Hierarchical Structures of Hydrogen-Bonded Liquid-Crystalline Side-Chain Diblock Copolymers in Nanoparticles

Hierarchical Structures of Hydrogen-Bonded Liquid-Crystalline Side-Chain Diblock Copolymers in NanoparticlesAntti J. Soininen,† Antti Rahikkala,† Juuso T. Korhonen,† Esko I. Kauppinen,† Raffaele Mezzenga,‡

Janne Raula,*,† and Janne Ruokolainen*,†

†Department of Applied Physics, Aalto University School of Science, 02150 Espoo, Finland‡Food & Soft Materials, Department of Health Science & Technology, ETH Zurich, 8092 Zurich, Switzerland

*S Supporting Information

ABSTRACT: Here we show that it is possible to control the overallmorphology as well as hierarchical microstructure of lamellae-forming blockcopolymers within nanoparticles by altering side-chain content andprocessing temperature. We used cholesteryl hemisuccinate (CholHS) ashydrogen-bonded side chains and poly(styrene)-block-poly(4-vinylpyridine)(PS−P4VP) as backbone to produce submicrometer particles with theaerosol method. With CholHS-to-P4VP repeat unit ratio of 0.25 and 0.50,we obtained onion-like particles with either single CholHS layerssandwiched between P4VP rich lamellae or smectic P4VP(CholHS) layers perpendicular to the polymer domain interfaces.When the fraction of CholHS was increased to 0.75, the onion-like structure broke down due to increased splay deformationenergy of the liquid crystalline P4VP(CholHS) domains. The onion-like structure could be re-established, however, when theparticles were produced at a higher temperature which made the CholHS molecules partially soluble into the PS phase. Becauseof the reversible nature of the hydrogen bonds, it was possible to selectively remove the CholHS side chains from the particles.

■ INTRODUCTION

Submicrometer-sized particles with inner structures havepromising applications for instance as templates for porousmaterials1 or as microenvironments for chemical reactions.2

Block copolymers offer a promising way to achieve innerstructures in the 10−100 nm length scale thanks to thepossibility to control their microphase separation.3 Themicrophase-separated structures can take the form of, forexample, alternating lamellae, hexagonally packed cylinders, orspheres in face-centered cubic lattice. The morphology dependson the relative volumes of the polymer blocks and the strengthof their mutual interactions. The block with the lowest surfaceenergy typically concentrates on the surface of the material.4−6

This controls the final microphase-separated structure of blockcopolymer in particles whose radius is smaller than thecoherence length of the structures. The surface-inducedorientation results in spherically symmetric structures inspherical particles. For instance, computational studies predictthat block copolymers with symmetric block volumes micro-phase separate into concentric shells of alternating do-mains.7−12 The morphology can be described as “onion-like”.However, if the size of the particles is incommensurate with theperiodicity of the domains, more exotic morphologies canarise.9,11,13 Since the early report by Thomas et al.,14 the onion-like morphology has been observed also experimentally in anumber of works.15−28 In particles prepared by the so-calledgood solvent-poor solvent method, the surface layer can becontrolled by the choice of surfactants.19 By further modifyingthe surface interactions, it is possible to turn the onion-likestructures to “tulip-bulb”-like or even change the shape of the

particles from spherical to prolate spheroidal with blockcopolymer lamellae stacked orthogonal to the long axis.22

Stacked lamellar configurations have also been prepared by thegood solvent-poor solvent method in temperatures below thesupposed glass transition temperature of the particles.25,29

Thermal annealing of these particles re-established theequilibrium onion-like structure,25,29 as solvent vapor annealingalso did.26

Blending homopolymers into the particles increases thethickness of the shells24 or alters the overall microphase-separated morphology.17,20,22 Gold nanoparticles have alsobeen blended with the block copolymers and shown to beselectively dispersed into the poly(styrene) domains ofpoly(styrene)-block-poly(isoprene) particles.30 The onion-likemorphology can be used to create concentric polymericcontainers by hydrogen bonding 3-pentadecylphenol to thepoly(4-vinylpyridine) blocks of poly(styrene)-block-poly(4-vinylpyridine) and selectively removing the 3-pentadecylphenolmolecules after formation of the particles.28 One efficientmethod to produce spherical block copolymer particles is in theaerosol reactor.14,23,31 In this method, a block copolymersolution is dispersed into droplets in a laminar gas flow aerosolreactor. The solvent evaporates quickly in the reactor whichresults in solid particles with diameters ranging from about 10nm to a few μm. With the aerosol method, it is easy toincorporate different elements to the particles such as drugs, for

Received: July 17, 2012Revised: October 15, 2012Published: October 23, 2012

Article

pubs.acs.org/Macromolecules

© 2012 American Chemical Society 8743 dx.doi.org/10.1021/ma301486p | Macromolecules 2012, 45, 8743−8751

pubs.acs.org/Macromolecules

example, if the particle is designed to be a carrier.32,33A blockcopolymer system may microphase separate hierarchically ifone of the blocks (called here the host block) carries sidechains.34−37 In hierarchical microphase separation the side-chain blocks organize into liquid crystalline structures while thecopolymer blocks microphase separate in an order ofmagnitude larger length scale. This leads to structure-within-structure morphologies where the smaller liquid crystallinestructure exists within one of the microphase-separated blockcopolymer domains. Hydrogen bonding or other weak bindingmechanism can be used to attach the side chains to the hostpolymer block.38−45 This makes the preparation of a materialeasy and gives control over the side-chain content, that is, theattachment ratio or the ratio of the side chains to the number ofrepeating units in the host block, and thus control over the finalmicrostructure. In some cases, liquid crystalline ordering ensueseven though the side chains do not show liquid crystallinebehavior alone.38−42,44 Additionally, the reversibility of weakbonds like the hydrogen bond can bring additional functionalityto the material such as switchable electrical conductivity46 orphotonic band gap.47,48

We have hydrogen-bonded cholesteryl hemisuccinate(CholHS) and poly(styrene)-block-poly(4-vinylpyridine)(PS−P4VP) to obtain side-chain diblock copolymers (PS−P4VP(CholHS)) and studied their hierarchical microphaseseparation as a function of side chain content.49 With a lowamount of CholHS side chains, the side chains microphaseseparated into a novel triple lamellar structure where singlelayers of side chains were sandwiched between host block richdomains. Binding additional side chains to the host blocksinduced smectic stacking of the blocks which led to thehierarchical smectic-within-lamellar microphase-separatedstructure commonly found in other side-chain diblockcopolymers.38,39,44,45 In this study, we address the microphaseseparation of these hierarchical structures in spherical particlesproduced by the aerosol method. Interestingly, we found that insome cases the block copolymer lamellae orient perpendicularto the particle surface instead of the usual onion-like

morphology. Our aim here is to show that this nononion-likemorphology is caused by the combined effects of the liquidcrystalline smectic stacking of the P4VP(CholHS) blocks andsurface effect and that it can be controlled by processingtemperature in the aerosol reactor.

■ EXPERIMENTAL SECTIONMaterials. PS−P4VP (PS: 31 900 g/mol; P4VP: 13 200 g/mol;

polydispersity: 1.08; Polymer Source, chemical structure shown inFigure 1a) and CholHS (molecular weight: 486.73 g/mol; Sigma-Aldrich) were dried overnight in a vacuum oven. CholHS wasdissolved in dimethylformamide (DMF, Sigma-Aldrich) as 1 wt %solution and mixed with PS−P4VP. Then more DMF was added tobring the final concentration of PS−P4VP(CholHS) to 1 wt %.

The Aerosol Method. The setup of the aerosol flow reactor usedfor the production of the particles has been described in detailpreviously.50 Briefly, a Collison type air jet atomizer was used in therecycling mode to atomize the PS−P4VP(CholHS) solution to aerosoldroplets that were transported to a heated tubular stainless steelreactor (inner diameter 26 mm and length 900 mm) with nitrogencarrier gas using a flow rate of 2.5 L/min. The generated aerosols weredried and heated to 160 or 200 °C with residence time of about 8 s.Four separate heating blocks were used to ensure uniform temperatureprofile. Downstream the reactor, the aerosol flow was cooled anddiluted with a large volume of nitrogen gas (30 L/min).The sampleswere collected on carbon-coated transmission electron microscopygrids (Agar Scientific) using an electrostatic precipitator (InToxProducts) and on aluminum sheets using a Berner low-pressureimpactor.51 A part of the particles collected on the aluminum sheetswere dispersed in ethanol (Altia) to selectively remove CholHS.

Transmission Electron Microscopy (TEM). TEM images of theparticles were taken with Ultrascan 4000 CCD camera (Gatan) usingJEM 3200FSC field emission microscope (Jeol) operated at 200 kV orat 300 kV in bright field mode while specimen temperature wasmaintained at −187 °C. Additional imaging was done with Ultrascan1000 CCD camera (Gatan) using Tecnai 12 microscope (FEI)operated at 80 kV or 120 kV in bright-field mode. Prior to imaging, thesamples were stained in vapor of iodine for 1−4 h. Iodine selectivelystains the P4VP blocks dark to provide contrast in the TEM images.Further, logarithm of the intensity is used in Figures 3, 5, 6, and 7 toenhance the otherwise dark color scale in the center of the particles.

Figure 1. (a) Chemical structures of PS−P4VP and CholHS and their mutual hydrogen bonding. (b) TEM images of the samples in bulk andschematic packing of the PS−P4VP and PS−P4VP(CholHS) with different amounts of CholHS side chains.

Macromolecules Article

dx.doi.org/10.1021/ma301486p | Macromolecules 2012, 45, 8743−87518744

The size of the microphase-separated morphologies in each of thesamples was determined by calculating the mean value of 20−75domain distances measured from particles in several TEM images byhand using Digital Micrograph software (Gatan). The standard errorsof the means were less than 1 nm. To assess the size distribution of theparticles, a series of TEM images were analyzed with the AnalyzeParticles tool in the ImageJ software.52

Scanning Electron Microscopy (SEM). SEM was performed on aJEOL JSM-7500FA field-emission microscope operated at 1.5 kV.Dispersions of the particles in ethanol were drop-cast on lacey carbongrids and stained 1 h in the vapor of iodine before imaging to enhancebackscattered electron contrast, and the samples were not otherwisecoated. An in-column type secondary electron detector was used toimage the topography of the particles.

■ RESULTS AND DISCUSSIONFigure 1a shows the chemical structures of PS−P4VP andCholHS as well as their mutual hydrogen bonding. A previousstudy on the hydrogen bonding using infrared spectroscopyshows that most of the CholHS molecules are bonded to thepyridine rings of P4VP repeat units with insignificant amount ofdimerization or nonbonded CholHS molecules.45 The samplesused in this study as well as their bulk morphologies, asdetermined according to our previous study,49 are summarizedin Table 1. Additionally, Figure 1b shows schematically thepacking of the copolymers and TEM images of thecorresponding bulk samples. To summarize the results, thepure PS−P4VP sample P-00 consisted of P4VP cylindershexagonally ordered in PS matrix as can be expected from theweight fraction (0.29) of the P4VP block. Attaching CholHSside chains to the P4VP blocks increases the volume fraction ofP4VP(CholHS) which results in lamellar morphologies withdifferent P4VP(CholHS) chain packing depending on theCholHS-to-P4VP repeat unit ratio. In the least CholHS-containing sample P-25, the CholHS side chains microphaseseparated into a single layer sandwiched between P4VP-richdomains. A higher amount of CholHS led to a gradual changeto smectic packing: in P-50, the single layered and smecticpacking coexisted while the packing was exclusively smectic inP-75.In the aerosol method, an atomizer is used to produce small

airborne droplets of a solution of starting materials. The solventevaporates quickly from the droplets, and the resulting particlesare conveyed into a heated reactor. The particles will then reachthermal equilibrium at the reactor if the components of theparticles are sufficiently mobile. The rapid cooling process backto room temperature after the reactor may kinetically trap themorphology of the particles into this state. This annealingprocess in the reactor renders the choice of solvent and theeffect of evaporation kinetics on the morphology largelyirrelevant. Thus, the final morphology in the case of blockcopolymer particles depends solely on the reactor temperatureif that temperature is above the glass transition temperatures ofthe blocks and the residence time is long enough. In our

experience, pure PS−P4VP particles, for instance, producedfrom DMF solutions with this method are left to non-equilibrium morphologies caused by the evaporation kinetics ofthe solvent if reactor temperature below the glass transitions ofPS and P4VP is used. These morphologies can be identified byconsiderably smaller microphase-separated structures than inbulk. However, equilibrium morphologies with lattice param-eters comparable with bulk are observed in the particles whenthe reactor temperature is above the glass transition temper-atures, that is, when the copolymer chains are mobile enough toreach thermal equilibrium.Figure 2a shows the size distribution of the P-00 particles

obtained from 1 wt % solution in DMF. Although particles with

diameters over 1 μm are present, most of the particles are lessthan 500 nm in diameter the median being 150 nm. Figure 2bshows a TEM image of five P-00 particles with diameters of80−260 nm produced at a reactor temperature of 160 °C,which is above the glass transition temperatures of both PS andP4VP. Inside the particles, worm-like domains of P4VP(stained dark by iodine) can be seen. The “periodicity” of theP4VP worms in the P-00 particles is ∼37 nm according to theTEM images which is reasonably close to the (001) latticeplane distance of 37.4 nm of P-00 in bulk (check for Table 1).Thus, we can conclude that the used reactor temperature of160 °C is sufficient to reach thermal equilibrium on thecopolymer chain scale. This should be even truer for the P-25,P-50, and P-75 particles with the CholHS side chains, sinceCholHS may act as plasticizer lowering the glass transitiontemperature of P4VP. On the mesoscale, however, the usedresidence time at the applied temperature results in thedisordered nature of the microphase-separated domains (nowell-ordered lattice observed). Because of this, we did not seeany of the intriguing domain shapes caused by the sphericalconfinement of the particles predicted by earlier simula-tions8,11,53,54 or observed experimentally.20,22 Later thermalannealing of the particles proved to be unsuccessful since theparticles melted and lost their shape or were lumped together

Table 1. Properties of the PS−P4VP(CholHS) Diblock Copolymer Samples

sampleCholHS-to-P4VP repeat unit

ratioP4VP(CholHS) block weight

fraction bulk morphology (TEM)(001) lattice plane distance

(nm)a

P-00 0.00 0.29 hexagonally packed P4VP cylinders in PS matrix 37.4P-25 0.25 0.47 single parallel CholHS layer within P4VP

lamellae29.5

P-50 0.50 0.58 mixed CholHS subphases within P4VP lamellae 29.5 and 36.0P-75 0.75 0.65 smectic P4VP(CholHS) lamellae 36.0

aAccording to our previous study.49

Figure 2. (a) Size distribution of P-00 particles produced at 160 °C.(b) TEM image of P-00 particles on a carbon support film showingthe worm-like P4VP domains.



immediately after the temperature was raised above the glasstransition temperatures of the copolymer blocks.Addition of CholHS in P-25 leads to the lamellar block

copolymer morphology shown in Figure 1b which organizesinto concentric shells in the particles. Figure 3 shows a TEMimage of a typical P-25 particle with the onion-like morphologyconsisting of three shells of PS−P4VP(CholHS) and aP4VP(CholHS) core. The shells in turn consist of alternatingPS lamellae and P4VP(CholHS) lamellae where a singleCholHS layer is packed within the P4VP-rich domains. To thebest of our knowledge, this is the first report of such ahierarchically organized onion-like lamellar phase confined insubmicrometer particles. Inside the spherical core of theparticle in Figure 3, the P4VP(CholHS) blocks pack into

smectic layers as opposed to the single layer packing in theouter shells. The smectic packing is probably due to thespherical shape of the core. The mean lamellar thickness of theP-25 particles is about 31 nm according to TEM images, whichis reasonably close to the bulk value of 29.5 nm. This indicatesthat the particles have reached thermal equilibrium in thereactor. The coherence length of the onion-like shells producedby our method is ∼4 times the lamellar period, and near thecenters of large particles (hundreds of nanometers or more indiameter) the morphology seems to be less ordered.Unfortunately, information from the morphology inside largeparticles is difficult to obtain from TEM images since theelectron beam intensity drops significantly inside the particlesleading to loss of contrast.PS covers the surface of pure PS−P4VP particles of P-00 due

to its lower surface energy compared to P4VP as can be seen inFigure 2b as well as in the onion-like PS-P4VP particle inFigure S1 of the Supporting Information. However, in the PS−P4VP(CholHS) particles, the CholHS side chains act assurfactants modifying the surface energy of the P4VP blocks.This allows the P4VP(CholHS) blocks to concentrate at theparticle/air interface instead of PS. This phenomenon can beverified by volumetric considerations and TEM. Figure 4ashows a sector of a P-25 particle as well as the intensity curve ofthe electron beam integrated in the azimuthal direction.Because of iodine staining, the P4VP domains appear dark,and the valleys in the intensity curve can be identified as P4VPdomains while the small peak in their middle can be attributedto the CholHS layers. In a lamellar A−B diblock copolymer, ifthe A blocks concentrated on a surface, the sequence of thelayers from inside of the material to the surface would be ...A-BB-AA-BB-A (see Figure 4b). Because of this, the A layer at thesurface is only about half of that in the inner layers where the

amount of A blocks is doubled as Figure 4b schematicallyshows. This is not the case for PS at the surface as can be seenfrom Figure 4a. On the contrary, the outermost PS shell isapproximately of the same thickness as the inner PS shell whichindicates that the surface layer must be of P4VP(CholHS). Forcomparison, the thickness of the outermost PS shell of the purePS−P4VP particle shown in Figure S1 is approximately onlyhalf of that of the inner PS shell. The concentration ofP4VP(CholHS) on the surface of the particles can be observedalso in the P-50 and P-75 particles. Similar control over thesurface chemistry of block copolymer particles by amphiphilicadditives has been demonstrated for example in particlesproduced in oil−water emulsions.22The size of the particles affects the number of shells and the

core domain confined in the center of the particles. Figure 5demonstrates the development of the morphology in small P-25 particles. The diameter of the particle in Figure 5a, 72 nm, isclose to 2.4 times the bulk lamellar period of 29.5 nm. Theparticle has a P4VP(CholHS) core domain which vaguelyresembles the smectic packing P4VP(CholHS) chains in thecenter of the larger particle shown in Figure 3. Figure 5b in turnshows a particle whose size is ∼3.8 times the lamellar period. In

Figure 3. TEM image of a 200 nm P-25 aerosol particle at the edge ofa carbon support film.

Figure 4. (a) Integrated electron intensity from a sector of a P-25particle shows that the outer PS shell is approximately of the samethickness as the inner PS shells. (b) A schematic arrangement of anA−B diblock copolymer near a surface shows that the thickness of theA layer at the surface must be about half of that inside the material.



this particle, there is one fully developed P4VP(CholHS) shelland a PS core. Finally, Figure 5c shows two larger particles: onewith P4VP(CholHS) core (diameter 4.3 times the lamellarperiod) and one with PS core (diameter 5.1 times the lamellarperiod). The smectic packing of P4VP(CholHS) seen inFigures 3 and 5a may be a general feature of the sphericalP4VP(CholHS) core domains, but this is difficult to prove fromthe TEM images since to observe the smectic layers, they needto be almost perfectly aligned to the electron beam. Bymeasuring the size of 55 particles larger than 70 nm in diameterand determining the block occupying the core domain, wecame to the conclusion that a new core emerges about everybulk period increase in the particle diameter. This is in contrastwith the Monte Carlo simulations done by Yu et al. where anew core domain emerges every 1.1 bulk periods.9

The increased amount of CholHS in P-50 compared to P-25turns the single CholHS layer morphology to the smecticpacking as shown in Figure 6 where the smectic layers can beseen inside the P4VP(CholHS) shells of the particle. However,the change is only partial, as a portion of the particles stillexhibited the single CholHS layer morphology. Also, mixedsingle layer and smectic packing were present in some particles.The lamellar period of the smectic packing particles is around35 nm, which is close to the bulk value, 36.0 nm. Also, themeasured period for the triple layer packing, 28 nm, does notdiffer appreciably from the bulk value of P-25. Thecomparability of the periodicities in the particles and in bulksignifies that both the smectic and single CholHS layermorphologies have equilibrated in the reactor. We tried to dosimilar particle size analysis considering the core domains as wedid with P-25 particles, but it proved to be difficult to identifythe core because of poor contrast in the center of the particlesand interference caused by the smectic P4VP(CholHS) aboveand below the center of the particles in the TEM. The lattereffect can be seen, for example, in the particle shown in Figure6.In addition to acting as surfactants, the CholHS side chains

can be used to completely change the onion-like ordering of theblock copolymer lamellae: strikingly in the P-75 particlesproduced at 160 °C, the onion-like structure breaks down. AsFigure 7 shows, the block copolymer domains do notmicrophase separate into the concentric shells seen in thecases of P-25 and P-50 particles. Instead of curving along thesurface of the particles, the block copolymer domains become

oriented perpendicular to the particle surface in the aerosolreactor. The P4VP(CholHS) blocks organize into smecticlayers inside their own domains. The period of the structures inthe P-75 particles is ∼37 nm, which is close to the bulk lamellarperiod of 36.0 nm. The onion-like morphology requiresextensive splay deformation of the smectic P4VP(CholHS)layers because of the small radiuses of curvature of the particles.This increase in the free energy probably makes the onion-likemorphology unfavorable in the P-75 particles. As the lessCholHS containing P-50 particles did not exhibit theperpendicular orientation, it may be reasonable to say thatthe splay deformations become more difficult as more CholHSside chains are bonded to the P4VP host block. In a way, theCholHS side chains make the P4VP(CholHS) lamellae “stiffer”.As stated earlier, P4VP(CholHS) concentrates preferentially atthe surface of the particles. The increased stiffness of theP4VP(CholHS) domains, however, prevents bending of theP4VP(CholHS) domain at the surface of the particle to acomplete spherical shell. Instead, the surface becomes patternedby alternating domains of PS and P4VP(CholHS) with theP4VP(CholHS) domains extending inside the particleperpendicularly to the surface. The perpendicular orientationobserved in the P-75 particles differs from the earlier

Figure 5. TEM images of a series of P-25 particles at the edge ofcarbon support films. Approximate particle diameters: (a) 72, (b) 113,and (c) 127, 151 nm.

Figure 6. TEM image of a P-50 particle with smectic P4VP(CholHS)packing attached to the edge of a carbon support film.

Figure 7. TEM image of a P-75 particle at the edge of a carbonsupport film.



structures22,25,29 in that it is essentially driven by the liquidcrystallinity induced stiffening of the P4VP(CholHS) lamellaeinstead of surface interactions during the formation of theparticles. Also, a recent computational study on the formationof perpendicular lamellar morphologies due to compatibility ofboth copolymer blocks with the surrounding media results inspheroidal-shaped particles.12 The shape of the P-75 particles,however, remains spherical, again emphasizing that theirperpendicular morphology originates from the increased splaydeformation. The morphology observed in the P-75 particlesshould be the thermally stable state at 160 °C, as we did notnotice difference in the morphology in TEM after the residencetime in the aerosol reactor was almost tripled from 8 to 23 s bycombining two reactors one after another.The spherical confinement combined with the perpendicular

surface orientation of the domains and the energetically costlysplay deformation of the lamellar morphology leads tofrustrated and less ordered microphase-separated structures.Figure 8 shows a collage of TEM images of P-75 particles. The

particles in Figures 8a,d seem to consist of two (Figure 8a) orthree (Figure 8d) flat P4VP(CholHS) lamellae sandwichedbetween PS lamellae. This kind of lamellar “stacks” was foundonly in particles of similar size, and larger particles, such as theone shown in Figure 7, have more “crooked” configurations.Further, Figure 8b shows a clearly different “star-like”P4VP(CholHS) domain configuration, meaning that thestacking of lamellae may not be the most favorableconfiguration energetically. No microphase-separated structuresappear in the particle in Figure 8c. However, the smectic layersof P4VP can be seen all over in the particle. This image mayhave been a result of a particle similar to the ones in Figure 8aor 8b being imaged along the stacking axis in the TEM. Lastly,Figure 8e shows two more particles with matching sizes butvery different and complex morphologies. This again underlinesthe fact that a morphology which satisfies all the constraintsimposed by the shape of the particles and the CholHS sidechain induced surface orientation is difficult to come by.

Since the perpendicular block copolymer lamellae orientationin the P-75 particles is essentially due to the smectic packing ofthe P4VP(CholHS) side chains, the onion-like structure shouldreturn if the particles were above the isotropization temper-ature. To test this hypothesis, we produced batches of theparticles at 200 °C, which is above the isotropizationtemperature of P4VP(CholHS) homopolymer45 but belowthe order−disorder transition temperature of PS−P4VP.55 Atthis temperature the structure of the particles should dependonly on the microphase separation of the copolymer blocks inthe molten state. As expected, no differences were observedbetween the microphase-separated structures produced at 160or 200 °C in the P-00 or P-50 particles. However, we weresurprised to find that the morphology of P-25 particles hadchanged to cylindrical as shown in Figure 9. Inside the

cylindrical P4VP(CholHS) domains, there are signs of CholHSconcentrating at the center of the domains similarly to thelamellar structure of P-25 in bulk and in particles produced at160 °C. The phenomenon resembles the temperature-depend-ent changes in morphology which have been observed before insimilar hydrogen-bonded side-chain diblock copolymer systemsand can be explained by the breakage of the hydrogen bondsand partial mixing of the side-chain molecules with the coilblocks.46 Probably in the case of P-25 as well, the CholHSmolecules partially swell into the PS domains, changing therelative volume fractions of the polymer blocks andconsequently the overall morphology. When the particles arerapidly cooled down before collecting this high-temperaturemorphology is “frozen” to a kinetically trapped state. Theperiodicity of the cylinders is 34 nm which is lower than the 37nm of P-00 which has similar morphology. One would expectthe periodicity to increase if molecules are added to a materialand the morphology stays the same. The situation resembleswhat we have observed earlier with hydrogen-bonded poly-(butadiene)-block-poly(2-vinylpyridine) and CholHS: thelamellar structure retains its period even though more CholHSis added to the system.49 It seems that the same holds forcylindrical morphology with an axial CholHS domain in thecenter of the P4VP(CholHS) cylinders.

Figure 8. TEM images of P-75 particles attached to each other or atthe edges of a carbon support film. Approximate particle diameters:(a) 119, (b) 126, (c) 162, (d) 156, and (e) 163, 165 nm.

Figure 9. TEM image of a P-25 particle produced at 200 °C.



In the P-75 particles produced at 200 °C, no perpendicularorientation was found. Instead, we found particles such as theone shown in Figure 10 with onion-like block copolymer

structure. The onion-like structure is again effectively frozen asthe cylindrical morphology in the P-25 particles because of thequick cooling in the aerosol reactor. Also this change from theperpendicular surface orientation to the onion-like morphologycan be attributed to the partial dissolution of CholHS into thePS domains. However, the original hypothesis that theperpendicular morphology breaks down due to the isotropiza-tion of the smectic P4VP(CholHS) domains may still play arole here, although the phenomenon is difficult to distinguishfrom the mixing of the CholHS in PS.Figure 11 shows schematically the hierarchical microphase-

separated structures found in the PS−P4VP(CholHS) particles.

With low amount of CholHS side chains as in P-25 and at 160°C, the CholHS side chains microphase separate into a singlelayer sandwiched between the P4VP domains. The lamellae onthe block copolymer level bend to onion-like concentric shellsdue to the preferential concentration of P4VP(CholHS) blockson the surface and spherical shape of the particles. Increasing

the amount CholHS side chains keeps the onion-like structureintact while changing the P4VP(CholHS) packing to smecticlayers oriented perpendicular to the block copolymer domaininterfaces as in P-50. Further increasing the amount of CholHSside chains leads to loss of the onion-like block copolymerordering, as observed in the case of P-75. The strikingbreakdown of the onion-like structure is due to the increased“rigidity” of the P4VP(CholHS) lamellae resulting from theincreased contribution of splay deformations to the free energy.The onion-like structure results again, if the particles areproduced at 200 °C which allows the CholHS molecules topartially mix with the PS domains “releasing” the rigidity of thesmectic lamellae. Also, the same mixing phenomenon leads tochange from lamellar to cylindrical morphology in P-25 whenproduced at 200 °C as the volume fraction of the PS domainsincreases as CholHS molecules diffuse into them. As themorphologies are kinetically trapped when cooled down rapidlyin the aerosol reactor, the method enables control over themicrophase-separated structure by reactor temperature. Thestructure of the P-50 particles stays effectively the same in bothreactor temperatures used.Because of the reversible nature of hydrogen bonds, it is

possible to remove the side chains by a solvent selective toCholHS. We were especially interested to find out whether theremoval of CholHS from the P-75 particles would result inplate-like objects or porous particles. P-75 particles produced at160 °C with the non-onion-like morphology (Figure 7) werethus dispersed in ethanol, a selective solvent for CholHS andP4VP and drop-cast on a TEM grid. Figure 12 shows SEM and

TEM images of the particles after the ethanol treatment. As canbe seen from the images, the particles do not break intoseparate platelets which suggest that the lamellae do not juststack or form “tulip bulb”-like structures22 but rather organizein more complex ways. This could open a door for porousparticles, but unfortunately, the wrinkly shape of the particles inthe SEM image in Figure 12a points to crumpling of theparticles after CholHS removal due to softness of the remainingPS−P4VP domains. The TEM image in Figure 12b alsosuggests that there are no voids inside the particles.The P-25 and P-50 particles were treated with ethanol as

well. The initially low amount of CholHS in the P-25 particlesled to onion-like particles with thin P4VP shells with occasionalvoids. In the P-50 particles, the removal of CholHS resulted inconcentric PS−P4VP shells resembling closely to previouslyreported particles of PS−P4VP and 3-pentadecylphenol afterselective removal of 3-pentadecylphenol.28

Figure 10. TEM image of a P-75 particle produced at 200 °C at theedge of a carbon support film.

Figure 11. Schematic presentation of the different microphaseseparated structures found in the PS−P4VP(CholHS) particles as afunction of CholHS content and aerosol reactor temperature.

Figure 12. (a) Secondary electron SEM image of an aggregation of P-75 particles at the edge of a carbon support film after selective removalof CholHS by ethanol. (b) TEM image of a particle on the same film.



■ CONCLUSIONSThe hierarchical microphase-separated lamellar structures thatwe identified in bulk of PS−P4VP(CholHS) previously49 canalso be found in the particles produced by the aerosol methodusing a reactor temperature of 160 °C. This temperature isabove the glass transition temperatures of both PS and P4VPproviding the chains sufficient mobility to reach thermalequilibrium inside the reactor rendering the effects of solventevaporation on the final morphology nonexistent. The blockcopolymer lamellae bend to concentric shells creating onion-like particles in the case of P-25 and P-50. The onion-likemorphology is a result of the preferential concentration ofCholHS at the air/particle interface combined with thespherical shape of the particles which force the lamellar blockcopolymer domains to organize into concentric shells. Thesmectic stacking of the P4VP(CholHS) blocks takes controlover the microphase separation with higher amount of CholHSside chains in P-75. In this case, the unaffordable deformationenergy of the smectic layers leads to the breakdown of theonion-like structure at the block copolymer level. Instead, theblock copolymer lamellae orient perpendicular to the surface ofthe particles, increasing the surface energy but reducing thesplay deformation energy associated with the onion-likemorphology. The onion-like structure re-emerges if the P-75particles are prepared at 200 °C. At this temperature, part ofthe CholHS molecules dissolve into the PS phase, reducing the“rigidness” of the P4VP(CholHS) lamellae and re-enabling theonion-like structure. This morphology is preserved at roomtemperature since the rapid cooling after the reactor kineticallytraps the morphology to this state. Thus, we have shown thatthe smectic packing of liquid crystalline P4VP(CholHS) blockscan guide the microphase separation on the block copolymerlevel in particles created with the aerosol method. The changein the solubility of CholHS in PS at 200 °C also leads to thechange from onion-like morphology to cylindrical morphologyin P-25 particles. Intriguingly, CholHS side chains pack into asingle domain in the center of the P4VP(CholHS) cylindersimilarly to the lamellar structure of P-25 particles produced at160 °C. Additionally, CholHS was selectively removed from theparticles by ethanol. In the onion-like particles, concentricshells that were partially attached to each other were formed. Inthe P-75 particles where the block copolymer structure wasperpendicular to the surface of the particles, the remaining PS−P4VP crumpled into wrinkly, nonporous particles.

■ ASSOCIATED CONTENT

*S Supporting InformationInformation on preparation of onion-like PS-P4VP particlesand associated Figure S1. This material is available free ofcharge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATION

Corresponding Author*E-mail: [email protected]; [email protected].

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

We acknowledge the funding from the Academy of Finland(projects 140303 and 140362). We also thank JoonasIivanainen for help in sample preparation.

■ REFERENCES(1) Wang, Y.; Price, A. D.; Caruso, F. J. Mater. Chem. 2009, 19,6451−6464.(2) Palivan, C. G.; Fischer-Onaca, O.; Delcea, M.; Itel, F.; Meier, W.Chem. Soc. Rev. 2011, 41, 2800−2823.(3) Hamley, I. W. In The Physics of Block Copolymers; OxfordUniversity Press: New York, 1998; p 424.(4) Thomas, H. R.; O’Malley, J. J. Macromolecules 1979, 12, 323−329.(5) Hasegawa, H.; Hashimoto, T.Macromolecules 1985, 18, 589−590.(6) Fredrickson, G. H. Macromolecules 1987, 20, 2535−2542.(7) He, X.; Song, M.; Liang, H.; Pan, C. J. Chem. Phys. 2001, 114,10510−10513.(8) Fraaije, J. G. E. M.; Sevink, G. J. A. Macromolecules 2003, 36,7891−7893.(9) Yu, B.; Li, B.; Jin, Q.; Ding, D.; Shi, A. Macromolecules 2007, 40,9133−9142.(10) Han, Y.; Cui, J.; Jiang, W. Macromolecules 2008, 41, 6239−6245.(11) Li, S.; Chen, P.; Zhang, L.; Liang, H. Langmuir 2011, 27, 5081−5089.(12) Chi, P.; Wang, Z.; Li, B.; Shi, A. C. Langmuir 2011, 27, 11683−11689.(13) Feng, J.; Liu, H.; Hu, Y. Fluid Phase Equilib. 2007, 261, 50−57.(14) Thomas, E. L.; Reffner, J. R.; Bellare, J. J. Phys., Colloq. 1990, 51,C7−363−C7−374.(15) Ding, J.; Liu, G. Macromolecules 1999, 32, 8413−8420.(16) Arsenault, A. C.; Rider, D. A.; Tetreault, N.; Chen, J. I. L.;Coombs, N.; Ozin, G. A.; Manners, I. J. Am. Chem. Soc. 2005, 127,9954−9955.(17) Okubo, M.; Saito, N.; Takekoh, R.; Kobayashi, H. Polymer 2005,46, 1151−1156.(18) Kagawa, Y.; Minami, H.; Okubo, M.; Zhou, J. Polymer 2005, 46,1045−1049.(19) Saito, N.; Takekoh, R.; Nakatsuru, R.; Okubo, M. Langmuir2007, 23, 5978−5983.(20) Jeon, S.; Yi, G.; Koo, C. M.; Yang, S. Macromolecules 2007, 40,8430−8439.(21) Higuchi, T.; Yabu, H.; Shimomura, M. J. Nanosci. Nanotechnol.2007, 7, 856−858.(22) Jeon, S.; Yi, G.; Yang, S. Adv. Mater. 2008, 20, 4103−4108.(23) Zhang, K.; Gao, L.; Chen, Y.; Yang, Z. Chem. Mater. 2008, 20,23−25.(24) Tanaka, T.; Saito, N.; Okubo, M. Macromolecules 2009, 42,7423−7429.(25) Higuchi, T.; Motoyoshi, K.; Sugimori, H.; Jinnai, H.; Yabu, H.;Shimomura, M. Macromol. Rapid Commun. 2010, 31, 1773−1778.(26) Li, L.; Matsunaga, K.; Zhu, J.; Higuchi, T.; Yabu, H.;Shimomura, M.; Jinnai, H.; Hayward, R. C.; Russell, T. P.Macromolecules 2010, 43, 7807−7812.(27) Staff, R. H.; Rupper, P.; Lieberwirth, I.; Landfester, K.; Crespy,D. Soft Matter 2011, 7, 10219−10226.(28) Deng, R.; Liu, S.; Li, J.; Liao, Y.; Tao, J.; Zhu, J. Adv. Mater.2012, 14, 1889−1893.(29) Higuchi, T.; Yabu, H.; Onoue, S.; Kunitake, T.; Shimomura, M.Colloids Surf., A 2008, 313−314, 87−90.(30) Yabu, H.; Jinno, T.; Koike, K.; Higuchi, T.; Shimomura, M.Macromolecules 2011, 44, 5868−5873.(31) Zhang, K.; Yu, X.; Gao, L.; Chen, Y.; Yang, Z. Langmuir 2008,24, 6542−6548.(32) Islam, N. In Colloids in Aerosol Drug Delivery Systems; Fanun, M.,Ed.; Surfactant Science Series; CRC Press: Boca Raton, FL, 2010; Vol.148, pp 479−500.(33) Labiris, N. R.; Bosco, A. P.; Dolovich, M. B. Nanoparticulates asdrug carriers. In Aerosols as Drug Carriers; Torchilin, V. P., Ed.; WorldScientific: Singapore, 2006; pp 367−395.(34) Fischer, H.; Poser, S. Acta Polym. 1996, 47, 413−428.(35) Thomas, E. L.; Chen, J. T.; O’Rourke, M. J. E.; Ober, C. K.;Mao, G. Macromol. Symp. 1997, 117, 241−256.



http://pubs.acs.org

mailto:[email protected]

mailto:[email protected]

(36) Poser, S.; Fischer, H.; Arnold, M. Prog. Polym. Sci. 1998, 23,1337−1379.(37) Walther, M.; Finkelmann, H. Prog. Polym. Sci. 1996, 21, 951−979.(38) Ruokolainen, J.; Saariaho, M.; Ikkala, O.; ten Brinke, G.;Thomas, E. L.; Torkkeli, M.; Serimaa, R. Macromolecules 1999, 32,1152−1158.(39) Ruokolainen, J.; ten Brinke, G.; Ikkala, O. Adv. Mater. 1999, 11,777−780.(40) Thunemann, A. F.; General, S. Macromolecules 2001, 34, 6978−6984.(41) Bondzic, S.; de Wit, J.; Polushkin, E.; Schouten, A. J.; ten Brinke,G.; Ruokolainen, J.; Ikkala, O.; Dolbnya, I.; Bras, W. Macromolecules2004, 37, 9517−9524.(42) Nandan, B.; Lee, C. H.; Chen, H. L.; Chen, W. C.Macromolecules 2005, 38, 10117−10126.(43) Hammond, M. R.; Mezzenga, R. Soft Matter 2008, 4, 952−961.(44) Chiang, W.; Lin, C.; Nandan, B.; Yeh, C.; Rahman, M. H.;Chen, W.; Chen, H. Macromolecules 2008, 41, 8138−8147.(45) Korhonen, J. T.; Verho, T.; Rannou, P.; Ikkala, O. Macro-molecules 2010, 43, 1507−1514.(46) Ruokolainen, J.; Makinen, R.; Torkkeli, M.; Makela, T.; Serimaa,R.; ten Brinke, G.; Ikkala, O. Science 1998, 280, 557−560.(47) Osuji, C.; Chao, C.; Bita, I.; Ober, C. K.; Thomas, E. L. Adv.Funct. Mater. 2002, 12, 753−758.(48) Valkama, S.; Kosonen, H.; Ruokolainen, J.; Haatainen, T.;Torkkeli, M.; Serimaa, R.; ten Brinke, G.; Ikkala, O. Nat. Mater. 2004,3, 872−876.(49) Soininen, A. J.; Tanionou, I.; ten Brummelhuis, N.; Schlaad, H.;Hadjichristidis, N.; Ikkala, O.; Raula, J.; Mezzenga, R.; Ruokolainen, J.Macromolecules 2012, 45, 7091−7097.(50) Eerikainen, H.; Kauppinen, E. I. Int. J. Pharm. 2003, 263, 69−83.(51) Hillamo, R. E.; Kauppinen, E. I. Aerosol Sci. Technol. 1991, 14,33−47.(52) Abramoff, M. D.; Magalhaes, P. J.; Ram, S. J. Biophotonics Int.2004, 11, 36−42.(53) Chen, P.; Liang, H.; Shi, A. Macromolecules 2008, 41, 8938−8943.(54) Yu, B.; Li, B.; Jin, Q.; Ding, D.; Shi, A. Soft Matter 2011, 7,10227−10240.(55) Zha, W.; Han, C. D.; Lee, D. H.; Han, S. H.; Kim, J. K.; Kang, J.H.; Park, C. Macromolecules 2007, 40, 2109−2119.



Hierarchical Structures of Hydrogen-Bonded Liquid-Crystalline Side-Chain Diblock Copolymers in Nanoparticles

Documents