journal article·2026-02Open access

Atomic‐Scale Mechanisms of Multi‐Resistance States in HfO x ‐Based RRAM: Evolution of Atomic Electric Fields and Oxygen Vacancies

Wen Sun, Yuyan Wang, Ruofei Hu, Yuyao Lu, Jun Xu, Xinyi Li, Bin Gao, He Qian, Jianshi Tang, Huaqiang Wu

Advanced Science· Volume 13· Issue 10

Abstract

Abstract The stability of multilevel resistive switching in HfO x ‐based RRAM is crucial for enhancing matrix‐vector multiplication efficiency in computing‐in‐memory architectures, yet precise control over conductive filament formation is limited by an incomplete understanding of oxygen vacancy dynamics. Using advanced transmission electron microscopy (TEM) techniques, this study reveals key correlations among crystallographic orientation, oxygen vacancy distribution, and resistive switching mechanisms. Distinct m‐phase orientations govern resistance states: the High Resistance State (HRS) is characterized by a [101] orientation without oxygen vacancies; the Medium and Low Resistance States (MRS/LRS) exhibit a [011] orientation with selectively formed vacancies. Thermally driven m‐phase urotation ([101] ↔ [011]) facilitates oxygen vacancy migration, with vacancies preferentially occupying specific lattice sites. This alters the atomic electric fields around Hf atoms and modifies electron transport pathways. The distribution of vacancies directly controls conduction mechanisms: Schottky emission in HRS, Poole‐Frenkel emission in MRS, and Ohmic conduction in LRS, corresponding to increasing vacancy concentrations. These findings demonstrate that resistive states emerge from coupled processes: crystallographic rotation and vacancy formation reconstruct atomic electric fields, which in turn determine macroscopic conduction. This framework establishes design principles for RRAM optimization by demonstrating that precise control of thermal conductivity and voltage modulation can regulate vacancy dynamics, ensuring reliable multilevel switching.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Advanced Memory and Neural Computing · Ferroelectric and Negative Capacitance Devices · Semiconductor materials and devices

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

School of Integrated Circuits Beijing Innovation Center for Future Chips BNRist Tsinghua University Beijing China

Funding

National Science and Technology Major Project · National Natural Science Foundation of China · Beijing Nova Program

References (34)

Show source-provided references
1. 10.1038/s41598‐023‐42214‐x10.1038/s41598‐023‐42214‐x ↗
2. 10.1039/D5TC01467K10.1039/D5TC01467K ↗
3. 10.1038/s41565‐020‐0655‐z10.1038/s41565‐020‐0655‐z ↗
4. 10.1038/s41467‐024‐45670‐910.1038/s41467‐024‐45670‐9 ↗
5. 10.1088/1361‐6641/abb84210.1088/1361‐6641/abb842 ↗
6. 10.1002/aisy.20220033810.1002/aisy.202200338 ↗
7. J.Zhang X.Ma Y.Xiet al. "A 28nm 4Mb Embedded RRAM IP with Record‐High Endurance of 107 Cycles and 10 Years@125°C Retention through Reliability‐Enhanced Design‐Technology Co‐Optimization " in2024 IEEE International Electron Devices Meeting (IEDM) (2024): 1‐4.10.1109/IEDM50854.2024.10873311 ↗
8. 10.1002/adma.20241391610.1002/adma.202413916 ↗
9. 10.1002/adfm.20170043210.1002/adfm.201700432 ↗
10. 10.1038/s41467‐021‐27575‐z10.1038/s41467‐021‐27575‐z ↗
11. 10.1002/adfm.20170692710.1002/adfm.201706927 ↗
12. 10.1103/PhysRevMaterials.2.11500210.1103/PhysRevMaterials.2.115002 ↗
13. 10.1038/nnano.2012.24010.1038/nnano.2012.240 ↗
14. 10.1002/adma.20220992510.1002/adma.202209925 ↗
15. A.Prakash J. S.Park J.Songet al. Multi‐state resistance switching and variability analysis of HfOx based RRAM for ultra‐high density memory applications "in2015 International Symposium on Next‐Generation Electronics (ISNE) (2015): 1‐2.10.1109/ISNE.2015.7132027 ↗
16. 10.1063/5.016010910.1063/5.0160109 ↗
17. 10.1063/1.501662610.1063/1.5016626 ↗
18. 10.1063/1.492866110.1063/1.4928661 ↗
19. 10.1021/acsaelm.3c0078210.1021/acsaelm.3c00782 ↗
20. 10.1038/nmat479910.1038/nmat4799 ↗
21. 10.1021/acsnano.8b0371210.1021/acsnano.8b03712 ↗
22. 10.1038/ncomms1563110.1038/ncomms15631 ↗
23. 10.1126/science.abq581510.1126/science.abq5815 ↗
24. 10.1002/adma.20210635910.1002/adma.202106359 ↗
25. 10.1103/PhysRevB.65.17411710.1103/PhysRevB.65.174117 ↗
26. W.Xia J.Cao D.Liu "Oxygen Vacancy Formation in Charged HfO2: An Ab Initio Study " in2022 International Conference on IC Design and Technology (ICICDT) (2022):16‐19.10.1109/ICICDT56182.2022.9933066 ↗
27. 10.1038/s41598‐019‐55219‐210.1038/s41598‐019‐55219‐2 ↗
28. 10.1063/1.296719410.1063/1.2967194 ↗
29. 10.1016/j.mssp.2015.02.07310.1016/j.mssp.2015.02.073 ↗
30. F.Xu B.Gao Y.Xiet al. "Atomic‐Device Hybrid Modeling of Relaxation Effect in Analog RRAM for Neuromorphic Computing " in2020 IEEE International Electron Devices Meeting (IEDM) (2020):13.12.11‐13.12.14.10.1109/IEDM13553.2020.9372114 ↗
31. 10.1063/1.492692110.1063/1.4926921 ↗
32. 10.1063/1.475124810.1063/1.4751248 ↗
33. D.Mu P.Liu Z.Zhouet al. "Multi‐Level RTN with Certain Regularities in Oxide‐RRAM: Experiments Defect Dynamics and 3D Multi‐Physics Modeling " in2025 IEEE International Reliability Physics Symposium (IRPS) (2025):1‐6.10.1109/IRPS48204.2025.10983298 ↗
34. 10.1088/0034‐4885/69/2/R0210.1088/0034‐4885/69/2/R02 ↗