Floating gate photo-memory devices based on van der Waals heterostructures for neuromorphic image recognition

Two-dimensional (2D) materials with reconfigurable properties show potential in neuromorphic hardware applications. However, most 2D materials-based neuromorphic hardware is volatile, which needs large energy to accomplish perception functions. Here, we report on nonvolatile floating gate photo-memory devices based on ReS2/h-BN/SnS2 van der Waals heterostructures. The devices exhibit a large memory window of similar to 60 V, a high program/erase current ratio of similar to 10(7) with excellent retention characteristics, a low off-state current of 7.4 x 10(-13) A, and a high detectivity of 1.98 x 10(13) cm Hz(1/2) W-1, allowing for multi-bit information storage. For the multi-level storage capacity, 27 photo-memory states are obtained by pulsed laser illumination. Moreover, a neuromorphic computing network is also constructed based on the photo-memory devices with a maximum recognition accuracy of up to 90%. This work paves the way for miniaturization and high-density integration of future optoelectronics for neuromorphic hardware applications.

Automated summary

This study reports nonvolatile floating gate photo-memory devices based on ReS2/h-BN/SnS2 van der Waals heterostructures, which exhibit a large memory window, a high program/erase current ratio, excellent retention characteristics, and the ability to achieve multi-bit information storage. Pulsed laser illumination allows for obtaining multiple photo-memory states. Additionally, a neuromorphic computing network is constructed based on these devices with high recognition accuracy. This work paves the way for miniaturization and high-density integration of future optoelectronics for neuromorphic hardware applications.