Search

Memristor, processing data storage, and logic operation all‐in‐one, is expected to create a new era of neuromorphic computing and digital logic. Here, this work demonstrates a BiFeO3‐based memristor with Ti‐doped indium tin oxide as the top electrode material, exhibiting high metrics such as a switching voltage of ≈0.15 V, coefficients of variations of low resistance state of ≈0.46% and a large on/off ratio of ≈103. What is more, the device exhibits a power consumption as low as ≈1.02 µW in set process with a self‐compliance current. To elucidate the resistive switching behavior, the role of TiInSnO electrode and the corresponding current conduction mechanism have been thoroughly investigated. In this end, the effect of Ti‐doping on the resistance behavior is demonstrated based on the current fitting analysis and first‐principles calculations based on density functional theory. It is clarified that the doped Ti ions may trap oxygen ions to concentrate around the interface between resistance switching layer and electrode, resulting in more stable and uniform performance. The findings suggest that, with TiInSnO film as the novel electrode of memristors, the present study can offer a significantly promising route to implement low‐power, highly uniform information storage, and neuromorphic computing technology. [ABSTRACT FROM AUTHOR]

In recent years, the potential of two-dimensional layered metal dichalcogenides for gas-sensing applications has been explored. Both defect engineering and noble metal modification are powerful strategies to achieve higher gas sensitivity. In this paper, a combined experimental and theoretical study on a SnS2-based sensor is performed to demonstrate the effect of individual or cooperative modulation of S-vacancy (Vs) defect and Ag nanoparticles (nano-Ag) on its gas-sensing performance. Gas-sensing measurements indicate that Ag/Vs-SnS2 shows the strongest response upon exposure to ethanol and a response as high as 94.7% when the ethanol concentration was 15 ppm. Especially, the limit of detection value of Ag/Vs-SnS2 calculated based on the low concentration response data is 2.12 ppb, implying its ultra-sensitive characteristic to ethanol vapor. Meanwhile, the response time and recovery time to 5 ppm ethanol are 7 and 45 s, respectively, and the response error is about 1.8%. It is clarified theoretically that both Vs modulation and Ag modification could enhance the adsorption strength of ethanol molecules on the SnS2 surface, and the SnS2 with co-modification of Vs and nano-Ag (Ag/Vs-SnS2) exhibits the strongest adsorption for the ethanol molecule. The potential mechanism for the enhanced response of nano-Ag/Vs-SnS2 was also further analyzed and demonstrated. This work provides an efficient strategy for developing high-quality gas sensors capable of sensitively detecting ethanol at room temperature. [ABSTRACT FROM AUTHOR]

The sensations of touch and pain are fundamental components of our daily life, which can transport vital information about the surroundings and provide protection to our bodies. In this study, the transmission process of sensing pressure stimuli to dorsal root neurons (nociceptors) was simulated using electronic devices. In this regard, we proposed and experimentally demonstrated a biomimetic nociceptor system with tactile perception. In this system, the sensing paper as E-skin simulates the biological skin to sense external pressure stimulation and generate electrical signals, while the threshold switching memristor simulates the biological nociceptor to receive and process the receptor signals. The W/VO2/Pt memristor exhibits all key features of nociceptors including threshold, relaxation, "no adaptation" and sensitization phenomena of allodynia and hyperalgesia. The E-skin shows high sensitivity and a broad sensing range and is capable of monitoring different human movements and physiological signals. With the bio-inspired artificial tactile nociceptive system, the threshold and sensitization properties under pressure stimuli are obtained successfully. Notably, this system could be used as an artificial tactile alarm system to demonstrate the potential applicability of humanoid robots. Thus, the present work is of great significance to the development of hardware architecture in artificial intelligence systems and replacement neuroprosthetics. [ABSTRACT FROM AUTHOR]

The sensations of touch and pain are fundamental components of our daily life, which can transport vital information about the surroundings and provide protection to our bodies. In this study, the transmission process of sensing pressure stimuli to dorsal root neurons (nociceptors) was simulated using electronic devices. In this regard, we proposed and experimentally demonstrated a biomimetic nociceptor system with tactile perception. In this system, the sensing paper as E-skin simulates the biological skin to sense external pressure stimulation and generate electrical signals, while the threshold switching memristor simulates the biological nociceptor to receive and process the receptor signals. The W/VO 2 /Pt memristor exhibits all key features of nociceptors including threshold, relaxation, "no adaptation" and sensitization phenomena of allodynia and hyperalgesia. The E-skin shows high sensitivity and a broad sensing range and is capable of monitoring different human movements and physiological signals. With the bio-inspired artificial tactile nociceptive system, the threshold and sensitization properties under pressure stimuli are obtained successfully. Notably, this system could be used as an artificial tactile alarm system to demonstrate the potential applicability of humanoid robots. Thus, the present work is of great significance to the development of hardware architecture in artificial intelligence systems and replacement neuroprosthetics.