Article pubs.acs.org/Langmuir
Formation and Frequency Response of Two-Dimensional Nanowire Lattices in an Applied Electric Field Sarah J. Boehm,† Lan Lin,‡ Kimberly Guzmán Betancourt,† Robyn Emery,† Jeffrey S. Mayer,‡ Theresa S. Mayer,*,‡ and Christine D. Keating*,† †
Departments of Chemistry and ‡Electrical Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802, United States S Supporting Information *
ABSTRACT: Ordered two-dimensional (2D) lattices were formed by assembling silica-coated solid and segmented Au nanowires between coplanar electrodes using alternating current (ac) electric fields. Dielectrophoretic forces from the ac field concentrated wires between the electrodes, with their long axis aligned parallel to the field lines. After reaching a sufficient particle density, field-induced dipolar interactions resulted in the assembly of dense 2D lattices that spanned the electrodes, a distance of at least ten wire lengths. The ends of neighboring Au wires or segments overlapped a fraction of their length to form lattice structures with a “running bond” brickwork-like pattern. The observed lattice structures were tunable in three distinct ways: (1) particle segmentation pattern, which fixed the lattice periodicity for a given field condition; (2) ac frequency, which varied lattice periodicity in real time; and (3) switching the field on/off, which converted between lattice and smectic particle organizations. Electric field simulations were performed to understand how the observed lattice periodicity depends on the assembly conditions and particle segmentation. Directed selfassembly of well-ordered 2D metallic nanowire lattices that can be designed by Au striping pattern and reconfigured by changes in field conditions could enable new types of switchable optical or electronic devices.
■
INTRODUCTION Nanowire assemblies are of interest for a wide variety of optical and electronic applications.1−4 However, achieving specific assembly organizations for desired device functions remains a largely unmet challenge.5−9 Approaches that take advantage of surface patterning and/or applied fields are perhaps the most promising.8−13 Such strategies offer new opportunities to incorporate functional components onto patterned substrates such as integrated circuit chips. For example, patterns of topography or chemical functionality have been used to control the location of particle assemblies on surfaces.14−18 Micropatterned electrodes allow localized application of electric fields, which can be used to control particle location and organization by inducing fluid flow, particle migration (e.g., dielectrophoresis), and field-dependent interparticle interactions.19 The resulting forces can be tuned in real time by varying the magnitude and frequency of the applied field, allowing assembly reorganization and optimization. Alternating current (ac) electric-field gradients generated by micropatterned electrodes have been used to form ordered assemblies of colloidal particles on-demand at predetermined locations on a substrate. For example, dielectrophoretic (DEP) forces have been used to position individual microns-long metallic or semiconducting nanowires and nanotubes across narrow electrode gaps (less than wire length) to make electrical contacts for subsequent device characterization.20−23 When gaps larger than the particle size are used, chains of nano- or © 2015 American Chemical Society
microparticles form to bridge the electrodes. Chaining has been observed in dielectric,24−27 metallic,28−34 and mixed-composition (e.g., Janus) particles.35−39 These assemblies rely on mutual DEP and field-induced dipolar interactions between adjacent particles to form chains parallel to the field lines.28,40 For dielectric particles, such as polystyrene, these chains have been shown to combine to create close-packed 2D colloidal crystals at high particle density.24−27,41,42 Open lattice-like structures were recently reported for polystyrene ellipsoid particles.24 In contrast, field-induced 2D crystallization for metallic particles has not, to our knowledge, been reported. The optical and electronic properties of metallic particles, and the responsiveness of these properties to particle organization, make 2D assemblies of metallic nanowires particularly interesting as potential device components. Here, we report assembly of silica-coated solid Au nanowires and partially etched nanowires (PENs) in ac electric fields applied between coplanar electrodes. PENs are produced by silica-coating Au−Ag striped nanowires and subsequently removing the Ag by acid etching.43,44 The presence of a ca. 30 nm silica shell prevented direct electrical contact between the Au nanowire cores. Induced interparticle interactions led to formation of well-ordered 2D nanowire lattices that filled the gap between electrodes over a 50 × nearly 100 μm area. Received: May 4, 2015 Published: May 15, 2015 5779
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
Article
Langmuir Electric field simulations of charge density for adjacent particles and electric-field force calculations as a function of wire position were performed to understand how field-induced dipoles led to the observed lattice periodicity. Here, electricfield force is used to describe forces due to DEP, electrostatics, and dipole−dipole interactions. Although the field response of PENs is dictated almost entirely by their Au core segments, the etched portions nonetheless impact the lattices that form by constraining their packing in the assembly. Differences in the frequency dependence of lattice periodicity were observed between solid Au wires and PENs because etched portions of the PENs prevented Au portions from optimizing their position in the assemblies, regardless of applied frequency. All silicacoated Au and PENs tested formed reversible 2D arrays between the electrodes in the applied ac field, providing a means of directed assembly in which the lattices formed depend on particle striping pattern and field conditions. Hence, lattice structures could be controlled in three ways: (1) particle segmentation pattern, which fixed the lattice periodicity for a given field condition; (2) frequency-dependent dipolar coupling, which varied lattice periodicity in real time; and (3) switching the field on/off, which converted between lattice and smectic particle organizations. The ability to formand rapidly reconfigureassemblies of colloidal nanowires at predetermined locations is of interest for optical and electronic applications.
■
developed in 101 A (18.2 MΩ·cm, using a Barnstead Nanopure filtration system. Alumina membranes with a nominal pore diameter of 0.2 μm were purchased from Whatman. Orotemp 24 and Silver Cyless R plating solutions were purchased from Technic Inc., and tetraethoxysilane was from Gelest. Nanowire Synthesis. Nanowires were fabricated through previously published methods.43,45−48 Briefly, particles were fabricated by templated electrodeposition of metal salts into a porous alumina membrane.46−48 After template release, amorphous silica coating was completed through a modified sol−gel process.43,45 Particles composed of multiple metal segments were selectively etched, resulting in partially etched nanowires. PENs are similar to particles formed by on-wire lithography (OWL)49 except that PEN core segments are held together by a cylindrical silica shell that surrounds the entire wire core. All wires were characterized by measurement of TEM images collected with a JEOL JEM 1200 EXII at 80 keV using an ORIUS 830 SC200 CCD camera. The nanowire concentration was determined using a Hausser Scientific Neubauer hemocytometer and adjusted as necessary to maintain, throughout all assemblies, a consistent ratio of nanowires to total electrode gap area. Electrode Fabrication. Electrodes were fabricated by photolithography. Circular glass coverslips (35 mm diameter, number 2 thickness, GlycoTech) were rinsed with DI water and ethanol then dried with nitrogen gas. PMGI SF6 (MicroChem) was spin-coated onto coverslips and baked at 200 °C for 1 min. Subsequently, SPR3012 (MicroChem) was spin-coated on top and baked at 95 °C for 1 min. Both resists used the same program (ramp rates in parentheses), where the resist was deposited during a 10 s dynamic period at 900 rpm (1000 rpm/s), spread for 1.5 s at 1500 rpm (10 000 rpm/s), thinned for 45 s at 4000 rpm (10 000 rpm/s), and finished for 1 s at 100 rpm (1000 rpm/s). For some electrodes, a Karl Suss MA/ BA6 contact exposure tool was used to expose the samples (64 W/ cm2) for 8 s using an alignment gap of 40 μm. For other electrodes, a GCA8000 Stepper exposure tool was used to expose samples for 0.56 s. Samples were developed in MF-CD-26 (2.4% tetramethylammonium hydroxide solution, MicroChem) for 1 min, rinsed in a water bath for 1 min, and then exposed for 480 s with an OAI Deep UV Flood Exposure tool. After deep UV exposure, samples were
■
RESULTS AND DISCUSSION Assemblies were prepared from silica-coated Au nanowires, uncoated Au nanowires, and several patterns of PENs, which contained Au segments adjacent to hollow, solvent filled segments, all fully encased within a silica shell.44,45 PENs are named based on their striping patterns; for example, 2.5Au-1E indicates a silica-coated particle in which the core is 2.5 μm of Au followed by 1 μm of etched region. By designing PENs with different Au segmentation patterns, different final assemblies could be produced. All solid Au nanowires and PENs used in this work were several microns in length and several hundred nanometers in width; transmission electron microscopy (TEM) sizing data for all particles are available in Table S1. In a typical assembly experiment, an aqueous suspension of particles was placed within a silicone spacer on top of the patterned microelectrodes, and the voltage/frequency of the applied bias was gradually increased to form nanowire assemblies that spanned the gap between electrodes. Assembly concentrations of aqueous particle suspensions were normalized to the wire length and electrode gap area to facilitate comparison across different experiments as described in the Methods section. Coplanar Au electrodes were lithographically patterned onto glass coverslips such that nanowires could be 5780
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
Article
Langmuir viewed from below during the assembly process by reflectance optical microscopy. We used an alternating parallel (stem) and circular (bulb) geometry (S/B) for the coplanar electrode gaps as illustrated in Figure 1A. Similar assemblies could be formed
Figure 1. Overview of assembly approach depicting electrodes used and sample preparation. (A) Side and top view representation of coplanar electrode setup, showing stem/bulb (S/B) geometry and optical micrograph of electrode gap (transmitted light; intensities in this image were inverted such that electrodes appear bright for clarity). Transmission electron micrographs of (B) 2.5Au and (C) 2.5Au-1E nanowires. (D) Reflectance optical micrograph as viewed from below of 2.5Au-1E nanowire assembly across stem region of electrode gap (330 V/cm, 900 kHz). Au segments appear bright whereas silica segments are indistinguishable from the background. Figure 2. Nanowire assembly between coplanar electrodes depicting the process of lattice formation (stem region). Sequential images taken at the same assembly location of (A) solid 2.5Au and (B) 2.5Au-1E nanowires, as applied field strength was increased over ca. 13 min time period. Field conditions apply to both panels. Particles first organized along electrode edges with increased voltage. They then began to form bridges, and eventually a lattice when frequency was increased.
using parallel electrodes without bulbs, with slightly more defects (Figure S1). The ac voltage was applied across the electrodes through contact to probe pads using a function generator. The conditions ranged from 200 to 2000 V/cm for peak field strength and 100 kHz−10 MHz for frequency. Electrode gap sizes (i.e., stem region) generally ranged from 50 to 60 μm (>10× nanowire length). Lattice Formation. Field-driven assembly of metallic particles across bare electrodes can result in electrical shorting, which can damage particles, removing the driving force for further assembly and can lead to poor quality assembled structures. This can be avoided by immediately shutting off the voltage or introducing a dielectric layer on the electrodes.22,23,28,31,34 Here, we overcame this limitation by coating the individual nanowires with a thin dielectric layer (30 nm silica). Figure S2 compares assemblies formed from Au wires with and without this silica shell as well as uncoated Au wires assembled on photoresist-coated electrodes. Only assemblies of glass coated nanowires were found to be reversible. Bare electrodes and silica-coated particles were used for the remainder of experiments in this article. The assembly process for silica-coated, solid Au nanowires can be seen at high magnification in Figure 2A and over a wider area in Figure 3A−C. Particles were allowed to sediment from suspension onto the surface for 2 min before the external ac voltage was applied. Each image shown in Figure 2 was collected after the respective field conditions had been applied for 1 min. We found that immediate application of high field conditions resulted in higher defect lattices; therefore, voltage and frequency were stepped up during the course of the assembly process. In the initial stages of assembly, dielectrophoresis led to migration of wires to the regions of highest
Figure 3. (A−C) Solid 2.5Au nanowires assembled within a stem/bulb electrode at constant field conditions (800 V/cm, 400 kHz, for the stem region). Arrow points to a chain of nanowires as it funnels from the bulb to the stem region of the electrode gap in less than 2 s. Frames taken from Movie S4. (D) Simulation showing field gradient at the top surface of the patterned stem/bulb electrodes using experimental bias conditions. Red indicates highest field gradient, and blue indicates lowest. The field gradient is highest in the stem region.
5781
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
Article
Langmuir
Figure 4. Cartoons of (A) solid 2.5Au nanowires and (B) 2.5Au-1E partially etched nanowires (PENs) illustrating induced dipole interactions between neighboring particles. (C) Three-dimensional simulation showing charge density within solid 2.5Au nanowires in a nonuniform electric field. Electric field: 340 V/cm at 900 kHz, nanowire overlap length set at 0.75 μm. Simulation results for 2.5Au-1E PENs can be found in the Supporting Information (Figure S4).
field gradient, resulting in their collection along the electrode edges with their long axes oriented parallel to the field lines (Figures 2A and 3A−C). Additional particles were then attracted to the ends of the assembled wires by mutual DEP,28,40 which ultimately resulted in the formation of particle chains that spanned across both stem and bulb areas as field strength was increased (Figure 3). The formation of 2D lattices that spanned the entire gap region between electrodes occurred as peak field strength and frequency increased (417 V/cm, 150−300 kHz), as shown in Figure 2A. All wires in the lattice structures were oriented with their long axes parallel to the field lines, which we will refer to as the x-direction. Wires did not closely approach each other end-to-end along the x-direction, but rather had nearest-neighbors in the y-direction offset such that only a small portion of the particle tips were overlapping. This led to a relatively open lattice structure similar to the common “running bond” brickwork pattern. The S/B electrode design facilitated formation of these 2D lattices by preorganization of chains within the bulb regions, which were then funneled into higher field stem regions (Figure S3 and Movie S3). Similar to 2D crystallization of dielectric particles in applied ac fields, field-induced dipole−dipole interactions between adjacent particles provide the driving force for lattice creation.24−26,41 This can be understood by considering the charge distribution on the nanowire array in a nonuniform electric field. Figure 4 shows the simulated charge density on a small portion of the Au nanowire array. In the initial phase of the assembly process, the polarized nanowires tend to produce chains with overlapping metal segments due to the attractive force between the oppositely charged regions of the wires. After a high enough wire density is reached, the chains crystallize into a “running bond” brickwork structure in which the fieldinduced charge alternates from positive to negative in both the x- and y-directions of the 2D lattice. Because the induced dipole forces are strong compared to other forces in this system (due to e.g., electrothermal fluid motion, Brownian motion, electrostatic repulsion, van der Waals attraction between particles, etc.),19,50 the observed structures depend on the factors that influence field-induced charge distribution, which include the wire geometry as well as the externally applied bias. Figure 2B shows the assembly of 2.5Au-1E PENs. These particles are identical to the 2.5 μm long solid Au wires in Figure 2A except for the presence of a 1 μm etched region on one end (Figure 1C). Both the assembly progression with field conditions and the final structures that form are similar to the
solid Au wires. This indicates that the assembly process is driven by the Au segments and that the etched segment does not interfere. This observation is supported by a simulation of charge density (Figure S4), which shows that the etched region participates negligibly in the dipolar interactions between adjacent particles, and its presence results in only a modest difference in the charge density distribution on the Au segments as compared to the solid Au particles in Figure 4. The 1 μm etched portions of these particles fit into the natural open spaces of the Au segment lattice, and consequently the observed assemblies appear essentially the same. Frequency-Dependent Lattice Restructuring. The xdirection packing density of the solid 2.5Au wire 2D lattice depended on the frequency of the applied ac voltage, with decreased end-to-end distances observed at lower frequency. Figures 5A and 5B show the same assembly at 100 and 900
Figure 5. Optical micrographs showing the change in particle overlap of 2.5Au solid nanowires at (A) 100 kHz and (B) 900 kHz but constant voltage (340 V/cm). As frequency increases, the overlap of neighboring particles decreases, creating a more open lattice. (C) Simulation of the electric-field force (x-component) on solid 2.5Au nanowires in an ac electric field (340 V/cm, 100 or 900 kHz) as a function of Au overlap with vertical neighbor, d/μm. The top and bottom wires in each simulation are fixed while the left and right wires are simultaneously moved in the x-direction. The stable nanowire array configuration occurs when the electric field force = 0. 5782
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
Article
Langmuir
Figure 6. Stem region assembly area of 2.5Au-1E partially etched nanowires. Electric field was turned off for 60 s, and images were taken every 15 s of particle reorganization into smectic rows. The electric field was (A) on at 333 V/cm, 900 kHz, (B) off for 15 s, (C) off for 30 s, (D) off for 45 s, (E) off for 60 s, and (F) back on at 333 V/cm, 900 kHz.
maintained at least 1 order of magnitude smaller than that of water. The electric field simulations show that nanowire overlap length decreases with increasing frequency when the water conductivity is higher than 7 × 10−4 S/m and the silica conductivity is in the range of 1 × 10−5−1 × 10−4 S/m. In contrast, when both the water and silica conductivities are either high or low, little to no frequency dependence is observed in the lattice packing density. This can be understood by considering a simplified electrical equivalent circuit model of the nanowire lattice, which is composed of an interconnected network of parallel resistor (R) and capacitor (C) circuit elements, as shown in Figure S6. The two dominant types of RC pairs are associated with the physical overlap between adjacent wires in the y-direction and the end-to-end spacing between adjacent wires in the x-direction of the lattice. The network approaches being purely resistive when the conductivity of both the water and porous silica shell are high and purely reactive (capacitive) when their conductivities are low. In these limits, the voltage drop (field distribution) across the network (lattice) is essentially independent of frequency, and hence the lattice remains unchanged as the frequency is varied. Between these limits, however, the ratio of the admittances for the two types of RC pairs changes significantly with frequency. Consequently, the voltage division (field distribution) between the RC pairs changes resulting in changes in the electrostatic forces and nanowire overlap. Using conductivity values of 4 × 10−3 S/m for water and 4 × 10−5 S/m for the silica shell, the simulated overlap length shifted from 0.75 μm at 100 kHz to 0.50 μm at 900 kHz. This agrees well with the 0.25 μm shift measured experimentally. We attribute the discrepancy in the absolute overlap lengths at the upper and lower frequency to the simplified lattice structure used in the simulation compared to that of the experiment. In contrast to the solid Au wires, clear frequency-dependent change in particle overlap was not observed for the 2.5Au-1E PEN system (Figure S7). These particles are asymmetric, and the Au segment can face toward the left or right electrode. Simulations showed that the preferred amount of overlap for adjacent particles depends on their orientation in the lattice, which is not experimentally controlled (Figure S8). Assembly Reorganization in Response to Cycling Field On/Off. In addition to the frequency response of the
kHz, respectively, under otherwise identical conditions. The lattice structure is more open at the higher frequency. We quantified this difference in the experimental data by measuring the length along the x-direction over which the tips of adjacent wires overlapped. This lattice overlap decreased from 1.00 ± 0.29 to 0.75 ± 0.25 μm as the frequency was increased from 100 to 900 kHz, at constant field strength. Histograms of the experimentally measured values of wire overlap length for these two frequencies are given in Figure S5, which show the difference is statistically significant (p < 0.001). This frequencydependent particle spacing provides a way to dynamically tune the structure of an assembly, specifically its x-axis lattice periodicity, in real time. The frequency dependence of the lattice packing density was studied theoretically by conducting 3D electric field simulations on a simplified four-wire portion of the structure. The Au wire core was considered to be a perfect conductor, while the silica shell and surrounding water were modeled as lossy dielectrics having a permittivity ε and a conductivity σ. The positions of the top and bottom wires were fixed within the electrode gap, while the left and right wires between them were moved along the x-axis (Figure 5C). The x-component of the electric field force (Fx) on the moving wires was calculated as a function of Au overlap length to determine the energetically stable nanowire array configuration where Fx = 0. Experimentally measured dielectric properties were not available for the suspending water or the sol−gel fabricated silica shell. Therefore, the simulation was repeated for a range of physically realistic dielectric properties to identify values that provided the best agreement with experiment. The water used in these experiments has a pH ∼3 and contains ions from the wire fabrication process, making it more conductive than pH 7 DI water.51 In addition, the sol−gel silica films are porous, resulting in a water filled silica matrix with higher conductivity and permittivity than high-quality thermal or plasma deposited oxide.43,52 To account for these factors, the conductivity of water was varied from 5 × 10−6 to 1 × 10−2 S/m and silica from 5 × 10−7 to 1 × 10−3 S/m. For all simulations, the real part of the permittivity for the water was fixed at 80. The silica shell permittivity was estimated using Maxwell’s mixing rule for a 90% silica and 10% water matrix, giving a value of 4.88.53 For a given simulation, the conductivity of the silica shell was always 5783
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
Article
Langmuir
been assembled previously in ac fields and shown to form interesting structures such as chains in which metal-coated portions are oriented toward each other, leaving uncoated dielectric portions facing outward.35−39 The PENs used here offer distinct field response for multicomponent materials similar to Janus particles but differ in two important aspects: (i) the different materials are located inside a uniform silica shell that evenly coats the entire particle, and (ii) design and fabrication of a large number of different particle patterns are straightforward. We reasoned that it should be possible to control lattice spacing in the x-direction by varying the segmentation pattern of the PENs from which the assemblies were generated. A series of particles were fabricated with similar total Au segment length (3 μm), distributed differently: 3Au2E, 3Au-2E′ (where the prime indicates a small, 120 nm Au cap on the far side of the etched region), 1.5Au-2.5E-1.5Au, and 2Au-1E-1Au. Field-induced assemblies for each of these PEN samples were then compared both at the same field conditions (Figure S9) and at individually optimized conditions (i.e., best conditions for producing a hole-free lattice that spanned the electrode gap) (Figure 8). Like the 2.5Au and 2.5Au-1E particles described in earlier sections, lattice structures spanning the electrode gap were observed for both 3Au-2E and 3Au-2E′ PENs. The lattice
lattice assemblies described above, it was also possible to induce a large-scale reorganization by switching on/off the field. Figure 6 shows a progression of images detailing the transition of 2.5Au-1E PENs between field on and off over a 60 s period, where images were taken every 15 s. Almost immediately after the field was turned off, the nanowires transitioned into smectic rows, where Au segments of neighboring particles aligned, similar to assemblies formed in the absence of fields. These rows can be attributed to the strong van der Waals interactions between Au segments of neighboring particles, causing them to preferentially align parallel to one another. We have previously observed such smectic rows for both solid Au wires and PENs at high particle density in the absence of any applied field.44,54 Here, particle number density was lower, and rows began to deteriorate over time due to diffusion because the particle population surrounding the electrode gap was too diffuse to support the local smectic structure (Figure 6D,E). Upon reapplication of the electric field, the lattice structure immediately re-formed, a process that can be replicated many times (Figure 6F and Movie S5). Reversible assemblies were achieved for all particle types evaluated (silica-coated Au and PENs). Assemblies could be annealed by cycling the field on and off, allowing particles to rearrange within the lattice to more favorable locations. Figure 7 highlights a defect area within the
Figure 7. Lattice areas of 2.5Au-1E partially etched nanowires (A) prior to and (B) after annealing by cycling the field on and off 5× at 333 V/cm, 900 kHz. Areas highlighted by white boxes are enlarged to the right, showing improved order after annealing.
lattice of misaligned 2.5Au-1E particles. The field was successively cycled on and off (10 s on, 10 s off each time) five times, revealing a more ordered structure with fewer defects. We found that the “off time” should not exceed 30 s because the wires diffuse too far from their original location, resulting in less ordering upon reapplication of the field, as can be seen in Figure 6F. Certain defects, predominantly caused by bundled particles cannot be removed this way, as they are not favorably excluded from the assembly. Control of Lattice Structures by PEN Au Segmentation Design. Mixed-composition (e.g., Janus) particles have
Figure 8. Assemblies of partially etched nanowires that all contain similar total Au volume but segments are distributed differently within each wire type. Representative transmission electron microscope (TEM) images of (A) 3Au-2E, (B) 3Au-2E′, (C) 1.5Au-2.5E-1.5Au, and, (D) 2Au-1E-1Au nanowires. Below their TEM image shows the most complete lattice structures formed by each particle type. Field conditions are as follows: (A) 800 V/cm, 400 kHz; (B) 500 V/cm, 750 kHz; (C) 2 kV/cm, 3.5 MHz; and (D) 700 V/cm, 1 MHz. All TEM image scale bars are 1 μm, and the 10 μm scale bar in (A) applies to all assembly images. 5784
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
Langmuir
■
structures appear similar to those formed from 2.5Au-1E PENs, but due to the longer Au segment, the x-direction periodicity of the assemblies differs. Measured distances in x between tips of Au segments are 2.3 ± 0.4 μm for the assembly in Figure 8A, as compared with 1.5 ± 0.2 μm for 2.5Au-1E PENs in Figure 6A. Because of the longer Au segment and similar overlap at the wire tips in adjacent particles, wider spacing is achieved. The presence of a small Au cap at the far end of the etched region did not markedly impact the observed assemblies, although somewhat more densely packed lattices were seen for the 3Au2E′ PENs (compare Figures 8A and 8B). This is consistent with the charge density simulation for the 3Au-2E′ PENs, which indicates the presence of a much smaller charge on the small Au cap as compared to the larger Au segment (Figure S10). This additional small charge may assist in lateral packing of the lattice structures but does not alter the general form of the assembly. Assemblies were also produced using PENs with two, more similarly sized Au segments separated by an etched region (1.5Au-2.5E-1.5Au and 2Au-1E-1Au). These have the same total amount of Au as the 3Au-2E particles in Figure 8A but give rise to very different lattice periodicity due to its different distribution along the length of the PENs. Measured distances in x between tips of Au segments were 0.99 ± 0.18 μm for the assembly in Figure 8C, less than half the tip−tip distance for the lattice in Figure 8A. The asymmetry in 2Au-1E-1Au particles makes for less regular periodicity in the x-direction, and hence this parameter was not quantified for Figure 8D. We note that field conditions required to form the assemblies in Figure 8 differed with the particle pattern, as noted in the caption. Under identical field conditions differences could be seen in the assemblies (Figure S9). PENs with larger Au segments assembled into the gap region more readily, while PENs with shorter Au segments had the most difficulty forming full lattices. This can be understood as a result of less effective mutual DEP as the particle chains grow outward from the electrode edges.
AUTHOR INFORMATION
Corresponding Authors
*E-mail
[email protected] (T.S.M.). *E-mail:
[email protected] (C.D.K.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
REFERENCES
This experimental research was primarily supported by the Penn State Materials Research Science and Engineering Center (MRSEC, NSF DMR-0820404 and DMR-1420620), with additional support from the Charles E. Kaufman Fund of The Pittsburgh Foundation. Undergraduates K.G.B. and R.E. were funded through the NSF MRSEC REU (DMR-062040) and NSF NNIN REU (ECCS-0335765), respectively. TEM images were acquired at the Penn State Microscopy and Cytometry Facility and electrodes were fabricated at Pennsylvania State University NSF NNIN Site.
(1) Kempa, T. J.; Day, R. W.; Kim, S.; Park, H.; Lieber, C. M. Semiconductor Nanowires: A Platform for Exploring Limits and Concepts for Nano-Enabled Solar Cells. Energy Environ. Sci. 2013, 6, 719−733. (2) Timko, B. P.; Cohen-Karni, T.; Qing, Q.; Tian, B.; Lieber, C. M. Design and Implementation of Functional Nanoelectronic Interfaces with Biomolecules, Cells, and Tissue Using Nanowire Device Arrays. IEEE Trans. Nanotechnol. 2010, 9, 269−280. (3) Liu, X.; Long, Y.-Z.; Liao, L.; Duan, X.; Fan, Z. Large-Scale Integration of Semiconductor Nanowires for High-Performance Flexible Electronics. ACS Nano 2012, 6, 1888−1900. (4) Yan, R.; Gargas, D.; Yang, P. Nanowire Photonics. Nat. Photonics 2009, 3, 569−576. (5) Pelaz, B.; Jaber, S.; Jimenez de Aberasturi, D.; Wulf, V.; Aida, T.; de la Fuente, J. M.; Feldman, J.; Gaub, H. E.; Josephson, L.; Kagan, C. R.; et al. The State of Nanoparticle-Based Nanoscience and Biotechnology: Progress, Promises and Challenges. ACS Nano 2012, 6, 8468−8483. (6) Velev, O. D.; Gupta, S. Materials Fabricated by Micro- and Nanoparticle Assembly− The Challenging Path from Science to Engineering. Adv. Mater. 2009, 21, 1897−1905. (7) Srivastava, S.; Kotov, N. A. Nanoparticle Assembly for 1D and 2D Ordered Structures. Soft Matter 2009, 5, 1146−1156. (8) Grzelczak, M.; Vermant, J.; Furst, E. M.; Liz-Marzán, L. M. Directed Self-Assembly of Nanoparticles. ACS Nano 2010, 4, 3591− 3605. (9) Furst, E. M. Directed Self-Assembly. Soft Matter 2013, 9, 9039− 9045. (10) Dziomkina, N. V.; Vancso, G. J. Colloidal Crystal Assembly on Topologically Patterned Templates. Soft Matter 2005, 1, 265−276. (11) Edwards, T. D.; Bevan, M. A. Controlling Colloidal Particles with Electric Fields. Langmuir 2014, 30, 10793−10803. (12) Pethig, R. Dielectrophoresis: Status of the Theory, Technology, and Applications. Biomicrofluidics 2010, 4, 022811. (13) van Blaaderen, A.; Dijkstra, M.; van Roij, R.; Imhof, A.; Kamp, M.; Kwaadgras, B. W.; Vissers, T.; Liu, B. Manipulating the Self Assembly of Colloids in Electric Fields. Eur. Phys. J.: Spec. Top. 2013, 222, 2895−2909. (14) Tzeng, S.; Lin, K.; Hu, J.; Chen, L.; Gwo, S. Templated SelfAssembly of Colloidal Nanoparticles Controlled by Electrostatic Patterning on a Si3N4/SiO2/Si Electret. Adv. Mater. 2006, 18, 1147− 1151. (15) Liu, Z.; Huang, H.; He, T. Large-Area 2D Gold Nanorod Arrays Assembled on Block Copolymer Templates. Small 2013, 9, 505−510.
■
CONCLUSIONS We have demonstrated versatile, on-demand particle assembly at predetermined sites on the substrate, dictated by the position of the coplanar electrodes. The overall lattice periodicity could be (1) designed by choice of Au segmentation of the PENs, (2) varied in real time by altering the applied frequency, and (3) completely reconfigured between these lattice structures and more traditional smectic rows by turning off/on the applied field. The silica-coated, solid Au-core nanowires and PENs used here go beyond previous work on ac field driven assembly of 2D colloidal crystals, which has largely focused on dielectric particles. It should ultimately be possible to take advantage of the optical properties of metal-containing particles in reconfigurable devices where changes in particle organization alter assembly properties, e.g., optical reflectance, absorbance, or transmittance.
■
Article
ASSOCIATED CONTENT
S Supporting Information *
Detailed particle dimensions, movies of assemblies, and supporting figures mentioned within the text. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b01633. 5785
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
Article
Langmuir
(37) Chaudhary, K.; Juarez, J. J.; Chen, Q.; Granick, S.; Lewis, J. A. Reconfigurable Assemblies of Janus Rods in AC Electric Fields. Soft Matter 2013, 10, 1320−1324. (38) Gangwal, S.; Cayre, O. J.; Velev, O. D. Dielectrophoretic Assembly of Metallodielectric Janus Particles in AC Electric Fields. Langmuir 2004, 24, 13312−13320. (39) Shah, A. A.; Shultz, B.; Zhang, W.; Glotzer, S. C.; Solomon, M. J. Actuation of Shape-Memory Colloidal Fibres of Janus Ellipsoids. Nat. Mater. 2015, 14, 117−124. (40) Pohl, H. A. Dielectrophoresis: The Behavior of Neutral Matter in Nonuniform Electric Fields; Cambridge University Press: New York, 1978. (41) Hernández-Navarro, S.; Ignés-Mullol, J.; Sagués, F.; Tierno, P. Role of Anisotropy in Electrdynamically Induced Colloidal Aggregates. Langmuir 2012, 28, 5981−5986. (42) McMullan, J. M.; Wagner, N. J. Directed Self-Assembly of Colloidal Crystals by Dielectrophoretic Ordering. Langmuir 2012, 28, 4123−4130. (43) Sioss, J. A.; Keating, C. D. Batch Preparation of Linear Au and Ag Nanoparticle Chains via Wet Chemistry. Nano Lett. 2005, 5, 1779− 1783. (44) Smith, B. D.; Kirby, D. J.; Rivera, I. O.; Keating, C. D. SelfAssembly of Segmented Anisotropic Particles: Tuning Compositional Anisotropy to Form Vertical or Horizontal Arrays. ACS Nano 2013, 7, 825−833. (45) Smith, B. D.; Kirby, J. D.; Keating, C. D. Vertical Arrays of Anisotropic Particles by Gravity-Driven Self-Assembly. Small 2011, 7, 781−787. (46) Nicewarner-Peña, S. R.; Freeman, R. G.; Reiss, B. D.; He, L.; Peña, D. J.; Walton, I. D.; Cromer, R.; Keating, C. D.; Natan, M. J. Submicrometer Metallic Barcodes. Science 2001, 294, 137−141. (47) Al-Mawlawi, D.; Liu, C. Z.; Moskovits, M. Nanowires Formed in Anodic Oxide Nanotemplates. J. Mater. Res. 1994, 9, 1014−1018. (48) Hulteen, J. C.; Martin, C. R. A General Template-Based Method for the Preparation of Nanomaterials. J. Mater. Chem. 1997, 7, 1075− 1087. (49) Qin, L.; Park, S.; Huang, L.; Mirkin, C. A. On-Wire Lithography. Science 2005, 309, 113−115. (50) Israelachvili, J. N. Intermolecular and Surface Forces, 2nd ed.; Academic Press Limited: London, 1992. (51) Morgan, H.; Green, N. G. AC Electrokinetics: Colloids and Nanoparticles; Research Studies Press: Philadelphia, PA, 2003; pp 15− 31. (52) Brinker, C. J.; Scherer, G. W. Sol-Gel Science, The Physics and Chemistry of Sol-Gel Processing; Academic Press Inc.: New York, 1990; Chapter 3, p 13. (53) Mohideen, U.; Gururaja, T.; Cross, L.; Roy, R. Ultra-Low Dielectric Constant Porous Silica Thick Films for High-Speed IC Packaging. IEEE Trans. Compon. Packag. Manuf. Technol. 1988, 11, 159−162. (54) Triplett, D. A.; Quimby, L. M.; Smith, B. D.; Rodríguez, D. H.; St. Angelo, S. K.; González, P.; Keating, C. D.; Fichthorn, K. A. Assembly of Gold Nanowires by Sedimentation from Suspension: Experiments and Simulation. J. Phys. Chem. C 2010, 114, 7346−7355.
(16) Kirby, D. J.; Smith, B. D.; Keating, C. D. Microwell-Directed Self-Assembly of Vertical Nanowire Arrays. Part. Part. Syst. Charact. 2014, 31, 492−499. (17) Ni, S.; Klein, M. J. K.; Spencer, N. D.; Wolf, H. Cascaded Assembly of Complex Multiparticle Patterns. Langmuir 2013, 30, 90− 95. (18) Hellstrom, S. L.; Kim, Y.; Fakonas, J. S.; Senesi, A. J.; Macfarlane, R. J.; Mirkin, C. A.; Atwater, H. A. Epitaxial Growth of DNA-Assembled Nanoparticle Superlattices on Patterned Substrates. Nano Lett. 2013, 13, 6084−6090. (19) Velev, O. D.; Bhatt, K. H. On-Chip Micromanipulation and Assembly of Colloidal Particles by Electric Fields. Soft Matter 2006, 2, 738−750. (20) Freer, E. M.; Grachev, O.; Duan, X.; Martin, S.; Stumbo, D. P. High-Yield Self-Limiting Single-Wire Assembly with Dielectrophoresis. Nat. Nanotechnol. 2010, 5, 525−530. (21) Smith, B. D.; Mayer, T. S.; Keating, C. K. Deterministic Assembly of Functional Nanostructures Using Nonuniform Electric Fields. Annu. Rev. Phys. Chem. 2012, 63, 241−263. (22) Smith, P. A.; Nordquist, C. D.; Jackson, T. N.; Mayer, T. S. Electric-Field Assisted Assembly and Alignment of Metallic Nanowires. Appl. Phys. Lett. 2000, 77, 1399−1401. (23) Li, M.; Bhildavala, R. B.; Morrow, T. J.; Sioss, J. A.; Lewm, K.; Redwing, J. M.; Keating, C. D.; Mayer, T. S. Bottom-Up Assembly of Large-Area Nanowire Resonator Arrays. Nat. Nanotechnol. 2008, 3, 88−92. (24) Singh, J. P.; Lele, P. P.; Nettesheim, F.; Wagner, N. J.; Furst, E. M. One- and Two-Dimensional Assembly of Colloidal Ellipsoids in AC Electric Fields. Phys. Rev. E 2009, 79, 050401. (25) Panczyk, M. M.; Park, J.; Wagner, N. J.; Furst, E. M. TwoDimensional Directed Assembly of Dicolloids. Langmuir 2013, 29, 75−81. (26) Lumsdom, S. O.; Kaler, E. W.; Velev, O. D. Two-Dimensional Crystallization of Microspheres by a Coplanar AC Electric Field. Langmuir 2004, 20, 2108−2116. (27) Mittal, M.; Lobo, R. F.; Furst, E. M. Externally Directed Assembly of Disk-Shaped Zeolite Particles by an Electric Field. J. Mater. Res. 2011, 26, 215−222. (28) Gierhart, B. C.; Howitt, D. G.; Chen, S. J.; Smith, R. L.; Collins, S. D. Frequency Dependence of Gold Nanoparticle Superassembly by Dielectrophoresis. Langmuir 2007, 23, 12450−12456. (29) Hermanson, K. D.; Lumsdon, S. O.; Williams, J. P.; Kaler, E. W.; Velev, O. D. Dielectrophoretic Assembly of Electrically Functional Microwires from Nanoparticle Suspensions. Science 2001, 294, 1082− 1085. (30) Fan, D. L.; Zhu, F. Q.; Cammarata, R. C.; Chien, C. L. Manipulation of Nanowires in Suspension by AC Electric Fields. Appl. Phys. Lett. 2004, 85, 4175−4177. (31) Bhatt, K. H.; Velev, O. D. Control and Modeling of Dielectrophoretic Assembly of On-Chip Nanoparticle Wires. Langmuir 2004, 20, 467−476. (32) Fan, D. L.; Zhu, F. Q.; Cammarata, R. C.; Chien, C. L. Efficiency of Assembly of Nanowires in Suspension by AC Electric Fields. Appl. Phys. Lett. 2006, 89, 223115. (33) Papadakis, S. J.; Gu, Z.; Gracias, D. H. Dielectrophoretic Assembly of Reversible and Irreversible Metal Nanowire Networks and Vertically Aligned Arrays. Appl. Phys. Lett. 2006, 88, 233118. (34) Leiterer, C.; Berg, S.; Eskelinen, A.; Csaki, A.; Urban, M.; Törmä, P.; Fritzsche, W. Assembling Gold Nanoparticle Chains Using an AC Electrical Field: Electrical Detection of Organic Thiols. Sens. Actuators, B 2013, 176, 368−373. (35) Gangwal, S.; Pawar, A.; Kretzschmar, I.; Velev, O. D. Programmed Assembly of Metallodielectric Patchy Particles in External AC Electric Fields. Soft Matter 2010, 6, 1413−1418. (36) Shields, C. W., IV; Zhu, S.; Yang, Y.; Bharti, B.; Liu, J.; Yellen, B. B.; Velev, O. D.; López, G. P. Field-Directed Assembly of Patchy Anisotropic Microparticles with Defined Shape. Soft Matter 2013, 9, 9219−9229. 5786
DOI: 10.1021/acs.langmuir.5b01633 Langmuir 2015, 31, 5779−5786
