A controlled approach to iron oxide nanoparticles functionalization for magnetic polymer brushes

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

A controlled approach to iron oxide nanoparticles functionalizationfor magnetic polymer brushes

Francesco Galeotti ⇑, Fabio Bertini, Guido Scavia, Alberto BolognesiCNR-Istituto per lo Studio delle Macromolecole (ISMAC), via E. Bassini 15, 20133 Milano, Italy and Polo Scientifico e Tecnologico (PTS), CNR, via Fantoli 16/15, 20138 Milano, Italy

a r t i c l e i n f o

Article history:Received 8 March 2011Accepted 20 April 2011Available online 29 April 2011

Keywords:Magnetite nanoparticleSurface modificationPolymer brush

a b s t r a c t

In this article, we report a detailed study of surface modification of magnetite nanoparticles by means ofthree different grafting agents, functional for the preparation of magnetic polymer brushes. 3-Aminopro-pyltriethoxysilane (APTES), 3-chloropropyltriethoxysilane (CPTES), and 2-(4-chlorosulfonylphenyl)ethyl-trichlorosilane (CTCS) were chosen as grafting models through which a wide range of polymer brushescan be obtained. By means of accurate thermogravimetric analysis a good control over the amount ofimmobilized molecules is achieved, and optimal operating conditions for each grafting agent are conse-quently determined. Graft densities ranging from approximately 4 to 7 molecules per nm2 are obtained,depending on the conditions used. In addition, the surface-initiated atom transfer radical polymerization(ATRP) of methyl methacrylate (MMA) carried out with CTCS-coated nanoparticles is presented as anexample of polymer brushes, leading to a well-defined and dense polymeric coating of around 0.6 PMMAchains per nm2.

� 2011 Elsevier Inc. All rights reserved.

1. Introduction

In the last few years, magnetic nanoparticles (MNPs) haveattracted increasing interest thanks to their ability to react inresponse to an external magnetic field. Among their potentialapplications, the most promising ones are probably in the biomed-ical field, as contrast enhancement agents for magnetic resonanceimaging [1,2], as drug deliverers into specific targets [3], and aslocalized magnetic hyperthermia agents for cancer therapy [4].Beyond the biomedical area, interesting studies have been reportedon the use of MNPs for ferrofluids [5,6], data storage [7], and catal-ysis [8].

Since bare MNPs tend to aggregate into bigger clusters whichhave modified properties with respect to the single nanocrystals,a surface modification able to prevent self-aggregation is necessaryfor most of the applications. Their stabilization with hydrophilic orlipophilic coating allows to prepare MNP dispersions in a liquidmedium and to keep them stable for long time without any aggre-gation. In addition, the coating can be used to add specific func-tionalities to the MNPs, such as fluorescent dyes or biologicalmarkers. In these circumstances, polymers play an important roleboth as dispersants and as linkers between the MNP and the exter-nal function. The two main strategies that can be followed tomodify the MNP with a polymer brush coating are the grafting toand the grafting from approach. In the former, a pre-synthesized

polymer with a suitable end-group is attached to the MNP byspecific reaction between the end-group and a grafting agentanchored on the MNP. In the latter, the polymer chain is growndirectly from an initiator which is pre-grafted to the MNP surface.In either case, a fine control of the surface functionalization withthe grafting agents is essential.

Even though several studies on MNP surface modification aredescribed in the literature [9], finding the right working conditionsfor a specific grafting agent is far from easy. In fact, reported dataare generally affected by a wide variability, especially concerningreaction time and concentration that must be used to achieve thedesired extent of grafting.

We focused our attention on the study of three differentgrafting agents of common use for the surface modification ofFe3O4 MNPs, with the aim of obtaining well-controlled functional-ized nanocrystals. We have chosen 3-aminopropyltriethoxysilane(APTES), 3-chloropropyltriethoxysilane (CPTES), and 2-(4-chloro-sulfonylphenyl)ethyltrichlorosilane (CTCS) as three grafting mod-els through which a wide range of polymer brushes can beobtained. For all of them, the surface modification is based onthe affinity of silane groups for the hydroxyls present on Fe3O4

surface, leading to the formation of SiAO bonds and leaving theterminal functional groups available for further steps. In particular,APTES and CPTES are suitable precursors of clickable coatings,which can be used in a grafting to approach, after transformationinto azide of the amino and chloride groups, respectively [10,11].The sulfonyl chloride moiety of CTCS is a universal initiator forthe atom transfer radical polymerization (ATRP). It was reported

0021-9797/$ - see front matter � 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.jcis.2011.04.076

⇑ Corresponding author. Fax: +39 02 70636400.E-mail address: [email protected] (F. Galeotti).

Journal of Colloid and Interface Science 360 (2011) 540–547

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

www.elsevier .com/locate / jc is


to be particularly effective in the polymerization of styrene [12],methylacrylates [13] and acrylates [14]. Therefore, CTCS can easilygive rise to polymer brushes on the MNPs, in a typical grafting fromapproach [15–18].

To overcome the lack of reproducibility which often affects thiskind of studies, mostly due to the differences in shape and size ofMNPs obtained by different authors, we utilized commercial nano-crystals of naked Fe3O4. A drawback of naked MNPs is that theyeasily form aggregates, thus they need to be sonicated long to givea homogeneous dispersion in liquid medium. On the other hand,they are chemically ready-to-use since they have no ligands tobe removed/exchanged, consequently the grafting process is cleanand direct. Thermogravimetric analysis (TGA) allowed us to care-fully monitor the grafting process, so that well-controlled surfacegrafted MNPs were obtained. In this article we discuss the immo-bilization of APTES, CPTES, and CTCS on magnetite MNP surface,giving an outline of the results obtained in different conditions,in comparison with already published data. Results concerningthe growth of PMMA brushes from the same modified MNPs arealso reported. The report aims at providing a general and versatilemethod for the stabilization and functionalization of Fe3O4 MNPswith polymeric brushes.

2. Experimental

2.1. Materials

Fe3O4 MNPs (average size: 20 nm, surface area > 60 m2/g) werepurchased from Aldrich. APTES and CPTES were purchased fromAldrich. CTCS (50% solution in CH2Cl2) was purchased from ABCR.Methyl methacrylate (MMA) was purchased from Aldrich and dis-tilled under reduced pressure over CaH2 before use. Toluene wasdistilled over Na before use. CH2Cl2 and EtOH were purchased fromRiedel-de Haën and used as received. CuBr, p-toluenesulfonylchlo-ride (TsCl), 2,20-bipyridine (bpy), and HF aqueous (47%) solutionwere purchased from Aldrich and used as received.

2.2. Measurements

FTIR spectra were recorded by a Bruker TENSOR27 spectropho-tometer. Gel Permeation Chromatography (GPC) measurementswere carried out on a Waters SEC system equipped with a 2414RI and a 490 UV detectors, 2 PL gel Mix C columns, THF as solventand PMMA as reference. GC analysis were performed using aAgilent Technologies 6890 N GC System. Atomic force microscopy(AFM) investigations were performed using a NT-MDT NTEGRAapparatus in tapping mode under ambient conditions. Samplesfor AFM were prepared by casting a 0.5 mg/mL CHCl3 solution ontop of a Si substrate, hence the morphology of resulting film wasstudied. Transmission electron microscopy (TEM) measurementshave been performed with a Zeiss HTEM Libra200. Samples forTEM were prepared by casting a diluted (0.1 mg/mL) MNP solutionin CHCl3 on top of a TEM copper grid, so that the only materialattached to the grid edges was analyzed. TGA were carried outon a Perkin Elmer TGA-7 instrument at a scan rate of 20 �C/minin nitrogen atmosphere at a flow rate of 50 ml/min. TGA and deri-vate thermogravimetry (DTG) curves were recorded from 50 up to700 �C.

2.3. Surface modification on Fe3O4 MNPs

MNPs surface was modified by reaction of APTES, CPTES andCTCS with hydroxyl groups located on Fe3O4 external surface. Toevaluate the extent of functionalization the reaction wasperformed using different conditions, which are reported in Table1.

In a general procedure, 20 mg of naked magnetite MNPs wereput in a Schlenk flask containing 10 mL of the required solventand kept in ultrasonic bath for 30 min to obtain a homogenousdark solution. After this time, required amounts of APTES or CPTESwere added drop by drop through a microsyringe without stoppingthe sonication. CTCS was added as 50% CH2Cl2 solution in the sameway, but keeping the system under nitrogen atmosphere.

Table 1Amounts of grafting agents anchored on magnetite nanoparticle surface, determined by TGA.

Sample Solvent Grafting agent amount(mmol/g MNP)

Reaction conditions TGA weightloss (%)

Calculated grafting success(mmol/g MNP)

Calculated grafting density(molecules/nm2)

MNP@APTES-a-20

EtOH 20 2 h sonication 3.8 0.33 3.4

MNP@APTES-b-20

EtOH 20 2 h sonication + 6 h stirring 4.5 0.42 4.3

MNP@APTES-c-20

EtOH 20 2 h sonication + 6 hstirring + 10 h stirring at 60 �C

5.1 0.50 5.1

MNP@APTES-c-5


4.3 0.39 4.0

MNP@APTES-c-100


7.0 0.74 7.6

MNP@CPTES-a-20

CH2Cl2 20 2 h sonication 4.4 0.33 3.4

MNP@CPTES-b-20

CH2Cl2 20 2 h sonication + 6 h stirring 5.0 0.39 4.0

MNP@CPTES-c-20

CH2Cl2 20 2 h sonication + 6 hstirring + 10 h stirring at 60 �C

7.6 0.68 6.9

MNP@CPTES-c-5


5.4 0.43 4.4

MNP@CPTES-c-100


6.8 0.59 6.0

MNP@CTCS-a-5

Toluene 5 2 h sonication 11.6 0.52 5.3

MNP@CTCS-a-20


MNP@CTCS-a-100


F. Galeotti et al. / Journal of Colloid and Interface Science 360 (2011) 540–547 541


Sonication was continued for 2 h, and then a magnetic stirrer wasadded to the Schlenk flasks for further reaction times. At the end ofreaction, MNPs were washed by five cycles of precipitation onmagnet and dispersion with fresh solvent: APTES and CPTES withthe same solvent used for reaction, CTCS with THF. Finally, theMNPs were dried under reduced pressure and stored under nitro-gen for further use. The amount of grafting was calculated fromTGA data, according to the following equations:

grafted mmol per g MNP ¼W=ð1�WÞ �W0

M � 10�3 ð1Þ

nm2 per g MNP ¼ SA

V � d ¼3

10�21 � r � dð2Þ

grafted molecules per nm2 ¼ ½W=ð1�WÞ �W0� � r � d � 6:023 � 102

M � 3ð3Þ

where W is the weight loss, W0 is weight loss of naked MNPs, M isthe molecular weight of the anchored grafting agent, SA is the

Scheme 1. Reaction scheme and organosilane structures employed for the functionalization of magnetite nanoparticle surface.

Fig. 1. FTIR spectra of naked Fe3O4 nanoparticles and their derivatization with threedifferent grafting agents.

Fig. 2. TGA and DTG curves of naked Fe3O4 nanoparticles (straight line) andMNP@CTCS-a-20 (dashed line).

542 F. Galeotti et al. / Journal of Colloid and Interface Science 360 (2011) 540–547


surface area of the MNP (assuming average spherical shape), V is itsvolume, d is density of Fe3O4, r is MNP radius in nm [19].

2.4. ATRP of MMA

The polymerization of MMA was carried following typical ATRPconditions, using as initiator CTCS anchored to the MNPs. In ageneral procedure, 100 mg of Fe3O4@CTCS MNPs were placed in aSchlenk tube kept under nitrogen, to which 10 mL of dry toluenewas added. The dispersion was sonicated for 30 min, then 50 mgof TsCl, 40 mg of CuBr, 44 mg of bpy and 8 mL of MMA were addedto the tube. The system was degassed by three freeze–pump-thaw

cycles and then kept at 90 �C under vigorous stirring for therequired reaction time. Aliquots of the solution were taken outafter 3, 6, 9 and 24 h of reaction for GC and GPC analysis. The solu-tion was cooled to room temperature, exposed to air and clarifiedby attracting the MNPs on one side with a magnet. 3 mL of thisclear solution were then filtered to Al2O3 and mixed with 30 mLof MeOH for the precipitation of the non-grafted PMMA. Afterpaper filtration, the solution was analyzed by GC to estimate themonomer conversion, which was calculated, for every sample, asthe ratio between the area peak of MMA and of toluene, taken asinternal standard.

The MNPs were washed by five cycles of magnet precipitation/redispersion in fresh toluene and then dried under reduced pres-sure. To determine the molecular weight of the graft PMMA, thiswas cleaved from the MNPs by decomposition of Fe3O4 with 48%aqueous HF. At this aim, 20 mg of MNPs were placed in a polycar-bonate tube and dispersed in 2 mL of toluene by sonication. 2 mL ofHF solution was added and the two phases were vigorously stirredfor 4 h. After this time, the organic phase was washed with NaHCO3

aqueous solution and then with water. The polymer obtained byprecipitation from the extracted organic phase into MeOH wasanalyzed by GPC to determine the molecular weight and its distri-bution. The graft density of the PMMA brushes was determinedfrom TGA data, using the same equation presented above (3), wereW0 is the weight loss of the MNPs coated by the initiator and M isthe number average molecular weight (Mn) of the cleaved polymer.

3. Results and discussion

3.1. Immobilization of grafting agents on Fe3O4 MNPs

The grafting agents were anchored to the MNP by reaction ofthe organosilane with the hydroxyl groups which cover the exter-nal surface of Fe3O4, as shown in Scheme 1. We tested differentreaction conditions for each grafting agent in order to be able toidentify the most suitable. Results of functionalization are summa-rized in Table 1. We used a specific solvent for each compound,according to our experience and to what reported in the literature.In particular, EtOH was used for APTES, CH2Cl2 was used for CPTESand dry toluene for CTCS. To evaluate the influence of reaction timeon the grafting efficiency, the same amount (20 mmol per gram ofMNPs) of grafting agent was let react initially for 2 h in ultrasoundbath at r.t., then for another 6 h under magnetic stirring at r.t., andfinally for another 10 h under stirring at 60 �C. After each reactiontime, part of the suspension was withdrawn to be analyzed.

The immobilization of grafting agents on Fe3O4 surface was con-firmed by FTIR, as shown in Fig. 1. Strong absorptions due to FeAObonds are evident at 587–635 cm�1 in all the samples. Naked MNP

Fig. 3. Plots of graft amount (j) and graft yield (e) as a function of the initialconcentration of organosilane compound. Different plots for APTES (top), CPTES(middle) and CTCS (bottom) are shown.

Fig. 4. TEM micrography (a) and AFM height images (b and c) of naked Fe3O4 MNPs.



spectrum shows no other significant peak, while derivatized MNPsshow absorbance due to alkyl groups at 2850 (not shown in figure)and between 1400 and 1500 cm�1. In addition, a large band ob-served at 1000–1150 cm�1 for all derivatized samples, is attributedto SiAO, SiAOASi and FeAOASi vibrations, attesting the formationof silica layer that entraps the MNP. MNP@CTCS shows a diagnosticpeak at 1390 cm�1 which can be assigned to S@O stretching.

TGA studies were carried out for the magnetic MNPs after eachstep of modification. In Fig. 2 the thermograms obtained forMNP@CTCS-a-20 are reported as examples and compared withthe curves of the pristine Fe3O4 MNPs. The TGA curve of the nakedMNP exhibits a small weight loss (about 1%) in the temperaturerange 50–300 �C due to the removal of physically and chemicallyadsorbed water. The thermogram of the CTCS-derivative MNPs

Fig. 5. TEM and AFM images of functionalized Fe3O4 MNPs. (a) TEM and (b) AFM height images of MNP@APTES. (c) AFM height and (d) corresponding phase image ofMNP@CPTES. (e and f) TEM and (g) AFM height image of MNP@CTCS.



sample shows a broad degradation region between 100 and 650 �C,typical of APTES, CPTES and CTCS functionalized MNPs. The DTGpattern, constituted by overlapping decomposition events, indi-cates that the organic products are released in steps, which reflectthe different involved mechanisms. The first weight loss occurringat lower temperature (100–300 �C) is associated with the evolutionof adsorbed water. The second degradation stage takes place in thetemperature range of 300–650 �C and arises from the decomposi-tion of CTCS moieties anchored on MNPs; it is characterized by aremarkable weight loss (9.4%) with two DTG maxima at 455 and535 �C.

By performing a quantitative TGA investigation we were able toevaluate the grafting process and accurately determine the graftingdensity on MNP surface. In the calculation (Eqs. (2) and (3)) weconsidered a density for Fe3O4 of 5.1 g/cm3 and an average diame-ter of 20 nm.

Both for APTES and CPTES, TGA evidenced an increase ofgrafting after every reaction step, indicating that anchoring oftriethoxysilane agent to Fe3O4 surface is rather slow and requireslong reaction times. On the other end, TGA did not record anyincrease in weight loss for CTCS after the first 2 h of reaction (datanot shown), which is in agreement with the higher reactivity oftrichlorosilane agents towards AOH groups, with respect to trieth-oxysilanes. The maximum grafting density reached was similar forall the compounds: approximately 5, 6 and 7 molecules of APTES,CTCS and CPTES, respectively, were anchored in 1 nm2. Eventhough not many comparable data are available in the literatureon this point, these densities are relatively high comparing to sim-ilar studies reported. In particular, Garcia et al. reported aboutgrafting 1.5 molecules/nm2 of CTCS [16], Campelj et al. grafted1.1 molecules/nm2 of APTES [20], and Babu and Dhamodhranobtained 2.0 molecules/nm2 for a phosphonic acid terminatedATRP initiator [21]. This confirms that the optimization of reactionconditions can lead to an increase in graft density. On the otherhand, Sun et al. obtained 4.6 and 6.3 molecules/nm2 for methacryl-oxypropyl trimethoxysilane and a modified APTES, respectively,which is in agreement with our results [22,23].

For estimating the dependence of grafting success from concen-tration, we adjusted the amount of grafting agent in order to have5, 20 and 100 mmol per g of MNPs in the reaction mixture. Consid-ering an average grafted amount of 0.5 mmol/g, these quantitiescorrespond to an excess of 10, 40 and 200 folds, respectively. Theresults are reported in Table 1 and plotted in Fig. 3.

The graft yield, calculated as the ratio between the graftedamount and the concentration of grafting agent at the beginningof reaction, decreases for all samples with the increase of organos-ilane initial concentration. Reaction with APTES showed an almostlinear increase of grafting with concentration, up to the value of0.74 mmol/g, which is consistent with a grafting density of7.6 molecules per nm2. Reaction with CPTES gave the best resultat 20 mmol/g (0.68 mmol/g) while an increase in reagent concen-tration did not produce any further grafting. Reaction with CTCSprovided similar results at every concentration tested, indicatingthat a 10� excess is enough to achieve a graft amount around0.6 mmol/g.

3.2. Morphological investigations on differently silanized MNP

Both naked and grafted Fe3O4 MNPs were analyzed by TEM andAFM. For naked MNPs, TEM investigation showed that they aremainly faceted (squared and polygonal) with dimensions rangingfrom 15 to 35 nm (Fig. 4a). Cast deposition of these particles pro-duces a layer with average roughness of 1 nm and some agglomer-ates coming out of the surface, as shown by AFM images (Fig. 4band c). The effect of silanization with APTES and CPTES is theformation of particles with a more spherical shape and irregularedges, probably due to the grafting agent coating on the Fe3O4 core,as shown by TEM image of MNP@APTES (Fig. 5a). Cast deposition ofthese silanized particles produces a series of terraces whosecompactness is probably aided by the inter-particle interactionsinduced by the organic coating, as shown by the 8 � 8 lm AFMscan reported in Fig. 5b. Isolated clusters are distributed onto theseterraces (see arrow in Fig. 5b). The effect of the grafting seems tolead to a higher local MNP self-assembly with respect to the nakedones, as revealed by the height and phase AFM images of aMNP@CPTES cast layer (Fig. 5c–d). After treatment with CTCS onthe other hand, MNPs predominantly maintain the original shapeof the naked Fe3O4, although the organic coating seems to mostlyform an envelope over clusters of agglomerated MNPs than

Fig. 6. FTIR spectra of CTCS derivatized MNPs before (top) and after (bottom) graftpolymerization of MMA.

Fig. 7. Plots (from top to bottom) of polydispersity, graft density, and monomerconversion vs. time for the polymerization of MMA on CTCS-coated magnetitenanoparticles. The dashed lines are set on the average values. The solid line is aguide for eyes.



covering the single particles (Fig. 5e and f). The contrast betweenthe Fe3O4 crystalline core (double arrow) and the overlying CTCSlayer (single arrow) is shown in Fig. 5f. The AFM image (Fig. 5g)shows a less compact layer compared to the other silanized MNPs,looking more similar to the naked Fe3O4 case. This different behav-ior may be explained by the fact that less homogeneously coatedMNPs have a decreased tendency to the auto-organization.

3.3. ATRP of MMA initiated on MNP surface

The CTCS initiator coated MNPs were afterward subjected tocopper mediated ATRP of MMA. For this purpose, we usedMNP@CTCS-a-100, because this was the sample with the highestfunctionalization density that we could achieve. We carried outthe polymerization in the presence of toluene as solvent and usingthe sacrificial free initiator TsCl. Solvent polymerization was pre-ferred to bulk because we wanted to maintain a good dispersionof nanoparticles in the reaction medium and to achieve a relativelyslow polymer growth, so that low molecular weight control waspossible. The role of the sacrificial initiator is to adjust the concen-tration of CuII complex, which is otherwise too low to reversiblydeactivate the growing polymer radical with a sufficiently highrate. For an exhaustive description of the so-called persistent rad-ical effect, see the work of Ejaz et al. [24,25]. The free polymer pro-duced was then easily separated through a few cycles of magnetprecipitation/redispersion. Bpy was preferred to pentamethyldi-ethylenetriamine (PMDETA) as ligand for Cu complex, given thatinefficient initiation problems related to the TsCl/CuBr/PMDETAsystem were previously reported [26]. The monomer/metal/ligandratio was fixed at 1:1:1 while the ratio initiator/TsCl was approxi-mately 1:10.

A comparison of FTIR spectra of MNPs before and after polymer-ization is shown in Fig. 6. The MNP@PMMA spectrum clearlyshows a new absorption peak centered at 1730 cm�1, characteristicof PMMA ester group, which proves the success of the polymeriza-tion. Fig. 7 displays the polymerization evolution within the time,through plots of data obtained from TGA, GC and GPC analysis. Themonomer conversion plot can be approximated by a straight lineup to 9 h of reaction, indicating that the concentration of propaga-tion species is kept constant. After that, the kinetic changes, prob-ably due to the reactivity decrease of the catalyst and to theincrease of system viscosity [16,18]. Nevertheless, the MW growsproportionally to the monomer conversion (Fig. 8) while polydis-persity (Fig. 7, top) remains around 1.25, which confirms that thegraft polymerization proceeds in a living fashion. In particular,we could achieve a good control of the MW in the region between9 and 14 KDa. Since the longer is the graft polymer chain, the thick-er is the coating which shields the Fe3O4 MNPs, the tuning of poly-

mer brush length in the low molecular weights interval is crucial toproduce well dispersible MNPs that retain their magnetic proper-ties. The polymer graft density, determined from TGA weight lossafter subtracting the contribution of bonded initiator and the Mn

of the grafted polymer chains (Eq. (3)), was nearly constant forall the samples, with an average value of 0.55 chains/nm2. Sincethe graft density of CTCS was 6.9 molecules/nm2, approximately10% of initiator molecules on the magnetite surface underwent topolymerization.

4. Conclusions

The surface functionalization of Fe3O4 MNPs with APTES, CPTESand CTCS was performed in different reaction conditions and graftdensities ranging from approximately 4 to 7 molecules per nm2

were obtained. TGA analysis was used to control the grafting effi-ciency, hence to adjust the operating conditions. Both APTES andCPTES surface functionalization are much more efficient when longreaction time is used and a temperature of 60 �C is applied. ForCTCS, on the other hand, reaction with Fe3O4 surface is very fast.While with CPTES and CTCS the maximum grafting yield is reachedwhen a 40-fold excess is used, APTES silanization increases almostlinearly up to 200-fold excess. By means of AFM and TEM morpho-logical analysis, the Fe3O4 surface was studied, revealing someessential differences between the differently silanized MNPs. Inparticular APTES and CPTES treated MNPs show a fairly homoge-neous coating while CTCS functionalization appears less uniformlydistributed and some enveloped clusters are visible.

To demonstrate that our functionalized NMPs are suitable forthe preparation of magnetic polymer brushes, PMMA chains weregrown in a living fashion through surface-initiated ATRP fromCTCS-coated nanoparticles, leading to a well-defined and densepolymeric coating of around 0.6 chains per nm2. Given the wideuse of particle surface modification as initial step for their applica-tion as functional material, the present study aims at providinguseful information for the optimization of this kind of approach.

Acknowledgments

This work is financially supported by Regione Lombardia, agree-ment Regione/CNR, project 4: ‘‘Nanoscienze per materiali e applic-azioni biomediche’’. We are grateful to Mr Daniele Piovani (ISMAC-CNR) for the GPC analysis.

References

[1] D.E. Sosnovik, R. Weissleder, Curr. Opin. Biotechnol. 18 (2007) 4.[2] R. Qiao, C. Yang, M. Gao, J. Mater. Chem. 19 (2009) 6274.[3] W. Zhao, H. Chen, Y. Li, L. Li, M. Lang, J. Shi, Adv. Funct. Mater. 18 (2008) 2780.[4] S. Mornet, S. Vasseur, F. Grasset, E. Duguet, J. Mater. Chem. 14 (2004) 2161.[5] P. Bian, T.J. McCarthy, Langmuir 26 (2010) 6145.[6] H. Xia, J. Wang, Y. Tian, Q.-D. Chen, X.-B. Du, Y.-L. Zhang, Y. He, H.-B. Sun, Adv.

Mater. 22 (2010) 3204.[7] Y. Kim, C. Lee, I. Shim, D. Wang, J. Cho, Adv. Mater. 22 (2010) 5140.[8] C.G. Tan, R.N. Grass, Chem. Commun. (2008) 4297.[9] R.l. Barbey, L. Lavanant, D. Paripovic, N. Schüwer, C. Sugnaux, S. Tugulu, H.-A.

Klok, Chem. Rev. 109 (2009) 5437.[10] L. Polito, D. Monti, E. Caneva, E. Delnevo, G. Russo, D. Prosperi, Chem. Commun.

(2008) 621.[11] A. Schlossbauer, D. Schaffert, J. Kecht, E. Wagner, T. Bein, J. Am. Chem. Soc. 130

(2008) 12558.[12] V. Percec, B. Barboiu, Macromolecules 28 (1995) 7970.[13] T. Grimaud, K. Matyjaszewski, Macromolecules 30 (1997) 2216.[14] V. Percec, B. Barboiu, H.J. Kim, J. Am. Chem. Soc. 120 (1998) 305.[15] I. Garcia, A. Tercjak, N.E. Zafeiropoulos, M. Stamm, I. Mondragon, J. Polym. Sci.

Part A: Polym. Chem. 45 (2007) 4744.[16] I. Garcia, N.E. Zafeiropoulos, A. Janke, A. Tercjak, A. Eceiza, M. Stamm, I.

Mondragon, J. Polym. Sci. Part A: Polym. Chem. 45 (2007) 925.[17] E. Marutani, S. Yamamoto, T. Ninjbadgar, Y. Tsujii, T. Fukuda, M. Takano,

Polymer 45 (2004) 2231.[18] K. Ohno, T. Morinaga, K. Koh, Y. Tsujii, T. Fukuda, Macromolecules 38 (2005)

2137.

Fig. 8. Plot of MW of graft polymer vs. the monomer conversion. The solid line is aguide for eyes.



[19] C. Bartholome, E. Beyou, E. Bourgeat-Lami, P. Chaumont, N. Zydowicz,Macromolecules 36 (2003) 7946.

[20] S. Campelj, D. Makovec, M. Drofenik, J. Magn. Magn. Mater. 321 (2009) 1346.[21] K. Babu, R. Dhamodharan, Nanoscale Res. Lett. 3 (2008) 109.[22] Y.B. Sun, X.B. Ding, Z.H. Zheng, X. Cheng, X.H. Hu, Y.X. Peng, Chem. Commun.

(2006) 2765.

[23] Y.B. Sun, X.B. Ding, Z.H. Zheng, X. Cheng, X.H. Hu, Y.X. Peng, Eur. Polym. J. 43(2007) 762.

[24] M. Ejaz, S. Yamamoto, K. Ohno, Y. Tsujii, T. Fukuda, Macromolecules 31 (1998)5934.

[25] M. Ejaz, K. Ohno, Y. Tsujii, T. Fukuda, Macromolecules 33 (2000) 2870.[26] P.A. Gurr, M.F. Mills, G.G. Qiao, D.H. Solomon, Polymer 46 (2005) 2097.


Date post:	28-Apr-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

A controlled approach to iron oxide nanoparticles functionalization for magnetic polymer brushes

Documents