Analog Synaptic Transistor with Al-Doped HfO2 Ferroelectric Thin Film

Growth of emergent simple pseudo-binary ferroelectrics and their potential in neuromorphic computing devices

Ferroelectric memory devices such as ferroelectric memristors, ferroelectric tunnel junctions, and field-effect transistors are considered among the most promising candidates for neuromorphic computing devices. The promise arises from their defect-independent switching mechanism, low energy consumption and high power efficiency, and important properties being aimed for are reliable switching at high speed, excellent endurance, retention, and compatibility with complementary metal-oxide-semiconductor (CMOS) technology. Binary or doped binary materials have emerged over conventional complex-composition ferroelectrics as an optimum solution, particularly in terms of CMOS compatibility. The current state-of-the-art route to achieving superlative ferroelectric performance of binary oxides is to induce ferroelectricity at the nanoscale, e.g., in ultra-thin films of doped HfO2, ZrO2, Zn1-xMgxO, Al-xScxN, and Bi1-xSmxO3. This short review article focuses on the materials science of emerging new ferroelectric materials, including their different properties such as remanent polarization, coercive field, endurance, etc. The potential of these materials is discussed for neuromorphic applications.

Hydrogel-Gated FETs in Neuromorphic Computing to Mimic Biological Signal: A Review

Hydrogel-gated synaptic transistors offer unique advantages, including biocompatibility, tunable electrical properties, being biodegradable, and having an ability to mimic biological synaptic plasticity. For processing massive data with ultralow power consumption due to high parallelism and human brain-like processing abilities, synaptic transistors have been widely considered for replacing von Neumann architecture-based traditional computers due to the parting of memory and control units. The crucial components mimic the complex biological signal, synaptic, and sensing systems. Hydrogel, as a gate dielectric, is the key factor for ionotropic devices owing to the excellent stability, ultra-high linearity, and extremely low operating voltage of the biodegradable and biocompatible polymers. Moreover, hydrogel exhibits ionotronic functions through a hybrid circuit of mobile ions and mobile electrons that can easily interface between machines and humans. To determine the high-efficiency neuromorphic chips, the development of synaptic devices based on organic field effect transistors (OFETs) with ultra-low power dissipation and very large-scale integration, including bio-friendly devices, is needed. This review highlights the latest advancements in neuromorphic computing by exploring synaptic transistor developments. Here, we focus on hydrogel-based ionic-gated three-terminal (3T) synaptic devices, their essential components, and their working principle, and summarize the essential neurodegenerative applications published recently. In addition, because hydrogel-gated FETs are the crucial members of neuromorphic devices in terms of cutting-edge synaptic progress and performances, the review will also summarize the biodegradable and biocompatible polymers with which such devices can be implemented. It is expected that neuromorphic devices might provide potential solutions for the future generation of interactive sensation, memory, and computation to facilitate the development of multimodal, large-scale, ultralow-power intelligent systems.

Synaptic Characteristics of Atomic Layer-Deposited Ferroelectric Lanthanum-Doped HfO2 (La:HfO2) and TaN-Based Artificial Synapses

Analog synaptic devices have made significant advances based on various electronic materials that can realize the biological synapse properties of neuromorphic computing. Ferroelectric (FE) HfO2-based materials with nonvolatile and low power consumption characteristics are being studied as promising materials for application to analog synaptic devices. The gradual reversal of FE multilevel polarization results in precise changes in the channel conductance and allows analogue synaptic weight updates. However, there have been few studies of FE synaptic devices doped with La, Y, and Gd. Furthermore, an investigation of interface quality is also crucial to enhance the remnant polarization (Pr), synaptic conductance linearity, and reliability characteristics. In this study, we demonstrate improved FE and artificial synaptic characteristics using an atomic layer-deposited (ALD) lanthanum-doped HfO2 (La:HfO2) and TaN electrode in the structure of an FE thin-film transistor (ITO/IGZO/La:HfO2/TaN), where indium-tin oxide (ITO) and indium-gallium-zinc oxide (IGZO) were used as source/drain and channel materials, respectively. Improved Pr and lower surface roughness were achieved by doped HfO2 and ALD TaN thin films. This synaptic transistor shows long-term potentiation and long-term depression with 200 levels of conductance states, high linearity (Ap, 0.97; Ad, 0.86), high Gmax/Gmin (∼6.1), and low cycle-to-cycle variability. In addition, a pattern recognition accuracy higher than 90% was achieved in an artificial neural network simulation.

High-performance van der Waals antiferroelectric CuCrP2S6-based memristors

Layered thio- and seleno-phosphate ferroelectrics, such as CuInP2S6, are promising building blocks for next-generation nonvolatile memory devices. However, because of the low Curie point, the CuInP2S6-based memory devices suffer from poor thermal stability (<42 °C). Here, exploiting the electric field-driven phase transition in the rarely studied antiferroelectric CuCrP2S6 crystals, we develop a nonvolatile memristor showing a sizable resistive-switching ratio of ~ 1000, high switching endurance up to 20,000 cycles, low cycle-to-cycle variation, and robust thermal stability up to 120 °C. The resistive switching is attributed to the ferroelectric polarization-modulated thermal emission accompanied by the Fowler-Nordheim tunneling across the interfaces. First-principles calculations reveal that the good device performances are associated with the exceptionally strong ferroelectric polarization in CuCrP2S6 crystal. Furthermore, the typical biological synaptic learning rules, such as long-term potentiation/depression and spike amplitude/spike time-dependent plasticity, are also demonstrated. The results highlight the great application potential of van der Waals antiferroelectrics in high-performance synaptic devices for neuromorphic computing.

A bi-functional three-terminal memristor applicable as an artificial synapse and neuron

Due to their significant resemblance to the biological brain, spiking neural networks (SNNs) show promise in handling spatiotemporal information with high time and energy efficiency. Two-terminal memristors have the capability to achieve both synaptic and neuronal functions; however, such memristors face asynchronous programming/reading operation issues. Here, a three-terminal memristor (3TM) based on oxygen ion migration is developed to function as both a synapse and a neuron. We demonstrate short-term plasticity such as pair-pulse facilitation and high-pass dynamic filtering in our devices. Additionally, a 'learning-forgetting-relearning' behavior is successfully mimicked, with lower power required for the relearning process than the first learning. Furthermore, by leveraging the short-term dynamics, the leaky-integrate-and-fire neuronal model is emulated by the 3TM without adopting an external capacitor to obtain the leakage property. The proposed bi-functional 3TM offers more process compatibility for integrating synaptic and neuronal components in the hardware implementation of an SNN.

Revival of Ferroelectric Memories Based on Emerging Fluorite-Structured Ferroelectrics

Over the last few decades, the research on ferroelectric memories has been limited due to their dimensional scalability and incompatibility with complementary metal-oxide-semiconductor (CMOS) technology. The discovery of ferroelectricity in fluorite-structured oxides revived interest in the research on ferroelectric memories, by inducing nanoscale nonvolatility in state-of-the-art gate insulators by minute doping and thermal treatment. The potential of this approach has been demonstrated by the fabrication of sub-30 nm electronic devices. Nonetheless, to realize practical applications, various technical limitations, such as insufficient reliability including endurance, retention, and imprint, as well as large device-to-device-variation, require urgent solutions. Furthermore, such limitations should be considered based on targeting devices as well as applications. Various types of ferroelectric memories including ferroelectric random-access-memory, ferroelectric field-effect-transistor, and ferroelectric tunnel junction should be considered for classical nonvolatile memories as well as emerging neuromorphic computing and processing-in-memory. Therefore, from the viewpoint of materials science, this review covers the recent research focusing on ferroelectric memories from the history of conventional approaches to future prospects.

Progress, Challenges, and Opportunities in Oxide Semiconductor Devices: A Key Building Block for Applications Ranging from Display Backplanes to 3D Integrated Semiconductor Chips

As Si has faced physical limits on further scaling down, novel semiconducting materials such as 2D transition metal dichalcogenides and oxide semiconductors (OSs) have gained tremendous attention to continue the ever-demanding downscaling represented by Moore's law. Among them, OS is considered to be the most promising alternative material because it has intriguing features such as modest mobility, extremely low off-current, great uniformity, and low-temperature processibility with conventional complementary-metal-oxide-semiconductor-compatible methods. In practice, OS has successfully replaced hydrogenated amorphous Si in high-end liquid crystal display devices and has now become a standard backplane electronic for organic light-emitting diode displays despite the short time since their invention in 2004. For OS to be implemented in next-generation electronics such as back-end-of-line transistor applications in monolithic 3D integration beyond the display applications, however, there is still much room for further study, such as high mobility, immune short-channel effects, low electrical contact properties, etc. This study reviews the brief history of OS and recent progress in device applications from a material science and device physics point of view. Simultaneously, remaining challenges and opportunities in OS for use in next-generation electronics are discussed.

Flexible aluminum-doped hafnium oxide ferroelectric synapse devices for neuromorphic computing

The HfO2-based ferroelectric tunnel junction has received outstanding attention owing to its high-speed and low-power characteristics. In this work, aluminum-doped HfO2 (HfAlO) ferroelectric thin films are deposited on a muscovite substrate (Mica). We investigate the bending effect on the ferroelectric characteristics of the Au/Ti/HfAlO/Pt/Ti/Mica device. After 1000 bending times, the ferroelectric properties and the fatigue characteristics are largely degraded. The finite element analysis indicates that crack formation is the main reason for the fatigue damage under threshold bending diameters. Moreover, the HfAlO-based ferroelectric synaptic device exhibits excellent performance of neuromorphic computing. The artificial synapse can mimic the paired-pulse facilitation and long-term potentiation/depression of biological synapses. Meanwhile, the accuracy of digit recognition is 88.8%. This research provides a new research idea for the further development of hafnium-based ferroelectric devices.

A high linearity and multilevel polymer-based conductive-bridging memristor for artificial synapses

Conductive-bridging memristors based on a metal ion redox mechanism have good application potential in future neuromorphic computing nanodevices owing to their high resistance switch ratio, fast operating speed, low power consumption and small size. Conductive-bridging memristor devices rely on the redox reaction of metal ions in the dielectric layer to form metal conductive filaments to control the conductance state. However, the migration of metal ions is uncontrollable by the applied bias, resulting in the random generation of conductive filaments, and the conductance state is difficult to accurately control. Herein, we report an organic polymer carboxylated chitosan-based memristor doped with a small amount of the conductive polymer PEDOT:PSS to improve the polymer ionic conductivity and regulate the redox of metal ions. The resulting device exhibits uniform conductive filaments during device operation, more than 100 and non-volatile conductance states with a ∼1 V range, and linear conductance regulation. Moreover, simulation using handwritten digital datasets shows that the recognition accuracy of the carboxylated chitosan-doped PEDOT:PSS memristor array can reach 93%. This work provides a path to facilitate the performance of metal ion-based memristors in artificial synapses.

Mechanism of the Wake-Up and the Split-Up in AlOx/Hf0.5Zr0.5Ox Film

Dielectric layers are widely used in ferroelectric applications such as memory and negative capacitance devices. The wake-up and the split-up phenomena in the ferroelectric hafnia are well-known challenges in early-stage device reliability. We found that the phenomena even occur in the bilayer, which is composed of the hafnia and the dielectrics. The phenomena are known to be affected mainly by oxygen vacancies of hafnia. Dielectric layers, which are often metal oxides, are also prone to be affected by oxygen vacancies. To study the effect of the dielectric layer on the wake-up and the split-up phenomena, we fabricated ferroelectric thin-film capacitors with dielectric layers of various thicknesses and measured their field-cycling behaviors. We found that the movement of oxygen vacancies in the dielectric layer was predominantly affected by the polarization state of the ferroelectric layer. In addition, the mechanism of the field-cycling behavior in the bilayer is similar to that in ferroelectric thin films. Our results can be applied in ferroelectric applications that use dielectric layers.

Ferroelectric Transistors for Memory and Neuromorphic Device Applications

Ferroelectric materials have been intensively investigated for high-performance nonvolatile memory devices in the past decades, owing to their nonvolatile polarization characteristics. Ferroelectric memory devices are expected to exhibit lower power consumption and higher speed than conventional memory devices. However, non-complementary metal-oxide-semiconductor (CMOS) compatibility and degradation due to fatigue of traditional perovskite-based ferroelectric materials have hindered the development of high-density and high-performance ferroelectric memories in the past. The recently developed hafnia-based ferroelectric materials have attracted immense attention in the development of advanced semiconductor devices. Because hafnia is typically used in CMOS processes, it can be directly incorporated into current semiconductor technologies. Additionally, hafnia-based ferroelectrics show high scalability and large coercive fields that are advantageous for high-density memory devices. This review summarizes the recent developments in ferroelectric devices, especially ferroelectric transistors, for next-generation memory and neuromorphic applications. First, the types of ferroelectric memories and their operation mechanisms are reviewed. Then, issues limiting the realization of high-performance ferroelectric transistors and possible solutions are discussed. The experimental demonstration of ferroelectric transistor arrays, including 3D ferroelectric NAND and its operation characteristics, are also reviewed. Finally, challenges and strategies toward the development of next-generation memory and neuromorphic applications based on ferroelectric transistors are outlined.

Proton-Gated Synaptic Transistors, Based on an Electron-Beam Patterned Nafion Electrolyte

Neuromorphic processors using artificial neural networks are the center of attention for energy-efficient analog computing. Artificial synapses act as building blocks in such neural networks for parallel information processing and data storage. Herein we describe the fabrication of a proton-gated synaptic transistor using a Nafion electrolyte thin film, which is patterned by electron-beam lithography (EBL). The device has an active channel of indium-zinc-oxide (IZO) between the source and drain electrodes, which shows Ohmic behavior with a conductance level on the order of 100 μS. Under voltage applications to the gate electrode, the channel conductance is changed due to the injection and extraction of protons between the IZO channel and the Nafion electrolyte, emulating various synaptic functions with short-term and long-term plasticity. When positive (negative) gate voltage pulses are consecutively applied, the device exhibits long-term potentiation (depression) at the same number of steps as the number of input pulses. Based on these characteristics, an artificial neural network using this transistor shows ∼84% image recognition accuracy for handwritten digits. The subject transistor also successfully mimics paired-pulse facilitation and depression, Hebbian spike-timing-dependent plasticity, and Pavlovian associative learning followed by extinction activities. Finally, dynamical pattern image memorization is demonstrated in a 5 × 5 array of these synaptic transistors. The results indicate that EBL patternable Nafion electrolytes have great potential for use in the fabrication and circuit-level integration of synaptic devices for neuromorphic computing applications.

Highly-scaled and fully-integrated 3-dimensional ferroelectric transistor array for hardware implementation of neural networks

Hardware-based neural networks (NNs) can provide a significant breakthrough in artificial intelligence applications due to their ability to extract features from unstructured data and learn from them. However, realizing complex NN models remains challenging because different tasks, such as feature extraction and classification, should be performed at different memory elements and arrays. This further increases the required number of memory arrays and chip size. Here, we propose a three-dimensional ferroelectric NAND (3D FeNAND) array for the area-efficient hardware implementation of NNs. Vector-matrix multiplication is successfully demonstrated using the integrated 3D FeNAND arrays, and excellent pattern classification is achieved. By allocating each array of vertical layers in 3D FeNAND as the hidden layer of NN, each layer can be used to perform different tasks, and the classification of color-mixed patterns is achieved. This work provides a practical strategy to realize high-performance and highly efficient NN systems by stacking computation components vertically.

Dielectric-Engineered High-Speed, Low-Power, Highly Reliable Charge Trap Flash-Based Synaptic Device for Neuromorphic Computing beyond Inference

The coming of the big-data era brought a need for power-efficient computing that cannot be realized in the Von Neumann architecture. Neuromorphic computing which is motivated by the human brain can greatly reduce power consumption through matrix multiplication, and a device that mimics a human synapse plays an important role. However, many synaptic devices suffer from limited linearity and symmetry without using incremental step pulse programming (ISPP). In this work, we demonstrated a charge-trap flash (CTF)-based synaptic transistor using trap-level engineered Al₂O₃/Ta₂O₅/Al₂O₃ gate stack for successful neuromorphic computing. This novel gate stack provided precise control of the conductance with more than 6 bits. We chose the appropriate bias for highly linear and symmetric modulation of conductance and realized it with very short (25 ns) identical pulses at low voltage, resulting in low power consumption and high reliability. Finally, we achieved high learning accuracy in the training of 60000 MNIST images.

Recent advances in neuromorphic transistors for artificial perception applications: FOCUS ISSUE REVIEW

Conventional von Neumann architecture is insufficient in establishing artificial intelligence (AI) in terms of energy efficiency, computing in memory and dynamic learning. Delightedly, rapid developments in neuromorphic computing provide a new paradigm to solve this dilemma. Furthermore, neuromorphic devices that can realize synaptic plasticity and neuromorphic function have extraordinary significance for neuromorphic system. A three-terminal neuromorphic transistor is one of the typical representatives. In addition, human body has five senses, including vision, touch, auditory sense, olfactory sense and gustatory sense, providing abundant information for brain. Inspired by the human perception system, developments in artificial perception system will give new vitality to intelligent robots. This review discusses the operation mechanism, function and application of neuromorphic transistors. The latest progresses in artificial perception systems based on neuromorphic transistors are provided. Finally, the opportunities and challenges of artificial perception systems are summarized.

Stretchable Temperature-Responsive Multimodal Neuromorphic Electronic Skin with Spontaneous Synaptic Plasticity Recovery

Multimodal electronic skin devices capable of detecting multimodal signals provide the possibility for health monitoring. Sensing and memory for temperature and deformation by human skin are of great significance for the perception and monitoring of physiological changes of the human body. Electronic skin is highly expected to have similar functions as human skin. Here, by implementing intrinsically stretchable neuromorphic transistors with mechanoreceptors and thermoreceptors in an array, we have realized stretchable temperature-responsive multimodal neuromorphic electronic skin (STRM-NES) with both sensory and memory functions, in which synaptic plasticity can be modulated by multiple modalities, in situ temperature variations, and stretching deformations. Temperature-responsive functions, spontaneous recovery, and temperature-dependent multitrial learning are proposed. Furthermore, a stretchable temperature neuromorphic array composed of multiple fully functional subcells is demonstrated to identify temperature distributions and variations at different regions and conditions after various strains of skin. The STRM-NES has temperature- and strain-responsive neuromorphic functions, excellent self-healing, and reusable capability, showing similar abilities as human skin to sense, transmit, memory, and recovery from external stimuli. It is expected to facilitate the development of wearable electronics, intelligent robotics, and prosthetic applications.