Research Article www.acsami.org
Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Silicones for Stretchable and Durable Soft Devices: Beyond Sylgard-184 Sungjune Park,† Kunal Mondal,† Robert M. Treadway, III,† Vikash Kumar,† Siyuan Ma,‡ James D. Holbery,‡ and Michael D. Dickey*,† †
Department of Chemical and Biomolecular Engineering, North Carolina State University, 911 Partners Way, Raleigh, North Carolina 27695, United States ‡ Applied Sciences Group, Microsoft Corporation, Redmond 98052, Washington, United States S Supporting Information *
ABSTRACT: This paper identifies and characterizes silicone elastomers that are well-suited for fabricating highly stretchable and tear-resistant devices that require interfacial bonding by plasma or UV ozone treatment. The ability to bond two or more pieces of molded silicone is important for creating microfluidic channels, chambers for pneumatically driven soft robotics, and other soft and stretchable devices. Sylgard-184 is a popular silicone, particularly for microfluidic applications. However, its low elongation at break (∼100% strain) and moderate tear strength (∼3 N/mm) make it unsuitable for emerging, mechanically demanding applications of silicone. In contrast, commercial silicones, such as Dragon Skin, have excellent mechanical properties yet are difficult to plasma-bond, likely because of the presence of silicone oils that soften the network yet migrate to the surface and interfere with plasma bonding. We found that extracting silicone oligomers from these soft networks allows these materials to bond but only when the Shore hardness exceeds a value of 15 A. It is also possible to mix highly stretchable silicones (Dragon Skin and Ecoflex) with Sylgard-184 to create silicones with intermediate mechanical properties; interestingly, these blends also only bond when the hardness exceeds 15 A. Eight different Pt-cured silicones were also screened; again, only those with Shore hardness above 15 A plasma-bond. The most promising silicones from this study are Sylgard-186 and Elastosil-M4130 and M4630, which exhibit a large deformation (>200% elongation at break), high tear strength (>12 N/mm), and strong plasma bonding. To illustrate the utility of these silicones, we created stretchable electrodes by injecting a liquid metal into microchannels created using such silicones, which may find use in soft robotics, electronic skin, and stretchable energy storage devices. KEYWORDS: soft robotics, stretchable electronics, silicones, microfluidics, soft lithography, liquid metals surface bonding.13,14 A typical process of Sylgard-184 involves mixing two liquid components from a kit, degassing the mixture, casting (pouring), and curing. The low viscosity of Sylgard-184 facilitates its handling throughout this process. Stretchable devices are often processed in a similar manner yet have more demanding mechanical requirements than conventional static microfluidic devices. Stretchable devices benefit from elastomers that can undergo high elongation at break and high tear resistance. Using these metrics, Sylgard-184 is no longer the best choice, yet it is often used for soft devices out of familiarity. A replacement for Sylgard-184 (for the fabrication of stretchable devices) should have improved mechanical properties yet remain commercially available, easy to process (low viscosity and ability to be molded), and in
1. INTRODUCTION There is a growing interest in soft actuators and electrical devices for a wide range of applications including stretchable electronics,1,2 soft robotics,3−6 and electronic skin.7 Many of these devices utilize silicones because they are soft, stretchable, commercially available, easy to process, biocompatible, and can be bonded together using a variety of techniques. For example, the ability to mold and bond silicone is a common way to create microfluidic channels (for manipulating fluids and patterning stretchable conductors such as liquid metals8), pneumatic cavities (for actuating soft robotics),9 and soft sensors (for electronic skin).3,10−12 Poly(dimethylsiloxane) (PDMS) is a common silicone for fabricating soft devices. In particular, Sylgard-184 is a silicone popularized by the microfluidics community because of several favorable properties including optical transparency, processability, moldability, moderate tear resistance (provided by silica particle fillers), and the ability to be sealed by plasma-activated © XXXX American Chemical Society
Received: December 3, 2017 Accepted: March 12, 2018
A
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Table 1. Mechanical Properties of Select Silicone Elastomers Recommended for Fabricating Stretchable Devices. Tear Strength Values of MED 10-6400 and ExSil 100 from Technical Data Sheets elastomers
elongation at break (%)
tear strength (N/mm)
Young’s modulus (MPa)
plasma bond
limitation
Sylgard-184 Ecoflex00‑10 Sylgard-186 Dragon Skin10 slow MED 10-6400 Elastosil-M4130 Elastosil-M4630 ExSil 100
100 697 195 507 484 449 787 1864
3.4 3.9 12.2 22.0 28.2 29.2 33.8 42.0
3.9 0.1 0.6 0.5 2.0 0.9 1.7 0.02
yes no yes no yes yes yes yes
poor tear strength poor tear strength, non-bonding by plasma high viscosity (70 000 cp) non-bonding by plasma processability (contains solvent) low optical transparency low optical transparency extremely soft (microchannels collapse)
lengthens the processing time. ExSil 100 is too soft (0.02 MPa of elastic modulus); molded structures collapse after release from the mold because of the softness of the elastomers.20,21 Even microchannels composed of Sylgard-184 (3.94 MPa elastic modulus) collapse or sag at aspect ratio values above 5 or below 0.5.18 The tendency of channels to collapse depends both on the modulus and the geometry of the channels, so we did not necessarily want to rule out silicones based on modulus alone, yet we found in practice that silicones with a modulus below 0.5 MPa collapsed using microchannels of dimensions typically used in our laboratory (50−100 μm). Nevertheless, silicones such as Ecoflex are popular for soft robotics, which use much larger features than microchannels. Thus, we did not eliminate all soft silicones from consideration even though they are likely to collapse for microfluidic applications. Taking all of these considerations into account, we focused on the elastomers listed in Table 1. In addition to screening alternative elastomers, we explored the trade-offs of blending Sylgard-184 with Dragon Skin and Ecoflex, two silicones with incredible mechanical properties that cannot plasma-bond on their own. This blending approach was explored previously to create devices 22 but not fully characterized. Herein, we characterized and compared the mechanical properties of promising silicones. We found that Sylgard-186 is appealing. It is from the same silicone “family” as Sylgard-184 (and the processing should therefore be familiar to users of Sylgard-184) but has a higher tear resistance and has a higher elongation at break. It does, however, have a large viscosity relative to 184. Elastosil-M4130 and M4630 also exhibit excellent mechanical properties and the ability to plasmabond; thus, they are also promising for fabricating stretchable and tear-resistant microfluidic devices, although they are not transparent. In addition, we provide general guidelines that are necessary to plasma-bond silicones. We identify a Shore hardness threshold below which a variety of silicones and silicone mixtures do not plasma-bond, presumably because of the ability of low-molecular weight “silicone oil” species to diffuse to the interface (and thus interfere with bonding) in the softer networks. Extracting these oligomers enables these silicones to bond but also modifies the desirable mechanical properties. To demonstrate the utility of such materials, we created microchannels and injected them with a liquid metal (EGaIn, eutectic gallium indium, 75% Ga and 25 % In by weight) because it is an attractive conductor due to its intrinsic stretchability.23 The ability to pattern liquid metals has been explored to realize stretchable wires,24 stretchable antennas,25 and self-healing interconnects.26 This liquid metal-incorporated highly stretchable microfluidic device can be implemented in
many cases, retain the ability to be sealed by plasma-activated surface bonding. There are several strategies for improving the mechanical properties of silicones. Silicones can be toughened by adding particle fillers (e.g., fumed silica), but this also increases the viscosity and often stiffens the silicone and decreases the elongation at break. When choosing silicones, we set an upper limit of 70 000 cp viscosity (Sylgard-184 has a viscosity of 3500 cp); although this limit is somewhat arbitrary, in practice, it is difficult to process silicones above these viscosities. It is possible to increase the extension at break of silicones by increasing the distance between cross-links (e.g., using bottle-brush polymers15 or polymerization strategies16). Silicones can also be toughened, in principle, by adding reversible bonds, although to date, this has only been employed in other elastomers.17 Silicones can also be toughened (and softened) by adding noncross-linked oligomers or cyclics into the formulation, but these silicones can diffuse to the surface and inhibit interfacial plasma bonding. For example, commercial elastomers such as Dragon Skin have fantastic mechanical properties, yet they are not easy to plasma-bond. One solution to this problem is to partially cure a layer of these materials to leave reactive groups that facilitate bonding between another silicone surface, but this approach is not trivial. If the materials are brought together too soon, the under cured layer can flow into the molded microchannels. If the materials are brought together too late, the interfacial bond may be weak. Finally, even if the elastomers can be bonded, if the silicone is too soft, then, the resulting channels are prone to collapse if the aspect ratio (width/height) is too large.18−20 We sought to identify a commercial, Pt-cured silicone elastomer (or combination of silicones) that has improved mechanical properties relative to Sylgard-184 yet would be easy to bond using conventional bonding techniques. We gave preference to commercial elastomers because they are readily available, familiar, and easy to process. Although there are many different commercially available silicones (so many that it is not possible to screen all of them), we narrowed the possibilities by consulting technical data sheets of commercial elastomers, by seeking recommendations from several companies who produce or distribute elastomers (Dow Corning, Wacker, NuSil, and Smooth-on), and by consulting the literature. We considered Sylgard-186 (Dow Corning), Elastosil M4130 and M4630 (Wacker), MED 10-6400 (NuSil), and the remarkably stretchable ExSil 100 (Gelest). Although several of these exhibit high elongations (up to 1864%) at break (Figure S1, Supporting Information) and high tear resistances (up to 42 N/mm) as reported in the technical data sheets, we ruled several of them out for a variety of reasons. MED 10-6400 is cast from a solvent, which complicates handling and B
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 1. (a) Plot of stress vs strain for various commercial silicones. The slope approaching zero strain provides the modulus. The maximum stress indicates tensile strength and the maximum strain indicates elongation at break. (b) Force−extension plots of the silicone elastomers utilized for determining tear resistance using ASTM standard D624-00.
Figure 2. (a) Stress−strain plots of Sylgard-184, Dragon Skin10 slow, and blended silicones as a function of mass ratio. (b) Young’s modulus of blended silicones as a function of weight percent of Dragon Skin10 slow. (c) Normal force−extension plots of Sylgard-184, Dragon Skin10 slow, and blended silicones as a function of mass ratio. (d) Tear resistance of blended silicones as a function of weight percent of Dragon Skin10 slow. The bars in the plot represent the standard deviation from three measurements.
stretchable and soft robotics,3,4,10 wearable sensors,27 and bioinspired electronics.28
Ecoflex00‑10 (697%), Dragon Skin10 slow (507%), ElastosilM4130 (449%), and Sylgard-186 (195%). For most stretchable devices, it is desirable to have soft elastomers, although microchannels can collapse if the modulus of the silicone is too low. The slope in the linear portion of the stress−strain profile (Figure 1a) corresponds to Young’s modulus. Sylgard-184 shows the highest Young’s modulus (3.94 MPa). Sylgard-186 exhibits a lower value (0.56 MPa), similar to that of Dragon Skin10 slow (0.50 MPa). Table 1 reports the modulus for the other silicones. This work is primarily motivated by identifying silicones with an improved tear resistance relative to Sylgard-184. Figure 1b reports the maximum forces required to tear silicones using ASTM standard D624-00. The tear resistance is the maximum force to initiate tearing divided by the thickness of the samples. Although Ecoflex and Dragon Skin both show a high elongation at break (Figure 1a), they exhibit completely different values of tear resistance (Figure 1b). Although Dragon Skin has a high
2. RESULTS AND DISCUSSION 2.1. Mechanical Properties. We compared several PDMS silicone elastomers to identify the best silicones for highly stretchable and tear-resistant microfluidic devices formed by plasma bonding. Stretchable devices should have large elongation at break values to avoid catastrophic failure during elongation. As shown in Figure 1a, we characterized the tensile stress−strain relationship of silicones using the ASTM D412 Type C standard. Sylgard-184, which provides a standard for the sake of comparison, exhibits 3.97 MPa of tensile stress at the maximum elongation (at break) of ∼100%. These results are comparable to the technical data sheet and previously reported mechanical properties.29 All silicones used in this work exhibit a higher elongation at break than Sylgard-184: Elastosil-M4630 (787%), C
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 3. (a) Stress−strain plots of Sylgard-184, Ecoflex00‑10, and blended silicones as a function of mass ratio. (b) Young’s modulus of blended silicones as a function of weight percent of Ecoflex00‑10. (c) Normal force−extension plots of Sylgard-184, Ecoflex00‑10, and blended silicones as a function of mass ratio. (d) Tear resistance of blended silicones as a function of weight percent of Ecoflex00‑10.
tear resistance (21.95 N/mm), Ecoflex tears at 3.89 N/mm, similar to that of Sylgard-184 (3.37 N/mm). Sylgard-186 exhibits an enhanced tear resistance (12.20 N/mm) compared to that of Sylgard-184, consistent with a higher concentration of silica particle fillers in the polymeric network (also consistent with the larger viscosity of Sylgard-186 prior to curing).30−34 Elastosil-M4630 and M4130 also exhibit remarkably high tear strengths. Resilience is also an important property for soft and stretchable devices.35 It enables silicones to return to their original shape after stretching. We therefore characterized the resilience of the silicones as a function of strain (Figures S2 and S3, Supporting Information). Resilience describes the energy recovered after the removal of stress divided by the total energy required for deformation, that is, the ratio of energy released in recovery to the energy that caused deformation. After strain cycling (i.e., stretching to near the elongation at break and relaxing to zero strain) one time to minimize Mullins effect,35 we performed four additional strain cycles. As shown in Figures S2 and S3 (Supporting Information), the silicones characterized in this work exhibit negligible hysteresis over the entire strain cycle with resilience values ≥92% even as the strain approaches the elongation at break. The lack of notable hysteresis during the stress−strain cycles indicates that the silicones are elastically reversible and that minimal permanent changes occur in the polymeric networks during strain cycling. Ecoflex, exhibits the lowest resilience at low strains (20 and 50% strain) and shows significant hysteresis during consecutive strain cycles at high strains (600%). 2.2. Mixing Strategy. Although Sylgard-184 is popular within the microfluidic community, its modest tear resistance is a drawback for applications requiring stretchable and deformable microfluidic devices. Previously, the mechanical properties of Sylgard-184 have been enhanced by mixing with Dragon Skin while preserving the ability to plasma-bond.22
Inspired by this work, we explored the blending approach to enhance the mechanical properties of Sylgard-184. The silicones used in this work employ similar curing chemistries; they cure via hydrosilyation reactions of vinyl groups in a silicone and a silicone hydride in a cross-linker in the presence of Pt catalyst.36,37 As shown in Figure 2a, a combination of 75% Dragon Skin and 25% Sylgard-184 exhibits an elongation at break similar to that of pure Sylgard-184 but with a decreased tensile stress. In other words, it provides no obvious benefit relative to pure Sylgard-184. For this reason, we did not explore mixtures with less than 75% Dragon Skin. As the amount of Dragon Skin increases, the elongation at break of the blend improves. However, there is little enhancement of elongation of the silicone blends with higher than 83% Dragon Skin, presumably because of the intrinsically limited elongation of Sylgard-184. Figure 2b shows the tear resistance of mixtures of Dragon Skin and Sylgard-184. Blending enhances the mechanical strength (elongation at break, tensile stress, and tear resistance) and decreases the stiffness of the materials. A mixture of Ecoflex with Sylgard-184 can also improve the elongation at break and decrease the elastic modulus compared to pure Sylgard-184, as shown in Figure 3a. However, there is no significant improvement of tear resistance in the blends because each elastomer has a similar tear resistance (Figure 3b). The mechanical properties of blended silicones are summarized in Table S1 in the Supporting Information. We also characterized the viscosity of blended silicones as a function of weight percent of Sylgard-184 (Figure S4, Supporting Information). Initially, Dragon Skin and Ecoflex exhibit relatively high viscosities (23 000 and 14 000 cp). Increasing the amount of Sylgard-184 therefore lowers the viscosity of the blended silicones. These results indicate that the viscosity and mechanical properties of silicones can be D
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 4. Shore A hardness of (a) blended silicones of Sylgard-184 with Dragon Skin10 slow and Ecoflex00‑10 as a function of weight percent of Sylgard184 and (b) various pure silicones used in this work. Gray area in (a) marks the plasma-bondable range (data of Shore A hardness of ExSil 100 and MED 10-6400 taken from their technical data sheet).
Figure 5. (a) Shore A hardness of Dragon Skin10 slow extracted by soaking it in toluene as a function of time. (b) Weight loss of Dragon Skin10 slow extracted by soaking it in toluene as a function of time. Gray area in (a) marks the plasma-bondable range.
modulus of 0.02 MPa yet it can plasma-bond (Shore A hardness of 15 A). It bonds, presumably, because it does not contain many silicone oils and because it achieves its mechanical properties from chemistries that provide large molecular weights between cross-links.16 To better understand the role of silicone “oils” on bonding, we extracted noncross-linked oils from Dragon Skin by soaking it in toluene. After soaking it for various durations of time, we measured the weight loss, new Shore A value, and whether it plasma-bonds. As shown in Figure 5a,b, Shore A hardness and weight loss of the silicones increase as the silicone soaks longer in toluene. As expected, this indicates that the extraction of oils increases Shore A hardness. Extracted silicones that have Shore hardness lower than 14 A do not plasma-bond. Interestingly, we observe that all silicones studied here only plasma-bond with a Shore A hardness value >15. This threshold occurs in blends (Figure 4a), commercial silicones (Figure 4b), and silicone networks with oils extracted by a solvent (Figure 5). During exposure to plasma, the surface of silicone networks (which are naturally hydrophobic) reacts to form hydroxyl groups that are critical to bonding. It is well-known that these hydrophilic surfaces can undergo “hydrophobic recovery” either through diffusion of silicone oils or reorientation of surface groups. Previous studies suggest that plasma treatment, similar to that used here, forms a thin silica-like layer on the surface, making it difficult for surfaces to reorient. Silicone oils, however, can still diffuse through this layer.44,45 Thus, we reason that silicones with increased Shore hardness can plasmabond because they have reduced amounts of “silicone oils” (and perhaps decreased mobility of oils) that would otherwise prevent plasma bonding.46 Intuitively, one may consider increasing the plasma dose to prevent diffusion of oils. However, prior studies have shown that increasing the duration of plasma exposure leads to the in situ creation of low-
manipulated via a simple blending approach as a function of mass ratio. 2.3. Plasma Bonding. An important contribution of this work is identifying silicones with desirable mechanical properties that can also bond. There are several approaches to bond silicones such as oxygen plasma bonding, corona discharge, partial curing, and uncured silicone adhesive technique.38 Among the strategies, oxygen plasma-induced surface activation has been used extensively in the fabrication of PDMS microfluidic devices for sealing microchannels. It creates reactive polar functional groups, that is, silanol group, to the surface. When brought together with another plasma-treated substrate, the reactive surfaces form covalent bonds to create excellent adhesion. It is common for silicone networks to contain uncross-linked silicone oils (low-molecular weight silicone species).39−43 These oils soften the networks. They also can migrate to the surface and interfere with plasma bonding. On the basis of this reasoning, we explored the blending approach to identify the greatest amount of Dragon Skin and Ecoflex that can be added to Sylgard-184 without losing the ability to plasma-bond. 2.4. Predicting Plasma Bonding. Interestingly, we found that both blends only plasma-bond if Shore A hardness is 17.5 A or greater, as shown in Figure 4a. Therefore, we hypothesized that Shore hardness might be a predictor of the ability to plasma-bond silicones. All of the elastomers tested with Shore A hardness values above 15 A bonded, as shown in Figure 4b. The softest silicone that bonded was ExSil 100, with a Shore A hardness of 15 A. Although this study is not exhaustive, it suggests that a general guideline for plasma bonding is to have silicones with Shore A hardness higher than 15 A. Interestingly, arranging the data by modulus does not provide the same clear distinction as predicted by Shore A hardness because of one exception: the softest silicone, ExSil 100, has a Young’s E
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 6. (a) Schematic illustration showing the fabrication of the microfluidic device consisting of silicone elastomers and a liquid metal electrode. (b) Image of the microfluidic device and the optical microscope image of liquid metal-filled 100 μm wide microchannels. (c,d) Photographs of the stretchable microfluidic device consisting of Sylgard-186 and the liquid metal electrode in the microchannels (c) before and (d) after stretching. (e,f) Optical microscope images of a microchannel 600 μm wide (e) before and (f) after stretching to 180% strain.
3. CONCLUSIONS In this paper, we report and compare several silicone elastomers to fabricate a highly stretchable and deformable microfluidic device via soft lithography. Sylgard-184 is commonly used for microfluidic applications; however, it has a limited elongation at break and low tear resistance. Silicones, such as Dragon Skin, exhibit promising mechanical properties, but they do not plasma-bond, which is important for adhering two or more layers together. Mixing of Sylgard-184 with Dragon Skin can improve the mechanical properties (relative to 184) while preserving the capability to plasma-bond as long as the ratio of Dragon Skin to 184 is between 3:1 and 7:1. In general, we found that silicones with Shore A hardness >15 can plasmabond. Understanding this threshold better is a future area of work; we reason that silicone networks above this threshold have decreased silicone “oils”, thus making it easier to plasmabond. We characterized additional several silicones and found that Sylgard-186, Elastosil-M4630, and M4130 are appealing for creating stretchable microfluidic devices because of their high elongation, high tear resistance, and plasma bonding ability. Elastosil-M4630 and M4130 have outstanding mechanical properties, but these are not transparent presumably because of oxide particles in the polymeric networks. Sylgard-186 is transparent but viscous. We demonstrated an application of microchannels composed of Sylgard-186 by injecting them with a stretchable liquid metal to create stretchable wires. These studies indicate that of the silicones studied here, Sylgard-186, Elastosil-M4630, and M4130 are the best suited for applications in soft robotics, stretchable electronics, and other soft/ stretchable devices that utilize silicones.
molecular weight surface species as well as microcracks that enable silicone oils to reach the surface.47,48 2.5. Choosing the Best Silicone. We compared several PDMS silicone elastomers and blends to seek better silicones to fabricate stretchable and tear-resistant devices that can seal by plasma bonding. We found Elastosil-elastomers (M4630 and M4130) and Sylgard-186 are appealing because of their high tear resistance, high elongation at break, and ability to plasmabond. These silicones are not without drawbacks: Sylgard-186 has a relatively large viscosity (increasing the processing time to cast and degas) but can still be processed for replica molding. It is transparent, whereas the Elastosil-elastomers (M4630 and M4130) are hazy. It is possible to reduce the viscosity of Sylgard-186 by the addition of a nonreactive silicon fluid diluent or Sylgard-184; however, it diminishes the mechanical properties. 2.6. Molding. To demonstrate the utility of such materials, we created microchannels by replica molding followed by plasma bonding. We followed the procedures typically used for processing Sylgard-184 as shown in Figure 6a. These microchannels can be utilized in a variety of ways. Here, we show it is possible to fill them with a liquid metal to fabricate stretchable microfluidic conductors (Figure 6b). After the oxygen plasma bonding, we filled the liquid metal (EGaIn, eutectic gallium indium, 75% Ga and 25% In by weight) into the microchannels by vacuum filling to pattern stretchable liquid metal conductors.49 The thin oxide layer on the surface of EGaIn stabilizes it within the channel despite the high tension of the metal.50 Figure 6c,d shows images of the stretchable microfluidic device consisting of Sylgard-186 and the liquid metal electrode in the microchannels before and after stretching, respectively. Figure 6e,f shows optical microscope images of the microchannels filled with the liquid metal embed in the microfluidic device. The elastomeric device stretches to larger strains without breaking, although there are some wrinkles on the surface of the metal that appear during stretching, as expected. 51 The electrical properties of stretchable liquid metal conductors have been demonstrated previously.23−25 We also fabricated stretchable devices using Elastosil-M4130 and M4630 (Figure S5). To prevent sticking of Elastosil-M4630 to the silicon mold, we cured it at room temperature.
4. EXPERIMENTAL METHODS 4.1. Materials. Various silicone elastomers such as Ecoflex00‑10, Dragon Skin10 slow (Smooth-on), Sylgard-184, Sylgard-186 (Dow Corning), MED 10-6400 (NuSil), ExSil 100 (Gelest), and ElastosilM4630 and M4130 (Wacker Chemicals) were used for the preparation of molds to characterize the mechanical properties. The materials were mixed with their curing agent as described in their technical note and thermally cured at 60 °C for 24 h. 4.2. Mechanical Characterization. The mechanical properties of the materials were characterized using Instron 5943 with a 1 kN load cell. To measure the tensile stress of silicones as a function of strain, the elastomers were poured into a “dog bone”-shaped mold as indicated in the ASTM D412 Type C standard. Two hydraulic grips held 1 in. sections of the material mold and stretched at a constant rate F
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces of 10 mm/min until it broke. To measure the tear resistance, we prepared smooth test pieces with a 90° notch on one side and with tab ends as indicated in ASTM D624-00. The test pieces were strained until they ruptured completely at a rate of grip separation of 500 mm/ min. The gauge length between the two grips was recorded. The measured forces were converted into tear strength based on the crosssectional area of the molds. Viscoelastic properties of blended silicones were characterized by a commercial rheometer (AR 2000). Shore A hardness of silicones was measured using a Shore A durometer (MegaBrand) as described in ASTM D2240. To measure the adhesion strength of the silicones induced by plasma treatment, we attempted to quantify it using a 90° peel test of two plasma-bonded slabs of elastomers.52−54 However, the results were binary: either the interface would delaminate readily (i.e., the silicones were not capable of plasma bonding) or the samples bonded so strongly that the substrates simply elongated rather than delaminating. 4.3. Fabrication of Microfluidic Devices. Thermally cured silicones with microchannels were replicated from a patterned Si mold produced by conventional photolithography. The elastomers were bonded to thin silicone layers by oxygen plasma treatment (0.5 mbar, 50 W, Diener Electronics) for 30 s. The microchannels were filled with the liquid metal (eutectic gallium indium, 75% Ga and 25% In by weight) by applying vacuum through punched holes made in the devices.49
■
(5) Palagi, S.; Mark, A. G.; Reigh, S. Y.; Melde, K.; Qiu, T.; Zeng, H.; Parmeggiani, C.; Martella, D.; Sanchez-Castillo, A.; Kapernaum, N.; Giesselmann, F.; Wiersma, D. S.; Lauga, E.; Fischer, P. Structured Light Enables Biomimetic Swimming and Versatile Locomotion of Photoresponsive Soft Microrobots. Nat. Mater. 2016, 15, 647−653. (6) Li, S.; Zhao, H.; Shepherd, R. F. Flexible and Stretchable Sensors for Fluidic Elastomer Actuated Soft Robots. MRS Bull. 2017, 42, 138− 142. (7) Benight, S. J.; Wang, C.; Tok, J. B. H.; Bao, Z. Stretchable and Self-Healing Polymers and Devices for Electronic Skin. Prog. Polym. Sci. 2013, 38, 1961−1977. (8) Dickey, M. D. Emerging Applications of Liquid Metals Featuring Surface Oxides. ACS Appl. Mater. Interfaces 2014, 6, 18369−18379. (9) Marchese, A. D.; Katzschmann, R. K.; Rus, D. A Recipe for Soft Fluidic Elastomer Robots. Soft Robot. 2015, 2, 7−25. (10) Bauer, S.; Bauer-Gogonea, S.; Graz, I.; Kaltenbrunner, M.; Keplinger, C.; Schwödiauer, R. 25th Anniversary Article: A Soft Future: From Robots and Sensor Skin to Energy Harvesters. Adv. Mater. 2014, 26, 149−162. (11) Majidi, C. Soft Robotics: A PerspectiveCurrent Trends and Prospects for the Future. Soft Robot. 2013, 1, 5−11. (12) Bao, Z. Skin-Inspired Organic Electronic Materials and Devices. MRS Bull. 2016, 41, 897−904. (13) McDonald, J. C.; Whitesides, G. M. Poly(dimethylsiloxane) as a Material for Fabricating Microfluidic Devices. Acc. Chem. Res. 2002, 35, 491−499. (14) Lee, J. N.; Park, C.; Whitesides, G. M. Solvent Compatibility of Poly(dimethylsiloxane)-Based Microfluidic Devices. Anal. Chem. 2003, 75, 6544−6554. (15) Vatankhah-Varnoosfaderani, M.; Daniel, W. F. M.; Zhushma, A. P.; Li, Q.; Morgan, B. J.; Matyjaszewski, K.; Armstrong, D. P.; Spontak, R. J.; Dobrynin, A. V.; Sheiko, S. S. Bottlebrush Elastomers: A New Platform for Freestanding Electroactuation. Adv. Mater. 2017, 29, 1604209. (16) Goff, J.; Sulaiman, S.; Arkles, B.; Lewicki, J. P. Soft Materials with Recoverable Shape Factors from Extreme Distortion States. Adv. Mater. 2016, 28, 2393−2398. (17) Wu, J.; Cai, L.-H.; Weitz, D. A. Tough Self-Healing Elastomers by Molecular Enforced Integration of Covalent and Reversible Networks. Adv. Mater. 2017, 29, 1702616. (18) Qin, D.; Xia, Y.; Whitesides, G. M. Soft Lithography for Microand Nanoscale Patterning. Nat. Protoc. 2010, 5, 491−502. (19) Xue, Y.; Kang, D.; Ma, Y.; Feng, X.; Rogers, J. A.; Huang, Y. Collapse of Microfluidic Channels/Reservoirs in Thin, Soft Epidermal Devices. Extreme Mech. Lett. 2017, 11, 18−23. (20) Huang, Y. Y.; Zhou, W.; Hsia, K. J.; Menard, E.; Park, J.-U.; Rogers, J. A.; Alleyne, A. G. Stamp Collapse in Soft Lithography. Langmuir 2005, 21, 8058−8068. (21) Hui, C. Y.; Jagota, A.; Lin, Y. Y.; Kramer, E. J. Constraints on Microcontact Printing Imposed by Stamp Deformation. Langmuir 2002, 18, 1394−1407. (22) Russo, S.; Ranzani, T.; Gafford, J.; Walsh, C. J.; Wood, R. J. Soft Pop-up Mechanisms for Micro Surgical Tools: Design and Characterization of Compliant Millimeter-Scale Articulated Structures. 2016 IEEE International Conference on Robotics and Automation (ICRA), 2016; pp 750−757. (23) Dickey, M. D. Stretchable and Soft Electronics Using Liquid Metals. Adv. Mater. 2017, 29, 1606425. (24) Zhu, S.; So, J.-H.; Mays, R.; Desai, S.; Barnes, W. R.; Pourdeyhimi, B.; Dickey, M. D. Ultrastretchable Fibers with Metallic Conductivity Using a Liquid Metal Alloy Core. Adv. Funct. Mater. 2013, 23, 2308−2314. (25) So, J.-H.; Thelen, J.; Qusba, A.; Hayes, G. J.; Lazzi, G.; Dickey, M. D. Reversibly Deformable and Mechanically Tunable Fluidic Antennas. Adv. Funct. Mater. 2009, 19, 3632−3637. (26) Palleau, E.; Reece, S.; Desai, S. C.; Smith, M. E.; Dickey, M. D. Self-Healing Stretchable Wires for Reconfigurable Circuit Wiring and 3D Microfluidics. Adv. Mater. 2013, 25, 1589−1592.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b18394. Mechanical properties of silicones, resilience of silicones, blending of silicones, and photos of stretchable microfluidic devices consisting of silicones filled with the liquid metal in microchannels (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Michael D. Dickey: 0000-0003-1251-1871 Funding
The authors gratefully acknowledge the support from Microsoft Corp. and the NSF-ASSIST Center for Advanced Self Powered Systems of Integrated Sensors and Technologies Center (EEC1160483). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors gratefully acknowledge Norman Riley at Wacker Chemical Corporation for providing Elastosil-M4630 and M4130 and helpful discussions, Prof. Jan Genzer (NC State) for helpful discussions, and Bruce Hilman at Dow-Corning for providing Sylgard-186 and helpful discussions.
■
REFERENCES
(1) Ahn, J.-H.; Je, J. H. Stretchable Electronics: Materials, Architectures and Integrations. J. Phys. Appl. Phys. 2012, 45, 103001. (2) Boley, J. W.; White, E. L.; Chiu, G. T.-C.; Kramer, R. K. Direct Writing of Gallium-Indium Alloy for Stretchable Electronics. Adv. Funct. Mater. 2014, 24, 3501−3507. (3) Kim, S.; Laschi, C.; Trimmer, B. Soft Robotics: A Bioinspired Evolution in Robotics. Trends Biotechnol. 2013, 31, 287−294. (4) Rus, D.; Tolley, M. T. Design, Fabrication and Control of Soft Robots. Nature 2015, 521, 467−475. G
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces (27) Kim, D.-H.; Lu, N.; Ma, R.; Kim, Y.-S.; Kim, R.-H.; Wang, S.; Wu, J.; Won, S. M.; Tao, H.; Islam, A.; Yu, K. J.; Kim, T.-I.; Chowdhury, R.; Ying, M.; Xu, L.; Li, M.; Chung, H.-J.; Keum, H.; McCormick, M.; Liu, P.; Zhang, Y.-W.; Omenetto, F. G.; Huang, Y.; Coleman, T.; Rogers, J. A. Epidermal Electronics. Science 2011, 333, 838−843. (28) Song, Y. M.; Xie, Y.; Malyarchuk, V.; Xiao, J.; Jung, I.; Choi, K.J.; Liu, Z.; Park, H.; Lu, C.; Kim, R.-H.; Li, R.; Crozier, K. B.; Huang, Y.; Rogers, J. A. Digital Cameras with Designs Inspired by the Arthropod Eye. Nature 2013, 497, 95−99. (29) Johnston, I. D.; McCluskey, D. K.; Tan, C. K. L.; Tracey, M. C. Mechanical Characterization of Bulk Sylgard 184 for Microfluidics and Microengineering. J. Micromech. Microeng. 2014, 24, 035017. (30) Winberg, P.; Eldrup, M.; Maurer, F. H. J. Nanoscopic Properties of Silica Filled Polydimethylsiloxane by Means of Positron Annihilation Lifetime Spectroscopy. Polymer 2004, 45, 8253−8264. (31) Vondrácě k, P.; Schätz, M. NH3-Modified Swelling of SilicaFilled Silicone Rubber. J. Appl. Polym. Sci. 1979, 23, 2681−2694. (32) Camenzind, A.; Schweizer, T.; Sztucki, M.; Pratsinis, S. E. Structure & Strength of Silica-PDMS Nanocomposites. Polymer 2010, 51, 1796−1804. (33) Sun, C.-C.; Mark, J. E. Comparisons among the Reinforcing Effects Provided by Various Silica-Based Fillers in a Siloxane Elastomer. Polymer 1989, 30, 104−106. (34) Bokobza, L. The Reinforcement of Elastomeric Networks by Fillers. Macromol. Mater. Eng. 2004, 289, 607−621. (35) Case, J. C.; White, E. L.; Kramer, R. K. Soft Material Characterization for Robotic Applications. Soft Robot. 2015, 2, 80−87. (36) Chambon, F.; Winter, H. H. Stopping of Crosslinking Reaction in a PDMS Polymer at the Gel Point. Polym. Bull. 1985, 13, 499−503. (37) Lewis, L. N.; Stein, Judith.; Gao, Y.; Colborn, R. E.; Hutchins, G. Platinum Catalysts Used in the Silicones Industry. Platinum Met. Rev. 1997, 41, 66−75. (38) Eddings, M. A.; Johnson, M. A.; Gale, B. K. Determining the optimal PDMS−PDMS bonding technique for microfluidic devices. J. Micromech. Microeng. 2008, 18, 067001. (39) Hourlier-Fargette, A.; Antkowiak, A.; Chateauminois, A.; Neukirch, S. Role of Uncrosslinked Chains in Droplets Dynamics on Silicone Elastomers. Soft Matter 2017, 13, 3484−3491. (40) Millet, L. J.; Stewart, M. E.; Sweedler, J. V.; Nuzzo, R. G.; Gillette, M. U. Microfluidic Devices for Culturing Primary Mammalian Neurons at Low Densities. Lab Chip 2007, 7, 987−994. (41) Eddington, D. T.; Puccinelli, J. P.; Beebe, D. J. Thermal Aging and Reduced Hydrophobic Recovery of Polydimethylsiloxane. Sens. Actuators, B 2006, 114, 170−172. (42) Hillborg, H.; Tomczak, N.; Olàh, A.; Schönherr, H.; Vancso, G. J. Nanoscale Hydrophobic Recovery: A Chemical Force Microscopy Study of UV/Ozone-Treated Cross-Linked Poly(dimethylsiloxane). Langmuir 2004, 20, 785−794. (43) Owen, M. J.; Smith, P. J. Plasma Treatment of Polydimethylsiloxane. J. Adhes. Sci. Technol. 1994, 8, 1063−1075. (44) Kim, J.; Chaudhury, M. K.; Owen, M. J. Hydrophobic Recovery of Polydimethylsiloxane Elastomer Exposed to Partial Electrical Discharge. J. Colloid Interface Sci. 2000, 226, 231−236. (45) Smith, P. J.; Owen, M. J.; Holm, P. H.; Toskey, G. A. Surface Studies of Corona-Treated Silicone Rubber High-Voltage Insulation. [Proceedings] 1992 Annual Report: Conference on Electrical Insulation and Dielectric Phenomena, 1992; pp 829−836. (46) Bodas, D.; Khan-Malek, C. Hydrophilization and Hydrophobic Recovery of PDMS by Oxygen Plasma and Chemical treatmentAn SEM Investigation. Sens. Actuators, B 2007, 123, 368−373. (47) Fritz, J. L.; Owen, M. J. Hydrophobic Recovery of PlasmaTreated Polydimethylsiloxane. J. Adhes. 1995, 54, 33−45. (48) Kim, J.; Chaudhury, M. K.; Owen, M. J.; Orbeck, T. The Mechanisms of Hydrophobic Recovery of Polydimethylsiloxane Elastomers Exposed to Partial Electrical Discharges. J. Colloid Interface Sci. 2001, 244, 200−207.
(49) Lin, Y.; Gordon, O.; Khan, M. R.; Vasquez, N.; Genzer, J.; Dickey, M. D. Vacuum Filling of Complex Microchannels with Liquid Metal. Lab Chip 2017, 17, 3043−3050. (50) Dickey, M. D.; Chiechi, R. C.; Larsen, R. J.; Weiss, E. A.; Weitz, D. A.; Whitesides, G. M. Eutectic Gallium-Indium (EGaIn): A Liquid Metal Alloy for the Formation of Stable Structures in Microchannels at Room Temperature. Adv. Funct. Mater. 2008, 18, 1097−1104. (51) Mohammed, M. G.; Dickey, M. D. Strain-Controlled Diffraction of Light from Stretchable Liquid Metal Micro-Components. Sens. Actuators, A 2013, 193, 246−250. (52) Kovalchick, C.; Molinari, A.; Ravichandran, G. Rate Dependent Adhesion Energy and Nonsteady Peeling of Inextensible Tapes. J. Appl. Mech. 2013, 81, 041016. (53) Eitner, U.; Rendler, L. C. The Mechanical Theory behind the Peel Test. Energy Procedia 2014, 55, 331−335. (54) Chen, H.; Feng, X.; Huang, Y.; Huang, Y.; Rogers, J. A. Experiments and Viscoelastic Analysis of Peel Test with Patterned Strips for Applications to Transfer Printing. J. Mech. Phys. Solid. 2013, 61, 1737−1752.
H
DOI: 10.1021/acsami.7b18394 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
