11940 DOI: 10.1021/la9016382 Langmuir 2009, 25(19), 11940–11946Published on Web 07/02/2009

pubs.acs.org/Langmuir

© 2009 American Chemical Society

Osmosis Based Method Drives the Self-Assembly of Polymeric

Chains into Micro- and Nanostructures

Laura Chronopoulou, Ilaria Fratoddi, Cleofe Palocci, Iole Venditti, and Maria V. Russo*

Department of Chemistry, University of Rome “La Sapienza”, P.le A. Moro 5, Rome 00185, Italy

Received May 8, 2009

Polymers derived from monomers with a variety of functionalities provide materials with a vast range of propertiesand applications. Worldwide research has recently developed a wide number of methods suitable for the preparation ofpolymeric materials of nanometric dimensions, in view of the fact that, at the nanoscale level, new and unexpectedproperties emerge and lead to innovative applications. In this framework, we have exploited an easy method for thegeneration of nanostructures, regardless of the chemical structure of the pristine amorphous polymers, that is,biopolymers (e.g., polysaccharides) and synthetic, functional, and structural polymers (i.e, polystyrene, polymethyl-methacrylates, polyacetylenes, and polymetallaynes). The nanostructure of these macromolecules, considered as theprototypes of various classes of polymeric materials, was achieved by using a simple and versatile procedure based on anosmotic method (OBM). Depending on the choice of solvent/nonsolvent pairs, the dialysis membrane molecular weightcutoff (MWCO), temperature, and polymer concentration, different morphologies can be obtained (e.g., spheres,sponges, disks, and fibers); also, a tuning of the nanoparticle dimensions ranging from the micro- to nanoscale has beenobtained.

Introduction

Materials with controllable dimensions and shape at thesub-micrometric scale are opening new perspectives in manyfields of science and technology.1 In particular, polymericmaterials, due to their peculiar characteristics, are of generalapplication in different areas. Synthetic polymers combinelightness, low cost, ease of processing, and functionalizationwith the characteristic properties of inorganic materials, forexample, conductivity,2,3 light propagation,4 linear and non-linear optical behavior,5 and sensitivity to chemical agents.6 Onthe other hand, biopolymers, in view of their biocompatibilityand biodegradability, are largely employed in biomedicine andpharmacy, for example, tissue engineering,7 diagnostics, anddrug delivery,8,9 in the food industry, agriculture, textiles, and inthe chemical and packaging industries. Such properties areexpected to be enhanced in nanostructured materials. In thiscontext, a simple strategy for nanostructured polymer produc-tion is desirable.

Nanostructured synthetic polymers are generally obtainable bya variety of chemical approaches, starting from themonomer thatis induced to growpolymeric chains which then self-assemble intonanostructured shapes. Several significant examples are brieflyoutlined hereafter.

Considering a widely studied polymer, polyaniline, nanotubes,nanofibers, nanorods, and hollow microspheres have beensynthesized by electrochemical routes, using acids as oxidants,with the aid of templates and by self-assembly approaches.10-17

Emulsion polymerization is one of the most common methodsthat allow the preparation of nanosized polymer particles, whoseprinciples concerning kinetics andmechanisms have been recentlyreported.11,12 This method usually requires the use of surfactantswhichmight constitute anadditional cost and pollute the product,though the development of controlled radical polymerizationmethods and surfactant-free emulsion polymerization is gradu-ally overcoming this issue.13,14 A different approach is based onmicrowave preparation of cross-linked nanopolymers,15 andanother simple method that uses routine laboratory chemicalsand equipment allows one to generate precisely shaped poly-styrene and polyvinyl alcohol micro- and nanoparticles.16

Elecrospinning is also a widely used and versatile techniquefor the preparation of polymeric nanofibers, which are suitablefor the incorporation of biological molecules, cells, or othernanoparticles.17

Moreover, recent approaches use self-assembly at liquid-liquid interfaces to drive nanoparticle formation, opening tonew industrial applications. For example, self-assembled nano-particles at a liquid-liquid interface serve as building blocks for

*Corresponding author. Telephone: þ390649913349. Fax: þ3906490324.E-mail: mariavittoria.russo@uniroma1.it.(1) Satyanarayana, V. N. T.; Kuchibhatla, A. S. K.; Debasis, B.; Seal, S. Prog.

Mater. Sci. 2007, 52, 699–913.(2) Lee, K.; Cho, S.; Park, S. H.; Heeger, A. J.; Lee, C.-W.; Lee, S.-H. Nature

2006, 441, 65–68.(3) Salleo, A. Mater. Today 2007, 10, 38–45.(4) Pursiainen, O. L.; Baumberg, J. J.; Winkler, H.; Viel, B.; Spahn, P.; Ruhl, T.

Opt. Express 2007, 15, 9553–9561.(5) Kippelen, B. Springer Ser. Opt. Sci. 2007, 114, 487–534.(6) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. Rev. 2000, 100,

2537–2574.(7) Biondi, M.; Ungano, F.; Quaglia, F.; Netti, P. A. Adv. Drug Delivery Rev.

2008, 60, 229–242.(8) Heath, F.; Haria, P.; Alexander, C. AAPS J. 2007, 9, E234–240.(9) Payne, G. F. Curr. Opin. Chem. Biol. 2007, 11(2), 214–219.

(10) Wan, M. X. In Encyclopedia of Nanoscience and Nanotechnology;Nalwa, H. S., Ed.; American Scientific Publishers: Los Angeles, 2004; Vol. 2,pp 153-169.

(11) Thickett, S. C.; Gilbert, R. G. Polymer 2007, 48, 6965–6991.(12) Capek, I. Adv. Colloid Interface Sci. 2002, 99, 77–162.(13) Voccia, S.; Ignatova, M.; Jerome, P.; Jerome, C. Langmuir 2006, 22,

8607–8613.(14) Ngai, T.; Wu, C. Langmuir 2005, 21, 8500–8525.(15) An, Z.; Tang,W.;Hawker, C. J.; Stucky,G.D. J. Am.Chem. Soc. 2006, 128,

15054–15055.(16) Champion, J. A.; Katare, Y.K.;Mitragotri, S.Proc. Natl. Acad. Sci. U.S.A.

2007, 104, 11901–11904.(17) Ramakrishna, S.; Fujihara, K.; Teo, W.-E.; Lim, T.-C.; Ma, Z. An

Introduction to Electrospinning and Nanofibers; World Scientific Publishing Co.Pte. Ltd.: Singapore, 2005.

DOI: 10.1021/la9016382 11941Langmuir 2009, 25(19), 11940–11946

Chronopoulou et al. Article

the achievement of new functional materials with unique physicalproperties.18-20

Owing to their peculiar structures, most nanostructured bio-polymers cannot be produced by “bottom-up” procedures. Inthose cases, a “top-down” approach must be used, starting fromthe macrometric polymer itself. The bioadhesive and self-assem-bling abilities of biopolymericmaterials are exploited in a versatileapproach for the nanofabrication of different biohybrids basedon interactions between inorganic and biomaterials, for example,silica and biopolymers.21,22 Methods for preparing polysacchar-ide basedmicro- and nanoparticles rely ona variety of approachesincluding interactions with counterions, chemical cross-linking,solvent evaporation, coating on preformed microparticles, andspray-drying.23

In the present work,we have focused our research on a range ofsynthetic and natural, functional, and structural polymers and we

have developed an osmosis based method (OBM)24 that is simpleand versatile for the control of both dimension and morphologyof polymeric materials at the sub-micrometric scale via self-assembly. This method is of general application and proved tobe suitable for both synthetic and biopolymers.

Experimental Section

Materials. Dialysis cellulose membranes (cellulose acetate,33 � 21 mm, MWCO = 12 and 2 KDa) were purchased fromAldrich. Reagent grade solvents (DMF, DMSO, EtOH, MeOH,CHCl3, THF) were used without further purification. PMMAand PS were purchased from Aldrich. Amorphous PPA wasprepared and characterized according to our previous studies25

aswell as amorphousPDMPAandPDMPAHCl,26 andPt-DEBPand Pd-DEBP.27 The biopolymers dextran (DX), pullulan (PU),and chitosan (CS) were purchased from Sigma-Aldrich. Hyaluro-nic acid derivatives, HYAFF 11 p100 and HYAFF 9 p75, wereprovided by Fidia pharmaceuticals. See Chart 1 for structures ofsynthetic and biopolymers.

OBM Method. The polymer under test was dissolved in asuitable solvent (the concentration was chosen depending uponthe desired kind of nanomorphology), and the solution (usually

Chart 1. Chemical Structures of Synthetic and Biopolymers: Polymethylmethacrylate (PMMA), Polystyrene (PS), Polyacetylenes (PA: PPA,

PDMPA, PDMPAHCl), Polymetallaynes (MPy: Pt-DEBP, Pd-DEBP), Hyaluronic Acid Derivatives (HYAFFs: HYAFF9, HYAFF11),

Chitosan (CS), Pullulan (PU), and Dextran (DX)

(18) Chen, H.; Dong, S. Langmuir 2007, 23, 12503–12507.(19) Deng, Z.; Peng, B.; Chen, D.; Tang, F.; Muscat, A. J. Langmuir 2008, 24,

11089–11095.(20) Luo, M.; Mazyar, O. A.; Zhu, Q.; Vaughn, M. W.; Hase, W. L.; Dai, L. L.

Langmuir 2006, 22, 6385–6390.(21) Masciangioli, T.; Zhang,W.X.Environ. Sci. Technol. 2003, 37, 102A–108A.(22) Palocci, C.; La Grotta, A.; Barbetta, A.; Dentini, M. Langmuir 2007, 23,

8243–8245.(23) Sinha, V. R.; Singla, A. K.; Wadhawan, S.; Kaushik, R.; Kumria, R.;

Bansal, K.; Dhawan, S. Int. J. Pharm. 2004, 274, 1–33.(24) Palocci, C.; Russo, M. V.; Belsito, C. M. A.; Cernia, E.; D’Amato, R.;

Fratoddi, I.; Panzavolta, F.; Soro, S.; Venditti, I. PCT/IT20057000653 Interna-tional Publication Number WO 2006-051572 A3.

(25) Furlani, A.; Napoletano, C.; Russo, M. V.; Camus, A.; Marsich, N. J.Polym. Sci., Part A: Polym. Chem. 1989, 27, 75–86.

(26) Russo, M. V.; Furlani, A.; Polzonetti, G.; Altamura, P.; Fratoddi, I.Polymer 1997, 38, 3677–3690.

(27) Fratoddi, I.; Gohlke, C.; Cametti, C.; Diociaiuti, M.; Russo,M. V.Polymer2008, 49, 3211–3216.

11942 DOI: 10.1021/la9016382 Langmuir 2009, 25(19), 11940–11946

Article Chronopoulou et al.

10 mL) was transferred into a dialysis bag and further immer-sed into the selected nonsolvent (usually 200 mL). A schematicrepresentation of the process is shown in Scheme 1. The solventand the nonsolvent must be miscible, and a nominal v/v solvent/nonsolvent volume ratio=1/20was chosen for the application ofthe OBM method. The procedure was carried out at constanttemperature (T=4 �C or T=25 �Cwas adopted). Precipitationof the nanoparticles occurs at times that are dependent on theexperimental conditions (nature of the polymer, solvent/nonsol-vent pair, temperature,MWCOof themembrane). After 72-96 h(osmotic equilibrium was reached), the polymer suspension wasrecovered, centrifuged twice after suspension into a nonsolvent,and cast deposited onto a corn glass substrate for the SEMinvestigation.

Procedure for the Production of Different Morphologies

of Polymers. PMMA, PS, PPA, PDMPA, PDMPAHCl,Pt-DEBP, Pd-DEBP, and HA derivatives (0.1-25 mg/mL) weredissolved in an organic solvent; DX and PU (0,1-1 mg/mL) weredissolved in distilled water (pH = 6.5), and CS (0.1 mg/mL)was dissolved in distilled water acidified with acetic acid (pH =5.5). The dialysis bag (cellulose acetate, MWCO = 12000 and2000 Da) containing the polymer solution was then immersed inthe selected nonsolvent (solvent/nonsolvent ratio, v/v 1/20) untilthermodynamic equilibrium was reached (48-72 h). The precipi-tated polymers were recovered, washed several times with thenonsolvent, and freeze-dried. GPC and 1H NMR measurementswere performed on synthetic polymers before and after beingprocessed by OBM, in order to verify the polymers’ structuralintegrity (data shown in the Supporting Information). 1H NMRmeasurements were also performed on biopolymers, proving thatOBM does not involve any structural changes in the materials.

Instruments. A Leo 1450 instrument was used for scanningelectron microscopy (SEM) on polymeric films cast or spindeposited from different solvents (H2O, CHCl3, DMF, toluene,EtOH) on gold or glass substrates. Evaporation rates werecontrolledbycarryingout depositions of the samples in controlledtemperature and relative humidity chambers.

The particle size distribution is expressed by the ratio Dw/Dn,namely, the polydispersity index (PI). Dw and Dn are the weight

average diameter and the number average diameter of particles,respectively, and they were calculated on 100 measurementsperformed on the same sample with the aid of the Scion imagesoftware directly on the SEM images.28

Results and Discussion

OBM: General Features. The OBM is based on the use of aphysical barrier, specifically dialysis membranes or commonsemipermeable membranes that allow the passive transport ofsolvents to slow down the mixing of a polymer solution with anonsolvent; the dialysis membrane contains the solution of thepolymer. The gradual mixing of the solvent and the nonsolventinside the bag causes the mixture to be progressively less able todissolve the polymer. The consequent increase in interfacialtension drives the molecules of the polymer to aggregate intospheroidal particles; their resultant shape allows a minimizationin energy of the system, and hence, a precipitation occurs. In somecases, nanostructured rods and fibers are obtained depending onthe physicochemical conditions of the system.

The advantages of the method, in addition to low cost andgeneral applicability, are manifold: the procedure is conductedunder mild conditions, avoiding the use of emulsifiers andcostabilizers, pure products are obtained, and solvents are easilyrecovered by distillation. After centrifugation and subsequentdrying of the polymer suspension, polymer yields are almostquantitative. Moreover, the method permits the nanoparticleformation simultaneously with the bioactive molecule immo-bilization.29 For example, molecules with biological activity(enzymes, nucleic acids, drugs, etc.) are expected to be linked tonanostructured polymers either with noncovalent or with cova-lent bonds, depending on thematching of the functional groups ofthe biomolecule with the selected polymer.30

The speed of solvent mixing could be one of the factors thatmodulate the product morphology as a function of severalparameters, among them the solvent/nonsolvent pair, the dialysismembrane molecular weight cutoff (MWCO), the temperature atwhich the process is carried out, and the polymer concentration.According to well established statements reported in the litera-ture, themorphology of the particles is then determined by kineticand thermodynamic factors.31 From a thermodynamic view-point, themost important parameter is the free interface energy.32

When the precipitation process is slow enough, themorphology isnot influenced by the kinetic parameters and the so-calledequilibrium morphology is reached, corresponding to the mini-mization of the free energy of the system. The kinetic parametersmainly affecting the particlemorphology are the diffusion and thephase rearrangement inside the particles themselves. Themobilityof polymeric chains is restricted during their aggregation, andhence, phase separation and rearrangement can be slower thanthe aggregation rate. When kinetic factors prevail over thermo-dynamic ones, nonequilibrium morphologies are obtained.The morphologies obtained by our OBM can be modulated ina variety of shapes: spongy, tubular rods (nonequilibrium mor-phologies) and spheres with dimensions ranging betweenmicrometers and nanometers, depending on the experimental

Scheme 1. Schematic Representation of the OBM Method

(28) Chen, C.-W.; Serizawa, T.; Akashi, M. Chem. Mater. 1999, 11, 1381–1389.

(29) Masotti, A.; Bordi, F.; Ortaggi, G.;Marino, F.; Palocci, C.Nanotechnology2008, 19, 055302–055307.

(30) Palocci, C.; Chronopoulou, L.; Venditti, I.; Cernia, E.; Diociaiuti, M.;Fratoddi, I.; Russo, M. V. Biomacromolecules 2007, 8, 3047–3053.

(31) Stubbs, J. M.; Sundberg, D. C. J. Appl. Polym. Sci. 2004, 9, 1538–1551.(32) Evans, D. F.; Wennerstrom, H. The Colloidal Domain: Where Physics,

Chemistry, Biology and Technology Meet; Wiley-VCH: New York, 1999; pp1-632.

DOI: 10.1021/la9016382 11943Langmuir 2009, 25(19), 11940–11946

Chronopoulou et al. Article

conditions. Typical examples of micro/nanomorphologies ob-tained in this work with the OBM procedure are reported inFigure 1 with the appropriate data collected in Table 1.

An experimental study has been carried out on selectedmacromolecules in an attempt to rationalize the nanoparticleformation as a function of polymer concentration, temperature,

Figure 1. SEM images of morphologies obtained for synthetic polymers (A, PMMA; B, PS; C, PPA; D, PDMPAHCl; E, PDMPA; F,Pt-DEBP) and biopolymers (G, CS; H, I, L, M, HYAFF9; N, DX). For experimental conditions, see Table 1.

Table 1. OBM Experimental Conditions for the Achievement of Natural and Synthetic Polymers Nano- and Microstructures

polymer or biopolymer SEM image T (�C) solvent pair conc [mg/mL] morphology mean particle size (nm) PIa

PMMA A 4 acetone/H2O 5 spheres 200-2000 1.13PS B 25 DMF/H2O 5 sponge 500PPA C 25 DMF/H2O 10 spheres 100-300 1.11PDMPAHCl D 25 THF/H2O 1 fibers 1000-3000; L = 10 μmb

PDMPA E 25 THF/H2O 1 spheres 800-1600 1.92Pt-DEBP F 25 CHCl3/EtOH 1 spheres 200-400 1.88CS G 4 CH3CN/H2O 0.5 disks 2000-3000HYAFF9 H 4 H2O/EtOH 0.5 flowerlike 2000-3000HYAFF9 I 4 H2O /MeOH 0.5 spheres 500-600 1.00HYAFF9 L 4 DMSO/MeOH 0.5 spheres 300-400 1.50HYAFF9 M 4 H2O/acetone 0.5 dandelion-like 8000-10000DX N 4 H2O/EtOH 0.5 spheres 200-1000 1.14

aPI = polydispersity index, calculated for spherical particles as in the Experimental Section. b L = fiber length.

11944 DOI: 10.1021/la9016382 Langmuir 2009, 25(19), 11940–11946

Article Chronopoulou et al.

solvent/nonsolvent pairs, and molecular weight membranecutoff.Effect of Concentration and Temperature on Nanopar-

ticle Size. We have earlier studied the emulsion polymerizationof PMMA leading to nanospheres with diameters in the range100-400 nmandvery lowpolydispersion.33 In thepresentwork, asystematic study of the experimental conditions for the formationof nanospheres by a different route was carried out. By using theOBM procedure, a linear correlation between polymer concen-tration and the diameter of the nanospheres was observed in therange 100-1000 nm (see Figure 2A and B).

Figure 2. Diameter-concentration profiles for spheres of poly-mers byOBMprocedure. (A) PMMA: solvent=acetone,T=4 �C ;(B) PMMA: solvent=DMF, T=25 �C; (C) PPA: solvent=DMF,T=4and25 �C.Errorbarsare calculatedas the product ofPI� 100for a graphical representation.

Figure 3. SEM images of sub-micrometric structures of Pd-DEBP(experimental conditions: CHCl3/MeOH 1 mg/mL, T= 4 �C).

Figure 4. SEM images of nanostructured PS obtained in differentexperimental conditions: (A) DMF/MeOH, 1.5 mg/mL, T =25 �C; (B) DMF/MeOH, 1.5 mg/mL, T=4 �C; (C) DMF/MeOH,3 mg/mL, T=4 �C.

(33) D’Amato, R.; Venditti, I.; Russo,M. V.; Falconieri, M. J. Appl. Polym. Sci.2006, 102, 4493–4499.

DOI: 10.1021/la9016382 11945Langmuir 2009, 25(19), 11940–11946

Chronopoulou et al. Article

In the case of PPA, the diameter of the nanospheres versusconcentration is shown in Figure 2C, in a concentration rangelower with respect to PMMA, due to lower solubility of PPA.

At low concentrations (1 mg/mL or less), lower temperatureleads to the formation of nanoparticles with smaller diametersthan those obtained at room temperature. This is probably due tothe formation of many nucleation centers in a fast precipitationprocess that occurs at higher temperatures. At higher concentra-tions, the influence of the temperature on the dimensions seems tobe less effective.

Nanosized Pt-DEBP spheres could also be formed by thestraightforward osmosis procedure (seeFigure 1F)with diametersin the range 200-600 nm, using as solvent pairCHCl3/EtOH.Theformation of spongelike structures was observed by using higherpolymer concentrations; superimposed structures based onspheres with a diameter of about 60 nmwere obtained by varyingthe metal center, that is, Pd(II) in the polymetallayne chemicalstructure (Figure 3).

In conclusion, by varying the experimental conditions such astemperature and concentration, parameters that influence theprecipitation rate, different morphologies, and size modulationcan be obtained. For example, in the case of PS, if all theparameters are kept constant, a temperature decrease induces

a decrease of the nanosphere diameter (see Figure 4A, B); whenthe concentration increases, a growth in diameter is observed(see Figure 4B, C).

Natural biopolymers (PU, DX, HYAFFs) were submitted toOBM at 4 and 25 �C using a fixed polymer concentration.Amorphous morphologies were obtained when the polymerswere processed at 25 �C, whereas at 4 �C different nanostructureswere observed (see Figure 1M and N for DX and HYAFF9,respectively).Effect of the Solvent/Nonsolvent Pair. Since the precipita-

tion depends, among other parameters, on the chemical pro-perties of the solvent/nonsolvent pair, a correlation betweenmorphology and solvents Δε (difference between the dielec-tric constant values of the solvent/nonsolvent pair) was studied.While the concentration mainly tunes the size of the spheres,as described in the previous paragraph, a systematic varia-tion of Δε affects both shape and size, and structures ofvarious morphologic features can be achieved. Consider-ing a completely hydrophobic and apolar polymer such asπ-conjugated PPA, a decrease of Δε leads to the formation ofbig spheres (Figure 5A, B), whereas an increase of this para-meter induces the formation of spongelike nanostructures.(Figure 5C and D).

Figure 5. Schematic representation of the dependence of the PPA morphology from Δε of the solvent pair at T=25 �C: (A) DMF/hexane,Δε=34, c=5 mg/mL; (B) DMF/H2O, Δε=42, c=10 mg/mL; (C) acetone/H2O, Δε=58, c=0.5 mg/mL; (D) THF/H2O, Δε=72,c=0.5 mg/mL (dimension bar=1 μm).

Figure 6. Schematic representation of the dependence of the morphology from Δε of the solvent pair. (Top) HYAFF 11, c=0. 5 mg/mL,T= 4 �C: (A) DMSO/MeOH, Δε= 14.2; (B) DMSO/EtOH, Δε= 22.9; (C) DMSO/H2O, Δε= 32.8 (dimension bar: A= 2 μm;B=200 nm ; C=1 μm). (Bottom) HYAFF 9, c=0. 5 mg/mL, T=4 �C: (D) DMSO/MeOH, Δε=14.2; (E) DMSO/EtOH, Δε=22.9;(F) H2O/MeOH, Δε=47.

11946 DOI: 10.1021/la9016382 Langmuir 2009, 25(19), 11940–11946

Article Chronopoulou et al.

Applying the OBM procedure and varying the solvent/non-solvent pair (CHCl3/hexane Δε = 2.9 or THF/H2O Δε = 72.5)for PDMPA, a conjugated polymer containing polar pendantgroups with different morphologies was also obtained. Sub-micrometric grains change their morphology, and spheres areobtained by increasing the Δε between solvent and nonsolventpairs (images in the Supporting Information).

The effect of the variation of Δε was then extended tobiopolymers. Hydrophilic and polar HA derivatives, (HYAFF11and HYAFF9) were investigated, and their behavior is reportedin Figure 6, where a variation of the sub-micrometric structurecan be observed: increasing the value of Δε, the formation ofspheres (micro and nano) is favored for both biopolymers.

CS was processed with OBM employing different solvent/nonsolvent pairs: H2O/EtOH (Δε = 55.7), H2O/acetone (Δε =59.3), and H2O/THF (Δε = 72.48); the images of the resultingmicro/nanostructures are not reported; as observed in the case ofHYAFFs, and in general for polymers with polar groups, also forCS, the increase of the Δε leads to the preferential formation ofspheres when THF is used as the nonsolvent (increase of Δε).

The above-reported results suggest that, for polymers oflow polarity, a decrease of the solvent/nonsolvent mixture polar-ity favors the formation of spheres. The opposite trend isfound for polymers containing highly polar groups, that is,

biopolymers. Therefore, when the polymer and the solvent/nonsolvent mixture have similar polarity, the equilibrium mor-phology is favored.Influence of the MWCO on Nanostructured Polymer

Dimensions. With the aim to evaluate the influence of physicalparameters on the polymer nanostructure dimensions, we useddialysis bags with different MWCOs.

Aqueous solutions of PU, DX, and CS were submitted toOBM by using dialysis membranes with MWCOs of 12000 and2000 Da. Figure 7 reports the SEM micrographs of (A) DX,(B) PU, and (C) CS micro- and nanospheres. Reducing the mem-brane MWCO, the mean diameter of the spheres decreases byabout 2 orders of magnitude, ranging from 1.5 μm (MWCO =12000 Da) to 20-30 nm (MWCO = 2000 Da). These experi-mental results confirm the importance of the mixing rate of thesolvent pairs inside the dialysis bag in determining the finalequilibrium morphology. By reducing the membrane MWCO,the mixing rate of the solvents decreases, thus favoring thermo-dynamic factors over kinetic ones in the organizationof polymericchains.

The evaluation of the influence of MWCO was carried outfor the synthetic polymer PPA at different concentrations ofthe polymer in DMF (0.5 and 0.1 mg/mL). The decrease of themembrane pore diameters induces a decrease of the mean particlediameter, in the range 80-100 nm, in analogy to the resultsobtained for biopolymers.

Conclusions

A novel and straightforward strategy to fabricate nanostruc-tured polymers by exploiting their solution properties at theinterface indifferentmedia is presented.Variations of the polymermorphologies and size modulation can be obtained by changingthe experimental conditions of the OBM process such as tem-perature, polymer concentration, solvent polarity, and dialysismembranemolecular weight cutoff. These physicochemical para-meters are able to control the osmotic process, thus influencingpolymer nucleation during nanostructure formation. Sphericalmorphologies can be easily obtained by controlling temperatureand polymer concentration or by changing the dielectric constantdifference between the solvent and nonsolvent. Other nonequili-brium morphologies are obtained, although their formationcannot be precisely foreseen through the control of all thephysicochemical parameters. The nanoparticles obtained are freeof contaminants such as surfactants, initiator residues, and theirdecomposition products. The methodology proposed can beapplicable to a broad range of polymers including those whichcannot be obtained by emulsion polymerization (biopolymers).Finally, the modulation in shape and dimensions of the nano-structured materials opens new strategies for application inbiotechnology and sensing areas.

Acknowledgment. The authors thank Dr. Daniela Ferrofor SEM images, Fidia Pharmaceuticals for providing hyalu-ronic acid derivatives, HYAFF 11 and HYAFF 9, and Dr. PeterPreston for valuable advice and language revision.

Supporting Information Available: NMR and GPC datafor polymers before and after OBM method and SEMimages. This material is available free of charge via theInternet at http://pubs.acs.org.

Figure 7. SEM micrographs of nanostructured biopolymers.(A) DX: H2O/EtOH, 0.5 mg/mL, MWCO 12000; (B) DX: H2O/EtOH, 0.5 mg/mL, MWCO 2000; (C) PU: H2O/EtOH, 1 mg/mL,MWCO 12000; (D) PU: H2O/EtOH, 1 mg/mL, MWCO 2000; (E)CS: H2O/EtOH, 0.5 mg/mL, MWCO 12000; (F) CS: H2O/EtOH,0.5 mg/mL, MWCO 2000.