Self-Patterned Stretchable Electrode Based on Silver Nanowire Bundle Mesh Developed by Liquid Bridge Evaporation

Abstract

A new strategy is required to realize a low-cost stretchable electrode while realizing high stretchability, conductivity, and manufacturability. In this study, we fabricated a self-patterned stretchable electrode using a simple and scalable process. The stretchable electrode is composed of a bridged square-shaped (BSS) AgNW bundle mesh developed by liquid bridge evaporation and a stretchable polymer matrix patterned with a microcavity array. Owing to the BSS structure and microcavity array, which effectively concentrate the applied strain on the deformable square region of the BSS structure under tensile stretching, the stretchable electrode exhibits high stretchability with a low ΔR/R₀ of 10.3 at a strain of 40%. Furthermore, by exploiting the self-patterning ability—attributable to the difference in the ability to form liquid bridges according to the distance between microstructures—we successfully demonstrated a stretchable AgNW bundle mesh with complex patterns without using additional patterning processes. In particular, stretchable electrodes were fabricated by spray coating and bar coating, which are widely used in industry for low-cost mass production. We believe that this study significantly contributes to the commercialization of stretchable electronics while achieving high performance and complex patterns, such as stretchable displays and electronic skin.

1. Introduction

Over the past decade, stretchable electronics have received significant attention because of their high potential for a wide range of human-friendly electronics applications, such as skin-attachable healthcare systems [1,2], wearable displays [3,4,5], and implantable devices [6,7,8]. Among the fundamental components of stretchable electronics, the fabrication of stretchable electrodes capable of maintaining their initial electrical conductivity under various modes of mechanical deformation remains a crucial research area. High stretchability, conductivity, mechanical durability, and low processing cost are essential to realizing stretchable electrodes [9]. Accordingly, many studies based on structural approaches [10] and material approaches have been reported [11,12,13]. To improve stretchability, novel structures suitable for mechanical deformation, such as serpentine structures [6,14], mesh-shaped structures [15,16], out-of-plane buckled structures [17,18], and kirigami structures [19,20] have been proposed. Although stretchable electrodes with deformable structures have exhibited high stretchability with minor resistance changes under large deformation, metal deposition under high vacuum and prestraining of the polymer substrate in their fabrication processes make mass production difficult, which is a critical requirement for commercialization and practical applications.

Low-dimensional carbon materials and metal nanomaterials [12,21,22] have been proposed to overcome these challenges. Among them, silver nanowires (AgNWs) have been widely used because of their high conductivity and flexibility induced by thinness [23,24]. In particular, AgNWs wrapped by polyvinylpyrrolidone are well dispersed in polar solvents such as water and alcohols [25]; thus, they have been readily applied to stretchable electrodes through solution processes that allow for straightforward mass production at low cost [26,27]. However, most AgNW-based stretchable electrodes fabricated by simple and scalable solution processes, such as spray and bar coating, exhibit limited stretchability because of the absence of an efficient deformable structure for delocalization and relaxation of localized stress for AgNWs under large deformations. Recently, a stretchable electrode based on a AgNW bundle with a microscaled random mesh form was reported [28,29]. The random mesh was formed by the coffee-ring effect-driven assembly of nanowires under spray coating. Owing to the deformability of the random mesh structure, the electrode exhibited a negligible increase in resistance against a considerable strain. Nevertheless, considering that a sophisticated pattern is essential for the application to stretchable displays, e-skin, and health-monitoring applications that are composed of multiple cells with complex device structures, there is a significant disadvantage in that additional photolithography or laser ablation for electrode patterning must be performed [30].

In this paper, we present a self-patterned stretchable electrode composed of a bridged square-shaped (BSS) AgNW bundle mesh embedded in a stretchable polymer substrate patterned with a microcavity array. By inducing a concave liquid bridge of AgNW dispersion between microstructures upon spray coating, which is widely used in industry for low-cost mass production, we obtained a regular and uniform BSS AgNW bundle mesh favorable to mechanical deformation. In particular, by inducing a microcavity array that was patterned inside the square region of the BSS bundle mesh, we achieved a stretchable electrode that exhibited high stretchability with a low ∆R/R0 of 10.3 at 40% strain. Furthermore, by exploiting the self-patterning ability based on the difference in the ability to form liquid bridges according to the distance between microstructures, we successfully demonstrated a stretchable AgNW bundle mesh with complex patterns without using additional patterning processes. We believe that this study significantly contributes to the commercialization of stretchable electronics with high performance and complex patterns, such as a stretchable antenna for skin attachable RFID patch [31] and battery-free electronic skin powered by Wi-Fi signal [32], and interconnectors of stretchable display with micro LEDs [10].

2. Materials and Methods

The microstructured substrates with the dimension of 30 mm × 30 mm were prepared by demolding a cured PDMS (Sylgard 184, Dow Corning, Midland, MI, USA) sheet from a prepatterned Si mold with an inversed micropyramidal cavity array [33]. On the substrate, pyramidal microstructures with the individual dimension of 10 μm × 10 μm (height ~7 μm) were patterned. To investigate the effect of the surface energy of the substrate on the formation of the mesh, fluorinated SAM-treated PDMS by chemical vapor deposition with trichloro(1H,1H,2H,2H-perfluorooctyl)silane (FOTS, Sigma Aldrich, St. Louis, MO, USA) [34] and UVO-treated PDMS by UV exposure in the air using an ozone cleaner (UVC-20, Japan Engineering Co., Hyogo, Japan) were also prepared. On the microstructured substrates, AgNW dispersion (Flexiowire2020c, D 25 nm L 25 μm, 0.5 wt%, SG Flexio Co., Ltd., Daejeon, Korea) diluted to 0.01 wt% in isopropyl alcohol (IPA, Sigma Aldrich) was sprayed using an automatic spray coater (ReVo-S) equipped with a nozzle (AM45, ATOMAX) with a diameter of 1 mm. The spray rate (i.e., the volume of sprayed dispersion per second), pressure, and distance from the nozzle tip to target substrates were 60 μL/s, 0.1 MPa, and 10 cm, respectively. After bar-coating with thermoplastic polyurethane (TPU, MIRATHANE E190, Miracll Chemical Co., Ltd., Yantai, China) solution dissolved in dimethylformamide (Sigma Aldrich) to 10 wt% and dried at 60 °C for 2 h in a vacuum, the stretchable electrode was gently peeled off using an adhesive PET frame. The deposition behavior of AgNWs was observed in situ using an inverted optical microscope (XDS 3FL4, Optika, Ponteranica, Italy) during spray coating. The morphologies after AgNW deposition or subsequently applied tensile strain were observed using an optical microscope (VK-X1000, KEYENCE) and FE-SEM (JSM-6701F, JEOL). The surface-modified substrates were characterized by X-ray photoelectron spectroscopy (XPS, K-Alpha, Thermo Fisher, Waltham, MA, USA) and a contact angle analyzer (SmartDrop Plus, Femtobiomed, Seongnam, Korea). Relative resistance changes (ΔR/R₀) were analyzed using an I-V meter (4200A-SCS, Keithley, Tacoma, WA, USA), and the sheet resistances were measured using a four-point probe module (PE100, MSTECH, Bath, PA, USA) connected to the I-V meter. The stress distribution on the BSS bundle mesh under 30% tensile strain was analyzed using finite element method-based simulation (COMSOL Multiphysics^® 5.2a). A solid mechanics module and nonlinear structural material module were utilized to stretch the AgNW bundle mesh and the TPU substrate. The result of the stress distribution was expressed as the von Mises stress (N/m) applied to the AgNW bundle mesh.

3. Results and Discussion

The self-patterned stretchable electrode was fabricated simply, as shown in Figure 1a. Detailed information on the fabrication of the stretchable electrode is described in Section 2. As it is fabricated by fabrication processes common to industry, that is, spray and bar coating, scalable and continuous production is feasible. Even if a PDMS substrate with a micropyramidal structure array is fabricated by a batch-type process such as lithography and etching, it can be semipermanently reusable. By spray coating the microstructured substrate with AgNW dispersion, a BSS bundle mesh was developed, as shown in Figure 1b. Interestingly, very few AgNWs were deposited on the surface of the micropyramidal structure. Most of the AgNWs were placed at the contact line between the microstructure and the basal plane (square region) and between the microstructures with a well-ordered form (bridge region). After bar coating the TPU solution and drying, the stretchable electrode was obtained by being peeled off from the microstructured PDMS substrate. The AgNW mesh was successfully transferred from the top surface of the microstructured PDMS substrate to the TPU matrix in an embedded form. In addition, an inversed pyramidal microcavity array was naturally patterned inside the square region of the AgNW mesh by demolding (Figure 1c). Owing to its thinness and stretchability, the stretchable electrode can be conformally attached to curved surfaces such as human skin. In addition, it was translucent (Figure 1d), allowing the stretchable electrode to be unnoticeable when attached to human skin as an electronic skin device.

To study the formation mechanism of the BSS bundle mesh, an in-situ observation of the deposition behavior was conducted using an inverted optical microscope during spray coating. As shown in Figure 2a, the microdroplet was first deposited on the microstructured substrate by spray coating. Then, the deposited droplets moved to and coalesced at the contact line between the microstructure and base surface by the air blown from the nozzle. The coalesced liquids placed outside the individual microstructure were interconnected in a bridge form by successive spray coating. Meanwhile, the interconnecting bridges composed of AgNW dispersion, called a liquid bridge, were developed according to the nearest-bridging rule [35]. After evaporation of the liquid, a BSS bundle mesh was finally formed. From the results of the in situ observation, we conclude that the formation of the BSS bundle mesh results from the evaporation of the liquid bridge and is based on the mechanism of evaporative lithography. Vakarelski et al. reported using evaporative lithography, which exploits various effects generated by solvent evaporation, to fabricate microwire networks and patterns [36]. After dropping the gold nanoparticle dispersion on the lyophilic particulate templates, the liquid bridge started to develop during evaporation. As evaporation proceeded, the shrinking pendular rings formed nodes at the base of the particles joined by thin liquid bridges that were in contact with the substrate [37]. With further evaporation and narrowing of the liquid bridges, the nanoparticles in the liquid bridge were forced to pack in one direction. In our study, a concave liquid bridge was formed between neighboring microstructures made of PDMS with moderately lyophilic properties to the IPA solvent of the dispersion. With the evaporation of the dispersion, narrowing of the liquid bridges occurred, and the AgNWs in the dispersion were aligned and packed. Consequently, the BSS bundle mesh was formed uniformly, as shown in Figure 2b.

To study the effect of the surface energy of the microstructured substrate on the formation of the BSS bundle mesh, microstructured substrates with various surface energies were spray coated. As shown in Figure 2c–f, in the case of FOTS-treated PDMS with lower surface energy compared to pristine PDMS [34], the bundle mesh structure was not formed, and the coating was nonuniform; thus, macroscaled aggregations were observed (Figure 2c). We attribute this result to the microdroplets, deposited on the substrate surface, moving too readily under airflow from the spray nozzle, thus coalescing with each other due to the low interaction with the FOTS-treated PDMS substrate. For the PDMS treated by UVO for 10 min (Figure 2e), the AgNW mesh structure appeared, but the AgNWs were not well aligned and packed. For the PDMS treated by UVO for 60 min (Figure 2f), AgNWs were deposited over the entire surface of the substrate, including the surface of the microstructure (Figure S1). This result is due to the full wetting of microdroplets under spraying, and pinning during evaporation [38]. All the surface-modified substrates without microstructures were characterized by XPS and contact angle analysis, as shown in Figures S2 and S3.

In contrast to substrates treated by FOTS or UVO, the BSS bundle mesh structure was clearly formed for the microstructured PDMS substrate without any surface treatment (Figure 2d). In addition, by virtue of a well-formed mesh structure with a packed and aligned AgNW bundle, it exhibited an even lower sheet resistance of 151 Ω/□ compared to other substrates, as shown in Figure 2g. Although there were some differences according to the spray conditions, such as temperature, pressure, and spraying rate, the trends were similar.

We employed a BSS bundle mesh as the conducting component in a stretchable electrode. As shown in Figure S4, the BSS bundle mesh is more efficient for deformation than a conventional mesh-shaped structure because the structure delocalizes the concentrated stress by the deformation of the square region; consequently, the structure allows the stress to be distributed more evenly. To determine the stretchability of the electrode, the relative resistance change (ΔR/R₀) with the tensile strain was measured. Notably, an inverted pyramidal microcavity array exists on the surface of the stretchable electrode, which is naturally created by demolding during the fabrication process. As shown in Figure 3a (solid circle), the value of ΔR/R₀ was only 10.3 with a strain of 40%. This high stretchability was also confirmed by the optical microscopic images of the stretchable electrodes taken under stretching (Figure 3b). A noticeable fracture of the BSS bundle could not be found up to 40% strain, and only a few were observed above a strain of 50% (Figure S5).

To study the causes of the high stretchability in depth, we prepared a stretchable electrode without a microcavity array on its surface. In order to fill up the microcavity array, we conducted over-coating of TPU solution on the stretchable electrode with a microcavity array. It exhibited lower stretchability and an even larger ΔR/R₀ increment than with the microcavity array, as shown in Figure 3a (hollow circle). Accordingly, many fractures of AgNW bundles were observed even at strains as low as 20% (Figure 3b). This difference in stretchability between the cases with and without the microcavity array is explained as follows. In addition to the contribution of the BSS mesh structure, the microcavity located inside the square region of AgNW bundle mesh serves to concentrate the tensile strain applied to the stretchable electrode to the square region which is easily deformed against tensile strain, thereby preventing and mitigating the concentration of stress at bridge region. This is because the thickness of the TPU substrate under the microcavity is thinner than the region without the microcavity. (Figure 3c,d). From these results, we conclude that the high stretchability of the stretchable electrode is due to the synergistic effect of the BSS bundle mesh structure and the microcavity array located in the square region of the BSS mesh.

To investigate the effect of the distance between microstructures on the formation of AgNW bundle mesh, we spray coated AgNW dispersions on microstructured PDMS substrates with various distances between adjacent structures of 10, 15, and 20 μm. As shown in Figure 4a–c, as the distance between adjacent microstructures increased, the width and number of bridges in the AgNW bundle mesh decreased, and no bridge connecting neighboring squares was formed even at a distance of 20 μm. It seems that the liquid bridges with a distance of 20 μm were not sustained because of the balance between the structural cohesive force and the capillary force [35,39]. Consistent with this observation, the sheet resistance sharply increased as the distance between the microstructures increased, reaching above 10⁸ Ω/□ at a distance of 20 μm (Figure 4d). These results indicate that if only the microstructured substrate—the target substrate of the spray coating—is deliberately designed by considering the self-patterning ability (i.e., deliberate electrical isolation by adjusting the distance between microstructures), it is possible to easily and repeatedly produce intricately patterned stretchable electrodes without additional patterning processes, such as photolithography and laser ablation. To demonstrate the self-patterning ability, we prepared the microstructured substrates that formed AgNW bundle mesh according to the shape of the letter “K”. The distance between the microstructures inside the letter was 10 μm, and that outside the letter was 20 μm. After spraying the AgNWs dispersion, the deposited AgNWs were transferred to the TPU substrate. As shown in Figure 4e, the BSS bundle mesh was only formed inside the character and not outside the character.

After peeling off the stretchable electrode, the microstructured substrate returned to its initial state without the AgNW or TPU residue as before the AgNW was spray coated (Figure S6a,b). From these observations, we fabricated five stretchable electrodes by repeatedly using the same microstructured substrate. As a result, the fabricated stretchable electrodes exhibited high reproducibility (Figure 4f), which means that the fabrication strategy presented in this work is applicable to continuous low-cost production when applied to roll-to-roll processes.

4. Conclusions

In this paper, we present a self-patterned stretchable electrode by liquid bridge evaporation. In the proposed process, the stretchable electrode is composed of a BSS AgNW bundle mesh embedded in a TPU matrix with a microcavity array. A regular and uniform bundle mesh is developed by the formation and evaporation of a liquid bridge located between microstructures, and the microcavity array on the surface of the TPU matrix is naturally created by demolding. Owing to the synergistic effect of the bridged square-shaped bundle mesh and microcavity array, the stretchable electrode in our experiments exhibited high stretchability with a low ΔR/R₀ of 10.3 at a strain of 40%. Furthermore, by exploiting the self-patterning ability based on the difference in the ability to form liquid bridges according to the distance between microstructures, we successfully demonstrated a stretchable AgNW bundle mesh with complex patterns without using additional patterning processes. In particular, stretchable electrodes were fabricated by spray coating and bar coating, which are widely used in industry for low-cost mass production.