Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Nov 7.
Published in final edited form as: J Mater Chem B. 2019 Jul 10;7(41):6293–6309. doi: 10.1039/c9tb01006h

Mechanochemical Engineering of 2D Materials for Multiscale Biointerfaces

Catherine E Machnicki a,b, Fanfan Fu c, Lin Jing c, Po-Yen Chen c,*, Ian Y Wong a,*
PMCID: PMC6812607  NIHMSID: NIHMS1041101  PMID: 31460549

Abstract

Atomically thin nanomaterials represent a unique paradigm for interfacing with biological systems due to their mechanical flexibility, exceptional interfacial area, and ease of chemical functionalization. In particular, these two-dimensional (2D) materials are able to bend, curve, and fold in response to biologically-generated forces or other external stimuli. Such origami-like folding of 2D materials into wrinkled or crumpled topographies allows them to withstand large deformations by accordion-like unfolding, with implications for stretchable and shape-changing devices. Here, we review how mechanically manipulated 2D materials can interact with biological systems across a multitude of length scales. We focus on recent work where wrinkling, crumpling, or bending of 2D materials permits new chemical and material properties, with four case studies: (i) programming biomolecular reactivity and enhanced sensing, (ii) directed adhesion and encapsulation of bacteria or mammalian cells, (iii) stimuli-responsive actuators and soft robotics, and (iv) stretchable barrier technologies and wearable human-scale sensors. Finally, we consider future directions for manufacturing, materials and systems integration, as well as biocompatibility. Taken together, these 2D materials may enable new avenues for ultrasensitive molecular detection, biomaterial scaffolds, soft machines, and wearable technologies.

Introduction

Interfacial interactions between artificial materials and biomolecules, cells, or tissues are critical for the functional performance of biomedical technologies.1 In particular, atomically-thin two-dimensional (2D) materials such as graphene and graphene oxide (GO) exhibit exceptional chemical, mechanical, electrical, photothermal, and optoelectronic properties that could potentially be harnessed for new biomaterials.2 Such nanomaterials can be patterned with lateral dimensions ranging from nanometers up to tens of centimeters, while their thickness can be tuned from a single monolayer to a multilayer stack comprised of hundreds and thousands of layers.3 As a consequence of their very high aspect ratios, 2D materials present extraordinary interfacial areas for interactions with biological entities. Indeed, the chemical groups presented on the basal surfaces and edges of these 2D materials serve as versatile “handles” for chemical functionalization and facile platform for nanomaterial assembly.4

Although 2D materials are extremely rigid when stretched in-plane, they are very deformable out-of-plane.5 Thus, 2D materials are easily bent or folded by relatively gentle mechanical stresses.6 This anisotropic mechanical behavior has remarkable consequences when 2D materials are deposited onto a compliant, pre-stretched substrate.7 When released, the mismatched mechanics of the stiff coating and soft substrate results in topographic patterns such as periodic wrinkles or disordered crumples.8 The 2D material is now spatially inhomogeneous, with highly localized regions of strain that distort the atomic lattice for altered electronic structure and chemical reactivity.9 This accordion-like topography can be unfolded in tension, allowing for the 2D material to be effectively stretched without fracture.10 Although early pioneering work utilized externally applied forces to manipulate 2D materials, there has been increased interest in “smart” micro/nano structures that self-actuate or fold in response to environmental stimuli.11 These dynamic behaviors can be programmed from the nanoscale up to the macroscale, enabling a wide range of biological and medical applications.

In this review, we focus on mechanical patterning and manipulation of 2D materials for interfacing with biological systems at varying length scales (Figure 1). First, we consider molecular length scales, where sharp curvatures can enhance chemical reactivity and optical sensing modalities. Second, we consider cellular length scales, where cell adhesion can be directed by 2D material topographies or cells can be encapsulated within 2D materials. Third, we consider centimeter length scales, where 2D materials can serve as stimuli-responsive actuators and derive new backbone materials for soft robotic applications. Fourth, we consider human length scales, where 2D materials can function as biotic-abiotic interfaces with enhanced skin-like multifunctionality. Finally, we address future directions for the field, including prototyping and manufacturing challenges across scales, compatibility with other material types, system integration, and biocompatibility.

Figure 1.

Figure 1.

A. Engineering of 2D materials for multiscale biointerfaces ranging from molecule (nanoscale), cell (microscale), actuator (macroscale) to wearable devices (human-scale). Adapted with permission from1214. Copyright 2018 Royal Chemical Society, John Wiley & Sons. B. Crumpled graphene oxide – gold nanoparticle for chemical sensing. Adapted with permission from15. Copyright 2015 American Chemical Society. C. Wrinkled graphene-based surface topographies for directing cell alignment and morphology. Adapted with permission from16. Copyright 2016 Elsevier. D. Wrinkle-crumple rGO pressure sensors applied in smart collision-aware surgical robots. Adapted with permission from17. Copyright 2019 American Chemical Society.

A Primer on the Chemistry and Mechanics of 2D Materials

The interfacial chemistry and mechanical flexibility of 2D materials enable unique interactions with biological systems, and some 2D materials of interest will be briefly highlighted here. In particular, we consider graphene, graphene oxide (GO), Xenes, transition metal dichalcogenides, MXenes, and hexagonal boron nitride (h-BN) (Table 1). For more comprehensive treatments of 2D materials chemistry, synthesis, and applications, we refer the reader to several excellent reviews.2, 4, 1821

Table 1.

Comparison of chemical structure, class, lattice structure, notable physical properties, and applications of various 2D materials.

Chemical structure 2D Class Lattice Structure Comments Applications
Graphene graphic file with name nihms-1041101-t0001.jpg Graphene Hexagonally arranged sp2 bonded carbons. High electric and thermal conductivity with a Young’s modulus of 1 TPa. Electronics, chemical barriers, fuel cells, catalysts, and sensors.
Graphene Oxide graphic file with name nihms-1041101-t0002.jpg Graphene Hexagonally arranged sp2 bonded carbons with regions of sp3 carbons where the basal (epoxides) and edge (carboxyl and hydroxyl) oxygen functional groups are attached. Hydrophilic, synthesized through the chemical oxidation and mechanical exfoliation of graphite. Wearable electronics, chemical barriers, catalysts, sensors, and biomedical devices.
Phosphorene graphic file with name nihms-1041101-t0003.jpg Xene
Groups IIIA - VA
Hexagonally arranged lattice (silicene, germanene, phosphorene, borophene, stanene, and antimone).22  Orthogonal with zigzag conformation out-of-plane. A semiconductor with anisotropic properties (electric, thermal, mechanical) due to the zigzag conformation. While most of these materials are synthetically grown, phosphorene is prepared from black phosphorus through mechanical or liquid exfoliation, similar to graphene. Batteries, transistors, optoelectronics, and sensors. Exceptional near-infrared adsorption.23,24
Molybdenum disulfide (MoS2)graphic file with name nihms-1041101-t0004.jpg Transition Metal Dichalcogenides (TMDs)
MX2;
M = transition metal, X = chalogen (S,Se, or Te)
Transition metal coordination of the chalcogen atoms by either octahedral or trigonal prismatic geometry, after exfoliation. Tunable electronic properties and semiconducting states through mechanical strain. MoS2 exhibits similar functional properties as GO, including large surface areas, strong near-infrared adsorption, as well as ease of chemical functionality, but can be degraded or cleared more rapidly in living systems.25 Energy storage, biomaterials, optoelectronics, electrochemical catalysis, and piezoelectrics.
Titanium carbide (Ti3C2Tx)graphic file with name nihms-1041101-t0005.jpg MXene
Mn+1XnTx;
M = transition metal, X = carbide or nitride, Tx = termination functional group (-F, -O, and -OH)
AB or ABC ordering of titanium atoms with carbon atoms in layers with terminal functional group either -F, -O, or -OH. High electric conductivity, metallic and hydrophilic. MXenes are synthesized by selective etching of A elements from the MAX phase using strong acids (normally contain HF) and exfoliation. Energy storage, hydrogen storage, and catalysis.
Hexagonal Boron Nitride graphic file with name nihms-1041101-t0006.jpg Boron Nitride Hexagonally arranged alternating boron and nitrogen atoms. 2D h-BN flakes can be controllably synthesized by mechanical/chemical exfoliation and chemical vapor deposition methods.26 Chemically inert, good thermal stability,27 mechanical property, outstanding thermal conductivity28 and biocompatibility,29 Nevertheless, it should be noted that h-BN may be more challenging to chemically functionalize, due to the inorganic boron and nitrogen surface groups.30 Insulating charge leak barrier in electronics and biomedical devices.

For clarity, we review some terminology for 2D materials based on length scale and chemistry, consistent with the proposed nomenclature defined elsewhere (Table 2).31 These definitions were originally proposed to graphene but can be generalized for other 2D materials.

Table 2.

Recommended nomenclature for 2D carbon materials, reproduced from31

Term Definition
Graphene A single-atom-thick sheet of hexagonally arranged, sp2-bonded carbon atoms that is not an integral part of a carbon material, but is freely suspended or adhered on a foreign substrate.
Multi-layer graphene
(MLG)
A 2D (sheet-like) material, either as a free-standing flake or substrate-bound coating, consisting of a small number (between 2 and about 10) of well-defined, countable, stacked graphene layers of extended lateral dimension.
Graphene nanosheet A single-atom-thick sheet of hexagonally arranged, sp2-bonded carbon atoms that is not an integral part of a carbon material, but is freely suspended or adhered on a foreign substrate and has a lateral dimension less than 100 nm
Graphene oxide (GO) Chemically modified graphene prepared by oxidation and exfoliation that is accompanied by extensive oxidative modification of the basal plane. Graphene oxide is a monolayer material with a high oxygen content, typically characterized by C/O atomic ratios less than 3.0 and typically closer to 2.0.
Reduced graphene oxide (rGO) Graphene oxide (as above) that has been reductively processed by chemical, thermal, microwave, photo-chemical, photo-thermal or microbial/bacterial methods to reduce its oxygen content.

2D materials can be prepared through top-down transfer and patterning approaches,20, 32 as well as bottom-up synthesis.21 For instance, Novoselov and Geim’s pioneering studies were performed by manually transferring monolayer or few layer graphene using the “Scotch tape” method,33 which yields high quality devices but is labor intensive. Another approach of producing 2D materials is through the synthesis on a catalytically active substrate (e.g., graphene on copper) by decomposition or reaction of gas, liquid, or solid. These bottom-up approaches enable extremely high quality 2D monolayers with few defects, but are limited to wafer scale for certain substrate materials, as well as the requirements for high temperature and vacuum, which may not be compatible with biomolecules and cells. Alternatively, bulk layered crystals can be mechanically exfoliated as single or few layer sheets in a solvent. The liquid-phase exfoliation produces a colloidal dispersion of 2D material nanosheets, but with considerable polydispersity and surface defects. R. Ruoff and coworkers demonstrated that dispersed 2D material nanosheets can be restacked on a substrate by dropcasting or vacuum filtration.34

Analogous to 3D crystalline lattices, 2D materials can exhibit point defects, lattice vacancies, and line defects.35, 36 Interestingly, point defects can occur in graphene through the formation of non-hexagonal rings without adding or removing any atoms from the lattice. As an example, the Stone−Wales (SW) defect, consists of two pentagons and two heptagons, which are transformed from four hexagons by rotating one of the C−C bonds by 90° (Figure 2A). Alternatively, vacancies occur due to the removal of one atom from the lattice, resulting in one pentagon bordering an empty region (Figure 2B). Moreover, line defects are boundaries that domains of different lattice orientations, frequently between grains (Figure 2C). Similarly, each graphene layer is terminated by edges with the edge atom being either free or passivated with hydrogen atoms (Figure 2D). Finally, graphene monolayers are terminated by edges in an armchair or zigzag orientation, often passivated with hydrogen atoms. Interestingly, these defects have been shown to have an increase in surface reactivity such as phenyl group reactivity to edge defects or to Stone-Wales defects.37 These defect sites have also been shown through computational studies to be preferential buckling sites when the 2D material are under compression or thermal agitation (Figure 2E).38 These defects may become localized by out-of-plane topographical features, which can be templated by depositing 2D materials over protruding nanostructures, such as arrays of pyramids or spherical nanoparticles (Figure 2F).3941

Figure 2. Schematic illustrations of lattice defects as well as mechanical deformations.

Figure 2.

A. Stone-Wales with 2 pentagons (green) and 2 heptagons (orange). B. Vacancy with one pentagon (green) and larger empty region (orange). C. Line defect. D. Edge defect. E. Point deformation. F. Nanoparticle strain. G. Planar/un-deformed substrate (green) with 2D material coating (gray). H. Wrinkled with sinusoidal profile (small strain buckling). I. “Crinkled” with sawtooth-like profile (large strain/postbuckling).

Mechanically, 2D materials are extremely stiff when stretched in-plane, but bend or fold very easily out-of-plane.5 As a consequence, freely suspended 2D materials often exhibit intrinsic ripples due to thermal fluctuations.42 Moreover, 2D materials often exhibit wrinkles due to mismatched thermal expansion coefficients with the underlying (metallic) substrate,43 as well as after being transferred to another substrate.44 These wrinkles can compromise electronic and barrier properties, and are considered undesirable for optoelectronic devices as well as anticorrosion coatings.6

Extrinsic wrinkling in a 2D material can also occur through mismatched deformation when adhered to a softer elastomeric substrate,7 resulting in out-of-plane buckling with a gentle sinusoidal profile (Figure 2G, 2H).8 In this small deformation regime with thick substrates, the characteristic wrinkling wavelength λ is approximated as λ=2πh(E¯c/3E¯s)1/3, where h is the coating thickness, E¯c and E¯s are the plane strain elastic moduli for the coating and substrate, respectively.45 Thus, the feature spacing λ can be directly tuned by changing the coating thickness h. With larger deformations, the 2D material coating can partially detach (i.e., the postbuckling regime), resulting in sharper sawtooth-like “crinkled” features (Figure 2I).46 For instance, monolayer graphene has been grown on copper wafers by chemical vapor deposition (CVD) method and subsequently transferred to a prestretched silicone substrate.32 X. Zhao’s group and others have demonstrated that the graphene coating forms wrinkled or crumpled morphologies after the substrate is relaxed uniaxially or biaxially, respectively.47, 48 Alternatively, our group and N. Koratkar’s group have shown that multilayered GO nanocoatings can be deposited on pre-stretched substrates through dropcasting or vacuum filtration, which similarly wrinkle or crumple after the substrate is relaxed uniaxially or biaxially, respectively.49, 50 Besides using the pre-stretched elastomer substrates, pre-stressed polystyrene (PS) “shrink films” can be used to deform the coated 2D materials by undergoing large contractions when heated above the glass transition temperature (~100 °C).51 Local heating of these PS films using an infrared laser can be used to locally manipulate graphene topographies.52 We have also shown that the application of multiple mechanical deformations can result in hierarchical GO architectures with wrinkled or crumpled features at multiple length scales, which differ based on the specific sequences of deformation modes.53

Nanoscale: Chemical Functionalization and Molecular Sensing

Mechanical distortion of the graphene lattice can modulate local electronic structure and chemical reactivity. R. Ruoff and coworkers transferred CVD-grown graphene over dispersed silica nanoparticles, and showed that aryl diazonium molecules selectively functionalized at regions of high curvature bear the nanoparticles.54 Subsequently, H. Ago and coworkers investigated how these aryl diazonium molecules covalently coupled to a CVD-grown graphene monolayer transferred to an elastomeric substrate, which was stretched from 0–15% strain (Figure 3A).55 Three different aryl molecules exhibited faster functionalization kinetics and final coverage for 15% strain relative to 0% strain. In particular, the methoxy-containing aryl molecule, which typically reacts poorly with graphene due to the electron-donating nature of the methoxy group, exhibited an order of magnitude speedup in reaction kinetics and greatly increased final coverage under 15% strain. The effect of mechanical stress was elucidated by density functional theory, which predicted the formation of an extended π orbital with a localized electron available for covalent bonding. Moreover, mechanical distortion of an imperfect graphene lattice containing Stone-Wales defects could also enhance surface functionalization. Overall, this facile mechanochemical mechanism may be widely applicable to selectively pattern biomolecules such as DNA or proteins on 2D materials.56

Figure 3.

Figure 3.

A. Mechanically enhanced chemical functionalization. Adapted with permission from55. Copyright 2013 American Chemical Society. B. Strain-induced electrical polarization in sharply kinked graphene films. Adapted with permission from58. Copyright 2018 Royal Society of London. C. Single file assembly of C60 fullerenes along graphite. Adapted with permission from59. Copyright 2007 Elsevier. D. Surface enhanced Raman spectroscopy of gold nanoparticles using plasmonic enhancement on wrinkled graphene. Adapted with permission from61. Copyright 2015 American Chemical Society.

Large mechanical strain gradients at the nanoscale can also drive an electrical polarization (i.e., flexoelectricity),57 which could result in highly localized regions of electrostatic charge. These large strain gradients may be associated with sharply crinkled topographies that exhibit a sawtooth-like profile.58 Kim et al. predicted using density functional theory that this local curvature could be localized within a 0.86 nm region, reaching a maximum value of 0.15 nm−1 at 3° deflection. As a consequence, the symmetry of the sp2 π-π electron orbitals is broken to generate a sp2-sp3 hybrid state (Figure 3B). Effectively, this segregates a line of electrical charge at the peak of a crinkle, with a polarization density of 0.12 e nm−1. It was proposed that this mechanism could be responsible for observations of single-file chaining of C60 fullerenes on highly oriented pyrolytic graphene (HOPG) (Figure 3C),59 which has been corroborated by preliminary experiments on graphene using fullerenes and DNA.60 Flexoelectricity in graphene represents an intriguing way to locally tune intermolecular interactions, although exceptionally controlled lattice deformations are required.

Crumpling of gold nanoparticle-decorated graphene was also demonstrated for surface enhanced Raman spectroscopy (SERS) based on plasmonic enhancement at regions of high curvature, which couples to the active charge transfer due to π electrons in graphene (Figure 3D).61 S. Nam and coworkers first deposited a thin film of gold onto a CVD-grown graphene monolayer, which dewet at elevated temperatures to form gold nanoparticles ~30 nm diameter. This composite film was then transferred onto a prestressed PS “shrink film”, which underwent uniform biaxial compression into a crumpled topography with an average height of ~300 nm and wavelength of ~500 nm. This roughened topography improved the SERS sensitivity by over an order of magnitude relative to a flat topography, which could be attributed to both the formation of plasmonic “hotspots” near regions of sharp curvature, as well as the placement of gold nanoparticles in closer proximity.62 Nam and coworkers further demonstrated that these composite films could be transferred onto a non-planar substrates for SERS sensing. Similarly, Z. Tang and coworkers showed that these gold nanoparticle-decorated crumpled graphene films could be reversibly actuated while maintaining good sensitivity for multiple analytes.63 Thus, topographically patterned 2D material films interfaced with nanoparticles enable new sensing functionalities in stretchable, non-planar geometries. Beyond graphene, chemical sensing has also been implemented using wrinkled MoS2,64 as well as the MXene Ti2C.65

Microscale: Manipulating Adhesion and Encapsulating Living Cells.

Microscale surface topography represents a powerful approach to direct the adhesion of mammalian cells or bacteria. Pioneering work utilized photolithography and semiconductor fabrication techniques to elucidate the effect of topographic features on cell shape and behavior.66 In particular, anisotropic architectures such as aligned grooves can mimic the fibrillar architecture of the extracellular matrix, particularly bundled collagen I.67 Mechanical patterning of silicone elastomers by X. Zhao and coworkers, as well as PS shrink films by M. Khine and coworkers, have been used to direct cell shape and differentiation.6871 Moreover, 2D materials exhibit distinct surface chemistries that can be easily functionalized,4 overcoming some inherent limitations of tissue culture plastic (e.g., PS) or glass. For instance, Loh and coworkers reported that stem cells exhibit enhanced growth and differentiation on planar GO (relative to tissue culture plastic) due to increased insulin adsorption.72 Emerging 2D materials such as silicate nanoclays may mimic the inorganic composition of bone, enhancing cell adhesion and differentiation.30 Nevertheless, the biocompatibility of emerging 2D materials in a wrinkled state remains to be further explored.

Our group has investigated the morphology and orientation of fibroblasts on wrinkled multilayer GO substrates.50 Multilayer GO coatings of varying thickness (10–200 nm) were deposited on elastomeric substrates and controllably wrinkled with varying wavelengths (2–30 μm), then gently heated to stabilize the multilayered structure. Mouse (NIH-3T3) and human (NHF) fibroblasts were then cultured on these substrates for 48 h and showed good viability. Fibroblasts exhibited a high degree of alignment on these wrinkled substrates (0°±13) relative to planar substrates (13°±57°), which were essentially random (Figure 4A). Moreover, fibroblasts were narrower and more elongated on wrinkled substrates, becoming more spread with a star-like (multipolar) morphology on planar substrates. Overall, we expect these aligned topographic features represent a 2.5D material for interrogating cancer cell invasion in confinement,73 as well as to modulate the foreign body response by macrophages.74 Moreover, controlled accordion-like wrinkling or unwrinkling of these 2D materials could be used to probe the dynamic cell response to alterations of substrate topography, elucidating the timescales of cell differentiation and other phenotypic changes.75

Figure 4.

Figure 4.

A. Mouse fibroblasts (NIH-3T3) align along wrinkled multilayered GO substrates. Inset: Representative wrinkled topography. Adapted with permission from50. Copyright 2016 Elsevier. B. Bacteria (S. auerus) contact wrinkled GO substrates and become encapsulated by GO nanosheets. Adapted with permission from76. Copyright 2017 American Chemical Society. C. Encapsulation of bacteria (B. subtilis) under a wrinkled graphene monolayer. Adapted with permission from79. Copyright 2016 American Chemical Society. D. Encapsulation of a live breast cancer cell (MDA-MB-231) for Raman spectroscopy. E. Hybrid graphene-PNIPAM-graphene-gold nanoparticle “skin” folds around a spherical object in response to thermal stimulus. Adapted with permission from81 under Creative Commons BY-NC license.

Alternatively, wrinkled multilayer GO substrates can be prepared with antibacterial activity. J. Lee and coworkers deposited GO nanosheets onto a pre-stretched filter using vacuum filtration, resulting in highly oriented wrinkles with a characteristic wavelength of 10 μm.76 Three types of bacteria (E. coli, M. smegmatic, and S. aureus) were then deposited on these substrates using a drop test. Bacteria typically became trapped between GO wrinkles, and adhered tightly to the GO coating (Figure 4B). Furthermore, the bacterial cell membranes appeared to be physically disrupted, which was attributed to destructive extraction of the membrane lipids by the GO nanosheets. Interestingly, GO nanosheets also appeared to dissociate from the substrate and partially encapsulate the bacteria, which could further disrupt nutrient transport and surface activity. This work highlights the potential cytotoxicity of 2D materials when allowed to dissociate and degrade in aqueous conditions, which is advantageous for antimicrobial coatings but may be problematic for biomedical implants. Moreover, the potential dispersal of GO nanosheets into the natural environment merits careful consideration.77, 78

Highly flexible 2D material monolayers can also be conformally wrapped around living cells in order to encapsulate them and template new nanostructures. Berry and coworkers cultured rod-shaped bacteria (B. subtilis) on a silicon wafer and transferred monolayer graphene on top.79 The bacteria was gently heated under a vacuum to remove water, causing a gradual volume contraction that drove the longitudinal wrinkling of the monolayer graphene (Figure 4C). These periodic wrinkles aligned along the long axis of the bacteria, with a typical wavelength of 33 nm. The electrical properties of this wrinkled graphene were further probed by dielectrophoretically aligning a bacterium between two opposing metal electrodes, revealing decreased electrical resistance in the longitudinal direction (parallel to the wrinkles), relative to the transverse direction (across the wrinkles). More generally, these bioelectrical interfaces could be used to interrogate or enhance electron transfer with microbes, building towards hybrid bionic devices.80

D. Gracias and coworkers have demonstrated that monolayer graphene can be functionalized with a temperature responsive polymer (PNIPAM brushes), which can contract and drive folding into three-dimensional shapes.81 For instance, upon gentle heating of initially planar flower-like shape from 35°C to 45°C, the respective pedals folded upwards to wrap around a living breast cancer cell (MDA-MB-231). In order to demonstrate conformal wrapping, a first graphene monolayer was again grafted with a temperature-responsive PNIPAM brush, followed by a second graphene monolayer, then decorated with plasmonic silver nanocubes.82 This planar multilayered “skin” could be patterned into various geometries, such as a dumbbell shape (i.e., two connected disks, each approximately 200 μm in diameter) (Figure 4D). Upon heating, the two opposing disks folded upwards and wrapped conformally around a 3D object. As a proof-of-concept, the graphene-PNIPAM-graphene “skin” was wrapped around an MDA-MB-231 cell (Figure 4E). The conformal contact of the silver nanocubes with the cell surface enabled SERS all around. In comparison, SERS could only resolve the bottom of the cell when adherent to the planar (unfolded) graphene-PNIPAM-graphene “skin.” More generally, analogous protective graphene monolayers can be used to contain liquid water in a high vacuum environment, enabling transmission electron microscopy of “wet” samples. Thus, encapsulation with 2D materials can be utilized to enhance the access and resolution of high resolution imaging modalities for subcellular processes.83, 84

Macroscale: Stimuli-Responsive 2D Material Actuators

Actuators based on 2D materials can dynamically shape-change and move in response to a wide range of natural stimuli (such as heat, light, and sound),8588 as well as also artificial stimuli (including electrochemical potential, solvent adsorption/desorption, and magnetic fields).8991 For example, various 2D materials, such as GO, MXenes, and TMDs, have been extensively investigated to serve as stimuli-responsive actuators because of their extraordinary electrical, mechanical, optical, and thermal characteristics.92 To date, multi-functional structures based on the aforementioned 2D materials have been fabricated to perform prompt and significant actuation responses to various stimuli including but not limited to humidity,93 light,94 electrochemical potentials,90, 91, 95, 96 and magnetic fields (Figure 5).97

Figure 5.

Figure 5.

Various stimuli-responsive actuators based on 2D materials. A. B. An actuator based on homogeneous GO film driven by moisture gradient, relying on the in situ formation of a bilayer structure induced by water adsorption, displays flipping locomotion on a moist substrate at 40 °C. Adapted with permission from93. Copyright 2018 Royal Society of Chemistry. C. An IR-transformable architecture assembled from precut r-GO/GO strips that performs reversible transformations between pre-determined poses upon ON/OFF switching of the IR lights. Adapted with permission from94. Copyright 2018 John Wiley & Sons. D. Electromechanical actuator based on stacked 1T MoS2 film shows substantial curvature changes induced by charge intercalation (E. and F.), leading to the fabrication of inverted-series connected bimorph actuator (G. and H.). Adapted with permission from95. Copyright 2017 Springer Nature. I. Magnetically responsive behaviors of multilayered 2D materials intercalated with paramagnetic holmium ions. J. Magnetized Ho3+-GO wheel can be actuated remotely by a magnet to perform designed actions, including rolling right, clockwise rotating, and rolling left. Adapted with permission from97. Copyright 2019 John Wiley & Sons.

Learning from the pinecones with aligned layers exhibiting nano/mesoscale conformation changes triggered by variation of humidity, Ge et al. developed a homogeneous GO actuator driven by moisture gradient, relying on the in situ formation of a bilayer structure induced by water adsorption (Figure 5A).93 Rendered by the homogeneous structure, the resulting GO actuator was highly sensitive to the stimuli of humidity gradients, displaying rapid and continuous actuation (Figure 5B). In addition, introducing an asymmetric design into the GO-based actuator could be an alternative strategy to enable their responsiveness to moisture stimuli. As Qiu et al. reported, a Janus GO film consisting of a wavy layer and a smooth layer was fabricated, which exhibited outstanding moisture-triggered responsiveness with maximum bending angle of ~1800° as the relative humidity changed.98 Moreover, specific actuation response could be achieved by designing complex 3D macroscale architecture based on the moisture-responsive GO units. For instance, Luo et al. constructed functional 3D structures by simply cutting and pasting the rGO/GO bilayer strips (Figure 5C).94 Benefiting from different degrees of water absorption/desorption between GO and rGO layers, reversible photo-thermal transformations between pre-determined poses were performed upon ON/OFF switching of the IR lights.

In addition, the exceptional mechanical property and electrochemical activity of 2D materials enable their potential application in electromechanical actuators, which can be further applied to adaptive wings for air craft, steerable catheters, and drag-reducing wind turbines. Recently, by stacking chemically exfoliated nanosheets of metallic 1T MoS2 onto thin plastic substrates (Figure 5D), Acerce et al. fabricated an electromechanical MoS2 actuator that generated substantial mechanical forces (with stress of ~17 MPa) over hundreds of cycles at frequencies up to 1 Hz, which can be controlled with the applied potential and scan rate.95 The instant and reversible bending behaviors of MoS2 actuator (Figures 5E, 5F) are attributed to the expansion and contraction during the charge-discharge processes, resulting from high electrical conductivity of 1T MoS2 and fast proton diffusion between MoS2 nanosheets. On top of that, an inverted-series connected bimorph actuator (Figures 5G, 5H) was further fabricated with pure vertical displacement, and the elliptic MoS2 spring could be considered as a single-piece, passive compliant actuator with no moving parts. On the other hand, as a representative of MXene, delaminated Ti3C2 paper has been found to show unique macroscopic deformation during the charge/discharge in various aqueous electrolytes, which strongly depends on the cation size and charge.90, 91 Different from conventional electrodes (e.g., graphite, silicon) that volumetrically expand upon redox intercalation of ions (e.g., Li+), the Ti3C2 paper undergoes large contraction during Mg2+, Na+ and Li+ intercalation, which can be easily adjusted by simply varying the electrolyte. This may inspire the exploration of electromechanical actuation behaviors of other MXene materials and further design of derivative electromechanical actuators.

Besides their intrinsic properties, we have shown that additional characteristics could be introduced to the multilayers of 2D materials by intercalating functional guest materials (e.g., metal ions), endowing their broader actuator applications.99 Very recently, Li et al. demonstrated a reversible magnetization method through de-/intercalation of paramagnetic holmium ions (Ho3+) into various multilayered 2D materials ranging from GO, montmorillonite (MMT), MXene, MoS2 to metal-organic framework (Figure 5I).97 The Ho3+ intercalation substantially increased the magnetic susceptibility of 2D materials and enabled their active response to relatively weak magnetic fields (~0.4 mT). One of the showcases Li et al. demonstrated was the magnetized GO wheel, which was remotely actuated by a magnet to perform designed actions, including rolling right, clockwise rotating, and rolling left (Figure 5J). Such Ho3+-intercalated 2D materials with active magnetic response are of great interest in the fabrication of remotely-controlled functional actuators/robots.

Macroscale: GO-Derived Backbone Materials for Soft (Origami) Robotics

Beyond simple actuation, 2D materials with high surface reactivity can be utilized as a useful template for synthesizing multifunctional backbones of soft actuators and robots. The confined galleries within multilayered 2D materials provide an alternative route for the patterning ultrathin architectures of various nanomaterials. Particularly, the GO multilayers have been adopted as sacrificial templates for the nucleation and growth of nanomaterials due to their rich surface functional groups, ease of aqueous processing, and enlarged interlayer spacings.100 With the intercalation of metal ion precursors, the interlayer galleries can guide the conversion of precursors into lamellar metal oxide (MO) structures during the high-temperature oxidation process. This GO-enabled intercalation approach enables the growth of various MO nanocrystals (e.g., ZnO,101, 102 TiO2, Fe2O3, and YBa2Cu3O7−δ (Y123)103) with complex micro- and meso-scale structures replicated from the GO templates, such as planar multilayers,20 wrinkled/crumpled textures,17 and freestanding strands.12 The efficient structural reproducibility achieved by the nanoscale GO templates makes it a promising approach to fabricating the multifunctional backbones for soft (origami) robots.

Yang et al. recently developed a GO-enabled templating synthesis method to produce complex MO origami structures from their paper origami templates with high structural replication.104 With further stabilization by a thin layer of elastomer, the resulting MO-elastomer origamis enabled the fabrication of MO robots with unconventional functionalities. The synthesis of MO robotic backbones is schematically illustrated in Figure 6A, which involves four major steps: (i) deposition of GO nanosheets onto cellulose paper origamis, (ii) spontaneous intercalation of hydrated metal ions, (iii) removal of carbonaceous templates via calcination, and (iv) stabilization of MO replicas with elastomer.

Figure 6.

Figure 6.

A. Schematic illustration of GO-enabled templating synthesis for transforming the backbone material of origami robot from cellulose paper to reconfigurable MO metamaterial. B. Photos of the Al3+-GO-cellulose origami assemblies and as-templated Al2O3 origami structures. C. Large deformability of Al2O3-elastomer auxetic hexagonal origami (180° bending, 180° twisting, 60% stretching). D. Pneumatic actuation of an Al2O3 bellows origami robot with superior fire retardancy. E. Fire-reborn concept of origami robot. A pneumatic paper robot can sacrifice itself in a fire scene to transform into a downsized Al2O3 robot that crawls through a narrow tunnel. Adapted with permission from104. Copyright 2019 American Chemical Society

For instance, various 3D origami structures in Al2O3, including auxetic hexagonal honeycomb and bellows tubes, were successfully synthesized by using this GO-enabled templating approach (Figure 6B). After infiltrated with dilute elastomer solution followed by the curing process, the templated Al2O3 origamis were stabilized with thin elastomer coating. The resulting Al2O3-elastomer origamis not only well replicated the folding patterns of their paper templates but also exhibited high deformability (180° folding, 180° twisting, 60% stretching) (Figure 6C). The MO-elastomer origamis can be categorized as reconfigurable metamaterials and integrated with different actuation systems (e.g., magnetic fields, shape-memory alloys, and pneumatics) towards the fabrication of soft MO robots. These unconventional MO backbones not only well replicated the structures of original paper origami templates with additional distinct features given by the synthesized MO, but also possessed lower weight, higher compliancy, and energy efficiency in comparison with conventional paper origami robots. Figure 6D shows that the pneumatic origami robot with Al2O3 backbones displayed cyclic locomotion to crawl forward, which even survived under the direct burning of an ethanol burner, attributing to the intrinsic fire retardancy of Al2O3 backbone.

Finally, the legendary phoenix-fire-reborn concept was realized in soft robotics (Figure 6E), where a paper origami robot sacrificed itself in a fire scene and transformed into a downsized Al2O3 robot. The resulting Al2O3 robot then crawled through a narrower tunnel where the original paper origami robot was not fit. The soft (origami) robots with MO backbones are expected to be competitive candidates to work in harsh environments (fire scenes and chemical spills). More fascinating functionalities could be incorporated into the MO origamis via the intercalation of various metal ions during the GO-enabled synthesis, which is expected to further expand the material library for construction of multi-functional soft robotics.

Human-scale: 2D Material Nanocoatings and Devices with Skin-Mimicking Capabilities

Human skin is a remarkable organ that provides necessary functionalities, such as protection, tactile sensation, heat regulation, communication, and excretion, to keep the human body behaving efficiently during daily activities.105 The mimicry of skin’s multifunctionality based on various nanomaterials has attracted tremendous research attention for stretchable and wearable electronics, which may also enable broader applications in soft robotics. Among diverse nanomaterials, 2D materials have attracted particular interest for mimicking the functionalities of human skin due to their intriguing physiochemical properties.106 Here, we summarize recent progress of applying 2D materials to fabricate stretchable nanocoatings and electronics with skin-mimicking capabilities, such as barrier protection and tactile sensors (Figure 7).

Figure 7.

Figure 7.

2D material nanocoatings and devices with skin-mimicking capabilities. To mimic high stretchability of human skin, stretchable 2D material nanocoatings/devices with hierarchical crumple/wrinkle structures have been fabricated by various approaches, such as A. “balloon-blowing”, B. two-step substrate deformation, and C. back-infiltration of elastomer. Adapted with permission from17, 107, 108. Copyright 2018, 2019 American Chemical Society, John Wiley & Sons. Based on the pre-textured 2D material nanocoatings, stretchable 2D material barriers possess exceptional impermeability to various small-molecule organic liquids (D.) and demonstrate high fire retardancy (E.). The stretchable protective skin allows the soft robots to work in fire scene (F.) and prevents nitrile gloves to be ignited under the contact with ethanol flames (G.). Adapted with permission from108, 109. Copyright 2018, 2019 American Chemical Society, John Wiley & Sons. Moreover, stretchable tactile sensors were fabricated by the use of 2D materials, and the skin-like 2D material devices were able to differentiate the strain and pressure signals, including pressure-insensitive strain sensor (H.), strain-insensitive pressure sensors (I. and J.) with fascinating application in surgical robots (J.). Adapted with permission from17, 113, 114. Copyright 2019 American Chemical Society.

To enable the realistic applications of 2D materials in mechanically dynamic system, the major obstacle for 2D materials is their limited stretchability. It is well noted that most of 2D materials possess superior mechanical strength and high Young’s modulus, while exhibited very limited ductility or stretchability. Therefore, 2D materials are very prone to undergo structural failure upon tension due to weak intersheet van der Waals forces. To address the intrinsic drawback of 2D materials, various approaches have been developed to fabricate stretchable 2D material devices. One of the promising methods is through the integration of higher dimensional 2D material nanocoating onto soft elastomer to achieve a bilayer hybrid structure. The textured 2D material nanocoating can tolerate the in-plane strains by undergoing reversible unfolding/crumpling processes, which allows the bilayer 2D material-elastomer structure to be stretched to the designed strains, enabling the development of stretchable 2D material device.

To achieve the higher dimensional textures of 2D materials, Wang et al. has reported an efficient “balloon-blowing” method with rGO as demonstration (Figure 7A).107 The basic steps for this approach include: (i) biaxial stretching an elastomeric substrate with areal strain of ~300% and covering a circular dish to form an enclosed space, (ii) heating the glass dish to increase the air pressure within and induce further expansion of the elastomer substrate, (iii) spray coating rGO nanosheets onto the expanded substrate, and (iv) relaxing the substrate by air cooling. The rGO film under multiaxial compressions formed hierarchical wrinkle patterns, which underwent reversible folding/unfolding behaviors to achieve enhanced cyclic stretchability. More recently, based on similar concept, Chang et al. utilized a thermally responsive PS substrate (i.e., shrink film) and a pre-stretched elastomer (VHB tape) to sequentially deform planar rGO nanocoating into multi-generational wrinkle-crumple rGO structure.17 As shown in Figure 7B, the GO suspension was deposited onto the PS films and further heated for biaxial shrinkage, forming isotropic GO crumples. The GO crumples were then reduced to rGO and transferred to a pre-stretched VHB tape. Upon the relaxation of uniaxial strain, larger wrinkles were formed on top of the pre-existing rGO crumples. This wrinkle-crumple rGO/VHB structure exhibited high stretchability (~100%) and strain-insensitive resistance profiles.

Besides the use of pre-fabricated elastomer substrate to achieve stretchable 2D material devices, another approach is to utilize shrink films to deform planar 2D material films into isotropic crumples followed by the back infiltration of uncured liquid elastomer (e.g., PDMS). After curing and dissolution of PS substrate, a freestanding 2D material-elastomer bilayer devices with high stretchability can be achieved (Figure 7C).108 It was demonstrated that the back-infiltration of uncured liquid elastomer maximized the contact area between rGO micro-textures and PDMS backing, resulting in an interlocked interface with strong adhesion. These crumpled rGO coatings remained mechanically stable, revealing they were resistant to unfolding under stretching and various cleaning stresses. Besides providing high mechanical stability, human skin also serves as the body’s first defense system and effectively prevents various types of threats such as bacteria, chemical agents, high temperature, and other environmental threats. Notably, 2D materials with intrinsic superior thermal and chemical stabilities are capable of mimicking the skin’s barrier functions to various external threats. As such, 2D materials have been widely investigated regarding their barrier applications in hindering the penetration of small molecules, reducing the spread of harmful biological substances, and enduring high temperature flame.

Chen et al. developed a textured GO-latex bilayer structure that can serve as ultrastretchable molecular barriers (Figure 7D).108 Large-area GO multilayer films were firstly deposited onto the pre-stretched latex substrates and subsequently underwent biaxial contraction. The resulting bilayer devices were able to undergo repeated stretching/relaxation processes up to a 1500% areal strain for 500 cycles and proved to be excellent barrier layers against various small-molecule organic liquids even under extreme deformations. As shown in Figure 7D, in contrast to the latex balloon that burst immediately, the GO-protected balloons were impermeable to nine diverse organic solvents under various areal strains. More impressively, textured GO coated glove was able to endure the permeation of bulk dichloromethane for more than 30 minutes. With exceptional barrier functions under strains, the ultrastretchable GO-latex bilayer structures are of great significance in potential barrier technologies, such as selective membranes, protective fabrics.

In addition to chemical resistance, human skin also serves to protect against injury due to excess heat. Inspired by this, Chang et al. reported a work using pre-textured MMT nanocoatings with high stretchability and fire retardancy (Figure 7E7G).109 By transferring the crumpled MMT coatings onto elastomers, the bilayer MMT-elastomer structure was obtained to perform high stretchability up to 225% areal strain and effective flame retardancy. Compared to the elastomer without MMT which directly burned within 5 seconds, the MMT-elastomer composite showed high flame retardancy over 60 seconds under uniaxial strain up to 100% (Figure 7E). The conformal MMT nanocoating was further utilized as flame-retardant skin for a soft robotic gripper, which was able to manipulate objects from a fire scene (Figure 7F). Moreover, with the protection of conformal MMT nanocoatings, the nitrile gloves can endure direct flame contact throughout the whole area without immediate ignition (Figure 7G). This fire-retardant protective skin is of great realistic use for soft robotics that can execute tasks in a fire scene, protective clothes for firefighters, and firefree gloves/furniture in daily life.

Another signature feature of human skin is its sensing capability to decouple tactile stimuli (between in-plane strain and normal pressure),110 which is difficult to achieve for conventional piezoresistive sensors that generate the combined signals from both strain and pressure.111, 112 To address this challenge, Oh et al. developed a pressure insensitive strain sensor (Figure 7H)113 relying on different structural changes in response to in-plane strains and compressive pressure. The uniaxial strain led to the formation of micro-cracks within CNT networks, whereas the compressive pressure was relieved by deforming the micro-pores, which barely affected the electrical pathways of CNT networks. As a result, the gauge factor for strain sensing can reach to ~56 under 70% strain whereas negligible change in resistance was observed even under pressure up to 140 kPa.

On the other hand, challenges remain in the fabrication of stretchable pressure sensor that performs stably under large strains without sacrificing its sensitivity. Recently, Choi et al. embedded AgNWs/rGO nanocomposites into polyurethane (PU) dielectric layer, which was further sandwiched between two stretchable electrodes on PDMS substrates (Figure 7I).114 The resulting AgNWs/rGO capacitive pressure sensors showed negligible change in resistance under 50% strain and exhibited touch sensing capability with differences in capacitance with respect to the distance from the touching points. An alternative approach to endow strain insensibility is to pre-fold/wrinkle the compliant conductive 2D materials, allowing the resulting devices to tolerate certain uniaxial strain without significant change in resistance. As Chang et al. just reported (Figure 7J),17 stretchable rGO electrodes with hierarchical Shar-Pei-like topographies have been fabricated by a two-step deformation method, as described in Figure 7B. The resulting rGO electrode exhibited stable resistance (<3%) that was insensitive to repeated uniaxial strain up to 100%. By further clamped two rGO electrodes face-to-face, the resulting pressure sensor can detect “soft” and “hard” presses with high sensitivities of 1.37, 1.30 and 0.98 kPa−1 under 0%, 30%, and 50% strains, respectively. Further applications of such pressure sensors have demonstrated in the transoral robotic surgery (TORS), which provided tactile “awareness” to sense unpredictable collisions and avoided any false alarms during the robotic surgery.

Discussion and Future Directions

Ultrathin 2D materials have emerged as a versatile modality for interfacing with biomolecules, cells, robotics, and human. Indeed, 2D materials are intrinsically multiscale, since individual nanosheets can be prepared as atomically thin monolayers, but with lateral dimensions on the order of centimeters. These length scales are sufficient to interface with biomolecules and living cells, and have benefited from developments in preparing high quality 2D materials at wafer scale for electronic and photonic devices.20 Nevertheless, interfacing between robotics and humans may require 2D material architectures that exceed tens of centimeters, requiring unconventional fabrication techniques better suited for stretchable, large area devices.10 Controlled stacking of nanosheets into multilayer structures through wet deposition or vacuum filtration represents a facile route towards large area 2D material coatings, which can be implemented (relatively) inexpensively but with a considerably higher defect density. However, one consideration for mass production of 2D material based devices is the overall consistency of nanosheets, which can vary considerably.115 Further work is necessary to characterize and optimize nanosheet processing for quality control, especially as 2D materials become more widely used for commercial devices. Moreover, large area and high quality nanocoatings will require 2D material nanocoatings require generalized and scalable deposition techniques, such as inkjet-printing and spin-spray coating.

Mechanochemical patterning of 2D materials typically occurs through interfacial templating against a soft substrate or a hard nanostructure. In the first instance, mismatched mechanical properties can result in buckling-like behaviors, which permit wrinkled or crumpled surface topographies. Under larger deformations, the 2D material coating can partially delaminate from the substrate, resulting in sharper features.47, 49, 50 Such delamination has been shown theoretically to depend on the adhesion energy or interfacial toughness between the stiff coating and soft substrate.46 Our previous work has demonstrated that multilayered GO coatings exhibit robust attachment to a silicone-based elastomer, which remains intact undergoing laundry or dishwashing cycles.108 Nevertheless, the adhesion of 2D materials with non-elastomeric materials of varying chemical composition merits further investigation. For instance, there is extensive interest in the use of water-swollen hydrogels for drug delivery, tissue engineering, and soft machines.116 An intriguing prospect is to integrate 2D materials with 3D printed hydrogels with dynamic functionalities.117119 Alternatively, 2D materials can be interfaced with lower dimensional nanoparticles or nanowires for new device architectures.20 In both these instances, stronger chemical attachment mechanisms beyond simple van der Waals interactions will be beneficial for sustained device performance in biological fluids.

Dynamic mechanical actuation of 2D material components without fracture enables integration into wearable technologies and soft robotic machines with skin-like multifunctionality (e.g., protection, regulation, tactile sensing) to augment interactions between humans and machines. One emerging application is to integrate the stretchable 2D material nanocoatings and devices onto soft robotics, and the 2D material-integrated soft robots can serve as a multifunctional system that can mimic the abilities of both human skin and muscle. For instance, the skin-like tactile sensors can continuously monitor the robotic movements and measure actuation parameters of the robotic bodies in real time. The applications of 2D materials that endows real-time tactile sensing provides both scientific merits and commercial impacts especially in surgical robots. The skin-like tactile sensors will enable surgeons to perform safer and more precise surgical procedures. With recent advances in wireless technology, these hybrid devices may be controlled and powered without external tethered interfaces, permitting less invasive procedures to access human physiology.120

Finally, the biological toxicity and environmental interactions of 2D materials remains an active area of study.77, 78 Many 2D materials with inorganic composition are relatively stable in aqueous environments, or can release toxic ionic species (e.g h-BN, WS2, etc.).121 This may be advantageous for antimicrobial applications, for instance, since 2D materials can penetrate or chemically compromise cell membranes over short timescales.76 Nevertheless, this could be a disadvantage of 2D materials relative to soft and polymeric materials that can be designed with controlled degradation chemistries. We believe that the addition of 2D materials can greatly augment the limited mechanical strength, conductivity, and stimuli-responsiveness of polymeric materials. Additional studies are necessary to elucidate the uptake and toxicity of 2D materials with varying composition, surface chemistry, size, and shape.122

The potential application of 2D materials for wearable technologies implies that potential exposure would occur through the skin. This represents a promising area for further investigation, since existing studies typically occur with skin keratinocytes or fibroblasts in.123 It is conceivable that 2D material degradation could result in skin irritation or other disorders, and certain applications may require chemistries that impart longer-term stability. More generally, the foreign body response occurs through sequential interactions with the immune system, via infiltration of neutrophils, macrophages, and fibroblasts, as well as encapsulation by giant cells. Interestingly, exposure to GO nanosheets has resulted in activation of the neutrophil “first responders,” including the release of neutrophil extracellular traps (NETs)124 and reactive oxygen species.125 In the latter case, inflammation-provoking GO nanosheets can be chemically degraded, although the stability of other 2D materials may vary. For instance, MoS2 nanosheets can dissolve in aqueous solution,126 which may be useful for transient devices implanted in vivo.127 Overall, 2D materials may elicit distinct biological responses after biodistribution in various organs, including the lungs, heart, GI system, nervous system, and reproductive organs.128 Further chemical stabilization and surface functionalization will likely to be necessary to utilize 2D materials safely for prolonged human use.

Conclusion

In summary, this article highlights innovative mechanochemical approaches to pattern and manipulate 2D materials across a wide range of biological length scales. We provided an introductory primer on the surface chemistry and physical properties of 2D materials, and the use of mechanical deformation to pattern distinct surface topographies with localized curvature. We then considered four case studies corresponding to distinct length scales and applications: First, we considered the use of strain engineering to modulate molecular interactions, including chemical functionalization and molecular sensing. In the limit of smaller strains, the 2D materials adopted a gently curved topography, resulting in altered electronic structure, local polarization, and enhanced chemical reactivity. With larger strains, the 2D materials formed sharper, sawtooth-like topographies, which were used for plasmonic surface enhancement of Raman spectroscopy in conjunction with gold nanoparticles. Second, we considered how mechanical manipulation of 2D materials could be used to interface with mammalian or bacterial cells. For instance, stabilized wrinkled GO substrates could be used to direct mammalian fibroblast adhesion and alignment, but non-stabilized wrinkled GO architectures had a cytotoxic effect on bacteria by disrupting their cell membranes. Moreover, graphene or GO could be conformally wrapped around living cells to facilitate electron microscopy or enhance Raman spectroscopy. Third, we reviewed the use of various 2D materials for the fabrication of soft actuators, which could be actuated by multiple external stimuli, including light, humidity, electrochemical potentials, and magnetic fields. These stimuli-responsive 2D material actuators are the essential components for smart soft robotics. Also, we discussed an emerging approach to synthesize new backbone materials in various MOs for soft robotic applications. The soft (origami) robotics with MO backbones demonstrated unconventional and advantageous functions compared with the conventional paper-based robots. Fourth, we consider human length scales, where the 2D material nanocoatings and devices can imitate the multifunctionality of human skin (e.g., mechanical stretchability, protection, and sensing). The skin-like 2D materials can function as biotic-abiotic interfaces to enhance the robot-robot and robot-human interactions. Ultimately, we envision that the exceptional physicochemical properties of 2D materials enable new interfacial modalities with living systems and may also facilitate new biologically-inspired functionalities for new hybrid materials, for energy storage and conversion, as well as environmental technologies.

Acknowledgements

We apologize to colleagues whose work could not be covered due to space limitations. We thank J. Daigle for preparing figures, as well as R.H. Hurt and K.S. Kim for useful discussions. We acknowledge funding from the U.S. National Cancer Institute’s Innovative Molecular Analysis Technologies (IMAT) Program (R21CA212932), as well as Brown University’s Office of the Vice President of Research. The authors acknowledge the financial support provided by the Faculty Research Committee (FRC) Start-Up Grant of University of Singapore R-279-000-515-133, the Ministry of Education (MOE) Academic Research Fund (AcRF) R-279-000-532-114, R279-000-551-114, R-397-000-227-112, the AME Young Investigator Research Grant R-279-000-546-305 (A*STAR Grant No. A1884c0017), and the Singapore-MIT Alliance for Research and Technology (SMART) Ignition Grant R-279-000-572-592.

Footnotes

Conflicts of Interest

I.Y. Wong, and P.Y. Chen are co-inventors with R.H. Hurt on a patent pending (“Substrate with Graphene-Based Layer,” US20190022975A1) filed by Brown University.

References

  • 1.Huebsch N and Mooney DJ, Nature, 2009, 462, 426–432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ferrari AC, Bonaccorso F, Fal’ko V, Novoselov KS, Roche S, Bøggild P, Borini S, Koppens FHL, Palermo V, Pugno N, Garrido JA, Sordan R, Bianco A, Ballerini L, Prato M, Lidorikis E, Kivioja J, Marinelli C, Ryhänen T, Morpurgo A, Coleman JN, Nicolosi V, Colombo L, Fert A, Garcia-Hernandez M, Bachtold A, Schneider GF, Guinea F, Dekker C, Barbone M, Sun Z, Galiotis C, Grigorenko AN, Konstantatos G, Kis A, Katsnelson M, Vandersypen L, Loiseau A, Morandi V, Neumaier D, Treossi E, Pellegrini V, Polini M, Tredicucci A, Williams GM, Hee Hong B, Ahn J-H, Min Kim J, Zirath H, van Wees BJ, van der Zant H, Occhipinti L, Di Matteo A, Kinloch IA, Seyller T, Quesnel E, Feng X, Teo K, Rupesinghe N, Hakonen P, Neil SRT, Tannock Q, Löfwander T and Kinaret J, Nanoscale, 2015, 7, 4598–4810. [DOI] [PubMed] [Google Scholar]
  • 3.Wick P, Louw-Gaume AE, Kucki M, Krug HF, Kostarelos K, Fadeel B, Dawson KA, Salvati A, Vazquez E, Ballerini L, Tretiach M, Benfenati F, Flahaut E, Gauthier L, Prato M and Bianco A, Angew. Chem., Int. Ed. Engl, 2014, 53, 7714–7718. [DOI] [PubMed] [Google Scholar]
  • 4.Eigler S and Hirsch A, Angew. Chem., Int. Ed. Engl, 2014, 53, 7720–7738. [DOI] [PubMed] [Google Scholar]
  • 5.Akinwande D, Brennan CJ, Bunch JS, Egberts P, Felts JR, Gao H, Huang R, Kim J-S, Li T, Li Y, Liechti KM, Lu N, Park HS, Reed EJ, Wang P, Yakobson BI, Zhang T, Zhang Y-W, Zhou Y and Zhu Y, Extreme Mech. Lett, 2017, 13, 42–77. [Google Scholar]
  • 6.Deng S and Berry V, Mater. Today, 2016, 19, 197–212. [Google Scholar]
  • 7.Chen P-Y, Liu M, Wang Z, Hurt RH and Wong IY, Adv. Mater, 2017, 29, 1605096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wang Q and Zhao X, MRS Bulletin, 2016, 41, 115–122. [Google Scholar]
  • 9.Deng S, Sumant AV and Berry V, Nano Today, 2018, 22, 14–35. [Google Scholar]
  • 10.Rogers JA, Someya T and Huang Y, Science, 2010, 327, 1603–1607. [DOI] [PubMed] [Google Scholar]
  • 11.Shenoy VB and Gracias DH, MRS Bulletin, 2012, 37, 847–854. [Google Scholar]
  • 12.Anichini C, Czepa W, Pakulski D, Aliprandi A, Ciesielski A and Samorì P, Chem. Soc. Rev, 2018, 47, 4860–4908. [DOI] [PubMed] [Google Scholar]
  • 13.Han B, Zhang Y-L, Zhu L, Li Y, Ma Z-C, Liu Y-Q, Zhang X-L, Cao X-W, Chen Q-D, Qiu C-W and Sun H-B, Adv. Mater, 2019, 31, 1806386. [DOI] [PubMed] [Google Scholar]
  • 14.Ha M, Lim S and Ko H, J. Mater. Chem. B, 2018, 6, 4043–4064. [DOI] [PubMed] [Google Scholar]
  • 15.Perkins FK, Friedman AL, Cobas E, Campbell P, Jernigan G and Jonker BT, Nano Lett., 2013, 13, 668–673. [DOI] [PubMed] [Google Scholar]
  • 16.Wang Z, Tonderys D, Leggett SE, Williams EK, Kiani MT, Spitz Steinberg R, Qiu Y, Wong IY and Hurt RH, Carbon, 2016, 97, 14–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Chang T-H, Tian Y, Li C, Gu X, Li K, Yang H, Sanghani P, Lim CM, Ren H and Chen P-Y, ACS Appl. Mater. Interfaces, 2019, 11, 10226–10236. [DOI] [PubMed] [Google Scholar]
  • 18.Dreyer DR, Park S, Bielawski CW and Ruoff RS, Chem. Soc. Rev, 2010, 39, 228–240. [DOI] [PubMed] [Google Scholar]
  • 19.Kurapati R, Kostarelos K, Prato M and Bianco A, Adv. Mater, 2016, 28, 6052–6074. [DOI] [PubMed] [Google Scholar]
  • 20.Liu Y, Huang Y and Duan X, Nature, 2019, 567, 323–333. [DOI] [PubMed] [Google Scholar]
  • 21.Mannix AJ, Kiraly B, Hersam MC and Guisinger NP, Nat Rev Chem, 2017, 1, 0014. [Google Scholar]
  • 22.Shi L and Zhao T, J. Mater. Chem. A, 2017, 5, 3735–3758. [DOI] [PubMed] [Google Scholar]
  • 23.Tao W, Kong N, Ji X, Zhang Y, Sharma A, Ouyang J, Qi B, Wang J, Xie N, Kang C, Zhang H, Farokhzad OC and Kim JS, Chem. Soc. Rev, 2019, 48, 2891–2912. [DOI] [PubMed] [Google Scholar]
  • 24.Tao W, Ji X, Zhu X, Li L, Wang J, Zhang Y, Saw PE, Li W, Kong N, Islam MA, Gan T, Zeng X, Zhang H, Mahmoudi M, Tearney GJ and Farokhzad OC, Adv. Mater, 2018, 30, 1802061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Liu T and Liu Z, Adv. Healthcare Mater, 2018, 7, 1701158. [Google Scholar]
  • 26.Yin J, Li J, Hang Y, Yu J, Tai G, Li X, Zhang Z and Guo W, Small, 2016, 12, 2942–2968. [DOI] [PubMed] [Google Scholar]
  • 27.Jing L, Tay RY, Li HL, Tsang SH, Huang JF, Tan DL, Zhang BW, Teo EHT and Tok AIY, Nanoscale, 2016, 8, 11114–11122. [DOI] [PubMed] [Google Scholar]
  • 28.Jing L, Samani MK, Liu B, Li H, Tay RY, Tsang SH, Cometto O, Nylander A, Liu J, Teo EHT and Tok AIY, ACS Appl. Mater. Interfaces, 2017, 9, 14555–14560. [DOI] [PubMed] [Google Scholar]
  • 29.Jing L, Li H, Tay RY, Sun B, Tsang SH, Cometto O, Lin J, Teo EHT and Tok AIY, ACS Nano, 2017, 11, 3742–3751. [DOI] [PubMed] [Google Scholar]
  • 30.Chimene D, Alge DL and Gaharwar AK, Adv. Mater, 2015, 27, 7261–7284. [DOI] [PubMed] [Google Scholar]
  • 31.Bianco A, Cheng H-M, Enoki T, Gogotsi Y, Hurt RH, Koratkar N, Kyotani T, Monthioux M, Park CR, Tascon JMD and Zhang J, Carbon, 2013, 65, 1–6. [Google Scholar]
  • 32.Carlson A, Bowen AM, Huang Y, Nuzzo RG and Rogers JA, Adv. Mater, 2012, 24, 5284–5318. [DOI] [PubMed] [Google Scholar]
  • 33.Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, Grigorieva IV and Firsov AA, Science, 2004, 306, 666–669. [DOI] [PubMed] [Google Scholar]
  • 34.Dikin DA, Stankovich S, Zimney EJ, Piner RD, Dommett GH, Evmenenko G, Nguyen ST and Ruoff RS, Nature, 2007, 448, 457–460. [DOI] [PubMed] [Google Scholar]
  • 35.Araujo PT, Terrones M and Dresselhaus MS, Mater. Today, 2012, 15, 98–109. [Google Scholar]
  • 36.Banhart F, Kotakoski J and Krasheninnikov AV, ACS Nano, 2011, 5, 26–41. [DOI] [PubMed] [Google Scholar]
  • 37.Denis PA and Iribarne F, J. Phys. Chem. C, 2013, 117, 19048–19055. [Google Scholar]
  • 38.Shen X, Lin X, Yousefi N, Jia J and Kim J-K, Carbon, 2014, 66, 84–92. [Google Scholar]
  • 39.Shim W, Brown KA, Zhou X, Rasin B, Liao X and Mirkin CA, Proc. Natl. Acad. Sci. U. S. A, 2012, 109, 18312–18317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Reserbat-Plantey A, Kalita D, Han Z, Ferlazzo L, Autier-Laurent S, Komatsu K, Li C, Weil R, Ralko A, Marty L, Gueron S, Bendiab N, Bouchiat H and Bouchiat V, Nano Lett., 2014, 14, 5044–5051. [DOI] [PubMed] [Google Scholar]
  • 41.Choi J, Kim HJ, Wang MC, Leem J, King WP and Nam S, Nano Lett., 2015, 15, 4525–4531. [DOI] [PubMed] [Google Scholar]
  • 42.Meyer JC, Geim AK, Katsnelson MI, Novoselov KS, Booth TJ and Roth S, Nature, 2007, 446, 60–63. [DOI] [PubMed] [Google Scholar]
  • 43.Obraztsov AN, Obraztsova EA, Tyurnina AV and Zolotukhin AA, Carbon, 2007, 45, 2017–2021. [Google Scholar]
  • 44.Calado VE, Schneider GF, Theulings AMMG, Dekker C and Vandersypen LMK, App. Phys. Lett, 2012, 101, 103116. [Google Scholar]
  • 45.Huang ZY, Hong W and Suo Z, J. Mech. Phys. Solids, 2005, 53, 2101–2118. [Google Scholar]
  • 46.Mei H, Landis CM and Huang R, Mech. Mater, 2011, 43, 627–642. [Google Scholar]
  • 47.Zang J, Ryu S, Pugno N, Wang Q, Tu Q, Buehler MJ and Zhao X, Nat. Mater, 2013, 12, 321–325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Xu P, Kang J, Choi JB, Suhr J, Yu J, Li F, Byun JH, Kim BS and Chou TW, ACS Nano, 2014, 8, 9437–9445. [DOI] [PubMed] [Google Scholar]
  • 49.Thomas AV, Andow BC, Suresh S, Eksik O, Yin J, Dyson AH and Koratkar N, Adv. Mater, 2015, 27, 3256–3265. [DOI] [PubMed] [Google Scholar]
  • 50.Wang Z, Tonderys D, Leggett SE, Williams EK, Kiani MT, Steinberg RS, Qiu Y, Wong IY and Hurt RH, Carbon N Y, 2016, 97, 14–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Lin S, Lee EK, Nguyen N and Khine M, Lab Chip, 2014, 14, 3475–3488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Wang MC, Chun S, Han RS, Ashraf A, Kang P and Nam S, Nano Lett., 2015, 15, 1829–1835. [DOI] [PubMed] [Google Scholar]
  • 53.Chen P-Y, Sodhi J, Qiu Y, Valentin TM, Steinberg RS, Wang Z, Hurt RH and Wong IY, Adv. Mater, 2016, 28, 3564–3571. [DOI] [PubMed] [Google Scholar]
  • 54.Wu Q, Wu Y, Hao Y, Geng J, Charlton M, Chen S, Ren Y, Ji H, Li H, Boukhvalov DW, Piner RD, Bielawski CW and Ruoff RS, Chem. Commun, 2013, 49, 677–679. [DOI] [PubMed] [Google Scholar]
  • 55.Bissett MA, Konabe S, Okada S, Tsuji M and Ago H, ACS Nano, 2013, 7, 10335–10343. [DOI] [PubMed] [Google Scholar]
  • 56.Hermanson GT and Hermanson GT, 2013.
  • 57.Zubko P, Catalan G and Tagantsev AK, Annu. Rev. Mater. Res, 2013, 43, 387–421. [Google Scholar]
  • 58.Kothari M, Cha M-H and Kim K-S, Proc. R. Soc. A, 2018, 474, 20180054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Klitgaard SK, Egeblad K, Haahr LT, Hansen MK, Hansen D, Svagin J and Christensen CH, Surf. Sci, 2007, 601, L35–L38. [Google Scholar]
  • 60.Li R, Kothari M, Landauer AK, Cha M-H, Kwon H and Kim K-S, MRS Adv, 2018, 3, 2763–2769. [Google Scholar]
  • 61.Leem J, Wang MC, Kang P and Nam S, Nano Lett, 2015, 15, 7684–7690. [DOI] [PubMed] [Google Scholar]
  • 62.Baumberg JJ, Aizpurua J, Mikkelsen MH and Smith DR, Nat. Mater, 2019, DOI: 10.1038/s41563-019-0290-y. [DOI] [PubMed] [Google Scholar]
  • 63.Chen W, Gui X, Zheng Y, Liang B, Lin Z, Zhao C, Chen H, Chen Z, Li X and Tang Z, Adv. Opt. Mater, 2017, 5, 1600715. [Google Scholar]
  • 64.Feng Q, Duan K, Ye X, Lu D, Du Y and Wang C, Sensor Actuat. B-Chem, 2014, 192, 1–8. [Google Scholar]
  • 65.Yu X-F, Li Y-C, Cheng J-B, Liu Z-B, Li Q-Z, Li W-Z, Yang X and Xiao B, ACS Appl. Mater. Interfaces, 2015, 7, 13707–13713. [DOI] [PubMed] [Google Scholar]
  • 66.Bettinger CJ, Langer R and Borenstein JT, Angew. Chem., Int. Ed. Engl, 2009, 48, 5406–5415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Kim DH, Provenzano PP, Smith CL and Levchenko A, J Cell Biol, 2012, 197, 351–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Chen A, Lieu DK, Freschauf L, Lew V, Sharma H, Wang J, Nguyen D, Karakikes I, Hajjar RJ, Gopinathan A, Botvinick E, Fowlkes CC, Li RA and Khine M, Adv. Mater, 2011, 23, 5785–5791. [DOI] [PubMed] [Google Scholar]
  • 69.Chen A, Lee E, Tu R, Santiago K, Grosberg A, Fowlkes C and Khine M, Biomaterials, 2014, 35, 675–683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Cao C, Chan HF, Zang J, Leong KW and Zhao X, Adv. Mater, 2014, 26, 1763–1770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Chan HF, Zhao R, Parada GA, Meng H, Leong KW, Griffith LG and Zhao X, Proc. Natl. Acad. Sci. U. S. A, 2018, 115, 7503–7508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Lee WC, Lim CH, Shi H, Tang LA, Wang Y, Lim CT and Loh KP, ACS Nano, 2011, 5, 7334–7341. [DOI] [PubMed] [Google Scholar]
  • 73.Paul CD, Hung WC, Wirtz D and Konstantopoulos K, Annu. Rev. Biomed. Eng, 2016, 18, 159–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.Chen S, Jones JA, Xu Y, Low HY, Anderson JM and Leong KW, Biomaterials, 2010, 31, 3479–3491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.Tibbitt MW and Anseth KS, Sci. Transl. Med, 2012, 4, 160ps124. [DOI] [PubMed] [Google Scholar]
  • 76.Zou F, Zhou H, Jeong DY, Kwon J, Eom SU, Park TJ, Hong SW and Lee J, ACS Appl. Mater. Interfaces, 2017, 9, 1343–1351. [DOI] [PubMed] [Google Scholar]
  • 77.Wang Z, Zhu W, Qiu Y, Yi X, von dem Bussche A, Kane A, Gao H, Koski K and Hurt R, Chem. Soc. Rev, 2016, 45, 1750–1780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.Fadeel B, Bussy C, Merino S, Vazquez E, Flahaut E, Mouchet F, Evariste L, Gauthier L, Koivisto AJ, Vogel U, Martin C, Delogu LG, Buerki-Thurnherr T, Wick P, Beloin-Saint-Pierre D, Hischier R, Pelin M, Candotto Carniel F, Tretiach M, Cesca F, Benfenati F, Scaini D, Ballerini L, Kostarelos K, Prato M and Bianco A, ACS Nano, 2018, 12, 10582–10620. [DOI] [PubMed] [Google Scholar]
  • 79.Deng S, Gao E, Wang Y, Sen S, Sreenivasan ST, Behura S, Kral P, Xu Z and Berry V, ACS Nano, 2016, 10, 8403–8412. [DOI] [PubMed] [Google Scholar]
  • 80.Hsu L, Deng P, Zhang Y, Nguyen HN and Jiang X, J. Mater. Chem. B, 2018, 6, 7144–7158. [DOI] [PubMed] [Google Scholar]
  • 81.Xu W, Qin Z, Chen C-T, Kwag HR, Ma Q, Sarkar A, Buehler MJ and Gracias DH, Sci. Adv, 2017, 3, e1701084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 82.Xu W, Paidi SK, Qin Z, Huang Q, Yu CH, Pagaduan JV, Buehler MJ, Barman I and Gracias DH, Nano Lett, 2019, 19, 1409–1417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 83.Mohanty N, Fahrenholtz M, Nagaraja A, Boyle D and Berry V, Nano Lett, 2011, 11, 1270–1275. [DOI] [PubMed] [Google Scholar]
  • 84.Wojcik M, Hauser M, Li W, Moon S and Xu K, Nat. Commun, 2015, 6, 7384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 85.Fu F, Shang L, Chen Z, Yu Y and Zhao Y, Sci. Robot, 2018, 3, eaar8580. [DOI] [PubMed] [Google Scholar]
  • 86.Mu J, Hou C, Wang H, Li Y, Zhang Q and Zhu M, Sci. Adv, 2015, 1, e1500533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 87.Yang L, Qi K, Chang L, Xu A, Hu Y, Zhai H and Lu P, J. Mater. Chem. B, 2018, 6, 5031–5038. [DOI] [PubMed] [Google Scholar]
  • 88.Chen L, Weng M, Zhou P, Huang F, Liu C, Fan S and Zhang W, Adv. Funct. Mater, 2019, 29, 1806057. [Google Scholar]
  • 89.Torop J, Palmre V, Arulepp M, Sugino T, Asaka K and Aabloo A, Carbon, 2011, 49, 3113–3119. [Google Scholar]
  • 90.Gao Q, Come J, Naguib M, Jesse S, Gogotsi Y and Balke N, Faraday Discuss, 2017, 199, 393–403. [DOI] [PubMed] [Google Scholar]
  • 91.Come J, Black JM, Lukatskaya MR, Naguib M, Beidaghi M, Rondinone AJ, Kalinin SV, Wesolowski DJ, Gogotsi Y and Balke N, Nano Energy, 2015, 17, 27–35. [Google Scholar]
  • 92.Zhou Y, Feng W, Qian X, Yu L, Han X, Fan G, Chen Y and Zhu J, ACS Appl. Mater. Interfaces, 2019, 11, 19712–19723. [DOI] [PubMed] [Google Scholar]
  • 93.Ge Y, Cao R, Ye S, Chen Z, Zhu Z, Tu Y, Ge D and Yang X, Chem. Commun, 2018, 54, 3126–3129. [DOI] [PubMed] [Google Scholar]
  • 94.Luo C, Yeh C-N, Baltazar JML, Tsai C-L and Huang J, Adv. Mater, 2018, 30, 1706229. [DOI] [PubMed] [Google Scholar]
  • 95.Acerce M, Akdoğan EK and Chhowalla M, Nature, 2017, 549, 370. [DOI] [PubMed] [Google Scholar]
  • 96.Li K, Shao Y, Yan H, Lu Z, Griffith KJ, Yan J, Wang G, Fan H, Lu J, Huang W, Bao B, Liu X, Hou C, Zhang Q, Li Y, Yu J and Wang H, Nat. Commun, 2018, 9, 4798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 97.Li K, Chang T-H, Xie Q, Cheng Y, Yang H, Chen J and Chen P-Y, Adv. Electron. Mater, 0, 1900040. [Google Scholar]
  • 98.Qiu Y, Wang M, Zhang W, Liu Y, Li YV and Pan K, Nanoscale, 2018, 10, 14060–14066. [DOI] [PubMed] [Google Scholar]
  • 99.Chen P-Y, Liu M, Valentin TM, Wang Z, Spitz Steinberg R, Sodhi J, Wong IY and Hurt RH, ACS Nano, 2016, 10, 10869–10879. [DOI] [PubMed] [Google Scholar]
  • 100.Li Z, Wu S, Lv W, Shao J-J, Kang F and Yang Q-H, Small, 2016, 12, 2674–2688. [DOI] [PubMed] [Google Scholar]
  • 101.Yeh C-N, Raidongia K, Shao J, Yang Q-H and Huang J, Nat. Chem, 2015, 7, 166–170. [DOI] [PubMed] [Google Scholar]
  • 102.Saito Y, Luo X, Zhao C, Pan W, Chen C, Gong J, Matsumoto H, Yao J and Wu H, Adv. Funct. Mater, 2015, 25, 5683–5690. [Google Scholar]
  • 103.Boston R, Bell A, Ting VP, Rhead AT, Nakayama T, Faul CFJ and Hall SR, Cryst. Eng. Comm, 2015, 17, 6094–6097. [Google Scholar]
  • 104.Yang H, Yeow BS, Chang T-H, Li K, Fu F, Ren H and Chen P-Y, ACS Nano, 2019, DOI: 10.1021/acsnano.9b00144. [DOI] [PubMed] [Google Scholar]
  • 105.Agache P, Lihoreau T, Mac-Mary S, Fanian F and Humbert P, Agache’s Measuring the Skin: Non-invasive Investigations, Physiology, Normal Constants, 2017, 1–4. [Google Scholar]
  • 106.Chang T-H, Li K, Yang H and Chen P-Y, Adv. Mater, 2018, 30, 1802418. [DOI] [PubMed] [Google Scholar]
  • 107.Mu J, Hou C, Wang G, Wang X, Zhang Q, Li Y, Wang H and Zhu M, Adv. Mater, 2016, 28, 9491–9497. [DOI] [PubMed] [Google Scholar]
  • 108.Chen P-Y, Zhang M, Liu M, Wong IY and Hurt RH, ACS Nano, 2018, 12, 234–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 109.Chang T-H, Tian Y, Wee DLY, Ren H and Chen P-Y, Small, 2018, 14, 1800596. [DOI] [PubMed] [Google Scholar]
  • 110.Bartolozzi C, Natale L, Nori F and Metta G, Nat. Mater, 2016, 15, 921. [DOI] [PubMed] [Google Scholar]
  • 111.Amjadi M, Kyung K-U, Park I and Sitti M, Adv. Funct. Mater, 2016, 26, 1678–1698. [Google Scholar]
  • 112.Yang T, Xie D, Li Z and Zhu H, Mater. Sci. Eng. R Rep, 2017, 115, 1–37. [Google Scholar]
  • 113.Oh J, Yang JC, Kim J-O, Park H, Kwon SY, Lee S, Sim JY, Oh HW, Kim J and Park S, ACS Nano, 2018, 12, 7546–7553. [DOI] [PubMed] [Google Scholar]
  • 114.Choi TY, Hwang B-U, Kim B-Y, Trung TQ, Nam YH, Kim D-N, Eom K and Lee N-E, ACS ACS Appl. Mater. Interfaces, 2017, 9, 18022–18030. [DOI] [PubMed] [Google Scholar]
  • 115.Kauling AP, Seefeldt AT, Pisoni DP, Pradeep RC, Bentini R, Oliveira RVB, Novoselov KS and Neto A. H. Castro, Adv. Mater, 2018, 30, e1803784. [DOI] [PubMed] [Google Scholar]
  • 116.Zhang YS and Khademhosseini A, Science, 2017, 356, eaaf3627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 117.Valentin TM, Leggett SE, Chen P-Y, Sodhi JK, Stephens LH, McClintock HD, Sim JY and Wong IY, Lab Chip, 2017, 17, 3474–3488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 118.Valentin TM, Landauer AK, Morales LC, DuBois EM, Shukla S, Liu M, Valentin LHS, Franck C, Chen P-Y and Wong IY, Carbon, 2019, 143, 447–456. [Google Scholar]
  • 119.Valentin TM, DuBois EM, Machnicki CE, Bhaskar D, Cui FR and Wong IY, Polym. Chem, 2019, 10, 2015–2028. [Google Scholar]
  • 120.Sarycheva A, Polemi A, Liu Y, Dandekar K, Anasori B and Gogotsi Y, Sci. Adv, 2018, 4, eaau0920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 121.Martín C, Kostarelos K, Prato M and Bianco A, Chem. Commun, 2019, 55, 5540–5546. [DOI] [PubMed] [Google Scholar]
  • 122.Zhang X, Hu W, Li J, Tao L and Wei Y, Toxicol. Res, 2012, 1, 62–68. [Google Scholar]
  • 123.Fadeel B, Bussy C, Merino S, Vázquez E, Flahaut E, Mouchet F, Evariste L, Gauthier L, Koivisto AJ, Vogel U, Martín C, Delogu LG, Buerki-Thurnherr T, Wick P, Beloin-Saint-Pierre D, Hischier R, Pelin M, Candotto Carniel F, Tretiach M, Cesca F, Benfenati F, Scaini D, Ballerini L, Kostarelos K, Prato M and Bianco A, ACS Nano, 2018, 12, 10582–10620. [DOI] [PubMed] [Google Scholar]
  • 124.Mukherjee SP, Lazzaretto B, Hultenby K, Newman L, Rodrigues AF, Lozano N, Kostarelos K, Malmberg P and Fadeel B, Chem, 2018, 4, 334–358. [Google Scholar]
  • 125.Kurapati R, Mukherjee SP, Martin C, Bepete G, Vazquez E, Penicaud A, Fadeel B and Bianco A, Angew. Chem., Int. Ed. Engl, 2018, 57, 11722–11727. [DOI] [PubMed] [Google Scholar]
  • 126.Wang Z, von dem Bussche A, Qiu Y, Valentin TM, Gion K, Kane AB and Hurt RH, Environ. Sci. Technol, 2016, 50, 7208–7217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 127.Chen X, Park YJ, Kang M, Kang SK, Koo J, Shinde SM, Shin J, Jeon S, Park G, Yan Y, MacEwan MR, Ray WZ, Lee KM, Rogers JA and Ahn JH, Nat. Commun, 2018, 9, 1690. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 128.Zhang X, Yin J, Peng C, Hu W, Zhu Z, Li W, Fan C and Huang Q, Carbon, 2011, 49, 986–995. [Google Scholar]

RESOURCES