Chained Iron Microparticles for Directionally ... - ACS Publications

Mar 28, 2017 - ... directionally responsive materials for applications in soft robotics. ... is made available by participants in Crossref's Cited-by ...
1 downloads 0 Views 4MB Size
Research Article www.acsami.org

Chained Iron Microparticles for Directionally Controlled Actuation of Soft Robots Marissa M. Schmauch,†,§ Sumeet R. Mishra,† Benjamin A. Evans,|| Orlin D. Velev,‡ and Joseph B. Tracy*,† †

Department of Materials Science and Engineering and ‡Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, North Carolina 27695, United States § Department of Chemistry and Biochemistry, University of Tulsa, Tulsa, Oklahoma 74104, United States || Department of Physics, Elon University, Elon, North Carolina 27244, United States S Supporting Information *

ABSTRACT: Magnetic field-directed self-assembly of magnetic particles in chains is useful for developing directionally responsive materials for applications in soft robotics. Using materials with greater complexity allows advanced functions, while still using simple device architectures. Elastomer films containing chained magnetic microparticles were prepared through solvent casting and formed into magnetically actuated lifters, accordions, valves, and pumps. Chaining both enhances actuation and imparts a directional response. Cantilevers used as lifters were able to lift up to 50 times the mass of the polymer film. We introduce the “specific torque”, the torque per field per mass of magnetic particles, as a figure of merit for assessing and comparing the performance of lifters and related devices. Devices in this work generated specific torques of 68 Nm/kgT, which is significantly higher than in previously reported actuators. Applying magnetic fields to folded accordion structures caused extension and compression, depending on the accordion’s orientation. In peristaltic pumps comprised of composite tubes containing embedded chains, magnetic fields caused a section of the tube to pinch closed where the field was applied. These results will facilitate both the further development of soft robots based on chained magnetic particles and efforts to engineer materials with higher specific torque. KEYWORDS: magnetic particles, chains, elastomer, anisotropy, actuation, lifter, valve, pump



INTRODUCTION Polymer composites that can be remotely actuated are of interest for applications in soft robotics, microelectromechanical systems (MEMS), lab-on-a-chip devices, and biomedical devices.1−3 Magnetic actuation in particular is an essential approach for remote actuation because it provides controlled, noncontact actuation. Soft polymer composites containing embedded magnetic particles, where applied magnetic fields modulate the mechanical properties of the composite, are known as magnetorheological elastomers (MREs),4 and magnetoactive elastomers (MAEs) are a subset of MREs that undergo shape changes in applied magnetic fields.5 In MAEs, magnetic actuation may be caused either by torques, when a homogeneous magnetic field acts to align magnetic moments, or by forces generated by magnetic field gradients.6−15 While these are fundamentally distinct effects, they may occur simultaneously in some magnetic field geometries. Furthermore, assembling the particles into chains within the polymer enhances the magnetic response and imparts directional actuation, since alignment of the chains in the direction of the magnetic field is energetically favored.16 This magnetic anisotropy arises from favorable head-to-tail coupling of © XXXX American Chemical Society

magnetic moments within the particle chains, which results when an external field oriented collinear with the particle chains aligns the magnetic moments of the particles in the same direction. The concept of using chained magnetic particles for directional and selective actuation has been established but has yet to be fully developed and optimized for applications in soft robotics.16−22 Chaining magnetic particles adds complexity to material actuation, enabling greater control and therefore enhancing the appeal of MAEs for applications in soft robotics. Artificial muscles, grippers, and wormlike undulating robots are some of the most common soft robotic devices because they offer high deformation, numerous and controllable degrees of freedom, and rapid response.23−25 Here, we report designs and demonstrations for magnetically actuated lifters, “accordions”, valves, and peristaltic pumps, where chained magnetic microparticles (MMPs) cause a response that depends on the direction of the applied field, allowing for more complex behaviors and stronger responses than uniformly dispersed Received: January 24, 2017 Accepted: February 24, 2017

A

DOI: 10.1021/acsami.7b01209 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

gravitational torque on the loaded cantilever. Magnetic torques (τ) ⃗ | = 0.5 Lmg sin θcom, where L and were therefore given by τ = |r⃗ × fgrav m are the length and mass of the cantilever, and θcom is the angle from vertical to the center of mass of the cantilever−Parafilm system. It should be noted that for larger Parafilm masses (≥7.8×), the center-ofmass angle was not the complement of the cantilever angle (θc + θcom ≠ 90°) due to the size of the attached Parafilm mass (see Figure S2 in the Supporting Information). Throughout this manuscript, reported angles refer to θc, the angle of the MAE cantilever with respect to the horizontal magnetic field. Accordion Experiments. Strips of polymer films (l = 60 mm, w = 16 mm) with the MMP chains oriented lengthwise were folded into “accordions” with between two and nine folds parallel to the width of the rectangular film. For creasing, the flat film was placed between two layers of aluminum foil. The foil and polymer composite multilayer was then creased and sandwiched between two glass slides, which were clamped and placed in an oven at 80 °C. After heating for 10 min, the slides were removed from the oven, and the clamps, glass, and aluminum foil were removed after cooling to room temperature. The film partially unfolded and retained the creases. The film was mounted on a glass slide (completely attaching one panel of the accordion to the slide) using double-sided tape, and the glass slide was attached to a wooden block. Two orientations of the accordion were investigated: Supported extension experiments were conducted by orienting the glass slide vertically and observing extension as the partially folded accordion extended horizontally with a horizontal applied field from the electromagnet, while placed on a horizontal support (wood). In unsupported compression experiments, the slide was placed in the field horizontally, and the accordion hung open because gravity caused significant vertical extension under its own weight. Application of horizontal magnetic fields pulled and folded the film upward, causing compression. For actuating the accordion structures, a maximum field of 7.9 kOe was applied. Pinch Valve and Peristaltic Pump Experiments. A pinch valve (circumference = 60 mm, w = 6 mm) was prepared by rolling a strip of composite film with chains oriented lengthwise and securing the ends together with double-sided tape. The valve was actuated in a uniform, horizontal magnetic field from the electromagnet of up to 2.11 kOe oriented across the valve opening. As the magnitude of the magnetic field increased, the valve flattened; its shape could be tuned by controlling the magnitude of the external applied magnetic field. For peristaltic pumps, three rectangular film samples (l = 60 mm, w = 16 mm) were prepared, which were identical except for the arrangement of the MMPs in the film. One sample was prepared outside the magnetic field so that chaining did not occur; the other two were prepared with the chains oriented along the width and the length of the rectangular film. To form cylindrical tubes that could serve as peristaltic pumps, the rectangular composite films were rolled widthwise and then taped together using double-sided tape to form cylinders. To study actuation as a peristaltic pump, the tubes were mounted on glass pipets using double sided tape, so that they could be passed through a horizontal magnetic field produced by two permanent magnets (0.5” diameter, 0.5” long cylindrical NdFeB magnets from Bunting Magnetics, N35P500500). The magnets were mounted 1.1 cm apart on a block of wood, and a field of ∼4 kOe was measured within the gap. Two small pieces of wood were placed near the faces of the magnets to reduce side-to-side movement of the tube as it was pulled through the gap.

MMPs. In our experiments, lifters were able to lift up to 50 times the mass of the polymer film. Applying magnetic fields to folded accordion structures caused extension and compression, depending on the orientation of the accordion in the field. In peristaltic pumps comprised of composite tubes containing embedded chains, magnetic fields caused a section of the tube to pinch closed where the field was applied. We have also developed a generalized figure of merit for assessing the performance of MAE actuators.26 These findings demonstrate the potential for remotely actuatable polymer composites, which will motivate research to further optimize the performance of these devices (including considerations of the structural design, the magnetic properties of individual MMPs, assembly into chains, and the properties of the polymer27), to investigate incorporation of MMP chains into other kinds of matrices and devices, and to make multifunctional devices through integration with other kinds of materials.



EXPERIMENTAL SECTION

Polymer Film Preparation. Polymer films were prepared via solvent casting Irogran PS455-203, a thermoplastic polyurethane (TPU) provided by Huntsman Corporation. Beads of the TPU were dissolved in tetrahydrofuran (THF, EMD, OmniSolv, Non-UV) in a ratio of 0.2 g of TPU per mL of THF, to which carbonyl iron microparticles (MMPs, Jilin Jien Nickel Industry, JCF2-2) obtained from Lord Corporation (average diameter of 4.2 μm) were added before solvent casting. The mass of MMPs was chosen to give a loading of 10.2 wt % MMPs in the composite films after solvent removal. THF was added to the MMP powder, and the slurry was then added and mixed into the TPU solution. For each film, 1 g of the mixture was poured into a custom-made, rectangular poly(tetrafluoroethylene) (PTFE) mold with a length of 60 mm and width of 16 mm. The solvent evaporated in approximately 2 h, yielding a film ∼100 μm thick. For chained samples, the mold was placed between the poles of a GMW 3472-70 electromagnet in a horizontal field of 300 Oe immediately after the dispersion was poured. When preparing samples for lifters and accordions, the PTFE mold was placed in the electromagnet with the long side of the mold parallel to the field direction. For unchained control samples, the same solvent casting process was performed but without placing the mold in the electromagnet. After solvent casting, the sides of the film had small vertical segments, where the dispersion had adhered to the side of the PTFE mold during solvent casting. These sections were manually removed with scissors, resulting in flat rectangular films. Lifting Experiments. Polymer films, prepared as described above, with the MMP chains oriented lengthwise, were cut into strips (l = 26 mm, w = 12 mm). To create a cantilever, the strip was affixed with double-sided tape to a glass slide mounted on the edge of a rigid, nonmagnetic platform, such as a wooden block. Two types of weights were connected to the cantilever using double-sided tape. In one experiment, small, dense objects were attached on the underside of the opposing (free) end. In another set of experiments for analysis of the torques, larger rolls of Parafilm were taped to the underside of the entire portion of the cantilever on the free side of the hinge. The cantilever was placed in the gap of the same electromagnet used to prepare the films with the lever arm hanging down nearly vertically, such that applying a horizontal magnetic field would rotate the lever arm up. The cantilever and attached weight were controllably lifted by gradually increasing the magnetic field strength up to a maximum of 5.7 kOe, beyond which further lifting was not observed. Analysis of Lifting Experiments. In the Parafilm lifting experiments, videos were recorded and analyzed to obtain the angle of the loaded cantilever with respect to horizontal, θc, as a function of the magnetic field strength. The experiment was repeated with different loadings, ranging from unloaded (0×) to a loading of just over 50 times the mass of the cantilever (50.1×). Measurements of the position of the cantilever were taken under equilibrium conditions. The magnetic torque was inferred by equating it with the opposing



RESULTS AND DISCUSSION Soft robotic devices were fabricated from thin films (∼100 μm thick) of a thermoplastic polyurethane (TPU) elastomer, Irogran, containing linear chains of 4 μm iron MMPs. While previous studies have employed a variety of sizes of magnetic particles, the actuation mechanism does not intrinsically depend on size, though size-dependent coercivity or superparamagnetism could affect the response. For this study, 4 μm iron particles were used because of their high saturation B

DOI: 10.1021/acsami.7b01209 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces magnetization, commercial availability, and lack of magnetic hysteresis (vide infra). We chose Irogran because it is a commercially available elastomer with sufficiently low elastic modulus that is compatible with sample preparation through solvent casting. The chained-MMP samples were prepared by solvent casting in an electromagnet, providing a uniform, horizontal field of 300 Oe, which drives chaining through magnetic field-directed self-assembly.28,29 During the assembly process, the magnetized chains also repel each other laterally, causing them to disperse uniformly throughout the film. When magnetic fields are subsequently applied to the chained samples, torques are generated within the chains, causing them to align with the direction of the field. The torque is caused by the favorable (negative) and unfavorable (positive) dipolar coupling energies, when chains of MMPs are magnetized head to tail and side by side, respectively. This torque to align the chains with the direction of the applied field may compete with other torques from gravity or from elastic energy in the polymer. In our electromagnet, the magnetic fields were always horizontal, and alignment of the chains within the composite films caused levers to lift near to horizontal, valves to close, and accordion structures to extend or compress. In photographs of the composite films, chains of MMPs are visible by eye, and individual MMPs can be imaged with an optical microscope (Figure 1). Magnetometry measurements at 300 K show that the samples have a saturation magnetization of 24.2 emu/g and exhibit magnetic anisotropy caused by chaining, such that they magnetize more easily parallel rather than perpendicular to the chain direction. The magnetization curves run through the origin, giving negligible coercivity and remanent magnetization, which is consistent with the use of MMPs originally developed for magnetorheological fluids. Consequently, effects of magnetic hysteresis are not anticipated in actuation studies. Lifter and Specific Torque. When a rectangular strip of the composite with chains oriented parallel to the length is held fixed at one end, gravity causes the free end of the flexible strip to bend down to nearly vertical. Subsequent application of a uniform magnetic field in the horizontal direction causes the strip to rotate upward until the magnetic and gravitational torques are balanced (Figure 2a,b and Supporting Information, Figure S1). The magnetic torque depends on both the magnitude of the applied field and the angle of cantilever with respect to the field, and in a single lifting experiment, it is not possible to independently determine the effect of each. However, by loading the lever arm and measuring the angle of the lifted cantilever with a series of progressively larger masses, from unloaded (0×) to 50 times the mass of the cantilever (50.1×; Figure 2c and Supporting Information, Figure S2), additional information is obtained to distinguish the two effects. From these data, the magnetic torque (τ) was calculated for a series of different loadings at a constant cantilever angle. These plots were repeated at fixed cantilever angles between 25° and 65° (Figure 2d). While these plots are linear at high fields above ∼0.15 T, where the magnetization of the MMPs has saturated, there is a short nonlinear segment at low fields, before the MMPs reach saturation. The slope of the line at high fields therefore represents a concise and effective characterization of the torque behavior of the material. We refer to this value, the torque per unit mass of MMP per unit field (Nm/ kgT), as the “specific torque”. Devices in this work generated specific torques of 68 ± 2 Nm/kgT, which is significantly

Figure 1. Optical micrographs of (a) iron MMPs (average d = 4.2 μm) dropcast onto a glass slide and (b) chained MMPs in a TPU thin film. (c) Magnetometry of chained thin films at 300 K measured perpendicular and parallel to chains with inset photograph of polymer thin film containing chained MMPs. (d) Photographs of MMP films used in lifter, accordion, and peristaltic pump devices.

higher than previously reported actuators. We present the specific torque as a figure of merit for evaluating rotational magnetic actuators in homogeneous magnetic fields. The specific torque of our chained-particle MAE material at different cantilever angles is plotted in Figure 2e. The torque is maximized when the cantilever is oriented at an angle of approximately 45° from the applied magnetic field, which is consistent with our previous experiments and modeling of magnetic cilia.8 Magnetic chains in our cantilever produce as much as 68 ± 2 Nm/kgT for fields ranging from 0.05 to 0.6 T, enabling a 2.6 cm cantilever to lift 50 times its own weight. By comparison, reports of unchained (homogeneously dispersed) iron oxide nanoparticles in silicone cantilevers have yielded only 0.9 Nm/kgT,8 and other instances of chained-particle cantilevers have demonstrated 16−28 Nm/kgT using iron C

DOI: 10.1021/acsami.7b01209 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Accordion for Compression and Extension. Remotely actuatable structures are needed that can compress or extend in a fashion similar to artificial muscles.31−33 Real muscles and many artificial muscles can operate only in compression or extension, or bending in the case of electroactive polymers.34 An “accordion” structure capable of both compression and extension was prepared by placing multiple folds in a rectangular strip of the TPU containing chained MMPs, where the extension or compression is determined solely by the orientation of the accordion in the applied field (Figure 3). In

Figure 3. Photographs of composite “accordions” (l = 60 mm, w = 12 mm), where one fold is anchored to the substrate: (a) supported, two folds, relaxed (0 kOe) and extended (2.7 kOe), (b) supported, five folds, relaxed (0 kOe) and extended (5.4 kOe), (c) unsupported, two folds, relaxed (0 kOe) and compressed (5.3 kOe), and (d) unsupported, five folds, relaxed (0 kOe) and compressed (7.9 kOe). α is angle of the hinge between adjacent panels. The magnetic field was applied in a horizontal direction. Movies S2−3 for extension and compression of accordions are included in the Supporting Information.

the accordion, between two and nine hinges were formed from folds along the width of a rectangular strip of the TPU containing MMP chains oriented along its length. After adding the folds, the accordion had a zigzag configuration whose initial configuration in zero field was partially unfolded, such that actuation could cause either compression or further extension. For actuation experiments in uniform, horizontal magnetic fields, the accordion was held fixed at one end and placed on a nonmagnetic horizontal support or allowed to hang vertically unsupported. As the field was applied, torques generated by the MMP chains caused each panel to align with the horizontal field direction. On the horizontal support, the panels unfolded and opened to large obtuse angles, resulting in extension. When unsupported, alignment of the MMP chains with the field caused the panels to close to small acute angles, resulting in compression in the vertical orientation, against the force of gravity. A related, bulkier accordion structure has been reported, constructed from panels with a thickness of 3 mm containing chained MMPs anchored to an elastomer film, for which compression was demonstrated, but with a hinge angle between panels (α) greater than 20° in the compressed configuration.35 During unsupported compression, misfolding was also possible, where two neighboring hinges could open to large obtuse angles instead of small acute angles, causing three adjacent panels to align horizontally. Misfolding occurred when panels were misoriented as they hung under their own weight in zero field. The angle at which each panel tilted in zero applied field determined whether the panel would rotate

Figure 2. Photographs of a cantilever made from a composite film (l = 26 mm, w = 12 mm) with MMP chains oriented lengthwise, held fixed on the left end and with a piece of Parafilm (1.67 g, 50.1×) attached beneath the cantilever in (a) zero magnetic field and (b) 6.4 kOe horizontal field. (c) Cantilever angle (θc) vs applied horizontal magnetic field for cantilevers loaded with Parafilm. Parafilm loadings range from unloaded (0×) to 50.1 times the mass of the MAE cantilever (50.1×). (d) Magnetic torque per unit mass of iron MMP vs applied magnetic field, calculated from the lifting experiments in (c). Each plot represents the magnetic torque at a different, fixed cantilever angle. The slope of each line is the “specific torque,” plotted in (e).

oxide nanoparticles in TPU (16 Nm/kgT),20 and iron (19 Nm/ kgT) or barium ferrite (28 Nm/kgT) microparticles in silicone.30 D

DOI: 10.1021/acsami.7b01209 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

circumference of the cylinder, application of a uniform field along the diameter aligns the chains with the field direction and pinches the valve closed. The favorable magnetostatic energy associated with aligning the MMP chains with the field direction and closing the valve overcomes the elastic energy that otherwise holds the valve open. A pinch valve can be extended into a peristaltic pump by elongating it into a tube and moving it with respect to a smaller magnetic field region that pinches a portion of the tube. Peristaltic pumps were fabricated containing MMPs chained perpendicular or parallel to the pump’s length or randomly dispersed (Figure 4b−d). When placed in the horizontal magnetic field (∼4 kOe) in the gap between two permanent magnets, a segment of the tube pinched closed. The closed region can be propagated along the tube by moving the tube through the gap between the magnets. Control samples confirm that MMP chains perpendicular to the length enhance actuation, because this configuration uniquely allows the chains to align with the field direction when the tube pinches closed. Both the unchained sample and the sample with chains parallel to the pump’s length close to a lesser extent than the sample with chains perpendicular to the pump’s length. Actuation of the unchained peristaltic pump can be attributed to field gradients pulling the sides of the pump toward the permanent magnets. For the pump with chains oriented parallel to its length and perpendicular to the applied field, pinching the tube does not alter the perpendicular alignment between the chains and the field, which produces energetically unfavorable side-by-side interactions among the chained MMPs. The weaker pinching observed for chains oriented parallel to the length of the pump is therefore also attributed to magnetic field gradient effects.

clockwise or counterclockwise to give alignment in the applied horizontal field without moving through the energetically unfavorable vertical orientation. It should also be noted that the field in the electromagnet is not perfectly uniform, and field gradient effects could also have an effect by applying forces to displace (rather than torques to rotate) segments of the sample near the pole caps. In our experience, these field gradient effects are negligible, however, if the sample is placed near the center of the gap. Accordions with different numbers of folds exhibit similar extension (supported) and compression (unsupported) behaviors. An accordion with more folds and shorter panels is more compact but is also more susceptible to misfolding during compression than an accordion with fewer folds and longer panels. Longer panels are potentially more susceptible to field gradient effects, however, because their larger size will bring them closer to the poles of the electromagnet than shorter panels. Valve and Peristaltic Pump. Valves and pumps are soft robotic components with applications in many areas, including medicine, microelectronic devices, and even space exploration.36 Magnetic actuation can also be used for pumping fluids, as has been demonstrated in peristaltic pumps based on ferrofluids37 and MAEs.38 The previously reported MAE-based peristaltic pump used unchained MMPs, was actuated using magnetic field gradients, and had a high loading of 67 wt % iron MMPs and walls with thicknesses of a few mm, thus requiring a significant amount of polymer and MMPs.38 Here, we report simple designs for a pinch valve and peristaltic pump, where chains of MMPs allow actuation using uniform fields, enhance the response, and provide pinching using thin (∼100 μm) walls at a loading of 10 wt % iron MMPs. A pinch valve was constructed from a TPU strip with chained MMPs oriented lengthwise by rolling the strip into a cylinder (Figure 4a). Because the chains are oriented around the



CONCLUSIONS



ASSOCIATED CONTENT

Chained MMPs embedded in thin elastomer films provide strong, directional responses to magnetic fields, arising from the magnetic anisotropy associated with dipolar coupling of MMPs within the chains. Chained MMPs can provide improved performance and more complex behavior for soft robotics than unchained MMPs. The specific torque, defined as the magnetic torque generated per mass of MMPs per unit field, is a useful figure of merit for quantifying the performance of chainedMMP elastomers. Devices in this work generated specific torques of 68 ± 2 Nm/kgT, which is significantly higher than in previously reported actuators (0.9−28 Nm/kgT). Elastomer actuators based on chained MMPs were fabricated to investigate and demonstrate their capabilities: lifters that can lift up to 50 times the device’s own weight, accordions that compress and extend in applied fields to strains up to several hundred percent, and valve structures that close upon the application of an external field. A design for a peristaltic pump was also demonstrated, which could be especially useful for noncontact in vivo applications. Additional potential applications of these devices include lab-on-a-chip devices and MEMS. Future work to improve the performance and expand the function of these materials will include optimizing the assembly of the MMPs39,40 to maximize the specific torque and exploring polymer matrices that also respond to chemical stimuli.

Figure 4. (a) Closing a pinch valve (circumference = 60 mm, w = 6 mm) with chains around its circumference, when a uniform, horizontal magnetic field is applied. Peristaltic pumps (b) with chains perpendicular to tube length and around the circumference, (c) with chains parallel to tube length, and (d) unchained. The magnetic field in the gap between two permanent magnets is ∼4 kOe. Movies S4−5 for (a,b) are included in the Supporting Information.

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b01209. E

DOI: 10.1021/acsami.7b01209 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces



(12) Benkoski, J. J.; Breidenich, J. L.; Uy, O. M.; Hayes, A. T.; Deacon, R. M.; Land, H. B.; Spicer, J. M.; Keng, P. Y.; Pyun, J. Dipolar Organization and Magnetic Actuation of Flagella-Like Nanoparticle Assemblies. J. Mater. Chem. 2011, 21, 7314−7325. (13) Yoonessi, M.; Peck, J. A.; Bail, J. L.; Rogers, R. B.; Lerch, B. A.; Meador, M. A. Transparent Large-Strain Thermoplastic Polyurethane Magnetoactive Nanocomposites. ACS Appl. Mater. Interfaces 2011, 3, 2686−2693. (14) Evans, B. A.; Fiser, B. L.; Prins, W. J.; Rapp, D. J.; Shields, A. R.; Glass, D. R.; Superfine, R. A highly tunable silicone-based magnetic elastomer with nanoscale homogeneity. J. Magn. Magn. Mater. 2012, 324, 501−507. (15) Vach, P. Actuation of Iron Oxide-Based Nanostructures by External Magnetic Fields. In Iron Oxides; Wiley-VCH Verlag GmbH & Co. KGaA: Berlin, 2016; pp 523−544. (16) Kim, J.; Chung, S. E.; Choi, S.-E.; Lee, H.; Kim, J.; Kwon, S. Programming Magnetic Anisotropy in Polymeric Microactuators. Nat. Mater. 2011, 10, 747−752. (17) Boczkowska, A.; Awietjan, S. F.; Wroblewski, R. Microstructure−Property Relationships of Urethane Magnetorheological Elastomers. Smart Mater. Struct. 2007, 16, 1924−1930. (18) Wu, J.; Gong, X.; Fan, Y.; Xia, H. Anisotropic Polyurethane Magnetorheological Elastomer Prepared through In Situ Polycondensation Under a Magnetic Field. Smart Mater. Struct. 2010, 19, 105007. (19) Kuo, J.-C.; Huang, H.-W.; Tung, S.-W.; Yang, Y.-J. A HydrogelBased Intravascular Microgripper Manipulated Using Magnetic Fields. Sens. Actuators, A 2014, 211, 121−130. (20) Mishra, S. R.; Dickey, M. D.; Velev, O. D.; Tracy, J. B. Selective and Directional Actuation of Elastomer Films using Chained Magnetic Nanoparticles. Nanoscale 2016, 8, 1309−1313. (21) Erb, R. M.; Martin, J. J.; Soheilian, R.; Pan, C.; Barber, J. R. Actuating Soft Matter with Magnetic Torque. Adv. Funct. Mater. 2016, 26, 3859−3880. (22) Lum, G. Z.; Ye, Z.; Dong, X.; Marvi, H.; Erin, O.; Hu, W.; Sitti, M. Shape-Programmable Magnetic Soft Matter. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, E6007−E6015. (23) Trivedi, D.; Rahn, C. D.; Kier, W. M.; Walker, I. D. Soft Robotics: Biological Inspiration, State of the Art, and Future Research. Appl. Bionics Biomech. 2008, 5, 99−117. (24) Kim, S.; Laschi, C.; Trimmer, B. Soft Robotics: A Bioinspired Evolution in Robotics. Trends Biotechnol. 2013, 31, 287−294. (25) Gao, W.; Wang, L.; Wang, X.; Liu, H. Magnetic Driving Flowerlike Soft Platform: Biomimetic Fabrication and External Regulation. ACS Appl. Mater. Interfaces 2016, 8, 14182−14189. (26) Zupan, M.; Ashby, M. F.; Fleck, N. A. Actuator Classification and Selection-The Development of a Database. Adv. Eng. Mater. 2002, 4, 933−940. (27) Morales, D.; Bharti, B.; Dickey, M. D.; Velev, O. D. Bending of Responsive Hydrogel Sheets Guided by Field-Assembled Microparticle Endoskeleton Structures. Small 2016, 12, 2283−2290. (28) Tracy, J. B.; Crawford, T. M. Magnetic Field-Directed SelfAssembly of Magnetic Nanoparticles. MRS Bull. 2013, 38, 915−920. (29) Tokarev, A.; Yatvin, J.; Trotsenko, O.; Locklin, J.; Minko, S. Nanostructured Soft Matter with Magnetic Nanoparticles. Adv. Funct. Mater. 2016, 26, 3761−3782. (30) von Lockette, P.; Lofland, S. E.; Biggs, J.; Roche, J.; Mineroff, J.; Babcock, M. Investigating New Symmetry Classes in Magnetorheological Elastomers: Cantilever Bending Behavior. Smart Mater. Struct. 2011, 20, 105022. (31) Pelrine, R.; Kornbluh, R. D.; Pei, Q.; Stanford, S.; Oh, S.; Eckerle, J.; Full, R. J.; Rosenthal, M. A.; Meijer, K. Dielectric Elastomer Artificial Muscle Actuators: Toward Biomimetic Motion. Proc. SPIE 2002, 4695, 126−137. (32) Mirfakhrai, T.; Madden, J. D. W.; Baughman, R. H. Polymer Artificial Muscles. Mater. Today 2007, 10, 30−38. (33) Haines, C. S.; Lima, M. D.; Li, N.; Spinks, G. M.; Foroughi, J.; Madden, J. D. W.; Kim, S. H.; Fang, S.; Jung de Andrade, M.; Göktepe, F.; Göktepe, Ö .; Mirvakili, S. M.; Naficy, S.; Lepró, X.; Oh, J.; Kozlov, M. E.; Kim, S. J.; Xu, X.; Swedlove, B. J.; Wallace, G. G.; Baughman, R.

Additional photographs of weight lifting in applied magnetic fields. (PDF) Movie S1 of lifter in a horizontal magnetic field (MPG) Movie S2 of extension of a supported accordion in a horizontal magnetic field (MPG) Movie S3 of compression of an unsupported accordion in a horizontal magnetic field (MPG) Movie S4 of closing of a pinch valve in a horizontal magnetic field (MPG) Movie S5 of pinching of a peristaltic pump with MMP chains around the circumference in the gap between two permanent magnets (MPG)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Orlin D. Velev: 0000-0003-0473-8056 Joseph B. Tracy: 0000-0002-3358-3703 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was supported by the Research Triangle MRSEC REU Program (National Science Foundation Grant DMR1121107) and National Science Foundation Grant DMR1056653. Lord Corporation is acknowledged for providing the MMPs and Huntsman Corporation for providing the elastomer used in this study. We gratefully acknowledge Tim Fornes, Joey Faulk, and Jonathan Gillen for helpful discussions.



REFERENCES

(1) Medeiros, S. F.; Santos, A. M.; Fessi, H.; Elaissari, A. StimuliResponsive Magnetic Particles for Biomedical Applications. Int. J. Pharm. 2011, 403, 139−161. (2) Thévenot, J.; Oliveira, H.; Sandre, O.; Lecommandoux, S. Magnetic Responsive Polymer Composite Materials. Chem. Soc. Rev. 2013, 42, 7099−7116. (3) Gray, B. L. A Review of Magnetic Composite Polymers Applied to Microfluidic Devices. J. Electrochem. Soc. 2014, 161, B3173−B3183. (4) Zrínyi, M. Magnetically Responsive Polymer Gels and Elastomers: Properties, Synthesis and Applications. In Smart Polymers and their Applications; Woodhead Publishing: Cambridge, United Kingdom, 2014; pp 134−165. (5) Farshad, M.; Benine, A. Magnetoactive Elastomer Composites. Polym. Test. 2004, 23, 347−353. (6) Zrínyi, M.; Barsi, L.; Büki, A. Deformation of Ferrogels Induced by Nonuniform Magnetic Fields. J. Chem. Phys. 1996, 104, 8750− 8756. (7) Ramanujan, R. V.; Lao, L. L. The Mechanical Behavior of Smart Magnet−Hydrogel Composites. Smart Mater. Struct. 2006, 15, 952− 956. (8) Evans, B. A.; Shields, A. R.; Carroll, R. L.; Washburn, S.; Falvo, M. R.; Superfine, R. Magnetically Actuated Nanorod Arrays as Biomimetic Cilia. Nano Lett. 2007, 7, 1428−1434. (9) Garstecki, P.; Tierno, P.; Weibel, D. B.; Sagués, F.; Whitesides, G. M. Propulsion of Flexible Polymer Structures in a Rotating Magnetic Field. J. Phys.: Condens. Matter 2009, 21, 204110. (10) Nguyen, V. Q.; Ramanujan, R. V. Novel Coiling Behavior in Magnet-Polymer Composites. Macromol. Chem. Phys. 2010, 211, 618− 626. (11) Timonen, J. V. I; Johans, C.; Kontturi, K.; Walther, A.; Ikkala, O.; Ras, R. H. A. A Facile Template-Free Approach to Magnetodriven, Multifunctional Artificial Cilia. ACS Appl. Mater. Interfaces 2010, 2, 2226−2230. F

DOI: 10.1021/acsami.7b01209 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces H. Artificial Muscles from Fishing Line and Sewing Thread. Science 2014, 343, 868−872. (34) Carpi, F.; Kornbluh, R.; Sommer-Larsen, P.; Alici, G. Electroactive Polymer Actuators as Artificial Muscles: Are They Ready for Bioinspired Applications? Bioinspiration Biomimetics 2011, 6, 045006. (35) von Lockette, P.; Sheridan, R. Folding Actuation and Locomotion of Novel Magneto-Active Elastomer (MAE) Composites. Proceedings of the ASME 2013 Conference on Smart Materials, Adaptive Structures, and Intelligent Systems 2013, V001T01A020. (36) Laser, D. J.; Santiago, J. G. A Review of Micropumps. J. Micromech. Microeng. 2004, 14, R35−R64. (37) Kim, E.-G.; Oh, J.-G.; Choi, B. A Study on the Development of a Continuous Peristaltic Micropump Using Magnetic Fluids. Sens. Actuators, A 2006, 128, 43−51. (38) Fuhrer, R.; Schumacher, C. M.; Zeltner, M.; Stark, W. J. Soft Iron/Silicon Composite Tubes for Magnetic Peristaltic Pumping: Frequency-Dependent Pressure and Volume Flow. Adv. Funct. Mater. 2013, 23, 3845−3849. (39) Ye, L.; Pearson, T.; Dolbashian, C.; Pstrak, P.; Mohtasebzadeh, A. R.; Fellows, B.; Mefford, O. T.; Crawford, T. M. Magnetic-FieldDirected Self-Assembly of Programmable Mesoscale Shapes. Adv. Funct. Mater. 2016, 26, 3983−3989. (40) Jiang, X.; Feng, J.; Huang, L.; Wu, Y.; Su, B.; Yang, W.; Mai, L.; Jiang, L. Bioinspired 1D Superparamagnetic Magnetite Arrays with Magnetic Field Perception. Adv. Mater. 2016, 28, 6952−6958.

G

DOI: 10.1021/acsami.7b01209 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX