High-Accuracy Deep Neural Networks Using a Contralateral-Gated Analog Synapse Composed of Ultrathin MoS₂ nFET and Nonvolatile Charge-Trap Memory

摘要

The development of high-accuracy analog synapse deep neural networks entails devising novel materials and innovative memory structures. We demonstrated an analog synapse with contralateral gates based on a two-dimensional (2D) field-effect transistor and nonvolatile charge-trapmemory. Vertical integration of a MoS2-channel FET with a charge-trapping layer provided excellent charge controllability and gate-tunable nonvolatile storage. In the proposed contralateral-gate design, the read andwrite operations were separated to mitigate read disturb degradation. Reducing the MoS(2)channel thickness to the ultrathin scale allowed large threshold voltage shifts and on-resistance (R-ON) modulations. This vertically integrated MoS(2)synapse device exhibited 55 conductance states, high conductance max-min ratio (G(MAX)/G(MIN); similar to 50), low nonlinearity of alpha(p) = -0.81 and alpha(d) = -0.31, near ideal asymmetry of 0.5, and free of read disturb degradation. High neural network accuracy (>87%) is also obtained.