Neuromorphic Chip Integrated with an LSI and Amorphous-metal-oxide Semiconductor Thin-film Synapse Devices

Artificial intelligences are promising in future societies, and neural networks are typical technologies with the advantages such as self-organization, self-learning, parallel distributed computing, and fault tolerance, but their size and power consumption are large. Neuromorphic systems are biomimetic systems from the hardware level, with the same advantages as living brains, especially compact size, low power, and robust operation, but some well-known ones are non-optimized systems, so the above benefits are only partially gained, for example, machine learning is processed elsewhere to download fixed parameters. To solve these problems, we are researching neuromorphic systems from various viewpoints. In this study, a neuromorphic chip integrated with an LSI and amorphous-metal-oxide semiconductor (AOS) thin-film synapse devices has been developed. The neuron elements are digital circuit, which are made in an LSI, and the synapse devices are analog devices, which are made of the AOS thin film and directly integrated on the LSI. This is the world's first hybrid chip where neuron elements and synapse devices of different functional semiconductors are integrated, and local autonomous learning is utilized, which becomes possible because the AOS thin film can be deposited without heat treatment and there is no damage to the underneath layer, and has all advantages of neuromorphic systems.

Fabrication of amorphous metal oxide/p-BiVO4 photocathode: understanding the role of entropy for reducing nitrate to ammonia

The controllable design of chemical microenvironment with the expected thermodynamics and kinetics for boosting catalytic activity and selectivity is still challenging. Herein, an amorphous metal oxide (A-MxOy) was employed to...

Heterostructure of core-shell IrCo@IrCoOx as efficient and stable catalysts for oxygen evolution reaction

Although researches on non-noble metal electrocatalysts have been made some progress recently, their performance in proton exchange membrane water electrolyzer (PEMWE) is still incomparable to that of noble-metal-based catalysts. Therefore, it is a more practical way to improve the utilization of precious metals in electrocatalysts for oxygen evolution reaction (OER) in the acidic medium. Herein, nanostructured IrCo@IrCoOx core-shell electrocatalysts composed of IrCo alloy core and IrCoOx shell were synthesized through a simple colloidally synthesis and calcination method. As expected, the hybrid IrCo-200 NPs with petal-like morphology show the best OER activities in acidic electrolytes. They deliver lower overpotential and better electrocatalytic kinetics than pristine IrCo alloy and commercial Ir/C, reaching a low overpotential (j = 10 mA/cm2) of 259 mV (vs. RHE) and a Tafel slope of 59 mV dec−1. The IrCo-200 NPs displayed robust durability with life time of about 55 h in acidic solution under a large current density of 50 mA/cm2. The enhanced electrocatalytic activity may be associated with the unique metal/amorphous metal oxide core-shell heterostructure, allowing the improved charge transferability. Moreover, the *OH-rich amorphous shell functions as the active site for OER and prevents the further dissolution of the metallic core and thus ensures high stability.

Scalable High-Speed Hybrid Complementary Integrated Circuits based on Solution-Processed Organic and Amorphous Metal Oxide Semiconductors

Solution-processed single-crystal organic semiconductors (OSCs) and amorphous metal oxide semiconductors (MOSs) are promising for high-mobility, p- and n-channel thin-film transistors (TFTs), respectively. Organic−inorganic hybrid complementary circuits hence have great potential to satisfy practical requirements; however, some chemical incompatibilities between OSCs and MOSs, such as heat and chemical resistance, conventionally make it difficult to rationally integrate TFTs based on solution-processed OSC and MOS into the same substrates. In this work, we achieved a rational integration method based on the solution-processed semiconductors by carefully managing the device configuration and the deposition and patterning techniques from materials point of view. The balanced high performances as well as the uniform fabrication of the TFTs led to densely integrated five-stage ring oscillators with the stage propagation delay of 1.3 µs, which is the fastest operation among ever reported complementary ring oscillators based on solution-processed semiconductors.

High-Entropy Oxides: Advanced Research on Electrical Properties

The concept of “high entropy” was first proposed while exploring the unknown center of the metal alloy phase diagram, and then expanded to oxides. The colossal dielectric constant found on the bulk high-entropy oxides (HEOs) reveals the potential application of the high-entropy oxides in the dielectric aspects. Despite the fact that known HEO thin films have not been reported in the field of dielectric properties so far, with the high-entropy effects and theoretical guidance of high entropy, it is predictable that they will be discovered. Currently, researchers are verifying that appropriately increasing the oxygen content in the oxide, raising the temperature and raising the pressure during preparation have an obvious influence on thin films’ resistivity, which may be the guidance on obtaining an HEO film large dielectric constant. Finally, it could composite a metal–insulator–metal capacitor, and contribute to sensors and energy storage devices’ development; alternatively, it could be put into application in emerging thin-film transistor technologies, such as those based on amorphous metal oxide semiconductors, semiconducting carbon nanotubes, and organic semiconductors.

Scalable High-Speed Hybrid Complementary Integrated Circuits based on Solution-Processed Organic and Inorganic Transistors

Printed electronics offer a cost-efficient way to realise flexible electronic devices. The combined use of p-type and n-type semiconductors would yield silicon-like integrated circuits with low power consumption and stability. However, printing complementary circuits is challenging due to a lack of suitable material systems. To counter this, we employed a hybrid system to integrate p-type organic semiconductors (OSCs) and n-type amorphous metal oxide semiconductors (MOSs). These damage-free patterned OSC- and MOS-based thin-film transistors with improved process durability allowed the fabrication of hybrid complementary circuits on flexible substrates. These inverters functioned well even after exposure to air for 5 months. A large noise margin and power gain of 38 were realised with a supply voltage as low as 7 V. Furthermore, a five-stage ring oscillator with a stage propagation delay of 1.3 µs was achieved, which is the fastest operation ever reported for printed, flexible complementary inverters.

Amorphous metal oxide semiconductor thin film, analog memristor, and autonomous local learning for neuromorphic systems

Artificial intelligence is a promising concept in modern and future societies. Presently, software programs are used but with a bulky computer size and large power consumption. Conversely, hardware systems named neuromorphic systems are suggested, with a compact computer size and low power consumption. An important factor is the number of processing elements that can be integrated. In the present study, three decisive technologies are proposed: (1) amorphous metal oxide semiconductor thin films, one of which, Ga–Sn–O (GTO) thin film, is used. GTO thin film does not contain rare metals and can be deposited by a simple process at room temperature. Here, oxygen-poor and oxygen-rich layers are stacked. GTO memristors are formed at cross points in a crossbar array; (2) analog memristor, in which, continuous and infinite information can be memorized in a single device. Here, the electrical conductance gradually changes when a voltage is applied to the GTO memristor. This is the effect of the drift and diffusion of the oxygen vacancies (Vo); and (3) autonomous local learning, i.e., extra control circuits are not required since a single device autonomously modifies its own electrical characteristic. Finally, a neuromorphic system is assembled using the abovementioned three technologies. The function of the letter recognition is confirmed, which can be regarded as an associative memory, a typical artificial intelligence application.