Self-Assembly of Colloidal Nanoparticles Inside Charged Droplets

Sep 25, 2013 - Department of Physics, Institut Teknologi Bandung, Jl Ganesha 10 Bandung 40132, Indonesia. ∥. Faculty of Chemical Engineering, Warsaw...
14 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Langmuir

Self-Assembly of Colloidal Nanoparticles Inside Charged Droplets during Spray-Drying in the Fabrication of Nanostructured Particles Asep Suhendi,† Asep Bayu Dani Nandiyanto,†,‡ Muhammad Miftahul Munir,§ Takashi Ogi,*,† Leon Gradon,∥ and Kikuo Okuyama† †

Department of Chemical Engineering, Graduate School of Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi Hiroshima 739-8527, Japan ‡ Departemen Kimia, Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam, Universitas Pendidikan Indonesia, Jl. Dr. Setiabudhi No. 229 Bandung 40154, Indonesia § Department of Physics, Institut Teknologi Bandung, Jl Ganesha 10 Bandung 40132, Indonesia ∥ Faculty of Chemical Engineering, Warsaw University of Technology, Waryńskiego 1, 00-645 Warsaw, Poland S Supporting Information *

ABSTRACT: Studies on self-assembly of colloidal nanoparticles during formation of nanostructured particles by spray-drying methods have attracted a large amount of attention. Understanding the self-assembly phenomenon allows the creation of creative materials with unique structures that may offer performance improvements in a variety of applications. However, current research on the self-assembly of colloidal nanoparticles have been conducted only on uncharged droplet systems. In this report, we first investigated the self-assembly processes of charged colloidal nanoparticles in charged droplets during spray-drying. Silica nanoparticles and polystyrene spheres are used as a model system. To induce a positive or a negative charge on the droplets, we used an electrospray method. Repulsive and attractive interactions between charged colloidal nanoparticles and droplet surface are found to control the self-assembly of colloidal nanoparticles inside the charged droplet. Interestingly, self-assembly of colloidal nanoparticles inside charged droplets under various processing parameters (i.e., droplet charge, droplet diameter, and surface charge, size, and composition of colloidal nanoparticles) allows the formation of unique nanostructured particles, including porous and hollow particles with control over the internal structure, external shape, number of hollow cavities, and shell thickness, in which this level of control cannot be achieved using conventional spray-drying method.

1. INTRODUCTION

and colloidal nanoparticles inside the droplet system, in which this droplet was produced from atomization of an initial solution/slurry.10 In general, when heat is added into the droplet, solvent evaporates, making the components inside the droplet to be self-assembled into their maximum and stable condition. Detailed information about the spray-drying method and their processing parameters are described in our previous works.9

Studies on self-assembly of colloidal nanoparticles during particle formation are of great fundamental scientific interest.1−4 Excellent performances that come from well-controlled self-assembly of nanoparticles make the final material useful in various applications, such as catalysts, thermal insulators, sensors, magnetic materials, printing materials, and drug delivery systems.5−8 One of the attractive studies on the self-assembly phenomena is investigation of the self-organization of components during spray-drying process. The spray-drying process is attractive because this method is facile and does not involve chemical reaction.9 Therefore, the process involves only transfer of heat © 2013 American Chemical Society

Received: June 18, 2013 Revised: September 24, 2013 Published: September 25, 2013 13152

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

Figure 1. Schematic of self-assembly of colloidal nanoparticles inside uncharged (a) and charged (b) droplets.

In the common spray-drying methods, droplet generation (atomization) involves a physical force. This physical force allows the generation of droplets without any charges (charge of droplets is neutral). Therefore, self-assembly of colloidal nanoparticles inside the droplet is mainly based on charge of colloidal nanoparticles only.11 Further, physical force also leads the generation of droplets with a broad size distribution and uncontrollable diameter, making most reports cannot discriminate the effect of droplet charge and diameter on the selfassembly of colloidal nanoparticles. In fact, charging of droplets and control of droplet diameter are likely to have a strong effect on the self-assembly of colloidal nanoparticles in the droplet. In our previous reports,11,12 we studied the self-assembly of colloidal nanoparticles using the spray-drying method and their effects on the formation of hollow and porous particles with controllable morphology. Effects of physicochemical properties of the colloidal nanoparticles (i.e., surface charge, size, and composition) were discussed and compared with theoretical calculation.11 However, our previous studies did not report about the influence of droplet charge and diameter yet. Information about self-assembly of colloidal nanoparticles inside charged droplet under various droplet diameters are virtually nonexistent. Here, the purpose of this study was to investigate the self-assembly of charged colloidal nanoparticles

inside charged droplet under various droplet diameters. To clarify our investigation in the self-assembly phenomenon inside a charged droplet, we varied droplet charge polarity, droplet diameter, and physicochemical properties of the colloidal nanoparticles (i.e., surface charge, size, and composition). For these investigations, we used a precursor containing a mixture of pure silica nanoparticles and surfactant-free polystyrene (PS) spheres. The combination of the silica and the PS colloidal nanoparticles was selected because this may allow the production of particles with a desirable nanostructure, such as porous and hollow structured particles, as reported in our previous works.11,13 In addition, to clarify the effect of the process parameters, the self-assembly process was conducted under surfactant-free conditions. To introduce a positive or a negative charge onto the droplet, we used an electrospray technique as an alternative to the conventional ultrasound or spinning disk atomization method.2,14−16 Using our electrospray method, monodispersed droplets with precise control on droplet diameter (narrow size distribution) and droplet charge polarity (positive or negative) can be generated, in which this can not be achieved using current existing spray-drying methods. Detailed information about how to control droplet charge and diameter in the electrospray is reported in our previous paper.17 To simplify the 13153

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

Figure 2. Experimental apparatuses for investigating self-assembly phenomena under charged droplet (a) (the electrospray-assisted spray-drying method and neutral droplet) and neutral droplet (b) (the ultrasound-assisted spray-drying methods).

addition to solvent evaporation and colloid−colloid interactions, there are additional effects, including: attractive and repulsive interactions between the colloidal particle and the droplet surface, ion emission, and mass reduction (Rayleigh limit theory).14,22 The combination of these factors may result in the formation of particles with various morphologies, which are different from a typical spray-drying method (sphericalshaped particles).

investigation into self-assembly of silica and PS colloidal nanoparticles during the formation of nanostructured particles, the method was then followed by a process to remove PS from the particles that were formed. In short, the results of selfassembly of colloidal nanoparticles could be more easily observed from the hole-structure of the particles remaining after removal of PS.11 Because to the best of our knowledge this report is the first to address self-assembly of charged colloidal nanoparticles inside charged droplets, we believe that this information may extend the application of self-assembly process to production of functional nanostructured materials.

3. EXPERIMENTAL SECTION 3.1. Precursor Preparation. To investigate self-assembly of colloidal nanoparticles inside a charged droplet, we used a precursor containing a mixture of pure silica nanoparticles (Nissan Chemical Co. Ltd., Japan; a mean size of 5 and 15 nm) and PS spheres (a mean size of 90 and 200 nm). PS spheres were synthesized from a simple polymerization of styrene monomer (styrene; Kanto Chemical Co., Inc., Japan) under surfactant-free conditions. Potassium persulfate (Sigma-Aldrich) or 2,2-azobis(isobutyramidine) dihydrochloride (Sigma-Aldrich) was used as an initiator in the styrene polymerization to produce PS spheres with a negative or a positive surface charge, respectively. To produce PS spheres with different diameters and charges, the concentrations of styrene and initiator were varied. A detailed synthetic method of PS is previously reported in our previous report.12 All chemicals were used without further purification. In brief, the precursor suspension was prepared by mixing suspensions of silica nanoparticles and PS spheres in ethanol with concentration of 0.15 wt %. To confirm the type of self-assembly process, we varied mass ratio of silica and PS colloidal nanoparticles. We also varied the size of both colloidal nanoparticles. This precursor was then transferred to the spray-drying apparatus. Two types of spray-drying equipment were used, namely, electrospray and ultrasound-assisted spray-drying methods (Figure 2). The electrospray-assisted spray-drying method was used to investigate self-assembly phenomena under charged droplet conditions, whereas the ultrasound-assisted spray-drying method to investigate selfassembly phenomena under neutral droplets.

2. HYPOTHETICAL SELF-ASSEMBLY OF COLLOIDAL NANOPARTICLES DURING SPRAY-DRYING Illustration of the self-assembly phenomena in the common spray-drying methods is shown in Figure 1a. For example, when particles are prepared from spray-drying of a precursor containing silica nanoparticles and PS spheres, the particle formation mechanism can be classified by two routes. A precursor containing a combination of negatively charged silica and anionic PS results in the formation of porous-structured particles, whereas a combination of negatively charged silica nanoparticles with cationic PS spheres gives formation of hollow particles.11 Although various hole-structured particles can be created, particles shapes are basically spherical.18−20 In general, the self-assembly phenomena involves the physicochemical properties of the colloidal nanoparticles. Detailed investigations about processing parameters are described in our previous works.11 When the spray-drying method is conducted under charged droplet conditions, additional factors in the colloidal selfassembly phenomena need to be considered (Figure 1b). Droplet charging adds electrostatic interactions to the colloidal nanoparticles during the self-assembly process.21 Therefore, in 13154

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

3.2. Self-Assembly of Colloidal Nanoparticles Inside Charged Droplet. To investigate self-assembly of colloidal nanoparticles inside charged droplets, the electrospray-assisted spray-drying method was used, shown in Figure 2a. This spray-drying system was similar to that used in our previous studies.17,23,24 Briefly, the apparatus consists of a syringe (Gastight 1000 μL; Hamilton; with a 27 gauge needle), a syringe pump (PHD2000; Harvard apparatus), a controllable DC high-voltage source (HVS; HER 10R3; Matsusada, Japan; conducted at 3.00 kV), a box chamber (a temperature of 40 °C and relative humidity of less than 30%), and a collection electrode (situated at 7.00 cm away from the needle and connected to the ground). After the precursor was added to the syringe, the precursor was sprayed, forming droplets. To control droplet diameters of 3.00, 3.50, 4.00, 4.50, 5.00, and 6.00 μm, we used flow rates of 1, 1.5, 2, 3, 4, and 6 μL/min, respectively. Droplets traveled to the collection electrode under an electrostatic force. As the droplet traveled, solvent evaporated, producing particles that were deposited on the collection electrode. The produced particles were then subjected to a PS removal process (a heat treatment at 500 °C; heating rate = 2 °C/min; holding time = 10 min). 3.3. Self-Assembly of Colloidal Nanoparticles Inside Uncharged Droplet. To investigate self-assembly of colloidal nanoparticles inside neutral droplet, the ultrasound-assisted spray-drying method was used (Figure 2b). Detailed synthetic apparatus and process are described in our previous works.12 Briefly, the apparatus consists of an ultrasonic nebulizer (model NE-U12, Omron Corp., Japan; equipped with a cyclone at an operating frequency of 1.7 MHz), a ceramic tubular furnace (diameter of 13 mm and length of 100 mm; two fixed temperature zones (200 and 600 °C)), and a filter. After the precursor was added to the ultrasonic nebulizer, micrometer-sized droplets were produced. Under a flow of nitrogen gas (0.75 L/min), the droplet passed through a tubular furnace (evaporating the solvent, removing the PS, and forming particles). The particles exiting the tubular furnace were collected by a filter. 3.4. Characterizations. The prepared particles were characterized using a scanning electron microscope (SEM; Hitachi S-5000, Hitachi, Japan; operated at 20 kV) and a transmission electron microscope (TEM; JEM-2010; JEOL, Japan; operated at 200 kV) to examine their size and morphology. The elemental and the chemical compositions of the prepared particles were evaluated using Fourier transmission infrared (FTIR; Spectrum One System; PerkinElmer; in the range of 600−4000 cm−1). The particles were also characterized using a thermal gravimetric and differential thermal analysis (TG-DTA; Exstar6000; Seiko Instruments Inc., Japan; heating rate of 5 °C/ min). The charges of the colloidal nanoparticles were measured using a zeta potential measurement system (Nano ZS; Malvern Zetasizer).

Figure 3. Physical properties of the initial colloidal nanoparticles used in the precursor. SEM images of colloidal nanoparticles: (a) 200 nm cationic PS spheres, (b) 200 nm anionic PS spheres, (c) 90 nm anionic PS spheres; (d) 5 nm silica nanoparticles; and (e) 15 nm silica nanoparticles. Panel (f) shows the zeta potential measurement results.

4. RESULTS 4.1. Investigation of Physicochemical Properties of Colloidal Nanoparticles. Physical properties of the colloidal nanoparticles in the precursor are presented in Figure 3. SEM images show monodispersed nanoparticles with a spherical shape. Zeta potential results indicated that silica nanoparticles with a mean diameter of 5 and 15 nm had a zeta value of −30 and −29 mV, respectively. The zeta potential of the cationic PS spheres with a mean diameter of 200 nm was +31 mV, whereas the anionic PS spheres with a mean diameter of 90 and 200 nm had a value of −36 and −32 mV, respectively. The properties of these colloidal nanoparticles were considered appropriate for investigation into self-assembly of colloidal nanoparticles in the particle formation.11 4.2. Effects of Droplet Diameter. Figure 4 shows lowmagnification SEM images of particles prepared with various droplet diameters. We found that the use of small droplet produced particles with a single spherical unit (Figure 4a) whereas increases in droplet diameter allow the production of particles with multi spherical units (Figure 4b and c).

Figure 4. Low-magnification SEM images of the electrosprayed particles using various droplet diameters. Samples were prepared using a droplet diameter of (a) 3.0, (b) 3.5 and (c) 4.0−5.0 μm.

The effects of droplet diameter on colloidal nanoparticle structure are detailed shown in Table 1. The number of spherical units (n) depended on adjustment of the droplet diameter, which was consistent with our previous work.23 Interestingly, when conducting the process using the same droplet diameter but different types of precursors (i.e., PS only, combination of PS and silica), an identical number of spherical components were found. However, the final particles featured changes to their outer surfaces, attributed additional silica nanoparticles in the PS solution giving higher surface 13155

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

Table 1. SEM Images and Schematic Models of Particles Prepared from a Precursor Containing Silica Nanoparticles and Anionic PS Spheres with Various Droplet Diametersa

a

Scale bars are 200 nm.

which the single spherical unit had a size of about 200 nm and rough surface. Different from produced particles in the conventional spraydrying method (shown in Figure 5c1 and c2), the morphological results in the spray-drying with charged droplet condition were not spherical. We also found that particles prepared under charged droplet have a good trend in producing monodispersed particles, compared with particles generated using the conventional spray-drying method (Figure 5c1 and c2). From the above results, the best condition to get monodispersed particles is when using positively charged droplet, in which this sample will be considered for further investigation in this study. In addition, we found free nanoparticles in the charged droplet product, in which this was due to particle ejection (Rayleigh limit theory). This result is unavoidable in the electrospray process. However, to reduce the number of free particles, we can optimize the process parameters, in which this will be done in our future work. 4.4. Effect of Physicochemical Properties of the Colloidal Nanoparticles. 4.4.1. Effects of Colloidal Surface Charge. Figure 6 shows effect of colloidal surface charge on self-assembly of colloidal nanoparticles examined by TEM

roughness. Experimental results for PS with opposite surface charge is shown in Table SI-1 in the Supporting Information. 4.3. Effect of Droplet Charge Polarity. Effects of droplet charge polarity (negative, positive, and neutral conditions) on the self-assembly of colloidal nanoparticles are shown in Figure 5. Figure 5a1 and a2 show particles prepared using positively charged droplets, whereas Figure 5b1 and b2 show particles prepared using negatively charged droplets. As a comparison, we also investigated the preparation of particles prepared using conventional spray-drying method (neutral charged droplets) (Figure 5c1 and c2). To confirm the effect of droplet charge on the self-assembly of colloidal nanoparticles, we used two types of PS spheres that were mixed with silica nanoparticles. The first type of PS is anionic PS spheres (Figure 5a1−c1), and the other type is cationic PS spheres (Figure 5a2−c2). The SEM images show various morphologies are produced. When using positively charged droplet, particles with daisy(Figure 5a1) and grapelike (Figure 5a2) structures were created. When negatively charged droplets were used, particles with irregular structures were produced (Figure 5b1 and b2). High-magnified SEM images in Figure 5a1, a2, b1, and b2 show that the final particles consisted of single spherical units, in 13156

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

Figure 5. SEM images of particles prepared using positive (a1, a2), negative (b1, b2), and neutral (c1, c2) droplet charge. Samples in a1−c1 were from combination of silica and anionic PS. Samples in a2−c2 were from combination of silica and cationic PS.

configurations and the formation of silica walls between holes (shown as red-dashed line in Figure 6a1 and b1). To confirm these colloidal charge effects, we also investigated the processes that occurred in droplets of different diameters. Changes in the droplet diameter influence the number of holes present in the particle but had no impact on the internal structure of the final hollow particles (shown in Figure 6a2 and b2). 4.4.2. Effect of Composition of Colloidal Nanoparticle. Figure 7 shows the electron microscope images of particles generated with various silica/PS mass ratios. All mass ratios resulted in formation of peanutlike particles with different shell structures, as shown in the SEM images in Figure 7a1 and b1. To confirm the structure inside the particle, TEM images in Figure 7a2 and b2 are shown. Particles containing two holes with different shell thicknesses were observed. Ferret analysis of

Figure 6. TEM images of the particles prepared with various colloidal particle surface charges and droplet diameters. Samples in a1 and a2 were prepared using anionic PS spheres, whereas samples in b1 and b2 were prepared using cationic PS spheres. Samples in a1 and b1 were prepared at a mean droplet diameter of 3.50 μm, whereas samples in a2 and b2 were 6.00 μm.

analysis. Particles having holes with diameters of about 200 nm were identified in all variations (i.e., colloidal surface charge and droplet diameter), verifying that these holes arose from removal of the PS spheres. Experimental results show different hole structures were found. When silica and anionic PS colloidal nanoparticles were electrosprayed, hollow particles with no silica walls were produced (Figure 6a1). However, when silica nanoparticles and cationic PS spheres were used, hollow particles with silica wall were formed (Figure 6b1). Changing the surface charge of the nanoparticles is found to have a significant impact on colloid−colloid interactions during the self-assembly process,11 as indicated by the different hole

Figure 7. SEM and TEM images of particles prepared with various silica/PS mass ratios. Samples in a1 and a2 were prepared using a silica/PS mass ratio of 0.33, whereas b1 and b2 were prepared with a ratio of 0.50. 13157

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

Figure 8. TEM images of the particles prepared with various silica/PS size ratios and droplet diameters. Samples were prepared using a silica/PS size ratio of (a, d) 15/200, (b) 5/200, and (c, e) 5/90, respectively. Samples in (a)−(c) were prepared at a mean droplet diameter of 3.50 μm, whereas samples in (d) and (e) were prepared at 6.00 μm.

the TEM images suggested that the shell thickness increased from 15 to 27 nm. The amount of silica attached to the surface of PS was increased with increasing silica/PS mass ratio from 0.33 to 0.50. 4.4.3. Effect of Size of Colloidal Nanoparticles. Figure 8 shows the TEM images of particles generated with various silica/PS size ratios. A strong correlation between component size and particle structure was found. When using a silica/PS size ratio of 15/200, hollow particles having hole sizes of 200 nm and a shell thickness of 42 nm were produced (Figure 8a). When using 5 nm silica nanoparticles, the shell thickness reduced to 15 nm, whereas the hole size was retained (Figure 8b). By changing the silica/PS size ratio to 5/90, hollow particles with hole diameters of 90 nm and a shell thickness of 15 nm were created (Figure 8c). If the PS size was varied but the silica size was maintained, no significant change in the shell thickness was observed. To confirm the colloidal size parameters, we also performed these processes using different droplet diameters (Figure 8d, e). Our TEM results showed that changing droplet diameter had no effect on hole diameter and shell thickness.

arrangement of PS spheres. Removal of the PS components leads to hollow-structured particles, in which this was confirmed by the appearance of hole in the broken particles shown in Table 1. This hollow structure was also confirmed using FTIR and TG-DTA analysis results, which are displayed in Figure SI-1 in the Supporting Information. The results demonstrate the possibility of producing porous and hollow particles with controllable internal structure, external shape, number of hollow cavities, and shell thickness by suitable changes in processing parameters. The particle inner structure could be controlled by modifying droplet and colloidal surface charges, whereas the particle diameter could be controlled in sizes ranging from hundreds to thousands of nanometers by changing droplet diameter. Hole size and shell thickness could be controlled by using colloidal nanoparticles with different physicochemical properties. 5.1. Effects of Droplet and Colloidal Surface Charges. Effects of droplet and colloidal surface charges on the selfassembly of colloidal nanoparticles inside the charged droplet are summarized in Figure 9. Figure 9 panels a and b show the cases of positively and negatively charged droplets, respectively, and Figure 9c shows the case of neutral droplets. Since the silica has a negative surface charge, we classify the particle formation into six possible routes, in which these routes are combinations of various droplet and PS surface charges. Route R1 describes the process that occurs from a combination of anionic PS and positively charged droplet. Silica and PS are in the same charge polarity, resulting repulsive interactions among colloidal nanoparticles inside the droplet. However, the charge of droplet surface and colloidal nanoparticles are in the opposite signs. Thus, attractive interactions between the colloidal nanoparticle and droplet surface happen. Most of colloidal nanoparticles are attracted to the surface of the droplet, and move to form a shell-like film. As the Brownian motion of the silica nanoparticles is faster than that of the PS spheres, the silica nanoparticles reach the surface earlier than the PS spheres. Therefore, the outer surface of the droplet is mostly composed of silica nanoparticles. When the solvent

5. DISCUSSION The work reported from the present study was primarily directed toward investigation of self-assembly of colloidal nanoparticles under various parameters (i.e., droplet charge, droplet diameters, and physicochemical properties of colloidal nanoparticles) in the spray-drying method. To clarify the effect of the process parameters, the self-assembly process was conducted under surfactant-free conditions. Based on the above results in Figure 5 and Table 1, all produced particles were from the arrangement of single spherical units. The size of a single spherical unit was identical to that of initial PS spheres. But, its outer surface was rougher than those of initial PS. These results confirm that the silica and the PS self-assemble during the spray-drying process. The PS spheres act as the major building block for the particle formation, while the silica components form a layer around the 13158

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

Figure 9. Schematic illustrations of colloidal nanoparticle self-assembly phenomena under different droplet and PS colloidal charges. Shown are the processes for (a) positive, (b) negative, and (c) neutral droplet charges, respectively. HP is defined as hollow particle.

R1, the Brownian motion of the colloidal nanoparticles also influences the self-assembly process. Unattached silica nanoparticles are likely going to the droplet surface faster than core− shell nanoparticles. After evaporating solvent, the core−shell are self-assembled and the unattached silica covers the surface of the arranged core−shell nanoparticles. As a result, hollow particles with silica walls are created, as shown in Figure 5a2, 6b1, and 6b2. Route R3 describes the process that occurs from a combination of anionic PS and negatively charged droplet. Silica, PS, and droplet surface are in the same charge polarity, resulting repulsive interactions among all charged components. All colloidal nanoparticles are repel each other. Due to repulsive action from droplet surface to colloidal nanoparticles, the colloidal nanoparticles are pushed to the center of the droplet. When the solvent evaporates, the droplet size shrinks and the droplet surface-to-particle distance shortens. This will increase the repulsive force between all charged components, causing expellant of some colloidal nanoparticles from their stable condition. Thus, the self-assembly of colloidal nanoparticles

evaporates, the shell that is formed from the self-assembly of colloidal nanoparticles is forced to the center of the droplet. The PS spheres are trapped and self-assemble at the center of droplet, while the silica nanoparticles collapse upon the arranged PS spheres covering them. As a consequence, hollow particles with no silica walls are formed in the final product, as shown in Figures 5a1, 6a1, and 6a2. Route R2 shows particle formation when using cationic PS spheres and positively charged droplets. In this route, silica and PS are in the opposite charge, resulting attractive interaction among colloidal nanoparticles since in the early stage of electrospraying process. Colloidal nanoparticle surface charge analysis can be found in the Figure SI-2 in the Supporting Information. Silica is attached on the surface of PS sphere, forming core−shell PS/silica nanoparticle in the precursor. Therefore, atomization process generates droplets containing core−shell and unattached silica nanoparticles. Because the charge of droplet surface and silica nanoparticles are in the opposite signs, the core−shell and unattached silica nanoparticles are attracted to the droplet surface. Similar to route 13159

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

Article

and size of colloidal nanoparticles). The effects of these parameters shown in Figures 7 and Figure 8 can be summarized in the following: (1) For the case of composition of colloidal nanoparticles (Figure 7), the colloidal composition influences the external diameter but has no impact on the external shape of the particle. Increases in the silica/PS colloidal composition are likely to increase the amount of silica attached on the surface of PS only. (2) For the case of size of colloidal nanoparticles (Figure 8), changes in PS size leads the modification of hole size, whereas changes in silica size permit the change in the shell thickness only. As a conclusion, the above results confirm that the PS and the silica govern the self-assembly process. During the selfassembly process, silica nanoparticles are going to the outer surface of the droplet faster than PS spheres (due to Brownian motion). Thus, when solvent evaporates, silica nanoparticles fill the outer part of the particle only. They do not exist in the deepest part of the particle (in the space between PS arrangements). Therefore, when using combination of silica and anionic PS spheres, the final particles are free of silica wall, and control of silica size and amount changes the shell thickness only. As silica and PS spheres were used as a model, other charged colloidal nanoparticles (e.g., metal particle) can be also potential to be used in initial solution. However, for conducting this process for other materials, we have to consider size, charge, and composition of colloidal nanoparticles used. Although the range of the colloidal composition and size studied here all showed successful synthesis of hollow particles, adjustment of size and composition of colloidal nanoparticles has some limitations. Large deviations from the optimal size and mass ratios led unsuccessful formation of hollow particles (shown in the Figure SI-3 in the Supporting Information).

resulting in the formation of particles with irregular shape, as shown in Figure 5b1. Route R4 shows particle formation when using cationic PS spheres and negatively charged droplets. Similar to route R2, core−shell PS/silica nanoparticles are produced from the beginning of spraying process. However, because the charged components (droplet surface, core−shell, and silica nanoparticle) are in the same signs, repulsive interactions between all charged components occur. As a result, expellant of some core−shell and silica nanoparticles happens. This results in the formation of hollow particles with various shapes, as shown in Figure 5b2. As a comparison for the self-assembly of colloidal nanoparticles in charged droplets, we also refer to the cases of particle formation in neutral droplets produced by conventional spray-drying method (routes R5 and R6). Similar to our previous reports,12 when silica nanoparticles and anionic PS spheres were used, porous structures were produced (route R5, Figure 5c1). In contrast, when silica nanoparticles and cationic PS spheres were employed, hollow particles with a spherical outer shape and smooth surfaces were obtained (route R6, Figure 5c2). In addition, because the droplets are neutral, when solvent evaporates, self-assembly process depends on physicochemical properties of colloidal nanoparticle only. Therefore, self-assembly can produce particles with the most stable structure. As a result all particles are spherical in shape. 5.2. Effect of Droplet Diameter. Experimental results in Table 1 show that the control of particle size can be achieved by adjusting the droplet diameter, driving the need of further investigations of this parameter. Droplet diameter of 3.00 μm leads the production of particles with a single spherical unit, and increases in droplet diameter can increase the number of spherical units. To confirm the relationship between droplet diameter and number of spherical units, the following equations can be used: ⎛ Q⎞ Dd = G(κ )⎜κεO ⎟ ⎝ K⎠

(1)

⎛ Q ⎞1/3 Dpeq = G(κ )⎜CκεO. ⎟ ⎝ K⎠

(2)

6. CONCLUSIONS The effects of processing parameters (i.e., droplet charge, droplet diameter, and surface charge, size, and composition of colloidal nanoparticles) on the self-assembly of colloidal nanoparticles in the formation of nanostructured particles have been studied. As a model of colloidal component, silica nanoparticles and PS spheres were used. Our experimental results show that changes in processing parameters affect on repulsive and attractive interactions among charged colloidal nanoparticles and droplet surface, making different selfassembly phenomena inside the droplet. The presented method allows the production of nanostructured particles with control over the external particle shape, the number of internal hollow cavities, shell thickness, and internal structure, in which this type of control cannot be achieved using common spray-drying methods. We believe that this information may extend the application of aerosol-assisted spray-drying processes to production of functional nanostructured particles.

1/3

Deq p

where Dd and are the diameter of droplet and the volume1/3 equivalent sphere diameter of final particle (Deq p = (6/πVp) ; Vp is the volume of a single particle), respectively. G(κ) is the experimental constant (−10.87κ −6/5 + 4.08κ −1/3). κ, K, and C are the dielectric, the conductivity, and the concentration of precursor, respectively. Q is the flow rate, and εO is the permittivity in vacuum. Substitution of above equations results Dpeq ≈ Dd (C)1/3

(3)

■

From eq 3, droplet diameter is proportional to particle size, and their relationship is linear, in which this result is in a good agreement with that in Table 1. Although the above calculation representation is effective for a specific condition (precursor concentration of 0.15 wt %, temperature of 40 °C, and humidity of less than 30%), this result offers a useful guide to the selection and the adjustment of droplet diameter for providing a process to produce particles with a specific size. 5.3. Effects of Composition and Size of Colloidal Nanoparticles. To confirm the self-assembly of colloidal nanoparticles inside charged droplet, we investigate effects of physical properties of colloidal nanoparticles (i.e., composition

ASSOCIATED CONTENT

* Supporting Information S

Effect of droplet diameter on the formation of particles prepared from a precursor containing silica nanoparticles and cationic PS spheres. FTIR and TG-DTA analysis result of samples containing “silica only”, “PS only”, and “silica and PS” (before and after the template removal process). Surface charge analysis of colloidal nanoparticles. Effect of silica/PS mass ratio on the particle formation. This material is available free of charge via the Internet at http://pubs.acs.org. 13160

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161

Langmuir

■

Article

(17) Suhendi, A.; Munir, M. M.; Suryamas, A. B.; Nandiyanto, A. B. D.; Ogi, T.; Okuyama, K. Control of cone-jet geometry during electrospray by an electric current. Adv. Powder Technol. 2013, 24 (2), 532−536. (18) Rosell-Llompart, J.; Fernandez de La Mora, J. Generation of monodisperse droplets 0.3 to 4 μm in diameter from electrified conejets of highly conducting and viscous liquids. J. Aerosol Sci. 1994, 25 (6), 1093−1119. (19) He, P.; Davis, S. S.; Illum, L. Chitosan microspheres prepared by spray drying. Int. J. Pharm. 1999, 187 (1), 53−65. (20) Wang, Y.; Zhang, F.; Xu, S.; Wang, X.; Evans, D. G.; Duan, X. Preparation of layered double hydroxide microspheres by spray drying. Ind. Eng. Chem. Res. 2008, 47 (15), 5746−5750. (21) Böttger, P. M.; Bi, Z.; Adolph, D.; Dick, K. A.; Karlsson, L. S.; Karlsson, M. N.; Wacaser, B. A.; Deppert, K. Electrospraying of colloidal nanoparticles for seeding of nanostructure growth. Nanotechnology 2007, 18 (10), 105304. (22) Kebarle, P.; Peschke, M. On the mechanisms by which the charged droplets produced by electrospray lead to gas phase ions. Anal. Chim. Acta 2000, 406 (1), 11−35. (23) Suhendi, A.; Nandiyanto, A. B. D.; Munir, M. M.; Ogi, T.; Okuyama, K. Preparation of agglomeration-free spherical hollow silica particles using an electrospray method with colloidal templating. Mater. Lett. 2013, 106, 432−435. (24) Suhendi, A.; Nandiyanto, A. B. D.; Ogi, T.; Okuyama, K. Agglomeration-free core-shell polystyrene/silica particles preparation using an electrospray method and additive-free cationic polystyrene core. Mater. Lett. 2013, 91, 161−164.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +81-82-424-7850. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledged the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan for providing scholarship for A.S. The authors thank R. Tsutsui and T. Iwaki for assistance in this research. This research was also supported by a Grant-in-Aid for Young Scientists (B) (No. 23760729) and Grant-in-Aid for Scientific Research (A) (No. 22246099) sponsored by MEXT.

■

REFERENCES

(1) Jana, N. R. Shape effect in nanoparticle self -assembly. Angew. Chem., Int. Ed. 2004, 43 (12), 1536−1540. (2) Sen, D.; Spalla, O.; Taché, O.; Haltebourg, P.; Thill, A. Slow drying of a spray of nanoparticles dispersion. In situ SAXS investigation. Langmuir 2007, 23 (8), 4296−4302. (3) Sen, D.; Mazumder, S.; Melo, J.; Khan, A.; Bhattyacharya, S.; D’souza, S. Evaporation driven self-assembly of a colloidal dispersion during spray drying: volume fraction dependent morphological transition. Langmuir 2009, 25 (12), 6690−6695. (4) Liu, W.; Wu, W. D.; Selomulya, C.; Chen, X. D. Facile spraydrying assembly of uniform microencapsulates with tunable core−shell structures and controlled release properties. Langmuir 2011, 27 (21), 12910−12915. (5) Taflin, D. C.; Ward, T. L.; Davis, E. J. Electrified droplet fission and the Rayleigh limit. Langmuir 1989, 5 (2), 376−384. (6) Bigioni, T. P.; Lin, X.-M.; Nguyen, T. T.; Corwin, E. I.; Witten, T. A.; Jaeger, H. M. Kinetically driven self assembly of highly ordered nanoparticle monolayers. Nat. Mater. 2006, 5 (4), 265−270. (7) Yang, J.; Lee, J.; Kang, J.; Lee, K.; Suh, J.-S.; Yoon, H.-G.; Huh, Y.-M.; Haam, S. Hollow silica nanocontainers as drug delivery vehicles. Langmuir 2008, 24 (7), 3417−3421. (8) Lee, J. H. Gas sensors using hierarchical and hollow oxide nanostructures: Overview. Sens. Actuators, B 2009, 140 (1), 319−336. (9) Nandiyanto, A. B. D.; Okuyama, K. Progress in developing spraydrying methods for the production of controlled morphology particles: From the nanometer to submicrometer size ranges. Adv. Powder Technol. 2011, 22 (1), 1−19. (10) Netting, D. I.; Hertzenberg, E. P. Method of preparing modified hollow, largely spherical particles by spray drying. US Patent US 3960583 A, 1976 (11) Nandiyanto, A. B. D.; Suhendi, A.; Arutanti, O.; Ogi, T.; Okuyama, K. Influences of surface charge, size, and concentration of colloidal nanoparticles on fabrication of self-organized porous silica in film and particle forms. Langmuir 2013, 29 (21), 6262−6270. (12) Nandiyanto, A. B. D.; Suhendi, A.; Ogi, T.; Iwaki, T.; Okuyama, K. Synthesis of additive-free cationic polystyrene particles with controllable size for hollow template applications. Colloids Surf., A 2012, 396, 96−105. (13) Blas, H.; Save, M.; Pasetto, P.; Boissiere, C.; Sanchez, C.; Charleux, B. Elaboration of monodisperse spherical hollow particles with ordered mesoporous silica shells via dual latex/surfactant templating: radial orientation of mesopore channels. Langmuir 2008, 24 (22), 13132−13137. (14) Gomez, A.; Tang, K. Charge and fission of droplets in electrostatic sprays. Phys. Fluids 1994, 6, 404. (15) Jaworek, A. Micro-and nanoparticle production by electrospraying. Powder Technol. 2007, 176 (1), 18−35. (16) Eninger, R. M.; Hogan, C. J., Jr.; Biswas, P.; Adhikari, A.; Reponen, T.; Grinshpun, S. A. Electrospray versus nebulization for aerosolization and filter testing with bacteriophage particles. Aerosol Sci. Technol. 2009, 43 (4), 298−304. 13161

dx.doi.org/10.1021/la403127e | Langmuir 2013, 29, 13152−13161