Dual-Mode Resistive Switching in Nanometer-Thick CeO <sub>2</sub> -Based Memristors for Flexible and Neuromorphic Applications
Memristive devices are promising candidates for next-generation electronic and neuromorphic systems due to their tunable resistive switching at the nanoscale. Here, we systematically investigated thickness-, current-, and phase-dependent switching in Ag/CeO2/ITO memristors. XRD, FESEM, and HRTEM analyses reveal the evolution of CeO2 from an amorphous to a crystalline state with increasing annealing temperature, resulting in uniform, dense nanometer-thick films with well-defined lattice fringes characteristic of the cubic fluorite structure. By tuning film thickness, compliance current (ICC), and annealing, devices exhibit a transition from forming-free volatile to forming-dependent volatile and nonvolatile switching. In amorphous films, switching is dominated by Ag filament formation and rupture, whereas crystalline films display a hybrid mechanism involving Ag+ migration and oxygen vacancy modulation. Mechanical reliability tests confirm excellent bending endurance with minimal variation under stress. We experimentally demonstrate that both neuronal and synaptic functionalities can be realized within a single amorphous CeO2 memristor: operating below the threshold limit, Ith (10–4 A) yields volatile neuronal-like spiking, while above the Ith, nonvolatile synaptic plasticity is observed. The ability to engineer volatile and nonvolatile modes within the same nanometer-thick CeO2 device highlights its multifunctionality for flexible memory and adaptive neuromorphic electronic applications, providing insights into structure–property–function relationships in oxide-based resistive switching systems.
Authors
- S. Angappane (ORCID: https://orcid.org/0000-0001-8475-0404)
- Bhagyalakshmi Kuzhimbattil (ORCID: https://orcid.org/0000-0002-3840-1331)
Institutions
- Centre for Nano and Soft Matter Sciences (IN)
- Academy of Scientific and Innovative Research (IN)
Publication Details
- Journal
- ACS Applied Nano Materials
- Published
- 2025-12-06
- DOI
- https://doi.org/10.1021/acsanm.5c05303
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00