Symposium SM06—Materials and Fabrication Schemes for Robotics

Helical nanobots driven by magnetic fields can provide mechanical information about their local surrounding with sub-micron spatial resolution, probe complex, heterogenous biological environments, such as the intracellular environment, as well as the extracellular matrix; and provide a rich and powerful platform towards futuristic drug delivery. The same system can be used for sub-micron colloidal cargo manipulation in microfluidic devices and on-chip assembly applications. Their fabrication is based on shadow evaporation allowing a wafer scale route towards integration of many different functionalities in a single nanostructure. Crucially, their properties and functionalities can be tuned with great control using intelligent design of material and geometry, and I will present few examples of the same.

Steering of nano-/microhelices by a rotating magnetic field is considered a promising technique for controlled navigation of tiny objects through viscous fluidic environments. It was recently demonstrated that simple geometrically achiral planar structures can also be steered quite efficiently [1]. Such planar propellers are interesting for practical reasons, as they can be mass-fabricated using standard photolithography techniques. Following the earlier development of a theory of driven rotation and propulsion of magnetized object of an arbitrary shape in an in-plane rotating magnetic field [2], we propose general symmetry arguments (involving parity and charge conjugation) establishing correspondence between propulsive solutions of simple planar V-shaped structures on orientation of the dipolar magnetic moment [3]. In particular, it can be shown that in-plane magnetization results in propulsion due to a spontaneous symmetry breaking, whereas the rotating motors swim either parallel or anti-parallel to the field rotation axis depending on their initial orientation. Particular off-plane magnetization yields unidirectional propulsion typically associated with chiral structures, such as helices. Since planar micro/nano-structures are prone to in-plane magnetization and their uniform off-plane magnetization is not an easy task, the interesting question is whether they can be steered in a controllable fashion? Here we demonstrate that actuation by a conical rotating magnetic field (i.e., superposition of an in-plane rotating field and constant field orthogonal to it) can yield efficient unidirectional propulsion of planar and in-plane magnetized structures [4]. In particular, we found that the symmetrical V-shape magnetized along its symmetry axis which exhibits no net propulsion in in-plane rotating field, shows unidirectional in-sync propulsion with a constant (frequency-independent) velocity when actuated by the conical field. When the constant field is imposed in plane of the rotating field, it results in the net propulsion accompanied by the drift orthogonal to the axis of the field rotation [5]. Such setup can potentially be used to achieve spatial control over motion of individual propellers and their swarms.

Until now most of the theories considered propulsion through purely viscous Newtonian fluids. However, most practical (e.g., biomedical) applications of micro-/nanopropellers involve complex heterogeneous non-Newtonian (viscoelastic) fluidic environments. Modeling of such complex media would typically require advanced numerical approaches. In this talk I shall present the hydrodynamic slender-body theory based on phenomenological “two-fluid model” first introduced by de Gennes to study entangled polymer solutions. The model describes coupled dynamics of two phases – the viscous solvent and the embedded elastic polymer network. We demonstrate that the type of the boundary condition employed for the polymer network at the surface of the particle (e.g., slip, stick or depletion layer) plays a crucial role in determining its hydrodynamic mobility.


Acknowledgement. This work was supported in part by the Israel Science Foundation (ISF) via the grant No. 1744/17.

References
[1] U. K. Cheang et al., Phys. Rev. E, 2014, 90, 033007.
[2] K. I. Morozov et al., Phys. Rev. Fluids, 2017, 2, 044202.
[3] J. Sachs et al. Phys Rev. E, 2018, 98, 063105
[4] K.-J. Cohen et al., Phys. Rev Appl. 2019, 12, 014025
[5] K. I. Morozov and A. M. Leshansky, Phys. Chem. Chem. Phys. 2020, 16407

Magnetic micromotors and microrobotic systems benefit from hard magnetic and biocompatible materials, as well as schemes to avoid adhesion and overcome biological barriers.
While chemical targeting enables tissue-specific delivery, magnetically propelled microrobots can navigate through complex biological media through surface functionalization^[1-2] and can transport material towards individual cells. However, many commonly used magnetic materials or coatings are not biocompatible (Ni, Co) or possess weak magnetic moments (Fe, Fe₃O₄).
Here, we discuss both non-cytotoxic materials^[3] and medically approved coatings^[1-2] for magnetically actuated microrobots. Exciting applications range from targeted delivery through real tissues and to cells, as well as mobile contrast agents. In addition to passive coatings, it is also possible to consider chemically-active coatings where enzymes are immobilized^[2] or assembled onto selected proteinaceous templates^[4] which can facilitate the transport through complex biological fluids. Fully biocompatible magnetic microrobots combined with a suitable surface-chemistry are promising tools for biomedical applications.

1. Wu, Z.; Troll, J.; Jeong, H.-H.; Wei, Q.; Stang, M.; Ziemssen, F.; Wang, Z.; Dong, M.; Schnichels, S.; Qiu, T.; Fischer, P., A swarm of slippery micropropellers penetrates the vitreous body of the eye. Science Advances 2018, 4 (11), eaat4388.
2. Walker, D.; Käsdorf, B. T.; Jeong, H.-H.; Lieleg, O.; Fischer, P., Enzymatically active biomimetic micropropellers for the penetration of mucin gels. Science Advances 2015, 1 (11), e1500501.
3. Kadiri, V.M., Bussi, C., Holle, A.W., Son, K., Kwon, H., Schütz, G., Gutierrez, M.G. and Fischer, P., Biocompatible magnetic micro and nanodevices: fabrication of FePt nanopropellers and cell transfection. Advanced Materials 2020: 1-9.
4. Alarcon-Correa M, Guenther JP, Troll J, Kadiri VM, Bill J, Fischer P, Rothenstein D. Self-assembled phage-based colloids for high localized enzymatic activity, ACS nano, 2019, 13(5):5810-5.

Vigorous research efforts have advanced the state-of-the-art nanosensors with ultrahigh sensitivity for bioanalysis. However, a dilemmatic challenge remains: it is extremely difficult to obtain nanosensors that are both sensitive and high-speed for the detection of low-concentration molecules in aqueous samples. In this work, we demonstrate substantially accelerated sensing speed of low-concentration DNA molecules in aqueous suspension with retained high sensitivity by rotationally motorizing microsensors. An improvement of at least 3 to 4 times has been obtained from a sensor rotating at 630-1200 rpm. Theoretical analysis and modeling concerning both the convective-diffusion and the diffusion-absorption processes unveil the underlying working mechanism and the application range and limitation. This work provides a device scheme that could be applied to alleviate the dilemmatic challenge in biochemical sensing. The understanding of the complex interactions of molecules and moving micro-objects may assist the design of desired microrobotic systems for the capture, translocation, sensing, and release of biocargoes.

Miniaturized soft robots capable of omnidirectional walking through rough terrain are particularly promising for cutting-edge applications such as planetary exploration, military reconnaissance and disaster rescue. Especially, inspired by natural animals with sophisticated motion systems, great efforts have been devoted to developing walking robots based on different mechanisms. Nevertheless, most of the reported mini-robots are incapable of omnidirectional walking, which limits their controllability, active obstacle-avoidance ability, full space utilization and dead-corners elimination. The essential difficulty lies in that mini-robots with simplified components and reduced load-bearing are incapable of breaking motion symmetry in all directions through a highly controlled manner. In this study, we fabricated a small-scale crab robot (CraBot, overall dimension 2.5 cm) that enables omnidirectional walking via in-situ laser synergistic integration (LSI) of multiple graphene actuators at specified positions as joints. In this way, for the first time, we manipulated a mini-robot through coupled magnetism- and light- fields. The magnetic field regulates the center-of-mass of the CraBot, leading to an asymmetrical energy storage within the robot; meanwhile, the addressable light field selectively triggers the deformation of specific joints, releasing the accumulated energy into mechanical walk. In this way, the light field produces effective walking gait under the guidance of magnetic field synchronously. As a proof-of-concept walking to the left/right/front/back, as well as 45^o off its lateral directions are presented as typical walking paths. Such dual-field-coupled manipulation strategy may serve as a versatile mechanism for future robot designs.

The last two decades have seen significant efforts in developing micro and nanoscale machines with the ability to locomote through fluids and realizing tasks that require small-scale precision. One of the primary motivations of this research has been the realization of machines that could navigate the human body's vessels and ultimately perform medical missions, such as targeted drug delivery, microsurgery, localized diagnosis, or tissue regeneration. To this end, the use of magnetically driven small-scale robots is among the most investigated strategies, as magnetic fields are biocompatible in a wide range of conditions and they allow for a rich spectrum of locomotion mechanisms. While traditionally, research on small-scale magnetic robots has been done with rigid magnetic micro- and nanostructures made of metals or ceramics, recent efforts are dedicated to creating softer versions with the incorporation of organic materials such as polymers or hydrogels. Soft materials are usually more biocompatible, their mechanical properties resemble those of biological tissues, and they can be easily processed in different sizes and shapes. Composites consisting of small magnetic particulates dispersed in organic matrices are typically used to render these materials magnetic. Another approach involves coating polymers with magnetic films. Yet, these strategies are usually not optimal for magnetic manipulation. Note that the force or torque exerted in a magnetic body is proportional to the magnetic material's volume. Composites comprise a reduced number of particulates that can be hosted in their matrices, while films are limited in thickness. Additionally, when polymers are coated, some of their potential functionalities such as cargo transportation, shape transformation, or degradation can be severely compromised or inhibited.
To exploit the full capabilities of both stiff magnetic and soft organic materials, we present a process to manufacture interlocked metal-organic micromachines. The devices are fabricated by combining template-assisted metal electrodeposition and polymer microcasting in 3D printed molds obtained by two-photon polymerization (2PP). Sophisticated micrometric architectures such as cage-bar-ring metallic structures connected to multiple polymeric legs or metallic helices framed by a polymer casing can be successfully manufactured. We demonstrate the versatility of our approach by fabricating micromachines with iron as the magnetic component, and three prototypical polymers used for biomedical applications such as shape-memory polyurethane-based polymer (NOA-63), polydimethylsiloxane, and pure gelatin. Note that iron is the platable element with the highest saturation magnetization and displays optimal biocompatibility characteristics. Importantly, as our method involves mold-casting of polymers, the need for toxic cross-linkable moieties or photoinitiators can be avoided. In this talk, we will show the detailed manufacturing of these devices, the rich locomotion mechanisms of several complex interlocked microrobots, as well as strategies to tailor the buoyancy of the magnetic micromachines by means of the solvophobicity of the threaded polymeric structures. Finally, prospective applications such as control over the agglomeration of magnetic swarms will be discussed.

Directly electrically driving soft actuators offers clear system-level advantages over pneumatic actuation, but adds a number of challenges. For a wide range of soft robotic applications, such as haptics, and mobile robots, getting rid of the compressor, with its bulk, noise and low overall energy efficiency is a key step towards untethered and portable operation.

Due to its high energy density, we have focused on electrostatic actuation, using high electric fields to deform elastomers or textile structures. By using thin dielectrics, high permittivity and high electrical breakdown strength films, and low stiffness electrodes, we can both push up energy density and reduced drive voltage.

For a fully stretchable device, we reduced the drive voltage of Dielectric Elastomer Actuators (DEAs) from several kV to 400 V, a level at which we can use SMD components for compact control electronics (mass 350 mg). We report “feel-through” untethered cutaneous haptics, with DEA actuators only 18 µm thick[1], so thin the user does not feel them mounted on his or her fingertip when they are off. However, when the 3 mm diameter devices are turned on, the user feels localized pulsation that allows receiving rich haptic information.

Using zipping electrostatic actuation, such as electro-ribbon from prof. Rossiter’s group and peano-HASELS from prof. Keplinger’s group allows for higher energy density than DEAs because thin flexible materials such a polyimide have generally much better electrical properties than thin elastomers. We have developed arrays of sub-mm thick flexible actuators that generate over 60% strain and operate at over 300 Hz [2]. A 6 mm diameter actuator generates 300 mN normal force and 500 µm displacement. This is achieved using fluidically coupled electrostatic zipping and a combination of flexible polymers with high breakdown field and central silicone elastomer region in which no electric field is applied. We have made 5x5 arrays, mounted them directly on the arm and on consumer products, and report on user feedback for notification, control and navigation.

Finally, I will present an additive manufacturing approach to such devices, using ink jet printing to freely deposit elastomer, electrode and sacrificial layers for fluidic channels[3]. I will close by going over the challenges of integrating these technologies in a glove or suit and the promise of this field of soft electrostatic actuators.

[1] X. Ji et al., “Untethered Feel-Through Haptics Using 18-µm Thick Dielectric Elastomer Actuators,” Advanced Functional Materials, p. 2006639, 2020,
[2] E. Leroy, R. Hinchet, and H. Shea, “Multimode Hydraulically Amplified Electrostatic Actuators for Wearable Haptics,” Advanced Materials Technologies, 2020.
[3] S. Schlatter, G. Grasso, S. Rosset, and H. Shea, “Inkjet Printing of Complex Soft Machines with Densely Integrated Electrostatic Actuators,” Advanced Intelligent Systems, p. 2000136, 2020.

Reversible actuation through external stimuli is a critical component for the realization of smart devices in micro-mechanical, micro-optical and microfluidic systems. Significant advances in additive manufacturing has allowed for the creation of three-dimensional (3D) structures with sub-100 nm resolution(1). These advances in fabrication have allowed for considerable progress in the realization of micro-robots, such as micro-swimmers (2) or micro-grippers(3). Soft polymeric materials, such as responsive hydrogels, have the potential of augmenting the capabilities of hard micro-robotic units through the introduction of soft modules that mimic actuation of natural organisms. Hydrogels are hydrophilic, crosslinked polymer networks, that can absorb large quantities of water. Introduction of stimuli-responsive units in their structure, allows for their reversible transition from a hydrophilic (swollen) to a hydrophobic (contracted) state under external stimulation, in response to, for example, pH, temperature, light and magnetic fields. However, the main drawback of osmotically driven hydrogel actuators is their low actuation speed, where actuation force and response time are inextricably linked. This represents one of the major challenges for stimuli-responsive hydrogels which rely on the diffusion of analytes or bioanalytes. As response rate is dependent on the square of the hydrogel’s characteristic dimensions, reduction of hydrogel size to a micro regime could yield an accelerated and augmented response.

Herein we demonstrate the use of Direct Laser Writing (DLW) by multi-photon polymerization for the realization of 3D micro-actuators and sensors based on stimuli-responsive hydrogels and soft polymeric materials. We present the high resolution fabrication of a variety of structures, in a wide range of materials, including acrylated monomers, liquid crystal elastomers (4) and poly(ionic liquids)(5). These 3D structures can increase their size by over 300% through the absorption of water or solvent and can expel a large amount of the hydration media upon photo-, thermal or chemical stimulation (4D effect). Furthermore, domains of varying stiffness (e.g. hinges, joints) are integrated on demand by altering the laser exposure. Regions of low stiffness (reduced laser exposure) allow for increased swelling of the hydrogel material whilst high stiffness (e.g. obtained via increased laser exposure) act as hinges to control the direction of the fold. The single fabrication step and single resist formulation used to tailor the mechanical properties of soft polymer structures enables a simple processing method for the fabrication of complex 4D architectures via DLW.

3D micro-fabrication by DLW coupled with intrinsic (molecular) alignment is also demonstrated for the realization of dual-responsive soft polymer micro-structures. In this example, a supramolecular cholesteric liquid crystalline (CLC) elastomer is used for the fabrication of 4D structures, where the on-demand actuation is coupled with a color change. The CLC networks exhibit a self-organized helical photonic structure that can selectively reflect light, giving the 3D structures an initial color. Upon stimulation, via humidity (directly) and temperature (indirectly) changes, the network expands triggering a change in pitch of the CLC’s helical structure, therefore resulting in a shift of the reflection band. The micro-actuator’s expansion (up to 42 % in height at 75 %RH and at 19 °C) is accompanied by a color change. Such dual-response would be beneficial in microrobotics, enabling real-time interrogation of actuator status, through observation of the intrinsic color of the material.

1. M. Farsari, B. N. Chichkov, Nature photonics 3, 450 (2009).
2. C. Alcântara et al., Small 15, 1805006 (2019).
3. M. Power et al., Small 14, 1703964 (2018).
4. M. Del Pozo et al., ACS Nano 14, 9832 (2020).
5. A. Tudor et al., Materials Today 21, 807 (2018).

Soft robots have been attracting great attention for their high flexibility and versatility. One of the most significant shortcomings of inflatable soft robots is the low bending stiffness due to the intrinsic soft nature of elastomers. As a result, it becomes challenging to maintain the shape of a soft robot against gravitational force when the aspect ratio of a soft robotic manipulator increases (e.g. long and slender). To address this inherent drawback, we present a soft robot with reconfigurable and deactivatable skeleton. It consists of a soft elastomeric body reinforced with a rigid shape memory polymer (SMP) skeleton. Exploiting viscoelasticity of the SMP, the stiffness of the SMP skeleton can be varied over two orders of magnitude from ~1.2 GPa to ~2.4 MPa by resistive heating through an embedded liquid metal channel. At room temperature, the rigid SMP skeleton provides structural stability, making it possible for a large-scale soft robot to maintain its shape against gravity as vertebrates do. When flexible soft robotic actuation is needed, resistive heating is applied to deactivate the SMP skeleton mechanically, thereby rendering the soft robot flexible and pneumatically actuatable. Furthermore, the SMP allows the soft robot to be reconfigured into different shapes. This approach enables simple soft robotic gripping actuation with a large length scale of ~ 50 cm. We also demonstrate an amphibious locomotion of a soft robot which can reconfigure its four soft robotic limbs from straight flap for swimming in water, to curved legs for multi-pedal gaiting on rough terrains, to wheel-liked geometry for driving on flat surfaces. The reconfigurable and deactivatable skeleton may expand the applications of soft robots to diverse practical and industrial uses.

Often the doors to major technological innovations are opened by the discovery of novel functional materials. Dielectric elastomer transducers (DET) are an emerging technology, which gained importance over the last years. For this technology, the need for high dielectric permittivity polymers is higher than ever before. They hold great promise as active components in actuators, sensors, energy harvesting, artificial muscles, and soft robotics.
This presentation will give an overview of how we synthesized high dielectric permittivity elastomers. One approach to increase the dielectric permittivity consists of blending highly polarizable fillers in matrices with good elasticity. This approach failed because the materials are inhomogeneous at the molecular level and the mechanical and electromechanical properties are deteriorated. The second approach uses a chemical modification of polymers with polar groups. This allows the formation of homogenous materials, but it has the downside that the glass transition temperature (T_g) increases too much and turns unattractive for practical applications.
Using a highly flexible polymer backbone, we have shown that it is possible to achieve polymers with increased dielectric permittivity and low T_g. Our strategy is to chemically modify polymers that have a very low T_g with polar groups. Despite the increase in T_g due to dipolar interactions, the T_g of the modified polymers is still attractively low. Using this approach, we were able to achieve the highest dielectric permittivity ever for a neat elastomer, which exhibit unprecedentedly high dielectric breakdown fields. Such elastomers respond to both low electric fields and low voltages.

BioRobotics is a successful approach for designing robots from and for living beings. Bioinspired robot design is based on principles observed and modelled on living beings. In a beneficial loop, bioinspired robots also contribute to gain insight in biology. Either bioinspired or not, biorobots are those applied in the biomedical field.
A lesson learnt from living beings is that the physical body has a more important role in shaping intelligence than we think. Behaviour is not only controlled by the computation happening in the nervous system but emerges from the interaction of the body with the environment. It then depends on the physical properties of the body itself, on its morphology, on the environment it is operating in. This concept of embodied intelligence brought to rethink bodyware in robotics. Compliance became key and either compliant joints or soft materials have been used.
In this scenario, recent advances in materials, smart materials and fabrication techniques are enabling the development of soft robots, with new technologies for actuation, sensing, control, energy supply. Among the many possible applications of soft robots are the biomedical ones, e.g. surgery, prosthetics, rehabilitation [1]. Soft robotics technologies can also be used for building biomimetic organs or physical models of human body parts. The EU HybridHeart project is developing a fully soft artificial heart. A physical model of vocal cords was developed for studying larynx physiology and pathologies.
Bioinspired soft robots find applications in explorations, in natural environments. They can access remote areas, confined spaces, complex or collapsed structures, both in land and at sea. Marine applications of soft robots give interesting challenges and opportunities for developing bioinspired swimming and locomotion. While current underwater robots operate in the water column, soft robots can afford the interaction with the seabed or underwater plants [2].
Soft robots open up unprecedent abilities. The classical robot abilities of manipulation and locomotion assume new forms and new abilities become possible, like morphing, growing, self-healing, biodegrading [3]. Soft robots enable a vision for life-like abilities and behaviour, thanks to the bioinspired principles they embed and to the materials they are built of. In this framework, bio-hybrid robots are another intriguing paradigm, pursuing the integration between artificial materials and living cells/tissues. The unique features of living cells, optimized by millions of years of natural evolution, can be actually exploited to enable specific robot abilities and to achieve robots scaling toward small dimensions [4].
Overall, soft biorobots support a vision for future robots where they have a full life cycle, in analogy to living beings [5]. They are born but grow in body and intelligence, they learn and adapt their bodies, find their energy, self-heal their bodies and biodegrade at the end of their life. The vision for life-like robot behaviour is still in its infancy but shows an enormous potential. It provides challenges and opportunities for revolutionizing robotics and for exploring new materials and fabrications schemes.

[1] M. Cianchetti, C. Laschi, “Pleasant to the touch”, IEEE Pulse Magazine, 3, 34-37, 2016
[2] G. Picardi, M. Chellapurath, S. Iacoponi, S. Stefanni, C. Laschi, M. Calisti. "Bioinspired underwater legged robot for seabed exploration with low environmental disturbance", Science Robotics 5(42), 2020
[3] C. Laschi, M. Cianchetti, B. Mazzolai, “Soft robotics: Technologies and systems pushing the boundaries of robot abilities”, Science Robotics 1(1), 2016
[4] L. Ricotti, B. Trimmer, A.W. Feinberg, et al., “Biohybrid actuators for robotics: A review of devices actuated by living cells”, Science Robotics 2(12), 2017
[5] B. Mazzolai, C. Laschi, “A vision for future bioinspired and biohybrid robots”, Science Robotics 5(1), 2020

Reliable and sustainable automated manufacturing is predicted to play an even greater role in future industrial value creation than today. Automated handling processes are currently facing unprecedented challenges in terms of miniaturization, sustainability, and demanding production environments. Systems in worldwide use are mechanical grippers, electromagnetic actuators or vacuum suction devices. Requiring electrical energy and other valuable resources, these systems leave a large environmental footprint. They are dysfunctional with sensitive materials that suffer damage easily or with micro objects, causing low yields and high scrap rates. In this talk, it will be demonstrated that the new paradigm involving the tailoring of surface microstructures can offer innovative gripping solutions. Inspired by natural examples such as geckos and insects, micropatterning of polymeric surfaces allows the creation of various structure-related functionalities including the controlled adhesion and release of objects. These effects are based on van der Waals interactions and are largely independent of – sometimes environmentally harmful - materials chemistry; in particular, microfibrils can impart switchable stickiness to a material which is intrinsically non-adhesive. The sticking action is provided by a multitude of fine polymeric fibers (imitating a gecko foot); unsticking is provoked by actively switching the orientation of these hairs, resulting in accurate placement of an object. An elegant mechanical approach is to apply a compressive overload to the adhesive structures: the resulting elastic instability due to buckling of the fibrils leads to contact loss and creates a transition to a non-adhesive state. Such a load-controlled stimulus constitutes a reliable strategy and is straightforward to implement in industrial robotic systems. Subsequently, multistep switchable adhesives were developed, in which fibrils of varying length enable tuning of the adhesion strength. Other actuation mechanisms involve thermal stimuli, e.g. in combination with trained shape memory polymers or magnetic fields, which can trigger the bending of fibrils and induce a loss of contact.

The resulting Gecomer® technology has decisive advantages over conventional technologies: It is highly energy saving, resource sustainable, noise-free, compatible with sensitive surfaces, and applicable under extreme conditions where conventional solutions fail, e.g. in micro dimensions and in vacuum (as in space applications). To ensure reliable adhesion and controlled release, the pick-and-place process has been modelled by considering the interfacial stress distribution in the contact area. Governing parameters are the compressive preload, the size, shape and modulus of the fibril, the assumed defect statistics and the properties of the backing layer. An emerging challenge in microfabrication is the controlled release of small objects, where the weight of the objects to be handled is insufficient for easy detachment; for this, novel microstructure designs were developed that minimize the residual adhesion while ensuring high lateral precision during the placement step. Current developments included in-situ monitoring and machine learning to improve the reliability of the handling process. The new, precise and universal system solutions provided for automation, robotics and handling have already proven their practical suitability in demanding pilot applications. Examples of successful implementation will be demonstrated.

This talk will discuss recent progress in utilizing liquid metals as conductors for stretchable, soft, and reconfigurable components for soft robotics. Alloys of gallium are noted for their low viscosity, low toxicity, and near-zero vapor pressure. Despite the large surface tension of the metal, it can be patterned into non-spherical 2D and 3D shapes due to the presence of an ultra-thin oxide skin that forms on its surface. Because it is a liquid, the metal is extremely soft and flows in response to stress to retain electrical continuity under extreme deformation. By embedding the metal into elastomeric or gel substrates, it is possible to form soft, flexible, and conformal electrical components, stretchable antennas, and ultra-stretchable wires that maintain metallic conductivity up to ~800% strain. Thus, these materials are well-suited for soft robotics because they decouple electrical conductivity and mechanical properties. In addition to introducing the advantages of these materials for soft robotics, this talk will focus on recent work to utilize liquid metal for tactile sensors and energy harvesters. These advances have implications for soft machines and robots that have ultra-soft mechanical properties.

Robotic surfaces that can reshape and react to external stimuli offer opportunities to create soft, versatile machines that can multi-task, while interacting safely with their surroundings. Such systems are useful in applications that range from haptic interfaces, wearable exoskeletons and reconfigurable medical supports. Key properties of robotic surfaces include their ability to control their local stiffness, reprogram their target shape and have sufficient mechanical loadbearing ability, to support weights and manipulate objects. In this talk, I will describe recent solutions developed in our group, to create structured fabrics that have tunable bending stiffness and robotic surfaces that allow for large, reprogrammable, and pliable shape morphing into smooth 3D geometries. To develop these solutions, we design layered, architected materials, consisting of interlocking particles or networks of layered, polymeric ribbons. We employ different actuation methods, including vacuum pressure and Joule heating, to control the response of the surfaces. We demonstrate the ability to fabricate fabrics that become >25 times stiffer than their relaxed configuration, when a small external pressure (~93 kPa) is applied. We also show that robotic surfaces consisting of layers of heat responsive liquid crystal elastomers (LCEs) can be reprogrammed to assume arbitrary shapes.

We have developed a photolithography-inspired method of transferring predetermined fold patterns to carbon fiber veneers via selective ultraviolet epoxy curing, which allows us to create deployable structural components for microrobots and microflyers. Our technique uses photosensitive epoxies that require ultraviolet light for curing with a patterned mask to inhibit photopolymerization along fold lines but to allow light exposure and therefore hardening reactions to take place elsewhere. Using this technique, we constructed origami-inspired fold patterns on samples with thicknesses of 200 microns and areal densities of 20 mg/cm² from bi-woven carbon fiber fabrics without the need for cuts, living hinges, or buckling lines. This work represents, to our knowledge, the first use of ultraviolet epoxies to make folding composite materials.

Significant research has been devoted to improving aerial vehicles and robots by decreasing their sizes and reducing their masses. Toward the first strategy, small untethered microflyers with diameters of just 2 cm have been demonstrated¹; however, these robots’ small sizes were achieved by sacrificing aerodynamic components that provide maneuverability. To overcome this, deployable structures have been proposed as methods to increase microflyer control while maintaining compact dormant footprints². Toward the second strategy, new materials have been explored that provide enhanced strength with lower densities. A prime example is carbon fiber³, which has been widely used to fabricate micro aerial vehicles with dimensions >10 cm. Examples combining both of these performance enhancement strategies, i.e., foldable composite sheets, are currently limited to large space assemblies and composite load bearing structures that have a single degree of freedom^3-5. Employing these methods for small-scale vehicles has required specialized joint actuation and other design limitations^6,7.

There is thus a significant need for lightweight structural components that can be folded in multiple directions to allow arbitrary modular robot and microflyer designs. To bridge this gap, we have developed a method of transferring predetermined fold lines onto carbon fiber sheets using photosensitive epoxy resin and light-obstructing mask patterns. This method results in veneers with areas of high stiffness where ultraviolet light was exposed but thin foldable elastic areas where radiation was blocked. Importantly, our folds do not require cuts or fractures; rather, due to the absence of cured epoxy in the hinge areas, the carbon fibers can undergo microbuckling in a repeatable fashion⁸. Moreover, our technique allows fold patterns, such as those previously used in space applications⁹, to be quickly imparted to carbon fiber sheets in a reproducible and mass producible fashion at a smaller scale than previous work.

To fabricate our folding veneers, we impregnated a bi-woven carbon fiber fabric with photocurable epoxy, exposed it to ultraviolet light through a 2-mm-wide frame, additively manufactured mask, and cleaned the resulting veneer to remove excess uncured resin. We are currently subjecting our samples to mechanical testing to identify their bending stiffnesses and radii of curvature and are designing complex fold patterns to demonstrate these veneers’ deployable functionality. Our results represent a significant step forward in modular design and assembly toward deployable and lightweight components for robots and microflyers at such a small scale.

1) Piccoli ICRA (2017)3328
2) Cha J. Microelectromech. Syst. 29(2020)1127
3) Shi IOP Conf. Ser., Mater. Sci. Eng. 531(2019)012067
4) Peterson AIAA (2013)1667
5) Leclerc AIAA (2019)1522
6) Dufour IROS (2016)1576
7) Jacob AIAA (2009)745
8) Jimenez AIAA (2009)2633
9) Miura Int. J. Space Struct. 8(1993)3

This material is based upon work supported by the Defense Advanced Research Projects Agency (DARPA) under Contract No. HR0011-19-C-0052. Approved for Public Release, Distribution Unlimited.

Textile assembly methods offer great possibilities to create complex, large-scale, multi-functional 2D materials (fabrics) by a continuous process of structuring yarns together, in an architected manner. By designing a specific pattern and using functionalized yarns the properties of such a fabric can enable a variety of roles for example actuation and mechanical stimuli. Moreover, actuation can be achieved in several directions as the textile assembly enables the construction of a network where yarns can be independently addressed in X and/or Y direction. These are advantages that can be utilized in the field of soft robotics in many ways. The requirements for human-robotic interactions call for soft and compliant materials that are safe for such collaborative interactions and involve several types of functionalities. Textiles are easily conformed to the body, whether that is a robotic or a human one. Here we report on the integration of novel functional actuating yarns in the purpose of creating pliable textile actuators that also exhibit versatile morphing capabilities. The yarns consist of three layers; two of which are made of thin poly (3, 4-ethylenedioxythiophene): poly (styrene sulfonate) (PEDOT: PSS) coatings that cover opposite sides of the third layer, an ionogel. This stretchable gel supplies the system with ions for the actuation mechanism and therefore enables in-air actuation. The yarns are transformed into fabrics by using woven assembly techniques. This is an additive method that structures one set of yarns in a parallel sequence that is perpendicular to another second set of yarns. By structuring a number of yarns together in parallel the performance in terms of force output including blocking force is shown to increase. The textile assembly process allows for two approaches, collective and individual addressing for the actuating yarns. For the former, arranging the yarns into different pre-determined segments enable collective actuation of each segment to change the overall shape of the textile structure. In regards to the latter, by individual addressing we show that a specific and targeted actuation can be achieved. Furthermore, the arrangement in which the yarns are interlaced in the fabric enables switching the modality of the actuation. This means that we can alter a motion specific to the yarns into another by their arrangement in the textile structure. With our developed textile assembly method, we are approaching low-cost, large-scale production of actuating systems for human-robotic applications.

The synthesis and mechanical properties of photoswitchable silsesquioxane/azobenzene hybrid 3D–polymers (“dynamic sponges”) are presented and discussed. The hybrid is capable of extensive macroscopic movement, and overcomes previously problematic crosslink locking issues. A hydride–functionalized Q–type silsesquioxane (Q₈M₈^H) was reacted with di-allyloxyazobenzene using hydrosilylation methods. The properties of the resulting materials are controlled via careful choice of starting material ratios and solvent, leading to gels or films. Both morphologies show pronounced photoresponsive behavior in and on the surfaces of different solvents. Photoactuation is tracked by microscopy, DMA and UV/Vis spectroscopy. The gel system has a porous structure similar to a sponge. It undergoes shrinkage in volume by 18.3% in toluene under UV irradiation, and shows excellent recovery to the swollen state after irradiation with visible light. These novel photodynamic materials offer reversible modulus switching from 160 kPa in the swollen state to 500 kPa in the “wrung–out” sponge. The sponges can engage in uptake and release of a range of substances (i.e., reversible hydrophobic sponging), with overall performance determined by substrate size and polarity. Such behavior gives these materials high potential for soft robotics applications and great promise as reusable environmental remediators.

Current structural materials for robotics generally possess mechanical properties that are static and can operate only within a narrow range of environmental conditions. In an effort to imitate living biological tissues that can modulate their properties dynamically in response to an environmental stimulus, we developed an electrochemical cell with a soluble porous anode (ECSPA). This cell allows material to be added to or removed from the structural electrode according to the duration, polarity and magnitude of the applied potential difference, bringing about precise and reversible modulation in the material’s modulus, strength and energy absorption characteristics over 3 orders of magnitude. In response to a complete fracture, the cell was shown to be able to self-heal, achieving ~ 50% of its original load-bearing capability with as little as 6 hours of treatment. A self-contained ECSPA setup was also demonstrated as a proof-of-concept for how the cell can possibly be integrated into a robotic assembly for real-world applications.

Fracture toughness and self-healing capabilities in synthetic soft materials are attractive features to prolong the lifetime of soft robotics and enable their usability in extreme or harsh environment. To achieve these two attributes, soft materials such as elastomers are often endowed with dynamic supramolecular bonds such as hydrogen bonds. In this work, we describe two toughening modes that arise from tuning the polymeric chain mobilities of carboxyl functionalized polyurethane (CPU) using plasticizers. Without plasticizers, intrinsic toughening was found to dominate as a result of strong hydrogen bonds that resist crack propagation. Under such conditions, CPU achieves a fracture toughness of (~105 kJ m^-2) at which a sharp crack tends to propagate when stretched. In contrast, upon the addition of plasticizers, extrinsic toughening mechanism dominates with energy dissipation from the breakage of weaker hydrogen bond interactions. Furthermore, large blunted cracks propagate within plasticized CPU with larger stretchability. By tuning the content of plasticizers within CPU, both toughening mechanisms can be combined, enhancing the fracture toughness of CPU (~122 kJ m^-2). To prove this hypothesis, we perform in-situ studies where notched CPU was stretched at various temperatures to modulate the chain mobilities. Also, at higher plasticizer content, CPU was found to display higher self-healing capabilities, retaining greater mechanical toughness. Overall, this work provides a fundamental mechanical understanding to tune the fracture toughness and self-healing of elastomers, making them attractive for soft robotic applications.

We report a framework for creating tough, 3D printable silicone double networks (SilDNs) by combining thiol-ene and condensation cured polydimethylsiloxanes. These two chemical reactions are "orthogonal"—each proceeds without hindering the other. Thus, photocurable thiol-ene network forms the shape during 3D printing and entraps the precursors to our latent condensation network which dominates the final mechanical properties. Thus, we can break the endemic process-structure-property trade-off in 3D printed polymers and create a family of soft (100kPa<E<1MPa), highly extensible (ε > 100%) elastomers. The slower condensation reaction can also form interfacial bonds with diverse materials including ceramics, metals, and other polymers. With this strategy, we can create mechanical gradients spanning almost seven orders of magnitude, a similar range found in natural systems. We demonstrate the utility of this system by building soft robotic devices cohesively affixed to electronic circuitry and commercial garments. Further, we use machine learning to generate mechanical metamaterial structures based on the assembled structures of 3D printed rigid bodies with SilDNs. Lastly, we demonstrate a ternary chemistry where this framework can be extended to create multimaterial foamed elastomers.

Successful designs of soft robots rely on both clever morphology and material properties. Many researchers have pursued the intelligent embodiment of smart materials into the soft systems, which results in the dynamic interaction of architecture, material, and environment. However, body designs of soft robots have lagged behind biological analogs because of the inherent complexities in both form and function in generating suitable environmental behaviors at desired scales.
Structural approaches are essential to increasing the systematic complexity and the functional diversity of soft material–based intelligent systems. Smart structures can represent unconventional but programmable mechanical properties, reacting to environmental changes in morphologically and functionally adaptive ways. These features distinguish smart structures from typical static structures with a primary purpose of providing load capacity. However, in most smart structures, most loads are focused on the flexible joints or hinges, and thus, combining multiple materials has limited synergistic effects in programming system-level mechanics to include both morphology and structural mechanics efficiently. Therefore, to increase programmable complexity and synergistic integration in 3D, more scalable and systematic approaches should be promoted because of combinatorial issues brought about by embedding multiple distinct materials in a seamless and synergistic way.
We propose the adoption of tensegrity as a class of metamaterial strategy for smart material-based robotic systems. Tensegrity systems are composed of both isolated compressive “struts” and a network of elastic “tendons” with a specific configuration of nodes. The main advantages of tensegrity structures include high stiffness-to-mass ratio, controllability, reliability, structural flexibility, and large deployment. Furthermore, by preserving the self-equilibrium of normal forces applied to its elements, tensegrities are easily stabilized as the structure stands or is deformed. This harmonious balancing of forces enables distinctive material components to be formed in a network, instead of piling up simple cellular units, and thus provides ample design space of morphology and mechanical properties. Despite its advantages, smart materials are rarely used as mechanical elements for the construction of smart tensegrity structures because of a lack of proper manufacturing processes allowing generation of multi-material parts with intricate 3D shapes.
In this work, we adopted the tensegrity for the development of soft structures with programmable mechanical responses in 3D. We endowed tensegrity with additional functionality by using magnetic materials as tendon components and used a 3D printing technology combined with sacrificial molding to fabricate tensegrities at a diverse scale. This method makes the construction of tensegrity a lot easier because it eliminates any post-assembly process of beam elements.
As a result of printing tensegrity with coordinated soft and stiff elements, we could use design parameters (such as geometry, topology, density, coordination number, and complexity) to program structure-level mechanics in a soft structure. On the basis of the programmed mechanics of tensegrity structures, we developed diverse smart structures and demonstrated a tensegrity robot capable of walking in any direction. We demonstrated several tensegrity actuators (such as auxetic behavior, locomotion, and intaking) by leveraging smart tendons with magnetic functionality. This physical realization of complex 3D metamaterials with multiple mechanical components can pave the way toward more analytical and algorithmic designs of scalable geometry and contribute to complex morphing for 3D soft systems. Furthermore, this may provide new form factors for 3D flexible devices in the fields of flexible electronics, biomedicine, and soft robotics.

Soft and stretchable integrated electronic systems have gained wide popularity recently due to their superior mechanical compliance and conformability, which distinguishes them from conventional rigid electronic devices. Cutting-edge technologies of stretchable, skin-mountable, and wearable electronics are able to effectively accommodate large strain when integrated onto soft, elastic and curved surfaces. Ultralow modulus and high stretchability of electronic systems are achieved by designing new structural layouts and developing novel materials. However, stretchable conductors made of metallic materials often suffer from cycling induced fatigue, which cannot be easily improved by new structural designs.
Recently, various wearable strain sensors with high stretchability have been developed for their broad applications in human motion detection, health care, human-machine interfaces. Strain sensors can be classified into two types: resistive-type and capacitive-type sensors. Resistive-type sensors are typically composed of composites combining electrically conductive sensing films with flexible substrates. When stretched, the microstructure of sensing films evolves, and the electrical resistance changes accordingly. On the other hand, a capacitive-type sensor consists of a highly compliant dielectric layer between a pair of stretchable electrodes. When stretched, two electrodes become closer, and the capacitance increases. Alternatively, stretchable and wearable strain sensors can also be classified according to their materials, into categories such as fiber, liquid metal (LM), and piezoelectricity based strain sensors. As a liquid-state conductor, the LM features excellent electrical conductivity, fatigue-free characteristics and extremely high deformability, thus is an ideal component for wearable strain sensors.
Moreover, materials that can heal/reheal like natural skins have also been developed in wearable electronics. Self-healing/rehealing can help wearable electronics to gain benefits of reliability, durability, cost and performance. While some self-healing mechanisms can automatically respond to damages without stimuli, most self-healing/rehealing materials still require moderate external stimulation—such as heat, light, water and chemicals—to trigger the healing process. Various strategies have also been investigated to achieve self-healability/rehealibility, including metal-ligand supramolecular, microvascular agents, hydrogen bonds, semiconducting polymers, and dynamic covalent bonds. Among the many self-healing mechanisms, dynamic covalent bonding in polymer networks is usually stronger than supramolecular interactions, thus making such materials more robust and can operate under a wider range of conditions⁴⁴. More recently, a series of dynamic covalent thermoset polyimine that can self-heal/reheal under modest external stimuli have been developed.
Here, we present a new type of flexible, highly stretchable, and rehealable strain sensing system enabled by eutectic LM alloy and dynamic covalent thermoset polyimine. As LM is a liquid, it doesn’t add rigidity and provides excellent deformability to the strain sensing system. Moreover, unlike conventional metal conductors, LM conductors doesn’t experience fatigue. Furthermore, a dynamic covalent thermoset polyimine matrix is not only highly stretchable, but also rehealable from damages. To provide prediction to the strain sensing system, we have also established an analytical model to describe the resistance change under applied strain, which shows good agreement with finite element simulations and experimental measurements.

Hygroresponsive soft materials can convert environmental humidity directly into mechanical motions, as can be mundanely observed in curling of wet paper, swelling of kitchen sponges, and opening of dry pine cones. Although the bending and coiling motions of such moisture-sensitive materials are recently explored for powering tiny robots, a lack of mechanistic and thermodynamic understanding of the soft actuation systems has hindered optimization of their mechanical designs and evaluation of energy conversion efficiency. Here, we construct a theoretical model to predict the temporal evolution of actuator shape and resulting force under environmental humidity change. We demonstrate how our mechanistic theory optimizes designs of various hygroresponsive machines developed in this work, including a crawler, a wheel, a seesaw, and a vehicle. Furthermore, we introduce a stress-volume diagram of soft power generators to calculate the thermodynamic efficiency of natural and artificial hygroresponsive engines.

Shape-changing hydrogels that can bend, twist, or actuate in response are critical elements of soft robots. Chemomechanical devices, which are controlled by chemical signals, have critical advantages for miniaturizing soft robots. The chemicals that control these devices can diffuse over large distances and into small or tortuous spaces, and the huge number of chemicals that can be synthesized offers unprecedented tunability and specificity. Chemomechanical devices require no batteries and can easily be miniaturized and integrated with other devices. We demonstrate hydrogels differentially responsive to different DNA signals. Specific DNA molecules can induce 100-fold volumetric expansion of hydrogels by successive extension of cross-links via the operation of polymerization motors. These structures can be photopatterned at size scales up to centimeters to create composite gels containing multiple domains. The resulting devices undergo different shape changes in response to different DNA sequences that selectively swell different domains. A simple design rule derived from experiments and material simulations suggests a means to control shape change. Methods of rational DNA design and control of DNA stoichiometry can, by altering polymerization motor operation, alter the amount and rate of swelling. Finally, multicomponent hydrogels using a variety of scaffold polymers, including polyethylene-glycol, acrylamide can be photopatterned into distinct shapes, have a broad range of mechanical properties, including tunable shear moduli across more than an order of magnitude, and enhanced biocompatibility. These materials offer a new possible means of building soft robots: because DNA molecules can be outputs and inputs to molecular sensors, amplifiers, and logic circuits, this strategy introduces the possibility of building soft devices autonomously controlled by chemical networks.

Biohybrid devices can be categorized into 4 groups: (i) biohybrid-sensors that can detect target molecules at highly selective and sensitive manner, (ii) biohybrid-reactors that mimic biological reaction in our body, and thus are useful for drug testing or tissue transplant for cell therapy, (iii) biohybrid-actuators that shows highly energy efficient motion and (iv) biohybrid-processors that achieve low-energy and highly parallel computing like our brain. In this talk, I would like to talk about a couple of our recent results regarding biohybrid sensors and actuators.

The mechanoreceptors buried in human skin is known for enabling the skin to be tactile sensitive. Meanwhile, the innervations of the mechanoreceptors, which function via three-dimensionally distributed nerves extending from deep skin upwards, also enable the skin to detect complex tactile stimuli like normal and shear forces. Here, we design a biomimetic e-skin sensor by embedding three-dimensional metal wire electrodes as ‘nerves’ in a low-modulus yet elastic self-healing foam named Artificially Innervated Foam (AiFoam). The three-dimensional electrodes in the sensing material enable the sensor to detect both normal and shear forces, as opposed to the conventional sensors with two-dimensional planar electrodes which can only detect normal forces. We also develop a unique bimodal piezo-impedance pressure sensing foam material with both piezoresistive and piezocapacitive sensing capabilities. The near-percolation metal particle foam composite can be obtained through a new one-step self-foaming process. The elastic foam material has a low modulus of 600 kPa and self-heals through strong dipole-dipole interactions and hydrogen bonding. In addition to the detection of tactile signals, the AiFoam e-skin can also function beyond the human skin by perceiving proximity.[1]
1. Guo H., Tan Y. J., et al. Artificially Innervated Self-healing Foams as Synthetic Piezo-Impedance Sensor Skins. Nat Commun. (in press).

Soft devices (including sensors, electronics, and machines) have attracted huge interest because they cannot only extend the scope of smart systems but also provide compliant and safer user experience. Particularly, power and electronic components that are self–healable, deformable, transparent, and even self–powered are highly desired for future robotic applications. Here, we present the first triboelectric–based energy–harvesting and self–powered robotic skin that is entirely, intrinsically, and autonomously self–healable and simultaneously highly transparent and extraordinarily stretchable. Not only can this energy–harvesting robotic skin serve as an untethered and robust power source for personal electronics, but it can also be used as elegant robotic sensing skin that combine all desired attributes including self–healing, self–powered, highly transparent, and super–stretchable. This is the first time that not only the structure of an slef-powered robitc skin is entirely and intrinsically self-healable at room conditions but also the device can be operated via self-generating electricity. Additionally, the self-powered robotic skin possesses a fast healing time (30 min, 100% efficiency at 900% strain), high transparency (88.6%), and extraordinary inherent stretchability (>900%). Even after 500 cutting-and-healing cycles or under extreme 900 %-strain, the energy-harvesting robotic skin can retain its functionality. The generated electricity form the robotic skin can be used directly or stored to power commercial electronics. Further, the robotic skin is designed for self-powered tactile-sensing matrix in diverse human−machine interfaces including smart glass, an epidermal controller, and cell phone panel. The unprecedented triboelectric-based robotic skin that is entirely and inherently ambient self-healable, highly transparent, and intrinsically stretchable, and possesses energy−harvesting and actively–sensing ability, can meet wide application needs ranging from deformable/portable/transparent electronics, smart interfaces, energy devices, artificial skins, to soft robotics.

[Ref]
[1] Paper cover: https://onlinelibrary.wiley.com/doi/10.1002/adfm.201970273
[2] Ying-Chih Lai,‡* Hsing-Mei Wu,‡ Heng-Chuan Lin, Chih-Li Chang, Ho-Hsiu Chou,* Yung-Chi Hsiao, and Yen-Cheng Wu, Entirely, Intrinsically, and Autonomously Self–Healable, Highly Transparent, and Superstretchable Triboelectric Nanogenerator for Personal Power Sources and Self–Powered Electronic Skins, Advanced Functional Materials, 2019, 190426. (Cover of the journal. VIP paper. This article was also selected in: Hot Topic: Flexible Electronics)
[News]
https://www.youtube.com/watch?v=Gzt4PKGItcY
https://www.youtube.com/watch?v=g1huAexxiY4
[Demo video]
https://www.youtube.com/watch?v=eYKW63GFheU

Molecular and supramolecular level engineering of soft matter that exhibits autonomous or externally guided locomotion and dynamic morphogenesis is a grand challenge for materials science. If this happens at macroscopic scales it would be a pathway to create robotic objects from soft materials; at the microscopic scales we could begin to emulate cell behaviors with chemical defined materials; and distantly organelles as we approach the nanoscale. In this lecture we report on bio-inspired systems synthesized in a chemistry and materials science laboratory to design bottom up macroscopic robots under water. These macroscopic robots have the capacity to respond synergistically to photons and magnetic fields and reveal locomotion under water at speeds that approach those of biological systems, namely one body length per second. We found that we could control speeds, trajectories, and gait on the fly by manipulating light intensity and magnetic fields, and most interestingly it is possible to predict theoretically the response of the robots to design their behaviors. The lecture will also present results on light responsive materials that exhibit origami-like morphogenesis using constructs with macromolecules in which it is possible to design their response to photons.

Current approaches to modify the material composition, and in this way the mechanical behavior, of physically cross-linked polymeric actuator materials are reliant on synthetic alteration of the polymer architecture.¹
Here, we introduce a reprocessable shape-memory polymer actuator consisting of a multiblock copolymer with poly(ε-caprolactone) and poly(L-lactide) (PLLA-PCL) segments, which is blended with a poly(D lactide) (PDLA) oligomeric additive. By careful molecular design, the incorporation of structural units capable of stereocomplexation enabled the formation of the stable physical network necessary for actuation. By varying the relative concentration of the two blend components, a material with tunable reversible bidirectional actuation performance and reprocessability was created.² Moreover, through the study of the strain recovery and stress relaxation the macroscopic deformation patterns of the investigated materials were attributed to the microscopic deformation mechanisms of PCL crystallites and PLA stereocomplexes. The strain and composition ranges could be defined, in which PLLA-PCL / PDLA blends having PLA stereocomplexes as physical netpoints maintain elastic behavior typically attributed to polymer networks. The polymer blend system is of potential relevance as thermally controlled artificial muscles in soft robotics.³

References:
A. Lendlein, M. Balk, N. Tarazona, O.E.C. Gould, Biomacromolecules, 20(19), 3627-3640 (2019)
V. Izraylit, O. E. C. Gould, T. Rudolph, K. Kratz, A. Lendlein, Biomacromolecules, 21(2), 338-348 (2020)
V. Izraylit, M. Heuchel, O.E.C. Gould, A. Lendlein, Polymer 209, 122984 (2020)

The proteins actin and myosin are well known for their roles as biological actuators, not only in muscle tissue, but also in the cytoskeleton of various non-muscle cells. When purified in a cell-free environment, these two proteins form the basis of an active gel, a material that is capable of undergoing deformation and exerting force. So far researchers have studied these gels to better understand actin-myosin contraction in living cells. Could these active gels find their way someday in robotics applications as actuators? In this talk, I will review the mechanical, force-generating, and rupture properties of active gels of actin and myosin. I will then explore the pros and cons of these materials as robotic actuators.

Recent research efforts have focused on developing soft, flexible, compliant materials for robotics, biointerfacing, and biosensing applications. Because of their intrinsic softness, these materials are susceptible to cut, puncture, scratch, and/or tear damage that compromise their physical integrity, and therefore self-healing properties are indispensable for soft machines and devices operating in dynamic environments. However, current self-healing materials have shortcomings that limit their practical application, such as low healing strength (below MPa) and long healing times (hours).
Here, we introduce high-strength, biodegradable protein-based materials that self-heal micro- and macro-scale mechanical damage within a second via reversible physical cross-linking. These proteins are systematically optimized to improve their hydrogen-bonded nanostructure and network morphology, with healing properties (~25 MPa strength after 1 second of healing) surpassing those found in other natural and synthetic soft materials by several orders of magnitude. We demonstrate soft gripper and artificial muscle prototypes integrating such biological materials. Such healing performance opens new opportunities for bioinspired materials design, and addresses current limitations in self-healing materials for soft robotics and wearable technology.

Soft robots, advantageous in their compliance and flexibility, have advanced smart soft materials towards multifunctionality with local sensing, control and powering capabilities that approach human dexterity. Mimicking biological neuromuscular systems’ sensory motion requires the unification of sensing and actuation in a singular artificial-muscle material, which must not only actuate but also sense their own motions and even has built-in feedback loop for basic-level control functions. Stimuli-responsive hydrogels are a class of synthetic materials that can change their volume upon environmental cues (temperature, light, and chemicals) to act as soft actuators and chemical sensors. They can also provide flexible porous scaffolds to load conductive components for constructing soft strain sensors, supercapacitors and other soft electronics. Their biocompatibility, easy functionalization, solution-based processing, and 3D printability are their additional merits for soft robotics. These together may lead to all-soft robots with higher flexibility, autonomy, and performance. Towards this goal, we have developed a series of material systems based on hydrogels and conducting polymers, including (i) beetle-inspired ultrafast colorimetric sensing of chemical and biological species (Adv. Mater. 2018; Adv. Opt. Mater. 2019), (ii) muscle-inspired high contractile tough actuators (Sci. Adv. 2020, ACS Appl. Mater. Inter. 2020), (iii) plant-mimetic adaptive light tracking and harvesting (Nat. Nanotech. 2019), (iv) phototaxic soft swimming robot (Sci. Robotics 2019; Sci. Adv. 2020), (v) high-performance soft strain sensors and supercapacitors (Adv. Funct. Mater. 2020, Matter 2020), and (vi) a soft somatosensitive actuating material utilizing an electrically-conductive and photothermally-responsive hydrogel, which combines the functions of piezoresistive strain/pressure sensing and photo/thermal actuation into a single material. Overall, these intelligent material systems represent a step towards next-generation soft robots with life-like adaptiveness and higher-level autonomy.

Despite the promise of operation in extreme environments and the use of high energy density power sources, strategies for thermoregulation remain underdeveloped in soft robots. Here, we present autonomically perspiring fluidic actuators based on multimaterial 3D printing of smart gels. We developed two custom hydrogel photochemistries—one based on poly-N-isopropylacrylamide (PNIPAm) and the other based on polyacrylamide (PAAm)—with opposing thermo-mechanical responses. At elevated temperatures, PAAm expands, and NIPAm contracts. Our actuator designs contain a porous dorsal layer of PAAm micropores that cap the fluidic channel embedded in the PNIPAm body. As the temperature rises above ~37°C, the PNIPAm body collapses to expel water through the dilated PAAm pores. This material response is reversible and occurs without any additional control from the robot to enable localized sweating. We then create the experimental protocols to measure the thermoregulatory performance of soft robots and further develop mathematical models using Newton's Law of Cooling to quantify this behavior and compare it to that of their non-sweating counterparts under a variety of conditions. Our sweating actuator exhibit a 600% enhancement in cooling rate with >100 W●kg^-1 of additional cooling capacity, comparable to the best animal systems. We further combine multiple actuators into a hand and demonstrate the ability to grasp and thermally manipulate objects in the environment.

(1) Mishra, A.K., Wallin, T.J., Pan, W., Xu, P., Wang, K., Giannelis, E.P., Mazzolai, B. and Shepherd, R.F., 2020. Autonomic perspiration in 3D-printed hydrogel actuators. Science Robotics, 5(38).

(2) Mishra, A.K, Pan, W., Giannelis, E.P. and Shepherd, R.F., Wallin, T.J., 2020. Making Bioinspired 3D Printed Autonomic Perspiring Hydrogel Actuators, Nature Protocols

We report on a novel kind of material robotics micro-swimmers, ones that could be assembled bottom-up. They are able to harvest energy from their environment to move, and recharge on demand. We quantify the movement and reveal their propulsion mechanism. The robots form a new type of active matter, that is sustainable, as they could be grown from single molecules and almost no energy or materials are wasted. In this invited talk I will report on the large collaborative effort that resulted in this discovery and directions in which new fields of study can develop.

Hydrogels with the ability to change shape in response to biochemical stimuli are important for biosensing, smart medicine, drug delivery, and soft robotics. Here, we create a new family of multicomponent DNA polymerization motor gels with different polymer backbones, including acrylamide-co-bis-acrylamide (Am-BIS); poly(ethylene glycol) diacrylate (PEGDA); and gelatin-methacryloyl (GelMA) that swell extensively in response to specific DNA sequences. A common actuation mechanism, a polymerization motor that induces swelling is driven by a cascade of DNA hairpin insertions into hydrogel crosslinks. These multicomponent hydrogels can be photopatterned into distinct shapes and structures, have a broad range of mechanical properties, tunable moduli, and enhanced biocompatibility. Human cells adhere to the GelMA-DNA gels and remain viable during approximately 70% volumetric swelling of the gel scaffold induced by DNA sequences. The results demonstrate the generality of sequential DNA hairpin insertion as a mechanism for inducing shape change in multicomponent hydrogels, suggesting new ways of building soft robotic devices and widespread applicability in biomaterials science and engineering.

Nature continually provides a riche and valuable source of inspiration for many fields of research. Looking to the natural development of life, researchers have been consistently inspired designing new functional and adaptive materials. Programmed and self-assembled deformations are widespread in nature, providing elegant paradigms to design self-morphing materials, responsive materials or engineered robots. In fact, it would be very powerful in different field of technology and, in particular for indwelling or external biomedical devices, to have elements capable of responding to external stimuli in a simple way. Within this context, there is a lot of interest in creating artificial structures that can walk, swim, move and perform various tasks answering to an external stimulus. Conventional robots are usually entirely composed of hard materials and require complex control systems to accomplish different tasks. Their fabrication process is long-lasting and complicated, inhibiting downscaling and constraining the device to a single predefined behavior. Recently, an emerging classes of soft robotics with flexible actuation, intelligent sensibility, and biomimetic functionality are driving advances in academic researches and commercial applications through the development in soft matter engineering and flexible actuation systems. One of the most challenging issues in developing enhanced robotic technology is making appropriate tools also for the full control of liquids. Liquid robots still represent an interesting outcome for the technology form micro to nano scale. In fact, liquid-manipulating systems extend to many fields with important industrial applications such as biomedicine, biotechnologies, food, chemistry, industry and cosmetics, to cite few. To build these systems, liquids, polymers, and soft matter require accurate, precise, and fully controllable methods to be handled and processed. So far, several engineering methods have been proposed to control and drive liquids in response to external magnetic, electric, or optical fields. However, until now, all the approaches proposed for locomotion and remote liquid control are complicated and very expensive. Here we introduce a new and simple working mechanism for actuation, liquid manipulation and a complete exploration of the opportunities of a multipurpose platform guided by physical intelligence. Physical intelligence is a new way of furnishing intelligent responses (outputs) as a function of environmental changes (inputs); in particular, this disruptive technology proposed is based on the use of a pyroelectric material (Lithium Niobate) that can provide an electric field (output) as a consequence of temperature changes (input). The electric field generated by the pyroelectric effect allows one to manage the on/off function of the system in order to command a fast response in a smart way, and to easily move units of liquid with different volumes. The pyroelectric platform allows to manage and displace liquid unit volumes from a starting position to the desired endpoint, controlling with good resolution any intermediate step until the final position. Beyond the guiding property, we prove additional functionalities like merging, stretching, mixing and jumping of liquid volumes and millimeter objects using a working distance of millimeter, bigger than the conventional distances used for liquid handling in classical digital microfluidic. Multiple volumes can be dragged over the surface, moved simultaneously and mixed in case of need. The proposed technology for locomotion and tweezing of liquids and particles could open a new route for soft robotics, biomedicine, material science, fluid dynamic and also for application in microgravity environment, where it is well known that the managing of liquid is difficult and have safety concerns.

Nasti, G., Coppola, S. et al. (2020). Pyroelectric tweezers for handling liquid unit volumes. Advanced Intelligent Systems, 2000044.