Symposium CP06—Smart Materials for Multifunctional Devices and Interfaces

We will concentrate on electrical-to-mechanical energy conversion using nanoporous metal-polymer composite materials. Nanoporous metallic actuators constitute a new class of low-voltage actuators that feature a unique combination of relatively large strain amplitudes, low operating voltages, and high specific stiffness and strength. These so-called ‘metallic muscles’ consist of ligaments and pores in the nanometer regime giving rise to a very high internal surface area.
The key obstacles to the integration of nanoporous metals into current fundamental concepts and technological applications (MEMS, NEMS) are (i) the presence of the aqueous electrolyte itself that is needed to inject electronic charge in the space-charge region at the metal/electrolyte interface. (ii) the rate of actuation due to the relatively low ionic conductivity of the electrolyte, and (iii), the magnitude of the actuating displacements. Here we discuss a novel approach to generate work from metallic muscles that overcome these hurdles. From an experimental viewpoint a new ultrafast, all-solid organometallic actuator has been designed, synthesized and tested. The tunable, semiconducting properties of conjugated polymers are exploited to inject charge into the metal. In addition, a new microstructural design based on a layered structure with enhanced actuation strokes has been developed. In the presentation also size effects of metallic muscles in particular will be discussed. In addition we have explored a new phenomenon, i.e. ion-induced bending phenomena which may serve as a versatile tool to shape and manipulate these nanostructured devices. The work shows that an in-depth understanding of focused electron and focused ion beams can play a predictive role in a quantitative control in for the nanofabrication of small-sized nano- products like metallic muscles.

We have recently designed, fabricated and tested shape memory alloy elements in spring, tube and sheet forms for use in thermal switches, connect-disconnect mechanisms, deployable membrane telescopes and torsional tube actuators. The work has benefitted from in situ neutron diffraction at Los Alamos National Laboratory and Oak Ridge National Laboratory at stress and temperature in carefully selected experiments. The first aspect of this work will address implications for training (thermomechanical cycling for stable strain recovery in use) these alloys for cyclic aerospace applications and rely on experiments performed under (i) isothermal (both uniaxial and torsional deformation at constant temperature), (ii) isobaric (thermal cycling to temperatures above and below the phase transformation temperatures under constant uniaxial stress), and (iii) isostrain (thermal cycling to temperatures above and below the phase transformation temperatures under constant uniaxial strain) conditions. Additionally, spatially resolved information was obtained in the case of the torsion specimens. The second aspect of this work will address the methodology adopted to design shape memory alloy springs along with results from comprehensive testing. The design considerations included multiple springs in limited operating volumes with requirements of forces exerted prior to actuation and following actuation, length, stroke and actuation temperature. Novel approaches include using springs for radial compression, innovative biasing configurations and using the R-phase transformation in NiTiFe springs. The discussion will also highlight appropriate modification of the theory in cases where substantial plastic deformation occurs during fabrication of the springs.

Actuators form the basis of the modern robotics industry, with applications ranging from microelectromechanical devices to medical or consumer electronics. Various actuator types including piezoelectric actuators, shape memory alloys and electrochemical actuators have been devised to meet the wide range of application needs. Especially, recent urge to develop humanoide robots and exoskeletons that mimic or aid human motions call for further development of a special kind of actuators called the artificial muscles. The requirements for successful artificial muscles include a large amount of strain (~10%) and large stress on the order of 100 MPa, preferably with low power consumption. Several novel actuator types have been suggested to serve as artificial muscles such as polymer-based actuators or electrochemical actuators. These actuators achieve successful integration into several robotics parts, with the characteristically large amount of strain, fast response time and below 3 V operation. Most artificial muscles developed so far, however, feature volatile actuation where voltage must be continuously supplied to maintain the actuated position; as soon as the input voltage is turned off, regardless of how big or small the operation voltage, the actuated position returns to its original position. Such volatile actuation is featured in most well-known actuators, including the piezoelectric, shape memory alloy-based, electrostrictive and ionic actuators. For artificial muscles, this volatility indicates a significant loss of efficiency, as mimicking human motions involve a series of constant actuated positions. Such non-volatile actuation capability, however, has not yet been realized with many well-known artificial muscles. Here, we develop a non-volatile actuator with Li_xGe electrodes for artificial muscle. The novel actuator type consists of ion insertion electrodes so that the inserted ions pose stress on the electrodes that sustain even after power shut-off. The device consists of lithium germanide (Li_xGe) thin films deposited on both sides of a flexible polyimide (PI) substrate. The demonstrated device operates at low power below 2 V and generates large stress per mass, a critical performance factor for actuators. The observed stress, computed via Stoney’s curvature formula, reaches 240 MPa for the 250 nm Li₃Ge thin films and volume expansion of 8.2%. The actuated position is maintained against gravity with 12.1% decay in the actuated distance after 10 minutes. The novel actuator type proves the potential use of lithium insertion materials as actuation materials and shows that non-volatile actuation can be realized with ion-insertion electrodes.

Bistable electroactive polymers (BSEPs) combine the properties of shape memory polymers and dielectric elastomer actuation at the rubbery state to generate rigid-to-rigid actuation. The variable stiffness could be obtained via glass transition or phase changing. The reversible melting-crystallization of polymer chains in the phase changing BSEP leads to a narrow temperature band to complete the stiffness change. A modulus change greater than 1000 fold can be achieved within a few degrees. Large actuation strains could be obtained. To drive the large-strain deformation, compliant electrode materials are also developed that maintain electrical conductivity at the large strains.

Liquid crystalline elastomers (LCEs) are soft, anisotropic materials that exhibit large shape transformations when subjected to various stimuli. This shape morphing results when the net order of the material is reduced by external stimuli, such as heat, light, or electric fields. Here, we demonstrate a facile approach, lamination, to enhance the out-of-plane work capacity of these materials by an order of magnitude, to nearly 20 J/kg. Additionally, these materials can displace loads more than 2500x heavier than its own weight over 0.5 mm. The enhancement in force output is enabled by the development of a room temperature polymerizable composition used both to prepare individual films, organized via directed self-assembly to retain arrays of topological defect profiles, as well as act as an adhesive to combine the LCE layers.<sup>1

1</sup>Guin, T.; Settle, M. J.; Kowalski, B. A.; Auguste, A. D.; Beblo, R. V; Reich, G. W.; White T. J. “Layered liquid crystal elastomer actuators,” Nature Communications, 2018, 9, 2531.

Shape memory alloys such as NiTi intermetallic compounds exhibit a shape memory effect and a pseudoelastic effect owing to the transition between B2 austenite and B19(prime) martensitic phases. Upon a large enough stress, dislocations will be formed and hinder the B2-B19(prime)-B2 transformation that will diminish the shape memory effect. To improve the shape-memory and pseudoelastic performance, NiTi-based minor-doped, multicomponent and even in recent years high-entropy intermetallic compounds have been developed. However, because of the complex atom configuration and large lattice distortion in high-entropy materials, the structure, phase transition, mechanical response and deformation behavior of NiTi-based high-entropy intermetallic compounds need further investigations. Thus in this work, NiTi-based low-entropy (binary), medium-entropy (quaternary) and high-entropy (senary) intermetallic compounds were prepared by arc melting and casting (thereafter with homogenization), and their structures, mechanical properties and deformation behavior were then characterized. X-ray diffractions indicated that all the samples had a B2 austenite structure, but an expansion of unit volume was obviously noted in the high-entropy intermetallic compounds. Nanoindentation tests on different-orientation grains revealed the loss of anisotropy of mechanical properties in the high-entropy intermetallic compounds. The in situ SEM compression of low-entropy micropillars presented a pseudoelastic response and, upon yielding at a high load, typical long-distance slip deformation. However, the in situ SEM compression of high-entropy micropillars showed homogeneous barrel-like deformation without clear slip lines or plastic heterogeneity. The cross-sectional TEM observation revealed a short-distance activity of abundant partial dislocations rather than long-distance gliding of few complete dislocations in the deformed high-entropy micropillars.

Particle-polymer composites are typically formulated for one of two purposes: to reinforce the polymer and improve its stiffness, strength and mechanical robustness; or to impart physical properties not commonly found in polymers, often electrical properties, to the resultant material. Within soft electronics, the latter is of common use, using highly deformable elastomers and carbon or metal particles to create stretchable conductive materials. However, reinforcement of the polymer still occurs, as the conductive material is significantly stiffer and more brittle than the polymer matrix.

Phase inversion techniques, commonly used to fabricate filtration membranes, offer a mechanism to induce micro and nanoscale porosity in polymer materials, reducing their stiffness and impacting their max elongation. Particulate additives can then be used to further modify the structure, and impart electrical properties. Together, macroscale properties of the composite can be tuned by controlling both the composition of the final material and its structure.

We demonstrate this using evaporation induced phase inversion of PVDF and explore the impact of non-solvent concentration and nano-scale ceramic particle loading on porosity and morphology. Particulate size and material are also varied, and show a dramatic impact on the structure of the pores present within the materials. This structure can be controlled from nano-scale voids to micro-scale foam-like openings. Testing shows how this morphology impacts the mechanical properties of these composites, determining how the materials can deform. Similar materials have already been used as battery separators. We look to further develop these porous composites to act as functional materials in soft electronics, using the filler to impart desired electrical properties such as conductivity, permittivity, and permeability, while maintaining low stiffness and high elongation.

Wearable technologies are capturing the attention of people – to turn obtrusive electronics into less obtrusive sources of information. The information could relate to health, safety, or social interactions. Such wearable devices need to conform to the skin, be ultra-light, and bio-compatible. All this while being accurate and high quality electronic technology.
Electronic devices typically use coatings of materials such as semiconductors or oxides – these materials are usually brittle. To enable wearable technologies, they would need to be combined with rubber-like membranes.
The presentation would discuss a variety of materials which have been demonstrated in stretchable applications. These include indium tin oxide, zinc oxide, titanium dioxide, strontium titanate, and vanadium oxide.
We have demonstrated a transfer process which is ubiquitous (can be used for any oxide coating), repeatable (and amenable to mass manufacture), scalable, capable of high integration densities (due to ability to micro/nano pattern) and which utilizes high performance oxide materials stretchable up to 15%. This process utilised the naturally weak adhesion of platinum to silicon, and this allows us to create electronics on a rigid substrate such as silicon and then peel off the layers to transfer onto a stretchable substrate. This process also results in a unique ‘micro-tectonic’ surface, creating opportunities to explore new stretchable device applications. This process has been successfully demonstrated using transparent indium tin oxide, zinc oxide thin films, and titanium dioxide thin films with stretchability of up to 15% which is exceptionally high for a brittle oxide. With this process, we have also demonstrated room temperature gas and UV sensors, mechanically tunable diffraction gratings as well as optical metasurfaces.
More recently, we have demonstrated stretchable memory devices. A stretchable non-volatile resistive memory is a fundamental element in realizing complex neuromorphic computing and compact logic application adaptable to wearable electronics. A room temperature deposited strontium titanate based resistive memory on stretchable substrate has been developed. The STO-based resistive memory does not require energy-intensive electroforming, exhibits stable complementary switching, long retention time, and reproducible endurance switching. The devices demonstrate the ability to operate under uniaxial tensile strain and extreme bending conditions.
The presentation will also discuss results from the various oxide based stretchable platforms for applications in sensing, optics, and memories.

Flexible electronics represents a fast-developing field and has a great potential to impact our daily life. In building up flexible electronics, the materials with controllable conduction, transparency, and good flexibility are required. van der Waals epitaxy (vdWE) involving two-dimensional layered materials can play a crucial role in the expansion of thin film epitaxy by overcoming the bottleneck of material combinations due to lattice/thermal matching conditions inherent to conventional epitaxy. Among the layered materials, mica is a well-known phyllosilicate mineral that can have a remarkable impact on flexible electronics. Due to the interplay of lattice, charge, orbital, and spin degrees of freedom, correlated electrons in oxides generate a rich spectrum of competing phases and physical properties. However, a generic approach to build up flexible electronics based on functional oxides is yet to be developed. In this study, we use a 2D material as the substrate. In this talk, we confine ourselves to the validity of vdWE of functional oxides on muscovite mica throughout this treatise. With such demonstrations, it is anticipated that MICAtronics, vdWE on mica, can reveal unusual properties and emergent phenomena in the realm of high-performance flexible device applications.

Smart materials and textiles with abilities to interact and respond to changes in their environment are becoming a need of the hour. Apart from being light-weight, flexible and portable, their fabrication into easy-to-read sensing devices can make the arena of diagnostics more accessible to the end-user. Among the options available today, engineered living materials, comprising a hybrid of living components that utilize energy sources to produce and/or secrete the bulk of the material and non-living components that confer novel properties, are gaining widespread popularity. One promising protein-based biomaterial we have studied in this project is the curli fiber system of E.coli. In this study, we report methods for effectively integrating curli nanofibers into commercially synthetic textiles to create functional wearable devices.

Curli fibers, in nature, constitute the non-pathogenic proteinaceous components of bacterial biofilms and can be genetically engineered to create a diverse range of stable and sensitive devices. These fibers form a high surface area porous network exhibiting unique self-assembly properties. They can be produced in various forms like gel, film, and powder while maintaining their properties. These systems have an edge over conventional organic polymers in that they are biocompatible and biodegradable while also being stable to enzymatic activities, detergent action and harsh chemical treatments. With properties that can be modulated through genetic, chemical and mechanical means, these materials find applications such as optical sensors in the diagnostic (e.g. wound infection) and personal care (e.g.blood glucose, stress hormone level) sectors.

A technique that showed great promise for integration of curli fibers into fabrics is vacuum filtration. In this work, we investigated the integration of protein curli fibers into various types of synthetic textile fabrics through the vacuum filtration technique. The filtration of curli-producing bacterial cultures onto the fabrics was followed by suitable washing and purification to separate the micro-organisms from the hybrid materials. With this method, we were able to load porous textile matrices with protein fibers throughout their cross-section, to create functional textile-protein composites. We optimized parameters affecting the extent of curli fiber incorporation onto the scaffold such as volume of the bacterial culture, porosity of the filtering fabric, and expression levels of the wild type or genetically engineered curli fibers. We also studied the microstructure and mechanical behavior of the fabrics before and after filtration. Microstructural analysis showed that the curli fibers were uniformly incorporated into the polymeric fibers of the fabrics over a large area. In some samples, self-assembly of curli fibers on textile fibers was also observed. For further optimization, we studied the viscosity and rheological properties of the curli fiber gel obtained via vacuum filtration, and we explored suitable printing techniques such as screen or ink jet printing, as an alternative method for incorporating protein fibers into textiles. Last but not the least, curli fibers expressing a fluorescent sensing moiety were incorporated onto the textile scaffold, and we studied changes in functional activities before and after integration for potential development of this system into a fluorescent and wearable textile-based biosensor that is both accurate and user-friendly.

Tomorrow’s Industry 4.0 environments raise a growing demand in self-sustained, flexible and low-cost sensors for versatile applications in process control as well as condition and energy monitoring. We present a new concept for direction-sensitive and flexible strain sensors based on the ferroelectric copolymer P(VDF:TrFE) and a single layer of interdigitated electrodes. Microstructured metal electrodes are formed on plastic substrates and embedded in the copolymer. Electric poling using the embedded electrodes (EE) allows the orientation of the ferroelectric domains primarily parallel to the P(VDF:TrFE)-substrate interface, thus having an increased piezoelectric coupling with respect to lateral strains present in the surface plane. We employ a combination of (self-aligned) photolithography and electroforming to produce 2-10µm wide and up to 4µm high electrodes that are fully covered by screen-printed P(VDF:TrFE). Alternatively, we present an elegant structuring method based on microfluidic channels hot embossed into P(VDF:TrFE) which are filled with conductive ink to form high aspect ratio EE.
Tensile tests with sensors having a footprint area of 1cm² were performed and compared with FEM simulations. The results show a highly linear charge response with respect to longitudinal strain, and the coupling strength depends significantly on the strain orientation in the surface plane. In conclusion, the use of the EE allows selectively resolving strain magnitude and orientation. Furthermore, bending at 90Hz excitation frequency yields an average power output of ~0.6µW per mm³ of active polymer volume, sufficient for stand-alone applications.

Magnetorheological elastomers (MREs) are unique smart materials of high elasticity and magnetic susceptibility. MREs are prominent for the high degree of mechanical deformation or changes in stiffness that can be induced by applying magnetic fields. While most MREs are made with thermoset elastomers, this research focuses on the development and testing of a thermoplastic magnetorheological elastomer (TMRE) for potential use as fused deposition modeling (FDM) filament. FDM, also known as 3D printing, is an additive manufacturing technique that consists of 1D viscous thermoplastic extrusions that create 2D layers that build up to a 3D part. This method of creating parts produces underlying anisotropies which can be tuned to control the properties of the final part. Printing could bring new functionalities and applications for TMREs due to new sub-structures that are available through FDM. Proposed applications include but are not limited to use in actuators, dampers, soft robotics, smart sensors and actuators, wearable technology and biomedical drug delivery. The first step in realizing such applications is the development of functional TMRE filament for FDM. Our TMRE was created utilizing solvent casting techniques to disperse isotropic magnetic particulate within a thermoplastic polyurethane matrix. A range of materials were explored to highlight the effects of particulate type and loading. Samples were created spanning two different magnetic particulate types (150µm iron & 2-4µm magnetite) and each with three different particulate loadings (20, 30, 40 wt%). The material was then extruded into FDM filaments with a Filastruder. Filament width of the extruded filament samples was tracked and compared to evaluate the consistency of the extrusion process used within the lab setting. Mechanical stress vs. strain curves of the extruded filaments were obtained using an MTS tensile tester. Magnetic hysteresis loops were acquired with a vibrating sample magnetometer (VSM). The analogous pure PU filaments were also extruded and tested as a control. Our testing indeed shows that altering the magnetic particulate type and its weight concentrations impacts both the magnetic and mechanical properties of the overall material. In general, the filament samples with magnetite particulate had higher diametric consistency and were stiffer than those with iron particulate. The heightened stiffness is likely due to the particulates smaller size and larger volume percentage of the magnetite given its lower density than iron. Additionally, samples with magnetite had higher magnetic susceptibility and coercivity but lower saturation magnetization than those with iron. Normalization of the saturation magnetization to compensate for the scaling of the particulate percentage indicates a consistent particulate dispersion and integration despite changes in its volume. Lastly, increasing particulate percentage increases both the mechanical stiffness and saturation magnetization of the samples, as expected. Future efforts will focus on further data analysis, magneto-mechanical testing, initial FDM printing tests, and cross-sectional imaging of the extruded samples to develop a robust explanation as to why the material functions in the way it does for insight and understanding when used in FDM. Despite being only the first step in understanding the functionality of this smart material, this step is critical for optimal utilization and realization in future applications.

Cellulose nanofibrils (CNFs) are abundant, naturally-derived nanomaterials with high mechanical strength-to-weight ratio enabled by their nano- to macroscale hierarchical structure. CNFs are also piezoelectric with recently demonstrated suitability for piezoelectric sensing [1]. In this work, we employ CNFs and 2,2,6,6-tetramethylpiperidine-1-oxyl radical (TEMPO)-oxidized CNFs (TOCNFs) as additives in poly(vinylidene fluoride) (PVDF) to prepare free-standing composite films doped with 0.5, 1, 2, 3, 4 and 5 wt% nanocellulose using doctor blade casting. The structure, morphology, surface wettability, as well as thermal, mechanical, and piezoelectric properties of the composites were evaluated with a suite of characterization techniques. For our processing conditions, the incorporation of CNFs and TOCNFs does not change the PVDF crystalline phase, which remains in a majority piezoelectric gamma phase, regardless of the amount of added cellulose. Addition of CNFs tends to promote crystallinity over that of the neat film with the highest increase in crystallinity (~10%) for the 5 wt% CNF/PVDF film, whereas TOCNFs reduce composite crystallinity. Further, CNFs and TOCNFs influence the surface wettability of the composites, as well as their thermal and mechanical behaviors. CNFs and TOCNFs reduce the tensile strength of the composites and the elongation at break indicating loss of ductility and onset of brittle fracture. An 18% increase of the tensile modulus over that of neat PVDF was observed for the 1 wt% TOCNF/PVDF composite. The piezoelectric response of the composite films was evaluated with the Berlincourt method, and electrical response of piezoelectric generators was evaluated under periodically applied loads. Permission to publish was granted by the Director of the Geotechnical and Structures Laboratory.
[1] S. Rajala, et al., ACS Appl. Mater. Interfaces 2016, 8, 15607−15614.

A set of environmental issues caused by spillage of hydrocarbons into the sea leads to an explosively growing interest in the technology of water-oil selective absorption. Recently, Janus water-oil selective absorbers that exhibit hydrophobic (or oleophilic) characteristics on one side and hydrophilic (or oleophobic) ones on the other side attract significant attention in scientific and industrial fields. These absorbers absorb one specific component from water-in-oil or oil-in-water emulsion, which makes them exclusively effective tools for removing small amount, more often final trace, of oil or water. Here, we modify the surface and even bulk characteristics of polydimethylsiloxane (PDMS) by physicochemical treatment where the formation of interconnected pore network and the addition of biocompatible surfactant (i.e., Silwet L-77) are made for PDMS as physical treatment and chemical one, respectively. The physical treatment is intended to create storage spaces for the absorbed liquid and also strengthen the hydrophobicity of PDMS, whereas the chemical one is designed to generate hydrophilic PDMS. 35 kinds of physicochemically treated PDMS samples with different pore sizes, d_ps, of 0.0, 92.2, 176.8, 355.5, and 634.3 µm and various Silwet L-77 concentrations, C_ss, of 0.0, 0.1, 0.5, 1.0, 2.0, 4.0, and 8.0 wt% are fabricated by casting Silwet L-77-added 10:1 (base:curing agent) PDMS on NaCl powder molds, which is prepared by pressure-assisted compaction, and then dissolving the NaCl mold in Silwet L-77 aqueous solution at critical micelle concentration. The Janus PDMS composed of hydrophobic PDMS with d_p = 92.2 µm and C_s = 0.0 wt% and hydrophilic PDMS with d_p = 634.3 µm and C_s = 8.0 wt% is also prepared. Experiments for oil, water, and their emulsion are carried out to quantitatively characterize the effect of physicochemical treatment, d_p and C_s, on the contact angle and absorption speed of PDMS. Water-oil selective absorption using our Janus PDMS is also demonstrated. The physicochemical treatment changes the water wettability of PDMS from hydrophobic to superhydrophobic or superhydrophilic. The formation of interconnected pore network increases the hydrophobicity of PDMS which is inversely proportional to d_p; the addition of Silwet L-77 makes PDMS hydrophilic and even superhydrophilic in a proportional way. The physicochemical treatment significantly affects the water absorption speed of PDMS which are in proportion to d_p and C_s. The physicochemical treatment, however, has almost no influence on the oil wettability and oil absorption speed of PDMS. This is because PDMS, even after physicochemical treatment, has still higher surface energy than oil. The difference between water absorption speed and oil absorption speed induced by physicochemical treatment makes it possible for the Janus PDMS to selectively absorb water and oil from water-in-oil or oil-in-water emulsion. In detail, the superhydrophilic half of the Janus PDMS absorbs water only and the superhydrophobic one absorbs oil only when a mixture of water and oil is in competition for absorption to the Janus PDMS. The Janus PDMS, needless to say, is reusable and also applicable to narrow niches owing to the high elasticity of PDMS. The findings of this study will result in a better understanding of the wetting dynamics of polymers and also an explosively growing use of physicochemically treated PDMS in the development of oil-water selective absorbers. This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT & Future Planning (NRF-2017R1A2B4010300).

Additive manufacturing (AM) enables flexible on-demand production of 3D structures with high shape complexity. Among the available AM techniques, fused deposition modeling (FDM) has attracted increasing attention due to its potential for high cost-efficiency. However, this extrusion-based AM method has some restrictions, the main ones being limited printable materials and poor mechanical performance of printed parts. When applied to crosslinked thermosetting polymers, the main limitation becomes the curing process, which typically prevents the use of these polymers for additive manufacturing as it becomes difficult for the material to maintain its printed form during curing. In addition, current printable materials lack the ability for tunability of material properties. We report a stimuli-responsive epoxy-based feedstock material for FDM. The novel epoxy-based polymer features reversible formation of a crosslinking network in response to temperature and light. At the first step, epoxy resin was oligomerized into stimuli-responsive building blocks which contained Diels-Alder (DA) chemistries (furan and maleimide) as well as photo-responsive cycloadducts (cinnamate). During the FDM process, the reversible and dynamic covalent bonding between furan and maleimide functional groups via DA reactions enables fast curing of the epoxy resin and improves interlayer adhesion. The photo-responsive cycloaddition reaction provides further control of material crosslinking network via exposure to short-wavelength UV irradiation. The resultant epoxy-based feed stock demonstrated fast curing that enables AM of crosslinked polymers via FDM, in which dynamic bonding controls material properties at local scale and provides unique anisotropic stimuli-responsiveness. The reported stimuli-responsive thermosetting polymers present a promising feedstock material for AM whose stiffness and/or shape can be manipulated by temperature and/or light.

Strategies to induce dynamic surface topography may enable new engineering devices where friction, fluid flow, or cell attachment is a property that needs to be controlled on-demand. There are multiple ways to create polymeric materials where the surface topography can be induced to change in response to a stimulus. For example, shape memory polymers can switch from smooth to rough on heating, but require mechanical programming, limiting this effect to only objects that can be easily deformed such as films. Also, liquid crystalline networks could induce surface topography changes without a programming step; however, they require a constant energy source in order to keep the surface topography from returning to its original state. They also require an initial alignment step, which further confines this ability. Here we describe a versatile strategy to significantly and permanently alter the surface topography of films, microstructures and microspheres without any mechanical programming step or initial alignment. Specifically, we use a radical thiol-ene reaction to build semi-crystalline polymer networks from low molecular weight monomers. Crystallization of the network occurs concurrently with polymerization and produces a material with built-in stresses. On heating through the melting point of the network, these initially smooth materials, having an average roughness of 10nm, form peaks and valleys, reaching an average roughness of 500 nm. We have been able to produce this effect in films, micro-molded structures, and microspheres. The magnitude of the roughness can be controlled by altering either the melting point of the network, the crosslink density, and/or the polymerization temperature. For example, we show that the closer the polymerization temperature is to the substrate's melting temperature, the less rough the substrate becomes after being heated to its melting temperature. This allows us to pattern roughness by simply polymerizing different portions of the substrate under various temperatures. This control will allow us to tackle major engineering challenges in the areas of micro fluidics, self-cleaning films, bio-adhesion and cell growth.

A series of Co_1-xZn_xFe₂O₄ ferrite with (x=0.0, 0.1, 0.2, 0.3, 0.4, 0.5) compositions were synthesized using the standard double sintering ceramic technique, sintered at 1050°C for 2 hours. Substituting Zn in place of Co influenced the structural, dielectric and magnetic properties of CoFe₂O₄ samples. Structural analyses were carried out using X-ray diffraction (XRD). The X-Ray diffraction pattern confirmed the single-phase cubic spinel structure and the sharp peak revealed that the samples are in good crystalline form. The lattice parameters were calculated for each composition and found to increase with Zn substitution. A significant increase in density and subsequent decrease in porosity was observed with increasing Zn content. The grain size of the samples was reduced by enhancing the Zn concentration. The dielectric constant (ε′) of the sample is found to decrease with increase in frequency exhibiting normal dielectric behavior. Dielectric relaxation peaks were observed for the frequency dependence of dielectric loss tangent curves. The observed dielectric properties were explained on the basis of electron conduction mechanism. The variation of the resistivity versus temperature was also studied and the dielectric constant of the system has a variation quite similar to that of the resistivity. VSM measurement confirmed that the magnetizations of all the samples were saturated. Saturation magnetization and coercivity were estimated with variation of Zn content. These effects are due to facilitation of demagnetization by substitution of the non-magnetic Zn ions. Permeability was found to decrease with increasing in Zn content. The characteristics of electromagnetism, excellent chemical stability, mechanical hardness, low coercivity, moderate saturation magnetization and high anisotropy constant suggested Cobalt Ferrite a good candidate for synthesizing and investigation to contribute as dielectric, magnetic, multiferroic material and in storage devices for the advancement of science and technology.

Microgels as carrier systems to transport guest molecules have attracted great attention in the past years. Their ability to bind and release guest molecules upon changing the outer stimuli is making them interesting for a wide range of applications especially in biomedical fields. It is well established that microgels interact with guest molecules via different pathways such as covalent binding, hydrophobic interactions and electrostatic forces. In case of the interaction between polyelectrolytes, electrostatic forces are of greater importance. Therefore we aimed to synthesize polyampholyte microgels as carriers to achieve a triggered uptake and release of guest molecules.
Polyampholyte microgels contain both acidic and basic functional groups and are able to change their charge as function of the pH. Their particle sizes under aqueous state swell at both high and low pH values. However, at relative mild pH ranges where both charges are present polyampholyte microgels undergo a de-swelling behavior according to the interaction between positive and negative charges. The existence of an isoelectrical point is another important characteristic feature of polyampholyte microgels revealing at which pH the charges are equal and the overall net charge become zero.
In this research we synthesized new microgels with defined charge localization and various architectures (random, core-shell and Janus-like microgels) with free radical precipitation polymerization. Together with the help of computer simulations, we investigated the swelling and collapse of the microgels with possible shape variation as function of different external stimuli. In addition we studied the interaction of the synthesized microgels with the model protein cytochrome c at different pH values and different internal structures.

In the accident at Fukushima Daiichi Nuclear Power Station, many radioactive materials were released into the environment field. In the future, materials with radiation shielding property will be required for a long period. We have developed and reported Anti-Sievert® Concrete with enhanced shielding function as construction material¹⁾. Anti-Sievert® Concrete has a shielding function of about 3 times as compared with ordinary concrete for X-ray (200 kV) and shows strength exceeding 40 N / mm^{2 1)}. For building materials other than concrete, the application of advanced shielding function is required.
In this report, we improved the function of shielding effects of radioactive materials and radiation on Rock-Wool, which is a typical heat insulating material widely used for many buildings. This material is a blend of Anti-Sievert® (high density ceramics material) with shielding function. For improving the shielding function, it is required to increase the content of Anti-Sievert®. However, in order to realize the heat insulating function, it is essential to be porous. We developed heat insulating materials with practical shielding function by properly balancing these conflicting design values. In the present formulation, it has a shielding function of about 0.2 to 0.3 mm (100 kV) equivalent to lead in Anti-Sievert® Rock-Wool with a thickness of 50 mm. This characteristic corresponds to the shielding function used for medical X-ray transmission photography equipment. Continue to improve the shielding function and report materials with shielding function of practical level that can be widely used for buildings. We think that this material is extremely useful in the Fukushima area.

１）I. Sato ，T. Hatakeyama, Y. Yamashita, Y. Mori, S. Hashimoto, I. Hatsumura, Y. Katakami, T. Nishitsunoi, T. O. Hirano, Development of X-ray Shield Concrete for Secure Safety Against the Impact of the Fukushima Accident,MRS Fall meeting 2017

Shape memory materials traditionally utilize temperature or environmentally dependent phase transitions in order to return from a deformed state. Alternatively, magnetic shape memory alloys have long been investigated, as they can be deformed by magnetic fields rather than environmental conditions; however, these alloys suffer from the drawback that they exhibit very small strains. This has led to recent developments in magnetorheological elastomers (MREs), which can exhibit significantly greater strains. Our research focuses on the processing of a thermoplastic MRE by solvent casting a polyurethane (PU) elastomer with magnetic particulate for fused deposition modelling (FDM) applications. FDM, commonly referred to as 3D printing, is a valuable manufacturing technique that allows for the creation of parts with inherent anisotropies. For the case of an MRE, FDM allows for the production of structures with tunable magnetic susceptibility along different axes. In these composites, the degree of particulate dispersion significantly affects the isotropy of material properties, which becomes increasingly important when small material volumes are used, such as in FDM. Incorporating solvent casting as a versatile method of producing polymer composites will allow for greater control over the particulate addition method, leading to a higher degree of dispersion when compared to a polymer melt. For our purposes, composite thin films were produced in order to examine effect of wet vs. dry addition of particulate on dispersion, while thicker samples were produced to study the porous structures obtainable by utilizing different drying methods. The solvent used in producing the PU composites was dimethylformamide (DMF). Preparation of polymer solutions included dissolution of PU in DMF to 20 w/v% followed by addition of the magnetic particulate. The particulates used were <150 µm iron powder and 2-4 µm magnetite powder. Composites solutions were made to concentrations of 20, 30, and 40 w/w% particulate to polymer by addition of either dry particulate or particulate pre-suspended in DMF. Drying conditions included room temperature air drying, using a dehydrator, or drying in a vacuum oven. Wet addition of magnetic particulate led to smaller clump sizes and improved dispersion, while dry addition led to larger clump sizes and poorer dispersion. All drying techniques prominently affected the degree of porosity in the resulting composite. Future inquiry will focus on the effects that the porosity of the composite, as a result of drying technique, has on further processing and manufacturing; ensuring that the effect of addition method on the dispersion of particulate in thicker samples is equivalent to that found in thin films; and understanding the impact that each of these factors will have on components manufactured by FDM.

Semiconductor nanowire ion-sensitive field effect transistor (NW-ISFET) has attracted considerable attention as a next-generation sensor platform for a wide range of chemical and biological applications. Despite their potential applications, highly reliable sensing performance of NW-ISFET is regarded as one of the main bottlenecks for commercialization of this technology. This issue may be mostly due to a lack of understanding of sensing mechanisms. To address this issue, we have performed 3D device modeling to capture the detailed aspects of NW-ISFET sensing mechanisms with the change of nanowire surface morphologies in chemical and biological applications. At the electrolyte interface between nanowire and liquid, a specific site binding model and Gouy-Chapman-Stern theories were applied in order to consider the realistic ion-screening effects via the electrical double layer. Meanwhile, 3D electronic transports on the nanowires with different surface morphologies were coupled with the electrochemical models. Finally, the simulation results are verified through comparison of our previous experimental data. It was concluded that the main aspects of NW-ISFET sensing performances were captured with our 3D numerical simulation according to the change of nanowire surface morphology. Finally, the detailed origins of the NW-ISFET sensing mechanism will be discussed. We believe that this work can give better insights to improve the reliability of chemical and biological NW-ISFET sensors.

3D printing is a novel fabrication technique for constructing structures with complex geometries and material characteristics. 3D printing of conductive structures could potentially create material systems with unique combination of properties including electrical conductivity, mechanical integrity, and electrochemical properties. Such unique properties render the structures excellent for applications such as micro-electronics, robotics, sensors and actuators, etc. Carbon nanotubes (CNTs) are excellent candidates for filler in inks to be used in 3D printing of conductive structures. The high aspect ratio of 1D CNTs makes them superior to zero-D materials such as metal nanoparticles and 2D materials such as graphene in applications such as fibrous constructs or linear actuators. Here we report 3D printing of carbon nanotube-based inks into geometries not possible with methods other than 3D printing. Our fabrication process consists of preparing a printable mixture of a polymer such as polydimethylsiloxane (PDMS) or poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT) and CNTs by ultrasonication, and then printing in a medium such as gelatin. We report unique properties ranging from mechanical to electrochemical including stiffness, elasticity, and capacitance.

The traditional Von Neumann computing architecture is reaching fundamental limits in terms of energy efficiency and scalability, especially with increasingly data-intensive applications. Hence, it is imperative to look for alternative computing paradigms with collocated computation and memory.¹ One promising example is neuromorphic hardware. These architectures are inspired by the massively interconnected architecture of biological brains, and have so far demonstrated improvements in scalability, robustness, and computational efficiency over traditional architectures. Resistive memories such as memristors (RRAM) and PCRAM are well suited for the artificial synapses of neuromorphic architectures, since they are non-volatile and can store multiple resistance states, similar to biological synapses.
In this work, a novel neuromorphic architecture called the Memristive Nanowire Neural Network (MN3) is proposed.^2,3 The MN3 consists of an array of neural nodes interconnected using a randomized network of memristive nanowires. Each nanowire comprises a conductive core, which transmits signals between neurons, and a memristive shell, which forms memristive synapses with the neural nodes in the array. The techniques for fabricating such an architecture will be demonstrated along with static and dynamical switching characteristics. Hafnium oxide (HfO_x) and tantalum oxide (TaO_x)-based RRAM memristors have been studied for this work. Detailed analysis using complementary techniques such as EDX, XPS, FE-SEM, and HR-TEM will be further presented to 1) understand how structural and microstructural properties are affected by the electrical characterization of such films and 2) identify important parameters to further improve the devices. Simulation results showing improvements in the scalability of the MN3 architecture over traditional crossbar architectures will also be presented.

1. Merolla et al. ‘A million spiking-neuron integrated circuit with a scalable communication network and interface’ Science, 345, 668 (2014)
2. Juan Claudo NINO, Jack D. KENDALL, ‘Memristive nanofiber neural networks’, WO2015195365A1 (2015)
3. Ross et al., ‘Memristive nanowires exhibit small-world connectivity’ Neural Networks 106,144 (2018)

Both experimental and theoretical efforts are unified to study the gas-sensing properties of hydrogen (H2) gas on ZnO films. The theoretical study is based on a combination of density-functional theory (DFT) with non-equilibrium Green’s function (NEGF) formalism. A focus was given to assess the catalytic effects of noble metals versus other low-electronegativity transition metals (e.g., Au, Pt, Ag, Pd, and Fe were put, separately, as ad-atoms on ZnO nano-ribbons (ZnO-NR)) on the transport properties and thus gas-sensing. While the results are in favor of chemisorption to take place in all the cases of the studied catalysts, Pt and Pd in particular have shown more coordination with the ZnO bed (stranger binding energies). Consequently, they both enormously affect the density of states at Fermi level and yield the largest deviations in IV-curve, from before to after the landing of H₂ molecule. In case of Pt catalyst, the sensor’s response is found to be the largest in detecting H₂ gas molecule. Tests of chemisorption using Pt catalyst to other gases (e.g., O₂, N₂, CO₂ and H₂S) have confirmed its great selectivity toward the detection of H₂ gas. The experimental findings are in excellent agreement with the theoretical predictions.

An increasing trend in the use of “green” energy sources for electricity generation has been observed in recent years. Despite this, electricity generation continues to be highly dependent on fossil fuel energy sources all around the world. As a result, the efficiency of energy conversion systems is critical for low emissions. Meticulous monitoring of the operating conditions of these systems can aid in achieving higher efficiencies. Piezoelectric materials are great candidates for this type of application due to their pressure and temperature sensing capabilities as well as their ability to withstand harsh environments. Barium titanate, BaTiO₃, is one of the most prominent piezoelectric ceramics. One of the main drawbacks of implementing piezoelectric ceramics in these systems are their geometry and size constraints. Current manufacturing methods of these ceramics do not allow for custom complex designs. A novel solution to this issue has been the implementation of additive manufacturing to fabricate piezoelectric ceramics with complex geometries. However, most of the recent efforts have yielded low density parts and therefore weak mechanical and electrical properties. The purpose of this research project was to fabricate BaTiO₃ using an ExOne M-Lab binder jetting 3D printer to study its dielectric and piezoelectric properties as a function of the printing direction. The morphology of the samples as well as the dielectric and piezoelectric properties were characterized in the normal and parallel orientations to the printing direction. Samples achieved a density of 36.8% after sintering. Meanwhile, samples that were electroded in the parallel direction to the printing layers presented an average dielectric constant of 34.6% and a piezoelectric constant of 70.5% of the theoretical properties of BaTiO₃. Additionally, improvements in dielectric and piezoelectric constants of 20% and 13.5%, respectively, were observed for the samples tested in the normal direction to the printing layers. These results show that despite low densities, ceramics with high piezoelectric coupling coefficients can be fabricated through binder jetting additive manufacturing. Further optimization in printing parameters and powder used for the fabrication must be performed to achieve higher densities in the final parts.

The solid state light emitting device (SSI-LED) was reported by the author's group recently (1). Upon the application of a gate voltage, the broad band warm white light is emitted from a large number of nano-sized conductive paths formed from the dielectric breakdown of a MOS capacitor on a p-type Si wafer. The light emission principle is the thermal excitation of the conductive path upon the passiage of a current (2). In spite of the difference in gate dielectric materials and process conditions (3), all lights are emitted over the same visible to near IR wavelength range. The SSI-LED also shows unique electric characteristics similar to those of an antifuse or a diode (4). However, since all SSI-LEDs on the same substrate share the common ground electrode, the crosstalk is unavoidable, which is a potential problem in practical applications.

In this paper, a new coplanar structured SSI-LED is reported. Each device has its own gate and ground electrodes both of which are located on the same side of the substrate. Since devices can be isolated with LOCOS, trench, or other methods, there is no issue of crosstalks among different devices. The emission spectrum of the new device is the same as that of the device with gate and electrodes located on different sides of the substrate. The former has a higher intensity of the emitted light than the latter has due to the formation of more conductive paths. The structure of the new device and the pattern of the current flow will be used to explain the improvement of the light emission. Potential applications of the new device will also be examined.

Shumao Zhang is acknowledged for preparation and characterization of SSI-LEDs in this paper.

1) Y. Kuo and C.-C. Lin, Appl. Phys. Letts., 102(3), 031117 (2013).
2) Y. Kuo and C.-C. Lin, Solid State Electronics, 89, 120 (2013).
3) Y. Kuo, IEDM 2014, 104 (2014).
4) Y. Kuo, ECS Trans., 79(1), 21 (2017).

VO₂ being a transition metal oxide is the most fascinating material of research as it shows a semiconductor to metal transition (SMT) near to room temperature (T_SMT ~ 68 ^οC). It undergoes a structural change from M1, monoclinic phase (Space group: P2₁/c) to R, tetragonal phase (Space group: P4₂/mnm).¹ Due to this phase transition, it is used for various applications like smart windows, metamaterial, bolometers, etc.² Herein, we report the synthesis of device quality VO₂ thin films. Various methods like home-built Ultrasonic Nebulised Spray Pyrolysis of Aqueous Combustion Mixture (UNSPACM)³, Chemical Vapor Deposition (CVD)⁴, Pulsed Laser Deposition (PLD) and DC Reactive sputtering have been employed for the synthesis of phase pure VO₂ thin films. The structural characterization has been carried out by XRD and Raman measurements which confirms the M1 phase of VO₂ at room temperature followed by high-temperature measurements to study the phase change. Temperature-dependent resistance measurements confirm the SMT transition at T_SMT ~68 ^οC with a three-four order of magnitude change in resistance. Temperature-dependent IR measurements show ~80% change in reflectance across the phase transition where the thin film acts as an IR reflector (T>T_SMT). Effect of dopant (W, Mo, Sc, Ce etc.) has been studied where the T_SMT was reduced to 25 ^οC with 2 at % W doping. Our preliminary attempts to make metamaterials will also be presented.

1. Goodenough, J. B., The two components of the crystallographic transition in VO₂. Journal of Solid State Chemistry 1971, 3 (4), 490-500.
2. Yang, Z.; Ko, C.; Ramanathan, S., Oxide Electronics Utilizing Ultrafast Metal-Insulator Transitions. Annual Review of Materials Research 2011, 41 (1), 337-367.
3. R. Bharathi, N. R., A. M. Umarji, Metal-insulator transition characteristics of vanadium dioxide thin films synthesized by ultrasonic nebulized spray pyrolysis of an aqueous combustion mixture. Journal of Physics D: Applied Physics 2015, 48 (30), 305103.
4. R. Bharathi; Pradhan, J. K.; Ramakrishna, S. A.; Umarji, A. M., Thermochromic VO₂ thin films on ITO-coated glass substrates for broadband high absorption at infra-red frequencies. Journal of Applied Physics 2017, 122 (16), 163107.

Composite films exhibiting remarkable shape memory effects are prepared via solvent casting of Metal-Organic Framework, Cu-BTC (HKUST-1) in poly(L-lactide) (PLLA) matrix as both reinforcement, and nucleation agent. Their enhanced mechanical properties and heat induced shape have been evaluated.
Porous Organic Polymers (POPs) are known for nearly two decades, and there are a lot of applications which they have shown many promises from catalysis to biomedicine .The time line of Metal Organic Frameworks (MOFs)’s development which are also known as porous coordination polymers is giving the promise of a turning point with the setting up of the first commercial MOF products for the storage and delivery of toxic gases. The potential of MOFs to influence material properties in a more fundamental basis can go further beyond its conventional fields of study. There are bottlenecks hindering the progress with many of the proposed next-generation applications of MOF materials which basically is the current lack of understanding of the fundamental framework mechanisms resulting in the flexibility (lattice dynamics) and mechanical stability. Despite this fact, there are still some simple aspects of MOFs which can be employed to get the new features of their polymeric composites which can bring exciting applications. <!--![endif]---->
<!--[endif]---->Due to their hybrid inorganic–organic nature, MOFs possess attractive characteristics that are absent from purely inorganic or organic systems, the organic parts allow them to interact chemically with polymeric chains to ensure chemical nucleation. Moreover, MOFs can exhibit, for example, higher surface areas while maintaining a degree of mechanical flexibility that is largely inaccessible to inorganic materials which can lead to enhanced nucleation of small polymer crystals.
To the best of our knowledge, this work is the first one to explore the MOF induced shape memory effect in polymers. It is found that HKUST-1 is acting as a nucleating agent to trigger the crystallization of PLLA for ensuring the entropy needed for the shape memory behavior; this results in recovery ratio, and fixity ratio to be about 100 %, and 95%, respectively with 5% MWt filler in elongation of 25%. Aspirin encapsulated in HKUST-1 reinforced PLLA is investigated to clarify the mechanism, and concluded that the supposed existing Metal−ligand coordinations are not the reason in fixing and releasing the temporary shape, instead having the transition of elastic to rubbery state in vicinity of cold crystallization temperature in semi-crystalline polymer provides HKUST-1 to act as a physical cross linking or netpoints. Moreover, the reinforcement accompanied by increased toughness using HKUST-1 offering a great combination of features which demonstrates the fact that how exciting features can be obtained with simply using a piece of reticular chemistry.

Piezoelectric materials (e.g. lead zirconate titanate or PZT and barium titanate) play a prominent role in the modern electroceramic industry as a result of their unique electromechanical coupling which render them ideal candidates for many applications such as sensors, actuators, ultrasound imagers and hydrophones. The incorporation of monolithic piezoelectric materials into a broader range of applications is often inhibited due to their limited inherent properties. As such, two different types of processes can be employed to selectively enhance certain characteristics and customize the piezoceramics to meet the design criteria of different applications. In the “additive” approach two or more constituents are added (as in piezoelectric composites) to enhance mechanical flexibility and piezoelectric response for optimized ultrasonic imaging. Conversely, in the “subtractive” approach, the introduction of controlled porosity (as in piezoelectric foams) in the matrix materials creates porous piezoelectrics with improved signal-to-noise ratio and impedance matching thus tailoring them for hydrophone devices and other related applications.

This presentation reviews the key results from the recent work in the field of computational modeling of novel piezoelectric foam structures and piezoelectric composite structures. Three-dimensional finite element models are developed to completely characterize the role of microstructural features such as foam shape, porosity shape, porosity aspect ratio, porosity distribution, and porosity volume fraction in determining the elastic, piezoelectric and dielectric properties of 3-0, 3-1, and 3-3 type piezoelectric foam structures. Unit-cell based finite element models (using python scripts) are also developed to systematically study the effects of fiber aspect ratio, fiber cross-sectional shape, fiber orientation, and fiber distribution on the electromechanical response of 3-0, 3-1, and 3-3 type piezoelectric composite (polymer [PVDF] – ceramic [BaTiO₃], ceramic [PZT-7A] -ceramic [BaTiO₃]) structures. The study on the effects of unit-cell boundary type on the electromechanical properties of piezoelectric composite structures will be reviewed.

To assess the suitability of these structures for such applications as hydrophones, bone implants, medical imaging, and diagnostic devices, five figures of merit are determined via the developed finite element model; the piezoelectric coupling constant (K_t), the acoustic impedance (Z), the piezoelectric charge coefficient (d_h), the hydrostatic voltage coefficient (g_h) and the hydrostatic figure of merit (d_hg_h). It is demonstrated that these figures of merit can be tailored and enhanced significantly by modifying the microstructural features in foam structures and piezoelectric composite structures to meet the criteria of different applications such as hydrophones, bone implants, medical imaging, and diagnostic devices.

Piezoelectric materials, with their unique electromechanical coupling characteristics, have found widespread use in many applications as sensors and actuators. With a continuing demand for macroscale piezoelectric devices with better performance characteristics, numerous efforts have been made to synthesize “bulk” monolithic and composite piezoelectric materials with better properties. More recently, within the context of microscale devices such as Smart MEMS, self-powered sensors, microbatteries, and energy harvesting devices, there is a growing interest in developing “thin-film” piezoelectric materials. While, advances in microfabrication technology have greatly helped the synthesis and fabrication of piezoelectric thin films and structures, there is a continuing need to develop small-scale test methods to characterize the properties of such thin film piezoelectric materials. While nano-indentation-based methods of property determination have been demonstrated to provide useful information about the mechanical properties – elastic, plastic, hardness, and fracture properties, of metallic materials in the thin film form, relatively fewer efforts have focused on developing the nanoindentation (or indentation) technique for understanding the electromechanical properties of thin-film piezoelectric materials.

The objectives of the present work are: (i) to obtain a comprehensive understanding of the indentation response of several classes of anisotropic piezoelectric materials; (ii) to elucidate the role of the indenter geometry and the indenter conductivity on the effective indentation response of anisotropic piezoelectric materials; (iii) to characterize the effects of electric fields on the indentation response of piezoelectric materials; (iv) to compare the indentation response of piezoelectric thin films with that of piezoelectric islands; (v) to differentiate between materials that are piezoelectrically active or passive, that is, poled and unpoled, and those that are piezoelectrically strong and weak; and (vi) to identify the principal poling directions in active materials.

Three-dimensional finite element models are developed to accurately capture the force–depth and charge–depth nanoindentation response of several classes of anisotropic piezoelectric materials such as relaxor ferroelectrics for which analytical models are at present unavailable. Upon validating the finite element model for transversely isotropic materials and with experimental results, it is demonstrated that the nanoindentation response of anisotropic piezoelectric materials displays a strong dependence on the nature of the indenter geometry and relatively weak dependence on the indenter conductivity. Furthermore, by recourse to “longitudinal” and “transverse” indentations, the nanoindentation method can also be used to identify the poling directions in piezoelectric materials as well. It is also demonstrated that the indentation of piezoelectric materials which are subjected to electric fields can be used to uniquely identify the piezoelectric characteristics of thin films that exhibit in-plane poling as well.

Piezoelectric transformers (PT) have been conventionally used in step-up high-voltage low-power applications such as LCD backlighting for portable electronic devices. Recently focus has shifted towards developing PTs for step-down applications such as power adapters for portable devices. In these high power applications, tuning the output impedance of the PT to meet the requirements of lower output voltages and higher output currents, implies the design of PTs with larger number of output layers. Further, these step-down applications require high efficiency operation of a PT under a variety of load conditions depending upon the load conditions. Thus, strategies to adapt the transformer to the variation of the load are sought. In this paper, a new tunable piezoelectric transformer (TPT) for AC-DC and DC-DC high power converters is presented. This TPT is designed to operate in the radial mode and is fabricated using cofiring multilayer process. By introducing control layers into the design, the TPT exhibits some unique features such as adjustable frequency response and flexible transfer ratio. Cofired multilayers with diameter of 48mm are successfully fabricated with platinum as inner electrode. Fabricated step-down TPT is tested for output power and efficiency. With control terminals open and short, the resonance frequency shift of the TPT can reach 3 kHz, which makes this TPT promising for working over wide frequency range. The experimental results show that TPT has a maximum output power of 75W with temperature rise of only 25 C. Maximum efficiency reached 97.3%, and the maximum voltage transformation is 0.45 (step-down).

Smart, functional composites based on suspensions of boron nitride nanotubes (BNNTs) embedded in a compliant matrix have potential to create a new class of mechanical, electrical and optical materials. Of particular interest is the ability of BNNTs to reinforce elastomers, in particular polydimethylsiloxane (PDMS), as well as their ability to induce novel properties including augmented thermal conductivity and piezoelectricity. This modified PDMS can be used as the basis for novel actuators, stimuli-responsive materials, and structures with controlled mechanical properties. Here we report mechanical, piezoelectric and thermal conductivity characterizations of a BNNT-PDMS composite prepared by a co-solvent dispersion process. Measurement of Young’s modulus via nano-indentation and dynamic mechanical analysis demonstrates an approximate doubling of Young’s modulus on the incorporation of just 6.3 weight percent of BNNT relative to PDMS. Similarly, incorporation of as little as 2 weight percent BNNT induces a clear piezoelectric response, as determined by measuring the displacement of samples subject to cyclic voltage using laser doppler vibrometry, in otherwise electrically inactive PDMS. Finally, time-domain thermoreflectance measurements of thin samples of BNNT-PDMS composite indicate a measurable increase in thermal conductivity with incorporation of BNNTs. Future exploration of alignment and texturing of BNNTs may allow for further augmentation of the above properties and allow for the development of advance soft, smart electroactive polymers for strain sensing and actuation or robust interface materials for thermal management and protection.

Flexible strain sensors have attracted attention in recent years for its applications in diverse field starting from soft robotics to human-machine interfaces and wearable health monitoring devices. We engineered a novel, lightweight, flexible, porous architecture composed of carbon nanotube (CNT) and silk which act as strain sensor. Such structures were fabricated using simple salt-leaching technique where the foam densities were varied from 0.06 to 0.14 gm cm^-3 by altering CNT and salt contents which eventually tunes the porosity of the structure up to 60%. The study showed that during compression, cell wall of the foams start moving towards each other reducing the gap between cells and form new conduction path and hence the resistance of the structure reduces along the compression direction with the increasing strain. At 50% compressive strain the normalized stress of CNT/silk-based foams structures were ranged from 30 to 60 MPa gm cm^-3. The relative change in resistance of the structures with increasing strain (up to 50% strain) follows a linear characteristic which provides an additional advantage for its application as piezoresistive strain sensor. The foam with highest porosity (60% porosity) and lowest electrical conductivity showed extremely high piezoresistive sensitivity with tuneable gauge factor ~ 250% at 50% strain. The study also revealed that with the increase in CNT content electrical conductivity increases whereas the gauge factor decreases. For compression over multiple cycles (> 200 cycles) the foam structures exhibited excellent stability and reversibility in terms of stress and resistance. The tuneable piezoresistive sensitivity and conductivity of the structures make them suitable candidate for the application as strain sensor in the field of soft robotics, flexible and wearable electronics etc.

Relaxor-ferroelectric multilayer heterostructure provides a high electric displacement or charge density, a large area to store the energy and fast discharge capacity. We have grown lead-free BaZr_0.2Ti_0.8O₃ (BZT)/ Ba_0.7Ca_0.3TiO₃ (BCT) multilayer heterostructures and studied the structural, dielectric, ferroelectric and energy density characteristics. The BZT/BCT multilayer epitaxial heterostructures were grown on La_0.67Sr_0.33MnO₃ (LSMO) buffered SrTiO₃ (STO) single crystal substrate by optimized pulsed laser deposition parameters such as substrate temperature, oxygen pressure and film thickness. The large angle x-ray scans showed only diffraction peaks from the substrate and pseudocubic reflections (00l) from the multilayer heterostructure, confirming that these films are phase pure and epitaxial in nature. The atomic force microscopy (AFM) studies indicate that the surface roughness is low and that film growth is of high quality. The low leakage current confirms the films are highly insulating in nature which is the basic requirement for the dielectric energy storage. The ferroelectric phase transitions have been proved above room temperature with relaxor behavior. The polarization versus electric field (P-E) measurement exhibits slim and well-saturated hysteresis loop with large polarization. The enhancement in polarization is due to a combination of space charge effect and interlayer charge coupling between two neighboring layers. Our lead-free relaxor-ferroelectric multilayer heterostructures exhibit good energy densities with better performance at low voltage above the room temperature which will be of great interest for researchers working on the physics and materials science.

Lead- free Potassium Sodium Niobate (K_xNa_1-xNbO₃ or KNN) has gained humungous popularity in the recent years due to increased environmental concerns and measures taken towards restraining the dominant use of toxic lead-based materials such as PZT in sensors and actuators for industrial applications. KNN, being eco-friendly, is considered to be a potential replacement to PZT due to its high Curie temperature ( 〉400^○C) providing broad temperature range for operation, and high piezoelectric and ferroelectric properties. As per the reports, a ferroelectric material exhibits the optimum electrical and piezoelectric properties at compositions especially near their morphotropic phase boundaries, where two ferroelectric phases are believed to co-exist. In K_xNa_1-xNbO₃, there exist three morphotropic phase boundaries (MPB) at x=0.17, x=0.35 and x=0.50 where the properties appear to be optimum. Although there are reports in literature on the KNN based ceramics, yet there is no report so far on the comparative study of the properties of K_xNa_1-xNbO₃ based thin films at composition near MPB. Attaining of the high dielectric and ferroelectric properties of KNN based thin films is significant for the realization of lead- free microwave-based devices for use in high frequency applications.
The present work investigates the electrical, dielectric and ferroelectric properties of K_xNa_1-xNbO₃ thin films near morphotropic phase boundaries [at compositions; KNN-35 (x= 0.35) and KNN-50 (x=0.50)]. The KNN thin films have been deposited using pulsed laser deposition (PLD) technique at optimized parameters. The dielectric properties and conduction behavior of the KNN-35 and KNN-50 thin films have been investigated and compared in a broad temperature range. The ferroelectric nature of the KNN-35 and KNN-50 thin films were confirmed by the obtained saturated Polarization- Electric field (P-E) hysteresis loops. The obtained results of the dielectric constant and remnant polarization for both KNN-35 and KNN-50 are promising for the integration of K_xNa_1-xNbO₃ thin films in MEMS based microwave tunable devices.

We report forming-free (FF) resistive random access memory (RRAM) with high endurance and retention for ZnO based thin films fabricated by dual ion beam sputtering (DIBS). We report the effect of interface anomalies such as disorder-induced interface states, Schottky barrier formation/dissolution for a resistive switch or memristor. Distribution of bulk defects with applied bias governs switching and also attributes to formation/dissolution of interfacial oxide. Further, conductance quantization is observed in the device.

Device shows excellent endurance and retention with forming-free behavior. Besides, conductive quantization as exhibited by device can be used to further work upon synaptic properties of device. The present work considerably contributes to further understand the conduction mechanisms of a wide range of resistive switches.

Atmospheric pressure and ambient temperature based micro-plasmas have been used in polarization and alignment of dipoles in ferroelectrics. The same phenomenon can be used to enhance the surface energy and surface characteristics of composite multifunctional thin films by means of surface modification. A significant increase in non-thermal atmospheric plasma applications, such as dielectric barrier discharge (DBD) and corona discharge plasmas have been steadily increasing in industry and in the research literature. The current work involves the use of corona micro-discharge for surface modification of ZnO/ZnO nanowire-Epoxy-Graphene based multi-functional and flexible thin film devices towards enhancement in surface bonding characteristics and variation in surface wettability and surface energy characteristics. The parameters being investigated in the experiments include plasma device characteristics such as voltage, current, and frequency, as well as other significant parameters such as displacement between thin film and DBD/Corona discharge source, treatment time, and temperature of the plasma. The modified surface micro-structure is analyzed using a scanning electron microscope, profilometry impedance tomography and the bulk electrical, dielectric, and piezoeelctric properties are characterized using impedance spectroscopy and piezoelectric strain coefficient measurements.

To face the pressing national challenges (health, social and economic) associated with obesity and overweight, significant preventive care efforts and educative programs are continuously being proposed and executed to reverse the trends of obesity prevalence. These efforts are often led by non-profit organizations, school administrator, university researchers, state agencies as well as federal agencies. For a long time the monitoring tool available to the population at large has been the low- cost weighing scale and only recently, as wearable electronics are becoming more widespread, new tools such as step and calorie counters are becoming available but remain expensive This work provides insights on how novel breath-gas sensors and skin wearable sensor nanotechnology may be used for monitoring the metabolic rate patterns of individuals in relation to their diet and exercise activities. It also addresses the need for human-computer interface systems that treat social network-based support groups as learning organizations. The latter can interface with the metabolic rate monitors to effectively promote positive health behavior. The synergy of the two technologies is expected to revolutionize personalized medicine and to radically improve health and well-being practices.

Diabetes is a chronic disease that effects a large part of the population in the United States and around the world. Real-time monitoring of blood glucose can detect early signs of diabetes by diagnosis of conditions such as impaired glucose tolerance. A wide range of analytical methods and techniques are being studied for non-invasive and real time testing/monitoring of blood glucose. These techniques include Optical Spectroscopy, Raman Spectroscopy, Photoacoustic Spectroscopy, Impedance Spectroscopy and Microwave based measurement methods among others. The following work deals with the analysis of a proof of concept hybrid monitoring system using a combination of optical and non-optical techniques such as Photoacoustic Spectroscopy and Impedance Spectroscopy. Data sets from the two methods are analyzed base on process parameters such as optical absorption, thermal expansion, acoustic velocity, specific heat, dielectric, and impedance characteristics. A system architecture based on multi-sensor and multivariable optimization is proposed. Further, to make it adaptive to different practical situations, AI or machine learning algorithms are to be investigated and implemented for reliability.

The impermeability of current hazmat suits restricts the thermoregulation of the user by blocking the water vapor, which can be fatal when used over relatively short period of time (~30 minutes) in absence of auxiliary cooling systems. Here, we develop a smart hazmat suit material based off stimuli responsive polymers that selectively swell upon action of hazardous chemical aerosols. Specifically, we develop a breathable, composite fabric that allows for perspiration cooling until they are exposed to a target range of hazardous chemicals. Once exposed, the polymer rapidly swells to close pores in the fabric turning into an impermeable membrane. Consequently, thermal stresses on the user are minimized by occurring only during actual exposure to hazardous materials. To achieve this, we develop a selectively swelling polymer, poly(N,N-butylphenylacrylamide),¹and integrate it with typical fabrics used in MOPP gear to create the adaptive, smart composite materials. We demonstrate the selective permeability of this smart composite fabric through a series of vapor transmission and droplet impingement experiments where we show that the fabric is permeable to water vapor but blocks droplets of target chemicals by swelling. We quantify the efficacy of this fabric in terms of response time and chemical containment. The response time largely depends upon the chemical permeation properties and polymer particle size/shape and varies inversely with it. Additionally, we develop a high-fidelity numerical model to capture the transient swelling dynamics and optimize the polymer shape and size. Specifically, we develop a finite element model to describe aerosol droplet-polymer interactions and polymer swelling kinematics. This model is based on implementation of concurrent fluid permeation and large deformation theory combined with constitutive relation based on Flory-Rehner theory.²In combination with the experimental data, we use this model to optimize the fabric design for maximum breathability while retaining the required protection levels for the user.
References:
1. Manning, K. C., Phadnis, A., Simonet, D., Burgin, T. P. & Rykaczewski, K. Development of a Nonelectrolytic Selectively Superabsorbent Polymer. Ind. Eng. Chem. Res.acs.iecr.8b02710 (2018).
2. Phadnis, A., Manning, K. C., Sanders, I., Burgin, T. P. & Rykaczewski, K. Droplet-train induced spatiotemporal swelling regimes in elastomers. Soft Matter14,5869–5877 (2018).

Biointerfaces of biological and engineered surfaces are relavant for many applications in biomedical, energy and structural applications. Tissue engineering is a promising technology for regenerating tissues and organs. The degrading scaffold in close molecular proximity of growing tissues representes a dynamic and intelligent interface that is controlled by genetic characteristics of cells as well as the physical and chemical characterictics of degrading scaffolds. We describe the use of a biomimetic mineralization process of mineralizing hydroxyapatite inside amino acid modified nanoclays for design of biomimetic and smart scaffolds for bone tissue engineering. The nanoclay scaffold enables growth proliferation and differenciation of human mesenchymal stem cells to generate tissue engineered bone with structural hierarchy characteristic of human bone, as well as the Ca/P stoichiometry characteristic of new remodeling bone. This type of bone is typically the niche to which prostate and breast cancer cells migrate to from their original location in a process called metastasis. At this point of bone metastasis, the prognosis is usually dismal for the patients and metastasis is the cause for most cancer deaths. We report the use of the smart scaffold to generate the first in vitro model of human prostate and breast cancer bone metastasis. The tumors at the metastasis site are investigated for gene expression that indicates that indeed the cancer is in late stage metastasis (also known as the mesenchymal to epithelial transition or MET). We report the use of FTIR and nanomechanics to capture disease progression at metastasis. The significance of the use of the scaffolds as cancer testbeds lies in the fact that human samples are unavailable and animal models fail, due to death of animal before metastasis to bone. Engineered biointerfaces of the smart scaffolds are thus the key to development of therapies advanced stages of metastasis.

Reducing microstructure dimensions down to the nanoscale can significantly improve the material strength. However, this is often accompanied by lower ductility, reduced fracture toughness and/or smaller fatigue crack growth resistance. There have been a number of successful strategies for achieving all-around superior mechanical properties, including the introduction of growth or deformation nanotwins, microstructure/nanostructure gradation, and various bioinspired strategies. Here a number of bioinspired nanostructure design strategies are proposed and discussed.
One particular strategy is inspired by the exceptionally strong and ductile structure of byssal threads found in certain mussels. The excellent mechanical properties are shown to be realized by structurally introducing sandwich structures at both the macro- and nano-scales.
On the other hand, nanoscale heterogeneity of bone has been identified to enhance inelastic energy dissipation. Experiments and simulations have been carried out to quantify the size-dependent heterogeneity, and understand the relative importance of elastic and inelastic heterogeneity. Parameters that affect energy dissipation are investigated under typical deformation modes of bone, and possible routes to enhance this effect are discussed.
Additional bioinspired microstructural design strategies are also discussed, including geometrically interlocking interface for enhanced strength as well as elastically mismatching surface layers for delayed catastrophic structure failure.

Numerous biological materials regenerate and self-heal in response to damage which allows them to strengthen areas under high mechanical stress and maintain their load-bearing capabilities for many decades. In bone, self-healing is enabled by the transport of nutrients and minerals through a vascular system embedded within a load-bearing cellular structure. In contrast, most human-made structural materials cannot autonomously heal or tune their mechanical strength. As a result, structures are designed with safety factors so that the strength of their constitutive materials is well above any stress they are expected to endure. Hence, not only are most human-made structures heavier and more voluminous than strictly necessary, but they also require regular monitoring and maintenance to guard against fatigue cracking, particularly in critical applications (aerospace vehicles and power plants, for example).

Self-healing materials have been developed to overcome the deficiencies of structural materials, but mostly rely on a “local healing” approach where the healing agent is available throughout the material so that cracks heal using material in the crack vicinity. Although successfully demonstrated in polymers, this method has proven impractical in metals without the external provision of heat, since metal atoms possess low diffusivities near room temperature.

Here, we overcome these limitations and report, for the first time, rapid and effective self-healing of metallic cellular materials at room temperature using electrochemistry as a method, and the structure of bones and their self-healing processes as an inspiration. In contrast to previous techniques, electrochemically-driven healing relies on the transport of metal as an ion through an electrolyte, which allows faster transport, as metal ion diffusivities in a room temperature electrolyte (~10^-9 m²/s) are at least three orders of magnitude greater than metal atom diffusivities in a room temperature solid. Consequently, our self-healing technique uses at least three orders of magnitude less energy to heal a crack of a given length compared to techniques which rely on heat-driven diffusion or precipitation in metallic alloys. We use tensile testing of dog-bone shaped open-cell nickel foams coated with a passivating layer, followed by electrochemical healing to understand the relationship between material structure and healing capabilities. Our nickel open-cell foams show full recovery of tensile strength and toughness using an external nickel source during healing. The use of a passivating polymer coating on the nickel foams improves the probability of achieving full recovery of tensile strength compared to uncoated nickel foams. Moreover, we show that by varying the maximum strain of the passivating coating, self-healing metals with three distinct functionalities can be achieved: materials that increase their local stiffness before failure, materials that heal microscale cracks and materials that only heal macroscale cracks above a threshold size. Finally, we design a nickel foam composite which uses an electrolyte embedded within an internal synthetic vascular system to redistribute load-bearing nickel to damaged areas and recover lost mechanical strength. Pathways for enabling a truly autonomous self-healing metallic material are discussed.

Cartilage is an important smart material which provides crucial mechanical properties for our body. Many diseases are associated with abnormal aggrecan degradation in articular cartilage. Aggrecan degradations are mainly controlled by matrix metalloproteinase-8 (MMP8). MMP8 cleaves at the catalytic cleavage site of Glu373-374Ala in the aggrecan core protein, with another potential cleavage site at Glu419-420Ala, however, left uncut. The catalytic mechanism of how the MMP8 recognizes the catalytic cleavage site has not yet been revealed. Understanding how nature design materials to be degraded at only specific regions can enable the design of new synthetic smart materials for many engineering applications.
To investigate this, we use a bottom-up computational mechanics approach to explore this conundrum. We found that the two key residues in the vicinity of the catalytic site, arginine in P2’ and glycine in P3’ play an important role in forming a stable binding pose of MMP8-Actual_peptide complex. For the potential cleavage site, the arginine is replaced with Threonine and the glycine is replaced with arginine, resulting in the unstable binding pose of MMP8-Potential_peptide complex. Our results suggest that MMP8 is able to recognize the molecular structure of the catalytic cleavage site and only cleave Glu373-374Ala in the aggrecan core protein. By calculating the binding affinity between MMP8 and aggrecan core protein, we find that the binding energy of MMP8-Actual_peptide complex is higher than the binding energy of MMP8-Potential_peptide complex. We hypothesize that the stable binding structure of the catalytic cleavage site of aggrecan core protein makes MMP8 stay in an “active” state, and then hydrolyze the scissile bond of aggrecan core protein. On the contrary, unstably binding between the potential cleavage site of aggrecan core protein and MMP8 makes MMP8 stay in an “inactive” state. Our results provide fundamental insights into the catalytic mechanism of biomaterials in cartilage at the molecular and nanoscale level.

Cartilage problems are becoming more frequent due some factors, such as population aging. Using biomimetics to understand how to minimize this problem, a biomaterial very sensitive to external stimuli has been studied as a possible substitute for worn cartilages: hydrogel, a biocompatible polymer. Thus, in order to create a basis on the feasibility of the application of this material, it is necessary to know different types of hydrogels, for instance, Poly (acrylic acid) and Polyethylene glycol, and to understand how they behave under external variations, such as pH and temperature. Accordingly, the work presented was based on recent bibliographical studies accessed mainly via Science Direct, Scientific Reports and Research Gate, for collecting fundamental information of this subject with the objective of producing visual simulation results via SolidWorks / Flow Simulation. Furthermore, the work will present objects composed of hydrogel printed in different formats by an experimental 4D printer developed at Federal University of ABC in collaboration with our research group 4DB. Moreover, it will be demonstrated how those printed objects behave in relation to external stimuli. It is hoped that the dissemination of the partial results of this ongoing research will contribute to the advancement of the subject in the interdisciplinary scientific community.

Precipitation hardening is one of the most efficient mechanisms to increase the yield strength of metallic alloys but accurate quantitative models to predict the size, shape and spatial distribution of precipitates during high temperature ageing are still lacking. In this presentation, a multiscale modelling strategy is presented to determine the precipitate stability as a function of temperature, the process of homogeneous and heterogenous nucleation and growth of precipitates. Precipitate stability is assessed from the calculation of the free formation energy obtained from first principles calculations of the formation enthalpy at 0K together plus the vibrational entropy contribution. Homogeneous and heterogeneous (on dislocations or other precipitates) precipitate nucleation is analysed by means of classical nucleation theory in which the different contributions to the Gibbs free energy (chemical interfacial and elastic) are determined from computational thermodynamics, first principles calculations and geometric considerations of the lattice distortion, respectively. Finally, precipitate growth up to the equilibrium shape was also determined by means of mesoscopic phase-field model.

The strategy was applied to analyse precipitation in Al-Cu alloys at high temperature, a regime in which three different kinds of precipitates can be found (θ'', θ' and θ). The results of the multiscale modelling strategy were in good agreement with the experimental data and it is envisaged that the strategy presented in this investigation can be used in the future to design optimum microstructures based on the information of the different energy contributions obtained from first principles calculations.

The fundamental understanding of the structure – property relationships of SMART materials at the atomistic level is a critical step towards rational design for targeted applications. On the experimental side, X-ray absorption spectroscopy (XAS) is a premier element-specific technique for materials characterization. Specifically, X-ray absorption near edge structure (XANES) carries rich local structural and chemical information around X-ray absorbing species, which makes it a powerful tool to probe materials structure and dynamic processes. As such, unraveling the local chemical environment from XAS spectra is akin to solving a challenging inverse problem. Because the structure - spectrum relationship is obscure, solving the inverse problem often requires prior empirical knowledge, which is qualitative or semi-quantitative and not transferable. Here we show how data science can be combined with first-principles theory to decipher the structure-spectrum relationship from XANES, with only a small amount of experimental data typically available for training. Such an approach can be applied in multiple contexts, including prediction of the 3D structures of metal nanoparticles using machine learning, local structure refinement of amorphous materials from a XANES database, and local chemical environment classification using XANES database and machine learning methods.

The influence of Co, Cu, Zn and Cd doping on Ni-Mn-Ga Heusler alloy exhibiting magnetic shape memory behavior is investigated using the first-principles exact muffin-tin orbital method in combination with the coherent-potential approximation. The control of material properties related to structural and magnetic phase transitions, especially martensitic transformation temperature T_M and Curie temperature T_C, is important for potential engineering applications in actuators, sensors, vibrational energy harvesters, or magnetic refrigeration systems. The energy difference between the austenite and the nonmodulated martensite phase ΔE_A-NM obtained from first-principles calculations can be used as a qualitative indicator of T_M whereas the energy difference between paramagnetic and ferromagnetic state ΔE_PM-FM serves for prediction of T_C. In addition, reducing of equilibrium tetragonality of nonmodulated martensite (c/a)_NM allows higher mobility of martensitic twins due to lower twinning stress, which further improve magnetic shape memory properties.
For Cu doping of a Ga-deficient alloy, we have found a strong increase in ΔE_A-NM, but just a small change of equilibrium (c/a)_NM. ΔE_PM-FM decreases in martensite phase and remains almost unchanged in austenite phase. If, on the other hand, Cu is instead doped at Mn sites, the increase of ΔE_A-NM is much smaller and a slight decrease of (c/a)_NM is seen. ΔE_PM-FM strongly decreases in both phases with increasing concentration of Cu. Co doping at Ni sites has much stronger effects, with increasing concentration of Co strongly decreasing both ΔE_A-NM and (c/a)_NM. These theoretical results are in good agreement with previous experimental findings that Cu doped in the Ga or Mn sublattice increases the martensitic transformation temperature T_M and Co in the Ni sublattice decreases it. Our results for Cu and Co doping also correspond to an empirical rule that T_M is correlated with the number of valence electrons per atom, e/a, in the alloy.
According to this rule Zn and Cd with more valence electron than Cu should provide stronger increase of T_M as well as ΔE_A-NM if these elements are used for doping instead of Ga or Mn. However, our results show that doping Zn or Cd in Ga-sublattice will results in the increase of T_M but weaker compare to Cu doping. T_C seems to be affected only weakly. The substitution of Mn atoms causes the decrease of both T_M and T_C. Thus, in contrast to Cu doping the calculation did not support the validity of the T_M ∼ e/a rule for Zn and Cd doping.
We have also investigated alloys where both elements Cu and Co were used in simultaneous doping. Our results show that the effects of simultaneous doping can be estimated using a linear superposition of the effects of individual dopants. This can be used for effective tuning of material properties; in particular, the decreasing of ΔE_A-NM (and T_M) arising from Co doping can be compensated by Cu doping at Ga sites. Moreover, the reduction of (c/a)_NM due to the Co doping is preserved because the Cu doping has just a small effect on this quantity.

Field effect transistors (FETs) with their Boolean operation have been the fundamental building block of digital computers. While this digital framework is excellent at performing complex arithmetic and logic calculations, it lags far behind the human brain in key areas such as adaptivity, generalization, and pattern recognition. Furthermore, as the size of these transistors is reduced, energy consumption and heat removal become very critical issues. Therefore, alternative technology initiatives such as analog switching are being explored. Neuromorphic computation using nanoscale adaptive oxide devices or memristors is a very promising alternative to this framework. Oxides of transition metals such as hafnium (HfO₂) are proven to be excellent candidate materials for these devices, which show non-volatile memory and analog switching, while using only two terminals, instead of three as in case of standard FETs. These devices are simple in terms of arrangement, wherein a few nanometers thick HfO₂ and an Hf metal layer are deposited between two electrodes on a substrate.
The naturally occurring HfO₂ has insulating properties. When it is applied with a voltage bias, Hf and O atoms dissociate, creating positively charged oxygen vacancies. While negatively charged oxygen ions migrate toward positively charged electrode and are temporarily stored in the adjacent pure Hf layer, a conducting filament of positively charged vacancies forms, which facilitates the flow of electrons through it. This is known as the on-state of the device or ‘set stage’. When the polarity reverses, oxygen ions move back into the oxide layers neutralizing the positively charged vacancies, thereby gradually creating insulating layers of HfO₂. As these layers are created, current flow through the device is drastically reduced and eventually an off-state is said to have occurred. This suppression of conducting filament is known as the reset stage. The extent of set/reset through applying alternate positive/negative voltage is temperature dependent transient process and is not completely characterized. Comprehensive understanding of coupled electrical current, ion/vacancy movement and heat transfer is essential to achieve the reliable and repeatable analog switching of this material.
This work focuses on the computational and experimental investigation of local and temporal variation of voltage, current and temperature, so that the coupled nature of heat transfer and current flow through these devices could be analyzed. By simulating the coupled heat transfer, current flow and vacancy/ion movement in COMSOL Multiphysics, the effect of substrate thickness and materials, set/reset voltages on relative on-state/off-state resistances, current magnitude and filament temperature is determined. Simultaneously, temperature of these devices is measured as a function of current and voltage using Microsanj transient temperature imaging (TTI) technique. The data from these experiments are used to develop useful information about the internal states of the devices. When the model results in terms of temperature and current history match well with those from the experiments, the COMSOL models are validated.
Overall, coupled electrical current and heat transfer modeling and simultaneous measurements of current, voltage and temperature are expected to provide insightful information on switching mechanisms of these adaptive oxides and warrant their scaled-up implementation for neuromorphic computing.

A novel concept of integrating a three dimensional (3D) strain scheme in epitaxial thin films^[1] is present in this work by combining 2-phase vertically aligned nanocomposite (VAN) with thin interlayers for a effective coupling of the vertical and the lateral interface strain. This 3D strain scheme takes advantages of both architectures and promotes various strain-driven physical properties, e.g., record high magnetoresistance values (MR%) as low field magnetoresistance properties have been demonstrated in the LSMO-CeO2 interlayered with different numbers of CeO2 interlayers. This work brings a new approach to achieve highly strained films beyond the critical thickness in epitaxy thin films and to demonstrate enhanced vertical strain coupling by the 3D strain scheme. This demonstration not only shows the power of 3D strain scheme in strain engineering, multifunctionality coupling, and flexibility in structural designs, but also fulfills the urgent demands of new material designs for future electronic devices.

[1] Sun, X.; Huang, J. J.; Jian, J.; Fan, M.; Wang, H.; Li, Q.; Mac Manus-Driscoll, J. L.; Lu, P.; Zhang, X. H.; Wang, H. Y., Three-dimensional strain engineering in epitaxial vertically aligned nanocomposite thin films with tunable magnetotransport properties. Mater. Horiz. 2018, 5 (3), 536-544.

Developing large area electrochromic smart windows based on nanoscale materials demands that trillions of nanoparticles modulate between transparent and colored states at the same rate. However, it is unclear how nanoparticle heterogeneity contributes to variable coloration dynamics. Here we demonstrate a single nanoparticle electrochromism approach to study optical modulation rates upon lithiation of isolated, clustered, and thin film tungsten oxide nanorod electrodes. We observe a particle-dependent waiting time for coloration (from 100 ms to 10 s) due to Li-ion insertion at optically inactive surface sites. Longer nanorods achieve higher OD modulation than shorter nanorods because they develop a Li-ion gradient that increases from the nanorod ends to the middle. Interestingly, electrochromic irreversibility increases monotonically with the number of particle-particle interactions due to ion trapping at nanoparticle interfaces. These findings lead us to propose a nanostructured electrode architecture that optimizes coloration magnitude, rate, and reversibility across large area elecrochromic smart windows.

The ability to reduce the size of antennae would enable a revolution in wearable and implantable devices. Multiferroic antennae, composed of individual ferromagnetic and piezoelectric phases, are posed to reduce antenna size by up to 5 orders of magnitude through the efficient coupling of magnetization and electric polarization via strain. However, this strategy requires a low-loss magnetic material with strong magnetoelastic coupling at high frequency.

Galfenol (Fe₈₄Ga₁₆ or FeGa) is a promising candidate material due to its large magnetostriction (>200 ppm), large piezomagnetic coefficient (>3 ppm/Oe), and high stiffness (>50 GPa), but it is highly lossy in the GHz regime. On the other hand, Permalloy (Ni₈₁Fe₁₉ or NiFe) is a soft magnetic material that has very low loss in the GHz regime (ferromagnetic resonance linewidth <20 Oe) but almost no magnetostriction. In this work, nanoscale laminates containing alternating layers of FeGa and NiFe were fabricated via DC magnetron sputtering to combine their complementary properties, yielding a small coercive field (<20 Oe), narrow FMR linewidth (<40 Oe), and high relative permeability (>700) (Rementer et al., 2017). These magnetic laminates were then grown on PMN-PT substrates and studied via polarized neutron reflectometry, demonstrating coherent rotation of the individual layers’ magnetization with an applied electric field, supported by micromagnetic and finite element simulations (Jamer et al, 2018).

In addition, optical magnetostriction measurements confirmed the presence of greatly enhanced magnetostriction relative to single-phase FeGa; these laminates represent a threefold increase in magnetostriction at saturation (~700 ppm) and an enhanced sensitivity at low bias magnetic fields (25 ppm/Oe). This enhancement in magnetoelasticity relative to single-phase FeGa was correlated to the microstructure of these composites using TEM. Recent efforts have further enhanced the high-frequency properties of these composites through insertion of ultrathin Al₂O₃ layers to reduce the conductivity and mitigate eddy current losses, and subsequent integration of these laminates into a strain-mediated multiferroic shear wave antennae successfully demonstrated the great potential of FeGa/NiFe laminates for use in microscale communications systems.

References:
1. C. R. Rementer, K. Fitzell, Q. Xu, et al., Applied Physics Letters, Vol. 110 (24), 242403 (2017)
2. M. E. Jamer, C. R. Rementer, A. Barra, A. J. Grutter, K. Fitzell, et al., Physical Review Applied, Vol. 10 (4), 044045 (2018)

Composite materials are engineered materials made from multiple different components. Each component has unique physical properties that combined together create complex structures for high-performance applications. Carbon fiber composites combine the strength and light weight of carbon fibers with a polymer resin. Composites are becoming more important in applications such as civil infrastructures, aircraft structures, prosthetic limb manufacturing.
Structural Health Monitoring (SHM) aims to make a structure self-sensing by automatically diagnosing its health at all times. There are many methods of SHM such as using imaging and pattern recognition techniques to find cracks or deformities, using traveling waves traversing a structure, and sensors that are aimed to monitor one type of damage. These techniques are costly and offer limited spatial resolution.
Composites enable a novel approach where SHM sensing is directly integrated into composite material. Approaches include piezoelectric wafers for aircraft SHM; Carbon Nano-Tube (CNT) threads embedded into steel structures or braided with fabric fibers to create smart fabrics; carbon fiber yarns in glass composites for windmill blades. These reports have been limited to discrete embedded sensors with limited spatial extent and resolution.
This project exploits the electrical conductivity of conventional woven carbon fibers to create a 2D strain sensor network for SHM. Carbon fiber is a material with high specific strength, electrically conductive, widely used in composites manufacturing, and it is considerably cheaper than CNTs. Thus, designing the sensor using carbon fiber allows for wide area sensing on surfaces or even embedded in the composite as one of the layers. The carbon fiber sheet provides both mechanical strength and electrical sensing, thereby creating a multifunctional smart self-sensing device.
The conductivity of carbon fiber was characterized through multiple experiments, starting with single fibers processing to yarns of various fiber quantity. Fiber filament diameter and resistivity were measured. Using the extracted carbon fiber resistivity, the number of fibers in a yarn was estimated and compared to manufacturer specification. In SHM, this method can be used to estimate the number of broken fibers in a bundle. The measured resistivity of a bundle was 23.67Ωμm, which is comparable to state of the art SHM.
Understanding the carbon fiber characteristics, a woven 2D carbon fiber laminate was modeled with an equivalent circuit model. Multiple laminates with different weave orientations were utilized. It was determined that, since the laminate is woven with orthogonal bundles crossing one another, it did not behave like an isotropic thin-film sheet resistor. Rather, the current follows the easiest path through the weave yarns. Thus, changing the orientation of the weave changes the current path making the sheet an anisotropic 2D resistor. Moreover, the size and contact point of probes and the spacing between probes changes resistance in a non-linear fashion. Various probing techniques and sheet orientations will be presented and compared to achieve the highest possible resolution of strain sensing for various applications.
Finally, this work will present the electrical measurement of laminates in a state of mechanical strain. Resin will be introduced to the laminate which will alter its mechanical properties by providing structure and making the carbon fiber more durable. Laminate configurations will be compared to reach an optimal smart material that can be easily deployed in current applications of carbon fiber laminate.
In summary, a novel approach to smart materials using 2D carbon fiber laminate strain sensing is presented. This method can be applied to advanced composite structures. This approach provides wide area sensing with relatively inexpensive material and little-added fabrication complexity.

Nanoscale particles are widely used in various applications like drug delivery, wastewater purification and thermal management areas. However, its inability to manage their alignment during processing limits its widespread applications. This technique will focus on the printing of nanoparticles-containing composites via additive manufacturing process with the promoted local alignment of nanoparticles assisted by the surface morphology and fluidic flows. Specifically, nanoparticle orientation will be controlled within a unit printing volume to obtain suitable mechanical and functional properties. Projection micro-stereo lithography (PµSLA)/fine fiber spinning based fused deposition modeling (FFS-FDM) were used for building for 3D triangular wave shape surface pattern of PVA (poly vinyl alcohol) transparent polymer. The liquid crustal 4-Cyano-4'-pentylbiphenyl (5CB) and halloysite nanoparticles are diffused together and this mixture is deposited over the transparent pattern to study its alignment. 5CB liquid crystal provides sufficient driving force for the unidirectional alignment of halloysite nanoparticles. The factors like fluid dynamics, colloidal inks, solidification thermodynamics and confinement systems will control the nanoparticle configurations. The orientation of nanoparticles due to the alignment of nematic liquid crystal in the groove direction were studied by polarized optical microscopy (POM), atomic force microscopy (AFM), x-ray diffraction (XRD), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Mechanical properties like static modulus, strength and toughness were measured. The orientation management of nanoparticles will improve the efficiency of drug delivery, optical or electrical signal transfers and the robustness of automobiles etc.

Controlled self-propelling of liquid transport is of significant interest for extensive applications, such as DNA microarrays, fog-harvesting, self-cleaning coatings and smart microfluidic devices. It is found that many biological structures possess unique structural features that demonstrate unidirectional wetting behavior, such as cactus with conical spines, spider silks with periodic spindle-knots and joints, and Nepenthes peristome with overlapping grooves. Investigations on the structure-property-function relationships of smart surfaces in Nature offer inspirations and insights into how to control the tendency of liquid self-transport without additional energy expenditure. In this work, we first applied the atmospheric pressure plasma (APP) treatment on inclined samples with a specific angle which resulted in gradient wettability and unidirectional wetting behavior was realized. To further improve the efficiency and durability, we further designed bio-inspired patterns consisting of arrays of asymmetric, spine-like structures and fabricated the negative molds by 3D-printing. Polydimethysiloxane (PDMS) was utilized to replicate the bio-inspired patterns. Then, hydrophilic polyethylene glycol (PEG) was grafted onto the surface of PDMS replica in a two-step process, including APP treatment and liquid phase deposition. The chemical composition was analyzed by X-ray photoelectron spectroscopy (XPS) and profile and microstructure of PDMS replica were characterized by 3D laser scanning confocal microscopy and SEM, respectively. The static and dynamic wetting behavior of liquids on biomimetic patterned surfaces were evaluate and observed by contact angle measurement with varying tilting angles and high-speed camera. Results showed that the serial liquid droplets move unidirectionally driven by the asymmetric structure and gradient surface and were pinned in the opposite direction. The unidirectional wetting behavior of liquids was observed even when the titling angle reached 90°. In summary, we successfully designed and synthesized biomimetic patterned surfaces by combining 3D printing-assisted replication, APP treatment and PEG grafting which possess effective unidirectional wettability and could be further applied in controlled fluid transportation, water/fog collection, microfluidic devices and smart textiles.

Chemical solution deposition of lead chalcogenide thin films on GaAs substrates has been shown to result in a wide range of film microstructures, from ultra-thin nanocrystalline films, via highly oriented polycrystalline films, to single crystal films with epitaxial relation with the underlying substrate.[1-6] Of particular interest are columnar thin films, which are comprised of nano-columns which are vertically aligned with respect to the substrate. This talk will highlight the solution deposition route to columnar thin films, which are interestingly formed via an oriented attachment growth mechanism.[7] These films are candidate absorber layers for nanomaterials based, solution deposited up-conversion night vision devices operating in the short wave infrared (SWIR) range.[8]


References

1. M. Shandalov and Y. Golan, Eur. Phys. J. Appl. Phys 24, (2003), 13.
2. A.Osherov, V. Ezersky and Y. Golan, Eur. Phys. J. Appl. Phys. 37 (2007) 39.
3. A.Osherov, M. Shandalov, V. Ezersky and Y. Golan, J. Crystal. Growth 304 (2007) 169.
4. A.Osherov, V. Ezersky and Y. Golan, J. Crystal Growth 308 (2007) 334.
5. J.J.H. Pijpers, R. Ulbricht, K.J. Tielrooij, A. Osherov, Y. Golan, C. Delerue, G. Allan and M. Bonn, Nature Physics 5 (2009) 811.
6. A. Osherov and Y. Golan, MRS Bulletin 35 (2010) 790.
7. T. Templeman, S. Sengupta, N. Maman, E. Bar-Or, M. Shandalov, V. Ezersky, E. Yahel, G. Sarusi, I. Visoly-Fisher and Y. Golan, Crystal Growth and Design 18 (2018) 1227.
8. G. Sarusi ; T. Templeman ; E. Hechster ; N. Nissim ; V. Vitenberg ; N. Maman ; A. Tal ; A. Solodar ; G. Makov ; I. Abdulhalim ; I. Visoly-Fisher ; Y. Golan, Proc. SPIE 9884, Nanophotonics VI, (2016) 98840L.

The elastocaloric cooling effect of natural rubbers has been demonstrated for cold storage, however, its programming by cold-drawing is possible only at high strain rate. We found that shape memory cross-linked poly(cyclooctene) (PCO) exhibits enhanced elastocaloric cooling effect at relatively moderate strain rate due to its strain-induced crystallization behavior during cold-drawing [1]. The elastocaloric heating and cooling effects of PCO were predicted using Green-Lagrangian strains obtained from 3D finite element analysis (FEA) with a suitable constitutive model, which was already developed for shape memory polymers in our group [2], and thermodynamic quantities, showing good agreements between simulation and experiments. In addition, carbon fiber reinforced PCO composites were also studied to investigate the effect of carbon fibers on the elastocaloric effects of PCO. For the modeling of PCO composites, anisotropic hyperelasticity theorem was used for modeling the mechanical behavior of carbon fabrics. The interaction between fiber and SMP matrix depending on temperature change was considered through thermal residual stress [3]. Finally, an application of shape memory PCO and its composite to the elastocaloric refrigerator will be studied to demonstrate that implementation of the current method into 3D FEA can facilitate the design study of a portable cooling device made of shape memory polymers.

[1] S B Hong et al., Characterization and modeling of elastocaloric effects of shape memory poly(cyclooctene), Applied Physics Letter, accepted, 2018.
[2] H Park et al., Three-dimensional constitutive model for shape memory polymers using multiplicative decomposition of the deformation gradient and shape memory strains, Mechanics of Materials, 2016, Vol 93, pp 43-62.
[3] S B Hong et al., Three-dimensional constitutive model of woven fabric-reinforced shape memory polymer composites considering thermal residual stress, Smart Materials and Structures, submitted, 2018.

Electronic (e-) skin has recently been widely pursued as state-of-the-art wearable electronics due to its skin-mimetic mechanical and sensing properties. Similar to human skin, which is stretchable, tough, and even self-healable with sensing capabilities, efforts to render such e-skin devices robust and durable to withstand constant mechanical damage have led to various reports of self-healing elastomers. Even though elastomers utilized in e-skin devices are stretchable, but unfortunately, they easily break along locations where damages incurred due to their low fracture energies of ~100 J/m² or lower. Furthermore, these damages are usually not self-healable¹⁰. Most self-healing materials are based on weak polymer systems with low fracture energies. Towards tough, self-healable and stretchable electronics for practical e-skin devices development, we report herein a polymer film cross-linked through multi-strength hydrogen bonding engineering. We further illustrate its application through the fabrication of highly stretchable and tough self-healing e-skin devices. Briefly, the underlying principle for our tough and self-healable film is achieved through the spontaneous formation of a mixture of both strong and weak cross-linking hydrogen bonds. The strong cross-linking bonds confer robustness and elasticity, while the weak bonds are able to dissipate energy through efficient reversible bonds breakage and reformation. Furthermore, the polymer film is autonomously self-healable and highly moldable into desired complex 3D structures. Our film is easily hybridized with conductive fillers, in which we proceed to fabricate a new generation of e-skin: highly tough and stretchable self-healing system.

Desirable phosphors for light emitting diodes (LED) must have good light absorption property, high concentration quenching, high quantum efficiency, and narrow color emission, etc. In this work, we first show that undoped yttrium hafnate Y₂Hf₂O₇ (YHO) nanoparticles (NPs) display dual blue and red bands after 330 nm light excitation. Based on Density functional theory (DFT) calculations, these two emission bands are correlated with the defect states arising in the band-gap region of YHO due to the presence of neutral and charged oxygen defects. Once doped with Eu³⁺ ions, the YHO NPs show bright red emission, long excited state lifetime and stable color coordinates upon near-UV and X-ray excitations. Concentration quenching is active when Eu³⁺ doping reaches 10mol% with a critical distance of ~4.43 Å. This phenomenon indicates a high Eu³⁺ solubility within the YHO host and the absence of Eu³⁺ clusters. The optical performance of the YHOE NPs has been further improved by lithium co-doping. Also, the YHOE NPs have been dispersed into PVA polymer to make nanocomposite films, which show strong red emission under excitations at 270 and 393 nm. Overall, high emission intensity and quantum efficiency under UV and X-ray excitations make the YHOE NPs suitable as phosphors, scintillators and luminescent probes for LED and life science applications.

Architectures of natural organisms especially plants largely determine their response to varying external conditions. Nature inspired shape transformation of artificial materials has motivated academic research for decades due to wide applications in smart textiles, actuators, soft robotics, and drug delivery. A “self-growth” method of controlling femtosecond laser scanning on the surface of a prestretched shape-memory polymer to realize microscale localized reconfigurable architectures transformation is introduced. It is discovered that microstructures can grow out of the original surface by intentional control of localized laser heating and ablation, and resultant structures can be further tuned by adopting an asymmetric laser scanning strategy. A distinguished paradigm of reconfigurable architectures is demonstrated by combining the flexible and programmable laser technique with a smart shape-memory polymer. Proof-of-concept experiments are performed respectively in information encryption/decryption, and microtarget capturing/release. The findings reveal new capacities of architectures with smart surfaces in various interdisciplinary fields including anti-counterfeiting, microstructure printing, and ultrasensitive detection.

The rapid expansion of sensors as a research field is continuously bringing to life novel materials, detection mechanisms and practical applications. During the last few decades a vast variety of sensors have been successfully developed at the laboratory scale demonstrating an enormous potential for solving global issues such as food and environmental pollution. Unfortunately, the fact that nearly 80% of the platforms developed at lab-scale never get to meet the market is not yet a common discussion within the field.

Challenges standing in the way of bringing sensors from the laboratory to the market include difficulties in reproducibility and repeatability. We used roll-to-roll processing to design field deployable flexible electrochemical sensors for the detection of environmental pollutants. Inkjet printed gold electrodes on flexible polyimide substrates were modified with a layer of specificity for organophosphorus pesticides and for mercury, respectively. We designed both platforms to transduce the target interaction signal by impedimetric (transfer charge resistance) response. For the pesticide sensor, a mesh of zirconium oxide fibers was deposited via roll-to-roll on the electrodes, leading to a detection of methyl parathion with a limit of detection (LOD) of 0.01 µM (r²= 0.952). The specificity was demonstrated vs nitrobenzene and 4-nitrophenol. Mercury ions were captured by a roll-to-roll deposited, thiol-functionalized aptameric sequence, highly specific for the target. The LOD for the Hg⁺ was of 0.1 ppm (r²= 0.995) and the possibility to be re-used without significant variation was demonstrated. The platform specificity was assessed vs arsenic, cadmium and lead as potential interferent ions.

The results presented by this work reveal not only the high sensibility and specificity of both capture strategies (zirconium oxide fibers and ss-DNA aptamer) but high stability, and repeatability in sensors response. We suggest it is now possible to incorporate roll-to-roll technology in sensor design, and take the high variety of already developed lab-scale devices to large-scale manufacturing.

The need for light-weight, high-strength, and easily machineable insect-repellant fabrics is of critical importance to the cessation of viral diseases, such as Zika, Malaria, and Dengue Fever due to climate change.

In this study, the viability of electrospinning nanofiber Nylon-6 for use in protective garments will be investigated and compared to conventional polyamide fabrics, Nylon-6 Tricot and Nylon-6,6 Knit.

Throughout this investigation, the surface coating efficiency was evaluated to determine the best method of uptake for Permethrin, as well as characterization of fabrics utilizing DSC, TGA, FTIR, FESEM, XPS, GC-MS, Porometry, Tensile Testing, Rheometry, UV degradation, and washing fastness.

Acknowledgement: This work made use of the Cornell Center for Materials Research Facilities supported by the National Science Foundation under Award Number DMR-1719875.