Search

Humans rely increasingly on sensors to address grand challenges and to improve quality of life in the era of digitalization and big data. For ubiquitous sensing, flexible sensors are developed to overcome the limitations of conventional rigid counterparts. Despite rapid advancement in bench-side research over the last decade, the market adoption of flexible sensors remains limited. To ease and to expedite their deployment, here, we identify bottlenecks hindering the maturation of flexible sensors and propose promising solutions. We first analyze challenges in achieving satisfactory sensing performance for real-world applications and then summarize issues in compatible sensor-biology interfaces, followed by brief discussions on powering and connecting sensor networks. Issues en route to commercialization and for sustainable growth of the sector are also analyzed, highlighting environmental concerns and emphasizing nontechnical issues such as business, regulatory, and ethical considerations. Additionally, we look at future intelligent flexible sensors. In proposing a comprehensive roadmap, we hope to steer research efforts towards common goals and to guide coordinated development strategies from disparate communities. Through such collaborative efforts, scientific breakthroughs can be made sooner and capitalized for the betterment of humanity. Agency for Science, Technology and Research (A*STAR) National Research Foundation (NRF) Submitted/Accepted version Y.L., Z.L., M.Z., and X.C. acknowledge the National Research Foundation, Singapore (NRF) under NRF’s Medium Sized Centre: Singapore Hybrid-Integrated Next-Generation μElectronics (SHINE) Centre funding programme, and AME programming funding scheme of Cyber Physiochemical Interface (CPI) project (no. A18A1b0045). Y.L. acknowledges National Natural Science Foundation of China (62201243). C.J. acknowledges funding support from the National Key R&D Program of China (no. 2019YFA0706100), the National Natural Science Foundation of China (82151305), Lingang Laboratory (LG-QS-202202-09). T.Q.T. and N.E.L. acknowledge support by the Basic Science Research Program (no. 2020R1A2C3013480) through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT. A.F. acknowledges the AFOSR (grant FA9550-22-1-0423). Y.L. and Y.Z. would like to acknowledge the NSF (award no. 2134664) and NIH (award no. R01HD108473) for financial support. X.F. acknowledges the support from the National Natural Science Foundation of China (grant no. U20A6001). L.Y. would like to thank the A*STAR Central Research Fund (CRF) and the AME Programmatic A18A1b0045 (Cyber Physiochemical Interfaces) for funding support. C.F.G. acknowledges the National Natural Science Foundation of China (no. T2225017). T.Q.T. acknowledges the Brain Pool Program (No. 2020H1D3A2A02111068) through the National Research Foundation (NRF) funded by the Ministry of Science. Z.L. acknowledges the support from RIE2020 AME Programmatic Grant funded by A*STAR-SERC, Singapore (Grant No. A18A1b0045). X.G. acknowledges funding support through the Shanghai Science and Technology Commission (grant no. 19JC1412400), the National Science Fund for Excellent Young Scholars (grant no. 61922057). C.D. acknowledges National Science Foundation CAREER: Conformable Piezoelectrics for Soft Tissue Imaging (grant no. 2044688) and MIT Media Lab Consortium funding. D.K. and O.G.S. acknowledge Leibniz Association and the German Research Foundation DFG (Gottfried Wilhelm Leibniz Program SCHM 1298/22-1, KA5051/1-1 and KA 5051/3-1), as well as the Leibniz association (Leibniz Transfer Program T62/2019). C.W. acknowledges the National Key Research and Development Program of China (grant no. 2021YFA1202600), National Natural Science Foundation of China (grant no. 62174082). A.V.-Y.T., E.Z., Y.Z., X.Z., and J.P. acknowledge the National Research Foundation, Singapore (NRF) under NRF’s Medium Sized Centre: Singapore Hybrid-Integrated Next-Generation μElectronics (SHINE) Centre funding programme, and AME programming funding scheme of Cyber Physiochemical Interface (CPI) project (no. A18A1b0045). R.Z. acknowledges National Natural Science Foundation of China (grant no. 51735007) and Beijing Natural Science Foundation (grant no. 3191001). N.M. acknowledges the support by JST PRESTO Grant Number JPMJPR20B7 and JST Adaptable and Seamless Technology transfer Program through Target-driven R&D (ASTEP) grant number JPMJTM22BK. C.P. acknowledges the Korean government (Ministry of Science and ICT, MSIT) (2022R1A4A3032923). M.W. acknowledges the National Key R&D Program of China under Grant (2021YFB3601200). X.Z. acknowledges National Natural Science Foundation of China (no. 62074029). S.X. acknowledges the 3M nontenured faculty award. T.-W.L. and D.-G.S. acknowledge the Pioneer Research Center Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT & Future Planning (grant no. NRF-2022M3C1A3081211). C.T.L. would like to acknowledge support from the Institute for Health Innovation and Technology (iHealthtech), the MechanoBioEngineering Laboratory at the Department of Biomedical Engineering and the Institute for Functional Intelligent Materials (I-FIM) at the National University of Singapore (NUS). C.T.L. also acknowledges support from the National Research Foundation and A*STAR, under its RIE2020 Industry Alignment Fund − Industry Collaboration Projects (IAF-ICP) (grant no. I2001E0059) − SIA-NUS Digital Aviation Corp Lab and the NUS ARTIC Research (grant no. HFM-RP1). X.Y. acknowledges funding support by City University of Hong Kong (grant no. 9667221). T.X. and X.Z. acknowledge National Natural Science Foundation of China (22234006). B.C.K.T. acknowledges Cyber-Physiochemical Interfaces CPI, A*STAR A18A1b0045. H.G. acknowledges a research start-up grant (002479-00001) from Nanyang Technological University and the Agency for Science, Technology and Research (A*STAR) in Singapore. W.G. acknowledges National Science Foundation grant 2145802. D.J.L. acknowledges support from the US National Science Foundation grant number CBET-2223566. G.Y. acknowledges support from The Welch Foundation award F-1861, and Camille Dreyfus Teacher-Scholar Award. M.D.D. acknowledges funding support from NSF (grant no. EEC1160483). J.-H.A acknowledges the National Research Foundation of Korea (NRF-2015R1A3A2066337). J.C. acknowledges the Henry Samueli School of Engineering & Applied Science and the Department of Bioengineering at the University of California, Los Angeles for startup support and a Brain & Behavior Research Foundation Young Investigator Grant. K.T. acknowledges JST AIP Accelerated Program (no. JPMJCR21U1) and JSPS KAKENHI (grant no. JP22H00594). P.S.W. acknowledges the National Science Foundation (CMMI1636136) for support. A.M.A., M.C.H., and P.S.W. thank the National Institute on Drug Abuse (DA045550) for support. S.M. and X.C. appreciated the support from the Smart Grippers for Soft Robotics (SGSR) Programme under the National Research Foundation, Prime Minister’s Office, Singapore under its Campus of Research Excellence and Technological Enterprise (CREATE) programme.

The cone photoreceptors in our eyes selectively transduce the natural light into spiking representations, which endows the brain with high energy-efficiency color vision. However, the cone-like device with color-selectivity and spike-encoding capability remains challenging. Here, we propose a vertical integrated spiking cone photoreceptor (VISCP) array with the structure of ITO/Ta2O5/Ag/IGZO/ITO, which can directly transduce persistent lights into spike trains at a certain rate according to the input wavelengths. The VISCPs have an ultralow power consumption of ≤400 pW per spike in visible light, which is very close to biological cones. In this work, lights with three wavelengths were exploited as pseudo three primary colors to form ‘colorful’ images for recognition tasks, and the device with the ability to discriminate mixed colors shows a better accuracy. Our results would enable hardware spiking neural networks with biologically plausible visual perception and provide great potential for the development of dynamic vision sensors.

Polarity often refers to the charge carrier type of a semiconductor or the charging state of a functional group, generally dominating their functionality and performance. Herein we uncover a spontaneous and stochastic polarity-flipping phenomenon in monolayer WSe2, which randomly switches between the n-type and p-type states and is essentially triggered by fluctuating carrier flows from or to the adjacent WS2 monolayer. We have traced such fluctuating carrier flows by interfacial photocurrent measurements in a zero-bias two-terminal device. Such polarity flipping results in switching between the negative and positive correlations between the emission intensities of WS2 and WSe2 in the heterobilayer, which is further well-controlled by the electrostatic gate-tuning experiments in a capacitor-structure device. Our work not only demonstrates giant and intermittent carrier flows through long-range coupling in 2D heterostructures and a consequent spontaneous polarity flipping phenomenon but also provides a two-emitter system with a switchable correlation sign that could project future applications in optical logic devices.

Animals execute intelligent and efficient interactions with their surroundings through neural pathways, exhibiting learning, memory, and cognition. Artificial autonomous devices that generate self-optimizing feedback mimicking biological systems are essential in pursuing future intelligent robots. Here, we report an artificial neural pathway (ANP) based on a memristor synapse to emulate neuromorphic learning behaviors. In our ANP, optical stimulations are detected and converted into electrical signals through a flexible perovskite photoreceptor. The acquired electrical signals are further processed in a zeolitic imidazolate frameworks-8 (ZIF-8)-based memristor device. By controlling the growth of the ZIF-8 nanoparticles, the conductance of the memristor can be finely modulated with electrical stimulations to mimic the modulation of synaptic plasticity. The device is employed in the ANP to implement synaptic functions of learning and memory. Subsequently, the synaptic feedbacks are used to direct a robotic arm to perform responding motions. Upon repeatedly "reviewing" the optical stimulation, the ANP is able to learn, memorize, and complete the specific motions. This work provides a promising strategy toward the design of intelligent autonomous devices and bioinspired robots through memristor-based systems. Agency for Science, Technology and Research (A*STAR) National Research Foundation (NRF) Submitted/Accepted version Financial support was provided by the Agency for Science, Technology and Research (A*STAR) under its AME Programmatic Funding Scheme (Project #A18A1b0045) and the National Research Foundation, Singapore (NRF) under NRF’s Medium Sized Centre: Singapore Hybrid-Integrated Next-Generation μ-Electronics (SHINE) Centre funding programme.

True random number generators (TRNGs) can generate unpredictable binary bitstream by exploiting the intrinsic stochasticity in physical variables. In a threshold switching memristor, the stochastic forming/rupture of conducting pathway has been proved to be a good random source, while further improvement of high randomness and throughput is still a challenge. Here, a crossbar array of amorphous indium–gallium–zinc–oxide (a-IGZO)-based threshold switching memristors was designed for high-throughput TRNGs. The intrinsic stochasticity of Ag conductive filament in IGZO memristor and the stochastic sneak paths in the crossbar array are the two sources of randomness in our TRNGs. In our design, one input pulse train can produce multi-channel random bits, which enables a high scalability for such TRNGs. In addition, the average energy consumption of the TRNGs can be further reduced by increasing the integration scale of the memristors. Such IGZO-based TRNGs are of great significance for security applications.

Physical reservoir computing (PRC) is thought to be a potential low training-cost temporal processing platform, which has been explored by the nonlinear and volatile dynamics of materials. An electric-double-layer (EDL) formed at the interface between a semiconductor and an electrolyte provided a great potential for building high energy-efficiency PRC. In this Letter, EDL coupled indium-gallium-zinc-oxide (IGZO) artificial synapses are used to implement reservoir computing (RC). Rich reservoir states can be obtained based the ionic relaxation-based time multiplexing mask process. Such an IGZO-based RC device exhibits nonlinearity, fade memory properties, and a low average power of ∼9.3 nW, well matching the requirement of a high energy-efficiency RC system. Recognition of handwritten digit and spoken-digit signals is simulated with an energy consumption per reservoir state of ∼1.9 nJ, and maximum accuracy of 90.86% and 100% can be achieved, respectively. Our results show a great potential of exploiting such EDL coupling for realizing a physical reservoir that would underlie a next-generation machine learning platform with a lightweight hardware structure.

The modern internet-of-things era has witnessed an increasing growth in the demand for advanced sensors to collect precise information. To meet this demand, extensive efforts have been devoted to exploring competent materials and designing rational architectures for the fabrication of sensing devices. Graphdiyne represents a promising material due to the attractive electronic, optical and electrochemical properties deriving from its unique molecular structure. In this review, we firstly provide the points of view on the architectures and work principles of the graphdiyne-based sensing devices with respect to resistive, electrochemical, photoelectrochemical and fluorescent categories. Secondly, we present the promising applications on biochemical sensing, such as the detection of DNA, micro-RNA, and glucose. Finally, the challenges and prospects of graphdiyne-based biochemical sensing platforms are also discussed, in order to provide a cornerstone for understanding this rapidly developing area.

Coupling myoelectric and mechanical signals during voluntary muscle contraction is paramount in human–machine interactions. Spatiotemporal differences in the two signals intrinsically arise from the muscular excitation–contraction process; however, current methods fail to deliver local electromechanical coupling of the process. Here we present the locally coupled electromechanical interface based on a quadra-layered ionotronic hybrid (named as CoupOn) that mimics the transmembrane cytoadhesion architecture. CoupOn simultaneously monitors mechanical strains with a gauge factor of ~34 and surface electromyogram with a signal-to-noise ratio of 32.2 dB. The resolved excitation–contraction signatures of forearm flexor muscles can recognize flexions of different fingers, hand grips of varying strength, and nervous and metabolic muscle fatigue. The orthogonal correlation of hand grip strength with speed is further exploited to manipulate robotic hands for recapitulating corresponding gesture dynamics. It can be envisioned that such locally coupled electromechanical interfaces would endow cyber–human interactions with unprecedented robustness and dexterity., Designing efficient systems capable emulating the muscular excitation-contraction signatures, remains a challenge. Here, the authors propose cytoadhesion-inspired hybrids as locally-coupled electromechanical interfaces capable retrieving the myoelectric and mechanical signals.

Non-invasive on-skin electrodes record the electrical potential changes from human skin, which reflect body condition and are applied for healthcare, sports management, and modern lifestyle. However, current on-skin electrodes have poor conformal properties under sweaty condition in real-life due to decreased electrode-skin adhesion with sweat film at the interface. Here, we fabricated bio-composite electrodes based on silk fibroin (SF) through interfacial polymerization which is applicable on sweaty skin. Interfacial polymerized conductive polypyrrole (PPy) and SF are structurally interlocked and endow the whole electrode with uniform stretchability. Existence of water results in similar Young’s modulus of SF to the skin and enhanced interfacial adhesion. It keeps the electrodes conformal to skin under sweaty condition and allows reliable collection of ambulatory electrophysiological signals during sports and sweating. Wearable devices with these electrodes were used to acquire continuous and stable real-time electrocardiography (ECG) signals during running for 2 hours. The collected signals can provide information for sports management and are also analyzed by artificial intelligence to show their potential for intelligent human emotion monitoring. Our strategy provides opportunities to record long-term continuous electrophysiological signals in real-life conditions for various smart monitoring systems. NRF (Natl Research Foundation, S’pore) ASTAR (Agency for Sci., Tech. and Research, S’pore) Accepted version

A new trend in brain-like computing is to develop more biologically realistic neurons in order to accurately replicate the computational power of the biological counterparts. The brain-like computing platforms based on these models that resemble the organization of neurons in brain would serve as a more efficient tool than traditional hardware. Recently, with the rapid development of neuromorphic devices, the realization of biologically-accurate neurons to fit these models becomes feasible. In view of this, we summarize the most recent progress on neuromorphic devices for achieving the essential neuronal information processing functions of a biological neuron. Despite its infancy, the biologically-accurate neuron already has exhibited an unprecedented potential to boost the hardware implementation of brain-like computation as well as leading a trend on neurorobotics and prosthetics. Challenges and perspectives are also addressed to shed light on future advances with respect to practical applications and t...

The biological visual system encodes information into spikes and processes them parallelly by the neural network, which enables the perception with high throughput of visual information processing at an energy budget of a few watts. The parallelism and efficiency of bio-visual system motivates electronic implementation of this biological computing paradigm, which is challenged by the lack of bionic devices, such as spiking neurons that can mimic its biological counterpart. Here, we present a highly bio-realistic spiking visual neuron based on an Ag/TaOX/ITO memristor. Such spiking visual neuron collects visual information by a photodetector, encodes them into action potentials through the memristive spiking encoder, and interprets them for recognition tasks based on a network of neuromorphic transistors. The firing spikes generated by the memristive spiking encoders have a frequency range of 1-200 Hz and sub-micro watts power consumption, very close to the biological counterparts. Furthermore, a spiking visual system is demonstrated, replicating the distance-dependent response and eye fatigue of biological visual systems. The mimicked depth perception shows a recognition improvement by adapting to sights at different distance. Our design presents a fundamental building block for energy-efficient and biologically plausible artificial visual systems.

High-performance amorphous indium-gallium-zinc-oxide thin-film transistors (a-IGZO TFTs) gated by Al2O3/HfO2 stacked dielectric films are investigated. The optimized TFTs with Al2O3 (2.0 nm)/HfO2 (13 nm) stacked gate dielectrics demonstrate the best performance, including low total trap density N t, low subthreshold swing voltage, large switching ratio I ON/OFF, high mobility μ FE, and low operating voltage, equal to 1.35 × 1012 cm−2, 88 mV/dec, 5.24 × 108, 14.2 cm2/V⋅s, and 2.0 V, respectively. Furthermore, a low-voltage-operated resistor-loaded inverter has been fabricated based on such an a-IGZO TFT, showing ideal full swing characteristics and high gain of ∼27 at 3.0 V. These results indicate a-IGZO TFTs gated by optimized Al2O3/HfO2 stacked dielectrics are of great interests for low-power, high performance, and large-area display and emerging electronics.

Synaptic plasticity divided into long-term and short-term categories is regarded as the origin of memory and learning, which also inspires the construction of neuromorphic systems. However, it is difficult to mimic the two behaviors monolithically, which is due to the lack of time-tailoring approaches for a certain synaptic device. In this Letter, indium-gallium-zinc-oxide (IGZO) nanofiber-based photoelectric transistors are proposed for realizing tunable photoelectric synaptic plasticity by the indium composition ratio. Notably, short-term plasticity to long-term plasticity transition can be realized by increasing the ratio of indium in the IGZO channel layer. The spatiotemporal dynamic logic and low energy consumption (

Noninvasive and seamless interfacing between the sensors and human skin is highly desired for wearable healthcare. Thin-film-based soft and stretchable sensors can to some extent form conformal contact with skin even under dynamic movements for high-fidelity signals acquisition. However, sweat accumulation underneath these sensors for long-term monitoring would compromise the thermal-wet comfort, electrode adherence to the skin, and signal fidelity. Here, we report the fabrication of a highly thermal-wet comfortable and conformal silk-based electrode, which can be used for on-skin electrophysiological measurement under sweaty conditions. It is realized through incorporating conducting polymers poly(3,4-ethylenedioxythiophene): polystyrenesulfonate (PEDOT:PSS) into glycerol-plasticized silk fiber mats. Glycerol plays the role of tuning the mechanical properties of silk fiber mats and enhancing the conductivity of PEDOT:PSS. Our silk-based electrodes show high stretchability (>250%), low thermal insulation (∼0.13 °C·m2·W-1), low evaporative resistance (∼23 Pa·m2·W-1, 10 times lower than ∼1.3 mm thick commercial gel electrodes), and high water-vapor transmission rate (∼117 g·m-2·h-1 under sweaty conditions, 2 times higher than skin water loss). These features enable a better electrocardiography signal quality than that of commercial gel electrodes without disturbing the heat dissipation during sweat evaporation and provide possibilities for textile integration to monitor the muscle activities under large deformation. Our glycerol-plasticized silk-based electrodes possessing superior physiological comfortability may further engage progress in on-skin electronics with sweat tolerance. Agency for Science, Technology and Research (A*STAR) National Research Foundation (NRF) Submitted/Accepted version X.C. would like to thank the financial support from the National Research Foundation, Prime Minister’s office, Singapore, under its NRF Investigatorship (NRF-NRFI2017- 07) and Agency for Science, Technology and Research (A*STAR) under its AME Programmable Funding Scheme (project no. A18A1b0045). K.-Q.Z. acknowledges the financial support from the National Key Research and Development Program of China (2017YFA0204600), the Natural Science Foundation of China (51873134), the Natural Science Foundation for Key Program of the Jiangsu Higher Education Institutions of China (17KJA540002). Y.C. and N.Z. are grateful for the support from the Agency for Science, Technology and Research (A*STAR) and the use of A*STAR Computational Resource Centre, Singapore (ACRC) and National Supercomputing Centre, Singapore (NSCC). Q.L. acknowledges the financial support from the China Scholarship Council (No. 201706920057).

Flexible electrodes that are multilayer, multimaterial, and conformal are pivotal for multifunctional wearable electronics. Traditional electronic circuits manufacturing requires substrate-supported transfer printing, which limits their multilayer integrity and device conformability on arbitrary surfaces. Herein, a "shrinkage-assisted patterning by evaporation" (SHAPE) method is reported, by employing evaporation-induced interfacial strain mismatch, to fabricate auto-detachable, freestanding, and patternable electrodes. The SHAPE method utilizes vacuum-filtration of polyaniline/bacterial cellulose (PANI/BC) ink through a masked filtration membrane to print high-resolution, patterned, and multilayer electrodes. The strong interlayer hydrogen bonding ensures robust multilayer integrity, while the controllable evaporative shrinking property of PANI/BC induces mismatch between the strains of the electrode and filtration membrane at the interface and thus autodetachment of electrodes. Notably, a 500-layer substrateless micro-supercapacitor fabricated using the SHAPE method exhibits an energy density of 350 mWh cm-2 at a power density of 40 mW cm-2 , 100 times higher than reported substrate-confined counterparts. Moreover, a digital circuit fabricated using the SHAPE method functions stably on a deformed glove, highlighting the broad wearable applications of the SHAPE method. Agency for Science, Technology and Research (A*STAR) Submitted/Accepted version This work was supported by the Agency for Science, Technology and Research (A*STAR) under its AME Programmatic Funding Scheme of Cyber-Physiochemical Interfaces Programme (Project No. A18A1b0045).

The human brain that relies on neural networks communicated by spikes is featured with ultralow energy consumption, which is more robust and adaptive than any digital system. Inspired by the spiking framework of the brain, spike-based neuromorphic systems have recently inspired intensive attention. Therefore, neuromorphic devices with spike-based synaptic functions are considered as the first step toward this aim. Photoelectric neuromorphic devices are promising candidates for spike-based synaptic devices with low latency, broad bandwidth, and superior parallelism. Here, the indium-gallium-zinc-oxide-based photoelectric neuromorphic transistors are fabricated for Morse coding based on spike processing, 405-nm light spikes are used as synaptic inputs, and some essential synaptic plasticity, including excitatory postsynaptic current, short-term plasticity, and high-pass filtering, can be mimicked. More interestingly, Morse codes encoded by light spikes are decoded using our devices and translated into amplitudes. Furthermore, such devices are compatible with standard integrated processes suitable for large-scale integrated neuromorphic systems.

The biological visual system encodes optical information into spikes and processes them by the neural network, which enables the perception with high throughput of visual processing with ultralow energy budget. This has inspired a wide spectrum of devices to imitate such neural process, while precise mimicking such procedure is still highly required. Here, a highly bio-realistic photoelectric spiking neuron for visual depth perception is presented. The firing spikes generated by the TaO

In recent years, the rapid increase in the data volume to be processed has led to urgent requirements for highly efficient computing paradigms. Brain-like computing that mimics the way the biological brain processes information has attracted growing interest due to extremely high energy efficiency. Particularly, dynamics play an essential role in neural spike information processing. Here, we offer a brief review and perspective in the field of electrolyte-gated neuromorphic transistors for brain-like dynamic computing. We first introduce the biological foundation of dynamic neural functions. Then dynamic synaptic plasticity, dynamic dendritic integration, dynamic neural functions, and bio-inspired somatosensory systems realized based on the electrolyte-gated neuromorphic transistors are presented. At last, conclusions and perspectives are given. Published version The authors acknowledge the financial support from the National Natural Science Foundation of China (Grant Nos. 61921005 and 61834001) and the National Key R&D Program of China (Grant No. 2019YFB2205400).

A hardware based artificial neural network (ANN), which holds the potential to alleviate the computation load and energy of a digital computer, has propelled the development of memory devices that can resemble the synapse. Memcapacitors, especially based on ferroelectric materials, with theoretically no static power, nondestructive readout, and multiple polarization states, are expected to have good energy efficiency and endurance as emerging artificial synapses. However, conventional ferroelectric devices are characterized with extremely high remnant polarization, which requires high energy for polarization state updating and always leads to low linearity and symmetry in updating properties. Here, we show a memcapacitive synapse based on an Au/HfZrOx (HZO)/Au ferroelectric memcapacitor with moderate remnant polarization that can offer unexceptionable updating properties for building an ANN. The memcapacitor demonstrates more than 64 weight states with an ultralow weight updating energy of ≤3.0 fJ/ μm2. Both potentiation and depression synaptic characteristics show an ultralow non-linearity of −2. Based on these properties, a two-layer restricted Boltzmann machine is built based on this memcapacitive synapse, and it can be trained to reconstruct incomplete images. The reconstructed images show reduced Euclidean distance to originals in comparison with that of the incomplete images. Furthermore, the memcapacitive synapse is also tested by a handwritten digits recognition task based on a simple perceptron, and the pattern recognition accuracy is as high as 93.4%. These results indicate that the HZO-based capacitive synapse devices have great potential for future high-efficiency neuromorphic systems.

Human behaviors are extremely sophisticated, relying on the adaptive, plastic and event-driven network of sensory neurons. Such neuronal system analyzes multiple sensory cues efficiently to establish accurate depiction of the environment. Here, we develop a bimodal artificial sensory neuron to implement the sensory fusion processes. Such a bimodal artificial sensory neuron collects optic and pressure information from the photodetector and pressure sensors respectively, transmits the bimodal information through an ionic cable, and integrates them into post-synaptic currents by a synaptic transistor. The sensory neuron can be excited in multiple levels by synchronizing the two sensory cues, which enables the manipulating of skeletal myotubes and a robotic hand. Furthermore, enhanced recognition capability achieved on fused visual/haptic cues is confirmed by simulation of a multi-transparency pattern recognition task. Our biomimetic design has the potential to advance technologies in cyborg and neuromorphic systems by endowing them with supramodal perceptual capabilities., Designing bioinspired perceptual system remains a challenge. Here, the authors report a bimodal artificial sensory neuron, integrating a resistive pressure sensor, a perovskite-based photodetector, a hydrogel-based ionic cable, and a synaptic transistor, to implement the visual-haptic fusion for motion control and patterns recognition.

Flexible electronics have witnessed exciting progress in academia over the past decade, but most of the research outcomes have yet to be translated into products or gain much market share. For mass production and commercialization, industrial adoption of newly developed functional materials and fabrication techniques is a prerequisite. However, due to the disparate features of academic laboratories and industrial plants, translating materials and manufacturing technologies from labs to fabs is notoriously difficult. Therefore, herein, key challenges in the materials manufacturing of flexible electronics are identified and discussed for its lab-to-fab translation, along the four stages in product manufacturing: design, materials supply, processing, and integration. Perspectives on industry-oriented strategies to overcome some of these obstacles are also proposed. Priorities for action are outlined, including standardization, iteration between basic and applied research, and adoption of smart manufacturing. With concerted efforts from academia and industry, flexible electronics will bring a bigger impact to society as promised. Agency for Science, Technology and Research (A*STAR) Ministry of Education (MOE) National Research Foundation (NRF) Accepted version The authors thank the financial support from the Agency for Science, Technology and Research (A*STAR) under its AME Programmatic Funding Scheme (project no. A18A1b0045), the National Research Foundation (NRF), Prime Minister’s Office, Singapore, under its NRF Investigatorship (NRF2016NRF-NRFI001-21), and Singapore Ministry of Education (MOE2019-T2-2-022).

Gesture recognition using machine learning methods is valuable in the development of advanced cybernetics, robotics, and healthcare systems, and typically relies on images or videos. To improve recognition accuracy, such visual data can be fused with data from other sensors, but this approach is limited by the quality of the sensor data and the incompatibility of the datasets. Here, we report a bioinspired data fusion architecture that can perform human gesture recognition by integrating visual data with somatosensory data from skin-like stretchable strain sensors. The learning architecture uses a convolutional neural network for visual processing, and then implements a sparse neural network for sensor data fusion and recognition. Our approach can achieve a recognition accuracy of 100%, and maintain recognition accuracy with noisy, under- or over-exposed images. We also show that our architecture can be implemented for robot navigation using hand gestures with a small error, even in the dark. Ministry of Education (MOE) National Research Foundation (NRF) Accepted version The project was partially supported by the National Research Foundation (NRF), Prime Minister’s office, Singapore, under its NRF Investigatorship (NRF2016NRF-NRF1001-21), Singapore Ministry of Education (MOE2015-T2-2-60), Advanced Manufacturing and Engineering (AME) Programmatic Grant (No. A19A1b0045), and the Australian Research Council (ARC) under Discovery Grant DP200100700. The authors thank all the volunteers for collecting data and also thank Dr. Ai Lin Chun for critically reading and editing the manuscript.

Sensory memory, formed at the beginning while perceiving and interacting with the environment, is considered a primary source of intelligence. Transferring such biological concepts into electronic implementation aims at achieving perceptual intelligence, which would profoundly advance a broad spectrum of applications, such as prosthetics, robotics, and cyborg systems. Here, the recent developments in the design and fabrication of artificial sensory memory devices are summarized and their applications in recognition, manipulation, and learning are highlighted. The emergence of such devices benefits from recent progress in both bioinspired sensing and neuromorphic engineering technologies and derives from abundant inspiration and benchmarks from an improved understanding of biological sensory processing. Increasing attention to this area would offer unprecedented opportunities toward new hardware architecture of artificial intelligence, which could extend the capabilities of digital systems with emotional/psychological attributes. Pending challenges are also addressed to aspects such as integration level, energy efficiency, and functionality, which would undoubtedly shed light on the future development of translational implementations. NRF (Natl Research Foundation, S’pore) ASTAR (Agency for Sci., Tech. and Research, S’pore) MOE (Min. of Education, S’pore) Accepted version

On‐skin flexible devices provide a noninvasive approach for continuous and real‐time acquisition of biological signals from the skin, which is essential for future chronic disease diagnosis and smart health monitoring. Great progress has been achieved in flexible devices to resolve the mechanical mismatching between conventional rigid devices and human skin. However, common materials used for flexible devices including silicon‐based elastomers and various metals exhibit no resistance to epidermal surface lipids (skin oil and grease), which restricts the long‐term and household usability. Herein, an on‐skin electrode with anti‐epidermal‐surface‐lipid function is reported, which is based on the grafting of a zwitterionic poly(2‐methacryl‐oyloxyethyl, methacryloyl‐oxyethyl, or meth‐acryloyloxyethyl phosphorylcholine) (PMPC) brush on top of gold‐coated poly(dimethylsiloxane) (Au/PDMS). Such an electrode allows the skin‐lipids‐fouled surface to be cleaned by simple water rinsing owing to the superhydrophilic zwitterionic groups. As a proof‐of‐concept, the PMPC‐Au/PDMS electrodes are employed for both electrocardiography (ECG) and electromyography (EMG) recording. The electrodes are able to maintain stable skin‐electrode impedance and good signal‐to noise ratio (SNR) by water rinsing alone. This work provides a material‐based solution to improve the long‐term reusability of on‐skin electronics and offers a unique prospective on developing next generation wearable healthcare devices. Agency for Science, Technology and Research (A*STAR) Ministry of Education (MOE) National Research Foundation (NRF) Accepted version This work was funded by the Agency for Science, Technology and Research (A*STAR) under its AME Programmatic Funding Scheme (Project #A18A1b0045) Cyber-Physiochemical Interfaces (CPI) Programme, the Singapore Ministry of Education (MOE) Tier 2 (MOE2019-T2-2-022) and National Research Foundation (NRF), Prime Minister's office, Singapore, under its Competitive Research Programme Funding Scheme (NRF-CRP13-2014-02), and NRF Investigatorship (NRF-NRFI2017-07).

Memristors constitute a promising functional component for information storage and in-memory computing in flexible and stretchable electronics including wearable devices, prosthetics, and soft robotics. Despite tremendous efforts made to adapt conventional rigid memristors to flexible and stretchable scenarios, stretchable and mechanical-damage-endurable memristors, which are critical for maintaining reliable functions under unexpected mechanical attack, have never been achieved. Here, the development of stretchable memristors with mechanical damage endurance based on a discrete structure design is reported. The memristors possess large stretchability (40%) and excellent deformability (half-fold), and retain stable performances under dynamic stretching and releasing. It is shown that the memristors maintain reliable functions and preserve information after extreme mechanical damage, including puncture (up to 100 times) and serious tearing situations (fully diagonally cut). The structural strategy offers new opportunities for next-generation stretchable memristors with mechanical damage endurance, which is vital to achieve reliable functions for flexible and stretchable electronics even in extreme and highly dynamic environments. Submitted/Accepted version T.W. and Z.C. contributed equally to this work. The project was supported by the National Research Foundation (NRF), Prime Minister’s office, Singapore, under its NRF Investigatorship (NRF-NRFI2017-07), Singapore Ministry of Education (MOE2017-T2-2-107 and MOE2019-T2-2-022), and the Agency for Science, Technology and Research (A*STAR) under its Advanced Manufacturing and Engineering (AME) Programmatic Funding Scheme (project #A18A1b0045).

Brain-inspired neuromorphic computing would bring a breakthrough to the classical computing paradigm through its massive parallelism and potential low power consumption advantages. Introduction of flexibility may bring vitality to this area by expanding its application areas to such as wearable and implantable electronics. At present, the development of flexible neuromorphic devices makes it a choice with wide prospect for next-generation wearable artificial neuromorphic computing. In this study, a freestanding graphene oxide/polyvinyl alcohol composite solid electrolyte membrane is utilized as the gate dielectric and support material, and indium-zinc-oxide (IZO) neuromorphic transistors are fabricated on such membrane. Based on the in-plane gate modulation, many key synaptic plasticity behaviors have been successfully emulated, including excitatory postsynaptic current, paired-pulse facilitation, high-pass filtering, and spatiotemporal signal processing. Moreover, transition of the spiking logic and the superlinear and sublinear dendritic integration function are realized. Our results indicate that these freestanding IZO-based neuromorphic transistors may of great significance for future flexible anthropomorphic robots, wearable bionic perception.

Neuromorphic devices are of great significance for capturing the computation power of the neural network with ultralow energy consumption and extremely high efficiency. Here, multigate indium-zinc-oxide-based neuron thin film transistors laterally coupled by sodium alginate electrolyte films were fabricated polyethylene terephthalate substrates for neuromorphic system application. The dynamic range of the synaptic response can be effectively modulated by the third modulatory input. More importantly, multiplicative operation was realized successfully based on the rate-coding mode. Such laterally coupled oxide-based neuron transistors could be used as the potential building block in neuromorphic systems.