Conductive dendrite engineering of single-crystalline two-dimensional dielectric memristors
Ultralow-power non-volatile memristors are key elements in electronics. Generally, power reduction of memristors compromises data retention, a challenge known as the "power-retention dilemma," due to the stochastic formation of conductive dendrites in resistive-switching materials. Here, we report the results of conductive dendrite engineering in single-crystalline two-dimensional (2D) dielectrics in which directional control of filamentary distribution is possible. We find that the single-vacancy density (n<sub>SV</sub>) of single-crystalline hexagonal boron nitride (h-BN) plays an essential role in regulating conductive dendrite growth, supported by scanning joule expansion microscopy (SJEM). With optimized n<sub>SV</sub>, random dendrite growth is largely limited, and electrons hop between the neighboring Ag nanoclusters in vertical channels. The corresponding model was established to probe the relationship between n<sub>SV</sub> and memristor operating voltage. The conductive channel confinement in the vertical orientation contributes to long-retention non-volatile memristors with ultralow switch voltages (set: 26 mV; reset: -135 mV), excellent power efficiency (4 fW standby and a switching energy of 72 pJ) while keeping a high on/off resistance ratio of 10<sup>8</sup>. Even at a record-low compliance current of 10 nA, memristors retains very robust non-volatile, multiple resistive states with an operating voltage less than 120 mV (the per-transition power low as 900 pW).
Authors
- Yuda Zhao
- Neng Wan (ORCID: https://orcid.org/0000-0003-1876-7149)
- Xingyu Zhai (ORCID: https://orcid.org/0000-0002-3035-9407)
- Baoshi Qiao (ORCID: https://orcid.org/0000-0002-5095-1344)
- Yu Kang (ORCID: https://orcid.org/0000-0002-8706-3252)
- Bin Yu (ORCID: https://orcid.org/0000-0003-0237-6311)
- Haohan Chen (ORCID: https://orcid.org/0000-0002-6515-4954)
- Yang Xu (ORCID: https://orcid.org/0000-0002-4943-6775)
- Huan Hu (ORCID: https://orcid.org/0000-0003-2181-9501)
- Wenchao Chen
- Chai Yang
- Quan Yang
- Ming Tian (ORCID: https://orcid.org/0000-0001-7949-0599)
- Zheng Bian
Institutions
- Hong Kong Polytechnic University (HK)
- Zhejiang University-University of Edinburgh Institute (CN)
- Zhejiang University of Technology (CN)
- Southeast University (CN)
Publication Details
- Journal
- The Innovation
- Published
- 2025-03-18
- DOI
- https://doi.org/10.1016/j.xinn.2025.100885
- Citations
- 8
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 5.84