Unveiling Phosphorus Vacancy-Mediated Resistive Switching in Cu/HfO2/BP/Pt Memristors: Experimental and First-Principles Investigations
Abstract Memristors integrating 2D materials are promising for neuromorphic computing, yet their microscopic resistive switching mechanisms require further clarification. In this study, a Cu/HfO2/BP/Pt memristor was fabricated to investigate its electrical characteristics and conductive filament (CF) dynamics. The device exhibits compliance-dependent behavior, enabling unipolar volatile switching at 100 μA and bipolar nonvolatile switching at 200 μA. In nonvolatile mode, it demonstrates a concentrated switching voltage distribution (Vset ≈ −1.491 V, Vreset ≈ 1.669 V), stable endurance exceeding 150 cycles, and an ON/OFF ratio of ∼104. The low- and high-resistance states follow ohmic conduction (a hybrid ohmic conduction of electron capture/decapture under negative bias) and Schottky emission, respectively. Furthermore, first-principles CI-NEB calculations indicate that intrinsic phosphorus vacancies (Vp) within the BP layer facilitate directional Cu+ migration (1.54 eV barrier) and spontaneous Pt atom trapping (0.1 eV barrier). This defect-mediated interfacial interaction promotes stable orbital hybridization, anchoring localized CFs. This work provides experimental and theoretical insights for designing oxide/2D material-based neuromorphic hardware.
Authors
- Huajun Sun (ORCID: https://orcid.org/0000-0003-0755-5247)
- Zuhao Shi (ORCID: https://orcid.org/0000-0003-4886-0328)
- Ouwen Zhang (ORCID: https://orcid.org/0000-0003-3102-1077)
- Niannian Yu (ORCID: https://orcid.org/0000-0002-2336-569X)
- Daiyang Jiang (ORCID: https://orcid.org/0009-0007-9185-7016)
- Changtian Cao
- Wenhao Li
- Xiangshui Miao
- Zihao Yan
Institutions
- Wuhan University of Technology (CN)
- Huazhong University of Science and Technology (CN)
Publication Details
- Journal
- The Journal of Physical Chemistry C
- Published
- 2026-10-09
- DOI
- https://doi.org/10.1021/acs.jpcc.6c05460
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00