Layered indium selenide (InSe) emerges as a promising candidate for next-generation optoelectronics, yet achieving dynamically reconfigurable near-infrared light sources and photodetectors in these devices remains challenging. Here, we demonstrate a strain-engineered bifunctional device based on a vertical graphene/γ-InSe/graphene Schottky junction on a flexible polypropylene (PP) substrate. By exploiting the large thermal expansion coefficient (CTE) mismatch between InSe and PP, reversible biaxial strain is precisely induced in situ via thermal control. Under electro-optic operation, compressive strain widens InSe’s bandgap via lattice distortion, generating a pronounced blueshift (up to 65 meV) in the electroluminescence (EL) peak. For opto-electronic conversion, strain-mediated modulation of Schottky barrier height suppresses dark current while modulating photocurrent and photoresponsivity under 532 nm and 980 nm illumination. This work establishes strain as a new strategy for reconfigurable InSe-based optoelectronic devices.
Emulating high-order visual cognitive functions in optoelectronic neuromorphic synapse devices is important for advanced artificial vision systems. Herein, high-quality lead-free vacancy-ordered double perovskite Cs 2 TiCl 6 single crystals are synthesized to construct optoelectronically co-modulated neuromorphic synapse devices. Electrically, the devices exhibit bipolar resistive switching behavior. Optically, the devices show prominent light-induced synaptic potentiation behaviors. Furthermore, three biomimetic high-order visual cognitive modes are implemented by designing specific asynchronous optoelectronic modulation sequences. Benefiting from the effective noise suppression capability of these modes, the pattern recognition accuracy of the neural network can be improved from the baseline of 31.95% to 74.71%, 69.55%, and 86.52%, respectively. This work provides an innovative material candidate and architectural platform for efficient and noise-tolerant advanced artificial visual perception systems.
