Related Experiment Video
Updated: Aug 5, 2026
07:20
Fabrication of Micro-Patterned Chip with Controlled Thickness for High-Throughput Cryogenic Electron Microscopy
Published on: April 21, 2022
Defect-engineered boron nitride memristors with enhanced uniformity and stability for CNN-based image encryption and
Xuan Chen1, Zi Li2, Tingting Guo3
1MIIT Key Laboratory of Advanced Display Materials and Devices, School of Materials Science and Engineering, Nanjing University of Science and Technology, Nanjing, Jiangsu 210094, China. xiufengsoong@njust.edu.cn.
Nanoscale
|July 29, 2026
Summary
Oxygen plasma treatment improves hexagonal boron nitride (h-BN) memristors for reliable resistive switching and secure data encryption. This defect engineering enhances uniformity and high-temperature stability, enabling new hardware security applications.
Area of Science:
- Materials Science
- Nanotechnology
- Solid-State Electronics
Background:
- Hexagonal boron nitride (h-BN) memristors show potential for resistive switching but suffer from poor device uniformity.
- Existing memristor technologies require improved stability and uniformity for practical applications.
Purpose of the Study:
- To enhance the device-to-device uniformity and thermal stability of h-BN memristors.
- To explore the application of engineered h-BN memristors in hardware security, specifically for generating true random keys.
Main Methods:
- Employing oxygen plasma treatment to introduce controlled defects (oxygen-substitutional doping, vacancies) in h-BN.
- Characterizing the resistive switching behavior and uniformity of the treated h-BN memristors.
- Validating the use of memristor-generated random keys for image encryption using a convolutional neural network (CNN).
Main Results:
- Significantly improved switching uniformity with a SET voltage coefficient of variation below 6.12%.
- Stable resistive switching performance maintained up to 613 K.
- Demonstrated generation of true random keys from the low-resistance state for effective image encryption, rendering encrypted images unrecognizable by CNNs.
Conclusions:
- Oxygen plasma treatment is an effective defect-engineering strategy to enhance h-BN memristor performance.
- Engineered h-BN memristors exhibit potential for high-temperature and hardware security applications, including secure data encryption.
- The controlled introduction of defects facilitates stable conductive filament formation, crucial for reliable memristor operation.
