Ferroelectric Neuromorphic Devices Enable Nonphotosensitive Neurons to Achieve Event-Driven Photosensitive Artificial Neural Interfaces
Abstract Neuromorphic devices integrated with biological neurons hold significant potential for rehabilitation and human-machine interfaces, thereby attracting extensive research interest. However, the effort to connect them through light stimulation is still in its very initial stages. Here, ferroelectric neuromorphic devices based on α-In2Se3 and P(VDF-TrFE) are proposed to construct hybrid neural interfaces. In contrast to neuromorphic devices employing other working principles, ferroelectricity plays a crucial role in connecting biological neurons, which enables nonphotosensitive neurons to respond to light stimuli. Various optoelectronic synaptic behaviors have been achieved. Under light pulses, the ferroelectric neuromorphic device induced a significant increase of up to 200% in the calcium ion signal of artificial neural interfaces. Patch-clamp was applied for further verification, revealing the influence of different light wavelengths and pulse frequencies on neuronal excitation signals, which is analogous to the human retina. Simulations demonstrate that the ferroelectric neural interface improves dynamic gesture recognition accuracy from 51% to 99%. Our approach offers a strategy to connect neuromorphic devices with neurons through light, which may offer a platform to investigate the fundamental principles of synaptic plasticity and learning in a hybrid setting for future remote therapy and human–machine interfaces.
Authors
- Weng Fu Io (ORCID: https://orcid.org/0000-0001-9339-5045)
- Jianhua Hao (ORCID: https://orcid.org/0000-0002-6186-5169)
- Zhaoying Dang (ORCID: https://orcid.org/0000-0002-5726-5208)
- Lei Sun (ORCID: https://orcid.org/0000-0001-7047-9529)
- Yong Wu (ORCID: https://orcid.org/0000-0002-9096-6946)
- Yunan Lin
- Feng Guo (ORCID: https://orcid.org/0000-0003-2830-4586)
- Di ZHAO (ORCID: https://orcid.org/0009-0005-5605-6331)
- Guoling Feng
Institutions
- Qingdao University (CN)
- Hong Kong Polytechnic University (HK)
Publication Details
- Journal
- ACS Nano
- Published
- 2026-11-13
- DOI
- https://doi.org/10.1021/acsnano.6c14245
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Research Grants Council, University Grants Committee
- Hong Kong Polytechnic University
- Natural Science Foundation of Shandong Province
- Taishan Scholar Project of Shandong Province