Low Energy Consumption Photoelectric Memristors with Multi‑Level Linear Conductance Modulation in Artificial Visual Systems Application

Published in Materials and Physics

Low Energy Consumption Photoelectric Memristors with Multi‑Level Linear Conductance Modulation in Artificial Visual Systems Application
Like

Share this post

Choose a social network to share with, or copy the URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Explore the Research

SpringerLink
SpringerLink SpringerLink

Low Energy Consumption Photoelectric Memristors with Multi-Level Linear Conductance Modulation in Artificial Visual Systems Application - Nano-Micro Letters

Optical synapses have an ability to perceive and remember visual information, making them expected to provide more intelligent and efficient visual solutions for humans. As a new type of artificial visual sensory devices, photoelectric memristors can fully simulate synaptic performance and have great prospects in the development of biological vision. However, due to the urgent problems of nonlinear conductance and high-energy consumption, its further application in high-precision control scenarios and integration is hindered. In this work, we report an optoelectronic memristor with a structure of TiN/CeO2/ZnO/ITO/Mica, which can achieve minimal energy consumption (187 pJ) at a single pulse (0.5 V, 5 ms). Under the stimulation of continuous pulses, linearity can be achieved up to 99.6%. In addition, the device has a variety of synaptic functions under the combined action of photoelectric, which can be used for advanced vision. By utilizing its typical long-term memory characteristics, we achieved image recognition and long-term memory in a 3 × 3 synaptic array and further achieved female facial feature extraction behavior with an activation rate of over 92%. Moreover, we also use the linear response characteristic of the device to design and implement the night meeting behavior of autonomous vehicles based on the hardware platform. This work highlights the potential of photoelectric memristors for advancing neuromorphic vision systems, offering a new direction for bionic eyes and visual automation technology.

As artificial intelligence races toward real-time, on-sensor vision, conventional CMOS cameras face a "power wall" and a "memory wall." Now, researchers from Hebei University, led by Professor Xiaobing Yan, have unveiled a low-energy photoelectric memristor that acts like an optical synapse, offering a one-chip solution to sense, store and process visual information the way our retina does.

Why Low-Energy Photo-Memristors Matter

  • Ultra-low energy: a single 0.5 V, 5 ms optical pulse consumes only 187 pJ—far below the ~900 pJ of a CMOS spike.
  • Linear learning: conductance updates show 99.6 % linearity, eliminating redundant weight tuning and cutting system power.
  • Retina-like functions: paired-pulse facilitation, long-term plasticity and "learning-experiencing" behaviors are reproduced under visible light.

Innovative Design & Features

  • TiN/CeO2/ZnO/ITO/Mica stack: the CeO2/ZnO interface traps photo-carriers, dynamically lowering the barrier and enabling multi-level, linear conductance modulation.
  • Two-terminal cross-point structure: compatible with dense 64 × 64 arrays and simple 1T1R or 1S1R back-end-of-line integration.
  • Wide-band response: defect-rich CeO2extends sensitivity from UV to visible, allowing outdoor and night-time operation without filters.

Applications & Future Outlook

  • Image memory: a 3 × 3 array memorizes the letter "L" for >8,000 s after only 10 exposures—mimicking human long-term visual memory.
  • Facial recognition: a 64 × 64 artificial retina achieves >92 % activation on female facial features and maintains high accuracy even under 20 % optical noise.
  • Autonomous night driving: a hardware demo uses linear optical synaptic weights to trigger safe bidirectional vehicle rendezvous without external illumination.

This work charts a clear roadmap for integrating sense-memory-compute pixels into bionic eyes, smart cameras and neuromorphic edge devices. Expect slimmer, smarter visual systems as the team scales the arrays toward megapixel chips and explores new oxide stacks for even lower energy and higher speed.

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Follow the Topic

Optoelectronic Devices
Physical Sciences > Physics and Astronomy > Optics and Photonics > Optoelectronic Devices
Optics and Photonics
Physical Sciences > Physics and Astronomy > Optics and Photonics
Materials for Devices
Physical Sciences > Materials Science > Materials for Devices
Visual system
Physical Sciences > Physics and Astronomy > Biophysics > Sensory Systems > Visual system
  • Nano-Micro Letters Nano-Micro Letters

    Nano-Micro Letters is a peer-reviewed, international, interdisciplinary and open-access journal that focus on science, experiments, engineering, technologies and applications of nano- or microscale structure and system in physics, chemistry, biology, material science, and pharmacy.

Your space to connect: The Polarised light Hub

A new Communities’ space to connect, collaborate, and explore research on Light-Matter Interaction, Optics and Photonics, Quantum Imaging and Sensing, Microscopy, and Spectroscopy!

Continue reading announcement