Perovskite memristors for neuromorphic applications
Loading...
Date
2026-10-07
Authors
Gion, Kalemai
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This research explores the development of memristive systems for neuromorphic computing by bridging biological synaptic plasticity with nanoscale device physics. The study investigates transition metal oxides, specifically addressing a stoichiometry paradox in molybdenum oxide devices where fully stoichiometric matrices unexpectedly yield superior resistive switching and high ON/OFF ratios by tightly confining filament percolation. To optimize analog update linearity and suppress stochastic noise, the research leverages space-charge-limited current (SCLC) and quantum mechanical tunneling via polymer interlayers like PMMA. Furthermore, the work evaluates halide perovskites, capitalizing on their unique mixed ionic-electronic transport for neuromorphic sensing. It mitigates inherent environmental and operational degradation through engineered 2D/3D heterojunctions and structural passivation strategies. By integrating theoretical defect chemistry with extensive experimental characterization, this framework advances the scalability and reliability of artificial synapses for next-generation hardware architectures.
Description
Keywords
Memristor, Perovskites, Neuromorphic applications