles, in particular thermosensitivegels like poly(N-isopropylacrylamide) (PNIPAM) that exhibit alower critical solution temperature (LCST) at 32 1C. However,PNIPAM and other polymers exhibiting a LCST are generallyless polar than PVA and can be inecient for trapping theMNPs due to the absence of hydrogen bonds and to a mesh size

of typically 1020 nm, i.e. larger than the size of superpara-magnetic iron oxide MNPs (510 nm).9 To address this issue,several strategies were proposed as schematically representedin Fig. 3, namely: a statistical copolymer network with achelating comonomer such as 2-acetoacetoxyethylmethacrylate(AEMA),10 a semi-interpenetrated network with alginate chainswrapping the MNPs11 or a composite network of PNIPAM andpoly(ethylene glycol) (PEG) using PEG400dimethacrylate ascrosslinker of NIPAM.12

With larger sized MNPs, i.e. of diameters around 2535 nmthat stand above the mesh size of the gel, a pure PNIPAM networkcan be used to design magnetic thermogels.13 In that case S. Goshand T. Cai nicely evidenced an actuationmechanism based on themagnetically-induced thermal deswelling mechanism. Like abimetal coil, a cylinder of this composite, loaded at 10 wt% ironoxide relatively to PNIPAM, bends at an angle that is related to themagnetic field intensity and frequency (Fig. 4).

2.2. Magneto-active elastomers: graduated deformations

Magnetic polymer actuators were also developed based on adirect eect of the magnetic field through the magnetic dipolarforces acting between the magnetic fillers (at the origin of thedemagnetizing and magnetostriction eects). In this case,microparticles (15 mm) of carbonyl iron (CI, i.e. pristine Fe,magnetically soft, made by thermal decomposition of Fe(CO)5)are dispersed in silicone-based elastomers referred as magneto-rheological (MR) elastomers,14 magneto-active (MA) elastomers,15

or magnet fillerpolymer matrix composites (Magpol).4

Fig. 2 Elongation, contraction or bending of a ferrogel in non-uniform mag-netic field created by the polar pieces guiding the field lines of an electromagnet:(a) no magnetic field applied; (b) the maximal field strength is located on the topsurface of the ferrogel; (c) the maximal field strength is focused in the middle ofthe gel along its axis; (d) the gel is located 15 cm from the magnet; (e) bendingon a toroidal permanent magnet (adapted from ref. 8 with permission. Copyright1996, American Institute of Physics. Pictures from the original publicationavailable on the authors website).

Fig. 3 Three strategies proposed to trap magnetic nanoparticles, around 10 nmin size (brown dots), inside a hydrogel network made of a LCST-polymer (e.g.PNIPAM, purple lines) which does not present interactions with the MNPs.Another type of chains (e.g. alginate or poly(acrylamide), green lines) stronglyadsorbing onto the surface of the iron oxide is introduced either as a randomcopolymer (a),10 a semi-interpenetrated network (b),11 or a fully interpenetratednetwork (c).12


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


Such materials elicit a low Youngs modulus (E B 106 Pa)together with a high magnetisation (M > 104 A m1) and thusexhibit large deformation, not only in field gradients but inhomogeneous magnetic fields as well. The significant variationof their tensile, compression, and shear moduli with theamplitude of the applied magnetic field (a property not foundfor ferrogels), makes them good candidates to design activedampers for anti-vibrating applications (e.g. in the automotiveindustry or to enhance the precision of rotating tools inmechanics workshops). Two types of MR/MA elastomers canbe prepared: either isotropic (unstructured) or anisotropic(structured), the latter samples being obtained by applying astatic magnetic field during the cross-linking reaction of thematrix. For magnetic silicones with a Youngs modulus E =200 kPa, the maximal magneto-elastic properties were foundfor the pre-aligned samples, in accordance with a theoreticalmodel based on the breaking and reorganisation of the dipolarchains at increasing strain (Fig. 5ab).14

Experiments with a less rigid CI formulation in silicone at aneven lower Youngs modulus, E = 16 kPa, reported that chainscould still form in the soft matrix after its preparation (Fig. 5d).In that case, pre-alignment during cross-linking was not aprerequisite to obtain large magneto-elastic eects. In particu-lar, a magnetic shape memory eect was demonstrated: suchMA elastomers keep their deformation (by stretching, compres-sion or bending) as long as a magnetic field is present, andrelax elastically to their initial shape only when the field isswitched o (Fig. 5ce). The interpretation of this magneticshape memory eect was also based on the variation of inter-particle distances and reorganisation of dipolar chains underfield.15 Other geometries of MA elastomers made of CI dis-persed in silicone include magnetoelastic membranes (Fig. 6)that can be used as pumps in microfluidic mixing devicespropelling two fluids back-and-forth (at 10 Hz) in order toaccelerate their mixing.16

MR/MA elastomers can also be obtained from superparamag-netic iron oxide MNPs dispersed in poly(dimethylsiloxane) (PDMS)thanks to prior functionalization of the particle surface with PDMSchains.17,18 The concept of magnetoelastic ratio (MER) wasintroduced by B. Evans et al. to quantify the propensity of a rod-like sample to bend in a uniform magnetic field independently ofthe rod length and diameter:17 it is defined as m0M

2r2/E where m0 isthe magnetic permeability of vacuum, M the magnetisation, r themass density, and E the Youngs modulus of the composite(the deflection angle is then given by the product of the MER bythe square of the rod aspect ratio). Fig. 7 represents the behaviourof such magnetoelastic samples at various magnetic loadsbased on ferrogels (E B 103104 Pa) or on silicone rubbers (E B106107 Pa). The MER enables rationalising the antagonist eectsof an increase of the magnetic load in the rubber which increasesmagnetisation but rigidifies the elastomer.

A last class of magnetic elastomers reported in the literaturetargets the tuning of optical properties with a magnetic field.Among these systems, poly(urethane) (PU) films doped at a lowmagnetic load, i.e. not more than 0.5 wt%, oer an acceptabletransparency in the visible spectrum combined with a magneto-elastic response.21

2.3. Anisotropic polymer particles for orientation control

When preparing macroscopic magnetoelastic samples, variousshapes can be obtained by the use of an appropriate mould.

Fig. 4 Magnetic PNIPAM ferrogel cylinder lying on a grooved substrate under-going bending under a RF magnetic field due to the strain dierence between thecontact area with the solid under tension and the free surface under compressiondue to gel de-swelling when temperature reaches the LCSTof PNIPAM bymagneticinduction (adapted from ref. 13 with permission from the IOP Publishing).

Fig. 5 Behaviours of MR/MA elastomers in homogeneous magnetic fields;(ac) MR silicone loaded at 25 vol% with soft iron pre-aligned under a field H =123 kA m1 during the curing step; (a) plot illustrating the influence of increaseof magnetic field intensity H on the traction curve expressed as the dierence ofstress Ds for a given strain e relatively to the curve without applied field H = 0; (b)optical micrograph showing chains of magnetic fillers;14 (ce) unstructured MAelastomer at 3035 vol% of magnetic phase in a uniform magnetic field; (c)stressstrain dependence at (1) H = 0, (2) H = 96 kA m1, and (3) H = 208 kA m1

showing a residual strain under magnetic field; (d) under an increasing appliedmagnetic field H, the embedded CI fillers form chains, before returning to theirinitial positions when the field is switched o; (e) picture of the initial shape (1)and of the deformed shapes retained after stretching (2), contraction (3) orbending (4) as long as the magnetic field is maintained; relaxation occurs whenthe field is switched o (adapted from ref. 14 and 15 with permission from theWorld Scientific Publishing Co. and IOP Publishing, respectively).


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


Practical issues can however be encountered when preparingparts of smaller dimensions (typically less than 10 mm) due to

oxygen poisoning of the cross-linking/curing reaction (be it aradical polymerization for acrylate-based hydrogels or a hydro-silylation for silicones). Thus an elegant method to prepareferrogel microparticles of various shapes (not only spherical)was introduced by P. S. Doyle et al.22,23 It consists in photo-polymerizing droplets containing a mixture of aqueous MNPsand poly(ethylene glycol diacrylate) (PEGDA) suspended inmineral oil inside a microfluidic channel illuminated withfocused UV light. The authors obtained not only magneticferrogel microspheres, but also microscopic disks and plugs,by squeezing the droplets in a constricting channel (with areduced vertical dimension),22 and even Janus particles (half-magnetic), using a particular flow-focusing chip.23 Unlikemacroscopic ferrogels which can deform (elongate, contractor deflect) when held by one end in a magnetic field gradient,24

the response of an assembly of microscopic ferrogels to anapplied magnetic field is at first chaining. Then, when micro-particles have a non-isotropic symmetry, their own orientationwithin a chain is determined by the demagnetising eect, i.e.the particles orient their largest dimension along the fielddirection, as illustrated for tile-like plugs and half-sphericalJanus micro-ferrogels in Fig. 8.

The microfluidic production of anisotropic magnetic micro-gels also enabled the study of their solid rotation when sub-jected to a rotating magnetic field, as shown in Fig. 9AH forthe Janus microgels of PEGDA. An analogous microfluidicapproach was developed by D. Weitz et al. using double-emulsion (MNPs in a monomer wrapped by another polymerphase and dispersed in a non-miscible oil) instead of a simpleone. This enabled the preparation of magnetic polystyrene (PS)cores encapsulated in a hydrophilic poly(acrylamide) (PAM)

Fig. 6 (a) Deflection of dierent membranes (B1.0 mm in diameter, 100 mm inthickness) under the same magnetic field (H = 80 kA m1, B = 100 mT) atincreasing weight ratios CI/PDMS from 0.5 to 4; (b) deflection of the samemembrane (weight ratio = 2) for increasing fields up to H = 112 kA m1. Suchmembranes can be use to design pumps or valves in a microchip (reprinted fromref. 16 with kind permission from Springer Science and Business Media).

Fig. 7 Magnetoelastic ratios (MER) from the literature. Solid curves represent aconstant magnetoelastic ratio as a function of magnetite load and Youngsmodulus, e.g. B102 for poly(acrylamide) (PAM) ferrogels19 and B104 formagnetic PDMS.17 The deflection angle y of a flexible magnetic rod is theproduct of MER by the square of the aspect ratio (rod length divided by itsdiameter) (adapted from ref. 17 and 20 with permission from Elsevier and JohnWiley and Sons, copyright 2012).

Fig. 8 Non-spherical magnetic micro-hydrogels prepared by photo-polymeriza-tion in microfluidics: plugs22 (a and b) and Janus particles23 (c and d) oriented byan in-plane (a and d) or out-of-plane magnetic field (b and c). The iron oxideloads are respectively 4.9 and 3.6 wt% relative to PEGDA. In both cases, theorientation is determined by the minimisation of the demagnetising field (largestdimension of the magnetic body along the applied magnetic field). Scale bars are25 (a and b) and 100 mm (c and d) (reprinted from ref. 22 and 23 with permissionfrom RSC Publishing and American Chemical Society, respectively, copyright2013).


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


shell (Fig. 9ae).25 The whole particle undergoes an originaleccentric rotation around a point in the magnetic core under arotating field, as shown by the trajectory of smaller particles(3 mm) surrounding the microgels. Another example deals withslightly larger PEGDA hydrogels prepared by a double emulsionprocess (carbonyl-iron fluidwatersilicone oil) implemented inimbricated capillaries.26 In addition to the classical translationin a field gradient and rotation in a rotating field (Fig. 9, 14),these coreshell microgels exhibit another original response tothe application of a magnet: above a threshold field gradient,the core is torn from the shell (Fig. 9, 5).

Other methods were reported to produce non-isotropicmagnetic polymer composites, in particular elongated colloids:magnetic polystyrene ellipsoids were prepared by dispersion in anelastomer membrane which was then stretched above its glasstransition temperature;27 magnetic fibres of poly(ethylene tere-phthalate) (PET) were electrospun;20 and cylindrical magneticneedles were self-assembled by clustering of hydrophilic MNPsunder a magnetic field, their electrostatic interaction with oppo-sitely charged polymers being finely tuned by the ionic strength.28

The application of a magnetic field during the self-assembly ofMNPs and polymers is a general route to produce rod-like MRPCs;for example, MnFe2O4 and g-Fe2O3 MNPs were co-assembled withpoly(maleic anhydride-alt-1-tetradecene) using a solvent-displace-ment method under field.29 In these worksreports, the usualapplications proposed for such rod-like micrometric MRPCs are:as magnetic stirrers for microfluidics,28 as probes for rheologicalanalysis of viscoelastic media including biological cells,29 or, lessoften, for field-responsive protection (presumably as a fabricwith adjustable flexural rigidity).20

3. Magnetically guided materials: magneticseparation and magnetic targeting

Magnetic guidance of polymeric composites can be describedas a controlled displacement of a system, composed of a

magnetic material associated to a polymeric matrix, uponexposure to a magnetic field gradient. Here, the polymericmatrix may serve as a multifunctional structure able to bindor adsorb dierent species, allowing their later separation orsimple displacement. This approach has been particularlyproductive in the field of biomedical applications, rangingfrom drug and cell delivery to diagnostic purposes. Nanosizedmagnetic composite materials still represent the majority of thedeveloped systems, with a variety of morphologies and applica-tions described. This particular subject was recently reviewedby Couvreur and colleagues.30 An overview concerningmagnetic guidance of polymeric composite materials will beprovided in the following paragraphs, using key examples toillustrate relevant strategies.

3.1. Magnetic guidance for drug delivery

It is widely accepted that the main limitations of conventionaldrug delivery are their diculty to overcome the naturalphysiological barriers and their lack of tissue/cell specificity.In an attempt to fill this gap, one could envisage a spatial andtemporal control of a cargo delivery through magnetic guidance,provided by an extracorporeal magnetic field.

The first obstacle that should be taken into account whenconsidering magnetic guidance is the high blood flow rate(greater than 10 cm s1 in arteries and 0.05 cm s1 in capillaries)encountered by the magnetic platforms upon intravenousadministration. It is thus paramount to develop powerfulmagnets that can provide strong magnetic fields, able tomanipulate particles against diusion and bloodstream. It isalso important to mention that these inhomogeneous magneticfields are usually eective at a distance of a few centimetresonly, which can limit their final application. Indeed, and aspresented in the following key examples, the majority ofmagnetic driven drug delivery strategies consider a fixedmagnetic field that will promote a locally increased residencetime, instead of a three dimensional guiding approach.

Orally administered proteins face many obstacles beforebeing finally absorbed into the circulation, implying adminis-tration of large doses in order to achieve significant eects.A known example to illustrate this problem concerns thedelivery of insulin, responsible for the regulation of bloodglucose levels, which has to be intravenously administered topatients suering from type-1 diabetes. In search of a lessinvasive insulin administration route, research has focusedon orally administered formulations, where the hormonewould be absorbed in the small intestine directly into circula-tion. As a means to increase the delivery system residence timein the small intestine, the team of Langer developed poly-(lactide-co-glycolide acid) microparticles loaded with both insulinand magnetite nanocrystals.31 Making use of the magneticmoieties of the system, the authors tested if the applicationof a magnetic field in the proximity of the small intestine wouldinduce an increase of blood insulin levels, with subsequentglucose level reduction. Indeed, under magnetic exposure, theadministered microparticles were able to consistently reduceglucose levels in mice, for as long as 20 h. Additionally, insulin

Fig. 9 Three examples of anisotropic magnetic polymer microparticles under arotating magnetic field: (AH) magnetic Janus PEGDA hydrogel in a field H =1.6 kA m1 rotating in-plane (B = 2 mT, the scale bar 50 mm);23 (ad); eccentricmagnetic PS core in a PAM hydrogel shell in the rotating field of a magneticstirrer (scale bar 50 mm); (e) visualisation of the flow velocity lines around theparticle;25 (14) CI core in a PEGDA hydrogel under a rotating field (scale bar200 mm, adapted from ref. 26 with permission from Elsevier, copyright 2013).


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


levels assessment in serum confirmed this magneticallyassisted insulin delivery.31

In another perspective, the use of metallic stents in theclinical practice has brought major therapeutic improvementsfor treatment of occlusive vascular diseases. However, in somecases a re-obstruction process, called in-stent restenosis, canoccur, with associated health complications. The current clin-ical solution consists in the use of drug eluting stents thatrelease potent antiproliferative drugs. Unfortunately, thisapproach has also led to some clinical complications. The teamof Levy addressed this issue using drug-loaded nanoparticlesmagnetically targeted to the stent region (see Fig. 10A and B).32

The developed poly(lactic acid) nanoparticles were loaded withboth magnetite nanocrystals and a non-proliferation drug(paclitaxel), and were shown to reduce stent restenosis, incomparison with non-treated control (see Fig. 10C and D,respectively), at 14 days post administration.

In this approach a paclitaxel dose of 7.5 mg per stent wasused, corresponding roughly to a 4 times lower dose as com-pared to current drug eluting stents,32 providing a safer andmore eective alternative to present methods.

In spite of great advances in cancer treatments, moderntherapeutics are still facing major limitations associated withnon-specific drug distribution and the inability to maintainecient drug concentrations. These barriers have been addressedby nanotechnology, as means to control drug delivery. In thisregard, the magnetically induced or assisted drug delivery canbe a way to improve treatment while reducing o-target eects.Additionally, by combining magnetite nanocrystals and anti-cancer drugs within a polymeric platform, one can envisage theformation of a hybrid therapy and diagnostic tool (theragnostic),

allowing guidance, controlled drug delivery and imaging. Inan elegant study, Arias et al. demonstrated the potential ofsqualene-based bioconjugates, containing both magnetitenanoparticles and the anticancer drug gemcitabine, for magneti-cally driven cancer therapy and imaging (i.e. magnetic reso-nance imaging, MRI).33 Under the influence of an externalmagnetic field, the intravenously administered nanoparticleswere capable of increasing drug accumulation in tumour tissuewith subsequent tumour growth inhibition (see Fig. 11C), whileallowing MR imaging (T2 weighted contrast, Fig. 11A and B).

33

This example emphasizes the interest of combining magneticresponsive materials with imaging and drug delivery, wideningthe toolbox for cancer treatment and diagnostics.

Gene therapy relies on the use of genes or sequences ofnucleic acids to produce or modulate protein expression intarget cells. In this case the genetic material can be faced as apro-drug, acting at the level of cellular gene expression, asmeans to normalize or optimize protein expression profiles. Inthe case of lung pathologies the administration of drugs in theform of aerosols is widely implemented, since it constitutes aminimally invasive approach towards drug administration.With that in mind Rudolph et al. associated superparamagneticiron oxide nanoparticles, coated with 25 kDa poly(ethyleneimine)to plasmid DNA, in order to produce a gene therapy system forlung diseases.34 The developed formulation was nebulisedduring inspiration by means of intratracheal intubation inmouse lungs, while a magnet was centred above one of thelungs. The authors demonstrated that, by action of themagnetic field, a significant increase of the particle depositionoccurred, associated to a decrease of particle exhalation.Additionally, a preferential accumulation was observed in thelung where the magnetic field was applied (2.5-fold increaseobserved for intact animals).34 Although drug ecacy was notevaluated in this study, this example broadens the use of

Fig. 10 Targeted local delivery of magnetic nanoparticles (MNPs) to a stainlesssteel stent mediated by a uniform field induced magnetisation eect. Theuniform field generated by paired electromagnets (A) both induces high gradi-ents on the stent and magnetises drug-loaded nanoparticles, thus creating amagnetic force driving MNPs to the stent and adjacent arterial tissue (B). Animalstreated with nanoparticles under magnetic vs. non-magnetic conditions weresacrificed, and the stented carotid segments were harvested 14 days post-surgery. As observed, upon treatment with nanoparticles under magneticconditions a significant restenosis reduction (intra-stent tissue thickness) wasobserved (C), in comparison to untreated controls (D). Adapted from ref. 32 withpermission.

Fig. 11 Examples of T2-weighted magnetic resonance images of tumoursobtained at 2 h-post injection of nanoparticles in the absence of externalmagnetic field (A) and the same nanoparticles guided by an external magneticfield (1.1 T) (B); in the latter case, nanoparticles were found dispersed all over thetumour tissue, as observed by the darkening of the tissue. In vivo anticanceractivity of nanoparticles (with extracorporeal 1.1 T magnetic field) comparing thefollowing groups: Untreated (K), placebo nanoparticles without drug but withmagnetite (J), free drug (B), nanoparticles without magnetite moieties but withdrug (m), drug/magnetite loaded nanoparticles with no extracorporeal magneticfield (n), drug/magnetite loaded nanoparticles with 1.1 T extracorporealmagnetic field (), in L1210 subcutaneous tumour bearing mice (C). Adaptedfrom ref. 33 with permission. Copyright 2013, American Chemical Society.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


magnetic guidance for drug delivery, taking us one step closerto real world clinical applications.

3.2. Magnetic cell delivery/manipulation

Cell therapy has for long been regarded as a forefront thera-peutic approach, holding promise for a wide range of humanpathologies. However, the control of cell localization, uponadministration in the body, is a major limitation to achievingclinical ecacy. In this regard, several groups have handledthis question by associating magnetic material to individualcells that, upon injection, would permit cell accumulation atthe target site by the use of a localised magnetic field. Levy andcolleagues have elegantly demonstrated the magnetic guidanceof endothelial cells bearing internalised magnetic nanoparticles.35

The magnetic platform consisted in albumin stabilised poly-(lactide) nanoparticles containing iron oxide nanocrystals.Bovine aortic endothelial cells were incubated with thedescribed nanoparticles and injected into the left ventricularcavity of rats bearing a stent in the carotid artery. The mainobjective was to induce stent re-endothelization, a majorunsolved problem in stent angioplasty procedures. Indeed,complete and uniform cell coverage was observed after 5 minutes,when a uniform magnetic field (100 mT E 80 kA m1) wasapplied. Conversely, in the absence of magnetic field, nomagnetic nanoparticle loaded cells were detected. The authorswent further and showed that upon cell injection through astented rat carotid, and using the same magnetic field asbefore, cells accumulated in the stent area only. Again, whenno magnetic field was applied, no signal was observed.35

Upon injury to the central nervous system, unrecoverableconsequences, including loss of functionality, may occur. It isfundamental to gain knowledge regarding strategies that canpotentiate neural regeneration. As an example, neurite growthis of significant importance in order to produce wired func-tional neuronal networks. In this regard, Fass and Oddefocused their attention on the development of magnetic driventools that could provide guided neurite formation.36 By the useof commercially available superparamagnetic poly(styrene)microbeads coated with anti-mouse antibodies (Dynabeads,Invitrogen), the authors demonstrated the ability to induceneurite growth in neuron cells using an external magnet. Theapplication of a constant force (450 pN) to the cell-attachedbead led to the formation and consequential neurite elongation(Fig. 12). Here the attachment to cells was achieved by coatingthe beads with an antibody that targeted a specific cellmembrane protein.36 This approach allows the application offorces to individual cells with great accuracy and precision, in aminimally invasive fashion. Fundamental studies consideringthe mechanism of neurite formation could be addressed.

The previously exposed examples strengthen the potential ofmagnetic guidance towards cell delivery and manipulation.Supported by their versatility and biocompatibility, magneticbased materials may find use in future tissue engineeringapplications, where three-dimensional guidance and manipu-lation may support new technological achievements.

3.3. Bio/chemo-separation applications

The separation or concentration of certain species in solution,by means of magnetic forces, opens up original perspectivestowards biomedical and environmental applications. Theenvisaged strategies can be applied to simple molecules or tolarge structures like cells, using dierent degrees of complexity.Again, this approach has been particularly prolific concerningbiomedical applications. Nevertheless, other examples, likeeuent treatment, could be found and will be briefly exposedin the following paragraphs.

Protein separation by means of magnetic field may provide asimple solution for sample concentration and purification,before downstream applications (e.g. diagnostics, production).Maintaining the native protein conformation is paramount formost applications, making mild separation methods, likemagnetic separation, the best choices for protein processing.Indeed, several commercial solutions are already available formagnetic protein separation, ranging from nano to micronsizes. In most cases the particle surface area limits the amountof protein adsorption, reducing separation eciency. Thefabrication of porous matrices has been used to circumventthis limitation. However, the use of porous materials impliesslower protein diusion rates that may increase the time forpurification. With that in mind, the team of Bruening devel-oped silica-coated magnetite nanoparticles bearing poly-(2-hydroxyethyl methacrylate) brushes on their surface.37 Thepolymer brushes were then derivatised in order to acquire theability to bind poly(histidine) tagged proteins. Poly(histidine)sequences are usually introduced by molecular biology techni-ques to provide recombinant proteins with a simple motif forsubsequent purification procedures. The brush conformationallowed an increase of the surface area, without compromisingdiusion rates or time of procedure, and translated intosuperior amounts of protein purified, higher than commerciallyavailable magnetic separation systems.37

Fig. 12 Neurite initiation and elongation in response to an applied force(450 pN) (A). Neurite length history corresponding to the sequence in A. Arrowsindicate time points corresponding to the images in A that were taken duringforce application (B). Adapted from ref. 36 with permission from Elsevier, copy-right 2013.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


Microfluidics present unique features that make this tech-nology best suited for the development of novel analyticalsystems. They allow unprecedented reduction in terms ofmaterials needed to provide measurable results, holding pro-mise for faster and cheaper diagnostic tools. In the absence ofapplied forces the transport of molecules through a micro-fluidic device is mainly dependent on diusion, which impliesa direct limitation in terms of process eciency. As a means toaddress this problem the team of Stayton and colleaguesdesigned a microfluidic approach where small magnetic nano-particles were used to transfer components from one flowstream to another, by means of an applied magnetic field.38

Magnetic nanoparticles consisted of biotinylated poly(metha-crylate-co-N-isopropylacrylamide) coated magnetic cores (5 nm,Fe(CO)5). The polymeric fraction is pH responsive, allowing themagnetic nanoparticles to aggregate at pH 7.3 and readilyredisperse at pH 8.4. Biotin moieties were used in this studydue to their high anity to streptavidin. The authors mixed themagnetic nanoparticles with a fluorescently labelled strepta-vidin and applied it to the microfluidic system (Fig. 13A). Aspreviously described, at pH 7.3 the complexes aggregated in theleft flow stream, therefore increasing their magnetic momentand allowing their sorting to the second (right) flow stream bymeans of a magnetic field. Once in the second stream, thecomplexes completely redisperse owing to the pH change,allowing their highly ecient recovery (ca. 80%, Fig. 13B).

Droplet microfluidics is another interesting approachtowards optimized diagnostic tools. By using antibody graftedmagnetic nanoparticles, the group of Malaquin and colleagues

recently developed a novel device for high-throughput nanoliterassays.39 The approach lies in the formation of nanoliter dropsthat travel immersed, under flow, in an oil phase inside amicrotube. Using a pair of soft magnetic tips, named tweezers,the suspended antibody-grafted nanoparticles can be mergedwith a serum-containing drop and used to isolate a specificcontent of the mixture (in this case the thyroid-stimulatinghormone (TSH), Fig. 14). After several steps of washing andconjugation with other antibodies, the final TSH concentrationcould be determined with high sensitivity.39 When comparedwith other standard methods (e.g. ELISA) this approach allowssimilar sensitivity while reducing the operation time (from2.5 h to 10 min) and sample volume by 1000 times.

Biomolecules sorting by means of a magnetic field mayenable unprecedented reduction of analyte volume and concen-tration, while diminishing very significantly operation times.This has very obvious implications at the level of diagnosticcosts, a major limitation in current medicine. It is expected thatmagnetic based technology may find use in a wide range ofhigh sensitive and high-throughput applications.

Cell separation, by the action of an external magnetic field,has long been regarded as a forefront approach in the bio-medical field. Indeed, several commercial products are availablefor pre-clinical research (e.g. DynaBeadss, Invitrogen; MACSs,Miltenyi Biotec) and diagnostics (Veridexs; Johnson & Johnson,Estapors, Merck, Adembeads, Ademtech). These approachesare based on the association of antibody grafted magnetic

Fig. 13 Target analyte separation, in a microfluidic channel, facilitated by pH-responsive magnetic nanoparticles (MNPs). The left stream (green) is the samplethat has been pre-incubated with MNPs. MNP aggregation is induced by using alower pH buer in this sample flow stream. The pH of the right stream (pink) ischosen to reverse MNP aggregation. A rare-earth magnet provides sucientmagnetic field to attract the aggregates laterally into the higher pH flow stream.The aggregates move out of the sample flow stream in the higher pH stream,where they return to a dispersed state, carrying the bound target analyte withthem (A). Fluorescence microscopy results showing continuous stream purificationof streptavidinMNP conjugates.When the magnetic field is applied, the conjugateaggregates moved laterally across to the higher pH flow stream, where theyredisperse. When the magnetic field is applied, most conjugates were moved intothe high pH stream and 80% of the conjugates were collected from the rightoutlet (B). Adapted from ref. 38 with permission of The Royal Society of Chemistry.

Fig. 14 Diagram illustrating the basic steps involved in the magnetic immu-noassay. Using an initial trap (i.e. magnetic tips, opposed triangles on thescheme), antibody-grafted magnetic nanoparticles contained in the first dropwere captured and transferred to a second sample drop composed of thethyroid-stimulating hormone (TSH) in serum (a, and b). After incubation inthe capillary loop following the first tweezers, the magnetic nanoparticles withthe antibody complexes were trapped again in the same tweezers and washedwith a droplet of buer to remove non-specifically adsorbed proteins (c). Thebeads were resuspended and incubated in a fourth droplet containing theenzyme-labelled secondary antibody (d). Washing was performed in the secondmagnetic trap with three successive buer droplets (e) to remove unboundsecondary antibodies. Finally, the particles were released and incubated with afluorogenic enzymatic substrate and the output was read (f). Adapted fromref. 39 with permission from John Wiley and Sons, copyright 2012.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


nanoparticles to a specific cell population (by means of theantibody specificity) and to their subsequent concentration/separation by the action of an externally applied gradientmagnetic field. The team of Soh took this approach one stepfurther and envisaged the use of multitarget magnetic cellsorting by exploring two dierent magnetic systems with dis-tinctive magnetisation saturation.40 The authors used a mixtureof 3 dierent E. coli bacteria one expressing the T7 peptide,the second a streptavidin-binding peptide and a third that wasnot labelled that were preincubated with two dierentmagnetic beads. One of the beads (4.5 mm in diameter) waslabelled with an anti-T7 monoclonal antibody and the second(2.8 mm in diameter) was grafted with streptavidin. The samplewas then loaded in a continuous-flow microfluidic device able toseparate labelled bacteria according to their magnetisation.Indeed, this process allowed the separation of the three popula-tions with good purity (>90%) in just one single pass, asdetermined by flow cytometry.40 Although a promisingapproach, the application of such a system to mammalian cellsstill faces some barriers. Larger structures imply a strong influ-ence on the flow drag, making ecient cell sorting impossiblewith current magnetic particles. It is then imperative to developmagnetic tags with higher saturation magnetisation i.e. highermagnetic payloads that could provide enough force for ecientcell separation. Indeed, most magnetic beads developed at thispoint have an iron oxide content of around 50 wt% (B10 vol%),corresponding to Msat B 30 kA m

1 at most.Finally, in the ecotoxicology field, the selective treatment of

industrial and domestic euents is of uttermost importance anda main issue for sustained societies. Wastewaters may containdistinct doses and types of contaminants, have inherent pH andionic strength properties andmay include solid residues as well. Itis then important to find versatile approaches, able to keep theseparation process cost eective. In this sense, the development ofa system that could extract contaminant, then be selectively andrapidly separated from the euent and able to be recycled formultiple uses is of major interest. Therefore, Bee et al. associatedthe adsorption capacity of activated carbon with magnetic ironoxide nanoparticles (maghemite, g-Fe2O3) by immersing them inan alginate matrix, forming 2.8 mm diameter composite beads.41

Model pollutants were used in order to test the system ecacy:positively charged methylene blue and the negatively chargedmethyl orange. The authors showed that the developed systemwas able to rapidly adsorb the proposed pollutants, independentlyof the pH, and to potentially serve for a water treatmentprotocol.41 The applicability of such a system using real euentshas still to be tested, but this approach holds promise for theimprovement of industrial and domestic euent treatment.

4. Magnetically actuated thermoresponsivematerials: controlled drug delivery andshape shifting

An interesting property of magnetic materials is their ability toproduce heat when exposed to an alternatingmagnetic field (AMF).

Already widely used for macroscopic magnetic devices, in forexample induction cooking, this phenomenon, called inductiveheating, can also be applied to polymer composites containingnanometric magnetic fillers. Indeed, when exposed to an AMF,magnetic nanoparticles generate heat through hysteresis losses(ferromagnetic particles) or through oscillation of theirmagnetic moment due to Neel and Brownian relaxations(superparamagnetic particles). Hence, this section will focuson temperature responsive polymer composites where thetemperature variation is obtained by inductive heating,generated by the exposure to high frequency AMF. Applicationswhere the temperature rise (hyperthermia) is the sole finalresponse are excluded from the scope of the review since thepolymer material plays no role in it.

One of the major advantages of magnetic induction is thepossibility of having a non-contact remote control of tempera-ture changes. Heat generated by the magnetic nanoparticlesunder exposure to AMF will trigger a phase or conformationaltransition of the polymer that will ultimately lead to theprogrammed response. This has been implemented mainly intwo fields: drug delivery and shape memory materials.

It is worth noting that single domain MNPs, which haveintrinsic superparamagnetic properties (above their blockingtemperature, TB), have been largely preferred in all the applica-tions presented here. Contrary to ferromagnetic particles, theyelicit no dipolar attraction in the absence of an externalmagnetic field. Consequently they have a better colloidal stabilitythat favours their storage, stability upon injection or homo-geneous dispersion within a polymer matrix.

4.1. Controlled drug release

A majority of the drug delivery systems (DDS) proposed so farelicits a release profile with an initial fast release followed by aslow continuous release. This could be a serious limitation fortheir application in vivo since the initial high concentrationmay constitute a toxicological risk and the slow release canresult in underdosing, especially when considering chemo-therapy. In other cases, for example when degradable particlesare used, a constant release can be observed. However it is moredicult to design systems that will exclusively release theircontent on demand. It is therefore paramount to improve thetemporal and spatial control of drug delivery.

Inductive heating can be used as a stimulus by promotingstructural changes in a thermosensitive material, accompaniedby the concomitant release of entrapped molecules. This strategyhas been applied to the design of on-demand controlled drugdelivery systems, mainly for cancer therapy in what has beencalled magneto-chemotherapy.42 Indeed, magnetic fieldactuation is a very attractive mechanism for potential clinicaluse, mainly due to its improved tissue penetration as comparedto light or heat flow. Two types of systems have been developed:particulate DDS (nano- andmicroparticles) for parenteral admin-istration and macroscopic materials (mostly membranes) forimplantable devices.

4.1.1 Nano- and microparticles. Nano- and microparticulatecomposites have been the most studied systems for drug


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


delivery applications, with a wide range of structures and drugrelease mechanisms being proposed. Here we present some keyexamples illustrating various strategies that have been envisionedso far.

A first strategy consists in covering the MNPs surface with athin polymer shell that will entrap the drug, as presented byLouguet et al. among others.43 In this work, the authors choseto use lanthanum strontium manganese oxide (LSMO) MNPscoated with a silica shell. Indeed, although iron oxide MNPs arethe most widely used (mainly due to their good biocompat-ibility), some safety concerns regarding their in vivo applicationcan arise due to potential tissue heating to above the necrosisthreshold temperature (45 1C), even though the actual tempera-ture reached may be limited by thermal dissipation. Thismotivated the study of other types of MNPs such as LSMO,whose Curie temperature (around 50 1C) is well below that ofiron oxide (B500 1C) and can be tuned by the composition ofLa and Sr in manganite, even though the latter exhibitsinherent toxicity. In Louguet et al. work, the polymer shellwas deposited through electrostatic interactions between thesilica surface and hydrophilic block copolymers containing ashort poly(lysine) block that covers the particles surface. Theother block is a poly(ether) segment that forms a polymerbrush, providing colloidal stability (and stealth properties),thermoresponsiveness (LCST) and a reservoir for drug loading(Fig. 15). To control the thermoresponsive properties of theassemblies, the ratio of ethylene oxide and propylene oxideunits could be varied in order to modify the apparent LCST ofthe polymer brush. Drug release was associated with the poly-mer brush shrinkage when a temperature above LCST wasreached. As shown by the authors, this could be controlledeither by raising the environment temperature or by applyingan AMF.

Recently, DDS presenting a capsule morphology have drawnattention due to their ability to carry both hydrophobic andhydrophilic payloads in their aqueous core and membrane,respectively. In a recent work, the team of Prof. San-Yuan Chen

developed a one-step double emulsion method for theproduction of nanocapsules whose membrane is constitutedby poly(vinyl alcohol) and magnetite (Fe3O4) nanoparticles(Fig. 16a).44 Both the preparation process (double emulsion)and the capsule morphology allow exceptionally high payloadsof hydrophilic and hydrophobic drugs. This was demonstratedfor two anticancer drugs, doxorubicin and paclitaxel, withloading eciencies of up to 60 and 95 wt%, respectively.Exposure to AMF drastically enhanced drug release in a way thatcan be tuned by the polymer molecular weight (Fig. 16b and c).The validity of such an approach was proved both in vitro andin vivo with enhanced cytotoxicity, or respective tumour volumeshrinkage, when AMF was applied. Moreover, the authorsshowed a synergistic eect when a combination of the twoanticancer agents, AMF and targeting was used (Fig. 16d).

In another study the same group prepared nanocapsules,based on poly(styrene)-b-poly(allyl alcohol), containing MNPsand capable of encapsulating DNA as well as hydrophobicmolecules.45 Interestingly, it was shown that the quantity ofDNA released under AMF (50 kHz) was field-strength-dependent.Up to 1.2 kA m1, the thermally accelerated release appeared tobe a reversible process with no release observed when the AMFwas stopped, accounting for a transitory permeabilisation ofthe polymer shell. But at 2.0 kA m1, DNA continued to bereleased even when the field was switched o, suggesting apermanent degradation of the polymer shell.

Also using a double emulsion method, Chiang et al. pre-pared poly(D,L-lactic-co-glycolic acid) (PLGA) microcapsules(13 mm in diameter) with a 250 nm thick membrane contain-ing a high density of iron oxide nanoparticles (up to 54 wt%).46

The aqueous cavity of these microspheres was loaded withdoxorubicin, to be used in cancer treatment via intratumoralinjection. The authors demonstrated that for MNPs loading ofat least 25 wt%, the application of an AMF (100 kHz, 2.5 kA m1)could induce heating above the Tg of PLGA (i.e. 40 1C). Thisallowed for the remote controlled pulsatile release of the drug,as shown in Fig. 17. This finely tuned, non-contact and

Fig. 15 Preparation of magnetic responsive drug delivery systems (DDS) based on lanthanum strontium manganese oxide (LSMO) MNPs and polymer brush with atunable LCST (top left). Chemical structures of the block copolymers used: (a) poly(ethylene glycol)-b-poly(lysine), and (b) poly(ethylene glycol-co-propylene glycol)-b-poly(lysine), with x = 6 and y = 29. Principle of drug loading and alternating magnetic field (AMF) triggered drug release (right). Reprinted from ref. 43 with permissionof The Royal Society of Chemistry.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


on-demand release of doxorubicin can allow an unprecedentedcontrol over drug concentration, by adjusting the duration ornumber of pulses according to the patients need or to stay inthe therapeutic window.

Another type of DDS presenting a capsule morphology ispolymer vesicles, or polymersomes, that are particularly inter-esting due to their biomimetic structure (bilayered membrane).In a theragnostic approach, our group has developed polymer-somes loaded with both maghemite (g-Fe2O3) nanoparticlesand doxorubicin for combined controlled drug delivery andmagnetic resonance imaging (MRI) contrast enhancement.42

When an alternating magnetic field (500 kHz, H = 2.12 kA m1)was applied to these particles, the drug release kinetics wassignificantly increased although no macroscopic heating wasmeasured in the dispersion. It was hypothesized that localheating at the vicinity of the MNPs leads to permeation of the

thin polymersome membrane that increased the drug diusionrate, resulting in a faster drug release. An enhanced cell deathwas recently demonstrated using these hybrid polymersomesin vitro.47 Polymersomes loaded with 6 wt% doxorubicin and30 wt% MNPs, internalised in HeLa cells, elicited an 18%increase in cell toxicity when a high frequency alternatingmagnetic field (750 kHz, H = 11.2 kA m1) was applied ascompared to the control without magnetic field exposure.

4.1.2. Implantable devices. Implantable drug deliverydevices are another type of DDS that can be modified forremote control by magnetic fields. For instance, compositesof PNIPAM and magnetite nanoparticles have been studied bySatarkar and Hilt for pulsatile drug release.48 They preparedrelatively thick (1500 mm for a diameter of 15 mm) disks inorder to reduce surface to volume ratio (S/V), i.e. reduce heattransfer to the surrounding medium through diusion, opti-mizing the maximum temperature that can be reached insidethe material. Vitamin B12, used as a model drug, was releasedfrom the composite following a Fickian profile in the absenceof AMF. Its release was largely accelerated when an AMF wasapplied (297 kHz, 5.3 kA m1). Indeed, the heat generatedunder AMF led the gel to de-swell and collapse (above LCST),causing the expulsion of water with a concomitant enhance-ment of the drug release, also favoured by the increased drugdiusivity at higher temperature. Short AMF pulses were usedto obtain on-demand stepwise burst releases. Nevertheless thenumber of bursts was limited due to a quick exhaustion of thedrug payload or to the complete collapse of the polymer matrix,which impaired further drug release.

One way to obtain sustained release over extended periods oftime and/or increase the number of repetitions is to use thistype of composite materials as a valve covering an implantabledrug reservoir that will provide access to a large quantity of thedrug. This approach was described by Hoare et al. who depositeda composite membrane, based on ethylcellulose and containing

Fig. 16 Schematic of one-step emulsion process for the preparation of PVA/ironoxide capsules with a PVA molecular weight of 16 000 or 19 000 g mol1 (a).Cumulative drug release of paclitaxel (PTX) and doxorubicin (DOXO) with orwithout application of the alternating magnetic field (MF, 16 kA m1) fromcapsules (DEC) prepared with PVA 16000 g mol1 (b) and 47 000 g mol1 (c),showing the influence of the polymer molecular weight on the drug release rate.Cell viability after incubation for 24 h with capsules loaded with PTX or DOXO orboth (PD), with or without MF (16 kA m1), targeted (with a cancer targetingpeptide, IVO24) or not (d). Experiments show the synergistic eect of chemo-therapeutics and application of the AMF. Adapted from ref. 44 with permissionfrom John Wiley and Sons, copyright 2013.

Fig. 17 Doxorubicin release profiles from the iron oxide/PLGA microcapsuleswith and without application of a high frequency magnetic field (HFMF) pulses(a). Fluorescence images showing a green-labelled microcapsule colour changeas doxorubicin is released into the medium (b). Reprinted from ref. 46 withpermission from John Wiley and Sons, copyright 2013.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


MNPs and thermoresponsive nanogels (eliciting a LCST) over adrug reservoir.49 This construction is schematically representedin Fig. 18a. Upon application of an AMF, the MNPs heat thenanogels, leading to their shrinkage and creating pores thatdramatically enhance the drug release rate. This release rate, aswell as the ono drug release ratio, could be tuned by themembrane thickness (Fig. 18b) and the nanogel loading(Fig. 18c). As expected, the frequency and power of the appliedAMF also aect the drug dosing by changing the steady statetemperature of the device. This tunability opens new avenues inthe development of adaptable therapeutic tools, providingunprecedented control over real time drug dosing.

4.2. Shape memory materials

Shape memory polymers are capable of performing importantshape changes, after application of a programming process usually a deformation at fixed temperature in order to recovertheir original shape. Generally they are thermoplastic reticu-lated materials with a glass and/or melting transition tempera-ture. After deformation at a temperature higher than theirtransition temperature, the temporary shape is fixed by rapidcooling (Fig. 19a). Then, the original shape can be recovered byheating again above the transition temperature (Fig. 19a and b).Such materials can be useful for applications in a wide range of

fields such as defence, smart textiles, packaging, aerospace,adaptive optics, robotics and biomedical engineering, likemagnetic actuators (see Section 1) or minimally invasive sur-gery devices.

Thermoplastic shape-memory polymers (SMP) have beenadapted for magnetic remote activation by filling them withMNPs. The thermally induced shape shifting of such materialscan then be triggered by exposing them to an alternatingmagnetic field. Indeed, the heat generated by the MNPs, underan alternating magnetic field, can trigger useful phase transi-tions of SMPs.

Early works concerned materials with dual-shape properties,i.e. one thermal transition at a temperature Ttrans that waseither a melting temperature (Tm)

50,51 or a glass transitiontemperature (Tg).

51 To prepare these composites, relatively highamounts of MNPs (at least 10 wt%) were dispersed in themonomer mixture before polymerization. To favour the pre-paration of homogeneous materials (evenly distributed filler),Mohr et al. used silica coated iron oxide NPs in order to reduceagglomeration into mm-sized clusters that could sedimentduring the polymerization process.51

In an elegant study, the Lendlein group prepared triple-shape polymer composites two thermal transitions containingpoly(e-caprolactone) (PCL) and poly(cyclohexyl methacrylate)(PCHMA) segments, for non-contact actuation.52 They showedthat MNP incorporation had little influence on the thermal

Fig. 18 Schematic representation of the composite membrane valve and itsfunctioning principle: under application of an alternating magnetic field (AMF),the magnetic nanoparticles release heat that reversibly shrinks the thermo-sensitive nanogels, enabling the release of a drug from a reservoir (a). Rate ofmass transfer of a model drug (fluorescein) as a function of membrane thickness(b) and nanogel loading (c). Adapted from ref. 49 with permission. Copyright2011, American Chemical Society.

Fig. 19 Schematic representation of the alternating magnetic field (AMF)induced shape memory eect in shape-memory polymers composites (a). Thepermanent shape (top) is transformed in a, second, temporary shape by aprogramming process (deformation at T > Ttrans). The temporary shape isstabilised by the crystalline phase of the polymer, represented as rigid segments(middle). Induction heating leads to a temperature increase inside the matrix,above the transition temperature (Ttrans) that induces the permanent shaperecovery (bottom). Series of pictures illustrating the previously described AMF(258 kHz, H = 30 kA m1) induced shape recovery process for a poly(etherurethane) composite (10 wt% magnetic nanoparticles) sample that wasdeformed into a spiral (b). Adapted from ref. 50 and 51 with permission.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


properties of the polymer but impacted upon the mechanicalproperties, significantly decreasing the elongation at break.Hence, the filler content should be a compromise betweenthe heating properties (higher reachable temperature) and themechanical properties. Another important parameter thatshould be taken into account in order to gain a good controlover shape recovery is the surface to volume ratio (S/V). Asobserved in Fig. 20, depending on the S/V ratio, the heat loss(by exchanges with the surrounding environment) can vary,which may imply large dierences in terms of the magneticfield strength necessary to attain the transition temperature.Finally, excellent triple shape properties with complete recoveryof the original shape were obtained for composites containing40 wt% of PCL. A step-wise increase of the magnetic fieldstrength allowed the recovery of the intermediate and originalshapes in a sequential manner.

More recently, the same group showed that it is possible toadjust the apparent switching temperature of a shape memorycomposite by combining contributions from two sources: theheat energy coming from the magnetic field actuation and theheat flow from the environment.53 Experimental measuresrevealed that environmental and inductive heating contributeadditively to the temperature in the composite material, the

latter being the environment temperature plus the contributionof the inductive heating. In a simple approach, the contributionof the inductive heating could be expressed as DTmag = kH

2

where H is the magnetic field strength and k is a materialrelated constant. Consequently, under AMF, the apparent environmental switching temperature of the shape memorypolymer can be decreased. To demonstrate this principleexperimentally, the authors built a fixation device consistingof thermoresponsive hook and lock (Fig. 21). To successfullyactivate the device, the hook should enter the fixation holebefore this part changes shape. As shown in Fig. 21, this wasachieved by filling the hook with MNPs and by applying theappropriate AMF. This work demonstrated that by combiningdierent heat sources it is possible to obtain a complexresponse from a thermosensitive composite system.

Lendlein and collaborators also showed that it was possibleto prepare magnetic memory composites by filling tempera-ture-memory polymers with MNPs.54 Temperature memorypolymers can remember the temperature at which they weredeformed. This eect results from the fact that they have broadtransitions and that the mechanical deformation is fixed by thevolume fraction of the domains associated with this transition(Tm or Tg). Since for a given composite material, the tempera-ture reached in the material corresponds to a given magneticfield strength, the temperature memory corresponds to amagnetic memory. This property was illustrated in an experi-ment where two identical samples of the same compositematerial were deformed at dierent field strengths. Eachsample recovers its original shape only when the field at whichit was deformed is reached.

The previous examples of shape-programming systems anddrug release triggered by ac magnetic fields illustrate the varietyand complexity of devices that can be thought of in future drug

Fig. 20 Pictures taken with an infrared camera of the shape recovery atdierent magnetic field strengths of two samples made of the same compositebut deformed using two dierent methods. The sample on the left side (ad) wasdeformed by elongation whereas the sample on the right side (eh) underwent abending process. The temperature reached inside the material appears todepend on the programming process, with recovery at higher field strength forthe sample deformed by elongation. This highlights the influence of the surfaceto volume ratio (S/V) in the design of shape memory polymer composite systems.Reprinted from ref. 52 with permission of The Royal Society of Chemistry.

Fig. 21 Principle of the fixation device or the active hook experiment: shapeprogramming of the two parts (a) and working principle of the fixation (b).Images taken at dierent environmental temperatures: environmental heating atH4 = 20.2 kA m

1 (c) and environmental heating without magnetic field (H = 0)(d). Reprinted from ref. 53 with permission from John Wiley and Sons, copyright2013.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


delivery and robotic applications, with ever increasing controlover the fine tuning of the magnetic actuation opening newperspectives for high precision applications.

5. Conclusions

To date, use of composites of polymer and magnetic nano-particles is the most elegant and ecient way to obtainmagnetic responsive polymer materials exhibiting high ampli-tude magneto-response. These MRPCs can be divided intothree classes with regard to their intended applications and/or mechanisms involved in the response to the magnetic field.The first group gathers ferrogels, microsized hydrogels andother elastomeric materials that can be deformed in a con-trolled manner in homogeneous fields or gradients. Anotherclass that we identified is composed of polymers designed forinteractions with biomolecules or biostructures (biocompatible,bioresponsive polymers). In this field, MRPCs were successfullyemployed for magnetic guidance of drug delivery systems,manipulation of individual cells and separation in complexbiological media or environmental samples. Finally, MRPCsprepared from thermoresponsive polymers can be remotelyactivated using alternating magnetic fields. This strategywas especially used for controlled drug delivery with injectableor implantable devices and in the area of shape-memorypolymers.

In order to perfectly anticipate the properties of theseMRPCs, many parameters need to be controlled (Table 1).Especially, the geometrical parameters of the devices (size,shape, surface/volume ratio, aspect ratio) and the nano-structure/composition of the composite (i.e. magnetic materialloading, compatibility between the polymer and magneticphases) are relevant parameters that drive the amplitude ofthe desired magneto-response. Therefore, a careful multiscaledesign is necessary to fully adjust the final properties. Inparticular, reducing the sample size can have a positive impacton the response kinetics. However, when considering heatexchanges, the decreased diusion time, associated with an

increased surface area (diusion time is scaling like the squareof the sample characteristic size), can also be a drawback,limiting the maximal temperature that should be reached bymagnetic induction with ac fields. Another important point isthe adjustment of polymer transition temperatures to just a fewdegrees above the body temperature. Finally, the obtention ofvery soft materials (e.g. low mechanical moduli) is alsorequested, a practical thumb-rule being the calculation of theMER factor described in Section 1.

Magnetic remote control is particularly interesting and hasalready demonstrated promising results for biomedical appli-cations. Indeed, even if the concept of MRPCs has initially beenestablished for material sciences and actuators, it is nowimpacting significantly the biomedical field. Most livingsystems are not sensitive to magnetic field, rendering thisactuation strategy specific and potentially harmless. In addi-tion, the possibility of having long distance eects, using amagnetic field gradient or alternating magnetic field, opens upnew avenues for realistic remote actuation of therapeutic ordiagnosis devices. In this sense, nanosized magnetic responsivecomposite materials present several characteristics which makethem optimal for therapeutic applications. Due to their smallsize, one can envisage their intravenous or localized adminis-tration, in a minimally invasive fashion. Additionally, andowing to the versatility of the polymeric structure the associa-tion with other components (e.g. drugs, tracing agents) can beeasily achieved. The magnetic moieties will then enable thespatial manipulation and trigger a relevant therapeutic eect(e.g. drug release, hyperthermia) by means of the application ofan external magnetic field. Moreover, due to the inherentnanoparticle high surface area and ease of manipulation, bymeans of magnetic gradients, such MRPC nanoparticles mayallow revolutionization of the fields of molecular manipulationand separation at the nanoscale in the near future.

Knowledge of the physical mechanisms associated with thepolymer materials response to magnetic actuation is of utmostimportance for bio-engineers involved in the inception of suchnew technologies, especially for devices implemented in the

Table 1 Main parameters for the control of MRPC properties depending on the intended application

Magneto-elastic deformations Magnetic guidance, magnetic separation AMF-induced thermal phase/shape transition

Magneticfillers

Controlled sizes:- Ferrogels: magnetic core size orclusters size superior to gel mesh-size- Silicones: micrometric size inducingparticle chaining when submitted to dcmagnetic field

- High loading content to reach highmagnetisation- Biocompatible magnetic materials(surface passivation required whenMNPs are not made of pure iron oxide toshield from the toxicity of hardnanomagnets)

- Homogeneous distribution in thepolymer matrix- High loading content- High specific absorption rate (SAR) tomaximise the heating source

Polymermatrix

- Material as soft as possible to optimizethe magneto-elastic ratio (balance ofmagnetisation vs. mechanical strength)

- Small MRPC particles size to increasehalf-life in blood circulation and mini-mize drag force- Controlled of surface properties tofinely tune the interactions withbiostructures depending on theapplication- High porosity for environmentalpurposes (remediation)

- Drug release:- Finely tuned Ttrans to meet the biologicalconstraints (4045 1C)- Control of the physico-chemical mechanismsleading to drug release: permeability vs.degradation- Shape memory devices:- Minimize the S/V ratioFor both: controlled heat dissipation(size, heat conductivities)


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


body or drug delivery vehicles embarked in the systemiccirculation. One can therefore anticipate future promisingdevelopments of MRPCs and their implementation in devicesthat will particularly impact the biomedical field.

Acknowledgements

Financial support from the European Commission under theseventh framework within the frame of the NanoTher project(CP-IP 213631-2), ESF RNP Precision Polymer Materials P2M,CNRS-MI G3N (NanoBlast), and COST actions TD1004 and1007 are acknowledged.

Notes and references

1 M. A. Cohen Stuart, W. T. S. Huck, J. Genzer, M. Muller,C. Ober, M. Stamm, G. B. Sukhorukov, I. Szleifer,V. V. Tsukruk, M. Urban, F. Winnik, S. Zauscher,I. Luzinov and S. Minko, Nat. Mater., 2010, 9, 101113.

2 P.-G. de Gennes, M. Hebert and R. Kant, Macromol. Symp.,1997, 113, 3949.

3 M. H. Li, P. Keller, J. Yang and P. A. Albouy, Adv. Mater.,2004, 16, 19221925.

4 V. Q. Nguyen, A. S. Ahmed and R. V. Ramanujan, Adv.Mater., 2012, 24, 40414054.

5 C. S. Brazel, Pharm. Res., 2009, 26, 644656.6 Q. Dai and A. Nelson, Chem. Soc. Rev., 2010, 39, 40574066.7 S. F. Medeiros, A. M. Santos, H. Fessi and A. Elassari, Int. J.

Pharm., 2011, 403, 139161.8 M. Zrinyi, L. Barsi and A. Buki, J. Chem. Phys., 1996, 104,87508756.

9 J. A. Galicia, F. Cousin, E. Dubois, O. Sandre, V. Cabuil andR. Perzynski, Soft Matter, 2009, 5, 26142624.

10 P. C. Papaphilippou, A. Pourgouris, O. Marinica, A. Taculescu,G. I. Athanasopoulos, L. Vekas and T. Krasia-Christoforou,J. Magn. Magn. Mater., 2011, 323, 557563.

11 Z. Li, J. Shen, H. Ma, X. Lu, M. Shi, N. Li and M. Ye, Polym.Bull., 2012, 68, 11531169.

12 N. S. Satarkar and J. Zach Hilt, Acta Biomater., 2007, 4, 1116.13 S. Ghosh and T. Cai, J. Phys. D: Appl. Phys., 2010, 43, 415504.14 C. Bellan and G. Bossis, Int. J. Mod. Phys. B, 2002, 16, 24472453.15 G. V. Stepanov, D. Y. Borin, L. R. Yu, P. V. Melenev and

N. S. Perov, J. Phys.: Condens. Matter, 2008, 20, 204121.16 J. Li, M. Zhang, L. Wang, W. Li, P. Sheng and W. Wen,

Microfluid. Nanofluid., 2011, 10, 919925.17 B. A. Evans, B. L. Fiser, W. J. Prins, D. J. Rapp, A. R. Shields,

D. R. Glass and R. Superfine, J. Magn. Magn. Mater., 2012,324, 501507.

18 C. Sotebier, A. Michel and J. Fresnais, Appl. Sci., 2012, 2,485495.

19 J. A. Galicia, O. Sandre, F. Cousin, D. Guemghar, C. Menagerand V. Cabuil, J. Phys.: Condens. Matter, 2003, 15, S1379S1402.

20 B. W. Ahn and T. J. Kang, J. Appl. Polym. Sci., 2012, 125,15671575.

21 M. Yoonessi, J. A. Peck, J. L. Bail, R. B. Rogers, B. A. Lerch andM. A. Meador, ACS Appl. Mater. Interfaces, 2011, 3, 26862693.

22 D. K. Hwang, D. Dendukuri and P. S. Doyle, Lab Chip, 2008,8, 16401647.

23 K. P. Yuet, D. K. Hwang, R. Haghgooie and P. S. Doyle,Langmuir, 2009, 26, 42814287.

24 M. Zrinyi, L. Barsi and A. Buki, Polym. Gels Networks, 1997,5, 415427.

25 C.-H. Chen, A. R. Abate, D. Lee, E. M. Terentjev andD. A. Weitz, Adv. Mater., 2009, 21, 32013204.

26 J. Rodriguez-Lopez, H. C. Shum, L. Elvira, F. Montero deEspinosa and D. A. Weitz, J. Magn. Magn. Mater., 2013, 326,220224.

27 C. Herrmann, M. B. Bannwarth, K. Landfester andD. Crespy, Macromol. Chem. Phys., 2012, 213, 829838.

28 J. Fresnais, J. F. Berret, B. Frka-Petesic, O. Sandre andR. Perzynski, Adv. Mater., 2008, 20, 38773881.

29 M. Allione, B. Torre, A. Casu, A. Falqui, P. Piacenza, R. DiCorato, T. Pellegrino and A. Diaspro, J. Appl. Phys., 2011,110, 064907.

30 L. H. Reddy, J. L. Arias, J. Nicolas and P. Couvreur, Chem.Rev., 2012, 112, 58185878.

31 J. Cheng, B. A. Teply, S. Y. Jeong, C. H. Yim, D. Ho, I. Sherifi,S. Jon, O. C. Farokhzad, A. Khademhosseini andR. S. Langer, Pharm. Res., 2006, 23, 557564.

32 M. Chorny, I. Fishbein, B. B. Yellen, I. S. Alferiev, M. Bakay,S. Ganta, R. Adamo, M. Amiji, G. Friedman and R. J. Levy,Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 83468351.

33 J. L. Arias, L. H. Reddy, M. Othman, B. Gillet, D. Desmaele,F. Zouhiri, F. Dosio, R. Gref and P. Couvreur, ACS Nano,2011, 5, 15131521.

34 P. Dames, B. Gleich, A. Flemmer, K. Hajek, N. Seidl,F. Wiekhorst, D. Eberbeck, I. Bittmann, C. Bergemann,T. Weyh, L. Trahms, J. Rosenecker and C. Rudolph, Nat.Nanotechnol., 2007, 2, 495499.

35 B. Polyak, I. Fishbein, M. Chorny, I. Alferiev, D. Williams,B. Yellen, G. Friedman and R. J. Levy, Proc. Natl. Acad. Sci.U. S. A., 2008, 105, 698703.

36 J. N. Fass and D. J. Odde, Biophys. J., 2003, 85, 623636.37 F. Xu, J. H. Geiger, G. L. Baker and M. L. Bruening,

Langmuir, 2011, 27, 31063112.38 J. J. Lai, K. E. Nelson, M. A. Nash, A. S. Homan, P. Yager

and P. S. Stayton, Lab Chip, 2009, 9, 19972002.39 A. Ali-Cherif, S. Begolo, S. Descroix, J. L. Viovy and

L. Malaquin, Angew. Chem., Int. Ed., 2012, 51, 1076510769.40 J. D. Adams, U. Kim and H. T. Soh, Proc. Natl. Acad. Sci.

U. S. A., 2008, 105, 1816518170.41 V. Rocher, J. M. Siaugue, V. Cabuil and A. Bee, Water Res.,

2008, 42, 12901298.42 C. Sanson, O. Diou, J. Thevenot, E. Ibarboure, A. Soum,

A. Brulet, S. Miraux, E. Thiaudie`re, S. Tan, A. Brisson,V. Dupuis, O. Sandre and S. Lecommandoux, ACS Nano,2011, 5, 11221140.

43 S. Louguet, B. Rousseau, R. Epherre, N. Guidolin, G. Goglio,S. Mornet, E. Duguet, S. Lecommandoux and C. Schatz,Polym. Chem., 2012, 3, 14081417.

44 S.-H. Hu, B.-J. Liao, C.-S. Chiang, P.-J. Chen, I. W. Chen andS.-Y. Chen, Adv. Mater., 2012, 24, 36273632.


Publ

ished

on

02 M

ay 2

013.

Dow

nloa

ded

by U

NIV

ERSI

DA

D D

E ZA

RAG

OZA

on

10/0

6/20

13 0

8:38

:59.

View Article Online


45 S.-H. Hu, S.-Y. Chen and X. Gao, ACS Nano, 2012, 6, 25582565.46 W.-L. Chiang, C.-J. Ke, Z.-X. Liao, S.-Y. Chen, F.-R. Chen,

C.-Y. Tsai, Y. Xia and H.-W. Sung, Small, 2012, 8, 35843588.47 H. Oliveira, E. Perez-Andres, J. Thevenot, O. Sandre, E. Berra

and S. Lecommandoux, J. Controlled Release, 2013, http://dx.doi.org/10.1016/j.jconrel.2013.01.013.

48 N. S. Satarkar and J. Z. Hilt, J. Controlled Release, 2008, 130,246251.

49 T. Hoare, B. P. Timko, J. Santamaria, G. F. Goya, S. Irusta,S. Lau, C. F. Stefanescu, D. Lin, R. Langer and D. S. Kohane,Nano Lett., 2011, 11, 13951400.

50 A. M. Schmidt, Macromol. Rapid Commun., 2006, 27,11681172.

51 R. Mohr, K. Kratz, T. Weigel, M. Lucka-Gabor, M. Monekeand A. Lendlein, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,35403545.

52 U. N. Kumar, K. Kratz, W. Wagermaier, M. Behl andA. Lendlein, J. Mater. Chem., 2010, 20, 34043415.

53 U. N. Kumar, K. Kratz, M. Heuchel, M. Behl and A. Lendlein,Adv. Mater., 2011, 23, 41574162.

54 M. Y. Razzaq, M. Behl and A. Lendlein, Adv. Funct. Mater.,2012, 22, 184

Magnetic Responsive Polymer Composite Materials

Documents

Transcript of Magnetic Responsive Polymer Composite Materials