London – Researchers from the Institute for the Study of Nanostructured Materials of the National Research Council of Italy (CNR-ISMN), the University of Milano-Bicocca, and RWTH Aachen University in Germany have developed an innovative photonic synaptic transistor based on organic molecules. This device, called the OPST (Organic Photonic Synaptic Transistor), can process and store information using light, marking a significant step toward neuromorphic systems that mimic the human brain.
Published in Materials Horizons by the Royal Society of Chemistry, the OPST can detect optical and electrical signals, process them, store them, and learn from them—just like neurons in the brain. The innovation lies in the molecular design of the materials: instead of modifying only the device’s structure or the nanomaterials used, researchers designed the material starting at the molecular level.
“The molecules we used are light-sensitive persistent organic radicals,” explains Prof. Luca Beverina of the University of Milano-Bicocca, who synthesized these molecules. “Their electronic structure allows precise control over how they react to light pulses.” When exposed to different wavelengths, such as blue or infrared light, the transistor exhibits varied behaviors, emulating short- and long-term memory processes similar to biological synapses.
By simulating the photophysical properties of the device’s active layer, researchers demonstrated a direct link between memory processes and the electronic states activated in the organic radicals. Francesca Santoro of Forschungszentrum Jülich and RWTH Aachen University adds, “Thanks to its ability to respond to multiple signals simultaneously, the OPST can recognize and classify complex patterns, simulating the parallel learning typical of neuronal dendrites.”
This breakthrough paves the way for a new generation of photonic neuromorphic systems, where light becomes the primary medium for information processing, increasing speed and efficiency. Stefano Toffanin from CNR-ISMN notes, “The multifunctionality of these radicals could also support the development of intelligent displays. In the near future, these devices could integrate sensing, light emission, and neuromorphic learning in a single system.”