Artificial synapses and neurons are crucial milestones for neuromorphic computing hardware, and memristors with resistive and threshold switching characteristics are regarded as the most promising candidates for the construction of hardware neural networks. However, most of the memristors can only operate in one mode, that is, resistive switching or threshold switching, and distinct memristors are required to construct fully memristive neuromorphic computing hardware, making it more complex for the fabrication and integration of the hardware. Herein, we propose a flexible dual‐mode memristor array based on core–shell CsPbBr3@graphdiyne nanocrystals, which features a 100% transition yield, small cycle‐to‐cycle and device‐to‐device variability, excellent flexibility, and environmental stability. Based on this dual‐mode memristor, homo‐material‐based fully memristive neuromorphic computing hardware—a power‐free artificial nociceptive signal processing system and a spiking neural network—are constructed for the first time. Our dual‐mode memristors greatly simplify the fabrication and integration of fully memristive neuromorphic systems.

Glucose uptake assay-guided fractionations of the methanol extract of Schisandra chinensis led to the isolation of the dibenzocyclooctadiene lignans: gomisin J (1), gomisin N (2), wuweizisu B (3), wuweizisu C (4), gomisin C (5), gomisin D (6), (+)-schisandrin A (7), schisandrin C (8), schisandrol A (9), gomisin H (10), angeloylgomisin H (11), gomisin A (12), and schizandrin (13). Among these, 1, 2, 7, and 8 significantly improved basal glucose uptake in HepG2 cells. Their improving effects were concentration-dependent. Compound 2 exhibited a stronger effect than that of rosiglitazone, which has been used as an anti-diabetic drug. The results suggest that these lignans may partially contribute to the anti-diabetic activity of Fructus Schisandrae Chinensis in traditional use by stimulating the glucose uptake into peripheral tissue, which may be responsible for reducing the level of blood glucose in circulation. Thus, these findings show the potential of these lignans for development as hypoglycemic drugs.

The integration and cooperation of nociceptors, neurons and synapses in the biological nervous system empower humans to efficiently perceive and process noxious information for avoiding dangers. Inspired by biological nervous systems, current artificial demonstrations include electrolyte-gate transistors, electrochemical metallization resistive switching devices and halide perovskite-based memristors. However, these devices suffer from integration difficulties and instability issues. Herein, we introduce a complementary metal oxide semiconductor (CMOS)-compatible simple and stable Pt/V 2 O 5 /Pt sandwich structure and carefully construct and modulate the suboxide V 2 O 5−x and Mott VO 2 nanochannels in the layered V 2 O 5 matrix to simulate brain-like processing and nervous pain perception functions, respectively. Simulation results demonstrate that the recognition accuracy of handwritten digits reaches 80% after only 5 training epochs and 89% after 52 epochs in a convolutional neural network based on the V 2 O 5−x nanochannel synaptic device. The nociceptor with all key characteristics is perfectly imitated based on the VO 2 nanochannel threshold switching device. Especially, an ultralow threshold level of 0.4 V and sub-millisecond incubation time are observed in the nociceptor simulation, which could be needed in special injury situations. The proposed drift and Mott nanochannels in one device hold a tremendous potential as synaptic and nociceptive emulators for artificial intelligence systems.

No abstract

No abstract