Molecular Crystal Memristor
Preprint
- 12 March 2025
- preprint
- Published by Springer Nature in Research Square
Abstract
The emergence of memristors offers a revolutionary solution for achieving in-memory computing at the hardware level. However, existing memristors suffer from the inherent channel materials damage during cyclical resistive switching, rendering excessive energy consumption and poor endurance. Herein we contribute an innovative Molecular Crystal Memristor, of which the representative channel material, Sb2O3, possesses a unique molecular crystal structure where molecular cages are interconnected via van der Waals forces. This distinct arrangement allows for the migration of ions through intermolecular spaces with minimal energy consumption while intactly preserving the pristine crystal structure after cyclical switching. Our Molecular Crystal Memristor thus exhibits incredibly low energy consumption of 26 zJ, with prominent endurance surpassing 109 cycles. Moreover, the memristors consistently demonstrate reconfigurable non-volatile and volatile resistive switching behavior, within a wide device scale from micrometer to nanometer. We also demonstrated their scalability by fabricating massive Mb-level crossbar arrays on an 8-inch wafer. Finally, we realized the Reservoir Computing functionality on a single CMOS-integrated chip made from our Molecular Crystal Memristor, achieving 100% accuracy in dynamic vision recognition. These breakthroughs herald the practical application of memristors in non-von Neumann computing architectures, with the potential to reshape the future of computing technologies.Keywords
All Related Versions
- Version 1, 2025-03-12, Research Square
- Published version: Nature Nanotechnology, 20 (11), 1641.
This publication has 53 references indexed in Scilit:
- Three‐Dimensional Reconstruction of Conductive Filaments in HfOx‐Based MemristorAdvanced Materials, 2023
- A Photoelectric Spiking Neuron for Visual Depth PerceptionAdvanced Materials, 2022
- High‐Performance Memristor Based on 2D Layered BiOI Nanosheet for Low‐Power Artificial Optoelectronic SynapsesAdvanced Functional Materials, 2022
- Effect of the Threshold Kinetics on the Filament Relaxation Behavior of Ag‐Based Diffusive MemristorsAdvanced Functional Materials, 2021
- Scalable Van der Waals Encapsulation by Inorganic Molecular CrystalsAdvanced Materials, 2021
- VASPKIT: A user-friendly interface facilitating high-throughput computing and analysis using VASP codeComputer Physics Communications, 2021
- Low‐Power Memristive Logic Device Enabled by Controllable Oxidation of 2D HfSe2 for In‐Memory ComputingAdvanced Science, 2021
- Performance‐Enhancing Selector via Symmetrical Multilayer DesignAdvanced Functional Materials, 2019
- Atomically Thin Femtojoule Memristive DeviceAdvanced Materials, 2017
- Eliminating Negative‐SET Behavior by Suppressing Nanofilament Overgrowth in Cation‐Based MemoryAdvanced Materials, 2016