Electric-field Assisted Assembly of Colloidal Particles into Ordered

Article pubs.acs.org/Langmuir

Electric-Field Assisted Assembly of Colloidal Particles into Ordered Nonclose-Packed Arrays Jingjing Gong and Ning Wu* Department of Chemical and Biological Engineering, Colorado School of Mines 1500 Illinois Street, Golden, Colorado 80401, United States S Supporting Information *

ABSTRACT: Nonclose-packed colloidal arrays have many potential applications ranging from plasmonic sensors, light trapping for photovoltaics, to transparent electrodes. However, scalable fabrication of those structures remains a challenge. In this Article, we investigate the robustness of an electric-field assisted approach systematically. A monolayer of nonclosepacked crystalline array is first created under a low-frequency alternating-current electric field in solution. We then apply a sequence of direct-current pulses to fix the particle array onto the substrate so that it remains intact even after both field removal and solvent evaporation. Key process parameters such as the alternating-current field strength, direct-current magnitude, particle concentration, and solvent-evaporation rate that affect both ordering and fixing of colloidal particles have been studied systematically. We find that direct currents with an intermediate magnitude induce electrophoretic motion of particles toward the substrate and facilitate their permanent adhesion on the substrate due to strong van der Waals attraction. A higher current, however, causes lateral aggregation of particles arising from electroosmotic flow of solvent and destroys the periodic ordering between particles. This approach, in principle, can be conveniently adapted into the continuous convective assembly process, thus making the fabrication of nonclose-packed colloidal arrays scalable.

1. INTRODUCTION Two-dimensional (2D) crystals of colloidal particles have elicited considerable interest in recent years1−3 because of their potential applications in sensing,4−6 photovoltaics,7−9 superhydrophobic surfaces,10 and transparent electrodes.11 Closepacked 2D crystals have been fabricated via different methods including spin-coating,12 interfacial assembly,13−15 and convective flow-coating.16,17 On the other hand, nonclose-packed colloidal arrays with tunable interparticle spacing are also important because they can better serve as masks for nanosphere lithography and as templates for photonic crystals18,19 and metamaterials. A few methods have been developed recently. However, they all suffer from one or more problems such as high production cost, multiple fabrication steps, and low throughput. For example, photolithography is a top-down strategy that provides precise control of both feature size and spacing. However, it is limited by the complex fabrication process and types of materials that can be used.20 Controlled etching of close-packed 2D colloidal crystals11,21 is a bottom-up method. One typically starts with the fabrication of a close-packed colloidal monolayer, followed by reactive ion or plasma etching. It provides a reasonable control of the interparticle spacing, although it is usually small compared with the original particle size. Moreover, the etching process can sometimes change the particle shape and surface morphology significantly.22,23 Spin-coating of a mixture of © 2017 American Chemical Society

silica nanoparticles and ethoxylated trimethylolpropane triacrylate (ETPTA) is a robust method to fabricate ordered, nonclose-packed silica crystals.24 However, this method only works for specified types of particles and requires subsequent ETPTA polymerization and plasma etching. Interface-assisted assembly can also provide large interparticle spacing (usually 3−8 times the particle diameter).25 Unfortunately, this method only applies to particles with specifically tailored surface wettability, and the subsequent transfer to other substrates is challenging. Thus, a low-cost and scalable method that has little restriction on the types of particles to be used is highly desired. The assembly of colloidal particles under external electrical fields has also been studied extensively.26−35 Various types of structures have been reported. For example, Gong et al.1,35 reported several 2D colloidal phases such as crystalline, wormlike, and honeycomb structures formed under different alternating-current (ac) electric field strengths and geometric confinement. Although ordered, nonclose-packed colloidal structures can be obtained, the particles were suspended in solution. The structures would disassemble once the field is removed. On the other hand, electrophoresis has been widely used to deposit colloidal particles onto a conducting substrate Received: February 17, 2017 Revised: May 11, 2017 Published: May 17, 2017 5769

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776

Article

Langmuir using direct-current (dc) electric field. However, this method usually results in densely packed layers with little control on the order of particle packing.36−38 Liu et al.34 fine-tuned the equilibrium distance of 2D colloidal arrays with both frequency of the ac electric field and temperature. The nonclose-packed array was further fixed by a sudden switch to a dc field. However, the impacts of both dc current and evaporation rate on structure preservation over a large length scale were not studied in detail. Here we investigate the robustness of a coating process that combines ac and dc electric fields systematically. An ordered, nonclose-packed colloidal array is first obtained by applying a low-frequency ac electric field, due to the induced dipolar repulsion between particles. Then a sequence of dc pulses is employed to induce electrophoretic motion of particles toward the substrate. We show that, once the particle−substrate separation is small enough, the van der Waals force can adhere the particle array permanently onto the substrate. The 2D array can resist strong capillary attraction during solvent evaporation, and the ordered patterns remain undisturbed. Because it is a physical process that does not depend on the material type of the particles, such a method can be used for creating large-area, 2D, nonclose-packed particle arrays for a variety of applications.

Figure 1. (a) Cross-sectional view of the experimental setup. (b) The electric field applied combines both a continuous alternating current and pulses of direct currents. during the application of electric fields was also recorded using a highspeed monochrome CCD camera (SV643M, EPIX) connected to the microscope. After the nonclose-packed array was fixed on the bottom substrate, solvent was allowed to evaporate at different rates. The whole evaporation process was also recorded using the monochrome CCD camera. We employed algorithms developed by Crocker and Grier40 and by Blair and Dufresne41 to track the particles’ locations, from which the local bond order parameter ψ6 was calculated.

2. EXPERIMENTAL SECTION Materials. Styrene (>99%), sodium 4-vinylbenzenesulfonate (>90%), 2,2′-azobis(2-methylpropionitrile) (AIBN) (98%), poly(sodium 4-styrenesulfonate) (PSS, average Mw ≈ 70 000, 30 wt % in water), and methanol (ACS reagent-grade) were purchased from Sigma-Aldrich. All chemicals were used as received except that styrene was purified by aluminum oxide to remove inhibitors before use. Polystyrene microspheres of 2 μm in diameter functionalized with sulfonate groups were purchased from Interfacial Dynamics Corporation. Polystyrene particles of 700 nm in diameter were synthesized by using dispersion polymerization.39 In brief, a mixture of styrene, sodium 4-vinylbenzenesulfonate, and methanol was heated up to 70 °C; AIBN was then added into the solution as the initiator. After polymerization for 4 h, the particles were centrifuged (3000 rpm, IEC HT Centrifuge) for 20 min and washed with deionized (DI) water three times before being resuspended in aqueous solutions with desired concentration. Indium tin oxide (ITO) glasses (R = 70−100 Ω) were purchased from Delta Technologies. Methods. We used ITO glass slides as electrodes. They were first ultrasonically cleaned in acetone and isopropanol for 15 min sequentially. They were then exposed in oxygen plasma for 1 min and finally coated in PSS solution (5 mg/mL) for 15 min. The purpose of coating ITO glass slides with PSS is to make sure that the surface is sufficiently charged negatively. Because the colloidal particles are also negatively charged, this coating prevents them from sticking to the substrate before the electric field is applied for assembly. A 10 μm thick polyester film was used as the spacer to control the separation between the two electrodes. Figure 1 shows a cross-sectional view of our experimental setup, where a droplet of colloidal suspension was placed between two ITO slides and particles were allowed to settle near the bottom electrode for 5 min. An alternating-current (ac) electric field was then applied between the two electrodes using a function generator (model DG 1022, RIGOL). After tuning both frequency and field strength to appropriate values, we applied a sequence of dc pulses with a programmable dc power supply (model PWS 4323, Tektronix). To measure the current, we connected a 10 Ω resistor in series with experimental load. The voltage drop across the resistor was then measured by a multifunction data-acquisition card (USB-6001, National Instruments) and recorded by a LabVIEW program. The current was then calculated accordingly. During the experiments, images were acquired using a charge-coupled device (CCD) camera (Qimaging, Retiga 2000R) that was connected to an inverted microscope (IX71, Olympus). The movement of particles

3. RESULTS AND DISCUSSION Particle Assembly under an ac Electric Field. Before the electric field was applied, colloidal particles were dispersed in solution and underwent Brownian motion. As time lapsed, some of them settled near the bottom electrode due to the gravitational force. The coating of PSS on the electrode, however, prevented them from sticking to the substrate. Once the ac field was turned on, all particles were attracted toward the electrode and formed into a 2D nonclose-packed hexagonal array (Figure 2). The particles were separated from each other at a constant distance because of the induced dipolar

Figure 2. Two-dimensional and nonclose-packed particle array under ac electric field only. The frequency is 1 kHz, and the peak-to-peak voltage is 10 V. Inset shows the fast Fourier transform of the image. Scale bar: 10 μm. 5770

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776

Article

Langmuir repulsion.1,35,42 When an ac electric field was applied, each particle acquired an induced dipole moment,

p = 4πεsε0R3K E

where the angular brackets represents the configurational average, rjk is the distance between particles j and k, θjk is the angle between rjk and a reference x-axis, Nr is the number of nearest neighbors of a particle j, and N is the total number of particles used in the analyses. As shown in Figure 4, ψ6 remains

(1)

where ε0 is the vacuum permittivity, εs is the solvent dielectric constant, R is the particle radius, and E represents the externally applied field. K is the polarization coefficient, which depends on R, frequency ω, Debye length κ−1, and particle zeta potential ζp.43,44 It can be calculated analytically using the modified Dukhin−Shilov theory.45 Two particles with induced dipoles experience a dipolar interaction that can be written as ⎛ 2R ⎞ 3 πεmR2 |K |2 Erms 2⎜ ⎟ [(3cos2 θ − 1)r ̂ ⎝ r ⎠ 4 + (sin 2θ)θ]̂ 4

Fdip =

(2)

where r is the center-to-center distance between particles, Erms = Vpp/2 2 H is the root-mean-square of the applied field, Vpp is the peak-to-peak voltage (in terms of volts), and θ is the angle between the particle centers and the field direction. Because θ = 90° in our experiments, the particles experience a repulsive force and are separated from each other. We further studied the impact of particle concentration and voltage on the ordering of the nonclose-packed arrays. As shown in Figure 3, for a given particle concentration, a more

Figure 4. Dependence of the orientational bond order parameter on both voltage and particle concentration.

at moderate values even at high voltages if the particle concentration is low. In contrast, at high particle concentrations ψ6 increases with voltage and approaches unity quickly, indicating a high degree of order. Figure 5 shows the impact of applied voltage and particle concentration on the separation between particles. Although

Figure 5. Dependence of interparticle separation (scaled by particle diameter) on both voltage and particle concentration.

Figure 3. Two-dimensional and nonclose-packed arrays formed at different voltages and particle concentrations. The frequency (ω = 1 kHz) is kept constant in all experiments. Scale bar (20 μm) applies for all images.

increasing voltage improves the order of arrays, it has little impact on the scaled interparticle spacing L/2R. On the other hand, if we keep the field strength constant but increase particle concentration from 0.4 vol % to 1 vol %, we can tune the spacing from 4 to 1.5. Particle separation could influence the order of nonclose-packed arrays too. For example, the consistently low degree of order at 0.4 vol % can be attributed to the large particle separation (∼8 μm). According to eq 2, the dipolar repulsion is inversely proportional to r3. At the distance of 8 μm, the dipolar interaction is about kBT, which is too weak for ordering. Smaller spacings between particles (such as 4 μm) result in a much stronger dipolar repulsion and create more ordered arrays, as can be seen in Figure 3c. Fixing the Particle Array with Additional dc Pulses. Although the nonclose-packed colloidal arrays with tunable

ordered array can be obtained with increasing voltage. It can be attributed to a stronger dipolar repulsion between neighboring particles, which dominates over particle Brownian motion. For example, when the field strength is 105 V/m (∼6 V peak-to peak across 10 μm gap), the ratio of the dipolar interaction energy to thermal energy,46 Λ = πεsε0R3|K|2Erms2/2kBT, is ∼200. The array ordering can be further quantified by the orientational bond order parameter, ψ6, which is defined by47 ψ6 =

1 N

N

∑ j=1

1 Nr

Nr

∑ exp[6iθjk(rjk)] k=1

(3) 5771

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776

Article

Langmuir

Figure 6. Impact of direct current on fixing the particle array onto the substrate. (a) One pulse of 1.5 mA current applied over 0.5 s. (b) One pulse of 2.0 mA current applied over 0.5 s. (c) One pulse of 3.0 mA current applied over 0.5 s. (d) Two sequential pulses of 1.5 mA current applied over 10 s. The black dots represent the measured current. The colored dots are the lateral displacements of 238 particles measured every 0.1 s. The colors indicate the percentage of particles displaced at a certain magnitude. The inset images are snapshots showing the particle array before and after applying the current pulse(s), respectively.

spacings can be maintained under an ac electric field, the structures are not fixed. When the field is turned off, particles simply disassemble. Because of the double layer repulsion arising from their surface charges, particles are still separated from the substrate. However, because the particles are negatively charged, they can undergo electrophoretic motion toward the substrate if a direct current with the right bias is applied. This additional electrophoretic force could potentially make the particle−substrate separation small enough so that the van der Waals attraction between them immobilizes the particles on the substrate permanently. Therefore, we applied a direct current that passes through the two electrodes briefly. As shown in the insets of Figure 6a, an ordered particle array was first formed under the ac field (ω = 1 kHz and 20 Vpp). Particles underwent reduced but nonzero Brownian motion because of the dipolar repulsion between them. We measured the lateral displacements of 238 individual particles every 0.1 s (which are within 0.25R) and plotted them in colored dots in Figure 6a. The colors indicate the percentage of particles that undergo a specific range of lateral displacement. Then a direct current (black dots) of 1.5 mA was applied over a period of 0.5 s. The particle array looks similar (insets of Figure 6a) before and after applying the current, and the magnitude of the particle’s lateral displacements also barely changes (Supporting Information Movie 1). Therefore, this current is too weak to immobilize the particles onto the substrate. When we increased the current to 2 mA while keeping the pulse width the same (0.5 s), most of the particles stopped moving as reflected by far fewer dots in Figure 6b after the current was off. Although a small portion of the particles were still mobile, the magnitude of their displacements became much smaller (Supporting Information Movie 2). As we further increased the current to 3 mA, we found that many particles experienced significant lateral movement, which can be as large as one particle radius during the application of current (Figure 6c). This is primarily due to an electroosmotic flow36,48,49 surrounding the particles initiated

by the dc current, which entrains neighboring ones and induces the formation of aggregates. When the current is turned off, particle displacements are essentially zero, which indicates that all particles are immobilized (Supporting Information Movie 3). However, because of the strong electroosmotic flow when current is on, the degree of array order decreases dramatically, as can be discerned from the insets of Figure 6c. Parts a−c of Figure 6 indicate that there exists an optimal current level within which the particle array can be fixed onto the substrate without significant lateral movement due to electroosmotic flow. If the current is too low, the electrophoretic force is not strong enough for particles to move across the energy barrier into the regime where van der Waals attraction dominates. On the other hand, if the current is too high, a strong electroosmotic flow will lead to significant particle displacement along the substrate. Although particles can be immobilized afterward, the original periodic spacing is disturbed. Inspired by these observations, we further tested to fix the particle arrays by applying a sequence of multiple low current pulses. Figure 6d shows a typical result. We applied two sequential pulses (1.5 mA over 0.5 s), between which there was a 10 s off period. It can be seen that, after the first pulse, some particles were immobilized while many others exhibited significantly reduced Brownian motion. After the second pulse, all particles were essentially immobilized as indicated by those zero-valued dots. The order of particle arrays can be preserved by this approach. Modeling the Vertical Movement of Particles. To understand the impact of direct current on the movement and potential fixing of colloidal particles onto the substrate, we developed a theoretical model. Applying a force balance on the particle with the inertia-free assumption yields50,51 α(h)6πμR 5772

dh = FEP + FvdW + FY + Fg dt

(4)

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776

Article

Langmuir The left-hand side of eq 4 is the drag force on a particle moving toward the substrate,52 where μ is the solvent viscosity and h is the separation between particle surface and substrate. The coefficient α(h) accounts for the confining effect of the substrate,53 α(h) =

6h2 + 9Rh + 2R2 6h2 + 2Rh

(5)

Four terms on the right-hand side of eq 4 are the electrophoretic, van der Waals, double layer (i.e., Yukawa), and buoyancy forces acting on the particle, respectively. Their expressions are readily available from literature52 FEP = β(h)6πεsε0ζpREdc

(6)

where β(h) = (h + 1.554R)/(h + 0.3R) is a coefficient that accounts for the wall effect.48 The electric field strength Edc is related to the dc current I, solvent conductivity σ, and the cross-sectional area of the suspension droplet A, i.e., Edc = I/Aσ. The van der Waals force is given by

FvdW = −

AHR 2

Figure 7. Calculated time evolution of the particle−substrate separation under (a) different currents with a constant substrate zeta potential ζs = −50 mV and (b) a constant current of 1.5 mA but with different substrate zeta potentials.

(7)

6h

−20

where AH = 2.8 × 10 J is the Hamaker constant for ITO/ polystyrene system.52 The double layer interaction between the particle and substrate has a Yukawa form54 FY = B exp( −κh)

substrate further. A similar result is obtained under a current of 1.5 mA, except that the particle−substrate separation is closer due to a larger driving force. With a further increased current of 2.0 mA, the particle’s vertical movement is qualitatively different. Although it moves toward the bottom substrate with a decreasing velocity initially, after it passes the energy barrier where the double layer repulsion dominates, both van der Waals and electrophoretic forces make the particle approach the substrate with an increasing velocity until the separation becomes zero, as illustrated by the inset of Figure 7a. These calculation results including the magnitude of current required for fixing the particles capture our experimental results well. Figure 7b shows the particle−substrate separation under a constant current (1.5 mA) but different substrate zeta potentials. When the substrate is highly charged (e.g., ζs = −60 mV), the particle eventually stays at a small separation from the substrate because of the strong double layer repulsion. With lower ζs, the repulsion becomes weaker and the particle can be fixed on the substrate eventually. This calculation result illustrates the importance of substrate zeta potential. Qualitatively similar results can be expected when changing the zeta potential of particles. Ideally, one would prefer using a weakly charged or neutral substrate because a lower current is more desirable for fixing the particle array. However, it increases the risk of unfavorable particle adhesion even before the application of ac electric field for assembly. In our experiments, we also find that multiple pulses of low dc current can fix all particles eventually. This can be explained by the gradual decrease of zeta potentials of both substrate and particle after each round. The applied dc current leads to water electrolysis, and the electrochemical reactions near the two electrodes in an acidic aqueous solution (because of dissolved carbon dioxide from ambient environment) are33 1 3H 2O → 2H3O+ + O2 ↑ + 2e− (anode) (11) 2

(8)

where B = 64πεsε0κR(kT/e) tanh(eζp/4kT) tanh(eζs/4kT), kT is the thermal energy, e is the elemental charge, and ζs is the substrate zeta potential. Finally, the buoyancy force is 4 Fg = − πR3(ρp − ρs )g (9) 3 2

where ρp and ρs are the densities of particle and solvent, respectively, and g is the gravitational acceleration. Substituting eqs 5−9 into eq 4 yields B exp( −κh0h ̅ ) AHR β (h ̅ ) dh ̅ = − + 2 2 dt ̅ α (h ̅ ) λ 6λh0 h ̅ −

4πR3(ρp − ρs )g 3λ

(10)

where λ = 6πα(h̅)Rεsε0ζpEdc, h̅ = h/h0, and t ̅ = t/t0. h0 is the equilibrium separation between particle and substrate under the ac electric field only. It can be solved by equating the sum of double layer, van der Waals, and gravitational forces to zero. Because the electrophoretic force is dominating during the particle’s vertical movement, the time scale t0 is chosen so that the first term in right-hand side of eq 10 is O(1). Therefore, t0 = μh0/εsε0ζpEdc. When h̅ approaches zero, all terms on the righthand side of eq 10 vanish except the second one. The dominant van der Waals force at small separations provides the mechanism for permanent fixing of the particles. We solve eq 10 numerically with the initial condition of t = 0, h = h0. Figure 7a shows the time evolution of the particle− substrate separation subject to different dc currents. We assume that the substrate zeta potential is similar to the particles, i.e., ζs = −50 mV. Under 1 mA, the particle initially moves toward the bottom substrate, but it eventually reaches a new equilibrium height where the double layer repulsion between the particle and the substrate is essentially balanced with the electrophoretic driving force, preventing it from approaching the

2H3O+ + 2e− → 2H 2O + H 2 ↑ (cathode) 5773

(12)

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776

Article

Langmuir

small clusters, the overall pattern quality was significantly improved. Moreover, the orientational bond order parameter ψ6 was much higher than those in Figure 8a and b. To further improve the array quality, we kept the ac field on while we let the solvent evaporate slowly. Figure 9a shows the

Both theoretical and experimental studies have shown that pH near electrodes can change dramatically.55−57 In particular, the concentration of protons will increase near the anode (i.e., the bottom substrate), leading to a lower pH. Because the pKa of poly(4-styrenesulfonic acid) is ∼1,58,59 it is likely that the surface charges on both substrate and particles (which are functionalized with sulfonate groups) decrease after each round of dc pulses. The reduced zeta potentials decrease the energy barrier that a particle needs to overcome before entering into the regime where van der Waals force dominates. We note that the same mechanism applies for fixing positively charged particles on the cathode too. In this regard, multiple pulses of low direct current produces the best result for fixing particle arrays because of the associated change of pH in solution and zeta potentials on substrate/particle. Solvent Evaporation. Although particles have been fixed onto the bottom substrate, they are still immersed in solvent. To obtain a dry particle array, the solvent needs to be evaporated. We have tested both fast and slow drying processes. For fast drying, we turned off the electric field, removed the top ITO glass, and let the solvent evaporate naturally at room temperature, which only took about 1 or 2 min. As shown in Figure 8a, the ordered particle array can be preserved in some

Figure 9. (a) Larger area of dried two-dimensional and nonclosepacked colloidal array. Optical images for multiple regimes across the sample are shown. (b) Array fabricated by 700 nm polystyrene particles. Scale bars: 20 μm.

optical images in multiple regimes over an area of ∼cm2 colloidal array. Clearly, the overall array order has been wellpreserved. Supporting Information Movie 4 shows the in situ observation of the solvent-evaporation dynamics. It can be seen that solvent evaporates in two stages. First, the meniscus that separates the bulk fluid and a much thinner layer of water sweeps gradually from the edge of the cell toward the center (from bottom to top in this movie). During this stage, no significant particle movement or disturbance is observed, indicating that the thickness of the water film is larger than that of the particle diameter. Next, the thin layer of water remaining on the substrate starts to evaporate, via a nucleation process that closely mimics the thin-film dewetting instability.60 It is clear that that some particles undergo aggregation quickly due to strong capillary forces, although most of them stay intact. The method we have developed can also be used to make colloidal arrays with smaller particles as long as the field strength (which is controlled by both the applied voltage and electrode separation) is sufficiently strong to generate dipolar repulsion that can overcome Brownian motion for array ordering. The fixing strategy by applying a sequence of dc pulses is also universal, and the resulting array can survive solvent evaporation. Figure 9b shows the representative optical images for a two-dimensional, nonclose-packed array fabricated from 700 nm polystyrene particles.

Figure 8. Representative optical images for the dried colloidal array in different areas on an ITO substrate under different drying conditions. (a) 1 dc pulse followed by fast solvent evaporation; (b) 1 dc pulse followed by slow solvent evaporation; and (c) 10 dc pulses followed by fast solvent evaporation. Scale bars: 20 μm.

areas. However, in many other areas, particles aggregated during solvent evaporation due to strong capillary attractions between neighboring particles as the water film became thinner. We have also tried a slow drying process. After applying the direct current and removing the ac field, the whole setup was left undisturbed and solvent evaporated slowly through the edge between the two electrodes, which took a few hours to complete. Figure 8b shows the representative optical images of the dried array in different areas. Although the degree of particle aggregation was reduced as compared with fast drying, the final pattern over a large area is still of poor quality. We then introduced multiple rounds of direct-current pulses before we removed the electric field and let the solvent evaporate. Figure 8c shows the final colloidal array under the fast drying condition. In general, most of the particles remained undisturbed. Although several of them still aggregated into 5774

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776

Article

Langmuir

(2) Yang, H.; Gozubenli, N.; Fang, Y.; Jiang, P. Generalized Fabrication of Monolayer Nonclose-Packed Colloidal Crystals with Tunable Lattice Spacing. Langmuir 2013, 29 (25), 7674−7681. (3) Cai, Z.; Liu, Y. J.; Leong, E. S. P.; Teng, J.; Lu, X. Highly Ordered and Gap Controllable Two-Dimensional Non-Close-Packed Colloidal Crystals and Plasmonic–Photonic Crystals with Enhanced Optical Transmission. J. Mater. Chem. 2012, 22 (47), 24668−24675. (4) Wang, Y.; Zhao, X.; Chen, L.; Chen, S.; Wei, M.; Gao, M.; Zhao, Y.; Wang, C.; Qu, X.; Zhang, Y.; et al. Ordered Nanocap Array Composed of SiO 2 -Isolated Ag Islands as SERS Platform. Langmuir 2014, 30 (50), 15285−15291. (5) Willets, K. A.; Van Duyne, R. P. Localized Surface Plasmon Resonance Spectroscopy and Sensing. Annu. Rev. Phys. Chem. 2007, 58 (1), 267−297. (6) Moskovits, M. Surface-Enhanced Spectroscopy. Rev. Mod. Phys. 1985, 57 (3), 783−826. (7) Zhu, J.; Yu, Z.; Fan, S.; Cui, Y. Nanostructured Photon Management for High Performance Solar Cells. Mater. Sci. Eng., R 2010, 70 (3−6), 330−340. (8) Atwater, H. a; Polman, A. Plasmonics for Improved Photovoltaic Devices. Nat. Mater. 2010, 9 (3), 205−213. (9) Brongersma, M. L.; Cui, Y.; Fan, S. Light Management for Photovoltaics Using High-Index Nanostructures. Nat. Mater. 2014, 13 (5), 451−460. (10) Yang, H.; Dou, X.; Fang, Y.; Jiang, P. Self-Assembled Biomimetic Superhydrophobic Hierarchical Arrays. J. Colloid Interface Sci. 2013, 405, 51−57. (11) Morfa, A. J.; Akinoglu, E. M.; Subbiah, J.; Giersig, M.; Mulvaney, P. Transparent Metal Electrodes from Ordered Nanosphere Arrays. J. Appl. Phys. 2013, 114 (5), 054502. (12) Deckman, H. W.; Dunsmuir, J. H. Natural Lithography. Appl. Phys. Lett. 1982, 41 (4), 377. (13) Zhang, J.; Wang, L.; Lamont, D. N.; Velankar, S. S.; Asher, S. A. Fabrication of Large-Area Two-Dimensional Colloidal Crystals. Angew. Chem., Int. Ed. 2012, 51 (25), 6117−6120. (14) Meng, X.; Qiu, D. Gas-Flow-Induced Reorientation to Centimeter-Sized Two-Dimensional Colloidal Single Crystal of Polystyrene Particle. Langmuir 2014, 30 (11), 3019−3023. (15) Dai, Z.; Li, Y.; Duan, G.; Jia, L.; Cai, W. Phase Diagram, Design of Monolayer Binary Colloidal Crystals, and Their Fabrication Based on Ethanol-Assisted Self-Assembly at the Air/Water Interface. ACS Nano 2012, 6 (8), 6706−6716. (16) Prevo, B. G.; Velev, O. D. Controlled, Rapid Deposition of Structured Coatings from Micro- and Nanoparticle Suspensions. Langmuir 2004, 20 (6), 2099−2107. (17) Kumnorkaew, P.; Ee, Y.-K.; Tansu, N.; Gilchrist, J. F. Investigation of the Deposition of Microsphere Monolayers for Fabrication of Microlens Arrays. Langmuir 2008, 24 (21), 12150− 12157. (18) Lin, S.-Y.; Chow, E.; Hietala, V.; Villeneuve, P. R.; Joannopoulos, J. D. Experimental Demonstration of Guiding and Bending of Electromagnetic Waves in a Photonic Crystal. Science 1998, 282 (5387), 274−276. (19) Hoogenboom, J. P.; Rétif, C.; De Bres, E.; Van De Boer, M.; Van Langen-Suurling, A. K.; Romijn, J.; Van Blaaderen, A. TemplateInduced Growth of Close-Packed and Non-Close-Packed Colloidal Crystals during Solvent Evaporation. Nano Lett. 2004, 4 (2), 205−208. (20) Tseng, A. A.; Chen, K.; Chen, C. D.; Ma, K. J. Electron Beam Lithography in Nanoscale Fabrication: Recent Development. IEEE Trans. Electron. Packag. Manuf. 2003, 26 (2), 141−149. (21) Hsu, C.; Connor, S. T.; Tang, M. X.; Cui, Y. Wafer-Scale Silicon Nanopillars and Nanocones by Langmuir−Blodgett Assembly and Etching. Appl. Phys. Lett. 2008, 93 (13), 133109. (22) Haginoya, C.; Ishibashi, M.; Koike, K. Nanostructure Array Fabrication with a Size-Controllable Natural Lithography. Appl. Phys. Lett. 1997, 71 (20), 2934. (23) Tan, B. J.-Y.; Sow, C.-H.; Lim, K.-Y.; Cheong, F.-C.; Chong, G.L.; Wee, A. T.-S.; Ong, C.-K. Fabrication of a Two-Dimensional

4. CONCLUSIONS We investigated an electric-field assisted coating approach to fabricate two-dimensional and nonclose-packed colloidal arrays onto an ITO glass. Ordered particle arrays were first obtained under low-frequency ac electric fields due to induced dipolar repulsion between neighboring particles. The impacts of both field strength and particle concentration on the array ordering and interparticle spacing have been studied systematically. After the array formation, we applied a direct-current pulse across both electrodes, which induced electrophoretic motion of the particles toward the substrate. For a single pulse, we found that currents with an intermediate magnitude were optimal for fixing the particle array onto the substrate, as a lower one is insufficient for the particles to move across the energy barrier where the double layer repulsion between the particle and substrate dominates. A higher current will also cause strong electroosmotic flow that induces lateral particle aggregation and destroys the array ordering. Alternatively, a sequence of multiple pulses of low current can also fix the particles onto the substrate. This can be attributed to the increase of proton concentration (i.e., decrease in pH) near the anode during electrolysis of water, which decreases the magnitude of zeta potentials for both particle and substrate. Under the condition of slow solvent evaporation, the particle array can resist strong capillary attraction and remain intact after removing the field. We note that such a process has the potential to be further adopted in a continuous convective assembly fashion16,61 for scalable production of nonclose-packed, two-dimensional particle arrays.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.7b00547. Full captions of the SI movie files (PDF) Particle array before and after applying a 1.5 mA dc current (AVI) Particle array before and after applying a 2.0 mA dc current (AVI) Particle array before and after applying a 3.0 mA dc current (AVI) Particle array during slow solvent evaporation (AVI)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected]. ORCID

Ning Wu: 0000-0002-2167-3621 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Bay Area Photovoltaic Consortium (BAPVC) under Award DE-EE0004946 and National Science Foundation (CBET-1336893).

■

REFERENCES

(1) Gong, T.; Wu, D. T.; Marr, D. W. M. Two-Dimensional Electrohydrodynamically Induced Colloidal Phases. Langmuir 2002, 18 (26), 10064−10067. 5775

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776

Article

Langmuir Periodic Non-Close-Packed Array of Polystyrene Particles. J. Phys. Chem. B 2004, 108 (48), 18575−18579. (24) Jiang, P.; McFarland, M. J. Large-Scale Fabrication of Wafer-Size Colloidal Crystals, Macroporous Polymers and Nanocomposites by Spin-Coating. J. Am. Chem. Soc. 2004, 126 (42), 13778−13786. (25) Isa, L.; Kumar, K.; Müller, M.; Grolig, J.; Textor, M.; Reimhult, E. Particle Lithography from Colloidal Self-Assembly at Liquid−Liquid Interfaces. ACS Nano 2010, 4 (10), 5665−5670. (26) Ma, F.; Wang, S.; Smith, L.; Wu, N. Two-Dimensional Assembly of Symmetric Colloidal Dimers under Electric Fields. Adv. Funct. Mater. 2012, 22 (20), 4334−4343. (27) Ma, F.; Wang, S.; Zhao, H.; Wu, D. T.; Wu, N. Colloidal Structures of Asymmetric Dimers via Orientation-Dependent Interactions. Soft Matter 2014, 10 (41), 8349−8357. (28) Ma, F.; Yang, X.; Zhao, H.; Wu, N. Inducing Propulsion of Colloidal Dimers by Breaking the Symmetry in Electrohydrodynamic Flow. Phys. Rev. Lett. 2015, 115 (20), 208302. (29) Ma, F.; Wang, S.; Wu, D. T.; Wu, N. Electric-Field−induced Assembly and Propulsion of Chiral Colloidal Clusters. Proc. Natl. Acad. Sci. U. S. A. 2015, 112 (20), 6307−6312. (30) Woehl, T. J.; Chen, B. J.; Heatley, K. L.; Talken, N. H.; Bukosky, S. C.; Dutcher, C. S.; Ristenpart, W. D. Bifurcation in the Steady-State Height of Colloidal Particles near an Electrode in Oscillatory Electric Fields: Evidence for a Tertiary Potential Minimum. Phys. Rev. X 2015, 5 (1), 1−11. (31) Dutcher, C. S.; Woehl, T. J.; Talken, N. H.; Ristenpart, W. D. Hexatic-to-Disorder Transition in Colloidal Crystals Near Electrodes: Rapid Annealing of Polycrystalline Domains. Phys. Rev. Lett. 2013, 111 (12), 128302. (32) Ristenpart, W. D.; Jiang, P.; Slowik, M. a.; Punckt, C.; Saville, D. a.; Aksay, I. a. Electrohydrodynamic Flow and Colloidal Patterning near Inhomogeneities on Electrodes. Langmuir 2008, 24 (21), 12172− 12180. (33) Ristenpart, W. D.; Aksay, I. a.; Saville, D. a. Electrically Driven Flow near a Colloidal Particle close to an Electrode with a Faradaic Current. Langmuir 2007, 23 (7), 4071−4080. (34) Liu, Y.; Xie, R.; Liu, X. Fine Tuning of Equilibrium Distance of Two-Dimensional Colloidal Assembly under an Alternating Electric Field. Appl. Phys. Lett. 2007, 91 (6), 063105. (35) Gong, T.; Marr, D. W. M. Electrically Switchable Colloidal Ordering in Confined Geometries. Langmuir 2001, 17 (8), 2301− 2304. (36) Solomentsev, Y.; Bevan, M.; Anderson, J. L. Aggregation Dynamics for Two Particles during Electrophoretic Deposition under Steady Fields. Langmuir 2000, 16 (24), 9208−9216. (37) Kim, J.; Guelcher, S. a.; Garoff, S.; Anderson, J. L. Two-Particle Dynamics on an Electrode in Ac Electric Fields. Adv. Colloid Interface Sci. 2002, 96 (1−3), 131−142. (38) Bazin, D.; Chevalier, S.; Saadaoui, H.; Santarelli, X.; Larpent, C.; Feracci, H.; Faure, C. Electrodeposition of Polymer Nanodots with Controlled Density and Their Reversible Functionalization by Polyhistidine-Tag Proteins. Langmuir 2012, 28 (39), 13968−13975. (39) Zhang, F.; Cao, L.; Yang, W. Preparation of Monodisperse and Anion-Charged Polystyrene Microspheres Stabilized with Polymerizable Sodium Styrene Sulfonate by Dispersion Polymerization. Macromol. Chem. Phys. 2010, 211 (7), 744−751. (40) Crocker, J. C.; Grier, D. G. Introduction, I. Methods of Digital Video Microscopy for Colloidal Studies. J. Colloid Interface Sci. 1996, 179 (179), 298−310. (41) Blair, D.; Dufresne, E. The Matlab Particle Tracking Code Repository http://site.physics.georgetown.edu/matlab/. (42) Mittal, M.; Lele, P. P.; Kaler, E. W.; Furst, E. M. Polarization and Interactions of Colloidal Particles in Ac Electric Fields. J. Chem. Phys. 2008, 129 (6), 064513. (43) Zhao, H. Double-Layer Polarization of a Non-Conducting Particle in an Alternating Current Field with Applications to Dielectrophoresis. Electrophoresis 2011, 32 (17), 2232−2244. (44) Zhou, H.; Preston, M. A.; Tilton, R. D.; White, L. R. Calculation of the Electric Polarizability of a Charged Spherical Dielectric Particle

by the Theory of Colloidal Electrokinetics. J. Colloid Interface Sci. 2005, 285 (2), 845−856. (45) Shilov, V. N.; Delgado, A. V.; Gonzalez-Caballero, F.; Grosse, C. Thin Double Layer Theory of the Wide-Frequency Range Dielectric Dispersion of Suspensions of Non-Conducting Spherical Particles Including Surface Conductivity of the Stagnant Layer. Colloids Surf., A 2001, 192 (1−3), 253−265. (46) Yethiraj, A. Tunable Colloids: Control of Colloidal Phase Transitions with Tunable Interactions. Soft Matter 2007, 3 (9), 1099. (47) Nelson, D. R.; Halperin, B. I. Dislocation-Mediated Melting in Two Dimensions. Phys. Rev. B: Condens. Matter Mater. Phys. 1979, 19 (5), 2457−2484. (48) Solomentsev, Y.; Bö hmer, M.; Anderson, J. L. Particle Clustering and Pattern Formation during Electrophoretic Deposition:A Hydrodynamic Model. Langmuir 1997, 13 (23), 6058−6068. (49) Böhmer. In Situ Observation of 2-Dimensional Clustering during Electrophoretic Deposition. Langmuir 1996, 12 (24), 5747− 5750. (50) Rock, R. M.; Sides, P. J.; Prieve, D. C. The Effect of Electrode Kinetics on Electrophoretic Forces. J. Colloid Interface Sci. 2013, 393 (1), 306−313. (51) Wirth, C. L.; Sides, P. J.; Prieve, D. C. The Imaging Ammeter. J. Colloid Interface Sci. 2011, 357 (1), 1−12. (52) Fagan, J. a.; Sides, P. J.; Prieve, D. C. Vertical Oscillatory Motion of a Single Colloidal Particle Adjacent to an Electrode in an Ac Electric Field. Langmuir 2002, 18 (21), 7810−7820. (53) Goldman, A. J.; Cox, R. G.; Brenner, H. Slow Viscous Motion of a Sphere Parallel to a Plane wallII Couette Flow. Chem. Eng. Sci. 1967, 22 (4), 653−660. (54) Israelachvili, J. N. Intermolecular and Surface Forces; Academic press, 2011. (55) De, D.; Nicholson, P. S. Role of Ionic Depletion in Deposition during Electrophoretic Deposition. J. Am. Ceram. Soc. 1999, 82 (11), 3031−3036. (56) Sarkar, P.; De, D.; Uchikochi, T.; Besra, L. Electrophoretic Deposition (EPD): Fundamentals and Novel Applications in Fabrication of Advanced Ceramic Microstructures. In Electrophoretic Deposition of Nanomaterials; Dickerson, J. H., Boccaccini, A. R., Eds.; Springer: New York: New York, NY, 2012; pp 181−215. (57) Besra, L.; Liu, M. A Review on Fundamentals and Applications of Electrophoretic Deposition (EPD). Prog. Mater. Sci. 2007, 52 (1), 1−61. (58) Hua, F.; Shi, J.; Lvov, Y.; Cui, T. Fabrication and Characterization of Metal Oxide Semiconductor Capacitor Based on Layer-byLayer Self-Assembled Thin Films. Nanotechnology 2003, 14 (4), 453− 457. (59) Kharlampieva, E.; Tsukruk, T.; Slocik, J. M.; Ko, H.; Poulsen, N.; Naik, R. R.; Kröger, N.; Tsukruk, V. V. Bioenabled SurfaceMediated Growth of Titania Nanoparticles. Adv. Mater. 2008, 20 (17), 3274−3279. (60) Thiele, U.; Mertig, M.; Pompe, W. Dewetting of an Evaporating Thin Liquid Film: Heterogeneous Nucleation and Surface Instability. Phys. Rev. Lett. 1998, 80 (13), 2869−2872. (61) Kleinert, J.; Kim, S.; Velev, O. D. Electric-Field-Assisted Convective Assembly of Colloidal Crystal Coatings. Langmuir 2010, 26 (12), 10380−10385.

5776

DOI: 10.1021/acs.langmuir.7b00547 Langmuir 2017, 33, 5769−5776