Two‐Dimensional SnS 2 Interlayers Enable Vacancy‐Guided Ion Migration for Stable Diffusive Memristors and Bioinspired Nociception
ABSTRACT The realization of sophisticated autonomous interaction in humanoid robots is currently constrained by the complexity and stochasticity of conventional sensory hardware. Volatile memristors provide a promising biomimetic alternative; however, their performance is often limited by uncontrolled ionic dynamics in disordered oxides. Here, we report two‐dimensional SnS 2 interlayers as an effective platform for vacancy‐guided ion migration in oxide‐based diffusive memristors. A solution‐processed SnS 2 layer with tunable sulfur vacancy density is introduced into an Ag/TaO x /SnS 2 /ITO heterostructure to regulate filament formation. Under electrical bias, sulfur vacancies of SnS 2 layer locally enhance the electric field and provide deterministic pathways for Ag + migration. This enables atomic‐scale confinement of conductive filaments, resulting in highly uniform threshold switching (variability ≈ 10.9%), ultralow switching energy (∼ 10.9 pJ), and stable operation over 5,000 cycles. Furthermore, modulation of the vacancy landscape allows dynamic reconfiguration of switching polarity, enabling a transition from unidirectional to bidirectional volatile switching. Beyond improved electrical performance, the devices reproduce key nociceptive behaviors and can be integrated into an 8 × 8 crossbar array to achieve spatially resolved perception of mechanical stimuli.
Authors
- Gangri Cai (ORCID: https://orcid.org/0000-0002-3021-8452)
- Zening Gao
- Jingzhou Shi (ORCID: https://orcid.org/0009-0008-6758-3198)
- Xianjin Feng (ORCID: https://orcid.org/0000-0001-5303-5678)
- Xinming Ma (ORCID: https://orcid.org/0009-0002-3263-7818)
- Jinshi Zhao
Institutions
- Tianjin University of Technology (CN)
- Shandong University (CN)
Publication Details
- Journal
- Angewandte Chemie International Edition
- Published
- 2026-09-21
- DOI
- https://doi.org/10.1002/anie.6605459
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00