(Invited) the Impact of Access Regions on Van Der Waals Heterostructure-Based Neuromorphic Devices for Tactile Sensing Applications
Shu-Ping Lin
摘要
In the somatosensory system, mechanoreceptors convert mechanical inputs into electrical signals that are transmitted to the brain. This study introduces a synaptic device designed to mimic Merkel discs, specialized mechanoreceptors found in human skin. The device employs a van der Waals heterostructure field-effect transistor with strategically positioned access regions (VHFET-AR). The morphological characteristics of the VHFET-AR devices were examined using high-resolution cross-sectional TEM, Raman spectroscopy, and AFM. Electrical performance was evaluated by measuring the currents after applying programming/erasing voltages of +50 V and -50 V for 1 second. Cycling endurance was tested at a drain-source voltage (V DS) of 50 mV, with repeated programming and erasing operations carried out under cyclic voltage pulses (V BG = ±50 V). The synaptic device successfully reproduces fundamental nervous system functions, including spike amplitude-dependent plasticity, spike duration-dependent plasticity, spike number-dependent plasticity, and slow adaptation. Additionally, it features inverse notch signaling and lateral inhibition at an approximate frequency of 11.23 Hz, aligning with the low-frequency stimuli (5–15 Hz) observed in biological Merkel discs. These functionalities are crucial for tactile localization and spatial discrimination. Through the synergistic integration of slow adaptation and inverse notch signaling with lateral inhibition, our device, incorporating stacked 2D materials and access regions, replicates the intricate functions of Merkel discs. Specifically developed to emulate lateral inhibition, this synaptic device enhances neuronal sensitivity to spatially discrete single-point stimulation while suppressing the responsiveness of surrounding areas. This advancement represents a significant step forward in mechanoreceptor technology, closely mimicking the complex mechanisms of the somatosensory system. Figure 1
