Article pubs.acs.org/Langmuir
Synthesis of Single and Multipatch Particles by Dip-Coating Method and Self-Assembly Thereof Manigandan Sabapathy, Sam David Christdoss Pushpam, Madivala G. Basavaraj, and Ethayaraja Mani* Polymer Engineering and Colloid Science Lab, Department of Chemical Engineering, Indian Institute of Technology Madras, Chennai 600036, Tamil Nadu, India S Supporting Information *
ABSTRACT: We report a simple strategy to produce single and multipatch particles via the conventional dip-coating process. In this method, a close-packed monolayer of micronsized silica particles is first formed at air−polymer solution interface, followed by dip coating of particles on a glass substrate. The simultaneous deposition of both polymer and particles on the substrate gives rise to a thin polymer layer and a monolayer of silica particles. Sonication of the substrate leads to the formation of a polymeric patch on one side of the particles. The patch shape depends on the aging of the polymer film prior to sonication. With aging time the patch evolves from ring-like to disk-like. This technique allows easy control of patch width by varying the concentration of polymer in the solution. We further show that the number of patches on the particle can be increased by controlling the concentration of silica particles at the interface such that surface coverage is less than that required for the formation of a close-packed monolayer. The single and multipatch particles are characterized by scanning electron and optical microscopy for the patch size, shape, and number distribution. The as-synthesized particles are used as a model to study self-assembly of colloids with electrostatic repulsion and patchy hydrophobic attractions due to polymeric patches. We find the formation of doublets and finite-sized clusters due to patchy interactions. Dip coating can be automated to produce large quantities of patchy particles, which is one of the major limitations of other methods of producing patchy particles.
1. INTRODUCTION Micron-sized colloidal particles have been used as an experimental model system to study various features of phase behavior such as self-assembly, gelation, and crystallization.1 Because colloids can be imaged in real space and in accessible time scales, several analogous atomic phenomena can be realized. Furthermore, attention has been given toward the synthesis of spherical and anisotropic colloids with different surface functionalities, commonly known as patchy particles.2−5 Such particles mimic shape and directional interactions like molecules and allow one to study the formation of diverse varieties of crystals and self-assembled structures.6 Another feature of these particles is their ability to spontaneously assemble into new kinds of supramolecular complex structures2,7 in addition to their potential applications in targeted drug delivery,8 controlled loading of catalyst,9 and photonic crystals.10 For instance, particles decorated with hydrophobic and hydrophilic regions on either half of the particle (Janus particles) surface provide different affinity to liquid phases and act as “particulate surfactant” in stabilizing what is known as Pickering emulsion.11−14 The particles can also be used as model system to study complex interaction of globular proteins, inorganic macromolecules such as polyoxometalates, and virus capside formation by mimicking anisotropic patchy interactions.15 © 2015 American Chemical Society
Patchy colloids can be synthesized by modifying a part of the particle surface through chemical or physical modifications. Different methods such as sputtering of metal films,16 stamp coating of polymer,17 particle lithography,18,19 and glancing angle deposition20,21 have been employed to prepare patchy particles. In the context of patchy colloid synthesis, Hong et al.12 demonstrated a procedure for synthesizing single-patch (or Janus) particles. They followed emulsion based method using wax−aqueous phase system. At low temperature, solidified wax locks a part of the particles in the wax phase so that selective chemical modification can be done to produce Janus or singlepatch particles. Multipatch colloidal particles are made through glancing angle deposition technique by thermal evaporation of metals on a monolayer of particles on a substrate.20,21 In this technique, the angle between the target (metal source) and the substrate containing a monolayer of particles is manipulated to create multiple patches. These metal patches can then be chemically functionalized (with alkanethiols) to create hydrophobic patches. All of the above synthesis protocols share common features such as (1) positioning of the particles in the form of close Received: September 4, 2014 Revised: January 1, 2015 Published: January 5, 2015 1255
DOI: 10.1021/la503531a Langmuir 2015, 31, 1255−1261
Article
Langmuir Scheme 1. Schematic of the Mechanism of Synthesis of Patchy Particles via Dip Coating
polystyrene (PS) of average molecular weight ∼35 000 are obtained from Alfa Aesar and Sigma-Aldrich, USA, respectively. These polymers are used to make patches over silica particles. Ethanol (99.9% pure) and toluene are received from Merck, Germany. Milli-Q deionized water (DI) with resistivity of 18 MΩ cm is used for all process requirements. Microscopy glass slide (substrate) is thoroughly cleaned by immersing in a piranha bath containing hydrogen peroxide and sulfuric acid in the volume ratio of 1:3. 2.2. Synthesis of Patchy Particles. A 0.1 wt % polymer (PMMA or PS) solution is first prepared by dissolving calculated amount of PMMA or PS in toluene. 10 mL of this polymeric solution is taken in a Petri dish. The substrate (cleaned microscopy glass slide) is attached to the dipping port of the dip coater (model: Xdip-MV1, APEX Instruments co., India) and is dipped vertically into the Petri dish containing polymer solution to a depth of 5 mm. To deposit a film area of 5 × 12.5 mm2, we use a clean glass Petri dishes of 4.5 cm diameter. A 4 wt % silica particle suspension is prepared by dispersing calculated amount of silica particles in ethanol. A known volume of silica particle suspension (100 to 200 μL) was carefully spread at the polymer solution−air interface. Thereby, particles are uniformly distributed at the interface in the form of a monolayer. After a waiting time of 5 min, the substrate is withdrawn from the polymer solution at a speed of 2 mm/min until the whole substrate is lifted above the polymeric solution. This ensures that a monolayer of the particles was transferred onto the glass substrate by dip coating. The speed of lifting is optimized to obtain a continuous film. All experiments are conducted at room temperature (∼22 °C). The temperature plays an important role in these experiments because it controls the vapor pressure of toluene and hence the evaporation rate. The particle-laden substrate is lifted and then air-dried for 5 min. The substrate is unmounted. The particle-laden substrate is immersed in a suitable medium and sonicated to detach the particles from the substrate to obtain patchy particles. The evolution of patch morphology is studied by redispersion of particles from the particleladen substrate at different aging time. 2.3. Characterization of Patchy Particles. The structural morphology of patchy colloids is studied using high-resolution scanning electron microscope (SEM, Hitachi S-4800, Japan) with operation voltage from 1 to 3 kV. Prior to the analysis, a thin layer of gold is sputter coated on the sample to render them electronically conductive. Inverted microscope (model: DMI3000B, Leica Microsystems, Germany) is used to characterize the particle monolayers prior to dip-coating and to observe structural morphology of particles during self-assembly investigation. Ultrasonicator (500GTI, Martin Walter Powersonic, Germany) is used for the separation and redispersion of particles into required liquid medium by operating it at a frequency of 132 kHz for 20 min. Electrophoretic light scattering (Horiba SZ-100, Japan) is performed to measure the zeta potential of single-patch, multipatch and bare silica particles.
packed monolayer on a substrate or on the surface of a droplet (in the case of emulsion based methods), (2) multistep functionalization (chemical or physical treatment), and (3) separation of particles from the substrate and redispersion in a liquid. We propose here a simple and scalable method to make single and multipatch particles from a simultaneous deposition of both polymer and particles on a substrate by dip-coating method. This method is free from use of any surface active agents, masking, and sequential functionalization, as the polymer itself can be used as a hydrophobic patch. The dipcoating process has been a well-known method used to produce thin films in various technologies such as photoresists films22 and lubricant layers for magnetic hard disks.23 The process involves immersion of a substrate into a reservoir of polymer solution, followed by the spreading of particles at the polymer solution−air interface. Then, the substrate is withdrawn from the reservoir, such that the codeposition of polymer film is followed by a monolayer of particles. In the final step, the silica particles are disengaged from the substrate by sonication, so that particles with polymeric cap on one side are formed. We demonstrate: (1) synthesis of single-patch particles of controlled patch size by varying polymer concentration, (2) synthesis of multipatch particles based on the formation of polymeric capillary bridges, and (3) self-assembly of as synthesized particles into clusters. Patchy particles made via this route possess a solid polymer patch on their surface. When two particles of this kind are placed close to each other in water, the patchy part (polymer) of the particle attract via hydrophobic interaction, while the bare particle (silica) part would repel electrostatically. In contrast with the glancing angle deposition method, the use of polymer patch avoids the functionalization step to induce hydrophobic interactions. The method is shown to produce poly(methyl methacrylate) (PMMA) patch and polystyrene (PS) patch on silica. The method can be generalized to coat any polymer−particle combination as long as the polymer adheres to the substrate during the dip-coating process and particles remain at the interface. These particles are used in studying self-assembly of patchy colloids with electrostatic and hydrophobic interactions.
2. EXPERIMENTAL SECTION 2.1. Materials and Methods. Monodispersed silica microspheres (SiO2) used in the experiments are purchased from Fiber Optic Centre, USA, in the form of fine powder. The average diameter of the particles is found to be 4.22 μm based on the measurements of scanning electron microscopic images of the particles. Poly(methyl methacrylate) (PMMA) of average molecular weight ∼650 000 and 1256
DOI: 10.1021/la503531a Langmuir 2015, 31, 1255−1261
Article
Langmuir
3. RESULTS AND DISCUSSION 3.1. Single Patch Particles. Scheme 1 shows the schematic diagram of the technique used and the possible mechanism of formtion of polymer patch on silica particles. Initially, monodispersed silica particles (200 μL of 4 wt % suspension in ethanol) are carefully spread at the polymer solution−air interface (step 1). At the interface, particles are packed into a hexagonal lattice such that particles are nearly in contact with each other. One of the methods to ensure the formation of a close packed monolayer of particles is to use slightly excess volume of particle suspension. It may also be possible that, while spreading, a small number of particles may reach the bulk as well (step 1). When the substrate is lifted (step 2), a thin film of polymer solution containing particles is carried by the moving substrate. The height of this capillary film depends on the interfacial tension between polymeric solution and the substrate and polymer concentration. As the solvent in the film evaporates, a thin layer of polymer is left on the substrate. Simultaneously, particles are deposited on top of the polymer film. Therefore, there is a thin polymer film just below every deposited particle (step 3). Because only part of the particle surface is exposed to polymer solution, the part exposed to air will not have any polymer deposition. We used fluorescently labeled PMMA to confirm that the particles do not rotate at the polymer-solution interface. (See Figure S1 in Supporting Information.) Furthermore, we found the zeta potential of bare and single-patch particles to be almost the same at 3 mM aqueous NaCl solution, as only a small part of the surface is covered by the patch. This proves that the particle is locked at the interface, which prevents free rotation, and the particle surface exposed to polymer solution did not acquire any polymer from the solution. If there were to be any adsorption of free polymer on the particle surface, the measured zeta potential would have been significantly different from that of bare particle because the polymer is uncharged. When the substrate containing particles trapped in polymer film is sonicated by immersion in a suitable solvent, the particles detach from the substrate, giving rise to single patch particles with a polymer patch. To demonstrate the existence of single patch on the particles, we used carbon tape for easy visualization of patch located at the bottom surface of the particles. Figure 1A shows a representative high-resolution SEM image of regular hexagonal arrangement of particles on the glass substrate after dip coating. It can be seen that the particles are arranged in the form of a close packed regular hexagonal lattice, which is essential for the synthesis of single patch particles.
Similar arrangement was found in the other parts of the substrate. The next step to obtain a dispersion of single patch particles is to immerse the substrate in a liquid medium so as to detach the particles from the substrate. The imaging of redispersed particles to characterize patch morphology and size distribution poses difficulties because all of the patches on a sample prepared for SEM imaging may not be accessible. Therefore, a carbon tape is gently pressed against the particles deposited on the substrate, and the tape is peeled off slowly. In this process, the silica particles are transferred from glass substrate to the carbon tape. Imaging the particles this way gives better visualization of the patches and measurement of patch size and shape. Figure 1 shows the bottom view of the particle after peeling them using a carbon tape after an aging time of 6 h after dip coating. We see a ring-like structure at the contact point between particles and glass substrate. Each particle in Figure 1A is a silica particle with a patch of PMMA. A control experiment without polymer in toluene did not show such a feature (see Figure S2 in Supporting Information), confirming that the patch is due to PMMA. We have repeated the experiments with polystyrene solutions and produced single-patch silica particles with PS patches. (See Figure S3A in Supporting Information.) The quality of patches can be characterized by its size and shape. The patch size varies with initial concentration of polymer in the dipping solution. By analyzing over 100 number of patchy particles, the average size (diameter) of the patch is found to be 1.13 μm and standard deviation of 0.09 μm for 0.1% PMMA solution. For 0.2% PMMA solution (Figure 1B), average size and standard deviation are 1.63 and 0.07 μm, respectively. Figure 2 shows the distribution of patch size for
Figure 2. Histogram of the patch size distribution on a single patch silica particle of diameter 4.22 μm using PMMA in toluene at concentration of 0.1 and 0.2 wt %.
single-patch particles produced from 0.1 and 0.2 wt % polymer solution. It can be seen that the patch size distribution is wellseparated for these two concentrations of polymer. Therefore, the size of the patch can be controlled by varying the concentration of the polymer in the solution. This is because the thickness of the polymer film increases with concentration of polymer, which, in turn, increases the depth of immersion of particles in the polymer film. While most of the patches appeared as a ring-like structure (Figure 1A, inset), the shape of patch is found to depend on aging time of the dip-coated sample. To assess the shape of the patch, we have followed the time evolution of patch shape after
Figure 1. HR SEM images of silica particles with controlled PMMA patch size obtained by using (A) 0.1% PMMA solution (average patch size: 1.13 ± 0.09 μm) and (B) 0.2% PMMA solution (average patch size: 1.63 ± 0.07μm). The inset shows a closer look at the patch morphology. 1257
DOI: 10.1021/la503531a Langmuir 2015, 31, 1255−1261
Article
Langmuir 6, 12, 24, and 48 h of aging, and the important results are shown in Figure 3. The shape of patch depends on the aging
Figure 4A shows that a particle in a hexagonal lattice has six capillary bridges near its equatorial plane and another patch
Figure 3. HR SEM images showing the transformation of rink-like polymer patch to disk-like patch at different time intervals of aging (A) 24 and (B) 48 h.
Figure 4. (A) HR SEM image of polymer bridge formation and (B) multipatch PMMA-silica particles after redispersion.
expected at the bottom, leading to a seven-patch particle. Similar bridge formation is also observed for particles spread at PS solution (0.1 wt %): air interface, leading to the formation of seven patches of PS on silica particles. (See Figure S3B in the Supporting Information.) After redispersing the particles in deionized water, the particles still retain all of the seven patches, as shown in Figure 4B. Furthermore, we found the zeta potential of bare and multipatch particles to be −49 and −24.4 mV, respectively, at 3 mM aqueous NaCl solution. Unlike the single-patch particles, multipatch particles have seven patches, and hence a substantial decrease in zeta potential is manifested by the polymeric patches. The area of the glass substrate used for the synthesis is 125 mm2 as previously noted. Because the area fraction of the particles on the glass substrate is 0.71, the calculated number of particles on the glass substrate is 6.34 × 106. All of the particles on the substrate are redispersed into DI water by sonication, which is confirmed by visualizing the substrate by microscopy after sonication. If these particles are dispersed in 1 mL of water, then the resulting particle concentration will be 0.043% wt/vol. The number of patches on the particles and the orientation distribution of patches due to capillary bridge were analyzed prior to redispersion of the particles. This is due to the fact that it would be highly difficult to locate the position and orientation of the patch after redispersion. The number distribution of patches (calculated by analyzing 270 particles) on the multipatch particles obtained from the analysis of SEM images is shown in Figure 5A. It must be noted that in an ideal case each multipatch particle should have seven patches: six patches due to capillary bridge (as a result of hexagonal arrangement of particles in the monolayer) and one patch due to the polymer film in between the particle and the substrate. The frequency of particles with five, six, and seven patches is higher, however, there were particles with lower number of patches, which is probably due to inhomogeneous distribution of particles in the monolayer or issues related to cohesive failure of capillary bridges. In Figure 5B, the orientation distribution of the six patches from the capillary bridge is plotted. The particles with fewer than six patches have been omitted from this analysis. The vertical lines in Figure 5B represent the patch angle for an ideal case, that is, a patch at every apex of a regular hexagon (due to hexagonal arrangement of particles). As shown in the inset, the position of patch is defined by the patch angle with respect to center of the particles. The angle θ12, for example, displayed in the inset refers to angle made by patches 1 and 2. A similar rule is applicable to remaining patches as well. In the same manner, the angular position of all patches
time of the sample after dip coating, that is, on the evaporation of the solvent from the polymer film and the consolidation of the polymer sandwiched between the particle and the glass substrate. From the SEM images in Figure 3, the patch shape evolves from ring-like to disk-like with aging time. For low aging times (up to 24 h), the polymer layer is not completely dried under the experimental conditions and forms a ring upon peeling as depicted in Figure 3A. In this case, as the solvent evaporates from the three-phase contact line (at the particle− polymer−glass substrate), solidified polymer is left as a ring. The polymer film at the interior that is yet fluid-like is not transferred to the particle surface. After 48 h of aging, the solvent evaporates completely leaving a solid-like polymer layer beneath the particle surface and a thin disk of polymer is formed (Figure 3B). The yield of single patch particles is 100%, meaning that all of the particles dipcoated from the interface carry a patch as long as we ensure the formation of a close-packed monolayer of particles at the polymer solution−air interface before dip coating. The quantity of single patch particles produced depends on the area of the interface and the dimensions of the substrate. In our experiments, we used 12.5 × 5 mm2 area of glass substrate, and the coating of particles occurs on both sides of the substrate, leading to an effective area of 125 mm2. Because the particles are in a 2D close-packed (90.7%) state, and as diameter of the particles is 4.22 μm, 8.1 × 106 particles per substrate are synthesized. If these particles are dispersed in 1 mL of water, then the resulting particle concentration will be 0.057% w/v. If one was to use smaller particles, say, 2 μm diameter, then the resulting concentration for this case will be 0.25% w/v. Because the dip coating is a scalable procedure and can be automated, it is possible to obtain sizable quantity of particles per synthesis. 3.2. Multipatch Particles. To create multipatch particles the presence of particle−particle contact, which is a requirement for the synthesis of single patch particles, must now be eliminated. For example, when a 100 μL suspension of 4 wt % silica particles of 4.22 μm diameter is spread at the interface of PMMA solution (0.1 wt %) and air, a monolayer with a surface coverage of 71.41% (by image analysis of microscopy images) is obtained. Because the surface coverage is
