Back‐End‐of‐Line SiC‐Based Memristor for Resistive Memory and Artificial Synapse

Abstract

Two‐terminal memristor has emerged as one of the most promising neuromorphic artificial electronic devices for their structural resemblance to biological synapses and ability to emulate many synaptic functions. In this work, a memristor based on the back‐end‐of‐line (BEOL) material silicon carbide (SiC) is developed. The thin film memristors demonstrate excellent binary resistive switching with compliance‐free and self‐rectifying characteristics which are advantageous for the implementation of high‐density 3D crossbar memory architectures. The conductance of this SiC‐based memristor can be modulated gradually through the application of both DC and AC signals. This behavior is demonstrated to further emulate several vital synaptic functions including paired‐pulse facilitation (PPF), post‐tetanic potentiation (PTP), short‐term potentiation (STP), and spike‐rate‐dependent plasticity (SRDP). The synaptic function of learning‐forgetting‐relearning processes is successfully emulated and demonstrated using a 3 × 3 artificial synapse array. This work presents an important advance in SiC‐based memristor and its application in both memory and neuromorphic computing.

AlGaN/GaN‐Based Optoelectronic Synaptic Devices for Neuromorphic Computing

“…The results show that the EPSC amplitude (A 2 ) stimulated by the second pulse is larger than that stimulated by the first one (A 1 ), demonstrating a facilitation process. [8,28] Figure 3c shows the dependence of the value of A 2 /A 1 (PPF index) on ΔT. The PPF index decays exponentially as ΔT increases, which is similar to the information processing in biological synapse.…”

“…ensure the reliability of data storage, the long-term memory time of optoelectronic synaptic devices should be greater than 10 years. Up to now, the memory time of optoelectronic synaptic devices is usually in the range from tens of seconds to more than 10 4 s. [7,8] In addition, a way to reduce the power consumption of optoelectronic synaptic devices is to reduce the width and optical power density of a stimulating optical pulse. [2] However, such a method leads to the degradation of device performance (e.g., smaller photocurrent and lower on-off ratio [9] ) due to the limited light sensitivity of the device.…”

A review of memristor: material and structure design, device performance, applications and prospects

“…Kapur et al fabricated a memristor based on silicon carbide (SiC) and the conductance could be modulated gradually. Moreover, the synaptic function of learning-forgetting-relearning processe was successfully emulated and demonstrated using a 3 × 3 array of SiC-based memristors ( Figure 8(b) ) [ 189 ]. Ali et al fabricated memristors with Ag/GeSe/Pt/Ti/SiO 2 structure and simulated synaptic functions, such as long time range enhancement (LTP), long time range inhibition (LTD) ( Figure 8(c) ), paired pulse facilitation (PPF) ( Figure 8(d) ) and others [ 190 ].…”

“…Our group prepared a Pt/HfO 2 /BFO/HfO 2 /TiN artificial synapse, realizing the physiological synapse STDP characteristics ( Figure 8(e) ) [ 16 ]. Furthermore, the memristors simulated other synaptic functions, such as short-term potentiation, post-tetanic potentiation, nonassociative learning, associative learning, synaptic scaling and spike-rate-dependent plasticity [ 189 ].

Figure 8.

…”

Non-volatile memristor-based artificial synaptic behavior of redox-active organic composites

“…Resistive memristors show different synaptic functions such as short-term plasticity (STP), long-term plasticity (LTP), paired pulse facilitation (PPF), learning, forgetting and relearning, etc. [25][26][27] Oxide-based resistive memory and neuromorphic devices have been demonstrated as potential candidates for application in the field of artificial intelligence. 28,29 The optimum current of oxide-based devices is high, which increases the overall power consumption of the system.…”