Tetrachromatic photonic synaptic arrays using WS2 monolayers for bioinspired neuromorphic retinas
image: Inspired by the butterfly tetrachromatic visual system, we develop an 8 × 8 retinal synapse array enabling color discrimination. The device demonstrates wavelength-sensitive synaptic behaviors with an ultralow power consumption of 2.28 aJ.
Credit: Nano Research Tsinghua University Press
How can machines “see” ultraviolet light and perceive the world like a butterfly? This vision is moving closer to reality with advances in neuromorphic vision systems. Researchers from Northeast Normal University in China, inspired by the superior visual system of butterflies, have successfully developed an artificial tetrachromatic vision optoelectronic synapse using a monolayer of the two-dimensional material tungsten disulfide (WS2). The work has been published in the journal Nano Research on February 23.
Human vision is trichromatic, based on red, green, and blue cone cells, and cannot perceive ultraviolet (UV) light. In contrast, many animals like butterflies, some birds, and fish possess tetrachromatic or even richer color vision. Their retinas contain more types of photoreceptor cells, allowing them to utilize the “invisible information” carried by UV light for efficient communication and survival. “Nature has already provided us with perfect solutions that surpass the limits of human vision,” explained corresponding author Prof. Haiyang Xu. “Our goal is to mimic this biological intelligence, endowing machine vision with the ability to sense and process ultraviolet spectral information, which will have profound impacts in environmental monitoring, medical diagnosis, autonomous driving, and more.”
The team chose monolayer WS2 as the core material due to its unique optoelectronic properties. “Monolayer WS2 is not only sensitive to visible light, but its band nesting effect and van Hove singularities in the Brillouin zone also give it a very strong absorption capacity for ultraviolet light,” said author Assoc Prof. Yuanzheng Li. “This means that with this single material, we can build a sensing unit responsive to a broad spectrum from UV to visible light. It greatly simplifies the device structure compared to previous approaches that required integrating complex additional UV-absorbing layers.”
The artificial synapse operates by mimicking information transfer between neurons in the biological brain. Light pulses act as presynaptic stimuli, generating photogenerated carriers in the WS2. By precisely controlling sulfur vacancy defects in the material, the team achieved controllable trapping and release of carriers, successfully emulating key learning and memory functions of biological synapses such as short-term potentiation (STP), long-term potentiation (LTP), and long-term depression (LTD). Remarkably, under 320 nm UV light stimulation, the energy consumption per synaptic event is as low as 2.28 aJ, significantly lower than the approximately 10 fJ consumption of human synaptic activity, demonstrating great potential for building ultra-low-power neuromorphic computing chips.
To verify its application potential in visual systems, the team fabricated an 8 × 8 retinal synaptic array comprising 64 synaptic units. The array successfully demonstrated the sensing and memory functions for images encoded by four colors: ultraviolet, red, green, and blue. More importantly, inspired by the information preprocessing mechanism of the biological retina, the array can simulate a visual attention mechanism, preferentially focusing on and extracting specific color features from an image. “For example, in a complex tetrachromatic image containing a red ‘2’, a green ‘4’, a blue ‘6’, and a purple ‘8’, our system can ‘focus its attention’ to recognize only the red feature, thereby clearly extracting the red digit ‘2’,” described Yuanzheng Li. This preprocessing effectively filters redundant information, dramatically increasing the subsequent neural network’s image recognition accuracy from about 45% without processing to over 98%. “This work not only demonstrates the great potential of monolayer WS2 in the field of bio-inspired optoelectronic integration but also provides a novel technological pathway for developing more efficient, intelligent, and lower-power next-generation machine vision systems,” concluded Assoc Prof. Yuanzheng Li.
Other contributors include Jixiu Li, Qingbin Wang, Chuxin Yan, Yongsheng Gao, Weixin, Weizhen Liu, and Yichun Liu from Northeast Normal University. The research was supported by the National Natural Science Fund for Distinguished Young Scholars (No. 52025022), the Program of National Natural Science Foundation of China (Nos. 12474421, 62275045, 12074060), the National Key R&D Program of China (No. 2023YFB3610200), and the Fund from Jilin Province (Nos. JJKH20241413KJ and 20240601049RC).
DOI Link:
https://doi.org/10.26599/NR.2026.94908375
About Nano Research
Nano Research is a peer-reviewed, open access, international and interdisciplinary research journal, sponsored by Tsinghua University and the Chinese Chemical Society, published by Tsinghua University Press on the platform SciOpen. It publishes original high-quality research and significant review articles on all aspects of nanoscience and nanotechnology, ranging from basic aspects of the science of nanoscale materials to practical applications of such materials. After 18 years of development, it has become one of the most influential academic journals in the nano field. Nano Research has published more than 1,000 papers every year from 2022, with its cumulative count surpassing 8,000 articles. In 2025 InCites Journal Citation Reports, its 2025 IF is 9.4 (8.3, 5 years), and it continues to be the Q1 area among the four subject classifications. Nano Research Award, established by Nano Research together with TUP and Springer Nature in 2013, and Nano Research Young Innovators (NR45) Awards, established by Nano Research in 2018, have become international academic awards with global influence.