Controlled Assembly of Nanoparticle Structures: Spherical and

May 12, 2009 - Safa'a Al-Rehili , Karim Fhayli , Mohamed Amen Hammami , Basem Moosa , Sachin Patil , Daliang Zhang , Ohoud Alharbi , Mohamed Nejib ...
0 downloads 0 Views 3MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Controlled Assembly of Nanoparticle Structures: Spherical and Toroidal Superlattices and Nanoparticle-Coated Polymeric Beads Tatsushi Isojima, Su Kyung Suh, John B. Vander Sande, and T. Alan Hatton* Departments of Chemical Engineering and Materials Science and Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139 Received February 11, 2009. Revised Manuscript Received April 22, 2009 The emulsion droplet solvent evaporation method has been used to prepare nanoclusters of monodisperse magnetite nanoparticles of varying morphologies depending on the temperature and rate of solvent evaporation and on the composition (solvent, presence of polymer, nanoparticle concentration, etc.) of the emulsion droplets. In the absence of a polymer, and with increasing solvent evaporation temperatures, the nanoparticles formed single- or multidomain crystalline superlattices, amorphous spherical aggregates, or toroidal clusters, as determined by the energetics and dynamics of the solvent evaporation process. When polymers that are incompatible with the nanoparticle coatings were included in the emulsion formulation, monolayer- and multilayer-coated polymer beads and partially coated Janus beads were prepared; the nanoparticles were expelled by the polymer as its concentration increased on evaporation of the solvent and accumulated on the surfaces of the beads in a well-ordered structure. The precise number of nanoparticle layers depended on the polymer/magnetic nanoparticle ratio in the oil droplet phase parent emulsion. The magnetic nanoparticle superstructures responded to the application of a modest magnetic field by forming regular chains with alignment of nonuniform structures (e.g., toroids and Janus beads) that are in accord with theoretical predictions and with observations in other systems.

Introduction The fabrication and directed assembly of functional nanoparticles comprised of magnetic materials, noble metals, and quantum dots are of increasing interest for a wide range of application areas, and there is currently a strong focus on the fundamental properties of such particle assemblies. The controlled clustering of nanoparticles with narrow size distribution, uniform shape, and tailored surface properties, in particular, is a promising approach for the development of new materials for a number of important applications.1-5 The clustering of nanoparticles on flat surfaces has been reported for the synthesis of single6-8 or binary nanoparticle superlattices,9 while supercrystalline colloidal particles,5 binary nanoparticle diamond lattice-like crystals,1 and anisotropic nanoparticles have been prepared in the bulk.10,11 Clustering at liquid-liquid interfaces has also been investigated as a route for the synthesis of new composite materials.12 Janus particles, which have anisotropic surface properties, have attracted much interest in recent years13 as they can have unique

processing advantages over their uniformly coated counterparts, depending on their amphiphilic character,14 surface charge distributions,15 dipole moments16,17 and responsiveness to changes in environmental conditions.10,11 These Janus-type particles can be self-assembled by exploiting their surface properties or through application of an external field.18 Methods for producing anisotropic colloid surfaces have been developed using vapor deposition,19 stamping,15,17 phase separation of a polymer mixture,20,21 adsorption at an air-water interface22,23 or on submicrometer silica particles,10,11 and seeded polymerization.24 Under an external magnetic field, clusters of magnetic nanoparticles align to form chains.5,25,26 The interaction between the magnetic nanoparticles affects the shapes of secondary structures formed under external magnetic fields. For instance, Singh modeled the behavior of magnetic rings in the presence of an external magnetic field and predicted chainlike structures but did not have the ability to test his analysis experimentally.27 We have also shown theoretically that the magnetic response, i.e., magnetic susceptibility, of core-shell type beads in which a polymeric core

*Corresponding author. E-mail: [email protected]. (1) Kalsin, A. M.; Fialkowski, M.; Paszewski, M.; Smoukov, S. K.; Bishop, K. J. M.; Grzybowski, B. A. Science 2006, 312, 420. (2) Klajn, R.; Bishop, K. J. M.; Fialkowski, M.; Paszewski, M.; Campbell, C. J.; Gray, T. P.; Grzybowski, B. A. Science 2007, 316, 261. (3) Ge, J.; Hu, Y.; Biasini, M.; Beyermann, W. P.; Yin, Y. Angew. Chem., Int. Ed. 2007, 46, 4342. (4) Ge, J.; Hu, Y.; Yin, Y. Angew. Chem., Int. Ed. 2007, 46, 7428. (5) Zhuang, J.; Wu, H.; Yang, Y.; Cao, Y. J. Am. Chem. Soc. 2007, 129, 14166. (6) Zhao, S.; Wang, S.; Kimura, K. Langmuir 2004, 20, 1977. (7) Sun, S.; Zeng, H.; Robinson, D. B.; Raoux, S.; Rice, P. M.; Wang, S. X.; Li, G. X. J. Am. Chem. Soc. 2004, 126, 273. (8) Yang, X.; Liu, C.; Ahner, J.; Yu, J.; Klemmer, T.; Johns, E.; Weller, D. J. Vac. Sci. Technol. B 2004, 22, 31. (9) Shevchenko, E. V.; Talapin, D. V.; Murray, C. B.; O’Brien, S. J. Am. Chem. Soc. 2006, 128, 3620. (10) Lattuada, M.; Hatton, T. A. J. Am. Chem. Soc. 2007, 129, 12878. (11) Isojima, T.; Lattuada, M.; Vander Sande, J. B.; Hatton, T. A. ACS Nano 2008, 2, 1799. (12) Boker, A.; He, J.; Emrick, T.; Russell, T. P. Soft Matter 2007, 3, 1231. (13) Perro, A.; Reculusa, S.; Ravaine, S.; Bourgeat-Lami, E. B.; Duguet, E. J. Mater. Chem. 2005, 15, 3745.

8292 DOI: 10.1021/la900522u

(14) Hong, L.; Cacciuto, A.; Luijten, E.; Granick, S. Langmuir 2008, 24, 621. (15) Cayre, O.; Paunov, V. N.; Velev, O. D. J. Mater. Chem. 2003, 13, 2445. (16) Velev, O. D.; Lenhoff, A. M.; Kaler, E. W. Science 2000, 287, 2240. (17) Cui, J. -Q.; Kretzschmar, I. Langmuir 2006, 22, 8281. (18) Gangwal, S.; Cayre, O. J.; Bazant, M. Z.; Velev, O. D. Phys. Rev. Lett. 2008, 100, 058302. (19) Correa-Duarte, M. A.; Salgueirino-Maceira, V.; Rodriguez-Gonzalez, B.; Liz-Marzan, L. M.; Kosiorek, A.; Kandulski, W.; Giersig, M. Adv. Mater. 2005, 17, 2014. (20) Saito, N.; Kagari, Y.; Okubo, M. Langmuir 2006, 22, 9397. (21) Kietzke, T.; Neher, D.; Kumke, M.; Ghazy, O.; Ziener, U.; Landfester, K. Small 2007, 3, 1041. (22) Fujimoto, K.; Nakahama, K.; Shidara, M.; Kawaguchi, H. Langmuir 1999, 15, 4630. (23) Petit, L.; Sellier, E.; Duguet, E.; Ravaine, S.; Mingotaud, C. J. Mater. Chem. 2000, 10, 253. (24) Kim, J. -W.; Larsen, R. J.; Weitz, D. A. J. Am. Chem. Soc. 2006, 128, 14374. (25) Singh, H.; Laibinis, P. E.; Hatton, T. A. Langmuir 2005, 21, 11500. (26) Singh, H.; Laibinis, P. E.; Hatton, T. A. Nano Lett. 2005, 5, 2149. (27) Singh, H. Ph.D Thesis, Massachusetts Institute of Technology, Cambridge, MA, 2005.

Published on Web 05/12/2009

Langmuir 2009, 25(14), 8292–8298

Isojima et al.

Article

Figure 1. Schematic of the synthesis procedure for the preparation of magnetic nanoparticle clusters. The OA/OAm-coated nanoparticles are dissolved in the solvent and emulsified with water, the solvent droplets being stabilized by SDS. The solvent is then evaporated, and the nanoparticles cluster to form the desired structures. When a polymer is added to the solvent mixture, the polymer precipitates out as a bead, and the rejected nanoparticles accumulate on the bead surface.

is coated with a monolayer of magnetic nanoparticles should be higher than that of the same concentration of nanoparticles distributed throughout polymeric particles, but this difference decreases as the number of layers increases.28 Such theories can not be tested, however, as uniformly coated beads are not available. Magnetic nanoparticles with narrow size distributions can now be obtained routinely,7,29,30 and their surface functionalities can be manipulated and controlled readily.31-36 Superparamagnetic magnetite (Fe3O4) nanoparticles are especially useful for a wide range of applications, including drug delivery, diagnostics, gene analysis, proteomics, separation and purification, hyperthermia therapy, in vivo imaging, and photonics.37-40 In many cases, the as-synthesized individual nanoparticles are not very useful for their intended applications, however, since the requirement that they be single-domain and superparamagnetic dictates that they be limited to about 15-20 nm in size. Particles of this size often do not respond sufficiently strongly or rapidly to magnetic fields to be useful in their intended applications area, and thus it is desirable to form small clusters of these nanoparticles that are more responsive to moderate magnetic fields and yet still retain their superparamagnetic properties. We describe here facile methods for producing magnetite nanoparticle clusters with varying geometry (spherical and toroidal clusters) and internal packing structure (single- or multidomain superlattices and amorphous aggregates) based on (28) Singh, H.; Hatton, T. A. J. Magn. Magn. Mater. 2007, 315, 53. (29) Sun, S.; Zeng, H. J. Am. Chem. Soc. 2002, 124, 8204. (30) Saita, S.; Maenosono, S. Chem. Mater. 2005, 17, 3705. (31) Wang, L.; Luo, J.; Fan, Q.; Suzuki, M.; Suzuki, I. S.; Engelhard, M. H.; Lin, Y.; Kim, N.; Wang, J. Q.; Zhong, C.-J. J. Phys. Chem. B 2005, 109, 21593. (32) Mikhaylova, M.; Kim, D. K.; Bobrysheva, N.; Osmolowsky, M.; Semenov, V.; Tsakalakos, T.; Muhammed, M. Langmuir 2004, 20, 2472. (33) Lattuada, M.; Hatton, T. A. Langmuir 2007, 23, 2158. (34) Shamim, N.; Hong, L.; Hidajat, K.; Uddin, M. S. Colloids Surf., B 2007, 55, 51. (35) Gravano, S. M.; Dumas, R.; Liu, K.; Patten, T. E. J. Polym. Sci., Polym. Chem. 2005, 43, 3675. (36) Tanaka, Y.; Saita, S.; Maenosono, S. Appl. Phys. Lett. 2008, 92, 093117. (37) Liu, X.; Novosad, V.; Rozhkova, E. A.; Chen, H.; Yefremenko, V.; Pearson, J.; Torno, M.; Bader, S. D.; Rosengart, A. J. IEEE Trans. Magn. 2007, 43, 2462. (38) Maeda, M.; Kuroda, C. S.; Shimura, T.; Tada, M.; Abe, M.; Yamamuro, S.; Sumiyama, K.; Handa, H. J. Appl. Phys. 2006, 99, 08H103. (39) Jun, Y.-W.; Huh, Y.-M.; Choi, J.-S.; Lee, J.-H.; Song, H.-T.; Kim, S.; Yoon, S.; Kim, K.-S.; Shin, J.-S.; Suh, J.-S.; Cheon, J. J. Am. Chem. Soc. 2005, 127, 5732. (40) Gijs, M. A. M. Microfluid. Nanofluid. 2004, 1, 22.

Langmuir 2009, 25(14), 8292–8298

the emulsion solvent evaporation method. We also prepare ∼100-200 nm polymer spheres coated with closely packed magnetic nanoparticle layer(s) with any desired degree of surface coverage (partial (i.e., Janus), mono- or multilayer) by a using a one-step solvent evaporation method. Many of these structures are reported in this paper for the first time. The approach we use is depicted schematically in Figure 1. Magnetic nanoparticles with hydrophobic surfaces are dispersed in an oil phase, which is then emulsified in a water continuous phase to provide a solid-in-oil-in-water (S/O/W) emulsion. On evaporation of the droplet solvent the particles are concentrated and form the targeted cluster structure. In the presence of a polymer that is incompatible with the nanoparticle coatings, the particles are excluded from the polymer-rich zone during the evaporation process and assemble on the surfaces of the resulting polymer beads. The emulsion solvent evaporation method has been used by others for the encapsulation of nanoparticles and other additives within desired polymer matrices37,41-43 and for the control of polymer particle morphologies.20,21 Recently, micrometer-sized colloidal crystal spheres of latex beads were prepared by the drying of simple droplets.16,44

Experimental Section Materials. Iron tri(acetylacetonate) (Fe(acac)3) (97%), 1,2-tetradecanediol (90%), oleic acid (OA) (90%), oleyl amine (OAm) (70%), benzyl ether (99%), sodium dodecyl sulfate (SDS) (99%), hexane (99%), and polyethylene (PE) (Mw: 35 000) were purchased from Sigma Aldrich. Methanol (99.8%) and chloroform (100%) were purchased from Mallinkrodt. Polystyrene (PS) (Mw: 125 000-250 000) was purchased from Alfa Aesar. All chemicals were used as received. All water utilized in the experiments was Milli-Q (Millipore) deionized water. Synthesis.

Preparation of the Magnetic Nanoparticles.

The Fe3O4 nanoparticles were synthesized using Sun’s method.7 Fe(acac)3 (2 mmol), 1,2-tetradecanediol (10 mmol), oleic acid (6 mmol), oleylamine (6 mmol), and benzyl ether (20 mL) were mixed under a constant flow of nitrogen for 30 min. The solution (41) Shang, H.; Chang, W.-S.; Kan, S.; Majetich, S. A.; Lee, G. U. Langmuir 2006, 22, 2516. (42) Omi, S.; Kanetaka, A.; Shimamori, Y.; Supsakulchai, A.; Nagai, M.; Ma, G. H. J. Microencapsul. 2001, 18, 749. (43) Andersson, N.; Kronberg, B.; Corkery, R.; Alberius, P. Langmuir 2007, 23, 1459. (44) Kuncicky, D. M.; Bose, K.; Costa, K. D.; Velev, O. D. Chem. Mater. 2007, 19, 141.

DOI: 10.1021/la900522u

8293

Article

Isojima et al.

Figure 2. Characterization of the magnetite nanoparticles synthesized using the approach of Sun et al.7,29,33 (a) TEM image demonstrates the monodispersity of the synthesized nanoparticles, (b) HRTEM image of a nanoparticle showing the single-domain crystal structure, (c) TEM image of nanoparticles on the edges of holes in the lacey carbon grids, showing the OA/OAm monolayer coating on the nanoparticles indicated by arrows, (d) FTIR spectrum showing the bands for Fe3O4 and for the methylene groups of the OA/OAm layer, and (e) TGA analysis of the coated nanoparticle showing that the OA/OAm layer accounts for ∼17% of the total mass of the coated nanoparticles. was heated gradually to 100 °C and kept at this temperature for 45 min. Next, the mixture was heated to 200 °C and maintained at this tempeature for 40 min. Finally, under a blanket of nitrogen, the mixture was heated to reflux (∼290 °C) for 1 h and then cooled to room temperature. The reaction solution was centrifuged with 40 mL methanol at 7000 rpm for 10 min using an Eppendorf Inc. centrifuge 5810R. The precipitate was dispersed in 20 mL of hexane and centrifuged again to remove any undispersed residue. The magnetic nanoparticles were precipitated with excess methanol, separated by an electromagnet (magnetic separator model L-1; S.G. Frantz Co. Inc.), and then dried at 80 °C. Nanoparticle Cluster Formation. A 0.3 mL colloidal suspension of magnetic nanoparticles (3.0 wt %) in hexane was dispersed in 10 mL of SDS aqueous solution (1.0 wt %) by ultrasonic homogenization using a Branson Sonifier 450 sonicator for 30 s. Finally, hexane was evaporated under mechanical stirring at the desired temperature and for a specified time. Preparation of Nanoparticle-Coated Polymer Beads. A 0.3 mL mixture of PS (either 3.0 or 0.7 wt %) and magnetic nanoparticles (3.0 wt %) in chloroform was dispersed in 10 mL of a 1.0 wt % SDS aqueous solution for the preparation of either mono- or multilayer-coated PS beads. For the synthesis of partially coated PS spheres, 0.3 mL of hexane was added to 0.3 mL of a dispersion of magnetic nanoparticles (either 1.0 or 1.5 wt %) and 3.0 wt % PS in chloroform. These mixtures were subjected to ultrasonic homogenization for 30 s before the chloroform (and hexane when used) was evaporated under mechanical stirring at 40 °C for 12 h.

Measurements. Transmission and Scanning Electron Microscopy (TEM and SEM). TEM was performed on JEOL 200-CX and 2010 transmission electron microscopes at an accelerating voltage of 200 kV. A JEOL 6320FV scanning electron microscope with a 5.0 kV accelerating voltage was used for the 8294 DOI: 10.1021/la900522u

SEM measurements. Samples for both TEM and SEM were prepared by placing drops of the nanoparticle dispersion on carbon and lacey carbon thin films supported by 200 mesh Cu grids made by Electron Microscopy Sciences and Structure Probe, Inc., respectively. For the observation of the array of magnetic nanoparticles cluster under an external magnetic field, the carbon-coated grid was placed between two parallel permanent magnets. The magnitude of the magnetic field measured by a Bell-5180 gaussmeter (Sypris Solutions, Inc.) was 160 mT. Samples were prepared under the external magnetic field by placing drops of the nanoparticle cluster dispersion on the grid. Fourier Transform Infrared Spectroscopy (FTIR). FTIR spectroscopy was performed on a NEXUS 870 FTIR spectrometer (Thermo Nicolet, Inc.). Spectra were recorded in the wavenumber range between 4000 and 400 cm-1 at a resolution of 2 cm-1 and reported as the average of 64 spectral scans. The sample was dried overnight at 80 °C in a vacuum oven and then ground and mixed with KBr to form the pellets used in the measurements. Thermogravimetric Analysis (TGA). TGA measurements were performed on a TGA Q50 (TA Instruments). All measurements were made under a constant flow of a gas mixture of nitrogen at 90 mL/min and helium at 10 mL/min. The temperature was increased from 30 to 600 °C at a rate of 15 °C/min. The sample was dried in a vacuum oven overnight at 80 °C prior to TGA analysis. Dynamic Light Scattering (DLS). DLS experiments were performed using a Brookhaven BI-200SM light scattering system (Brookhaven Instruments Corp.) at a detection angle of 90°. The Contin program was used for hydrodynamic diameter determinations from the DLS correlation functions. Samples were measured for 3 min, and the reported diameters are the average of three repeated measurements. Langmuir 2009, 25(14), 8292–8298

Isojima et al.

Article

Zeta Potential Measurements. All zeta potential measurements were performed using a Brookhaven ZetaPALS zeta potential analyzer (Brookhaven Instruments Corp.). The Smoluchowski equation was used to calculate the zeta potential from the electrophoretic mobility. The reported zeta potential values are an average of five measurements, each of which was obtained over 20 electrode cycles.

Results and Discussion Preparation of Monodisperse Magnetic Nanoparticles. High-resolution TEM (HR-TEM) images of the magnetic nanoparticles synthesized in the work, stabilized in a nonpolar organic solvent by a mixed oleic acid and oleyl amine (OA/OAm) surface monolayer, are shown in Figure 2. These images indicate (a) the monodispersity of the nanoparticles, which is important for their use as building blocks in the fabrication of crystalline nanoparticle superlattice structures, and (b) their single-domain crystal lattice structure, which ensures that the nanoparticles and their clusters retain their superparamagnetic properties. The OA/OAm coating on the nanoparticles is seen clearly in (c); since the contrast between the coating of the nanoparticles and the carbon film of a traditional TEM grid is poor, we used lacey carbon grids, i.e., thin carbon films perforated with holes, in which the structures of coated nanoparticles located on the peripheries of the holes and protruding into the holes could be clearly discerned. The value of the OA/OAm shell suggests a layer thickness of about 0.2-0.5 nm. The FTIR spectrum of the OA/OAm magnetic nanoparticles shown in Figure 2d exhibits a magnetite (Fe3O4) adsorption band at 586 cm-1, while the bands at 2852 and 2927 cm-1 are characteristic of the methylene groups of the OA/OAm layer. Thermogravimetric analysis (TGA) of the OA/ OAm-coated magnetic nanoparticles indicated that OA/OAm shell contributed 17 wt % to the total mass of the particles (Figure 2e). Such an OA/OAm shell acts to disperse the magnetic nanoparticles in a nonpolar organic solvent, which dispersion is then used as the droplet phase in a S/O/W type emulsion. Controlled Clustering of Magnetic Nanoparticles. The controlled formaton of dense clusters of the 7 nm magnetic nanoparticles was accomplished by the solvent evaporation method illustrated schematically in Figure 1. The emulsion formed by dispersing a nanoparticle-laden solvent phase as droplets in an aqueous contnuous phase was heated to drive off the solvent preferentially. The solvent has a low but finite solubility in the aqueous phase, and once dissolved in this phase diffuses, until it reaches the emulsion/air interface where it can escape as a vapor. The stirring of the emulsion during this process reduces the concentration boundary layer at the vapor/liquid interface and hence accelerates the loss by evaporation of the solvent phase. As the nanoparticles within the emulsion droplets became more and more concentrated with continuing solvent loss, they packed into dense clusters dispersed within the water phase, as illustrated schematically in Figure 1. The packing characteristics within these clusters were determined by the temperature or rate at which the solvent evaporated, as shown in the TEM images in Figure 3. With gentle solvent evaporation for 3 days at room temperature or slightly above (∼25-40 °C), the particles clustered to form well-ordered crystalline nanoparticle superlattice structures (Figure 3a), which was confirmed by the FFT (fast Fourier transform) image shown in Figure 3b; the single crystal diffraction pattern indicates the (110) plane of a BCC (body-centered cubic) structure. With an increase in the solvent evaporation temperature to 50 °C, a multidomain polycrystalline structure was formed (Figure 3c) while at 60 °C (near the hexane boiling point of 69 °C) for 12 h the relatively rapid evaporation of the oil phase prevented Langmuir 2009, 25(14), 8292–8298

Figure 3. Magnetic nanoparticle clusters formed at different rates of hexane evaporation: (a) TEM image of a single-domain crystalline superlattice formed at low temperature (25 °C), (b) fast Fourier transform diffraction pattern for particles in (a) showing BCC (110) structure, (c) TEM image of a multidomain crystalline superlattice formed at intermediate temperatures (50 °C), (d) TEM image of amorphous cluster formed at higher temperatures (60 °C), and (e) TEM image and (f) SEM image of toroidal structures formed at 80 °C, above the solvent boiling point. The TEM image in (e) was paired with an image of the same cluster tilted at an angle of 30° for stereoscopic visualization to show that the toroidal aggregates have a true doughnut-like structure.

the particles from arraying in any well-ordered manner, resulting in an amorphous structure (Figure 3d), as confirmed by an FFT analysis of the TEM images (not shown). After 8 h at 80 °C, which is significantly higher than the solvent boiling point, toroidal or doughnut-shaped clusters formed; Figure 3e shows one such cluster, and Figure 3f shows an SEM image of a similar cluster. Stereoscopic visualization of the toroidal cluster shown in Figure 3e paired with an image of the same cluster tilted at an angle of 30° using a type F-71 stereoscope from Forestry Suppliers Inc. confirmed the full three-dimensional toroidal structure. We speculate that a hexane vapor bubble forms within the droplet, forcing the nanoparticles to concentrate in a shell around the vapor core. Eventually the bubble breaks through the shell at its weakest point and erupts from the droplet, thereby forming the hole in the toroidal cluster, while surface tension forces cause the shell region to adopt the regular curved surface characteristics of the doughnut-like ring structure. It appears that the cluster is not purely toroidal in nature since there is a layer or two of nanoparticles forming a membrane-like film across the DOI: 10.1021/la900522u

8295

Article

Isojima et al.

Figure 4. Nanoparticle-coated polystyrene beads prepared by the solvent evaporation method using choroform as the solvent for both polystyrene and the nanoparticles: (a, b) beads without nanoparticles, (c, d) monolayer-coated beads, (e, f) bilayer-coated beads, (g-j) Janus beads with different surface coverages depending on the initial nanoparticle/polystyrene ratio. In (g)-(j), the solvent used was a mixture of hexane and chloroform.

hole of the toroid. We attribute this effect to an initial asymmetric growth or oscillation of the vapor bubble as it is forming which causes a thinning of the particle/solvent layers at opposite poles of the vapor bubble, and the subsequent ejection of the solvent vapors from one of these poles leaves the other and its layer of particles somewhat intact. The initial emulsion droplets were typically fairly polydisperse with a mean diameter on the order of 800 nm to 1 μm, measured by dynamic light scattering, and thus the clusters were also polydispersed, with mean diameters of about 150-180, nm as anticipated from calculations based on the initial volume fraction of nanoparticles in the dispersed emulsion phase. These clusters were stabilized electrostatically against flocculation by the anionic surfactant sodium dodecyl sulfate (SDS) used originally to stabilize the parent emulsion droplets.; the zeta potentials of the clusters ranged between -60 and -90 mV. The alkyl chains on the surfaces of magnetic nanoparticles participated in both nanoparticle-nanoparticle and, on the surface, nanoparticlesurfactant interactions, both of which acted to stabilize the selfassembled clusters. It is inevitable that the excess surfactant released as the emulsion droplet surface area decreased during the evaporation not only formed micelles in the continuous aqueous phase but also coated some of the individual nanoparticles to form a charged surfactant bilayer that allowed for the stable dispersion of some individual nanoparticles in the continuous phase. Magnetic Nanoparticle-Coated Polymer Beads. We also prepared magnetic nanoparticle-coated polymer beads via the solvent evaporation method. In this case the emulsion droplets contained both the OA/OAm-coated nanoparticles and the beadforming polymer PS (or, in some instances, PE) dissolved uniformly in chloroform, which was used as the nonpolar volatile organic solvent since the polymers have poor solubility in hexane. Because the nanoparticles with their exposed alkyl tails are incompatible with the polymers, they phase-separated from the PS (or PE) as the concentrations of the two components increased during the solvent evaporation process; the nanoparticles were rejected by the polymer core and were displayed on the surfaces of the resulting polymer beads. We used appropriate concentrations of PS and magnetic nanoparticles to ensure the desired final 8296 DOI: 10.1021/la900522u

structures, namely monolayer or multilayer-coated beads. The beads formed with polymer alone, in the absence of nanoparticles, are shown in Figure 4a,b. For PS beads of 100 nm diameter and 7 nm magnetic nanoparticles, the required volume ratio of magnetic nanoparticles to PS is about 5 for a monolayer coating, which is equivalent to a mass ratio of ∼1.0 (assuming a PS density of 1.0 g/cm3 and Fe3O4 density of 5.2 g/cm3). The resulting bead structure is shown in Figure 4c,d, where it is clearly evident that the nanoparticles assembled to form a well-ordered monolayer on each bead surface; that it was a monolayer is evident from the thickness of the nanoparticle coating seen on the edges of the beads. Figure 4e,f shows multilayer-coated PS beads obtained when the volume ratio of PS to magnetic nanoparticles was about 1, consistent with a double-layer coverage of the beads with nanoparticles. Again, the layers on the edges of the bead image indicate such multilayer ordered coverage of the beads. When the nanoparticle concentration used was less than that required to form a monolayer, the particles were distributed randomly over the bead surface, with no discernible ordering (not shown). To form Janus beads, we used a mixture of chloroform and hexane as the nonpolar organic solvent. Chloroform has a lower boiling point (61.4 °C) than does hexane (69 °C) and therefore evaporates more quickly. Thus, the solvent composition ratio changes during the evaporation process, and the increasingly hexane-rich solvent phase becomes a poorer solvent for the polystyrene, which ultimately precipitates to form the PS bead. While we have not established the complete phase diagram for this system, we have observed that such a phase separation occurs at room temperature when the choroform:hexane ratio approaches about 20:80. With further evaporation of the solvents, a dewetting at the bead surface by the hexane/magnetic nanoparticle dispersion occurs, and the rejected hexane phase is speculated to accumulate at one pole of the polymer bead. On further evaporation of the hexane, the magnetic nanoparticles accumulate as a well-ordered and compact structured layer on just one half of the polymer particle surface to form the Janus beads shown in Figure 4g,h. When the initial concentration of the nanoparticles is reduced even further, the particles are concentrated over a commensurately smaller area on the particle surface, located at one of the poles of the bead, as shown in Figure 4i,j. Langmuir 2009, 25(14), 8292–8298

Isojima et al.

Article

Figure 5. Chaining of magnetic nanoparticle clusters under a magnetic field of 160 mT: (a, b) crystalline superlattices, (c, d) toroidal clusters, and (e, f) Janus beads.

The structures obtained here were not confined to the formation of coated polystyrene beads only; we observed the same structures when we used poyethylene in the emulsion evaporation processes (results not shown). Chaining of Clusters and Beads under an Applied Magnetic Field. The magnetic nanoparticle clusters and coated beads can form chains under weak external magnetic fields, as shown in Figure 5 for the three major categories of nanoparticle assemblies: crystalline nanoparticle superlattices, toroidal nanoparticle clusters, and Janus beads. These structures were formed when a magnetic field of 160 mT was applied to the suspension during the preparation of the TEM grids. Figure 5a,c,e shows the chain structures for the different nanoparticles assemblies, while Figure 5b,d,f shows higher magnifications of these particles in each of the chains. The crystalline superlattices shown in Figure 5b generally retained their structure, although there is an amorphous accumulation of nanoparticles around each point of contact of the clusters. These amoprhous regions could be due to a breakdown of the superlattice structure in these regions, which is held together primarily by van der Waals and hydrophobic forces, or due to an accumulation at the juncture points of the individually dispersed nanoparticles The strong magnetic field gradients established in the high curvature contact regions would provide the driving force for the particles to accumulate there. Similar observations are made for the toroidal doughnut-shaped clusters, with the added feature that the clusters formed chains edge to edge, in accord with the predictions of Singh, who showed that magnetic rings are linked together along their edges in the direction of the applied magnetic field as this provides the minimum-energy configuration.27 The Janus beads aligned along their equators with the externally applied magnetic field as shown in Figure 5e,f, and, on (45) Zhao, N.; Gao, M. Y. Adv. Mater. 2009, 21, 184–187.

Langmuir 2009, 25(14), 8292–8298

average, with each other in a zigzag fashion, as has been observed in other systems.45 The chains are not entirely uniform in their structures because of polydispersity effects associated with the polydispersity of the parent emulsion droplets. Future work will address this issue.

Conclusions We have reported the synthesis of nanoclusters of monodisperse 7 nm magnetite nanoparticles coated with OA/OAm stabilizing layers via the emulsion droplet solvent evaporation method. A range of morphologies, many novel, can be generated depending on the temperature and rate of solvent evaporation and on the composition (solvent, presence of polymer, nanoparticle concentration, etc.) of the emulsion droplets. In the absence of a polymer, and with increasing solvent evaporation temperatures, the nanoparticles formed single- or multidomain crystalline superlattices, amorphous spherical aggregates, or toroidal clusters, as determined by the energetics and dynamics of the solvent evaporation process. Monolayer- and multilayercoated polymer beads, and partially coated Janus beads, were prepared by relying on the incompatibility of the concentrated polymer solution following solvent evaporation with the alkyl tail groups of the stabilizing OA/OAm stabilizing layers on the nanoparticles; the nanoparticles were expelled by the polymer and accumulated at the surface of the beads in a well-ordered structure, the precise number of layers depending on the polymer/ magnetic nanoparticle ratio in the oil droplet phase parent emulsion. The magnetic nanoparticle superstructures responded to the application of a modest magnetic field by forming regular chains with alignment of nonuniform structures (e.g., toroids and Janus beads) that are in accord with theoretical predictions and with observations in other systems. DOI: 10.1021/la900522u

8297

Article

While we have focused on the assembly of magnetic nanoparticles in this work, it should be emphasized that this simple technique is a versatile and powerful method for creating unique self-assembled nanoparticle structures through manipulation of the surface interactions between the nanoparticles, solvent, and polymer. Continuing work is focused on generating monodisperse clusters and beads, and their subsequent stabilization by cross-linking of the

8298 DOI: 10.1021/la900522u

Isojima et al.

nanoaprticle within the structures, in addition to the generation of other morphologies using the approaches delineated here. Acknowledgment. This work was supported by the Mitsubishi Chemical Corp. and the Singapore-MIT Alliance (SMA). The assistance of Dr. Yong Zhang with the high-resolution TEM measurements is gratefully acknowledged.

Langmuir 2009, 25(14), 8292–8298