Achieving analog and digital resistive switching for quasi-2D perovskite memristors by compositional regulation and crystalline manipulation
With the advent of the information age, the demand for memristors in neuromorphic circuits and artificial intelligence has become increasingly urgent. In this Letter, analog- and digital-type resistive switching is concurrently obtained in quasi-2D perovskite memristors by regulating elemental composition and crystallinity. For the high Pb sample treated with MACl, continuous enhancement and suppression of analog-type conductance are achieved under direct current voltage sweep and voltage pulse stimulation, which shows 62.2% enhancement and 68.4% inhibition after long-term synaptic plasticity. The analog-type memristor also presents good spike voltage-dependent and frequency-dependent plasticities with a paired-pulse facilitation index of 55.5%. In contrast, for the low Pb sample treated with MACl, the lattice distortion and crystal defects are further decreased, leading to a sudden change of digital-type resistance in the current−voltage curve. The digital-type memristor has very good non-volatile memory performance with a resistance-to-switch ratio of 5 × 103, an endurance of more than 2100 cycles, a retention characteristic of more than 104 s, a statistical set voltage of 0.55 V, and a reset voltage of −1 V. This study elucidates analog and digital resistive switching mechanism in quasi-2D perovskites and demonstrates their significant potential and prospects in neuronal circuit and storage-computing technologies.
Authors
- Wen Lei (ORCID: https://orcid.org/0000-0001-5473-5350)
- Chenhui Su
- Cuiying Hu
- Shuyan Fang
- Xuemiao Wen
- Xingui Tang
- Tianjiao Liu (ORCID: https://orcid.org/0000-0002-6430-0163)
- Y. F. Fan
- Lintao Hou
- Dongping Yang
- Guanglu Lin
- Zhenfang Tang
- Yibo Peng
- Renqiang Yang
- Zhichao Chen
Institutions
- Guangdong University of Technology (CN)
- Jinan University (CN)
- China Mobile (China) (CN)
- Jianghan University (CN)
Publication Details
- Journal
- Applied Physics Letters
- Published
- 2026-01-26
- DOI
- https://doi.org/10.1063/5.0276050
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00