Search

It is of great significance to develop multifunction-integrated photoelectric synaptic devices with the simple structure to imitate the functions of eyes. In this work, we reported the multimodal photoelectric neuromorphic thin film transistors (TFTs) with single gate and two input terminals using sorted semiconducting single-walled carbon nanotubes (sc-SWCNTs) mixed with lead-free perovskite (CsBi3I10) and lightly n-doped silicon as active layers and gate electrodes. The results showed that optical and electrical responses of SWCNT TFTs could increase 26 and 23 times after addition of CsBi3I10 under the same pulse sequence stimulation, respectively. Apart from emulating a series of typical synaptic functionalities, an artificial neural network based on SWCNT phototransistors was simulated to be trained for the recognition of handwritten digits in the Modified National Institute of Standards and Technology with the maximum accuracy of about 85.46%. Moreover, our synaptic devices can imitate traditional Pavlovian conditioning, and show NOR logic, reproducible flash memory functions under the optical and electrical programming stimulations. It is noted that the erasing voltage (3 V, 0.2 s) is one of the best value among the reported phototransistor memory. Furthermore, our synaptic devices can work well after more than 150 programming/erasing operation cycles and over 40 days testing.

Summary Memristive devices offer desirable voltage-regulated conductance switching and promise to address zettabyte storage challenges in the big-data era. Generally, most of these reported devices use amorphous solids, where the structural and compositional inhomogeneity is regarded as the origin of stochastic variability. Self-assembling peptide crystals with solution-processed fabrication, controllable morphologies, and structural stability are therefore a promising candidate to address the reliability issues. Here we report a planar diffusive memristor that possesses reliable switching characteristics based on a quasi-one-dimensional crystallized material: diphenylalanine (FF) microrod (MR). This element offers a preferential ion migration path, confining conductive filaments in a defined crystalline surface and therefore reducing programming stochasticity. Ion transport confinement along FF MR was observed via in situ electrostatic force microscopy technique. Additionally, FF MR memristor with high switching uniformity and reproducible relaxation dynamic provides an ideal hardware platform for reliable nociceptor emulation and hardware identification of four-bit decimal numbers.

As the power density of the electronics is on increasing, improving the heat dissipation performance of electronic packaging materials will play a positive role in promoting the performance of modern electronics. In this work, the three-dimensional carbon fiber (CF)-MXenes foam, in which the vertically aligned CF constructed the heat transport paths, were prepared by simple freeze-drying method. As a result, the thermal conductivity (TC) of CF-M/epoxy composites was improved to an ultra-high level (9.68 W/mK) at 30.2 wt% hybrid fillers, increasing by 4509% enhancement compared with that of neat epoxy. In addition, the thermal properties of composites, such as glass transition temperature (Tg), coefficient of thermal expansion (CTE) were investigated. All the results indicated that the CF-MXenes/epoxy composite was supposed to be a promising heat dissipation material and will be used in the field of electronic packaging.

Inspired by the parallel performing of storage and processing information in biological synapse, artificial synapse that can control ionic and protonic currents are ideal means for attracted tremendous attention as a promising option to break the von Neumann bottleneck. Here we demonstrate an artificial synapse with tunable synaptic behavior based on solution-processed two-dimensional (2D) C3N/polyvinylpyrrolidone (PVPy). The proton modulated memristive characteristics of synaptic device are verified by ambient pressure X-ray photoelectron spectroscopy under different H2O gas atmosphere. The highly proton conducting of C3N/PVPy matrix stems from the hydrogen bonding network between C3N and PVPy, as well as the large amount of ordered nitrogen atoms in C3N. The artificial synapse ensure a direct imitation of short-term and long-term plasticity in biological synapses including excitatory post-synaptic current (EPSC), paired-pulse facilitation (PPF), paired-pulse depression (PPD), PPF following PPD and post-tetanic potentiation (PTP). The C3N/PVPy matrix-based memristor which can mimic the synapse cleft based on proton conducting mechanism may find further applications in artificial intelligence.

Natural biomaterials are capable of producing biodegradable and/or biocompatible electronic devices for vast applications such as next-generation green data storage. Here, we demonstrate an environment-friendly alkali lignin as a favourable candidate for resistive switching memory storage applications. The resistive memory devices based on gold nanoparticles embedded alkali lignin (Au NPs:lignin) exhibits a typical write-once-read-many-times memory (WORM) resistive switching behaviour with a large on-off ratio (>104), and long data retention characteristics (>103 s) under low power operation (4.7 V). Au NPs acting as trapping sites in the active layer are responsible for the resistive switching mechanism. The active layer is fabricated on organic substrates through facile solution-processed methods under normal ambient conditions. The skin-inspired polylactide (PLA) film is utilized as an ultrathin substrate endowing the memory devices with good mechanical support and biodegradability. This work opens up new avenues towards renewable and environmentally benign lignin-based materials for biodegradable electronic devices.

Increasing environmental crises caused by exploring and using fossil fuels have compelled human being to develop innovative technologies to utilize renewable and sustainable energy sources. For this purpose, photovoltaic conversion of solar energy into electricity with solar cells is a promising and attracting way in that solar energy is clean and inexhaustible. Nowadays, the bottleneck in the application of solar cells on a large scale to sustainable energy generation still lies in lacking an efficient, stable and low-cost materials system for photon-to-electricity conversion. Perovskite materials are a class of materials widely applied in solar cells. Many evidences showed that the perovskite materials have both ferroelectric and photovoltaic properties, offering a special system called photoferroelectric materials. A built-in electric field established in these materials due to the ferroelectric property is more helpful for the separation of e-h pairs and enhancing the power conversion efficiency during photovoltaic process in solar cells. Here, we review the recent photoferroelectric perovskite solar cells (PPSCs). After giving a brief description of the structure and property of photoferroelectric perovskite materials, the device structures, working principles and characterization of PPSCs are introduced, followed by the state-of-the-art advances and the insights for the PPSCs based on oxide and halide perovskite materials. Finally, the main challenges in developing efficient PPSCs are discussed.

Electronic synapses implementing in-memory computing system could overcome the developing limitation on the energy efficiency of traditional von Neumann architecture. Compared with the high sensitivity of biological synapses, lower responsivity of the memristive synapses was found via the electrical stimulations. Here, poly{2,2-(2,5-bis(2-octyldodecyl)-3,6-dioxo-2,3,5,6- tetrahydropyrrolo[3,4-c]pyrrole-1,4-diyl)-dithieno[3,2-b]thiophene-5,5-diyl-alt-thiophen-2,5-diyl} (PDPPBTT)/zinc oxide (ZnO) based heterojunction is found to exhibit stable memristive switching behavior, which originates from the confined formation/rupture of filament in the two-layer interface region as the ions migrate with different transport rates in two layers. The implementing synaptic functions in the sensitive memristive device can realize the short-term plasticity and long-term plasticity when stimulated by the applied electrical signals with different stimulating rate. Similar to the biological synapse, the memory loss, memory transition, and the critical role of stimulation rate on the transition process, can be achieved in the as-prepared memristor device. The systematic demonstrations on the synaptic emulation may facilitate building bio-inspired device-level neuromorphic systems.

Artificial retina perception system is significant to pattern recognition and visual function emulation. However, the recent artificial retina system is mainly reported on the integration of three-terminal transistors, whose structural limitations may result in low processing speeds and high energy consumption due to a low array density and complex line design. Furthermore, the external power source is required to drive devices so that the power consumption of the system would increase. Here we present a self-powered artificial retina perception system by utilizing two-terminal solar cells as artificial neurons and perovskite-based memristors as artificial synapses, ensuring the bio-inspired retina system with extendable crossbar array structure for high-density and low power consumption neural networks. By a light stimulation with various wavelengths and intensities, the electrical pre-synaptic signal is generated in the solar cell and subsequently transferred to the perovskite-based memristor to perform further information preprocessing. Especially, the applicability of the artificial retina system to neuromorphic computing is demonstrated for contrast enhancement and noise reduction. The retina perception system is capable of feature extraction by to implement partial functions of convolutional neural networks (CNNs) on the hardware level with improved recognition rate, boosted recognition speed, and reduced energy consumption.

Coupling integration devices have been widely investigated in order to realize advanced artificial intelligence. In particular, by varying the mechanical energy of the sensory memory, the external force causes the alteration of internal energy and exerts an effect on its electrical properties. The goal of converting artificial touch into electronic implementation is to achieve perceptual intelligence, and it would profoundly advance a wide range of applications, such as robotic grasping tools, prosthetics, and human-computer interaction. Here, we focus on the recent progresses in electromechanical coupling memory and their applications in tactile sensory memory and artificial synapses. Also, electromechanical sensory memories are analyzed in terms of deformation-involved physical mechanisms, including the piezoelectric effect, piezoresistive effect, and triboelectric effect. Deformation-involved functional sensors integrated with memristors are also discussed for the artificial tactile memory applications. In addition, pressure-tunable synaptic functions and artificial memory skin emulated by electromechanical memories are discussed in details. At last, the application prospect is shown and challenges are also exist in aspects such as multilevel storage, energy consumption, and sensitive degree, which would further disclosure the merit and demerit of electromechanical integrated devices.

Artificial neuron is an important part of constructing neuromorphic network in which information can be computed with high parallelism and efficiency like in the human brain. However, owing to the poor biological plausibility, artificial neurons based on traditional complementary metal-oxide-semiconductor (CMOS) platform fail to reveal the rich ion dynamics of the biological counterparts. Organic–inorganic halide perovskites (OHPs) are prospective for imitating the ion dynamics on the membranes of biological neurons because of its intrinsic ion migration. Herein, a diffusive CH3NH3PbI3(MAPbI3)-based memristor with superior amplitude-frequency characteristics and highly linear conductivity modulation for more than 1000 states have been fabricated for the construction of a leaky integrate-and-fire (LIF) bio-inspired neuron. The as-designed LIF model can successfully emulate the leakage, spatiotemporal integration and firing functions in a biological neuron. Moreover, by connecting LIF neurons with a 2*2 non-volatile Al2O3-based synaptic array, a simple spiking neural network (SNN) which is called the 3rd generation of neural network has been implemented at the hardware level to study the cognitive performance of the network. The SNN exhibits outstanding selective sensitivity to particular input sequence, indicating the excellent adaptability and versatility of the network for future applications of neuromorphic computing by utilizing novel ionotropic device.

Lead-free halide perovskites are emerging as promising candidate for practical application of optoelectronic devices due to their nontoxicity. Unfortunately, previously-reported lead-free halide perovskites-based resistive switching devices suffer from high leakage and operating current stemmed from the intrinsic nature of polycrystalline film with a great amount of grain boundaries and pin-holes. Here, we report for the first time a monocrystalline lead-free Cs3Sb2Br9 perovskite nanoflake based lateral-structured device capable of combining nonvolatile bipolar switching and threshold switching with record-low switching electric field of 2.2 × 105 V m−1. Confirmed by elemental analysis and theoretical calculation, migration of highly mobile Br vacancy with low activation energy in defects-free monocrystalline Cs3Sb2Br9 is believed to be responsible for resistive switching. Short-term Ca2+ dynamics of biological synapses were then imitated by Cs3Sb2Br9 resistive switching devices which were further implemented as an effective reservoir element. The construction of neural network-based reservoir computing system to efficiently process temporal information can be realized since its conductance states are determined by the history of external simulation.

Developing new-style non-volatile memory is an important part to solve von Neumann bottleneck. Among numerous studies of memory device, resistive random-access memory (RRAM) is widely researched for their simple geometrical structure and high integration density. In recent decades, organic RRAM catches many researchers’ attention for their superiorities such as outstanding flexibility, simple solution processability, low-cost fabrication and tunable functionalities by changing the chemical groups. Herein, a series of organic small molecules 2,7-di(9H-carbazol-9-yl)-9H-fluoren-9-one (FLOCZ) and 2,7-bis(3,6-di-tert-butyl-9H-carbazol-9-yl)-9H-fluoren-9-one (FLOBCZ) are synthesized and their band structures and electronic states are systematically studied with the assistance of density function theory (DFT). Then, using simple solution-processing approach, FLOCZ/FLOBCZ based RRAMs are fabricated and their memory characteristics are studied. The as-fabricated Ag/FLOBCZ/ITO RRAM exhibits non-volatile memory effect with a memory window up to 7 × 103 and data retention time of more than 2 × 104 s, which can meet the requirement of data storage. Meanwhile, by controlling compliance current on the RRAM precisely, multi-level data storage can be implemented experimentally. Our work extends the applications of organic small molecule and promotes the evolution of organic RRAM.

Near infrared (NIR) synaptic devices offer a remote-control approach to implement neuromorphic computing for data safety and artificial retinal system applications. In upconverting nanoparticles (UCNPs)-mediated optogenetics biosystems, NIR regulation of membrane ion channels allows remote and selective control of the Ca2+ flux to modulate synaptic plasticity behaviors. Inspired by the upconversion optogenetics, we proposed a NIR artificial synapse based on a UCNPs-MoS2 floating gate phototransistor in which MoS2 acts as light-sensitive ion channels to reabsorb the visible light emitted from UCNPs under NIR illumination. As a result, the synaptic device exhibits stable persistent photocurrent (PPC) effect up to 353 K and ultrahigh photogain (~108 electrons per photon), ensuring the long-term potentiation (LTP) behavior. Simulations using the handwritten digit data sets indicate good recognition accuracy of the light-controlled artificial neuron network. Overall, this design concept combining biology, optics and electronics opens up a new avenue for developing optogenetics-inspired neuromorphic technology in the future.