The moiré materials, formed by stacking two-dimensional atomic crystals, are new artificial quantum materials with novel strong correlation and topological properties. In stark contrast to conventional quantum materials, the moiré materials host exotic quantum states (such as correlated insulators, orbital magnetism, interfacial ferroelectricity, etc.) and can be tuned by manipulating accessible external stimuli such as electrical fields, light, or strain. This makes the moiré material systems become ideal platforms for not only exploring novel quantum states but also developing moiré electronics. Current related research mainly focuses on the exploration and tuning of new quantum states in the moiré materials. How to use the unique quantum states and tunability of the moiré materials to design moiré electronic devices is a topic of widespread concern.
The team co-led by Professor Shi-Jun Liang and Professor Feng Miao from the School of Physics at Nanjing University reported a moiré synaptic transistor for the first time. This transistor has been fabricated by using h-BN/graphene/h-BN moiré heterostructure, and can faithfully emulate the long-term plasticity and short-term plasticity of the biological synapse. By exploiting the tunable dynamics of moiré synaptic transistor, they proposed a full-moiré physical neural network to implement homogeneous-architecture reservoir computing, which solves the long-standing challenge that traditional reservoir computing must be implemented with heterostructure architecture. This work represents the initial effort in developing neuromorphic computing hardware based on the moiré quantum materials and opens up a promising avenue for the development of moiré electronics. The related research result, titled " Moiré synaptic transistor for homogeneous-architecture reservoir computing," was published in Express Letter column of Chinese Physics Letters on October 12, 2023.
Figure. The synaptic transistor is fabricated based on a h-BN/bilayer graphene/h-BN moiré heterostructure, exhibiting ferroelectricity-like hysteretic gate voltage dependence of resistance. Varying the magnitude of the gate voltage enables the moiré transistor to be switched between long-term memory and short-term memory with nonlinear dynamics.