Randomized trial demonstrates enhanced performance in memristors, indicating potential for neuromorphic computing.
With the increasing demand for low-power, high-reliability integrated storage and computing hardware in neuromorphic computing, traditional single-layer memristors have long faced challenges in balancing key performances such as operating voltage, switching ratio, and cycling stability. This study successfully developed MoO3/HfO2 heterostructure memristors and systematically investigated their performance and underlying mechanisms. The device exhibits remarkable performance, including a low operating voltage (∼0.2 V), a high switching ratio (∼105), and good cycling stability. In addition, the resistive switching behavior is driven by the controlled formation and rupture of silver conductive filaments, verified by conductive atomic force microscopy (CAFM). According to theoretical calculations, it is found that the MoO3 layer accelerates filament dynamics through its fast ion migration ability, significantly lowering the operating barrier. Meanwhile, the HfO2 layer restricts ion diffusion and stabilizes the remaining filament segments, effectively anchoring filament growth in the MoO3 layer to ensure consistent and reliable switching behavior. This synergistic strategy of high-speed ion migration and physical filament restriction proposed in this work not only offers an approach to overcoming the performance limitations of traditional single-layer memristors but also deepens the understanding of the role of heterointerface engineering in modulating memristive behavior.
No takes yet. Share an insight, caveat, or question.
Zhou et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: